Namespace(aa='rand-m9-mstd0.5-inc1', auto_resume=True, batch_size=256, cfg_path='iFormer_l2_distill.yaml', clip_grad=None, color_jitter=0.4, crop_pct=None, cutmix=1.0, cutmix_minmax=None, data_path='imagenet', data_set='IMNET', device='cuda', disable_eval=False, dist_backend='nccl', dist_eval=True, dist_on_itp=False, dist_url='env://', distillation_alpha=0.5, distillation_tau=1.0, distillation_type='hard', distributed=True, drop_path=0, enable_wandb=False, epochs=450, eval=False, eval_data_path=None, finetune='', gpu=0, head_init_scale=1.0, imagenet_default_mean_and_std=True, input_size=224, layer_decay=1.0, layer_scale_init_value=0, local_rank=-1, log_dir=None, lr=0.004, min_lr=1e-06, mixup=0.8, mixup_mode='batch', mixup_prob=1.0, mixup_switch_prob=0.5, model='iFormer_l2', model_ema=True, model_ema_decay=0.9999, model_ema_eval=True, model_ema_force_cpu=False, model_key='model|module', model_prefix='', momentum=0.9, nb_classes=1000, num_workers=16, opt='adamw', opt_betas=None, opt_eps=1e-08, output_dir='', pin_mem=True, project='iFormer', rank=0, recount=1, remode='pixel', reprob=0.25, resplit=False, resume='', save_ckpt=True, save_ckpt_freq=1, save_ckpt_num=3, seed=0, smoothing=0.1, start_epoch=0, teacher_model='regnety_160', teacher_path='regnety_160-a5fe301d.pth', train_interpolation='bicubic', update_freq=1, use_amp=False, wandb_ckpt=False, warmup_epochs=20, warmup_steps=-1, weight_decay=0.05, weight_decay_end=None, world_size=16) Transform = RandomResizedCropAndInterpolation(size=(224, 224), scale=(0.08, 1.0), ratio=(0.75, 1.3333), interpolation=PIL.Image.BICUBIC) RandomHorizontalFlip(p=0.5) ToTensor() Normalize(mean=tensor([0.4850, 0.4560, 0.4060]), std=tensor([0.2290, 0.2240, 0.2250])) --------------------------- reading from datapath imagenet Number of the class = 1000 Transform = Resize(size=256, interpolation=bicubic, max_size=None, antialias=warn) CenterCrop(size=(224, 224)) ToTensor() Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)) --------------------------- reading from datapath imagenet Number of the class = 1000 Sampler_train = Mixup is activated! Using EMA with decay = 0.99990000 Model = iFormer( (downsample_layers): ModuleList( (0): Sequential( (0): Conv2d_BN( (c): Conv2d(3, 32, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), bias=False) (bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): EdgeResidual( (conv_exp_bn1): Conv2d_BN( (c): Conv2d(32, 128, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (act): GELU(approximate='none') (conv_pwl_bn2): Conv2d_BN( (c): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) (1): Sequential( (0): Conv2d_BN( (c): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (2): Sequential( (0): Conv2d_BN( (c): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) (3): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) (stages): ModuleList( (0): Sequential( (0): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(64, 64, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=64, bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (1): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(64, 64, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=64, bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (2): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(64, 64, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=64, bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) ) (1): Sequential( (0): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (1): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (2): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(128, 128, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=128, bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) ) (2): Sequential( (0): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (1): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (2): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (3): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (4): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (5): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (6): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (7): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (8): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (9): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (10): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (11): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (12): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (13): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (14): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (15): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (16): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (17): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (18): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (19): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (20): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (21): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (22): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (23): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (24): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (25): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (26): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (27): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (28): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (29): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (30): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (31): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (32): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (33): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (34): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (35): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (36): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (37): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (38): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (39): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (40): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (41): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (42): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (43): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(256, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (44): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 768, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(768, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(768, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (45): BasicBlock( (block): ConvBlock( (token_channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(256, 256, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=256, bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): Conv2d_BN( (c): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (2): GELU(approximate='none') (3): Conv2d_BN( (c): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) ) (3): Sequential( (0): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (1): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (2): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(512, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(1536, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (3): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (4): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (5): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(512, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(1536, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (6): BasicBlock( (block): RepCPE( (cpe): Residual( (m): Conv2d_BN( (c): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) (7): BasicBlock( (block): SHMABlock( (token_channel_mixer): Residual( (m): GAU2dv2( (q): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (k): Conv2d_BN( (c): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (gate_act): Sigmoid() (attn_drop): Dropout(p=0.0, inplace=False) (proj): Conv2d_BN( (c): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (v_gate): Conv2d_BN( (c): Conv2d(512, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) (8): BasicBlock( (block): FFN2d( (channel_mixer): Residual( (m): Sequential( (0): Conv2d_BN( (c): Conv2d(512, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(1536, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) (1): GELU(approximate='none') (2): Conv2d_BN( (c): Conv2d(1536, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) ) ) ) ) ) ) ) (classifier): Classfier( (classifier): BN_Linear( (bn): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (l): Linear(in_features=512, out_features=1000, bias=True) ) (classifier_dist): BN_Linear( (bn): BatchNorm1d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) (l): Linear(in_features=512, out_features=1000, bias=True) ) ) ) number of params: 25033008 LR = 0.00400000 Batch size = 4096 Update frequent = 1 Number of training examples = 1281167 Number of training training per epoch = 312 Param groups = { "decay": { "weight_decay": 0.05, "params": [ "downsample_layers.0.0.c.weight", "downsample_layers.0.2.conv_exp_bn1.c.weight", "downsample_layers.0.2.conv_pwl_bn2.c.weight", "downsample_layers.1.0.c.weight", "downsample_layers.2.0.c.weight", "downsample_layers.3.0.c.weight", "stages.0.0.block.token_channel_mixer.m.0.c.weight", "stages.0.0.block.token_channel_mixer.m.1.c.weight", "stages.0.0.block.token_channel_mixer.m.3.c.weight", "stages.0.1.block.token_channel_mixer.m.0.c.weight", "stages.0.1.block.token_channel_mixer.m.1.c.weight", "stages.0.1.block.token_channel_mixer.m.3.c.weight", "stages.0.2.block.token_channel_mixer.m.0.c.weight", "stages.0.2.block.token_channel_mixer.m.1.c.weight", "stages.0.2.block.token_channel_mixer.m.3.c.weight", "stages.1.0.block.token_channel_mixer.m.0.c.weight", "stages.1.0.block.token_channel_mixer.m.1.c.weight", "stages.1.0.block.token_channel_mixer.m.3.c.weight", "stages.1.1.block.token_channel_mixer.m.0.c.weight", "stages.1.1.block.token_channel_mixer.m.1.c.weight", "stages.1.1.block.token_channel_mixer.m.3.c.weight", "stages.1.2.block.token_channel_mixer.m.0.c.weight", "stages.1.2.block.token_channel_mixer.m.1.c.weight", "stages.1.2.block.token_channel_mixer.m.3.c.weight", "stages.2.0.block.token_channel_mixer.m.0.c.weight", "stages.2.0.block.token_channel_mixer.m.1.c.weight", "stages.2.0.block.token_channel_mixer.m.3.c.weight", "stages.2.1.block.token_channel_mixer.m.0.c.weight", "stages.2.1.block.token_channel_mixer.m.1.c.weight", "stages.2.1.block.token_channel_mixer.m.3.c.weight", "stages.2.2.block.token_channel_mixer.m.0.c.weight", "stages.2.2.block.token_channel_mixer.m.1.c.weight", "stages.2.2.block.token_channel_mixer.m.3.c.weight", "stages.2.3.block.token_channel_mixer.m.0.c.weight", "stages.2.3.block.token_channel_mixer.m.1.c.weight", "stages.2.3.block.token_channel_mixer.m.3.c.weight", "stages.2.4.block.token_channel_mixer.m.0.c.weight", "stages.2.4.block.token_channel_mixer.m.1.c.weight", "stages.2.4.block.token_channel_mixer.m.3.c.weight", "stages.2.5.block.token_channel_mixer.m.0.c.weight", "stages.2.5.block.token_channel_mixer.m.1.c.weight", "stages.2.5.block.token_channel_mixer.m.3.c.weight", "stages.2.6.block.token_channel_mixer.m.0.c.weight", "stages.2.6.block.token_channel_mixer.m.1.c.weight", "stages.2.6.block.token_channel_mixer.m.3.c.weight", "stages.2.7.block.token_channel_mixer.m.0.c.weight", "stages.2.7.block.token_channel_mixer.m.1.c.weight", "stages.2.7.block.token_channel_mixer.m.3.c.weight", "stages.2.8.block.token_channel_mixer.m.0.c.weight", "stages.2.8.block.token_channel_mixer.m.1.c.weight", "stages.2.8.block.token_channel_mixer.m.3.c.weight", "stages.2.9.block.token_channel_mixer.m.0.c.weight", "stages.2.9.block.token_channel_mixer.m.1.c.weight", "stages.2.9.block.token_channel_mixer.m.3.c.weight", "stages.2.10.block.token_channel_mixer.m.0.c.weight", "stages.2.10.block.token_channel_mixer.m.1.c.weight", "stages.2.10.block.token_channel_mixer.m.3.c.weight", "stages.2.11.block.token_channel_mixer.m.0.c.weight", "stages.2.11.block.token_channel_mixer.m.1.c.weight", "stages.2.11.block.token_channel_mixer.m.3.c.weight", "stages.2.12.block.cpe.m.c.weight", "stages.2.13.block.token_channel_mixer.m.q.c.weight", "stages.2.13.block.token_channel_mixer.m.k.c.weight", "stages.2.13.block.token_channel_mixer.m.proj.c.weight", "stages.2.13.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.14.block.channel_mixer.m.0.c.weight", "stages.2.14.block.channel_mixer.m.2.c.weight", "stages.2.15.block.cpe.m.c.weight", "stages.2.16.block.token_channel_mixer.m.q.c.weight", "stages.2.16.block.token_channel_mixer.m.k.c.weight", "stages.2.16.block.token_channel_mixer.m.proj.c.weight", "stages.2.16.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.17.block.channel_mixer.m.0.c.weight", "stages.2.17.block.channel_mixer.m.2.c.weight", "stages.2.18.block.cpe.m.c.weight", "stages.2.19.block.token_channel_mixer.m.q.c.weight", "stages.2.19.block.token_channel_mixer.m.k.c.weight", "stages.2.19.block.token_channel_mixer.m.proj.c.weight", "stages.2.19.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.20.block.channel_mixer.m.0.c.weight", "stages.2.20.block.channel_mixer.m.2.c.weight", "stages.2.21.block.cpe.m.c.weight", "stages.2.22.block.token_channel_mixer.m.q.c.weight", "stages.2.22.block.token_channel_mixer.m.k.c.weight", "stages.2.22.block.token_channel_mixer.m.proj.c.weight", "stages.2.22.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.23.block.channel_mixer.m.0.c.weight", "stages.2.23.block.channel_mixer.m.2.c.weight", "stages.2.24.block.cpe.m.c.weight", "stages.2.25.block.token_channel_mixer.m.q.c.weight", "stages.2.25.block.token_channel_mixer.m.k.c.weight", "stages.2.25.block.token_channel_mixer.m.proj.c.weight", "stages.2.25.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.26.block.channel_mixer.m.0.c.weight", "stages.2.26.block.channel_mixer.m.2.c.weight", "stages.2.27.block.cpe.m.c.weight", "stages.2.28.block.token_channel_mixer.m.q.c.weight", "stages.2.28.block.token_channel_mixer.m.k.c.weight", "stages.2.28.block.token_channel_mixer.m.proj.c.weight", "stages.2.28.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.29.block.channel_mixer.m.0.c.weight", "stages.2.29.block.channel_mixer.m.2.c.weight", "stages.2.30.block.cpe.m.c.weight", "stages.2.31.block.token_channel_mixer.m.q.c.weight", "stages.2.31.block.token_channel_mixer.m.k.c.weight", "stages.2.31.block.token_channel_mixer.m.proj.c.weight", "stages.2.31.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.32.block.channel_mixer.m.0.c.weight", "stages.2.32.block.channel_mixer.m.2.c.weight", "stages.2.33.block.cpe.m.c.weight", "stages.2.34.block.token_channel_mixer.m.q.c.weight", "stages.2.34.block.token_channel_mixer.m.k.c.weight", "stages.2.34.block.token_channel_mixer.m.proj.c.weight", "stages.2.34.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.35.block.channel_mixer.m.0.c.weight", "stages.2.35.block.channel_mixer.m.2.c.weight", "stages.2.36.block.cpe.m.c.weight", "stages.2.37.block.token_channel_mixer.m.q.c.weight", "stages.2.37.block.token_channel_mixer.m.k.c.weight", "stages.2.37.block.token_channel_mixer.m.proj.c.weight", "stages.2.37.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.38.block.channel_mixer.m.0.c.weight", "stages.2.38.block.channel_mixer.m.2.c.weight", "stages.2.39.block.cpe.m.c.weight", "stages.2.40.block.token_channel_mixer.m.q.c.weight", "stages.2.40.block.token_channel_mixer.m.k.c.weight", "stages.2.40.block.token_channel_mixer.m.proj.c.weight", "stages.2.40.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.41.block.channel_mixer.m.0.c.weight", "stages.2.41.block.channel_mixer.m.2.c.weight", "stages.2.42.block.cpe.m.c.weight", "stages.2.43.block.token_channel_mixer.m.q.c.weight", "stages.2.43.block.token_channel_mixer.m.k.c.weight", "stages.2.43.block.token_channel_mixer.m.proj.c.weight", "stages.2.43.block.token_channel_mixer.m.v_gate.c.weight", "stages.2.44.block.channel_mixer.m.0.c.weight", "stages.2.44.block.channel_mixer.m.2.c.weight", "stages.2.45.block.token_channel_mixer.m.0.c.weight", "stages.2.45.block.token_channel_mixer.m.1.c.weight", "stages.2.45.block.token_channel_mixer.m.3.c.weight", "stages.3.0.block.cpe.m.c.weight", "stages.3.1.block.token_channel_mixer.m.q.c.weight", "stages.3.1.block.token_channel_mixer.m.k.c.weight", "stages.3.1.block.token_channel_mixer.m.proj.c.weight", "stages.3.1.block.token_channel_mixer.m.v_gate.c.weight", "stages.3.2.block.channel_mixer.m.0.c.weight", "stages.3.2.block.channel_mixer.m.2.c.weight", "stages.3.3.block.cpe.m.c.weight", "stages.3.4.block.token_channel_mixer.m.q.c.weight", "stages.3.4.block.token_channel_mixer.m.k.c.weight", "stages.3.4.block.token_channel_mixer.m.proj.c.weight", "stages.3.4.block.token_channel_mixer.m.v_gate.c.weight", "stages.3.5.block.channel_mixer.m.0.c.weight", "stages.3.5.block.channel_mixer.m.2.c.weight", "stages.3.6.block.cpe.m.c.weight", "stages.3.7.block.token_channel_mixer.m.q.c.weight", "stages.3.7.block.token_channel_mixer.m.k.c.weight", "stages.3.7.block.token_channel_mixer.m.proj.c.weight", "stages.3.7.block.token_channel_mixer.m.v_gate.c.weight", "stages.3.8.block.channel_mixer.m.0.c.weight", "stages.3.8.block.channel_mixer.m.2.c.weight", "classifier.classifier.l.weight", "classifier.classifier_dist.l.weight" ], "lr_scale": 1.0 }, "no_decay": { "weight_decay": 0.0, "params": [ "downsample_layers.0.0.bn.weight", "downsample_layers.0.0.bn.bias", "downsample_layers.0.2.conv_exp_bn1.bn.weight", "downsample_layers.0.2.conv_exp_bn1.bn.bias", "downsample_layers.0.2.conv_pwl_bn2.bn.weight", "downsample_layers.0.2.conv_pwl_bn2.bn.bias", "downsample_layers.1.0.bn.weight", "downsample_layers.1.0.bn.bias", "downsample_layers.2.0.bn.weight", "downsample_layers.2.0.bn.bias", "downsample_layers.3.0.bn.weight", "downsample_layers.3.0.bn.bias", "stages.0.0.block.token_channel_mixer.m.0.bn.weight", "stages.0.0.block.token_channel_mixer.m.0.bn.bias", "stages.0.0.block.token_channel_mixer.m.1.bn.weight", "stages.0.0.block.token_channel_mixer.m.1.bn.bias", "stages.0.0.block.token_channel_mixer.m.3.bn.weight", "stages.0.0.block.token_channel_mixer.m.3.bn.bias", "stages.0.1.block.token_channel_mixer.m.0.bn.weight", "stages.0.1.block.token_channel_mixer.m.0.bn.bias", "stages.0.1.block.token_channel_mixer.m.1.bn.weight", "stages.0.1.block.token_channel_mixer.m.1.bn.bias", "stages.0.1.block.token_channel_mixer.m.3.bn.weight", "stages.0.1.block.token_channel_mixer.m.3.bn.bias", "stages.0.2.block.token_channel_mixer.m.0.bn.weight", "stages.0.2.block.token_channel_mixer.m.0.bn.bias", "stages.0.2.block.token_channel_mixer.m.1.bn.weight", "stages.0.2.block.token_channel_mixer.m.1.bn.bias", "stages.0.2.block.token_channel_mixer.m.3.bn.weight", "stages.0.2.block.token_channel_mixer.m.3.bn.bias", "stages.1.0.block.token_channel_mixer.m.0.bn.weight", "stages.1.0.block.token_channel_mixer.m.0.bn.bias", "stages.1.0.block.token_channel_mixer.m.1.bn.weight", "stages.1.0.block.token_channel_mixer.m.1.bn.bias", "stages.1.0.block.token_channel_mixer.m.3.bn.weight", "stages.1.0.block.token_channel_mixer.m.3.bn.bias", "stages.1.1.block.token_channel_mixer.m.0.bn.weight", "stages.1.1.block.token_channel_mixer.m.0.bn.bias", "stages.1.1.block.token_channel_mixer.m.1.bn.weight", "stages.1.1.block.token_channel_mixer.m.1.bn.bias", "stages.1.1.block.token_channel_mixer.m.3.bn.weight", "stages.1.1.block.token_channel_mixer.m.3.bn.bias", "stages.1.2.block.token_channel_mixer.m.0.bn.weight", "stages.1.2.block.token_channel_mixer.m.0.bn.bias", "stages.1.2.block.token_channel_mixer.m.1.bn.weight", "stages.1.2.block.token_channel_mixer.m.1.bn.bias", "stages.1.2.block.token_channel_mixer.m.3.bn.weight", "stages.1.2.block.token_channel_mixer.m.3.bn.bias", "stages.2.0.block.token_channel_mixer.m.0.bn.weight", "stages.2.0.block.token_channel_mixer.m.0.bn.bias", "stages.2.0.block.token_channel_mixer.m.1.bn.weight", "stages.2.0.block.token_channel_mixer.m.1.bn.bias", "stages.2.0.block.token_channel_mixer.m.3.bn.weight", "stages.2.0.block.token_channel_mixer.m.3.bn.bias", "stages.2.1.block.token_channel_mixer.m.0.bn.weight", "stages.2.1.block.token_channel_mixer.m.0.bn.bias", "stages.2.1.block.token_channel_mixer.m.1.bn.weight", "stages.2.1.block.token_channel_mixer.m.1.bn.bias", "stages.2.1.block.token_channel_mixer.m.3.bn.weight", "stages.2.1.block.token_channel_mixer.m.3.bn.bias", "stages.2.2.block.token_channel_mixer.m.0.bn.weight", "stages.2.2.block.token_channel_mixer.m.0.bn.bias", "stages.2.2.block.token_channel_mixer.m.1.bn.weight", "stages.2.2.block.token_channel_mixer.m.1.bn.bias", "stages.2.2.block.token_channel_mixer.m.3.bn.weight", "stages.2.2.block.token_channel_mixer.m.3.bn.bias", "stages.2.3.block.token_channel_mixer.m.0.bn.weight", "stages.2.3.block.token_channel_mixer.m.0.bn.bias", "stages.2.3.block.token_channel_mixer.m.1.bn.weight", "stages.2.3.block.token_channel_mixer.m.1.bn.bias", "stages.2.3.block.token_channel_mixer.m.3.bn.weight", "stages.2.3.block.token_channel_mixer.m.3.bn.bias", "stages.2.4.block.token_channel_mixer.m.0.bn.weight", "stages.2.4.block.token_channel_mixer.m.0.bn.bias", "stages.2.4.block.token_channel_mixer.m.1.bn.weight", "stages.2.4.block.token_channel_mixer.m.1.bn.bias", "stages.2.4.block.token_channel_mixer.m.3.bn.weight", "stages.2.4.block.token_channel_mixer.m.3.bn.bias", "stages.2.5.block.token_channel_mixer.m.0.bn.weight", "stages.2.5.block.token_channel_mixer.m.0.bn.bias", "stages.2.5.block.token_channel_mixer.m.1.bn.weight", "stages.2.5.block.token_channel_mixer.m.1.bn.bias", "stages.2.5.block.token_channel_mixer.m.3.bn.weight", "stages.2.5.block.token_channel_mixer.m.3.bn.bias", "stages.2.6.block.token_channel_mixer.m.0.bn.weight", "stages.2.6.block.token_channel_mixer.m.0.bn.bias", "stages.2.6.block.token_channel_mixer.m.1.bn.weight", "stages.2.6.block.token_channel_mixer.m.1.bn.bias", "stages.2.6.block.token_channel_mixer.m.3.bn.weight", "stages.2.6.block.token_channel_mixer.m.3.bn.bias", "stages.2.7.block.token_channel_mixer.m.0.bn.weight", "stages.2.7.block.token_channel_mixer.m.0.bn.bias", "stages.2.7.block.token_channel_mixer.m.1.bn.weight", "stages.2.7.block.token_channel_mixer.m.1.bn.bias", "stages.2.7.block.token_channel_mixer.m.3.bn.weight", "stages.2.7.block.token_channel_mixer.m.3.bn.bias", "stages.2.8.block.token_channel_mixer.m.0.bn.weight", "stages.2.8.block.token_channel_mixer.m.0.bn.bias", "stages.2.8.block.token_channel_mixer.m.1.bn.weight", "stages.2.8.block.token_channel_mixer.m.1.bn.bias", "stages.2.8.block.token_channel_mixer.m.3.bn.weight", "stages.2.8.block.token_channel_mixer.m.3.bn.bias", "stages.2.9.block.token_channel_mixer.m.0.bn.weight", "stages.2.9.block.token_channel_mixer.m.0.bn.bias", "stages.2.9.block.token_channel_mixer.m.1.bn.weight", "stages.2.9.block.token_channel_mixer.m.1.bn.bias", "stages.2.9.block.token_channel_mixer.m.3.bn.weight", "stages.2.9.block.token_channel_mixer.m.3.bn.bias", "stages.2.10.block.token_channel_mixer.m.0.bn.weight", "stages.2.10.block.token_channel_mixer.m.0.bn.bias", "stages.2.10.block.token_channel_mixer.m.1.bn.weight", "stages.2.10.block.token_channel_mixer.m.1.bn.bias", "stages.2.10.block.token_channel_mixer.m.3.bn.weight", "stages.2.10.block.token_channel_mixer.m.3.bn.bias", "stages.2.11.block.token_channel_mixer.m.0.bn.weight", "stages.2.11.block.token_channel_mixer.m.0.bn.bias", "stages.2.11.block.token_channel_mixer.m.1.bn.weight", "stages.2.11.block.token_channel_mixer.m.1.bn.bias", "stages.2.11.block.token_channel_mixer.m.3.bn.weight", "stages.2.11.block.token_channel_mixer.m.3.bn.bias", "stages.2.12.block.cpe.m.bn.weight", "stages.2.12.block.cpe.m.bn.bias", "stages.2.13.block.token_channel_mixer.m.q.bn.weight", "stages.2.13.block.token_channel_mixer.m.q.bn.bias", "stages.2.13.block.token_channel_mixer.m.k.bn.weight", "stages.2.13.block.token_channel_mixer.m.k.bn.bias", "stages.2.13.block.token_channel_mixer.m.proj.bn.weight", "stages.2.13.block.token_channel_mixer.m.proj.bn.bias", "stages.2.13.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.13.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.14.block.channel_mixer.m.0.bn.weight", "stages.2.14.block.channel_mixer.m.0.bn.bias", "stages.2.14.block.channel_mixer.m.2.bn.weight", "stages.2.14.block.channel_mixer.m.2.bn.bias", "stages.2.15.block.cpe.m.bn.weight", "stages.2.15.block.cpe.m.bn.bias", "stages.2.16.block.token_channel_mixer.m.q.bn.weight", "stages.2.16.block.token_channel_mixer.m.q.bn.bias", "stages.2.16.block.token_channel_mixer.m.k.bn.weight", "stages.2.16.block.token_channel_mixer.m.k.bn.bias", "stages.2.16.block.token_channel_mixer.m.proj.bn.weight", "stages.2.16.block.token_channel_mixer.m.proj.bn.bias", "stages.2.16.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.16.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.17.block.channel_mixer.m.0.bn.weight", "stages.2.17.block.channel_mixer.m.0.bn.bias", "stages.2.17.block.channel_mixer.m.2.bn.weight", "stages.2.17.block.channel_mixer.m.2.bn.bias", "stages.2.18.block.cpe.m.bn.weight", "stages.2.18.block.cpe.m.bn.bias", "stages.2.19.block.token_channel_mixer.m.q.bn.weight", "stages.2.19.block.token_channel_mixer.m.q.bn.bias", "stages.2.19.block.token_channel_mixer.m.k.bn.weight", "stages.2.19.block.token_channel_mixer.m.k.bn.bias", "stages.2.19.block.token_channel_mixer.m.proj.bn.weight", "stages.2.19.block.token_channel_mixer.m.proj.bn.bias", "stages.2.19.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.19.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.20.block.channel_mixer.m.0.bn.weight", "stages.2.20.block.channel_mixer.m.0.bn.bias", "stages.2.20.block.channel_mixer.m.2.bn.weight", "stages.2.20.block.channel_mixer.m.2.bn.bias", "stages.2.21.block.cpe.m.bn.weight", "stages.2.21.block.cpe.m.bn.bias", "stages.2.22.block.token_channel_mixer.m.q.bn.weight", "stages.2.22.block.token_channel_mixer.m.q.bn.bias", "stages.2.22.block.token_channel_mixer.m.k.bn.weight", "stages.2.22.block.token_channel_mixer.m.k.bn.bias", "stages.2.22.block.token_channel_mixer.m.proj.bn.weight", "stages.2.22.block.token_channel_mixer.m.proj.bn.bias", "stages.2.22.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.22.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.23.block.channel_mixer.m.0.bn.weight", "stages.2.23.block.channel_mixer.m.0.bn.bias", "stages.2.23.block.channel_mixer.m.2.bn.weight", "stages.2.23.block.channel_mixer.m.2.bn.bias", "stages.2.24.block.cpe.m.bn.weight", "stages.2.24.block.cpe.m.bn.bias", "stages.2.25.block.token_channel_mixer.m.q.bn.weight", "stages.2.25.block.token_channel_mixer.m.q.bn.bias", "stages.2.25.block.token_channel_mixer.m.k.bn.weight", "stages.2.25.block.token_channel_mixer.m.k.bn.bias", "stages.2.25.block.token_channel_mixer.m.proj.bn.weight", "stages.2.25.block.token_channel_mixer.m.proj.bn.bias", "stages.2.25.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.25.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.26.block.channel_mixer.m.0.bn.weight", "stages.2.26.block.channel_mixer.m.0.bn.bias", "stages.2.26.block.channel_mixer.m.2.bn.weight", "stages.2.26.block.channel_mixer.m.2.bn.bias", "stages.2.27.block.cpe.m.bn.weight", "stages.2.27.block.cpe.m.bn.bias", "stages.2.28.block.token_channel_mixer.m.q.bn.weight", "stages.2.28.block.token_channel_mixer.m.q.bn.bias", "stages.2.28.block.token_channel_mixer.m.k.bn.weight", "stages.2.28.block.token_channel_mixer.m.k.bn.bias", "stages.2.28.block.token_channel_mixer.m.proj.bn.weight", "stages.2.28.block.token_channel_mixer.m.proj.bn.bias", "stages.2.28.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.28.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.29.block.channel_mixer.m.0.bn.weight", "stages.2.29.block.channel_mixer.m.0.bn.bias", "stages.2.29.block.channel_mixer.m.2.bn.weight", "stages.2.29.block.channel_mixer.m.2.bn.bias", "stages.2.30.block.cpe.m.bn.weight", "stages.2.30.block.cpe.m.bn.bias", "stages.2.31.block.token_channel_mixer.m.q.bn.weight", "stages.2.31.block.token_channel_mixer.m.q.bn.bias", "stages.2.31.block.token_channel_mixer.m.k.bn.weight", "stages.2.31.block.token_channel_mixer.m.k.bn.bias", "stages.2.31.block.token_channel_mixer.m.proj.bn.weight", "stages.2.31.block.token_channel_mixer.m.proj.bn.bias", "stages.2.31.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.31.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.32.block.channel_mixer.m.0.bn.weight", "stages.2.32.block.channel_mixer.m.0.bn.bias", "stages.2.32.block.channel_mixer.m.2.bn.weight", "stages.2.32.block.channel_mixer.m.2.bn.bias", "stages.2.33.block.cpe.m.bn.weight", "stages.2.33.block.cpe.m.bn.bias", "stages.2.34.block.token_channel_mixer.m.q.bn.weight", "stages.2.34.block.token_channel_mixer.m.q.bn.bias", "stages.2.34.block.token_channel_mixer.m.k.bn.weight", "stages.2.34.block.token_channel_mixer.m.k.bn.bias", "stages.2.34.block.token_channel_mixer.m.proj.bn.weight", "stages.2.34.block.token_channel_mixer.m.proj.bn.bias", "stages.2.34.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.34.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.35.block.channel_mixer.m.0.bn.weight", "stages.2.35.block.channel_mixer.m.0.bn.bias", "stages.2.35.block.channel_mixer.m.2.bn.weight", "stages.2.35.block.channel_mixer.m.2.bn.bias", "stages.2.36.block.cpe.m.bn.weight", "stages.2.36.block.cpe.m.bn.bias", "stages.2.37.block.token_channel_mixer.m.q.bn.weight", "stages.2.37.block.token_channel_mixer.m.q.bn.bias", "stages.2.37.block.token_channel_mixer.m.k.bn.weight", "stages.2.37.block.token_channel_mixer.m.k.bn.bias", "stages.2.37.block.token_channel_mixer.m.proj.bn.weight", "stages.2.37.block.token_channel_mixer.m.proj.bn.bias", "stages.2.37.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.37.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.38.block.channel_mixer.m.0.bn.weight", "stages.2.38.block.channel_mixer.m.0.bn.bias", "stages.2.38.block.channel_mixer.m.2.bn.weight", "stages.2.38.block.channel_mixer.m.2.bn.bias", "stages.2.39.block.cpe.m.bn.weight", "stages.2.39.block.cpe.m.bn.bias", "stages.2.40.block.token_channel_mixer.m.q.bn.weight", "stages.2.40.block.token_channel_mixer.m.q.bn.bias", "stages.2.40.block.token_channel_mixer.m.k.bn.weight", "stages.2.40.block.token_channel_mixer.m.k.bn.bias", "stages.2.40.block.token_channel_mixer.m.proj.bn.weight", "stages.2.40.block.token_channel_mixer.m.proj.bn.bias", "stages.2.40.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.40.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.41.block.channel_mixer.m.0.bn.weight", "stages.2.41.block.channel_mixer.m.0.bn.bias", "stages.2.41.block.channel_mixer.m.2.bn.weight", "stages.2.41.block.channel_mixer.m.2.bn.bias", "stages.2.42.block.cpe.m.bn.weight", "stages.2.42.block.cpe.m.bn.bias", "stages.2.43.block.token_channel_mixer.m.q.bn.weight", "stages.2.43.block.token_channel_mixer.m.q.bn.bias", "stages.2.43.block.token_channel_mixer.m.k.bn.weight", "stages.2.43.block.token_channel_mixer.m.k.bn.bias", "stages.2.43.block.token_channel_mixer.m.proj.bn.weight", "stages.2.43.block.token_channel_mixer.m.proj.bn.bias", "stages.2.43.block.token_channel_mixer.m.v_gate.bn.weight", "stages.2.43.block.token_channel_mixer.m.v_gate.bn.bias", "stages.2.44.block.channel_mixer.m.0.bn.weight", "stages.2.44.block.channel_mixer.m.0.bn.bias", "stages.2.44.block.channel_mixer.m.2.bn.weight", "stages.2.44.block.channel_mixer.m.2.bn.bias", "stages.2.45.block.token_channel_mixer.m.0.bn.weight", "stages.2.45.block.token_channel_mixer.m.0.bn.bias", "stages.2.45.block.token_channel_mixer.m.1.bn.weight", "stages.2.45.block.token_channel_mixer.m.1.bn.bias", "stages.2.45.block.token_channel_mixer.m.3.bn.weight", "stages.2.45.block.token_channel_mixer.m.3.bn.bias", "stages.3.0.block.cpe.m.bn.weight", "stages.3.0.block.cpe.m.bn.bias", "stages.3.1.block.token_channel_mixer.m.q.bn.weight", "stages.3.1.block.token_channel_mixer.m.q.bn.bias", "stages.3.1.block.token_channel_mixer.m.k.bn.weight", "stages.3.1.block.token_channel_mixer.m.k.bn.bias", "stages.3.1.block.token_channel_mixer.m.proj.bn.weight", "stages.3.1.block.token_channel_mixer.m.proj.bn.bias", "stages.3.1.block.token_channel_mixer.m.v_gate.bn.weight", "stages.3.1.block.token_channel_mixer.m.v_gate.bn.bias", "stages.3.2.block.channel_mixer.m.0.bn.weight", "stages.3.2.block.channel_mixer.m.0.bn.bias", "stages.3.2.block.channel_mixer.m.2.bn.weight", "stages.3.2.block.channel_mixer.m.2.bn.bias", "stages.3.3.block.cpe.m.bn.weight", "stages.3.3.block.cpe.m.bn.bias", "stages.3.4.block.token_channel_mixer.m.q.bn.weight", "stages.3.4.block.token_channel_mixer.m.q.bn.bias", "stages.3.4.block.token_channel_mixer.m.k.bn.weight", "stages.3.4.block.token_channel_mixer.m.k.bn.bias", "stages.3.4.block.token_channel_mixer.m.proj.bn.weight", "stages.3.4.block.token_channel_mixer.m.proj.bn.bias", "stages.3.4.block.token_channel_mixer.m.v_gate.bn.weight", "stages.3.4.block.token_channel_mixer.m.v_gate.bn.bias", "stages.3.5.block.channel_mixer.m.0.bn.weight", "stages.3.5.block.channel_mixer.m.0.bn.bias", "stages.3.5.block.channel_mixer.m.2.bn.weight", "stages.3.5.block.channel_mixer.m.2.bn.bias", "stages.3.6.block.cpe.m.bn.weight", "stages.3.6.block.cpe.m.bn.bias", "stages.3.7.block.token_channel_mixer.m.q.bn.weight", "stages.3.7.block.token_channel_mixer.m.q.bn.bias", "stages.3.7.block.token_channel_mixer.m.k.bn.weight", "stages.3.7.block.token_channel_mixer.m.k.bn.bias", "stages.3.7.block.token_channel_mixer.m.proj.bn.weight", "stages.3.7.block.token_channel_mixer.m.proj.bn.bias", "stages.3.7.block.token_channel_mixer.m.v_gate.bn.weight", "stages.3.7.block.token_channel_mixer.m.v_gate.bn.bias", "stages.3.8.block.channel_mixer.m.0.bn.weight", "stages.3.8.block.channel_mixer.m.0.bn.bias", "stages.3.8.block.channel_mixer.m.2.bn.weight", "stages.3.8.block.channel_mixer.m.2.bn.bias", "classifier.classifier.bn.weight", "classifier.classifier.bn.bias", "classifier.classifier.l.bias", "classifier.classifier_dist.bn.weight", "classifier.classifier_dist.bn.bias", "classifier.classifier_dist.l.bias" ], "lr_scale": 1.0 } } Use Cosine LR scheduler Set warmup steps = 6240 Set warmup steps = 0 Max WD = 0.0500000, Min WD = 0.0500000 Creating teacher model: regnety_160 criterion = SoftTargetCrossEntropy() Auto resume checkpoint: Start training for 450 epochs Epoch: [0] [ 0/312] eta: 5:39:08 lr: 0.000000 min_lr: 0.000000 loss: 6.9641 (6.9641) weight_decay: 0.0500 (0.0500) time: 65.2204 data: 8.7530 max mem: 64948 Epoch: [0] [ 10/312] eta: 0:37:18 lr: 0.000006 min_lr: 0.000006 loss: 7.0114 (7.0138) weight_decay: 0.0500 (0.0500) time: 7.4111 data: 0.7961 max mem: 64948 Epoch: [0] [ 20/312] eta: 0:20:29 lr: 0.000013 min_lr: 0.000013 loss: 7.0053 (7.0042) weight_decay: 0.0500 (0.0500) time: 1.1610 data: 0.0004 max mem: 64948 Epoch: [0] [ 30/312] eta: 0:14:27 lr: 0.000019 min_lr: 0.000019 loss: 6.9849 (6.9917) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [0] [ 40/312] eta: 0:11:20 lr: 0.000026 min_lr: 0.000026 loss: 6.9504 (6.9806) weight_decay: 0.0500 (0.0500) time: 0.7101 data: 0.0005 max mem: 64948 Epoch: [0] [ 50/312] eta: 0:09:23 lr: 0.000032 min_lr: 0.000032 loss: 6.9362 (6.9704) weight_decay: 0.0500 (0.0500) time: 0.7115 data: 0.0005 max mem: 64948 Epoch: [0] [ 60/312] eta: 0:08:02 lr: 0.000038 min_lr: 0.000038 loss: 6.9144 (6.9607) weight_decay: 0.0500 (0.0500) time: 0.7073 data: 0.0005 max mem: 64948 Epoch: [0] [ 70/312] eta: 0:07:01 lr: 0.000045 min_lr: 0.000045 loss: 6.8924 (6.9479) weight_decay: 0.0500 (0.0500) time: 0.7046 data: 0.0005 max mem: 64948 Epoch: [0] [ 80/312] eta: 0:06:14 lr: 0.000051 min_lr: 0.000051 loss: 6.8624 (6.9364) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [0] [ 90/312] eta: 0:05:35 lr: 0.000058 min_lr: 0.000058 loss: 6.8476 (6.9271) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0005 max mem: 64948 Epoch: [0] [100/312] eta: 0:05:03 lr: 0.000064 min_lr: 0.000064 loss: 6.8434 (6.9185) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0005 max mem: 64948 Epoch: [0] [110/312] eta: 0:04:35 lr: 0.000071 min_lr: 0.000071 loss: 6.8359 (6.9097) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [0] [120/312] eta: 0:04:11 lr: 0.000077 min_lr: 0.000077 loss: 6.8039 (6.9010) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [0] [130/312] eta: 0:03:49 lr: 0.000083 min_lr: 0.000083 loss: 6.7753 (6.8913) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [0] [140/312] eta: 0:03:30 lr: 0.000090 min_lr: 0.000090 loss: 6.7658 (6.8823) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [0] [150/312] eta: 0:03:12 lr: 0.000096 min_lr: 0.000096 loss: 6.7280 (6.8716) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [0] [160/312] eta: 0:02:55 lr: 0.000103 min_lr: 0.000103 loss: 6.7198 (6.8630) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [0] [170/312] eta: 0:02:40 lr: 0.000109 min_lr: 0.000109 loss: 6.7318 (6.8544) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [0] [180/312] eta: 0:02:25 lr: 0.000115 min_lr: 0.000115 loss: 6.6879 (6.8442) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0004 max mem: 64948 Epoch: [0] [190/312] eta: 0:02:12 lr: 0.000122 min_lr: 0.000122 loss: 6.6401 (6.8346) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [0] [200/312] eta: 0:01:59 lr: 0.000128 min_lr: 0.000128 loss: 6.6149 (6.8240) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [0] [210/312] eta: 0:01:46 lr: 0.000135 min_lr: 0.000135 loss: 6.5961 (6.8128) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [0] [220/312] eta: 0:01:34 lr: 0.000141 min_lr: 0.000141 loss: 6.5923 (6.8028) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [0] [230/312] eta: 0:01:23 lr: 0.000147 min_lr: 0.000147 loss: 6.5923 (6.7930) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [0] [240/312] eta: 0:01:12 lr: 0.000154 min_lr: 0.000154 loss: 6.5460 (6.7832) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [0] [250/312] eta: 0:01:01 lr: 0.000160 min_lr: 0.000160 loss: 6.5362 (6.7729) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [0] [260/312] eta: 0:00:50 lr: 0.000167 min_lr: 0.000167 loss: 6.4988 (6.7637) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [0] [270/312] eta: 0:00:40 lr: 0.000173 min_lr: 0.000173 loss: 6.4818 (6.7534) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [0] [280/312] eta: 0:00:30 lr: 0.000180 min_lr: 0.000180 loss: 6.5205 (6.7469) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [0] [290/312] eta: 0:00:20 lr: 0.000186 min_lr: 0.000186 loss: 6.5259 (6.7376) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [0] [300/312] eta: 0:00:11 lr: 0.000192 min_lr: 0.000192 loss: 6.4584 (6.7283) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [0] [310/312] eta: 0:00:01 lr: 0.000199 min_lr: 0.000199 loss: 6.4581 (6.7206) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [0] [311/312] eta: 0:00:00 lr: 0.000199 min_lr: 0.000199 loss: 6.4760 (6.7202) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [0] Total time: 0:04:52 (0.9359 s / it) Averaged stats: lr: 0.000199 min_lr: 0.000199 loss: 6.4760 (6.7249) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:02:05 loss: 5.7016 (5.7016) acc1: 2.0833 (2.0833) acc5: 9.8958 (9.8958) time: 13.9680 data: 4.4908 max mem: 64948 Test: [8/9] eta: 0:00:02 loss: 6.0172 (5.8642) acc1: 2.0833 (2.6880) acc5: 9.1146 (9.5360) time: 2.0976 data: 0.4991 max mem: 64948 Test: Total time: 0:00:18 (2.1103 s / it) * Acc@1 2.740 Acc@5 9.388 loss 5.883 Accuracy of the model on the 50000 test images: 2.7% Max accuracy: 2.74% Test: [0/9] eta: 0:00:40 loss: 6.8976 (6.8976) acc1: 0.7812 (0.7812) acc5: 0.7812 (0.7812) time: 4.4780 data: 4.2364 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9109 (6.9098) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6486 data: 0.4709 max mem: 64948 Test: Total time: 0:00:05 (0.6560 s / it) * Acc@1 0.098 Acc@5 0.492 loss 6.910 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.10% Epoch: [1] [ 0/312] eta: 1:31:33 lr: 0.000200 min_lr: 0.000200 loss: 6.5261 (6.5261) weight_decay: 0.0500 (0.0500) time: 17.6070 data: 7.8573 max mem: 64948 Epoch: [1] [ 10/312] eta: 0:14:55 lr: 0.000206 min_lr: 0.000206 loss: 6.5210 (6.5069) weight_decay: 0.0500 (0.0500) time: 2.9667 data: 0.7147 max mem: 64948 Epoch: [1] [ 20/312] eta: 0:11:02 lr: 0.000213 min_lr: 0.000213 loss: 6.5205 (6.4844) weight_decay: 0.0500 (0.0500) time: 1.5029 data: 0.0004 max mem: 64948 Epoch: [1] [ 30/312] eta: 0:09:30 lr: 0.000219 min_lr: 0.000219 loss: 6.4529 (6.4751) weight_decay: 0.0500 (0.0500) time: 1.5046 data: 0.0004 max mem: 64948 Epoch: [1] [ 40/312] eta: 0:08:31 lr: 0.000226 min_lr: 0.000226 loss: 6.4197 (6.4546) weight_decay: 0.0500 (0.0500) time: 1.4710 data: 0.0004 max mem: 64948 Epoch: [1] [ 50/312] eta: 0:07:11 lr: 0.000232 min_lr: 0.000232 loss: 6.3689 (6.4442) weight_decay: 0.0500 (0.0500) time: 1.0644 data: 0.0004 max mem: 64948 Epoch: [1] [ 60/312] eta: 0:06:15 lr: 0.000238 min_lr: 0.000238 loss: 6.4235 (6.4324) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [1] [ 70/312] eta: 0:05:33 lr: 0.000245 min_lr: 0.000245 loss: 6.3591 (6.4152) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [1] [ 80/312] eta: 0:05:00 lr: 0.000251 min_lr: 0.000251 loss: 6.2564 (6.4016) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [1] [ 90/312] eta: 0:04:32 lr: 0.000258 min_lr: 0.000258 loss: 6.3928 (6.4028) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [1] [100/312] eta: 0:04:09 lr: 0.000264 min_lr: 0.000264 loss: 6.3733 (6.3940) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [1] [110/312] eta: 0:03:48 lr: 0.000271 min_lr: 0.000271 loss: 6.2579 (6.3807) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [1] [120/312] eta: 0:03:30 lr: 0.000277 min_lr: 0.000277 loss: 6.2802 (6.3793) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0004 max mem: 64948 Epoch: [1] [130/312] eta: 0:03:13 lr: 0.000283 min_lr: 0.000283 loss: 6.3372 (6.3709) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [1] [140/312] eta: 0:02:58 lr: 0.000290 min_lr: 0.000290 loss: 6.1925 (6.3571) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [1] [150/312] eta: 0:02:44 lr: 0.000296 min_lr: 0.000296 loss: 6.1307 (6.3426) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [1] [160/312] eta: 0:02:31 lr: 0.000303 min_lr: 0.000303 loss: 6.1317 (6.3301) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [1] [170/312] eta: 0:02:18 lr: 0.000309 min_lr: 0.000309 loss: 6.1808 (6.3237) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0003 max mem: 64948 Epoch: [1] [180/312] eta: 0:02:07 lr: 0.000315 min_lr: 0.000315 loss: 6.2102 (6.3173) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [1] [190/312] eta: 0:01:55 lr: 0.000322 min_lr: 0.000322 loss: 6.1586 (6.3072) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [1] [200/312] eta: 0:01:44 lr: 0.000328 min_lr: 0.000328 loss: 6.2210 (6.3063) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [1] [210/312] eta: 0:01:34 lr: 0.000335 min_lr: 0.000335 loss: 6.2604 (6.2965) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [1] [220/312] eta: 0:01:24 lr: 0.000341 min_lr: 0.000341 loss: 6.1649 (6.2888) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [1] [230/312] eta: 0:01:14 lr: 0.000347 min_lr: 0.000347 loss: 6.1879 (6.2808) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [1] [240/312] eta: 0:01:04 lr: 0.000354 min_lr: 0.000354 loss: 6.1409 (6.2733) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [1] [250/312] eta: 0:00:55 lr: 0.000360 min_lr: 0.000360 loss: 6.1554 (6.2697) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [1] [260/312] eta: 0:00:45 lr: 0.000367 min_lr: 0.000367 loss: 6.1567 (6.2631) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [1] [270/312] eta: 0:00:36 lr: 0.000373 min_lr: 0.000373 loss: 6.1799 (6.2585) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [1] [280/312] eta: 0:00:27 lr: 0.000380 min_lr: 0.000380 loss: 6.1006 (6.2509) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [1] [290/312] eta: 0:00:18 lr: 0.000386 min_lr: 0.000386 loss: 6.0232 (6.2418) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [1] [300/312] eta: 0:00:10 lr: 0.000392 min_lr: 0.000392 loss: 6.0025 (6.2354) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [1] [310/312] eta: 0:00:01 lr: 0.000399 min_lr: 0.000399 loss: 6.0582 (6.2310) weight_decay: 0.0500 (0.0500) time: 0.6899 data: 0.0001 max mem: 64948 Epoch: [1] [311/312] eta: 0:00:00 lr: 0.000399 min_lr: 0.000399 loss: 6.0582 (6.2306) weight_decay: 0.0500 (0.0500) time: 0.6898 data: 0.0001 max mem: 64948 Epoch: [1] Total time: 0:04:25 (0.8523 s / it) Averaged stats: lr: 0.000399 min_lr: 0.000399 loss: 6.0582 (6.2212) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 4.8255 (4.8255) acc1: 9.3750 (9.3750) acc5: 25.5208 (25.5208) time: 4.8435 data: 4.6243 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 4.9804 (5.0610) acc1: 9.3750 (9.2800) acc5: 22.3958 (22.2080) time: 0.6901 data: 0.5139 max mem: 64948 Test: Total time: 0:00:06 (0.7151 s / it) * Acc@1 8.634 Acc@5 22.130 loss 5.057 Accuracy of the model on the 50000 test images: 8.6% Max accuracy: 8.63% Test: [0/9] eta: 0:00:38 loss: 6.8983 (6.8983) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.2607 data: 4.0520 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9083 (6.9088) acc1: 0.0000 (0.0960) acc5: 0.7812 (0.5120) time: 0.6247 data: 0.4503 max mem: 64948 Test: Total time: 0:00:05 (0.6332 s / it) * Acc@1 0.096 Acc@5 0.496 loss 6.909 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [2] [ 0/312] eta: 0:56:08 lr: 0.000400 min_lr: 0.000400 loss: 6.1766 (6.1766) weight_decay: 0.0500 (0.0500) time: 10.7980 data: 8.5169 max mem: 64948 Epoch: [2] [ 10/312] eta: 0:08:14 lr: 0.000406 min_lr: 0.000406 loss: 5.9574 (6.0291) weight_decay: 0.0500 (0.0500) time: 1.6389 data: 0.7746 max mem: 64948 Epoch: [2] [ 20/312] eta: 0:05:47 lr: 0.000413 min_lr: 0.000413 loss: 5.8908 (5.9292) weight_decay: 0.0500 (0.0500) time: 0.7091 data: 0.0004 max mem: 64948 Epoch: [2] [ 30/312] eta: 0:04:50 lr: 0.000419 min_lr: 0.000419 loss: 5.8680 (5.9370) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [2] [ 40/312] eta: 0:04:17 lr: 0.000426 min_lr: 0.000426 loss: 5.8068 (5.9025) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [2] [ 50/312] eta: 0:03:55 lr: 0.000432 min_lr: 0.000432 loss: 5.9242 (5.9325) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [2] [ 60/312] eta: 0:03:37 lr: 0.000439 min_lr: 0.000439 loss: 5.9280 (5.9185) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [2] [ 70/312] eta: 0:03:23 lr: 0.000445 min_lr: 0.000445 loss: 5.7682 (5.9039) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [2] [ 80/312] eta: 0:03:10 lr: 0.000451 min_lr: 0.000451 loss: 5.6714 (5.8880) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [2] [ 90/312] eta: 0:02:59 lr: 0.000458 min_lr: 0.000458 loss: 5.6492 (5.8750) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [2] [100/312] eta: 0:02:49 lr: 0.000464 min_lr: 0.000464 loss: 5.6582 (5.8630) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [2] [110/312] eta: 0:02:39 lr: 0.000471 min_lr: 0.000471 loss: 5.7958 (5.8668) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [2] [120/312] eta: 0:02:30 lr: 0.000477 min_lr: 0.000477 loss: 5.9868 (5.8691) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [2] [130/312] eta: 0:02:21 lr: 0.000483 min_lr: 0.000483 loss: 5.9884 (5.8599) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [2] [140/312] eta: 0:02:12 lr: 0.000490 min_lr: 0.000490 loss: 5.8823 (5.8640) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [2] [150/312] eta: 0:02:03 lr: 0.000496 min_lr: 0.000496 loss: 5.8823 (5.8606) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [2] [160/312] eta: 0:01:55 lr: 0.000503 min_lr: 0.000503 loss: 5.8420 (5.8507) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [2] [170/312] eta: 0:01:47 lr: 0.000509 min_lr: 0.000509 loss: 5.7791 (5.8500) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [2] [180/312] eta: 0:01:39 lr: 0.000515 min_lr: 0.000515 loss: 5.7421 (5.8423) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [2] [190/312] eta: 0:01:31 lr: 0.000522 min_lr: 0.000522 loss: 5.5447 (5.8286) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [2] [200/312] eta: 0:01:23 lr: 0.000528 min_lr: 0.000528 loss: 5.5447 (5.8238) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [2] [210/312] eta: 0:01:15 lr: 0.000535 min_lr: 0.000535 loss: 5.5521 (5.8090) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0004 max mem: 64948 Epoch: [2] [220/312] eta: 0:01:08 lr: 0.000541 min_lr: 0.000541 loss: 5.5112 (5.7988) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0004 max mem: 64948 Epoch: [2] [230/312] eta: 0:01:00 lr: 0.000548 min_lr: 0.000548 loss: 5.5891 (5.7970) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [2] [240/312] eta: 0:00:53 lr: 0.000554 min_lr: 0.000554 loss: 5.6775 (5.7892) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [2] [250/312] eta: 0:00:45 lr: 0.000560 min_lr: 0.000560 loss: 5.6083 (5.7842) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [2] [260/312] eta: 0:00:38 lr: 0.000567 min_lr: 0.000567 loss: 5.5897 (5.7731) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [2] [270/312] eta: 0:00:30 lr: 0.000573 min_lr: 0.000573 loss: 5.6048 (5.7658) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0004 max mem: 64948 Epoch: [2] [280/312] eta: 0:00:23 lr: 0.000580 min_lr: 0.000580 loss: 5.6265 (5.7597) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0009 max mem: 64948 Epoch: [2] [290/312] eta: 0:00:16 lr: 0.000586 min_lr: 0.000586 loss: 5.6265 (5.7526) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [2] [300/312] eta: 0:00:08 lr: 0.000592 min_lr: 0.000592 loss: 5.7194 (5.7515) weight_decay: 0.0500 (0.0500) time: 0.6902 data: 0.0001 max mem: 64948 Epoch: [2] [310/312] eta: 0:00:01 lr: 0.000599 min_lr: 0.000599 loss: 5.7194 (5.7455) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [2] [311/312] eta: 0:00:00 lr: 0.000599 min_lr: 0.000599 loss: 5.6915 (5.7429) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [2] Total time: 0:03:47 (0.7304 s / it) Averaged stats: lr: 0.000599 min_lr: 0.000599 loss: 5.6915 (5.7572) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 3.9916 (3.9916) acc1: 21.6146 (21.6146) acc5: 44.2708 (44.2708) time: 4.4977 data: 4.2836 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 4.1620 (4.1048) acc1: 17.1875 (17.7600) acc5: 40.1042 (38.5600) time: 0.6511 data: 0.4761 max mem: 64948 Test: Total time: 0:00:05 (0.6648 s / it) * Acc@1 18.248 Acc@5 39.142 loss 4.090 Accuracy of the model on the 50000 test images: 18.2% Max accuracy: 18.25% Test: [0/9] eta: 0:00:38 loss: 6.8968 (6.8968) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.2820 data: 4.0683 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9086 (6.9089) acc1: 0.0000 (0.0960) acc5: 0.7812 (0.5120) time: 0.6302 data: 0.4553 max mem: 64948 Test: Total time: 0:00:05 (0.6380 s / it) * Acc@1 0.096 Acc@5 0.490 loss 6.909 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [3] [ 0/312] eta: 0:58:08 lr: 0.000600 min_lr: 0.000600 loss: 5.7364 (5.7364) weight_decay: 0.0500 (0.0500) time: 11.1825 data: 8.2174 max mem: 64948 Epoch: [3] [ 10/312] eta: 0:08:28 lr: 0.000607 min_lr: 0.000607 loss: 5.5148 (5.4290) weight_decay: 0.0500 (0.0500) time: 1.6838 data: 0.7474 max mem: 64948 Epoch: [3] [ 20/312] eta: 0:05:54 lr: 0.000613 min_lr: 0.000613 loss: 5.3720 (5.4041) weight_decay: 0.0500 (0.0500) time: 0.7153 data: 0.0004 max mem: 64948 Epoch: [3] [ 30/312] eta: 0:04:55 lr: 0.000619 min_lr: 0.000619 loss: 5.4869 (5.4653) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [3] [ 40/312] eta: 0:04:22 lr: 0.000626 min_lr: 0.000626 loss: 5.6429 (5.5000) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [3] [ 50/312] eta: 0:03:58 lr: 0.000632 min_lr: 0.000632 loss: 5.7443 (5.4998) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [3] [ 60/312] eta: 0:03:40 lr: 0.000639 min_lr: 0.000639 loss: 5.6748 (5.4922) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [3] [ 70/312] eta: 0:03:25 lr: 0.000645 min_lr: 0.000645 loss: 5.4975 (5.4834) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [3] [ 80/312] eta: 0:03:12 lr: 0.000651 min_lr: 0.000651 loss: 5.5522 (5.4889) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [3] [ 90/312] eta: 0:03:01 lr: 0.000658 min_lr: 0.000658 loss: 5.5521 (5.4722) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [3] [100/312] eta: 0:02:50 lr: 0.000664 min_lr: 0.000664 loss: 5.4079 (5.4635) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [3] [110/312] eta: 0:02:40 lr: 0.000671 min_lr: 0.000671 loss: 5.5089 (5.4590) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [3] [120/312] eta: 0:02:30 lr: 0.000677 min_lr: 0.000677 loss: 5.5323 (5.4666) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [3] [130/312] eta: 0:02:21 lr: 0.000683 min_lr: 0.000683 loss: 5.4440 (5.4543) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [3] [140/312] eta: 0:02:12 lr: 0.000690 min_lr: 0.000690 loss: 5.3227 (5.4397) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [3] [150/312] eta: 0:02:04 lr: 0.000696 min_lr: 0.000696 loss: 5.3227 (5.4312) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [3] [160/312] eta: 0:01:55 lr: 0.000703 min_lr: 0.000703 loss: 5.4904 (5.4246) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [3] [170/312] eta: 0:01:47 lr: 0.000709 min_lr: 0.000709 loss: 5.4012 (5.4191) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [3] [180/312] eta: 0:01:39 lr: 0.000715 min_lr: 0.000715 loss: 5.4402 (5.4197) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [3] [190/312] eta: 0:01:31 lr: 0.000722 min_lr: 0.000722 loss: 5.4402 (5.4155) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [3] [200/312] eta: 0:01:23 lr: 0.000728 min_lr: 0.000728 loss: 5.2091 (5.4079) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0004 max mem: 64948 Epoch: [3] [210/312] eta: 0:01:16 lr: 0.000735 min_lr: 0.000735 loss: 5.3651 (5.4039) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [3] [220/312] eta: 0:01:08 lr: 0.000741 min_lr: 0.000741 loss: 5.4066 (5.3998) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [3] [230/312] eta: 0:01:00 lr: 0.000748 min_lr: 0.000748 loss: 5.2227 (5.3873) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [3] [240/312] eta: 0:00:53 lr: 0.000754 min_lr: 0.000754 loss: 5.4322 (5.3943) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [3] [250/312] eta: 0:00:45 lr: 0.000760 min_lr: 0.000760 loss: 5.5815 (5.3996) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [3] [260/312] eta: 0:00:38 lr: 0.000767 min_lr: 0.000767 loss: 5.3993 (5.3902) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [3] [270/312] eta: 0:00:30 lr: 0.000773 min_lr: 0.000773 loss: 5.2826 (5.3871) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [3] [280/312] eta: 0:00:23 lr: 0.000780 min_lr: 0.000780 loss: 5.4054 (5.3812) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0006 max mem: 64948 Epoch: [3] [290/312] eta: 0:00:16 lr: 0.000786 min_lr: 0.000786 loss: 5.1632 (5.3734) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0005 max mem: 64948 Epoch: [3] [300/312] eta: 0:00:08 lr: 0.000792 min_lr: 0.000792 loss: 5.0725 (5.3656) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [3] [310/312] eta: 0:00:01 lr: 0.000799 min_lr: 0.000799 loss: 5.2403 (5.3619) weight_decay: 0.0500 (0.0500) time: 0.6900 data: 0.0001 max mem: 64948 Epoch: [3] [311/312] eta: 0:00:00 lr: 0.000799 min_lr: 0.000799 loss: 5.2403 (5.3606) weight_decay: 0.0500 (0.0500) time: 0.6899 data: 0.0001 max mem: 64948 Epoch: [3] Total time: 0:03:48 (0.7318 s / it) Averaged stats: lr: 0.000799 min_lr: 0.000799 loss: 5.2403 (5.3397) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 3.3484 (3.3484) acc1: 32.8125 (32.8125) acc5: 55.7292 (55.7292) time: 4.5535 data: 4.3386 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 3.5256 (3.4865) acc1: 27.6042 (26.2400) acc5: 51.5625 (50.4000) time: 0.6578 data: 0.4822 max mem: 64948 Test: Total time: 0:00:06 (0.6791 s / it) * Acc@1 26.864 Acc@5 50.834 loss 3.489 Accuracy of the model on the 50000 test images: 26.9% Max accuracy: 26.86% Test: [0/9] eta: 0:00:40 loss: 6.8946 (6.8946) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.4737 data: 4.2673 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9096 (6.9094) acc1: 0.0000 (0.1280) acc5: 0.0000 (0.5120) time: 0.6483 data: 0.4743 max mem: 64948 Test: Total time: 0:00:05 (0.6563 s / it) * Acc@1 0.098 Acc@5 0.494 loss 6.910 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [4] [ 0/312] eta: 0:58:25 lr: 0.000800 min_lr: 0.000800 loss: 4.4332 (4.4332) weight_decay: 0.0500 (0.0500) time: 11.2364 data: 7.4269 max mem: 64948 Epoch: [4] [ 10/312] eta: 0:08:26 lr: 0.000807 min_lr: 0.000807 loss: 5.2925 (5.0961) weight_decay: 0.0500 (0.0500) time: 1.6785 data: 0.6756 max mem: 64948 Epoch: [4] [ 20/312] eta: 0:05:53 lr: 0.000813 min_lr: 0.000813 loss: 5.3320 (5.1874) weight_decay: 0.0500 (0.0500) time: 0.7077 data: 0.0004 max mem: 64948 Epoch: [4] [ 30/312] eta: 0:04:55 lr: 0.000819 min_lr: 0.000819 loss: 5.2723 (5.1882) weight_decay: 0.0500 (0.0500) time: 0.7021 data: 0.0003 max mem: 64948 Epoch: [4] [ 40/312] eta: 0:04:21 lr: 0.000826 min_lr: 0.000826 loss: 5.2011 (5.1938) weight_decay: 0.0500 (0.0500) time: 0.7033 data: 0.0003 max mem: 64948 Epoch: [4] [ 50/312] eta: 0:03:58 lr: 0.000832 min_lr: 0.000832 loss: 5.1986 (5.1742) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [4] [ 60/312] eta: 0:03:40 lr: 0.000839 min_lr: 0.000839 loss: 5.1040 (5.1460) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [4] [ 70/312] eta: 0:03:25 lr: 0.000845 min_lr: 0.000845 loss: 4.8561 (5.1074) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [4] [ 80/312] eta: 0:03:12 lr: 0.000851 min_lr: 0.000851 loss: 4.8737 (5.1021) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [4] [ 90/312] eta: 0:03:00 lr: 0.000858 min_lr: 0.000858 loss: 5.2494 (5.1060) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [4] [100/312] eta: 0:02:50 lr: 0.000864 min_lr: 0.000864 loss: 5.2494 (5.1052) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [4] [110/312] eta: 0:02:40 lr: 0.000871 min_lr: 0.000871 loss: 5.1946 (5.1117) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [4] [120/312] eta: 0:02:30 lr: 0.000877 min_lr: 0.000877 loss: 5.3267 (5.1275) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [4] [130/312] eta: 0:02:21 lr: 0.000883 min_lr: 0.000883 loss: 5.3267 (5.1289) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [4] [140/312] eta: 0:02:12 lr: 0.000890 min_lr: 0.000890 loss: 5.1889 (5.1266) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [4] [150/312] eta: 0:02:04 lr: 0.000896 min_lr: 0.000896 loss: 5.0814 (5.1158) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [4] [160/312] eta: 0:01:55 lr: 0.000903 min_lr: 0.000903 loss: 5.0747 (5.1173) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [4] [170/312] eta: 0:01:47 lr: 0.000909 min_lr: 0.000909 loss: 5.0747 (5.1120) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [4] [180/312] eta: 0:01:39 lr: 0.000916 min_lr: 0.000916 loss: 4.9368 (5.0997) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [4] [190/312] eta: 0:01:31 lr: 0.000922 min_lr: 0.000922 loss: 4.9278 (5.0915) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [4] [200/312] eta: 0:01:23 lr: 0.000928 min_lr: 0.000928 loss: 5.1165 (5.0895) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [4] [210/312] eta: 0:01:16 lr: 0.000935 min_lr: 0.000935 loss: 5.1450 (5.0842) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [4] [220/312] eta: 0:01:08 lr: 0.000941 min_lr: 0.000941 loss: 5.0666 (5.0826) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [4] [230/312] eta: 0:01:00 lr: 0.000948 min_lr: 0.000948 loss: 5.0322 (5.0699) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [4] [240/312] eta: 0:00:53 lr: 0.000954 min_lr: 0.000954 loss: 4.5138 (5.0523) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [4] [250/312] eta: 0:00:45 lr: 0.000960 min_lr: 0.000960 loss: 4.5138 (5.0363) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [4] [260/312] eta: 0:00:38 lr: 0.000967 min_lr: 0.000967 loss: 4.8505 (5.0382) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [4] [270/312] eta: 0:00:30 lr: 0.000973 min_lr: 0.000973 loss: 5.1177 (5.0345) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [4] [280/312] eta: 0:00:23 lr: 0.000980 min_lr: 0.000980 loss: 4.9260 (5.0270) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [4] [290/312] eta: 0:00:16 lr: 0.000986 min_lr: 0.000986 loss: 4.8723 (5.0231) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [4] [300/312] eta: 0:00:08 lr: 0.000992 min_lr: 0.000992 loss: 4.7255 (5.0120) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [4] [310/312] eta: 0:00:01 lr: 0.000999 min_lr: 0.000999 loss: 4.8879 (5.0082) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [4] [311/312] eta: 0:00:00 lr: 0.001000 min_lr: 0.001000 loss: 4.6027 (5.0059) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [4] Total time: 0:03:48 (0.7325 s / it) Averaged stats: lr: 0.001000 min_lr: 0.001000 loss: 4.6027 (4.9810) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 2.7782 (2.7782) acc1: 41.1458 (41.1458) acc5: 64.3229 (64.3229) time: 4.5964 data: 4.3769 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 3.2527 (3.0790) acc1: 33.8542 (33.3120) acc5: 58.0729 (57.1200) time: 0.6624 data: 0.4864 max mem: 64948 Test: Total time: 0:00:06 (0.6721 s / it) * Acc@1 33.274 Acc@5 58.950 loss 3.063 Accuracy of the model on the 50000 test images: 33.3% Max accuracy: 33.27% Test: [0/9] eta: 0:00:40 loss: 6.8929 (6.8929) acc1: 0.0000 (0.0000) acc5: 1.5625 (1.5625) time: 4.4659 data: 4.2617 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9090 (6.9100) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.6475 data: 0.4736 max mem: 64948 Test: Total time: 0:00:05 (0.6551 s / it) * Acc@1 0.098 Acc@5 0.492 loss 6.910 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [5] [ 0/312] eta: 0:57:59 lr: 0.001000 min_lr: 0.001000 loss: 3.9632 (3.9632) weight_decay: 0.0500 (0.0500) time: 11.1516 data: 7.7647 max mem: 64948 Epoch: [5] [ 10/312] eta: 0:08:21 lr: 0.001007 min_lr: 0.001007 loss: 4.3230 (4.5106) weight_decay: 0.0500 (0.0500) time: 1.6609 data: 0.7063 max mem: 64948 Epoch: [5] [ 20/312] eta: 0:05:51 lr: 0.001013 min_lr: 0.001013 loss: 4.7119 (4.6241) weight_decay: 0.0500 (0.0500) time: 0.7072 data: 0.0004 max mem: 64948 Epoch: [5] [ 30/312] eta: 0:04:53 lr: 0.001019 min_lr: 0.001019 loss: 4.8221 (4.6990) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0003 max mem: 64948 Epoch: [5] [ 40/312] eta: 0:04:20 lr: 0.001026 min_lr: 0.001026 loss: 4.9512 (4.7840) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0003 max mem: 64948 Epoch: [5] [ 50/312] eta: 0:03:57 lr: 0.001032 min_lr: 0.001032 loss: 4.9906 (4.7954) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0003 max mem: 64948 Epoch: [5] [ 60/312] eta: 0:03:39 lr: 0.001039 min_lr: 0.001039 loss: 4.7888 (4.7773) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [5] [ 70/312] eta: 0:03:25 lr: 0.001045 min_lr: 0.001045 loss: 4.7256 (4.7878) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [5] [ 80/312] eta: 0:03:12 lr: 0.001051 min_lr: 0.001051 loss: 4.9060 (4.7883) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [5] [ 90/312] eta: 0:03:00 lr: 0.001058 min_lr: 0.001058 loss: 4.5003 (4.7392) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [5] [100/312] eta: 0:02:49 lr: 0.001064 min_lr: 0.001064 loss: 4.3380 (4.7090) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [5] [110/312] eta: 0:02:39 lr: 0.001071 min_lr: 0.001071 loss: 4.5303 (4.7108) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [5] [120/312] eta: 0:02:30 lr: 0.001077 min_lr: 0.001077 loss: 4.7854 (4.7046) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [5] [130/312] eta: 0:02:21 lr: 0.001084 min_lr: 0.001084 loss: 4.9042 (4.7079) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [5] [140/312] eta: 0:02:12 lr: 0.001090 min_lr: 0.001090 loss: 4.9940 (4.7262) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [5] [150/312] eta: 0:02:04 lr: 0.001096 min_lr: 0.001096 loss: 4.7112 (4.7162) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [5] [160/312] eta: 0:01:55 lr: 0.001103 min_lr: 0.001103 loss: 4.4978 (4.7091) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [5] [170/312] eta: 0:01:47 lr: 0.001109 min_lr: 0.001109 loss: 4.9536 (4.7210) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [5] [180/312] eta: 0:01:39 lr: 0.001116 min_lr: 0.001116 loss: 4.9536 (4.7107) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [5] [190/312] eta: 0:01:31 lr: 0.001122 min_lr: 0.001122 loss: 4.3542 (4.6998) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [5] [200/312] eta: 0:01:23 lr: 0.001128 min_lr: 0.001128 loss: 4.4751 (4.6989) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [5] [210/312] eta: 0:01:16 lr: 0.001135 min_lr: 0.001135 loss: 4.9296 (4.7001) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [5] [220/312] eta: 0:01:08 lr: 0.001141 min_lr: 0.001141 loss: 4.9296 (4.7035) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [5] [230/312] eta: 0:01:00 lr: 0.001148 min_lr: 0.001148 loss: 4.8076 (4.6964) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [5] [240/312] eta: 0:00:53 lr: 0.001154 min_lr: 0.001154 loss: 4.7840 (4.6997) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [5] [250/312] eta: 0:00:45 lr: 0.001160 min_lr: 0.001160 loss: 4.7017 (4.6897) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [5] [260/312] eta: 0:00:38 lr: 0.001167 min_lr: 0.001167 loss: 4.5209 (4.6899) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [5] [270/312] eta: 0:00:30 lr: 0.001173 min_lr: 0.001173 loss: 4.9091 (4.6931) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [5] [280/312] eta: 0:00:23 lr: 0.001180 min_lr: 0.001180 loss: 4.7987 (4.6896) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0009 max mem: 64948 Epoch: [5] [290/312] eta: 0:00:16 lr: 0.001186 min_lr: 0.001186 loss: 4.6886 (4.6801) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [5] [300/312] eta: 0:00:08 lr: 0.001192 min_lr: 0.001192 loss: 4.5408 (4.6759) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [5] [310/312] eta: 0:00:01 lr: 0.001199 min_lr: 0.001199 loss: 4.4676 (4.6643) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [5] [311/312] eta: 0:00:00 lr: 0.001200 min_lr: 0.001200 loss: 4.4676 (4.6630) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [5] Total time: 0:03:48 (0.7312 s / it) Averaged stats: lr: 0.001200 min_lr: 0.001200 loss: 4.4676 (4.6763) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 2.3281 (2.3281) acc1: 49.7396 (49.7396) acc5: 72.3958 (72.3958) time: 4.6497 data: 4.4383 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.8920 (2.6816) acc1: 41.1458 (40.6720) acc5: 63.2812 (65.0560) time: 0.6679 data: 0.4932 max mem: 64948 Test: Total time: 0:00:06 (0.6945 s / it) * Acc@1 40.890 Acc@5 66.126 loss 2.651 Accuracy of the model on the 50000 test images: 40.9% Max accuracy: 40.89% Test: [0/9] eta: 0:00:41 loss: 6.8908 (6.8908) acc1: 0.0000 (0.0000) acc5: 1.5625 (1.5625) time: 4.5876 data: 4.3753 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9070 (6.9107) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.6610 data: 0.4862 max mem: 64948 Test: Total time: 0:00:06 (0.6697 s / it) * Acc@1 0.100 Acc@5 0.500 loss 6.911 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.10% Epoch: [6] [ 0/312] eta: 0:48:10 lr: 0.001200 min_lr: 0.001200 loss: 5.1594 (5.1594) weight_decay: 0.0500 (0.0500) time: 9.2649 data: 8.4056 max mem: 64948 Epoch: [6] [ 10/312] eta: 0:07:34 lr: 0.001207 min_lr: 0.001207 loss: 4.8626 (4.6829) weight_decay: 0.0500 (0.0500) time: 1.5038 data: 0.7645 max mem: 64948 Epoch: [6] [ 20/312] eta: 0:05:26 lr: 0.001213 min_lr: 0.001213 loss: 4.7204 (4.5514) weight_decay: 0.0500 (0.0500) time: 0.7121 data: 0.0004 max mem: 64948 Epoch: [6] [ 30/312] eta: 0:04:36 lr: 0.001219 min_lr: 0.001219 loss: 4.4221 (4.5283) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [6] [ 40/312] eta: 0:04:08 lr: 0.001226 min_lr: 0.001226 loss: 4.5417 (4.5391) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [6] [ 50/312] eta: 0:03:47 lr: 0.001232 min_lr: 0.001232 loss: 4.1833 (4.4608) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [6] [ 60/312] eta: 0:03:31 lr: 0.001239 min_lr: 0.001239 loss: 4.2126 (4.4441) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [6] [ 70/312] eta: 0:03:18 lr: 0.001245 min_lr: 0.001245 loss: 4.4028 (4.4470) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [6] [ 80/312] eta: 0:03:06 lr: 0.001251 min_lr: 0.001251 loss: 4.4993 (4.4255) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [6] [ 90/312] eta: 0:02:55 lr: 0.001258 min_lr: 0.001258 loss: 4.4012 (4.4207) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [6] [100/312] eta: 0:02:45 lr: 0.001264 min_lr: 0.001264 loss: 4.4012 (4.4060) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [6] [110/312] eta: 0:02:36 lr: 0.001271 min_lr: 0.001271 loss: 4.3314 (4.3977) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [6] [120/312] eta: 0:02:27 lr: 0.001277 min_lr: 0.001277 loss: 4.3314 (4.3991) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [6] [130/312] eta: 0:02:18 lr: 0.001284 min_lr: 0.001284 loss: 4.4679 (4.3961) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [6] [140/312] eta: 0:02:10 lr: 0.001290 min_lr: 0.001290 loss: 4.5429 (4.4151) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [6] [150/312] eta: 0:02:02 lr: 0.001296 min_lr: 0.001296 loss: 4.6057 (4.4198) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [6] [160/312] eta: 0:01:54 lr: 0.001303 min_lr: 0.001303 loss: 4.7549 (4.4340) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [6] [170/312] eta: 0:01:46 lr: 0.001309 min_lr: 0.001309 loss: 4.6617 (4.4365) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [6] [180/312] eta: 0:01:38 lr: 0.001316 min_lr: 0.001316 loss: 4.3954 (4.4356) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [6] [190/312] eta: 0:01:30 lr: 0.001322 min_lr: 0.001322 loss: 4.2749 (4.4218) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [6] [200/312] eta: 0:01:22 lr: 0.001328 min_lr: 0.001328 loss: 4.3736 (4.4285) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [6] [210/312] eta: 0:01:15 lr: 0.001335 min_lr: 0.001335 loss: 4.3736 (4.4153) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [6] [220/312] eta: 0:01:07 lr: 0.001341 min_lr: 0.001341 loss: 4.6009 (4.4295) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [6] [230/312] eta: 0:01:00 lr: 0.001348 min_lr: 0.001348 loss: 4.6422 (4.4206) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [6] [240/312] eta: 0:00:52 lr: 0.001354 min_lr: 0.001354 loss: 4.1303 (4.4166) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [6] [250/312] eta: 0:00:45 lr: 0.001360 min_lr: 0.001360 loss: 4.5324 (4.4166) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [6] [260/312] eta: 0:00:37 lr: 0.001367 min_lr: 0.001367 loss: 4.5608 (4.4189) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [6] [270/312] eta: 0:00:30 lr: 0.001373 min_lr: 0.001373 loss: 4.5658 (4.4216) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [6] [280/312] eta: 0:00:23 lr: 0.001380 min_lr: 0.001380 loss: 4.4724 (4.4192) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [6] [290/312] eta: 0:00:15 lr: 0.001386 min_lr: 0.001386 loss: 4.5180 (4.4188) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [6] [300/312] eta: 0:00:08 lr: 0.001393 min_lr: 0.001393 loss: 4.5180 (4.4149) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [6] [310/312] eta: 0:00:01 lr: 0.001399 min_lr: 0.001399 loss: 4.3044 (4.4139) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [6] [311/312] eta: 0:00:00 lr: 0.001400 min_lr: 0.001400 loss: 4.2049 (4.4116) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [6] Total time: 0:03:46 (0.7256 s / it) Averaged stats: lr: 0.001400 min_lr: 0.001400 loss: 4.2049 (4.4215) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 2.0659 (2.0659) acc1: 52.0833 (52.0833) acc5: 76.0417 (76.0417) time: 4.5106 data: 4.3013 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.5768 (2.4194) acc1: 43.4896 (44.6400) acc5: 70.8333 (70.4640) time: 0.6524 data: 0.4780 max mem: 64948 Test: Total time: 0:00:06 (0.6763 s / it) * Acc@1 45.436 Acc@5 71.202 loss 2.387 Accuracy of the model on the 50000 test images: 45.4% Max accuracy: 45.44% Test: [0/9] eta: 0:00:41 loss: 6.8887 (6.8887) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.6106 data: 4.3927 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9053 (6.9116) acc1: 0.0000 (0.0960) acc5: 0.7812 (0.4800) time: 0.6655 data: 0.4882 max mem: 64948 Test: Total time: 0:00:06 (0.6735 s / it) * Acc@1 0.102 Acc@5 0.496 loss 6.911 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.10% Epoch: [7] [ 0/312] eta: 0:51:13 lr: 0.001400 min_lr: 0.001400 loss: 4.6919 (4.6919) weight_decay: 0.0500 (0.0500) time: 9.8524 data: 9.1000 max mem: 64948 Epoch: [7] [ 10/312] eta: 0:07:46 lr: 0.001407 min_lr: 0.001407 loss: 4.6224 (4.4688) weight_decay: 0.0500 (0.0500) time: 1.5436 data: 0.8276 max mem: 64948 Epoch: [7] [ 20/312] eta: 0:05:32 lr: 0.001413 min_lr: 0.001413 loss: 4.5221 (4.2798) weight_decay: 0.0500 (0.0500) time: 0.7031 data: 0.0004 max mem: 64948 Epoch: [7] [ 30/312] eta: 0:04:40 lr: 0.001419 min_lr: 0.001419 loss: 4.3042 (4.2841) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [7] [ 40/312] eta: 0:04:10 lr: 0.001426 min_lr: 0.001426 loss: 4.1411 (4.2133) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [7] [ 50/312] eta: 0:03:49 lr: 0.001432 min_lr: 0.001432 loss: 4.1183 (4.1997) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [7] [ 60/312] eta: 0:03:33 lr: 0.001439 min_lr: 0.001439 loss: 4.0857 (4.2096) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [7] [ 70/312] eta: 0:03:20 lr: 0.001445 min_lr: 0.001445 loss: 4.0802 (4.2015) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [7] [ 80/312] eta: 0:03:08 lr: 0.001452 min_lr: 0.001452 loss: 4.2470 (4.2103) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [7] [ 90/312] eta: 0:02:57 lr: 0.001458 min_lr: 0.001458 loss: 4.4225 (4.2300) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [7] [100/312] eta: 0:02:47 lr: 0.001464 min_lr: 0.001464 loss: 4.3313 (4.2138) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [7] [110/312] eta: 0:02:37 lr: 0.001471 min_lr: 0.001471 loss: 4.1901 (4.2106) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [7] [120/312] eta: 0:02:28 lr: 0.001477 min_lr: 0.001477 loss: 4.2635 (4.2111) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [7] [130/312] eta: 0:02:19 lr: 0.001484 min_lr: 0.001484 loss: 4.3537 (4.2197) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [7] [140/312] eta: 0:02:10 lr: 0.001490 min_lr: 0.001490 loss: 4.4988 (4.2319) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [7] [150/312] eta: 0:02:02 lr: 0.001496 min_lr: 0.001496 loss: 4.4988 (4.2383) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [7] [160/312] eta: 0:01:54 lr: 0.001503 min_lr: 0.001503 loss: 4.3916 (4.2319) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [7] [170/312] eta: 0:01:46 lr: 0.001509 min_lr: 0.001509 loss: 4.2031 (4.2298) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [7] [180/312] eta: 0:01:38 lr: 0.001516 min_lr: 0.001516 loss: 4.1531 (4.2197) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [7] [190/312] eta: 0:01:30 lr: 0.001522 min_lr: 0.001522 loss: 4.0857 (4.2146) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [7] [200/312] eta: 0:01:23 lr: 0.001528 min_lr: 0.001528 loss: 4.3210 (4.2189) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [7] [210/312] eta: 0:01:15 lr: 0.001535 min_lr: 0.001535 loss: 4.1683 (4.2068) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [7] [220/312] eta: 0:01:07 lr: 0.001541 min_lr: 0.001541 loss: 4.2141 (4.2152) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [7] [230/312] eta: 0:01:00 lr: 0.001548 min_lr: 0.001548 loss: 4.4968 (4.2123) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [7] [240/312] eta: 0:00:52 lr: 0.001554 min_lr: 0.001554 loss: 4.0060 (4.2055) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [7] [250/312] eta: 0:00:45 lr: 0.001561 min_lr: 0.001561 loss: 4.0205 (4.2039) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [7] [260/312] eta: 0:00:37 lr: 0.001567 min_lr: 0.001567 loss: 4.3100 (4.2046) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [7] [270/312] eta: 0:00:30 lr: 0.001573 min_lr: 0.001573 loss: 4.3868 (4.2116) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [7] [280/312] eta: 0:00:23 lr: 0.001580 min_lr: 0.001580 loss: 4.0245 (4.1966) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0010 max mem: 64948 Epoch: [7] [290/312] eta: 0:00:15 lr: 0.001586 min_lr: 0.001586 loss: 3.8533 (4.1937) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0009 max mem: 64948 Epoch: [7] [300/312] eta: 0:00:08 lr: 0.001593 min_lr: 0.001593 loss: 4.2492 (4.1899) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [7] [310/312] eta: 0:00:01 lr: 0.001599 min_lr: 0.001599 loss: 4.0268 (4.1893) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [7] [311/312] eta: 0:00:00 lr: 0.001600 min_lr: 0.001600 loss: 4.0268 (4.1873) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [7] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.001600 min_lr: 0.001600 loss: 4.0268 (4.2104) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.9199 (1.9199) acc1: 58.8542 (58.8542) acc5: 77.6042 (77.6042) time: 4.7347 data: 4.5210 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.4245 (2.1890) acc1: 47.9167 (50.4640) acc5: 74.2188 (74.4000) time: 0.6773 data: 0.5024 max mem: 64948 Test: Total time: 0:00:06 (0.7106 s / it) * Acc@1 50.170 Acc@5 74.870 loss 2.185 Accuracy of the model on the 50000 test images: 50.2% Max accuracy: 50.17% Test: [0/9] eta: 0:00:40 loss: 6.8864 (6.8864) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.5173 data: 4.2995 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9041 (6.9125) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6535 data: 0.4778 max mem: 64948 Test: Total time: 0:00:05 (0.6612 s / it) * Acc@1 0.102 Acc@5 0.496 loss 6.912 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.10% Epoch: [8] [ 0/312] eta: 0:46:35 lr: 0.001600 min_lr: 0.001600 loss: 3.7399 (3.7399) weight_decay: 0.0500 (0.0500) time: 8.9605 data: 8.1753 max mem: 64948 Epoch: [8] [ 10/312] eta: 0:07:50 lr: 0.001607 min_lr: 0.001607 loss: 3.7537 (3.9588) weight_decay: 0.0500 (0.0500) time: 1.5574 data: 0.8333 max mem: 64948 Epoch: [8] [ 20/312] eta: 0:05:35 lr: 0.001613 min_lr: 0.001613 loss: 4.0093 (4.0984) weight_decay: 0.0500 (0.0500) time: 0.7595 data: 0.0497 max mem: 64948 Epoch: [8] [ 30/312] eta: 0:04:42 lr: 0.001619 min_lr: 0.001619 loss: 4.3207 (4.1702) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0003 max mem: 64948 Epoch: [8] [ 40/312] eta: 0:04:12 lr: 0.001626 min_lr: 0.001626 loss: 4.3134 (4.1785) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [8] [ 50/312] eta: 0:03:51 lr: 0.001632 min_lr: 0.001632 loss: 3.9850 (4.1400) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [8] [ 60/312] eta: 0:03:34 lr: 0.001639 min_lr: 0.001639 loss: 4.3743 (4.1872) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [8] [ 70/312] eta: 0:03:20 lr: 0.001645 min_lr: 0.001645 loss: 4.3743 (4.1927) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [8] [ 80/312] eta: 0:03:08 lr: 0.001652 min_lr: 0.001652 loss: 4.2309 (4.1809) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [8] [ 90/312] eta: 0:02:57 lr: 0.001658 min_lr: 0.001658 loss: 4.1414 (4.1737) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [8] [100/312] eta: 0:02:47 lr: 0.001664 min_lr: 0.001664 loss: 4.0133 (4.1451) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [8] [110/312] eta: 0:02:37 lr: 0.001671 min_lr: 0.001671 loss: 4.0133 (4.1464) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [8] [120/312] eta: 0:02:28 lr: 0.001677 min_lr: 0.001677 loss: 4.2066 (4.1285) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [8] [130/312] eta: 0:02:19 lr: 0.001684 min_lr: 0.001684 loss: 4.2104 (4.1326) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [8] [140/312] eta: 0:02:11 lr: 0.001690 min_lr: 0.001690 loss: 4.2104 (4.1268) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [8] [150/312] eta: 0:02:02 lr: 0.001696 min_lr: 0.001696 loss: 4.3111 (4.1355) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [8] [160/312] eta: 0:01:54 lr: 0.001703 min_lr: 0.001703 loss: 4.4150 (4.1247) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [8] [170/312] eta: 0:01:46 lr: 0.001709 min_lr: 0.001709 loss: 4.0019 (4.1207) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [8] [180/312] eta: 0:01:38 lr: 0.001716 min_lr: 0.001716 loss: 4.0452 (4.1232) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [8] [190/312] eta: 0:01:30 lr: 0.001722 min_lr: 0.001722 loss: 4.3245 (4.1262) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [8] [200/312] eta: 0:01:23 lr: 0.001728 min_lr: 0.001728 loss: 4.2925 (4.1251) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [8] [210/312] eta: 0:01:15 lr: 0.001735 min_lr: 0.001735 loss: 4.3398 (4.1357) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [8] [220/312] eta: 0:01:07 lr: 0.001741 min_lr: 0.001741 loss: 4.1957 (4.1266) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [8] [230/312] eta: 0:01:00 lr: 0.001748 min_lr: 0.001748 loss: 4.1091 (4.1299) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [8] [240/312] eta: 0:00:52 lr: 0.001754 min_lr: 0.001754 loss: 4.0976 (4.1217) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [8] [250/312] eta: 0:00:45 lr: 0.001761 min_lr: 0.001761 loss: 3.8463 (4.1112) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [8] [260/312] eta: 0:00:38 lr: 0.001767 min_lr: 0.001767 loss: 3.9674 (4.1031) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [8] [270/312] eta: 0:00:30 lr: 0.001773 min_lr: 0.001773 loss: 4.1091 (4.1032) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [8] [280/312] eta: 0:00:23 lr: 0.001780 min_lr: 0.001780 loss: 4.2167 (4.1023) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [8] [290/312] eta: 0:00:16 lr: 0.001786 min_lr: 0.001786 loss: 3.6866 (4.0896) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [8] [300/312] eta: 0:00:08 lr: 0.001793 min_lr: 0.001793 loss: 3.6866 (4.0816) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [8] [310/312] eta: 0:00:01 lr: 0.001799 min_lr: 0.001799 loss: 3.9818 (4.0788) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [8] [311/312] eta: 0:00:00 lr: 0.001800 min_lr: 0.001800 loss: 3.9734 (4.0784) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [8] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.001800 min_lr: 0.001800 loss: 3.9734 (4.0290) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.6442 (1.6442) acc1: 62.5000 (62.5000) acc5: 83.8542 (83.8542) time: 4.6716 data: 4.4524 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.1496 (1.9905) acc1: 52.6042 (53.7280) acc5: 75.0000 (77.2160) time: 0.6705 data: 0.4948 max mem: 64948 Test: Total time: 0:00:06 (0.6920 s / it) * Acc@1 53.732 Acc@5 77.994 loss 1.973 Accuracy of the model on the 50000 test images: 53.7% Max accuracy: 53.73% Test: [0/9] eta: 0:00:39 loss: 6.8842 (6.8842) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.3449 data: 4.1357 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9034 (6.9135) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6361 data: 0.4618 max mem: 64948 Test: Total time: 0:00:05 (0.6439 s / it) * Acc@1 0.102 Acc@5 0.498 loss 6.913 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [9] [ 0/312] eta: 0:50:22 lr: 0.001800 min_lr: 0.001800 loss: 4.1827 (4.1827) weight_decay: 0.0500 (0.0500) time: 9.6871 data: 8.0049 max mem: 64948 Epoch: [9] [ 10/312] eta: 0:07:51 lr: 0.001807 min_lr: 0.001807 loss: 4.1827 (4.0067) weight_decay: 0.0500 (0.0500) time: 1.5612 data: 0.7282 max mem: 64948 Epoch: [9] [ 20/312] eta: 0:05:35 lr: 0.001813 min_lr: 0.001813 loss: 4.1407 (3.9834) weight_decay: 0.0500 (0.0500) time: 0.7208 data: 0.0004 max mem: 64948 Epoch: [9] [ 30/312] eta: 0:04:42 lr: 0.001820 min_lr: 0.001820 loss: 3.8948 (3.9413) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [9] [ 40/312] eta: 0:04:12 lr: 0.001826 min_lr: 0.001826 loss: 3.8018 (3.8990) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [9] [ 50/312] eta: 0:03:51 lr: 0.001832 min_lr: 0.001832 loss: 3.5658 (3.8340) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [9] [ 60/312] eta: 0:03:34 lr: 0.001839 min_lr: 0.001839 loss: 3.5658 (3.8197) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [9] [ 70/312] eta: 0:03:20 lr: 0.001845 min_lr: 0.001845 loss: 3.8064 (3.8193) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [9] [ 80/312] eta: 0:03:08 lr: 0.001852 min_lr: 0.001852 loss: 3.6765 (3.7991) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [9] [ 90/312] eta: 0:02:57 lr: 0.001858 min_lr: 0.001858 loss: 3.9903 (3.8347) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [9] [100/312] eta: 0:02:47 lr: 0.001864 min_lr: 0.001864 loss: 3.9774 (3.8224) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [9] [110/312] eta: 0:02:37 lr: 0.001871 min_lr: 0.001871 loss: 3.8079 (3.8390) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [9] [120/312] eta: 0:02:28 lr: 0.001877 min_lr: 0.001877 loss: 4.1857 (3.8491) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [9] [130/312] eta: 0:02:19 lr: 0.001884 min_lr: 0.001884 loss: 4.0268 (3.8631) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [9] [140/312] eta: 0:02:11 lr: 0.001890 min_lr: 0.001890 loss: 3.9634 (3.8512) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [9] [150/312] eta: 0:02:03 lr: 0.001896 min_lr: 0.001896 loss: 3.7773 (3.8409) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [9] [160/312] eta: 0:01:54 lr: 0.001903 min_lr: 0.001903 loss: 3.8757 (3.8413) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [9] [170/312] eta: 0:01:46 lr: 0.001909 min_lr: 0.001909 loss: 4.0457 (3.8541) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [9] [180/312] eta: 0:01:38 lr: 0.001916 min_lr: 0.001916 loss: 3.9030 (3.8425) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [9] [190/312] eta: 0:01:30 lr: 0.001922 min_lr: 0.001922 loss: 3.8886 (3.8507) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [9] [200/312] eta: 0:01:23 lr: 0.001929 min_lr: 0.001929 loss: 3.8886 (3.8443) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [9] [210/312] eta: 0:01:15 lr: 0.001935 min_lr: 0.001935 loss: 3.8548 (3.8529) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [9] [220/312] eta: 0:01:07 lr: 0.001941 min_lr: 0.001941 loss: 3.8531 (3.8417) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [9] [230/312] eta: 0:01:00 lr: 0.001948 min_lr: 0.001948 loss: 3.8531 (3.8507) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0003 max mem: 64948 Epoch: [9] [240/312] eta: 0:00:52 lr: 0.001954 min_lr: 0.001954 loss: 4.0661 (3.8475) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [9] [250/312] eta: 0:00:45 lr: 0.001961 min_lr: 0.001961 loss: 3.9555 (3.8482) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [9] [260/312] eta: 0:00:38 lr: 0.001967 min_lr: 0.001967 loss: 3.9551 (3.8412) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [9] [270/312] eta: 0:00:30 lr: 0.001973 min_lr: 0.001973 loss: 3.8614 (3.8402) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [9] [280/312] eta: 0:00:23 lr: 0.001980 min_lr: 0.001980 loss: 3.7552 (3.8316) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [9] [290/312] eta: 0:00:16 lr: 0.001986 min_lr: 0.001986 loss: 3.8239 (3.8370) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [9] [300/312] eta: 0:00:08 lr: 0.001993 min_lr: 0.001993 loss: 3.9797 (3.8354) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [9] [310/312] eta: 0:00:01 lr: 0.001999 min_lr: 0.001999 loss: 3.8054 (3.8377) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [9] [311/312] eta: 0:00:00 lr: 0.002000 min_lr: 0.002000 loss: 3.8054 (3.8385) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [9] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.002000 min_lr: 0.002000 loss: 3.8054 (3.8689) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.5413 (1.5413) acc1: 66.4062 (66.4062) acc5: 85.9375 (85.9375) time: 4.6944 data: 4.4831 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.0403 (1.8829) acc1: 54.1667 (55.4560) acc5: 78.3854 (79.2640) time: 0.6729 data: 0.4982 max mem: 64948 Test: Total time: 0:00:06 (0.6972 s / it) * Acc@1 56.408 Acc@5 79.950 loss 1.857 Accuracy of the model on the 50000 test images: 56.4% Max accuracy: 56.41% Test: [0/9] eta: 0:00:42 loss: 6.8822 (6.8822) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.7558 data: 4.5409 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9031 (6.9145) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6797 data: 0.5046 max mem: 64948 Test: Total time: 0:00:06 (0.6877 s / it) * Acc@1 0.104 Acc@5 0.500 loss 6.914 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.10% Epoch: [10] [ 0/312] eta: 0:48:55 lr: 0.002000 min_lr: 0.002000 loss: 3.9491 (3.9491) weight_decay: 0.0500 (0.0500) time: 9.4078 data: 8.1185 max mem: 64948 Epoch: [10] [ 10/312] eta: 0:07:38 lr: 0.002007 min_lr: 0.002007 loss: 3.7608 (3.7554) weight_decay: 0.0500 (0.0500) time: 1.5180 data: 0.7384 max mem: 64948 Epoch: [10] [ 20/312] eta: 0:05:28 lr: 0.002013 min_lr: 0.002013 loss: 3.6570 (3.6268) weight_decay: 0.0500 (0.0500) time: 0.7104 data: 0.0003 max mem: 64948 Epoch: [10] [ 30/312] eta: 0:04:38 lr: 0.002020 min_lr: 0.002020 loss: 3.7208 (3.6766) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [10] [ 40/312] eta: 0:04:08 lr: 0.002026 min_lr: 0.002026 loss: 3.9339 (3.6885) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [10] [ 50/312] eta: 0:03:48 lr: 0.002032 min_lr: 0.002032 loss: 3.8401 (3.7114) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [10] [ 60/312] eta: 0:03:32 lr: 0.002039 min_lr: 0.002039 loss: 3.8401 (3.7340) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [10] [ 70/312] eta: 0:03:19 lr: 0.002045 min_lr: 0.002045 loss: 3.7502 (3.7187) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [10] [ 80/312] eta: 0:03:07 lr: 0.002052 min_lr: 0.002052 loss: 3.8807 (3.7410) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [10] [ 90/312] eta: 0:02:56 lr: 0.002058 min_lr: 0.002058 loss: 4.0570 (3.7460) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [10] [100/312] eta: 0:02:46 lr: 0.002064 min_lr: 0.002064 loss: 3.8982 (3.7608) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [10] [110/312] eta: 0:02:36 lr: 0.002071 min_lr: 0.002071 loss: 3.8962 (3.7664) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [10] [120/312] eta: 0:02:27 lr: 0.002077 min_lr: 0.002077 loss: 4.0424 (3.8026) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [10] [130/312] eta: 0:02:19 lr: 0.002084 min_lr: 0.002084 loss: 4.0977 (3.7985) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [10] [140/312] eta: 0:02:10 lr: 0.002090 min_lr: 0.002090 loss: 3.6049 (3.7954) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [10] [150/312] eta: 0:02:02 lr: 0.002096 min_lr: 0.002096 loss: 3.6863 (3.7795) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [10] [160/312] eta: 0:01:54 lr: 0.002103 min_lr: 0.002103 loss: 3.6111 (3.7744) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [10] [170/312] eta: 0:01:46 lr: 0.002109 min_lr: 0.002109 loss: 3.6329 (3.7696) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [10] [180/312] eta: 0:01:38 lr: 0.002116 min_lr: 0.002116 loss: 3.6771 (3.7717) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [10] [190/312] eta: 0:01:30 lr: 0.002122 min_lr: 0.002122 loss: 3.9398 (3.7810) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [10] [200/312] eta: 0:01:22 lr: 0.002129 min_lr: 0.002129 loss: 3.8445 (3.7792) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [10] [210/312] eta: 0:01:15 lr: 0.002135 min_lr: 0.002135 loss: 3.6495 (3.7630) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [10] [220/312] eta: 0:01:07 lr: 0.002141 min_lr: 0.002141 loss: 3.4037 (3.7487) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [10] [230/312] eta: 0:01:00 lr: 0.002148 min_lr: 0.002148 loss: 3.7211 (3.7511) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [10] [240/312] eta: 0:00:52 lr: 0.002154 min_lr: 0.002154 loss: 3.8966 (3.7522) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [10] [250/312] eta: 0:00:45 lr: 0.002161 min_lr: 0.002161 loss: 3.8201 (3.7501) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [10] [260/312] eta: 0:00:37 lr: 0.002167 min_lr: 0.002167 loss: 3.8307 (3.7511) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [10] [270/312] eta: 0:00:30 lr: 0.002173 min_lr: 0.002173 loss: 3.9853 (3.7550) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [10] [280/312] eta: 0:00:23 lr: 0.002180 min_lr: 0.002180 loss: 3.6177 (3.7435) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [10] [290/312] eta: 0:00:15 lr: 0.002186 min_lr: 0.002186 loss: 3.7190 (3.7488) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0008 max mem: 64948 Epoch: [10] [300/312] eta: 0:00:08 lr: 0.002193 min_lr: 0.002193 loss: 4.0141 (3.7464) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [10] [310/312] eta: 0:00:01 lr: 0.002199 min_lr: 0.002199 loss: 3.7494 (3.7408) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [10] [311/312] eta: 0:00:00 lr: 0.002200 min_lr: 0.002200 loss: 3.7494 (3.7400) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [10] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.002200 min_lr: 0.002200 loss: 3.7494 (3.7359) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.3362 (1.3362) acc1: 69.5312 (69.5312) acc5: 89.0625 (89.0625) time: 4.7500 data: 4.5379 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.9719 (1.7624) acc1: 55.9896 (57.6320) acc5: 79.9479 (81.4400) time: 0.6791 data: 0.5043 max mem: 64948 Test: Total time: 0:00:06 (0.7021 s / it) * Acc@1 58.356 Acc@5 81.352 loss 1.761 Accuracy of the model on the 50000 test images: 58.4% Max accuracy: 58.36% Test: [0/9] eta: 0:00:40 loss: 6.8808 (6.8808) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.5030 data: 4.2916 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9025 (6.9157) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.6516 data: 0.4770 max mem: 64948 Test: Total time: 0:00:05 (0.6613 s / it) * Acc@1 0.100 Acc@5 0.500 loss 6.916 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [11] [ 0/312] eta: 0:51:04 lr: 0.002200 min_lr: 0.002200 loss: 4.1319 (4.1319) weight_decay: 0.0500 (0.0500) time: 9.8210 data: 7.6184 max mem: 64948 Epoch: [11] [ 10/312] eta: 0:07:53 lr: 0.002207 min_lr: 0.002207 loss: 3.4732 (3.5464) weight_decay: 0.0500 (0.0500) time: 1.5688 data: 0.6930 max mem: 64948 Epoch: [11] [ 20/312] eta: 0:05:37 lr: 0.002213 min_lr: 0.002213 loss: 3.4732 (3.5728) weight_decay: 0.0500 (0.0500) time: 0.7210 data: 0.0004 max mem: 64948 Epoch: [11] [ 30/312] eta: 0:04:43 lr: 0.002220 min_lr: 0.002220 loss: 3.8891 (3.6205) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [11] [ 40/312] eta: 0:04:13 lr: 0.002226 min_lr: 0.002226 loss: 3.4629 (3.5459) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [11] [ 50/312] eta: 0:03:51 lr: 0.002232 min_lr: 0.002232 loss: 3.4629 (3.5528) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [11] [ 60/312] eta: 0:03:35 lr: 0.002239 min_lr: 0.002239 loss: 3.4961 (3.5731) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [11] [ 70/312] eta: 0:03:21 lr: 0.002245 min_lr: 0.002245 loss: 3.7666 (3.5828) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [11] [ 80/312] eta: 0:03:09 lr: 0.002252 min_lr: 0.002252 loss: 3.7666 (3.5955) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [11] [ 90/312] eta: 0:02:58 lr: 0.002258 min_lr: 0.002258 loss: 3.5469 (3.5799) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [11] [100/312] eta: 0:02:47 lr: 0.002264 min_lr: 0.002264 loss: 3.5469 (3.5972) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [11] [110/312] eta: 0:02:38 lr: 0.002271 min_lr: 0.002271 loss: 3.4825 (3.5779) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [11] [120/312] eta: 0:02:28 lr: 0.002277 min_lr: 0.002277 loss: 3.4789 (3.5779) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [11] [130/312] eta: 0:02:19 lr: 0.002284 min_lr: 0.002284 loss: 3.8582 (3.5973) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [11] [140/312] eta: 0:02:11 lr: 0.002290 min_lr: 0.002290 loss: 3.8547 (3.6051) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [11] [150/312] eta: 0:02:02 lr: 0.002297 min_lr: 0.002297 loss: 3.8244 (3.6159) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [11] [160/312] eta: 0:01:54 lr: 0.002303 min_lr: 0.002303 loss: 3.7994 (3.6177) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [11] [170/312] eta: 0:01:46 lr: 0.002309 min_lr: 0.002309 loss: 3.7744 (3.6101) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [11] [180/312] eta: 0:01:38 lr: 0.002316 min_lr: 0.002316 loss: 3.7098 (3.6173) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [11] [190/312] eta: 0:01:30 lr: 0.002322 min_lr: 0.002322 loss: 3.7623 (3.6202) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [11] [200/312] eta: 0:01:23 lr: 0.002329 min_lr: 0.002329 loss: 3.7623 (3.6315) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [11] [210/312] eta: 0:01:15 lr: 0.002335 min_lr: 0.002335 loss: 3.7486 (3.6314) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [11] [220/312] eta: 0:01:07 lr: 0.002341 min_lr: 0.002341 loss: 3.7033 (3.6385) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [11] [230/312] eta: 0:01:00 lr: 0.002348 min_lr: 0.002348 loss: 3.8654 (3.6360) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [11] [240/312] eta: 0:00:52 lr: 0.002354 min_lr: 0.002354 loss: 3.8368 (3.6372) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [11] [250/312] eta: 0:00:45 lr: 0.002361 min_lr: 0.002361 loss: 3.7715 (3.6376) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [11] [260/312] eta: 0:00:38 lr: 0.002367 min_lr: 0.002367 loss: 3.6774 (3.6362) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [11] [270/312] eta: 0:00:30 lr: 0.002373 min_lr: 0.002373 loss: 3.6670 (3.6346) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [11] [280/312] eta: 0:00:23 lr: 0.002380 min_lr: 0.002380 loss: 3.6459 (3.6266) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [11] [290/312] eta: 0:00:16 lr: 0.002386 min_lr: 0.002386 loss: 3.6579 (3.6331) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [11] [300/312] eta: 0:00:08 lr: 0.002393 min_lr: 0.002393 loss: 3.8907 (3.6327) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [11] [310/312] eta: 0:00:01 lr: 0.002399 min_lr: 0.002399 loss: 3.8253 (3.6366) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [11] [311/312] eta: 0:00:00 lr: 0.002400 min_lr: 0.002400 loss: 3.8253 (3.6362) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [11] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.002400 min_lr: 0.002400 loss: 3.8253 (3.6139) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:50 loss: 1.3895 (1.3895) acc1: 68.2292 (68.2292) acc5: 85.6771 (85.6771) time: 5.5854 data: 5.3654 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.7819 (1.6699) acc1: 58.8542 (59.9680) acc5: 83.0729 (83.0080) time: 0.7723 data: 0.5962 max mem: 64948 Test: Total time: 0:00:07 (0.7930 s / it) * Acc@1 60.346 Acc@5 82.828 loss 1.665 Accuracy of the model on the 50000 test images: 60.3% Max accuracy: 60.35% Test: [0/9] eta: 0:00:44 loss: 6.8799 (6.8799) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.9567 data: 4.7487 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9015 (6.9170) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.7020 data: 0.5277 max mem: 64948 Test: Total time: 0:00:06 (0.7131 s / it) * Acc@1 0.100 Acc@5 0.504 loss 6.917 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [12] [ 0/312] eta: 0:46:04 lr: 0.002400 min_lr: 0.002400 loss: 3.9237 (3.9237) weight_decay: 0.0500 (0.0500) time: 8.8606 data: 7.1017 max mem: 64948 Epoch: [12] [ 10/312] eta: 0:07:28 lr: 0.002407 min_lr: 0.002407 loss: 3.3884 (3.3660) weight_decay: 0.0500 (0.0500) time: 1.4838 data: 0.6461 max mem: 64948 Epoch: [12] [ 20/312] eta: 0:05:23 lr: 0.002413 min_lr: 0.002413 loss: 3.4870 (3.5078) weight_decay: 0.0500 (0.0500) time: 0.7204 data: 0.0004 max mem: 64948 Epoch: [12] [ 30/312] eta: 0:04:35 lr: 0.002420 min_lr: 0.002420 loss: 3.7581 (3.5100) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [12] [ 40/312] eta: 0:04:06 lr: 0.002426 min_lr: 0.002426 loss: 3.6299 (3.5377) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [12] [ 50/312] eta: 0:03:46 lr: 0.002432 min_lr: 0.002432 loss: 3.6299 (3.5491) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [12] [ 60/312] eta: 0:03:31 lr: 0.002439 min_lr: 0.002439 loss: 3.5719 (3.5452) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [12] [ 70/312] eta: 0:03:18 lr: 0.002445 min_lr: 0.002445 loss: 3.4616 (3.5179) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [12] [ 80/312] eta: 0:03:06 lr: 0.002452 min_lr: 0.002452 loss: 3.5976 (3.5325) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [12] [ 90/312] eta: 0:02:55 lr: 0.002458 min_lr: 0.002458 loss: 3.7559 (3.5407) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [12] [100/312] eta: 0:02:45 lr: 0.002464 min_lr: 0.002464 loss: 3.5824 (3.5269) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [12] [110/312] eta: 0:02:36 lr: 0.002471 min_lr: 0.002471 loss: 3.6702 (3.5353) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [12] [120/312] eta: 0:02:27 lr: 0.002477 min_lr: 0.002477 loss: 3.6652 (3.5219) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [12] [130/312] eta: 0:02:18 lr: 0.002484 min_lr: 0.002484 loss: 3.6308 (3.5264) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [12] [140/312] eta: 0:02:10 lr: 0.002490 min_lr: 0.002490 loss: 3.5986 (3.5116) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [12] [150/312] eta: 0:02:02 lr: 0.002497 min_lr: 0.002497 loss: 3.4191 (3.5069) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [12] [160/312] eta: 0:01:53 lr: 0.002503 min_lr: 0.002503 loss: 3.5003 (3.4965) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [12] [170/312] eta: 0:01:45 lr: 0.002509 min_lr: 0.002509 loss: 3.5003 (3.5005) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [12] [180/312] eta: 0:01:38 lr: 0.002516 min_lr: 0.002516 loss: 3.5778 (3.4967) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [12] [190/312] eta: 0:01:30 lr: 0.002522 min_lr: 0.002522 loss: 3.5778 (3.4935) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [12] [200/312] eta: 0:01:22 lr: 0.002529 min_lr: 0.002529 loss: 3.6389 (3.4857) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [12] [210/312] eta: 0:01:15 lr: 0.002535 min_lr: 0.002535 loss: 3.5391 (3.4814) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [12] [220/312] eta: 0:01:07 lr: 0.002541 min_lr: 0.002541 loss: 3.5650 (3.4894) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [12] [230/312] eta: 0:01:00 lr: 0.002548 min_lr: 0.002548 loss: 3.6340 (3.4928) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [12] [240/312] eta: 0:00:52 lr: 0.002554 min_lr: 0.002554 loss: 3.5650 (3.4956) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [12] [250/312] eta: 0:00:45 lr: 0.002561 min_lr: 0.002561 loss: 3.6506 (3.4992) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [12] [260/312] eta: 0:00:37 lr: 0.002567 min_lr: 0.002567 loss: 3.6760 (3.4994) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [12] [270/312] eta: 0:00:30 lr: 0.002573 min_lr: 0.002573 loss: 3.4890 (3.4913) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [12] [280/312] eta: 0:00:23 lr: 0.002580 min_lr: 0.002580 loss: 3.5549 (3.4987) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [12] [290/312] eta: 0:00:15 lr: 0.002586 min_lr: 0.002586 loss: 3.5116 (3.4931) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [12] [300/312] eta: 0:00:08 lr: 0.002593 min_lr: 0.002593 loss: 3.5116 (3.5015) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [12] [310/312] eta: 0:00:01 lr: 0.002599 min_lr: 0.002599 loss: 3.7974 (3.5078) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [12] [311/312] eta: 0:00:00 lr: 0.002600 min_lr: 0.002600 loss: 3.7974 (3.5093) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [12] Total time: 0:03:46 (0.7257 s / it) Averaged stats: lr: 0.002600 min_lr: 0.002600 loss: 3.7974 (3.5096) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 1.3197 (1.3197) acc1: 67.9688 (67.9688) acc5: 87.5000 (87.5000) time: 4.8618 data: 4.6473 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.7759 (1.6340) acc1: 57.8125 (60.6080) acc5: 83.3333 (83.4880) time: 0.6921 data: 0.5164 max mem: 64948 Test: Total time: 0:00:06 (0.7159 s / it) * Acc@1 61.310 Acc@5 83.728 loss 1.621 Accuracy of the model on the 50000 test images: 61.3% Max accuracy: 61.31% Test: [0/9] eta: 0:00:41 loss: 6.8806 (6.8806) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.6568 data: 4.4409 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9002 (6.9184) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6687 data: 0.4935 max mem: 64948 Test: Total time: 0:00:06 (0.6757 s / it) * Acc@1 0.098 Acc@5 0.502 loss 6.918 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [13] [ 0/312] eta: 0:55:38 lr: 0.002600 min_lr: 0.002600 loss: 3.6646 (3.6646) weight_decay: 0.0500 (0.0500) time: 10.6995 data: 6.8097 max mem: 64948 Epoch: [13] [ 10/312] eta: 0:08:12 lr: 0.002607 min_lr: 0.002607 loss: 3.2804 (3.2933) weight_decay: 0.0500 (0.0500) time: 1.6302 data: 0.6195 max mem: 64948 Epoch: [13] [ 20/312] eta: 0:05:46 lr: 0.002613 min_lr: 0.002613 loss: 3.2804 (3.3264) weight_decay: 0.0500 (0.0500) time: 0.7095 data: 0.0004 max mem: 64948 Epoch: [13] [ 30/312] eta: 0:04:49 lr: 0.002620 min_lr: 0.002620 loss: 3.2252 (3.2793) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [13] [ 40/312] eta: 0:04:17 lr: 0.002626 min_lr: 0.002626 loss: 3.2840 (3.3184) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [13] [ 50/312] eta: 0:03:55 lr: 0.002632 min_lr: 0.002632 loss: 3.3154 (3.2730) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [13] [ 60/312] eta: 0:03:37 lr: 0.002639 min_lr: 0.002639 loss: 3.2118 (3.2981) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0003 max mem: 64948 Epoch: [13] [ 70/312] eta: 0:03:23 lr: 0.002645 min_lr: 0.002645 loss: 3.4702 (3.3115) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [13] [ 80/312] eta: 0:03:10 lr: 0.002652 min_lr: 0.002652 loss: 3.3852 (3.3002) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [13] [ 90/312] eta: 0:02:59 lr: 0.002658 min_lr: 0.002658 loss: 3.3494 (3.3119) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [13] [100/312] eta: 0:02:49 lr: 0.002665 min_lr: 0.002665 loss: 3.5114 (3.3347) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [13] [110/312] eta: 0:02:39 lr: 0.002671 min_lr: 0.002671 loss: 3.5719 (3.3473) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [13] [120/312] eta: 0:02:29 lr: 0.002677 min_lr: 0.002677 loss: 3.6331 (3.3675) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [13] [130/312] eta: 0:02:20 lr: 0.002684 min_lr: 0.002684 loss: 3.7340 (3.3731) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [13] [140/312] eta: 0:02:12 lr: 0.002690 min_lr: 0.002690 loss: 3.4902 (3.3758) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [13] [150/312] eta: 0:02:03 lr: 0.002697 min_lr: 0.002697 loss: 3.5288 (3.3774) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [13] [160/312] eta: 0:01:55 lr: 0.002703 min_lr: 0.002703 loss: 3.5730 (3.3851) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [13] [170/312] eta: 0:01:47 lr: 0.002709 min_lr: 0.002709 loss: 3.6295 (3.3874) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [13] [180/312] eta: 0:01:39 lr: 0.002716 min_lr: 0.002716 loss: 3.4562 (3.3770) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [13] [190/312] eta: 0:01:31 lr: 0.002722 min_lr: 0.002722 loss: 3.4562 (3.3868) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [13] [200/312] eta: 0:01:23 lr: 0.002729 min_lr: 0.002729 loss: 3.8077 (3.4003) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [13] [210/312] eta: 0:01:15 lr: 0.002735 min_lr: 0.002735 loss: 3.5081 (3.3896) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [13] [220/312] eta: 0:01:08 lr: 0.002741 min_lr: 0.002741 loss: 3.2139 (3.3873) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [13] [230/312] eta: 0:01:00 lr: 0.002748 min_lr: 0.002748 loss: 3.3789 (3.3888) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [13] [240/312] eta: 0:00:53 lr: 0.002754 min_lr: 0.002754 loss: 3.3789 (3.3924) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [13] [250/312] eta: 0:00:45 lr: 0.002761 min_lr: 0.002761 loss: 3.3516 (3.3829) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [13] [260/312] eta: 0:00:38 lr: 0.002767 min_lr: 0.002767 loss: 3.5388 (3.3918) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [13] [270/312] eta: 0:00:30 lr: 0.002774 min_lr: 0.002774 loss: 3.6404 (3.3977) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [13] [280/312] eta: 0:00:23 lr: 0.002780 min_lr: 0.002780 loss: 3.5917 (3.3900) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0009 max mem: 64948 Epoch: [13] [290/312] eta: 0:00:16 lr: 0.002786 min_lr: 0.002786 loss: 3.2092 (3.3869) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [13] [300/312] eta: 0:00:08 lr: 0.002793 min_lr: 0.002793 loss: 3.5178 (3.3890) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [13] [310/312] eta: 0:00:01 lr: 0.002799 min_lr: 0.002799 loss: 3.5178 (3.3846) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [13] [311/312] eta: 0:00:00 lr: 0.002800 min_lr: 0.002800 loss: 3.5178 (3.3842) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [13] Total time: 0:03:47 (0.7304 s / it) Averaged stats: lr: 0.002800 min_lr: 0.002800 loss: 3.5178 (3.4320) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 1.1857 (1.1857) acc1: 72.1354 (72.1354) acc5: 89.8438 (89.8438) time: 4.6350 data: 4.4250 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.7021 (1.5530) acc1: 60.9375 (61.5040) acc5: 86.4583 (84.3840) time: 0.6663 data: 0.4918 max mem: 64948 Test: Total time: 0:00:06 (0.6911 s / it) * Acc@1 62.720 Acc@5 84.716 loss 1.544 Accuracy of the model on the 50000 test images: 62.7% Max accuracy: 62.72% Test: [0/9] eta: 0:00:40 loss: 6.8845 (6.8845) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.5393 data: 4.3260 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.8981 (6.9201) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6556 data: 0.4808 max mem: 64948 Test: Total time: 0:00:05 (0.6635 s / it) * Acc@1 0.098 Acc@5 0.498 loss 6.920 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [14] [ 0/312] eta: 0:56:04 lr: 0.002800 min_lr: 0.002800 loss: 3.5693 (3.5693) weight_decay: 0.0500 (0.0500) time: 10.7837 data: 7.0041 max mem: 64948 Epoch: [14] [ 10/312] eta: 0:08:13 lr: 0.002807 min_lr: 0.002807 loss: 3.5561 (3.4025) weight_decay: 0.0500 (0.0500) time: 1.6337 data: 0.6372 max mem: 64948 Epoch: [14] [ 20/312] eta: 0:05:46 lr: 0.002813 min_lr: 0.002813 loss: 3.4544 (3.3788) weight_decay: 0.0500 (0.0500) time: 0.7061 data: 0.0004 max mem: 64948 Epoch: [14] [ 30/312] eta: 0:04:49 lr: 0.002820 min_lr: 0.002820 loss: 3.6058 (3.4585) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [14] [ 40/312] eta: 0:04:17 lr: 0.002826 min_lr: 0.002826 loss: 3.6476 (3.4812) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [14] [ 50/312] eta: 0:03:55 lr: 0.002833 min_lr: 0.002833 loss: 3.3804 (3.4127) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [14] [ 60/312] eta: 0:03:37 lr: 0.002839 min_lr: 0.002839 loss: 3.2504 (3.3968) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [14] [ 70/312] eta: 0:03:23 lr: 0.002845 min_lr: 0.002845 loss: 3.7052 (3.4191) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [14] [ 80/312] eta: 0:03:10 lr: 0.002852 min_lr: 0.002852 loss: 3.6701 (3.4311) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [14] [ 90/312] eta: 0:02:59 lr: 0.002858 min_lr: 0.002858 loss: 3.4691 (3.4008) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [14] [100/312] eta: 0:02:49 lr: 0.002865 min_lr: 0.002865 loss: 3.4691 (3.4215) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [14] [110/312] eta: 0:02:39 lr: 0.002871 min_lr: 0.002871 loss: 3.6785 (3.4242) weight_decay: 0.0500 (0.0500) time: 0.7020 data: 0.0004 max mem: 64948 Epoch: [14] [120/312] eta: 0:02:29 lr: 0.002877 min_lr: 0.002877 loss: 3.5110 (3.4129) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [14] [130/312] eta: 0:02:20 lr: 0.002884 min_lr: 0.002884 loss: 3.5471 (3.4083) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [14] [140/312] eta: 0:02:12 lr: 0.002890 min_lr: 0.002890 loss: 3.5584 (3.4075) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [14] [150/312] eta: 0:02:03 lr: 0.002897 min_lr: 0.002897 loss: 3.5336 (3.4012) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [14] [160/312] eta: 0:01:55 lr: 0.002903 min_lr: 0.002903 loss: 3.5435 (3.4074) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [14] [170/312] eta: 0:01:47 lr: 0.002909 min_lr: 0.002909 loss: 3.3720 (3.3961) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [14] [180/312] eta: 0:01:39 lr: 0.002916 min_lr: 0.002916 loss: 3.3720 (3.3993) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [14] [190/312] eta: 0:01:31 lr: 0.002922 min_lr: 0.002922 loss: 3.4350 (3.4063) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [14] [200/312] eta: 0:01:23 lr: 0.002929 min_lr: 0.002929 loss: 3.6577 (3.4078) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [14] [210/312] eta: 0:01:15 lr: 0.002935 min_lr: 0.002935 loss: 3.5311 (3.4091) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [14] [220/312] eta: 0:01:08 lr: 0.002941 min_lr: 0.002941 loss: 3.5311 (3.4072) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [14] [230/312] eta: 0:01:00 lr: 0.002948 min_lr: 0.002948 loss: 3.4418 (3.4044) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [14] [240/312] eta: 0:00:53 lr: 0.002954 min_lr: 0.002954 loss: 3.2424 (3.3888) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [14] [250/312] eta: 0:00:45 lr: 0.002961 min_lr: 0.002961 loss: 3.2585 (3.3936) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [14] [260/312] eta: 0:00:38 lr: 0.002967 min_lr: 0.002967 loss: 3.4505 (3.3905) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [14] [270/312] eta: 0:00:30 lr: 0.002974 min_lr: 0.002974 loss: 3.4505 (3.3925) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [14] [280/312] eta: 0:00:23 lr: 0.002980 min_lr: 0.002980 loss: 3.3637 (3.3886) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [14] [290/312] eta: 0:00:16 lr: 0.002986 min_lr: 0.002986 loss: 3.5920 (3.3961) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [14] [300/312] eta: 0:00:08 lr: 0.002993 min_lr: 0.002993 loss: 3.6449 (3.3993) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [14] [310/312] eta: 0:00:01 lr: 0.002999 min_lr: 0.002999 loss: 3.4933 (3.3975) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [14] [311/312] eta: 0:00:00 lr: 0.003000 min_lr: 0.003000 loss: 3.4933 (3.3995) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [14] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.003000 min_lr: 0.003000 loss: 3.4933 (3.3694) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.3080 (1.3080) acc1: 70.5729 (70.5729) acc5: 89.0625 (89.0625) time: 4.7347 data: 4.5242 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.7332 (1.5314) acc1: 60.6771 (61.8880) acc5: 83.0729 (84.4160) time: 0.6774 data: 0.5028 max mem: 64948 Test: Total time: 0:00:06 (0.7020 s / it) * Acc@1 63.054 Acc@5 84.786 loss 1.524 Accuracy of the model on the 50000 test images: 63.1% Max accuracy: 63.05% Test: [0/9] eta: 0:00:41 loss: 6.8897 (6.8897) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.5612 data: 4.3546 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.8956 (6.9222) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6581 data: 0.4840 max mem: 64948 Test: Total time: 0:00:05 (0.6654 s / it) * Acc@1 0.098 Acc@5 0.498 loss 6.922 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [15] [ 0/312] eta: 0:52:10 lr: 0.003000 min_lr: 0.003000 loss: 3.3028 (3.3028) weight_decay: 0.0500 (0.0500) time: 10.0346 data: 8.1154 max mem: 64948 Epoch: [15] [ 10/312] eta: 0:08:08 lr: 0.003007 min_lr: 0.003007 loss: 3.5630 (3.4482) weight_decay: 0.0500 (0.0500) time: 1.6179 data: 0.7382 max mem: 64948 Epoch: [15] [ 20/312] eta: 0:05:43 lr: 0.003013 min_lr: 0.003013 loss: 3.6099 (3.4578) weight_decay: 0.0500 (0.0500) time: 0.7348 data: 0.0004 max mem: 64948 Epoch: [15] [ 30/312] eta: 0:04:48 lr: 0.003020 min_lr: 0.003020 loss: 3.2689 (3.3733) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [15] [ 40/312] eta: 0:04:16 lr: 0.003026 min_lr: 0.003026 loss: 3.3351 (3.3888) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [15] [ 50/312] eta: 0:03:54 lr: 0.003033 min_lr: 0.003033 loss: 3.5269 (3.3534) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [15] [ 60/312] eta: 0:03:37 lr: 0.003039 min_lr: 0.003039 loss: 3.5099 (3.3860) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [15] [ 70/312] eta: 0:03:22 lr: 0.003045 min_lr: 0.003045 loss: 3.5789 (3.3803) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [15] [ 80/312] eta: 0:03:10 lr: 0.003052 min_lr: 0.003052 loss: 3.4640 (3.3583) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [15] [ 90/312] eta: 0:02:59 lr: 0.003058 min_lr: 0.003058 loss: 3.4304 (3.3464) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [15] [100/312] eta: 0:02:48 lr: 0.003065 min_lr: 0.003065 loss: 3.2549 (3.3394) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [15] [110/312] eta: 0:02:39 lr: 0.003071 min_lr: 0.003071 loss: 3.4491 (3.3404) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [15] [120/312] eta: 0:02:29 lr: 0.003077 min_lr: 0.003077 loss: 3.4491 (3.3325) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [15] [130/312] eta: 0:02:20 lr: 0.003084 min_lr: 0.003084 loss: 3.3221 (3.3247) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [15] [140/312] eta: 0:02:12 lr: 0.003090 min_lr: 0.003090 loss: 3.3825 (3.3244) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [15] [150/312] eta: 0:02:03 lr: 0.003097 min_lr: 0.003097 loss: 3.2441 (3.3106) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [15] [160/312] eta: 0:01:55 lr: 0.003103 min_lr: 0.003103 loss: 3.2441 (3.3203) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [15] [170/312] eta: 0:01:47 lr: 0.003109 min_lr: 0.003109 loss: 3.5351 (3.3189) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [15] [180/312] eta: 0:01:39 lr: 0.003116 min_lr: 0.003116 loss: 3.0425 (3.3015) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [15] [190/312] eta: 0:01:31 lr: 0.003122 min_lr: 0.003122 loss: 2.9775 (3.2862) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [15] [200/312] eta: 0:01:23 lr: 0.003129 min_lr: 0.003129 loss: 3.0963 (3.2845) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0003 max mem: 64948 Epoch: [15] [210/312] eta: 0:01:15 lr: 0.003135 min_lr: 0.003135 loss: 3.2930 (3.2842) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [15] [220/312] eta: 0:01:08 lr: 0.003142 min_lr: 0.003142 loss: 3.0713 (3.2756) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [15] [230/312] eta: 0:01:00 lr: 0.003148 min_lr: 0.003148 loss: 3.1284 (3.2730) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [15] [240/312] eta: 0:00:53 lr: 0.003154 min_lr: 0.003154 loss: 3.3132 (3.2791) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [15] [250/312] eta: 0:00:45 lr: 0.003161 min_lr: 0.003161 loss: 3.3132 (3.2792) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [15] [260/312] eta: 0:00:38 lr: 0.003167 min_lr: 0.003167 loss: 3.4355 (3.2889) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [15] [270/312] eta: 0:00:30 lr: 0.003174 min_lr: 0.003174 loss: 3.4355 (3.2871) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [15] [280/312] eta: 0:00:23 lr: 0.003180 min_lr: 0.003180 loss: 3.3398 (3.2862) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0009 max mem: 64948 Epoch: [15] [290/312] eta: 0:00:16 lr: 0.003186 min_lr: 0.003186 loss: 3.3120 (3.2804) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0008 max mem: 64948 Epoch: [15] [300/312] eta: 0:00:08 lr: 0.003193 min_lr: 0.003193 loss: 3.4501 (3.2906) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [15] [310/312] eta: 0:00:01 lr: 0.003199 min_lr: 0.003199 loss: 3.5277 (3.2914) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [15] [311/312] eta: 0:00:00 lr: 0.003200 min_lr: 0.003200 loss: 3.4501 (3.2916) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [15] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.003200 min_lr: 0.003200 loss: 3.4501 (3.2895) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 1.1387 (1.1387) acc1: 71.8750 (71.8750) acc5: 90.3646 (90.3646) time: 4.6431 data: 4.4247 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.5766 (1.4745) acc1: 64.0625 (63.8720) acc5: 87.2396 (86.2400) time: 0.6672 data: 0.4917 max mem: 64948 Test: Total time: 0:00:06 (0.6881 s / it) * Acc@1 64.130 Acc@5 85.794 loss 1.477 Accuracy of the model on the 50000 test images: 64.1% Max accuracy: 64.13% Test: [0/9] eta: 0:00:42 loss: 6.8974 (6.8974) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.7222 data: 4.5045 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.8974 (6.9249) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6760 data: 0.5006 max mem: 64948 Test: Total time: 0:00:06 (0.6847 s / it) * Acc@1 0.098 Acc@5 0.500 loss 6.925 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [16] [ 0/312] eta: 0:54:05 lr: 0.003201 min_lr: 0.003201 loss: 3.1925 (3.1925) weight_decay: 0.0500 (0.0500) time: 10.4007 data: 9.5587 max mem: 64948 Epoch: [16] [ 10/312] eta: 0:08:10 lr: 0.003207 min_lr: 0.003207 loss: 3.4315 (3.3697) weight_decay: 0.0500 (0.0500) time: 1.6256 data: 0.8693 max mem: 64948 Epoch: [16] [ 20/312] eta: 0:05:44 lr: 0.003213 min_lr: 0.003213 loss: 3.4315 (3.3180) weight_decay: 0.0500 (0.0500) time: 0.7204 data: 0.0004 max mem: 64948 Epoch: [16] [ 30/312] eta: 0:04:49 lr: 0.003220 min_lr: 0.003220 loss: 3.3800 (3.2861) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [16] [ 40/312] eta: 0:04:16 lr: 0.003226 min_lr: 0.003226 loss: 3.1884 (3.2511) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [16] [ 50/312] eta: 0:03:54 lr: 0.003233 min_lr: 0.003233 loss: 3.3087 (3.2437) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [16] [ 60/312] eta: 0:03:37 lr: 0.003239 min_lr: 0.003239 loss: 3.2177 (3.2334) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [16] [ 70/312] eta: 0:03:23 lr: 0.003245 min_lr: 0.003245 loss: 3.2402 (3.2619) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [16] [ 80/312] eta: 0:03:10 lr: 0.003252 min_lr: 0.003252 loss: 3.4644 (3.2413) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [16] [ 90/312] eta: 0:02:59 lr: 0.003258 min_lr: 0.003258 loss: 3.4078 (3.2566) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [16] [100/312] eta: 0:02:48 lr: 0.003265 min_lr: 0.003265 loss: 3.3760 (3.2352) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [16] [110/312] eta: 0:02:39 lr: 0.003271 min_lr: 0.003271 loss: 3.1857 (3.2348) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [16] [120/312] eta: 0:02:29 lr: 0.003277 min_lr: 0.003277 loss: 3.3034 (3.2353) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [16] [130/312] eta: 0:02:20 lr: 0.003284 min_lr: 0.003284 loss: 3.1622 (3.2292) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [16] [140/312] eta: 0:02:12 lr: 0.003290 min_lr: 0.003290 loss: 3.1873 (3.2434) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [16] [150/312] eta: 0:02:03 lr: 0.003297 min_lr: 0.003297 loss: 3.4552 (3.2559) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [16] [160/312] eta: 0:01:55 lr: 0.003303 min_lr: 0.003303 loss: 3.3172 (3.2476) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [16] [170/312] eta: 0:01:47 lr: 0.003310 min_lr: 0.003310 loss: 3.2347 (3.2565) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [16] [180/312] eta: 0:01:39 lr: 0.003316 min_lr: 0.003316 loss: 3.2347 (3.2502) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [16] [190/312] eta: 0:01:31 lr: 0.003322 min_lr: 0.003322 loss: 3.3825 (3.2584) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [16] [200/312] eta: 0:01:23 lr: 0.003329 min_lr: 0.003329 loss: 3.2890 (3.2466) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [16] [210/312] eta: 0:01:15 lr: 0.003335 min_lr: 0.003335 loss: 3.2645 (3.2538) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [16] [220/312] eta: 0:01:08 lr: 0.003342 min_lr: 0.003342 loss: 3.4024 (3.2553) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [16] [230/312] eta: 0:01:00 lr: 0.003348 min_lr: 0.003348 loss: 3.3929 (3.2624) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [16] [240/312] eta: 0:00:53 lr: 0.003354 min_lr: 0.003354 loss: 3.3751 (3.2567) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [16] [250/312] eta: 0:00:45 lr: 0.003361 min_lr: 0.003361 loss: 3.3751 (3.2636) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [16] [260/312] eta: 0:00:38 lr: 0.003367 min_lr: 0.003367 loss: 3.4704 (3.2629) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [16] [270/312] eta: 0:00:30 lr: 0.003374 min_lr: 0.003374 loss: 3.4214 (3.2638) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [16] [280/312] eta: 0:00:23 lr: 0.003380 min_lr: 0.003380 loss: 3.4316 (3.2646) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0010 max mem: 64948 Epoch: [16] [290/312] eta: 0:00:16 lr: 0.003386 min_lr: 0.003386 loss: 3.4392 (3.2621) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [16] [300/312] eta: 0:00:08 lr: 0.003393 min_lr: 0.003393 loss: 3.3658 (3.2650) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [16] [310/312] eta: 0:00:01 lr: 0.003399 min_lr: 0.003399 loss: 3.4439 (3.2711) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [16] [311/312] eta: 0:00:00 lr: 0.003400 min_lr: 0.003400 loss: 3.4439 (3.2715) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [16] Total time: 0:03:47 (0.7303 s / it) Averaged stats: lr: 0.003400 min_lr: 0.003400 loss: 3.4439 (3.2410) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.2400 (1.2400) acc1: 73.1771 (73.1771) acc5: 88.2812 (88.2812) time: 4.6822 data: 4.4606 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.6418 (1.5506) acc1: 63.0208 (63.3600) acc5: 85.1562 (84.8960) time: 0.6722 data: 0.4957 max mem: 64948 Test: Total time: 0:00:06 (0.6964 s / it) * Acc@1 63.562 Acc@5 85.048 loss 1.538 Accuracy of the model on the 50000 test images: 63.6% Max accuracy: 64.13% Test: [0/9] eta: 0:00:45 loss: 6.9076 (6.9076) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 5.0992 data: 4.8952 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9076 (6.9280) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.7178 data: 0.5440 max mem: 64948 Test: Total time: 0:00:06 (0.7271 s / it) * Acc@1 0.098 Acc@5 0.498 loss 6.928 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [17] [ 0/312] eta: 0:50:34 lr: 0.003401 min_lr: 0.003401 loss: 2.8088 (2.8088) weight_decay: 0.0500 (0.0500) time: 9.7263 data: 8.3021 max mem: 64948 Epoch: [17] [ 10/312] eta: 0:07:49 lr: 0.003407 min_lr: 0.003407 loss: 2.9167 (3.0329) weight_decay: 0.0500 (0.0500) time: 1.5558 data: 0.7552 max mem: 64948 Epoch: [17] [ 20/312] eta: 0:05:34 lr: 0.003413 min_lr: 0.003413 loss: 3.0028 (3.0873) weight_decay: 0.0500 (0.0500) time: 0.7154 data: 0.0004 max mem: 64948 Epoch: [17] [ 30/312] eta: 0:04:41 lr: 0.003420 min_lr: 0.003420 loss: 3.0028 (3.0760) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0003 max mem: 64948 Epoch: [17] [ 40/312] eta: 0:04:11 lr: 0.003426 min_lr: 0.003426 loss: 3.3699 (3.1320) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [17] [ 50/312] eta: 0:03:50 lr: 0.003433 min_lr: 0.003433 loss: 3.4577 (3.1747) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [17] [ 60/312] eta: 0:03:34 lr: 0.003439 min_lr: 0.003439 loss: 3.3661 (3.1550) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [17] [ 70/312] eta: 0:03:20 lr: 0.003445 min_lr: 0.003445 loss: 3.0051 (3.1472) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [17] [ 80/312] eta: 0:03:08 lr: 0.003452 min_lr: 0.003452 loss: 3.0959 (3.1408) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [17] [ 90/312] eta: 0:02:57 lr: 0.003458 min_lr: 0.003458 loss: 3.2465 (3.1491) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [17] [100/312] eta: 0:02:47 lr: 0.003465 min_lr: 0.003465 loss: 3.3564 (3.1722) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [17] [110/312] eta: 0:02:37 lr: 0.003471 min_lr: 0.003471 loss: 3.3564 (3.1774) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [17] [120/312] eta: 0:02:28 lr: 0.003477 min_lr: 0.003477 loss: 3.0954 (3.1610) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [17] [130/312] eta: 0:02:19 lr: 0.003484 min_lr: 0.003484 loss: 3.2327 (3.1693) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [17] [140/312] eta: 0:02:11 lr: 0.003490 min_lr: 0.003490 loss: 3.3804 (3.1731) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [17] [150/312] eta: 0:02:02 lr: 0.003497 min_lr: 0.003497 loss: 3.3083 (3.1773) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [17] [160/312] eta: 0:01:54 lr: 0.003503 min_lr: 0.003503 loss: 3.5258 (3.1846) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [17] [170/312] eta: 0:01:46 lr: 0.003510 min_lr: 0.003510 loss: 3.4687 (3.1919) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [17] [180/312] eta: 0:01:38 lr: 0.003516 min_lr: 0.003516 loss: 3.0629 (3.1765) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [17] [190/312] eta: 0:01:30 lr: 0.003522 min_lr: 0.003522 loss: 3.1970 (3.1846) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [17] [200/312] eta: 0:01:23 lr: 0.003529 min_lr: 0.003529 loss: 3.3436 (3.1866) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [17] [210/312] eta: 0:01:15 lr: 0.003535 min_lr: 0.003535 loss: 3.0326 (3.1830) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [17] [220/312] eta: 0:01:07 lr: 0.003542 min_lr: 0.003542 loss: 3.3301 (3.1895) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [17] [230/312] eta: 0:01:00 lr: 0.003548 min_lr: 0.003548 loss: 3.3301 (3.1805) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [17] [240/312] eta: 0:00:52 lr: 0.003554 min_lr: 0.003554 loss: 3.1536 (3.1762) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [17] [250/312] eta: 0:00:45 lr: 0.003561 min_lr: 0.003561 loss: 3.1891 (3.1720) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [17] [260/312] eta: 0:00:38 lr: 0.003567 min_lr: 0.003567 loss: 3.1912 (3.1722) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [17] [270/312] eta: 0:00:30 lr: 0.003574 min_lr: 0.003574 loss: 2.9585 (3.1649) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [17] [280/312] eta: 0:00:23 lr: 0.003580 min_lr: 0.003580 loss: 3.1368 (3.1658) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [17] [290/312] eta: 0:00:15 lr: 0.003586 min_lr: 0.003586 loss: 3.3895 (3.1698) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [17] [300/312] eta: 0:00:08 lr: 0.003593 min_lr: 0.003593 loss: 3.3110 (3.1646) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [17] [310/312] eta: 0:00:01 lr: 0.003599 min_lr: 0.003599 loss: 3.3214 (3.1662) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [17] [311/312] eta: 0:00:00 lr: 0.003600 min_lr: 0.003600 loss: 3.3329 (3.1669) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [17] Total time: 0:03:46 (0.7275 s / it) Averaged stats: lr: 0.003600 min_lr: 0.003600 loss: 3.3329 (3.1834) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.1334 (1.1334) acc1: 76.0417 (76.0417) acc5: 89.8438 (89.8438) time: 4.6981 data: 4.4859 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.6146 (1.4612) acc1: 63.5417 (65.1200) acc5: 85.9375 (86.4000) time: 0.6733 data: 0.4985 max mem: 64948 Test: Total time: 0:00:06 (0.6952 s / it) * Acc@1 65.188 Acc@5 86.030 loss 1.455 Accuracy of the model on the 50000 test images: 65.2% Max accuracy: 65.19% Test: [0/9] eta: 0:00:41 loss: 6.9183 (6.9183) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.6337 data: 4.4159 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9183 (6.9315) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.6666 data: 0.4908 max mem: 64948 Test: Total time: 0:00:06 (0.6748 s / it) * Acc@1 0.100 Acc@5 0.502 loss 6.932 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [18] [ 0/312] eta: 0:54:34 lr: 0.003601 min_lr: 0.003601 loss: 2.5746 (2.5746) weight_decay: 0.0500 (0.0500) time: 10.4946 data: 7.6018 max mem: 64948 Epoch: [18] [ 10/312] eta: 0:08:07 lr: 0.003607 min_lr: 0.003607 loss: 3.4369 (3.2213) weight_decay: 0.0500 (0.0500) time: 1.6129 data: 0.6915 max mem: 64948 Epoch: [18] [ 20/312] eta: 0:05:43 lr: 0.003613 min_lr: 0.003613 loss: 3.4369 (3.3169) weight_decay: 0.0500 (0.0500) time: 0.7097 data: 0.0004 max mem: 64948 Epoch: [18] [ 30/312] eta: 0:04:48 lr: 0.003620 min_lr: 0.003620 loss: 3.3836 (3.2228) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [18] [ 40/312] eta: 0:04:16 lr: 0.003626 min_lr: 0.003626 loss: 3.0779 (3.1841) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [18] [ 50/312] eta: 0:03:54 lr: 0.003633 min_lr: 0.003633 loss: 2.9476 (3.1663) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [18] [ 60/312] eta: 0:03:37 lr: 0.003639 min_lr: 0.003639 loss: 3.0712 (3.1612) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [18] [ 70/312] eta: 0:03:22 lr: 0.003645 min_lr: 0.003645 loss: 3.2518 (3.1656) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [18] [ 80/312] eta: 0:03:10 lr: 0.003652 min_lr: 0.003652 loss: 3.2518 (3.1750) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [18] [ 90/312] eta: 0:02:59 lr: 0.003658 min_lr: 0.003658 loss: 3.3652 (3.1796) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [18] [100/312] eta: 0:02:48 lr: 0.003665 min_lr: 0.003665 loss: 3.1774 (3.1527) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [18] [110/312] eta: 0:02:38 lr: 0.003671 min_lr: 0.003671 loss: 2.9556 (3.1510) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [18] [120/312] eta: 0:02:29 lr: 0.003678 min_lr: 0.003678 loss: 2.9631 (3.1393) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [18] [130/312] eta: 0:02:20 lr: 0.003684 min_lr: 0.003684 loss: 3.1076 (3.1383) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [18] [140/312] eta: 0:02:11 lr: 0.003690 min_lr: 0.003690 loss: 3.1416 (3.1434) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [18] [150/312] eta: 0:02:03 lr: 0.003697 min_lr: 0.003697 loss: 3.2281 (3.1477) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [18] [160/312] eta: 0:01:55 lr: 0.003703 min_lr: 0.003703 loss: 3.2591 (3.1446) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [18] [170/312] eta: 0:01:47 lr: 0.003710 min_lr: 0.003710 loss: 3.2196 (3.1369) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [18] [180/312] eta: 0:01:39 lr: 0.003716 min_lr: 0.003716 loss: 3.0069 (3.1309) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0003 max mem: 64948 Epoch: [18] [190/312] eta: 0:01:31 lr: 0.003722 min_lr: 0.003722 loss: 3.0220 (3.1349) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [18] [200/312] eta: 0:01:23 lr: 0.003729 min_lr: 0.003729 loss: 3.1235 (3.1231) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [18] [210/312] eta: 0:01:15 lr: 0.003735 min_lr: 0.003735 loss: 3.4000 (3.1416) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [18] [220/312] eta: 0:01:08 lr: 0.003742 min_lr: 0.003742 loss: 3.5065 (3.1503) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [18] [230/312] eta: 0:01:00 lr: 0.003748 min_lr: 0.003748 loss: 3.2943 (3.1471) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [18] [240/312] eta: 0:00:53 lr: 0.003754 min_lr: 0.003754 loss: 3.0179 (3.1416) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [18] [250/312] eta: 0:00:45 lr: 0.003761 min_lr: 0.003761 loss: 3.2816 (3.1530) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [18] [260/312] eta: 0:00:38 lr: 0.003767 min_lr: 0.003767 loss: 3.4686 (3.1664) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [18] [270/312] eta: 0:00:30 lr: 0.003774 min_lr: 0.003774 loss: 3.3926 (3.1696) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [18] [280/312] eta: 0:00:23 lr: 0.003780 min_lr: 0.003780 loss: 3.0078 (3.1568) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [18] [290/312] eta: 0:00:16 lr: 0.003787 min_lr: 0.003787 loss: 2.8264 (3.1543) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [18] [300/312] eta: 0:00:08 lr: 0.003793 min_lr: 0.003793 loss: 3.1697 (3.1523) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [18] [310/312] eta: 0:00:01 lr: 0.003799 min_lr: 0.003799 loss: 3.2731 (3.1531) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [18] [311/312] eta: 0:00:00 lr: 0.003800 min_lr: 0.003800 loss: 3.2792 (3.1540) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [18] Total time: 0:03:47 (0.7296 s / it) Averaged stats: lr: 0.003800 min_lr: 0.003800 loss: 3.2792 (3.1533) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 1.1477 (1.1477) acc1: 71.3542 (71.3542) acc5: 89.5833 (89.5833) time: 4.8289 data: 4.6058 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.6131 (1.4692) acc1: 63.8021 (63.8720) acc5: 84.3750 (85.4720) time: 0.6878 data: 0.5118 max mem: 64948 Test: Total time: 0:00:06 (0.7162 s / it) * Acc@1 64.564 Acc@5 85.920 loss 1.464 Accuracy of the model on the 50000 test images: 64.6% Max accuracy: 65.19% Test: [0/9] eta: 0:00:46 loss: 6.9323 (6.9323) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.1160 data: 4.8981 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9323 (6.9359) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.7197 data: 0.5443 max mem: 64948 Test: Total time: 0:00:06 (0.7293 s / it) * Acc@1 0.100 Acc@5 0.498 loss 6.936 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [19] [ 0/312] eta: 0:53:31 lr: 0.003801 min_lr: 0.003801 loss: 3.4560 (3.4560) weight_decay: 0.0500 (0.0500) time: 10.2925 data: 8.7166 max mem: 64948 Epoch: [19] [ 10/312] eta: 0:08:02 lr: 0.003807 min_lr: 0.003807 loss: 3.3813 (3.2538) weight_decay: 0.0500 (0.0500) time: 1.5972 data: 0.7929 max mem: 64948 Epoch: [19] [ 20/312] eta: 0:05:41 lr: 0.003813 min_lr: 0.003813 loss: 3.1979 (3.1798) weight_decay: 0.0500 (0.0500) time: 0.7138 data: 0.0004 max mem: 64948 Epoch: [19] [ 30/312] eta: 0:04:46 lr: 0.003820 min_lr: 0.003820 loss: 3.0394 (3.1026) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0003 max mem: 64948 Epoch: [19] [ 40/312] eta: 0:04:15 lr: 0.003826 min_lr: 0.003826 loss: 3.1200 (3.1494) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0003 max mem: 64948 Epoch: [19] [ 50/312] eta: 0:03:53 lr: 0.003833 min_lr: 0.003833 loss: 3.3926 (3.1397) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [19] [ 60/312] eta: 0:03:36 lr: 0.003839 min_lr: 0.003839 loss: 3.2410 (3.1459) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [19] [ 70/312] eta: 0:03:22 lr: 0.003845 min_lr: 0.003845 loss: 3.0786 (3.1347) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [19] [ 80/312] eta: 0:03:10 lr: 0.003852 min_lr: 0.003852 loss: 3.2565 (3.1466) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [19] [ 90/312] eta: 0:02:58 lr: 0.003858 min_lr: 0.003858 loss: 3.2834 (3.1440) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [19] [100/312] eta: 0:02:48 lr: 0.003865 min_lr: 0.003865 loss: 3.0914 (3.1255) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [19] [110/312] eta: 0:02:38 lr: 0.003871 min_lr: 0.003871 loss: 2.8555 (3.1096) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [19] [120/312] eta: 0:02:29 lr: 0.003878 min_lr: 0.003878 loss: 2.8657 (3.1067) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [19] [130/312] eta: 0:02:20 lr: 0.003884 min_lr: 0.003884 loss: 3.0599 (3.1052) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [19] [140/312] eta: 0:02:11 lr: 0.003890 min_lr: 0.003890 loss: 3.3294 (3.1262) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [19] [150/312] eta: 0:02:03 lr: 0.003897 min_lr: 0.003897 loss: 3.3762 (3.1368) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [19] [160/312] eta: 0:01:55 lr: 0.003903 min_lr: 0.003903 loss: 3.2695 (3.1345) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [19] [170/312] eta: 0:01:47 lr: 0.003910 min_lr: 0.003910 loss: 3.1686 (3.1309) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [19] [180/312] eta: 0:01:39 lr: 0.003916 min_lr: 0.003916 loss: 2.8218 (3.1226) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [19] [190/312] eta: 0:01:31 lr: 0.003922 min_lr: 0.003922 loss: 2.9700 (3.1219) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0003 max mem: 64948 Epoch: [19] [200/312] eta: 0:01:23 lr: 0.003929 min_lr: 0.003929 loss: 3.0162 (3.1151) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [19] [210/312] eta: 0:01:15 lr: 0.003935 min_lr: 0.003935 loss: 2.8183 (3.1023) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [19] [220/312] eta: 0:01:08 lr: 0.003942 min_lr: 0.003942 loss: 2.8220 (3.0967) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [19] [230/312] eta: 0:01:00 lr: 0.003948 min_lr: 0.003948 loss: 2.9994 (3.0955) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [19] [240/312] eta: 0:00:53 lr: 0.003954 min_lr: 0.003954 loss: 3.1035 (3.0989) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [19] [250/312] eta: 0:00:45 lr: 0.003961 min_lr: 0.003961 loss: 3.3066 (3.1079) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [19] [260/312] eta: 0:00:38 lr: 0.003967 min_lr: 0.003967 loss: 3.3425 (3.1098) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [19] [270/312] eta: 0:00:30 lr: 0.003974 min_lr: 0.003974 loss: 3.3163 (3.1137) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [19] [280/312] eta: 0:00:23 lr: 0.003980 min_lr: 0.003980 loss: 3.1018 (3.1048) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [19] [290/312] eta: 0:00:16 lr: 0.003987 min_lr: 0.003987 loss: 2.7086 (3.0987) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [19] [300/312] eta: 0:00:08 lr: 0.003993 min_lr: 0.003993 loss: 3.0181 (3.0967) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [19] [310/312] eta: 0:00:01 lr: 0.003999 min_lr: 0.003999 loss: 3.2158 (3.1028) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [19] [311/312] eta: 0:00:00 lr: 0.004000 min_lr: 0.004000 loss: 3.2158 (3.1035) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [19] Total time: 0:03:47 (0.7298 s / it) Averaged stats: lr: 0.004000 min_lr: 0.004000 loss: 3.2158 (3.0984) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 1.1770 (1.1770) acc1: 73.9583 (73.9583) acc5: 89.5833 (89.5833) time: 4.6268 data: 4.4070 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.5870 (1.4891) acc1: 62.5000 (64.2240) acc5: 84.8958 (84.8320) time: 0.6660 data: 0.4898 max mem: 64948 Test: Total time: 0:00:06 (0.6892 s / it) * Acc@1 64.312 Acc@5 85.934 loss 1.474 Accuracy of the model on the 50000 test images: 64.3% Max accuracy: 65.19% Test: [0/9] eta: 0:00:44 loss: 6.9483 (6.9483) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.9738 data: 4.7561 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9483 (6.9411) acc1: 0.0000 (0.0960) acc5: 0.2604 (0.5120) time: 0.7040 data: 0.5286 max mem: 64948 Test: Total time: 0:00:06 (0.7129 s / it) * Acc@1 0.096 Acc@5 0.504 loss 6.941 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [20] [ 0/312] eta: 0:52:38 lr: 0.004000 min_lr: 0.004000 loss: 3.0685 (3.0685) weight_decay: 0.0500 (0.0500) time: 10.1230 data: 8.7917 max mem: 64948 Epoch: [20] [ 10/312] eta: 0:07:59 lr: 0.004000 min_lr: 0.004000 loss: 3.4298 (3.3054) weight_decay: 0.0500 (0.0500) time: 1.5871 data: 0.7998 max mem: 64948 Epoch: [20] [ 20/312] eta: 0:05:39 lr: 0.004000 min_lr: 0.004000 loss: 3.1678 (3.2037) weight_decay: 0.0500 (0.0500) time: 0.7151 data: 0.0005 max mem: 64948 Epoch: [20] [ 30/312] eta: 0:04:46 lr: 0.004000 min_lr: 0.004000 loss: 3.2218 (3.2521) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0003 max mem: 64948 Epoch: [20] [ 40/312] eta: 0:04:14 lr: 0.004000 min_lr: 0.004000 loss: 3.2788 (3.1969) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [20] [ 50/312] eta: 0:03:52 lr: 0.004000 min_lr: 0.004000 loss: 3.1936 (3.1880) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [20] [ 60/312] eta: 0:03:36 lr: 0.004000 min_lr: 0.004000 loss: 3.2246 (3.2116) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [20] [ 70/312] eta: 0:03:21 lr: 0.004000 min_lr: 0.004000 loss: 3.4636 (3.2399) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [20] [ 80/312] eta: 0:03:09 lr: 0.004000 min_lr: 0.004000 loss: 3.4041 (3.2091) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [20] [ 90/312] eta: 0:02:58 lr: 0.004000 min_lr: 0.004000 loss: 3.1784 (3.1970) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [20] [100/312] eta: 0:02:48 lr: 0.004000 min_lr: 0.004000 loss: 3.0396 (3.1834) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [20] [110/312] eta: 0:02:38 lr: 0.004000 min_lr: 0.004000 loss: 3.1368 (3.1705) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [20] [120/312] eta: 0:02:29 lr: 0.004000 min_lr: 0.004000 loss: 3.2553 (3.1603) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [20] [130/312] eta: 0:02:20 lr: 0.004000 min_lr: 0.004000 loss: 3.2043 (3.1587) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [20] [140/312] eta: 0:02:11 lr: 0.004000 min_lr: 0.004000 loss: 3.2043 (3.1655) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [20] [150/312] eta: 0:02:03 lr: 0.004000 min_lr: 0.004000 loss: 3.0828 (3.1568) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [20] [160/312] eta: 0:01:54 lr: 0.004000 min_lr: 0.004000 loss: 2.9912 (3.1482) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [20] [170/312] eta: 0:01:46 lr: 0.004000 min_lr: 0.004000 loss: 2.9912 (3.1428) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [20] [180/312] eta: 0:01:38 lr: 0.004000 min_lr: 0.004000 loss: 3.2312 (3.1535) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [20] [190/312] eta: 0:01:31 lr: 0.004000 min_lr: 0.004000 loss: 3.2040 (3.1424) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [20] [200/312] eta: 0:01:23 lr: 0.004000 min_lr: 0.004000 loss: 3.0277 (3.1342) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [20] [210/312] eta: 0:01:15 lr: 0.004000 min_lr: 0.004000 loss: 2.9955 (3.1209) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0005 max mem: 64948 Epoch: [20] [220/312] eta: 0:01:08 lr: 0.004000 min_lr: 0.004000 loss: 2.6977 (3.1092) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0005 max mem: 64948 Epoch: [20] [230/312] eta: 0:01:00 lr: 0.004000 min_lr: 0.004000 loss: 2.8404 (3.1001) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [20] [240/312] eta: 0:00:52 lr: 0.004000 min_lr: 0.004000 loss: 2.8469 (3.0880) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [20] [250/312] eta: 0:00:45 lr: 0.004000 min_lr: 0.004000 loss: 2.8469 (3.0850) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [20] [260/312] eta: 0:00:38 lr: 0.004000 min_lr: 0.004000 loss: 3.0782 (3.0764) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [20] [270/312] eta: 0:00:30 lr: 0.004000 min_lr: 0.004000 loss: 3.2194 (3.0814) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [20] [280/312] eta: 0:00:23 lr: 0.004000 min_lr: 0.004000 loss: 3.1990 (3.0793) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [20] [290/312] eta: 0:00:16 lr: 0.004000 min_lr: 0.004000 loss: 3.0862 (3.0810) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0008 max mem: 64948 Epoch: [20] [300/312] eta: 0:00:08 lr: 0.004000 min_lr: 0.004000 loss: 3.2148 (3.0868) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [20] [310/312] eta: 0:00:01 lr: 0.004000 min_lr: 0.004000 loss: 3.2148 (3.0899) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [20] [311/312] eta: 0:00:00 lr: 0.004000 min_lr: 0.004000 loss: 3.2349 (3.0910) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [20] Total time: 0:03:47 (0.7293 s / it) Averaged stats: lr: 0.004000 min_lr: 0.004000 loss: 3.2349 (3.0573) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.3162 (1.3162) acc1: 72.6562 (72.6562) acc5: 90.3646 (90.3646) time: 4.7463 data: 4.5321 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.6338 (1.5720) acc1: 64.5833 (63.3920) acc5: 83.8542 (84.8640) time: 0.6787 data: 0.5037 max mem: 64948 Test: Total time: 0:00:06 (0.7025 s / it) * Acc@1 64.144 Acc@5 85.188 loss 1.564 Accuracy of the model on the 50000 test images: 64.1% Max accuracy: 65.19% Test: [0/9] eta: 0:00:41 loss: 6.9653 (6.9653) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.6650 data: 4.4472 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9652 (6.9470) acc1: 0.0000 (0.0960) acc5: 0.2604 (0.5120) time: 0.6696 data: 0.4942 max mem: 64948 Test: Total time: 0:00:06 (0.6776 s / it) * Acc@1 0.096 Acc@5 0.498 loss 6.947 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [21] [ 0/312] eta: 0:54:02 lr: 0.004000 min_lr: 0.004000 loss: 2.2990 (2.2990) weight_decay: 0.0500 (0.0500) time: 10.3920 data: 6.4853 max mem: 64948 Epoch: [21] [ 10/312] eta: 0:08:10 lr: 0.004000 min_lr: 0.004000 loss: 2.8184 (2.9411) weight_decay: 0.0500 (0.0500) time: 1.6257 data: 0.5900 max mem: 64948 Epoch: [21] [ 20/312] eta: 0:05:45 lr: 0.004000 min_lr: 0.004000 loss: 3.1521 (3.0810) weight_decay: 0.0500 (0.0500) time: 0.7214 data: 0.0004 max mem: 64948 Epoch: [21] [ 30/312] eta: 0:04:48 lr: 0.004000 min_lr: 0.004000 loss: 3.2084 (3.0032) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [21] [ 40/312] eta: 0:04:17 lr: 0.004000 min_lr: 0.004000 loss: 2.9850 (3.0246) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [21] [ 50/312] eta: 0:03:54 lr: 0.004000 min_lr: 0.004000 loss: 3.0857 (3.0192) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [21] [ 60/312] eta: 0:03:37 lr: 0.004000 min_lr: 0.004000 loss: 2.9603 (2.9871) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [21] [ 70/312] eta: 0:03:23 lr: 0.004000 min_lr: 0.004000 loss: 3.0530 (3.0208) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [21] [ 80/312] eta: 0:03:10 lr: 0.004000 min_lr: 0.004000 loss: 3.1280 (2.9987) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [21] [ 90/312] eta: 0:02:59 lr: 0.004000 min_lr: 0.004000 loss: 3.1586 (3.0021) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [21] [100/312] eta: 0:02:49 lr: 0.004000 min_lr: 0.004000 loss: 3.1322 (2.9959) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [21] [110/312] eta: 0:02:39 lr: 0.004000 min_lr: 0.004000 loss: 2.9447 (2.9857) weight_decay: 0.0500 (0.0500) time: 0.7038 data: 0.0004 max mem: 64948 Epoch: [21] [120/312] eta: 0:02:29 lr: 0.004000 min_lr: 0.004000 loss: 2.9099 (2.9687) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [21] [130/312] eta: 0:02:20 lr: 0.004000 min_lr: 0.004000 loss: 2.8015 (2.9570) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [21] [140/312] eta: 0:02:12 lr: 0.004000 min_lr: 0.004000 loss: 2.9773 (2.9461) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [21] [150/312] eta: 0:02:03 lr: 0.004000 min_lr: 0.004000 loss: 3.1138 (2.9618) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [21] [160/312] eta: 0:01:55 lr: 0.004000 min_lr: 0.004000 loss: 3.1738 (2.9683) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [21] [170/312] eta: 0:01:47 lr: 0.004000 min_lr: 0.004000 loss: 3.1165 (2.9850) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [21] [180/312] eta: 0:01:39 lr: 0.004000 min_lr: 0.004000 loss: 3.0985 (2.9888) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [21] [190/312] eta: 0:01:31 lr: 0.004000 min_lr: 0.004000 loss: 2.8764 (2.9796) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [21] [200/312] eta: 0:01:23 lr: 0.004000 min_lr: 0.004000 loss: 2.6041 (2.9671) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [21] [210/312] eta: 0:01:15 lr: 0.004000 min_lr: 0.004000 loss: 2.7279 (2.9668) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [21] [220/312] eta: 0:01:08 lr: 0.004000 min_lr: 0.004000 loss: 2.7279 (2.9604) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [21] [230/312] eta: 0:01:00 lr: 0.004000 min_lr: 0.004000 loss: 3.1122 (2.9670) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [21] [240/312] eta: 0:00:53 lr: 0.004000 min_lr: 0.004000 loss: 3.2994 (2.9806) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [21] [250/312] eta: 0:00:45 lr: 0.004000 min_lr: 0.004000 loss: 3.2102 (2.9812) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [21] [260/312] eta: 0:00:38 lr: 0.004000 min_lr: 0.004000 loss: 3.0567 (2.9781) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [21] [270/312] eta: 0:00:30 lr: 0.004000 min_lr: 0.004000 loss: 3.0091 (2.9796) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [21] [280/312] eta: 0:00:23 lr: 0.004000 min_lr: 0.004000 loss: 3.0056 (2.9783) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [21] [290/312] eta: 0:00:16 lr: 0.004000 min_lr: 0.004000 loss: 3.1035 (2.9827) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [21] [300/312] eta: 0:00:08 lr: 0.004000 min_lr: 0.004000 loss: 2.8536 (2.9748) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [21] [310/312] eta: 0:00:01 lr: 0.004000 min_lr: 0.004000 loss: 2.7677 (2.9729) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [21] [311/312] eta: 0:00:00 lr: 0.004000 min_lr: 0.004000 loss: 2.6863 (2.9717) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [21] Total time: 0:03:47 (0.7304 s / it) Averaged stats: lr: 0.004000 min_lr: 0.004000 loss: 2.6863 (2.9977) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.2168 (1.2168) acc1: 71.3542 (71.3542) acc5: 88.0208 (88.0208) time: 4.6988 data: 4.4773 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4215 (1.3705) acc1: 64.3229 (65.6000) acc5: 88.0208 (86.7520) time: 0.6734 data: 0.4976 max mem: 64948 Test: Total time: 0:00:06 (0.7071 s / it) * Acc@1 66.968 Acc@5 87.284 loss 1.367 Accuracy of the model on the 50000 test images: 67.0% Max accuracy: 66.97% Test: [0/9] eta: 0:00:39 loss: 6.9854 (6.9854) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.4053 data: 4.2031 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9722 (6.9540) acc1: 0.0000 (0.0960) acc5: 0.2604 (0.5120) time: 0.6407 data: 0.4671 max mem: 64948 Test: Total time: 0:00:05 (0.6484 s / it) * Acc@1 0.098 Acc@5 0.504 loss 6.954 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [22] [ 0/312] eta: 0:57:12 lr: 0.004000 min_lr: 0.004000 loss: 2.5685 (2.5685) weight_decay: 0.0500 (0.0500) time: 11.0018 data: 8.1329 max mem: 64948 Epoch: [22] [ 10/312] eta: 0:08:21 lr: 0.004000 min_lr: 0.004000 loss: 2.9140 (3.0338) weight_decay: 0.0500 (0.0500) time: 1.6616 data: 0.7398 max mem: 64948 Epoch: [22] [ 20/312] eta: 0:05:51 lr: 0.004000 min_lr: 0.004000 loss: 2.8798 (2.9051) weight_decay: 0.0500 (0.0500) time: 0.7148 data: 0.0004 max mem: 64948 Epoch: [22] [ 30/312] eta: 0:04:53 lr: 0.004000 min_lr: 0.004000 loss: 2.8798 (2.9337) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0003 max mem: 64948 Epoch: [22] [ 40/312] eta: 0:04:20 lr: 0.004000 min_lr: 0.004000 loss: 3.0137 (2.9037) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [22] [ 50/312] eta: 0:03:57 lr: 0.004000 min_lr: 0.004000 loss: 2.9266 (2.9153) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [22] [ 60/312] eta: 0:03:39 lr: 0.004000 min_lr: 0.004000 loss: 2.9266 (2.9150) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [22] [ 70/312] eta: 0:03:24 lr: 0.004000 min_lr: 0.004000 loss: 2.9661 (2.9351) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [22] [ 80/312] eta: 0:03:11 lr: 0.004000 min_lr: 0.004000 loss: 2.8957 (2.9229) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [22] [ 90/312] eta: 0:03:00 lr: 0.004000 min_lr: 0.004000 loss: 2.7959 (2.9088) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [22] [100/312] eta: 0:02:49 lr: 0.004000 min_lr: 0.004000 loss: 3.0409 (2.9254) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [22] [110/312] eta: 0:02:39 lr: 0.004000 min_lr: 0.004000 loss: 2.9790 (2.9161) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [22] [120/312] eta: 0:02:30 lr: 0.004000 min_lr: 0.004000 loss: 2.9632 (2.9262) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [22] [130/312] eta: 0:02:21 lr: 0.004000 min_lr: 0.004000 loss: 2.9557 (2.9252) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [22] [140/312] eta: 0:02:12 lr: 0.004000 min_lr: 0.004000 loss: 2.8274 (2.9013) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [22] [150/312] eta: 0:02:04 lr: 0.004000 min_lr: 0.004000 loss: 2.6795 (2.9053) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [22] [160/312] eta: 0:01:55 lr: 0.004000 min_lr: 0.004000 loss: 3.1234 (2.9063) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [22] [170/312] eta: 0:01:47 lr: 0.004000 min_lr: 0.004000 loss: 3.1218 (2.9113) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [22] [180/312] eta: 0:01:39 lr: 0.004000 min_lr: 0.004000 loss: 3.1094 (2.9025) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [22] [190/312] eta: 0:01:31 lr: 0.004000 min_lr: 0.004000 loss: 2.9934 (2.9014) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [22] [200/312] eta: 0:01:23 lr: 0.004000 min_lr: 0.004000 loss: 3.1589 (2.9119) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [22] [210/312] eta: 0:01:16 lr: 0.004000 min_lr: 0.004000 loss: 3.0399 (2.9074) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [22] [220/312] eta: 0:01:08 lr: 0.004000 min_lr: 0.004000 loss: 2.9246 (2.9017) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [22] [230/312] eta: 0:01:00 lr: 0.004000 min_lr: 0.004000 loss: 2.9983 (2.9097) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [22] [240/312] eta: 0:00:53 lr: 0.004000 min_lr: 0.004000 loss: 3.0734 (2.9124) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [22] [250/312] eta: 0:00:45 lr: 0.004000 min_lr: 0.004000 loss: 2.8311 (2.9075) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [22] [260/312] eta: 0:00:38 lr: 0.004000 min_lr: 0.004000 loss: 2.7205 (2.9053) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [22] [270/312] eta: 0:00:30 lr: 0.004000 min_lr: 0.004000 loss: 2.9718 (2.9105) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [22] [280/312] eta: 0:00:23 lr: 0.004000 min_lr: 0.004000 loss: 3.1415 (2.9118) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [22] [290/312] eta: 0:00:16 lr: 0.004000 min_lr: 0.004000 loss: 3.2241 (2.9265) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [22] [300/312] eta: 0:00:08 lr: 0.004000 min_lr: 0.004000 loss: 3.3161 (2.9358) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [22] [310/312] eta: 0:00:01 lr: 0.004000 min_lr: 0.004000 loss: 3.2465 (2.9380) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [22] [311/312] eta: 0:00:00 lr: 0.004000 min_lr: 0.004000 loss: 3.2465 (2.9402) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [22] Total time: 0:03:48 (0.7327 s / it) Averaged stats: lr: 0.004000 min_lr: 0.004000 loss: 3.2465 (2.9542) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 1.1754 (1.1754) acc1: 73.6979 (73.6979) acc5: 90.1042 (90.1042) time: 4.5774 data: 4.3697 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.6257 (1.4888) acc1: 65.1042 (65.3440) acc5: 84.8958 (85.4720) time: 0.6606 data: 0.4856 max mem: 64948 Test: Total time: 0:00:06 (0.6871 s / it) * Acc@1 65.110 Acc@5 86.044 loss 1.486 Accuracy of the model on the 50000 test images: 65.1% Max accuracy: 66.97% Test: [0/9] eta: 0:00:46 loss: 7.0064 (7.0064) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.1946 data: 4.9768 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9791 (6.9619) acc1: 0.0000 (0.0960) acc5: 0.2604 (0.5120) time: 0.7286 data: 0.5531 max mem: 64948 Test: Total time: 0:00:06 (0.7376 s / it) * Acc@1 0.098 Acc@5 0.498 loss 6.962 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [23] [ 0/312] eta: 0:54:44 lr: 0.004000 min_lr: 0.004000 loss: 3.0903 (3.0903) weight_decay: 0.0500 (0.0500) time: 10.5279 data: 7.1328 max mem: 64948 Epoch: [23] [ 10/312] eta: 0:08:09 lr: 0.004000 min_lr: 0.004000 loss: 2.5825 (2.5953) weight_decay: 0.0500 (0.0500) time: 1.6196 data: 0.6488 max mem: 64948 Epoch: [23] [ 20/312] eta: 0:05:44 lr: 0.003999 min_lr: 0.003999 loss: 2.8324 (2.8101) weight_decay: 0.0500 (0.0500) time: 0.7113 data: 0.0004 max mem: 64948 Epoch: [23] [ 30/312] eta: 0:04:48 lr: 0.003999 min_lr: 0.003999 loss: 2.9396 (2.7879) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [23] [ 40/312] eta: 0:04:16 lr: 0.003999 min_lr: 0.003999 loss: 2.9084 (2.8347) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [23] [ 50/312] eta: 0:03:54 lr: 0.003999 min_lr: 0.003999 loss: 2.9960 (2.8592) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [23] [ 60/312] eta: 0:03:37 lr: 0.003999 min_lr: 0.003999 loss: 3.0478 (2.8842) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [23] [ 70/312] eta: 0:03:23 lr: 0.003999 min_lr: 0.003999 loss: 3.0067 (2.8933) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [23] [ 80/312] eta: 0:03:10 lr: 0.003999 min_lr: 0.003999 loss: 3.1213 (2.9108) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [23] [ 90/312] eta: 0:02:59 lr: 0.003999 min_lr: 0.003999 loss: 3.1872 (2.9508) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [23] [100/312] eta: 0:02:48 lr: 0.003999 min_lr: 0.003999 loss: 3.2123 (2.9535) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [23] [110/312] eta: 0:02:38 lr: 0.003999 min_lr: 0.003999 loss: 2.8489 (2.9260) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [23] [120/312] eta: 0:02:29 lr: 0.003999 min_lr: 0.003999 loss: 2.9456 (2.9300) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [23] [130/312] eta: 0:02:20 lr: 0.003999 min_lr: 0.003999 loss: 2.9456 (2.9309) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [23] [140/312] eta: 0:02:11 lr: 0.003999 min_lr: 0.003999 loss: 2.9362 (2.9406) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [23] [150/312] eta: 0:02:03 lr: 0.003999 min_lr: 0.003999 loss: 2.8896 (2.9330) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [23] [160/312] eta: 0:01:55 lr: 0.003999 min_lr: 0.003999 loss: 2.8237 (2.9261) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [23] [170/312] eta: 0:01:47 lr: 0.003999 min_lr: 0.003999 loss: 3.0382 (2.9392) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [23] [180/312] eta: 0:01:39 lr: 0.003999 min_lr: 0.003999 loss: 3.0461 (2.9365) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [23] [190/312] eta: 0:01:31 lr: 0.003999 min_lr: 0.003999 loss: 2.8502 (2.9309) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [23] [200/312] eta: 0:01:23 lr: 0.003999 min_lr: 0.003999 loss: 2.9134 (2.9338) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [23] [210/312] eta: 0:01:15 lr: 0.003999 min_lr: 0.003999 loss: 3.0107 (2.9344) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [23] [220/312] eta: 0:01:08 lr: 0.003999 min_lr: 0.003999 loss: 2.9427 (2.9323) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [23] [230/312] eta: 0:01:00 lr: 0.003999 min_lr: 0.003999 loss: 2.8651 (2.9244) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [23] [240/312] eta: 0:00:53 lr: 0.003999 min_lr: 0.003999 loss: 2.8888 (2.9226) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [23] [250/312] eta: 0:00:45 lr: 0.003999 min_lr: 0.003999 loss: 2.9409 (2.9247) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [23] [260/312] eta: 0:00:38 lr: 0.003999 min_lr: 0.003999 loss: 2.8448 (2.9172) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [23] [270/312] eta: 0:00:30 lr: 0.003999 min_lr: 0.003999 loss: 2.6408 (2.9127) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [23] [280/312] eta: 0:00:23 lr: 0.003999 min_lr: 0.003999 loss: 2.8078 (2.9100) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0009 max mem: 64948 Epoch: [23] [290/312] eta: 0:00:16 lr: 0.003999 min_lr: 0.003999 loss: 3.0373 (2.9157) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [23] [300/312] eta: 0:00:08 lr: 0.003999 min_lr: 0.003999 loss: 3.0373 (2.9138) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [23] [310/312] eta: 0:00:01 lr: 0.003999 min_lr: 0.003999 loss: 2.8376 (2.9131) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [23] [311/312] eta: 0:00:00 lr: 0.003999 min_lr: 0.003999 loss: 2.8130 (2.9107) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [23] Total time: 0:03:47 (0.7300 s / it) Averaged stats: lr: 0.003999 min_lr: 0.003999 loss: 2.8130 (2.9191) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 1.0932 (1.0932) acc1: 76.0417 (76.0417) acc5: 91.4062 (91.4062) time: 4.4796 data: 4.2650 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4648 (1.3885) acc1: 65.6250 (66.5920) acc5: 87.5000 (87.1360) time: 0.6491 data: 0.4740 max mem: 64948 Test: Total time: 0:00:06 (0.6728 s / it) * Acc@1 66.682 Acc@5 87.300 loss 1.402 Accuracy of the model on the 50000 test images: 66.7% Max accuracy: 66.97% Test: [0/9] eta: 0:00:43 loss: 7.0268 (7.0268) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.8505 data: 4.6325 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9868 (6.9703) acc1: 0.0000 (0.0960) acc5: 0.5208 (0.5120) time: 0.6958 data: 0.5203 max mem: 64948 Test: Total time: 0:00:06 (0.7121 s / it) * Acc@1 0.098 Acc@5 0.498 loss 6.970 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [24] [ 0/312] eta: 0:56:13 lr: 0.003999 min_lr: 0.003999 loss: 3.0782 (3.0782) weight_decay: 0.0500 (0.0500) time: 10.8113 data: 6.1346 max mem: 64948 Epoch: [24] [ 10/312] eta: 0:08:13 lr: 0.003999 min_lr: 0.003999 loss: 3.0601 (2.9554) weight_decay: 0.0500 (0.0500) time: 1.6355 data: 0.5581 max mem: 64948 Epoch: [24] [ 20/312] eta: 0:05:46 lr: 0.003999 min_lr: 0.003999 loss: 2.8456 (2.8352) weight_decay: 0.0500 (0.0500) time: 0.7066 data: 0.0004 max mem: 64948 Epoch: [24] [ 30/312] eta: 0:04:50 lr: 0.003999 min_lr: 0.003999 loss: 2.8104 (2.8491) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [24] [ 40/312] eta: 0:04:17 lr: 0.003999 min_lr: 0.003999 loss: 2.9296 (2.8436) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [24] [ 50/312] eta: 0:03:55 lr: 0.003999 min_lr: 0.003999 loss: 2.7517 (2.8168) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [24] [ 60/312] eta: 0:03:38 lr: 0.003999 min_lr: 0.003999 loss: 2.9963 (2.8692) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [24] [ 70/312] eta: 0:03:23 lr: 0.003999 min_lr: 0.003999 loss: 2.9963 (2.8420) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [24] [ 80/312] eta: 0:03:11 lr: 0.003999 min_lr: 0.003999 loss: 2.7125 (2.8168) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [24] [ 90/312] eta: 0:02:59 lr: 0.003999 min_lr: 0.003999 loss: 2.8668 (2.8364) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [24] [100/312] eta: 0:02:49 lr: 0.003999 min_lr: 0.003999 loss: 2.9541 (2.8453) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [24] [110/312] eta: 0:02:39 lr: 0.003999 min_lr: 0.003999 loss: 2.8407 (2.8427) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [24] [120/312] eta: 0:02:29 lr: 0.003999 min_lr: 0.003999 loss: 2.8981 (2.8492) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [24] [130/312] eta: 0:02:20 lr: 0.003999 min_lr: 0.003999 loss: 3.1041 (2.8474) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [24] [140/312] eta: 0:02:12 lr: 0.003999 min_lr: 0.003999 loss: 3.1041 (2.8550) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [24] [150/312] eta: 0:02:03 lr: 0.003999 min_lr: 0.003999 loss: 3.0286 (2.8546) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [24] [160/312] eta: 0:01:55 lr: 0.003999 min_lr: 0.003999 loss: 2.8041 (2.8562) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [24] [170/312] eta: 0:01:47 lr: 0.003999 min_lr: 0.003999 loss: 2.9774 (2.8582) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [24] [180/312] eta: 0:01:39 lr: 0.003999 min_lr: 0.003999 loss: 3.0409 (2.8710) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [24] [190/312] eta: 0:01:31 lr: 0.003999 min_lr: 0.003999 loss: 3.0409 (2.8644) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [24] [200/312] eta: 0:01:23 lr: 0.003999 min_lr: 0.003999 loss: 2.8638 (2.8696) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [24] [210/312] eta: 0:01:15 lr: 0.003999 min_lr: 0.003999 loss: 2.8803 (2.8745) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [24] [220/312] eta: 0:01:08 lr: 0.003999 min_lr: 0.003999 loss: 3.1332 (2.8854) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [24] [230/312] eta: 0:01:00 lr: 0.003999 min_lr: 0.003999 loss: 2.6842 (2.8701) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [24] [240/312] eta: 0:00:53 lr: 0.003999 min_lr: 0.003999 loss: 2.5658 (2.8614) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [24] [250/312] eta: 0:00:45 lr: 0.003999 min_lr: 0.003999 loss: 2.8103 (2.8619) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [24] [260/312] eta: 0:00:38 lr: 0.003999 min_lr: 0.003999 loss: 2.8099 (2.8579) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [24] [270/312] eta: 0:00:30 lr: 0.003999 min_lr: 0.003999 loss: 2.7639 (2.8559) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [24] [280/312] eta: 0:00:23 lr: 0.003999 min_lr: 0.003999 loss: 2.7278 (2.8539) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [24] [290/312] eta: 0:00:16 lr: 0.003999 min_lr: 0.003999 loss: 2.7278 (2.8520) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [24] [300/312] eta: 0:00:08 lr: 0.003999 min_lr: 0.003999 loss: 2.8089 (2.8536) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [24] [310/312] eta: 0:00:01 lr: 0.003999 min_lr: 0.003999 loss: 2.8089 (2.8528) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [24] [311/312] eta: 0:00:00 lr: 0.003999 min_lr: 0.003999 loss: 2.8089 (2.8502) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [24] Total time: 0:03:48 (0.7313 s / it) Averaged stats: lr: 0.003999 min_lr: 0.003999 loss: 2.8089 (2.8860) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 1.0894 (1.0894) acc1: 74.7396 (74.7396) acc5: 91.6667 (91.6667) time: 4.4708 data: 4.2563 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4626 (1.3762) acc1: 66.6667 (66.8800) acc5: 85.6771 (87.2000) time: 0.6480 data: 0.4730 max mem: 64948 Test: Total time: 0:00:06 (0.6696 s / it) * Acc@1 67.792 Acc@5 87.918 loss 1.342 Accuracy of the model on the 50000 test images: 67.8% Max accuracy: 67.79% Test: [0/9] eta: 0:00:40 loss: 7.0505 (7.0505) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.4598 data: 4.2495 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9971 (6.9795) acc1: 0.0000 (0.1280) acc5: 0.5208 (0.5120) time: 0.6468 data: 0.4723 max mem: 64948 Test: Total time: 0:00:05 (0.6538 s / it) * Acc@1 0.100 Acc@5 0.500 loss 6.980 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [25] [ 0/312] eta: 0:52:34 lr: 0.003999 min_lr: 0.003999 loss: 3.4730 (3.4730) weight_decay: 0.0500 (0.0500) time: 10.1091 data: 9.3020 max mem: 64948 Epoch: [25] [ 10/312] eta: 0:07:59 lr: 0.003999 min_lr: 0.003999 loss: 2.9352 (2.9563) weight_decay: 0.0500 (0.0500) time: 1.5876 data: 0.8460 max mem: 64948 Epoch: [25] [ 20/312] eta: 0:05:39 lr: 0.003999 min_lr: 0.003999 loss: 2.9352 (2.9686) weight_decay: 0.0500 (0.0500) time: 0.7152 data: 0.0004 max mem: 64948 Epoch: [25] [ 30/312] eta: 0:04:45 lr: 0.003999 min_lr: 0.003999 loss: 2.9593 (2.8894) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [25] [ 40/312] eta: 0:04:14 lr: 0.003999 min_lr: 0.003999 loss: 2.8555 (2.8583) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [25] [ 50/312] eta: 0:03:52 lr: 0.003999 min_lr: 0.003999 loss: 3.1417 (2.8918) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [25] [ 60/312] eta: 0:03:35 lr: 0.003999 min_lr: 0.003999 loss: 2.9729 (2.8629) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [25] [ 70/312] eta: 0:03:21 lr: 0.003999 min_lr: 0.003999 loss: 2.9757 (2.8968) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [25] [ 80/312] eta: 0:03:09 lr: 0.003999 min_lr: 0.003999 loss: 3.0450 (2.8903) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [25] [ 90/312] eta: 0:02:58 lr: 0.003999 min_lr: 0.003999 loss: 2.9561 (2.8830) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [25] [100/312] eta: 0:02:48 lr: 0.003998 min_lr: 0.003998 loss: 3.0118 (2.8909) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [25] [110/312] eta: 0:02:38 lr: 0.003998 min_lr: 0.003998 loss: 3.0118 (2.9016) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [25] [120/312] eta: 0:02:29 lr: 0.003998 min_lr: 0.003998 loss: 2.7808 (2.8779) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [25] [130/312] eta: 0:02:20 lr: 0.003998 min_lr: 0.003998 loss: 2.7767 (2.8758) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [25] [140/312] eta: 0:02:11 lr: 0.003998 min_lr: 0.003998 loss: 2.8865 (2.8775) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [25] [150/312] eta: 0:02:03 lr: 0.003998 min_lr: 0.003998 loss: 2.9150 (2.8809) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [25] [160/312] eta: 0:01:54 lr: 0.003998 min_lr: 0.003998 loss: 2.9078 (2.8749) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [25] [170/312] eta: 0:01:46 lr: 0.003998 min_lr: 0.003998 loss: 2.7505 (2.8600) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [25] [180/312] eta: 0:01:38 lr: 0.003998 min_lr: 0.003998 loss: 2.6921 (2.8574) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [25] [190/312] eta: 0:01:31 lr: 0.003998 min_lr: 0.003998 loss: 2.7093 (2.8577) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [25] [200/312] eta: 0:01:23 lr: 0.003998 min_lr: 0.003998 loss: 2.8586 (2.8566) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [25] [210/312] eta: 0:01:15 lr: 0.003998 min_lr: 0.003998 loss: 2.8134 (2.8526) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [25] [220/312] eta: 0:01:08 lr: 0.003998 min_lr: 0.003998 loss: 2.8258 (2.8534) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [25] [230/312] eta: 0:01:00 lr: 0.003998 min_lr: 0.003998 loss: 2.6689 (2.8457) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [25] [240/312] eta: 0:00:52 lr: 0.003998 min_lr: 0.003998 loss: 2.9890 (2.8503) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [25] [250/312] eta: 0:00:45 lr: 0.003998 min_lr: 0.003998 loss: 3.0299 (2.8410) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [25] [260/312] eta: 0:00:38 lr: 0.003998 min_lr: 0.003998 loss: 2.6960 (2.8371) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [25] [270/312] eta: 0:00:30 lr: 0.003998 min_lr: 0.003998 loss: 2.9919 (2.8438) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [25] [280/312] eta: 0:00:23 lr: 0.003998 min_lr: 0.003998 loss: 3.1070 (2.8461) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0009 max mem: 64948 Epoch: [25] [290/312] eta: 0:00:16 lr: 0.003998 min_lr: 0.003998 loss: 3.0459 (2.8502) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0008 max mem: 64948 Epoch: [25] [300/312] eta: 0:00:08 lr: 0.003998 min_lr: 0.003998 loss: 2.8924 (2.8500) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [25] [310/312] eta: 0:00:01 lr: 0.003998 min_lr: 0.003998 loss: 2.7308 (2.8466) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0001 max mem: 64948 Epoch: [25] [311/312] eta: 0:00:00 lr: 0.003998 min_lr: 0.003998 loss: 2.6930 (2.8462) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0001 max mem: 64948 Epoch: [25] Total time: 0:03:47 (0.7296 s / it) Averaged stats: lr: 0.003998 min_lr: 0.003998 loss: 2.6930 (2.8687) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.0708 (1.0708) acc1: 74.4792 (74.4792) acc5: 91.9271 (91.9271) time: 4.7663 data: 4.5472 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3223 (1.3897) acc1: 69.2708 (66.9440) acc5: 87.5000 (87.5520) time: 0.6808 data: 0.5053 max mem: 64948 Test: Total time: 0:00:06 (0.6911 s / it) * Acc@1 67.450 Acc@5 87.562 loss 1.376 Accuracy of the model on the 50000 test images: 67.5% Max accuracy: 67.79% Test: [0/9] eta: 0:00:42 loss: 7.0707 (7.0707) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.7446 data: 4.5267 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0075 (6.9895) acc1: 0.0000 (0.1280) acc5: 0.0000 (0.5120) time: 0.6786 data: 0.5031 max mem: 64948 Test: Total time: 0:00:06 (0.6891 s / it) * Acc@1 0.100 Acc@5 0.504 loss 6.990 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [26] [ 0/312] eta: 0:50:38 lr: 0.003998 min_lr: 0.003998 loss: 3.3374 (3.3374) weight_decay: 0.0500 (0.0500) time: 9.7395 data: 8.7410 max mem: 64948 Epoch: [26] [ 10/312] eta: 0:07:48 lr: 0.003998 min_lr: 0.003998 loss: 2.9815 (2.8666) weight_decay: 0.0500 (0.0500) time: 1.5528 data: 0.7951 max mem: 64948 Epoch: [26] [ 20/312] eta: 0:05:33 lr: 0.003998 min_lr: 0.003998 loss: 2.8799 (2.8359) weight_decay: 0.0500 (0.0500) time: 0.7135 data: 0.0004 max mem: 64948 Epoch: [26] [ 30/312] eta: 0:04:42 lr: 0.003998 min_lr: 0.003998 loss: 2.6814 (2.8159) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [26] [ 40/312] eta: 0:04:12 lr: 0.003998 min_lr: 0.003998 loss: 2.7682 (2.8012) weight_decay: 0.0500 (0.0500) time: 0.7006 data: 0.0003 max mem: 64948 Epoch: [26] [ 50/312] eta: 0:03:51 lr: 0.003998 min_lr: 0.003998 loss: 2.8101 (2.7811) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [26] [ 60/312] eta: 0:03:34 lr: 0.003998 min_lr: 0.003998 loss: 2.8458 (2.8018) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [26] [ 70/312] eta: 0:03:20 lr: 0.003998 min_lr: 0.003998 loss: 2.9095 (2.8022) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [26] [ 80/312] eta: 0:03:08 lr: 0.003998 min_lr: 0.003998 loss: 2.9432 (2.8103) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [26] [ 90/312] eta: 0:02:57 lr: 0.003998 min_lr: 0.003998 loss: 2.8716 (2.8098) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [26] [100/312] eta: 0:02:47 lr: 0.003998 min_lr: 0.003998 loss: 2.8426 (2.8148) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [26] [110/312] eta: 0:02:37 lr: 0.003998 min_lr: 0.003998 loss: 2.8961 (2.8163) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [26] [120/312] eta: 0:02:28 lr: 0.003998 min_lr: 0.003998 loss: 2.5949 (2.7949) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [26] [130/312] eta: 0:02:19 lr: 0.003998 min_lr: 0.003998 loss: 2.4990 (2.7875) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [26] [140/312] eta: 0:02:11 lr: 0.003998 min_lr: 0.003998 loss: 2.8553 (2.7933) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [26] [150/312] eta: 0:02:02 lr: 0.003998 min_lr: 0.003998 loss: 3.0224 (2.8043) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [26] [160/312] eta: 0:01:54 lr: 0.003998 min_lr: 0.003998 loss: 3.0122 (2.8175) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [26] [170/312] eta: 0:01:46 lr: 0.003998 min_lr: 0.003998 loss: 2.9428 (2.8167) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [26] [180/312] eta: 0:01:38 lr: 0.003998 min_lr: 0.003998 loss: 2.7863 (2.8236) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [26] [190/312] eta: 0:01:30 lr: 0.003998 min_lr: 0.003998 loss: 2.7551 (2.8155) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [26] [200/312] eta: 0:01:23 lr: 0.003998 min_lr: 0.003998 loss: 2.6546 (2.8109) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [26] [210/312] eta: 0:01:15 lr: 0.003998 min_lr: 0.003998 loss: 2.6842 (2.8138) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [26] [220/312] eta: 0:01:07 lr: 0.003998 min_lr: 0.003998 loss: 2.8286 (2.8108) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [26] [230/312] eta: 0:01:00 lr: 0.003998 min_lr: 0.003998 loss: 2.7613 (2.8130) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [26] [240/312] eta: 0:00:52 lr: 0.003998 min_lr: 0.003998 loss: 2.8578 (2.8082) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [26] [250/312] eta: 0:00:45 lr: 0.003998 min_lr: 0.003998 loss: 2.9829 (2.8193) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [26] [260/312] eta: 0:00:38 lr: 0.003998 min_lr: 0.003998 loss: 3.1198 (2.8170) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [26] [270/312] eta: 0:00:30 lr: 0.003997 min_lr: 0.003997 loss: 2.8683 (2.8205) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [26] [280/312] eta: 0:00:23 lr: 0.003997 min_lr: 0.003997 loss: 2.8683 (2.8224) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [26] [290/312] eta: 0:00:16 lr: 0.003997 min_lr: 0.003997 loss: 2.8013 (2.8223) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [26] [300/312] eta: 0:00:08 lr: 0.003997 min_lr: 0.003997 loss: 2.7495 (2.8248) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [26] [310/312] eta: 0:00:01 lr: 0.003997 min_lr: 0.003997 loss: 2.7517 (2.8195) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [26] [311/312] eta: 0:00:00 lr: 0.003997 min_lr: 0.003997 loss: 2.7517 (2.8203) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [26] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.003997 min_lr: 0.003997 loss: 2.7517 (2.8286) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 1.0163 (1.0163) acc1: 77.6042 (77.6042) acc5: 91.4062 (91.4062) time: 4.4459 data: 4.2369 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4993 (1.3729) acc1: 68.2292 (67.4880) acc5: 86.9792 (86.8160) time: 0.6453 data: 0.4709 max mem: 64948 Test: Total time: 0:00:05 (0.6664 s / it) * Acc@1 67.410 Acc@5 87.518 loss 1.346 Accuracy of the model on the 50000 test images: 67.4% Max accuracy: 67.79% Test: [0/9] eta: 0:00:45 loss: 7.0919 (7.0919) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.0187 data: 4.8009 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0211 (7.0003) acc1: 0.0000 (0.1280) acc5: 0.0000 (0.5120) time: 0.7095 data: 0.5335 max mem: 64948 Test: Total time: 0:00:06 (0.7188 s / it) * Acc@1 0.100 Acc@5 0.506 loss 7.000 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [27] [ 0/312] eta: 0:54:23 lr: 0.003997 min_lr: 0.003997 loss: 2.6709 (2.6709) weight_decay: 0.0500 (0.0500) time: 10.4596 data: 6.9560 max mem: 64948 Epoch: [27] [ 10/312] eta: 0:08:07 lr: 0.003997 min_lr: 0.003997 loss: 2.5734 (2.6269) weight_decay: 0.0500 (0.0500) time: 1.6128 data: 0.6328 max mem: 64948 Epoch: [27] [ 20/312] eta: 0:05:43 lr: 0.003997 min_lr: 0.003997 loss: 2.8357 (2.7942) weight_decay: 0.0500 (0.0500) time: 0.7108 data: 0.0004 max mem: 64948 Epoch: [27] [ 30/312] eta: 0:04:47 lr: 0.003997 min_lr: 0.003997 loss: 2.9533 (2.8404) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [27] [ 40/312] eta: 0:04:16 lr: 0.003997 min_lr: 0.003997 loss: 3.0071 (2.8352) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [27] [ 50/312] eta: 0:03:53 lr: 0.003997 min_lr: 0.003997 loss: 2.7230 (2.7898) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [27] [ 60/312] eta: 0:03:36 lr: 0.003997 min_lr: 0.003997 loss: 2.5502 (2.7575) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [27] [ 70/312] eta: 0:03:22 lr: 0.003997 min_lr: 0.003997 loss: 2.7644 (2.7947) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [27] [ 80/312] eta: 0:03:10 lr: 0.003997 min_lr: 0.003997 loss: 3.0410 (2.8167) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [27] [ 90/312] eta: 0:02:59 lr: 0.003997 min_lr: 0.003997 loss: 2.9585 (2.8239) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [27] [100/312] eta: 0:02:48 lr: 0.003997 min_lr: 0.003997 loss: 2.8997 (2.8224) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [27] [110/312] eta: 0:02:38 lr: 0.003997 min_lr: 0.003997 loss: 2.9012 (2.8258) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [27] [120/312] eta: 0:02:29 lr: 0.003997 min_lr: 0.003997 loss: 2.9092 (2.8362) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [27] [130/312] eta: 0:02:20 lr: 0.003997 min_lr: 0.003997 loss: 3.0590 (2.8420) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [27] [140/312] eta: 0:02:12 lr: 0.003997 min_lr: 0.003997 loss: 3.0612 (2.8455) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [27] [150/312] eta: 0:02:03 lr: 0.003997 min_lr: 0.003997 loss: 2.8687 (2.8434) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [27] [160/312] eta: 0:01:55 lr: 0.003997 min_lr: 0.003997 loss: 2.8297 (2.8289) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [27] [170/312] eta: 0:01:47 lr: 0.003997 min_lr: 0.003997 loss: 2.8548 (2.8449) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [27] [180/312] eta: 0:01:39 lr: 0.003997 min_lr: 0.003997 loss: 3.0302 (2.8463) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [27] [190/312] eta: 0:01:31 lr: 0.003997 min_lr: 0.003997 loss: 2.9859 (2.8486) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [27] [200/312] eta: 0:01:23 lr: 0.003997 min_lr: 0.003997 loss: 2.9026 (2.8529) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [27] [210/312] eta: 0:01:15 lr: 0.003997 min_lr: 0.003997 loss: 2.8938 (2.8478) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [27] [220/312] eta: 0:01:08 lr: 0.003997 min_lr: 0.003997 loss: 2.8766 (2.8521) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [27] [230/312] eta: 0:01:00 lr: 0.003997 min_lr: 0.003997 loss: 2.8373 (2.8464) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [27] [240/312] eta: 0:00:53 lr: 0.003997 min_lr: 0.003997 loss: 2.6726 (2.8336) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [27] [250/312] eta: 0:00:45 lr: 0.003997 min_lr: 0.003997 loss: 2.7361 (2.8317) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [27] [260/312] eta: 0:00:38 lr: 0.003997 min_lr: 0.003997 loss: 2.6719 (2.8204) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [27] [270/312] eta: 0:00:30 lr: 0.003997 min_lr: 0.003997 loss: 2.5019 (2.8164) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [27] [280/312] eta: 0:00:23 lr: 0.003997 min_lr: 0.003997 loss: 3.0012 (2.8227) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0006 max mem: 64948 Epoch: [27] [290/312] eta: 0:00:16 lr: 0.003997 min_lr: 0.003997 loss: 2.9107 (2.8173) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [27] [300/312] eta: 0:00:08 lr: 0.003997 min_lr: 0.003997 loss: 2.7278 (2.8220) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [27] [310/312] eta: 0:00:01 lr: 0.003997 min_lr: 0.003997 loss: 2.9000 (2.8214) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [27] [311/312] eta: 0:00:00 lr: 0.003997 min_lr: 0.003997 loss: 2.7991 (2.8212) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [27] Total time: 0:03:48 (0.7308 s / it) Averaged stats: lr: 0.003997 min_lr: 0.003997 loss: 2.7991 (2.8090) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 1.0556 (1.0556) acc1: 76.5625 (76.5625) acc5: 92.4479 (92.4479) time: 4.5784 data: 4.3589 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3920 (1.3119) acc1: 67.7083 (67.8080) acc5: 87.2396 (87.9040) time: 0.6606 data: 0.4844 max mem: 64948 Test: Total time: 0:00:06 (0.6851 s / it) * Acc@1 68.126 Acc@5 88.254 loss 1.329 Accuracy of the model on the 50000 test images: 68.1% Max accuracy: 68.13% Test: [0/9] eta: 0:00:42 loss: 7.1164 (7.1164) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.7025 data: 4.4847 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0364 (7.0123) acc1: 0.0000 (0.1280) acc5: 0.0000 (0.5120) time: 0.6744 data: 0.4984 max mem: 64948 Test: Total time: 0:00:06 (0.6828 s / it) * Acc@1 0.100 Acc@5 0.504 loss 7.012 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [28] [ 0/312] eta: 0:52:25 lr: 0.003997 min_lr: 0.003997 loss: 3.4712 (3.4712) weight_decay: 0.0500 (0.0500) time: 10.0805 data: 6.4622 max mem: 64948 Epoch: [28] [ 10/312] eta: 0:08:01 lr: 0.003997 min_lr: 0.003997 loss: 2.9630 (2.9212) weight_decay: 0.0500 (0.0500) time: 1.5942 data: 0.5881 max mem: 64948 Epoch: [28] [ 20/312] eta: 0:05:40 lr: 0.003997 min_lr: 0.003997 loss: 2.8712 (2.9133) weight_decay: 0.0500 (0.0500) time: 0.7197 data: 0.0005 max mem: 64948 Epoch: [28] [ 30/312] eta: 0:04:46 lr: 0.003997 min_lr: 0.003997 loss: 2.9722 (2.9066) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [28] [ 40/312] eta: 0:04:14 lr: 0.003996 min_lr: 0.003996 loss: 2.9722 (2.8778) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [28] [ 50/312] eta: 0:03:53 lr: 0.003996 min_lr: 0.003996 loss: 2.8873 (2.8697) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [28] [ 60/312] eta: 0:03:36 lr: 0.003996 min_lr: 0.003996 loss: 2.9042 (2.8658) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [28] [ 70/312] eta: 0:03:21 lr: 0.003996 min_lr: 0.003996 loss: 2.7567 (2.8323) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [28] [ 80/312] eta: 0:03:09 lr: 0.003996 min_lr: 0.003996 loss: 2.6945 (2.8259) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [28] [ 90/312] eta: 0:02:58 lr: 0.003996 min_lr: 0.003996 loss: 2.8297 (2.8285) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [28] [100/312] eta: 0:02:48 lr: 0.003996 min_lr: 0.003996 loss: 3.0103 (2.8329) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [28] [110/312] eta: 0:02:38 lr: 0.003996 min_lr: 0.003996 loss: 2.9329 (2.8370) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [28] [120/312] eta: 0:02:29 lr: 0.003996 min_lr: 0.003996 loss: 2.7627 (2.8287) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [28] [130/312] eta: 0:02:20 lr: 0.003996 min_lr: 0.003996 loss: 2.7346 (2.8202) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [28] [140/312] eta: 0:02:11 lr: 0.003996 min_lr: 0.003996 loss: 2.8168 (2.8139) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [28] [150/312] eta: 0:02:03 lr: 0.003996 min_lr: 0.003996 loss: 2.8444 (2.8147) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [28] [160/312] eta: 0:01:55 lr: 0.003996 min_lr: 0.003996 loss: 2.8657 (2.8075) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [28] [170/312] eta: 0:01:46 lr: 0.003996 min_lr: 0.003996 loss: 2.7072 (2.7945) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [28] [180/312] eta: 0:01:38 lr: 0.003996 min_lr: 0.003996 loss: 2.8095 (2.8006) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [28] [190/312] eta: 0:01:31 lr: 0.003996 min_lr: 0.003996 loss: 2.9280 (2.8063) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [28] [200/312] eta: 0:01:23 lr: 0.003996 min_lr: 0.003996 loss: 2.9402 (2.8139) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [28] [210/312] eta: 0:01:15 lr: 0.003996 min_lr: 0.003996 loss: 2.7294 (2.8044) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [28] [220/312] eta: 0:01:08 lr: 0.003996 min_lr: 0.003996 loss: 2.6237 (2.7987) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [28] [230/312] eta: 0:01:00 lr: 0.003996 min_lr: 0.003996 loss: 3.0526 (2.8138) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [28] [240/312] eta: 0:00:52 lr: 0.003996 min_lr: 0.003996 loss: 3.0526 (2.8053) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [28] [250/312] eta: 0:00:45 lr: 0.003996 min_lr: 0.003996 loss: 2.8262 (2.8063) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [28] [260/312] eta: 0:00:38 lr: 0.003996 min_lr: 0.003996 loss: 2.8319 (2.8015) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [28] [270/312] eta: 0:00:30 lr: 0.003996 min_lr: 0.003996 loss: 2.6264 (2.7893) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [28] [280/312] eta: 0:00:23 lr: 0.003996 min_lr: 0.003996 loss: 2.6201 (2.7866) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [28] [290/312] eta: 0:00:16 lr: 0.003996 min_lr: 0.003996 loss: 2.9347 (2.7919) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [28] [300/312] eta: 0:00:08 lr: 0.003996 min_lr: 0.003996 loss: 2.9347 (2.7909) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [28] [310/312] eta: 0:00:01 lr: 0.003996 min_lr: 0.003996 loss: 2.6085 (2.7862) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [28] [311/312] eta: 0:00:00 lr: 0.003996 min_lr: 0.003996 loss: 2.5974 (2.7844) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [28] Total time: 0:03:47 (0.7298 s / it) Averaged stats: lr: 0.003996 min_lr: 0.003996 loss: 2.5974 (2.7703) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.9457 (0.9457) acc1: 79.1667 (79.1667) acc5: 91.9271 (91.9271) time: 4.6324 data: 4.4183 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3707 (1.2498) acc1: 67.7083 (68.7680) acc5: 89.0625 (88.2240) time: 0.6660 data: 0.4910 max mem: 64948 Test: Total time: 0:00:06 (0.6895 s / it) * Acc@1 69.142 Acc@5 88.968 loss 1.250 Accuracy of the model on the 50000 test images: 69.1% Max accuracy: 69.14% Test: [0/9] eta: 0:00:41 loss: 7.1363 (7.1363) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.5732 data: 4.3695 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0495 (7.0234) acc1: 0.0000 (0.1280) acc5: 0.0000 (0.5120) time: 0.6594 data: 0.4856 max mem: 64948 Test: Total time: 0:00:06 (0.6678 s / it) * Acc@1 0.100 Acc@5 0.508 loss 7.023 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [29] [ 0/312] eta: 0:55:38 lr: 0.003996 min_lr: 0.003996 loss: 2.0945 (2.0945) weight_decay: 0.0500 (0.0500) time: 10.7011 data: 6.9223 max mem: 64948 Epoch: [29] [ 10/312] eta: 0:08:11 lr: 0.003996 min_lr: 0.003996 loss: 2.8467 (2.6847) weight_decay: 0.0500 (0.0500) time: 1.6277 data: 0.6297 max mem: 64948 Epoch: [29] [ 20/312] eta: 0:05:45 lr: 0.003996 min_lr: 0.003996 loss: 2.8467 (2.7046) weight_decay: 0.0500 (0.0500) time: 0.7064 data: 0.0004 max mem: 64948 Epoch: [29] [ 30/312] eta: 0:04:48 lr: 0.003996 min_lr: 0.003996 loss: 2.6627 (2.7264) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0003 max mem: 64948 Epoch: [29] [ 40/312] eta: 0:04:17 lr: 0.003996 min_lr: 0.003996 loss: 2.8058 (2.7395) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [29] [ 50/312] eta: 0:03:54 lr: 0.003996 min_lr: 0.003996 loss: 2.7872 (2.7374) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0003 max mem: 64948 Epoch: [29] [ 60/312] eta: 0:03:37 lr: 0.003995 min_lr: 0.003995 loss: 2.7872 (2.7682) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [29] [ 70/312] eta: 0:03:23 lr: 0.003995 min_lr: 0.003995 loss: 2.8166 (2.7423) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [29] [ 80/312] eta: 0:03:10 lr: 0.003995 min_lr: 0.003995 loss: 2.6629 (2.7384) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0004 max mem: 64948 Epoch: [29] [ 90/312] eta: 0:02:59 lr: 0.003995 min_lr: 0.003995 loss: 2.6629 (2.7265) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [29] [100/312] eta: 0:02:49 lr: 0.003995 min_lr: 0.003995 loss: 2.7207 (2.7261) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [29] [110/312] eta: 0:02:39 lr: 0.003995 min_lr: 0.003995 loss: 2.6029 (2.7154) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [29] [120/312] eta: 0:02:30 lr: 0.003995 min_lr: 0.003995 loss: 2.6029 (2.7311) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [29] [130/312] eta: 0:02:21 lr: 0.003995 min_lr: 0.003995 loss: 2.8076 (2.7184) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [29] [140/312] eta: 0:02:12 lr: 0.003995 min_lr: 0.003995 loss: 2.7019 (2.7151) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [29] [150/312] eta: 0:02:03 lr: 0.003995 min_lr: 0.003995 loss: 2.7352 (2.7177) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [29] [160/312] eta: 0:01:55 lr: 0.003995 min_lr: 0.003995 loss: 2.8682 (2.7188) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [29] [170/312] eta: 0:01:47 lr: 0.003995 min_lr: 0.003995 loss: 3.0372 (2.7357) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [29] [180/312] eta: 0:01:39 lr: 0.003995 min_lr: 0.003995 loss: 2.9623 (2.7348) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [29] [190/312] eta: 0:01:31 lr: 0.003995 min_lr: 0.003995 loss: 2.6333 (2.7340) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [29] [200/312] eta: 0:01:23 lr: 0.003995 min_lr: 0.003995 loss: 2.6333 (2.7236) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [29] [210/312] eta: 0:01:15 lr: 0.003995 min_lr: 0.003995 loss: 2.7344 (2.7280) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [29] [220/312] eta: 0:01:08 lr: 0.003995 min_lr: 0.003995 loss: 2.8302 (2.7331) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [29] [230/312] eta: 0:01:00 lr: 0.003995 min_lr: 0.003995 loss: 2.9581 (2.7396) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [29] [240/312] eta: 0:00:53 lr: 0.003995 min_lr: 0.003995 loss: 2.8626 (2.7389) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [29] [250/312] eta: 0:00:45 lr: 0.003995 min_lr: 0.003995 loss: 2.7510 (2.7330) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [29] [260/312] eta: 0:00:38 lr: 0.003995 min_lr: 0.003995 loss: 2.8257 (2.7389) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [29] [270/312] eta: 0:00:30 lr: 0.003995 min_lr: 0.003995 loss: 2.8257 (2.7345) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [29] [280/312] eta: 0:00:23 lr: 0.003995 min_lr: 0.003995 loss: 2.6543 (2.7301) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [29] [290/312] eta: 0:00:16 lr: 0.003995 min_lr: 0.003995 loss: 2.6886 (2.7272) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [29] [300/312] eta: 0:00:08 lr: 0.003995 min_lr: 0.003995 loss: 2.6886 (2.7247) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [29] [310/312] eta: 0:00:01 lr: 0.003995 min_lr: 0.003995 loss: 2.8625 (2.7272) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [29] [311/312] eta: 0:00:00 lr: 0.003995 min_lr: 0.003995 loss: 2.8625 (2.7279) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [29] Total time: 0:03:48 (0.7309 s / it) Averaged stats: lr: 0.003995 min_lr: 0.003995 loss: 2.8625 (2.7459) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 1.0216 (1.0216) acc1: 74.7396 (74.7396) acc5: 91.1458 (91.1458) time: 4.5299 data: 4.3216 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4680 (1.3602) acc1: 64.8438 (67.1680) acc5: 87.5000 (87.5520) time: 0.6546 data: 0.4803 max mem: 64948 Test: Total time: 0:00:06 (0.6784 s / it) * Acc@1 68.266 Acc@5 88.108 loss 1.349 Accuracy of the model on the 50000 test images: 68.3% Max accuracy: 69.14% Test: [0/9] eta: 0:00:44 loss: 7.1549 (7.1549) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.9126 data: 4.7085 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0540 (7.0358) acc1: 0.0000 (0.1280) acc5: 0.0000 (0.5120) time: 0.6971 data: 0.5233 max mem: 64948 Test: Total time: 0:00:06 (0.7064 s / it) * Acc@1 0.100 Acc@5 0.508 loss 7.035 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [30] [ 0/312] eta: 0:52:43 lr: 0.003995 min_lr: 0.003995 loss: 3.1130 (3.1130) weight_decay: 0.0500 (0.0500) time: 10.1384 data: 9.1651 max mem: 64948 Epoch: [30] [ 10/312] eta: 0:08:16 lr: 0.003995 min_lr: 0.003995 loss: 2.9085 (2.8696) weight_decay: 0.0500 (0.0500) time: 1.6429 data: 0.8336 max mem: 64948 Epoch: [30] [ 20/312] eta: 0:05:48 lr: 0.003995 min_lr: 0.003995 loss: 2.8222 (2.7775) weight_decay: 0.0500 (0.0500) time: 0.7448 data: 0.0004 max mem: 64948 Epoch: [30] [ 30/312] eta: 0:04:51 lr: 0.003995 min_lr: 0.003995 loss: 2.7453 (2.7575) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0003 max mem: 64948 Epoch: [30] [ 40/312] eta: 0:04:18 lr: 0.003995 min_lr: 0.003995 loss: 2.9359 (2.7646) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0003 max mem: 64948 Epoch: [30] [ 50/312] eta: 0:03:56 lr: 0.003994 min_lr: 0.003994 loss: 3.0260 (2.8012) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [30] [ 60/312] eta: 0:03:38 lr: 0.003994 min_lr: 0.003994 loss: 3.0287 (2.8169) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [30] [ 70/312] eta: 0:03:24 lr: 0.003994 min_lr: 0.003994 loss: 2.9320 (2.8181) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [30] [ 80/312] eta: 0:03:11 lr: 0.003994 min_lr: 0.003994 loss: 2.7969 (2.8121) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [30] [ 90/312] eta: 0:02:59 lr: 0.003994 min_lr: 0.003994 loss: 2.7969 (2.8117) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [30] [100/312] eta: 0:02:49 lr: 0.003994 min_lr: 0.003994 loss: 2.5683 (2.7816) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [30] [110/312] eta: 0:02:39 lr: 0.003994 min_lr: 0.003994 loss: 2.6712 (2.7955) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [30] [120/312] eta: 0:02:30 lr: 0.003994 min_lr: 0.003994 loss: 2.9537 (2.8003) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [30] [130/312] eta: 0:02:21 lr: 0.003994 min_lr: 0.003994 loss: 2.8451 (2.8031) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [30] [140/312] eta: 0:02:12 lr: 0.003994 min_lr: 0.003994 loss: 2.7906 (2.8055) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [30] [150/312] eta: 0:02:03 lr: 0.003994 min_lr: 0.003994 loss: 2.3997 (2.7719) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [30] [160/312] eta: 0:01:55 lr: 0.003994 min_lr: 0.003994 loss: 2.3272 (2.7604) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [30] [170/312] eta: 0:01:47 lr: 0.003994 min_lr: 0.003994 loss: 2.6977 (2.7629) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [30] [180/312] eta: 0:01:39 lr: 0.003994 min_lr: 0.003994 loss: 2.8483 (2.7703) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [30] [190/312] eta: 0:01:31 lr: 0.003994 min_lr: 0.003994 loss: 2.7324 (2.7652) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [30] [200/312] eta: 0:01:23 lr: 0.003994 min_lr: 0.003994 loss: 2.6871 (2.7658) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [30] [210/312] eta: 0:01:15 lr: 0.003994 min_lr: 0.003994 loss: 2.9254 (2.7686) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [30] [220/312] eta: 0:01:08 lr: 0.003994 min_lr: 0.003994 loss: 2.7033 (2.7629) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [30] [230/312] eta: 0:01:00 lr: 0.003994 min_lr: 0.003994 loss: 2.7033 (2.7616) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [30] [240/312] eta: 0:00:53 lr: 0.003994 min_lr: 0.003994 loss: 2.7780 (2.7571) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [30] [250/312] eta: 0:00:45 lr: 0.003994 min_lr: 0.003994 loss: 2.7726 (2.7580) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [30] [260/312] eta: 0:00:38 lr: 0.003994 min_lr: 0.003994 loss: 2.6436 (2.7517) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [30] [270/312] eta: 0:00:30 lr: 0.003994 min_lr: 0.003994 loss: 2.7669 (2.7581) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [30] [280/312] eta: 0:00:23 lr: 0.003994 min_lr: 0.003994 loss: 2.7032 (2.7487) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [30] [290/312] eta: 0:00:16 lr: 0.003994 min_lr: 0.003994 loss: 2.5621 (2.7496) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [30] [300/312] eta: 0:00:08 lr: 0.003994 min_lr: 0.003994 loss: 2.7682 (2.7516) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [30] [310/312] eta: 0:00:01 lr: 0.003994 min_lr: 0.003994 loss: 2.8635 (2.7592) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [30] [311/312] eta: 0:00:00 lr: 0.003994 min_lr: 0.003994 loss: 2.8635 (2.7596) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [30] Total time: 0:03:48 (0.7311 s / it) Averaged stats: lr: 0.003994 min_lr: 0.003994 loss: 2.8635 (2.7344) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 1.1122 (1.1122) acc1: 74.7396 (74.7396) acc5: 91.6667 (91.6667) time: 4.4828 data: 4.2787 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3918 (1.2992) acc1: 69.7917 (68.8640) acc5: 88.2812 (88.3520) time: 0.6494 data: 0.4755 max mem: 64948 Test: Total time: 0:00:06 (0.6723 s / it) * Acc@1 68.938 Acc@5 88.704 loss 1.305 Accuracy of the model on the 50000 test images: 68.9% Max accuracy: 69.14% Test: [0/9] eta: 0:00:45 loss: 7.1716 (7.1716) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.0157 data: 4.7978 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0619 (7.0497) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.7087 data: 0.5332 max mem: 64948 Test: Total time: 0:00:06 (0.7196 s / it) * Acc@1 0.098 Acc@5 0.506 loss 7.048 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [31] [ 0/312] eta: 0:54:00 lr: 0.003994 min_lr: 0.003994 loss: 2.4572 (2.4572) weight_decay: 0.0500 (0.0500) time: 10.3872 data: 6.7225 max mem: 64948 Epoch: [31] [ 10/312] eta: 0:08:03 lr: 0.003994 min_lr: 0.003994 loss: 2.8991 (2.8819) weight_decay: 0.0500 (0.0500) time: 1.6014 data: 0.6116 max mem: 64948 Epoch: [31] [ 20/312] eta: 0:05:41 lr: 0.003993 min_lr: 0.003993 loss: 2.8517 (2.8088) weight_decay: 0.0500 (0.0500) time: 0.7091 data: 0.0004 max mem: 64948 Epoch: [31] [ 30/312] eta: 0:04:46 lr: 0.003993 min_lr: 0.003993 loss: 2.8517 (2.8281) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [31] [ 40/312] eta: 0:04:15 lr: 0.003993 min_lr: 0.003993 loss: 2.8723 (2.8452) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [31] [ 50/312] eta: 0:03:53 lr: 0.003993 min_lr: 0.003993 loss: 2.7789 (2.7881) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [31] [ 60/312] eta: 0:03:36 lr: 0.003993 min_lr: 0.003993 loss: 2.8430 (2.7968) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [31] [ 70/312] eta: 0:03:22 lr: 0.003993 min_lr: 0.003993 loss: 2.8629 (2.7930) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [31] [ 80/312] eta: 0:03:09 lr: 0.003993 min_lr: 0.003993 loss: 2.6833 (2.7681) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [31] [ 90/312] eta: 0:02:58 lr: 0.003993 min_lr: 0.003993 loss: 2.6833 (2.7650) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [31] [100/312] eta: 0:02:48 lr: 0.003993 min_lr: 0.003993 loss: 2.6630 (2.7448) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0003 max mem: 64948 Epoch: [31] [110/312] eta: 0:02:38 lr: 0.003993 min_lr: 0.003993 loss: 2.6146 (2.7399) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0003 max mem: 64948 Epoch: [31] [120/312] eta: 0:02:29 lr: 0.003993 min_lr: 0.003993 loss: 2.7530 (2.7432) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [31] [130/312] eta: 0:02:20 lr: 0.003993 min_lr: 0.003993 loss: 2.7530 (2.7370) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [31] [140/312] eta: 0:02:11 lr: 0.003993 min_lr: 0.003993 loss: 2.7208 (2.7263) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [31] [150/312] eta: 0:02:03 lr: 0.003993 min_lr: 0.003993 loss: 2.7785 (2.7259) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [31] [160/312] eta: 0:01:55 lr: 0.003993 min_lr: 0.003993 loss: 2.7785 (2.7225) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [31] [170/312] eta: 0:01:47 lr: 0.003993 min_lr: 0.003993 loss: 2.6919 (2.7211) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [31] [180/312] eta: 0:01:39 lr: 0.003993 min_lr: 0.003993 loss: 2.5219 (2.7127) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [31] [190/312] eta: 0:01:31 lr: 0.003993 min_lr: 0.003993 loss: 2.7339 (2.7128) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [31] [200/312] eta: 0:01:23 lr: 0.003993 min_lr: 0.003993 loss: 2.7339 (2.7105) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [31] [210/312] eta: 0:01:15 lr: 0.003993 min_lr: 0.003993 loss: 2.8131 (2.7172) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [31] [220/312] eta: 0:01:08 lr: 0.003993 min_lr: 0.003993 loss: 2.8924 (2.7189) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [31] [230/312] eta: 0:01:00 lr: 0.003993 min_lr: 0.003993 loss: 2.7127 (2.7087) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [31] [240/312] eta: 0:00:53 lr: 0.003993 min_lr: 0.003993 loss: 2.7127 (2.7120) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [31] [250/312] eta: 0:00:45 lr: 0.003993 min_lr: 0.003993 loss: 2.9377 (2.7149) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [31] [260/312] eta: 0:00:38 lr: 0.003993 min_lr: 0.003993 loss: 2.9395 (2.7163) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [31] [270/312] eta: 0:00:30 lr: 0.003992 min_lr: 0.003992 loss: 2.8852 (2.7162) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [31] [280/312] eta: 0:00:23 lr: 0.003992 min_lr: 0.003992 loss: 2.6935 (2.7159) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [31] [290/312] eta: 0:00:16 lr: 0.003992 min_lr: 0.003992 loss: 2.6010 (2.7093) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0008 max mem: 64948 Epoch: [31] [300/312] eta: 0:00:08 lr: 0.003992 min_lr: 0.003992 loss: 2.5873 (2.7100) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [31] [310/312] eta: 0:00:01 lr: 0.003992 min_lr: 0.003992 loss: 2.8976 (2.7139) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [31] [311/312] eta: 0:00:00 lr: 0.003992 min_lr: 0.003992 loss: 2.8976 (2.7129) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [31] Total time: 0:03:47 (0.7298 s / it) Averaged stats: lr: 0.003992 min_lr: 0.003992 loss: 2.8976 (2.7053) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 1.0027 (1.0027) acc1: 77.3438 (77.3438) acc5: 91.6667 (91.6667) time: 4.4992 data: 4.2790 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3895 (1.2688) acc1: 67.9688 (69.6960) acc5: 89.3229 (88.6720) time: 0.6519 data: 0.4755 max mem: 64948 Test: Total time: 0:00:06 (0.6752 s / it) * Acc@1 69.516 Acc@5 89.138 loss 1.266 Accuracy of the model on the 50000 test images: 69.5% Max accuracy: 69.52% Test: [0/9] eta: 0:00:41 loss: 7.1825 (7.1825) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.5941 data: 4.3763 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0654 (7.0636) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.6648 data: 0.4864 max mem: 64948 Test: Total time: 0:00:06 (0.6719 s / it) * Acc@1 0.098 Acc@5 0.512 loss 7.062 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [32] [ 0/312] eta: 0:54:29 lr: 0.003992 min_lr: 0.003992 loss: 2.0187 (2.0187) weight_decay: 0.0500 (0.0500) time: 10.4782 data: 7.2238 max mem: 64948 Epoch: [32] [ 10/312] eta: 0:08:17 lr: 0.003992 min_lr: 0.003992 loss: 2.4524 (2.5024) weight_decay: 0.0500 (0.0500) time: 1.6464 data: 0.6571 max mem: 64948 Epoch: [32] [ 20/312] eta: 0:05:48 lr: 0.003992 min_lr: 0.003992 loss: 2.6285 (2.5858) weight_decay: 0.0500 (0.0500) time: 0.7290 data: 0.0004 max mem: 64948 Epoch: [32] [ 30/312] eta: 0:04:51 lr: 0.003992 min_lr: 0.003992 loss: 2.8305 (2.6632) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [32] [ 40/312] eta: 0:04:18 lr: 0.003992 min_lr: 0.003992 loss: 2.8305 (2.6833) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [32] [ 50/312] eta: 0:03:55 lr: 0.003992 min_lr: 0.003992 loss: 2.7801 (2.7350) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [32] [ 60/312] eta: 0:03:38 lr: 0.003992 min_lr: 0.003992 loss: 2.9227 (2.7344) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [32] [ 70/312] eta: 0:03:24 lr: 0.003992 min_lr: 0.003992 loss: 2.7871 (2.7307) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [32] [ 80/312] eta: 0:03:11 lr: 0.003992 min_lr: 0.003992 loss: 2.8563 (2.7570) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [32] [ 90/312] eta: 0:03:00 lr: 0.003992 min_lr: 0.003992 loss: 2.8850 (2.7635) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0003 max mem: 64948 Epoch: [32] [100/312] eta: 0:02:49 lr: 0.003992 min_lr: 0.003992 loss: 2.7549 (2.7659) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0003 max mem: 64948 Epoch: [32] [110/312] eta: 0:02:39 lr: 0.003992 min_lr: 0.003992 loss: 2.7869 (2.7636) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [32] [120/312] eta: 0:02:30 lr: 0.003992 min_lr: 0.003992 loss: 2.8143 (2.7521) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [32] [130/312] eta: 0:02:21 lr: 0.003992 min_lr: 0.003992 loss: 2.8143 (2.7593) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [32] [140/312] eta: 0:02:12 lr: 0.003992 min_lr: 0.003992 loss: 2.8315 (2.7630) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [32] [150/312] eta: 0:02:03 lr: 0.003992 min_lr: 0.003992 loss: 2.6663 (2.7550) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [32] [160/312] eta: 0:01:55 lr: 0.003992 min_lr: 0.003992 loss: 2.7262 (2.7512) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [32] [170/312] eta: 0:01:47 lr: 0.003992 min_lr: 0.003992 loss: 2.7547 (2.7473) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [32] [180/312] eta: 0:01:39 lr: 0.003992 min_lr: 0.003992 loss: 2.5082 (2.7338) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [32] [190/312] eta: 0:01:31 lr: 0.003992 min_lr: 0.003992 loss: 2.6998 (2.7448) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [32] [200/312] eta: 0:01:23 lr: 0.003991 min_lr: 0.003991 loss: 2.9063 (2.7432) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [32] [210/312] eta: 0:01:15 lr: 0.003991 min_lr: 0.003991 loss: 2.8751 (2.7452) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [32] [220/312] eta: 0:01:08 lr: 0.003991 min_lr: 0.003991 loss: 2.7909 (2.7491) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [32] [230/312] eta: 0:01:00 lr: 0.003991 min_lr: 0.003991 loss: 2.8128 (2.7464) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [32] [240/312] eta: 0:00:53 lr: 0.003991 min_lr: 0.003991 loss: 2.8672 (2.7408) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [32] [250/312] eta: 0:00:45 lr: 0.003991 min_lr: 0.003991 loss: 2.6233 (2.7326) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [32] [260/312] eta: 0:00:38 lr: 0.003991 min_lr: 0.003991 loss: 2.7022 (2.7308) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [32] [270/312] eta: 0:00:30 lr: 0.003991 min_lr: 0.003991 loss: 2.8883 (2.7446) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [32] [280/312] eta: 0:00:23 lr: 0.003991 min_lr: 0.003991 loss: 3.0090 (2.7505) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0006 max mem: 64948 Epoch: [32] [290/312] eta: 0:00:16 lr: 0.003991 min_lr: 0.003991 loss: 2.7509 (2.7391) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0005 max mem: 64948 Epoch: [32] [300/312] eta: 0:00:08 lr: 0.003991 min_lr: 0.003991 loss: 2.7509 (2.7417) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [32] [310/312] eta: 0:00:01 lr: 0.003991 min_lr: 0.003991 loss: 2.8174 (2.7387) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [32] [311/312] eta: 0:00:00 lr: 0.003991 min_lr: 0.003991 loss: 2.8376 (2.7399) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [32] Total time: 0:03:48 (0.7313 s / it) Averaged stats: lr: 0.003991 min_lr: 0.003991 loss: 2.8376 (2.6940) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 1.0200 (1.0200) acc1: 75.5208 (75.5208) acc5: 90.6250 (90.6250) time: 4.7833 data: 4.5734 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3485 (1.2834) acc1: 69.7917 (68.8640) acc5: 88.5417 (88.3840) time: 0.6828 data: 0.5082 max mem: 64948 Test: Total time: 0:00:06 (0.7111 s / it) * Acc@1 69.552 Acc@5 88.908 loss 1.269 Accuracy of the model on the 50000 test images: 69.6% Max accuracy: 69.55% Test: [0/9] eta: 0:00:39 loss: 7.1851 (7.1851) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.3736 data: 4.1705 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0650 (7.0763) acc1: 0.0000 (0.0960) acc5: 0.7812 (0.6080) time: 0.6372 data: 0.4635 max mem: 64948 Test: Total time: 0:00:05 (0.6447 s / it) * Acc@1 0.102 Acc@5 0.506 loss 7.074 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [33] [ 0/312] eta: 0:52:17 lr: 0.003991 min_lr: 0.003991 loss: 2.3757 (2.3757) weight_decay: 0.0500 (0.0500) time: 10.0560 data: 6.9512 max mem: 64948 Epoch: [33] [ 10/312] eta: 0:08:07 lr: 0.003991 min_lr: 0.003991 loss: 2.8991 (2.8993) weight_decay: 0.0500 (0.0500) time: 1.6141 data: 0.6324 max mem: 64948 Epoch: [33] [ 20/312] eta: 0:05:43 lr: 0.003991 min_lr: 0.003991 loss: 2.6587 (2.6465) weight_decay: 0.0500 (0.0500) time: 0.7328 data: 0.0005 max mem: 64948 Epoch: [33] [ 30/312] eta: 0:04:48 lr: 0.003991 min_lr: 0.003991 loss: 2.4507 (2.6399) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [33] [ 40/312] eta: 0:04:16 lr: 0.003991 min_lr: 0.003991 loss: 2.6595 (2.6530) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [33] [ 50/312] eta: 0:03:54 lr: 0.003991 min_lr: 0.003991 loss: 2.7316 (2.6764) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [33] [ 60/312] eta: 0:03:36 lr: 0.003991 min_lr: 0.003991 loss: 2.8081 (2.6912) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [33] [ 70/312] eta: 0:03:22 lr: 0.003991 min_lr: 0.003991 loss: 2.7437 (2.6872) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [33] [ 80/312] eta: 0:03:10 lr: 0.003991 min_lr: 0.003991 loss: 2.6438 (2.6596) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [33] [ 90/312] eta: 0:02:59 lr: 0.003991 min_lr: 0.003991 loss: 2.2526 (2.6348) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [33] [100/312] eta: 0:02:48 lr: 0.003991 min_lr: 0.003991 loss: 2.5334 (2.6470) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [33] [110/312] eta: 0:02:38 lr: 0.003990 min_lr: 0.003990 loss: 2.5039 (2.6311) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [33] [120/312] eta: 0:02:29 lr: 0.003990 min_lr: 0.003990 loss: 2.4717 (2.6425) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [33] [130/312] eta: 0:02:20 lr: 0.003990 min_lr: 0.003990 loss: 2.6788 (2.6414) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [33] [140/312] eta: 0:02:11 lr: 0.003990 min_lr: 0.003990 loss: 2.5735 (2.6366) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [33] [150/312] eta: 0:02:03 lr: 0.003990 min_lr: 0.003990 loss: 2.5868 (2.6331) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [33] [160/312] eta: 0:01:55 lr: 0.003990 min_lr: 0.003990 loss: 2.6350 (2.6359) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [33] [170/312] eta: 0:01:47 lr: 0.003990 min_lr: 0.003990 loss: 2.5744 (2.6321) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [33] [180/312] eta: 0:01:39 lr: 0.003990 min_lr: 0.003990 loss: 2.8211 (2.6453) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [33] [190/312] eta: 0:01:31 lr: 0.003990 min_lr: 0.003990 loss: 2.8211 (2.6482) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [33] [200/312] eta: 0:01:23 lr: 0.003990 min_lr: 0.003990 loss: 2.6265 (2.6443) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [33] [210/312] eta: 0:01:15 lr: 0.003990 min_lr: 0.003990 loss: 2.7531 (2.6521) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [33] [220/312] eta: 0:01:08 lr: 0.003990 min_lr: 0.003990 loss: 2.7785 (2.6548) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [33] [230/312] eta: 0:01:00 lr: 0.003990 min_lr: 0.003990 loss: 2.6935 (2.6519) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [33] [240/312] eta: 0:00:53 lr: 0.003990 min_lr: 0.003990 loss: 2.4195 (2.6531) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [33] [250/312] eta: 0:00:45 lr: 0.003990 min_lr: 0.003990 loss: 2.4195 (2.6496) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [33] [260/312] eta: 0:00:38 lr: 0.003990 min_lr: 0.003990 loss: 2.7752 (2.6533) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [33] [270/312] eta: 0:00:30 lr: 0.003990 min_lr: 0.003990 loss: 2.4251 (2.6396) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [33] [280/312] eta: 0:00:23 lr: 0.003990 min_lr: 0.003990 loss: 2.3701 (2.6422) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0010 max mem: 64948 Epoch: [33] [290/312] eta: 0:00:16 lr: 0.003990 min_lr: 0.003990 loss: 2.7951 (2.6468) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0008 max mem: 64948 Epoch: [33] [300/312] eta: 0:00:08 lr: 0.003990 min_lr: 0.003990 loss: 2.5095 (2.6380) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [33] [310/312] eta: 0:00:01 lr: 0.003990 min_lr: 0.003990 loss: 2.5095 (2.6377) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [33] [311/312] eta: 0:00:00 lr: 0.003990 min_lr: 0.003990 loss: 2.5095 (2.6392) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [33] Total time: 0:03:47 (0.7301 s / it) Averaged stats: lr: 0.003990 min_lr: 0.003990 loss: 2.5095 (2.6809) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.9891 (0.9891) acc1: 76.5625 (76.5625) acc5: 91.1458 (91.1458) time: 4.4811 data: 4.2616 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3426 (1.2826) acc1: 66.1458 (67.2000) acc5: 88.5417 (89.1200) time: 0.6498 data: 0.4736 max mem: 64948 Test: Total time: 0:00:05 (0.6657 s / it) * Acc@1 69.152 Acc@5 89.158 loss 1.267 Accuracy of the model on the 50000 test images: 69.2% Max accuracy: 69.55% Test: [0/9] eta: 0:00:44 loss: 7.1836 (7.1836) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.9635 data: 4.7383 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0658 (7.0896) acc1: 0.0000 (0.0960) acc5: 0.7812 (0.5120) time: 0.7073 data: 0.5266 max mem: 64948 Test: Total time: 0:00:06 (0.7231 s / it) * Acc@1 0.100 Acc@5 0.500 loss 7.087 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [34] [ 0/312] eta: 0:52:10 lr: 0.003990 min_lr: 0.003990 loss: 3.0414 (3.0414) weight_decay: 0.0500 (0.0500) time: 10.0342 data: 5.7690 max mem: 64948 Epoch: [34] [ 10/312] eta: 0:07:56 lr: 0.003990 min_lr: 0.003990 loss: 2.5354 (2.5799) weight_decay: 0.0500 (0.0500) time: 1.5782 data: 0.5249 max mem: 64948 Epoch: [34] [ 20/312] eta: 0:05:37 lr: 0.003989 min_lr: 0.003989 loss: 2.5125 (2.5600) weight_decay: 0.0500 (0.0500) time: 0.7132 data: 0.0004 max mem: 64948 Epoch: [34] [ 30/312] eta: 0:04:44 lr: 0.003989 min_lr: 0.003989 loss: 2.9022 (2.6704) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [34] [ 40/312] eta: 0:04:13 lr: 0.003989 min_lr: 0.003989 loss: 2.8090 (2.6357) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [34] [ 50/312] eta: 0:03:51 lr: 0.003989 min_lr: 0.003989 loss: 2.4864 (2.6168) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [34] [ 60/312] eta: 0:03:35 lr: 0.003989 min_lr: 0.003989 loss: 2.5171 (2.6254) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [34] [ 70/312] eta: 0:03:21 lr: 0.003989 min_lr: 0.003989 loss: 2.7820 (2.6331) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [34] [ 80/312] eta: 0:03:09 lr: 0.003989 min_lr: 0.003989 loss: 2.7976 (2.6658) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [34] [ 90/312] eta: 0:02:58 lr: 0.003989 min_lr: 0.003989 loss: 2.7854 (2.6709) weight_decay: 0.0500 (0.0500) time: 0.7016 data: 0.0004 max mem: 64948 Epoch: [34] [100/312] eta: 0:02:48 lr: 0.003989 min_lr: 0.003989 loss: 2.6619 (2.6800) weight_decay: 0.0500 (0.0500) time: 0.7011 data: 0.0004 max mem: 64948 Epoch: [34] [110/312] eta: 0:02:38 lr: 0.003989 min_lr: 0.003989 loss: 2.8101 (2.6693) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [34] [120/312] eta: 0:02:29 lr: 0.003989 min_lr: 0.003989 loss: 2.6870 (2.6658) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [34] [130/312] eta: 0:02:20 lr: 0.003989 min_lr: 0.003989 loss: 2.7267 (2.6739) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [34] [140/312] eta: 0:02:11 lr: 0.003989 min_lr: 0.003989 loss: 2.7873 (2.6760) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [34] [150/312] eta: 0:02:03 lr: 0.003989 min_lr: 0.003989 loss: 2.7202 (2.6779) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [34] [160/312] eta: 0:01:54 lr: 0.003989 min_lr: 0.003989 loss: 2.5700 (2.6739) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [34] [170/312] eta: 0:01:46 lr: 0.003989 min_lr: 0.003989 loss: 2.5512 (2.6670) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [34] [180/312] eta: 0:01:38 lr: 0.003989 min_lr: 0.003989 loss: 2.7067 (2.6717) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [34] [190/312] eta: 0:01:31 lr: 0.003989 min_lr: 0.003989 loss: 2.7067 (2.6561) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [34] [200/312] eta: 0:01:23 lr: 0.003989 min_lr: 0.003989 loss: 2.6132 (2.6579) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [34] [210/312] eta: 0:01:15 lr: 0.003989 min_lr: 0.003989 loss: 2.8076 (2.6609) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [34] [220/312] eta: 0:01:08 lr: 0.003988 min_lr: 0.003988 loss: 2.4943 (2.6492) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [34] [230/312] eta: 0:01:00 lr: 0.003988 min_lr: 0.003988 loss: 2.5777 (2.6516) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [34] [240/312] eta: 0:00:52 lr: 0.003988 min_lr: 0.003988 loss: 2.8026 (2.6512) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [34] [250/312] eta: 0:00:45 lr: 0.003988 min_lr: 0.003988 loss: 2.8026 (2.6535) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [34] [260/312] eta: 0:00:38 lr: 0.003988 min_lr: 0.003988 loss: 2.8684 (2.6573) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [34] [270/312] eta: 0:00:30 lr: 0.003988 min_lr: 0.003988 loss: 2.7964 (2.6597) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [34] [280/312] eta: 0:00:23 lr: 0.003988 min_lr: 0.003988 loss: 2.7964 (2.6596) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [34] [290/312] eta: 0:00:16 lr: 0.003988 min_lr: 0.003988 loss: 2.7145 (2.6548) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [34] [300/312] eta: 0:00:08 lr: 0.003988 min_lr: 0.003988 loss: 2.6814 (2.6566) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [34] [310/312] eta: 0:00:01 lr: 0.003988 min_lr: 0.003988 loss: 2.7085 (2.6514) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [34] [311/312] eta: 0:00:00 lr: 0.003988 min_lr: 0.003988 loss: 2.7085 (2.6504) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [34] Total time: 0:03:47 (0.7291 s / it) Averaged stats: lr: 0.003988 min_lr: 0.003988 loss: 2.7085 (2.6715) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.8945 (0.8945) acc1: 77.6042 (77.6042) acc5: 93.4896 (93.4896) time: 4.5138 data: 4.3071 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2014 (1.2038) acc1: 71.8750 (70.2400) acc5: 90.6250 (89.1520) time: 0.6528 data: 0.4787 max mem: 64948 Test: Total time: 0:00:06 (0.6723 s / it) * Acc@1 70.188 Acc@5 89.298 loss 1.212 Accuracy of the model on the 50000 test images: 70.2% Max accuracy: 70.19% Test: [0/9] eta: 0:00:40 loss: 7.1803 (7.1803) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.5132 data: 4.3061 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0678 (7.1039) acc1: 0.0000 (0.0960) acc5: 0.7812 (0.5120) time: 0.6527 data: 0.4786 max mem: 64948 Test: Total time: 0:00:05 (0.6610 s / it) * Acc@1 0.100 Acc@5 0.506 loss 7.100 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [35] [ 0/312] eta: 0:53:18 lr: 0.003988 min_lr: 0.003988 loss: 2.3119 (2.3119) weight_decay: 0.0500 (0.0500) time: 10.2515 data: 7.1859 max mem: 64948 Epoch: [35] [ 10/312] eta: 0:08:05 lr: 0.003988 min_lr: 0.003988 loss: 2.5668 (2.5823) weight_decay: 0.0500 (0.0500) time: 1.6075 data: 0.6537 max mem: 64948 Epoch: [35] [ 20/312] eta: 0:05:42 lr: 0.003988 min_lr: 0.003988 loss: 2.5222 (2.5714) weight_decay: 0.0500 (0.0500) time: 0.7190 data: 0.0004 max mem: 64948 Epoch: [35] [ 30/312] eta: 0:04:47 lr: 0.003988 min_lr: 0.003988 loss: 2.7186 (2.6459) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [35] [ 40/312] eta: 0:04:15 lr: 0.003988 min_lr: 0.003988 loss: 2.8216 (2.6313) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [35] [ 50/312] eta: 0:03:53 lr: 0.003988 min_lr: 0.003988 loss: 2.8182 (2.6505) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [35] [ 60/312] eta: 0:03:36 lr: 0.003988 min_lr: 0.003988 loss: 2.8182 (2.6663) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [35] [ 70/312] eta: 0:03:22 lr: 0.003988 min_lr: 0.003988 loss: 2.6266 (2.6420) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [35] [ 80/312] eta: 0:03:09 lr: 0.003988 min_lr: 0.003988 loss: 2.6108 (2.6453) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [35] [ 90/312] eta: 0:02:58 lr: 0.003988 min_lr: 0.003988 loss: 2.7784 (2.6700) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [35] [100/312] eta: 0:02:48 lr: 0.003987 min_lr: 0.003987 loss: 2.8795 (2.6897) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [35] [110/312] eta: 0:02:38 lr: 0.003987 min_lr: 0.003987 loss: 2.7390 (2.6887) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [35] [120/312] eta: 0:02:29 lr: 0.003987 min_lr: 0.003987 loss: 2.6361 (2.6781) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [35] [130/312] eta: 0:02:20 lr: 0.003987 min_lr: 0.003987 loss: 2.6046 (2.6818) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [35] [140/312] eta: 0:02:11 lr: 0.003987 min_lr: 0.003987 loss: 2.7283 (2.6710) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [35] [150/312] eta: 0:02:03 lr: 0.003987 min_lr: 0.003987 loss: 2.7871 (2.6773) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [35] [160/312] eta: 0:01:55 lr: 0.003987 min_lr: 0.003987 loss: 2.7348 (2.6674) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [35] [170/312] eta: 0:01:46 lr: 0.003987 min_lr: 0.003987 loss: 2.4719 (2.6633) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [35] [180/312] eta: 0:01:39 lr: 0.003987 min_lr: 0.003987 loss: 2.7473 (2.6696) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [35] [190/312] eta: 0:01:31 lr: 0.003987 min_lr: 0.003987 loss: 2.7473 (2.6618) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [35] [200/312] eta: 0:01:23 lr: 0.003987 min_lr: 0.003987 loss: 2.6220 (2.6580) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [35] [210/312] eta: 0:01:15 lr: 0.003987 min_lr: 0.003987 loss: 2.8070 (2.6653) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [35] [220/312] eta: 0:01:08 lr: 0.003987 min_lr: 0.003987 loss: 2.7795 (2.6612) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [35] [230/312] eta: 0:01:00 lr: 0.003987 min_lr: 0.003987 loss: 2.7904 (2.6710) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [35] [240/312] eta: 0:00:53 lr: 0.003987 min_lr: 0.003987 loss: 2.7904 (2.6668) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [35] [250/312] eta: 0:00:45 lr: 0.003987 min_lr: 0.003987 loss: 2.6131 (2.6649) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [35] [260/312] eta: 0:00:38 lr: 0.003987 min_lr: 0.003987 loss: 2.5984 (2.6611) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [35] [270/312] eta: 0:00:30 lr: 0.003987 min_lr: 0.003987 loss: 2.5576 (2.6544) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [35] [280/312] eta: 0:00:23 lr: 0.003987 min_lr: 0.003987 loss: 2.7769 (2.6582) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [35] [290/312] eta: 0:00:16 lr: 0.003986 min_lr: 0.003986 loss: 2.8000 (2.6542) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [35] [300/312] eta: 0:00:08 lr: 0.003986 min_lr: 0.003986 loss: 2.6797 (2.6537) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [35] [310/312] eta: 0:00:01 lr: 0.003986 min_lr: 0.003986 loss: 2.7855 (2.6579) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [35] [311/312] eta: 0:00:00 lr: 0.003986 min_lr: 0.003986 loss: 2.7738 (2.6582) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [35] Total time: 0:03:47 (0.7296 s / it) Averaged stats: lr: 0.003986 min_lr: 0.003986 loss: 2.7738 (2.6361) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.9495 (0.9495) acc1: 75.0000 (75.0000) acc5: 92.4479 (92.4479) time: 4.6122 data: 4.3921 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3104 (1.2244) acc1: 69.2708 (69.5360) acc5: 89.8438 (89.0880) time: 0.6643 data: 0.4881 max mem: 64948 Test: Total time: 0:00:06 (0.6881 s / it) * Acc@1 70.472 Acc@5 89.698 loss 1.211 Accuracy of the model on the 50000 test images: 70.5% Max accuracy: 70.47% Test: [0/9] eta: 0:00:40 loss: 7.1701 (7.1701) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.4608 data: 4.2571 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0653 (7.1176) acc1: 0.0000 (0.0960) acc5: 0.7812 (0.5120) time: 0.6469 data: 0.4731 max mem: 64948 Test: Total time: 0:00:05 (0.6544 s / it) * Acc@1 0.106 Acc@5 0.508 loss 7.113 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.11% Epoch: [36] [ 0/312] eta: 0:54:05 lr: 0.003986 min_lr: 0.003986 loss: 2.9856 (2.9856) weight_decay: 0.0500 (0.0500) time: 10.4012 data: 9.0762 max mem: 64948 Epoch: [36] [ 10/312] eta: 0:08:05 lr: 0.003986 min_lr: 0.003986 loss: 2.5950 (2.5881) weight_decay: 0.0500 (0.0500) time: 1.6084 data: 0.8254 max mem: 64948 Epoch: [36] [ 20/312] eta: 0:05:42 lr: 0.003986 min_lr: 0.003986 loss: 2.5802 (2.5709) weight_decay: 0.0500 (0.0500) time: 0.7114 data: 0.0003 max mem: 64948 Epoch: [36] [ 30/312] eta: 0:04:48 lr: 0.003986 min_lr: 0.003986 loss: 2.6861 (2.5590) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0003 max mem: 64948 Epoch: [36] [ 40/312] eta: 0:04:16 lr: 0.003986 min_lr: 0.003986 loss: 2.6861 (2.5775) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0003 max mem: 64948 Epoch: [36] [ 50/312] eta: 0:03:54 lr: 0.003986 min_lr: 0.003986 loss: 2.4981 (2.5541) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [36] [ 60/312] eta: 0:03:36 lr: 0.003986 min_lr: 0.003986 loss: 2.5948 (2.5874) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [36] [ 70/312] eta: 0:03:22 lr: 0.003986 min_lr: 0.003986 loss: 2.8098 (2.6058) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [36] [ 80/312] eta: 0:03:10 lr: 0.003986 min_lr: 0.003986 loss: 2.8098 (2.6112) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [36] [ 90/312] eta: 0:02:58 lr: 0.003986 min_lr: 0.003986 loss: 2.6505 (2.6132) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [36] [100/312] eta: 0:02:48 lr: 0.003986 min_lr: 0.003986 loss: 2.6138 (2.6067) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [36] [110/312] eta: 0:02:38 lr: 0.003986 min_lr: 0.003986 loss: 2.6487 (2.6241) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0003 max mem: 64948 Epoch: [36] [120/312] eta: 0:02:29 lr: 0.003986 min_lr: 0.003986 loss: 2.7605 (2.6334) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [36] [130/312] eta: 0:02:20 lr: 0.003986 min_lr: 0.003986 loss: 2.7629 (2.6327) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [36] [140/312] eta: 0:02:11 lr: 0.003986 min_lr: 0.003986 loss: 2.8365 (2.6398) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [36] [150/312] eta: 0:02:03 lr: 0.003986 min_lr: 0.003986 loss: 2.8065 (2.6434) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [36] [160/312] eta: 0:01:55 lr: 0.003985 min_lr: 0.003985 loss: 2.7243 (2.6402) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [36] [170/312] eta: 0:01:47 lr: 0.003985 min_lr: 0.003985 loss: 2.6199 (2.6364) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [36] [180/312] eta: 0:01:39 lr: 0.003985 min_lr: 0.003985 loss: 2.5877 (2.6330) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [36] [190/312] eta: 0:01:31 lr: 0.003985 min_lr: 0.003985 loss: 2.8954 (2.6464) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [36] [200/312] eta: 0:01:23 lr: 0.003985 min_lr: 0.003985 loss: 2.8692 (2.6495) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [36] [210/312] eta: 0:01:15 lr: 0.003985 min_lr: 0.003985 loss: 2.4565 (2.6438) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [36] [220/312] eta: 0:01:08 lr: 0.003985 min_lr: 0.003985 loss: 2.7203 (2.6488) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [36] [230/312] eta: 0:01:00 lr: 0.003985 min_lr: 0.003985 loss: 2.7314 (2.6529) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [36] [240/312] eta: 0:00:53 lr: 0.003985 min_lr: 0.003985 loss: 2.7314 (2.6531) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [36] [250/312] eta: 0:00:45 lr: 0.003985 min_lr: 0.003985 loss: 2.8191 (2.6557) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [36] [260/312] eta: 0:00:38 lr: 0.003985 min_lr: 0.003985 loss: 2.8489 (2.6635) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [36] [270/312] eta: 0:00:30 lr: 0.003985 min_lr: 0.003985 loss: 2.9255 (2.6740) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [36] [280/312] eta: 0:00:23 lr: 0.003985 min_lr: 0.003985 loss: 2.8315 (2.6793) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [36] [290/312] eta: 0:00:16 lr: 0.003985 min_lr: 0.003985 loss: 2.7682 (2.6719) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0008 max mem: 64948 Epoch: [36] [300/312] eta: 0:00:08 lr: 0.003985 min_lr: 0.003985 loss: 2.5234 (2.6669) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [36] [310/312] eta: 0:00:01 lr: 0.003985 min_lr: 0.003985 loss: 2.7245 (2.6688) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [36] [311/312] eta: 0:00:00 lr: 0.003985 min_lr: 0.003985 loss: 2.6472 (2.6669) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [36] Total time: 0:03:47 (0.7302 s / it) Averaged stats: lr: 0.003985 min_lr: 0.003985 loss: 2.6472 (2.6374) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 1.0047 (1.0047) acc1: 75.7812 (75.7812) acc5: 91.9271 (91.9271) time: 4.7212 data: 4.5057 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4224 (1.2572) acc1: 65.1042 (67.4240) acc5: 88.2812 (88.3520) time: 0.6758 data: 0.5007 max mem: 64948 Test: Total time: 0:00:06 (0.7006 s / it) * Acc@1 69.184 Acc@5 88.910 loss 1.248 Accuracy of the model on the 50000 test images: 69.2% Max accuracy: 70.47% Test: [0/9] eta: 0:00:41 loss: 7.1609 (7.1609) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.6438 data: 4.4238 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0627 (7.1324) acc1: 0.0000 (0.0960) acc5: 0.7812 (0.6080) time: 0.6675 data: 0.4916 max mem: 64948 Test: Total time: 0:00:06 (0.6756 s / it) * Acc@1 0.108 Acc@5 0.512 loss 7.127 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.11% Epoch: [37] [ 0/312] eta: 0:48:42 lr: 0.003985 min_lr: 0.003985 loss: 3.1787 (3.1787) weight_decay: 0.0500 (0.0500) time: 9.3665 data: 7.7472 max mem: 64948 Epoch: [37] [ 10/312] eta: 0:07:44 lr: 0.003985 min_lr: 0.003985 loss: 2.7574 (2.7676) weight_decay: 0.0500 (0.0500) time: 1.5368 data: 0.7047 max mem: 64948 Epoch: [37] [ 20/312] eta: 0:05:31 lr: 0.003984 min_lr: 0.003984 loss: 2.7056 (2.6520) weight_decay: 0.0500 (0.0500) time: 0.7249 data: 0.0004 max mem: 64948 Epoch: [37] [ 30/312] eta: 0:04:40 lr: 0.003984 min_lr: 0.003984 loss: 2.4886 (2.6278) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [37] [ 40/312] eta: 0:04:10 lr: 0.003984 min_lr: 0.003984 loss: 2.4886 (2.5711) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [37] [ 50/312] eta: 0:03:49 lr: 0.003984 min_lr: 0.003984 loss: 2.5779 (2.5648) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [37] [ 60/312] eta: 0:03:33 lr: 0.003984 min_lr: 0.003984 loss: 2.6244 (2.5941) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [37] [ 70/312] eta: 0:03:19 lr: 0.003984 min_lr: 0.003984 loss: 2.7251 (2.5944) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [37] [ 80/312] eta: 0:03:07 lr: 0.003984 min_lr: 0.003984 loss: 2.4930 (2.5763) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [37] [ 90/312] eta: 0:02:56 lr: 0.003984 min_lr: 0.003984 loss: 2.5604 (2.5787) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [37] [100/312] eta: 0:02:46 lr: 0.003984 min_lr: 0.003984 loss: 2.5843 (2.5906) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [37] [110/312] eta: 0:02:37 lr: 0.003984 min_lr: 0.003984 loss: 2.8099 (2.5876) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [37] [120/312] eta: 0:02:28 lr: 0.003984 min_lr: 0.003984 loss: 2.8283 (2.6108) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [37] [130/312] eta: 0:02:19 lr: 0.003984 min_lr: 0.003984 loss: 2.4985 (2.6008) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [37] [140/312] eta: 0:02:10 lr: 0.003984 min_lr: 0.003984 loss: 2.4368 (2.5964) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [37] [150/312] eta: 0:02:02 lr: 0.003984 min_lr: 0.003984 loss: 2.6528 (2.6012) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [37] [160/312] eta: 0:01:54 lr: 0.003984 min_lr: 0.003984 loss: 2.6528 (2.6040) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [37] [170/312] eta: 0:01:46 lr: 0.003984 min_lr: 0.003984 loss: 2.6550 (2.6052) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [37] [180/312] eta: 0:01:38 lr: 0.003984 min_lr: 0.003984 loss: 2.5700 (2.5921) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [37] [190/312] eta: 0:01:30 lr: 0.003983 min_lr: 0.003983 loss: 2.4682 (2.5920) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [37] [200/312] eta: 0:01:23 lr: 0.003983 min_lr: 0.003983 loss: 2.6824 (2.5949) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [37] [210/312] eta: 0:01:15 lr: 0.003983 min_lr: 0.003983 loss: 2.6121 (2.5860) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [37] [220/312] eta: 0:01:07 lr: 0.003983 min_lr: 0.003983 loss: 2.6918 (2.5913) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [37] [230/312] eta: 0:01:00 lr: 0.003983 min_lr: 0.003983 loss: 2.6632 (2.5875) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [37] [240/312] eta: 0:00:52 lr: 0.003983 min_lr: 0.003983 loss: 2.5843 (2.5926) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [37] [250/312] eta: 0:00:45 lr: 0.003983 min_lr: 0.003983 loss: 2.5556 (2.5877) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [37] [260/312] eta: 0:00:37 lr: 0.003983 min_lr: 0.003983 loss: 2.5806 (2.5931) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [37] [270/312] eta: 0:00:30 lr: 0.003983 min_lr: 0.003983 loss: 2.7895 (2.6007) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [37] [280/312] eta: 0:00:23 lr: 0.003983 min_lr: 0.003983 loss: 2.7911 (2.6026) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [37] [290/312] eta: 0:00:15 lr: 0.003983 min_lr: 0.003983 loss: 2.5892 (2.5981) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [37] [300/312] eta: 0:00:08 lr: 0.003983 min_lr: 0.003983 loss: 2.6129 (2.5982) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [37] [310/312] eta: 0:00:01 lr: 0.003983 min_lr: 0.003983 loss: 2.7554 (2.6051) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [37] [311/312] eta: 0:00:00 lr: 0.003983 min_lr: 0.003983 loss: 2.7257 (2.6020) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [37] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.003983 min_lr: 0.003983 loss: 2.7257 (2.6213) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.9111 (0.9111) acc1: 78.9062 (78.9062) acc5: 92.9688 (92.9688) time: 4.7652 data: 4.5601 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3254 (1.1990) acc1: 69.0104 (69.5680) acc5: 89.3229 (88.8960) time: 0.6807 data: 0.5068 max mem: 64948 Test: Total time: 0:00:06 (0.6986 s / it) * Acc@1 71.138 Acc@5 90.044 loss 1.162 Accuracy of the model on the 50000 test images: 71.1% Max accuracy: 71.14% Test: [0/9] eta: 0:00:42 loss: 7.1623 (7.1623) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.7334 data: 4.5281 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0570 (7.1503) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6772 data: 0.5032 max mem: 64948 Test: Total time: 0:00:06 (0.6854 s / it) * Acc@1 0.108 Acc@5 0.504 loss 7.143 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [38] [ 0/312] eta: 0:54:20 lr: 0.003983 min_lr: 0.003983 loss: 1.9030 (1.9030) weight_decay: 0.0500 (0.0500) time: 10.4508 data: 7.7130 max mem: 64948 Epoch: [38] [ 10/312] eta: 0:08:23 lr: 0.003983 min_lr: 0.003983 loss: 2.4832 (2.3794) weight_decay: 0.0500 (0.0500) time: 1.6663 data: 0.7015 max mem: 64948 Epoch: [38] [ 20/312] eta: 0:05:52 lr: 0.003983 min_lr: 0.003983 loss: 2.5267 (2.5028) weight_decay: 0.0500 (0.0500) time: 0.7462 data: 0.0004 max mem: 64948 Epoch: [38] [ 30/312] eta: 0:04:54 lr: 0.003983 min_lr: 0.003983 loss: 2.7211 (2.4957) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0003 max mem: 64948 Epoch: [38] [ 40/312] eta: 0:04:20 lr: 0.003982 min_lr: 0.003982 loss: 2.7095 (2.5291) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [38] [ 50/312] eta: 0:03:57 lr: 0.003982 min_lr: 0.003982 loss: 2.7095 (2.5614) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [38] [ 60/312] eta: 0:03:39 lr: 0.003982 min_lr: 0.003982 loss: 2.7233 (2.5585) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [38] [ 70/312] eta: 0:03:24 lr: 0.003982 min_lr: 0.003982 loss: 2.7343 (2.5759) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [38] [ 80/312] eta: 0:03:12 lr: 0.003982 min_lr: 0.003982 loss: 2.5090 (2.5581) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [38] [ 90/312] eta: 0:03:00 lr: 0.003982 min_lr: 0.003982 loss: 2.5952 (2.5690) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [38] [100/312] eta: 0:02:49 lr: 0.003982 min_lr: 0.003982 loss: 2.6885 (2.5626) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [38] [110/312] eta: 0:02:40 lr: 0.003982 min_lr: 0.003982 loss: 2.6885 (2.5742) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [38] [120/312] eta: 0:02:30 lr: 0.003982 min_lr: 0.003982 loss: 2.6940 (2.5820) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [38] [130/312] eta: 0:02:21 lr: 0.003982 min_lr: 0.003982 loss: 2.4877 (2.5681) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [38] [140/312] eta: 0:02:12 lr: 0.003982 min_lr: 0.003982 loss: 2.4877 (2.5698) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [38] [150/312] eta: 0:02:04 lr: 0.003982 min_lr: 0.003982 loss: 2.8465 (2.5786) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [38] [160/312] eta: 0:01:55 lr: 0.003982 min_lr: 0.003982 loss: 2.7416 (2.5827) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [38] [170/312] eta: 0:01:47 lr: 0.003982 min_lr: 0.003982 loss: 2.5344 (2.5809) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [38] [180/312] eta: 0:01:39 lr: 0.003982 min_lr: 0.003982 loss: 2.7055 (2.5835) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [38] [190/312] eta: 0:01:31 lr: 0.003982 min_lr: 0.003982 loss: 2.7155 (2.5860) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [38] [200/312] eta: 0:01:23 lr: 0.003981 min_lr: 0.003981 loss: 2.7296 (2.5908) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [38] [210/312] eta: 0:01:16 lr: 0.003981 min_lr: 0.003981 loss: 2.7296 (2.5940) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [38] [220/312] eta: 0:01:08 lr: 0.003981 min_lr: 0.003981 loss: 2.7821 (2.6044) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [38] [230/312] eta: 0:01:00 lr: 0.003981 min_lr: 0.003981 loss: 2.6457 (2.6003) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [38] [240/312] eta: 0:00:53 lr: 0.003981 min_lr: 0.003981 loss: 2.5758 (2.5915) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [38] [250/312] eta: 0:00:45 lr: 0.003981 min_lr: 0.003981 loss: 2.3750 (2.5874) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [38] [260/312] eta: 0:00:38 lr: 0.003981 min_lr: 0.003981 loss: 2.4741 (2.5894) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [38] [270/312] eta: 0:00:30 lr: 0.003981 min_lr: 0.003981 loss: 2.5614 (2.5836) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [38] [280/312] eta: 0:00:23 lr: 0.003981 min_lr: 0.003981 loss: 2.3383 (2.5755) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [38] [290/312] eta: 0:00:16 lr: 0.003981 min_lr: 0.003981 loss: 2.4676 (2.5745) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [38] [300/312] eta: 0:00:08 lr: 0.003981 min_lr: 0.003981 loss: 2.4925 (2.5727) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [38] [310/312] eta: 0:00:01 lr: 0.003981 min_lr: 0.003981 loss: 2.5352 (2.5738) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [38] [311/312] eta: 0:00:00 lr: 0.003981 min_lr: 0.003981 loss: 2.5370 (2.5736) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [38] Total time: 0:03:48 (0.7321 s / it) Averaged stats: lr: 0.003981 min_lr: 0.003981 loss: 2.5370 (2.5936) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.9310 (0.9310) acc1: 80.2083 (80.2083) acc5: 92.4479 (92.4479) time: 4.7964 data: 4.5927 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3175 (1.1486) acc1: 70.5729 (71.3280) acc5: 88.5417 (89.8880) time: 0.6842 data: 0.5104 max mem: 64948 Test: Total time: 0:00:06 (0.7094 s / it) * Acc@1 71.566 Acc@5 90.284 loss 1.153 Accuracy of the model on the 50000 test images: 71.6% Max accuracy: 71.57% Test: [0/9] eta: 0:00:41 loss: 7.1604 (7.1604) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.5773 data: 4.3594 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0637 (7.1684) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.6605 data: 0.4845 max mem: 64948 Test: Total time: 0:00:06 (0.6693 s / it) * Acc@1 0.112 Acc@5 0.524 loss 7.160 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.11% Epoch: [39] [ 0/312] eta: 0:49:46 lr: 0.003981 min_lr: 0.003981 loss: 2.8318 (2.8318) weight_decay: 0.0500 (0.0500) time: 9.5726 data: 8.7912 max mem: 64948 Epoch: [39] [ 10/312] eta: 0:07:41 lr: 0.003981 min_lr: 0.003981 loss: 2.4213 (2.5165) weight_decay: 0.0500 (0.0500) time: 1.5277 data: 0.7996 max mem: 64948 Epoch: [39] [ 20/312] eta: 0:05:30 lr: 0.003981 min_lr: 0.003981 loss: 2.3183 (2.4796) weight_decay: 0.0500 (0.0500) time: 0.7082 data: 0.0004 max mem: 64948 Epoch: [39] [ 30/312] eta: 0:04:38 lr: 0.003981 min_lr: 0.003981 loss: 2.3183 (2.4698) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [39] [ 40/312] eta: 0:04:10 lr: 0.003981 min_lr: 0.003981 loss: 2.3568 (2.4631) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0003 max mem: 64948 Epoch: [39] [ 50/312] eta: 0:03:50 lr: 0.003980 min_lr: 0.003980 loss: 2.5162 (2.4781) weight_decay: 0.0500 (0.0500) time: 0.7057 data: 0.0004 max mem: 64948 Epoch: [39] [ 60/312] eta: 0:03:34 lr: 0.003980 min_lr: 0.003980 loss: 2.5162 (2.4897) weight_decay: 0.0500 (0.0500) time: 0.7054 data: 0.0003 max mem: 64948 Epoch: [39] [ 70/312] eta: 0:03:20 lr: 0.003980 min_lr: 0.003980 loss: 2.5743 (2.5084) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0003 max mem: 64948 Epoch: [39] [ 80/312] eta: 0:03:08 lr: 0.003980 min_lr: 0.003980 loss: 2.7307 (2.5042) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [39] [ 90/312] eta: 0:02:57 lr: 0.003980 min_lr: 0.003980 loss: 2.6282 (2.5280) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [39] [100/312] eta: 0:02:47 lr: 0.003980 min_lr: 0.003980 loss: 2.7925 (2.5404) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [39] [110/312] eta: 0:02:37 lr: 0.003980 min_lr: 0.003980 loss: 2.6083 (2.5294) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [39] [120/312] eta: 0:02:28 lr: 0.003980 min_lr: 0.003980 loss: 2.6083 (2.5447) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [39] [130/312] eta: 0:02:19 lr: 0.003980 min_lr: 0.003980 loss: 2.7317 (2.5520) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [39] [140/312] eta: 0:02:11 lr: 0.003980 min_lr: 0.003980 loss: 2.6942 (2.5544) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [39] [150/312] eta: 0:02:02 lr: 0.003980 min_lr: 0.003980 loss: 2.5444 (2.5575) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [39] [160/312] eta: 0:01:54 lr: 0.003980 min_lr: 0.003980 loss: 2.7359 (2.5699) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [39] [170/312] eta: 0:01:46 lr: 0.003980 min_lr: 0.003980 loss: 2.7359 (2.5628) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [39] [180/312] eta: 0:01:38 lr: 0.003980 min_lr: 0.003980 loss: 2.7213 (2.5702) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [39] [190/312] eta: 0:01:30 lr: 0.003980 min_lr: 0.003980 loss: 2.6780 (2.5662) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [39] [200/312] eta: 0:01:23 lr: 0.003979 min_lr: 0.003979 loss: 2.5118 (2.5532) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [39] [210/312] eta: 0:01:15 lr: 0.003979 min_lr: 0.003979 loss: 2.4529 (2.5503) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [39] [220/312] eta: 0:01:07 lr: 0.003979 min_lr: 0.003979 loss: 2.6043 (2.5515) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [39] [230/312] eta: 0:01:00 lr: 0.003979 min_lr: 0.003979 loss: 2.5237 (2.5450) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [39] [240/312] eta: 0:00:52 lr: 0.003979 min_lr: 0.003979 loss: 2.5237 (2.5514) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [39] [250/312] eta: 0:00:45 lr: 0.003979 min_lr: 0.003979 loss: 2.5051 (2.5464) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [39] [260/312] eta: 0:00:37 lr: 0.003979 min_lr: 0.003979 loss: 2.5226 (2.5494) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [39] [270/312] eta: 0:00:30 lr: 0.003979 min_lr: 0.003979 loss: 2.5029 (2.5406) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [39] [280/312] eta: 0:00:23 lr: 0.003979 min_lr: 0.003979 loss: 2.5292 (2.5402) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [39] [290/312] eta: 0:00:15 lr: 0.003979 min_lr: 0.003979 loss: 2.5395 (2.5349) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [39] [300/312] eta: 0:00:08 lr: 0.003979 min_lr: 0.003979 loss: 2.5395 (2.5381) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [39] [310/312] eta: 0:00:01 lr: 0.003979 min_lr: 0.003979 loss: 2.6121 (2.5373) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [39] [311/312] eta: 0:00:00 lr: 0.003979 min_lr: 0.003979 loss: 2.6316 (2.5376) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [39] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.003979 min_lr: 0.003979 loss: 2.6316 (2.5931) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.9288 (0.9288) acc1: 79.9479 (79.9479) acc5: 92.4479 (92.4479) time: 4.5403 data: 4.3206 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2571 (1.1957) acc1: 69.2708 (69.8560) acc5: 89.3229 (89.6960) time: 0.6560 data: 0.4802 max mem: 64948 Test: Total time: 0:00:06 (0.6766 s / it) * Acc@1 71.160 Acc@5 90.080 loss 1.177 Accuracy of the model on the 50000 test images: 71.2% Max accuracy: 71.57% Test: [0/9] eta: 0:00:45 loss: 7.1575 (7.1575) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.0068 data: 4.7962 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.0861 (7.1878) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.4800) time: 0.7076 data: 0.5330 max mem: 64948 Test: Total time: 0:00:06 (0.7149 s / it) * Acc@1 0.114 Acc@5 0.502 loss 7.178 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.11% Epoch: [40] [ 0/312] eta: 0:48:36 lr: 0.003979 min_lr: 0.003979 loss: 2.2867 (2.2867) weight_decay: 0.0500 (0.0500) time: 9.3475 data: 8.5644 max mem: 64948 Epoch: [40] [ 10/312] eta: 0:07:52 lr: 0.003979 min_lr: 0.003979 loss: 2.4573 (2.4503) weight_decay: 0.0500 (0.0500) time: 1.5640 data: 0.8060 max mem: 64948 Epoch: [40] [ 20/312] eta: 0:05:36 lr: 0.003979 min_lr: 0.003979 loss: 2.5582 (2.5278) weight_decay: 0.0500 (0.0500) time: 0.7410 data: 0.0152 max mem: 64948 Epoch: [40] [ 30/312] eta: 0:04:43 lr: 0.003978 min_lr: 0.003978 loss: 2.7181 (2.5616) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [40] [ 40/312] eta: 0:04:12 lr: 0.003978 min_lr: 0.003978 loss: 2.7461 (2.5977) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [40] [ 50/312] eta: 0:03:51 lr: 0.003978 min_lr: 0.003978 loss: 2.7335 (2.6200) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [40] [ 60/312] eta: 0:03:34 lr: 0.003978 min_lr: 0.003978 loss: 2.6560 (2.6006) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [40] [ 70/312] eta: 0:03:20 lr: 0.003978 min_lr: 0.003978 loss: 2.5577 (2.5979) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [40] [ 80/312] eta: 0:03:08 lr: 0.003978 min_lr: 0.003978 loss: 2.6315 (2.5851) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [40] [ 90/312] eta: 0:02:57 lr: 0.003978 min_lr: 0.003978 loss: 2.5715 (2.5770) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [40] [100/312] eta: 0:02:47 lr: 0.003978 min_lr: 0.003978 loss: 2.7559 (2.5954) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [40] [110/312] eta: 0:02:37 lr: 0.003978 min_lr: 0.003978 loss: 2.8366 (2.6119) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [40] [120/312] eta: 0:02:28 lr: 0.003978 min_lr: 0.003978 loss: 2.6722 (2.6104) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [40] [130/312] eta: 0:02:19 lr: 0.003978 min_lr: 0.003978 loss: 2.5054 (2.5984) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [40] [140/312] eta: 0:02:11 lr: 0.003978 min_lr: 0.003978 loss: 2.5830 (2.5987) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [40] [150/312] eta: 0:02:02 lr: 0.003978 min_lr: 0.003978 loss: 2.7747 (2.6004) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [40] [160/312] eta: 0:01:54 lr: 0.003978 min_lr: 0.003978 loss: 2.6099 (2.5969) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [40] [170/312] eta: 0:01:46 lr: 0.003978 min_lr: 0.003978 loss: 2.5990 (2.5969) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [40] [180/312] eta: 0:01:38 lr: 0.003977 min_lr: 0.003977 loss: 2.5941 (2.6007) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [40] [190/312] eta: 0:01:30 lr: 0.003977 min_lr: 0.003977 loss: 2.6509 (2.6054) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [40] [200/312] eta: 0:01:23 lr: 0.003977 min_lr: 0.003977 loss: 2.7510 (2.6130) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [40] [210/312] eta: 0:01:15 lr: 0.003977 min_lr: 0.003977 loss: 2.7510 (2.6116) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [40] [220/312] eta: 0:01:07 lr: 0.003977 min_lr: 0.003977 loss: 2.7049 (2.6106) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [40] [230/312] eta: 0:01:00 lr: 0.003977 min_lr: 0.003977 loss: 2.7085 (2.6072) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [40] [240/312] eta: 0:00:52 lr: 0.003977 min_lr: 0.003977 loss: 2.7085 (2.6144) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [40] [250/312] eta: 0:00:45 lr: 0.003977 min_lr: 0.003977 loss: 2.7739 (2.6155) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [40] [260/312] eta: 0:00:38 lr: 0.003977 min_lr: 0.003977 loss: 2.6641 (2.6158) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [40] [270/312] eta: 0:00:30 lr: 0.003977 min_lr: 0.003977 loss: 2.4109 (2.6061) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [40] [280/312] eta: 0:00:23 lr: 0.003977 min_lr: 0.003977 loss: 2.4771 (2.6063) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0009 max mem: 64948 Epoch: [40] [290/312] eta: 0:00:16 lr: 0.003977 min_lr: 0.003977 loss: 2.7261 (2.6145) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [40] [300/312] eta: 0:00:08 lr: 0.003977 min_lr: 0.003977 loss: 2.7089 (2.6091) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [40] [310/312] eta: 0:00:01 lr: 0.003977 min_lr: 0.003977 loss: 2.4577 (2.6040) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [40] [311/312] eta: 0:00:00 lr: 0.003977 min_lr: 0.003977 loss: 2.3901 (2.6033) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [40] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.003977 min_lr: 0.003977 loss: 2.3901 (2.5873) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.9438 (0.9438) acc1: 79.1667 (79.1667) acc5: 92.1875 (92.1875) time: 4.7287 data: 4.5095 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2374 (1.1554) acc1: 70.3125 (70.6560) acc5: 90.8854 (90.1440) time: 0.6772 data: 0.5011 max mem: 64948 Test: Total time: 0:00:06 (0.6991 s / it) * Acc@1 71.296 Acc@5 90.200 loss 1.138 Accuracy of the model on the 50000 test images: 71.3% Max accuracy: 71.57% Test: [0/9] eta: 0:00:44 loss: 7.1505 (7.1505) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.9441 data: 4.7261 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.1078 (7.2082) acc1: 0.0000 (0.1920) acc5: 0.0000 (0.4800) time: 0.7007 data: 0.5252 max mem: 64948 Test: Total time: 0:00:06 (0.7104 s / it) * Acc@1 0.128 Acc@5 0.522 loss 7.198 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.13% Epoch: [41] [ 0/312] eta: 0:50:25 lr: 0.003977 min_lr: 0.003977 loss: 2.0970 (2.0970) weight_decay: 0.0500 (0.0500) time: 9.6982 data: 8.3150 max mem: 64948 Epoch: [41] [ 10/312] eta: 0:07:44 lr: 0.003976 min_lr: 0.003976 loss: 2.2351 (2.3906) weight_decay: 0.0500 (0.0500) time: 1.5383 data: 0.7562 max mem: 64948 Epoch: [41] [ 20/312] eta: 0:05:32 lr: 0.003976 min_lr: 0.003976 loss: 2.6782 (2.5388) weight_decay: 0.0500 (0.0500) time: 0.7105 data: 0.0003 max mem: 64948 Epoch: [41] [ 30/312] eta: 0:04:40 lr: 0.003976 min_lr: 0.003976 loss: 2.7642 (2.5467) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0003 max mem: 64948 Epoch: [41] [ 40/312] eta: 0:04:10 lr: 0.003976 min_lr: 0.003976 loss: 2.6601 (2.5961) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [41] [ 50/312] eta: 0:03:50 lr: 0.003976 min_lr: 0.003976 loss: 2.5494 (2.5589) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [41] [ 60/312] eta: 0:03:33 lr: 0.003976 min_lr: 0.003976 loss: 2.5322 (2.5860) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [41] [ 70/312] eta: 0:03:20 lr: 0.003976 min_lr: 0.003976 loss: 2.3825 (2.5528) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [41] [ 80/312] eta: 0:03:08 lr: 0.003976 min_lr: 0.003976 loss: 2.4403 (2.5751) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [41] [ 90/312] eta: 0:02:57 lr: 0.003976 min_lr: 0.003976 loss: 2.6661 (2.5785) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [41] [100/312] eta: 0:02:47 lr: 0.003976 min_lr: 0.003976 loss: 2.7052 (2.5804) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [41] [110/312] eta: 0:02:37 lr: 0.003976 min_lr: 0.003976 loss: 2.8022 (2.5918) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [41] [120/312] eta: 0:02:28 lr: 0.003976 min_lr: 0.003976 loss: 2.7520 (2.5960) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [41] [130/312] eta: 0:02:19 lr: 0.003976 min_lr: 0.003976 loss: 2.7417 (2.6077) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [41] [140/312] eta: 0:02:11 lr: 0.003975 min_lr: 0.003975 loss: 2.5382 (2.5887) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [41] [150/312] eta: 0:02:02 lr: 0.003975 min_lr: 0.003975 loss: 2.4174 (2.5930) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [41] [160/312] eta: 0:01:54 lr: 0.003975 min_lr: 0.003975 loss: 2.6509 (2.5900) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [41] [170/312] eta: 0:01:46 lr: 0.003975 min_lr: 0.003975 loss: 2.6509 (2.5964) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [41] [180/312] eta: 0:01:38 lr: 0.003975 min_lr: 0.003975 loss: 2.7096 (2.5966) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [41] [190/312] eta: 0:01:30 lr: 0.003975 min_lr: 0.003975 loss: 2.6006 (2.5865) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [41] [200/312] eta: 0:01:23 lr: 0.003975 min_lr: 0.003975 loss: 2.4640 (2.5814) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [41] [210/312] eta: 0:01:15 lr: 0.003975 min_lr: 0.003975 loss: 2.4705 (2.5731) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [41] [220/312] eta: 0:01:07 lr: 0.003975 min_lr: 0.003975 loss: 2.7308 (2.5782) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [41] [230/312] eta: 0:01:00 lr: 0.003975 min_lr: 0.003975 loss: 2.8107 (2.5916) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [41] [240/312] eta: 0:00:52 lr: 0.003975 min_lr: 0.003975 loss: 2.6913 (2.5908) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [41] [250/312] eta: 0:00:45 lr: 0.003975 min_lr: 0.003975 loss: 2.6208 (2.5924) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [41] [260/312] eta: 0:00:38 lr: 0.003975 min_lr: 0.003975 loss: 2.6733 (2.5972) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [41] [270/312] eta: 0:00:30 lr: 0.003975 min_lr: 0.003975 loss: 2.6560 (2.5955) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [41] [280/312] eta: 0:00:23 lr: 0.003974 min_lr: 0.003974 loss: 2.3982 (2.5880) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [41] [290/312] eta: 0:00:15 lr: 0.003974 min_lr: 0.003974 loss: 2.4554 (2.5869) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [41] [300/312] eta: 0:00:08 lr: 0.003974 min_lr: 0.003974 loss: 2.8356 (2.5927) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [41] [310/312] eta: 0:00:01 lr: 0.003974 min_lr: 0.003974 loss: 2.8604 (2.5980) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [41] [311/312] eta: 0:00:00 lr: 0.003974 min_lr: 0.003974 loss: 2.8604 (2.5993) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [41] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.003974 min_lr: 0.003974 loss: 2.8604 (2.5683) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.9649 (0.9649) acc1: 77.0833 (77.0833) acc5: 91.6667 (91.6667) time: 4.7649 data: 4.5535 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3224 (1.2309) acc1: 68.4896 (69.3120) acc5: 89.3229 (89.6000) time: 0.6807 data: 0.5060 max mem: 64948 Test: Total time: 0:00:06 (0.7043 s / it) * Acc@1 70.942 Acc@5 89.860 loss 1.190 Accuracy of the model on the 50000 test images: 70.9% Max accuracy: 71.57% Test: [0/9] eta: 0:00:45 loss: 7.1397 (7.1397) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.0749 data: 4.8663 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.1317 (7.2296) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.5120) time: 0.7151 data: 0.5408 max mem: 64948 Test: Total time: 0:00:06 (0.7248 s / it) * Acc@1 0.106 Acc@5 0.514 loss 7.219 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [42] [ 0/312] eta: 0:55:59 lr: 0.003974 min_lr: 0.003974 loss: 2.9998 (2.9998) weight_decay: 0.0500 (0.0500) time: 10.7661 data: 8.6785 max mem: 64948 Epoch: [42] [ 10/312] eta: 0:08:14 lr: 0.003974 min_lr: 0.003974 loss: 2.4084 (2.4488) weight_decay: 0.0500 (0.0500) time: 1.6369 data: 0.7893 max mem: 64948 Epoch: [42] [ 20/312] eta: 0:05:46 lr: 0.003974 min_lr: 0.003974 loss: 2.4084 (2.5092) weight_decay: 0.0500 (0.0500) time: 0.7083 data: 0.0004 max mem: 64948 Epoch: [42] [ 30/312] eta: 0:04:49 lr: 0.003974 min_lr: 0.003974 loss: 2.5748 (2.4945) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [42] [ 40/312] eta: 0:04:17 lr: 0.003974 min_lr: 0.003974 loss: 2.5965 (2.5263) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [42] [ 50/312] eta: 0:03:55 lr: 0.003974 min_lr: 0.003974 loss: 2.8204 (2.5924) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [42] [ 60/312] eta: 0:03:37 lr: 0.003974 min_lr: 0.003974 loss: 2.5646 (2.5627) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [42] [ 70/312] eta: 0:03:23 lr: 0.003974 min_lr: 0.003974 loss: 2.4696 (2.5561) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [42] [ 80/312] eta: 0:03:10 lr: 0.003974 min_lr: 0.003974 loss: 2.6777 (2.5726) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [42] [ 90/312] eta: 0:02:59 lr: 0.003974 min_lr: 0.003974 loss: 2.5681 (2.5475) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [42] [100/312] eta: 0:02:49 lr: 0.003973 min_lr: 0.003973 loss: 2.3933 (2.5470) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [42] [110/312] eta: 0:02:39 lr: 0.003973 min_lr: 0.003973 loss: 2.7963 (2.5758) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [42] [120/312] eta: 0:02:29 lr: 0.003973 min_lr: 0.003973 loss: 2.8093 (2.5772) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [42] [130/312] eta: 0:02:20 lr: 0.003973 min_lr: 0.003973 loss: 2.4240 (2.5800) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [42] [140/312] eta: 0:02:12 lr: 0.003973 min_lr: 0.003973 loss: 2.4014 (2.5573) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [42] [150/312] eta: 0:02:03 lr: 0.003973 min_lr: 0.003973 loss: 2.4911 (2.5611) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [42] [160/312] eta: 0:01:55 lr: 0.003973 min_lr: 0.003973 loss: 2.7076 (2.5663) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [42] [170/312] eta: 0:01:47 lr: 0.003973 min_lr: 0.003973 loss: 2.4957 (2.5544) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [42] [180/312] eta: 0:01:39 lr: 0.003973 min_lr: 0.003973 loss: 2.3182 (2.5530) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [42] [190/312] eta: 0:01:31 lr: 0.003973 min_lr: 0.003973 loss: 2.7291 (2.5624) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [42] [200/312] eta: 0:01:23 lr: 0.003973 min_lr: 0.003973 loss: 2.6764 (2.5520) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [42] [210/312] eta: 0:01:15 lr: 0.003973 min_lr: 0.003973 loss: 2.3114 (2.5434) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [42] [220/312] eta: 0:01:08 lr: 0.003973 min_lr: 0.003973 loss: 2.4902 (2.5475) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [42] [230/312] eta: 0:01:00 lr: 0.003972 min_lr: 0.003972 loss: 2.6106 (2.5471) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [42] [240/312] eta: 0:00:53 lr: 0.003972 min_lr: 0.003972 loss: 2.6106 (2.5487) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [42] [250/312] eta: 0:00:45 lr: 0.003972 min_lr: 0.003972 loss: 2.6491 (2.5492) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [42] [260/312] eta: 0:00:38 lr: 0.003972 min_lr: 0.003972 loss: 2.5911 (2.5495) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [42] [270/312] eta: 0:00:30 lr: 0.003972 min_lr: 0.003972 loss: 2.7930 (2.5575) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [42] [280/312] eta: 0:00:23 lr: 0.003972 min_lr: 0.003972 loss: 2.7991 (2.5622) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0009 max mem: 64948 Epoch: [42] [290/312] eta: 0:00:16 lr: 0.003972 min_lr: 0.003972 loss: 2.5226 (2.5553) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [42] [300/312] eta: 0:00:08 lr: 0.003972 min_lr: 0.003972 loss: 2.2289 (2.5430) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [42] [310/312] eta: 0:00:01 lr: 0.003972 min_lr: 0.003972 loss: 2.2289 (2.5417) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [42] [311/312] eta: 0:00:00 lr: 0.003972 min_lr: 0.003972 loss: 2.2686 (2.5417) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0001 max mem: 64948 Epoch: [42] Total time: 0:03:48 (0.7311 s / it) Averaged stats: lr: 0.003972 min_lr: 0.003972 loss: 2.2686 (2.5698) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.9089 (0.9089) acc1: 80.4688 (80.4688) acc5: 93.7500 (93.7500) time: 4.7328 data: 4.5228 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4161 (1.2616) acc1: 68.4896 (70.3040) acc5: 88.5417 (88.6400) time: 0.6771 data: 0.5026 max mem: 64948 Test: Total time: 0:00:06 (0.7025 s / it) * Acc@1 70.752 Acc@5 89.786 loss 1.245 Accuracy of the model on the 50000 test images: 70.8% Max accuracy: 71.57% Test: [0/9] eta: 0:00:45 loss: 7.1301 (7.1301) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.0206 data: 4.8028 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.1556 (7.2501) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.6080) time: 0.7092 data: 0.5337 max mem: 64948 Test: Total time: 0:00:06 (0.7170 s / it) * Acc@1 0.114 Acc@5 0.526 loss 7.240 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [43] [ 0/312] eta: 0:52:32 lr: 0.003972 min_lr: 0.003972 loss: 2.6103 (2.6103) weight_decay: 0.0500 (0.0500) time: 10.1039 data: 8.9078 max mem: 64948 Epoch: [43] [ 10/312] eta: 0:07:56 lr: 0.003972 min_lr: 0.003972 loss: 2.6103 (2.4183) weight_decay: 0.0500 (0.0500) time: 1.5794 data: 0.8102 max mem: 64948 Epoch: [43] [ 20/312] eta: 0:05:38 lr: 0.003972 min_lr: 0.003972 loss: 2.6710 (2.5836) weight_decay: 0.0500 (0.0500) time: 0.7106 data: 0.0004 max mem: 64948 Epoch: [43] [ 30/312] eta: 0:04:44 lr: 0.003972 min_lr: 0.003972 loss: 2.7019 (2.6185) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [43] [ 40/312] eta: 0:04:13 lr: 0.003972 min_lr: 0.003972 loss: 2.6044 (2.5968) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [43] [ 50/312] eta: 0:03:52 lr: 0.003971 min_lr: 0.003971 loss: 2.7184 (2.6168) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [43] [ 60/312] eta: 0:03:35 lr: 0.003971 min_lr: 0.003971 loss: 2.7184 (2.5972) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [43] [ 70/312] eta: 0:03:21 lr: 0.003971 min_lr: 0.003971 loss: 2.3277 (2.5639) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [43] [ 80/312] eta: 0:03:09 lr: 0.003971 min_lr: 0.003971 loss: 2.5692 (2.5768) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [43] [ 90/312] eta: 0:02:58 lr: 0.003971 min_lr: 0.003971 loss: 2.7043 (2.5734) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [43] [100/312] eta: 0:02:48 lr: 0.003971 min_lr: 0.003971 loss: 2.6366 (2.5692) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [43] [110/312] eta: 0:02:38 lr: 0.003971 min_lr: 0.003971 loss: 2.3475 (2.5280) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [43] [120/312] eta: 0:02:29 lr: 0.003971 min_lr: 0.003971 loss: 2.3673 (2.5386) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [43] [130/312] eta: 0:02:20 lr: 0.003971 min_lr: 0.003971 loss: 2.4663 (2.5271) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [43] [140/312] eta: 0:02:11 lr: 0.003971 min_lr: 0.003971 loss: 2.4663 (2.5298) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [43] [150/312] eta: 0:02:03 lr: 0.003971 min_lr: 0.003971 loss: 2.5279 (2.5222) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [43] [160/312] eta: 0:01:54 lr: 0.003971 min_lr: 0.003971 loss: 2.4213 (2.5247) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [43] [170/312] eta: 0:01:46 lr: 0.003970 min_lr: 0.003970 loss: 2.6177 (2.5247) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [43] [180/312] eta: 0:01:38 lr: 0.003970 min_lr: 0.003970 loss: 2.4002 (2.5148) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [43] [190/312] eta: 0:01:31 lr: 0.003970 min_lr: 0.003970 loss: 2.2098 (2.5143) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [43] [200/312] eta: 0:01:23 lr: 0.003970 min_lr: 0.003970 loss: 2.6751 (2.5156) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [43] [210/312] eta: 0:01:15 lr: 0.003970 min_lr: 0.003970 loss: 2.6892 (2.5209) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [43] [220/312] eta: 0:01:08 lr: 0.003970 min_lr: 0.003970 loss: 2.6775 (2.5172) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [43] [230/312] eta: 0:01:00 lr: 0.003970 min_lr: 0.003970 loss: 2.3049 (2.5065) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [43] [240/312] eta: 0:00:52 lr: 0.003970 min_lr: 0.003970 loss: 2.1186 (2.4939) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [43] [250/312] eta: 0:00:45 lr: 0.003970 min_lr: 0.003970 loss: 2.3248 (2.4973) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [43] [260/312] eta: 0:00:38 lr: 0.003970 min_lr: 0.003970 loss: 2.7065 (2.5035) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [43] [270/312] eta: 0:00:30 lr: 0.003970 min_lr: 0.003970 loss: 2.7192 (2.5061) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [43] [280/312] eta: 0:00:23 lr: 0.003970 min_lr: 0.003970 loss: 2.6522 (2.5062) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [43] [290/312] eta: 0:00:16 lr: 0.003970 min_lr: 0.003970 loss: 2.4224 (2.5029) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [43] [300/312] eta: 0:00:08 lr: 0.003969 min_lr: 0.003969 loss: 2.5752 (2.5087) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [43] [310/312] eta: 0:00:01 lr: 0.003969 min_lr: 0.003969 loss: 2.6968 (2.5116) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [43] [311/312] eta: 0:00:00 lr: 0.003969 min_lr: 0.003969 loss: 2.6487 (2.5094) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [43] Total time: 0:03:47 (0.7288 s / it) Averaged stats: lr: 0.003969 min_lr: 0.003969 loss: 2.6487 (2.5358) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.8766 (0.8766) acc1: 79.1667 (79.1667) acc5: 94.0104 (94.0104) time: 4.4919 data: 4.2755 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2390 (1.1792) acc1: 70.8333 (70.5920) acc5: 90.8854 (90.3360) time: 0.6504 data: 0.4751 max mem: 64948 Test: Total time: 0:00:06 (0.6742 s / it) * Acc@1 70.942 Acc@5 89.948 loss 1.193 Accuracy of the model on the 50000 test images: 70.9% Max accuracy: 71.57% Test: [0/9] eta: 0:00:46 loss: 7.1136 (7.1136) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.1886 data: 4.9853 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.1799 (7.2723) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.6080) time: 0.7278 data: 0.5540 max mem: 64948 Test: Total time: 0:00:06 (0.7634 s / it) * Acc@1 0.100 Acc@5 0.534 loss 7.263 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [44] [ 0/312] eta: 0:53:02 lr: 0.003969 min_lr: 0.003969 loss: 2.6021 (2.6021) weight_decay: 0.0500 (0.0500) time: 10.2000 data: 6.8251 max mem: 64948 Epoch: [44] [ 10/312] eta: 0:08:03 lr: 0.003969 min_lr: 0.003969 loss: 2.5461 (2.5326) weight_decay: 0.0500 (0.0500) time: 1.6016 data: 0.6210 max mem: 64948 Epoch: [44] [ 20/312] eta: 0:05:41 lr: 0.003969 min_lr: 0.003969 loss: 2.5461 (2.5723) weight_decay: 0.0500 (0.0500) time: 0.7196 data: 0.0005 max mem: 64948 Epoch: [44] [ 30/312] eta: 0:04:46 lr: 0.003969 min_lr: 0.003969 loss: 2.5686 (2.5796) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [44] [ 40/312] eta: 0:04:15 lr: 0.003969 min_lr: 0.003969 loss: 2.4956 (2.5690) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [44] [ 50/312] eta: 0:03:53 lr: 0.003969 min_lr: 0.003969 loss: 2.7929 (2.6157) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [44] [ 60/312] eta: 0:03:36 lr: 0.003969 min_lr: 0.003969 loss: 2.7929 (2.5946) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [44] [ 70/312] eta: 0:03:22 lr: 0.003969 min_lr: 0.003969 loss: 2.4687 (2.5924) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [44] [ 80/312] eta: 0:03:10 lr: 0.003969 min_lr: 0.003969 loss: 2.3403 (2.5606) weight_decay: 0.0500 (0.0500) time: 0.7012 data: 0.0003 max mem: 64948 Epoch: [44] [ 90/312] eta: 0:02:58 lr: 0.003969 min_lr: 0.003969 loss: 2.4361 (2.5446) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [44] [100/312] eta: 0:02:48 lr: 0.003969 min_lr: 0.003969 loss: 2.5319 (2.5377) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [44] [110/312] eta: 0:02:38 lr: 0.003968 min_lr: 0.003968 loss: 2.5359 (2.5369) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [44] [120/312] eta: 0:02:29 lr: 0.003968 min_lr: 0.003968 loss: 2.6935 (2.5336) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [44] [130/312] eta: 0:02:20 lr: 0.003968 min_lr: 0.003968 loss: 2.6327 (2.5475) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [44] [140/312] eta: 0:02:11 lr: 0.003968 min_lr: 0.003968 loss: 2.5973 (2.5449) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [44] [150/312] eta: 0:02:03 lr: 0.003968 min_lr: 0.003968 loss: 2.5533 (2.5422) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [44] [160/312] eta: 0:01:55 lr: 0.003968 min_lr: 0.003968 loss: 2.4539 (2.5328) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [44] [170/312] eta: 0:01:46 lr: 0.003968 min_lr: 0.003968 loss: 2.6599 (2.5394) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [44] [180/312] eta: 0:01:39 lr: 0.003968 min_lr: 0.003968 loss: 2.6881 (2.5372) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [44] [190/312] eta: 0:01:31 lr: 0.003968 min_lr: 0.003968 loss: 2.5314 (2.5361) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [44] [200/312] eta: 0:01:23 lr: 0.003968 min_lr: 0.003968 loss: 2.5243 (2.5316) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [44] [210/312] eta: 0:01:15 lr: 0.003968 min_lr: 0.003968 loss: 2.5378 (2.5298) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [44] [220/312] eta: 0:01:08 lr: 0.003968 min_lr: 0.003968 loss: 2.5378 (2.5252) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [44] [230/312] eta: 0:01:00 lr: 0.003967 min_lr: 0.003967 loss: 2.4358 (2.5191) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [44] [240/312] eta: 0:00:53 lr: 0.003967 min_lr: 0.003967 loss: 2.5197 (2.5175) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [44] [250/312] eta: 0:00:45 lr: 0.003967 min_lr: 0.003967 loss: 2.6101 (2.5132) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [44] [260/312] eta: 0:00:38 lr: 0.003967 min_lr: 0.003967 loss: 2.6101 (2.5197) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [44] [270/312] eta: 0:00:30 lr: 0.003967 min_lr: 0.003967 loss: 2.7727 (2.5318) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [44] [280/312] eta: 0:00:23 lr: 0.003967 min_lr: 0.003967 loss: 2.7722 (2.5307) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [44] [290/312] eta: 0:00:16 lr: 0.003967 min_lr: 0.003967 loss: 2.5329 (2.5332) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0008 max mem: 64948 Epoch: [44] [300/312] eta: 0:00:08 lr: 0.003967 min_lr: 0.003967 loss: 2.5332 (2.5337) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [44] [310/312] eta: 0:00:01 lr: 0.003967 min_lr: 0.003967 loss: 2.4712 (2.5327) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [44] [311/312] eta: 0:00:00 lr: 0.003967 min_lr: 0.003967 loss: 2.4712 (2.5332) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [44] Total time: 0:03:47 (0.7296 s / it) Averaged stats: lr: 0.003967 min_lr: 0.003967 loss: 2.4712 (2.5352) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.8756 (0.8756) acc1: 79.6875 (79.6875) acc5: 92.7083 (92.7083) time: 4.5125 data: 4.3049 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1374 (1.1086) acc1: 70.8333 (70.9760) acc5: 91.6667 (91.0080) time: 0.6527 data: 0.4784 max mem: 64948 Test: Total time: 0:00:06 (0.6762 s / it) * Acc@1 72.732 Acc@5 91.052 loss 1.084 Accuracy of the model on the 50000 test images: 72.7% Max accuracy: 72.73% Test: [0/9] eta: 0:00:40 loss: 7.0922 (7.0922) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.5104 data: 4.3017 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.2042 (7.2912) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.6080) time: 0.6524 data: 0.4781 max mem: 64948 Test: Total time: 0:00:05 (0.6591 s / it) * Acc@1 0.104 Acc@5 0.540 loss 7.283 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [45] [ 0/312] eta: 0:54:42 lr: 0.003967 min_lr: 0.003967 loss: 1.8510 (1.8510) weight_decay: 0.0500 (0.0500) time: 10.5196 data: 9.7443 max mem: 64948 Epoch: [45] [ 10/312] eta: 0:08:06 lr: 0.003967 min_lr: 0.003967 loss: 2.6268 (2.4851) weight_decay: 0.0500 (0.0500) time: 1.6117 data: 0.8862 max mem: 64948 Epoch: [45] [ 20/312] eta: 0:05:43 lr: 0.003967 min_lr: 0.003967 loss: 2.4275 (2.4455) weight_decay: 0.0500 (0.0500) time: 0.7080 data: 0.0004 max mem: 64948 Epoch: [45] [ 30/312] eta: 0:04:47 lr: 0.003966 min_lr: 0.003966 loss: 2.5945 (2.5456) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [45] [ 40/312] eta: 0:04:16 lr: 0.003966 min_lr: 0.003966 loss: 2.8026 (2.5716) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [45] [ 50/312] eta: 0:03:54 lr: 0.003966 min_lr: 0.003966 loss: 2.6946 (2.6212) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [45] [ 60/312] eta: 0:03:36 lr: 0.003966 min_lr: 0.003966 loss: 2.5822 (2.6038) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [45] [ 70/312] eta: 0:03:22 lr: 0.003966 min_lr: 0.003966 loss: 2.3303 (2.5776) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [45] [ 80/312] eta: 0:03:10 lr: 0.003966 min_lr: 0.003966 loss: 2.3139 (2.5601) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [45] [ 90/312] eta: 0:02:59 lr: 0.003966 min_lr: 0.003966 loss: 2.4940 (2.5597) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0003 max mem: 64948 Epoch: [45] [100/312] eta: 0:02:48 lr: 0.003966 min_lr: 0.003966 loss: 2.6928 (2.5636) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [45] [110/312] eta: 0:02:39 lr: 0.003966 min_lr: 0.003966 loss: 2.6932 (2.5739) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [45] [120/312] eta: 0:02:29 lr: 0.003966 min_lr: 0.003966 loss: 2.6874 (2.5700) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [45] [130/312] eta: 0:02:20 lr: 0.003966 min_lr: 0.003966 loss: 2.4903 (2.5616) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [45] [140/312] eta: 0:02:11 lr: 0.003966 min_lr: 0.003966 loss: 2.4903 (2.5509) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [45] [150/312] eta: 0:02:03 lr: 0.003965 min_lr: 0.003965 loss: 2.6112 (2.5662) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [45] [160/312] eta: 0:01:55 lr: 0.003965 min_lr: 0.003965 loss: 2.7156 (2.5728) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [45] [170/312] eta: 0:01:47 lr: 0.003965 min_lr: 0.003965 loss: 2.3864 (2.5500) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [45] [180/312] eta: 0:01:39 lr: 0.003965 min_lr: 0.003965 loss: 2.1089 (2.5390) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [45] [190/312] eta: 0:01:31 lr: 0.003965 min_lr: 0.003965 loss: 2.5721 (2.5359) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [45] [200/312] eta: 0:01:23 lr: 0.003965 min_lr: 0.003965 loss: 2.6168 (2.5400) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [45] [210/312] eta: 0:01:15 lr: 0.003965 min_lr: 0.003965 loss: 2.6330 (2.5427) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [45] [220/312] eta: 0:01:08 lr: 0.003965 min_lr: 0.003965 loss: 2.4658 (2.5384) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [45] [230/312] eta: 0:01:00 lr: 0.003965 min_lr: 0.003965 loss: 2.4097 (2.5361) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [45] [240/312] eta: 0:00:53 lr: 0.003965 min_lr: 0.003965 loss: 2.4368 (2.5335) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [45] [250/312] eta: 0:00:45 lr: 0.003965 min_lr: 0.003965 loss: 2.5607 (2.5403) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [45] [260/312] eta: 0:00:38 lr: 0.003964 min_lr: 0.003964 loss: 2.5607 (2.5392) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [45] [270/312] eta: 0:00:30 lr: 0.003964 min_lr: 0.003964 loss: 2.2764 (2.5312) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [45] [280/312] eta: 0:00:23 lr: 0.003964 min_lr: 0.003964 loss: 2.5127 (2.5356) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0010 max mem: 64948 Epoch: [45] [290/312] eta: 0:00:16 lr: 0.003964 min_lr: 0.003964 loss: 2.6126 (2.5355) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [45] [300/312] eta: 0:00:08 lr: 0.003964 min_lr: 0.003964 loss: 2.5298 (2.5334) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [45] [310/312] eta: 0:00:01 lr: 0.003964 min_lr: 0.003964 loss: 2.5758 (2.5358) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [45] [311/312] eta: 0:00:00 lr: 0.003964 min_lr: 0.003964 loss: 2.5758 (2.5366) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [45] Total time: 0:03:47 (0.7303 s / it) Averaged stats: lr: 0.003964 min_lr: 0.003964 loss: 2.5758 (2.5332) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:45 loss: 0.9799 (0.9799) acc1: 78.6458 (78.6458) acc5: 91.9271 (91.9271) time: 5.0332 data: 4.8254 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2862 (1.1924) acc1: 69.2708 (70.1760) acc5: 88.5417 (89.5360) time: 0.7105 data: 0.5362 max mem: 64948 Test: Total time: 0:00:06 (0.7260 s / it) * Acc@1 71.508 Acc@5 90.254 loss 1.163 Accuracy of the model on the 50000 test images: 71.5% Max accuracy: 72.73% Test: [0/9] eta: 0:00:45 loss: 7.0629 (7.0629) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.0409 data: 4.8230 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.2286 (7.3135) acc1: 0.0000 (0.0960) acc5: 0.0000 (0.6400) time: 0.7117 data: 0.5360 max mem: 64948 Test: Total time: 0:00:06 (0.7207 s / it) * Acc@1 0.100 Acc@5 0.540 loss 7.308 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [46] [ 0/312] eta: 0:55:10 lr: 0.003964 min_lr: 0.003964 loss: 2.6045 (2.6045) weight_decay: 0.0500 (0.0500) time: 10.6116 data: 8.3966 max mem: 64948 Epoch: [46] [ 10/312] eta: 0:08:10 lr: 0.003964 min_lr: 0.003964 loss: 2.6232 (2.4888) weight_decay: 0.0500 (0.0500) time: 1.6239 data: 0.7638 max mem: 64948 Epoch: [46] [ 20/312] eta: 0:05:44 lr: 0.003964 min_lr: 0.003964 loss: 2.6232 (2.5356) weight_decay: 0.0500 (0.0500) time: 0.7092 data: 0.0004 max mem: 64948 Epoch: [46] [ 30/312] eta: 0:04:49 lr: 0.003964 min_lr: 0.003964 loss: 2.7510 (2.5652) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [46] [ 40/312] eta: 0:04:17 lr: 0.003964 min_lr: 0.003964 loss: 2.6896 (2.5599) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [46] [ 50/312] eta: 0:03:54 lr: 0.003964 min_lr: 0.003964 loss: 2.5712 (2.5881) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [46] [ 60/312] eta: 0:03:37 lr: 0.003964 min_lr: 0.003964 loss: 2.7011 (2.5859) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [46] [ 70/312] eta: 0:03:23 lr: 0.003963 min_lr: 0.003963 loss: 2.6237 (2.5773) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [46] [ 80/312] eta: 0:03:10 lr: 0.003963 min_lr: 0.003963 loss: 2.6196 (2.5832) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [46] [ 90/312] eta: 0:02:59 lr: 0.003963 min_lr: 0.003963 loss: 2.6196 (2.5657) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [46] [100/312] eta: 0:02:48 lr: 0.003963 min_lr: 0.003963 loss: 2.7212 (2.5818) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [46] [110/312] eta: 0:02:39 lr: 0.003963 min_lr: 0.003963 loss: 2.7644 (2.5851) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [46] [120/312] eta: 0:02:29 lr: 0.003963 min_lr: 0.003963 loss: 2.7948 (2.6058) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [46] [130/312] eta: 0:02:20 lr: 0.003963 min_lr: 0.003963 loss: 2.6860 (2.5968) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [46] [140/312] eta: 0:02:12 lr: 0.003963 min_lr: 0.003963 loss: 2.6304 (2.6045) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [46] [150/312] eta: 0:02:03 lr: 0.003963 min_lr: 0.003963 loss: 2.6304 (2.5988) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [46] [160/312] eta: 0:01:55 lr: 0.003963 min_lr: 0.003963 loss: 2.5414 (2.5955) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [46] [170/312] eta: 0:01:47 lr: 0.003963 min_lr: 0.003963 loss: 2.5988 (2.5982) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [46] [180/312] eta: 0:01:39 lr: 0.003962 min_lr: 0.003962 loss: 2.6668 (2.5959) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [46] [190/312] eta: 0:01:31 lr: 0.003962 min_lr: 0.003962 loss: 2.6213 (2.5916) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [46] [200/312] eta: 0:01:23 lr: 0.003962 min_lr: 0.003962 loss: 2.4576 (2.5906) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [46] [210/312] eta: 0:01:15 lr: 0.003962 min_lr: 0.003962 loss: 2.4911 (2.5831) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [46] [220/312] eta: 0:01:08 lr: 0.003962 min_lr: 0.003962 loss: 2.5690 (2.5803) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [46] [230/312] eta: 0:01:00 lr: 0.003962 min_lr: 0.003962 loss: 2.5360 (2.5806) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [46] [240/312] eta: 0:00:53 lr: 0.003962 min_lr: 0.003962 loss: 2.5855 (2.5785) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [46] [250/312] eta: 0:00:45 lr: 0.003962 min_lr: 0.003962 loss: 2.7098 (2.5823) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [46] [260/312] eta: 0:00:38 lr: 0.003962 min_lr: 0.003962 loss: 2.8100 (2.5931) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [46] [270/312] eta: 0:00:30 lr: 0.003962 min_lr: 0.003962 loss: 2.8054 (2.5931) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [46] [280/312] eta: 0:00:23 lr: 0.003962 min_lr: 0.003962 loss: 2.5877 (2.5889) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [46] [290/312] eta: 0:00:16 lr: 0.003961 min_lr: 0.003961 loss: 2.6187 (2.5848) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [46] [300/312] eta: 0:00:08 lr: 0.003961 min_lr: 0.003961 loss: 2.5032 (2.5803) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [46] [310/312] eta: 0:00:01 lr: 0.003961 min_lr: 0.003961 loss: 2.5436 (2.5813) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [46] [311/312] eta: 0:00:00 lr: 0.003961 min_lr: 0.003961 loss: 2.6517 (2.5824) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [46] Total time: 0:03:48 (0.7310 s / it) Averaged stats: lr: 0.003961 min_lr: 0.003961 loss: 2.6517 (2.5272) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.9118 (0.9118) acc1: 78.3854 (78.3854) acc5: 92.9688 (92.9688) time: 4.6015 data: 4.3947 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1929 (1.1661) acc1: 71.6146 (70.4320) acc5: 89.5833 (90.1120) time: 0.6628 data: 0.4884 max mem: 64948 Test: Total time: 0:00:06 (0.6878 s / it) * Acc@1 71.776 Acc@5 90.392 loss 1.133 Accuracy of the model on the 50000 test images: 71.8% Max accuracy: 72.73% Test: [0/9] eta: 0:00:43 loss: 7.0342 (7.0342) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 4.8785 data: 4.6726 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.2508 (7.3314) acc1: 0.0000 (0.1280) acc5: 0.0000 (0.6720) time: 0.6933 data: 0.5193 max mem: 64948 Test: Total time: 0:00:06 (0.7017 s / it) * Acc@1 0.102 Acc@5 0.558 loss 7.329 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [47] [ 0/312] eta: 0:56:26 lr: 0.003961 min_lr: 0.003961 loss: 2.8197 (2.8197) weight_decay: 0.0500 (0.0500) time: 10.8536 data: 7.1715 max mem: 64948 Epoch: [47] [ 10/312] eta: 0:08:15 lr: 0.003961 min_lr: 0.003961 loss: 2.5997 (2.4640) weight_decay: 0.0500 (0.0500) time: 1.6414 data: 0.6523 max mem: 64948 Epoch: [47] [ 20/312] eta: 0:05:48 lr: 0.003961 min_lr: 0.003961 loss: 2.4213 (2.4155) weight_decay: 0.0500 (0.0500) time: 0.7102 data: 0.0004 max mem: 64948 Epoch: [47] [ 30/312] eta: 0:04:50 lr: 0.003961 min_lr: 0.003961 loss: 2.1133 (2.3357) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [47] [ 40/312] eta: 0:04:18 lr: 0.003961 min_lr: 0.003961 loss: 2.2467 (2.3482) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0003 max mem: 64948 Epoch: [47] [ 50/312] eta: 0:03:55 lr: 0.003961 min_lr: 0.003961 loss: 2.4400 (2.3717) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [47] [ 60/312] eta: 0:03:38 lr: 0.003961 min_lr: 0.003961 loss: 2.5034 (2.3790) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [47] [ 70/312] eta: 0:03:23 lr: 0.003961 min_lr: 0.003961 loss: 2.5467 (2.3937) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [47] [ 80/312] eta: 0:03:11 lr: 0.003960 min_lr: 0.003960 loss: 2.4553 (2.4090) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [47] [ 90/312] eta: 0:02:59 lr: 0.003960 min_lr: 0.003960 loss: 2.5275 (2.4209) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [47] [100/312] eta: 0:02:49 lr: 0.003960 min_lr: 0.003960 loss: 2.6020 (2.4185) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [47] [110/312] eta: 0:02:39 lr: 0.003960 min_lr: 0.003960 loss: 2.2357 (2.4085) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [47] [120/312] eta: 0:02:29 lr: 0.003960 min_lr: 0.003960 loss: 2.2708 (2.4066) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [47] [130/312] eta: 0:02:21 lr: 0.003960 min_lr: 0.003960 loss: 2.6243 (2.4312) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [47] [140/312] eta: 0:02:12 lr: 0.003960 min_lr: 0.003960 loss: 2.6243 (2.4416) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [47] [150/312] eta: 0:02:03 lr: 0.003960 min_lr: 0.003960 loss: 2.6054 (2.4461) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [47] [160/312] eta: 0:01:55 lr: 0.003960 min_lr: 0.003960 loss: 2.2014 (2.4306) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [47] [170/312] eta: 0:01:47 lr: 0.003960 min_lr: 0.003960 loss: 2.3585 (2.4443) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [47] [180/312] eta: 0:01:39 lr: 0.003960 min_lr: 0.003960 loss: 2.7254 (2.4550) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [47] [190/312] eta: 0:01:31 lr: 0.003959 min_lr: 0.003959 loss: 2.5884 (2.4596) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [47] [200/312] eta: 0:01:23 lr: 0.003959 min_lr: 0.003959 loss: 2.5600 (2.4637) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [47] [210/312] eta: 0:01:15 lr: 0.003959 min_lr: 0.003959 loss: 2.5171 (2.4660) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [47] [220/312] eta: 0:01:08 lr: 0.003959 min_lr: 0.003959 loss: 2.4869 (2.4662) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [47] [230/312] eta: 0:01:00 lr: 0.003959 min_lr: 0.003959 loss: 2.4869 (2.4655) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [47] [240/312] eta: 0:00:53 lr: 0.003959 min_lr: 0.003959 loss: 2.6777 (2.4655) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [47] [250/312] eta: 0:00:45 lr: 0.003959 min_lr: 0.003959 loss: 2.6798 (2.4754) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [47] [260/312] eta: 0:00:38 lr: 0.003959 min_lr: 0.003959 loss: 2.7469 (2.4775) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [47] [270/312] eta: 0:00:30 lr: 0.003959 min_lr: 0.003959 loss: 2.4569 (2.4719) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [47] [280/312] eta: 0:00:23 lr: 0.003959 min_lr: 0.003959 loss: 2.4569 (2.4712) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0010 max mem: 64948 Epoch: [47] [290/312] eta: 0:00:16 lr: 0.003959 min_lr: 0.003959 loss: 2.5423 (2.4768) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0008 max mem: 64948 Epoch: [47] [300/312] eta: 0:00:08 lr: 0.003958 min_lr: 0.003958 loss: 2.5914 (2.4802) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0002 max mem: 64948 Epoch: [47] [310/312] eta: 0:00:01 lr: 0.003958 min_lr: 0.003958 loss: 2.4417 (2.4745) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [47] [311/312] eta: 0:00:00 lr: 0.003958 min_lr: 0.003958 loss: 2.4417 (2.4761) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [47] Total time: 0:03:48 (0.7311 s / it) Averaged stats: lr: 0.003958 min_lr: 0.003958 loss: 2.4417 (2.5131) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.8032 (0.8032) acc1: 82.0312 (82.0312) acc5: 94.2708 (94.2708) time: 4.5919 data: 4.3805 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1374 (1.1154) acc1: 71.8750 (71.5200) acc5: 93.4896 (91.1360) time: 0.6615 data: 0.4868 max mem: 64948 Test: Total time: 0:00:06 (0.6845 s / it) * Acc@1 72.242 Acc@5 91.158 loss 1.114 Accuracy of the model on the 50000 test images: 72.2% Max accuracy: 72.73% Test: [0/9] eta: 0:00:45 loss: 6.9938 (6.9938) acc1: 0.0000 (0.0000) acc5: 0.0000 (0.0000) time: 5.0898 data: 4.8726 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.2709 (7.3424) acc1: 0.0000 (0.1600) acc5: 0.0000 (0.6720) time: 0.7187 data: 0.5415 max mem: 64948 Test: Total time: 0:00:06 (0.7313 s / it) * Acc@1 0.106 Acc@5 0.582 loss 7.343 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [48] [ 0/312] eta: 0:53:37 lr: 0.003958 min_lr: 0.003958 loss: 2.3193 (2.3193) weight_decay: 0.0500 (0.0500) time: 10.3112 data: 9.0271 max mem: 64948 Epoch: [48] [ 10/312] eta: 0:08:03 lr: 0.003958 min_lr: 0.003958 loss: 2.4847 (2.4702) weight_decay: 0.0500 (0.0500) time: 1.5997 data: 0.8211 max mem: 64948 Epoch: [48] [ 20/312] eta: 0:05:41 lr: 0.003958 min_lr: 0.003958 loss: 2.5663 (2.5012) weight_decay: 0.0500 (0.0500) time: 0.7125 data: 0.0004 max mem: 64948 Epoch: [48] [ 30/312] eta: 0:04:46 lr: 0.003958 min_lr: 0.003958 loss: 2.6308 (2.5659) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [48] [ 40/312] eta: 0:04:15 lr: 0.003958 min_lr: 0.003958 loss: 2.7049 (2.5348) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [48] [ 50/312] eta: 0:03:53 lr: 0.003958 min_lr: 0.003958 loss: 2.5749 (2.5149) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [48] [ 60/312] eta: 0:03:36 lr: 0.003958 min_lr: 0.003958 loss: 2.5749 (2.5090) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [48] [ 70/312] eta: 0:03:22 lr: 0.003958 min_lr: 0.003958 loss: 2.5886 (2.5019) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [48] [ 80/312] eta: 0:03:10 lr: 0.003958 min_lr: 0.003958 loss: 2.2847 (2.4878) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [48] [ 90/312] eta: 0:02:58 lr: 0.003957 min_lr: 0.003957 loss: 2.2723 (2.4652) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [48] [100/312] eta: 0:02:48 lr: 0.003957 min_lr: 0.003957 loss: 2.5424 (2.4742) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [48] [110/312] eta: 0:02:38 lr: 0.003957 min_lr: 0.003957 loss: 2.5476 (2.4728) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [48] [120/312] eta: 0:02:29 lr: 0.003957 min_lr: 0.003957 loss: 2.5351 (2.4889) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [48] [130/312] eta: 0:02:20 lr: 0.003957 min_lr: 0.003957 loss: 2.5249 (2.4776) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [48] [140/312] eta: 0:02:11 lr: 0.003957 min_lr: 0.003957 loss: 2.6514 (2.4936) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [48] [150/312] eta: 0:02:03 lr: 0.003957 min_lr: 0.003957 loss: 2.6868 (2.4962) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [48] [160/312] eta: 0:01:55 lr: 0.003957 min_lr: 0.003957 loss: 2.4097 (2.4869) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [48] [170/312] eta: 0:01:47 lr: 0.003957 min_lr: 0.003957 loss: 2.5053 (2.4996) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [48] [180/312] eta: 0:01:39 lr: 0.003957 min_lr: 0.003957 loss: 2.6842 (2.5057) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [48] [190/312] eta: 0:01:31 lr: 0.003956 min_lr: 0.003956 loss: 2.5830 (2.5083) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [48] [200/312] eta: 0:01:23 lr: 0.003956 min_lr: 0.003956 loss: 2.4952 (2.5094) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [48] [210/312] eta: 0:01:15 lr: 0.003956 min_lr: 0.003956 loss: 2.6220 (2.5149) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [48] [220/312] eta: 0:01:08 lr: 0.003956 min_lr: 0.003956 loss: 2.6082 (2.5127) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [48] [230/312] eta: 0:01:00 lr: 0.003956 min_lr: 0.003956 loss: 2.4737 (2.5115) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [48] [240/312] eta: 0:00:53 lr: 0.003956 min_lr: 0.003956 loss: 2.4737 (2.5106) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [48] [250/312] eta: 0:00:45 lr: 0.003956 min_lr: 0.003956 loss: 2.6258 (2.5151) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [48] [260/312] eta: 0:00:38 lr: 0.003956 min_lr: 0.003956 loss: 2.6258 (2.5157) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [48] [270/312] eta: 0:00:30 lr: 0.003956 min_lr: 0.003956 loss: 2.6356 (2.5182) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [48] [280/312] eta: 0:00:23 lr: 0.003956 min_lr: 0.003956 loss: 2.6356 (2.5181) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [48] [290/312] eta: 0:00:16 lr: 0.003956 min_lr: 0.003956 loss: 2.6440 (2.5184) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [48] [300/312] eta: 0:00:08 lr: 0.003955 min_lr: 0.003955 loss: 2.7015 (2.5223) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [48] [310/312] eta: 0:00:01 lr: 0.003955 min_lr: 0.003955 loss: 2.7180 (2.5220) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [48] [311/312] eta: 0:00:00 lr: 0.003955 min_lr: 0.003955 loss: 2.7035 (2.5216) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [48] Total time: 0:03:47 (0.7300 s / it) Averaged stats: lr: 0.003955 min_lr: 0.003955 loss: 2.7035 (2.5178) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.9178 (0.9178) acc1: 78.9062 (78.9062) acc5: 93.4896 (93.4896) time: 4.6364 data: 4.4242 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2793 (1.1419) acc1: 69.2708 (70.9760) acc5: 89.0625 (90.4960) time: 0.6665 data: 0.4917 max mem: 64948 Test: Total time: 0:00:06 (0.6905 s / it) * Acc@1 72.068 Acc@5 90.684 loss 1.113 Accuracy of the model on the 50000 test images: 72.1% Max accuracy: 72.73% Test: [0/9] eta: 0:00:47 loss: 6.9531 (6.9531) acc1: 0.0000 (0.0000) acc5: 0.2604 (0.2604) time: 5.2637 data: 5.0456 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.3070 (7.3519) acc1: 0.0000 (0.1600) acc5: 0.0000 (0.7040) time: 0.7363 data: 0.5608 max mem: 64948 Test: Total time: 0:00:06 (0.7527 s / it) * Acc@1 0.108 Acc@5 0.620 loss 7.356 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [49] [ 0/312] eta: 0:51:47 lr: 0.003955 min_lr: 0.003955 loss: 2.9116 (2.9116) weight_decay: 0.0500 (0.0500) time: 9.9593 data: 7.3900 max mem: 64948 Epoch: [49] [ 10/312] eta: 0:07:55 lr: 0.003955 min_lr: 0.003955 loss: 2.6679 (2.6363) weight_decay: 0.0500 (0.0500) time: 1.5756 data: 0.6723 max mem: 64948 Epoch: [49] [ 20/312] eta: 0:05:37 lr: 0.003955 min_lr: 0.003955 loss: 2.5001 (2.4575) weight_decay: 0.0500 (0.0500) time: 0.7160 data: 0.0005 max mem: 64948 Epoch: [49] [ 30/312] eta: 0:04:44 lr: 0.003955 min_lr: 0.003955 loss: 2.5001 (2.4663) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [49] [ 40/312] eta: 0:04:13 lr: 0.003955 min_lr: 0.003955 loss: 2.5224 (2.4223) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [49] [ 50/312] eta: 0:03:52 lr: 0.003955 min_lr: 0.003955 loss: 2.3336 (2.4109) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [49] [ 60/312] eta: 0:03:35 lr: 0.003955 min_lr: 0.003955 loss: 2.3188 (2.4096) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [49] [ 70/312] eta: 0:03:21 lr: 0.003955 min_lr: 0.003955 loss: 2.5950 (2.4595) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [49] [ 80/312] eta: 0:03:09 lr: 0.003954 min_lr: 0.003954 loss: 2.6425 (2.4703) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [49] [ 90/312] eta: 0:02:58 lr: 0.003954 min_lr: 0.003954 loss: 2.5937 (2.4864) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [49] [100/312] eta: 0:02:47 lr: 0.003954 min_lr: 0.003954 loss: 2.5739 (2.4926) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [49] [110/312] eta: 0:02:38 lr: 0.003954 min_lr: 0.003954 loss: 2.3983 (2.4703) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [49] [120/312] eta: 0:02:28 lr: 0.003954 min_lr: 0.003954 loss: 2.1841 (2.4658) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [49] [130/312] eta: 0:02:20 lr: 0.003954 min_lr: 0.003954 loss: 2.5241 (2.4759) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [49] [140/312] eta: 0:02:11 lr: 0.003954 min_lr: 0.003954 loss: 2.4883 (2.4677) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [49] [150/312] eta: 0:02:03 lr: 0.003954 min_lr: 0.003954 loss: 2.3173 (2.4715) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [49] [160/312] eta: 0:01:54 lr: 0.003954 min_lr: 0.003954 loss: 2.5981 (2.4787) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [49] [170/312] eta: 0:01:46 lr: 0.003954 min_lr: 0.003954 loss: 2.5981 (2.4726) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [49] [180/312] eta: 0:01:38 lr: 0.003953 min_lr: 0.003953 loss: 2.2761 (2.4666) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [49] [190/312] eta: 0:01:30 lr: 0.003953 min_lr: 0.003953 loss: 2.2761 (2.4635) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [49] [200/312] eta: 0:01:23 lr: 0.003953 min_lr: 0.003953 loss: 2.4970 (2.4727) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [49] [210/312] eta: 0:01:15 lr: 0.003953 min_lr: 0.003953 loss: 2.7403 (2.4844) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [49] [220/312] eta: 0:01:07 lr: 0.003953 min_lr: 0.003953 loss: 2.7001 (2.4833) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [49] [230/312] eta: 0:01:00 lr: 0.003953 min_lr: 0.003953 loss: 2.6764 (2.4887) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [49] [240/312] eta: 0:00:52 lr: 0.003953 min_lr: 0.003953 loss: 2.7177 (2.4970) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [49] [250/312] eta: 0:00:45 lr: 0.003953 min_lr: 0.003953 loss: 2.6760 (2.4955) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [49] [260/312] eta: 0:00:38 lr: 0.003953 min_lr: 0.003953 loss: 2.7187 (2.5005) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [49] [270/312] eta: 0:00:30 lr: 0.003953 min_lr: 0.003953 loss: 2.6440 (2.4981) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [49] [280/312] eta: 0:00:23 lr: 0.003952 min_lr: 0.003952 loss: 2.6330 (2.5038) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [49] [290/312] eta: 0:00:16 lr: 0.003952 min_lr: 0.003952 loss: 2.7126 (2.5086) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [49] [300/312] eta: 0:00:08 lr: 0.003952 min_lr: 0.003952 loss: 2.5518 (2.5009) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [49] [310/312] eta: 0:00:01 lr: 0.003952 min_lr: 0.003952 loss: 2.2058 (2.4975) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [49] [311/312] eta: 0:00:00 lr: 0.003952 min_lr: 0.003952 loss: 2.5359 (2.4983) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [49] Total time: 0:03:47 (0.7286 s / it) Averaged stats: lr: 0.003952 min_lr: 0.003952 loss: 2.5359 (2.4969) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.9425 (0.9425) acc1: 77.8646 (77.8646) acc5: 93.4896 (93.4896) time: 4.5976 data: 4.3780 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2252 (1.1929) acc1: 71.0938 (70.0800) acc5: 89.5833 (89.5360) time: 0.6622 data: 0.4865 max mem: 64948 Test: Total time: 0:00:06 (0.6852 s / it) * Acc@1 71.616 Acc@5 90.184 loss 1.174 Accuracy of the model on the 50000 test images: 71.6% Max accuracy: 72.73% Test: [0/9] eta: 0:00:43 loss: 6.9133 (6.9133) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.8040 data: 4.5967 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.3367 (7.3596) acc1: 0.0000 (0.1600) acc5: 0.0000 (0.7680) time: 0.6850 data: 0.5109 max mem: 64948 Test: Total time: 0:00:06 (0.6954 s / it) * Acc@1 0.106 Acc@5 0.682 loss 7.370 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [50] [ 0/312] eta: 0:50:38 lr: 0.003952 min_lr: 0.003952 loss: 2.3905 (2.3905) weight_decay: 0.0500 (0.0500) time: 9.7383 data: 7.1286 max mem: 64948 Epoch: [50] [ 10/312] eta: 0:07:56 lr: 0.003952 min_lr: 0.003952 loss: 2.5126 (2.4585) weight_decay: 0.0500 (0.0500) time: 1.5790 data: 0.6513 max mem: 64948 Epoch: [50] [ 20/312] eta: 0:05:38 lr: 0.003952 min_lr: 0.003952 loss: 2.5568 (2.5598) weight_decay: 0.0500 (0.0500) time: 0.7292 data: 0.0019 max mem: 64948 Epoch: [50] [ 30/312] eta: 0:04:44 lr: 0.003952 min_lr: 0.003952 loss: 2.6366 (2.5820) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [50] [ 40/312] eta: 0:04:13 lr: 0.003952 min_lr: 0.003952 loss: 2.5305 (2.5504) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [50] [ 50/312] eta: 0:03:52 lr: 0.003952 min_lr: 0.003952 loss: 2.2741 (2.4925) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [50] [ 60/312] eta: 0:03:35 lr: 0.003952 min_lr: 0.003952 loss: 2.4444 (2.5110) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [50] [ 70/312] eta: 0:03:21 lr: 0.003951 min_lr: 0.003951 loss: 2.5737 (2.5034) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [50] [ 80/312] eta: 0:03:09 lr: 0.003951 min_lr: 0.003951 loss: 2.5182 (2.4997) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [50] [ 90/312] eta: 0:02:58 lr: 0.003951 min_lr: 0.003951 loss: 2.3614 (2.4794) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [50] [100/312] eta: 0:02:47 lr: 0.003951 min_lr: 0.003951 loss: 2.5457 (2.4768) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [50] [110/312] eta: 0:02:38 lr: 0.003951 min_lr: 0.003951 loss: 2.5729 (2.4860) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [50] [120/312] eta: 0:02:28 lr: 0.003951 min_lr: 0.003951 loss: 2.6591 (2.4906) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [50] [130/312] eta: 0:02:20 lr: 0.003951 min_lr: 0.003951 loss: 2.4607 (2.4811) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [50] [140/312] eta: 0:02:11 lr: 0.003951 min_lr: 0.003951 loss: 2.4607 (2.4834) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [50] [150/312] eta: 0:02:03 lr: 0.003951 min_lr: 0.003951 loss: 2.7647 (2.4969) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [50] [160/312] eta: 0:01:54 lr: 0.003951 min_lr: 0.003951 loss: 2.5864 (2.5009) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [50] [170/312] eta: 0:01:46 lr: 0.003950 min_lr: 0.003950 loss: 2.5846 (2.4996) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [50] [180/312] eta: 0:01:38 lr: 0.003950 min_lr: 0.003950 loss: 2.5846 (2.4944) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [50] [190/312] eta: 0:01:31 lr: 0.003950 min_lr: 0.003950 loss: 2.2366 (2.4919) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [50] [200/312] eta: 0:01:23 lr: 0.003950 min_lr: 0.003950 loss: 2.6135 (2.4997) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [50] [210/312] eta: 0:01:15 lr: 0.003950 min_lr: 0.003950 loss: 2.6135 (2.4971) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [50] [220/312] eta: 0:01:07 lr: 0.003950 min_lr: 0.003950 loss: 2.5526 (2.5022) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [50] [230/312] eta: 0:01:00 lr: 0.003950 min_lr: 0.003950 loss: 2.5526 (2.5048) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [50] [240/312] eta: 0:00:52 lr: 0.003950 min_lr: 0.003950 loss: 2.5451 (2.5085) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [50] [250/312] eta: 0:00:45 lr: 0.003950 min_lr: 0.003950 loss: 2.5451 (2.5047) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [50] [260/312] eta: 0:00:38 lr: 0.003949 min_lr: 0.003949 loss: 2.6133 (2.5087) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [50] [270/312] eta: 0:00:30 lr: 0.003949 min_lr: 0.003949 loss: 2.6749 (2.5107) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [50] [280/312] eta: 0:00:23 lr: 0.003949 min_lr: 0.003949 loss: 2.6817 (2.5169) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0010 max mem: 64948 Epoch: [50] [290/312] eta: 0:00:16 lr: 0.003949 min_lr: 0.003949 loss: 2.6817 (2.5188) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [50] [300/312] eta: 0:00:08 lr: 0.003949 min_lr: 0.003949 loss: 2.6039 (2.5203) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [50] [310/312] eta: 0:00:01 lr: 0.003949 min_lr: 0.003949 loss: 2.4762 (2.5179) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [50] [311/312] eta: 0:00:00 lr: 0.003949 min_lr: 0.003949 loss: 2.4314 (2.5153) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [50] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.003949 min_lr: 0.003949 loss: 2.4314 (2.4943) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.8576 (0.8576) acc1: 78.6458 (78.6458) acc5: 92.9688 (92.9688) time: 4.7498 data: 4.5465 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1932 (1.1104) acc1: 72.1354 (72.3840) acc5: 91.4062 (91.3600) time: 0.6790 data: 0.5052 max mem: 64948 Test: Total time: 0:00:06 (0.7014 s / it) * Acc@1 72.640 Acc@5 91.228 loss 1.109 Accuracy of the model on the 50000 test images: 72.6% Max accuracy: 72.73% Test: [0/9] eta: 0:00:44 loss: 6.8674 (6.8674) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.9796 data: 4.7618 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.3531 (7.3580) acc1: 0.0000 (0.1280) acc5: 0.2604 (0.8640) time: 0.7052 data: 0.5292 max mem: 64948 Test: Total time: 0:00:06 (0.7160 s / it) * Acc@1 0.106 Acc@5 0.702 loss 7.375 Accuracy of the model EMA on 50000 test images: 0.1% Epoch: [51] [ 0/312] eta: 0:52:08 lr: 0.003949 min_lr: 0.003949 loss: 2.7044 (2.7044) weight_decay: 0.0500 (0.0500) time: 10.0279 data: 7.8637 max mem: 64948 Epoch: [51] [ 10/312] eta: 0:07:55 lr: 0.003949 min_lr: 0.003949 loss: 2.5079 (2.3999) weight_decay: 0.0500 (0.0500) time: 1.5741 data: 0.7154 max mem: 64948 Epoch: [51] [ 20/312] eta: 0:05:37 lr: 0.003949 min_lr: 0.003949 loss: 2.1561 (2.2520) weight_decay: 0.0500 (0.0500) time: 0.7109 data: 0.0004 max mem: 64948 Epoch: [51] [ 30/312] eta: 0:04:43 lr: 0.003949 min_lr: 0.003949 loss: 2.3656 (2.3338) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [51] [ 40/312] eta: 0:04:12 lr: 0.003949 min_lr: 0.003949 loss: 2.4579 (2.3380) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [51] [ 50/312] eta: 0:03:51 lr: 0.003948 min_lr: 0.003948 loss: 2.4273 (2.3566) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [51] [ 60/312] eta: 0:03:34 lr: 0.003948 min_lr: 0.003948 loss: 2.6060 (2.3680) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [51] [ 70/312] eta: 0:03:21 lr: 0.003948 min_lr: 0.003948 loss: 2.5957 (2.3624) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [51] [ 80/312] eta: 0:03:08 lr: 0.003948 min_lr: 0.003948 loss: 2.3065 (2.3599) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [51] [ 90/312] eta: 0:02:57 lr: 0.003948 min_lr: 0.003948 loss: 2.4603 (2.3845) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [51] [100/312] eta: 0:02:47 lr: 0.003948 min_lr: 0.003948 loss: 2.5363 (2.3815) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [51] [110/312] eta: 0:02:37 lr: 0.003948 min_lr: 0.003948 loss: 2.4959 (2.3994) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [51] [120/312] eta: 0:02:28 lr: 0.003948 min_lr: 0.003948 loss: 2.4959 (2.3979) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [51] [130/312] eta: 0:02:19 lr: 0.003948 min_lr: 0.003948 loss: 2.4640 (2.4023) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [51] [140/312] eta: 0:02:11 lr: 0.003947 min_lr: 0.003947 loss: 2.4640 (2.4047) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [51] [150/312] eta: 0:02:03 lr: 0.003947 min_lr: 0.003947 loss: 2.3885 (2.4053) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [51] [160/312] eta: 0:01:54 lr: 0.003947 min_lr: 0.003947 loss: 2.4499 (2.4135) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [51] [170/312] eta: 0:01:46 lr: 0.003947 min_lr: 0.003947 loss: 2.7264 (2.4301) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [51] [180/312] eta: 0:01:38 lr: 0.003947 min_lr: 0.003947 loss: 2.6315 (2.4263) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [51] [190/312] eta: 0:01:30 lr: 0.003947 min_lr: 0.003947 loss: 2.5207 (2.4242) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [51] [200/312] eta: 0:01:23 lr: 0.003947 min_lr: 0.003947 loss: 2.3551 (2.4241) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [51] [210/312] eta: 0:01:15 lr: 0.003947 min_lr: 0.003947 loss: 2.5533 (2.4298) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [51] [220/312] eta: 0:01:07 lr: 0.003947 min_lr: 0.003947 loss: 2.5300 (2.4308) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [51] [230/312] eta: 0:01:00 lr: 0.003946 min_lr: 0.003946 loss: 2.6142 (2.4408) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [51] [240/312] eta: 0:00:52 lr: 0.003946 min_lr: 0.003946 loss: 2.7457 (2.4486) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [51] [250/312] eta: 0:00:45 lr: 0.003946 min_lr: 0.003946 loss: 2.5652 (2.4514) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [51] [260/312] eta: 0:00:38 lr: 0.003946 min_lr: 0.003946 loss: 2.5716 (2.4503) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [51] [270/312] eta: 0:00:30 lr: 0.003946 min_lr: 0.003946 loss: 2.4196 (2.4414) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [51] [280/312] eta: 0:00:23 lr: 0.003946 min_lr: 0.003946 loss: 2.3989 (2.4450) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [51] [290/312] eta: 0:00:16 lr: 0.003946 min_lr: 0.003946 loss: 2.6145 (2.4513) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [51] [300/312] eta: 0:00:08 lr: 0.003946 min_lr: 0.003946 loss: 2.6068 (2.4547) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [51] [310/312] eta: 0:00:01 lr: 0.003946 min_lr: 0.003946 loss: 2.6291 (2.4539) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [51] [311/312] eta: 0:00:00 lr: 0.003946 min_lr: 0.003946 loss: 2.6291 (2.4539) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [51] Total time: 0:03:47 (0.7284 s / it) Averaged stats: lr: 0.003946 min_lr: 0.003946 loss: 2.6291 (2.4917) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.9032 (0.9032) acc1: 80.9896 (80.9896) acc5: 93.2292 (93.2292) time: 4.4352 data: 4.2194 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2678 (1.1371) acc1: 67.9688 (71.0720) acc5: 89.5833 (90.2400) time: 0.6448 data: 0.4689 max mem: 64948 Test: Total time: 0:00:06 (0.6699 s / it) * Acc@1 72.778 Acc@5 91.116 loss 1.098 Accuracy of the model on the 50000 test images: 72.8% Max accuracy: 72.78% Test: [0/9] eta: 0:00:41 loss: 6.8188 (6.8188) acc1: 0.0000 (0.0000) acc5: 0.7812 (0.7812) time: 4.6240 data: 4.4100 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.3632 (7.3466) acc1: 0.0000 (0.1920) acc5: 0.2604 (0.9280) time: 0.6651 data: 0.4901 max mem: 64948 Test: Total time: 0:00:06 (0.6742 s / it) * Acc@1 0.130 Acc@5 0.724 loss 7.372 Accuracy of the model EMA on 50000 test images: 0.1% Max EMA accuracy: 0.13% Epoch: [52] [ 0/312] eta: 0:53:34 lr: 0.003946 min_lr: 0.003946 loss: 2.5143 (2.5143) weight_decay: 0.0500 (0.0500) time: 10.3016 data: 8.0960 max mem: 64948 Epoch: [52] [ 10/312] eta: 0:07:58 lr: 0.003945 min_lr: 0.003945 loss: 2.5211 (2.4308) weight_decay: 0.0500 (0.0500) time: 1.5838 data: 0.7363 max mem: 64948 Epoch: [52] [ 20/312] eta: 0:05:38 lr: 0.003945 min_lr: 0.003945 loss: 2.5118 (2.4074) weight_decay: 0.0500 (0.0500) time: 0.7019 data: 0.0003 max mem: 64948 Epoch: [52] [ 30/312] eta: 0:04:44 lr: 0.003945 min_lr: 0.003945 loss: 2.2532 (2.3713) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [52] [ 40/312] eta: 0:04:13 lr: 0.003945 min_lr: 0.003945 loss: 2.4228 (2.4247) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [52] [ 50/312] eta: 0:03:52 lr: 0.003945 min_lr: 0.003945 loss: 2.5147 (2.4021) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [52] [ 60/312] eta: 0:03:35 lr: 0.003945 min_lr: 0.003945 loss: 2.4057 (2.4243) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [52] [ 70/312] eta: 0:03:21 lr: 0.003945 min_lr: 0.003945 loss: 2.4911 (2.4238) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [52] [ 80/312] eta: 0:03:09 lr: 0.003945 min_lr: 0.003945 loss: 2.3572 (2.4110) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [52] [ 90/312] eta: 0:02:58 lr: 0.003945 min_lr: 0.003945 loss: 2.5427 (2.4319) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [52] [100/312] eta: 0:02:48 lr: 0.003945 min_lr: 0.003945 loss: 2.5450 (2.4329) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [52] [110/312] eta: 0:02:38 lr: 0.003944 min_lr: 0.003944 loss: 2.5104 (2.4309) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [52] [120/312] eta: 0:02:29 lr: 0.003944 min_lr: 0.003944 loss: 2.5379 (2.4250) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [52] [130/312] eta: 0:02:20 lr: 0.003944 min_lr: 0.003944 loss: 2.2343 (2.4200) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [52] [140/312] eta: 0:02:11 lr: 0.003944 min_lr: 0.003944 loss: 2.5318 (2.4271) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [52] [150/312] eta: 0:02:03 lr: 0.003944 min_lr: 0.003944 loss: 2.6973 (2.4280) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [52] [160/312] eta: 0:01:54 lr: 0.003944 min_lr: 0.003944 loss: 2.2867 (2.4212) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [52] [170/312] eta: 0:01:46 lr: 0.003944 min_lr: 0.003944 loss: 2.5059 (2.4283) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [52] [180/312] eta: 0:01:38 lr: 0.003944 min_lr: 0.003944 loss: 2.5059 (2.4265) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [52] [190/312] eta: 0:01:31 lr: 0.003944 min_lr: 0.003944 loss: 2.4034 (2.4363) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [52] [200/312] eta: 0:01:23 lr: 0.003943 min_lr: 0.003943 loss: 2.4007 (2.4308) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [52] [210/312] eta: 0:01:15 lr: 0.003943 min_lr: 0.003943 loss: 2.2996 (2.4311) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [52] [220/312] eta: 0:01:08 lr: 0.003943 min_lr: 0.003943 loss: 2.5692 (2.4486) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [52] [230/312] eta: 0:01:00 lr: 0.003943 min_lr: 0.003943 loss: 2.4848 (2.4407) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [52] [240/312] eta: 0:00:52 lr: 0.003943 min_lr: 0.003943 loss: 2.3366 (2.4381) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [52] [250/312] eta: 0:00:45 lr: 0.003943 min_lr: 0.003943 loss: 2.5925 (2.4440) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [52] [260/312] eta: 0:00:38 lr: 0.003943 min_lr: 0.003943 loss: 2.4899 (2.4466) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [52] [270/312] eta: 0:00:30 lr: 0.003943 min_lr: 0.003943 loss: 2.4492 (2.4476) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [52] [280/312] eta: 0:00:23 lr: 0.003943 min_lr: 0.003943 loss: 2.6431 (2.4535) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [52] [290/312] eta: 0:00:16 lr: 0.003942 min_lr: 0.003942 loss: 2.7876 (2.4618) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [52] [300/312] eta: 0:00:08 lr: 0.003942 min_lr: 0.003942 loss: 2.6561 (2.4590) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [52] [310/312] eta: 0:00:01 lr: 0.003942 min_lr: 0.003942 loss: 2.3802 (2.4580) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [52] [311/312] eta: 0:00:00 lr: 0.003942 min_lr: 0.003942 loss: 2.3386 (2.4553) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [52] Total time: 0:03:47 (0.7289 s / it) Averaged stats: lr: 0.003942 min_lr: 0.003942 loss: 2.3386 (2.4773) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.8353 (0.8353) acc1: 80.7292 (80.7292) acc5: 94.0104 (94.0104) time: 4.5661 data: 4.3522 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1736 (1.0879) acc1: 72.1354 (72.7680) acc5: 91.6667 (90.9760) time: 0.6586 data: 0.4837 max mem: 64948 Test: Total time: 0:00:06 (0.6832 s / it) * Acc@1 73.192 Acc@5 91.444 loss 1.060 Accuracy of the model on the 50000 test images: 73.2% Max accuracy: 73.19% Test: [0/9] eta: 0:00:41 loss: 6.7663 (6.7663) acc1: 0.5208 (0.5208) acc5: 1.0417 (1.0417) time: 4.5841 data: 4.3800 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.3510 (7.3194) acc1: 0.0000 (0.3200) acc5: 0.7812 (1.0560) time: 0.6606 data: 0.4868 max mem: 64948 Test: Total time: 0:00:06 (0.6678 s / it) * Acc@1 0.174 Acc@5 0.818 loss 7.357 Accuracy of the model EMA on 50000 test images: 0.2% Max EMA accuracy: 0.17% Epoch: [53] [ 0/312] eta: 0:47:40 lr: 0.003942 min_lr: 0.003942 loss: 2.7517 (2.7517) weight_decay: 0.0500 (0.0500) time: 9.1674 data: 7.2873 max mem: 64948 Epoch: [53] [ 10/312] eta: 0:07:31 lr: 0.003942 min_lr: 0.003942 loss: 2.6095 (2.4437) weight_decay: 0.0500 (0.0500) time: 1.4943 data: 0.6629 max mem: 64948 Epoch: [53] [ 20/312] eta: 0:05:25 lr: 0.003942 min_lr: 0.003942 loss: 2.2715 (2.3221) weight_decay: 0.0500 (0.0500) time: 0.7112 data: 0.0004 max mem: 64948 Epoch: [53] [ 30/312] eta: 0:04:36 lr: 0.003942 min_lr: 0.003942 loss: 2.2758 (2.3709) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [53] [ 40/312] eta: 0:04:07 lr: 0.003942 min_lr: 0.003942 loss: 2.4797 (2.3934) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [53] [ 50/312] eta: 0:03:47 lr: 0.003942 min_lr: 0.003942 loss: 2.6942 (2.4635) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [53] [ 60/312] eta: 0:03:31 lr: 0.003941 min_lr: 0.003941 loss: 2.6942 (2.4277) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [53] [ 70/312] eta: 0:03:18 lr: 0.003941 min_lr: 0.003941 loss: 2.3651 (2.4385) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [53] [ 80/312] eta: 0:03:06 lr: 0.003941 min_lr: 0.003941 loss: 2.4620 (2.4401) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [53] [ 90/312] eta: 0:02:56 lr: 0.003941 min_lr: 0.003941 loss: 2.5424 (2.4395) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [53] [100/312] eta: 0:02:46 lr: 0.003941 min_lr: 0.003941 loss: 2.4905 (2.4411) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [53] [110/312] eta: 0:02:36 lr: 0.003941 min_lr: 0.003941 loss: 2.5576 (2.4500) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [53] [120/312] eta: 0:02:27 lr: 0.003941 min_lr: 0.003941 loss: 2.5822 (2.4419) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0003 max mem: 64948 Epoch: [53] [130/312] eta: 0:02:18 lr: 0.003941 min_lr: 0.003941 loss: 2.5428 (2.4418) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [53] [140/312] eta: 0:02:10 lr: 0.003941 min_lr: 0.003941 loss: 2.5428 (2.4408) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [53] [150/312] eta: 0:02:02 lr: 0.003940 min_lr: 0.003940 loss: 2.3815 (2.4427) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [53] [160/312] eta: 0:01:54 lr: 0.003940 min_lr: 0.003940 loss: 2.6579 (2.4546) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [53] [170/312] eta: 0:01:46 lr: 0.003940 min_lr: 0.003940 loss: 2.7032 (2.4672) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [53] [180/312] eta: 0:01:38 lr: 0.003940 min_lr: 0.003940 loss: 2.5874 (2.4658) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [53] [190/312] eta: 0:01:30 lr: 0.003940 min_lr: 0.003940 loss: 2.4593 (2.4558) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [53] [200/312] eta: 0:01:22 lr: 0.003940 min_lr: 0.003940 loss: 2.4313 (2.4532) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0004 max mem: 64948 Epoch: [53] [210/312] eta: 0:01:15 lr: 0.003940 min_lr: 0.003940 loss: 2.5542 (2.4585) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [53] [220/312] eta: 0:01:07 lr: 0.003940 min_lr: 0.003940 loss: 2.6332 (2.4648) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [53] [230/312] eta: 0:01:00 lr: 0.003940 min_lr: 0.003940 loss: 2.7449 (2.4733) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [53] [240/312] eta: 0:00:52 lr: 0.003939 min_lr: 0.003939 loss: 2.6897 (2.4842) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [53] [250/312] eta: 0:00:45 lr: 0.003939 min_lr: 0.003939 loss: 2.5483 (2.4777) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [53] [260/312] eta: 0:00:37 lr: 0.003939 min_lr: 0.003939 loss: 2.2533 (2.4667) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [53] [270/312] eta: 0:00:30 lr: 0.003939 min_lr: 0.003939 loss: 2.3262 (2.4672) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [53] [280/312] eta: 0:00:23 lr: 0.003939 min_lr: 0.003939 loss: 2.4933 (2.4576) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [53] [290/312] eta: 0:00:15 lr: 0.003939 min_lr: 0.003939 loss: 2.5026 (2.4631) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [53] [300/312] eta: 0:00:08 lr: 0.003939 min_lr: 0.003939 loss: 2.6965 (2.4696) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [53] [310/312] eta: 0:00:01 lr: 0.003939 min_lr: 0.003939 loss: 2.6328 (2.4721) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [53] [311/312] eta: 0:00:00 lr: 0.003939 min_lr: 0.003939 loss: 2.6328 (2.4739) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [53] Total time: 0:03:46 (0.7264 s / it) Averaged stats: lr: 0.003939 min_lr: 0.003939 loss: 2.6328 (2.4761) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.8638 (0.8638) acc1: 79.1667 (79.1667) acc5: 92.1875 (92.1875) time: 4.7791 data: 4.5591 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1979 (1.1237) acc1: 71.0938 (71.4880) acc5: 90.6250 (91.0400) time: 0.6823 data: 0.5066 max mem: 64948 Test: Total time: 0:00:06 (0.7085 s / it) * Acc@1 72.876 Acc@5 91.200 loss 1.111 Accuracy of the model on the 50000 test images: 72.9% Max accuracy: 73.19% Test: [0/9] eta: 0:00:44 loss: 6.7157 (6.7157) acc1: 0.5208 (0.5208) acc5: 1.0417 (1.0417) time: 4.9044 data: 4.6863 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.3184 (7.2788) acc1: 0.0000 (0.2880) acc5: 0.7812 (1.2160) time: 0.6962 data: 0.5208 max mem: 64948 Test: Total time: 0:00:06 (0.7075 s / it) * Acc@1 0.210 Acc@5 0.944 loss 7.333 Accuracy of the model EMA on 50000 test images: 0.2% Max EMA accuracy: 0.21% Epoch: [54] [ 0/312] eta: 0:45:03 lr: 0.003939 min_lr: 0.003939 loss: 2.9679 (2.9679) weight_decay: 0.0500 (0.0500) time: 8.6666 data: 7.6823 max mem: 64948 Epoch: [54] [ 10/312] eta: 0:07:19 lr: 0.003939 min_lr: 0.003939 loss: 2.4012 (2.3501) weight_decay: 0.0500 (0.0500) time: 1.4552 data: 0.6989 max mem: 64948 Epoch: [54] [ 20/312] eta: 0:05:18 lr: 0.003938 min_lr: 0.003938 loss: 2.4302 (2.4681) weight_decay: 0.0500 (0.0500) time: 0.7137 data: 0.0004 max mem: 64948 Epoch: [54] [ 30/312] eta: 0:04:31 lr: 0.003938 min_lr: 0.003938 loss: 2.6478 (2.4711) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [54] [ 40/312] eta: 0:04:04 lr: 0.003938 min_lr: 0.003938 loss: 2.4916 (2.4525) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [54] [ 50/312] eta: 0:03:44 lr: 0.003938 min_lr: 0.003938 loss: 2.3727 (2.4474) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [54] [ 60/312] eta: 0:03:29 lr: 0.003938 min_lr: 0.003938 loss: 2.3514 (2.4105) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [54] [ 70/312] eta: 0:03:16 lr: 0.003938 min_lr: 0.003938 loss: 2.5862 (2.4397) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [54] [ 80/312] eta: 0:03:05 lr: 0.003938 min_lr: 0.003938 loss: 2.5553 (2.4191) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [54] [ 90/312] eta: 0:02:54 lr: 0.003938 min_lr: 0.003938 loss: 2.2776 (2.4102) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [54] [100/312] eta: 0:02:45 lr: 0.003937 min_lr: 0.003937 loss: 2.5034 (2.4324) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [54] [110/312] eta: 0:02:35 lr: 0.003937 min_lr: 0.003937 loss: 2.6206 (2.4425) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [54] [120/312] eta: 0:02:26 lr: 0.003937 min_lr: 0.003937 loss: 2.5254 (2.4409) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [54] [130/312] eta: 0:02:18 lr: 0.003937 min_lr: 0.003937 loss: 2.5690 (2.4497) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [54] [140/312] eta: 0:02:09 lr: 0.003937 min_lr: 0.003937 loss: 2.4217 (2.4365) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [54] [150/312] eta: 0:02:01 lr: 0.003937 min_lr: 0.003937 loss: 2.5623 (2.4421) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [54] [160/312] eta: 0:01:53 lr: 0.003937 min_lr: 0.003937 loss: 2.5904 (2.4423) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [54] [170/312] eta: 0:01:45 lr: 0.003937 min_lr: 0.003937 loss: 2.5342 (2.4368) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [54] [180/312] eta: 0:01:37 lr: 0.003937 min_lr: 0.003937 loss: 2.2885 (2.4283) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0003 max mem: 64948 Epoch: [54] [190/312] eta: 0:01:30 lr: 0.003936 min_lr: 0.003936 loss: 2.1269 (2.4219) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0003 max mem: 64948 Epoch: [54] [200/312] eta: 0:01:22 lr: 0.003936 min_lr: 0.003936 loss: 2.4441 (2.4259) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [54] [210/312] eta: 0:01:14 lr: 0.003936 min_lr: 0.003936 loss: 2.4630 (2.4238) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [54] [220/312] eta: 0:01:07 lr: 0.003936 min_lr: 0.003936 loss: 2.4520 (2.4241) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [54] [230/312] eta: 0:00:59 lr: 0.003936 min_lr: 0.003936 loss: 2.5000 (2.4264) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [54] [240/312] eta: 0:00:52 lr: 0.003936 min_lr: 0.003936 loss: 2.6258 (2.4371) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [54] [250/312] eta: 0:00:45 lr: 0.003936 min_lr: 0.003936 loss: 2.6257 (2.4363) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [54] [260/312] eta: 0:00:37 lr: 0.003936 min_lr: 0.003936 loss: 2.4451 (2.4395) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [54] [270/312] eta: 0:00:30 lr: 0.003935 min_lr: 0.003935 loss: 2.7061 (2.4485) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [54] [280/312] eta: 0:00:23 lr: 0.003935 min_lr: 0.003935 loss: 2.4361 (2.4408) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0011 max mem: 64948 Epoch: [54] [290/312] eta: 0:00:15 lr: 0.003935 min_lr: 0.003935 loss: 2.4361 (2.4430) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0010 max mem: 64948 Epoch: [54] [300/312] eta: 0:00:08 lr: 0.003935 min_lr: 0.003935 loss: 2.6517 (2.4441) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [54] [310/312] eta: 0:00:01 lr: 0.003935 min_lr: 0.003935 loss: 2.6517 (2.4491) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [54] [311/312] eta: 0:00:00 lr: 0.003935 min_lr: 0.003935 loss: 2.6517 (2.4508) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [54] Total time: 0:03:45 (0.7243 s / it) Averaged stats: lr: 0.003935 min_lr: 0.003935 loss: 2.6517 (2.4732) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.9293 (0.9293) acc1: 80.4688 (80.4688) acc5: 93.4896 (93.4896) time: 4.6119 data: 4.3974 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2792 (1.1703) acc1: 71.0938 (71.7760) acc5: 90.1042 (90.4960) time: 0.6637 data: 0.4887 max mem: 64948 Test: Total time: 0:00:06 (0.6885 s / it) * Acc@1 72.486 Acc@5 90.860 loss 1.162 Accuracy of the model on the 50000 test images: 72.5% Max accuracy: 73.19% Test: [0/9] eta: 0:00:40 loss: 6.6591 (6.6591) acc1: 0.7812 (0.7812) acc5: 1.3021 (1.3021) time: 4.5368 data: 4.3244 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.2519 (7.2140) acc1: 0.2604 (0.4160) acc5: 0.7812 (1.3760) time: 0.6554 data: 0.4806 max mem: 64948 Test: Total time: 0:00:06 (0.6683 s / it) * Acc@1 0.274 Acc@5 1.098 loss 7.293 Accuracy of the model EMA on 50000 test images: 0.3% Max EMA accuracy: 0.27% Epoch: [55] [ 0/312] eta: 0:47:17 lr: 0.003935 min_lr: 0.003935 loss: 2.9852 (2.9852) weight_decay: 0.0500 (0.0500) time: 9.0957 data: 7.5101 max mem: 64948 Epoch: [55] [ 10/312] eta: 0:07:32 lr: 0.003935 min_lr: 0.003935 loss: 2.5375 (2.4871) weight_decay: 0.0500 (0.0500) time: 1.5000 data: 0.6943 max mem: 64948 Epoch: [55] [ 20/312] eta: 0:05:26 lr: 0.003935 min_lr: 0.003935 loss: 2.4174 (2.4356) weight_decay: 0.0500 (0.0500) time: 0.7192 data: 0.0065 max mem: 64948 Epoch: [55] [ 30/312] eta: 0:04:36 lr: 0.003935 min_lr: 0.003935 loss: 2.2547 (2.3266) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [55] [ 40/312] eta: 0:04:08 lr: 0.003935 min_lr: 0.003935 loss: 2.3473 (2.3722) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [55] [ 50/312] eta: 0:03:48 lr: 0.003934 min_lr: 0.003934 loss: 2.4702 (2.3935) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [55] [ 60/312] eta: 0:03:32 lr: 0.003934 min_lr: 0.003934 loss: 2.4677 (2.3931) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [55] [ 70/312] eta: 0:03:18 lr: 0.003934 min_lr: 0.003934 loss: 2.3412 (2.3760) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [55] [ 80/312] eta: 0:03:06 lr: 0.003934 min_lr: 0.003934 loss: 2.2730 (2.3872) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [55] [ 90/312] eta: 0:02:56 lr: 0.003934 min_lr: 0.003934 loss: 2.5412 (2.4090) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [55] [100/312] eta: 0:02:46 lr: 0.003934 min_lr: 0.003934 loss: 2.5193 (2.4250) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [55] [110/312] eta: 0:02:36 lr: 0.003934 min_lr: 0.003934 loss: 2.4305 (2.4240) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [55] [120/312] eta: 0:02:27 lr: 0.003934 min_lr: 0.003934 loss: 2.5093 (2.4229) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [55] [130/312] eta: 0:02:18 lr: 0.003933 min_lr: 0.003933 loss: 2.5706 (2.4397) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [55] [140/312] eta: 0:02:10 lr: 0.003933 min_lr: 0.003933 loss: 2.5190 (2.4408) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [55] [150/312] eta: 0:02:02 lr: 0.003933 min_lr: 0.003933 loss: 2.5190 (2.4473) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [55] [160/312] eta: 0:01:54 lr: 0.003933 min_lr: 0.003933 loss: 2.4828 (2.4444) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [55] [170/312] eta: 0:01:46 lr: 0.003933 min_lr: 0.003933 loss: 2.4884 (2.4496) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [55] [180/312] eta: 0:01:38 lr: 0.003933 min_lr: 0.003933 loss: 2.5075 (2.4445) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [55] [190/312] eta: 0:01:30 lr: 0.003933 min_lr: 0.003933 loss: 2.3580 (2.4426) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [55] [200/312] eta: 0:01:22 lr: 0.003933 min_lr: 0.003933 loss: 2.5487 (2.4448) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [55] [210/312] eta: 0:01:15 lr: 0.003932 min_lr: 0.003932 loss: 2.5487 (2.4460) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [55] [220/312] eta: 0:01:07 lr: 0.003932 min_lr: 0.003932 loss: 2.4712 (2.4382) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [55] [230/312] eta: 0:01:00 lr: 0.003932 min_lr: 0.003932 loss: 2.2777 (2.4365) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [55] [240/312] eta: 0:00:52 lr: 0.003932 min_lr: 0.003932 loss: 2.5740 (2.4407) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [55] [250/312] eta: 0:00:45 lr: 0.003932 min_lr: 0.003932 loss: 2.5069 (2.4444) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [55] [260/312] eta: 0:00:37 lr: 0.003932 min_lr: 0.003932 loss: 2.5857 (2.4509) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [55] [270/312] eta: 0:00:30 lr: 0.003932 min_lr: 0.003932 loss: 2.6422 (2.4521) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [55] [280/312] eta: 0:00:23 lr: 0.003932 min_lr: 0.003932 loss: 2.5795 (2.4555) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [55] [290/312] eta: 0:00:15 lr: 0.003932 min_lr: 0.003932 loss: 2.4017 (2.4478) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [55] [300/312] eta: 0:00:08 lr: 0.003931 min_lr: 0.003931 loss: 2.1525 (2.4398) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [55] [310/312] eta: 0:00:01 lr: 0.003931 min_lr: 0.003931 loss: 2.5088 (2.4433) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [55] [311/312] eta: 0:00:00 lr: 0.003931 min_lr: 0.003931 loss: 2.5088 (2.4406) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [55] Total time: 0:03:46 (0.7257 s / it) Averaged stats: lr: 0.003931 min_lr: 0.003931 loss: 2.5088 (2.4509) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.9617 (0.9617) acc1: 78.6458 (78.6458) acc5: 94.2708 (94.2708) time: 4.5982 data: 4.3937 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2258 (1.1996) acc1: 72.1354 (71.4880) acc5: 89.0625 (90.4640) time: 0.6622 data: 0.4883 max mem: 64948 Test: Total time: 0:00:06 (0.6873 s / it) * Acc@1 71.774 Acc@5 90.576 loss 1.177 Accuracy of the model on the 50000 test images: 71.8% Max accuracy: 73.19% Test: [0/9] eta: 0:00:44 loss: 6.5898 (6.5898) acc1: 0.7812 (0.7812) acc5: 1.8229 (1.8229) time: 4.9503 data: 4.7323 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 7.1244 (7.1194) acc1: 0.2604 (0.5120) acc5: 1.8229 (1.8560) time: 0.7020 data: 0.5259 max mem: 64948 Test: Total time: 0:00:06 (0.7108 s / it) * Acc@1 0.344 Acc@5 1.300 loss 7.233 Accuracy of the model EMA on 50000 test images: 0.3% Max EMA accuracy: 0.34% Epoch: [56] [ 0/312] eta: 0:46:18 lr: 0.003931 min_lr: 0.003931 loss: 2.2876 (2.2876) weight_decay: 0.0500 (0.0500) time: 8.9051 data: 7.5980 max mem: 64948 Epoch: [56] [ 10/312] eta: 0:07:52 lr: 0.003931 min_lr: 0.003931 loss: 2.4926 (2.4166) weight_decay: 0.0500 (0.0500) time: 1.5636 data: 0.7557 max mem: 64948 Epoch: [56] [ 20/312] eta: 0:05:35 lr: 0.003931 min_lr: 0.003931 loss: 2.6007 (2.4567) weight_decay: 0.0500 (0.0500) time: 0.7621 data: 0.0359 max mem: 64948 Epoch: [56] [ 30/312] eta: 0:04:42 lr: 0.003931 min_lr: 0.003931 loss: 2.3918 (2.3843) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [56] [ 40/312] eta: 0:04:12 lr: 0.003931 min_lr: 0.003931 loss: 2.3918 (2.4098) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [56] [ 50/312] eta: 0:03:51 lr: 0.003931 min_lr: 0.003931 loss: 2.6816 (2.4698) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [56] [ 60/312] eta: 0:03:34 lr: 0.003931 min_lr: 0.003931 loss: 2.7045 (2.4448) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [56] [ 70/312] eta: 0:03:20 lr: 0.003930 min_lr: 0.003930 loss: 2.6330 (2.4586) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [56] [ 80/312] eta: 0:03:08 lr: 0.003930 min_lr: 0.003930 loss: 2.3153 (2.4359) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [56] [ 90/312] eta: 0:02:57 lr: 0.003930 min_lr: 0.003930 loss: 2.2320 (2.4207) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [56] [100/312] eta: 0:02:47 lr: 0.003930 min_lr: 0.003930 loss: 2.5750 (2.4415) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [56] [110/312] eta: 0:02:37 lr: 0.003930 min_lr: 0.003930 loss: 2.6156 (2.4417) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [56] [120/312] eta: 0:02:28 lr: 0.003930 min_lr: 0.003930 loss: 2.5840 (2.4512) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [56] [130/312] eta: 0:02:19 lr: 0.003930 min_lr: 0.003930 loss: 2.5831 (2.4548) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [56] [140/312] eta: 0:02:11 lr: 0.003930 min_lr: 0.003930 loss: 2.4792 (2.4535) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [56] [150/312] eta: 0:02:02 lr: 0.003929 min_lr: 0.003929 loss: 2.4926 (2.4505) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [56] [160/312] eta: 0:01:54 lr: 0.003929 min_lr: 0.003929 loss: 2.3919 (2.4416) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [56] [170/312] eta: 0:01:46 lr: 0.003929 min_lr: 0.003929 loss: 2.4689 (2.4521) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [56] [180/312] eta: 0:01:38 lr: 0.003929 min_lr: 0.003929 loss: 2.4689 (2.4467) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [56] [190/312] eta: 0:01:30 lr: 0.003929 min_lr: 0.003929 loss: 2.4427 (2.4500) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [56] [200/312] eta: 0:01:23 lr: 0.003929 min_lr: 0.003929 loss: 2.6295 (2.4549) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [56] [210/312] eta: 0:01:15 lr: 0.003929 min_lr: 0.003929 loss: 2.6270 (2.4513) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [56] [220/312] eta: 0:01:07 lr: 0.003929 min_lr: 0.003929 loss: 2.0988 (2.4373) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [56] [230/312] eta: 0:01:00 lr: 0.003928 min_lr: 0.003928 loss: 2.3048 (2.4425) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [56] [240/312] eta: 0:00:52 lr: 0.003928 min_lr: 0.003928 loss: 2.4969 (2.4429) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [56] [250/312] eta: 0:00:45 lr: 0.003928 min_lr: 0.003928 loss: 2.3943 (2.4426) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [56] [260/312] eta: 0:00:38 lr: 0.003928 min_lr: 0.003928 loss: 2.3943 (2.4414) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [56] [270/312] eta: 0:00:30 lr: 0.003928 min_lr: 0.003928 loss: 2.5898 (2.4458) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [56] [280/312] eta: 0:00:23 lr: 0.003928 min_lr: 0.003928 loss: 2.5898 (2.4493) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [56] [290/312] eta: 0:00:16 lr: 0.003928 min_lr: 0.003928 loss: 2.5941 (2.4542) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0008 max mem: 64948 Epoch: [56] [300/312] eta: 0:00:08 lr: 0.003928 min_lr: 0.003928 loss: 2.6625 (2.4558) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0001 max mem: 64948 Epoch: [56] [310/312] eta: 0:00:01 lr: 0.003927 min_lr: 0.003927 loss: 2.5393 (2.4538) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0001 max mem: 64948 Epoch: [56] [311/312] eta: 0:00:00 lr: 0.003927 min_lr: 0.003927 loss: 2.5393 (2.4540) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0001 max mem: 64948 Epoch: [56] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.003927 min_lr: 0.003927 loss: 2.5393 (2.4512) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.8117 (0.8117) acc1: 80.7292 (80.7292) acc5: 94.2708 (94.2708) time: 4.8457 data: 4.6265 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1948 (1.1084) acc1: 71.0938 (72.1920) acc5: 90.8854 (90.7200) time: 0.6897 data: 0.5141 max mem: 64948 Test: Total time: 0:00:06 (0.7003 s / it) * Acc@1 72.680 Acc@5 91.076 loss 1.095 Accuracy of the model on the 50000 test images: 72.7% Max accuracy: 73.19% Test: [0/9] eta: 0:00:41 loss: 6.4838 (6.4838) acc1: 0.7812 (0.7812) acc5: 2.8646 (2.8646) time: 4.6313 data: 4.4282 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.9259 (6.9762) acc1: 0.5208 (0.6720) acc5: 2.8646 (3.1360) time: 0.6659 data: 0.4921 max mem: 64948 Test: Total time: 0:00:06 (0.6731 s / it) * Acc@1 0.468 Acc@5 1.806 loss 7.135 Accuracy of the model EMA on 50000 test images: 0.5% Max EMA accuracy: 0.47% Epoch: [57] [ 0/312] eta: 0:52:00 lr: 0.003927 min_lr: 0.003927 loss: 1.7152 (1.7152) weight_decay: 0.0500 (0.0500) time: 10.0021 data: 9.1941 max mem: 64948 Epoch: [57] [ 10/312] eta: 0:07:50 lr: 0.003927 min_lr: 0.003927 loss: 2.5257 (2.3286) weight_decay: 0.0500 (0.0500) time: 1.5585 data: 0.8362 max mem: 64948 Epoch: [57] [ 20/312] eta: 0:05:34 lr: 0.003927 min_lr: 0.003927 loss: 2.5257 (2.3887) weight_decay: 0.0500 (0.0500) time: 0.7028 data: 0.0004 max mem: 64948 Epoch: [57] [ 30/312] eta: 0:04:41 lr: 0.003927 min_lr: 0.003927 loss: 2.4705 (2.4110) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0004 max mem: 64948 Epoch: [57] [ 40/312] eta: 0:04:11 lr: 0.003927 min_lr: 0.003927 loss: 2.3493 (2.3580) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [57] [ 50/312] eta: 0:03:50 lr: 0.003927 min_lr: 0.003927 loss: 2.2563 (2.3692) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [57] [ 60/312] eta: 0:03:34 lr: 0.003927 min_lr: 0.003927 loss: 2.4901 (2.3872) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [57] [ 70/312] eta: 0:03:20 lr: 0.003927 min_lr: 0.003927 loss: 2.4271 (2.3777) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [57] [ 80/312] eta: 0:03:08 lr: 0.003926 min_lr: 0.003926 loss: 2.2894 (2.3773) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [57] [ 90/312] eta: 0:02:57 lr: 0.003926 min_lr: 0.003926 loss: 2.6339 (2.4047) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [57] [100/312] eta: 0:02:47 lr: 0.003926 min_lr: 0.003926 loss: 2.6353 (2.3953) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [57] [110/312] eta: 0:02:37 lr: 0.003926 min_lr: 0.003926 loss: 2.5462 (2.4026) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [57] [120/312] eta: 0:02:28 lr: 0.003926 min_lr: 0.003926 loss: 2.5750 (2.4072) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [57] [130/312] eta: 0:02:19 lr: 0.003926 min_lr: 0.003926 loss: 2.6017 (2.4209) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [57] [140/312] eta: 0:02:11 lr: 0.003926 min_lr: 0.003926 loss: 2.6195 (2.4276) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [57] [150/312] eta: 0:02:02 lr: 0.003926 min_lr: 0.003926 loss: 2.3391 (2.4152) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [57] [160/312] eta: 0:01:54 lr: 0.003925 min_lr: 0.003925 loss: 2.4253 (2.4258) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [57] [170/312] eta: 0:01:46 lr: 0.003925 min_lr: 0.003925 loss: 2.4212 (2.4156) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [57] [180/312] eta: 0:01:38 lr: 0.003925 min_lr: 0.003925 loss: 2.2487 (2.4137) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [57] [190/312] eta: 0:01:30 lr: 0.003925 min_lr: 0.003925 loss: 2.5397 (2.4215) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [57] [200/312] eta: 0:01:23 lr: 0.003925 min_lr: 0.003925 loss: 2.5002 (2.4098) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [57] [210/312] eta: 0:01:15 lr: 0.003925 min_lr: 0.003925 loss: 2.2098 (2.4060) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [57] [220/312] eta: 0:01:07 lr: 0.003925 min_lr: 0.003925 loss: 2.5430 (2.4120) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [57] [230/312] eta: 0:01:00 lr: 0.003924 min_lr: 0.003924 loss: 2.5792 (2.4149) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [57] [240/312] eta: 0:00:52 lr: 0.003924 min_lr: 0.003924 loss: 2.5234 (2.4202) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [57] [250/312] eta: 0:00:45 lr: 0.003924 min_lr: 0.003924 loss: 2.4639 (2.4145) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [57] [260/312] eta: 0:00:38 lr: 0.003924 min_lr: 0.003924 loss: 2.3449 (2.4145) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [57] [270/312] eta: 0:00:30 lr: 0.003924 min_lr: 0.003924 loss: 2.4874 (2.4096) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [57] [280/312] eta: 0:00:23 lr: 0.003924 min_lr: 0.003924 loss: 2.4874 (2.4123) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0009 max mem: 64948 Epoch: [57] [290/312] eta: 0:00:15 lr: 0.003924 min_lr: 0.003924 loss: 2.6436 (2.4193) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [57] [300/312] eta: 0:00:08 lr: 0.003924 min_lr: 0.003924 loss: 2.6198 (2.4221) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [57] [310/312] eta: 0:00:01 lr: 0.003923 min_lr: 0.003923 loss: 2.6628 (2.4303) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [57] [311/312] eta: 0:00:00 lr: 0.003923 min_lr: 0.003923 loss: 2.6628 (2.4321) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [57] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.003923 min_lr: 0.003923 loss: 2.6628 (2.4581) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.8493 (0.8493) acc1: 78.6458 (78.6458) acc5: 95.3125 (95.3125) time: 4.6293 data: 4.4102 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1636 (1.0840) acc1: 73.6979 (72.5760) acc5: 90.8854 (91.5200) time: 0.6656 data: 0.4901 max mem: 64948 Test: Total time: 0:00:06 (0.6889 s / it) * Acc@1 73.102 Acc@5 91.210 loss 1.080 Accuracy of the model on the 50000 test images: 73.1% Max accuracy: 73.19% Test: [0/9] eta: 0:00:43 loss: 6.3221 (6.3221) acc1: 1.3021 (1.3021) acc5: 4.4271 (4.4271) time: 4.7835 data: 4.5719 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.7259 (6.7751) acc1: 1.3021 (1.3440) acc5: 4.4271 (4.3520) time: 0.6860 data: 0.5081 max mem: 64948 Test: Total time: 0:00:06 (0.6935 s / it) * Acc@1 0.764 Acc@5 2.580 loss 6.993 Accuracy of the model EMA on 50000 test images: 0.8% Max EMA accuracy: 0.76% Epoch: [58] [ 0/312] eta: 0:50:38 lr: 0.003923 min_lr: 0.003923 loss: 2.6837 (2.6837) weight_decay: 0.0500 (0.0500) time: 9.7375 data: 7.6334 max mem: 64948 Epoch: [58] [ 10/312] eta: 0:07:46 lr: 0.003923 min_lr: 0.003923 loss: 2.6598 (2.5345) weight_decay: 0.0500 (0.0500) time: 1.5431 data: 0.6943 max mem: 64948 Epoch: [58] [ 20/312] eta: 0:05:32 lr: 0.003923 min_lr: 0.003923 loss: 2.6481 (2.4933) weight_decay: 0.0500 (0.0500) time: 0.7089 data: 0.0004 max mem: 64948 Epoch: [58] [ 30/312] eta: 0:04:40 lr: 0.003923 min_lr: 0.003923 loss: 2.5424 (2.4813) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [58] [ 40/312] eta: 0:04:10 lr: 0.003923 min_lr: 0.003923 loss: 2.3783 (2.4492) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [58] [ 50/312] eta: 0:03:50 lr: 0.003923 min_lr: 0.003923 loss: 2.3879 (2.4417) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [58] [ 60/312] eta: 0:03:33 lr: 0.003923 min_lr: 0.003923 loss: 2.5370 (2.4356) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [58] [ 70/312] eta: 0:03:20 lr: 0.003923 min_lr: 0.003923 loss: 2.5375 (2.4330) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0003 max mem: 64948 Epoch: [58] [ 80/312] eta: 0:03:08 lr: 0.003922 min_lr: 0.003922 loss: 2.5474 (2.4514) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [58] [ 90/312] eta: 0:02:57 lr: 0.003922 min_lr: 0.003922 loss: 2.5560 (2.4450) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [58] [100/312] eta: 0:02:46 lr: 0.003922 min_lr: 0.003922 loss: 2.5560 (2.4529) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [58] [110/312] eta: 0:02:37 lr: 0.003922 min_lr: 0.003922 loss: 2.5118 (2.4601) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [58] [120/312] eta: 0:02:28 lr: 0.003922 min_lr: 0.003922 loss: 2.5389 (2.4739) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [58] [130/312] eta: 0:02:19 lr: 0.003922 min_lr: 0.003922 loss: 2.5882 (2.4752) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [58] [140/312] eta: 0:02:10 lr: 0.003922 min_lr: 0.003922 loss: 2.6546 (2.4805) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [58] [150/312] eta: 0:02:02 lr: 0.003921 min_lr: 0.003921 loss: 2.6631 (2.4872) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [58] [160/312] eta: 0:01:54 lr: 0.003921 min_lr: 0.003921 loss: 2.7056 (2.4958) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [58] [170/312] eta: 0:01:46 lr: 0.003921 min_lr: 0.003921 loss: 2.6672 (2.5040) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [58] [180/312] eta: 0:01:38 lr: 0.003921 min_lr: 0.003921 loss: 2.6700 (2.5098) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [58] [190/312] eta: 0:01:30 lr: 0.003921 min_lr: 0.003921 loss: 2.6830 (2.5095) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [58] [200/312] eta: 0:01:22 lr: 0.003921 min_lr: 0.003921 loss: 2.3059 (2.4996) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [58] [210/312] eta: 0:01:15 lr: 0.003921 min_lr: 0.003921 loss: 2.3059 (2.5011) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [58] [220/312] eta: 0:01:07 lr: 0.003921 min_lr: 0.003921 loss: 2.4382 (2.4972) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [58] [230/312] eta: 0:01:00 lr: 0.003920 min_lr: 0.003920 loss: 2.5580 (2.4977) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [58] [240/312] eta: 0:00:52 lr: 0.003920 min_lr: 0.003920 loss: 2.5679 (2.4978) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [58] [250/312] eta: 0:00:45 lr: 0.003920 min_lr: 0.003920 loss: 2.2874 (2.4875) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [58] [260/312] eta: 0:00:37 lr: 0.003920 min_lr: 0.003920 loss: 2.1204 (2.4838) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [58] [270/312] eta: 0:00:30 lr: 0.003920 min_lr: 0.003920 loss: 2.4029 (2.4820) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [58] [280/312] eta: 0:00:23 lr: 0.003920 min_lr: 0.003920 loss: 2.4590 (2.4828) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0009 max mem: 64948 Epoch: [58] [290/312] eta: 0:00:15 lr: 0.003920 min_lr: 0.003920 loss: 2.6300 (2.4875) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [58] [300/312] eta: 0:00:08 lr: 0.003920 min_lr: 0.003920 loss: 2.7007 (2.4881) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [58] [310/312] eta: 0:00:01 lr: 0.003919 min_lr: 0.003919 loss: 2.6144 (2.4892) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [58] [311/312] eta: 0:00:00 lr: 0.003919 min_lr: 0.003919 loss: 2.5681 (2.4893) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [58] Total time: 0:03:46 (0.7271 s / it) Averaged stats: lr: 0.003919 min_lr: 0.003919 loss: 2.5681 (2.4495) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7732 (0.7732) acc1: 79.6875 (79.6875) acc5: 95.0521 (95.0521) time: 4.5856 data: 4.3761 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1979 (1.0715) acc1: 72.1354 (72.7680) acc5: 92.4479 (91.6480) time: 0.6607 data: 0.4863 max mem: 64948 Test: Total time: 0:00:06 (0.6839 s / it) * Acc@1 73.674 Acc@5 91.468 loss 1.069 Accuracy of the model on the 50000 test images: 73.7% Max accuracy: 73.67% Test: [0/9] eta: 0:00:43 loss: 6.0982 (6.0982) acc1: 2.8646 (2.8646) acc5: 7.8125 (7.8125) time: 4.8662 data: 4.6507 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.4624 (6.5121) acc1: 2.8646 (2.3360) acc5: 7.5472 (6.7520) time: 0.6919 data: 0.5168 max mem: 64948 Test: Total time: 0:00:06 (0.6990 s / it) * Acc@1 1.294 Acc@5 4.124 loss 6.799 Accuracy of the model EMA on 50000 test images: 1.3% Max EMA accuracy: 1.29% Epoch: [59] [ 0/312] eta: 0:47:52 lr: 0.003919 min_lr: 0.003919 loss: 2.4669 (2.4669) weight_decay: 0.0500 (0.0500) time: 9.2077 data: 8.4250 max mem: 64948 Epoch: [59] [ 10/312] eta: 0:07:47 lr: 0.003919 min_lr: 0.003919 loss: 2.7713 (2.5499) weight_decay: 0.0500 (0.0500) time: 1.5477 data: 0.8282 max mem: 64948 Epoch: [59] [ 20/312] eta: 0:05:34 lr: 0.003919 min_lr: 0.003919 loss: 2.3417 (2.4192) weight_decay: 0.0500 (0.0500) time: 0.7412 data: 0.0344 max mem: 64948 Epoch: [59] [ 30/312] eta: 0:04:41 lr: 0.003919 min_lr: 0.003919 loss: 2.2873 (2.4156) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0003 max mem: 64948 Epoch: [59] [ 40/312] eta: 0:04:11 lr: 0.003919 min_lr: 0.003919 loss: 2.5642 (2.4038) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [59] [ 50/312] eta: 0:03:50 lr: 0.003919 min_lr: 0.003919 loss: 2.2672 (2.3977) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [59] [ 60/312] eta: 0:03:34 lr: 0.003919 min_lr: 0.003919 loss: 2.2672 (2.3788) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [59] [ 70/312] eta: 0:03:20 lr: 0.003918 min_lr: 0.003918 loss: 2.5282 (2.4274) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [59] [ 80/312] eta: 0:03:08 lr: 0.003918 min_lr: 0.003918 loss: 2.6949 (2.4210) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [59] [ 90/312] eta: 0:02:57 lr: 0.003918 min_lr: 0.003918 loss: 2.3784 (2.4302) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [59] [100/312] eta: 0:02:47 lr: 0.003918 min_lr: 0.003918 loss: 2.5599 (2.4339) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [59] [110/312] eta: 0:02:37 lr: 0.003918 min_lr: 0.003918 loss: 2.5599 (2.4503) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [59] [120/312] eta: 0:02:28 lr: 0.003918 min_lr: 0.003918 loss: 2.5425 (2.4536) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [59] [130/312] eta: 0:02:19 lr: 0.003918 min_lr: 0.003918 loss: 2.2670 (2.4389) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [59] [140/312] eta: 0:02:11 lr: 0.003918 min_lr: 0.003918 loss: 2.1279 (2.4265) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [59] [150/312] eta: 0:02:02 lr: 0.003917 min_lr: 0.003917 loss: 2.4964 (2.4385) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [59] [160/312] eta: 0:01:54 lr: 0.003917 min_lr: 0.003917 loss: 2.5206 (2.4265) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [59] [170/312] eta: 0:01:46 lr: 0.003917 min_lr: 0.003917 loss: 2.2642 (2.4289) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [59] [180/312] eta: 0:01:38 lr: 0.003917 min_lr: 0.003917 loss: 2.3595 (2.4258) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [59] [190/312] eta: 0:01:30 lr: 0.003917 min_lr: 0.003917 loss: 2.3595 (2.4236) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [59] [200/312] eta: 0:01:23 lr: 0.003917 min_lr: 0.003917 loss: 2.5800 (2.4286) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [59] [210/312] eta: 0:01:15 lr: 0.003917 min_lr: 0.003917 loss: 2.5813 (2.4290) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [59] [220/312] eta: 0:01:07 lr: 0.003916 min_lr: 0.003916 loss: 2.3670 (2.4156) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [59] [230/312] eta: 0:01:00 lr: 0.003916 min_lr: 0.003916 loss: 2.0796 (2.4159) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [59] [240/312] eta: 0:00:52 lr: 0.003916 min_lr: 0.003916 loss: 2.5440 (2.4223) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [59] [250/312] eta: 0:00:45 lr: 0.003916 min_lr: 0.003916 loss: 2.4903 (2.4236) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [59] [260/312] eta: 0:00:38 lr: 0.003916 min_lr: 0.003916 loss: 2.4650 (2.4243) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [59] [270/312] eta: 0:00:30 lr: 0.003916 min_lr: 0.003916 loss: 2.4469 (2.4183) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [59] [280/312] eta: 0:00:23 lr: 0.003916 min_lr: 0.003916 loss: 2.4794 (2.4222) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [59] [290/312] eta: 0:00:16 lr: 0.003916 min_lr: 0.003916 loss: 2.6385 (2.4289) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [59] [300/312] eta: 0:00:08 lr: 0.003915 min_lr: 0.003915 loss: 2.5754 (2.4243) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [59] [310/312] eta: 0:00:01 lr: 0.003915 min_lr: 0.003915 loss: 2.2883 (2.4207) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [59] [311/312] eta: 0:00:00 lr: 0.003915 min_lr: 0.003915 loss: 2.2883 (2.4201) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [59] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.003915 min_lr: 0.003915 loss: 2.2883 (2.4396) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7986 (0.7986) acc1: 81.5104 (81.5104) acc5: 94.0104 (94.0104) time: 4.6027 data: 4.3987 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0399 (1.0134) acc1: 74.4792 (73.6640) acc5: 92.4479 (92.3520) time: 0.6627 data: 0.4888 max mem: 64948 Test: Total time: 0:00:06 (0.6862 s / it) * Acc@1 74.190 Acc@5 91.940 loss 1.016 Accuracy of the model on the 50000 test images: 74.2% Max accuracy: 74.19% Test: [0/9] eta: 0:00:40 loss: 5.8042 (5.8042) acc1: 5.2083 (5.2083) acc5: 11.4583 (11.4583) time: 4.4728 data: 4.2550 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 6.1438 (6.1863) acc1: 3.9062 (3.8400) acc5: 11.1979 (10.4960) time: 0.6487 data: 0.4729 max mem: 64948 Test: Total time: 0:00:05 (0.6565 s / it) * Acc@1 2.254 Acc@5 6.616 loss 6.544 Accuracy of the model EMA on 50000 test images: 2.3% Max EMA accuracy: 2.25% Epoch: [60] [ 0/312] eta: 0:46:49 lr: 0.003915 min_lr: 0.003915 loss: 2.4547 (2.4547) weight_decay: 0.0500 (0.0500) time: 9.0041 data: 8.1406 max mem: 64948 Epoch: [60] [ 10/312] eta: 0:07:35 lr: 0.003915 min_lr: 0.003915 loss: 2.4547 (2.4157) weight_decay: 0.0500 (0.0500) time: 1.5096 data: 0.7688 max mem: 64948 Epoch: [60] [ 20/312] eta: 0:05:27 lr: 0.003915 min_lr: 0.003915 loss: 2.3578 (2.3729) weight_decay: 0.0500 (0.0500) time: 0.7267 data: 0.0160 max mem: 64948 Epoch: [60] [ 30/312] eta: 0:04:37 lr: 0.003915 min_lr: 0.003915 loss: 2.3246 (2.3657) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [60] [ 40/312] eta: 0:04:08 lr: 0.003915 min_lr: 0.003915 loss: 2.3096 (2.3508) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [60] [ 50/312] eta: 0:03:48 lr: 0.003915 min_lr: 0.003915 loss: 2.4287 (2.3895) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [60] [ 60/312] eta: 0:03:32 lr: 0.003914 min_lr: 0.003914 loss: 2.5985 (2.4074) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [60] [ 70/312] eta: 0:03:18 lr: 0.003914 min_lr: 0.003914 loss: 2.5035 (2.4184) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [60] [ 80/312] eta: 0:03:07 lr: 0.003914 min_lr: 0.003914 loss: 2.6228 (2.4392) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0003 max mem: 64948 Epoch: [60] [ 90/312] eta: 0:02:56 lr: 0.003914 min_lr: 0.003914 loss: 2.6325 (2.4458) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [60] [100/312] eta: 0:02:46 lr: 0.003914 min_lr: 0.003914 loss: 2.4492 (2.4351) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [60] [110/312] eta: 0:02:37 lr: 0.003914 min_lr: 0.003914 loss: 2.3226 (2.4321) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [60] [120/312] eta: 0:02:27 lr: 0.003914 min_lr: 0.003914 loss: 2.5206 (2.4304) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [60] [130/312] eta: 0:02:19 lr: 0.003913 min_lr: 0.003913 loss: 2.4464 (2.4124) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [60] [140/312] eta: 0:02:10 lr: 0.003913 min_lr: 0.003913 loss: 2.4015 (2.4158) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [60] [150/312] eta: 0:02:02 lr: 0.003913 min_lr: 0.003913 loss: 2.6518 (2.4241) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [60] [160/312] eta: 0:01:54 lr: 0.003913 min_lr: 0.003913 loss: 2.6900 (2.4275) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [60] [170/312] eta: 0:01:46 lr: 0.003913 min_lr: 0.003913 loss: 2.6573 (2.4435) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [60] [180/312] eta: 0:01:38 lr: 0.003913 min_lr: 0.003913 loss: 2.6163 (2.4477) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [60] [190/312] eta: 0:01:30 lr: 0.003913 min_lr: 0.003913 loss: 2.4374 (2.4401) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [60] [200/312] eta: 0:01:22 lr: 0.003913 min_lr: 0.003913 loss: 2.4374 (2.4378) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [60] [210/312] eta: 0:01:15 lr: 0.003912 min_lr: 0.003912 loss: 2.5991 (2.4450) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [60] [220/312] eta: 0:01:07 lr: 0.003912 min_lr: 0.003912 loss: 2.4799 (2.4348) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [60] [230/312] eta: 0:01:00 lr: 0.003912 min_lr: 0.003912 loss: 2.3163 (2.4354) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [60] [240/312] eta: 0:00:52 lr: 0.003912 min_lr: 0.003912 loss: 2.2619 (2.4255) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [60] [250/312] eta: 0:00:45 lr: 0.003912 min_lr: 0.003912 loss: 2.3697 (2.4297) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [60] [260/312] eta: 0:00:37 lr: 0.003912 min_lr: 0.003912 loss: 2.5260 (2.4339) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [60] [270/312] eta: 0:00:30 lr: 0.003912 min_lr: 0.003912 loss: 2.5644 (2.4292) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [60] [280/312] eta: 0:00:23 lr: 0.003911 min_lr: 0.003911 loss: 2.5644 (2.4283) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [60] [290/312] eta: 0:00:15 lr: 0.003911 min_lr: 0.003911 loss: 2.5930 (2.4330) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [60] [300/312] eta: 0:00:08 lr: 0.003911 min_lr: 0.003911 loss: 2.5551 (2.4328) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [60] [310/312] eta: 0:00:01 lr: 0.003911 min_lr: 0.003911 loss: 2.4671 (2.4298) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [60] [311/312] eta: 0:00:00 lr: 0.003911 min_lr: 0.003911 loss: 2.5245 (2.4307) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [60] Total time: 0:03:46 (0.7267 s / it) Averaged stats: lr: 0.003911 min_lr: 0.003911 loss: 2.5245 (2.4412) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.8451 (0.8451) acc1: 80.2083 (80.2083) acc5: 93.7500 (93.7500) time: 4.7768 data: 4.5639 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1325 (1.0741) acc1: 72.9167 (72.5120) acc5: 91.9271 (91.0400) time: 0.6820 data: 0.5072 max mem: 64948 Test: Total time: 0:00:06 (0.7033 s / it) * Acc@1 73.288 Acc@5 91.390 loss 1.065 Accuracy of the model on the 50000 test images: 73.3% Max accuracy: 74.19% Test: [0/9] eta: 0:00:43 loss: 5.4658 (5.4658) acc1: 8.0729 (8.0729) acc5: 16.9271 (16.9271) time: 4.7880 data: 4.5848 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 5.7901 (5.8146) acc1: 7.2917 (6.4640) acc5: 15.8854 (15.1360) time: 0.6834 data: 0.5095 max mem: 64948 Test: Total time: 0:00:06 (0.6917 s / it) * Acc@1 3.876 Acc@5 10.520 loss 6.231 Accuracy of the model EMA on 50000 test images: 3.9% Max EMA accuracy: 3.88% Epoch: [61] [ 0/312] eta: 0:49:28 lr: 0.003911 min_lr: 0.003911 loss: 2.8744 (2.8744) weight_decay: 0.0500 (0.0500) time: 9.5131 data: 8.7366 max mem: 64948 Epoch: [61] [ 10/312] eta: 0:07:40 lr: 0.003911 min_lr: 0.003911 loss: 2.0825 (2.2388) weight_decay: 0.0500 (0.0500) time: 1.5263 data: 0.7946 max mem: 64948 Epoch: [61] [ 20/312] eta: 0:05:29 lr: 0.003911 min_lr: 0.003911 loss: 2.2230 (2.3276) weight_decay: 0.0500 (0.0500) time: 0.7107 data: 0.0004 max mem: 64948 Epoch: [61] [ 30/312] eta: 0:04:39 lr: 0.003911 min_lr: 0.003911 loss: 2.5098 (2.3929) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [61] [ 40/312] eta: 0:04:09 lr: 0.003910 min_lr: 0.003910 loss: 2.6008 (2.4332) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [61] [ 50/312] eta: 0:03:49 lr: 0.003910 min_lr: 0.003910 loss: 2.6095 (2.4496) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [61] [ 60/312] eta: 0:03:33 lr: 0.003910 min_lr: 0.003910 loss: 2.3970 (2.4197) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [61] [ 70/312] eta: 0:03:19 lr: 0.003910 min_lr: 0.003910 loss: 2.4653 (2.4522) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [61] [ 80/312] eta: 0:03:07 lr: 0.003910 min_lr: 0.003910 loss: 2.6356 (2.4702) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [61] [ 90/312] eta: 0:02:56 lr: 0.003910 min_lr: 0.003910 loss: 2.7126 (2.4638) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [61] [100/312] eta: 0:02:46 lr: 0.003910 min_lr: 0.003910 loss: 2.2704 (2.4353) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [61] [110/312] eta: 0:02:37 lr: 0.003909 min_lr: 0.003909 loss: 2.2704 (2.4283) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [61] [120/312] eta: 0:02:28 lr: 0.003909 min_lr: 0.003909 loss: 2.4562 (2.4319) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [61] [130/312] eta: 0:02:19 lr: 0.003909 min_lr: 0.003909 loss: 2.4691 (2.4212) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [61] [140/312] eta: 0:02:10 lr: 0.003909 min_lr: 0.003909 loss: 2.5832 (2.4346) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [61] [150/312] eta: 0:02:02 lr: 0.003909 min_lr: 0.003909 loss: 2.6586 (2.4397) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [61] [160/312] eta: 0:01:54 lr: 0.003909 min_lr: 0.003909 loss: 2.6213 (2.4408) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [61] [170/312] eta: 0:01:46 lr: 0.003909 min_lr: 0.003909 loss: 2.5535 (2.4453) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [61] [180/312] eta: 0:01:38 lr: 0.003908 min_lr: 0.003908 loss: 2.5535 (2.4441) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [61] [190/312] eta: 0:01:30 lr: 0.003908 min_lr: 0.003908 loss: 2.5210 (2.4530) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [61] [200/312] eta: 0:01:22 lr: 0.003908 min_lr: 0.003908 loss: 2.5465 (2.4622) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [61] [210/312] eta: 0:01:15 lr: 0.003908 min_lr: 0.003908 loss: 2.5511 (2.4616) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [61] [220/312] eta: 0:01:07 lr: 0.003908 min_lr: 0.003908 loss: 2.5404 (2.4529) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [61] [230/312] eta: 0:01:00 lr: 0.003908 min_lr: 0.003908 loss: 2.5048 (2.4502) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [61] [240/312] eta: 0:00:52 lr: 0.003908 min_lr: 0.003908 loss: 2.3341 (2.4454) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [61] [250/312] eta: 0:00:45 lr: 0.003907 min_lr: 0.003907 loss: 2.1981 (2.4376) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [61] [260/312] eta: 0:00:37 lr: 0.003907 min_lr: 0.003907 loss: 2.4260 (2.4402) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [61] [270/312] eta: 0:00:30 lr: 0.003907 min_lr: 0.003907 loss: 2.5696 (2.4410) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [61] [280/312] eta: 0:00:23 lr: 0.003907 min_lr: 0.003907 loss: 2.4898 (2.4408) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [61] [290/312] eta: 0:00:15 lr: 0.003907 min_lr: 0.003907 loss: 2.3609 (2.4375) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0008 max mem: 64948 Epoch: [61] [300/312] eta: 0:00:08 lr: 0.003907 min_lr: 0.003907 loss: 2.3131 (2.4318) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [61] [310/312] eta: 0:00:01 lr: 0.003907 min_lr: 0.003907 loss: 2.3542 (2.4339) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [61] [311/312] eta: 0:00:00 lr: 0.003907 min_lr: 0.003907 loss: 2.3789 (2.4342) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [61] Total time: 0:03:46 (0.7268 s / it) Averaged stats: lr: 0.003907 min_lr: 0.003907 loss: 2.3789 (2.4382) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.8066 (0.8066) acc1: 80.2083 (80.2083) acc5: 94.0104 (94.0104) time: 4.4785 data: 4.2665 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1699 (1.0814) acc1: 71.8750 (72.7360) acc5: 92.4479 (91.1040) time: 0.6489 data: 0.4741 max mem: 64948 Test: Total time: 0:00:05 (0.6589 s / it) * Acc@1 73.670 Acc@5 91.530 loss 1.054 Accuracy of the model on the 50000 test images: 73.7% Max accuracy: 74.19% Test: [0/9] eta: 0:00:44 loss: 5.0855 (5.0855) acc1: 11.4583 (11.4583) acc5: 23.6979 (23.6979) time: 4.9676 data: 4.7645 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 5.4136 (5.3999) acc1: 10.4167 (9.8880) acc5: 21.3542 (22.0800) time: 0.7032 data: 0.5295 max mem: 64948 Test: Total time: 0:00:06 (0.7110 s / it) * Acc@1 6.338 Acc@5 15.710 loss 5.854 Accuracy of the model EMA on 50000 test images: 6.3% Max EMA accuracy: 6.34% Epoch: [62] [ 0/312] eta: 0:54:04 lr: 0.003907 min_lr: 0.003907 loss: 2.5352 (2.5352) weight_decay: 0.0500 (0.0500) time: 10.3990 data: 9.5962 max mem: 64948 Epoch: [62] [ 10/312] eta: 0:08:00 lr: 0.003906 min_lr: 0.003906 loss: 2.5056 (2.4550) weight_decay: 0.0500 (0.0500) time: 1.5915 data: 0.8727 max mem: 64948 Epoch: [62] [ 20/312] eta: 0:05:39 lr: 0.003906 min_lr: 0.003906 loss: 2.5056 (2.4488) weight_decay: 0.0500 (0.0500) time: 0.7015 data: 0.0004 max mem: 64948 Epoch: [62] [ 30/312] eta: 0:04:45 lr: 0.003906 min_lr: 0.003906 loss: 2.3121 (2.3612) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [62] [ 40/312] eta: 0:04:14 lr: 0.003906 min_lr: 0.003906 loss: 2.2956 (2.3546) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [62] [ 50/312] eta: 0:03:52 lr: 0.003906 min_lr: 0.003906 loss: 2.3070 (2.3205) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [62] [ 60/312] eta: 0:03:36 lr: 0.003906 min_lr: 0.003906 loss: 2.3070 (2.3490) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [62] [ 70/312] eta: 0:03:21 lr: 0.003906 min_lr: 0.003906 loss: 2.5287 (2.3660) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [62] [ 80/312] eta: 0:03:09 lr: 0.003905 min_lr: 0.003905 loss: 2.6603 (2.4140) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [62] [ 90/312] eta: 0:02:58 lr: 0.003905 min_lr: 0.003905 loss: 2.5968 (2.4097) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [62] [100/312] eta: 0:02:48 lr: 0.003905 min_lr: 0.003905 loss: 2.4730 (2.4226) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [62] [110/312] eta: 0:02:38 lr: 0.003905 min_lr: 0.003905 loss: 2.5813 (2.4315) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [62] [120/312] eta: 0:02:29 lr: 0.003905 min_lr: 0.003905 loss: 2.5721 (2.4326) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [62] [130/312] eta: 0:02:20 lr: 0.003905 min_lr: 0.003905 loss: 2.5749 (2.4359) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [62] [140/312] eta: 0:02:11 lr: 0.003905 min_lr: 0.003905 loss: 2.6320 (2.4468) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [62] [150/312] eta: 0:02:03 lr: 0.003904 min_lr: 0.003904 loss: 2.6151 (2.4484) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [62] [160/312] eta: 0:01:55 lr: 0.003904 min_lr: 0.003904 loss: 2.5288 (2.4405) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [62] [170/312] eta: 0:01:46 lr: 0.003904 min_lr: 0.003904 loss: 2.4336 (2.4317) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [62] [180/312] eta: 0:01:38 lr: 0.003904 min_lr: 0.003904 loss: 2.5394 (2.4325) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [62] [190/312] eta: 0:01:31 lr: 0.003904 min_lr: 0.003904 loss: 2.4458 (2.4352) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [62] [200/312] eta: 0:01:23 lr: 0.003904 min_lr: 0.003904 loss: 2.5883 (2.4395) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [62] [210/312] eta: 0:01:15 lr: 0.003904 min_lr: 0.003904 loss: 2.5883 (2.4398) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [62] [220/312] eta: 0:01:08 lr: 0.003903 min_lr: 0.003903 loss: 2.4583 (2.4400) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [62] [230/312] eta: 0:01:00 lr: 0.003903 min_lr: 0.003903 loss: 2.4583 (2.4448) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [62] [240/312] eta: 0:00:52 lr: 0.003903 min_lr: 0.003903 loss: 2.5230 (2.4405) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [62] [250/312] eta: 0:00:45 lr: 0.003903 min_lr: 0.003903 loss: 2.4006 (2.4407) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [62] [260/312] eta: 0:00:38 lr: 0.003903 min_lr: 0.003903 loss: 2.4361 (2.4409) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [62] [270/312] eta: 0:00:30 lr: 0.003903 min_lr: 0.003903 loss: 2.4949 (2.4420) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [62] [280/312] eta: 0:00:23 lr: 0.003903 min_lr: 0.003903 loss: 2.4277 (2.4430) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [62] [290/312] eta: 0:00:16 lr: 0.003902 min_lr: 0.003902 loss: 2.4590 (2.4437) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [62] [300/312] eta: 0:00:08 lr: 0.003902 min_lr: 0.003902 loss: 2.4590 (2.4399) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [62] [310/312] eta: 0:00:01 lr: 0.003902 min_lr: 0.003902 loss: 2.4655 (2.4354) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [62] [311/312] eta: 0:00:00 lr: 0.003902 min_lr: 0.003902 loss: 2.4655 (2.4337) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [62] Total time: 0:03:47 (0.7291 s / it) Averaged stats: lr: 0.003902 min_lr: 0.003902 loss: 2.4655 (2.4268) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.8503 (0.8503) acc1: 77.8646 (77.8646) acc5: 93.2292 (93.2292) time: 4.6872 data: 4.4673 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1368 (1.0642) acc1: 71.6146 (72.6080) acc5: 91.1458 (91.2960) time: 0.6721 data: 0.4964 max mem: 64948 Test: Total time: 0:00:06 (0.6970 s / it) * Acc@1 73.774 Acc@5 91.666 loss 1.050 Accuracy of the model on the 50000 test images: 73.8% Max accuracy: 74.19% Test: [0/9] eta: 0:00:42 loss: 4.6906 (4.6906) acc1: 14.8438 (14.8438) acc5: 31.2500 (31.2500) time: 4.7189 data: 4.4956 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 5.0430 (4.9648) acc1: 13.2812 (13.6960) acc5: 29.4271 (29.7280) time: 0.6760 data: 0.4996 max mem: 64948 Test: Total time: 0:00:06 (0.6870 s / it) * Acc@1 9.752 Acc@5 22.174 loss 5.434 Accuracy of the model EMA on 50000 test images: 9.8% Max EMA accuracy: 9.75% Epoch: [63] [ 0/312] eta: 0:49:50 lr: 0.003902 min_lr: 0.003902 loss: 2.0222 (2.0222) weight_decay: 0.0500 (0.0500) time: 9.5850 data: 8.7959 max mem: 64948 Epoch: [63] [ 10/312] eta: 0:07:52 lr: 0.003902 min_lr: 0.003902 loss: 2.6140 (2.4038) weight_decay: 0.0500 (0.0500) time: 1.5650 data: 0.8277 max mem: 64948 Epoch: [63] [ 20/312] eta: 0:05:35 lr: 0.003902 min_lr: 0.003902 loss: 2.5861 (2.3946) weight_decay: 0.0500 (0.0500) time: 0.7284 data: 0.0156 max mem: 64948 Epoch: [63] [ 30/312] eta: 0:04:42 lr: 0.003902 min_lr: 0.003902 loss: 2.2813 (2.3915) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [63] [ 40/312] eta: 0:04:12 lr: 0.003902 min_lr: 0.003902 loss: 2.4627 (2.3976) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [63] [ 50/312] eta: 0:03:51 lr: 0.003901 min_lr: 0.003901 loss: 2.6086 (2.4317) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [63] [ 60/312] eta: 0:03:34 lr: 0.003901 min_lr: 0.003901 loss: 2.4279 (2.3840) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [63] [ 70/312] eta: 0:03:20 lr: 0.003901 min_lr: 0.003901 loss: 2.4110 (2.4292) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [63] [ 80/312] eta: 0:03:08 lr: 0.003901 min_lr: 0.003901 loss: 2.6171 (2.4299) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [63] [ 90/312] eta: 0:02:57 lr: 0.003901 min_lr: 0.003901 loss: 2.4734 (2.4276) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [63] [100/312] eta: 0:02:47 lr: 0.003901 min_lr: 0.003901 loss: 2.4237 (2.4170) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [63] [110/312] eta: 0:02:37 lr: 0.003901 min_lr: 0.003901 loss: 2.5272 (2.4326) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [63] [120/312] eta: 0:02:28 lr: 0.003900 min_lr: 0.003900 loss: 2.4347 (2.4184) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [63] [130/312] eta: 0:02:19 lr: 0.003900 min_lr: 0.003900 loss: 2.2389 (2.4154) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [63] [140/312] eta: 0:02:11 lr: 0.003900 min_lr: 0.003900 loss: 2.3803 (2.4032) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [63] [150/312] eta: 0:02:02 lr: 0.003900 min_lr: 0.003900 loss: 2.4501 (2.4035) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [63] [160/312] eta: 0:01:54 lr: 0.003900 min_lr: 0.003900 loss: 2.3907 (2.3962) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [63] [170/312] eta: 0:01:46 lr: 0.003900 min_lr: 0.003900 loss: 2.3501 (2.3904) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [63] [180/312] eta: 0:01:38 lr: 0.003900 min_lr: 0.003900 loss: 2.5330 (2.3962) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [63] [190/312] eta: 0:01:30 lr: 0.003899 min_lr: 0.003899 loss: 2.4456 (2.3947) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [63] [200/312] eta: 0:01:23 lr: 0.003899 min_lr: 0.003899 loss: 2.4311 (2.3885) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [63] [210/312] eta: 0:01:15 lr: 0.003899 min_lr: 0.003899 loss: 2.4674 (2.3955) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [63] [220/312] eta: 0:01:07 lr: 0.003899 min_lr: 0.003899 loss: 2.4357 (2.3937) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [63] [230/312] eta: 0:01:00 lr: 0.003899 min_lr: 0.003899 loss: 2.6154 (2.4036) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [63] [240/312] eta: 0:00:52 lr: 0.003899 min_lr: 0.003899 loss: 2.6545 (2.4083) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [63] [250/312] eta: 0:00:45 lr: 0.003898 min_lr: 0.003898 loss: 2.4041 (2.4065) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [63] [260/312] eta: 0:00:38 lr: 0.003898 min_lr: 0.003898 loss: 2.4041 (2.4090) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [63] [270/312] eta: 0:00:30 lr: 0.003898 min_lr: 0.003898 loss: 2.4269 (2.4108) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [63] [280/312] eta: 0:00:23 lr: 0.003898 min_lr: 0.003898 loss: 2.4376 (2.4155) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0010 max mem: 64948 Epoch: [63] [290/312] eta: 0:00:16 lr: 0.003898 min_lr: 0.003898 loss: 2.4944 (2.4131) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0008 max mem: 64948 Epoch: [63] [300/312] eta: 0:00:08 lr: 0.003898 min_lr: 0.003898 loss: 2.5172 (2.4177) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [63] [310/312] eta: 0:00:01 lr: 0.003898 min_lr: 0.003898 loss: 2.5172 (2.4172) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [63] [311/312] eta: 0:00:00 lr: 0.003898 min_lr: 0.003898 loss: 2.4632 (2.4166) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [63] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.003898 min_lr: 0.003898 loss: 2.4632 (2.4183) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.8435 (0.8435) acc1: 79.9479 (79.9479) acc5: 94.0104 (94.0104) time: 4.7916 data: 4.5725 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0642 (1.0591) acc1: 75.0000 (72.9280) acc5: 91.6667 (91.3920) time: 0.6840 data: 0.5081 max mem: 64948 Test: Total time: 0:00:06 (0.7083 s / it) * Acc@1 73.916 Acc@5 91.660 loss 1.037 Accuracy of the model on the 50000 test images: 73.9% Max accuracy: 74.19% Test: [0/9] eta: 0:00:43 loss: 4.3008 (4.3008) acc1: 18.7500 (18.7500) acc5: 39.8438 (39.8438) time: 4.8656 data: 4.6618 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 4.6182 (4.5377) acc1: 19.0104 (18.0480) acc5: 38.0208 (37.3440) time: 0.6919 data: 0.5181 max mem: 64948 Test: Total time: 0:00:06 (0.7023 s / it) * Acc@1 13.902 Acc@5 29.368 loss 4.993 Accuracy of the model EMA on 50000 test images: 13.9% Max EMA accuracy: 13.90% Epoch: [64] [ 0/312] eta: 0:44:54 lr: 0.003898 min_lr: 0.003898 loss: 2.3087 (2.3087) weight_decay: 0.0500 (0.0500) time: 8.6363 data: 7.8439 max mem: 64948 Epoch: [64] [ 10/312] eta: 0:07:36 lr: 0.003897 min_lr: 0.003897 loss: 2.4235 (2.3969) weight_decay: 0.0500 (0.0500) time: 1.5129 data: 0.7825 max mem: 64948 Epoch: [64] [ 20/312] eta: 0:05:28 lr: 0.003897 min_lr: 0.003897 loss: 2.3779 (2.3341) weight_decay: 0.0500 (0.0500) time: 0.7488 data: 0.0383 max mem: 64948 Epoch: [64] [ 30/312] eta: 0:04:38 lr: 0.003897 min_lr: 0.003897 loss: 2.2893 (2.3376) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [64] [ 40/312] eta: 0:04:08 lr: 0.003897 min_lr: 0.003897 loss: 2.3387 (2.3504) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [64] [ 50/312] eta: 0:03:48 lr: 0.003897 min_lr: 0.003897 loss: 2.4645 (2.3836) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [64] [ 60/312] eta: 0:03:32 lr: 0.003897 min_lr: 0.003897 loss: 2.4823 (2.3887) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [64] [ 70/312] eta: 0:03:18 lr: 0.003897 min_lr: 0.003897 loss: 2.5418 (2.4281) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [64] [ 80/312] eta: 0:03:07 lr: 0.003896 min_lr: 0.003896 loss: 2.5688 (2.4395) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [64] [ 90/312] eta: 0:02:56 lr: 0.003896 min_lr: 0.003896 loss: 2.5461 (2.4309) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [64] [100/312] eta: 0:02:46 lr: 0.003896 min_lr: 0.003896 loss: 2.5684 (2.4448) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [64] [110/312] eta: 0:02:36 lr: 0.003896 min_lr: 0.003896 loss: 2.6533 (2.4642) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [64] [120/312] eta: 0:02:27 lr: 0.003896 min_lr: 0.003896 loss: 2.4239 (2.4534) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [64] [130/312] eta: 0:02:18 lr: 0.003896 min_lr: 0.003896 loss: 2.5015 (2.4696) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [64] [140/312] eta: 0:02:10 lr: 0.003895 min_lr: 0.003895 loss: 2.5251 (2.4612) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [64] [150/312] eta: 0:02:02 lr: 0.003895 min_lr: 0.003895 loss: 2.2619 (2.4432) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [64] [160/312] eta: 0:01:54 lr: 0.003895 min_lr: 0.003895 loss: 2.2993 (2.4385) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [64] [170/312] eta: 0:01:46 lr: 0.003895 min_lr: 0.003895 loss: 2.3870 (2.4350) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [64] [180/312] eta: 0:01:38 lr: 0.003895 min_lr: 0.003895 loss: 2.3870 (2.4320) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [64] [190/312] eta: 0:01:30 lr: 0.003895 min_lr: 0.003895 loss: 2.5073 (2.4352) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [64] [200/312] eta: 0:01:22 lr: 0.003895 min_lr: 0.003895 loss: 2.4884 (2.4302) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [64] [210/312] eta: 0:01:15 lr: 0.003894 min_lr: 0.003894 loss: 2.3201 (2.4270) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [64] [220/312] eta: 0:01:07 lr: 0.003894 min_lr: 0.003894 loss: 2.2518 (2.4215) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [64] [230/312] eta: 0:01:00 lr: 0.003894 min_lr: 0.003894 loss: 2.2544 (2.4213) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [64] [240/312] eta: 0:00:52 lr: 0.003894 min_lr: 0.003894 loss: 2.3553 (2.4166) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [64] [250/312] eta: 0:00:45 lr: 0.003894 min_lr: 0.003894 loss: 2.3420 (2.4149) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [64] [260/312] eta: 0:00:37 lr: 0.003894 min_lr: 0.003894 loss: 2.3494 (2.4155) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [64] [270/312] eta: 0:00:30 lr: 0.003894 min_lr: 0.003894 loss: 2.5047 (2.4158) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [64] [280/312] eta: 0:00:23 lr: 0.003893 min_lr: 0.003893 loss: 2.5479 (2.4190) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0009 max mem: 64948 Epoch: [64] [290/312] eta: 0:00:15 lr: 0.003893 min_lr: 0.003893 loss: 2.5197 (2.4121) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [64] [300/312] eta: 0:00:08 lr: 0.003893 min_lr: 0.003893 loss: 2.1867 (2.4090) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [64] [310/312] eta: 0:00:01 lr: 0.003893 min_lr: 0.003893 loss: 2.4113 (2.4072) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [64] [311/312] eta: 0:00:00 lr: 0.003893 min_lr: 0.003893 loss: 2.4513 (2.4080) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [64] Total time: 0:03:46 (0.7268 s / it) Averaged stats: lr: 0.003893 min_lr: 0.003893 loss: 2.4513 (2.4173) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.9049 (0.9049) acc1: 79.1667 (79.1667) acc5: 92.9688 (92.9688) time: 4.6965 data: 4.4863 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1322 (1.0986) acc1: 73.9583 (72.4160) acc5: 91.4062 (91.2640) time: 0.6731 data: 0.4986 max mem: 64948 Test: Total time: 0:00:06 (0.6948 s / it) * Acc@1 73.616 Acc@5 91.510 loss 1.078 Accuracy of the model on the 50000 test images: 73.6% Max accuracy: 74.19% Test: [0/9] eta: 0:00:47 loss: 3.9225 (3.9225) acc1: 24.2188 (24.2188) acc5: 45.0521 (45.0521) time: 5.2719 data: 5.0580 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 4.1787 (4.1294) acc1: 22.6562 (22.6880) acc5: 43.3962 (44.1920) time: 0.7370 data: 0.5621 max mem: 64948 Test: Total time: 0:00:06 (0.7461 s / it) * Acc@1 18.640 Acc@5 37.152 loss 4.546 Accuracy of the model EMA on 50000 test images: 18.6% Max EMA accuracy: 18.64% Epoch: [65] [ 0/312] eta: 0:49:59 lr: 0.003893 min_lr: 0.003893 loss: 2.8513 (2.8513) weight_decay: 0.0500 (0.0500) time: 9.6132 data: 8.8435 max mem: 64948 Epoch: [65] [ 10/312] eta: 0:07:39 lr: 0.003893 min_lr: 0.003893 loss: 2.5198 (2.4604) weight_decay: 0.0500 (0.0500) time: 1.5201 data: 0.8043 max mem: 64948 Epoch: [65] [ 20/312] eta: 0:05:29 lr: 0.003893 min_lr: 0.003893 loss: 2.5198 (2.4656) weight_decay: 0.0500 (0.0500) time: 0.7044 data: 0.0004 max mem: 64948 Epoch: [65] [ 30/312] eta: 0:04:39 lr: 0.003892 min_lr: 0.003892 loss: 2.5070 (2.4174) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0003 max mem: 64948 Epoch: [65] [ 40/312] eta: 0:04:10 lr: 0.003892 min_lr: 0.003892 loss: 2.3887 (2.4067) weight_decay: 0.0500 (0.0500) time: 0.7017 data: 0.0003 max mem: 64948 Epoch: [65] [ 50/312] eta: 0:03:49 lr: 0.003892 min_lr: 0.003892 loss: 2.3887 (2.4134) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [65] [ 60/312] eta: 0:03:33 lr: 0.003892 min_lr: 0.003892 loss: 2.4676 (2.4455) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [65] [ 70/312] eta: 0:03:19 lr: 0.003892 min_lr: 0.003892 loss: 2.2553 (2.4006) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [65] [ 80/312] eta: 0:03:07 lr: 0.003892 min_lr: 0.003892 loss: 2.0928 (2.3759) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [65] [ 90/312] eta: 0:02:56 lr: 0.003892 min_lr: 0.003892 loss: 2.2432 (2.3999) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [65] [100/312] eta: 0:02:46 lr: 0.003891 min_lr: 0.003891 loss: 2.5121 (2.4004) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [65] [110/312] eta: 0:02:37 lr: 0.003891 min_lr: 0.003891 loss: 2.4659 (2.4091) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [65] [120/312] eta: 0:02:28 lr: 0.003891 min_lr: 0.003891 loss: 2.4550 (2.4047) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [65] [130/312] eta: 0:02:19 lr: 0.003891 min_lr: 0.003891 loss: 2.2637 (2.3933) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [65] [140/312] eta: 0:02:10 lr: 0.003891 min_lr: 0.003891 loss: 2.4218 (2.4010) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [65] [150/312] eta: 0:02:02 lr: 0.003891 min_lr: 0.003891 loss: 2.5470 (2.4094) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [65] [160/312] eta: 0:01:54 lr: 0.003890 min_lr: 0.003890 loss: 2.4687 (2.4038) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [65] [170/312] eta: 0:01:46 lr: 0.003890 min_lr: 0.003890 loss: 2.2916 (2.4002) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [65] [180/312] eta: 0:01:38 lr: 0.003890 min_lr: 0.003890 loss: 2.4876 (2.4088) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [65] [190/312] eta: 0:01:30 lr: 0.003890 min_lr: 0.003890 loss: 2.5115 (2.4092) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [65] [200/312] eta: 0:01:22 lr: 0.003890 min_lr: 0.003890 loss: 2.4971 (2.4148) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [65] [210/312] eta: 0:01:15 lr: 0.003890 min_lr: 0.003890 loss: 2.5310 (2.4152) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [65] [220/312] eta: 0:01:07 lr: 0.003890 min_lr: 0.003890 loss: 2.2601 (2.4065) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [65] [230/312] eta: 0:01:00 lr: 0.003889 min_lr: 0.003889 loss: 2.3600 (2.4072) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [65] [240/312] eta: 0:00:52 lr: 0.003889 min_lr: 0.003889 loss: 2.5432 (2.4123) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [65] [250/312] eta: 0:00:45 lr: 0.003889 min_lr: 0.003889 loss: 2.5488 (2.4112) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [65] [260/312] eta: 0:00:37 lr: 0.003889 min_lr: 0.003889 loss: 2.5987 (2.4186) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [65] [270/312] eta: 0:00:30 lr: 0.003889 min_lr: 0.003889 loss: 2.5137 (2.4198) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [65] [280/312] eta: 0:00:23 lr: 0.003889 min_lr: 0.003889 loss: 2.2668 (2.4133) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [65] [290/312] eta: 0:00:15 lr: 0.003888 min_lr: 0.003888 loss: 2.5210 (2.4202) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [65] [300/312] eta: 0:00:08 lr: 0.003888 min_lr: 0.003888 loss: 2.5721 (2.4199) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [65] [310/312] eta: 0:00:01 lr: 0.003888 min_lr: 0.003888 loss: 2.5097 (2.4246) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [65] [311/312] eta: 0:00:00 lr: 0.003888 min_lr: 0.003888 loss: 2.5097 (2.4228) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [65] Total time: 0:03:46 (0.7268 s / it) Averaged stats: lr: 0.003888 min_lr: 0.003888 loss: 2.5097 (2.4233) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.8785 (0.8785) acc1: 78.3854 (78.3854) acc5: 93.7500 (93.7500) time: 4.7213 data: 4.5014 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0651 (1.0471) acc1: 73.9583 (72.9600) acc5: 91.6667 (91.4560) time: 0.6762 data: 0.5003 max mem: 64948 Test: Total time: 0:00:06 (0.7013 s / it) * Acc@1 73.794 Acc@5 91.644 loss 1.053 Accuracy of the model on the 50000 test images: 73.8% Max accuracy: 74.19% Test: [0/9] eta: 0:00:43 loss: 3.5786 (3.5786) acc1: 29.9479 (29.9479) acc5: 51.8229 (51.8229) time: 4.7884 data: 4.5790 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 3.8147 (3.7590) acc1: 25.7812 (27.4240) acc5: 48.4375 (50.5600) time: 0.6834 data: 0.5089 max mem: 64948 Test: Total time: 0:00:06 (0.6937 s / it) * Acc@1 23.622 Acc@5 43.908 loss 4.127 Accuracy of the model EMA on 50000 test images: 23.6% Max EMA accuracy: 23.62% Epoch: [66] [ 0/312] eta: 0:52:03 lr: 0.003888 min_lr: 0.003888 loss: 2.5636 (2.5636) weight_decay: 0.0500 (0.0500) time: 10.0105 data: 9.2307 max mem: 64948 Epoch: [66] [ 10/312] eta: 0:08:00 lr: 0.003888 min_lr: 0.003888 loss: 2.5183 (2.5072) weight_decay: 0.0500 (0.0500) time: 1.5902 data: 0.8395 max mem: 64948 Epoch: [66] [ 20/312] eta: 0:05:39 lr: 0.003888 min_lr: 0.003888 loss: 2.4952 (2.5173) weight_decay: 0.0500 (0.0500) time: 0.7204 data: 0.0003 max mem: 64948 Epoch: [66] [ 30/312] eta: 0:04:45 lr: 0.003888 min_lr: 0.003888 loss: 2.5139 (2.4995) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [66] [ 40/312] eta: 0:04:14 lr: 0.003888 min_lr: 0.003888 loss: 2.4812 (2.4637) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [66] [ 50/312] eta: 0:03:52 lr: 0.003887 min_lr: 0.003887 loss: 2.4398 (2.4384) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [66] [ 60/312] eta: 0:03:35 lr: 0.003887 min_lr: 0.003887 loss: 2.5324 (2.4601) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [66] [ 70/312] eta: 0:03:21 lr: 0.003887 min_lr: 0.003887 loss: 2.5926 (2.4572) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [66] [ 80/312] eta: 0:03:09 lr: 0.003887 min_lr: 0.003887 loss: 2.4100 (2.4381) weight_decay: 0.0500 (0.0500) time: 0.7013 data: 0.0004 max mem: 64948 Epoch: [66] [ 90/312] eta: 0:02:58 lr: 0.003887 min_lr: 0.003887 loss: 2.4456 (2.4495) weight_decay: 0.0500 (0.0500) time: 0.7026 data: 0.0003 max mem: 64948 Epoch: [66] [100/312] eta: 0:02:48 lr: 0.003887 min_lr: 0.003887 loss: 2.4456 (2.4414) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [66] [110/312] eta: 0:02:38 lr: 0.003886 min_lr: 0.003886 loss: 2.4710 (2.4404) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [66] [120/312] eta: 0:02:29 lr: 0.003886 min_lr: 0.003886 loss: 2.4678 (2.4385) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [66] [130/312] eta: 0:02:20 lr: 0.003886 min_lr: 0.003886 loss: 2.4095 (2.4340) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [66] [140/312] eta: 0:02:11 lr: 0.003886 min_lr: 0.003886 loss: 2.3778 (2.4344) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [66] [150/312] eta: 0:02:03 lr: 0.003886 min_lr: 0.003886 loss: 2.2952 (2.4265) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [66] [160/312] eta: 0:01:55 lr: 0.003886 min_lr: 0.003886 loss: 2.4606 (2.4317) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [66] [170/312] eta: 0:01:46 lr: 0.003885 min_lr: 0.003885 loss: 2.4860 (2.4336) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [66] [180/312] eta: 0:01:38 lr: 0.003885 min_lr: 0.003885 loss: 2.4860 (2.4350) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [66] [190/312] eta: 0:01:31 lr: 0.003885 min_lr: 0.003885 loss: 2.5201 (2.4347) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [66] [200/312] eta: 0:01:23 lr: 0.003885 min_lr: 0.003885 loss: 2.5893 (2.4393) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [66] [210/312] eta: 0:01:15 lr: 0.003885 min_lr: 0.003885 loss: 2.2843 (2.4251) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [66] [220/312] eta: 0:01:08 lr: 0.003885 min_lr: 0.003885 loss: 2.6004 (2.4334) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [66] [230/312] eta: 0:01:00 lr: 0.003885 min_lr: 0.003885 loss: 2.6144 (2.4358) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [66] [240/312] eta: 0:00:52 lr: 0.003884 min_lr: 0.003884 loss: 2.4860 (2.4342) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [66] [250/312] eta: 0:00:45 lr: 0.003884 min_lr: 0.003884 loss: 2.5609 (2.4393) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [66] [260/312] eta: 0:00:38 lr: 0.003884 min_lr: 0.003884 loss: 2.3793 (2.4320) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [66] [270/312] eta: 0:00:30 lr: 0.003884 min_lr: 0.003884 loss: 2.3793 (2.4266) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [66] [280/312] eta: 0:00:23 lr: 0.003884 min_lr: 0.003884 loss: 2.4196 (2.4246) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0010 max mem: 64948 Epoch: [66] [290/312] eta: 0:00:16 lr: 0.003884 min_lr: 0.003884 loss: 2.3467 (2.4241) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [66] [300/312] eta: 0:00:08 lr: 0.003883 min_lr: 0.003883 loss: 2.4735 (2.4259) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [66] [310/312] eta: 0:00:01 lr: 0.003883 min_lr: 0.003883 loss: 2.4735 (2.4255) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [66] [311/312] eta: 0:00:00 lr: 0.003883 min_lr: 0.003883 loss: 2.5049 (2.4267) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [66] Total time: 0:03:47 (0.7290 s / it) Averaged stats: lr: 0.003883 min_lr: 0.003883 loss: 2.5049 (2.4116) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.8926 (0.8926) acc1: 78.1250 (78.1250) acc5: 93.2292 (93.2292) time: 4.6832 data: 4.4768 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1251 (1.0668) acc1: 74.7396 (74.0160) acc5: 92.4479 (91.6160) time: 0.6716 data: 0.4975 max mem: 64948 Test: Total time: 0:00:06 (0.6955 s / it) * Acc@1 74.290 Acc@5 92.002 loss 1.038 Accuracy of the model on the 50000 test images: 74.3% Max accuracy: 74.29% Test: [0/9] eta: 0:00:41 loss: 3.2735 (3.2735) acc1: 33.8542 (33.8542) acc5: 55.4688 (55.4688) time: 4.6498 data: 4.4284 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 3.5250 (3.4306) acc1: 29.9479 (31.9360) acc5: 53.9062 (55.9360) time: 0.6682 data: 0.4921 max mem: 64948 Test: Total time: 0:00:06 (0.6759 s / it) * Acc@1 28.376 Acc@5 50.124 loss 3.752 Accuracy of the model EMA on 50000 test images: 28.4% Max EMA accuracy: 28.38% Epoch: [67] [ 0/312] eta: 0:50:16 lr: 0.003883 min_lr: 0.003883 loss: 2.2368 (2.2368) weight_decay: 0.0500 (0.0500) time: 9.6689 data: 8.2822 max mem: 64948 Epoch: [67] [ 10/312] eta: 0:07:43 lr: 0.003883 min_lr: 0.003883 loss: 2.2368 (2.3268) weight_decay: 0.0500 (0.0500) time: 1.5345 data: 0.7559 max mem: 64948 Epoch: [67] [ 20/312] eta: 0:05:31 lr: 0.003883 min_lr: 0.003883 loss: 2.5003 (2.3943) weight_decay: 0.0500 (0.0500) time: 0.7072 data: 0.0018 max mem: 64948 Epoch: [67] [ 30/312] eta: 0:04:39 lr: 0.003883 min_lr: 0.003883 loss: 2.5128 (2.3559) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [67] [ 40/312] eta: 0:04:10 lr: 0.003883 min_lr: 0.003883 loss: 2.1342 (2.2979) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [67] [ 50/312] eta: 0:03:49 lr: 0.003882 min_lr: 0.003882 loss: 2.4200 (2.3463) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [67] [ 60/312] eta: 0:03:33 lr: 0.003882 min_lr: 0.003882 loss: 2.5412 (2.3378) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [67] [ 70/312] eta: 0:03:19 lr: 0.003882 min_lr: 0.003882 loss: 2.5073 (2.3330) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [67] [ 80/312] eta: 0:03:07 lr: 0.003882 min_lr: 0.003882 loss: 2.3241 (2.3365) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [67] [ 90/312] eta: 0:02:56 lr: 0.003882 min_lr: 0.003882 loss: 2.3241 (2.3415) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [67] [100/312] eta: 0:02:46 lr: 0.003882 min_lr: 0.003882 loss: 2.5064 (2.3606) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [67] [110/312] eta: 0:02:37 lr: 0.003882 min_lr: 0.003882 loss: 2.4105 (2.3609) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0003 max mem: 64948 Epoch: [67] [120/312] eta: 0:02:28 lr: 0.003881 min_lr: 0.003881 loss: 2.2597 (2.3437) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [67] [130/312] eta: 0:02:19 lr: 0.003881 min_lr: 0.003881 loss: 2.3029 (2.3459) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [67] [140/312] eta: 0:02:10 lr: 0.003881 min_lr: 0.003881 loss: 2.4711 (2.3541) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [67] [150/312] eta: 0:02:02 lr: 0.003881 min_lr: 0.003881 loss: 2.4444 (2.3532) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [67] [160/312] eta: 0:01:54 lr: 0.003881 min_lr: 0.003881 loss: 2.4444 (2.3624) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [67] [170/312] eta: 0:01:46 lr: 0.003881 min_lr: 0.003881 loss: 2.4349 (2.3552) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [67] [180/312] eta: 0:01:38 lr: 0.003880 min_lr: 0.003880 loss: 2.2800 (2.3523) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [67] [190/312] eta: 0:01:30 lr: 0.003880 min_lr: 0.003880 loss: 2.5766 (2.3559) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [67] [200/312] eta: 0:01:22 lr: 0.003880 min_lr: 0.003880 loss: 2.5530 (2.3559) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [67] [210/312] eta: 0:01:15 lr: 0.003880 min_lr: 0.003880 loss: 2.3804 (2.3535) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [67] [220/312] eta: 0:01:07 lr: 0.003880 min_lr: 0.003880 loss: 2.2191 (2.3457) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [67] [230/312] eta: 0:01:00 lr: 0.003880 min_lr: 0.003880 loss: 2.3198 (2.3470) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [67] [240/312] eta: 0:00:52 lr: 0.003879 min_lr: 0.003879 loss: 2.3700 (2.3498) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [67] [250/312] eta: 0:00:45 lr: 0.003879 min_lr: 0.003879 loss: 2.6251 (2.3598) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [67] [260/312] eta: 0:00:37 lr: 0.003879 min_lr: 0.003879 loss: 2.6065 (2.3622) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [67] [270/312] eta: 0:00:30 lr: 0.003879 min_lr: 0.003879 loss: 2.2930 (2.3602) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [67] [280/312] eta: 0:00:23 lr: 0.003879 min_lr: 0.003879 loss: 2.3190 (2.3599) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0010 max mem: 64948 Epoch: [67] [290/312] eta: 0:00:15 lr: 0.003879 min_lr: 0.003879 loss: 2.5796 (2.3705) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [67] [300/312] eta: 0:00:08 lr: 0.003878 min_lr: 0.003878 loss: 2.7146 (2.3738) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [67] [310/312] eta: 0:00:01 lr: 0.003878 min_lr: 0.003878 loss: 2.5462 (2.3786) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [67] [311/312] eta: 0:00:00 lr: 0.003878 min_lr: 0.003878 loss: 2.5462 (2.3804) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [67] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.003878 min_lr: 0.003878 loss: 2.5462 (2.4040) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.8528 (0.8528) acc1: 78.1250 (78.1250) acc5: 92.9688 (92.9688) time: 4.5261 data: 4.3182 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1409 (1.1063) acc1: 73.4375 (71.8400) acc5: 92.1875 (90.8160) time: 0.6542 data: 0.4799 max mem: 64948 Test: Total time: 0:00:06 (0.6768 s / it) * Acc@1 73.076 Acc@5 91.144 loss 1.094 Accuracy of the model on the 50000 test images: 73.1% Max accuracy: 74.29% Test: [0/9] eta: 0:00:43 loss: 2.9995 (2.9995) acc1: 39.0625 (39.0625) acc5: 59.8958 (59.8958) time: 4.8146 data: 4.5965 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 3.2688 (3.1404) acc1: 34.3750 (36.3520) acc5: 58.3333 (61.2160) time: 0.6864 data: 0.5108 max mem: 64948 Test: Total time: 0:00:06 (0.6956 s / it) * Acc@1 32.730 Acc@5 55.890 loss 3.420 Accuracy of the model EMA on 50000 test images: 32.7% Max EMA accuracy: 32.73% Epoch: [68] [ 0/312] eta: 0:48:23 lr: 0.003878 min_lr: 0.003878 loss: 2.3320 (2.3320) weight_decay: 0.0500 (0.0500) time: 9.3075 data: 8.5312 max mem: 64948 Epoch: [68] [ 10/312] eta: 0:07:32 lr: 0.003878 min_lr: 0.003878 loss: 2.4343 (2.3550) weight_decay: 0.0500 (0.0500) time: 1.4980 data: 0.7760 max mem: 64948 Epoch: [68] [ 20/312] eta: 0:05:26 lr: 0.003878 min_lr: 0.003878 loss: 2.4552 (2.3618) weight_decay: 0.0500 (0.0500) time: 0.7088 data: 0.0004 max mem: 64948 Epoch: [68] [ 30/312] eta: 0:04:36 lr: 0.003878 min_lr: 0.003878 loss: 2.4570 (2.3911) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0003 max mem: 64948 Epoch: [68] [ 40/312] eta: 0:04:07 lr: 0.003878 min_lr: 0.003878 loss: 2.4570 (2.3924) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [68] [ 50/312] eta: 0:03:47 lr: 0.003877 min_lr: 0.003877 loss: 2.4234 (2.3863) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [68] [ 60/312] eta: 0:03:31 lr: 0.003877 min_lr: 0.003877 loss: 2.4234 (2.3925) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [68] [ 70/312] eta: 0:03:18 lr: 0.003877 min_lr: 0.003877 loss: 2.5197 (2.4091) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [68] [ 80/312] eta: 0:03:06 lr: 0.003877 min_lr: 0.003877 loss: 2.2985 (2.3824) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [68] [ 90/312] eta: 0:02:56 lr: 0.003877 min_lr: 0.003877 loss: 2.1812 (2.3603) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [68] [100/312] eta: 0:02:46 lr: 0.003877 min_lr: 0.003877 loss: 1.8688 (2.3207) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [68] [110/312] eta: 0:02:36 lr: 0.003877 min_lr: 0.003877 loss: 2.1625 (2.3350) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [68] [120/312] eta: 0:02:27 lr: 0.003876 min_lr: 0.003876 loss: 2.5156 (2.3499) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [68] [130/312] eta: 0:02:18 lr: 0.003876 min_lr: 0.003876 loss: 2.5197 (2.3589) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [68] [140/312] eta: 0:02:10 lr: 0.003876 min_lr: 0.003876 loss: 2.5197 (2.3682) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [68] [150/312] eta: 0:02:02 lr: 0.003876 min_lr: 0.003876 loss: 2.5624 (2.3818) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [68] [160/312] eta: 0:01:54 lr: 0.003876 min_lr: 0.003876 loss: 2.5339 (2.3867) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [68] [170/312] eta: 0:01:46 lr: 0.003876 min_lr: 0.003876 loss: 2.4061 (2.3804) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [68] [180/312] eta: 0:01:38 lr: 0.003875 min_lr: 0.003875 loss: 2.2753 (2.3826) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [68] [190/312] eta: 0:01:30 lr: 0.003875 min_lr: 0.003875 loss: 2.2966 (2.3801) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [68] [200/312] eta: 0:01:22 lr: 0.003875 min_lr: 0.003875 loss: 2.2966 (2.3787) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [68] [210/312] eta: 0:01:15 lr: 0.003875 min_lr: 0.003875 loss: 2.3092 (2.3763) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [68] [220/312] eta: 0:01:07 lr: 0.003875 min_lr: 0.003875 loss: 2.3092 (2.3752) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [68] [230/312] eta: 0:01:00 lr: 0.003875 min_lr: 0.003875 loss: 2.2641 (2.3728) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [68] [240/312] eta: 0:00:52 lr: 0.003874 min_lr: 0.003874 loss: 2.2987 (2.3724) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [68] [250/312] eta: 0:00:45 lr: 0.003874 min_lr: 0.003874 loss: 2.4373 (2.3664) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [68] [260/312] eta: 0:00:37 lr: 0.003874 min_lr: 0.003874 loss: 2.4582 (2.3705) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [68] [270/312] eta: 0:00:30 lr: 0.003874 min_lr: 0.003874 loss: 2.5820 (2.3745) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [68] [280/312] eta: 0:00:23 lr: 0.003874 min_lr: 0.003874 loss: 2.3545 (2.3704) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [68] [290/312] eta: 0:00:15 lr: 0.003874 min_lr: 0.003874 loss: 2.1560 (2.3660) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [68] [300/312] eta: 0:00:08 lr: 0.003873 min_lr: 0.003873 loss: 2.4282 (2.3630) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [68] [310/312] eta: 0:00:01 lr: 0.003873 min_lr: 0.003873 loss: 2.5240 (2.3676) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [68] [311/312] eta: 0:00:00 lr: 0.003873 min_lr: 0.003873 loss: 2.5273 (2.3684) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [68] Total time: 0:03:46 (0.7260 s / it) Averaged stats: lr: 0.003873 min_lr: 0.003873 loss: 2.5273 (2.4028) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.8128 (0.8128) acc1: 79.9479 (79.9479) acc5: 94.0104 (94.0104) time: 4.8623 data: 4.6431 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0595 (1.0391) acc1: 74.7396 (73.4080) acc5: 92.9688 (91.9360) time: 0.6918 data: 0.5160 max mem: 64948 Test: Total time: 0:00:06 (0.7176 s / it) * Acc@1 74.354 Acc@5 92.230 loss 1.015 Accuracy of the model on the 50000 test images: 74.4% Max accuracy: 74.35% Test: [0/9] eta: 0:00:41 loss: 2.7553 (2.7553) acc1: 42.4479 (42.4479) acc5: 63.8021 (63.8021) time: 4.6217 data: 4.4039 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 3.0033 (2.8895) acc1: 37.2396 (40.2240) acc5: 63.8021 (65.7280) time: 0.6659 data: 0.4894 max mem: 64948 Test: Total time: 0:00:06 (0.6751 s / it) * Acc@1 36.892 Acc@5 60.746 loss 3.132 Accuracy of the model EMA on 50000 test images: 36.9% Max EMA accuracy: 36.89% Epoch: [69] [ 0/312] eta: 0:44:43 lr: 0.003873 min_lr: 0.003873 loss: 1.9461 (1.9461) weight_decay: 0.0500 (0.0500) time: 8.6011 data: 7.3326 max mem: 64948 Epoch: [69] [ 10/312] eta: 0:07:38 lr: 0.003873 min_lr: 0.003873 loss: 2.0257 (2.1769) weight_decay: 0.0500 (0.0500) time: 1.5191 data: 0.7058 max mem: 64948 Epoch: [69] [ 20/312] eta: 0:05:28 lr: 0.003873 min_lr: 0.003873 loss: 2.2358 (2.2470) weight_decay: 0.0500 (0.0500) time: 0.7516 data: 0.0217 max mem: 64948 Epoch: [69] [ 30/312] eta: 0:04:38 lr: 0.003873 min_lr: 0.003873 loss: 2.2920 (2.2474) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [69] [ 40/312] eta: 0:04:09 lr: 0.003873 min_lr: 0.003873 loss: 2.4843 (2.3171) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [69] [ 50/312] eta: 0:03:48 lr: 0.003872 min_lr: 0.003872 loss: 2.5539 (2.3453) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [69] [ 60/312] eta: 0:03:32 lr: 0.003872 min_lr: 0.003872 loss: 2.5220 (2.3510) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [69] [ 70/312] eta: 0:03:19 lr: 0.003872 min_lr: 0.003872 loss: 2.2942 (2.3119) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [69] [ 80/312] eta: 0:03:07 lr: 0.003872 min_lr: 0.003872 loss: 2.0680 (2.3083) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [69] [ 90/312] eta: 0:02:56 lr: 0.003872 min_lr: 0.003872 loss: 2.1754 (2.2985) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [69] [100/312] eta: 0:02:46 lr: 0.003872 min_lr: 0.003872 loss: 2.4485 (2.3239) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [69] [110/312] eta: 0:02:36 lr: 0.003871 min_lr: 0.003871 loss: 2.6574 (2.3346) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [69] [120/312] eta: 0:02:27 lr: 0.003871 min_lr: 0.003871 loss: 2.5664 (2.3446) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [69] [130/312] eta: 0:02:19 lr: 0.003871 min_lr: 0.003871 loss: 2.3545 (2.3394) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [69] [140/312] eta: 0:02:10 lr: 0.003871 min_lr: 0.003871 loss: 2.2004 (2.3374) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [69] [150/312] eta: 0:02:02 lr: 0.003871 min_lr: 0.003871 loss: 2.3609 (2.3509) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [69] [160/312] eta: 0:01:54 lr: 0.003871 min_lr: 0.003871 loss: 2.4537 (2.3529) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [69] [170/312] eta: 0:01:46 lr: 0.003870 min_lr: 0.003870 loss: 2.5333 (2.3606) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [69] [180/312] eta: 0:01:38 lr: 0.003870 min_lr: 0.003870 loss: 2.4690 (2.3634) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [69] [190/312] eta: 0:01:30 lr: 0.003870 min_lr: 0.003870 loss: 2.4651 (2.3741) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [69] [200/312] eta: 0:01:22 lr: 0.003870 min_lr: 0.003870 loss: 2.4651 (2.3765) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [69] [210/312] eta: 0:01:15 lr: 0.003870 min_lr: 0.003870 loss: 2.4052 (2.3827) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [69] [220/312] eta: 0:01:07 lr: 0.003870 min_lr: 0.003870 loss: 2.3911 (2.3797) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [69] [230/312] eta: 0:01:00 lr: 0.003869 min_lr: 0.003869 loss: 2.5185 (2.3843) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [69] [240/312] eta: 0:00:52 lr: 0.003869 min_lr: 0.003869 loss: 2.6269 (2.3871) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [69] [250/312] eta: 0:00:45 lr: 0.003869 min_lr: 0.003869 loss: 2.6125 (2.3858) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [69] [260/312] eta: 0:00:37 lr: 0.003869 min_lr: 0.003869 loss: 2.5433 (2.3897) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [69] [270/312] eta: 0:00:30 lr: 0.003869 min_lr: 0.003869 loss: 2.3739 (2.3798) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [69] [280/312] eta: 0:00:23 lr: 0.003869 min_lr: 0.003869 loss: 2.5929 (2.3871) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0012 max mem: 64948 Epoch: [69] [290/312] eta: 0:00:15 lr: 0.003868 min_lr: 0.003868 loss: 2.6754 (2.3950) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0010 max mem: 64948 Epoch: [69] [300/312] eta: 0:00:08 lr: 0.003868 min_lr: 0.003868 loss: 2.6754 (2.4012) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [69] [310/312] eta: 0:00:01 lr: 0.003868 min_lr: 0.003868 loss: 2.5543 (2.3976) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [69] [311/312] eta: 0:00:00 lr: 0.003868 min_lr: 0.003868 loss: 2.4895 (2.3967) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [69] Total time: 0:03:46 (0.7259 s / it) Averaged stats: lr: 0.003868 min_lr: 0.003868 loss: 2.4895 (2.4044) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.7544 (0.7544) acc1: 80.4688 (80.4688) acc5: 95.0521 (95.0521) time: 4.8090 data: 4.5947 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1003 (1.0224) acc1: 74.2188 (73.6320) acc5: 91.9271 (91.7120) time: 0.6856 data: 0.5106 max mem: 64948 Test: Total time: 0:00:06 (0.7114 s / it) * Acc@1 74.168 Acc@5 91.896 loss 1.015 Accuracy of the model on the 50000 test images: 74.2% Max accuracy: 74.35% Test: [0/9] eta: 0:00:44 loss: 2.5378 (2.5378) acc1: 45.5729 (45.5729) acc5: 69.0104 (69.0104) time: 4.9161 data: 4.6991 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.7570 (2.6670) acc1: 40.6250 (43.7120) acc5: 67.1875 (68.8000) time: 0.6975 data: 0.5222 max mem: 64948 Test: Total time: 0:00:06 (0.7079 s / it) * Acc@1 40.778 Acc@5 64.892 loss 2.877 Accuracy of the model EMA on 50000 test images: 40.8% Max EMA accuracy: 40.78% Epoch: [70] [ 0/312] eta: 0:47:44 lr: 0.003868 min_lr: 0.003868 loss: 2.6496 (2.6496) weight_decay: 0.0500 (0.0500) time: 9.1821 data: 7.8829 max mem: 64948 Epoch: [70] [ 10/312] eta: 0:07:50 lr: 0.003868 min_lr: 0.003868 loss: 2.6060 (2.2978) weight_decay: 0.0500 (0.0500) time: 1.5594 data: 0.7549 max mem: 64948 Epoch: [70] [ 20/312] eta: 0:05:34 lr: 0.003868 min_lr: 0.003868 loss: 2.1326 (2.2872) weight_decay: 0.0500 (0.0500) time: 0.7448 data: 0.0212 max mem: 64948 Epoch: [70] [ 30/312] eta: 0:04:42 lr: 0.003868 min_lr: 0.003868 loss: 2.3973 (2.3603) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [70] [ 40/312] eta: 0:04:11 lr: 0.003867 min_lr: 0.003867 loss: 2.5233 (2.3801) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [70] [ 50/312] eta: 0:03:51 lr: 0.003867 min_lr: 0.003867 loss: 2.5116 (2.3657) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [70] [ 60/312] eta: 0:03:34 lr: 0.003867 min_lr: 0.003867 loss: 2.5582 (2.4023) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [70] [ 70/312] eta: 0:03:20 lr: 0.003867 min_lr: 0.003867 loss: 2.4498 (2.3679) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [70] [ 80/312] eta: 0:03:08 lr: 0.003867 min_lr: 0.003867 loss: 2.3163 (2.3866) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [70] [ 90/312] eta: 0:02:57 lr: 0.003867 min_lr: 0.003867 loss: 2.5522 (2.3930) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [70] [100/312] eta: 0:02:47 lr: 0.003866 min_lr: 0.003866 loss: 2.4188 (2.3889) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [70] [110/312] eta: 0:02:37 lr: 0.003866 min_lr: 0.003866 loss: 2.4121 (2.3777) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [70] [120/312] eta: 0:02:28 lr: 0.003866 min_lr: 0.003866 loss: 2.3486 (2.3799) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [70] [130/312] eta: 0:02:19 lr: 0.003866 min_lr: 0.003866 loss: 2.2909 (2.3732) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [70] [140/312] eta: 0:02:11 lr: 0.003866 min_lr: 0.003866 loss: 2.3850 (2.3790) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [70] [150/312] eta: 0:02:02 lr: 0.003866 min_lr: 0.003866 loss: 2.5371 (2.3867) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [70] [160/312] eta: 0:01:54 lr: 0.003865 min_lr: 0.003865 loss: 2.4843 (2.3863) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [70] [170/312] eta: 0:01:46 lr: 0.003865 min_lr: 0.003865 loss: 2.6152 (2.3970) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [70] [180/312] eta: 0:01:38 lr: 0.003865 min_lr: 0.003865 loss: 2.6259 (2.3949) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [70] [190/312] eta: 0:01:30 lr: 0.003865 min_lr: 0.003865 loss: 2.3059 (2.3884) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [70] [200/312] eta: 0:01:23 lr: 0.003865 min_lr: 0.003865 loss: 2.3187 (2.3805) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [70] [210/312] eta: 0:01:15 lr: 0.003865 min_lr: 0.003865 loss: 2.3616 (2.3814) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [70] [220/312] eta: 0:01:07 lr: 0.003864 min_lr: 0.003864 loss: 2.4278 (2.3813) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [70] [230/312] eta: 0:01:00 lr: 0.003864 min_lr: 0.003864 loss: 2.4811 (2.3905) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [70] [240/312] eta: 0:00:52 lr: 0.003864 min_lr: 0.003864 loss: 2.4811 (2.3855) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [70] [250/312] eta: 0:00:45 lr: 0.003864 min_lr: 0.003864 loss: 2.2043 (2.3829) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [70] [260/312] eta: 0:00:38 lr: 0.003864 min_lr: 0.003864 loss: 2.1501 (2.3809) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [70] [270/312] eta: 0:00:30 lr: 0.003864 min_lr: 0.003864 loss: 2.4535 (2.3849) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [70] [280/312] eta: 0:00:23 lr: 0.003863 min_lr: 0.003863 loss: 2.4535 (2.3803) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0010 max mem: 64948 Epoch: [70] [290/312] eta: 0:00:16 lr: 0.003863 min_lr: 0.003863 loss: 2.1869 (2.3796) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [70] [300/312] eta: 0:00:08 lr: 0.003863 min_lr: 0.003863 loss: 2.1935 (2.3777) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [70] [310/312] eta: 0:00:01 lr: 0.003863 min_lr: 0.003863 loss: 2.4402 (2.3816) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [70] [311/312] eta: 0:00:00 lr: 0.003863 min_lr: 0.003863 loss: 2.4402 (2.3833) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [70] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.003863 min_lr: 0.003863 loss: 2.4402 (2.4028) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.8188 (0.8188) acc1: 78.9062 (78.9062) acc5: 94.2708 (94.2708) time: 4.3268 data: 4.1077 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1192 (1.0473) acc1: 73.6979 (73.3120) acc5: 92.4479 (91.8400) time: 0.6321 data: 0.4565 max mem: 64948 Test: Total time: 0:00:05 (0.6654 s / it) * Acc@1 74.098 Acc@5 91.866 loss 1.037 Accuracy of the model on the 50000 test images: 74.1% Max accuracy: 74.35% Test: [0/9] eta: 0:00:44 loss: 2.3442 (2.3442) acc1: 49.2188 (49.2188) acc5: 71.3542 (71.3542) time: 4.9519 data: 4.7357 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.5398 (2.4746) acc1: 45.0521 (46.6240) acc5: 69.5312 (71.1680) time: 0.7015 data: 0.5263 max mem: 64948 Test: Total time: 0:00:06 (0.7110 s / it) * Acc@1 44.178 Acc@5 68.456 loss 2.656 Accuracy of the model EMA on 50000 test images: 44.2% Max EMA accuracy: 44.18% Epoch: [71] [ 0/312] eta: 0:43:40 lr: 0.003863 min_lr: 0.003863 loss: 1.9268 (1.9268) weight_decay: 0.0500 (0.0500) time: 8.3999 data: 6.8989 max mem: 64948 Epoch: [71] [ 10/312] eta: 0:07:38 lr: 0.003863 min_lr: 0.003863 loss: 2.1760 (2.1775) weight_decay: 0.0500 (0.0500) time: 1.5171 data: 0.6806 max mem: 64948 Epoch: [71] [ 20/312] eta: 0:05:28 lr: 0.003862 min_lr: 0.003862 loss: 2.5061 (2.3271) weight_decay: 0.0500 (0.0500) time: 0.7625 data: 0.0295 max mem: 64948 Epoch: [71] [ 30/312] eta: 0:04:38 lr: 0.003862 min_lr: 0.003862 loss: 2.6502 (2.4149) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [71] [ 40/312] eta: 0:04:09 lr: 0.003862 min_lr: 0.003862 loss: 2.6631 (2.3869) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [71] [ 50/312] eta: 0:03:48 lr: 0.003862 min_lr: 0.003862 loss: 2.4304 (2.3793) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [71] [ 60/312] eta: 0:03:32 lr: 0.003862 min_lr: 0.003862 loss: 2.4304 (2.3702) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [71] [ 70/312] eta: 0:03:19 lr: 0.003862 min_lr: 0.003862 loss: 2.5228 (2.3911) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [71] [ 80/312] eta: 0:03:07 lr: 0.003861 min_lr: 0.003861 loss: 2.3865 (2.3779) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [71] [ 90/312] eta: 0:02:56 lr: 0.003861 min_lr: 0.003861 loss: 2.3391 (2.3739) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [71] [100/312] eta: 0:02:46 lr: 0.003861 min_lr: 0.003861 loss: 2.4343 (2.3849) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0003 max mem: 64948 Epoch: [71] [110/312] eta: 0:02:36 lr: 0.003861 min_lr: 0.003861 loss: 2.5271 (2.4051) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [71] [120/312] eta: 0:02:27 lr: 0.003861 min_lr: 0.003861 loss: 2.6128 (2.4232) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [71] [130/312] eta: 0:02:19 lr: 0.003861 min_lr: 0.003861 loss: 2.6128 (2.4344) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [71] [140/312] eta: 0:02:10 lr: 0.003860 min_lr: 0.003860 loss: 2.4117 (2.4209) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [71] [150/312] eta: 0:02:02 lr: 0.003860 min_lr: 0.003860 loss: 2.1687 (2.4096) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [71] [160/312] eta: 0:01:54 lr: 0.003860 min_lr: 0.003860 loss: 2.2533 (2.4117) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [71] [170/312] eta: 0:01:46 lr: 0.003860 min_lr: 0.003860 loss: 2.5379 (2.4125) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [71] [180/312] eta: 0:01:38 lr: 0.003860 min_lr: 0.003860 loss: 2.6282 (2.4261) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [71] [190/312] eta: 0:01:30 lr: 0.003860 min_lr: 0.003860 loss: 2.5996 (2.4277) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [71] [200/312] eta: 0:01:22 lr: 0.003859 min_lr: 0.003859 loss: 2.4024 (2.4242) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [71] [210/312] eta: 0:01:15 lr: 0.003859 min_lr: 0.003859 loss: 2.4069 (2.4291) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [71] [220/312] eta: 0:01:07 lr: 0.003859 min_lr: 0.003859 loss: 2.4261 (2.4244) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [71] [230/312] eta: 0:01:00 lr: 0.003859 min_lr: 0.003859 loss: 2.4791 (2.4271) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [71] [240/312] eta: 0:00:52 lr: 0.003859 min_lr: 0.003859 loss: 2.5335 (2.4285) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [71] [250/312] eta: 0:00:45 lr: 0.003859 min_lr: 0.003859 loss: 2.5264 (2.4294) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [71] [260/312] eta: 0:00:37 lr: 0.003858 min_lr: 0.003858 loss: 2.5104 (2.4229) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [71] [270/312] eta: 0:00:30 lr: 0.003858 min_lr: 0.003858 loss: 2.4745 (2.4248) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [71] [280/312] eta: 0:00:23 lr: 0.003858 min_lr: 0.003858 loss: 2.4782 (2.4259) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0010 max mem: 64948 Epoch: [71] [290/312] eta: 0:00:15 lr: 0.003858 min_lr: 0.003858 loss: 2.5509 (2.4276) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [71] [300/312] eta: 0:00:08 lr: 0.003858 min_lr: 0.003858 loss: 2.5760 (2.4306) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0002 max mem: 64948 Epoch: [71] [310/312] eta: 0:00:01 lr: 0.003857 min_lr: 0.003857 loss: 2.5430 (2.4326) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [71] [311/312] eta: 0:00:00 lr: 0.003857 min_lr: 0.003857 loss: 2.5430 (2.4337) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [71] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.003857 min_lr: 0.003857 loss: 2.5430 (2.3992) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.8221 (0.8221) acc1: 79.4271 (79.4271) acc5: 93.2292 (93.2292) time: 4.6293 data: 4.4065 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0941 (1.0207) acc1: 76.0417 (74.3680) acc5: 90.8854 (91.2960) time: 0.6660 data: 0.4897 max mem: 64948 Test: Total time: 0:00:06 (0.6851 s / it) * Acc@1 74.332 Acc@5 91.908 loss 1.019 Accuracy of the model on the 50000 test images: 74.3% Max accuracy: 74.35% Test: [0/9] eta: 0:00:44 loss: 2.1830 (2.1830) acc1: 52.0833 (52.0833) acc5: 74.7396 (74.7396) time: 4.9504 data: 4.7323 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.3627 (2.3175) acc1: 47.9167 (49.3760) acc5: 71.3542 (73.3760) time: 0.7015 data: 0.5259 max mem: 64948 Test: Total time: 0:00:06 (0.7150 s / it) * Acc@1 47.010 Acc@5 71.296 loss 2.476 Accuracy of the model EMA on 50000 test images: 47.0% Max EMA accuracy: 47.01% Epoch: [72] [ 0/312] eta: 0:47:06 lr: 0.003857 min_lr: 0.003857 loss: 2.0003 (2.0003) weight_decay: 0.0500 (0.0500) time: 9.0604 data: 8.2936 max mem: 64948 Epoch: [72] [ 10/312] eta: 0:07:26 lr: 0.003857 min_lr: 0.003857 loss: 2.4575 (2.2623) weight_decay: 0.0500 (0.0500) time: 1.4771 data: 0.7543 max mem: 64948 Epoch: [72] [ 20/312] eta: 0:05:22 lr: 0.003857 min_lr: 0.003857 loss: 2.4575 (2.3345) weight_decay: 0.0500 (0.0500) time: 0.7079 data: 0.0004 max mem: 64948 Epoch: [72] [ 30/312] eta: 0:04:34 lr: 0.003857 min_lr: 0.003857 loss: 2.3787 (2.3144) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0003 max mem: 64948 Epoch: [72] [ 40/312] eta: 0:04:06 lr: 0.003857 min_lr: 0.003857 loss: 2.3077 (2.3460) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [72] [ 50/312] eta: 0:03:46 lr: 0.003857 min_lr: 0.003857 loss: 2.4729 (2.4012) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [72] [ 60/312] eta: 0:03:30 lr: 0.003856 min_lr: 0.003856 loss: 2.4729 (2.4037) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [72] [ 70/312] eta: 0:03:17 lr: 0.003856 min_lr: 0.003856 loss: 2.3987 (2.3966) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [72] [ 80/312] eta: 0:03:06 lr: 0.003856 min_lr: 0.003856 loss: 2.3516 (2.3814) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [72] [ 90/312] eta: 0:02:55 lr: 0.003856 min_lr: 0.003856 loss: 2.3420 (2.3859) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [72] [100/312] eta: 0:02:45 lr: 0.003856 min_lr: 0.003856 loss: 2.3420 (2.3721) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [72] [110/312] eta: 0:02:36 lr: 0.003856 min_lr: 0.003856 loss: 2.3585 (2.3841) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [72] [120/312] eta: 0:02:27 lr: 0.003855 min_lr: 0.003855 loss: 2.5212 (2.3825) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [72] [130/312] eta: 0:02:18 lr: 0.003855 min_lr: 0.003855 loss: 2.2158 (2.3576) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [72] [140/312] eta: 0:02:10 lr: 0.003855 min_lr: 0.003855 loss: 2.2719 (2.3688) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [72] [150/312] eta: 0:02:02 lr: 0.003855 min_lr: 0.003855 loss: 2.5433 (2.3739) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [72] [160/312] eta: 0:01:53 lr: 0.003855 min_lr: 0.003855 loss: 2.5906 (2.3783) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [72] [170/312] eta: 0:01:45 lr: 0.003854 min_lr: 0.003854 loss: 2.6303 (2.3906) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [72] [180/312] eta: 0:01:38 lr: 0.003854 min_lr: 0.003854 loss: 2.5989 (2.3921) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [72] [190/312] eta: 0:01:30 lr: 0.003854 min_lr: 0.003854 loss: 2.5578 (2.4026) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [72] [200/312] eta: 0:01:22 lr: 0.003854 min_lr: 0.003854 loss: 2.5491 (2.4019) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [72] [210/312] eta: 0:01:15 lr: 0.003854 min_lr: 0.003854 loss: 2.2151 (2.3943) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [72] [220/312] eta: 0:01:07 lr: 0.003854 min_lr: 0.003854 loss: 2.1720 (2.3881) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [72] [230/312] eta: 0:01:00 lr: 0.003853 min_lr: 0.003853 loss: 2.2908 (2.3782) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [72] [240/312] eta: 0:00:52 lr: 0.003853 min_lr: 0.003853 loss: 2.0616 (2.3766) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [72] [250/312] eta: 0:00:45 lr: 0.003853 min_lr: 0.003853 loss: 2.3555 (2.3730) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [72] [260/312] eta: 0:00:37 lr: 0.003853 min_lr: 0.003853 loss: 2.4160 (2.3732) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [72] [270/312] eta: 0:00:30 lr: 0.003853 min_lr: 0.003853 loss: 2.3870 (2.3757) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [72] [280/312] eta: 0:00:23 lr: 0.003853 min_lr: 0.003853 loss: 2.4326 (2.3803) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [72] [290/312] eta: 0:00:15 lr: 0.003852 min_lr: 0.003852 loss: 2.4940 (2.3820) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0008 max mem: 64948 Epoch: [72] [300/312] eta: 0:00:08 lr: 0.003852 min_lr: 0.003852 loss: 2.5113 (2.3832) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [72] [310/312] eta: 0:00:01 lr: 0.003852 min_lr: 0.003852 loss: 2.4644 (2.3781) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [72] [311/312] eta: 0:00:00 lr: 0.003852 min_lr: 0.003852 loss: 2.4644 (2.3774) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [72] Total time: 0:03:46 (0.7259 s / it) Averaged stats: lr: 0.003852 min_lr: 0.003852 loss: 2.4644 (2.3831) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7797 (0.7797) acc1: 79.9479 (79.9479) acc5: 94.7917 (94.7917) time: 4.5792 data: 4.3690 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1128 (1.0545) acc1: 74.4792 (73.5040) acc5: 93.2292 (91.9040) time: 0.6601 data: 0.4855 max mem: 64948 Test: Total time: 0:00:06 (0.6790 s / it) * Acc@1 74.202 Acc@5 91.978 loss 1.031 Accuracy of the model on the 50000 test images: 74.2% Max accuracy: 74.35% Test: [0/9] eta: 0:00:44 loss: 2.0347 (2.0347) acc1: 54.1667 (54.1667) acc5: 77.3438 (77.3438) time: 4.9734 data: 4.7658 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.2018 (2.1746) acc1: 50.5208 (51.6800) acc5: 73.4375 (75.3920) time: 0.7043 data: 0.5296 max mem: 64948 Test: Total time: 0:00:06 (0.7120 s / it) * Acc@1 49.682 Acc@5 73.882 loss 2.313 Accuracy of the model EMA on 50000 test images: 49.7% Max EMA accuracy: 49.68% Epoch: [73] [ 0/312] eta: 0:45:11 lr: 0.003852 min_lr: 0.003852 loss: 3.0287 (3.0287) weight_decay: 0.0500 (0.0500) time: 8.6919 data: 7.9198 max mem: 64948 Epoch: [73] [ 10/312] eta: 0:07:18 lr: 0.003852 min_lr: 0.003852 loss: 2.5762 (2.5958) weight_decay: 0.0500 (0.0500) time: 1.4513 data: 0.7203 max mem: 64948 Epoch: [73] [ 20/312] eta: 0:05:18 lr: 0.003852 min_lr: 0.003852 loss: 2.4869 (2.4341) weight_decay: 0.0500 (0.0500) time: 0.7117 data: 0.0003 max mem: 64948 Epoch: [73] [ 30/312] eta: 0:04:31 lr: 0.003851 min_lr: 0.003851 loss: 2.5596 (2.5012) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [73] [ 40/312] eta: 0:04:04 lr: 0.003851 min_lr: 0.003851 loss: 2.6246 (2.4781) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [73] [ 50/312] eta: 0:03:45 lr: 0.003851 min_lr: 0.003851 loss: 2.3520 (2.4143) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [73] [ 60/312] eta: 0:03:29 lr: 0.003851 min_lr: 0.003851 loss: 2.2186 (2.4068) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [73] [ 70/312] eta: 0:03:17 lr: 0.003851 min_lr: 0.003851 loss: 2.4035 (2.4046) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [73] [ 80/312] eta: 0:03:05 lr: 0.003851 min_lr: 0.003851 loss: 2.5260 (2.4100) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [73] [ 90/312] eta: 0:02:54 lr: 0.003850 min_lr: 0.003850 loss: 2.4347 (2.3954) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [73] [100/312] eta: 0:02:45 lr: 0.003850 min_lr: 0.003850 loss: 2.1371 (2.3722) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [73] [110/312] eta: 0:02:35 lr: 0.003850 min_lr: 0.003850 loss: 2.2191 (2.3697) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [73] [120/312] eta: 0:02:26 lr: 0.003850 min_lr: 0.003850 loss: 2.4325 (2.3743) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [73] [130/312] eta: 0:02:18 lr: 0.003850 min_lr: 0.003850 loss: 2.5260 (2.3685) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [73] [140/312] eta: 0:02:09 lr: 0.003849 min_lr: 0.003849 loss: 2.3822 (2.3797) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [73] [150/312] eta: 0:02:01 lr: 0.003849 min_lr: 0.003849 loss: 2.3571 (2.3731) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [73] [160/312] eta: 0:01:53 lr: 0.003849 min_lr: 0.003849 loss: 2.3810 (2.3729) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [73] [170/312] eta: 0:01:45 lr: 0.003849 min_lr: 0.003849 loss: 2.4505 (2.3758) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [73] [180/312] eta: 0:01:37 lr: 0.003849 min_lr: 0.003849 loss: 2.4726 (2.3730) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [73] [190/312] eta: 0:01:30 lr: 0.003849 min_lr: 0.003849 loss: 2.4737 (2.3770) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [73] [200/312] eta: 0:01:22 lr: 0.003848 min_lr: 0.003848 loss: 2.3471 (2.3735) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [73] [210/312] eta: 0:01:14 lr: 0.003848 min_lr: 0.003848 loss: 2.2897 (2.3755) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [73] [220/312] eta: 0:01:07 lr: 0.003848 min_lr: 0.003848 loss: 2.5419 (2.3787) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [73] [230/312] eta: 0:00:59 lr: 0.003848 min_lr: 0.003848 loss: 2.5763 (2.3810) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [73] [240/312] eta: 0:00:52 lr: 0.003848 min_lr: 0.003848 loss: 2.6104 (2.3888) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [73] [250/312] eta: 0:00:45 lr: 0.003848 min_lr: 0.003848 loss: 2.4497 (2.3834) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [73] [260/312] eta: 0:00:37 lr: 0.003847 min_lr: 0.003847 loss: 2.2750 (2.3814) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [73] [270/312] eta: 0:00:30 lr: 0.003847 min_lr: 0.003847 loss: 2.4130 (2.3810) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [73] [280/312] eta: 0:00:23 lr: 0.003847 min_lr: 0.003847 loss: 2.2175 (2.3743) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [73] [290/312] eta: 0:00:15 lr: 0.003847 min_lr: 0.003847 loss: 2.1881 (2.3714) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [73] [300/312] eta: 0:00:08 lr: 0.003847 min_lr: 0.003847 loss: 2.3194 (2.3691) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [73] [310/312] eta: 0:00:01 lr: 0.003846 min_lr: 0.003846 loss: 2.5255 (2.3766) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [73] [311/312] eta: 0:00:00 lr: 0.003846 min_lr: 0.003846 loss: 2.5247 (2.3741) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [73] Total time: 0:03:46 (0.7247 s / it) Averaged stats: lr: 0.003846 min_lr: 0.003846 loss: 2.5247 (2.3914) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.8949 (0.8949) acc1: 79.1667 (79.1667) acc5: 93.4896 (93.4896) time: 4.9026 data: 4.6963 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1622 (1.0965) acc1: 72.6562 (72.6080) acc5: 92.1875 (91.2000) time: 0.6960 data: 0.5219 max mem: 64948 Test: Total time: 0:00:06 (0.7301 s / it) * Acc@1 73.514 Acc@5 91.590 loss 1.085 Accuracy of the model on the 50000 test images: 73.5% Max accuracy: 74.35% Test: [0/9] eta: 0:00:41 loss: 1.9033 (1.9033) acc1: 57.5521 (57.5521) acc5: 78.1250 (78.1250) time: 4.6044 data: 4.3865 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 2.0674 (2.0537) acc1: 53.6458 (54.2400) acc5: 76.0417 (76.9600) time: 0.6632 data: 0.4875 max mem: 64948 Test: Total time: 0:00:06 (0.6724 s / it) * Acc@1 52.006 Acc@5 76.036 loss 2.175 Accuracy of the model EMA on 50000 test images: 52.0% Max EMA accuracy: 52.01% Epoch: [74] [ 0/312] eta: 0:50:32 lr: 0.003846 min_lr: 0.003846 loss: 2.2767 (2.2767) weight_decay: 0.0500 (0.0500) time: 9.7186 data: 8.9150 max mem: 64948 Epoch: [74] [ 10/312] eta: 0:07:44 lr: 0.003846 min_lr: 0.003846 loss: 2.4810 (2.3457) weight_decay: 0.0500 (0.0500) time: 1.5392 data: 0.8108 max mem: 64948 Epoch: [74] [ 20/312] eta: 0:05:32 lr: 0.003846 min_lr: 0.003846 loss: 2.5792 (2.4428) weight_decay: 0.0500 (0.0500) time: 0.7098 data: 0.0004 max mem: 64948 Epoch: [74] [ 30/312] eta: 0:04:40 lr: 0.003846 min_lr: 0.003846 loss: 2.5055 (2.3623) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [74] [ 40/312] eta: 0:04:11 lr: 0.003846 min_lr: 0.003846 loss: 2.3080 (2.3802) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [74] [ 50/312] eta: 0:03:50 lr: 0.003845 min_lr: 0.003845 loss: 2.3080 (2.3633) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [74] [ 60/312] eta: 0:03:33 lr: 0.003845 min_lr: 0.003845 loss: 2.4094 (2.3731) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [74] [ 70/312] eta: 0:03:20 lr: 0.003845 min_lr: 0.003845 loss: 2.4533 (2.3776) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [74] [ 80/312] eta: 0:03:08 lr: 0.003845 min_lr: 0.003845 loss: 2.3655 (2.3541) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [74] [ 90/312] eta: 0:02:57 lr: 0.003845 min_lr: 0.003845 loss: 2.3655 (2.3604) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [74] [100/312] eta: 0:02:47 lr: 0.003845 min_lr: 0.003845 loss: 2.3719 (2.3480) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [74] [110/312] eta: 0:02:37 lr: 0.003844 min_lr: 0.003844 loss: 2.5084 (2.3702) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [74] [120/312] eta: 0:02:28 lr: 0.003844 min_lr: 0.003844 loss: 2.5519 (2.3652) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [74] [130/312] eta: 0:02:19 lr: 0.003844 min_lr: 0.003844 loss: 2.4939 (2.3733) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [74] [140/312] eta: 0:02:10 lr: 0.003844 min_lr: 0.003844 loss: 2.5763 (2.3817) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [74] [150/312] eta: 0:02:02 lr: 0.003844 min_lr: 0.003844 loss: 2.4839 (2.3910) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [74] [160/312] eta: 0:01:54 lr: 0.003844 min_lr: 0.003844 loss: 2.4150 (2.3857) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [74] [170/312] eta: 0:01:46 lr: 0.003843 min_lr: 0.003843 loss: 2.3502 (2.3852) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [74] [180/312] eta: 0:01:38 lr: 0.003843 min_lr: 0.003843 loss: 2.4626 (2.3954) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [74] [190/312] eta: 0:01:30 lr: 0.003843 min_lr: 0.003843 loss: 2.5024 (2.3953) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [74] [200/312] eta: 0:01:23 lr: 0.003843 min_lr: 0.003843 loss: 2.2638 (2.3797) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [74] [210/312] eta: 0:01:15 lr: 0.003843 min_lr: 0.003843 loss: 2.3014 (2.3772) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [74] [220/312] eta: 0:01:07 lr: 0.003842 min_lr: 0.003842 loss: 2.4781 (2.3811) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [74] [230/312] eta: 0:01:00 lr: 0.003842 min_lr: 0.003842 loss: 2.5818 (2.3856) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [74] [240/312] eta: 0:00:52 lr: 0.003842 min_lr: 0.003842 loss: 2.4821 (2.3871) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [74] [250/312] eta: 0:00:45 lr: 0.003842 min_lr: 0.003842 loss: 2.3985 (2.3829) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [74] [260/312] eta: 0:00:37 lr: 0.003842 min_lr: 0.003842 loss: 2.3985 (2.3881) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [74] [270/312] eta: 0:00:30 lr: 0.003842 min_lr: 0.003842 loss: 2.4784 (2.3847) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [74] [280/312] eta: 0:00:23 lr: 0.003841 min_lr: 0.003841 loss: 2.4784 (2.3860) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [74] [290/312] eta: 0:00:15 lr: 0.003841 min_lr: 0.003841 loss: 2.4952 (2.3835) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [74] [300/312] eta: 0:00:08 lr: 0.003841 min_lr: 0.003841 loss: 2.4946 (2.3822) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [74] [310/312] eta: 0:00:01 lr: 0.003841 min_lr: 0.003841 loss: 2.3909 (2.3801) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [74] [311/312] eta: 0:00:00 lr: 0.003841 min_lr: 0.003841 loss: 2.0105 (2.3775) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [74] Total time: 0:03:46 (0.7272 s / it) Averaged stats: lr: 0.003841 min_lr: 0.003841 loss: 2.0105 (2.3873) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7385 (0.7385) acc1: 82.5521 (82.5521) acc5: 94.5312 (94.5312) time: 4.5437 data: 4.3384 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1388 (1.0557) acc1: 73.4375 (73.7600) acc5: 92.4528 (91.9680) time: 0.6562 data: 0.4822 max mem: 64948 Test: Total time: 0:00:06 (0.6788 s / it) * Acc@1 74.142 Acc@5 92.008 loss 1.050 Accuracy of the model on the 50000 test images: 74.1% Max accuracy: 74.35% Test: [0/9] eta: 0:00:40 loss: 1.7860 (1.7860) acc1: 59.3750 (59.3750) acc5: 79.6875 (79.6875) time: 4.4613 data: 4.2487 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.9467 (1.9476) acc1: 56.5104 (55.6800) acc5: 77.0833 (78.1120) time: 0.6511 data: 0.4762 max mem: 64948 Test: Total time: 0:00:05 (0.6599 s / it) * Acc@1 54.240 Acc@5 77.932 loss 2.052 Accuracy of the model EMA on 50000 test images: 54.2% Max EMA accuracy: 54.24% Epoch: [75] [ 0/312] eta: 0:47:46 lr: 0.003841 min_lr: 0.003841 loss: 2.6185 (2.6185) weight_decay: 0.0500 (0.0500) time: 9.1871 data: 8.3913 max mem: 64948 Epoch: [75] [ 10/312] eta: 0:07:46 lr: 0.003841 min_lr: 0.003841 loss: 2.3663 (2.2456) weight_decay: 0.0500 (0.0500) time: 1.5433 data: 0.8267 max mem: 64948 Epoch: [75] [ 20/312] eta: 0:05:32 lr: 0.003840 min_lr: 0.003840 loss: 2.1590 (2.2327) weight_decay: 0.0500 (0.0500) time: 0.7357 data: 0.0353 max mem: 64948 Epoch: [75] [ 30/312] eta: 0:04:40 lr: 0.003840 min_lr: 0.003840 loss: 2.4394 (2.2397) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [75] [ 40/312] eta: 0:04:10 lr: 0.003840 min_lr: 0.003840 loss: 2.3445 (2.2569) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [75] [ 50/312] eta: 0:03:50 lr: 0.003840 min_lr: 0.003840 loss: 2.3428 (2.2814) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [75] [ 60/312] eta: 0:03:33 lr: 0.003840 min_lr: 0.003840 loss: 2.4675 (2.3212) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [75] [ 70/312] eta: 0:03:20 lr: 0.003839 min_lr: 0.003839 loss: 2.5319 (2.3241) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [75] [ 80/312] eta: 0:03:08 lr: 0.003839 min_lr: 0.003839 loss: 2.3776 (2.3350) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [75] [ 90/312] eta: 0:02:57 lr: 0.003839 min_lr: 0.003839 loss: 2.3776 (2.3341) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [75] [100/312] eta: 0:02:47 lr: 0.003839 min_lr: 0.003839 loss: 2.2433 (2.3101) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [75] [110/312] eta: 0:02:37 lr: 0.003839 min_lr: 0.003839 loss: 2.0385 (2.3170) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [75] [120/312] eta: 0:02:28 lr: 0.003839 min_lr: 0.003839 loss: 2.4971 (2.3221) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [75] [130/312] eta: 0:02:19 lr: 0.003838 min_lr: 0.003838 loss: 2.4058 (2.3254) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [75] [140/312] eta: 0:02:11 lr: 0.003838 min_lr: 0.003838 loss: 2.4465 (2.3260) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [75] [150/312] eta: 0:02:02 lr: 0.003838 min_lr: 0.003838 loss: 2.4233 (2.3350) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [75] [160/312] eta: 0:01:54 lr: 0.003838 min_lr: 0.003838 loss: 2.3714 (2.3458) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [75] [170/312] eta: 0:01:46 lr: 0.003838 min_lr: 0.003838 loss: 2.4698 (2.3486) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [75] [180/312] eta: 0:01:38 lr: 0.003837 min_lr: 0.003837 loss: 2.4653 (2.3560) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [75] [190/312] eta: 0:01:30 lr: 0.003837 min_lr: 0.003837 loss: 2.4196 (2.3581) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [75] [200/312] eta: 0:01:23 lr: 0.003837 min_lr: 0.003837 loss: 2.4196 (2.3644) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [75] [210/312] eta: 0:01:15 lr: 0.003837 min_lr: 0.003837 loss: 2.5906 (2.3758) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [75] [220/312] eta: 0:01:07 lr: 0.003837 min_lr: 0.003837 loss: 2.4972 (2.3786) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [75] [230/312] eta: 0:01:00 lr: 0.003836 min_lr: 0.003836 loss: 2.4247 (2.3770) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [75] [240/312] eta: 0:00:52 lr: 0.003836 min_lr: 0.003836 loss: 2.4269 (2.3782) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [75] [250/312] eta: 0:00:45 lr: 0.003836 min_lr: 0.003836 loss: 2.3646 (2.3742) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [75] [260/312] eta: 0:00:38 lr: 0.003836 min_lr: 0.003836 loss: 2.4142 (2.3717) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [75] [270/312] eta: 0:00:30 lr: 0.003836 min_lr: 0.003836 loss: 2.5082 (2.3765) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [75] [280/312] eta: 0:00:23 lr: 0.003836 min_lr: 0.003836 loss: 2.6285 (2.3854) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [75] [290/312] eta: 0:00:15 lr: 0.003835 min_lr: 0.003835 loss: 2.6285 (2.3842) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [75] [300/312] eta: 0:00:08 lr: 0.003835 min_lr: 0.003835 loss: 2.3023 (2.3831) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [75] [310/312] eta: 0:00:01 lr: 0.003835 min_lr: 0.003835 loss: 2.3359 (2.3817) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [75] [311/312] eta: 0:00:00 lr: 0.003835 min_lr: 0.003835 loss: 2.3359 (2.3823) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [75] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.003835 min_lr: 0.003835 loss: 2.3359 (2.3831) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.8546 (0.8546) acc1: 80.7292 (80.7292) acc5: 92.7083 (92.7083) time: 4.5442 data: 4.3246 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0890 (1.0515) acc1: 74.4792 (73.5360) acc5: 91.6667 (91.5520) time: 0.6568 data: 0.4806 max mem: 64948 Test: Total time: 0:00:06 (0.6811 s / it) * Acc@1 73.968 Acc@5 91.802 loss 1.030 Accuracy of the model on the 50000 test images: 74.0% Max accuracy: 74.35% Test: [0/9] eta: 0:00:45 loss: 1.6815 (1.6815) acc1: 61.9792 (61.9792) acc5: 82.2917 (82.2917) time: 5.0345 data: 4.8166 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.8407 (1.8536) acc1: 57.5521 (57.5680) acc5: 79.6875 (80.0000) time: 0.7107 data: 0.5353 max mem: 64948 Test: Total time: 0:00:06 (0.7187 s / it) * Acc@1 56.178 Acc@5 79.696 loss 1.943 Accuracy of the model EMA on 50000 test images: 56.2% Max EMA accuracy: 56.18% Epoch: [76] [ 0/312] eta: 0:43:32 lr: 0.003835 min_lr: 0.003835 loss: 2.5450 (2.5450) weight_decay: 0.0500 (0.0500) time: 8.3741 data: 7.1236 max mem: 64948 Epoch: [76] [ 10/312] eta: 0:07:27 lr: 0.003835 min_lr: 0.003835 loss: 2.5450 (2.4926) weight_decay: 0.0500 (0.0500) time: 1.4813 data: 0.7106 max mem: 64948 Epoch: [76] [ 20/312] eta: 0:05:23 lr: 0.003835 min_lr: 0.003835 loss: 2.3483 (2.3735) weight_decay: 0.0500 (0.0500) time: 0.7446 data: 0.0348 max mem: 64948 Epoch: [76] [ 30/312] eta: 0:04:34 lr: 0.003834 min_lr: 0.003834 loss: 2.3103 (2.3679) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [76] [ 40/312] eta: 0:04:06 lr: 0.003834 min_lr: 0.003834 loss: 2.3456 (2.3607) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [76] [ 50/312] eta: 0:03:46 lr: 0.003834 min_lr: 0.003834 loss: 2.3500 (2.3618) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0003 max mem: 64948 Epoch: [76] [ 60/312] eta: 0:03:31 lr: 0.003834 min_lr: 0.003834 loss: 2.3928 (2.3666) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [76] [ 70/312] eta: 0:03:17 lr: 0.003834 min_lr: 0.003834 loss: 2.2596 (2.3562) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [76] [ 80/312] eta: 0:03:06 lr: 0.003833 min_lr: 0.003833 loss: 2.2248 (2.3353) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [76] [ 90/312] eta: 0:02:55 lr: 0.003833 min_lr: 0.003833 loss: 2.2727 (2.3296) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [76] [100/312] eta: 0:02:45 lr: 0.003833 min_lr: 0.003833 loss: 2.3820 (2.3294) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [76] [110/312] eta: 0:02:36 lr: 0.003833 min_lr: 0.003833 loss: 2.3820 (2.3279) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [76] [120/312] eta: 0:02:27 lr: 0.003833 min_lr: 0.003833 loss: 2.4581 (2.3443) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [76] [130/312] eta: 0:02:18 lr: 0.003833 min_lr: 0.003833 loss: 2.5369 (2.3445) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [76] [140/312] eta: 0:02:10 lr: 0.003832 min_lr: 0.003832 loss: 2.2370 (2.3368) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [76] [150/312] eta: 0:02:01 lr: 0.003832 min_lr: 0.003832 loss: 2.3391 (2.3454) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [76] [160/312] eta: 0:01:53 lr: 0.003832 min_lr: 0.003832 loss: 2.5695 (2.3576) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [76] [170/312] eta: 0:01:45 lr: 0.003832 min_lr: 0.003832 loss: 2.5077 (2.3564) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [76] [180/312] eta: 0:01:38 lr: 0.003832 min_lr: 0.003832 loss: 2.5077 (2.3620) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [76] [190/312] eta: 0:01:30 lr: 0.003831 min_lr: 0.003831 loss: 2.5015 (2.3622) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [76] [200/312] eta: 0:01:22 lr: 0.003831 min_lr: 0.003831 loss: 2.4014 (2.3589) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [76] [210/312] eta: 0:01:15 lr: 0.003831 min_lr: 0.003831 loss: 2.4931 (2.3639) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [76] [220/312] eta: 0:01:07 lr: 0.003831 min_lr: 0.003831 loss: 2.4931 (2.3686) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [76] [230/312] eta: 0:01:00 lr: 0.003831 min_lr: 0.003831 loss: 2.3788 (2.3706) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [76] [240/312] eta: 0:00:52 lr: 0.003830 min_lr: 0.003830 loss: 2.3788 (2.3729) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [76] [250/312] eta: 0:00:45 lr: 0.003830 min_lr: 0.003830 loss: 2.5075 (2.3785) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [76] [260/312] eta: 0:00:37 lr: 0.003830 min_lr: 0.003830 loss: 2.4476 (2.3747) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [76] [270/312] eta: 0:00:30 lr: 0.003830 min_lr: 0.003830 loss: 2.4476 (2.3793) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [76] [280/312] eta: 0:00:23 lr: 0.003830 min_lr: 0.003830 loss: 2.3819 (2.3781) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0009 max mem: 64948 Epoch: [76] [290/312] eta: 0:00:15 lr: 0.003830 min_lr: 0.003830 loss: 2.4933 (2.3804) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [76] [300/312] eta: 0:00:08 lr: 0.003829 min_lr: 0.003829 loss: 2.5829 (2.3874) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [76] [310/312] eta: 0:00:01 lr: 0.003829 min_lr: 0.003829 loss: 2.5060 (2.3829) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [76] [311/312] eta: 0:00:00 lr: 0.003829 min_lr: 0.003829 loss: 2.5208 (2.3845) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [76] Total time: 0:03:46 (0.7257 s / it) Averaged stats: lr: 0.003829 min_lr: 0.003829 loss: 2.5208 (2.3806) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.7963 (0.7963) acc1: 79.9479 (79.9479) acc5: 93.2292 (93.2292) time: 4.3148 data: 4.1019 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0741 (1.0261) acc1: 72.9167 (73.3120) acc5: 93.2292 (91.8080) time: 0.6308 data: 0.4559 max mem: 64948 Test: Total time: 0:00:05 (0.6526 s / it) * Acc@1 74.664 Acc@5 92.140 loss 0.992 Accuracy of the model on the 50000 test images: 74.7% Max accuracy: 74.66% Test: [0/9] eta: 0:00:39 loss: 1.5915 (1.5915) acc1: 65.1042 (65.1042) acc5: 83.3333 (83.3333) time: 4.3434 data: 4.1255 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.7520 (1.7744) acc1: 60.4167 (59.2320) acc5: 80.9896 (81.0880) time: 0.6339 data: 0.4585 max mem: 64948 Test: Total time: 0:00:05 (0.6431 s / it) * Acc@1 57.834 Acc@5 81.096 loss 1.849 Accuracy of the model EMA on 50000 test images: 57.8% Max EMA accuracy: 57.83% Epoch: [77] [ 0/312] eta: 0:50:54 lr: 0.003829 min_lr: 0.003829 loss: 2.9686 (2.9686) weight_decay: 0.0500 (0.0500) time: 9.7896 data: 9.0097 max mem: 64948 Epoch: [77] [ 10/312] eta: 0:07:44 lr: 0.003829 min_lr: 0.003829 loss: 2.0078 (2.0847) weight_decay: 0.0500 (0.0500) time: 1.5385 data: 0.8194 max mem: 64948 Epoch: [77] [ 20/312] eta: 0:05:31 lr: 0.003829 min_lr: 0.003829 loss: 2.2355 (2.2336) weight_decay: 0.0500 (0.0500) time: 0.7041 data: 0.0003 max mem: 64948 Epoch: [77] [ 30/312] eta: 0:04:40 lr: 0.003829 min_lr: 0.003829 loss: 2.4918 (2.3082) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [77] [ 40/312] eta: 0:04:10 lr: 0.003828 min_lr: 0.003828 loss: 2.4819 (2.2995) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [77] [ 50/312] eta: 0:03:49 lr: 0.003828 min_lr: 0.003828 loss: 2.3077 (2.3005) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [77] [ 60/312] eta: 0:03:33 lr: 0.003828 min_lr: 0.003828 loss: 2.3529 (2.2950) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [77] [ 70/312] eta: 0:03:20 lr: 0.003828 min_lr: 0.003828 loss: 2.5397 (2.3403) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [77] [ 80/312] eta: 0:03:08 lr: 0.003828 min_lr: 0.003828 loss: 2.5510 (2.3463) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [77] [ 90/312] eta: 0:02:57 lr: 0.003827 min_lr: 0.003827 loss: 2.4349 (2.3494) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [77] [100/312] eta: 0:02:47 lr: 0.003827 min_lr: 0.003827 loss: 2.4349 (2.3348) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [77] [110/312] eta: 0:02:37 lr: 0.003827 min_lr: 0.003827 loss: 2.3918 (2.3359) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [77] [120/312] eta: 0:02:28 lr: 0.003827 min_lr: 0.003827 loss: 2.4170 (2.3510) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [77] [130/312] eta: 0:02:19 lr: 0.003827 min_lr: 0.003827 loss: 2.4828 (2.3513) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [77] [140/312] eta: 0:02:10 lr: 0.003826 min_lr: 0.003826 loss: 2.5250 (2.3681) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [77] [150/312] eta: 0:02:02 lr: 0.003826 min_lr: 0.003826 loss: 2.4923 (2.3559) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [77] [160/312] eta: 0:01:54 lr: 0.003826 min_lr: 0.003826 loss: 2.4338 (2.3680) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [77] [170/312] eta: 0:01:46 lr: 0.003826 min_lr: 0.003826 loss: 2.4559 (2.3653) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [77] [180/312] eta: 0:01:38 lr: 0.003826 min_lr: 0.003826 loss: 2.5015 (2.3691) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [77] [190/312] eta: 0:01:30 lr: 0.003825 min_lr: 0.003825 loss: 2.5299 (2.3701) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [77] [200/312] eta: 0:01:23 lr: 0.003825 min_lr: 0.003825 loss: 2.5022 (2.3680) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [77] [210/312] eta: 0:01:15 lr: 0.003825 min_lr: 0.003825 loss: 2.3308 (2.3600) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [77] [220/312] eta: 0:01:07 lr: 0.003825 min_lr: 0.003825 loss: 2.3038 (2.3614) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [77] [230/312] eta: 0:01:00 lr: 0.003825 min_lr: 0.003825 loss: 2.2926 (2.3606) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [77] [240/312] eta: 0:00:52 lr: 0.003825 min_lr: 0.003825 loss: 2.4876 (2.3673) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [77] [250/312] eta: 0:00:45 lr: 0.003824 min_lr: 0.003824 loss: 2.4846 (2.3700) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [77] [260/312] eta: 0:00:37 lr: 0.003824 min_lr: 0.003824 loss: 2.4125 (2.3685) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [77] [270/312] eta: 0:00:30 lr: 0.003824 min_lr: 0.003824 loss: 2.3530 (2.3667) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [77] [280/312] eta: 0:00:23 lr: 0.003824 min_lr: 0.003824 loss: 2.4256 (2.3721) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [77] [290/312] eta: 0:00:15 lr: 0.003824 min_lr: 0.003824 loss: 2.5204 (2.3762) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [77] [300/312] eta: 0:00:08 lr: 0.003823 min_lr: 0.003823 loss: 2.5173 (2.3792) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [77] [310/312] eta: 0:00:01 lr: 0.003823 min_lr: 0.003823 loss: 2.3789 (2.3758) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [77] [311/312] eta: 0:00:00 lr: 0.003823 min_lr: 0.003823 loss: 2.4247 (2.3765) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [77] Total time: 0:03:46 (0.7272 s / it) Averaged stats: lr: 0.003823 min_lr: 0.003823 loss: 2.4247 (2.3830) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.8844 (0.8844) acc1: 80.9896 (80.9896) acc5: 93.4896 (93.4896) time: 4.8866 data: 4.6673 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0741 (1.0195) acc1: 74.2188 (74.1120) acc5: 91.9271 (91.8720) time: 0.6949 data: 0.5187 max mem: 64948 Test: Total time: 0:00:06 (0.7157 s / it) * Acc@1 75.000 Acc@5 91.992 loss 1.011 Accuracy of the model on the 50000 test images: 75.0% Max accuracy: 75.00% Test: [0/9] eta: 0:00:41 loss: 1.5158 (1.5158) acc1: 67.4479 (67.4479) acc5: 84.1146 (84.1146) time: 4.5649 data: 4.3540 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.6738 (1.7042) acc1: 62.2396 (60.4800) acc5: 82.2917 (82.1760) time: 0.6591 data: 0.4839 max mem: 64948 Test: Total time: 0:00:06 (0.6673 s / it) * Acc@1 59.442 Acc@5 82.282 loss 1.767 Accuracy of the model EMA on 50000 test images: 59.4% Max EMA accuracy: 59.44% Epoch: [78] [ 0/312] eta: 0:48:19 lr: 0.003823 min_lr: 0.003823 loss: 1.5318 (1.5318) weight_decay: 0.0500 (0.0500) time: 9.2925 data: 8.4895 max mem: 64948 Epoch: [78] [ 10/312] eta: 0:07:42 lr: 0.003823 min_lr: 0.003823 loss: 2.4604 (2.3670) weight_decay: 0.0500 (0.0500) time: 1.5328 data: 0.7721 max mem: 64948 Epoch: [78] [ 20/312] eta: 0:05:30 lr: 0.003823 min_lr: 0.003823 loss: 2.4205 (2.3015) weight_decay: 0.0500 (0.0500) time: 0.7247 data: 0.0004 max mem: 64948 Epoch: [78] [ 30/312] eta: 0:04:39 lr: 0.003823 min_lr: 0.003823 loss: 2.3041 (2.3175) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [78] [ 40/312] eta: 0:04:09 lr: 0.003822 min_lr: 0.003822 loss: 2.2914 (2.3277) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [78] [ 50/312] eta: 0:03:49 lr: 0.003822 min_lr: 0.003822 loss: 2.0767 (2.2815) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [78] [ 60/312] eta: 0:03:33 lr: 0.003822 min_lr: 0.003822 loss: 2.1331 (2.3112) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [78] [ 70/312] eta: 0:03:19 lr: 0.003822 min_lr: 0.003822 loss: 2.4224 (2.3179) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [78] [ 80/312] eta: 0:03:07 lr: 0.003822 min_lr: 0.003822 loss: 2.4603 (2.3365) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [78] [ 90/312] eta: 0:02:57 lr: 0.003821 min_lr: 0.003821 loss: 2.3523 (2.3234) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [78] [100/312] eta: 0:02:46 lr: 0.003821 min_lr: 0.003821 loss: 2.3025 (2.3210) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [78] [110/312] eta: 0:02:37 lr: 0.003821 min_lr: 0.003821 loss: 2.2765 (2.3131) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [78] [120/312] eta: 0:02:28 lr: 0.003821 min_lr: 0.003821 loss: 2.0216 (2.3037) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [78] [130/312] eta: 0:02:19 lr: 0.003821 min_lr: 0.003821 loss: 2.1939 (2.3095) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [78] [140/312] eta: 0:02:10 lr: 0.003820 min_lr: 0.003820 loss: 2.4838 (2.3225) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [78] [150/312] eta: 0:02:02 lr: 0.003820 min_lr: 0.003820 loss: 2.3327 (2.3231) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [78] [160/312] eta: 0:01:54 lr: 0.003820 min_lr: 0.003820 loss: 2.4552 (2.3289) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [78] [170/312] eta: 0:01:46 lr: 0.003820 min_lr: 0.003820 loss: 2.4552 (2.3326) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [78] [180/312] eta: 0:01:38 lr: 0.003820 min_lr: 0.003820 loss: 2.4555 (2.3355) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [78] [190/312] eta: 0:01:30 lr: 0.003819 min_lr: 0.003819 loss: 2.4670 (2.3378) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [78] [200/312] eta: 0:01:22 lr: 0.003819 min_lr: 0.003819 loss: 2.5220 (2.3477) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [78] [210/312] eta: 0:01:15 lr: 0.003819 min_lr: 0.003819 loss: 2.5737 (2.3551) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [78] [220/312] eta: 0:01:07 lr: 0.003819 min_lr: 0.003819 loss: 2.5195 (2.3573) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [78] [230/312] eta: 0:01:00 lr: 0.003819 min_lr: 0.003819 loss: 2.4142 (2.3544) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [78] [240/312] eta: 0:00:52 lr: 0.003819 min_lr: 0.003819 loss: 2.5754 (2.3651) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [78] [250/312] eta: 0:00:45 lr: 0.003818 min_lr: 0.003818 loss: 2.5464 (2.3623) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [78] [260/312] eta: 0:00:37 lr: 0.003818 min_lr: 0.003818 loss: 2.3206 (2.3654) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [78] [270/312] eta: 0:00:30 lr: 0.003818 min_lr: 0.003818 loss: 2.3019 (2.3594) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [78] [280/312] eta: 0:00:23 lr: 0.003818 min_lr: 0.003818 loss: 2.1957 (2.3567) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [78] [290/312] eta: 0:00:15 lr: 0.003818 min_lr: 0.003818 loss: 2.4410 (2.3636) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [78] [300/312] eta: 0:00:08 lr: 0.003817 min_lr: 0.003817 loss: 2.5568 (2.3659) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [78] [310/312] eta: 0:00:01 lr: 0.003817 min_lr: 0.003817 loss: 2.4080 (2.3675) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [78] [311/312] eta: 0:00:00 lr: 0.003817 min_lr: 0.003817 loss: 2.4080 (2.3686) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [78] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.003817 min_lr: 0.003817 loss: 2.4080 (2.3657) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.8186 (0.8186) acc1: 81.5104 (81.5104) acc5: 93.2292 (93.2292) time: 4.6094 data: 4.3902 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1153 (1.0466) acc1: 73.9583 (72.8960) acc5: 91.9271 (91.9680) time: 0.6640 data: 0.4879 max mem: 64948 Test: Total time: 0:00:06 (0.6875 s / it) * Acc@1 73.852 Acc@5 92.036 loss 1.036 Accuracy of the model on the 50000 test images: 73.9% Max accuracy: 75.00% Test: [0/9] eta: 0:00:45 loss: 1.4452 (1.4452) acc1: 69.2708 (69.2708) acc5: 85.6771 (85.6771) time: 5.0211 data: 4.8073 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.6046 (1.6406) acc1: 63.5417 (61.5360) acc5: 83.0729 (83.1680) time: 0.7092 data: 0.5342 max mem: 64948 Test: Total time: 0:00:06 (0.7165 s / it) * Acc@1 60.822 Acc@5 83.352 loss 1.692 Accuracy of the model EMA on 50000 test images: 60.8% Max EMA accuracy: 60.82% Epoch: [79] [ 0/312] eta: 0:47:19 lr: 0.003817 min_lr: 0.003817 loss: 1.9548 (1.9548) weight_decay: 0.0500 (0.0500) time: 9.1004 data: 7.6194 max mem: 64948 Epoch: [79] [ 10/312] eta: 0:07:29 lr: 0.003817 min_lr: 0.003817 loss: 2.6812 (2.4743) weight_decay: 0.0500 (0.0500) time: 1.4886 data: 0.6933 max mem: 64948 Epoch: [79] [ 20/312] eta: 0:05:24 lr: 0.003817 min_lr: 0.003817 loss: 2.5751 (2.4457) weight_decay: 0.0500 (0.0500) time: 0.7102 data: 0.0005 max mem: 64948 Epoch: [79] [ 30/312] eta: 0:04:35 lr: 0.003817 min_lr: 0.003817 loss: 2.4319 (2.4082) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [79] [ 40/312] eta: 0:04:06 lr: 0.003816 min_lr: 0.003816 loss: 2.3019 (2.3957) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [79] [ 50/312] eta: 0:03:46 lr: 0.003816 min_lr: 0.003816 loss: 2.4706 (2.4015) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [79] [ 60/312] eta: 0:03:31 lr: 0.003816 min_lr: 0.003816 loss: 2.4287 (2.3811) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [79] [ 70/312] eta: 0:03:18 lr: 0.003816 min_lr: 0.003816 loss: 2.4879 (2.3906) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [79] [ 80/312] eta: 0:03:06 lr: 0.003816 min_lr: 0.003816 loss: 2.5188 (2.3912) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [79] [ 90/312] eta: 0:02:55 lr: 0.003815 min_lr: 0.003815 loss: 2.6005 (2.4215) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [79] [100/312] eta: 0:02:45 lr: 0.003815 min_lr: 0.003815 loss: 2.6124 (2.4427) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [79] [110/312] eta: 0:02:36 lr: 0.003815 min_lr: 0.003815 loss: 2.5258 (2.4442) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [79] [120/312] eta: 0:02:27 lr: 0.003815 min_lr: 0.003815 loss: 2.4171 (2.4407) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [79] [130/312] eta: 0:02:18 lr: 0.003815 min_lr: 0.003815 loss: 2.4509 (2.4478) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [79] [140/312] eta: 0:02:10 lr: 0.003814 min_lr: 0.003814 loss: 2.1042 (2.4155) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [79] [150/312] eta: 0:02:02 lr: 0.003814 min_lr: 0.003814 loss: 2.0755 (2.4083) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [79] [160/312] eta: 0:01:53 lr: 0.003814 min_lr: 0.003814 loss: 2.2663 (2.3988) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [79] [170/312] eta: 0:01:45 lr: 0.003814 min_lr: 0.003814 loss: 2.4177 (2.4005) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [79] [180/312] eta: 0:01:38 lr: 0.003814 min_lr: 0.003814 loss: 2.5083 (2.3953) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [79] [190/312] eta: 0:01:30 lr: 0.003813 min_lr: 0.003813 loss: 2.2601 (2.3921) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [79] [200/312] eta: 0:01:22 lr: 0.003813 min_lr: 0.003813 loss: 2.3810 (2.3918) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [79] [210/312] eta: 0:01:15 lr: 0.003813 min_lr: 0.003813 loss: 2.4178 (2.3962) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [79] [220/312] eta: 0:01:07 lr: 0.003813 min_lr: 0.003813 loss: 2.5194 (2.3981) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [79] [230/312] eta: 0:01:00 lr: 0.003813 min_lr: 0.003813 loss: 2.5063 (2.3942) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [79] [240/312] eta: 0:00:52 lr: 0.003812 min_lr: 0.003812 loss: 2.5297 (2.3998) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [79] [250/312] eta: 0:00:45 lr: 0.003812 min_lr: 0.003812 loss: 2.5830 (2.3913) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [79] [260/312] eta: 0:00:37 lr: 0.003812 min_lr: 0.003812 loss: 2.2261 (2.3881) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [79] [270/312] eta: 0:00:30 lr: 0.003812 min_lr: 0.003812 loss: 2.2261 (2.3840) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [79] [280/312] eta: 0:00:23 lr: 0.003812 min_lr: 0.003812 loss: 2.3007 (2.3854) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0007 max mem: 64948 Epoch: [79] [290/312] eta: 0:00:15 lr: 0.003811 min_lr: 0.003811 loss: 2.5306 (2.3827) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0005 max mem: 64948 Epoch: [79] [300/312] eta: 0:00:08 lr: 0.003811 min_lr: 0.003811 loss: 2.4782 (2.3861) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [79] [310/312] eta: 0:00:01 lr: 0.003811 min_lr: 0.003811 loss: 2.5007 (2.3934) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [79] [311/312] eta: 0:00:00 lr: 0.003811 min_lr: 0.003811 loss: 2.5007 (2.3944) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [79] Total time: 0:03:46 (0.7254 s / it) Averaged stats: lr: 0.003811 min_lr: 0.003811 loss: 2.5007 (2.3702) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.8231 (0.8231) acc1: 80.2083 (80.2083) acc5: 94.0104 (94.0104) time: 4.5622 data: 4.3563 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0969 (1.0530) acc1: 72.9167 (74.2080) acc5: 92.9688 (92.1600) time: 0.6583 data: 0.4841 max mem: 64948 Test: Total time: 0:00:06 (0.6843 s / it) * Acc@1 74.666 Acc@5 92.150 loss 1.041 Accuracy of the model on the 50000 test images: 74.7% Max accuracy: 75.00% Test: [0/9] eta: 0:00:42 loss: 1.3826 (1.3826) acc1: 69.2708 (69.2708) acc5: 86.7188 (86.7188) time: 4.7683 data: 4.5630 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.5443 (1.5835) acc1: 64.5833 (62.4640) acc5: 83.8542 (84.0960) time: 0.6815 data: 0.5071 max mem: 64948 Test: Total time: 0:00:06 (0.6910 s / it) * Acc@1 61.964 Acc@5 84.218 loss 1.625 Accuracy of the model EMA on 50000 test images: 62.0% Max EMA accuracy: 61.96% Epoch: [80] [ 0/312] eta: 0:48:19 lr: 0.003811 min_lr: 0.003811 loss: 2.6068 (2.6068) weight_decay: 0.0500 (0.0500) time: 9.2940 data: 8.0060 max mem: 64948 Epoch: [80] [ 10/312] eta: 0:07:51 lr: 0.003811 min_lr: 0.003811 loss: 2.4736 (2.3506) weight_decay: 0.0500 (0.0500) time: 1.5622 data: 0.7563 max mem: 64948 Epoch: [80] [ 20/312] eta: 0:05:35 lr: 0.003811 min_lr: 0.003811 loss: 2.4736 (2.4499) weight_decay: 0.0500 (0.0500) time: 0.7411 data: 0.0159 max mem: 64948 Epoch: [80] [ 30/312] eta: 0:04:42 lr: 0.003810 min_lr: 0.003810 loss: 2.3980 (2.3642) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [80] [ 40/312] eta: 0:04:12 lr: 0.003810 min_lr: 0.003810 loss: 2.4526 (2.3959) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0003 max mem: 64948 Epoch: [80] [ 50/312] eta: 0:03:51 lr: 0.003810 min_lr: 0.003810 loss: 2.4995 (2.4095) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0003 max mem: 64948 Epoch: [80] [ 60/312] eta: 0:03:35 lr: 0.003810 min_lr: 0.003810 loss: 2.4122 (2.3940) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0003 max mem: 64948 Epoch: [80] [ 70/312] eta: 0:03:21 lr: 0.003810 min_lr: 0.003810 loss: 2.4439 (2.3947) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [80] [ 80/312] eta: 0:03:09 lr: 0.003809 min_lr: 0.003809 loss: 2.5452 (2.4006) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [80] [ 90/312] eta: 0:02:57 lr: 0.003809 min_lr: 0.003809 loss: 2.4245 (2.3890) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [80] [100/312] eta: 0:02:47 lr: 0.003809 min_lr: 0.003809 loss: 2.2593 (2.3656) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [80] [110/312] eta: 0:02:38 lr: 0.003809 min_lr: 0.003809 loss: 2.0901 (2.3519) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [80] [120/312] eta: 0:02:28 lr: 0.003809 min_lr: 0.003809 loss: 2.2618 (2.3487) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [80] [130/312] eta: 0:02:19 lr: 0.003808 min_lr: 0.003808 loss: 2.4006 (2.3534) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [80] [140/312] eta: 0:02:11 lr: 0.003808 min_lr: 0.003808 loss: 2.4244 (2.3533) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [80] [150/312] eta: 0:02:02 lr: 0.003808 min_lr: 0.003808 loss: 2.3748 (2.3475) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [80] [160/312] eta: 0:01:54 lr: 0.003808 min_lr: 0.003808 loss: 1.9719 (2.3255) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [80] [170/312] eta: 0:01:46 lr: 0.003808 min_lr: 0.003808 loss: 2.0592 (2.3275) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [80] [180/312] eta: 0:01:38 lr: 0.003807 min_lr: 0.003807 loss: 2.4175 (2.3286) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [80] [190/312] eta: 0:01:30 lr: 0.003807 min_lr: 0.003807 loss: 2.4175 (2.3264) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [80] [200/312] eta: 0:01:23 lr: 0.003807 min_lr: 0.003807 loss: 2.3050 (2.3235) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [80] [210/312] eta: 0:01:15 lr: 0.003807 min_lr: 0.003807 loss: 2.3931 (2.3299) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [80] [220/312] eta: 0:01:07 lr: 0.003807 min_lr: 0.003807 loss: 2.5186 (2.3333) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [80] [230/312] eta: 0:01:00 lr: 0.003806 min_lr: 0.003806 loss: 2.2405 (2.3234) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [80] [240/312] eta: 0:00:52 lr: 0.003806 min_lr: 0.003806 loss: 2.4533 (2.3331) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [80] [250/312] eta: 0:00:45 lr: 0.003806 min_lr: 0.003806 loss: 2.5576 (2.3438) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [80] [260/312] eta: 0:00:38 lr: 0.003806 min_lr: 0.003806 loss: 2.5726 (2.3531) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [80] [270/312] eta: 0:00:30 lr: 0.003806 min_lr: 0.003806 loss: 2.5434 (2.3572) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [80] [280/312] eta: 0:00:23 lr: 0.003805 min_lr: 0.003805 loss: 2.5258 (2.3592) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [80] [290/312] eta: 0:00:16 lr: 0.003805 min_lr: 0.003805 loss: 2.5294 (2.3615) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [80] [300/312] eta: 0:00:08 lr: 0.003805 min_lr: 0.003805 loss: 2.3574 (2.3585) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [80] [310/312] eta: 0:00:01 lr: 0.003805 min_lr: 0.003805 loss: 2.2835 (2.3548) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [80] [311/312] eta: 0:00:00 lr: 0.003805 min_lr: 0.003805 loss: 2.2835 (2.3531) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [80] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.003805 min_lr: 0.003805 loss: 2.2835 (2.3781) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.8801 (0.8801) acc1: 81.2500 (81.2500) acc5: 93.4896 (93.4896) time: 4.5830 data: 4.3646 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0220 (1.0060) acc1: 73.6979 (74.7200) acc5: 92.4479 (92.5760) time: 0.6605 data: 0.4850 max mem: 64948 Test: Total time: 0:00:06 (0.6868 s / it) * Acc@1 74.976 Acc@5 92.440 loss 0.987 Accuracy of the model on the 50000 test images: 75.0% Max accuracy: 75.00% Test: [0/9] eta: 0:00:46 loss: 1.3259 (1.3259) acc1: 70.5729 (70.5729) acc5: 87.5000 (87.5000) time: 5.2153 data: 5.0122 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4879 (1.5302) acc1: 66.4062 (63.5200) acc5: 84.3750 (84.8960) time: 0.7310 data: 0.5570 max mem: 64948 Test: Total time: 0:00:06 (0.7382 s / it) * Acc@1 63.076 Acc@5 85.124 loss 1.563 Accuracy of the model EMA on 50000 test images: 63.1% Max EMA accuracy: 63.08% Epoch: [81] [ 0/312] eta: 0:47:16 lr: 0.003805 min_lr: 0.003805 loss: 2.9141 (2.9141) weight_decay: 0.0500 (0.0500) time: 9.0902 data: 8.2960 max mem: 64948 Epoch: [81] [ 10/312] eta: 0:07:30 lr: 0.003804 min_lr: 0.003804 loss: 2.6096 (2.5290) weight_decay: 0.0500 (0.0500) time: 1.4904 data: 0.7547 max mem: 64948 Epoch: [81] [ 20/312] eta: 0:05:24 lr: 0.003804 min_lr: 0.003804 loss: 2.4165 (2.4342) weight_decay: 0.0500 (0.0500) time: 0.7118 data: 0.0005 max mem: 64948 Epoch: [81] [ 30/312] eta: 0:04:35 lr: 0.003804 min_lr: 0.003804 loss: 2.4165 (2.4266) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [81] [ 40/312] eta: 0:04:06 lr: 0.003804 min_lr: 0.003804 loss: 2.4250 (2.4063) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [81] [ 50/312] eta: 0:03:46 lr: 0.003804 min_lr: 0.003804 loss: 2.3410 (2.3907) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [81] [ 60/312] eta: 0:03:31 lr: 0.003803 min_lr: 0.003803 loss: 2.3345 (2.3793) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [81] [ 70/312] eta: 0:03:17 lr: 0.003803 min_lr: 0.003803 loss: 2.4879 (2.3696) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [81] [ 80/312] eta: 0:03:06 lr: 0.003803 min_lr: 0.003803 loss: 2.5441 (2.3908) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [81] [ 90/312] eta: 0:02:55 lr: 0.003803 min_lr: 0.003803 loss: 2.5860 (2.3959) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [81] [100/312] eta: 0:02:45 lr: 0.003803 min_lr: 0.003803 loss: 2.4381 (2.3799) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [81] [110/312] eta: 0:02:36 lr: 0.003802 min_lr: 0.003802 loss: 2.3300 (2.3634) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [81] [120/312] eta: 0:02:27 lr: 0.003802 min_lr: 0.003802 loss: 2.3111 (2.3654) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [81] [130/312] eta: 0:02:18 lr: 0.003802 min_lr: 0.003802 loss: 2.3778 (2.3662) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [81] [140/312] eta: 0:02:10 lr: 0.003802 min_lr: 0.003802 loss: 2.3572 (2.3598) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [81] [150/312] eta: 0:02:01 lr: 0.003802 min_lr: 0.003802 loss: 2.3314 (2.3472) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [81] [160/312] eta: 0:01:53 lr: 0.003801 min_lr: 0.003801 loss: 2.3911 (2.3539) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [81] [170/312] eta: 0:01:45 lr: 0.003801 min_lr: 0.003801 loss: 2.5097 (2.3708) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [81] [180/312] eta: 0:01:38 lr: 0.003801 min_lr: 0.003801 loss: 2.5792 (2.3748) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [81] [190/312] eta: 0:01:30 lr: 0.003801 min_lr: 0.003801 loss: 2.3408 (2.3680) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [81] [200/312] eta: 0:01:22 lr: 0.003801 min_lr: 0.003801 loss: 2.0682 (2.3585) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [81] [210/312] eta: 0:01:15 lr: 0.003800 min_lr: 0.003800 loss: 2.4730 (2.3623) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [81] [220/312] eta: 0:01:07 lr: 0.003800 min_lr: 0.003800 loss: 2.5473 (2.3633) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [81] [230/312] eta: 0:01:00 lr: 0.003800 min_lr: 0.003800 loss: 2.0866 (2.3538) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [81] [240/312] eta: 0:00:52 lr: 0.003800 min_lr: 0.003800 loss: 2.1024 (2.3499) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [81] [250/312] eta: 0:00:45 lr: 0.003800 min_lr: 0.003800 loss: 2.3082 (2.3461) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [81] [260/312] eta: 0:00:37 lr: 0.003799 min_lr: 0.003799 loss: 2.0275 (2.3337) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [81] [270/312] eta: 0:00:30 lr: 0.003799 min_lr: 0.003799 loss: 2.0275 (2.3306) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [81] [280/312] eta: 0:00:23 lr: 0.003799 min_lr: 0.003799 loss: 2.3837 (2.3317) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [81] [290/312] eta: 0:00:15 lr: 0.003799 min_lr: 0.003799 loss: 2.4863 (2.3289) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [81] [300/312] eta: 0:00:08 lr: 0.003799 min_lr: 0.003799 loss: 2.4868 (2.3325) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [81] [310/312] eta: 0:00:01 lr: 0.003798 min_lr: 0.003798 loss: 2.4366 (2.3348) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [81] [311/312] eta: 0:00:00 lr: 0.003798 min_lr: 0.003798 loss: 2.4366 (2.3352) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [81] Total time: 0:03:46 (0.7251 s / it) Averaged stats: lr: 0.003798 min_lr: 0.003798 loss: 2.4366 (2.3512) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7527 (0.7527) acc1: 82.0312 (82.0312) acc5: 95.3125 (95.3125) time: 4.6000 data: 4.3754 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0139 (1.0109) acc1: 75.2604 (73.6000) acc5: 92.7083 (92.5120) time: 0.6628 data: 0.4862 max mem: 64948 Test: Total time: 0:00:06 (0.6878 s / it) * Acc@1 74.642 Acc@5 92.172 loss 1.015 Accuracy of the model on the 50000 test images: 74.6% Max accuracy: 75.00% Test: [0/9] eta: 0:00:45 loss: 1.2764 (1.2764) acc1: 71.6146 (71.6146) acc5: 88.0208 (88.0208) time: 5.0182 data: 4.8049 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.4376 (1.4822) acc1: 66.6667 (64.3520) acc5: 85.9375 (85.8240) time: 0.7088 data: 0.5340 max mem: 64948 Test: Total time: 0:00:06 (0.7180 s / it) * Acc@1 64.108 Acc@5 85.882 loss 1.508 Accuracy of the model EMA on 50000 test images: 64.1% Max EMA accuracy: 64.11% Epoch: [82] [ 0/312] eta: 0:53:29 lr: 0.003798 min_lr: 0.003798 loss: 2.0876 (2.0876) weight_decay: 0.0500 (0.0500) time: 10.2867 data: 9.5220 max mem: 64948 Epoch: [82] [ 10/312] eta: 0:08:05 lr: 0.003798 min_lr: 0.003798 loss: 2.5679 (2.4433) weight_decay: 0.0500 (0.0500) time: 1.6067 data: 0.8659 max mem: 64948 Epoch: [82] [ 20/312] eta: 0:05:42 lr: 0.003798 min_lr: 0.003798 loss: 2.2926 (2.3110) weight_decay: 0.0500 (0.0500) time: 0.7156 data: 0.0003 max mem: 64948 Epoch: [82] [ 30/312] eta: 0:04:46 lr: 0.003798 min_lr: 0.003798 loss: 2.3877 (2.3824) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [82] [ 40/312] eta: 0:04:15 lr: 0.003798 min_lr: 0.003798 loss: 2.4678 (2.3691) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [82] [ 50/312] eta: 0:03:53 lr: 0.003797 min_lr: 0.003797 loss: 2.3673 (2.3578) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [82] [ 60/312] eta: 0:03:36 lr: 0.003797 min_lr: 0.003797 loss: 2.3771 (2.3274) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [82] [ 70/312] eta: 0:03:22 lr: 0.003797 min_lr: 0.003797 loss: 2.3771 (2.3269) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [82] [ 80/312] eta: 0:03:09 lr: 0.003797 min_lr: 0.003797 loss: 2.4100 (2.3244) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [82] [ 90/312] eta: 0:02:58 lr: 0.003797 min_lr: 0.003797 loss: 2.4100 (2.3168) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [82] [100/312] eta: 0:02:48 lr: 0.003796 min_lr: 0.003796 loss: 2.0768 (2.2979) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [82] [110/312] eta: 0:02:38 lr: 0.003796 min_lr: 0.003796 loss: 2.0768 (2.2902) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [82] [120/312] eta: 0:02:29 lr: 0.003796 min_lr: 0.003796 loss: 2.2926 (2.3015) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [82] [130/312] eta: 0:02:20 lr: 0.003796 min_lr: 0.003796 loss: 2.4808 (2.3084) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [82] [140/312] eta: 0:02:11 lr: 0.003795 min_lr: 0.003795 loss: 2.4943 (2.3228) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [82] [150/312] eta: 0:02:03 lr: 0.003795 min_lr: 0.003795 loss: 2.5931 (2.3346) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [82] [160/312] eta: 0:01:55 lr: 0.003795 min_lr: 0.003795 loss: 2.3954 (2.3352) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [82] [170/312] eta: 0:01:46 lr: 0.003795 min_lr: 0.003795 loss: 2.3904 (2.3273) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [82] [180/312] eta: 0:01:38 lr: 0.003795 min_lr: 0.003795 loss: 2.3904 (2.3237) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [82] [190/312] eta: 0:01:31 lr: 0.003794 min_lr: 0.003794 loss: 2.3231 (2.3193) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [82] [200/312] eta: 0:01:23 lr: 0.003794 min_lr: 0.003794 loss: 2.2849 (2.3194) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [82] [210/312] eta: 0:01:15 lr: 0.003794 min_lr: 0.003794 loss: 2.2225 (2.3177) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [82] [220/312] eta: 0:01:08 lr: 0.003794 min_lr: 0.003794 loss: 2.2194 (2.3248) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [82] [230/312] eta: 0:01:00 lr: 0.003794 min_lr: 0.003794 loss: 2.3170 (2.3263) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [82] [240/312] eta: 0:00:53 lr: 0.003793 min_lr: 0.003793 loss: 2.3170 (2.3274) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [82] [250/312] eta: 0:00:45 lr: 0.003793 min_lr: 0.003793 loss: 2.4600 (2.3289) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [82] [260/312] eta: 0:00:38 lr: 0.003793 min_lr: 0.003793 loss: 2.2533 (2.3304) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [82] [270/312] eta: 0:00:30 lr: 0.003793 min_lr: 0.003793 loss: 2.5655 (2.3405) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [82] [280/312] eta: 0:00:23 lr: 0.003793 min_lr: 0.003793 loss: 2.6060 (2.3410) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [82] [290/312] eta: 0:00:16 lr: 0.003792 min_lr: 0.003792 loss: 2.6164 (2.3469) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [82] [300/312] eta: 0:00:08 lr: 0.003792 min_lr: 0.003792 loss: 2.1331 (2.3355) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [82] [310/312] eta: 0:00:01 lr: 0.003792 min_lr: 0.003792 loss: 2.0775 (2.3312) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [82] [311/312] eta: 0:00:00 lr: 0.003792 min_lr: 0.003792 loss: 2.0775 (2.3310) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [82] Total time: 0:03:47 (0.7294 s / it) Averaged stats: lr: 0.003792 min_lr: 0.003792 loss: 2.0775 (2.3520) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7597 (0.7597) acc1: 84.1146 (84.1146) acc5: 94.0104 (94.0104) time: 4.5517 data: 4.3389 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0619 (0.9915) acc1: 74.7396 (74.6560) acc5: 94.0104 (92.7360) time: 0.6570 data: 0.4822 max mem: 64948 Test: Total time: 0:00:06 (0.6801 s / it) * Acc@1 75.500 Acc@5 92.522 loss 0.973 Accuracy of the model on the 50000 test images: 75.5% Max accuracy: 75.50% Test: [0/9] eta: 0:00:41 loss: 1.2316 (1.2316) acc1: 72.6562 (72.6562) acc5: 88.8021 (88.8021) time: 4.6050 data: 4.3991 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3917 (1.4386) acc1: 67.4479 (65.2480) acc5: 86.1979 (86.3040) time: 0.6633 data: 0.4889 max mem: 64948 Test: Total time: 0:00:06 (0.6701 s / it) * Acc@1 65.086 Acc@5 86.510 loss 1.459 Accuracy of the model EMA on 50000 test images: 65.1% Max EMA accuracy: 65.09% Epoch: [83] [ 0/312] eta: 0:48:45 lr: 0.003792 min_lr: 0.003792 loss: 2.3090 (2.3090) weight_decay: 0.0500 (0.0500) time: 9.3751 data: 8.5941 max mem: 64948 Epoch: [83] [ 10/312] eta: 0:07:42 lr: 0.003792 min_lr: 0.003792 loss: 2.2265 (2.1051) weight_decay: 0.0500 (0.0500) time: 1.5325 data: 0.7817 max mem: 64948 Epoch: [83] [ 20/312] eta: 0:05:30 lr: 0.003791 min_lr: 0.003791 loss: 2.3006 (2.2290) weight_decay: 0.0500 (0.0500) time: 0.7207 data: 0.0004 max mem: 64948 Epoch: [83] [ 30/312] eta: 0:04:40 lr: 0.003791 min_lr: 0.003791 loss: 2.3611 (2.2752) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [83] [ 40/312] eta: 0:04:10 lr: 0.003791 min_lr: 0.003791 loss: 2.3161 (2.2741) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [83] [ 50/312] eta: 0:03:49 lr: 0.003791 min_lr: 0.003791 loss: 2.4628 (2.2912) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [83] [ 60/312] eta: 0:03:33 lr: 0.003791 min_lr: 0.003791 loss: 2.1861 (2.2679) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [83] [ 70/312] eta: 0:03:19 lr: 0.003790 min_lr: 0.003790 loss: 2.1861 (2.2868) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [83] [ 80/312] eta: 0:03:07 lr: 0.003790 min_lr: 0.003790 loss: 2.3763 (2.2812) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [83] [ 90/312] eta: 0:02:56 lr: 0.003790 min_lr: 0.003790 loss: 2.2271 (2.2701) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [83] [100/312] eta: 0:02:46 lr: 0.003790 min_lr: 0.003790 loss: 2.3192 (2.2682) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [83] [110/312] eta: 0:02:37 lr: 0.003790 min_lr: 0.003790 loss: 2.4025 (2.2754) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [83] [120/312] eta: 0:02:28 lr: 0.003789 min_lr: 0.003789 loss: 2.3893 (2.2698) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [83] [130/312] eta: 0:02:19 lr: 0.003789 min_lr: 0.003789 loss: 2.3873 (2.2846) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [83] [140/312] eta: 0:02:10 lr: 0.003789 min_lr: 0.003789 loss: 2.3070 (2.2841) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [83] [150/312] eta: 0:02:02 lr: 0.003789 min_lr: 0.003789 loss: 2.2758 (2.2827) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [83] [160/312] eta: 0:01:54 lr: 0.003789 min_lr: 0.003789 loss: 2.1763 (2.2800) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [83] [170/312] eta: 0:01:46 lr: 0.003788 min_lr: 0.003788 loss: 2.1763 (2.2775) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [83] [180/312] eta: 0:01:38 lr: 0.003788 min_lr: 0.003788 loss: 2.3821 (2.2826) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [83] [190/312] eta: 0:01:30 lr: 0.003788 min_lr: 0.003788 loss: 2.4707 (2.2933) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [83] [200/312] eta: 0:01:22 lr: 0.003788 min_lr: 0.003788 loss: 2.5272 (2.2957) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [83] [210/312] eta: 0:01:15 lr: 0.003788 min_lr: 0.003788 loss: 2.3120 (2.2946) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [83] [220/312] eta: 0:01:07 lr: 0.003787 min_lr: 0.003787 loss: 2.2536 (2.2930) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [83] [230/312] eta: 0:01:00 lr: 0.003787 min_lr: 0.003787 loss: 2.3232 (2.2953) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [83] [240/312] eta: 0:00:52 lr: 0.003787 min_lr: 0.003787 loss: 2.4335 (2.2938) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [83] [250/312] eta: 0:00:45 lr: 0.003787 min_lr: 0.003787 loss: 2.2322 (2.2906) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [83] [260/312] eta: 0:00:37 lr: 0.003786 min_lr: 0.003786 loss: 2.2322 (2.2913) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [83] [270/312] eta: 0:00:30 lr: 0.003786 min_lr: 0.003786 loss: 2.4249 (2.2911) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [83] [280/312] eta: 0:00:23 lr: 0.003786 min_lr: 0.003786 loss: 2.4672 (2.2999) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [83] [290/312] eta: 0:00:15 lr: 0.003786 min_lr: 0.003786 loss: 2.5649 (2.3041) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [83] [300/312] eta: 0:00:08 lr: 0.003786 min_lr: 0.003786 loss: 2.3493 (2.3044) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [83] [310/312] eta: 0:00:01 lr: 0.003785 min_lr: 0.003785 loss: 2.2687 (2.3010) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [83] [311/312] eta: 0:00:00 lr: 0.003785 min_lr: 0.003785 loss: 2.3116 (2.3017) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [83] Total time: 0:03:46 (0.7268 s / it) Averaged stats: lr: 0.003785 min_lr: 0.003785 loss: 2.3116 (2.3555) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.8027 (0.8027) acc1: 81.2500 (81.2500) acc5: 94.2708 (94.2708) time: 4.3791 data: 4.1624 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0653 (1.0214) acc1: 74.7396 (73.9840) acc5: 92.9688 (92.2560) time: 0.6379 data: 0.4626 max mem: 64948 Test: Total time: 0:00:05 (0.6612 s / it) * Acc@1 74.740 Acc@5 92.288 loss 1.004 Accuracy of the model on the 50000 test images: 74.7% Max accuracy: 75.50% Test: [0/9] eta: 0:00:43 loss: 1.1904 (1.1904) acc1: 73.4375 (73.4375) acc5: 89.5833 (89.5833) time: 4.7849 data: 4.5778 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3499 (1.3977) acc1: 68.7500 (65.8880) acc5: 86.7188 (86.9440) time: 0.6829 data: 0.5087 max mem: 64948 Test: Total time: 0:00:06 (0.6922 s / it) * Acc@1 65.914 Acc@5 87.038 loss 1.413 Accuracy of the model EMA on 50000 test images: 65.9% Max EMA accuracy: 65.91% Epoch: [84] [ 0/312] eta: 0:55:03 lr: 0.003785 min_lr: 0.003785 loss: 2.0681 (2.0681) weight_decay: 0.0500 (0.0500) time: 10.5879 data: 9.8122 max mem: 64948 Epoch: [84] [ 10/312] eta: 0:08:06 lr: 0.003785 min_lr: 0.003785 loss: 2.5012 (2.3446) weight_decay: 0.0500 (0.0500) time: 1.6112 data: 0.8923 max mem: 64948 Epoch: [84] [ 20/312] eta: 0:05:42 lr: 0.003785 min_lr: 0.003785 loss: 2.4932 (2.3529) weight_decay: 0.0500 (0.0500) time: 0.7036 data: 0.0003 max mem: 64948 Epoch: [84] [ 30/312] eta: 0:04:47 lr: 0.003785 min_lr: 0.003785 loss: 2.3745 (2.3197) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [84] [ 40/312] eta: 0:04:15 lr: 0.003785 min_lr: 0.003785 loss: 2.3745 (2.3410) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [84] [ 50/312] eta: 0:03:53 lr: 0.003784 min_lr: 0.003784 loss: 2.5764 (2.3810) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [84] [ 60/312] eta: 0:03:36 lr: 0.003784 min_lr: 0.003784 loss: 2.5574 (2.3871) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [84] [ 70/312] eta: 0:03:22 lr: 0.003784 min_lr: 0.003784 loss: 2.4194 (2.3919) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [84] [ 80/312] eta: 0:03:09 lr: 0.003784 min_lr: 0.003784 loss: 2.4750 (2.4133) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [84] [ 90/312] eta: 0:02:58 lr: 0.003783 min_lr: 0.003783 loss: 2.4159 (2.3907) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [84] [100/312] eta: 0:02:48 lr: 0.003783 min_lr: 0.003783 loss: 2.2991 (2.3938) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [84] [110/312] eta: 0:02:38 lr: 0.003783 min_lr: 0.003783 loss: 2.3134 (2.3776) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [84] [120/312] eta: 0:02:29 lr: 0.003783 min_lr: 0.003783 loss: 2.3134 (2.3842) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [84] [130/312] eta: 0:02:20 lr: 0.003783 min_lr: 0.003783 loss: 2.5093 (2.3821) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [84] [140/312] eta: 0:02:11 lr: 0.003782 min_lr: 0.003782 loss: 2.4874 (2.3873) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [84] [150/312] eta: 0:02:03 lr: 0.003782 min_lr: 0.003782 loss: 2.3403 (2.3768) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [84] [160/312] eta: 0:01:55 lr: 0.003782 min_lr: 0.003782 loss: 2.3485 (2.3795) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [84] [170/312] eta: 0:01:47 lr: 0.003782 min_lr: 0.003782 loss: 2.4832 (2.3821) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [84] [180/312] eta: 0:01:39 lr: 0.003782 min_lr: 0.003782 loss: 2.4743 (2.3834) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [84] [190/312] eta: 0:01:31 lr: 0.003781 min_lr: 0.003781 loss: 2.2382 (2.3795) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [84] [200/312] eta: 0:01:23 lr: 0.003781 min_lr: 0.003781 loss: 2.2093 (2.3738) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [84] [210/312] eta: 0:01:15 lr: 0.003781 min_lr: 0.003781 loss: 2.5202 (2.3794) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [84] [220/312] eta: 0:01:08 lr: 0.003781 min_lr: 0.003781 loss: 2.5346 (2.3787) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [84] [230/312] eta: 0:01:00 lr: 0.003780 min_lr: 0.003780 loss: 2.5471 (2.3852) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [84] [240/312] eta: 0:00:53 lr: 0.003780 min_lr: 0.003780 loss: 2.4398 (2.3825) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [84] [250/312] eta: 0:00:45 lr: 0.003780 min_lr: 0.003780 loss: 2.3781 (2.3835) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [84] [260/312] eta: 0:00:38 lr: 0.003780 min_lr: 0.003780 loss: 2.4463 (2.3841) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [84] [270/312] eta: 0:00:30 lr: 0.003780 min_lr: 0.003780 loss: 2.2351 (2.3733) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [84] [280/312] eta: 0:00:23 lr: 0.003779 min_lr: 0.003779 loss: 2.1452 (2.3715) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [84] [290/312] eta: 0:00:16 lr: 0.003779 min_lr: 0.003779 loss: 2.2371 (2.3691) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [84] [300/312] eta: 0:00:08 lr: 0.003779 min_lr: 0.003779 loss: 2.4545 (2.3715) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [84] [310/312] eta: 0:00:01 lr: 0.003779 min_lr: 0.003779 loss: 2.4670 (2.3672) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [84] [311/312] eta: 0:00:00 lr: 0.003779 min_lr: 0.003779 loss: 2.4670 (2.3685) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [84] Total time: 0:03:47 (0.7297 s / it) Averaged stats: lr: 0.003779 min_lr: 0.003779 loss: 2.4670 (2.3398) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7449 (0.7449) acc1: 82.0312 (82.0312) acc5: 93.4896 (93.4896) time: 4.5547 data: 4.3386 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0192 (0.9734) acc1: 74.2188 (74.8480) acc5: 92.4479 (92.8000) time: 0.6574 data: 0.4822 max mem: 64948 Test: Total time: 0:00:06 (0.6802 s / it) * Acc@1 75.144 Acc@5 92.718 loss 0.979 Accuracy of the model on the 50000 test images: 75.1% Max accuracy: 75.50% Test: [0/9] eta: 0:00:44 loss: 1.1523 (1.1523) acc1: 74.7396 (74.7396) acc5: 89.8438 (89.8438) time: 4.9424 data: 4.7322 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.3130 (1.3598) acc1: 69.2708 (66.5280) acc5: 86.9792 (87.3280) time: 0.7022 data: 0.5259 max mem: 64948 Test: Total time: 0:00:06 (0.7117 s / it) * Acc@1 66.756 Acc@5 87.590 loss 1.372 Accuracy of the model EMA on 50000 test images: 66.8% Max EMA accuracy: 66.76% Epoch: [85] [ 0/312] eta: 0:52:20 lr: 0.003779 min_lr: 0.003779 loss: 2.2326 (2.2326) weight_decay: 0.0500 (0.0500) time: 10.0660 data: 9.2781 max mem: 64948 Epoch: [85] [ 10/312] eta: 0:07:53 lr: 0.003779 min_lr: 0.003779 loss: 2.2926 (2.3958) weight_decay: 0.0500 (0.0500) time: 1.5668 data: 0.8438 max mem: 64948 Epoch: [85] [ 20/312] eta: 0:05:36 lr: 0.003778 min_lr: 0.003778 loss: 2.4314 (2.4397) weight_decay: 0.0500 (0.0500) time: 0.7061 data: 0.0003 max mem: 64948 Epoch: [85] [ 30/312] eta: 0:04:43 lr: 0.003778 min_lr: 0.003778 loss: 2.4602 (2.4339) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [85] [ 40/312] eta: 0:04:12 lr: 0.003778 min_lr: 0.003778 loss: 2.4582 (2.4088) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [85] [ 50/312] eta: 0:03:51 lr: 0.003778 min_lr: 0.003778 loss: 2.4633 (2.4165) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [85] [ 60/312] eta: 0:03:34 lr: 0.003777 min_lr: 0.003777 loss: 2.3684 (2.4001) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [85] [ 70/312] eta: 0:03:21 lr: 0.003777 min_lr: 0.003777 loss: 2.3684 (2.3992) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [85] [ 80/312] eta: 0:03:08 lr: 0.003777 min_lr: 0.003777 loss: 2.3574 (2.3796) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [85] [ 90/312] eta: 0:02:57 lr: 0.003777 min_lr: 0.003777 loss: 2.3214 (2.3634) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [85] [100/312] eta: 0:02:47 lr: 0.003777 min_lr: 0.003777 loss: 2.2155 (2.3536) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [85] [110/312] eta: 0:02:37 lr: 0.003776 min_lr: 0.003776 loss: 2.2394 (2.3457) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [85] [120/312] eta: 0:02:28 lr: 0.003776 min_lr: 0.003776 loss: 2.2142 (2.3266) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [85] [130/312] eta: 0:02:19 lr: 0.003776 min_lr: 0.003776 loss: 2.2317 (2.3326) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [85] [140/312] eta: 0:02:11 lr: 0.003776 min_lr: 0.003776 loss: 2.4867 (2.3551) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [85] [150/312] eta: 0:02:02 lr: 0.003776 min_lr: 0.003776 loss: 2.5783 (2.3671) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [85] [160/312] eta: 0:01:54 lr: 0.003775 min_lr: 0.003775 loss: 2.5205 (2.3743) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [85] [170/312] eta: 0:01:46 lr: 0.003775 min_lr: 0.003775 loss: 2.4003 (2.3715) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [85] [180/312] eta: 0:01:38 lr: 0.003775 min_lr: 0.003775 loss: 2.4003 (2.3759) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [85] [190/312] eta: 0:01:30 lr: 0.003775 min_lr: 0.003775 loss: 2.4045 (2.3764) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [85] [200/312] eta: 0:01:23 lr: 0.003774 min_lr: 0.003774 loss: 2.4045 (2.3787) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [85] [210/312] eta: 0:01:15 lr: 0.003774 min_lr: 0.003774 loss: 2.4284 (2.3814) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0003 max mem: 64948 Epoch: [85] [220/312] eta: 0:01:07 lr: 0.003774 min_lr: 0.003774 loss: 2.3909 (2.3777) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [85] [230/312] eta: 0:01:00 lr: 0.003774 min_lr: 0.003774 loss: 2.1639 (2.3768) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [85] [240/312] eta: 0:00:52 lr: 0.003774 min_lr: 0.003774 loss: 2.3618 (2.3692) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [85] [250/312] eta: 0:00:45 lr: 0.003773 min_lr: 0.003773 loss: 2.3618 (2.3704) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [85] [260/312] eta: 0:00:38 lr: 0.003773 min_lr: 0.003773 loss: 2.4040 (2.3695) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [85] [270/312] eta: 0:00:30 lr: 0.003773 min_lr: 0.003773 loss: 2.5247 (2.3740) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [85] [280/312] eta: 0:00:23 lr: 0.003773 min_lr: 0.003773 loss: 2.4580 (2.3727) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [85] [290/312] eta: 0:00:15 lr: 0.003772 min_lr: 0.003772 loss: 2.3222 (2.3767) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [85] [300/312] eta: 0:00:08 lr: 0.003772 min_lr: 0.003772 loss: 2.4175 (2.3781) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [85] [310/312] eta: 0:00:01 lr: 0.003772 min_lr: 0.003772 loss: 2.5577 (2.3826) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [85] [311/312] eta: 0:00:00 lr: 0.003772 min_lr: 0.003772 loss: 2.4945 (2.3797) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [85] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.003772 min_lr: 0.003772 loss: 2.4945 (2.3531) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7163 (0.7163) acc1: 81.5104 (81.5104) acc5: 95.0521 (95.0521) time: 4.6248 data: 4.4056 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9923 (0.9682) acc1: 74.7396 (74.2080) acc5: 93.7500 (93.0240) time: 0.6652 data: 0.4896 max mem: 64948 Test: Total time: 0:00:06 (0.6785 s / it) * Acc@1 75.326 Acc@5 92.698 loss 0.966 Accuracy of the model on the 50000 test images: 75.3% Max accuracy: 75.50% Test: [0/9] eta: 0:00:44 loss: 1.1188 (1.1188) acc1: 75.5208 (75.5208) acc5: 90.3646 (90.3646) time: 4.9257 data: 4.7078 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2795 (1.3251) acc1: 70.0521 (67.3280) acc5: 88.0208 (87.8080) time: 0.6986 data: 0.5232 max mem: 64948 Test: Total time: 0:00:06 (0.7082 s / it) * Acc@1 67.558 Acc@5 88.072 loss 1.334 Accuracy of the model EMA on 50000 test images: 67.6% Max EMA accuracy: 67.56% Epoch: [86] [ 0/312] eta: 0:49:20 lr: 0.003772 min_lr: 0.003772 loss: 1.5938 (1.5938) weight_decay: 0.0500 (0.0500) time: 9.4898 data: 8.1000 max mem: 64948 Epoch: [86] [ 10/312] eta: 0:07:49 lr: 0.003772 min_lr: 0.003772 loss: 2.2432 (2.2085) weight_decay: 0.0500 (0.0500) time: 1.5558 data: 0.7368 max mem: 64948 Epoch: [86] [ 20/312] eta: 0:05:34 lr: 0.003772 min_lr: 0.003772 loss: 2.3022 (2.3065) weight_decay: 0.0500 (0.0500) time: 0.7288 data: 0.0004 max mem: 64948 Epoch: [86] [ 30/312] eta: 0:04:42 lr: 0.003771 min_lr: 0.003771 loss: 2.4970 (2.3705) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0003 max mem: 64948 Epoch: [86] [ 40/312] eta: 0:04:12 lr: 0.003771 min_lr: 0.003771 loss: 2.4970 (2.3848) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [86] [ 50/312] eta: 0:03:51 lr: 0.003771 min_lr: 0.003771 loss: 2.4698 (2.3983) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [86] [ 60/312] eta: 0:03:34 lr: 0.003771 min_lr: 0.003771 loss: 2.4604 (2.3654) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [86] [ 70/312] eta: 0:03:20 lr: 0.003770 min_lr: 0.003770 loss: 2.5006 (2.3885) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [86] [ 80/312] eta: 0:03:08 lr: 0.003770 min_lr: 0.003770 loss: 2.3824 (2.3748) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [86] [ 90/312] eta: 0:02:57 lr: 0.003770 min_lr: 0.003770 loss: 2.3824 (2.3756) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [86] [100/312] eta: 0:02:47 lr: 0.003770 min_lr: 0.003770 loss: 2.4178 (2.3624) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [86] [110/312] eta: 0:02:37 lr: 0.003770 min_lr: 0.003770 loss: 2.3610 (2.3636) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [86] [120/312] eta: 0:02:28 lr: 0.003769 min_lr: 0.003769 loss: 2.4492 (2.3718) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [86] [130/312] eta: 0:02:19 lr: 0.003769 min_lr: 0.003769 loss: 2.5290 (2.3732) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [86] [140/312] eta: 0:02:11 lr: 0.003769 min_lr: 0.003769 loss: 2.4365 (2.3698) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [86] [150/312] eta: 0:02:02 lr: 0.003769 min_lr: 0.003769 loss: 2.3209 (2.3596) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [86] [160/312] eta: 0:01:54 lr: 0.003769 min_lr: 0.003769 loss: 2.3911 (2.3602) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [86] [170/312] eta: 0:01:46 lr: 0.003768 min_lr: 0.003768 loss: 2.4113 (2.3676) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [86] [180/312] eta: 0:01:38 lr: 0.003768 min_lr: 0.003768 loss: 2.3869 (2.3598) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [86] [190/312] eta: 0:01:30 lr: 0.003768 min_lr: 0.003768 loss: 2.1966 (2.3587) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [86] [200/312] eta: 0:01:23 lr: 0.003768 min_lr: 0.003768 loss: 2.4145 (2.3644) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [86] [210/312] eta: 0:01:15 lr: 0.003767 min_lr: 0.003767 loss: 2.4875 (2.3603) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [86] [220/312] eta: 0:01:07 lr: 0.003767 min_lr: 0.003767 loss: 2.4087 (2.3643) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [86] [230/312] eta: 0:01:00 lr: 0.003767 min_lr: 0.003767 loss: 2.3260 (2.3604) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [86] [240/312] eta: 0:00:52 lr: 0.003767 min_lr: 0.003767 loss: 2.3260 (2.3553) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [86] [250/312] eta: 0:00:45 lr: 0.003767 min_lr: 0.003767 loss: 2.5149 (2.3616) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [86] [260/312] eta: 0:00:38 lr: 0.003766 min_lr: 0.003766 loss: 2.5408 (2.3701) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [86] [270/312] eta: 0:00:30 lr: 0.003766 min_lr: 0.003766 loss: 2.4188 (2.3681) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [86] [280/312] eta: 0:00:23 lr: 0.003766 min_lr: 0.003766 loss: 2.4562 (2.3745) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [86] [290/312] eta: 0:00:15 lr: 0.003766 min_lr: 0.003766 loss: 2.5570 (2.3773) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [86] [300/312] eta: 0:00:08 lr: 0.003765 min_lr: 0.003765 loss: 2.3730 (2.3763) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [86] [310/312] eta: 0:00:01 lr: 0.003765 min_lr: 0.003765 loss: 2.4418 (2.3764) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [86] [311/312] eta: 0:00:00 lr: 0.003765 min_lr: 0.003765 loss: 2.4418 (2.3743) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [86] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.003765 min_lr: 0.003765 loss: 2.4418 (2.3487) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7306 (0.7306) acc1: 81.7708 (81.7708) acc5: 93.7500 (93.7500) time: 4.6428 data: 4.4233 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0814 (0.9864) acc1: 74.7396 (74.0480) acc5: 92.9688 (92.4480) time: 0.6672 data: 0.4916 max mem: 64948 Test: Total time: 0:00:06 (0.6921 s / it) * Acc@1 75.200 Acc@5 92.612 loss 0.973 Accuracy of the model on the 50000 test images: 75.2% Max accuracy: 75.50% Test: [0/9] eta: 0:00:44 loss: 1.0897 (1.0897) acc1: 75.7812 (75.7812) acc5: 90.6250 (90.6250) time: 4.9963 data: 4.7917 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2513 (1.2930) acc1: 70.3125 (67.9680) acc5: 88.0208 (88.2240) time: 0.7064 data: 0.5325 max mem: 64948 Test: Total time: 0:00:06 (0.7161 s / it) * Acc@1 68.240 Acc@5 88.570 loss 1.299 Accuracy of the model EMA on 50000 test images: 68.2% Max EMA accuracy: 68.24% Epoch: [87] [ 0/312] eta: 0:47:51 lr: 0.003765 min_lr: 0.003765 loss: 2.1211 (2.1211) weight_decay: 0.0500 (0.0500) time: 9.2033 data: 8.2235 max mem: 64948 Epoch: [87] [ 10/312] eta: 0:07:31 lr: 0.003765 min_lr: 0.003765 loss: 2.5532 (2.5428) weight_decay: 0.0500 (0.0500) time: 1.4965 data: 0.7480 max mem: 64948 Epoch: [87] [ 20/312] eta: 0:05:25 lr: 0.003765 min_lr: 0.003765 loss: 2.4710 (2.4912) weight_decay: 0.0500 (0.0500) time: 0.7118 data: 0.0004 max mem: 64948 Epoch: [87] [ 30/312] eta: 0:04:36 lr: 0.003765 min_lr: 0.003765 loss: 2.3190 (2.3985) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [87] [ 40/312] eta: 0:04:07 lr: 0.003764 min_lr: 0.003764 loss: 2.2857 (2.3675) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [87] [ 50/312] eta: 0:03:47 lr: 0.003764 min_lr: 0.003764 loss: 2.4421 (2.3781) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [87] [ 60/312] eta: 0:03:31 lr: 0.003764 min_lr: 0.003764 loss: 2.5005 (2.3804) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [87] [ 70/312] eta: 0:03:18 lr: 0.003764 min_lr: 0.003764 loss: 2.3568 (2.3487) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [87] [ 80/312] eta: 0:03:06 lr: 0.003763 min_lr: 0.003763 loss: 2.1944 (2.3270) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [87] [ 90/312] eta: 0:02:55 lr: 0.003763 min_lr: 0.003763 loss: 2.1227 (2.3043) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [87] [100/312] eta: 0:02:45 lr: 0.003763 min_lr: 0.003763 loss: 2.0208 (2.2896) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [87] [110/312] eta: 0:02:36 lr: 0.003763 min_lr: 0.003763 loss: 2.3662 (2.3097) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [87] [120/312] eta: 0:02:27 lr: 0.003763 min_lr: 0.003763 loss: 2.4370 (2.3231) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [87] [130/312] eta: 0:02:18 lr: 0.003762 min_lr: 0.003762 loss: 2.4051 (2.3288) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [87] [140/312] eta: 0:02:10 lr: 0.003762 min_lr: 0.003762 loss: 2.3320 (2.3271) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [87] [150/312] eta: 0:02:02 lr: 0.003762 min_lr: 0.003762 loss: 2.4568 (2.3270) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [87] [160/312] eta: 0:01:53 lr: 0.003762 min_lr: 0.003762 loss: 2.3058 (2.3243) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [87] [170/312] eta: 0:01:45 lr: 0.003761 min_lr: 0.003761 loss: 2.3058 (2.3210) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [87] [180/312] eta: 0:01:38 lr: 0.003761 min_lr: 0.003761 loss: 2.3813 (2.3338) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [87] [190/312] eta: 0:01:30 lr: 0.003761 min_lr: 0.003761 loss: 2.4384 (2.3307) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [87] [200/312] eta: 0:01:22 lr: 0.003761 min_lr: 0.003761 loss: 2.2707 (2.3227) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [87] [210/312] eta: 0:01:15 lr: 0.003761 min_lr: 0.003761 loss: 2.2198 (2.3202) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [87] [220/312] eta: 0:01:07 lr: 0.003760 min_lr: 0.003760 loss: 2.3381 (2.3273) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [87] [230/312] eta: 0:01:00 lr: 0.003760 min_lr: 0.003760 loss: 2.3381 (2.3282) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [87] [240/312] eta: 0:00:52 lr: 0.003760 min_lr: 0.003760 loss: 2.2980 (2.3297) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [87] [250/312] eta: 0:00:45 lr: 0.003760 min_lr: 0.003760 loss: 2.5068 (2.3369) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [87] [260/312] eta: 0:00:37 lr: 0.003759 min_lr: 0.003759 loss: 2.4324 (2.3368) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [87] [270/312] eta: 0:00:30 lr: 0.003759 min_lr: 0.003759 loss: 2.4324 (2.3353) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [87] [280/312] eta: 0:00:23 lr: 0.003759 min_lr: 0.003759 loss: 2.3883 (2.3336) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [87] [290/312] eta: 0:00:15 lr: 0.003759 min_lr: 0.003759 loss: 2.3312 (2.3321) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [87] [300/312] eta: 0:00:08 lr: 0.003759 min_lr: 0.003759 loss: 2.3646 (2.3323) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [87] [310/312] eta: 0:00:01 lr: 0.003758 min_lr: 0.003758 loss: 2.5474 (2.3320) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [87] [311/312] eta: 0:00:00 lr: 0.003758 min_lr: 0.003758 loss: 2.4634 (2.3324) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [87] Total time: 0:03:46 (0.7254 s / it) Averaged stats: lr: 0.003758 min_lr: 0.003758 loss: 2.4634 (2.3467) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.7385 (0.7385) acc1: 81.5104 (81.5104) acc5: 95.0521 (95.0521) time: 4.4068 data: 4.1953 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0787 (1.0183) acc1: 75.5208 (74.8160) acc5: 92.4479 (92.1920) time: 0.6409 data: 0.4662 max mem: 64948 Test: Total time: 0:00:05 (0.6651 s / it) * Acc@1 74.924 Acc@5 92.308 loss 1.020 Accuracy of the model on the 50000 test images: 74.9% Max accuracy: 75.50% Test: [0/9] eta: 0:00:45 loss: 1.0645 (1.0645) acc1: 76.5625 (76.5625) acc5: 91.4062 (91.4062) time: 5.0388 data: 4.8211 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.2247 (1.2647) acc1: 70.5729 (68.5120) acc5: 88.5417 (88.6400) time: 0.7115 data: 0.5358 max mem: 64948 Test: Total time: 0:00:06 (0.7181 s / it) * Acc@1 68.864 Acc@5 88.988 loss 1.268 Accuracy of the model EMA on 50000 test images: 68.9% Max EMA accuracy: 68.86% Epoch: [88] [ 0/312] eta: 0:47:49 lr: 0.003758 min_lr: 0.003758 loss: 1.6014 (1.6014) weight_decay: 0.0500 (0.0500) time: 9.1957 data: 8.3478 max mem: 64948 Epoch: [88] [ 10/312] eta: 0:07:34 lr: 0.003758 min_lr: 0.003758 loss: 2.1255 (2.1555) weight_decay: 0.0500 (0.0500) time: 1.5066 data: 0.7763 max mem: 64948 Epoch: [88] [ 20/312] eta: 0:05:26 lr: 0.003758 min_lr: 0.003758 loss: 2.2731 (2.2332) weight_decay: 0.0500 (0.0500) time: 0.7159 data: 0.0098 max mem: 64948 Epoch: [88] [ 30/312] eta: 0:04:37 lr: 0.003758 min_lr: 0.003758 loss: 2.3665 (2.2841) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [88] [ 40/312] eta: 0:04:08 lr: 0.003757 min_lr: 0.003757 loss: 2.4522 (2.3031) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [88] [ 50/312] eta: 0:03:47 lr: 0.003757 min_lr: 0.003757 loss: 2.3700 (2.2976) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [88] [ 60/312] eta: 0:03:32 lr: 0.003757 min_lr: 0.003757 loss: 2.2875 (2.2858) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [88] [ 70/312] eta: 0:03:18 lr: 0.003757 min_lr: 0.003757 loss: 2.4548 (2.3024) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [88] [ 80/312] eta: 0:03:07 lr: 0.003756 min_lr: 0.003756 loss: 2.5003 (2.3227) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [88] [ 90/312] eta: 0:02:56 lr: 0.003756 min_lr: 0.003756 loss: 2.5622 (2.3356) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [88] [100/312] eta: 0:02:46 lr: 0.003756 min_lr: 0.003756 loss: 2.5372 (2.3482) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [88] [110/312] eta: 0:02:36 lr: 0.003756 min_lr: 0.003756 loss: 2.4671 (2.3421) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [88] [120/312] eta: 0:02:27 lr: 0.003756 min_lr: 0.003756 loss: 2.2516 (2.3317) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [88] [130/312] eta: 0:02:18 lr: 0.003755 min_lr: 0.003755 loss: 2.3286 (2.3395) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [88] [140/312] eta: 0:02:10 lr: 0.003755 min_lr: 0.003755 loss: 2.3260 (2.3301) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [88] [150/312] eta: 0:02:02 lr: 0.003755 min_lr: 0.003755 loss: 2.2594 (2.3336) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [88] [160/312] eta: 0:01:54 lr: 0.003755 min_lr: 0.003755 loss: 2.3549 (2.3306) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [88] [170/312] eta: 0:01:46 lr: 0.003754 min_lr: 0.003754 loss: 2.3135 (2.3273) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [88] [180/312] eta: 0:01:38 lr: 0.003754 min_lr: 0.003754 loss: 2.3828 (2.3333) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [88] [190/312] eta: 0:01:30 lr: 0.003754 min_lr: 0.003754 loss: 2.3469 (2.3241) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [88] [200/312] eta: 0:01:22 lr: 0.003754 min_lr: 0.003754 loss: 2.1856 (2.3304) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [88] [210/312] eta: 0:01:15 lr: 0.003754 min_lr: 0.003754 loss: 2.5291 (2.3384) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [88] [220/312] eta: 0:01:07 lr: 0.003753 min_lr: 0.003753 loss: 2.3946 (2.3266) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [88] [230/312] eta: 0:01:00 lr: 0.003753 min_lr: 0.003753 loss: 2.2012 (2.3292) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [88] [240/312] eta: 0:00:52 lr: 0.003753 min_lr: 0.003753 loss: 2.4455 (2.3322) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [88] [250/312] eta: 0:00:45 lr: 0.003753 min_lr: 0.003753 loss: 2.4455 (2.3365) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [88] [260/312] eta: 0:00:37 lr: 0.003752 min_lr: 0.003752 loss: 2.3861 (2.3313) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [88] [270/312] eta: 0:00:30 lr: 0.003752 min_lr: 0.003752 loss: 2.3996 (2.3327) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [88] [280/312] eta: 0:00:23 lr: 0.003752 min_lr: 0.003752 loss: 2.2504 (2.3300) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [88] [290/312] eta: 0:00:15 lr: 0.003752 min_lr: 0.003752 loss: 2.2401 (2.3308) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [88] [300/312] eta: 0:00:08 lr: 0.003752 min_lr: 0.003752 loss: 2.4329 (2.3260) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [88] [310/312] eta: 0:00:01 lr: 0.003751 min_lr: 0.003751 loss: 2.2496 (2.3235) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [88] [311/312] eta: 0:00:00 lr: 0.003751 min_lr: 0.003751 loss: 2.1887 (2.3213) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [88] Total time: 0:03:46 (0.7262 s / it) Averaged stats: lr: 0.003751 min_lr: 0.003751 loss: 2.1887 (2.3446) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7395 (0.7395) acc1: 81.7708 (81.7708) acc5: 93.7500 (93.7500) time: 4.6187 data: 4.4091 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9715 (0.9644) acc1: 76.0417 (75.0080) acc5: 92.9688 (92.7360) time: 0.6644 data: 0.4900 max mem: 64948 Test: Total time: 0:00:06 (0.6871 s / it) * Acc@1 75.532 Acc@5 92.720 loss 0.950 Accuracy of the model on the 50000 test images: 75.5% Max accuracy: 75.53% Test: [0/9] eta: 0:00:42 loss: 1.0416 (1.0416) acc1: 76.5625 (76.5625) acc5: 91.9271 (91.9271) time: 4.6682 data: 4.4541 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1987 (1.2377) acc1: 70.8333 (68.8320) acc5: 89.5833 (89.0880) time: 0.6700 data: 0.4950 max mem: 64948 Test: Total time: 0:00:06 (0.6791 s / it) * Acc@1 69.406 Acc@5 89.342 loss 1.239 Accuracy of the model EMA on 50000 test images: 69.4% Max EMA accuracy: 69.41% Epoch: [89] [ 0/312] eta: 0:55:49 lr: 0.003751 min_lr: 0.003751 loss: 2.7000 (2.7000) weight_decay: 0.0500 (0.0500) time: 10.7351 data: 9.6335 max mem: 64948 Epoch: [89] [ 10/312] eta: 0:08:10 lr: 0.003751 min_lr: 0.003751 loss: 2.4756 (2.4197) weight_decay: 0.0500 (0.0500) time: 1.6246 data: 0.8761 max mem: 64948 Epoch: [89] [ 20/312] eta: 0:05:45 lr: 0.003751 min_lr: 0.003751 loss: 2.3959 (2.3620) weight_decay: 0.0500 (0.0500) time: 0.7047 data: 0.0004 max mem: 64948 Epoch: [89] [ 30/312] eta: 0:04:49 lr: 0.003751 min_lr: 0.003751 loss: 2.3469 (2.3572) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [89] [ 40/312] eta: 0:04:17 lr: 0.003750 min_lr: 0.003750 loss: 2.4151 (2.3872) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [89] [ 50/312] eta: 0:03:55 lr: 0.003750 min_lr: 0.003750 loss: 2.5271 (2.3773) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [89] [ 60/312] eta: 0:03:37 lr: 0.003750 min_lr: 0.003750 loss: 2.5271 (2.3908) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [89] [ 70/312] eta: 0:03:23 lr: 0.003750 min_lr: 0.003750 loss: 2.4417 (2.3829) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [89] [ 80/312] eta: 0:03:10 lr: 0.003749 min_lr: 0.003749 loss: 2.4417 (2.3953) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [89] [ 90/312] eta: 0:02:59 lr: 0.003749 min_lr: 0.003749 loss: 2.5528 (2.4027) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [89] [100/312] eta: 0:02:49 lr: 0.003749 min_lr: 0.003749 loss: 2.5548 (2.4118) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [89] [110/312] eta: 0:02:39 lr: 0.003749 min_lr: 0.003749 loss: 2.4829 (2.4116) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [89] [120/312] eta: 0:02:29 lr: 0.003749 min_lr: 0.003749 loss: 2.3166 (2.4016) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [89] [130/312] eta: 0:02:20 lr: 0.003748 min_lr: 0.003748 loss: 2.2669 (2.3858) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [89] [140/312] eta: 0:02:12 lr: 0.003748 min_lr: 0.003748 loss: 2.2173 (2.3694) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [89] [150/312] eta: 0:02:03 lr: 0.003748 min_lr: 0.003748 loss: 2.2214 (2.3698) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [89] [160/312] eta: 0:01:55 lr: 0.003748 min_lr: 0.003748 loss: 2.3237 (2.3695) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [89] [170/312] eta: 0:01:47 lr: 0.003747 min_lr: 0.003747 loss: 2.3141 (2.3651) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [89] [180/312] eta: 0:01:39 lr: 0.003747 min_lr: 0.003747 loss: 2.4173 (2.3715) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [89] [190/312] eta: 0:01:31 lr: 0.003747 min_lr: 0.003747 loss: 2.4173 (2.3647) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [89] [200/312] eta: 0:01:23 lr: 0.003747 min_lr: 0.003747 loss: 2.4391 (2.3688) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [89] [210/312] eta: 0:01:15 lr: 0.003746 min_lr: 0.003746 loss: 2.4391 (2.3686) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [89] [220/312] eta: 0:01:08 lr: 0.003746 min_lr: 0.003746 loss: 2.2836 (2.3625) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [89] [230/312] eta: 0:01:00 lr: 0.003746 min_lr: 0.003746 loss: 2.2404 (2.3618) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [89] [240/312] eta: 0:00:53 lr: 0.003746 min_lr: 0.003746 loss: 2.3753 (2.3640) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [89] [250/312] eta: 0:00:45 lr: 0.003746 min_lr: 0.003746 loss: 2.4181 (2.3611) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [89] [260/312] eta: 0:00:38 lr: 0.003745 min_lr: 0.003745 loss: 2.0879 (2.3518) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [89] [270/312] eta: 0:00:30 lr: 0.003745 min_lr: 0.003745 loss: 2.3484 (2.3545) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [89] [280/312] eta: 0:00:23 lr: 0.003745 min_lr: 0.003745 loss: 2.5266 (2.3576) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [89] [290/312] eta: 0:00:16 lr: 0.003745 min_lr: 0.003745 loss: 2.4411 (2.3575) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [89] [300/312] eta: 0:00:08 lr: 0.003744 min_lr: 0.003744 loss: 2.3833 (2.3522) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [89] [310/312] eta: 0:00:01 lr: 0.003744 min_lr: 0.003744 loss: 2.4208 (2.3563) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [89] [311/312] eta: 0:00:00 lr: 0.003744 min_lr: 0.003744 loss: 2.4208 (2.3547) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [89] Total time: 0:03:47 (0.7300 s / it) Averaged stats: lr: 0.003744 min_lr: 0.003744 loss: 2.4208 (2.3330) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.7715 (0.7715) acc1: 83.3333 (83.3333) acc5: 94.5312 (94.5312) time: 4.6671 data: 4.4628 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0403 (0.9879) acc1: 75.2604 (74.1760) acc5: 92.9688 (92.7680) time: 0.6699 data: 0.4960 max mem: 64948 Test: Total time: 0:00:06 (0.6931 s / it) * Acc@1 75.198 Acc@5 92.540 loss 0.972 Accuracy of the model on the 50000 test images: 75.2% Max accuracy: 75.53% Test: [0/9] eta: 0:00:43 loss: 1.0201 (1.0201) acc1: 77.0833 (77.0833) acc5: 92.1875 (92.1875) time: 4.8755 data: 4.6680 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1746 (1.2121) acc1: 70.8333 (69.7280) acc5: 89.5833 (89.4080) time: 0.6930 data: 0.5188 max mem: 64948 Test: Total time: 0:00:06 (0.7019 s / it) * Acc@1 70.008 Acc@5 89.684 loss 1.212 Accuracy of the model EMA on 50000 test images: 70.0% Max EMA accuracy: 70.01% Epoch: [90] [ 0/312] eta: 0:53:20 lr: 0.003744 min_lr: 0.003744 loss: 1.9084 (1.9084) weight_decay: 0.0500 (0.0500) time: 10.2580 data: 9.4726 max mem: 64948 Epoch: [90] [ 10/312] eta: 0:07:57 lr: 0.003744 min_lr: 0.003744 loss: 2.4534 (2.5681) weight_decay: 0.0500 (0.0500) time: 1.5827 data: 0.8615 max mem: 64948 Epoch: [90] [ 20/312] eta: 0:05:39 lr: 0.003744 min_lr: 0.003744 loss: 2.3978 (2.3518) weight_decay: 0.0500 (0.0500) time: 0.7071 data: 0.0003 max mem: 64948 Epoch: [90] [ 30/312] eta: 0:04:45 lr: 0.003743 min_lr: 0.003743 loss: 2.2396 (2.3548) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [90] [ 40/312] eta: 0:04:14 lr: 0.003743 min_lr: 0.003743 loss: 2.2418 (2.3460) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [90] [ 50/312] eta: 0:03:52 lr: 0.003743 min_lr: 0.003743 loss: 2.2418 (2.3292) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0003 max mem: 64948 Epoch: [90] [ 60/312] eta: 0:03:35 lr: 0.003743 min_lr: 0.003743 loss: 2.1081 (2.2898) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [90] [ 70/312] eta: 0:03:21 lr: 0.003743 min_lr: 0.003743 loss: 2.3784 (2.3294) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [90] [ 80/312] eta: 0:03:09 lr: 0.003742 min_lr: 0.003742 loss: 2.5474 (2.3236) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [90] [ 90/312] eta: 0:02:58 lr: 0.003742 min_lr: 0.003742 loss: 2.4555 (2.3295) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [90] [100/312] eta: 0:02:48 lr: 0.003742 min_lr: 0.003742 loss: 2.3922 (2.3340) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [90] [110/312] eta: 0:02:38 lr: 0.003742 min_lr: 0.003742 loss: 2.2864 (2.3261) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [90] [120/312] eta: 0:02:29 lr: 0.003741 min_lr: 0.003741 loss: 2.2810 (2.3240) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [90] [130/312] eta: 0:02:20 lr: 0.003741 min_lr: 0.003741 loss: 2.3590 (2.3256) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [90] [140/312] eta: 0:02:11 lr: 0.003741 min_lr: 0.003741 loss: 2.2236 (2.3146) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [90] [150/312] eta: 0:02:03 lr: 0.003741 min_lr: 0.003741 loss: 2.3473 (2.3308) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [90] [160/312] eta: 0:01:54 lr: 0.003740 min_lr: 0.003740 loss: 2.5077 (2.3246) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [90] [170/312] eta: 0:01:46 lr: 0.003740 min_lr: 0.003740 loss: 2.3216 (2.3230) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [90] [180/312] eta: 0:01:38 lr: 0.003740 min_lr: 0.003740 loss: 2.4185 (2.3361) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [90] [190/312] eta: 0:01:31 lr: 0.003740 min_lr: 0.003740 loss: 2.4185 (2.3414) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [90] [200/312] eta: 0:01:23 lr: 0.003740 min_lr: 0.003740 loss: 2.3890 (2.3440) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [90] [210/312] eta: 0:01:15 lr: 0.003739 min_lr: 0.003739 loss: 2.2141 (2.3346) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [90] [220/312] eta: 0:01:08 lr: 0.003739 min_lr: 0.003739 loss: 2.1921 (2.3331) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [90] [230/312] eta: 0:01:00 lr: 0.003739 min_lr: 0.003739 loss: 2.4262 (2.3357) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [90] [240/312] eta: 0:00:52 lr: 0.003739 min_lr: 0.003739 loss: 2.3487 (2.3305) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [90] [250/312] eta: 0:00:45 lr: 0.003738 min_lr: 0.003738 loss: 2.3487 (2.3333) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [90] [260/312] eta: 0:00:38 lr: 0.003738 min_lr: 0.003738 loss: 2.3103 (2.3284) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [90] [270/312] eta: 0:00:30 lr: 0.003738 min_lr: 0.003738 loss: 2.1686 (2.3189) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [90] [280/312] eta: 0:00:23 lr: 0.003738 min_lr: 0.003738 loss: 2.3386 (2.3236) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [90] [290/312] eta: 0:00:16 lr: 0.003737 min_lr: 0.003737 loss: 2.5272 (2.3255) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [90] [300/312] eta: 0:00:08 lr: 0.003737 min_lr: 0.003737 loss: 2.2883 (2.3259) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [90] [310/312] eta: 0:00:01 lr: 0.003737 min_lr: 0.003737 loss: 2.2385 (2.3231) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [90] [311/312] eta: 0:00:00 lr: 0.003737 min_lr: 0.003737 loss: 2.2385 (2.3217) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [90] Total time: 0:03:47 (0.7292 s / it) Averaged stats: lr: 0.003737 min_lr: 0.003737 loss: 2.2385 (2.3404) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7488 (0.7488) acc1: 79.4271 (79.4271) acc5: 94.5312 (94.5312) time: 4.5535 data: 4.3337 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0196 (0.9778) acc1: 75.5208 (74.7840) acc5: 92.4479 (92.6720) time: 0.6575 data: 0.4816 max mem: 64948 Test: Total time: 0:00:06 (0.6825 s / it) * Acc@1 75.656 Acc@5 92.740 loss 0.955 Accuracy of the model on the 50000 test images: 75.7% Max accuracy: 75.66% Test: [0/9] eta: 0:00:41 loss: 1.0008 (1.0008) acc1: 77.3438 (77.3438) acc5: 92.4479 (92.4479) time: 4.6277 data: 4.4166 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1513 (1.1879) acc1: 71.3542 (70.0160) acc5: 90.1042 (89.7600) time: 0.6655 data: 0.4908 max mem: 64948 Test: Total time: 0:00:06 (0.6751 s / it) * Acc@1 70.560 Acc@5 89.994 loss 1.187 Accuracy of the model EMA on 50000 test images: 70.6% Max EMA accuracy: 70.56% Epoch: [91] [ 0/312] eta: 0:52:01 lr: 0.003737 min_lr: 0.003737 loss: 2.1401 (2.1401) weight_decay: 0.0500 (0.0500) time: 10.0033 data: 9.1433 max mem: 64948 Epoch: [91] [ 10/312] eta: 0:07:49 lr: 0.003737 min_lr: 0.003737 loss: 2.4013 (2.2554) weight_decay: 0.0500 (0.0500) time: 1.5562 data: 0.8316 max mem: 64948 Epoch: [91] [ 20/312] eta: 0:05:34 lr: 0.003737 min_lr: 0.003737 loss: 2.2762 (2.2626) weight_decay: 0.0500 (0.0500) time: 0.7027 data: 0.0004 max mem: 64948 Epoch: [91] [ 30/312] eta: 0:04:42 lr: 0.003736 min_lr: 0.003736 loss: 2.2047 (2.2670) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [91] [ 40/312] eta: 0:04:12 lr: 0.003736 min_lr: 0.003736 loss: 2.3747 (2.2587) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [91] [ 50/312] eta: 0:03:51 lr: 0.003736 min_lr: 0.003736 loss: 2.1569 (2.2489) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [91] [ 60/312] eta: 0:03:34 lr: 0.003736 min_lr: 0.003736 loss: 2.2648 (2.2553) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [91] [ 70/312] eta: 0:03:20 lr: 0.003735 min_lr: 0.003735 loss: 2.3208 (2.2737) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [91] [ 80/312] eta: 0:03:08 lr: 0.003735 min_lr: 0.003735 loss: 2.4249 (2.2839) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [91] [ 90/312] eta: 0:02:57 lr: 0.003735 min_lr: 0.003735 loss: 2.3606 (2.2785) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [91] [100/312] eta: 0:02:47 lr: 0.003735 min_lr: 0.003735 loss: 2.3606 (2.2853) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [91] [110/312] eta: 0:02:37 lr: 0.003734 min_lr: 0.003734 loss: 2.4593 (2.2936) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [91] [120/312] eta: 0:02:28 lr: 0.003734 min_lr: 0.003734 loss: 2.4695 (2.3097) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [91] [130/312] eta: 0:02:19 lr: 0.003734 min_lr: 0.003734 loss: 2.4287 (2.3076) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [91] [140/312] eta: 0:02:11 lr: 0.003734 min_lr: 0.003734 loss: 2.3215 (2.3094) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [91] [150/312] eta: 0:02:02 lr: 0.003733 min_lr: 0.003733 loss: 2.3771 (2.3175) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [91] [160/312] eta: 0:01:54 lr: 0.003733 min_lr: 0.003733 loss: 2.3761 (2.3189) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [91] [170/312] eta: 0:01:46 lr: 0.003733 min_lr: 0.003733 loss: 2.4738 (2.3302) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [91] [180/312] eta: 0:01:38 lr: 0.003733 min_lr: 0.003733 loss: 2.4958 (2.3254) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [91] [190/312] eta: 0:01:30 lr: 0.003733 min_lr: 0.003733 loss: 2.4521 (2.3311) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [91] [200/312] eta: 0:01:23 lr: 0.003732 min_lr: 0.003732 loss: 2.3838 (2.3297) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [91] [210/312] eta: 0:01:15 lr: 0.003732 min_lr: 0.003732 loss: 2.4263 (2.3325) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [91] [220/312] eta: 0:01:07 lr: 0.003732 min_lr: 0.003732 loss: 2.4612 (2.3375) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [91] [230/312] eta: 0:01:00 lr: 0.003732 min_lr: 0.003732 loss: 2.4840 (2.3480) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [91] [240/312] eta: 0:00:52 lr: 0.003731 min_lr: 0.003731 loss: 2.4548 (2.3441) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [91] [250/312] eta: 0:00:45 lr: 0.003731 min_lr: 0.003731 loss: 2.3820 (2.3429) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [91] [260/312] eta: 0:00:38 lr: 0.003731 min_lr: 0.003731 loss: 2.5394 (2.3417) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [91] [270/312] eta: 0:00:30 lr: 0.003731 min_lr: 0.003731 loss: 2.4313 (2.3461) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [91] [280/312] eta: 0:00:23 lr: 0.003730 min_lr: 0.003730 loss: 2.3143 (2.3403) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [91] [290/312] eta: 0:00:16 lr: 0.003730 min_lr: 0.003730 loss: 2.2225 (2.3423) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [91] [300/312] eta: 0:00:08 lr: 0.003730 min_lr: 0.003730 loss: 2.3840 (2.3453) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [91] [310/312] eta: 0:00:01 lr: 0.003730 min_lr: 0.003730 loss: 2.4261 (2.3487) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [91] [311/312] eta: 0:00:00 lr: 0.003730 min_lr: 0.003730 loss: 2.4929 (2.3495) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [91] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.003730 min_lr: 0.003730 loss: 2.4929 (2.3328) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7763 (0.7763) acc1: 81.5104 (81.5104) acc5: 94.5312 (94.5312) time: 4.5475 data: 4.3261 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0517 (0.9978) acc1: 75.7812 (74.3360) acc5: 91.9271 (92.3520) time: 0.6573 data: 0.4808 max mem: 64948 Test: Total time: 0:00:06 (0.6807 s / it) * Acc@1 75.062 Acc@5 92.456 loss 0.979 Accuracy of the model on the 50000 test images: 75.1% Max accuracy: 75.66% Test: [0/9] eta: 0:00:45 loss: 0.9825 (0.9825) acc1: 77.6042 (77.6042) acc5: 92.1875 (92.1875) time: 5.0004 data: 4.7826 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1312 (1.1661) acc1: 71.3542 (70.4000) acc5: 89.8438 (89.7280) time: 0.7111 data: 0.5315 max mem: 64948 Test: Total time: 0:00:06 (0.7203 s / it) * Acc@1 70.976 Acc@5 90.294 loss 1.164 Accuracy of the model EMA on 50000 test images: 71.0% Max EMA accuracy: 70.98% Epoch: [92] [ 0/312] eta: 0:51:29 lr: 0.003730 min_lr: 0.003730 loss: 2.8234 (2.8234) weight_decay: 0.0500 (0.0500) time: 9.9021 data: 9.1003 max mem: 64948 Epoch: [92] [ 10/312] eta: 0:07:47 lr: 0.003729 min_lr: 0.003729 loss: 2.3129 (2.2463) weight_decay: 0.0500 (0.0500) time: 1.5490 data: 0.8277 max mem: 64948 Epoch: [92] [ 20/312] eta: 0:05:33 lr: 0.003729 min_lr: 0.003729 loss: 2.4734 (2.3667) weight_decay: 0.0500 (0.0500) time: 0.7036 data: 0.0004 max mem: 64948 Epoch: [92] [ 30/312] eta: 0:04:41 lr: 0.003729 min_lr: 0.003729 loss: 2.6477 (2.4460) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [92] [ 40/312] eta: 0:04:11 lr: 0.003729 min_lr: 0.003729 loss: 2.4745 (2.4004) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [92] [ 50/312] eta: 0:03:50 lr: 0.003729 min_lr: 0.003729 loss: 2.3464 (2.3955) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [92] [ 60/312] eta: 0:03:34 lr: 0.003728 min_lr: 0.003728 loss: 2.3464 (2.3518) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0003 max mem: 64948 Epoch: [92] [ 70/312] eta: 0:03:20 lr: 0.003728 min_lr: 0.003728 loss: 2.2333 (2.3257) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [92] [ 80/312] eta: 0:03:08 lr: 0.003728 min_lr: 0.003728 loss: 2.1644 (2.3221) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [92] [ 90/312] eta: 0:02:57 lr: 0.003728 min_lr: 0.003728 loss: 2.3285 (2.3274) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [92] [100/312] eta: 0:02:47 lr: 0.003727 min_lr: 0.003727 loss: 2.4329 (2.3306) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [92] [110/312] eta: 0:02:37 lr: 0.003727 min_lr: 0.003727 loss: 2.3107 (2.3215) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [92] [120/312] eta: 0:02:28 lr: 0.003727 min_lr: 0.003727 loss: 2.3107 (2.3232) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [92] [130/312] eta: 0:02:19 lr: 0.003727 min_lr: 0.003727 loss: 2.4744 (2.3286) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [92] [140/312] eta: 0:02:11 lr: 0.003726 min_lr: 0.003726 loss: 2.4385 (2.3230) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [92] [150/312] eta: 0:02:02 lr: 0.003726 min_lr: 0.003726 loss: 2.4385 (2.3363) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [92] [160/312] eta: 0:01:54 lr: 0.003726 min_lr: 0.003726 loss: 2.5413 (2.3453) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [92] [170/312] eta: 0:01:46 lr: 0.003726 min_lr: 0.003726 loss: 2.5036 (2.3530) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [92] [180/312] eta: 0:01:38 lr: 0.003725 min_lr: 0.003725 loss: 2.3774 (2.3457) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [92] [190/312] eta: 0:01:30 lr: 0.003725 min_lr: 0.003725 loss: 2.2725 (2.3345) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [92] [200/312] eta: 0:01:23 lr: 0.003725 min_lr: 0.003725 loss: 2.2868 (2.3333) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [92] [210/312] eta: 0:01:15 lr: 0.003725 min_lr: 0.003725 loss: 2.3596 (2.3328) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [92] [220/312] eta: 0:01:07 lr: 0.003724 min_lr: 0.003724 loss: 2.2623 (2.3242) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [92] [230/312] eta: 0:01:00 lr: 0.003724 min_lr: 0.003724 loss: 2.2761 (2.3276) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [92] [240/312] eta: 0:00:52 lr: 0.003724 min_lr: 0.003724 loss: 2.2761 (2.3233) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [92] [250/312] eta: 0:00:45 lr: 0.003724 min_lr: 0.003724 loss: 2.1668 (2.3181) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [92] [260/312] eta: 0:00:38 lr: 0.003724 min_lr: 0.003724 loss: 2.3723 (2.3229) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [92] [270/312] eta: 0:00:30 lr: 0.003723 min_lr: 0.003723 loss: 2.4489 (2.3216) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [92] [280/312] eta: 0:00:23 lr: 0.003723 min_lr: 0.003723 loss: 2.4076 (2.3271) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [92] [290/312] eta: 0:00:16 lr: 0.003723 min_lr: 0.003723 loss: 2.4691 (2.3255) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [92] [300/312] eta: 0:00:08 lr: 0.003723 min_lr: 0.003723 loss: 2.4576 (2.3276) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [92] [310/312] eta: 0:00:01 lr: 0.003722 min_lr: 0.003722 loss: 2.4056 (2.3245) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [92] [311/312] eta: 0:00:00 lr: 0.003722 min_lr: 0.003722 loss: 2.3843 (2.3234) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [92] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.003722 min_lr: 0.003722 loss: 2.3843 (2.3331) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.7209 (0.7209) acc1: 83.3333 (83.3333) acc5: 95.3125 (95.3125) time: 4.3874 data: 4.1764 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0460 (0.9625) acc1: 75.7812 (75.5520) acc5: 92.4479 (92.3200) time: 0.6388 data: 0.4641 max mem: 64948 Test: Total time: 0:00:05 (0.6624 s / it) * Acc@1 75.198 Acc@5 92.418 loss 0.974 Accuracy of the model on the 50000 test images: 75.2% Max accuracy: 75.66% Test: [0/9] eta: 0:00:44 loss: 0.9649 (0.9649) acc1: 77.3438 (77.3438) acc5: 92.4479 (92.4479) time: 4.9965 data: 4.7918 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1121 (1.1454) acc1: 71.6146 (70.7520) acc5: 90.6250 (90.1120) time: 0.7067 data: 0.5325 max mem: 64948 Test: Total time: 0:00:06 (0.7157 s / it) * Acc@1 71.388 Acc@5 90.608 loss 1.142 Accuracy of the model EMA on 50000 test images: 71.4% Max EMA accuracy: 71.39% Epoch: [93] [ 0/312] eta: 0:52:35 lr: 0.003722 min_lr: 0.003722 loss: 2.6481 (2.6481) weight_decay: 0.0500 (0.0500) time: 10.1151 data: 9.3148 max mem: 64948 Epoch: [93] [ 10/312] eta: 0:07:53 lr: 0.003722 min_lr: 0.003722 loss: 2.5058 (2.3661) weight_decay: 0.0500 (0.0500) time: 1.5692 data: 0.8471 max mem: 64948 Epoch: [93] [ 20/312] eta: 0:05:36 lr: 0.003722 min_lr: 0.003722 loss: 2.3569 (2.3337) weight_decay: 0.0500 (0.0500) time: 0.7048 data: 0.0004 max mem: 64948 Epoch: [93] [ 30/312] eta: 0:04:43 lr: 0.003722 min_lr: 0.003722 loss: 2.3063 (2.3287) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [93] [ 40/312] eta: 0:04:13 lr: 0.003721 min_lr: 0.003721 loss: 2.3897 (2.3359) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [93] [ 50/312] eta: 0:03:51 lr: 0.003721 min_lr: 0.003721 loss: 2.3996 (2.3521) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [93] [ 60/312] eta: 0:03:35 lr: 0.003721 min_lr: 0.003721 loss: 2.3134 (2.3234) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [93] [ 70/312] eta: 0:03:21 lr: 0.003721 min_lr: 0.003721 loss: 2.3942 (2.3411) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [93] [ 80/312] eta: 0:03:08 lr: 0.003720 min_lr: 0.003720 loss: 2.3942 (2.3343) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [93] [ 90/312] eta: 0:02:57 lr: 0.003720 min_lr: 0.003720 loss: 2.2892 (2.3215) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [93] [100/312] eta: 0:02:47 lr: 0.003720 min_lr: 0.003720 loss: 2.3151 (2.3256) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [93] [110/312] eta: 0:02:37 lr: 0.003720 min_lr: 0.003720 loss: 2.2921 (2.3172) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [93] [120/312] eta: 0:02:28 lr: 0.003719 min_lr: 0.003719 loss: 2.3619 (2.3133) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [93] [130/312] eta: 0:02:19 lr: 0.003719 min_lr: 0.003719 loss: 2.4172 (2.3230) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [93] [140/312] eta: 0:02:11 lr: 0.003719 min_lr: 0.003719 loss: 2.4021 (2.3227) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [93] [150/312] eta: 0:02:02 lr: 0.003719 min_lr: 0.003719 loss: 2.3492 (2.3197) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [93] [160/312] eta: 0:01:54 lr: 0.003718 min_lr: 0.003718 loss: 2.2289 (2.3122) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [93] [170/312] eta: 0:01:46 lr: 0.003718 min_lr: 0.003718 loss: 2.2361 (2.3128) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [93] [180/312] eta: 0:01:38 lr: 0.003718 min_lr: 0.003718 loss: 2.2298 (2.3071) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [93] [190/312] eta: 0:01:30 lr: 0.003718 min_lr: 0.003718 loss: 2.4541 (2.3127) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [93] [200/312] eta: 0:01:23 lr: 0.003718 min_lr: 0.003718 loss: 2.4307 (2.3086) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [93] [210/312] eta: 0:01:15 lr: 0.003717 min_lr: 0.003717 loss: 2.4104 (2.3077) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [93] [220/312] eta: 0:01:07 lr: 0.003717 min_lr: 0.003717 loss: 2.4501 (2.3105) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [93] [230/312] eta: 0:01:00 lr: 0.003717 min_lr: 0.003717 loss: 2.4289 (2.3059) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [93] [240/312] eta: 0:00:52 lr: 0.003717 min_lr: 0.003717 loss: 2.4289 (2.3133) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [93] [250/312] eta: 0:00:45 lr: 0.003716 min_lr: 0.003716 loss: 2.3289 (2.3061) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [93] [260/312] eta: 0:00:38 lr: 0.003716 min_lr: 0.003716 loss: 2.3614 (2.3156) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [93] [270/312] eta: 0:00:30 lr: 0.003716 min_lr: 0.003716 loss: 2.5651 (2.3254) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [93] [280/312] eta: 0:00:23 lr: 0.003716 min_lr: 0.003716 loss: 2.5328 (2.3313) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0006 max mem: 64948 Epoch: [93] [290/312] eta: 0:00:16 lr: 0.003715 min_lr: 0.003715 loss: 2.5066 (2.3299) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0005 max mem: 64948 Epoch: [93] [300/312] eta: 0:00:08 lr: 0.003715 min_lr: 0.003715 loss: 2.3688 (2.3315) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [93] [310/312] eta: 0:00:01 lr: 0.003715 min_lr: 0.003715 loss: 2.3840 (2.3320) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [93] [311/312] eta: 0:00:00 lr: 0.003715 min_lr: 0.003715 loss: 2.3688 (2.3315) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [93] Total time: 0:03:47 (0.7287 s / it) Averaged stats: lr: 0.003715 min_lr: 0.003715 loss: 2.3688 (2.3307) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7090 (0.7090) acc1: 81.2500 (81.2500) acc5: 95.8333 (95.8333) time: 4.6164 data: 4.4009 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0493 (1.0113) acc1: 74.4792 (74.2720) acc5: 92.9688 (92.4800) time: 0.6642 data: 0.4891 max mem: 64948 Test: Total time: 0:00:06 (0.6888 s / it) * Acc@1 75.240 Acc@5 92.534 loss 0.981 Accuracy of the model on the 50000 test images: 75.2% Max accuracy: 75.66% Test: [0/9] eta: 0:00:43 loss: 0.9485 (0.9485) acc1: 77.6042 (77.6042) acc5: 92.7083 (92.7083) time: 4.7816 data: 4.5742 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0941 (1.1263) acc1: 72.6562 (71.2640) acc5: 90.8854 (90.3360) time: 0.6826 data: 0.5084 max mem: 64948 Test: Total time: 0:00:06 (0.6952 s / it) * Acc@1 71.786 Acc@5 90.770 loss 1.122 Accuracy of the model EMA on 50000 test images: 71.8% Max EMA accuracy: 71.79% Epoch: [94] [ 0/312] eta: 0:45:56 lr: 0.003715 min_lr: 0.003715 loss: 2.5936 (2.5936) weight_decay: 0.0500 (0.0500) time: 8.8351 data: 7.9647 max mem: 64948 Epoch: [94] [ 10/312] eta: 0:07:24 lr: 0.003715 min_lr: 0.003715 loss: 2.4785 (2.5028) weight_decay: 0.0500 (0.0500) time: 1.4721 data: 0.7245 max mem: 64948 Epoch: [94] [ 20/312] eta: 0:05:22 lr: 0.003714 min_lr: 0.003714 loss: 2.4677 (2.4930) weight_decay: 0.0500 (0.0500) time: 0.7170 data: 0.0004 max mem: 64948 Epoch: [94] [ 30/312] eta: 0:04:33 lr: 0.003714 min_lr: 0.003714 loss: 2.4212 (2.4049) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [94] [ 40/312] eta: 0:04:05 lr: 0.003714 min_lr: 0.003714 loss: 2.2147 (2.3646) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [94] [ 50/312] eta: 0:03:46 lr: 0.003714 min_lr: 0.003714 loss: 2.3137 (2.3599) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [94] [ 60/312] eta: 0:03:30 lr: 0.003713 min_lr: 0.003713 loss: 2.3307 (2.3208) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [94] [ 70/312] eta: 0:03:17 lr: 0.003713 min_lr: 0.003713 loss: 2.3835 (2.3447) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [94] [ 80/312] eta: 0:03:05 lr: 0.003713 min_lr: 0.003713 loss: 2.4878 (2.3300) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [94] [ 90/312] eta: 0:02:55 lr: 0.003713 min_lr: 0.003713 loss: 2.3208 (2.3285) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [94] [100/312] eta: 0:02:45 lr: 0.003712 min_lr: 0.003712 loss: 2.4015 (2.3280) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [94] [110/312] eta: 0:02:36 lr: 0.003712 min_lr: 0.003712 loss: 2.5199 (2.3289) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [94] [120/312] eta: 0:02:27 lr: 0.003712 min_lr: 0.003712 loss: 2.1870 (2.3111) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [94] [130/312] eta: 0:02:18 lr: 0.003712 min_lr: 0.003712 loss: 2.2333 (2.3141) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [94] [140/312] eta: 0:02:10 lr: 0.003711 min_lr: 0.003711 loss: 2.5275 (2.3198) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [94] [150/312] eta: 0:02:01 lr: 0.003711 min_lr: 0.003711 loss: 2.5247 (2.3288) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [94] [160/312] eta: 0:01:53 lr: 0.003711 min_lr: 0.003711 loss: 2.4972 (2.3267) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [94] [170/312] eta: 0:01:45 lr: 0.003711 min_lr: 0.003711 loss: 2.4457 (2.3361) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [94] [180/312] eta: 0:01:38 lr: 0.003710 min_lr: 0.003710 loss: 2.4580 (2.3413) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [94] [190/312] eta: 0:01:30 lr: 0.003710 min_lr: 0.003710 loss: 2.3800 (2.3359) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [94] [200/312] eta: 0:01:22 lr: 0.003710 min_lr: 0.003710 loss: 2.2582 (2.3313) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [94] [210/312] eta: 0:01:15 lr: 0.003710 min_lr: 0.003710 loss: 2.3103 (2.3286) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [94] [220/312] eta: 0:01:07 lr: 0.003709 min_lr: 0.003709 loss: 2.3337 (2.3230) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [94] [230/312] eta: 0:01:00 lr: 0.003709 min_lr: 0.003709 loss: 2.3844 (2.3257) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [94] [240/312] eta: 0:00:52 lr: 0.003709 min_lr: 0.003709 loss: 2.3376 (2.3266) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [94] [250/312] eta: 0:00:45 lr: 0.003709 min_lr: 0.003709 loss: 2.3474 (2.3323) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [94] [260/312] eta: 0:00:37 lr: 0.003709 min_lr: 0.003709 loss: 2.3499 (2.3299) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [94] [270/312] eta: 0:00:30 lr: 0.003708 min_lr: 0.003708 loss: 2.3311 (2.3278) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0003 max mem: 64948 Epoch: [94] [280/312] eta: 0:00:23 lr: 0.003708 min_lr: 0.003708 loss: 2.2322 (2.3236) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [94] [290/312] eta: 0:00:15 lr: 0.003708 min_lr: 0.003708 loss: 2.2092 (2.3180) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0008 max mem: 64948 Epoch: [94] [300/312] eta: 0:00:08 lr: 0.003708 min_lr: 0.003708 loss: 2.0930 (2.3117) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [94] [310/312] eta: 0:00:01 lr: 0.003707 min_lr: 0.003707 loss: 2.3276 (2.3209) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [94] [311/312] eta: 0:00:00 lr: 0.003707 min_lr: 0.003707 loss: 2.5144 (2.3221) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [94] Total time: 0:03:46 (0.7247 s / it) Averaged stats: lr: 0.003707 min_lr: 0.003707 loss: 2.5144 (2.3261) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7128 (0.7128) acc1: 82.8125 (82.8125) acc5: 94.5312 (94.5312) time: 4.4783 data: 4.2588 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0491 (0.9937) acc1: 75.2604 (74.8160) acc5: 93.4896 (92.7680) time: 0.6492 data: 0.4733 max mem: 64948 Test: Total time: 0:00:06 (0.6680 s / it) * Acc@1 75.434 Acc@5 92.662 loss 0.978 Accuracy of the model on the 50000 test images: 75.4% Max accuracy: 75.66% Test: [0/9] eta: 0:00:43 loss: 0.9326 (0.9326) acc1: 77.6042 (77.6042) acc5: 92.7083 (92.7083) time: 4.7934 data: 4.5786 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0776 (1.1082) acc1: 72.9167 (71.6480) acc5: 91.1458 (90.4640) time: 0.6839 data: 0.5088 max mem: 64948 Test: Total time: 0:00:06 (0.6919 s / it) * Acc@1 72.200 Acc@5 90.980 loss 1.104 Accuracy of the model EMA on 50000 test images: 72.2% Max EMA accuracy: 72.20% Epoch: [95] [ 0/312] eta: 0:52:58 lr: 0.003707 min_lr: 0.003707 loss: 2.4162 (2.4162) weight_decay: 0.0500 (0.0500) time: 10.1889 data: 8.7899 max mem: 64948 Epoch: [95] [ 10/312] eta: 0:08:00 lr: 0.003707 min_lr: 0.003707 loss: 2.2889 (2.3478) weight_decay: 0.0500 (0.0500) time: 1.5912 data: 0.7994 max mem: 64948 Epoch: [95] [ 20/312] eta: 0:05:39 lr: 0.003707 min_lr: 0.003707 loss: 2.2140 (2.2941) weight_decay: 0.0500 (0.0500) time: 0.7121 data: 0.0003 max mem: 64948 Epoch: [95] [ 30/312] eta: 0:04:45 lr: 0.003707 min_lr: 0.003707 loss: 2.3919 (2.3493) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [95] [ 40/312] eta: 0:04:14 lr: 0.003706 min_lr: 0.003706 loss: 2.3919 (2.3516) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [95] [ 50/312] eta: 0:03:52 lr: 0.003706 min_lr: 0.003706 loss: 2.4309 (2.3728) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [95] [ 60/312] eta: 0:03:35 lr: 0.003706 min_lr: 0.003706 loss: 2.5696 (2.3772) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [95] [ 70/312] eta: 0:03:21 lr: 0.003706 min_lr: 0.003706 loss: 2.4013 (2.3638) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [95] [ 80/312] eta: 0:03:09 lr: 0.003705 min_lr: 0.003705 loss: 2.4013 (2.3669) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [95] [ 90/312] eta: 0:02:58 lr: 0.003705 min_lr: 0.003705 loss: 2.3109 (2.3548) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [95] [100/312] eta: 0:02:48 lr: 0.003705 min_lr: 0.003705 loss: 2.2920 (2.3416) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [95] [110/312] eta: 0:02:38 lr: 0.003705 min_lr: 0.003705 loss: 2.2662 (2.3393) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [95] [120/312] eta: 0:02:29 lr: 0.003704 min_lr: 0.003704 loss: 2.2526 (2.3290) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [95] [130/312] eta: 0:02:20 lr: 0.003704 min_lr: 0.003704 loss: 2.3776 (2.3316) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [95] [140/312] eta: 0:02:11 lr: 0.003704 min_lr: 0.003704 loss: 2.2602 (2.3106) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [95] [150/312] eta: 0:02:03 lr: 0.003704 min_lr: 0.003704 loss: 2.2291 (2.3165) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [95] [160/312] eta: 0:01:54 lr: 0.003703 min_lr: 0.003703 loss: 2.3728 (2.3228) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [95] [170/312] eta: 0:01:46 lr: 0.003703 min_lr: 0.003703 loss: 2.4058 (2.3294) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [95] [180/312] eta: 0:01:38 lr: 0.003703 min_lr: 0.003703 loss: 2.4299 (2.3254) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [95] [190/312] eta: 0:01:31 lr: 0.003703 min_lr: 0.003703 loss: 2.4299 (2.3275) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [95] [200/312] eta: 0:01:23 lr: 0.003702 min_lr: 0.003702 loss: 2.4821 (2.3308) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [95] [210/312] eta: 0:01:15 lr: 0.003702 min_lr: 0.003702 loss: 2.3749 (2.3310) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [95] [220/312] eta: 0:01:08 lr: 0.003702 min_lr: 0.003702 loss: 2.3053 (2.3307) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [95] [230/312] eta: 0:01:00 lr: 0.003702 min_lr: 0.003702 loss: 2.2969 (2.3276) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [95] [240/312] eta: 0:00:52 lr: 0.003701 min_lr: 0.003701 loss: 2.3815 (2.3281) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [95] [250/312] eta: 0:00:45 lr: 0.003701 min_lr: 0.003701 loss: 2.3679 (2.3215) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [95] [260/312] eta: 0:00:38 lr: 0.003701 min_lr: 0.003701 loss: 2.3073 (2.3203) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [95] [270/312] eta: 0:00:30 lr: 0.003701 min_lr: 0.003701 loss: 2.4071 (2.3244) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [95] [280/312] eta: 0:00:23 lr: 0.003700 min_lr: 0.003700 loss: 2.4243 (2.3205) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [95] [290/312] eta: 0:00:16 lr: 0.003700 min_lr: 0.003700 loss: 2.3685 (2.3188) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [95] [300/312] eta: 0:00:08 lr: 0.003700 min_lr: 0.003700 loss: 2.3246 (2.3187) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [95] [310/312] eta: 0:00:01 lr: 0.003700 min_lr: 0.003700 loss: 2.4259 (2.3229) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [95] [311/312] eta: 0:00:00 lr: 0.003700 min_lr: 0.003700 loss: 2.4836 (2.3234) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [95] Total time: 0:03:47 (0.7290 s / it) Averaged stats: lr: 0.003700 min_lr: 0.003700 loss: 2.4836 (2.3329) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.8023 (0.8023) acc1: 80.7292 (80.7292) acc5: 94.0104 (94.0104) time: 4.8186 data: 4.6041 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0220 (0.9907) acc1: 75.0000 (73.8560) acc5: 92.4479 (92.6720) time: 0.6867 data: 0.5116 max mem: 64948 Test: Total time: 0:00:06 (0.7093 s / it) * Acc@1 75.582 Acc@5 92.796 loss 0.963 Accuracy of the model on the 50000 test images: 75.6% Max accuracy: 75.66% Test: [0/9] eta: 0:00:44 loss: 0.9174 (0.9174) acc1: 78.1250 (78.1250) acc5: 92.9688 (92.9688) time: 4.9346 data: 4.7254 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0621 (1.0913) acc1: 72.6562 (72.0640) acc5: 91.1458 (90.6560) time: 0.6996 data: 0.5251 max mem: 64948 Test: Total time: 0:00:06 (0.7089 s / it) * Acc@1 72.576 Acc@5 91.178 loss 1.086 Accuracy of the model EMA on 50000 test images: 72.6% Max EMA accuracy: 72.58% Epoch: [96] [ 0/312] eta: 0:49:22 lr: 0.003700 min_lr: 0.003700 loss: 3.0032 (3.0032) weight_decay: 0.0500 (0.0500) time: 9.4953 data: 8.7003 max mem: 64948 Epoch: [96] [ 10/312] eta: 0:07:48 lr: 0.003699 min_lr: 0.003699 loss: 2.4528 (2.3304) weight_decay: 0.0500 (0.0500) time: 1.5506 data: 0.7913 max mem: 64948 Epoch: [96] [ 20/312] eta: 0:05:33 lr: 0.003699 min_lr: 0.003699 loss: 2.4981 (2.4415) weight_decay: 0.0500 (0.0500) time: 0.7255 data: 0.0004 max mem: 64948 Epoch: [96] [ 30/312] eta: 0:04:41 lr: 0.003699 min_lr: 0.003699 loss: 2.4981 (2.3498) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [96] [ 40/312] eta: 0:04:11 lr: 0.003699 min_lr: 0.003699 loss: 2.3348 (2.3599) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [96] [ 50/312] eta: 0:03:50 lr: 0.003698 min_lr: 0.003698 loss: 2.4963 (2.3741) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [96] [ 60/312] eta: 0:03:33 lr: 0.003698 min_lr: 0.003698 loss: 2.2346 (2.3522) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [96] [ 70/312] eta: 0:03:20 lr: 0.003698 min_lr: 0.003698 loss: 2.3213 (2.3460) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [96] [ 80/312] eta: 0:03:08 lr: 0.003698 min_lr: 0.003698 loss: 2.3228 (2.3356) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [96] [ 90/312] eta: 0:02:57 lr: 0.003697 min_lr: 0.003697 loss: 2.1830 (2.3171) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [96] [100/312] eta: 0:02:47 lr: 0.003697 min_lr: 0.003697 loss: 2.3456 (2.3249) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [96] [110/312] eta: 0:02:37 lr: 0.003697 min_lr: 0.003697 loss: 2.5004 (2.3383) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [96] [120/312] eta: 0:02:28 lr: 0.003697 min_lr: 0.003697 loss: 2.3858 (2.3274) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [96] [130/312] eta: 0:02:19 lr: 0.003696 min_lr: 0.003696 loss: 2.1081 (2.3165) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [96] [140/312] eta: 0:02:11 lr: 0.003696 min_lr: 0.003696 loss: 2.2075 (2.3106) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [96] [150/312] eta: 0:02:02 lr: 0.003696 min_lr: 0.003696 loss: 2.4083 (2.3294) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [96] [160/312] eta: 0:01:54 lr: 0.003696 min_lr: 0.003696 loss: 2.5399 (2.3359) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [96] [170/312] eta: 0:01:46 lr: 0.003695 min_lr: 0.003695 loss: 2.5261 (2.3416) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [96] [180/312] eta: 0:01:38 lr: 0.003695 min_lr: 0.003695 loss: 2.5360 (2.3498) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [96] [190/312] eta: 0:01:30 lr: 0.003695 min_lr: 0.003695 loss: 2.4467 (2.3471) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [96] [200/312] eta: 0:01:23 lr: 0.003695 min_lr: 0.003695 loss: 2.3952 (2.3452) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [96] [210/312] eta: 0:01:15 lr: 0.003694 min_lr: 0.003694 loss: 2.3541 (2.3439) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [96] [220/312] eta: 0:01:07 lr: 0.003694 min_lr: 0.003694 loss: 2.4048 (2.3435) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [96] [230/312] eta: 0:01:00 lr: 0.003694 min_lr: 0.003694 loss: 2.4279 (2.3412) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [96] [240/312] eta: 0:00:52 lr: 0.003694 min_lr: 0.003694 loss: 2.4526 (2.3451) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [96] [250/312] eta: 0:00:45 lr: 0.003693 min_lr: 0.003693 loss: 2.3187 (2.3424) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [96] [260/312] eta: 0:00:38 lr: 0.003693 min_lr: 0.003693 loss: 2.3517 (2.3476) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [96] [270/312] eta: 0:00:30 lr: 0.003693 min_lr: 0.003693 loss: 2.1828 (2.3394) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [96] [280/312] eta: 0:00:23 lr: 0.003693 min_lr: 0.003693 loss: 2.0660 (2.3347) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [96] [290/312] eta: 0:00:15 lr: 0.003692 min_lr: 0.003692 loss: 2.2434 (2.3336) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [96] [300/312] eta: 0:00:08 lr: 0.003692 min_lr: 0.003692 loss: 2.4486 (2.3373) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [96] [310/312] eta: 0:00:01 lr: 0.003692 min_lr: 0.003692 loss: 2.4898 (2.3399) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [96] [311/312] eta: 0:00:00 lr: 0.003692 min_lr: 0.003692 loss: 2.4898 (2.3397) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [96] Total time: 0:03:46 (0.7274 s / it) Averaged stats: lr: 0.003692 min_lr: 0.003692 loss: 2.4898 (2.3312) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7736 (0.7736) acc1: 80.4688 (80.4688) acc5: 95.8333 (95.8333) time: 4.5747 data: 4.3562 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0622 (0.9996) acc1: 75.2604 (73.7280) acc5: 94.0104 (92.9920) time: 0.6595 data: 0.4841 max mem: 64948 Test: Total time: 0:00:06 (0.6832 s / it) * Acc@1 74.944 Acc@5 92.756 loss 0.992 Accuracy of the model on the 50000 test images: 74.9% Max accuracy: 75.66% Test: [0/9] eta: 0:00:43 loss: 0.9030 (0.9030) acc1: 78.6458 (78.6458) acc5: 92.9688 (92.9688) time: 4.8250 data: 4.6209 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0478 (1.0759) acc1: 73.1771 (72.5440) acc5: 91.1458 (90.8800) time: 0.6874 data: 0.5135 max mem: 64948 Test: Total time: 0:00:06 (0.6945 s / it) * Acc@1 72.892 Acc@5 91.356 loss 1.070 Accuracy of the model EMA on 50000 test images: 72.9% Max EMA accuracy: 72.89% Epoch: [97] [ 0/312] eta: 0:53:26 lr: 0.003692 min_lr: 0.003692 loss: 2.3859 (2.3859) weight_decay: 0.0500 (0.0500) time: 10.2768 data: 9.4102 max mem: 64948 Epoch: [97] [ 10/312] eta: 0:07:58 lr: 0.003692 min_lr: 0.003692 loss: 2.3131 (2.2804) weight_decay: 0.0500 (0.0500) time: 1.5829 data: 0.8558 max mem: 64948 Epoch: [97] [ 20/312] eta: 0:05:38 lr: 0.003691 min_lr: 0.003691 loss: 2.2367 (2.2605) weight_decay: 0.0500 (0.0500) time: 0.7041 data: 0.0004 max mem: 64948 Epoch: [97] [ 30/312] eta: 0:04:44 lr: 0.003691 min_lr: 0.003691 loss: 2.2367 (2.2404) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [97] [ 40/312] eta: 0:04:13 lr: 0.003691 min_lr: 0.003691 loss: 2.3644 (2.2911) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [97] [ 50/312] eta: 0:03:52 lr: 0.003691 min_lr: 0.003691 loss: 2.5281 (2.2969) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [97] [ 60/312] eta: 0:03:35 lr: 0.003690 min_lr: 0.003690 loss: 2.3777 (2.3113) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [97] [ 70/312] eta: 0:03:21 lr: 0.003690 min_lr: 0.003690 loss: 2.3971 (2.3200) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [97] [ 80/312] eta: 0:03:09 lr: 0.003690 min_lr: 0.003690 loss: 2.4017 (2.3273) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [97] [ 90/312] eta: 0:02:58 lr: 0.003690 min_lr: 0.003690 loss: 2.3097 (2.3150) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [97] [100/312] eta: 0:02:48 lr: 0.003689 min_lr: 0.003689 loss: 2.2757 (2.3306) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [97] [110/312] eta: 0:02:38 lr: 0.003689 min_lr: 0.003689 loss: 2.3316 (2.3233) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [97] [120/312] eta: 0:02:29 lr: 0.003689 min_lr: 0.003689 loss: 2.2787 (2.3151) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [97] [130/312] eta: 0:02:20 lr: 0.003689 min_lr: 0.003689 loss: 2.2787 (2.3161) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [97] [140/312] eta: 0:02:11 lr: 0.003688 min_lr: 0.003688 loss: 2.3148 (2.3078) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [97] [150/312] eta: 0:02:03 lr: 0.003688 min_lr: 0.003688 loss: 2.1153 (2.2986) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [97] [160/312] eta: 0:01:54 lr: 0.003688 min_lr: 0.003688 loss: 2.1422 (2.3068) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [97] [170/312] eta: 0:01:46 lr: 0.003688 min_lr: 0.003688 loss: 2.4776 (2.3131) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [97] [180/312] eta: 0:01:38 lr: 0.003687 min_lr: 0.003687 loss: 2.4776 (2.3198) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [97] [190/312] eta: 0:01:31 lr: 0.003687 min_lr: 0.003687 loss: 2.3431 (2.3129) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [97] [200/312] eta: 0:01:23 lr: 0.003687 min_lr: 0.003687 loss: 2.2905 (2.3151) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [97] [210/312] eta: 0:01:15 lr: 0.003687 min_lr: 0.003687 loss: 2.3865 (2.3153) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [97] [220/312] eta: 0:01:08 lr: 0.003686 min_lr: 0.003686 loss: 2.3865 (2.3242) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [97] [230/312] eta: 0:01:00 lr: 0.003686 min_lr: 0.003686 loss: 2.4448 (2.3239) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [97] [240/312] eta: 0:00:52 lr: 0.003686 min_lr: 0.003686 loss: 2.2638 (2.3192) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [97] [250/312] eta: 0:00:45 lr: 0.003686 min_lr: 0.003686 loss: 2.0056 (2.3129) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [97] [260/312] eta: 0:00:38 lr: 0.003685 min_lr: 0.003685 loss: 2.0380 (2.3070) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [97] [270/312] eta: 0:00:30 lr: 0.003685 min_lr: 0.003685 loss: 2.0409 (2.3010) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [97] [280/312] eta: 0:00:23 lr: 0.003685 min_lr: 0.003685 loss: 2.1248 (2.3024) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0012 max mem: 64948 Epoch: [97] [290/312] eta: 0:00:16 lr: 0.003685 min_lr: 0.003685 loss: 2.4875 (2.3045) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0011 max mem: 64948 Epoch: [97] [300/312] eta: 0:00:08 lr: 0.003684 min_lr: 0.003684 loss: 2.2595 (2.2996) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [97] [310/312] eta: 0:00:01 lr: 0.003684 min_lr: 0.003684 loss: 2.1301 (2.2958) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [97] [311/312] eta: 0:00:00 lr: 0.003684 min_lr: 0.003684 loss: 2.1301 (2.2967) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [97] Total time: 0:03:47 (0.7291 s / it) Averaged stats: lr: 0.003684 min_lr: 0.003684 loss: 2.1301 (2.3201) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7966 (0.7966) acc1: 79.4271 (79.4271) acc5: 95.0521 (95.0521) time: 4.5641 data: 4.3446 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0025 (0.9661) acc1: 76.3021 (74.6240) acc5: 93.4896 (93.2160) time: 0.6584 data: 0.4828 max mem: 64948 Test: Total time: 0:00:06 (0.6816 s / it) * Acc@1 75.696 Acc@5 93.048 loss 0.944 Accuracy of the model on the 50000 test images: 75.7% Max accuracy: 75.70% Test: [0/9] eta: 0:00:40 loss: 0.8898 (0.8898) acc1: 78.6458 (78.6458) acc5: 92.9688 (92.9688) time: 4.5471 data: 4.3301 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0332 (1.0615) acc1: 73.9583 (72.8960) acc5: 91.4062 (90.9120) time: 0.6566 data: 0.4812 max mem: 64948 Test: Total time: 0:00:05 (0.6640 s / it) * Acc@1 73.220 Acc@5 91.594 loss 1.054 Accuracy of the model EMA on 50000 test images: 73.2% Max EMA accuracy: 73.22% Epoch: [98] [ 0/312] eta: 0:47:42 lr: 0.003684 min_lr: 0.003684 loss: 2.3832 (2.3832) weight_decay: 0.0500 (0.0500) time: 9.1734 data: 8.3687 max mem: 64948 Epoch: [98] [ 10/312] eta: 0:07:30 lr: 0.003684 min_lr: 0.003684 loss: 2.3705 (2.3534) weight_decay: 0.0500 (0.0500) time: 1.4910 data: 0.7621 max mem: 64948 Epoch: [98] [ 20/312] eta: 0:05:24 lr: 0.003684 min_lr: 0.003684 loss: 2.3521 (2.3388) weight_decay: 0.0500 (0.0500) time: 0.7076 data: 0.0009 max mem: 64948 Epoch: [98] [ 30/312] eta: 0:04:35 lr: 0.003683 min_lr: 0.003683 loss: 2.4520 (2.3399) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0003 max mem: 64948 Epoch: [98] [ 40/312] eta: 0:04:06 lr: 0.003683 min_lr: 0.003683 loss: 2.4520 (2.3501) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [98] [ 50/312] eta: 0:03:46 lr: 0.003683 min_lr: 0.003683 loss: 2.4629 (2.3547) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [98] [ 60/312] eta: 0:03:31 lr: 0.003683 min_lr: 0.003683 loss: 2.2565 (2.3322) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [98] [ 70/312] eta: 0:03:18 lr: 0.003682 min_lr: 0.003682 loss: 2.1835 (2.3138) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [98] [ 80/312] eta: 0:03:06 lr: 0.003682 min_lr: 0.003682 loss: 2.3803 (2.3474) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [98] [ 90/312] eta: 0:02:55 lr: 0.003682 min_lr: 0.003682 loss: 2.5550 (2.3628) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [98] [100/312] eta: 0:02:45 lr: 0.003681 min_lr: 0.003681 loss: 2.4583 (2.3450) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [98] [110/312] eta: 0:02:36 lr: 0.003681 min_lr: 0.003681 loss: 2.0162 (2.3158) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [98] [120/312] eta: 0:02:27 lr: 0.003681 min_lr: 0.003681 loss: 2.1074 (2.3228) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [98] [130/312] eta: 0:02:18 lr: 0.003681 min_lr: 0.003681 loss: 2.4847 (2.3215) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [98] [140/312] eta: 0:02:10 lr: 0.003680 min_lr: 0.003680 loss: 2.4997 (2.3302) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [98] [150/312] eta: 0:02:02 lr: 0.003680 min_lr: 0.003680 loss: 2.3882 (2.3298) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [98] [160/312] eta: 0:01:53 lr: 0.003680 min_lr: 0.003680 loss: 2.3105 (2.3274) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [98] [170/312] eta: 0:01:45 lr: 0.003680 min_lr: 0.003680 loss: 2.3666 (2.3273) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [98] [180/312] eta: 0:01:38 lr: 0.003679 min_lr: 0.003679 loss: 2.3892 (2.3322) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [98] [190/312] eta: 0:01:30 lr: 0.003679 min_lr: 0.003679 loss: 2.4236 (2.3348) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [98] [200/312] eta: 0:01:22 lr: 0.003679 min_lr: 0.003679 loss: 2.4248 (2.3393) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [98] [210/312] eta: 0:01:15 lr: 0.003679 min_lr: 0.003679 loss: 2.3623 (2.3383) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [98] [220/312] eta: 0:01:07 lr: 0.003678 min_lr: 0.003678 loss: 2.4179 (2.3454) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [98] [230/312] eta: 0:01:00 lr: 0.003678 min_lr: 0.003678 loss: 2.4554 (2.3492) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [98] [240/312] eta: 0:00:52 lr: 0.003678 min_lr: 0.003678 loss: 2.5346 (2.3634) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [98] [250/312] eta: 0:00:45 lr: 0.003678 min_lr: 0.003678 loss: 2.6054 (2.3718) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [98] [260/312] eta: 0:00:37 lr: 0.003677 min_lr: 0.003677 loss: 2.5446 (2.3701) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [98] [270/312] eta: 0:00:30 lr: 0.003677 min_lr: 0.003677 loss: 2.3706 (2.3679) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [98] [280/312] eta: 0:00:23 lr: 0.003677 min_lr: 0.003677 loss: 2.3322 (2.3674) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [98] [290/312] eta: 0:00:15 lr: 0.003677 min_lr: 0.003677 loss: 2.5802 (2.3747) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [98] [300/312] eta: 0:00:08 lr: 0.003676 min_lr: 0.003676 loss: 2.5309 (2.3771) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [98] [310/312] eta: 0:00:01 lr: 0.003676 min_lr: 0.003676 loss: 2.4013 (2.3744) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [98] [311/312] eta: 0:00:00 lr: 0.003676 min_lr: 0.003676 loss: 2.4463 (2.3756) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [98] Total time: 0:03:46 (0.7258 s / it) Averaged stats: lr: 0.003676 min_lr: 0.003676 loss: 2.4463 (2.3173) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.7988 (0.7988) acc1: 80.9896 (80.9896) acc5: 94.5312 (94.5312) time: 4.8274 data: 4.6077 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9514 (0.9846) acc1: 75.7812 (74.5600) acc5: 92.7083 (92.6400) time: 0.6881 data: 0.5121 max mem: 64948 Test: Total time: 0:00:06 (0.7127 s / it) * Acc@1 75.598 Acc@5 92.940 loss 0.952 Accuracy of the model on the 50000 test images: 75.6% Max accuracy: 75.70% Test: [0/9] eta: 0:00:42 loss: 0.8774 (0.8774) acc1: 78.6458 (78.6458) acc5: 92.9688 (92.9688) time: 4.7508 data: 4.5329 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0192 (1.0480) acc1: 74.4792 (73.2160) acc5: 92.1875 (91.1360) time: 0.6796 data: 0.5038 max mem: 64948 Test: Total time: 0:00:06 (0.6904 s / it) * Acc@1 73.514 Acc@5 91.724 loss 1.040 Accuracy of the model EMA on 50000 test images: 73.5% Max EMA accuracy: 73.51% Epoch: [99] [ 0/312] eta: 0:56:18 lr: 0.003676 min_lr: 0.003676 loss: 2.4366 (2.4366) weight_decay: 0.0500 (0.0500) time: 10.8293 data: 10.1048 max mem: 64948 Epoch: [99] [ 10/312] eta: 0:08:13 lr: 0.003676 min_lr: 0.003676 loss: 2.2828 (2.1819) weight_decay: 0.0500 (0.0500) time: 1.6335 data: 0.9189 max mem: 64948 Epoch: [99] [ 20/312] eta: 0:05:46 lr: 0.003676 min_lr: 0.003676 loss: 2.2828 (2.2628) weight_decay: 0.0500 (0.0500) time: 0.7047 data: 0.0003 max mem: 64948 Epoch: [99] [ 30/312] eta: 0:04:49 lr: 0.003675 min_lr: 0.003675 loss: 2.4098 (2.3162) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [99] [ 40/312] eta: 0:04:17 lr: 0.003675 min_lr: 0.003675 loss: 2.3403 (2.3030) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [99] [ 50/312] eta: 0:03:55 lr: 0.003675 min_lr: 0.003675 loss: 2.4558 (2.3102) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [99] [ 60/312] eta: 0:03:37 lr: 0.003675 min_lr: 0.003675 loss: 2.4558 (2.3026) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [99] [ 70/312] eta: 0:03:23 lr: 0.003674 min_lr: 0.003674 loss: 2.3907 (2.2913) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [99] [ 80/312] eta: 0:03:11 lr: 0.003674 min_lr: 0.003674 loss: 2.3907 (2.2849) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [99] [ 90/312] eta: 0:02:59 lr: 0.003674 min_lr: 0.003674 loss: 2.3357 (2.2990) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [99] [100/312] eta: 0:02:49 lr: 0.003674 min_lr: 0.003674 loss: 2.3336 (2.2928) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [99] [110/312] eta: 0:02:39 lr: 0.003673 min_lr: 0.003673 loss: 2.2910 (2.2913) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [99] [120/312] eta: 0:02:30 lr: 0.003673 min_lr: 0.003673 loss: 2.1344 (2.2796) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [99] [130/312] eta: 0:02:21 lr: 0.003673 min_lr: 0.003673 loss: 2.2674 (2.2833) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [99] [140/312] eta: 0:02:12 lr: 0.003673 min_lr: 0.003673 loss: 2.3054 (2.2852) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [99] [150/312] eta: 0:02:03 lr: 0.003672 min_lr: 0.003672 loss: 2.3267 (2.2882) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [99] [160/312] eta: 0:01:55 lr: 0.003672 min_lr: 0.003672 loss: 2.4448 (2.3075) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [99] [170/312] eta: 0:01:47 lr: 0.003672 min_lr: 0.003672 loss: 2.5658 (2.3148) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [99] [180/312] eta: 0:01:39 lr: 0.003671 min_lr: 0.003671 loss: 2.5546 (2.3239) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0005 max mem: 64948 Epoch: [99] [190/312] eta: 0:01:31 lr: 0.003671 min_lr: 0.003671 loss: 2.3007 (2.3205) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0005 max mem: 64948 Epoch: [99] [200/312] eta: 0:01:23 lr: 0.003671 min_lr: 0.003671 loss: 2.3946 (2.3297) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0005 max mem: 64948 Epoch: [99] [210/312] eta: 0:01:15 lr: 0.003671 min_lr: 0.003671 loss: 2.3946 (2.3329) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0005 max mem: 64948 Epoch: [99] [220/312] eta: 0:01:08 lr: 0.003670 min_lr: 0.003670 loss: 2.4500 (2.3381) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0005 max mem: 64948 Epoch: [99] [230/312] eta: 0:01:00 lr: 0.003670 min_lr: 0.003670 loss: 2.4402 (2.3381) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0005 max mem: 64948 Epoch: [99] [240/312] eta: 0:00:53 lr: 0.003670 min_lr: 0.003670 loss: 2.4835 (2.3468) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0006 max mem: 64948 Epoch: [99] [250/312] eta: 0:00:45 lr: 0.003670 min_lr: 0.003670 loss: 2.5309 (2.3524) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0006 max mem: 64948 Epoch: [99] [260/312] eta: 0:00:38 lr: 0.003669 min_lr: 0.003669 loss: 2.4596 (2.3499) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0006 max mem: 64948 Epoch: [99] [270/312] eta: 0:00:30 lr: 0.003669 min_lr: 0.003669 loss: 2.4335 (2.3506) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0006 max mem: 64948 Epoch: [99] [280/312] eta: 0:00:23 lr: 0.003669 min_lr: 0.003669 loss: 2.3330 (2.3446) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0015 max mem: 64948 Epoch: [99] [290/312] eta: 0:00:16 lr: 0.003669 min_lr: 0.003669 loss: 2.2185 (2.3405) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0013 max mem: 64948 Epoch: [99] [300/312] eta: 0:00:08 lr: 0.003668 min_lr: 0.003668 loss: 2.2321 (2.3378) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0002 max mem: 64948 Epoch: [99] [310/312] eta: 0:00:01 lr: 0.003668 min_lr: 0.003668 loss: 2.2278 (2.3337) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [99] [311/312] eta: 0:00:00 lr: 0.003668 min_lr: 0.003668 loss: 2.2278 (2.3347) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [99] Total time: 0:03:48 (0.7309 s / it) Averaged stats: lr: 0.003668 min_lr: 0.003668 loss: 2.2278 (2.3107) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.8398 (0.8398) acc1: 80.2083 (80.2083) acc5: 93.7500 (93.7500) time: 4.7409 data: 4.5372 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0569 (1.0032) acc1: 75.5208 (74.1440) acc5: 92.4479 (92.3520) time: 0.6781 data: 0.5042 max mem: 64948 Test: Total time: 0:00:06 (0.7041 s / it) * Acc@1 75.366 Acc@5 92.422 loss 0.983 Accuracy of the model on the 50000 test images: 75.4% Max accuracy: 75.70% Test: [0/9] eta: 0:00:41 loss: 0.8658 (0.8658) acc1: 78.9062 (78.9062) acc5: 92.9688 (92.9688) time: 4.5848 data: 4.3667 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0063 (1.0348) acc1: 74.4792 (73.3760) acc5: 92.4479 (91.3600) time: 0.6691 data: 0.4936 max mem: 64948 Test: Total time: 0:00:06 (0.6872 s / it) * Acc@1 73.816 Acc@5 91.922 loss 1.026 Accuracy of the model EMA on 50000 test images: 73.8% Max EMA accuracy: 73.82% Epoch: [100] [ 0/312] eta: 0:46:41 lr: 0.003668 min_lr: 0.003668 loss: 1.6584 (1.6584) weight_decay: 0.0500 (0.0500) time: 8.9777 data: 7.0920 max mem: 64948 Epoch: [100] [ 10/312] eta: 0:07:27 lr: 0.003668 min_lr: 0.003668 loss: 2.1777 (2.0996) weight_decay: 0.0500 (0.0500) time: 1.4817 data: 0.6452 max mem: 64948 Epoch: [100] [ 20/312] eta: 0:05:23 lr: 0.003668 min_lr: 0.003668 loss: 2.2256 (2.2011) weight_decay: 0.0500 (0.0500) time: 0.7140 data: 0.0005 max mem: 64948 Epoch: [100] [ 30/312] eta: 0:04:34 lr: 0.003667 min_lr: 0.003667 loss: 2.3597 (2.2535) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [100] [ 40/312] eta: 0:04:06 lr: 0.003667 min_lr: 0.003667 loss: 2.3597 (2.2772) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [100] [ 50/312] eta: 0:03:46 lr: 0.003667 min_lr: 0.003667 loss: 2.4500 (2.2933) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [100] [ 60/312] eta: 0:03:30 lr: 0.003667 min_lr: 0.003667 loss: 2.4266 (2.2990) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [100] [ 70/312] eta: 0:03:17 lr: 0.003666 min_lr: 0.003666 loss: 2.3979 (2.3044) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [100] [ 80/312] eta: 0:03:06 lr: 0.003666 min_lr: 0.003666 loss: 2.4775 (2.3200) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [100] [ 90/312] eta: 0:02:55 lr: 0.003666 min_lr: 0.003666 loss: 2.3092 (2.3156) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [100] [100/312] eta: 0:02:45 lr: 0.003665 min_lr: 0.003665 loss: 2.1179 (2.2917) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [100] [110/312] eta: 0:02:36 lr: 0.003665 min_lr: 0.003665 loss: 2.2572 (2.2928) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [100] [120/312] eta: 0:02:27 lr: 0.003665 min_lr: 0.003665 loss: 2.1082 (2.2708) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [100] [130/312] eta: 0:02:18 lr: 0.003665 min_lr: 0.003665 loss: 2.2891 (2.2849) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [100] [140/312] eta: 0:02:10 lr: 0.003664 min_lr: 0.003664 loss: 2.5140 (2.3063) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [100] [150/312] eta: 0:02:01 lr: 0.003664 min_lr: 0.003664 loss: 2.3924 (2.2973) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [100] [160/312] eta: 0:01:53 lr: 0.003664 min_lr: 0.003664 loss: 2.3174 (2.2999) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [100] [170/312] eta: 0:01:45 lr: 0.003664 min_lr: 0.003664 loss: 2.4521 (2.3014) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [100] [180/312] eta: 0:01:38 lr: 0.003663 min_lr: 0.003663 loss: 2.3244 (2.3007) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [100] [190/312] eta: 0:01:30 lr: 0.003663 min_lr: 0.003663 loss: 2.4782 (2.3126) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [100] [200/312] eta: 0:01:22 lr: 0.003663 min_lr: 0.003663 loss: 2.4148 (2.3055) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [100] [210/312] eta: 0:01:15 lr: 0.003663 min_lr: 0.003663 loss: 2.2984 (2.3045) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [100] [220/312] eta: 0:01:07 lr: 0.003662 min_lr: 0.003662 loss: 2.3608 (2.3096) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [100] [230/312] eta: 0:01:00 lr: 0.003662 min_lr: 0.003662 loss: 2.5195 (2.3175) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [100] [240/312] eta: 0:00:52 lr: 0.003662 min_lr: 0.003662 loss: 2.6306 (2.3270) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [100] [250/312] eta: 0:00:45 lr: 0.003662 min_lr: 0.003662 loss: 2.3985 (2.3241) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [100] [260/312] eta: 0:00:37 lr: 0.003661 min_lr: 0.003661 loss: 2.3537 (2.3276) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [100] [270/312] eta: 0:00:30 lr: 0.003661 min_lr: 0.003661 loss: 2.3796 (2.3252) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [100] [280/312] eta: 0:00:23 lr: 0.003661 min_lr: 0.003661 loss: 2.3687 (2.3290) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [100] [290/312] eta: 0:00:15 lr: 0.003661 min_lr: 0.003661 loss: 2.4111 (2.3277) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [100] [300/312] eta: 0:00:08 lr: 0.003660 min_lr: 0.003660 loss: 2.4111 (2.3292) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [100] [310/312] eta: 0:00:01 lr: 0.003660 min_lr: 0.003660 loss: 2.3725 (2.3273) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [100] [311/312] eta: 0:00:00 lr: 0.003660 min_lr: 0.003660 loss: 2.3245 (2.3254) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [100] Total time: 0:03:46 (0.7253 s / it) Averaged stats: lr: 0.003660 min_lr: 0.003660 loss: 2.3245 (2.3136) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:45 loss: 0.8472 (0.8472) acc1: 79.4271 (79.4271) acc5: 94.5312 (94.5312) time: 5.0629 data: 4.8437 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0387 (0.9859) acc1: 73.6979 (75.1040) acc5: 93.2292 (92.4800) time: 0.7140 data: 0.5383 max mem: 64948 Test: Total time: 0:00:06 (0.7398 s / it) * Acc@1 75.098 Acc@5 92.642 loss 0.974 Accuracy of the model on the 50000 test images: 75.1% Max accuracy: 75.70% Test: [0/9] eta: 0:00:44 loss: 0.8545 (0.8545) acc1: 79.1667 (79.1667) acc5: 93.4896 (93.4896) time: 4.9751 data: 4.7624 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9956 (1.0225) acc1: 74.4792 (73.5680) acc5: 92.7083 (91.4880) time: 0.7041 data: 0.5292 max mem: 64948 Test: Total time: 0:00:06 (0.7187 s / it) * Acc@1 74.088 Acc@5 92.036 loss 1.014 Accuracy of the model EMA on 50000 test images: 74.1% Max EMA accuracy: 74.09% Epoch: [101] [ 0/312] eta: 0:46:22 lr: 0.003660 min_lr: 0.003660 loss: 2.3725 (2.3725) weight_decay: 0.0500 (0.0500) time: 8.9186 data: 8.1467 max mem: 64948 Epoch: [101] [ 10/312] eta: 0:07:35 lr: 0.003660 min_lr: 0.003660 loss: 2.3623 (2.4134) weight_decay: 0.0500 (0.0500) time: 1.5083 data: 0.7410 max mem: 64948 Epoch: [101] [ 20/312] eta: 0:05:27 lr: 0.003659 min_lr: 0.003659 loss: 2.3356 (2.3481) weight_decay: 0.0500 (0.0500) time: 0.7313 data: 0.0004 max mem: 64948 Epoch: [101] [ 30/312] eta: 0:04:37 lr: 0.003659 min_lr: 0.003659 loss: 2.3101 (2.3104) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [101] [ 40/312] eta: 0:04:08 lr: 0.003659 min_lr: 0.003659 loss: 1.9031 (2.2611) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [101] [ 50/312] eta: 0:03:48 lr: 0.003659 min_lr: 0.003659 loss: 2.2018 (2.2525) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [101] [ 60/312] eta: 0:03:32 lr: 0.003658 min_lr: 0.003658 loss: 2.3186 (2.2694) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [101] [ 70/312] eta: 0:03:18 lr: 0.003658 min_lr: 0.003658 loss: 2.5066 (2.3028) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [101] [ 80/312] eta: 0:03:07 lr: 0.003658 min_lr: 0.003658 loss: 2.5292 (2.3153) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [101] [ 90/312] eta: 0:02:56 lr: 0.003658 min_lr: 0.003658 loss: 2.3381 (2.3069) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [101] [100/312] eta: 0:02:46 lr: 0.003657 min_lr: 0.003657 loss: 2.1661 (2.2857) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [101] [110/312] eta: 0:02:36 lr: 0.003657 min_lr: 0.003657 loss: 2.0709 (2.2844) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [101] [120/312] eta: 0:02:27 lr: 0.003657 min_lr: 0.003657 loss: 2.2658 (2.2808) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [101] [130/312] eta: 0:02:19 lr: 0.003657 min_lr: 0.003657 loss: 2.3202 (2.2927) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [101] [140/312] eta: 0:02:10 lr: 0.003656 min_lr: 0.003656 loss: 2.4678 (2.3005) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [101] [150/312] eta: 0:02:02 lr: 0.003656 min_lr: 0.003656 loss: 2.4210 (2.2965) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [101] [160/312] eta: 0:01:54 lr: 0.003656 min_lr: 0.003656 loss: 2.4210 (2.3041) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [101] [170/312] eta: 0:01:46 lr: 0.003656 min_lr: 0.003656 loss: 2.4931 (2.3171) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [101] [180/312] eta: 0:01:38 lr: 0.003655 min_lr: 0.003655 loss: 2.3717 (2.3119) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [101] [190/312] eta: 0:01:30 lr: 0.003655 min_lr: 0.003655 loss: 2.3805 (2.3227) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [101] [200/312] eta: 0:01:22 lr: 0.003655 min_lr: 0.003655 loss: 2.4866 (2.3248) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [101] [210/312] eta: 0:01:15 lr: 0.003654 min_lr: 0.003654 loss: 2.3354 (2.3225) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [101] [220/312] eta: 0:01:07 lr: 0.003654 min_lr: 0.003654 loss: 2.3354 (2.3202) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [101] [230/312] eta: 0:01:00 lr: 0.003654 min_lr: 0.003654 loss: 2.2891 (2.3164) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [101] [240/312] eta: 0:00:52 lr: 0.003654 min_lr: 0.003654 loss: 2.2891 (2.3235) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [101] [250/312] eta: 0:00:45 lr: 0.003653 min_lr: 0.003653 loss: 2.4697 (2.3272) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [101] [260/312] eta: 0:00:37 lr: 0.003653 min_lr: 0.003653 loss: 2.4737 (2.3312) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [101] [270/312] eta: 0:00:30 lr: 0.003653 min_lr: 0.003653 loss: 2.4800 (2.3343) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [101] [280/312] eta: 0:00:23 lr: 0.003653 min_lr: 0.003653 loss: 2.4586 (2.3372) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [101] [290/312] eta: 0:00:15 lr: 0.003652 min_lr: 0.003652 loss: 2.4468 (2.3390) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [101] [300/312] eta: 0:00:08 lr: 0.003652 min_lr: 0.003652 loss: 2.3460 (2.3359) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [101] [310/312] eta: 0:00:01 lr: 0.003652 min_lr: 0.003652 loss: 2.3414 (2.3358) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [101] [311/312] eta: 0:00:00 lr: 0.003652 min_lr: 0.003652 loss: 2.3158 (2.3351) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [101] Total time: 0:03:46 (0.7265 s / it) Averaged stats: lr: 0.003652 min_lr: 0.003652 loss: 2.3158 (2.3188) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7494 (0.7494) acc1: 80.7292 (80.7292) acc5: 94.2708 (94.2708) time: 4.4606 data: 4.2476 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0433 (0.9732) acc1: 76.0417 (74.6560) acc5: 92.9688 (92.2880) time: 0.6469 data: 0.4720 max mem: 64948 Test: Total time: 0:00:06 (0.6717 s / it) * Acc@1 75.760 Acc@5 92.748 loss 0.953 Accuracy of the model on the 50000 test images: 75.8% Max accuracy: 75.76% Test: [0/9] eta: 0:00:40 loss: 0.8427 (0.8427) acc1: 79.4271 (79.4271) acc5: 93.4896 (93.4896) time: 4.4572 data: 4.2392 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9852 (1.0109) acc1: 75.2604 (73.9840) acc5: 92.9688 (91.5840) time: 0.6466 data: 0.4711 max mem: 64948 Test: Total time: 0:00:05 (0.6541 s / it) * Acc@1 74.368 Acc@5 92.204 loss 1.001 Accuracy of the model EMA on 50000 test images: 74.4% Max EMA accuracy: 74.37% Epoch: [102] [ 0/312] eta: 0:45:04 lr: 0.003652 min_lr: 0.003652 loss: 2.7797 (2.7797) weight_decay: 0.0500 (0.0500) time: 8.6682 data: 7.6489 max mem: 64948 Epoch: [102] [ 10/312] eta: 0:07:18 lr: 0.003652 min_lr: 0.003652 loss: 2.4234 (2.3964) weight_decay: 0.0500 (0.0500) time: 1.4506 data: 0.6959 max mem: 64948 Epoch: [102] [ 20/312] eta: 0:05:18 lr: 0.003651 min_lr: 0.003651 loss: 2.2827 (2.3040) weight_decay: 0.0500 (0.0500) time: 0.7117 data: 0.0005 max mem: 64948 Epoch: [102] [ 30/312] eta: 0:04:32 lr: 0.003651 min_lr: 0.003651 loss: 2.1903 (2.2592) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [102] [ 40/312] eta: 0:04:04 lr: 0.003651 min_lr: 0.003651 loss: 2.2846 (2.2710) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [102] [ 50/312] eta: 0:03:45 lr: 0.003650 min_lr: 0.003650 loss: 2.4230 (2.2824) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [102] [ 60/312] eta: 0:03:29 lr: 0.003650 min_lr: 0.003650 loss: 2.3148 (2.2708) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [102] [ 70/312] eta: 0:03:16 lr: 0.003650 min_lr: 0.003650 loss: 2.1094 (2.2609) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [102] [ 80/312] eta: 0:03:05 lr: 0.003650 min_lr: 0.003650 loss: 2.1872 (2.2660) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [102] [ 90/312] eta: 0:02:54 lr: 0.003649 min_lr: 0.003649 loss: 2.3470 (2.2760) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [102] [100/312] eta: 0:02:45 lr: 0.003649 min_lr: 0.003649 loss: 2.4606 (2.2944) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [102] [110/312] eta: 0:02:35 lr: 0.003649 min_lr: 0.003649 loss: 2.4359 (2.3032) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [102] [120/312] eta: 0:02:26 lr: 0.003649 min_lr: 0.003649 loss: 2.4472 (2.3070) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [102] [130/312] eta: 0:02:18 lr: 0.003648 min_lr: 0.003648 loss: 2.4714 (2.3132) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [102] [140/312] eta: 0:02:09 lr: 0.003648 min_lr: 0.003648 loss: 2.3263 (2.3014) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [102] [150/312] eta: 0:02:01 lr: 0.003648 min_lr: 0.003648 loss: 2.2707 (2.2932) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [102] [160/312] eta: 0:01:53 lr: 0.003648 min_lr: 0.003648 loss: 1.9573 (2.2776) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [102] [170/312] eta: 0:01:45 lr: 0.003647 min_lr: 0.003647 loss: 2.0993 (2.2809) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [102] [180/312] eta: 0:01:37 lr: 0.003647 min_lr: 0.003647 loss: 2.3807 (2.2812) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [102] [190/312] eta: 0:01:30 lr: 0.003647 min_lr: 0.003647 loss: 2.1973 (2.2739) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [102] [200/312] eta: 0:01:22 lr: 0.003646 min_lr: 0.003646 loss: 2.3148 (2.2779) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [102] [210/312] eta: 0:01:14 lr: 0.003646 min_lr: 0.003646 loss: 2.3492 (2.2790) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [102] [220/312] eta: 0:01:07 lr: 0.003646 min_lr: 0.003646 loss: 2.1689 (2.2728) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [102] [230/312] eta: 0:00:59 lr: 0.003646 min_lr: 0.003646 loss: 2.2106 (2.2722) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [102] [240/312] eta: 0:00:52 lr: 0.003645 min_lr: 0.003645 loss: 2.3137 (2.2755) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [102] [250/312] eta: 0:00:45 lr: 0.003645 min_lr: 0.003645 loss: 2.3650 (2.2803) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [102] [260/312] eta: 0:00:37 lr: 0.003645 min_lr: 0.003645 loss: 2.4175 (2.2832) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [102] [270/312] eta: 0:00:30 lr: 0.003645 min_lr: 0.003645 loss: 2.4141 (2.2846) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [102] [280/312] eta: 0:00:23 lr: 0.003644 min_lr: 0.003644 loss: 2.4141 (2.2893) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [102] [290/312] eta: 0:00:15 lr: 0.003644 min_lr: 0.003644 loss: 2.4067 (2.2918) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [102] [300/312] eta: 0:00:08 lr: 0.003644 min_lr: 0.003644 loss: 2.5093 (2.3016) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [102] [310/312] eta: 0:00:01 lr: 0.003644 min_lr: 0.003644 loss: 2.4406 (2.2987) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [102] [311/312] eta: 0:00:00 lr: 0.003644 min_lr: 0.003644 loss: 2.4235 (2.2976) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [102] Total time: 0:03:46 (0.7244 s / it) Averaged stats: lr: 0.003644 min_lr: 0.003644 loss: 2.4235 (2.3129) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7082 (0.7082) acc1: 82.2917 (82.2917) acc5: 94.5312 (94.5312) time: 4.5582 data: 4.3539 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9651 (0.9915) acc1: 74.7396 (74.5920) acc5: 92.4479 (92.6720) time: 0.6578 data: 0.4839 max mem: 64948 Test: Total time: 0:00:06 (0.6813 s / it) * Acc@1 75.148 Acc@5 92.642 loss 0.978 Accuracy of the model on the 50000 test images: 75.1% Max accuracy: 75.76% Test: [0/9] eta: 0:00:44 loss: 0.8305 (0.8305) acc1: 79.4271 (79.4271) acc5: 93.4896 (93.4896) time: 4.9868 data: 4.7688 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9753 (0.9999) acc1: 75.2604 (74.2720) acc5: 92.9688 (91.7120) time: 0.7055 data: 0.5300 max mem: 64948 Test: Total time: 0:00:06 (0.7147 s / it) * Acc@1 74.660 Acc@5 92.352 loss 0.989 Accuracy of the model EMA on 50000 test images: 74.7% Max EMA accuracy: 74.66% Epoch: [103] [ 0/312] eta: 0:51:33 lr: 0.003643 min_lr: 0.003643 loss: 2.7424 (2.7424) weight_decay: 0.0500 (0.0500) time: 9.9155 data: 9.1351 max mem: 64948 Epoch: [103] [ 10/312] eta: 0:07:50 lr: 0.003643 min_lr: 0.003643 loss: 2.4703 (2.4456) weight_decay: 0.0500 (0.0500) time: 1.5570 data: 0.8308 max mem: 64948 Epoch: [103] [ 20/312] eta: 0:05:35 lr: 0.003643 min_lr: 0.003643 loss: 2.4664 (2.4155) weight_decay: 0.0500 (0.0500) time: 0.7101 data: 0.0003 max mem: 64948 Epoch: [103] [ 30/312] eta: 0:04:42 lr: 0.003643 min_lr: 0.003643 loss: 2.4664 (2.4141) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0003 max mem: 64948 Epoch: [103] [ 40/312] eta: 0:04:12 lr: 0.003642 min_lr: 0.003642 loss: 2.4247 (2.4064) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [103] [ 50/312] eta: 0:03:50 lr: 0.003642 min_lr: 0.003642 loss: 2.4247 (2.4069) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [103] [ 60/312] eta: 0:03:34 lr: 0.003642 min_lr: 0.003642 loss: 2.4594 (2.4034) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [103] [ 70/312] eta: 0:03:20 lr: 0.003642 min_lr: 0.003642 loss: 2.3572 (2.3848) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [103] [ 80/312] eta: 0:03:08 lr: 0.003641 min_lr: 0.003641 loss: 2.3572 (2.3822) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [103] [ 90/312] eta: 0:02:57 lr: 0.003641 min_lr: 0.003641 loss: 2.3527 (2.3819) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [103] [100/312] eta: 0:02:47 lr: 0.003641 min_lr: 0.003641 loss: 2.3491 (2.3614) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [103] [110/312] eta: 0:02:37 lr: 0.003641 min_lr: 0.003641 loss: 2.3245 (2.3665) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [103] [120/312] eta: 0:02:28 lr: 0.003640 min_lr: 0.003640 loss: 2.3381 (2.3599) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [103] [130/312] eta: 0:02:19 lr: 0.003640 min_lr: 0.003640 loss: 2.4390 (2.3589) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [103] [140/312] eta: 0:02:11 lr: 0.003640 min_lr: 0.003640 loss: 2.4994 (2.3614) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [103] [150/312] eta: 0:02:02 lr: 0.003639 min_lr: 0.003639 loss: 2.2303 (2.3522) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [103] [160/312] eta: 0:01:54 lr: 0.003639 min_lr: 0.003639 loss: 2.2303 (2.3476) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [103] [170/312] eta: 0:01:46 lr: 0.003639 min_lr: 0.003639 loss: 2.3774 (2.3517) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [103] [180/312] eta: 0:01:38 lr: 0.003639 min_lr: 0.003639 loss: 2.4938 (2.3583) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [103] [190/312] eta: 0:01:30 lr: 0.003638 min_lr: 0.003638 loss: 2.5305 (2.3571) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [103] [200/312] eta: 0:01:23 lr: 0.003638 min_lr: 0.003638 loss: 2.4265 (2.3608) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [103] [210/312] eta: 0:01:15 lr: 0.003638 min_lr: 0.003638 loss: 2.5088 (2.3687) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [103] [220/312] eta: 0:01:07 lr: 0.003638 min_lr: 0.003638 loss: 2.4045 (2.3616) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [103] [230/312] eta: 0:01:00 lr: 0.003637 min_lr: 0.003637 loss: 2.2882 (2.3560) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [103] [240/312] eta: 0:00:52 lr: 0.003637 min_lr: 0.003637 loss: 2.2179 (2.3557) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [103] [250/312] eta: 0:00:45 lr: 0.003637 min_lr: 0.003637 loss: 2.3092 (2.3578) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [103] [260/312] eta: 0:00:38 lr: 0.003637 min_lr: 0.003637 loss: 2.3092 (2.3525) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [103] [270/312] eta: 0:00:30 lr: 0.003636 min_lr: 0.003636 loss: 2.4120 (2.3541) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [103] [280/312] eta: 0:00:23 lr: 0.003636 min_lr: 0.003636 loss: 2.4150 (2.3480) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0010 max mem: 64948 Epoch: [103] [290/312] eta: 0:00:16 lr: 0.003636 min_lr: 0.003636 loss: 2.4122 (2.3493) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [103] [300/312] eta: 0:00:08 lr: 0.003635 min_lr: 0.003635 loss: 2.4122 (2.3444) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [103] [310/312] eta: 0:00:01 lr: 0.003635 min_lr: 0.003635 loss: 2.4323 (2.3485) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [103] [311/312] eta: 0:00:00 lr: 0.003635 min_lr: 0.003635 loss: 2.4323 (2.3467) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [103] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.003635 min_lr: 0.003635 loss: 2.4323 (2.3064) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7309 (0.7309) acc1: 83.0729 (83.0729) acc5: 95.8333 (95.8333) time: 4.5821 data: 4.3670 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0085 (1.0200) acc1: 75.0000 (74.4960) acc5: 92.9688 (92.5760) time: 0.6605 data: 0.4853 max mem: 64948 Test: Total time: 0:00:06 (0.6799 s / it) * Acc@1 75.076 Acc@5 92.510 loss 0.990 Accuracy of the model on the 50000 test images: 75.1% Max accuracy: 75.76% Test: [0/9] eta: 0:00:45 loss: 0.8196 (0.8196) acc1: 79.4271 (79.4271) acc5: 94.2708 (94.2708) time: 5.0968 data: 4.8790 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9660 (0.9894) acc1: 75.2604 (74.4000) acc5: 92.9688 (91.9680) time: 0.7178 data: 0.5422 max mem: 64948 Test: Total time: 0:00:06 (0.7269 s / it) * Acc@1 74.872 Acc@5 92.476 loss 0.978 Accuracy of the model EMA on 50000 test images: 74.9% Max EMA accuracy: 74.87% Epoch: [104] [ 0/312] eta: 0:51:37 lr: 0.003635 min_lr: 0.003635 loss: 1.7400 (1.7400) weight_decay: 0.0500 (0.0500) time: 9.9276 data: 9.1416 max mem: 64948 Epoch: [104] [ 10/312] eta: 0:07:49 lr: 0.003635 min_lr: 0.003635 loss: 2.1123 (2.1626) weight_decay: 0.0500 (0.0500) time: 1.5551 data: 0.8314 max mem: 64948 Epoch: [104] [ 20/312] eta: 0:05:35 lr: 0.003635 min_lr: 0.003635 loss: 2.1612 (2.2096) weight_decay: 0.0500 (0.0500) time: 0.7094 data: 0.0004 max mem: 64948 Epoch: [104] [ 30/312] eta: 0:04:42 lr: 0.003634 min_lr: 0.003634 loss: 2.4680 (2.2641) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [104] [ 40/312] eta: 0:04:12 lr: 0.003634 min_lr: 0.003634 loss: 2.3961 (2.2260) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [104] [ 50/312] eta: 0:03:51 lr: 0.003634 min_lr: 0.003634 loss: 2.0708 (2.2176) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [104] [ 60/312] eta: 0:03:34 lr: 0.003634 min_lr: 0.003634 loss: 2.3250 (2.2306) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [104] [ 70/312] eta: 0:03:20 lr: 0.003633 min_lr: 0.003633 loss: 2.1976 (2.2252) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [104] [ 80/312] eta: 0:03:08 lr: 0.003633 min_lr: 0.003633 loss: 2.1984 (2.2304) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [104] [ 90/312] eta: 0:02:57 lr: 0.003633 min_lr: 0.003633 loss: 2.3202 (2.2383) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [104] [100/312] eta: 0:02:47 lr: 0.003632 min_lr: 0.003632 loss: 2.3205 (2.2344) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [104] [110/312] eta: 0:02:37 lr: 0.003632 min_lr: 0.003632 loss: 2.2092 (2.2240) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [104] [120/312] eta: 0:02:28 lr: 0.003632 min_lr: 0.003632 loss: 2.3070 (2.2432) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [104] [130/312] eta: 0:02:19 lr: 0.003632 min_lr: 0.003632 loss: 2.4942 (2.2577) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [104] [140/312] eta: 0:02:11 lr: 0.003631 min_lr: 0.003631 loss: 2.3822 (2.2624) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [104] [150/312] eta: 0:02:02 lr: 0.003631 min_lr: 0.003631 loss: 2.2349 (2.2575) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [104] [160/312] eta: 0:01:54 lr: 0.003631 min_lr: 0.003631 loss: 2.1805 (2.2565) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [104] [170/312] eta: 0:01:46 lr: 0.003631 min_lr: 0.003631 loss: 2.1968 (2.2557) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [104] [180/312] eta: 0:01:38 lr: 0.003630 min_lr: 0.003630 loss: 2.0835 (2.2571) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [104] [190/312] eta: 0:01:30 lr: 0.003630 min_lr: 0.003630 loss: 2.4148 (2.2736) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [104] [200/312] eta: 0:01:23 lr: 0.003630 min_lr: 0.003630 loss: 2.5594 (2.2783) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [104] [210/312] eta: 0:01:15 lr: 0.003629 min_lr: 0.003629 loss: 2.2815 (2.2736) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [104] [220/312] eta: 0:01:07 lr: 0.003629 min_lr: 0.003629 loss: 2.2110 (2.2778) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [104] [230/312] eta: 0:01:00 lr: 0.003629 min_lr: 0.003629 loss: 2.3588 (2.2808) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [104] [240/312] eta: 0:00:52 lr: 0.003629 min_lr: 0.003629 loss: 2.3576 (2.2816) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [104] [250/312] eta: 0:00:45 lr: 0.003628 min_lr: 0.003628 loss: 2.2943 (2.2783) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [104] [260/312] eta: 0:00:38 lr: 0.003628 min_lr: 0.003628 loss: 2.2020 (2.2780) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [104] [270/312] eta: 0:00:30 lr: 0.003628 min_lr: 0.003628 loss: 2.2907 (2.2802) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [104] [280/312] eta: 0:00:23 lr: 0.003628 min_lr: 0.003628 loss: 2.3800 (2.2861) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [104] [290/312] eta: 0:00:16 lr: 0.003627 min_lr: 0.003627 loss: 2.5208 (2.2903) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [104] [300/312] eta: 0:00:08 lr: 0.003627 min_lr: 0.003627 loss: 2.3652 (2.2882) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [104] [310/312] eta: 0:00:01 lr: 0.003627 min_lr: 0.003627 loss: 2.2461 (2.2876) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [104] [311/312] eta: 0:00:00 lr: 0.003627 min_lr: 0.003627 loss: 2.2461 (2.2888) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [104] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.003627 min_lr: 0.003627 loss: 2.2461 (2.3161) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7457 (0.7457) acc1: 83.3333 (83.3333) acc5: 93.7500 (93.7500) time: 4.6513 data: 4.4407 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0002 (0.9446) acc1: 75.5208 (74.8160) acc5: 93.7500 (92.9600) time: 0.6685 data: 0.4935 max mem: 64948 Test: Total time: 0:00:06 (0.6926 s / it) * Acc@1 75.714 Acc@5 92.814 loss 0.943 Accuracy of the model on the 50000 test images: 75.7% Max accuracy: 75.76% Test: [0/9] eta: 0:00:45 loss: 0.8087 (0.8087) acc1: 79.4271 (79.4271) acc5: 94.2708 (94.2708) time: 5.0412 data: 4.8287 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9567 (0.9789) acc1: 75.2604 (74.5920) acc5: 92.7083 (91.9680) time: 0.7147 data: 0.5366 max mem: 64948 Test: Total time: 0:00:06 (0.7229 s / it) * Acc@1 75.074 Acc@5 92.614 loss 0.967 Accuracy of the model EMA on 50000 test images: 75.1% Max EMA accuracy: 75.07% Epoch: [105] [ 0/312] eta: 0:48:52 lr: 0.003627 min_lr: 0.003627 loss: 1.5944 (1.5944) weight_decay: 0.0500 (0.0500) time: 9.3986 data: 8.5957 max mem: 64948 Epoch: [105] [ 10/312] eta: 0:07:36 lr: 0.003626 min_lr: 0.003626 loss: 2.3870 (2.2647) weight_decay: 0.0500 (0.0500) time: 1.5117 data: 0.7818 max mem: 64948 Epoch: [105] [ 20/312] eta: 0:05:27 lr: 0.003626 min_lr: 0.003626 loss: 2.3870 (2.3313) weight_decay: 0.0500 (0.0500) time: 0.7073 data: 0.0004 max mem: 64948 Epoch: [105] [ 30/312] eta: 0:04:37 lr: 0.003626 min_lr: 0.003626 loss: 2.3425 (2.3240) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [105] [ 40/312] eta: 0:04:08 lr: 0.003626 min_lr: 0.003626 loss: 2.3775 (2.3302) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [105] [ 50/312] eta: 0:03:48 lr: 0.003625 min_lr: 0.003625 loss: 2.3478 (2.3011) weight_decay: 0.0500 (0.0500) time: 0.7013 data: 0.0004 max mem: 64948 Epoch: [105] [ 60/312] eta: 0:03:32 lr: 0.003625 min_lr: 0.003625 loss: 2.3478 (2.3175) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [105] [ 70/312] eta: 0:03:19 lr: 0.003625 min_lr: 0.003625 loss: 2.5821 (2.3552) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [105] [ 80/312] eta: 0:03:07 lr: 0.003624 min_lr: 0.003624 loss: 2.5908 (2.3369) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [105] [ 90/312] eta: 0:02:56 lr: 0.003624 min_lr: 0.003624 loss: 2.4939 (2.3604) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [105] [100/312] eta: 0:02:46 lr: 0.003624 min_lr: 0.003624 loss: 2.5014 (2.3653) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [105] [110/312] eta: 0:02:37 lr: 0.003624 min_lr: 0.003624 loss: 2.3761 (2.3547) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [105] [120/312] eta: 0:02:27 lr: 0.003623 min_lr: 0.003623 loss: 2.1992 (2.3367) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [105] [130/312] eta: 0:02:19 lr: 0.003623 min_lr: 0.003623 loss: 2.1992 (2.3319) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [105] [140/312] eta: 0:02:10 lr: 0.003623 min_lr: 0.003623 loss: 2.4003 (2.3248) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [105] [150/312] eta: 0:02:02 lr: 0.003623 min_lr: 0.003623 loss: 2.3756 (2.3252) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [105] [160/312] eta: 0:01:54 lr: 0.003622 min_lr: 0.003622 loss: 2.3136 (2.3251) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [105] [170/312] eta: 0:01:46 lr: 0.003622 min_lr: 0.003622 loss: 2.3131 (2.3162) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [105] [180/312] eta: 0:01:38 lr: 0.003622 min_lr: 0.003622 loss: 2.4118 (2.3290) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [105] [190/312] eta: 0:01:30 lr: 0.003621 min_lr: 0.003621 loss: 2.4319 (2.3257) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [105] [200/312] eta: 0:01:22 lr: 0.003621 min_lr: 0.003621 loss: 2.1889 (2.3177) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [105] [210/312] eta: 0:01:15 lr: 0.003621 min_lr: 0.003621 loss: 2.1938 (2.3161) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [105] [220/312] eta: 0:01:07 lr: 0.003621 min_lr: 0.003621 loss: 2.0906 (2.3079) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [105] [230/312] eta: 0:01:00 lr: 0.003620 min_lr: 0.003620 loss: 2.1045 (2.3044) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [105] [240/312] eta: 0:00:52 lr: 0.003620 min_lr: 0.003620 loss: 2.2890 (2.3004) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [105] [250/312] eta: 0:00:45 lr: 0.003620 min_lr: 0.003620 loss: 2.3567 (2.3063) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [105] [260/312] eta: 0:00:37 lr: 0.003620 min_lr: 0.003620 loss: 2.3567 (2.3054) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [105] [270/312] eta: 0:00:30 lr: 0.003619 min_lr: 0.003619 loss: 2.4810 (2.3110) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [105] [280/312] eta: 0:00:23 lr: 0.003619 min_lr: 0.003619 loss: 2.4909 (2.3126) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0010 max mem: 64948 Epoch: [105] [290/312] eta: 0:00:15 lr: 0.003619 min_lr: 0.003619 loss: 2.4883 (2.3218) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [105] [300/312] eta: 0:00:08 lr: 0.003618 min_lr: 0.003618 loss: 2.4793 (2.3266) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [105] [310/312] eta: 0:00:01 lr: 0.003618 min_lr: 0.003618 loss: 2.2770 (2.3181) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [105] [311/312] eta: 0:00:00 lr: 0.003618 min_lr: 0.003618 loss: 2.3839 (2.3184) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [105] Total time: 0:03:46 (0.7265 s / it) Averaged stats: lr: 0.003618 min_lr: 0.003618 loss: 2.3839 (2.2996) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7325 (0.7325) acc1: 81.7708 (81.7708) acc5: 94.2708 (94.2708) time: 4.6257 data: 4.4064 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0179 (0.9989) acc1: 74.4792 (74.3360) acc5: 92.1875 (92.2240) time: 0.6659 data: 0.4897 max mem: 64948 Test: Total time: 0:00:06 (0.6896 s / it) * Acc@1 75.378 Acc@5 92.666 loss 0.976 Accuracy of the model on the 50000 test images: 75.4% Max accuracy: 75.76% Test: [0/9] eta: 0:00:44 loss: 0.7987 (0.7987) acc1: 79.9479 (79.9479) acc5: 94.2708 (94.2708) time: 4.9863 data: 4.7684 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9472 (0.9690) acc1: 75.5208 (74.8160) acc5: 92.4479 (92.0320) time: 0.7053 data: 0.5299 max mem: 64948 Test: Total time: 0:00:06 (0.7146 s / it) * Acc@1 75.264 Acc@5 92.706 loss 0.957 Accuracy of the model EMA on 50000 test images: 75.3% Max EMA accuracy: 75.26% Epoch: [106] [ 0/312] eta: 0:49:20 lr: 0.003618 min_lr: 0.003618 loss: 1.8698 (1.8698) weight_decay: 0.0500 (0.0500) time: 9.4899 data: 7.5074 max mem: 64948 Epoch: [106] [ 10/312] eta: 0:07:39 lr: 0.003618 min_lr: 0.003618 loss: 2.3810 (2.3446) weight_decay: 0.0500 (0.0500) time: 1.5205 data: 0.6829 max mem: 64948 Epoch: [106] [ 20/312] eta: 0:05:28 lr: 0.003618 min_lr: 0.003618 loss: 2.5120 (2.3924) weight_decay: 0.0500 (0.0500) time: 0.7077 data: 0.0004 max mem: 64948 Epoch: [106] [ 30/312] eta: 0:04:38 lr: 0.003617 min_lr: 0.003617 loss: 2.3907 (2.3360) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0003 max mem: 64948 Epoch: [106] [ 40/312] eta: 0:04:09 lr: 0.003617 min_lr: 0.003617 loss: 2.4396 (2.3804) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [106] [ 50/312] eta: 0:03:48 lr: 0.003617 min_lr: 0.003617 loss: 2.4396 (2.3666) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [106] [ 60/312] eta: 0:03:32 lr: 0.003616 min_lr: 0.003616 loss: 2.3566 (2.3504) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [106] [ 70/312] eta: 0:03:19 lr: 0.003616 min_lr: 0.003616 loss: 2.0992 (2.2997) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [106] [ 80/312] eta: 0:03:07 lr: 0.003616 min_lr: 0.003616 loss: 2.0704 (2.2830) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [106] [ 90/312] eta: 0:02:56 lr: 0.003616 min_lr: 0.003616 loss: 2.3511 (2.2914) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [106] [100/312] eta: 0:02:46 lr: 0.003615 min_lr: 0.003615 loss: 2.3511 (2.2952) weight_decay: 0.0500 (0.0500) time: 0.7010 data: 0.0004 max mem: 64948 Epoch: [106] [110/312] eta: 0:02:37 lr: 0.003615 min_lr: 0.003615 loss: 2.4014 (2.2989) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [106] [120/312] eta: 0:02:27 lr: 0.003615 min_lr: 0.003615 loss: 2.4435 (2.3010) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [106] [130/312] eta: 0:02:19 lr: 0.003615 min_lr: 0.003615 loss: 2.2903 (2.2979) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [106] [140/312] eta: 0:02:10 lr: 0.003614 min_lr: 0.003614 loss: 2.2060 (2.2915) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [106] [150/312] eta: 0:02:02 lr: 0.003614 min_lr: 0.003614 loss: 2.1642 (2.2849) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0003 max mem: 64948 Epoch: [106] [160/312] eta: 0:01:54 lr: 0.003614 min_lr: 0.003614 loss: 2.2648 (2.2882) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [106] [170/312] eta: 0:01:46 lr: 0.003613 min_lr: 0.003613 loss: 2.3898 (2.3008) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [106] [180/312] eta: 0:01:38 lr: 0.003613 min_lr: 0.003613 loss: 2.4490 (2.2980) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [106] [190/312] eta: 0:01:30 lr: 0.003613 min_lr: 0.003613 loss: 2.2380 (2.2999) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [106] [200/312] eta: 0:01:22 lr: 0.003613 min_lr: 0.003613 loss: 2.2380 (2.3019) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [106] [210/312] eta: 0:01:15 lr: 0.003612 min_lr: 0.003612 loss: 2.2416 (2.3024) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [106] [220/312] eta: 0:01:07 lr: 0.003612 min_lr: 0.003612 loss: 2.4046 (2.3056) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [106] [230/312] eta: 0:01:00 lr: 0.003612 min_lr: 0.003612 loss: 2.4046 (2.3107) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [106] [240/312] eta: 0:00:52 lr: 0.003611 min_lr: 0.003611 loss: 2.3246 (2.3089) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [106] [250/312] eta: 0:00:45 lr: 0.003611 min_lr: 0.003611 loss: 2.4549 (2.3168) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [106] [260/312] eta: 0:00:37 lr: 0.003611 min_lr: 0.003611 loss: 2.5064 (2.3218) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [106] [270/312] eta: 0:00:30 lr: 0.003611 min_lr: 0.003611 loss: 2.4084 (2.3203) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [106] [280/312] eta: 0:00:23 lr: 0.003610 min_lr: 0.003610 loss: 2.4379 (2.3254) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0006 max mem: 64948 Epoch: [106] [290/312] eta: 0:00:15 lr: 0.003610 min_lr: 0.003610 loss: 2.3218 (2.3210) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0005 max mem: 64948 Epoch: [106] [300/312] eta: 0:00:08 lr: 0.003610 min_lr: 0.003610 loss: 2.0901 (2.3140) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [106] [310/312] eta: 0:00:01 lr: 0.003610 min_lr: 0.003610 loss: 2.2151 (2.3189) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [106] [311/312] eta: 0:00:00 lr: 0.003610 min_lr: 0.003610 loss: 2.2357 (2.3188) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [106] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.003610 min_lr: 0.003610 loss: 2.2357 (2.3049) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7333 (0.7333) acc1: 82.2917 (82.2917) acc5: 94.5312 (94.5312) time: 4.5108 data: 4.2999 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.1308 (1.0175) acc1: 74.7396 (74.5600) acc5: 91.4062 (91.7760) time: 0.6525 data: 0.4779 max mem: 64948 Test: Total time: 0:00:06 (0.6763 s / it) * Acc@1 75.350 Acc@5 92.606 loss 0.986 Accuracy of the model on the 50000 test images: 75.4% Max accuracy: 75.76% Test: [0/9] eta: 0:00:42 loss: 0.7896 (0.7896) acc1: 79.9479 (79.9479) acc5: 94.2708 (94.2708) time: 4.7260 data: 4.5184 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9386 (0.9597) acc1: 75.2604 (74.9440) acc5: 92.7083 (92.1280) time: 0.6866 data: 0.5098 max mem: 64948 Test: Total time: 0:00:06 (0.6951 s / it) * Acc@1 75.524 Acc@5 92.812 loss 0.946 Accuracy of the model EMA on 50000 test images: 75.5% Max EMA accuracy: 75.52% Epoch: [107] [ 0/312] eta: 0:48:31 lr: 0.003609 min_lr: 0.003609 loss: 2.6121 (2.6121) weight_decay: 0.0500 (0.0500) time: 9.3311 data: 8.4184 max mem: 64948 Epoch: [107] [ 10/312] eta: 0:07:38 lr: 0.003609 min_lr: 0.003609 loss: 2.3824 (2.2936) weight_decay: 0.0500 (0.0500) time: 1.5176 data: 0.7657 max mem: 64948 Epoch: [107] [ 20/312] eta: 0:05:28 lr: 0.003609 min_lr: 0.003609 loss: 2.2846 (2.2826) weight_decay: 0.0500 (0.0500) time: 0.7151 data: 0.0004 max mem: 64948 Epoch: [107] [ 30/312] eta: 0:04:38 lr: 0.003609 min_lr: 0.003609 loss: 2.1017 (2.2015) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [107] [ 40/312] eta: 0:04:09 lr: 0.003608 min_lr: 0.003608 loss: 2.3124 (2.2285) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [107] [ 50/312] eta: 0:03:48 lr: 0.003608 min_lr: 0.003608 loss: 2.3434 (2.2201) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [107] [ 60/312] eta: 0:03:32 lr: 0.003608 min_lr: 0.003608 loss: 2.2514 (2.2194) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [107] [ 70/312] eta: 0:03:19 lr: 0.003608 min_lr: 0.003608 loss: 2.2514 (2.2213) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [107] [ 80/312] eta: 0:03:07 lr: 0.003607 min_lr: 0.003607 loss: 2.2299 (2.2333) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [107] [ 90/312] eta: 0:02:56 lr: 0.003607 min_lr: 0.003607 loss: 2.4616 (2.2658) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [107] [100/312] eta: 0:02:46 lr: 0.003607 min_lr: 0.003607 loss: 2.4624 (2.2905) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [107] [110/312] eta: 0:02:36 lr: 0.003606 min_lr: 0.003606 loss: 2.4351 (2.2819) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [107] [120/312] eta: 0:02:27 lr: 0.003606 min_lr: 0.003606 loss: 2.4158 (2.3007) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [107] [130/312] eta: 0:02:19 lr: 0.003606 min_lr: 0.003606 loss: 2.4638 (2.3048) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [107] [140/312] eta: 0:02:10 lr: 0.003606 min_lr: 0.003606 loss: 2.4025 (2.3090) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [107] [150/312] eta: 0:02:02 lr: 0.003605 min_lr: 0.003605 loss: 2.4095 (2.3143) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [107] [160/312] eta: 0:01:54 lr: 0.003605 min_lr: 0.003605 loss: 2.4112 (2.3152) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [107] [170/312] eta: 0:01:46 lr: 0.003605 min_lr: 0.003605 loss: 2.2249 (2.3080) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [107] [180/312] eta: 0:01:38 lr: 0.003604 min_lr: 0.003604 loss: 2.2249 (2.3137) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [107] [190/312] eta: 0:01:30 lr: 0.003604 min_lr: 0.003604 loss: 2.5264 (2.3196) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [107] [200/312] eta: 0:01:22 lr: 0.003604 min_lr: 0.003604 loss: 2.4097 (2.3189) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [107] [210/312] eta: 0:01:15 lr: 0.003604 min_lr: 0.003604 loss: 2.3986 (2.3228) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [107] [220/312] eta: 0:01:07 lr: 0.003603 min_lr: 0.003603 loss: 2.3892 (2.3230) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [107] [230/312] eta: 0:01:00 lr: 0.003603 min_lr: 0.003603 loss: 2.5134 (2.3318) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [107] [240/312] eta: 0:00:52 lr: 0.003603 min_lr: 0.003603 loss: 2.5134 (2.3314) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [107] [250/312] eta: 0:00:45 lr: 0.003603 min_lr: 0.003603 loss: 2.4150 (2.3325) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [107] [260/312] eta: 0:00:37 lr: 0.003602 min_lr: 0.003602 loss: 2.2514 (2.3197) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [107] [270/312] eta: 0:00:30 lr: 0.003602 min_lr: 0.003602 loss: 2.1456 (2.3161) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [107] [280/312] eta: 0:00:23 lr: 0.003602 min_lr: 0.003602 loss: 2.3013 (2.3182) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [107] [290/312] eta: 0:00:15 lr: 0.003601 min_lr: 0.003601 loss: 2.3354 (2.3158) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [107] [300/312] eta: 0:00:08 lr: 0.003601 min_lr: 0.003601 loss: 2.2091 (2.3114) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [107] [310/312] eta: 0:00:01 lr: 0.003601 min_lr: 0.003601 loss: 2.1794 (2.3076) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [107] [311/312] eta: 0:00:00 lr: 0.003601 min_lr: 0.003601 loss: 2.1811 (2.3074) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [107] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.003601 min_lr: 0.003601 loss: 2.1811 (2.2950) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7307 (0.7307) acc1: 80.7292 (80.7292) acc5: 95.5729 (95.5729) time: 4.5801 data: 4.3727 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0067 (0.9790) acc1: 75.5208 (74.4000) acc5: 94.2708 (93.5360) time: 0.6605 data: 0.4859 max mem: 64948 Test: Total time: 0:00:06 (0.7135 s / it) * Acc@1 75.462 Acc@5 92.880 loss 0.965 Accuracy of the model on the 50000 test images: 75.5% Max accuracy: 75.76% Test: [0/9] eta: 0:00:44 loss: 0.7810 (0.7810) acc1: 79.9479 (79.9479) acc5: 94.2708 (94.2708) time: 4.9792 data: 4.7740 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9314 (0.9509) acc1: 75.2604 (75.0720) acc5: 93.2292 (92.3520) time: 0.7047 data: 0.5306 max mem: 64948 Test: Total time: 0:00:06 (0.7137 s / it) * Acc@1 75.754 Acc@5 92.922 loss 0.937 Accuracy of the model EMA on 50000 test images: 75.8% Max EMA accuracy: 75.75% Epoch: [108] [ 0/312] eta: 0:47:12 lr: 0.003601 min_lr: 0.003601 loss: 2.6306 (2.6306) weight_decay: 0.0500 (0.0500) time: 9.0791 data: 8.2653 max mem: 64948 Epoch: [108] [ 10/312] eta: 0:07:26 lr: 0.003601 min_lr: 0.003601 loss: 2.4499 (2.2751) weight_decay: 0.0500 (0.0500) time: 1.4769 data: 0.7519 max mem: 64948 Epoch: [108] [ 20/312] eta: 0:05:22 lr: 0.003600 min_lr: 0.003600 loss: 2.3919 (2.2844) weight_decay: 0.0500 (0.0500) time: 0.7057 data: 0.0005 max mem: 64948 Epoch: [108] [ 30/312] eta: 0:04:34 lr: 0.003600 min_lr: 0.003600 loss: 2.1399 (2.1866) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [108] [ 40/312] eta: 0:04:06 lr: 0.003600 min_lr: 0.003600 loss: 2.1203 (2.1947) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [108] [ 50/312] eta: 0:03:46 lr: 0.003599 min_lr: 0.003599 loss: 2.3694 (2.2567) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [108] [ 60/312] eta: 0:03:31 lr: 0.003599 min_lr: 0.003599 loss: 2.2593 (2.2219) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [108] [ 70/312] eta: 0:03:17 lr: 0.003599 min_lr: 0.003599 loss: 2.0857 (2.2111) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [108] [ 80/312] eta: 0:03:06 lr: 0.003599 min_lr: 0.003599 loss: 2.1597 (2.2220) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [108] [ 90/312] eta: 0:02:55 lr: 0.003598 min_lr: 0.003598 loss: 2.2360 (2.2196) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [108] [100/312] eta: 0:02:45 lr: 0.003598 min_lr: 0.003598 loss: 2.2384 (2.2185) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [108] [110/312] eta: 0:02:36 lr: 0.003598 min_lr: 0.003598 loss: 2.4446 (2.2539) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [108] [120/312] eta: 0:02:27 lr: 0.003597 min_lr: 0.003597 loss: 2.4359 (2.2518) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [108] [130/312] eta: 0:02:18 lr: 0.003597 min_lr: 0.003597 loss: 2.3668 (2.2542) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [108] [140/312] eta: 0:02:10 lr: 0.003597 min_lr: 0.003597 loss: 2.4025 (2.2607) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [108] [150/312] eta: 0:02:01 lr: 0.003597 min_lr: 0.003597 loss: 2.4046 (2.2669) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [108] [160/312] eta: 0:01:53 lr: 0.003596 min_lr: 0.003596 loss: 2.4358 (2.2758) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [108] [170/312] eta: 0:01:45 lr: 0.003596 min_lr: 0.003596 loss: 2.4517 (2.2889) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [108] [180/312] eta: 0:01:38 lr: 0.003596 min_lr: 0.003596 loss: 2.4517 (2.2963) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [108] [190/312] eta: 0:01:30 lr: 0.003595 min_lr: 0.003595 loss: 2.3206 (2.2887) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [108] [200/312] eta: 0:01:22 lr: 0.003595 min_lr: 0.003595 loss: 2.3447 (2.2973) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [108] [210/312] eta: 0:01:15 lr: 0.003595 min_lr: 0.003595 loss: 2.3521 (2.2906) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [108] [220/312] eta: 0:01:07 lr: 0.003595 min_lr: 0.003595 loss: 2.1180 (2.2848) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [108] [230/312] eta: 0:01:00 lr: 0.003594 min_lr: 0.003594 loss: 2.1680 (2.2821) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [108] [240/312] eta: 0:00:52 lr: 0.003594 min_lr: 0.003594 loss: 2.3911 (2.2878) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [108] [250/312] eta: 0:00:45 lr: 0.003594 min_lr: 0.003594 loss: 2.3911 (2.2811) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [108] [260/312] eta: 0:00:37 lr: 0.003593 min_lr: 0.003593 loss: 1.9553 (2.2697) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [108] [270/312] eta: 0:00:30 lr: 0.003593 min_lr: 0.003593 loss: 2.0345 (2.2752) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [108] [280/312] eta: 0:00:23 lr: 0.003593 min_lr: 0.003593 loss: 2.4231 (2.2780) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0009 max mem: 64948 Epoch: [108] [290/312] eta: 0:00:15 lr: 0.003593 min_lr: 0.003593 loss: 2.2980 (2.2756) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [108] [300/312] eta: 0:00:08 lr: 0.003592 min_lr: 0.003592 loss: 2.2980 (2.2744) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [108] [310/312] eta: 0:00:01 lr: 0.003592 min_lr: 0.003592 loss: 2.4721 (2.2782) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [108] [311/312] eta: 0:00:00 lr: 0.003592 min_lr: 0.003592 loss: 2.2506 (2.2780) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [108] Total time: 0:03:46 (0.7251 s / it) Averaged stats: lr: 0.003592 min_lr: 0.003592 loss: 2.2506 (2.3002) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7213 (0.7213) acc1: 82.8125 (82.8125) acc5: 95.0521 (95.0521) time: 4.5658 data: 4.3508 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0631 (0.9755) acc1: 74.7396 (74.9120) acc5: 93.4896 (92.4160) time: 0.6586 data: 0.4835 max mem: 64948 Test: Total time: 0:00:06 (0.6834 s / it) * Acc@1 76.004 Acc@5 92.958 loss 0.946 Accuracy of the model on the 50000 test images: 76.0% Max accuracy: 76.00% Test: [0/9] eta: 0:00:39 loss: 0.7727 (0.7727) acc1: 80.2083 (80.2083) acc5: 94.2708 (94.2708) time: 4.3837 data: 4.1740 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9239 (0.9424) acc1: 75.2604 (75.1360) acc5: 93.2292 (92.4800) time: 0.6384 data: 0.4639 max mem: 64948 Test: Total time: 0:00:05 (0.6473 s / it) * Acc@1 75.934 Acc@5 93.000 loss 0.928 Accuracy of the model EMA on 50000 test images: 75.9% Max EMA accuracy: 75.93% Epoch: [109] [ 0/312] eta: 0:47:52 lr: 0.003592 min_lr: 0.003592 loss: 1.6667 (1.6667) weight_decay: 0.0500 (0.0500) time: 9.2066 data: 8.4363 max mem: 64948 Epoch: [109] [ 10/312] eta: 0:07:33 lr: 0.003592 min_lr: 0.003592 loss: 2.1813 (2.2248) weight_decay: 0.0500 (0.0500) time: 1.5013 data: 0.7673 max mem: 64948 Epoch: [109] [ 20/312] eta: 0:05:26 lr: 0.003591 min_lr: 0.003591 loss: 2.2804 (2.2434) weight_decay: 0.0500 (0.0500) time: 0.7134 data: 0.0004 max mem: 64948 Epoch: [109] [ 30/312] eta: 0:04:36 lr: 0.003591 min_lr: 0.003591 loss: 2.2956 (2.2910) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [109] [ 40/312] eta: 0:04:08 lr: 0.003591 min_lr: 0.003591 loss: 2.4050 (2.2884) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [109] [ 50/312] eta: 0:03:47 lr: 0.003591 min_lr: 0.003591 loss: 2.3950 (2.2829) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [109] [ 60/312] eta: 0:03:31 lr: 0.003590 min_lr: 0.003590 loss: 2.2788 (2.2918) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [109] [ 70/312] eta: 0:03:18 lr: 0.003590 min_lr: 0.003590 loss: 2.2887 (2.2981) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [109] [ 80/312] eta: 0:03:06 lr: 0.003590 min_lr: 0.003590 loss: 2.1119 (2.2792) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0003 max mem: 64948 Epoch: [109] [ 90/312] eta: 0:02:56 lr: 0.003589 min_lr: 0.003589 loss: 2.3972 (2.3041) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [109] [100/312] eta: 0:02:46 lr: 0.003589 min_lr: 0.003589 loss: 2.4226 (2.2882) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [109] [110/312] eta: 0:02:36 lr: 0.003589 min_lr: 0.003589 loss: 2.2340 (2.2845) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [109] [120/312] eta: 0:02:27 lr: 0.003589 min_lr: 0.003589 loss: 2.3213 (2.2849) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [109] [130/312] eta: 0:02:18 lr: 0.003588 min_lr: 0.003588 loss: 2.4794 (2.3037) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [109] [140/312] eta: 0:02:10 lr: 0.003588 min_lr: 0.003588 loss: 2.3355 (2.2890) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [109] [150/312] eta: 0:02:02 lr: 0.003588 min_lr: 0.003588 loss: 2.2016 (2.2976) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [109] [160/312] eta: 0:01:54 lr: 0.003587 min_lr: 0.003587 loss: 2.4711 (2.2989) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [109] [170/312] eta: 0:01:46 lr: 0.003587 min_lr: 0.003587 loss: 2.1883 (2.2915) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [109] [180/312] eta: 0:01:38 lr: 0.003587 min_lr: 0.003587 loss: 2.3344 (2.3006) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [109] [190/312] eta: 0:01:30 lr: 0.003587 min_lr: 0.003587 loss: 2.2668 (2.2838) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [109] [200/312] eta: 0:01:22 lr: 0.003586 min_lr: 0.003586 loss: 1.9560 (2.2846) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [109] [210/312] eta: 0:01:15 lr: 0.003586 min_lr: 0.003586 loss: 2.2781 (2.2840) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [109] [220/312] eta: 0:01:07 lr: 0.003586 min_lr: 0.003586 loss: 2.0051 (2.2686) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [109] [230/312] eta: 0:01:00 lr: 0.003585 min_lr: 0.003585 loss: 2.1784 (2.2776) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [109] [240/312] eta: 0:00:52 lr: 0.003585 min_lr: 0.003585 loss: 2.4657 (2.2770) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [109] [250/312] eta: 0:00:45 lr: 0.003585 min_lr: 0.003585 loss: 2.3121 (2.2732) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [109] [260/312] eta: 0:00:37 lr: 0.003585 min_lr: 0.003585 loss: 2.3069 (2.2743) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [109] [270/312] eta: 0:00:30 lr: 0.003584 min_lr: 0.003584 loss: 2.3953 (2.2780) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [109] [280/312] eta: 0:00:23 lr: 0.003584 min_lr: 0.003584 loss: 2.3953 (2.2817) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [109] [290/312] eta: 0:00:15 lr: 0.003584 min_lr: 0.003584 loss: 2.2849 (2.2779) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [109] [300/312] eta: 0:00:08 lr: 0.003583 min_lr: 0.003583 loss: 2.2189 (2.2811) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [109] [310/312] eta: 0:00:01 lr: 0.003583 min_lr: 0.003583 loss: 2.4429 (2.2830) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [109] [311/312] eta: 0:00:00 lr: 0.003583 min_lr: 0.003583 loss: 2.4429 (2.2831) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [109] Total time: 0:03:46 (0.7262 s / it) Averaged stats: lr: 0.003583 min_lr: 0.003583 loss: 2.4429 (2.2987) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.8177 (0.8177) acc1: 79.9479 (79.9479) acc5: 93.7500 (93.7500) time: 4.6852 data: 4.4641 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0404 (0.9911) acc1: 74.4792 (74.8800) acc5: 92.9688 (92.5760) time: 0.6723 data: 0.4961 max mem: 64948 Test: Total time: 0:00:06 (0.6963 s / it) * Acc@1 75.800 Acc@5 92.670 loss 0.968 Accuracy of the model on the 50000 test images: 75.8% Max accuracy: 76.00% Test: [0/9] eta: 0:00:44 loss: 0.7646 (0.7646) acc1: 80.2083 (80.2083) acc5: 94.2708 (94.2708) time: 4.9096 data: 4.6929 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9174 (0.9343) acc1: 75.5208 (75.3920) acc5: 93.2292 (92.6400) time: 0.6974 data: 0.5215 max mem: 64948 Test: Total time: 0:00:06 (0.7113 s / it) * Acc@1 76.138 Acc@5 93.090 loss 0.920 Accuracy of the model EMA on 50000 test images: 76.1% Max EMA accuracy: 76.14% Epoch: [110] [ 0/312] eta: 0:51:56 lr: 0.003583 min_lr: 0.003583 loss: 2.4807 (2.4807) weight_decay: 0.0500 (0.0500) time: 9.9878 data: 9.2088 max mem: 64948 Epoch: [110] [ 10/312] eta: 0:07:51 lr: 0.003583 min_lr: 0.003583 loss: 2.4814 (2.4182) weight_decay: 0.0500 (0.0500) time: 1.5628 data: 0.8375 max mem: 64948 Epoch: [110] [ 20/312] eta: 0:05:35 lr: 0.003583 min_lr: 0.003583 loss: 2.3417 (2.2320) weight_decay: 0.0500 (0.0500) time: 0.7081 data: 0.0003 max mem: 64948 Epoch: [110] [ 30/312] eta: 0:04:43 lr: 0.003582 min_lr: 0.003582 loss: 2.2147 (2.2236) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0003 max mem: 64948 Epoch: [110] [ 40/312] eta: 0:04:12 lr: 0.003582 min_lr: 0.003582 loss: 2.4369 (2.2692) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0003 max mem: 64948 Epoch: [110] [ 50/312] eta: 0:03:51 lr: 0.003582 min_lr: 0.003582 loss: 2.4456 (2.3056) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [110] [ 60/312] eta: 0:03:35 lr: 0.003581 min_lr: 0.003581 loss: 2.5001 (2.3196) weight_decay: 0.0500 (0.0500) time: 0.7025 data: 0.0004 max mem: 64948 Epoch: [110] [ 70/312] eta: 0:03:21 lr: 0.003581 min_lr: 0.003581 loss: 2.3212 (2.3149) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [110] [ 80/312] eta: 0:03:09 lr: 0.003581 min_lr: 0.003581 loss: 2.2260 (2.3260) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [110] [ 90/312] eta: 0:02:58 lr: 0.003581 min_lr: 0.003581 loss: 2.4296 (2.3403) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [110] [100/312] eta: 0:02:47 lr: 0.003580 min_lr: 0.003580 loss: 2.3914 (2.3195) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [110] [110/312] eta: 0:02:38 lr: 0.003580 min_lr: 0.003580 loss: 2.1623 (2.3049) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [110] [120/312] eta: 0:02:28 lr: 0.003580 min_lr: 0.003580 loss: 2.2305 (2.3045) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [110] [130/312] eta: 0:02:20 lr: 0.003579 min_lr: 0.003579 loss: 2.4337 (2.3096) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [110] [140/312] eta: 0:02:11 lr: 0.003579 min_lr: 0.003579 loss: 2.3394 (2.3060) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [110] [150/312] eta: 0:02:03 lr: 0.003579 min_lr: 0.003579 loss: 2.2291 (2.2952) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [110] [160/312] eta: 0:01:54 lr: 0.003579 min_lr: 0.003579 loss: 2.2923 (2.2997) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [110] [170/312] eta: 0:01:46 lr: 0.003578 min_lr: 0.003578 loss: 2.4434 (2.3113) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [110] [180/312] eta: 0:01:38 lr: 0.003578 min_lr: 0.003578 loss: 2.4434 (2.3110) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [110] [190/312] eta: 0:01:30 lr: 0.003578 min_lr: 0.003578 loss: 2.2877 (2.3023) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [110] [200/312] eta: 0:01:23 lr: 0.003577 min_lr: 0.003577 loss: 2.2726 (2.3045) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [110] [210/312] eta: 0:01:15 lr: 0.003577 min_lr: 0.003577 loss: 2.2726 (2.3003) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [110] [220/312] eta: 0:01:07 lr: 0.003577 min_lr: 0.003577 loss: 2.3190 (2.3008) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [110] [230/312] eta: 0:01:00 lr: 0.003576 min_lr: 0.003576 loss: 2.4085 (2.3037) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [110] [240/312] eta: 0:00:52 lr: 0.003576 min_lr: 0.003576 loss: 2.4661 (2.3021) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [110] [250/312] eta: 0:00:45 lr: 0.003576 min_lr: 0.003576 loss: 2.2937 (2.2987) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [110] [260/312] eta: 0:00:38 lr: 0.003576 min_lr: 0.003576 loss: 2.2838 (2.2972) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [110] [270/312] eta: 0:00:30 lr: 0.003575 min_lr: 0.003575 loss: 2.3510 (2.2996) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [110] [280/312] eta: 0:00:23 lr: 0.003575 min_lr: 0.003575 loss: 2.2989 (2.2961) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0010 max mem: 64948 Epoch: [110] [290/312] eta: 0:00:16 lr: 0.003575 min_lr: 0.003575 loss: 2.2077 (2.2931) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [110] [300/312] eta: 0:00:08 lr: 0.003574 min_lr: 0.003574 loss: 2.1768 (2.2864) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [110] [310/312] eta: 0:00:01 lr: 0.003574 min_lr: 0.003574 loss: 2.2778 (2.2837) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [110] [311/312] eta: 0:00:00 lr: 0.003574 min_lr: 0.003574 loss: 2.3030 (2.2844) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [110] Total time: 0:03:47 (0.7284 s / it) Averaged stats: lr: 0.003574 min_lr: 0.003574 loss: 2.3030 (2.2932) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.8241 (0.8241) acc1: 78.9062 (78.9062) acc5: 93.7500 (93.7500) time: 4.3679 data: 4.1517 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0841 (0.9936) acc1: 75.2604 (75.1360) acc5: 91.9271 (92.5760) time: 0.6366 data: 0.4614 max mem: 64948 Test: Total time: 0:00:05 (0.6584 s / it) * Acc@1 75.850 Acc@5 92.726 loss 0.987 Accuracy of the model on the 50000 test images: 75.9% Max accuracy: 76.00% Test: [0/9] eta: 0:00:45 loss: 0.7570 (0.7570) acc1: 80.4688 (80.4688) acc5: 94.5312 (94.5312) time: 5.0937 data: 4.8759 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9106 (0.9266) acc1: 75.5208 (75.5840) acc5: 93.2292 (92.7360) time: 0.7179 data: 0.5419 max mem: 64948 Test: Total time: 0:00:06 (0.7274 s / it) * Acc@1 76.324 Acc@5 93.184 loss 0.912 Accuracy of the model EMA on 50000 test images: 76.3% Max EMA accuracy: 76.32% Epoch: [111] [ 0/312] eta: 0:54:00 lr: 0.003574 min_lr: 0.003574 loss: 2.5307 (2.5307) weight_decay: 0.0500 (0.0500) time: 10.3875 data: 9.6026 max mem: 64948 Epoch: [111] [ 10/312] eta: 0:07:59 lr: 0.003574 min_lr: 0.003574 loss: 2.0360 (2.1416) weight_decay: 0.0500 (0.0500) time: 1.5865 data: 0.8733 max mem: 64948 Epoch: [111] [ 20/312] eta: 0:05:39 lr: 0.003574 min_lr: 0.003574 loss: 2.0423 (2.1776) weight_decay: 0.0500 (0.0500) time: 0.7016 data: 0.0003 max mem: 64948 Epoch: [111] [ 30/312] eta: 0:04:45 lr: 0.003573 min_lr: 0.003573 loss: 2.2335 (2.1892) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [111] [ 40/312] eta: 0:04:14 lr: 0.003573 min_lr: 0.003573 loss: 2.3524 (2.2439) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [111] [ 50/312] eta: 0:03:52 lr: 0.003573 min_lr: 0.003573 loss: 2.3524 (2.2312) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [111] [ 60/312] eta: 0:03:35 lr: 0.003572 min_lr: 0.003572 loss: 2.4415 (2.2893) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [111] [ 70/312] eta: 0:03:21 lr: 0.003572 min_lr: 0.003572 loss: 2.5512 (2.3181) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [111] [ 80/312] eta: 0:03:09 lr: 0.003572 min_lr: 0.003572 loss: 2.5251 (2.3453) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [111] [ 90/312] eta: 0:02:58 lr: 0.003572 min_lr: 0.003572 loss: 2.4660 (2.3374) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [111] [100/312] eta: 0:02:48 lr: 0.003571 min_lr: 0.003571 loss: 2.3638 (2.3397) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [111] [110/312] eta: 0:02:38 lr: 0.003571 min_lr: 0.003571 loss: 2.2624 (2.3226) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [111] [120/312] eta: 0:02:29 lr: 0.003571 min_lr: 0.003571 loss: 2.2253 (2.3263) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [111] [130/312] eta: 0:02:20 lr: 0.003570 min_lr: 0.003570 loss: 2.4274 (2.3342) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [111] [140/312] eta: 0:02:11 lr: 0.003570 min_lr: 0.003570 loss: 2.5092 (2.3395) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [111] [150/312] eta: 0:02:03 lr: 0.003570 min_lr: 0.003570 loss: 2.4552 (2.3464) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [111] [160/312] eta: 0:01:55 lr: 0.003569 min_lr: 0.003569 loss: 2.2874 (2.3328) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [111] [170/312] eta: 0:01:46 lr: 0.003569 min_lr: 0.003569 loss: 2.0412 (2.3257) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [111] [180/312] eta: 0:01:38 lr: 0.003569 min_lr: 0.003569 loss: 2.3552 (2.3282) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [111] [190/312] eta: 0:01:31 lr: 0.003569 min_lr: 0.003569 loss: 2.3372 (2.3257) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [111] [200/312] eta: 0:01:23 lr: 0.003568 min_lr: 0.003568 loss: 2.3313 (2.3276) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [111] [210/312] eta: 0:01:15 lr: 0.003568 min_lr: 0.003568 loss: 2.4006 (2.3300) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [111] [220/312] eta: 0:01:08 lr: 0.003568 min_lr: 0.003568 loss: 2.4006 (2.3311) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [111] [230/312] eta: 0:01:00 lr: 0.003567 min_lr: 0.003567 loss: 2.3920 (2.3332) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [111] [240/312] eta: 0:00:52 lr: 0.003567 min_lr: 0.003567 loss: 2.3920 (2.3340) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [111] [250/312] eta: 0:00:45 lr: 0.003567 min_lr: 0.003567 loss: 2.3351 (2.3309) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [111] [260/312] eta: 0:00:38 lr: 0.003567 min_lr: 0.003567 loss: 2.1156 (2.3244) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [111] [270/312] eta: 0:00:30 lr: 0.003566 min_lr: 0.003566 loss: 2.2077 (2.3259) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [111] [280/312] eta: 0:00:23 lr: 0.003566 min_lr: 0.003566 loss: 2.1641 (2.3217) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0011 max mem: 64948 Epoch: [111] [290/312] eta: 0:00:16 lr: 0.003566 min_lr: 0.003566 loss: 2.2778 (2.3236) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0010 max mem: 64948 Epoch: [111] [300/312] eta: 0:00:08 lr: 0.003565 min_lr: 0.003565 loss: 2.3683 (2.3267) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [111] [310/312] eta: 0:00:01 lr: 0.003565 min_lr: 0.003565 loss: 2.3018 (2.3226) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [111] [311/312] eta: 0:00:00 lr: 0.003565 min_lr: 0.003565 loss: 2.3018 (2.3237) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [111] Total time: 0:03:47 (0.7293 s / it) Averaged stats: lr: 0.003565 min_lr: 0.003565 loss: 2.3018 (2.2945) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7846 (0.7846) acc1: 81.7708 (81.7708) acc5: 93.7500 (93.7500) time: 4.5830 data: 4.3722 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0246 (0.9826) acc1: 75.5208 (74.7200) acc5: 93.7500 (92.6080) time: 0.6610 data: 0.4859 max mem: 64948 Test: Total time: 0:00:06 (0.6823 s / it) * Acc@1 75.870 Acc@5 92.898 loss 0.962 Accuracy of the model on the 50000 test images: 75.9% Max accuracy: 76.00% Test: [0/9] eta: 0:00:44 loss: 0.7502 (0.7502) acc1: 80.7292 (80.7292) acc5: 94.7917 (94.7917) time: 4.9967 data: 4.7785 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9035 (0.9188) acc1: 75.5208 (75.6800) acc5: 93.7500 (92.9600) time: 0.7066 data: 0.5310 max mem: 64948 Test: Total time: 0:00:06 (0.7163 s / it) * Acc@1 76.482 Acc@5 93.254 loss 0.904 Accuracy of the model EMA on 50000 test images: 76.5% Max EMA accuracy: 76.48% Epoch: [112] [ 0/312] eta: 0:49:57 lr: 0.003565 min_lr: 0.003565 loss: 2.5099 (2.5099) weight_decay: 0.0500 (0.0500) time: 9.6064 data: 8.8248 max mem: 64948 Epoch: [112] [ 10/312] eta: 0:07:41 lr: 0.003565 min_lr: 0.003565 loss: 2.5099 (2.5385) weight_decay: 0.0500 (0.0500) time: 1.5267 data: 0.8026 max mem: 64948 Epoch: [112] [ 20/312] eta: 0:05:30 lr: 0.003564 min_lr: 0.003564 loss: 2.4870 (2.4790) weight_decay: 0.0500 (0.0500) time: 0.7093 data: 0.0004 max mem: 64948 Epoch: [112] [ 30/312] eta: 0:04:39 lr: 0.003564 min_lr: 0.003564 loss: 2.4217 (2.4379) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [112] [ 40/312] eta: 0:04:10 lr: 0.003564 min_lr: 0.003564 loss: 2.2522 (2.3455) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [112] [ 50/312] eta: 0:03:49 lr: 0.003564 min_lr: 0.003564 loss: 2.1012 (2.3202) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [112] [ 60/312] eta: 0:03:33 lr: 0.003563 min_lr: 0.003563 loss: 2.3251 (2.3192) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [112] [ 70/312] eta: 0:03:19 lr: 0.003563 min_lr: 0.003563 loss: 2.2351 (2.2883) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [112] [ 80/312] eta: 0:03:07 lr: 0.003563 min_lr: 0.003563 loss: 2.1583 (2.2759) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [112] [ 90/312] eta: 0:02:56 lr: 0.003562 min_lr: 0.003562 loss: 2.1944 (2.2696) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [112] [100/312] eta: 0:02:46 lr: 0.003562 min_lr: 0.003562 loss: 2.3630 (2.2669) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [112] [110/312] eta: 0:02:37 lr: 0.003562 min_lr: 0.003562 loss: 2.3252 (2.2688) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [112] [120/312] eta: 0:02:28 lr: 0.003562 min_lr: 0.003562 loss: 2.1939 (2.2642) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [112] [130/312] eta: 0:02:19 lr: 0.003561 min_lr: 0.003561 loss: 2.4229 (2.2725) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [112] [140/312] eta: 0:02:10 lr: 0.003561 min_lr: 0.003561 loss: 2.1892 (2.2585) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [112] [150/312] eta: 0:02:02 lr: 0.003561 min_lr: 0.003561 loss: 2.2997 (2.2710) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [112] [160/312] eta: 0:01:54 lr: 0.003560 min_lr: 0.003560 loss: 2.3416 (2.2644) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [112] [170/312] eta: 0:01:46 lr: 0.003560 min_lr: 0.003560 loss: 2.3553 (2.2722) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [112] [180/312] eta: 0:01:38 lr: 0.003560 min_lr: 0.003560 loss: 2.4664 (2.2880) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [112] [190/312] eta: 0:01:30 lr: 0.003560 min_lr: 0.003560 loss: 2.4910 (2.2979) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [112] [200/312] eta: 0:01:22 lr: 0.003559 min_lr: 0.003559 loss: 2.4147 (2.2999) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [112] [210/312] eta: 0:01:15 lr: 0.003559 min_lr: 0.003559 loss: 2.4147 (2.3016) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [112] [220/312] eta: 0:01:07 lr: 0.003559 min_lr: 0.003559 loss: 2.4308 (2.3038) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [112] [230/312] eta: 0:01:00 lr: 0.003558 min_lr: 0.003558 loss: 2.4812 (2.3080) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [112] [240/312] eta: 0:00:52 lr: 0.003558 min_lr: 0.003558 loss: 2.4812 (2.3129) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [112] [250/312] eta: 0:00:45 lr: 0.003558 min_lr: 0.003558 loss: 2.3813 (2.3140) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [112] [260/312] eta: 0:00:37 lr: 0.003557 min_lr: 0.003557 loss: 2.4416 (2.3200) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [112] [270/312] eta: 0:00:30 lr: 0.003557 min_lr: 0.003557 loss: 2.4813 (2.3216) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [112] [280/312] eta: 0:00:23 lr: 0.003557 min_lr: 0.003557 loss: 2.3633 (2.3213) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0006 max mem: 64948 Epoch: [112] [290/312] eta: 0:00:15 lr: 0.003557 min_lr: 0.003557 loss: 2.3521 (2.3228) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0005 max mem: 64948 Epoch: [112] [300/312] eta: 0:00:08 lr: 0.003556 min_lr: 0.003556 loss: 2.4724 (2.3261) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [112] [310/312] eta: 0:00:01 lr: 0.003556 min_lr: 0.003556 loss: 2.4894 (2.3317) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [112] [311/312] eta: 0:00:00 lr: 0.003556 min_lr: 0.003556 loss: 2.4792 (2.3297) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [112] Total time: 0:03:46 (0.7268 s / it) Averaged stats: lr: 0.003556 min_lr: 0.003556 loss: 2.4792 (2.2952) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.8154 (0.8154) acc1: 79.9479 (79.9479) acc5: 93.4896 (93.4896) time: 4.5346 data: 4.3153 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0659 (0.9940) acc1: 75.0000 (73.6960) acc5: 93.4896 (92.2240) time: 0.6551 data: 0.4796 max mem: 64948 Test: Total time: 0:00:06 (0.6768 s / it) * Acc@1 75.430 Acc@5 92.588 loss 0.968 Accuracy of the model on the 50000 test images: 75.4% Max accuracy: 76.00% Test: [0/9] eta: 0:00:44 loss: 0.7427 (0.7427) acc1: 80.7292 (80.7292) acc5: 94.7917 (94.7917) time: 4.9194 data: 4.7014 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8981 (0.9118) acc1: 75.5208 (75.9040) acc5: 93.7500 (92.9920) time: 0.6980 data: 0.5225 max mem: 64948 Test: Total time: 0:00:06 (0.7056 s / it) * Acc@1 76.636 Acc@5 93.322 loss 0.897 Accuracy of the model EMA on 50000 test images: 76.6% Max EMA accuracy: 76.64% Epoch: [113] [ 0/312] eta: 0:48:20 lr: 0.003556 min_lr: 0.003556 loss: 2.7228 (2.7228) weight_decay: 0.0500 (0.0500) time: 9.2973 data: 8.5170 max mem: 64948 Epoch: [113] [ 10/312] eta: 0:07:40 lr: 0.003556 min_lr: 0.003556 loss: 2.3723 (2.2544) weight_decay: 0.0500 (0.0500) time: 1.5250 data: 0.7949 max mem: 64948 Epoch: [113] [ 20/312] eta: 0:05:30 lr: 0.003555 min_lr: 0.003555 loss: 2.2479 (2.2355) weight_decay: 0.0500 (0.0500) time: 0.7234 data: 0.0115 max mem: 64948 Epoch: [113] [ 30/312] eta: 0:04:39 lr: 0.003555 min_lr: 0.003555 loss: 2.2114 (2.2079) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [113] [ 40/312] eta: 0:04:09 lr: 0.003555 min_lr: 0.003555 loss: 2.1945 (2.2036) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [113] [ 50/312] eta: 0:03:49 lr: 0.003554 min_lr: 0.003554 loss: 2.2137 (2.2225) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [113] [ 60/312] eta: 0:03:33 lr: 0.003554 min_lr: 0.003554 loss: 2.2137 (2.2347) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [113] [ 70/312] eta: 0:03:19 lr: 0.003554 min_lr: 0.003554 loss: 2.1140 (2.2229) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [113] [ 80/312] eta: 0:03:07 lr: 0.003554 min_lr: 0.003554 loss: 2.2637 (2.2234) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [113] [ 90/312] eta: 0:02:56 lr: 0.003553 min_lr: 0.003553 loss: 2.1926 (2.2123) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [113] [100/312] eta: 0:02:46 lr: 0.003553 min_lr: 0.003553 loss: 2.1126 (2.1981) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [113] [110/312] eta: 0:02:37 lr: 0.003553 min_lr: 0.003553 loss: 2.0365 (2.1967) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [113] [120/312] eta: 0:02:28 lr: 0.003552 min_lr: 0.003552 loss: 2.2547 (2.1962) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [113] [130/312] eta: 0:02:19 lr: 0.003552 min_lr: 0.003552 loss: 2.2837 (2.2009) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [113] [140/312] eta: 0:02:10 lr: 0.003552 min_lr: 0.003552 loss: 1.9609 (2.1798) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [113] [150/312] eta: 0:02:02 lr: 0.003552 min_lr: 0.003552 loss: 1.9609 (2.1829) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [113] [160/312] eta: 0:01:54 lr: 0.003551 min_lr: 0.003551 loss: 2.3391 (2.1903) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [113] [170/312] eta: 0:01:46 lr: 0.003551 min_lr: 0.003551 loss: 2.2596 (2.1963) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [113] [180/312] eta: 0:01:38 lr: 0.003551 min_lr: 0.003551 loss: 2.3583 (2.2044) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [113] [190/312] eta: 0:01:30 lr: 0.003550 min_lr: 0.003550 loss: 2.3583 (2.2095) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [113] [200/312] eta: 0:01:22 lr: 0.003550 min_lr: 0.003550 loss: 2.2217 (2.2087) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [113] [210/312] eta: 0:01:15 lr: 0.003550 min_lr: 0.003550 loss: 2.2661 (2.2181) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [113] [220/312] eta: 0:01:07 lr: 0.003549 min_lr: 0.003549 loss: 2.4274 (2.2280) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [113] [230/312] eta: 0:01:00 lr: 0.003549 min_lr: 0.003549 loss: 2.4022 (2.2305) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [113] [240/312] eta: 0:00:52 lr: 0.003549 min_lr: 0.003549 loss: 2.4022 (2.2335) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [113] [250/312] eta: 0:00:45 lr: 0.003549 min_lr: 0.003549 loss: 2.3138 (2.2314) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [113] [260/312] eta: 0:00:37 lr: 0.003548 min_lr: 0.003548 loss: 2.3138 (2.2352) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [113] [270/312] eta: 0:00:30 lr: 0.003548 min_lr: 0.003548 loss: 2.3723 (2.2381) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [113] [280/312] eta: 0:00:23 lr: 0.003548 min_lr: 0.003548 loss: 1.9619 (2.2303) weight_decay: 0.0500 (0.0500) time: 0.7008 data: 0.0009 max mem: 64948 Epoch: [113] [290/312] eta: 0:00:15 lr: 0.003547 min_lr: 0.003547 loss: 2.2026 (2.2377) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0008 max mem: 64948 Epoch: [113] [300/312] eta: 0:00:08 lr: 0.003547 min_lr: 0.003547 loss: 2.4908 (2.2385) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [113] [310/312] eta: 0:00:01 lr: 0.003547 min_lr: 0.003547 loss: 2.3843 (2.2407) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [113] [311/312] eta: 0:00:00 lr: 0.003547 min_lr: 0.003547 loss: 2.4021 (2.2424) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [113] Total time: 0:03:46 (0.7276 s / it) Averaged stats: lr: 0.003547 min_lr: 0.003547 loss: 2.4021 (2.2860) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7751 (0.7751) acc1: 80.7292 (80.7292) acc5: 93.7500 (93.7500) time: 4.4911 data: 4.2711 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9692 (0.9579) acc1: 75.0000 (75.3920) acc5: 93.2292 (93.1200) time: 0.6505 data: 0.4747 max mem: 64948 Test: Total time: 0:00:06 (0.6750 s / it) * Acc@1 76.342 Acc@5 93.108 loss 0.938 Accuracy of the model on the 50000 test images: 76.3% Max accuracy: 76.34% Test: [0/9] eta: 0:00:39 loss: 0.7352 (0.7352) acc1: 80.9896 (80.9896) acc5: 94.7917 (94.7917) time: 4.3351 data: 4.1280 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8927 (0.9048) acc1: 76.3021 (76.1280) acc5: 93.7500 (93.0240) time: 0.6330 data: 0.4588 max mem: 64948 Test: Total time: 0:00:05 (0.6419 s / it) * Acc@1 76.808 Acc@5 93.376 loss 0.890 Accuracy of the model EMA on 50000 test images: 76.8% Max EMA accuracy: 76.81% Epoch: [114] [ 0/312] eta: 0:49:23 lr: 0.003547 min_lr: 0.003547 loss: 1.8111 (1.8111) weight_decay: 0.0500 (0.0500) time: 9.4985 data: 8.7183 max mem: 64948 Epoch: [114] [ 10/312] eta: 0:07:50 lr: 0.003546 min_lr: 0.003546 loss: 2.4345 (2.3620) weight_decay: 0.0500 (0.0500) time: 1.5578 data: 0.7930 max mem: 64948 Epoch: [114] [ 20/312] eta: 0:05:35 lr: 0.003546 min_lr: 0.003546 loss: 2.4453 (2.3264) weight_decay: 0.0500 (0.0500) time: 0.7319 data: 0.0004 max mem: 64948 Epoch: [114] [ 30/312] eta: 0:04:42 lr: 0.003546 min_lr: 0.003546 loss: 2.4572 (2.3593) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0003 max mem: 64948 Epoch: [114] [ 40/312] eta: 0:04:12 lr: 0.003546 min_lr: 0.003546 loss: 2.4981 (2.4069) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [114] [ 50/312] eta: 0:03:50 lr: 0.003545 min_lr: 0.003545 loss: 2.4493 (2.3666) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [114] [ 60/312] eta: 0:03:34 lr: 0.003545 min_lr: 0.003545 loss: 2.3760 (2.3966) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [114] [ 70/312] eta: 0:03:20 lr: 0.003545 min_lr: 0.003545 loss: 2.4296 (2.3730) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [114] [ 80/312] eta: 0:03:08 lr: 0.003544 min_lr: 0.003544 loss: 2.1429 (2.3275) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [114] [ 90/312] eta: 0:02:57 lr: 0.003544 min_lr: 0.003544 loss: 2.3011 (2.3383) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [114] [100/312] eta: 0:02:47 lr: 0.003544 min_lr: 0.003544 loss: 2.3249 (2.3117) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [114] [110/312] eta: 0:02:37 lr: 0.003543 min_lr: 0.003543 loss: 2.0995 (2.3082) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [114] [120/312] eta: 0:02:28 lr: 0.003543 min_lr: 0.003543 loss: 2.4407 (2.2992) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [114] [130/312] eta: 0:02:19 lr: 0.003543 min_lr: 0.003543 loss: 2.4628 (2.3026) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [114] [140/312] eta: 0:02:11 lr: 0.003543 min_lr: 0.003543 loss: 2.3508 (2.2977) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [114] [150/312] eta: 0:02:02 lr: 0.003542 min_lr: 0.003542 loss: 1.9786 (2.2756) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [114] [160/312] eta: 0:01:54 lr: 0.003542 min_lr: 0.003542 loss: 2.1827 (2.2848) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [114] [170/312] eta: 0:01:46 lr: 0.003542 min_lr: 0.003542 loss: 2.2950 (2.2758) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [114] [180/312] eta: 0:01:38 lr: 0.003541 min_lr: 0.003541 loss: 2.3430 (2.2815) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [114] [190/312] eta: 0:01:30 lr: 0.003541 min_lr: 0.003541 loss: 2.5177 (2.2906) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [114] [200/312] eta: 0:01:23 lr: 0.003541 min_lr: 0.003541 loss: 2.2956 (2.2695) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [114] [210/312] eta: 0:01:15 lr: 0.003540 min_lr: 0.003540 loss: 2.0895 (2.2700) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [114] [220/312] eta: 0:01:07 lr: 0.003540 min_lr: 0.003540 loss: 2.2190 (2.2742) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [114] [230/312] eta: 0:01:00 lr: 0.003540 min_lr: 0.003540 loss: 2.4303 (2.2803) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [114] [240/312] eta: 0:00:52 lr: 0.003540 min_lr: 0.003540 loss: 2.4702 (2.2805) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [114] [250/312] eta: 0:00:45 lr: 0.003539 min_lr: 0.003539 loss: 2.2200 (2.2796) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [114] [260/312] eta: 0:00:38 lr: 0.003539 min_lr: 0.003539 loss: 2.2237 (2.2834) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [114] [270/312] eta: 0:00:30 lr: 0.003539 min_lr: 0.003539 loss: 2.3472 (2.2843) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [114] [280/312] eta: 0:00:23 lr: 0.003538 min_lr: 0.003538 loss: 2.3472 (2.2847) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [114] [290/312] eta: 0:00:15 lr: 0.003538 min_lr: 0.003538 loss: 2.3278 (2.2791) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [114] [300/312] eta: 0:00:08 lr: 0.003538 min_lr: 0.003538 loss: 2.1302 (2.2735) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [114] [310/312] eta: 0:00:01 lr: 0.003537 min_lr: 0.003537 loss: 2.2401 (2.2749) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [114] [311/312] eta: 0:00:00 lr: 0.003537 min_lr: 0.003537 loss: 2.2401 (2.2737) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [114] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.003537 min_lr: 0.003537 loss: 2.2401 (2.2824) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:47 loss: 0.7402 (0.7402) acc1: 83.0729 (83.0729) acc5: 94.2708 (94.2708) time: 5.3307 data: 5.1114 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9904 (0.9644) acc1: 74.7396 (74.5920) acc5: 93.2292 (92.9280) time: 0.7440 data: 0.5680 max mem: 64948 Test: Total time: 0:00:06 (0.7540 s / it) * Acc@1 75.380 Acc@5 92.520 loss 0.960 Accuracy of the model on the 50000 test images: 75.4% Max accuracy: 76.34% Test: [0/9] eta: 0:00:46 loss: 0.7287 (0.7287) acc1: 81.2500 (81.2500) acc5: 94.7917 (94.7917) time: 5.1863 data: 4.9789 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8878 (0.8981) acc1: 76.3021 (76.2560) acc5: 93.7500 (93.0560) time: 0.7275 data: 0.5533 max mem: 64948 Test: Total time: 0:00:06 (0.7349 s / it) * Acc@1 76.958 Acc@5 93.452 loss 0.883 Accuracy of the model EMA on 50000 test images: 77.0% Max EMA accuracy: 76.96% Epoch: [115] [ 0/312] eta: 0:51:30 lr: 0.003537 min_lr: 0.003537 loss: 1.7282 (1.7282) weight_decay: 0.0500 (0.0500) time: 9.9058 data: 7.4099 max mem: 64948 Epoch: [115] [ 10/312] eta: 0:07:50 lr: 0.003537 min_lr: 0.003537 loss: 1.7673 (1.9631) weight_decay: 0.0500 (0.0500) time: 1.5567 data: 0.6740 max mem: 64948 Epoch: [115] [ 20/312] eta: 0:05:35 lr: 0.003537 min_lr: 0.003537 loss: 1.9968 (2.0705) weight_decay: 0.0500 (0.0500) time: 0.7104 data: 0.0003 max mem: 64948 Epoch: [115] [ 30/312] eta: 0:04:42 lr: 0.003537 min_lr: 0.003537 loss: 2.3605 (2.1686) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [115] [ 40/312] eta: 0:04:12 lr: 0.003536 min_lr: 0.003536 loss: 2.3532 (2.1936) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [115] [ 50/312] eta: 0:03:50 lr: 0.003536 min_lr: 0.003536 loss: 2.2329 (2.1724) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [115] [ 60/312] eta: 0:03:34 lr: 0.003536 min_lr: 0.003536 loss: 2.2987 (2.2225) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [115] [ 70/312] eta: 0:03:20 lr: 0.003535 min_lr: 0.003535 loss: 2.3794 (2.2289) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [115] [ 80/312] eta: 0:03:08 lr: 0.003535 min_lr: 0.003535 loss: 2.4395 (2.2629) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [115] [ 90/312] eta: 0:02:57 lr: 0.003535 min_lr: 0.003535 loss: 2.4165 (2.2443) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [115] [100/312] eta: 0:02:47 lr: 0.003534 min_lr: 0.003534 loss: 1.8981 (2.2144) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [115] [110/312] eta: 0:02:37 lr: 0.003534 min_lr: 0.003534 loss: 2.1199 (2.2268) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [115] [120/312] eta: 0:02:28 lr: 0.003534 min_lr: 0.003534 loss: 2.3251 (2.2148) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [115] [130/312] eta: 0:02:19 lr: 0.003534 min_lr: 0.003534 loss: 2.1905 (2.2232) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [115] [140/312] eta: 0:02:11 lr: 0.003533 min_lr: 0.003533 loss: 2.2095 (2.2160) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [115] [150/312] eta: 0:02:02 lr: 0.003533 min_lr: 0.003533 loss: 2.3311 (2.2298) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [115] [160/312] eta: 0:01:54 lr: 0.003533 min_lr: 0.003533 loss: 2.3548 (2.2343) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [115] [170/312] eta: 0:01:46 lr: 0.003532 min_lr: 0.003532 loss: 2.3578 (2.2441) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [115] [180/312] eta: 0:01:38 lr: 0.003532 min_lr: 0.003532 loss: 2.2770 (2.2452) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [115] [190/312] eta: 0:01:30 lr: 0.003532 min_lr: 0.003532 loss: 2.1556 (2.2429) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [115] [200/312] eta: 0:01:23 lr: 0.003531 min_lr: 0.003531 loss: 2.0642 (2.2336) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [115] [210/312] eta: 0:01:15 lr: 0.003531 min_lr: 0.003531 loss: 2.3003 (2.2385) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [115] [220/312] eta: 0:01:07 lr: 0.003531 min_lr: 0.003531 loss: 2.3003 (2.2367) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [115] [230/312] eta: 0:01:00 lr: 0.003530 min_lr: 0.003530 loss: 2.1776 (2.2383) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [115] [240/312] eta: 0:00:52 lr: 0.003530 min_lr: 0.003530 loss: 2.2157 (2.2360) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [115] [250/312] eta: 0:00:45 lr: 0.003530 min_lr: 0.003530 loss: 2.1895 (2.2338) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [115] [260/312] eta: 0:00:38 lr: 0.003530 min_lr: 0.003530 loss: 2.3392 (2.2355) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [115] [270/312] eta: 0:00:30 lr: 0.003529 min_lr: 0.003529 loss: 2.3329 (2.2337) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [115] [280/312] eta: 0:00:23 lr: 0.003529 min_lr: 0.003529 loss: 2.4011 (2.2433) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [115] [290/312] eta: 0:00:16 lr: 0.003529 min_lr: 0.003529 loss: 2.4011 (2.2421) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [115] [300/312] eta: 0:00:08 lr: 0.003528 min_lr: 0.003528 loss: 2.1521 (2.2405) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [115] [310/312] eta: 0:00:01 lr: 0.003528 min_lr: 0.003528 loss: 2.1521 (2.2442) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [115] [311/312] eta: 0:00:00 lr: 0.003528 min_lr: 0.003528 loss: 2.1938 (2.2447) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [115] Total time: 0:03:47 (0.7278 s / it) Averaged stats: lr: 0.003528 min_lr: 0.003528 loss: 2.1938 (2.2869) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.7256 (0.7256) acc1: 83.0729 (83.0729) acc5: 94.5312 (94.5312) time: 4.6829 data: 4.4629 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0335 (1.0364) acc1: 74.7396 (74.0160) acc5: 93.2292 (91.8080) time: 0.6720 data: 0.4960 max mem: 64948 Test: Total time: 0:00:06 (0.6960 s / it) * Acc@1 74.866 Acc@5 92.492 loss 1.003 Accuracy of the model on the 50000 test images: 74.9% Max accuracy: 76.34% Test: [0/9] eta: 0:00:42 loss: 0.7225 (0.7225) acc1: 81.5104 (81.5104) acc5: 94.7917 (94.7917) time: 4.7468 data: 4.5340 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8824 (0.8919) acc1: 76.3021 (76.4480) acc5: 94.0104 (93.1200) time: 0.6787 data: 0.5039 max mem: 64948 Test: Total time: 0:00:06 (0.6930 s / it) * Acc@1 77.082 Acc@5 93.502 loss 0.877 Accuracy of the model EMA on 50000 test images: 77.1% Max EMA accuracy: 77.08% Epoch: [116] [ 0/312] eta: 0:52:57 lr: 0.003528 min_lr: 0.003528 loss: 2.2936 (2.2936) weight_decay: 0.0500 (0.0500) time: 10.1848 data: 9.3924 max mem: 64948 Epoch: [116] [ 10/312] eta: 0:07:53 lr: 0.003528 min_lr: 0.003528 loss: 2.2266 (2.1628) weight_decay: 0.0500 (0.0500) time: 1.5667 data: 0.8542 max mem: 64948 Epoch: [116] [ 20/312] eta: 0:05:36 lr: 0.003527 min_lr: 0.003527 loss: 2.2266 (2.2233) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [116] [ 30/312] eta: 0:04:42 lr: 0.003527 min_lr: 0.003527 loss: 2.3665 (2.2423) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [116] [ 40/312] eta: 0:04:12 lr: 0.003527 min_lr: 0.003527 loss: 2.3111 (2.2104) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [116] [ 50/312] eta: 0:03:51 lr: 0.003527 min_lr: 0.003527 loss: 2.2935 (2.2273) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [116] [ 60/312] eta: 0:03:35 lr: 0.003526 min_lr: 0.003526 loss: 2.2889 (2.2329) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [116] [ 70/312] eta: 0:03:21 lr: 0.003526 min_lr: 0.003526 loss: 2.2889 (2.2502) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [116] [ 80/312] eta: 0:03:08 lr: 0.003526 min_lr: 0.003526 loss: 2.3731 (2.2713) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [116] [ 90/312] eta: 0:02:57 lr: 0.003525 min_lr: 0.003525 loss: 2.3840 (2.2903) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [116] [100/312] eta: 0:02:47 lr: 0.003525 min_lr: 0.003525 loss: 2.3817 (2.2890) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [116] [110/312] eta: 0:02:38 lr: 0.003525 min_lr: 0.003525 loss: 2.5384 (2.2959) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [116] [120/312] eta: 0:02:28 lr: 0.003524 min_lr: 0.003524 loss: 2.0653 (2.2809) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [116] [130/312] eta: 0:02:19 lr: 0.003524 min_lr: 0.003524 loss: 2.0021 (2.2684) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [116] [140/312] eta: 0:02:11 lr: 0.003524 min_lr: 0.003524 loss: 2.1068 (2.2677) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [116] [150/312] eta: 0:02:03 lr: 0.003523 min_lr: 0.003523 loss: 2.1068 (2.2539) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [116] [160/312] eta: 0:01:54 lr: 0.003523 min_lr: 0.003523 loss: 2.1922 (2.2561) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [116] [170/312] eta: 0:01:46 lr: 0.003523 min_lr: 0.003523 loss: 2.3144 (2.2523) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [116] [180/312] eta: 0:01:38 lr: 0.003523 min_lr: 0.003523 loss: 2.3114 (2.2568) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [116] [190/312] eta: 0:01:30 lr: 0.003522 min_lr: 0.003522 loss: 2.3782 (2.2598) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [116] [200/312] eta: 0:01:23 lr: 0.003522 min_lr: 0.003522 loss: 2.2360 (2.2497) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [116] [210/312] eta: 0:01:15 lr: 0.003522 min_lr: 0.003522 loss: 2.2163 (2.2509) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [116] [220/312] eta: 0:01:07 lr: 0.003521 min_lr: 0.003521 loss: 2.4079 (2.2525) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [116] [230/312] eta: 0:01:00 lr: 0.003521 min_lr: 0.003521 loss: 2.4497 (2.2545) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [116] [240/312] eta: 0:00:52 lr: 0.003521 min_lr: 0.003521 loss: 2.3897 (2.2573) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [116] [250/312] eta: 0:00:45 lr: 0.003520 min_lr: 0.003520 loss: 2.4504 (2.2645) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [116] [260/312] eta: 0:00:38 lr: 0.003520 min_lr: 0.003520 loss: 2.4498 (2.2671) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [116] [270/312] eta: 0:00:30 lr: 0.003520 min_lr: 0.003520 loss: 2.2358 (2.2648) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [116] [280/312] eta: 0:00:23 lr: 0.003520 min_lr: 0.003520 loss: 2.1784 (2.2599) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0009 max mem: 64948 Epoch: [116] [290/312] eta: 0:00:16 lr: 0.003519 min_lr: 0.003519 loss: 2.0472 (2.2534) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [116] [300/312] eta: 0:00:08 lr: 0.003519 min_lr: 0.003519 loss: 2.3633 (2.2573) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [116] [310/312] eta: 0:00:01 lr: 0.003519 min_lr: 0.003519 loss: 2.3938 (2.2592) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [116] [311/312] eta: 0:00:00 lr: 0.003519 min_lr: 0.003519 loss: 2.3368 (2.2590) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [116] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.003519 min_lr: 0.003519 loss: 2.3368 (2.2820) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6941 (0.6941) acc1: 83.0729 (83.0729) acc5: 96.3542 (96.3542) time: 4.4483 data: 4.2282 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9688 (0.9414) acc1: 77.3438 (76.2240) acc5: 93.4896 (93.1840) time: 0.6456 data: 0.4699 max mem: 64948 Test: Total time: 0:00:06 (0.6690 s / it) * Acc@1 76.494 Acc@5 93.378 loss 0.914 Accuracy of the model on the 50000 test images: 76.5% Max accuracy: 76.49% Test: [0/9] eta: 0:00:41 loss: 0.7161 (0.7161) acc1: 82.2917 (82.2917) acc5: 94.7917 (94.7917) time: 4.6299 data: 4.4244 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8775 (0.8857) acc1: 76.5625 (76.7360) acc5: 94.2708 (93.1840) time: 0.6658 data: 0.4917 max mem: 64948 Test: Total time: 0:00:06 (0.6731 s / it) * Acc@1 77.246 Acc@5 93.560 loss 0.871 Accuracy of the model EMA on 50000 test images: 77.2% Max EMA accuracy: 77.25% Epoch: [117] [ 0/312] eta: 0:49:58 lr: 0.003519 min_lr: 0.003519 loss: 2.8022 (2.8022) weight_decay: 0.0500 (0.0500) time: 9.6091 data: 7.8401 max mem: 64948 Epoch: [117] [ 10/312] eta: 0:07:48 lr: 0.003518 min_lr: 0.003518 loss: 2.2923 (2.3142) weight_decay: 0.0500 (0.0500) time: 1.5522 data: 0.7132 max mem: 64948 Epoch: [117] [ 20/312] eta: 0:05:34 lr: 0.003518 min_lr: 0.003518 loss: 2.3080 (2.3654) weight_decay: 0.0500 (0.0500) time: 0.7238 data: 0.0004 max mem: 64948 Epoch: [117] [ 30/312] eta: 0:04:42 lr: 0.003518 min_lr: 0.003518 loss: 2.2508 (2.2890) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [117] [ 40/312] eta: 0:04:11 lr: 0.003517 min_lr: 0.003517 loss: 2.1941 (2.2974) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [117] [ 50/312] eta: 0:03:50 lr: 0.003517 min_lr: 0.003517 loss: 2.2046 (2.2772) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [117] [ 60/312] eta: 0:03:34 lr: 0.003517 min_lr: 0.003517 loss: 2.1658 (2.2581) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [117] [ 70/312] eta: 0:03:20 lr: 0.003516 min_lr: 0.003516 loss: 2.3643 (2.2845) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [117] [ 80/312] eta: 0:03:08 lr: 0.003516 min_lr: 0.003516 loss: 2.4404 (2.2783) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [117] [ 90/312] eta: 0:02:57 lr: 0.003516 min_lr: 0.003516 loss: 2.4130 (2.2905) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [117] [100/312] eta: 0:02:47 lr: 0.003516 min_lr: 0.003516 loss: 2.4498 (2.3103) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [117] [110/312] eta: 0:02:37 lr: 0.003515 min_lr: 0.003515 loss: 2.4413 (2.3006) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [117] [120/312] eta: 0:02:28 lr: 0.003515 min_lr: 0.003515 loss: 2.3202 (2.2917) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [117] [130/312] eta: 0:02:19 lr: 0.003515 min_lr: 0.003515 loss: 2.2063 (2.2808) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [117] [140/312] eta: 0:02:11 lr: 0.003514 min_lr: 0.003514 loss: 2.1083 (2.2784) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [117] [150/312] eta: 0:02:02 lr: 0.003514 min_lr: 0.003514 loss: 2.3546 (2.2851) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [117] [160/312] eta: 0:01:54 lr: 0.003514 min_lr: 0.003514 loss: 2.4197 (2.2924) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [117] [170/312] eta: 0:01:46 lr: 0.003513 min_lr: 0.003513 loss: 2.3962 (2.2929) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [117] [180/312] eta: 0:01:38 lr: 0.003513 min_lr: 0.003513 loss: 2.2679 (2.2817) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [117] [190/312] eta: 0:01:30 lr: 0.003513 min_lr: 0.003513 loss: 2.2654 (2.2805) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [117] [200/312] eta: 0:01:23 lr: 0.003512 min_lr: 0.003512 loss: 2.2654 (2.2748) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [117] [210/312] eta: 0:01:15 lr: 0.003512 min_lr: 0.003512 loss: 2.1860 (2.2654) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [117] [220/312] eta: 0:01:07 lr: 0.003512 min_lr: 0.003512 loss: 2.3501 (2.2731) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [117] [230/312] eta: 0:01:00 lr: 0.003512 min_lr: 0.003512 loss: 2.3231 (2.2638) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [117] [240/312] eta: 0:00:52 lr: 0.003511 min_lr: 0.003511 loss: 2.3167 (2.2661) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0003 max mem: 64948 Epoch: [117] [250/312] eta: 0:00:45 lr: 0.003511 min_lr: 0.003511 loss: 2.4169 (2.2725) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [117] [260/312] eta: 0:00:37 lr: 0.003511 min_lr: 0.003511 loss: 2.4683 (2.2811) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [117] [270/312] eta: 0:00:30 lr: 0.003510 min_lr: 0.003510 loss: 2.4162 (2.2813) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [117] [280/312] eta: 0:00:23 lr: 0.003510 min_lr: 0.003510 loss: 2.2446 (2.2772) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [117] [290/312] eta: 0:00:15 lr: 0.003510 min_lr: 0.003510 loss: 2.1520 (2.2813) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [117] [300/312] eta: 0:00:08 lr: 0.003509 min_lr: 0.003509 loss: 2.2604 (2.2785) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [117] [310/312] eta: 0:00:01 lr: 0.003509 min_lr: 0.003509 loss: 2.2604 (2.2728) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [117] [311/312] eta: 0:00:00 lr: 0.003509 min_lr: 0.003509 loss: 2.2604 (2.2717) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [117] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.003509 min_lr: 0.003509 loss: 2.2604 (2.2851) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7083 (0.7083) acc1: 82.0312 (82.0312) acc5: 95.3125 (95.3125) time: 4.4492 data: 4.2356 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9856 (0.9373) acc1: 75.2604 (74.9440) acc5: 94.2708 (93.1840) time: 0.6456 data: 0.4707 max mem: 64948 Test: Total time: 0:00:06 (0.6708 s / it) * Acc@1 76.446 Acc@5 93.332 loss 0.917 Accuracy of the model on the 50000 test images: 76.4% Max accuracy: 76.49% Test: [0/9] eta: 0:00:41 loss: 0.7099 (0.7099) acc1: 82.2917 (82.2917) acc5: 95.0521 (95.0521) time: 4.6218 data: 4.4037 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8737 (0.8802) acc1: 76.3021 (76.7040) acc5: 94.2708 (93.2800) time: 0.6707 data: 0.4953 max mem: 64948 Test: Total time: 0:00:06 (0.6787 s / it) * Acc@1 77.368 Acc@5 93.622 loss 0.866 Accuracy of the model EMA on 50000 test images: 77.4% Max EMA accuracy: 77.37% Epoch: [118] [ 0/312] eta: 0:52:30 lr: 0.003509 min_lr: 0.003509 loss: 2.6878 (2.6878) weight_decay: 0.0500 (0.0500) time: 10.0985 data: 9.2928 max mem: 64948 Epoch: [118] [ 10/312] eta: 0:07:53 lr: 0.003509 min_lr: 0.003509 loss: 2.4514 (2.3996) weight_decay: 0.0500 (0.0500) time: 1.5666 data: 0.8452 max mem: 64948 Epoch: [118] [ 20/312] eta: 0:05:36 lr: 0.003508 min_lr: 0.003508 loss: 2.3614 (2.2858) weight_decay: 0.0500 (0.0500) time: 0.7036 data: 0.0004 max mem: 64948 Epoch: [118] [ 30/312] eta: 0:04:42 lr: 0.003508 min_lr: 0.003508 loss: 2.3923 (2.3172) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [118] [ 40/312] eta: 0:04:12 lr: 0.003508 min_lr: 0.003508 loss: 2.3294 (2.2681) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [118] [ 50/312] eta: 0:03:51 lr: 0.003507 min_lr: 0.003507 loss: 2.2741 (2.2871) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [118] [ 60/312] eta: 0:03:34 lr: 0.003507 min_lr: 0.003507 loss: 2.3360 (2.2812) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [118] [ 70/312] eta: 0:03:21 lr: 0.003507 min_lr: 0.003507 loss: 2.3360 (2.2884) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [118] [ 80/312] eta: 0:03:08 lr: 0.003507 min_lr: 0.003507 loss: 2.3701 (2.2994) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [118] [ 90/312] eta: 0:02:57 lr: 0.003506 min_lr: 0.003506 loss: 2.3057 (2.2809) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [118] [100/312] eta: 0:02:47 lr: 0.003506 min_lr: 0.003506 loss: 2.3057 (2.2925) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [118] [110/312] eta: 0:02:38 lr: 0.003506 min_lr: 0.003506 loss: 2.3550 (2.2799) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [118] [120/312] eta: 0:02:28 lr: 0.003505 min_lr: 0.003505 loss: 2.3897 (2.3060) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [118] [130/312] eta: 0:02:20 lr: 0.003505 min_lr: 0.003505 loss: 2.3897 (2.2895) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [118] [140/312] eta: 0:02:11 lr: 0.003505 min_lr: 0.003505 loss: 2.1438 (2.2885) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [118] [150/312] eta: 0:02:02 lr: 0.003504 min_lr: 0.003504 loss: 2.3291 (2.3028) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [118] [160/312] eta: 0:01:54 lr: 0.003504 min_lr: 0.003504 loss: 2.3963 (2.2987) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [118] [170/312] eta: 0:01:46 lr: 0.003504 min_lr: 0.003504 loss: 2.3963 (2.3021) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [118] [180/312] eta: 0:01:38 lr: 0.003503 min_lr: 0.003503 loss: 2.4341 (2.3104) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [118] [190/312] eta: 0:01:30 lr: 0.003503 min_lr: 0.003503 loss: 2.4123 (2.3090) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [118] [200/312] eta: 0:01:23 lr: 0.003503 min_lr: 0.003503 loss: 2.3698 (2.3061) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [118] [210/312] eta: 0:01:15 lr: 0.003503 min_lr: 0.003503 loss: 2.2779 (2.2975) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [118] [220/312] eta: 0:01:07 lr: 0.003502 min_lr: 0.003502 loss: 2.3485 (2.2996) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [118] [230/312] eta: 0:01:00 lr: 0.003502 min_lr: 0.003502 loss: 2.3058 (2.2931) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [118] [240/312] eta: 0:00:52 lr: 0.003502 min_lr: 0.003502 loss: 2.2668 (2.2951) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [118] [250/312] eta: 0:00:45 lr: 0.003501 min_lr: 0.003501 loss: 2.3981 (2.2938) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [118] [260/312] eta: 0:00:38 lr: 0.003501 min_lr: 0.003501 loss: 2.3936 (2.2945) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [118] [270/312] eta: 0:00:30 lr: 0.003501 min_lr: 0.003501 loss: 2.3620 (2.2941) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [118] [280/312] eta: 0:00:23 lr: 0.003500 min_lr: 0.003500 loss: 2.2153 (2.2906) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0009 max mem: 64948 Epoch: [118] [290/312] eta: 0:00:16 lr: 0.003500 min_lr: 0.003500 loss: 2.1727 (2.2879) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0008 max mem: 64948 Epoch: [118] [300/312] eta: 0:00:08 lr: 0.003500 min_lr: 0.003500 loss: 2.2483 (2.2930) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [118] [310/312] eta: 0:00:01 lr: 0.003499 min_lr: 0.003499 loss: 2.2448 (2.2866) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [118] [311/312] eta: 0:00:00 lr: 0.003499 min_lr: 0.003499 loss: 2.1870 (2.2863) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [118] Total time: 0:03:47 (0.7284 s / it) Averaged stats: lr: 0.003499 min_lr: 0.003499 loss: 2.1870 (2.2748) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7175 (0.7175) acc1: 83.3333 (83.3333) acc5: 95.0521 (95.0521) time: 4.5918 data: 4.3723 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0024 (0.9745) acc1: 76.0417 (74.8800) acc5: 92.7083 (92.7040) time: 0.6621 data: 0.4859 max mem: 64948 Test: Total time: 0:00:06 (0.6842 s / it) * Acc@1 76.228 Acc@5 93.108 loss 0.946 Accuracy of the model on the 50000 test images: 76.2% Max accuracy: 76.49% Test: [0/9] eta: 0:00:46 loss: 0.7038 (0.7038) acc1: 82.2917 (82.2917) acc5: 95.0521 (95.0521) time: 5.2098 data: 5.0003 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8698 (0.8750) acc1: 76.5625 (76.8000) acc5: 94.5312 (93.4400) time: 0.7302 data: 0.5557 max mem: 64948 Test: Total time: 0:00:06 (0.7380 s / it) * Acc@1 77.490 Acc@5 93.682 loss 0.860 Accuracy of the model EMA on 50000 test images: 77.5% Max EMA accuracy: 77.49% Epoch: [119] [ 0/312] eta: 0:52:29 lr: 0.003499 min_lr: 0.003499 loss: 2.5330 (2.5330) weight_decay: 0.0500 (0.0500) time: 10.0951 data: 8.0878 max mem: 64948 Epoch: [119] [ 10/312] eta: 0:07:53 lr: 0.003499 min_lr: 0.003499 loss: 2.6083 (2.4783) weight_decay: 0.0500 (0.0500) time: 1.5694 data: 0.7356 max mem: 64948 Epoch: [119] [ 20/312] eta: 0:05:36 lr: 0.003499 min_lr: 0.003499 loss: 2.3932 (2.3104) weight_decay: 0.0500 (0.0500) time: 0.7063 data: 0.0004 max mem: 64948 Epoch: [119] [ 30/312] eta: 0:04:43 lr: 0.003498 min_lr: 0.003498 loss: 2.2745 (2.3277) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [119] [ 40/312] eta: 0:04:13 lr: 0.003498 min_lr: 0.003498 loss: 2.3066 (2.2945) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [119] [ 50/312] eta: 0:03:51 lr: 0.003498 min_lr: 0.003498 loss: 2.4305 (2.3287) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [119] [ 60/312] eta: 0:03:35 lr: 0.003498 min_lr: 0.003498 loss: 2.4305 (2.3142) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [119] [ 70/312] eta: 0:03:21 lr: 0.003497 min_lr: 0.003497 loss: 2.0403 (2.2721) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [119] [ 80/312] eta: 0:03:09 lr: 0.003497 min_lr: 0.003497 loss: 2.0226 (2.2408) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [119] [ 90/312] eta: 0:02:57 lr: 0.003497 min_lr: 0.003497 loss: 2.2518 (2.2607) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [119] [100/312] eta: 0:02:47 lr: 0.003496 min_lr: 0.003496 loss: 2.3287 (2.2614) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [119] [110/312] eta: 0:02:38 lr: 0.003496 min_lr: 0.003496 loss: 2.2506 (2.2644) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [119] [120/312] eta: 0:02:28 lr: 0.003496 min_lr: 0.003496 loss: 2.1778 (2.2580) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [119] [130/312] eta: 0:02:19 lr: 0.003495 min_lr: 0.003495 loss: 2.1404 (2.2596) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [119] [140/312] eta: 0:02:11 lr: 0.003495 min_lr: 0.003495 loss: 2.2519 (2.2633) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [119] [150/312] eta: 0:02:03 lr: 0.003495 min_lr: 0.003495 loss: 2.2519 (2.2683) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [119] [160/312] eta: 0:01:54 lr: 0.003494 min_lr: 0.003494 loss: 2.4125 (2.2770) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [119] [170/312] eta: 0:01:46 lr: 0.003494 min_lr: 0.003494 loss: 2.3941 (2.2764) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [119] [180/312] eta: 0:01:38 lr: 0.003494 min_lr: 0.003494 loss: 2.3437 (2.2777) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [119] [190/312] eta: 0:01:30 lr: 0.003493 min_lr: 0.003493 loss: 2.3762 (2.2859) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [119] [200/312] eta: 0:01:23 lr: 0.003493 min_lr: 0.003493 loss: 2.3674 (2.2907) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [119] [210/312] eta: 0:01:15 lr: 0.003493 min_lr: 0.003493 loss: 2.2395 (2.2843) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [119] [220/312] eta: 0:01:07 lr: 0.003493 min_lr: 0.003493 loss: 2.3244 (2.2904) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [119] [230/312] eta: 0:01:00 lr: 0.003492 min_lr: 0.003492 loss: 2.4106 (2.2922) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [119] [240/312] eta: 0:00:52 lr: 0.003492 min_lr: 0.003492 loss: 2.3010 (2.2891) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [119] [250/312] eta: 0:00:45 lr: 0.003492 min_lr: 0.003492 loss: 2.1128 (2.2809) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [119] [260/312] eta: 0:00:38 lr: 0.003491 min_lr: 0.003491 loss: 2.2318 (2.2840) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [119] [270/312] eta: 0:00:30 lr: 0.003491 min_lr: 0.003491 loss: 2.4978 (2.2894) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [119] [280/312] eta: 0:00:23 lr: 0.003491 min_lr: 0.003491 loss: 2.2071 (2.2843) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0010 max mem: 64948 Epoch: [119] [290/312] eta: 0:00:16 lr: 0.003490 min_lr: 0.003490 loss: 2.3000 (2.2858) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [119] [300/312] eta: 0:00:08 lr: 0.003490 min_lr: 0.003490 loss: 2.3248 (2.2864) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [119] [310/312] eta: 0:00:01 lr: 0.003490 min_lr: 0.003490 loss: 2.2577 (2.2817) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [119] [311/312] eta: 0:00:00 lr: 0.003490 min_lr: 0.003490 loss: 2.2577 (2.2842) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [119] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.003490 min_lr: 0.003490 loss: 2.2577 (2.2765) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6581 (0.6581) acc1: 84.3750 (84.3750) acc5: 95.3125 (95.3125) time: 4.5864 data: 4.3620 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0105 (0.9317) acc1: 76.3021 (75.5840) acc5: 94.0104 (93.2480) time: 0.6614 data: 0.4847 max mem: 64948 Test: Total time: 0:00:06 (0.6832 s / it) * Acc@1 76.396 Acc@5 93.228 loss 0.926 Accuracy of the model on the 50000 test images: 76.4% Max accuracy: 76.49% Test: [0/9] eta: 0:00:41 loss: 0.6980 (0.6980) acc1: 82.5521 (82.5521) acc5: 95.3125 (95.3125) time: 4.6590 data: 4.4562 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8647 (0.8695) acc1: 76.5625 (76.9600) acc5: 94.5312 (93.5680) time: 0.6690 data: 0.4952 max mem: 64948 Test: Total time: 0:00:06 (0.6779 s / it) * Acc@1 77.632 Acc@5 93.754 loss 0.855 Accuracy of the model EMA on 50000 test images: 77.6% Max EMA accuracy: 77.63% Epoch: [120] [ 0/312] eta: 0:51:15 lr: 0.003490 min_lr: 0.003490 loss: 2.3875 (2.3875) weight_decay: 0.0500 (0.0500) time: 9.8582 data: 7.9931 max mem: 64948 Epoch: [120] [ 10/312] eta: 0:07:48 lr: 0.003489 min_lr: 0.003489 loss: 2.1452 (2.1749) weight_decay: 0.0500 (0.0500) time: 1.5512 data: 0.7270 max mem: 64948 Epoch: [120] [ 20/312] eta: 0:05:33 lr: 0.003489 min_lr: 0.003489 loss: 2.1452 (2.1734) weight_decay: 0.0500 (0.0500) time: 0.7070 data: 0.0004 max mem: 64948 Epoch: [120] [ 30/312] eta: 0:04:41 lr: 0.003489 min_lr: 0.003489 loss: 2.3690 (2.1835) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [120] [ 40/312] eta: 0:04:11 lr: 0.003488 min_lr: 0.003488 loss: 2.2746 (2.1981) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0003 max mem: 64948 Epoch: [120] [ 50/312] eta: 0:03:50 lr: 0.003488 min_lr: 0.003488 loss: 2.2232 (2.1932) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0003 max mem: 64948 Epoch: [120] [ 60/312] eta: 0:03:34 lr: 0.003488 min_lr: 0.003488 loss: 2.1916 (2.1848) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0003 max mem: 64948 Epoch: [120] [ 70/312] eta: 0:03:20 lr: 0.003487 min_lr: 0.003487 loss: 2.2405 (2.1980) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [120] [ 80/312] eta: 0:03:08 lr: 0.003487 min_lr: 0.003487 loss: 2.2079 (2.1938) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [120] [ 90/312] eta: 0:02:57 lr: 0.003487 min_lr: 0.003487 loss: 2.2079 (2.2027) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [120] [100/312] eta: 0:02:47 lr: 0.003487 min_lr: 0.003487 loss: 2.2874 (2.2142) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [120] [110/312] eta: 0:02:37 lr: 0.003486 min_lr: 0.003486 loss: 2.4441 (2.2219) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [120] [120/312] eta: 0:02:28 lr: 0.003486 min_lr: 0.003486 loss: 2.4069 (2.2256) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [120] [130/312] eta: 0:02:19 lr: 0.003486 min_lr: 0.003486 loss: 2.2965 (2.2260) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [120] [140/312] eta: 0:02:11 lr: 0.003485 min_lr: 0.003485 loss: 2.2965 (2.2333) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [120] [150/312] eta: 0:02:02 lr: 0.003485 min_lr: 0.003485 loss: 2.4095 (2.2424) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [120] [160/312] eta: 0:01:54 lr: 0.003485 min_lr: 0.003485 loss: 2.2569 (2.2284) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [120] [170/312] eta: 0:01:46 lr: 0.003484 min_lr: 0.003484 loss: 2.3682 (2.2344) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [120] [180/312] eta: 0:01:38 lr: 0.003484 min_lr: 0.003484 loss: 2.3736 (2.2321) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [120] [190/312] eta: 0:01:30 lr: 0.003484 min_lr: 0.003484 loss: 2.3736 (2.2325) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [120] [200/312] eta: 0:01:23 lr: 0.003483 min_lr: 0.003483 loss: 2.4538 (2.2463) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [120] [210/312] eta: 0:01:15 lr: 0.003483 min_lr: 0.003483 loss: 2.5600 (2.2563) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [120] [220/312] eta: 0:01:07 lr: 0.003483 min_lr: 0.003483 loss: 2.3629 (2.2569) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [120] [230/312] eta: 0:01:00 lr: 0.003482 min_lr: 0.003482 loss: 2.3437 (2.2568) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [120] [240/312] eta: 0:00:52 lr: 0.003482 min_lr: 0.003482 loss: 2.1721 (2.2490) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [120] [250/312] eta: 0:00:45 lr: 0.003482 min_lr: 0.003482 loss: 2.2003 (2.2528) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [120] [260/312] eta: 0:00:38 lr: 0.003482 min_lr: 0.003482 loss: 2.3557 (2.2567) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [120] [270/312] eta: 0:00:30 lr: 0.003481 min_lr: 0.003481 loss: 2.3181 (2.2567) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [120] [280/312] eta: 0:00:23 lr: 0.003481 min_lr: 0.003481 loss: 2.3652 (2.2626) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [120] [290/312] eta: 0:00:16 lr: 0.003481 min_lr: 0.003481 loss: 2.3804 (2.2651) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [120] [300/312] eta: 0:00:08 lr: 0.003480 min_lr: 0.003480 loss: 2.2350 (2.2625) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [120] [310/312] eta: 0:00:01 lr: 0.003480 min_lr: 0.003480 loss: 1.9900 (2.2514) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [120] [311/312] eta: 0:00:00 lr: 0.003480 min_lr: 0.003480 loss: 1.9251 (2.2498) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [120] Total time: 0:03:47 (0.7278 s / it) Averaged stats: lr: 0.003480 min_lr: 0.003480 loss: 1.9251 (2.2722) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.7259 (0.7259) acc1: 82.8125 (82.8125) acc5: 94.5312 (94.5312) time: 4.6757 data: 4.4710 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9988 (0.9330) acc1: 76.5625 (75.9040) acc5: 94.5312 (93.3120) time: 0.6709 data: 0.4969 max mem: 64948 Test: Total time: 0:00:06 (0.6937 s / it) * Acc@1 76.376 Acc@5 93.262 loss 0.914 Accuracy of the model on the 50000 test images: 76.4% Max accuracy: 76.49% Test: [0/9] eta: 0:00:42 loss: 0.6926 (0.6926) acc1: 83.0729 (83.0729) acc5: 95.3125 (95.3125) time: 4.6878 data: 4.4746 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8592 (0.8643) acc1: 76.5625 (77.0560) acc5: 94.5312 (93.6640) time: 0.6722 data: 0.4973 max mem: 64948 Test: Total time: 0:00:06 (0.6808 s / it) * Acc@1 77.720 Acc@5 93.820 loss 0.850 Accuracy of the model EMA on 50000 test images: 77.7% Max EMA accuracy: 77.72% Epoch: [121] [ 0/312] eta: 0:51:27 lr: 0.003480 min_lr: 0.003480 loss: 2.0463 (2.0463) weight_decay: 0.0500 (0.0500) time: 9.8959 data: 8.8014 max mem: 64948 Epoch: [121] [ 10/312] eta: 0:07:47 lr: 0.003480 min_lr: 0.003480 loss: 2.4072 (2.3288) weight_decay: 0.0500 (0.0500) time: 1.5492 data: 0.8005 max mem: 64948 Epoch: [121] [ 20/312] eta: 0:05:33 lr: 0.003479 min_lr: 0.003479 loss: 2.4072 (2.3204) weight_decay: 0.0500 (0.0500) time: 0.7044 data: 0.0004 max mem: 64948 Epoch: [121] [ 30/312] eta: 0:04:41 lr: 0.003479 min_lr: 0.003479 loss: 2.1366 (2.2826) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [121] [ 40/312] eta: 0:04:11 lr: 0.003479 min_lr: 0.003479 loss: 2.1720 (2.2600) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [121] [ 50/312] eta: 0:03:50 lr: 0.003478 min_lr: 0.003478 loss: 2.2747 (2.2837) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [121] [ 60/312] eta: 0:03:33 lr: 0.003478 min_lr: 0.003478 loss: 2.2413 (2.2492) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [121] [ 70/312] eta: 0:03:20 lr: 0.003478 min_lr: 0.003478 loss: 2.2121 (2.2549) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [121] [ 80/312] eta: 0:03:08 lr: 0.003477 min_lr: 0.003477 loss: 2.2121 (2.2426) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [121] [ 90/312] eta: 0:02:57 lr: 0.003477 min_lr: 0.003477 loss: 2.1836 (2.2337) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [121] [100/312] eta: 0:02:47 lr: 0.003477 min_lr: 0.003477 loss: 2.3503 (2.2518) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [121] [110/312] eta: 0:02:37 lr: 0.003476 min_lr: 0.003476 loss: 2.3781 (2.2502) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [121] [120/312] eta: 0:02:28 lr: 0.003476 min_lr: 0.003476 loss: 2.3279 (2.2516) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [121] [130/312] eta: 0:02:19 lr: 0.003476 min_lr: 0.003476 loss: 2.5408 (2.2739) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [121] [140/312] eta: 0:02:11 lr: 0.003475 min_lr: 0.003475 loss: 2.5582 (2.2737) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [121] [150/312] eta: 0:02:02 lr: 0.003475 min_lr: 0.003475 loss: 2.3345 (2.2621) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [121] [160/312] eta: 0:01:54 lr: 0.003475 min_lr: 0.003475 loss: 2.2807 (2.2629) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [121] [170/312] eta: 0:01:46 lr: 0.003475 min_lr: 0.003475 loss: 2.3487 (2.2670) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [121] [180/312] eta: 0:01:38 lr: 0.003474 min_lr: 0.003474 loss: 2.3964 (2.2694) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [121] [190/312] eta: 0:01:30 lr: 0.003474 min_lr: 0.003474 loss: 2.3964 (2.2652) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [121] [200/312] eta: 0:01:23 lr: 0.003474 min_lr: 0.003474 loss: 1.9440 (2.2518) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [121] [210/312] eta: 0:01:15 lr: 0.003473 min_lr: 0.003473 loss: 2.3526 (2.2599) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [121] [220/312] eta: 0:01:07 lr: 0.003473 min_lr: 0.003473 loss: 2.3643 (2.2534) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [121] [230/312] eta: 0:01:00 lr: 0.003473 min_lr: 0.003473 loss: 2.1699 (2.2527) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [121] [240/312] eta: 0:00:52 lr: 0.003472 min_lr: 0.003472 loss: 2.2761 (2.2523) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [121] [250/312] eta: 0:00:45 lr: 0.003472 min_lr: 0.003472 loss: 2.4035 (2.2556) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [121] [260/312] eta: 0:00:38 lr: 0.003472 min_lr: 0.003472 loss: 2.2514 (2.2485) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [121] [270/312] eta: 0:00:30 lr: 0.003471 min_lr: 0.003471 loss: 2.2514 (2.2504) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [121] [280/312] eta: 0:00:23 lr: 0.003471 min_lr: 0.003471 loss: 2.2669 (2.2467) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [121] [290/312] eta: 0:00:15 lr: 0.003471 min_lr: 0.003471 loss: 2.3494 (2.2457) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [121] [300/312] eta: 0:00:08 lr: 0.003470 min_lr: 0.003470 loss: 2.3494 (2.2445) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [121] [310/312] eta: 0:00:01 lr: 0.003470 min_lr: 0.003470 loss: 2.2563 (2.2441) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [121] [311/312] eta: 0:00:00 lr: 0.003470 min_lr: 0.003470 loss: 2.2563 (2.2438) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [121] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.003470 min_lr: 0.003470 loss: 2.2563 (2.2666) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6920 (0.6920) acc1: 82.0312 (82.0312) acc5: 95.0521 (95.0521) time: 4.5632 data: 4.3562 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0209 (0.9737) acc1: 75.5208 (75.0720) acc5: 92.4479 (92.4800) time: 0.6583 data: 0.4841 max mem: 64948 Test: Total time: 0:00:06 (0.6820 s / it) * Acc@1 75.818 Acc@5 92.646 loss 0.962 Accuracy of the model on the 50000 test images: 75.8% Max accuracy: 76.49% Test: [0/9] eta: 0:00:43 loss: 0.6881 (0.6881) acc1: 83.3333 (83.3333) acc5: 95.3125 (95.3125) time: 4.8886 data: 4.6822 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8531 (0.8588) acc1: 77.0833 (77.1200) acc5: 94.5312 (93.6960) time: 0.6944 data: 0.5203 max mem: 64948 Test: Total time: 0:00:06 (0.7032 s / it) * Acc@1 77.854 Acc@5 93.872 loss 0.845 Accuracy of the model EMA on 50000 test images: 77.9% Max EMA accuracy: 77.85% Epoch: [122] [ 0/312] eta: 0:48:56 lr: 0.003470 min_lr: 0.003470 loss: 2.5073 (2.5073) weight_decay: 0.0500 (0.0500) time: 9.4109 data: 8.0171 max mem: 64948 Epoch: [122] [ 10/312] eta: 0:07:40 lr: 0.003470 min_lr: 0.003470 loss: 2.2958 (2.1584) weight_decay: 0.0500 (0.0500) time: 1.5263 data: 0.7549 max mem: 64948 Epoch: [122] [ 20/312] eta: 0:05:30 lr: 0.003469 min_lr: 0.003469 loss: 2.1354 (2.1493) weight_decay: 0.0500 (0.0500) time: 0.7168 data: 0.0145 max mem: 64948 Epoch: [122] [ 30/312] eta: 0:04:39 lr: 0.003469 min_lr: 0.003469 loss: 2.2706 (2.2089) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [122] [ 40/312] eta: 0:04:10 lr: 0.003469 min_lr: 0.003469 loss: 2.3756 (2.2704) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0003 max mem: 64948 Epoch: [122] [ 50/312] eta: 0:03:49 lr: 0.003468 min_lr: 0.003468 loss: 2.3756 (2.2720) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [122] [ 60/312] eta: 0:03:33 lr: 0.003468 min_lr: 0.003468 loss: 2.4029 (2.2862) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [122] [ 70/312] eta: 0:03:19 lr: 0.003468 min_lr: 0.003468 loss: 2.4499 (2.2841) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [122] [ 80/312] eta: 0:03:07 lr: 0.003467 min_lr: 0.003467 loss: 2.3444 (2.2891) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [122] [ 90/312] eta: 0:02:57 lr: 0.003467 min_lr: 0.003467 loss: 2.4492 (2.3030) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [122] [100/312] eta: 0:02:46 lr: 0.003467 min_lr: 0.003467 loss: 2.3718 (2.3054) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [122] [110/312] eta: 0:02:37 lr: 0.003467 min_lr: 0.003467 loss: 2.3066 (2.3041) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [122] [120/312] eta: 0:02:28 lr: 0.003466 min_lr: 0.003466 loss: 2.2498 (2.2909) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [122] [130/312] eta: 0:02:19 lr: 0.003466 min_lr: 0.003466 loss: 2.2798 (2.2862) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [122] [140/312] eta: 0:02:10 lr: 0.003466 min_lr: 0.003466 loss: 2.3144 (2.2891) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [122] [150/312] eta: 0:02:02 lr: 0.003465 min_lr: 0.003465 loss: 2.4089 (2.2973) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [122] [160/312] eta: 0:01:54 lr: 0.003465 min_lr: 0.003465 loss: 2.3522 (2.2856) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [122] [170/312] eta: 0:01:46 lr: 0.003465 min_lr: 0.003465 loss: 2.3522 (2.2919) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [122] [180/312] eta: 0:01:38 lr: 0.003464 min_lr: 0.003464 loss: 2.3836 (2.2920) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [122] [190/312] eta: 0:01:30 lr: 0.003464 min_lr: 0.003464 loss: 2.3260 (2.2927) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [122] [200/312] eta: 0:01:23 lr: 0.003464 min_lr: 0.003464 loss: 2.3155 (2.2920) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [122] [210/312] eta: 0:01:15 lr: 0.003463 min_lr: 0.003463 loss: 2.4326 (2.2962) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [122] [220/312] eta: 0:01:07 lr: 0.003463 min_lr: 0.003463 loss: 2.3682 (2.2982) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [122] [230/312] eta: 0:01:00 lr: 0.003463 min_lr: 0.003463 loss: 2.2618 (2.2919) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [122] [240/312] eta: 0:00:52 lr: 0.003462 min_lr: 0.003462 loss: 2.1255 (2.2869) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [122] [250/312] eta: 0:00:45 lr: 0.003462 min_lr: 0.003462 loss: 2.1305 (2.2801) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [122] [260/312] eta: 0:00:37 lr: 0.003462 min_lr: 0.003462 loss: 2.1651 (2.2746) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [122] [270/312] eta: 0:00:30 lr: 0.003461 min_lr: 0.003461 loss: 2.1283 (2.2703) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [122] [280/312] eta: 0:00:23 lr: 0.003461 min_lr: 0.003461 loss: 2.1140 (2.2669) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [122] [290/312] eta: 0:00:15 lr: 0.003461 min_lr: 0.003461 loss: 2.1559 (2.2663) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0008 max mem: 64948 Epoch: [122] [300/312] eta: 0:00:08 lr: 0.003460 min_lr: 0.003460 loss: 2.2533 (2.2662) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [122] [310/312] eta: 0:00:01 lr: 0.003460 min_lr: 0.003460 loss: 2.2938 (2.2670) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [122] [311/312] eta: 0:00:00 lr: 0.003460 min_lr: 0.003460 loss: 2.2938 (2.2652) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [122] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.003460 min_lr: 0.003460 loss: 2.2938 (2.2737) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7325 (0.7325) acc1: 81.7708 (81.7708) acc5: 94.7917 (94.7917) time: 4.6207 data: 4.4008 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0488 (0.9471) acc1: 73.9583 (75.2320) acc5: 92.9688 (92.5440) time: 0.6648 data: 0.4891 max mem: 64948 Test: Total time: 0:00:06 (0.6882 s / it) * Acc@1 76.372 Acc@5 93.008 loss 0.930 Accuracy of the model on the 50000 test images: 76.4% Max accuracy: 76.49% Test: [0/9] eta: 0:00:41 loss: 0.6828 (0.6828) acc1: 83.3333 (83.3333) acc5: 95.3125 (95.3125) time: 4.6123 data: 4.3959 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8479 (0.8534) acc1: 77.3438 (77.2160) acc5: 94.5312 (93.7600) time: 0.6723 data: 0.4971 max mem: 64948 Test: Total time: 0:00:06 (0.6799 s / it) * Acc@1 77.948 Acc@5 93.930 loss 0.840 Accuracy of the model EMA on 50000 test images: 77.9% Max EMA accuracy: 77.95% Epoch: [123] [ 0/312] eta: 0:49:26 lr: 0.003460 min_lr: 0.003460 loss: 1.9776 (1.9776) weight_decay: 0.0500 (0.0500) time: 9.5087 data: 8.3042 max mem: 64948 Epoch: [123] [ 10/312] eta: 0:07:47 lr: 0.003460 min_lr: 0.003460 loss: 2.4106 (2.1840) weight_decay: 0.0500 (0.0500) time: 1.5494 data: 0.7554 max mem: 64948 Epoch: [123] [ 20/312] eta: 0:05:33 lr: 0.003459 min_lr: 0.003459 loss: 2.4181 (2.2628) weight_decay: 0.0500 (0.0500) time: 0.7227 data: 0.0004 max mem: 64948 Epoch: [123] [ 30/312] eta: 0:04:41 lr: 0.003459 min_lr: 0.003459 loss: 2.4279 (2.1946) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [123] [ 40/312] eta: 0:04:11 lr: 0.003459 min_lr: 0.003459 loss: 2.0571 (2.1790) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [123] [ 50/312] eta: 0:03:50 lr: 0.003458 min_lr: 0.003458 loss: 2.3432 (2.2111) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [123] [ 60/312] eta: 0:03:34 lr: 0.003458 min_lr: 0.003458 loss: 2.4049 (2.2276) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [123] [ 70/312] eta: 0:03:20 lr: 0.003458 min_lr: 0.003458 loss: 2.2612 (2.2133) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [123] [ 80/312] eta: 0:03:08 lr: 0.003458 min_lr: 0.003458 loss: 2.2009 (2.2079) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [123] [ 90/312] eta: 0:02:57 lr: 0.003457 min_lr: 0.003457 loss: 2.2074 (2.2109) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [123] [100/312] eta: 0:02:47 lr: 0.003457 min_lr: 0.003457 loss: 2.3186 (2.2190) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [123] [110/312] eta: 0:02:37 lr: 0.003457 min_lr: 0.003457 loss: 2.2675 (2.2118) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [123] [120/312] eta: 0:02:28 lr: 0.003456 min_lr: 0.003456 loss: 2.2151 (2.2146) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [123] [130/312] eta: 0:02:19 lr: 0.003456 min_lr: 0.003456 loss: 2.4489 (2.2329) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [123] [140/312] eta: 0:02:11 lr: 0.003456 min_lr: 0.003456 loss: 2.4761 (2.2448) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [123] [150/312] eta: 0:02:02 lr: 0.003455 min_lr: 0.003455 loss: 2.4012 (2.2464) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [123] [160/312] eta: 0:01:54 lr: 0.003455 min_lr: 0.003455 loss: 2.2180 (2.2441) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [123] [170/312] eta: 0:01:46 lr: 0.003455 min_lr: 0.003455 loss: 2.0569 (2.2389) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [123] [180/312] eta: 0:01:38 lr: 0.003454 min_lr: 0.003454 loss: 2.2174 (2.2391) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [123] [190/312] eta: 0:01:30 lr: 0.003454 min_lr: 0.003454 loss: 2.2242 (2.2365) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [123] [200/312] eta: 0:01:23 lr: 0.003454 min_lr: 0.003454 loss: 2.3662 (2.2427) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [123] [210/312] eta: 0:01:15 lr: 0.003453 min_lr: 0.003453 loss: 2.4013 (2.2386) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [123] [220/312] eta: 0:01:07 lr: 0.003453 min_lr: 0.003453 loss: 2.2405 (2.2376) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [123] [230/312] eta: 0:01:00 lr: 0.003453 min_lr: 0.003453 loss: 2.2835 (2.2418) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [123] [240/312] eta: 0:00:52 lr: 0.003452 min_lr: 0.003452 loss: 2.3790 (2.2423) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [123] [250/312] eta: 0:00:45 lr: 0.003452 min_lr: 0.003452 loss: 2.3287 (2.2445) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [123] [260/312] eta: 0:00:37 lr: 0.003452 min_lr: 0.003452 loss: 2.3241 (2.2453) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [123] [270/312] eta: 0:00:30 lr: 0.003451 min_lr: 0.003451 loss: 2.2937 (2.2439) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [123] [280/312] eta: 0:00:23 lr: 0.003451 min_lr: 0.003451 loss: 2.2729 (2.2443) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [123] [290/312] eta: 0:00:15 lr: 0.003451 min_lr: 0.003451 loss: 2.1754 (2.2405) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [123] [300/312] eta: 0:00:08 lr: 0.003450 min_lr: 0.003450 loss: 2.2704 (2.2407) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [123] [310/312] eta: 0:00:01 lr: 0.003450 min_lr: 0.003450 loss: 2.2693 (2.2391) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [123] [311/312] eta: 0:00:00 lr: 0.003450 min_lr: 0.003450 loss: 2.2704 (2.2394) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [123] Total time: 0:03:46 (0.7271 s / it) Averaged stats: lr: 0.003450 min_lr: 0.003450 loss: 2.2704 (2.2692) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6659 (0.6659) acc1: 83.5938 (83.5938) acc5: 96.3542 (96.3542) time: 4.6705 data: 4.4570 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0305 (0.9429) acc1: 75.7812 (76.1600) acc5: 93.4896 (92.9920) time: 0.6702 data: 0.4953 max mem: 64948 Test: Total time: 0:00:06 (0.6961 s / it) * Acc@1 76.590 Acc@5 93.278 loss 0.926 Accuracy of the model on the 50000 test images: 76.6% Max accuracy: 76.59% Test: [0/9] eta: 0:00:41 loss: 0.6780 (0.6780) acc1: 82.8125 (82.8125) acc5: 95.3125 (95.3125) time: 4.6121 data: 4.4088 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8437 (0.8485) acc1: 77.3438 (77.2800) acc5: 94.5312 (93.8240) time: 0.6637 data: 0.4900 max mem: 64948 Test: Total time: 0:00:06 (0.6728 s / it) * Acc@1 78.030 Acc@5 93.994 loss 0.835 Accuracy of the model EMA on 50000 test images: 78.0% Max EMA accuracy: 78.03% Epoch: [124] [ 0/312] eta: 0:49:23 lr: 0.003450 min_lr: 0.003450 loss: 1.7149 (1.7149) weight_decay: 0.0500 (0.0500) time: 9.4977 data: 8.1311 max mem: 64948 Epoch: [124] [ 10/312] eta: 0:07:39 lr: 0.003450 min_lr: 0.003450 loss: 2.3402 (2.2228) weight_decay: 0.0500 (0.0500) time: 1.5223 data: 0.7396 max mem: 64948 Epoch: [124] [ 20/312] eta: 0:05:29 lr: 0.003449 min_lr: 0.003449 loss: 2.3402 (2.2397) weight_decay: 0.0500 (0.0500) time: 0.7098 data: 0.0004 max mem: 64948 Epoch: [124] [ 30/312] eta: 0:04:38 lr: 0.003449 min_lr: 0.003449 loss: 2.4420 (2.2513) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [124] [ 40/312] eta: 0:04:09 lr: 0.003449 min_lr: 0.003449 loss: 2.3363 (2.2313) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [124] [ 50/312] eta: 0:03:48 lr: 0.003448 min_lr: 0.003448 loss: 2.3808 (2.2854) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [124] [ 60/312] eta: 0:03:32 lr: 0.003448 min_lr: 0.003448 loss: 2.3808 (2.2809) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [124] [ 70/312] eta: 0:03:19 lr: 0.003448 min_lr: 0.003448 loss: 2.2227 (2.2696) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [124] [ 80/312] eta: 0:03:07 lr: 0.003447 min_lr: 0.003447 loss: 1.9984 (2.2362) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [124] [ 90/312] eta: 0:02:56 lr: 0.003447 min_lr: 0.003447 loss: 2.0473 (2.2344) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [124] [100/312] eta: 0:02:46 lr: 0.003447 min_lr: 0.003447 loss: 2.3491 (2.2310) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [124] [110/312] eta: 0:02:36 lr: 0.003446 min_lr: 0.003446 loss: 2.3936 (2.2437) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [124] [120/312] eta: 0:02:27 lr: 0.003446 min_lr: 0.003446 loss: 2.4125 (2.2392) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [124] [130/312] eta: 0:02:19 lr: 0.003446 min_lr: 0.003446 loss: 2.0699 (2.2259) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [124] [140/312] eta: 0:02:10 lr: 0.003446 min_lr: 0.003446 loss: 1.9398 (2.2089) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [124] [150/312] eta: 0:02:02 lr: 0.003445 min_lr: 0.003445 loss: 1.9449 (2.2053) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [124] [160/312] eta: 0:01:54 lr: 0.003445 min_lr: 0.003445 loss: 2.1734 (2.2002) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [124] [170/312] eta: 0:01:46 lr: 0.003445 min_lr: 0.003445 loss: 2.1839 (2.1935) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [124] [180/312] eta: 0:01:38 lr: 0.003444 min_lr: 0.003444 loss: 2.3984 (2.2060) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [124] [190/312] eta: 0:01:30 lr: 0.003444 min_lr: 0.003444 loss: 2.4141 (2.2143) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [124] [200/312] eta: 0:01:22 lr: 0.003444 min_lr: 0.003444 loss: 2.4141 (2.2218) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [124] [210/312] eta: 0:01:15 lr: 0.003443 min_lr: 0.003443 loss: 2.4170 (2.2296) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [124] [220/312] eta: 0:01:07 lr: 0.003443 min_lr: 0.003443 loss: 2.3667 (2.2367) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [124] [230/312] eta: 0:01:00 lr: 0.003443 min_lr: 0.003443 loss: 2.3639 (2.2401) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [124] [240/312] eta: 0:00:52 lr: 0.003442 min_lr: 0.003442 loss: 2.2563 (2.2348) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [124] [250/312] eta: 0:00:45 lr: 0.003442 min_lr: 0.003442 loss: 2.1474 (2.2337) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [124] [260/312] eta: 0:00:37 lr: 0.003442 min_lr: 0.003442 loss: 2.2278 (2.2327) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [124] [270/312] eta: 0:00:30 lr: 0.003441 min_lr: 0.003441 loss: 2.5433 (2.2450) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [124] [280/312] eta: 0:00:23 lr: 0.003441 min_lr: 0.003441 loss: 2.5296 (2.2409) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0009 max mem: 64948 Epoch: [124] [290/312] eta: 0:00:15 lr: 0.003441 min_lr: 0.003441 loss: 2.1064 (2.2407) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [124] [300/312] eta: 0:00:08 lr: 0.003440 min_lr: 0.003440 loss: 2.3998 (2.2491) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [124] [310/312] eta: 0:00:01 lr: 0.003440 min_lr: 0.003440 loss: 2.3785 (2.2492) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [124] [311/312] eta: 0:00:00 lr: 0.003440 min_lr: 0.003440 loss: 2.3254 (2.2480) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [124] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.003440 min_lr: 0.003440 loss: 2.3254 (2.2687) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6815 (0.6815) acc1: 83.3333 (83.3333) acc5: 94.7917 (94.7917) time: 4.6284 data: 4.4085 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9220 (0.9156) acc1: 77.3438 (75.7440) acc5: 91.9271 (93.1200) time: 0.6658 data: 0.4899 max mem: 64948 Test: Total time: 0:00:06 (0.6874 s / it) * Acc@1 76.538 Acc@5 93.242 loss 0.914 Accuracy of the model on the 50000 test images: 76.5% Max accuracy: 76.59% Test: [0/9] eta: 0:00:44 loss: 0.6734 (0.6734) acc1: 83.3333 (83.3333) acc5: 95.3125 (95.3125) time: 4.9792 data: 4.7613 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8389 (0.8436) acc1: 77.3438 (77.3760) acc5: 94.5312 (93.8240) time: 0.7046 data: 0.5291 max mem: 64948 Test: Total time: 0:00:06 (0.7126 s / it) * Acc@1 78.114 Acc@5 94.020 loss 0.831 Accuracy of the model EMA on 50000 test images: 78.1% Max EMA accuracy: 78.11% Epoch: [125] [ 0/312] eta: 0:53:04 lr: 0.003440 min_lr: 0.003440 loss: 2.5124 (2.5124) weight_decay: 0.0500 (0.0500) time: 10.2064 data: 9.4266 max mem: 64948 Epoch: [125] [ 10/312] eta: 0:07:54 lr: 0.003440 min_lr: 0.003440 loss: 2.0837 (2.1611) weight_decay: 0.0500 (0.0500) time: 1.5708 data: 0.8573 max mem: 64948 Epoch: [125] [ 20/312] eta: 0:05:36 lr: 0.003439 min_lr: 0.003439 loss: 2.0507 (2.1251) weight_decay: 0.0500 (0.0500) time: 0.7012 data: 0.0004 max mem: 64948 Epoch: [125] [ 30/312] eta: 0:04:43 lr: 0.003439 min_lr: 0.003439 loss: 2.2760 (2.1750) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [125] [ 40/312] eta: 0:04:13 lr: 0.003439 min_lr: 0.003439 loss: 2.2913 (2.1877) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [125] [ 50/312] eta: 0:03:51 lr: 0.003438 min_lr: 0.003438 loss: 2.5236 (2.2481) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [125] [ 60/312] eta: 0:03:34 lr: 0.003438 min_lr: 0.003438 loss: 2.3976 (2.2264) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [125] [ 70/312] eta: 0:03:21 lr: 0.003438 min_lr: 0.003438 loss: 2.3194 (2.2421) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [125] [ 80/312] eta: 0:03:08 lr: 0.003437 min_lr: 0.003437 loss: 2.3969 (2.2617) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [125] [ 90/312] eta: 0:02:57 lr: 0.003437 min_lr: 0.003437 loss: 2.4709 (2.2719) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [125] [100/312] eta: 0:02:47 lr: 0.003437 min_lr: 0.003437 loss: 2.3702 (2.2599) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [125] [110/312] eta: 0:02:38 lr: 0.003436 min_lr: 0.003436 loss: 2.3885 (2.2758) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [125] [120/312] eta: 0:02:28 lr: 0.003436 min_lr: 0.003436 loss: 2.3987 (2.2765) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [125] [130/312] eta: 0:02:19 lr: 0.003436 min_lr: 0.003436 loss: 2.3470 (2.2711) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [125] [140/312] eta: 0:02:11 lr: 0.003435 min_lr: 0.003435 loss: 2.2714 (2.2686) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [125] [150/312] eta: 0:02:02 lr: 0.003435 min_lr: 0.003435 loss: 2.1620 (2.2596) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [125] [160/312] eta: 0:01:54 lr: 0.003435 min_lr: 0.003435 loss: 2.3321 (2.2568) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [125] [170/312] eta: 0:01:46 lr: 0.003434 min_lr: 0.003434 loss: 2.2787 (2.2534) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [125] [180/312] eta: 0:01:38 lr: 0.003434 min_lr: 0.003434 loss: 2.3976 (2.2702) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [125] [190/312] eta: 0:01:30 lr: 0.003434 min_lr: 0.003434 loss: 2.4032 (2.2688) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [125] [200/312] eta: 0:01:23 lr: 0.003433 min_lr: 0.003433 loss: 2.2994 (2.2762) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [125] [210/312] eta: 0:01:15 lr: 0.003433 min_lr: 0.003433 loss: 2.3589 (2.2730) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [125] [220/312] eta: 0:01:07 lr: 0.003433 min_lr: 0.003433 loss: 2.3971 (2.2749) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [125] [230/312] eta: 0:01:00 lr: 0.003432 min_lr: 0.003432 loss: 2.4071 (2.2727) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [125] [240/312] eta: 0:00:52 lr: 0.003432 min_lr: 0.003432 loss: 2.2943 (2.2711) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [125] [250/312] eta: 0:00:45 lr: 0.003432 min_lr: 0.003432 loss: 2.2967 (2.2710) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [125] [260/312] eta: 0:00:38 lr: 0.003431 min_lr: 0.003431 loss: 2.4663 (2.2759) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [125] [270/312] eta: 0:00:30 lr: 0.003431 min_lr: 0.003431 loss: 2.2959 (2.2730) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [125] [280/312] eta: 0:00:23 lr: 0.003431 min_lr: 0.003431 loss: 2.2148 (2.2691) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0009 max mem: 64948 Epoch: [125] [290/312] eta: 0:00:16 lr: 0.003430 min_lr: 0.003430 loss: 2.1588 (2.2652) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [125] [300/312] eta: 0:00:08 lr: 0.003430 min_lr: 0.003430 loss: 2.1427 (2.2606) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [125] [310/312] eta: 0:00:01 lr: 0.003430 min_lr: 0.003430 loss: 2.1831 (2.2658) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [125] [311/312] eta: 0:00:00 lr: 0.003430 min_lr: 0.003430 loss: 2.1831 (2.2639) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [125] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.003430 min_lr: 0.003430 loss: 2.1831 (2.2766) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.7901 (0.7901) acc1: 82.2917 (82.2917) acc5: 93.4896 (93.4896) time: 4.6907 data: 4.4780 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9812 (0.9505) acc1: 76.0417 (76.1280) acc5: 93.2292 (92.8000) time: 0.6724 data: 0.4976 max mem: 64948 Test: Total time: 0:00:06 (0.6979 s / it) * Acc@1 76.194 Acc@5 93.070 loss 0.938 Accuracy of the model on the 50000 test images: 76.2% Max accuracy: 76.59% Test: [0/9] eta: 0:00:44 loss: 0.6687 (0.6687) acc1: 83.5938 (83.5938) acc5: 95.3125 (95.3125) time: 4.9233 data: 4.7072 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8339 (0.8393) acc1: 77.3438 (77.4400) acc5: 94.7917 (93.8880) time: 0.7049 data: 0.5231 max mem: 64948 Test: Total time: 0:00:06 (0.7228 s / it) * Acc@1 78.190 Acc@5 94.066 loss 0.826 Accuracy of the model EMA on 50000 test images: 78.2% Max EMA accuracy: 78.19% Epoch: [126] [ 0/312] eta: 0:49:48 lr: 0.003430 min_lr: 0.003430 loss: 2.0895 (2.0895) weight_decay: 0.0500 (0.0500) time: 9.5795 data: 8.2083 max mem: 64948 Epoch: [126] [ 10/312] eta: 0:07:44 lr: 0.003429 min_lr: 0.003429 loss: 2.2528 (2.3212) weight_decay: 0.0500 (0.0500) time: 1.5394 data: 0.7466 max mem: 64948 Epoch: [126] [ 20/312] eta: 0:05:31 lr: 0.003429 min_lr: 0.003429 loss: 2.2332 (2.1851) weight_decay: 0.0500 (0.0500) time: 0.7136 data: 0.0004 max mem: 64948 Epoch: [126] [ 30/312] eta: 0:04:40 lr: 0.003429 min_lr: 0.003429 loss: 1.9686 (2.1761) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [126] [ 40/312] eta: 0:04:10 lr: 0.003428 min_lr: 0.003428 loss: 2.2211 (2.2235) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [126] [ 50/312] eta: 0:03:49 lr: 0.003428 min_lr: 0.003428 loss: 2.4004 (2.2393) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [126] [ 60/312] eta: 0:03:33 lr: 0.003428 min_lr: 0.003428 loss: 2.2911 (2.2307) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [126] [ 70/312] eta: 0:03:19 lr: 0.003427 min_lr: 0.003427 loss: 2.3564 (2.2342) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [126] [ 80/312] eta: 0:03:07 lr: 0.003427 min_lr: 0.003427 loss: 2.3394 (2.2265) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [126] [ 90/312] eta: 0:02:56 lr: 0.003427 min_lr: 0.003427 loss: 2.3690 (2.2578) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [126] [100/312] eta: 0:02:46 lr: 0.003426 min_lr: 0.003426 loss: 2.4137 (2.2590) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [126] [110/312] eta: 0:02:37 lr: 0.003426 min_lr: 0.003426 loss: 2.4137 (2.2702) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [126] [120/312] eta: 0:02:28 lr: 0.003426 min_lr: 0.003426 loss: 2.3953 (2.2680) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [126] [130/312] eta: 0:02:19 lr: 0.003426 min_lr: 0.003426 loss: 2.2005 (2.2584) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [126] [140/312] eta: 0:02:10 lr: 0.003425 min_lr: 0.003425 loss: 2.2521 (2.2635) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [126] [150/312] eta: 0:02:02 lr: 0.003425 min_lr: 0.003425 loss: 2.3436 (2.2643) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [126] [160/312] eta: 0:01:54 lr: 0.003425 min_lr: 0.003425 loss: 2.1856 (2.2636) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [126] [170/312] eta: 0:01:46 lr: 0.003424 min_lr: 0.003424 loss: 2.4122 (2.2700) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [126] [180/312] eta: 0:01:38 lr: 0.003424 min_lr: 0.003424 loss: 2.3487 (2.2641) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0003 max mem: 64948 Epoch: [126] [190/312] eta: 0:01:30 lr: 0.003424 min_lr: 0.003424 loss: 2.2954 (2.2699) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [126] [200/312] eta: 0:01:22 lr: 0.003423 min_lr: 0.003423 loss: 2.4195 (2.2726) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [126] [210/312] eta: 0:01:15 lr: 0.003423 min_lr: 0.003423 loss: 2.4957 (2.2841) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [126] [220/312] eta: 0:01:07 lr: 0.003423 min_lr: 0.003423 loss: 2.4957 (2.2824) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [126] [230/312] eta: 0:01:00 lr: 0.003422 min_lr: 0.003422 loss: 2.1522 (2.2751) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [126] [240/312] eta: 0:00:52 lr: 0.003422 min_lr: 0.003422 loss: 2.1096 (2.2686) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [126] [250/312] eta: 0:00:45 lr: 0.003422 min_lr: 0.003422 loss: 2.3409 (2.2658) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [126] [260/312] eta: 0:00:37 lr: 0.003421 min_lr: 0.003421 loss: 2.3965 (2.2670) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [126] [270/312] eta: 0:00:30 lr: 0.003421 min_lr: 0.003421 loss: 2.4267 (2.2748) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [126] [280/312] eta: 0:00:23 lr: 0.003421 min_lr: 0.003421 loss: 2.4873 (2.2794) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [126] [290/312] eta: 0:00:15 lr: 0.003420 min_lr: 0.003420 loss: 2.3994 (2.2804) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [126] [300/312] eta: 0:00:08 lr: 0.003420 min_lr: 0.003420 loss: 2.3140 (2.2773) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [126] [310/312] eta: 0:00:01 lr: 0.003420 min_lr: 0.003420 loss: 2.3374 (2.2797) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [126] [311/312] eta: 0:00:00 lr: 0.003420 min_lr: 0.003420 loss: 2.3880 (2.2802) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [126] Total time: 0:03:46 (0.7272 s / it) Averaged stats: lr: 0.003420 min_lr: 0.003420 loss: 2.3880 (2.2645) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7853 (0.7853) acc1: 81.2500 (81.2500) acc5: 94.5312 (94.5312) time: 4.5835 data: 4.3735 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0342 (0.9897) acc1: 76.8229 (75.2640) acc5: 93.4896 (92.7360) time: 0.6610 data: 0.4860 max mem: 64948 Test: Total time: 0:00:06 (0.6865 s / it) * Acc@1 75.722 Acc@5 92.768 loss 0.965 Accuracy of the model on the 50000 test images: 75.7% Max accuracy: 76.59% Test: [0/9] eta: 0:00:44 loss: 0.6646 (0.6646) acc1: 83.5938 (83.5938) acc5: 95.3125 (95.3125) time: 4.9842 data: 4.7665 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8296 (0.8350) acc1: 77.6042 (77.5360) acc5: 94.7917 (93.9520) time: 0.7051 data: 0.5297 max mem: 64948 Test: Total time: 0:00:06 (0.7129 s / it) * Acc@1 78.264 Acc@5 94.110 loss 0.822 Accuracy of the model EMA on 50000 test images: 78.3% Max EMA accuracy: 78.26% Epoch: [127] [ 0/312] eta: 0:54:20 lr: 0.003420 min_lr: 0.003420 loss: 2.2309 (2.2309) weight_decay: 0.0500 (0.0500) time: 10.4497 data: 9.7342 max mem: 64948 Epoch: [127] [ 10/312] eta: 0:08:03 lr: 0.003419 min_lr: 0.003419 loss: 2.2309 (2.2334) weight_decay: 0.0500 (0.0500) time: 1.6020 data: 0.8852 max mem: 64948 Epoch: [127] [ 20/312] eta: 0:05:41 lr: 0.003419 min_lr: 0.003419 loss: 2.4147 (2.2824) weight_decay: 0.0500 (0.0500) time: 0.7054 data: 0.0003 max mem: 64948 Epoch: [127] [ 30/312] eta: 0:04:46 lr: 0.003419 min_lr: 0.003419 loss: 2.4147 (2.2812) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [127] [ 40/312] eta: 0:04:15 lr: 0.003418 min_lr: 0.003418 loss: 2.1279 (2.2460) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [127] [ 50/312] eta: 0:03:53 lr: 0.003418 min_lr: 0.003418 loss: 2.3404 (2.2574) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [127] [ 60/312] eta: 0:03:36 lr: 0.003418 min_lr: 0.003418 loss: 2.4222 (2.2731) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [127] [ 70/312] eta: 0:03:22 lr: 0.003417 min_lr: 0.003417 loss: 2.3805 (2.2791) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [127] [ 80/312] eta: 0:03:09 lr: 0.003417 min_lr: 0.003417 loss: 2.3652 (2.2683) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [127] [ 90/312] eta: 0:02:58 lr: 0.003417 min_lr: 0.003417 loss: 2.2588 (2.2625) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [127] [100/312] eta: 0:02:48 lr: 0.003416 min_lr: 0.003416 loss: 2.4092 (2.2792) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [127] [110/312] eta: 0:02:38 lr: 0.003416 min_lr: 0.003416 loss: 2.4368 (2.2804) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [127] [120/312] eta: 0:02:29 lr: 0.003416 min_lr: 0.003416 loss: 2.2883 (2.2732) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [127] [130/312] eta: 0:02:20 lr: 0.003415 min_lr: 0.003415 loss: 2.3087 (2.2719) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [127] [140/312] eta: 0:02:11 lr: 0.003415 min_lr: 0.003415 loss: 2.3353 (2.2709) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [127] [150/312] eta: 0:02:03 lr: 0.003415 min_lr: 0.003415 loss: 2.2706 (2.2681) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [127] [160/312] eta: 0:01:55 lr: 0.003414 min_lr: 0.003414 loss: 2.2167 (2.2582) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [127] [170/312] eta: 0:01:46 lr: 0.003414 min_lr: 0.003414 loss: 2.1567 (2.2520) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [127] [180/312] eta: 0:01:38 lr: 0.003414 min_lr: 0.003414 loss: 2.2275 (2.2497) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [127] [190/312] eta: 0:01:31 lr: 0.003413 min_lr: 0.003413 loss: 2.4052 (2.2632) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [127] [200/312] eta: 0:01:23 lr: 0.003413 min_lr: 0.003413 loss: 2.3811 (2.2604) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [127] [210/312] eta: 0:01:15 lr: 0.003413 min_lr: 0.003413 loss: 2.3253 (2.2617) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [127] [220/312] eta: 0:01:08 lr: 0.003412 min_lr: 0.003412 loss: 2.3253 (2.2529) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [127] [230/312] eta: 0:01:00 lr: 0.003412 min_lr: 0.003412 loss: 2.2853 (2.2578) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [127] [240/312] eta: 0:00:52 lr: 0.003412 min_lr: 0.003412 loss: 2.2853 (2.2583) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [127] [250/312] eta: 0:00:45 lr: 0.003411 min_lr: 0.003411 loss: 2.3839 (2.2668) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [127] [260/312] eta: 0:00:38 lr: 0.003411 min_lr: 0.003411 loss: 2.3711 (2.2584) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [127] [270/312] eta: 0:00:30 lr: 0.003411 min_lr: 0.003411 loss: 2.0098 (2.2548) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [127] [280/312] eta: 0:00:23 lr: 0.003410 min_lr: 0.003410 loss: 2.0203 (2.2542) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [127] [290/312] eta: 0:00:16 lr: 0.003410 min_lr: 0.003410 loss: 2.3528 (2.2546) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [127] [300/312] eta: 0:00:08 lr: 0.003410 min_lr: 0.003410 loss: 2.2702 (2.2557) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [127] [310/312] eta: 0:00:01 lr: 0.003409 min_lr: 0.003409 loss: 2.2579 (2.2524) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [127] [311/312] eta: 0:00:00 lr: 0.003409 min_lr: 0.003409 loss: 2.2579 (2.2510) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [127] Total time: 0:03:47 (0.7291 s / it) Averaged stats: lr: 0.003409 min_lr: 0.003409 loss: 2.2579 (2.2595) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7518 (0.7518) acc1: 82.2917 (82.2917) acc5: 94.2708 (94.2708) time: 4.5588 data: 4.3390 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9585 (0.9155) acc1: 78.1250 (76.4800) acc5: 92.7083 (93.1520) time: 0.6580 data: 0.4822 max mem: 64948 Test: Total time: 0:00:06 (0.6830 s / it) * Acc@1 76.760 Acc@5 93.546 loss 0.903 Accuracy of the model on the 50000 test images: 76.8% Max accuracy: 76.76% Test: [0/9] eta: 0:00:42 loss: 0.6613 (0.6613) acc1: 83.8542 (83.8542) acc5: 95.3125 (95.3125) time: 4.7559 data: 4.5381 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8255 (0.8311) acc1: 77.8646 (77.6960) acc5: 94.7917 (93.9200) time: 0.6799 data: 0.5043 max mem: 64948 Test: Total time: 0:00:06 (0.6878 s / it) * Acc@1 78.334 Acc@5 94.128 loss 0.818 Accuracy of the model EMA on 50000 test images: 78.3% Max EMA accuracy: 78.33% Epoch: [128] [ 0/312] eta: 0:51:24 lr: 0.003409 min_lr: 0.003409 loss: 2.4906 (2.4906) weight_decay: 0.0500 (0.0500) time: 9.8852 data: 9.0860 max mem: 64948 Epoch: [128] [ 10/312] eta: 0:07:48 lr: 0.003409 min_lr: 0.003409 loss: 2.4912 (2.4380) weight_decay: 0.0500 (0.0500) time: 1.5526 data: 0.8263 max mem: 64948 Epoch: [128] [ 20/312] eta: 0:05:35 lr: 0.003409 min_lr: 0.003409 loss: 2.3377 (2.3536) weight_decay: 0.0500 (0.0500) time: 0.7132 data: 0.0003 max mem: 64948 Epoch: [128] [ 30/312] eta: 0:04:43 lr: 0.003408 min_lr: 0.003408 loss: 2.1705 (2.2921) weight_decay: 0.0500 (0.0500) time: 0.7065 data: 0.0003 max mem: 64948 Epoch: [128] [ 40/312] eta: 0:04:13 lr: 0.003408 min_lr: 0.003408 loss: 2.1705 (2.2823) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0003 max mem: 64948 Epoch: [128] [ 50/312] eta: 0:03:51 lr: 0.003408 min_lr: 0.003408 loss: 2.3182 (2.3065) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [128] [ 60/312] eta: 0:03:35 lr: 0.003407 min_lr: 0.003407 loss: 2.4131 (2.3165) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [128] [ 70/312] eta: 0:03:21 lr: 0.003407 min_lr: 0.003407 loss: 2.4131 (2.3060) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [128] [ 80/312] eta: 0:03:08 lr: 0.003407 min_lr: 0.003407 loss: 2.1552 (2.2918) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [128] [ 90/312] eta: 0:02:57 lr: 0.003406 min_lr: 0.003406 loss: 1.9935 (2.2631) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [128] [100/312] eta: 0:02:47 lr: 0.003406 min_lr: 0.003406 loss: 1.9971 (2.2632) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [128] [110/312] eta: 0:02:38 lr: 0.003406 min_lr: 0.003406 loss: 2.2570 (2.2507) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [128] [120/312] eta: 0:02:28 lr: 0.003405 min_lr: 0.003405 loss: 2.1256 (2.2474) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [128] [130/312] eta: 0:02:19 lr: 0.003405 min_lr: 0.003405 loss: 2.2047 (2.2548) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [128] [140/312] eta: 0:02:11 lr: 0.003405 min_lr: 0.003405 loss: 2.4671 (2.2601) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [128] [150/312] eta: 0:02:02 lr: 0.003404 min_lr: 0.003404 loss: 2.3969 (2.2546) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [128] [160/312] eta: 0:01:54 lr: 0.003404 min_lr: 0.003404 loss: 2.3872 (2.2670) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [128] [170/312] eta: 0:01:46 lr: 0.003404 min_lr: 0.003404 loss: 2.4755 (2.2620) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [128] [180/312] eta: 0:01:38 lr: 0.003403 min_lr: 0.003403 loss: 2.3603 (2.2659) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [128] [190/312] eta: 0:01:30 lr: 0.003403 min_lr: 0.003403 loss: 2.2622 (2.2591) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [128] [200/312] eta: 0:01:23 lr: 0.003403 min_lr: 0.003403 loss: 2.1869 (2.2588) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [128] [210/312] eta: 0:01:15 lr: 0.003402 min_lr: 0.003402 loss: 2.3033 (2.2665) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [128] [220/312] eta: 0:01:07 lr: 0.003402 min_lr: 0.003402 loss: 2.3033 (2.2654) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [128] [230/312] eta: 0:01:00 lr: 0.003402 min_lr: 0.003402 loss: 2.3145 (2.2655) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [128] [240/312] eta: 0:00:52 lr: 0.003401 min_lr: 0.003401 loss: 2.3451 (2.2661) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [128] [250/312] eta: 0:00:45 lr: 0.003401 min_lr: 0.003401 loss: 2.2321 (2.2630) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [128] [260/312] eta: 0:00:38 lr: 0.003401 min_lr: 0.003401 loss: 2.2379 (2.2631) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [128] [270/312] eta: 0:00:30 lr: 0.003400 min_lr: 0.003400 loss: 2.2709 (2.2663) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [128] [280/312] eta: 0:00:23 lr: 0.003400 min_lr: 0.003400 loss: 2.2663 (2.2636) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [128] [290/312] eta: 0:00:16 lr: 0.003400 min_lr: 0.003400 loss: 2.2298 (2.2629) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [128] [300/312] eta: 0:00:08 lr: 0.003399 min_lr: 0.003399 loss: 2.2298 (2.2592) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [128] [310/312] eta: 0:00:01 lr: 0.003399 min_lr: 0.003399 loss: 2.1076 (2.2549) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [128] [311/312] eta: 0:00:00 lr: 0.003399 min_lr: 0.003399 loss: 2.1076 (2.2558) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [128] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.003399 min_lr: 0.003399 loss: 2.1076 (2.2623) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7853 (0.7853) acc1: 79.9479 (79.9479) acc5: 95.5729 (95.5729) time: 4.5184 data: 4.3139 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0168 (0.9666) acc1: 73.4375 (75.0080) acc5: 94.2708 (93.1200) time: 0.6533 data: 0.4794 max mem: 64948 Test: Total time: 0:00:06 (0.6826 s / it) * Acc@1 76.218 Acc@5 92.976 loss 0.930 Accuracy of the model on the 50000 test images: 76.2% Max accuracy: 76.76% Test: [0/9] eta: 0:00:41 loss: 0.6575 (0.6575) acc1: 83.8542 (83.8542) acc5: 95.3125 (95.3125) time: 4.5662 data: 4.3482 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8214 (0.8272) acc1: 78.3854 (77.8240) acc5: 94.5312 (93.9520) time: 0.6855 data: 0.5101 max mem: 64948 Test: Total time: 0:00:06 (0.6955 s / it) * Acc@1 78.410 Acc@5 94.166 loss 0.815 Accuracy of the model EMA on 50000 test images: 78.4% Max EMA accuracy: 78.41% Epoch: [129] [ 0/312] eta: 0:49:31 lr: 0.003399 min_lr: 0.003399 loss: 2.6465 (2.6465) weight_decay: 0.0500 (0.0500) time: 9.5251 data: 7.4067 max mem: 64948 Epoch: [129] [ 10/312] eta: 0:07:50 lr: 0.003398 min_lr: 0.003398 loss: 2.2758 (2.3130) weight_decay: 0.0500 (0.0500) time: 1.5594 data: 0.7188 max mem: 64948 Epoch: [129] [ 20/312] eta: 0:05:34 lr: 0.003398 min_lr: 0.003398 loss: 2.3961 (2.3791) weight_decay: 0.0500 (0.0500) time: 0.7281 data: 0.0252 max mem: 64948 Epoch: [129] [ 30/312] eta: 0:04:42 lr: 0.003398 min_lr: 0.003398 loss: 2.3961 (2.3690) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [129] [ 40/312] eta: 0:04:12 lr: 0.003397 min_lr: 0.003397 loss: 2.3265 (2.3877) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [129] [ 50/312] eta: 0:03:50 lr: 0.003397 min_lr: 0.003397 loss: 2.3736 (2.3921) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [129] [ 60/312] eta: 0:03:34 lr: 0.003397 min_lr: 0.003397 loss: 2.3943 (2.3911) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [129] [ 70/312] eta: 0:03:20 lr: 0.003396 min_lr: 0.003396 loss: 2.3688 (2.3659) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [129] [ 80/312] eta: 0:03:08 lr: 0.003396 min_lr: 0.003396 loss: 2.2585 (2.3402) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [129] [ 90/312] eta: 0:02:57 lr: 0.003396 min_lr: 0.003396 loss: 2.2737 (2.3219) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [129] [100/312] eta: 0:02:47 lr: 0.003395 min_lr: 0.003395 loss: 2.2528 (2.3053) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [129] [110/312] eta: 0:02:37 lr: 0.003395 min_lr: 0.003395 loss: 2.1850 (2.2980) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [129] [120/312] eta: 0:02:28 lr: 0.003395 min_lr: 0.003395 loss: 2.2073 (2.2942) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [129] [130/312] eta: 0:02:19 lr: 0.003394 min_lr: 0.003394 loss: 2.2073 (2.2953) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [129] [140/312] eta: 0:02:11 lr: 0.003394 min_lr: 0.003394 loss: 2.3962 (2.3042) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [129] [150/312] eta: 0:02:02 lr: 0.003394 min_lr: 0.003394 loss: 2.3962 (2.3001) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [129] [160/312] eta: 0:01:54 lr: 0.003393 min_lr: 0.003393 loss: 2.0454 (2.2834) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [129] [170/312] eta: 0:01:46 lr: 0.003393 min_lr: 0.003393 loss: 2.0758 (2.2785) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [129] [180/312] eta: 0:01:38 lr: 0.003393 min_lr: 0.003393 loss: 2.2331 (2.2803) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [129] [190/312] eta: 0:01:30 lr: 0.003392 min_lr: 0.003392 loss: 2.2400 (2.2744) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [129] [200/312] eta: 0:01:23 lr: 0.003392 min_lr: 0.003392 loss: 2.2567 (2.2772) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [129] [210/312] eta: 0:01:15 lr: 0.003392 min_lr: 0.003392 loss: 2.4226 (2.2835) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [129] [220/312] eta: 0:01:07 lr: 0.003391 min_lr: 0.003391 loss: 2.4632 (2.2914) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [129] [230/312] eta: 0:01:00 lr: 0.003391 min_lr: 0.003391 loss: 2.4285 (2.2926) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [129] [240/312] eta: 0:00:52 lr: 0.003391 min_lr: 0.003391 loss: 2.3307 (2.2963) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [129] [250/312] eta: 0:00:45 lr: 0.003390 min_lr: 0.003390 loss: 2.2894 (2.2937) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [129] [260/312] eta: 0:00:38 lr: 0.003390 min_lr: 0.003390 loss: 2.2685 (2.2933) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [129] [270/312] eta: 0:00:30 lr: 0.003390 min_lr: 0.003390 loss: 2.4270 (2.2952) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [129] [280/312] eta: 0:00:23 lr: 0.003389 min_lr: 0.003389 loss: 2.3945 (2.2962) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [129] [290/312] eta: 0:00:15 lr: 0.003389 min_lr: 0.003389 loss: 2.2126 (2.2919) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [129] [300/312] eta: 0:00:08 lr: 0.003389 min_lr: 0.003389 loss: 2.1692 (2.2848) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [129] [310/312] eta: 0:00:01 lr: 0.003388 min_lr: 0.003388 loss: 1.9643 (2.2776) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [129] [311/312] eta: 0:00:00 lr: 0.003388 min_lr: 0.003388 loss: 2.0084 (2.2768) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [129] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.003388 min_lr: 0.003388 loss: 2.0084 (2.2607) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7159 (0.7159) acc1: 82.2917 (82.2917) acc5: 94.0104 (94.0104) time: 4.6340 data: 4.4135 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9367 (0.9242) acc1: 75.2604 (75.5840) acc5: 94.0104 (93.3120) time: 0.6663 data: 0.4905 max mem: 64948 Test: Total time: 0:00:06 (0.6919 s / it) * Acc@1 76.650 Acc@5 93.362 loss 0.906 Accuracy of the model on the 50000 test images: 76.7% Max accuracy: 76.76% Test: [0/9] eta: 0:00:42 loss: 0.6542 (0.6542) acc1: 84.1146 (84.1146) acc5: 95.3125 (95.3125) time: 4.7305 data: 4.5122 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8174 (0.8235) acc1: 78.3854 (78.0800) acc5: 94.5312 (93.9520) time: 0.6777 data: 0.5015 max mem: 64948 Test: Total time: 0:00:06 (0.6933 s / it) * Acc@1 78.470 Acc@5 94.202 loss 0.811 Accuracy of the model EMA on 50000 test images: 78.5% Max EMA accuracy: 78.47% Epoch: [130] [ 0/312] eta: 0:51:55 lr: 0.003388 min_lr: 0.003388 loss: 2.0647 (2.0647) weight_decay: 0.0500 (0.0500) time: 9.9854 data: 9.1968 max mem: 64948 Epoch: [130] [ 10/312] eta: 0:08:00 lr: 0.003388 min_lr: 0.003388 loss: 2.4004 (2.3828) weight_decay: 0.0500 (0.0500) time: 1.5912 data: 0.8365 max mem: 64948 Epoch: [130] [ 20/312] eta: 0:05:39 lr: 0.003388 min_lr: 0.003388 loss: 2.3363 (2.3302) weight_decay: 0.0500 (0.0500) time: 0.7216 data: 0.0004 max mem: 64948 Epoch: [130] [ 30/312] eta: 0:04:45 lr: 0.003387 min_lr: 0.003387 loss: 2.2967 (2.2809) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0003 max mem: 64948 Epoch: [130] [ 40/312] eta: 0:04:13 lr: 0.003387 min_lr: 0.003387 loss: 2.2967 (2.2762) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [130] [ 50/312] eta: 0:03:52 lr: 0.003387 min_lr: 0.003387 loss: 2.3311 (2.2635) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [130] [ 60/312] eta: 0:03:35 lr: 0.003386 min_lr: 0.003386 loss: 2.3311 (2.2656) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [130] [ 70/312] eta: 0:03:21 lr: 0.003386 min_lr: 0.003386 loss: 2.2879 (2.2591) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [130] [ 80/312] eta: 0:03:09 lr: 0.003386 min_lr: 0.003386 loss: 2.2879 (2.2606) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [130] [ 90/312] eta: 0:02:58 lr: 0.003385 min_lr: 0.003385 loss: 2.2792 (2.2585) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [130] [100/312] eta: 0:02:47 lr: 0.003385 min_lr: 0.003385 loss: 2.2486 (2.2539) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [130] [110/312] eta: 0:02:38 lr: 0.003385 min_lr: 0.003385 loss: 2.3687 (2.2681) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [130] [120/312] eta: 0:02:29 lr: 0.003384 min_lr: 0.003384 loss: 2.3431 (2.2610) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [130] [130/312] eta: 0:02:20 lr: 0.003384 min_lr: 0.003384 loss: 2.1067 (2.2487) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [130] [140/312] eta: 0:02:11 lr: 0.003384 min_lr: 0.003384 loss: 2.1067 (2.2385) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [130] [150/312] eta: 0:02:03 lr: 0.003383 min_lr: 0.003383 loss: 2.3207 (2.2400) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [130] [160/312] eta: 0:01:54 lr: 0.003383 min_lr: 0.003383 loss: 2.3770 (2.2459) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [130] [170/312] eta: 0:01:46 lr: 0.003383 min_lr: 0.003383 loss: 2.3770 (2.2512) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [130] [180/312] eta: 0:01:38 lr: 0.003382 min_lr: 0.003382 loss: 2.2261 (2.2460) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [130] [190/312] eta: 0:01:31 lr: 0.003382 min_lr: 0.003382 loss: 2.2865 (2.2484) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [130] [200/312] eta: 0:01:23 lr: 0.003382 min_lr: 0.003382 loss: 2.2792 (2.2434) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [130] [210/312] eta: 0:01:15 lr: 0.003381 min_lr: 0.003381 loss: 2.1274 (2.2386) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [130] [220/312] eta: 0:01:08 lr: 0.003381 min_lr: 0.003381 loss: 2.2208 (2.2369) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [130] [230/312] eta: 0:01:00 lr: 0.003381 min_lr: 0.003381 loss: 2.1792 (2.2314) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [130] [240/312] eta: 0:00:52 lr: 0.003380 min_lr: 0.003380 loss: 2.1764 (2.2307) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [130] [250/312] eta: 0:00:45 lr: 0.003380 min_lr: 0.003380 loss: 2.3437 (2.2334) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [130] [260/312] eta: 0:00:38 lr: 0.003380 min_lr: 0.003380 loss: 2.3423 (2.2331) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [130] [270/312] eta: 0:00:30 lr: 0.003379 min_lr: 0.003379 loss: 2.1687 (2.2256) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [130] [280/312] eta: 0:00:23 lr: 0.003379 min_lr: 0.003379 loss: 1.9617 (2.2203) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0011 max mem: 64948 Epoch: [130] [290/312] eta: 0:00:16 lr: 0.003378 min_lr: 0.003378 loss: 2.2553 (2.2201) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0009 max mem: 64948 Epoch: [130] [300/312] eta: 0:00:08 lr: 0.003378 min_lr: 0.003378 loss: 2.2746 (2.2241) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [130] [310/312] eta: 0:00:01 lr: 0.003378 min_lr: 0.003378 loss: 2.4230 (2.2292) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [130] [311/312] eta: 0:00:00 lr: 0.003378 min_lr: 0.003378 loss: 2.4260 (2.2302) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [130] Total time: 0:03:47 (0.7286 s / it) Averaged stats: lr: 0.003378 min_lr: 0.003378 loss: 2.4260 (2.2548) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.7049 (0.7049) acc1: 81.2500 (81.2500) acc5: 95.8333 (95.8333) time: 4.6999 data: 4.4871 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9438 (0.9225) acc1: 77.8646 (76.1600) acc5: 94.0104 (93.4080) time: 0.6735 data: 0.4986 max mem: 64948 Test: Total time: 0:00:06 (0.7034 s / it) * Acc@1 76.660 Acc@5 93.284 loss 0.918 Accuracy of the model on the 50000 test images: 76.7% Max accuracy: 76.76% Test: [0/9] eta: 0:00:43 loss: 0.6508 (0.6508) acc1: 84.1146 (84.1146) acc5: 95.3125 (95.3125) time: 4.7779 data: 4.5667 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8144 (0.8197) acc1: 78.3854 (78.1440) acc5: 94.5312 (94.0160) time: 0.6890 data: 0.5075 max mem: 64948 Test: Total time: 0:00:06 (0.7023 s / it) * Acc@1 78.528 Acc@5 94.244 loss 0.808 Accuracy of the model EMA on 50000 test images: 78.5% Max EMA accuracy: 78.53% Epoch: [131] [ 0/312] eta: 0:45:17 lr: 0.003378 min_lr: 0.003378 loss: 1.6724 (1.6724) weight_decay: 0.0500 (0.0500) time: 8.7100 data: 7.2864 max mem: 64948 Epoch: [131] [ 10/312] eta: 0:07:52 lr: 0.003377 min_lr: 0.003377 loss: 2.2288 (2.1516) weight_decay: 0.0500 (0.0500) time: 1.5660 data: 0.7914 max mem: 64948 Epoch: [131] [ 20/312] eta: 0:05:35 lr: 0.003377 min_lr: 0.003377 loss: 2.1357 (2.0987) weight_decay: 0.0500 (0.0500) time: 0.7724 data: 0.0711 max mem: 64948 Epoch: [131] [ 30/312] eta: 0:04:43 lr: 0.003377 min_lr: 0.003377 loss: 2.3001 (2.2039) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [131] [ 40/312] eta: 0:04:12 lr: 0.003376 min_lr: 0.003376 loss: 2.4377 (2.1679) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [131] [ 50/312] eta: 0:03:51 lr: 0.003376 min_lr: 0.003376 loss: 2.1626 (2.1911) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0004 max mem: 64948 Epoch: [131] [ 60/312] eta: 0:03:35 lr: 0.003376 min_lr: 0.003376 loss: 2.1795 (2.1895) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [131] [ 70/312] eta: 0:03:21 lr: 0.003375 min_lr: 0.003375 loss: 2.0431 (2.1686) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [131] [ 80/312] eta: 0:03:09 lr: 0.003375 min_lr: 0.003375 loss: 2.1980 (2.1755) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [131] [ 90/312] eta: 0:02:57 lr: 0.003375 min_lr: 0.003375 loss: 2.2367 (2.1791) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [131] [100/312] eta: 0:02:47 lr: 0.003374 min_lr: 0.003374 loss: 2.2949 (2.1827) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [131] [110/312] eta: 0:02:38 lr: 0.003374 min_lr: 0.003374 loss: 2.3127 (2.1889) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [131] [120/312] eta: 0:02:28 lr: 0.003374 min_lr: 0.003374 loss: 2.2899 (2.1989) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [131] [130/312] eta: 0:02:19 lr: 0.003373 min_lr: 0.003373 loss: 2.2899 (2.2011) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [131] [140/312] eta: 0:02:11 lr: 0.003373 min_lr: 0.003373 loss: 2.2620 (2.2078) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [131] [150/312] eta: 0:02:02 lr: 0.003373 min_lr: 0.003373 loss: 2.3681 (2.2170) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [131] [160/312] eta: 0:01:54 lr: 0.003372 min_lr: 0.003372 loss: 2.4664 (2.2303) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [131] [170/312] eta: 0:01:46 lr: 0.003372 min_lr: 0.003372 loss: 2.4027 (2.2307) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [131] [180/312] eta: 0:01:38 lr: 0.003372 min_lr: 0.003372 loss: 2.3746 (2.2345) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [131] [190/312] eta: 0:01:30 lr: 0.003371 min_lr: 0.003371 loss: 2.1492 (2.2308) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [131] [200/312] eta: 0:01:23 lr: 0.003371 min_lr: 0.003371 loss: 2.2484 (2.2376) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [131] [210/312] eta: 0:01:15 lr: 0.003371 min_lr: 0.003371 loss: 2.3502 (2.2400) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [131] [220/312] eta: 0:01:07 lr: 0.003370 min_lr: 0.003370 loss: 2.3760 (2.2432) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [131] [230/312] eta: 0:01:00 lr: 0.003370 min_lr: 0.003370 loss: 2.3386 (2.2475) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [131] [240/312] eta: 0:00:52 lr: 0.003370 min_lr: 0.003370 loss: 2.3386 (2.2520) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [131] [250/312] eta: 0:00:45 lr: 0.003369 min_lr: 0.003369 loss: 2.4556 (2.2558) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [131] [260/312] eta: 0:00:38 lr: 0.003369 min_lr: 0.003369 loss: 2.3977 (2.2604) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [131] [270/312] eta: 0:00:30 lr: 0.003369 min_lr: 0.003369 loss: 2.4287 (2.2700) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [131] [280/312] eta: 0:00:23 lr: 0.003368 min_lr: 0.003368 loss: 2.4798 (2.2752) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [131] [290/312] eta: 0:00:16 lr: 0.003368 min_lr: 0.003368 loss: 2.3106 (2.2734) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0008 max mem: 64948 Epoch: [131] [300/312] eta: 0:00:08 lr: 0.003368 min_lr: 0.003368 loss: 2.3106 (2.2734) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [131] [310/312] eta: 0:00:01 lr: 0.003367 min_lr: 0.003367 loss: 2.2362 (2.2732) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [131] [311/312] eta: 0:00:00 lr: 0.003367 min_lr: 0.003367 loss: 2.2362 (2.2735) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [131] Total time: 0:03:47 (0.7286 s / it) Averaged stats: lr: 0.003367 min_lr: 0.003367 loss: 2.2362 (2.2425) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7776 (0.7776) acc1: 80.2083 (80.2083) acc5: 92.9688 (92.9688) time: 4.4676 data: 4.2549 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9279 (0.9242) acc1: 76.0417 (76.0320) acc5: 94.0104 (93.5360) time: 0.6477 data: 0.4729 max mem: 64948 Test: Total time: 0:00:06 (0.6709 s / it) * Acc@1 76.614 Acc@5 93.500 loss 0.913 Accuracy of the model on the 50000 test images: 76.6% Max accuracy: 76.76% Test: [0/9] eta: 0:00:42 loss: 0.6472 (0.6472) acc1: 83.8542 (83.8542) acc5: 95.3125 (95.3125) time: 4.6667 data: 4.4486 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8118 (0.8159) acc1: 78.6458 (78.1120) acc5: 94.7917 (94.0800) time: 0.6704 data: 0.4944 max mem: 64948 Test: Total time: 0:00:06 (0.6865 s / it) * Acc@1 78.614 Acc@5 94.302 loss 0.804 Accuracy of the model EMA on 50000 test images: 78.6% Max EMA accuracy: 78.61% Epoch: [132] [ 0/312] eta: 0:52:04 lr: 0.003367 min_lr: 0.003367 loss: 1.6396 (1.6396) weight_decay: 0.0500 (0.0500) time: 10.0153 data: 8.9456 max mem: 64948 Epoch: [132] [ 10/312] eta: 0:07:51 lr: 0.003367 min_lr: 0.003367 loss: 2.2757 (2.2087) weight_decay: 0.0500 (0.0500) time: 1.5624 data: 0.8136 max mem: 64948 Epoch: [132] [ 20/312] eta: 0:05:35 lr: 0.003366 min_lr: 0.003366 loss: 2.2781 (2.2560) weight_decay: 0.0500 (0.0500) time: 0.7051 data: 0.0003 max mem: 64948 Epoch: [132] [ 30/312] eta: 0:04:42 lr: 0.003366 min_lr: 0.003366 loss: 2.4240 (2.2614) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [132] [ 40/312] eta: 0:04:12 lr: 0.003366 min_lr: 0.003366 loss: 1.9840 (2.2069) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [132] [ 50/312] eta: 0:03:51 lr: 0.003365 min_lr: 0.003365 loss: 2.0371 (2.1948) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [132] [ 60/312] eta: 0:03:34 lr: 0.003365 min_lr: 0.003365 loss: 2.3198 (2.1863) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [132] [ 70/312] eta: 0:03:20 lr: 0.003365 min_lr: 0.003365 loss: 2.2969 (2.1783) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [132] [ 80/312] eta: 0:03:08 lr: 0.003364 min_lr: 0.003364 loss: 2.1756 (2.1682) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [132] [ 90/312] eta: 0:02:57 lr: 0.003364 min_lr: 0.003364 loss: 2.2404 (2.1898) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [132] [100/312] eta: 0:02:47 lr: 0.003364 min_lr: 0.003364 loss: 2.4160 (2.2160) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [132] [110/312] eta: 0:02:37 lr: 0.003363 min_lr: 0.003363 loss: 2.4294 (2.2349) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [132] [120/312] eta: 0:02:28 lr: 0.003363 min_lr: 0.003363 loss: 2.3762 (2.2357) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [132] [130/312] eta: 0:02:19 lr: 0.003363 min_lr: 0.003363 loss: 2.3762 (2.2350) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [132] [140/312] eta: 0:02:11 lr: 0.003362 min_lr: 0.003362 loss: 2.2785 (2.2315) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [132] [150/312] eta: 0:02:02 lr: 0.003362 min_lr: 0.003362 loss: 2.3318 (2.2316) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [132] [160/312] eta: 0:01:54 lr: 0.003362 min_lr: 0.003362 loss: 2.3318 (2.2248) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [132] [170/312] eta: 0:01:46 lr: 0.003361 min_lr: 0.003361 loss: 2.2075 (2.2203) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [132] [180/312] eta: 0:01:38 lr: 0.003361 min_lr: 0.003361 loss: 2.3506 (2.2304) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [132] [190/312] eta: 0:01:30 lr: 0.003361 min_lr: 0.003361 loss: 2.5202 (2.2374) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [132] [200/312] eta: 0:01:23 lr: 0.003360 min_lr: 0.003360 loss: 2.3338 (2.2388) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [132] [210/312] eta: 0:01:15 lr: 0.003360 min_lr: 0.003360 loss: 2.2812 (2.2425) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [132] [220/312] eta: 0:01:07 lr: 0.003360 min_lr: 0.003360 loss: 2.3151 (2.2437) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [132] [230/312] eta: 0:01:00 lr: 0.003359 min_lr: 0.003359 loss: 2.2579 (2.2469) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [132] [240/312] eta: 0:00:52 lr: 0.003359 min_lr: 0.003359 loss: 2.1470 (2.2387) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [132] [250/312] eta: 0:00:45 lr: 0.003359 min_lr: 0.003359 loss: 2.1245 (2.2404) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [132] [260/312] eta: 0:00:38 lr: 0.003358 min_lr: 0.003358 loss: 2.3917 (2.2464) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [132] [270/312] eta: 0:00:30 lr: 0.003358 min_lr: 0.003358 loss: 2.3917 (2.2442) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [132] [280/312] eta: 0:00:23 lr: 0.003358 min_lr: 0.003358 loss: 2.1095 (2.2349) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [132] [290/312] eta: 0:00:15 lr: 0.003357 min_lr: 0.003357 loss: 2.1042 (2.2348) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [132] [300/312] eta: 0:00:08 lr: 0.003357 min_lr: 0.003357 loss: 2.3267 (2.2365) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [132] [310/312] eta: 0:00:01 lr: 0.003356 min_lr: 0.003356 loss: 2.3983 (2.2362) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [132] [311/312] eta: 0:00:00 lr: 0.003356 min_lr: 0.003356 loss: 2.3983 (2.2372) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [132] Total time: 0:03:46 (0.7275 s / it) Averaged stats: lr: 0.003356 min_lr: 0.003356 loss: 2.3983 (2.2583) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.7084 (0.7084) acc1: 83.5938 (83.5938) acc5: 95.5729 (95.5729) time: 4.7670 data: 4.5494 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9412 (0.9273) acc1: 76.8229 (76.4480) acc5: 94.7917 (93.5040) time: 0.6809 data: 0.5056 max mem: 64948 Test: Total time: 0:00:06 (0.7045 s / it) * Acc@1 76.914 Acc@5 93.452 loss 0.900 Accuracy of the model on the 50000 test images: 76.9% Max accuracy: 76.91% Test: [0/9] eta: 0:00:40 loss: 0.6442 (0.6442) acc1: 84.1146 (84.1146) acc5: 95.0521 (95.0521) time: 4.5306 data: 4.3127 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8083 (0.8124) acc1: 78.3854 (78.2720) acc5: 94.7917 (94.1120) time: 0.6547 data: 0.4793 max mem: 64948 Test: Total time: 0:00:05 (0.6625 s / it) * Acc@1 78.664 Acc@5 94.354 loss 0.801 Accuracy of the model EMA on 50000 test images: 78.7% Max EMA accuracy: 78.66% Epoch: [133] [ 0/312] eta: 0:47:50 lr: 0.003356 min_lr: 0.003356 loss: 1.4510 (1.4510) weight_decay: 0.0500 (0.0500) time: 9.1997 data: 7.9958 max mem: 64948 Epoch: [133] [ 10/312] eta: 0:07:45 lr: 0.003356 min_lr: 0.003356 loss: 2.2354 (2.0797) weight_decay: 0.0500 (0.0500) time: 1.5400 data: 0.7837 max mem: 64948 Epoch: [133] [ 20/312] eta: 0:05:32 lr: 0.003356 min_lr: 0.003356 loss: 2.3488 (2.1685) weight_decay: 0.0500 (0.0500) time: 0.7357 data: 0.0314 max mem: 64948 Epoch: [133] [ 30/312] eta: 0:04:40 lr: 0.003355 min_lr: 0.003355 loss: 2.4242 (2.1378) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [133] [ 40/312] eta: 0:04:10 lr: 0.003355 min_lr: 0.003355 loss: 2.2359 (2.1585) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [133] [ 50/312] eta: 0:03:49 lr: 0.003355 min_lr: 0.003355 loss: 2.2708 (2.1552) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [133] [ 60/312] eta: 0:03:33 lr: 0.003354 min_lr: 0.003354 loss: 2.2369 (2.1622) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [133] [ 70/312] eta: 0:03:20 lr: 0.003354 min_lr: 0.003354 loss: 2.3710 (2.2079) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [133] [ 80/312] eta: 0:03:08 lr: 0.003354 min_lr: 0.003354 loss: 2.3234 (2.2157) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [133] [ 90/312] eta: 0:02:57 lr: 0.003353 min_lr: 0.003353 loss: 2.3112 (2.2291) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [133] [100/312] eta: 0:02:47 lr: 0.003353 min_lr: 0.003353 loss: 2.3389 (2.2326) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [133] [110/312] eta: 0:02:37 lr: 0.003353 min_lr: 0.003353 loss: 2.2028 (2.2134) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [133] [120/312] eta: 0:02:28 lr: 0.003352 min_lr: 0.003352 loss: 2.1338 (2.2128) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [133] [130/312] eta: 0:02:19 lr: 0.003352 min_lr: 0.003352 loss: 2.3609 (2.2171) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [133] [140/312] eta: 0:02:10 lr: 0.003352 min_lr: 0.003352 loss: 2.2295 (2.2070) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [133] [150/312] eta: 0:02:02 lr: 0.003351 min_lr: 0.003351 loss: 1.9575 (2.1921) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [133] [160/312] eta: 0:01:54 lr: 0.003351 min_lr: 0.003351 loss: 2.1355 (2.1979) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [133] [170/312] eta: 0:01:46 lr: 0.003351 min_lr: 0.003351 loss: 2.2678 (2.1983) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [133] [180/312] eta: 0:01:38 lr: 0.003350 min_lr: 0.003350 loss: 2.3266 (2.1997) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [133] [190/312] eta: 0:01:30 lr: 0.003350 min_lr: 0.003350 loss: 2.2214 (2.1948) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [133] [200/312] eta: 0:01:23 lr: 0.003350 min_lr: 0.003350 loss: 2.2214 (2.2024) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [133] [210/312] eta: 0:01:15 lr: 0.003349 min_lr: 0.003349 loss: 2.2634 (2.1991) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [133] [220/312] eta: 0:01:07 lr: 0.003349 min_lr: 0.003349 loss: 2.1932 (2.1997) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [133] [230/312] eta: 0:01:00 lr: 0.003348 min_lr: 0.003348 loss: 2.1932 (2.1977) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [133] [240/312] eta: 0:00:52 lr: 0.003348 min_lr: 0.003348 loss: 2.2322 (2.1968) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [133] [250/312] eta: 0:00:45 lr: 0.003348 min_lr: 0.003348 loss: 2.2768 (2.1973) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [133] [260/312] eta: 0:00:37 lr: 0.003347 min_lr: 0.003347 loss: 2.2662 (2.1939) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [133] [270/312] eta: 0:00:30 lr: 0.003347 min_lr: 0.003347 loss: 2.2433 (2.1952) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [133] [280/312] eta: 0:00:23 lr: 0.003347 min_lr: 0.003347 loss: 2.2144 (2.1933) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [133] [290/312] eta: 0:00:15 lr: 0.003346 min_lr: 0.003346 loss: 2.2637 (2.1967) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0008 max mem: 64948 Epoch: [133] [300/312] eta: 0:00:08 lr: 0.003346 min_lr: 0.003346 loss: 2.3153 (2.2009) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [133] [310/312] eta: 0:00:01 lr: 0.003346 min_lr: 0.003346 loss: 2.1595 (2.1969) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [133] [311/312] eta: 0:00:00 lr: 0.003346 min_lr: 0.003346 loss: 2.1685 (2.1971) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [133] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.003346 min_lr: 0.003346 loss: 2.1685 (2.2418) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7151 (0.7151) acc1: 82.5521 (82.5521) acc5: 95.0521 (95.0521) time: 4.6098 data: 4.3956 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9617 (0.9507) acc1: 76.3021 (75.8400) acc5: 94.5312 (93.2160) time: 0.6635 data: 0.4885 max mem: 64948 Test: Total time: 0:00:06 (0.6856 s / it) * Acc@1 76.348 Acc@5 93.042 loss 0.932 Accuracy of the model on the 50000 test images: 76.3% Max accuracy: 76.91% Test: [0/9] eta: 0:00:46 loss: 0.6410 (0.6410) acc1: 84.1146 (84.1146) acc5: 95.0521 (95.0521) time: 5.1246 data: 4.9068 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8049 (0.8093) acc1: 78.3854 (78.3040) acc5: 95.0521 (94.2400) time: 0.7241 data: 0.5453 max mem: 64948 Test: Total time: 0:00:06 (0.7334 s / it) * Acc@1 78.724 Acc@5 94.408 loss 0.798 Accuracy of the model EMA on 50000 test images: 78.7% Max EMA accuracy: 78.72% Epoch: [134] [ 0/312] eta: 0:47:15 lr: 0.003346 min_lr: 0.003346 loss: 2.6904 (2.6904) weight_decay: 0.0500 (0.0500) time: 9.0886 data: 7.2208 max mem: 64948 Epoch: [134] [ 10/312] eta: 0:07:37 lr: 0.003345 min_lr: 0.003345 loss: 2.3939 (2.2176) weight_decay: 0.0500 (0.0500) time: 1.5157 data: 0.6568 max mem: 64948 Epoch: [134] [ 20/312] eta: 0:05:28 lr: 0.003345 min_lr: 0.003345 loss: 2.1823 (2.1508) weight_decay: 0.0500 (0.0500) time: 0.7260 data: 0.0004 max mem: 64948 Epoch: [134] [ 30/312] eta: 0:04:37 lr: 0.003345 min_lr: 0.003345 loss: 1.8129 (2.1098) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [134] [ 40/312] eta: 0:04:08 lr: 0.003344 min_lr: 0.003344 loss: 2.1509 (2.1437) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [134] [ 50/312] eta: 0:03:48 lr: 0.003344 min_lr: 0.003344 loss: 2.3642 (2.1595) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [134] [ 60/312] eta: 0:03:32 lr: 0.003344 min_lr: 0.003344 loss: 2.2293 (2.1608) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [134] [ 70/312] eta: 0:03:18 lr: 0.003343 min_lr: 0.003343 loss: 2.2293 (2.1740) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [134] [ 80/312] eta: 0:03:07 lr: 0.003343 min_lr: 0.003343 loss: 2.3221 (2.1837) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [134] [ 90/312] eta: 0:02:56 lr: 0.003343 min_lr: 0.003343 loss: 2.3328 (2.1888) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [134] [100/312] eta: 0:02:46 lr: 0.003342 min_lr: 0.003342 loss: 2.4454 (2.2221) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [134] [110/312] eta: 0:02:36 lr: 0.003342 min_lr: 0.003342 loss: 2.4742 (2.2208) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [134] [120/312] eta: 0:02:27 lr: 0.003341 min_lr: 0.003341 loss: 2.1477 (2.2156) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [134] [130/312] eta: 0:02:19 lr: 0.003341 min_lr: 0.003341 loss: 2.1768 (2.2112) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [134] [140/312] eta: 0:02:10 lr: 0.003341 min_lr: 0.003341 loss: 2.2656 (2.2203) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [134] [150/312] eta: 0:02:02 lr: 0.003340 min_lr: 0.003340 loss: 2.2784 (2.2314) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [134] [160/312] eta: 0:01:54 lr: 0.003340 min_lr: 0.003340 loss: 2.3216 (2.2371) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [134] [170/312] eta: 0:01:46 lr: 0.003340 min_lr: 0.003340 loss: 2.4006 (2.2358) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [134] [180/312] eta: 0:01:38 lr: 0.003339 min_lr: 0.003339 loss: 2.4327 (2.2517) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [134] [190/312] eta: 0:01:30 lr: 0.003339 min_lr: 0.003339 loss: 2.4327 (2.2492) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [134] [200/312] eta: 0:01:22 lr: 0.003339 min_lr: 0.003339 loss: 2.2887 (2.2532) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [134] [210/312] eta: 0:01:15 lr: 0.003338 min_lr: 0.003338 loss: 2.2946 (2.2471) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [134] [220/312] eta: 0:01:07 lr: 0.003338 min_lr: 0.003338 loss: 2.2723 (2.2476) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [134] [230/312] eta: 0:01:00 lr: 0.003338 min_lr: 0.003338 loss: 2.2723 (2.2551) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [134] [240/312] eta: 0:00:52 lr: 0.003337 min_lr: 0.003337 loss: 2.3432 (2.2578) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [134] [250/312] eta: 0:00:45 lr: 0.003337 min_lr: 0.003337 loss: 2.3472 (2.2593) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [134] [260/312] eta: 0:00:37 lr: 0.003337 min_lr: 0.003337 loss: 2.3475 (2.2618) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [134] [270/312] eta: 0:00:30 lr: 0.003336 min_lr: 0.003336 loss: 2.2622 (2.2592) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [134] [280/312] eta: 0:00:23 lr: 0.003336 min_lr: 0.003336 loss: 2.2879 (2.2583) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [134] [290/312] eta: 0:00:15 lr: 0.003336 min_lr: 0.003336 loss: 2.2798 (2.2533) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [134] [300/312] eta: 0:00:08 lr: 0.003335 min_lr: 0.003335 loss: 2.1828 (2.2512) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0002 max mem: 64948 Epoch: [134] [310/312] eta: 0:00:01 lr: 0.003335 min_lr: 0.003335 loss: 2.2082 (2.2569) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [134] [311/312] eta: 0:00:00 lr: 0.003335 min_lr: 0.003335 loss: 2.2082 (2.2557) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [134] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.003335 min_lr: 0.003335 loss: 2.2082 (2.2505) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6664 (0.6664) acc1: 82.5521 (82.5521) acc5: 95.5729 (95.5729) time: 4.4586 data: 4.2347 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0233 (0.9654) acc1: 75.0000 (75.4560) acc5: 92.4528 (93.1840) time: 0.6467 data: 0.4706 max mem: 64948 Test: Total time: 0:00:05 (0.6561 s / it) * Acc@1 76.172 Acc@5 93.262 loss 0.935 Accuracy of the model on the 50000 test images: 76.2% Max accuracy: 76.91% Test: [0/9] eta: 0:00:41 loss: 0.6385 (0.6385) acc1: 84.1146 (84.1146) acc5: 95.0521 (95.0521) time: 4.6144 data: 4.4115 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8009 (0.8066) acc1: 78.6458 (78.3360) acc5: 95.0521 (94.2720) time: 0.6640 data: 0.4903 max mem: 64948 Test: Total time: 0:00:06 (0.6717 s / it) * Acc@1 78.780 Acc@5 94.452 loss 0.794 Accuracy of the model EMA on 50000 test images: 78.8% Max EMA accuracy: 78.78% Epoch: [135] [ 0/312] eta: 0:46:18 lr: 0.003335 min_lr: 0.003335 loss: 2.7544 (2.7544) weight_decay: 0.0500 (0.0500) time: 8.9044 data: 7.6432 max mem: 64948 Epoch: [135] [ 10/312] eta: 0:07:25 lr: 0.003334 min_lr: 0.003334 loss: 2.2558 (2.1953) weight_decay: 0.0500 (0.0500) time: 1.4757 data: 0.6953 max mem: 64948 Epoch: [135] [ 20/312] eta: 0:05:22 lr: 0.003334 min_lr: 0.003334 loss: 2.1643 (2.2755) weight_decay: 0.0500 (0.0500) time: 0.7130 data: 0.0005 max mem: 64948 Epoch: [135] [ 30/312] eta: 0:04:33 lr: 0.003334 min_lr: 0.003334 loss: 2.2929 (2.2596) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [135] [ 40/312] eta: 0:04:05 lr: 0.003333 min_lr: 0.003333 loss: 2.1934 (2.2112) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [135] [ 50/312] eta: 0:03:46 lr: 0.003333 min_lr: 0.003333 loss: 2.1934 (2.1981) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [135] [ 60/312] eta: 0:03:31 lr: 0.003333 min_lr: 0.003333 loss: 2.3373 (2.2344) weight_decay: 0.0500 (0.0500) time: 0.7026 data: 0.0004 max mem: 64948 Epoch: [135] [ 70/312] eta: 0:03:17 lr: 0.003332 min_lr: 0.003332 loss: 2.4328 (2.2429) weight_decay: 0.0500 (0.0500) time: 0.7018 data: 0.0004 max mem: 64948 Epoch: [135] [ 80/312] eta: 0:03:06 lr: 0.003332 min_lr: 0.003332 loss: 2.5008 (2.2636) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [135] [ 90/312] eta: 0:02:55 lr: 0.003332 min_lr: 0.003332 loss: 2.2770 (2.2469) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [135] [100/312] eta: 0:02:45 lr: 0.003331 min_lr: 0.003331 loss: 2.2628 (2.2512) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [135] [110/312] eta: 0:02:36 lr: 0.003331 min_lr: 0.003331 loss: 2.2843 (2.2535) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [135] [120/312] eta: 0:02:27 lr: 0.003331 min_lr: 0.003331 loss: 2.2843 (2.2577) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [135] [130/312] eta: 0:02:18 lr: 0.003330 min_lr: 0.003330 loss: 2.2576 (2.2502) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [135] [140/312] eta: 0:02:10 lr: 0.003330 min_lr: 0.003330 loss: 2.2918 (2.2672) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [135] [150/312] eta: 0:02:02 lr: 0.003330 min_lr: 0.003330 loss: 2.2565 (2.2458) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [135] [160/312] eta: 0:01:53 lr: 0.003329 min_lr: 0.003329 loss: 2.0476 (2.2435) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [135] [170/312] eta: 0:01:45 lr: 0.003329 min_lr: 0.003329 loss: 2.2778 (2.2385) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [135] [180/312] eta: 0:01:38 lr: 0.003329 min_lr: 0.003329 loss: 2.2778 (2.2392) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [135] [190/312] eta: 0:01:30 lr: 0.003328 min_lr: 0.003328 loss: 2.3535 (2.2447) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [135] [200/312] eta: 0:01:22 lr: 0.003328 min_lr: 0.003328 loss: 2.3627 (2.2517) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [135] [210/312] eta: 0:01:15 lr: 0.003327 min_lr: 0.003327 loss: 2.4607 (2.2607) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [135] [220/312] eta: 0:01:07 lr: 0.003327 min_lr: 0.003327 loss: 2.3775 (2.2603) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [135] [230/312] eta: 0:01:00 lr: 0.003327 min_lr: 0.003327 loss: 2.2978 (2.2574) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [135] [240/312] eta: 0:00:52 lr: 0.003326 min_lr: 0.003326 loss: 2.2475 (2.2539) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [135] [250/312] eta: 0:00:45 lr: 0.003326 min_lr: 0.003326 loss: 2.2868 (2.2531) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [135] [260/312] eta: 0:00:37 lr: 0.003326 min_lr: 0.003326 loss: 2.2868 (2.2520) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [135] [270/312] eta: 0:00:30 lr: 0.003325 min_lr: 0.003325 loss: 2.2210 (2.2491) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [135] [280/312] eta: 0:00:23 lr: 0.003325 min_lr: 0.003325 loss: 2.2210 (2.2516) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [135] [290/312] eta: 0:00:15 lr: 0.003325 min_lr: 0.003325 loss: 2.3759 (2.2550) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [135] [300/312] eta: 0:00:08 lr: 0.003324 min_lr: 0.003324 loss: 2.2557 (2.2531) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [135] [310/312] eta: 0:00:01 lr: 0.003324 min_lr: 0.003324 loss: 2.2941 (2.2544) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [135] [311/312] eta: 0:00:00 lr: 0.003324 min_lr: 0.003324 loss: 2.2941 (2.2548) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [135] Total time: 0:03:46 (0.7251 s / it) Averaged stats: lr: 0.003324 min_lr: 0.003324 loss: 2.2941 (2.2422) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7254 (0.7254) acc1: 81.7708 (81.7708) acc5: 96.3542 (96.3542) time: 4.5757 data: 4.3544 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0087 (0.9553) acc1: 76.5625 (75.7760) acc5: 92.7083 (92.9920) time: 0.6598 data: 0.4839 max mem: 64948 Test: Total time: 0:00:06 (0.6846 s / it) * Acc@1 76.376 Acc@5 92.960 loss 0.939 Accuracy of the model on the 50000 test images: 76.4% Max accuracy: 76.91% Test: [0/9] eta: 0:00:43 loss: 0.6354 (0.6354) acc1: 84.1146 (84.1146) acc5: 95.0521 (95.0521) time: 4.8746 data: 4.6654 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7979 (0.8036) acc1: 78.9062 (78.4320) acc5: 95.0521 (94.3040) time: 0.6934 data: 0.5185 max mem: 64948 Test: Total time: 0:00:06 (0.7027 s / it) * Acc@1 78.860 Acc@5 94.500 loss 0.791 Accuracy of the model EMA on 50000 test images: 78.9% Max EMA accuracy: 78.86% Epoch: [136] [ 0/312] eta: 0:50:57 lr: 0.003324 min_lr: 0.003324 loss: 2.5528 (2.5528) weight_decay: 0.0500 (0.0500) time: 9.7992 data: 9.0141 max mem: 64948 Epoch: [136] [ 10/312] eta: 0:07:45 lr: 0.003324 min_lr: 0.003324 loss: 2.2591 (2.1808) weight_decay: 0.0500 (0.0500) time: 1.5425 data: 0.8198 max mem: 64948 Epoch: [136] [ 20/312] eta: 0:05:32 lr: 0.003323 min_lr: 0.003323 loss: 2.2774 (2.2629) weight_decay: 0.0500 (0.0500) time: 0.7047 data: 0.0004 max mem: 64948 Epoch: [136] [ 30/312] eta: 0:04:40 lr: 0.003323 min_lr: 0.003323 loss: 2.3372 (2.2648) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [136] [ 40/312] eta: 0:04:10 lr: 0.003322 min_lr: 0.003322 loss: 2.2865 (2.2405) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [136] [ 50/312] eta: 0:03:50 lr: 0.003322 min_lr: 0.003322 loss: 2.2671 (2.2564) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [136] [ 60/312] eta: 0:03:33 lr: 0.003322 min_lr: 0.003322 loss: 2.3977 (2.2568) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [136] [ 70/312] eta: 0:03:20 lr: 0.003321 min_lr: 0.003321 loss: 2.4139 (2.2859) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [136] [ 80/312] eta: 0:03:08 lr: 0.003321 min_lr: 0.003321 loss: 2.4432 (2.2863) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [136] [ 90/312] eta: 0:02:57 lr: 0.003321 min_lr: 0.003321 loss: 2.2884 (2.2825) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [136] [100/312] eta: 0:02:46 lr: 0.003320 min_lr: 0.003320 loss: 2.1412 (2.2694) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [136] [110/312] eta: 0:02:37 lr: 0.003320 min_lr: 0.003320 loss: 2.2722 (2.2712) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [136] [120/312] eta: 0:02:28 lr: 0.003320 min_lr: 0.003320 loss: 2.3037 (2.2663) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [136] [130/312] eta: 0:02:19 lr: 0.003319 min_lr: 0.003319 loss: 1.8996 (2.2361) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [136] [140/312] eta: 0:02:10 lr: 0.003319 min_lr: 0.003319 loss: 2.2159 (2.2475) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [136] [150/312] eta: 0:02:02 lr: 0.003319 min_lr: 0.003319 loss: 2.2677 (2.2426) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [136] [160/312] eta: 0:01:54 lr: 0.003318 min_lr: 0.003318 loss: 2.2589 (2.2405) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [136] [170/312] eta: 0:01:46 lr: 0.003318 min_lr: 0.003318 loss: 2.3152 (2.2357) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [136] [180/312] eta: 0:01:38 lr: 0.003318 min_lr: 0.003318 loss: 2.1675 (2.2317) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [136] [190/312] eta: 0:01:30 lr: 0.003317 min_lr: 0.003317 loss: 2.1675 (2.2346) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [136] [200/312] eta: 0:01:22 lr: 0.003317 min_lr: 0.003317 loss: 2.2357 (2.2345) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [136] [210/312] eta: 0:01:15 lr: 0.003317 min_lr: 0.003317 loss: 2.3937 (2.2406) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [136] [220/312] eta: 0:01:07 lr: 0.003316 min_lr: 0.003316 loss: 2.4708 (2.2478) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [136] [230/312] eta: 0:01:00 lr: 0.003316 min_lr: 0.003316 loss: 2.4058 (2.2414) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [136] [240/312] eta: 0:00:52 lr: 0.003315 min_lr: 0.003315 loss: 2.2636 (2.2370) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [136] [250/312] eta: 0:00:45 lr: 0.003315 min_lr: 0.003315 loss: 2.2809 (2.2385) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [136] [260/312] eta: 0:00:37 lr: 0.003315 min_lr: 0.003315 loss: 2.1807 (2.2349) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [136] [270/312] eta: 0:00:30 lr: 0.003314 min_lr: 0.003314 loss: 2.1807 (2.2343) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [136] [280/312] eta: 0:00:23 lr: 0.003314 min_lr: 0.003314 loss: 2.1037 (2.2314) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0010 max mem: 64948 Epoch: [136] [290/312] eta: 0:00:15 lr: 0.003314 min_lr: 0.003314 loss: 2.1037 (2.2332) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [136] [300/312] eta: 0:00:08 lr: 0.003313 min_lr: 0.003313 loss: 2.3773 (2.2366) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0001 max mem: 64948 Epoch: [136] [310/312] eta: 0:00:01 lr: 0.003313 min_lr: 0.003313 loss: 2.4910 (2.2416) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [136] [311/312] eta: 0:00:00 lr: 0.003313 min_lr: 0.003313 loss: 2.4910 (2.2422) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [136] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.003313 min_lr: 0.003313 loss: 2.4910 (2.2401) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7411 (0.7411) acc1: 83.5938 (83.5938) acc5: 95.8333 (95.8333) time: 4.4533 data: 4.2334 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0026 (0.9801) acc1: 77.0833 (75.3920) acc5: 92.9688 (92.9600) time: 0.6466 data: 0.4705 max mem: 64948 Test: Total time: 0:00:06 (0.6686 s / it) * Acc@1 76.308 Acc@5 93.208 loss 0.955 Accuracy of the model on the 50000 test images: 76.3% Max accuracy: 76.91% Test: [0/9] eta: 0:00:42 loss: 0.6334 (0.6334) acc1: 83.8542 (83.8542) acc5: 95.0521 (95.0521) time: 4.6713 data: 4.4670 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7954 (0.8008) acc1: 78.3854 (78.3360) acc5: 95.0521 (94.4000) time: 0.6703 data: 0.4964 max mem: 64948 Test: Total time: 0:00:06 (0.6826 s / it) * Acc@1 78.872 Acc@5 94.544 loss 0.788 Accuracy of the model EMA on 50000 test images: 78.9% Max EMA accuracy: 78.87% Epoch: [137] [ 0/312] eta: 0:47:40 lr: 0.003313 min_lr: 0.003313 loss: 2.5127 (2.5127) weight_decay: 0.0500 (0.0500) time: 9.1684 data: 8.3552 max mem: 64948 Epoch: [137] [ 10/312] eta: 0:07:30 lr: 0.003313 min_lr: 0.003313 loss: 2.2881 (2.0878) weight_decay: 0.0500 (0.0500) time: 1.4917 data: 0.7599 max mem: 64948 Epoch: [137] [ 20/312] eta: 0:05:24 lr: 0.003312 min_lr: 0.003312 loss: 2.2881 (2.1821) weight_decay: 0.0500 (0.0500) time: 0.7090 data: 0.0003 max mem: 64948 Epoch: [137] [ 30/312] eta: 0:04:35 lr: 0.003312 min_lr: 0.003312 loss: 2.3170 (2.1856) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [137] [ 40/312] eta: 0:04:07 lr: 0.003311 min_lr: 0.003311 loss: 2.2666 (2.1677) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [137] [ 50/312] eta: 0:03:47 lr: 0.003311 min_lr: 0.003311 loss: 2.0995 (2.1681) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [137] [ 60/312] eta: 0:03:31 lr: 0.003311 min_lr: 0.003311 loss: 2.2878 (2.2017) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [137] [ 70/312] eta: 0:03:18 lr: 0.003310 min_lr: 0.003310 loss: 2.1896 (2.1823) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [137] [ 80/312] eta: 0:03:06 lr: 0.003310 min_lr: 0.003310 loss: 2.2259 (2.2056) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [137] [ 90/312] eta: 0:02:55 lr: 0.003310 min_lr: 0.003310 loss: 2.2948 (2.2001) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [137] [100/312] eta: 0:02:45 lr: 0.003309 min_lr: 0.003309 loss: 2.2899 (2.2095) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [137] [110/312] eta: 0:02:36 lr: 0.003309 min_lr: 0.003309 loss: 2.3015 (2.2040) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [137] [120/312] eta: 0:02:27 lr: 0.003309 min_lr: 0.003309 loss: 2.3346 (2.2207) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [137] [130/312] eta: 0:02:18 lr: 0.003308 min_lr: 0.003308 loss: 2.3783 (2.2267) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [137] [140/312] eta: 0:02:10 lr: 0.003308 min_lr: 0.003308 loss: 2.3309 (2.2230) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [137] [150/312] eta: 0:02:02 lr: 0.003308 min_lr: 0.003308 loss: 1.9013 (2.2074) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [137] [160/312] eta: 0:01:53 lr: 0.003307 min_lr: 0.003307 loss: 2.2140 (2.2145) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [137] [170/312] eta: 0:01:46 lr: 0.003307 min_lr: 0.003307 loss: 2.3618 (2.2199) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [137] [180/312] eta: 0:01:38 lr: 0.003307 min_lr: 0.003307 loss: 2.2161 (2.2125) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [137] [190/312] eta: 0:01:30 lr: 0.003306 min_lr: 0.003306 loss: 2.1058 (2.2091) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [137] [200/312] eta: 0:01:22 lr: 0.003306 min_lr: 0.003306 loss: 1.9812 (2.2005) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [137] [210/312] eta: 0:01:15 lr: 0.003305 min_lr: 0.003305 loss: 2.0750 (2.2080) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [137] [220/312] eta: 0:01:07 lr: 0.003305 min_lr: 0.003305 loss: 2.1545 (2.2038) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [137] [230/312] eta: 0:01:00 lr: 0.003305 min_lr: 0.003305 loss: 2.1545 (2.2034) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [137] [240/312] eta: 0:00:52 lr: 0.003304 min_lr: 0.003304 loss: 2.3471 (2.2071) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [137] [250/312] eta: 0:00:45 lr: 0.003304 min_lr: 0.003304 loss: 2.3451 (2.2009) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [137] [260/312] eta: 0:00:37 lr: 0.003304 min_lr: 0.003304 loss: 2.2167 (2.2042) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [137] [270/312] eta: 0:00:30 lr: 0.003303 min_lr: 0.003303 loss: 2.2278 (2.2074) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [137] [280/312] eta: 0:00:23 lr: 0.003303 min_lr: 0.003303 loss: 2.2278 (2.2055) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [137] [290/312] eta: 0:00:15 lr: 0.003303 min_lr: 0.003303 loss: 2.3117 (2.2116) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [137] [300/312] eta: 0:00:08 lr: 0.003302 min_lr: 0.003302 loss: 2.3632 (2.2131) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [137] [310/312] eta: 0:00:01 lr: 0.003302 min_lr: 0.003302 loss: 2.2086 (2.2138) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [137] [311/312] eta: 0:00:00 lr: 0.003302 min_lr: 0.003302 loss: 2.2086 (2.2134) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [137] Total time: 0:03:46 (0.7253 s / it) Averaged stats: lr: 0.003302 min_lr: 0.003302 loss: 2.2086 (2.2462) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6823 (0.6823) acc1: 82.5521 (82.5521) acc5: 95.8333 (95.8333) time: 4.5491 data: 4.3307 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9520 (0.9264) acc1: 75.5208 (76.3520) acc5: 92.9688 (93.2160) time: 0.6568 data: 0.4813 max mem: 64948 Test: Total time: 0:00:06 (0.6772 s / it) * Acc@1 76.688 Acc@5 93.356 loss 0.911 Accuracy of the model on the 50000 test images: 76.7% Max accuracy: 76.91% Test: [0/9] eta: 0:00:45 loss: 0.6315 (0.6315) acc1: 83.8542 (83.8542) acc5: 95.0521 (95.0521) time: 5.0150 data: 4.8078 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7935 (0.7980) acc1: 78.6458 (78.4320) acc5: 95.0521 (94.4960) time: 0.7086 data: 0.5343 max mem: 64948 Test: Total time: 0:00:06 (0.7166 s / it) * Acc@1 78.964 Acc@5 94.592 loss 0.786 Accuracy of the model EMA on 50000 test images: 79.0% Max EMA accuracy: 78.96% Epoch: [138] [ 0/312] eta: 0:45:08 lr: 0.003302 min_lr: 0.003302 loss: 2.5157 (2.5157) weight_decay: 0.0500 (0.0500) time: 8.6806 data: 7.7114 max mem: 64948 Epoch: [138] [ 10/312] eta: 0:07:25 lr: 0.003301 min_lr: 0.003301 loss: 2.3757 (2.3007) weight_decay: 0.0500 (0.0500) time: 1.4757 data: 0.7138 max mem: 64948 Epoch: [138] [ 20/312] eta: 0:05:22 lr: 0.003301 min_lr: 0.003301 loss: 2.3757 (2.3304) weight_decay: 0.0500 (0.0500) time: 0.7270 data: 0.0072 max mem: 64948 Epoch: [138] [ 30/312] eta: 0:04:34 lr: 0.003301 min_lr: 0.003301 loss: 2.4048 (2.3372) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [138] [ 40/312] eta: 0:04:06 lr: 0.003300 min_lr: 0.003300 loss: 2.3491 (2.3263) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [138] [ 50/312] eta: 0:03:46 lr: 0.003300 min_lr: 0.003300 loss: 2.3491 (2.2968) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [138] [ 60/312] eta: 0:03:30 lr: 0.003300 min_lr: 0.003300 loss: 2.3607 (2.2896) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [138] [ 70/312] eta: 0:03:17 lr: 0.003299 min_lr: 0.003299 loss: 2.1108 (2.2618) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [138] [ 80/312] eta: 0:03:06 lr: 0.003299 min_lr: 0.003299 loss: 1.9833 (2.2295) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [138] [ 90/312] eta: 0:02:55 lr: 0.003299 min_lr: 0.003299 loss: 2.3575 (2.2474) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [138] [100/312] eta: 0:02:45 lr: 0.003298 min_lr: 0.003298 loss: 2.3575 (2.2313) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [138] [110/312] eta: 0:02:36 lr: 0.003298 min_lr: 0.003298 loss: 2.1707 (2.2235) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [138] [120/312] eta: 0:02:27 lr: 0.003298 min_lr: 0.003298 loss: 2.3695 (2.2408) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [138] [130/312] eta: 0:02:18 lr: 0.003297 min_lr: 0.003297 loss: 2.4799 (2.2475) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [138] [140/312] eta: 0:02:10 lr: 0.003297 min_lr: 0.003297 loss: 2.3537 (2.2475) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [138] [150/312] eta: 0:02:01 lr: 0.003297 min_lr: 0.003297 loss: 2.2975 (2.2500) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [138] [160/312] eta: 0:01:53 lr: 0.003296 min_lr: 0.003296 loss: 2.3180 (2.2532) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [138] [170/312] eta: 0:01:45 lr: 0.003296 min_lr: 0.003296 loss: 2.3180 (2.2512) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [138] [180/312] eta: 0:01:38 lr: 0.003295 min_lr: 0.003295 loss: 2.3566 (2.2607) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [138] [190/312] eta: 0:01:30 lr: 0.003295 min_lr: 0.003295 loss: 2.3566 (2.2658) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [138] [200/312] eta: 0:01:22 lr: 0.003295 min_lr: 0.003295 loss: 2.4680 (2.2766) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [138] [210/312] eta: 0:01:15 lr: 0.003294 min_lr: 0.003294 loss: 2.5100 (2.2733) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [138] [220/312] eta: 0:01:07 lr: 0.003294 min_lr: 0.003294 loss: 2.2447 (2.2749) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [138] [230/312] eta: 0:01:00 lr: 0.003294 min_lr: 0.003294 loss: 2.3107 (2.2778) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [138] [240/312] eta: 0:00:52 lr: 0.003293 min_lr: 0.003293 loss: 2.3570 (2.2859) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [138] [250/312] eta: 0:00:45 lr: 0.003293 min_lr: 0.003293 loss: 2.2338 (2.2718) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [138] [260/312] eta: 0:00:37 lr: 0.003293 min_lr: 0.003293 loss: 2.1579 (2.2745) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [138] [270/312] eta: 0:00:30 lr: 0.003292 min_lr: 0.003292 loss: 2.3849 (2.2721) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [138] [280/312] eta: 0:00:23 lr: 0.003292 min_lr: 0.003292 loss: 2.2773 (2.2726) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [138] [290/312] eta: 0:00:15 lr: 0.003292 min_lr: 0.003292 loss: 2.2773 (2.2753) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [138] [300/312] eta: 0:00:08 lr: 0.003291 min_lr: 0.003291 loss: 2.3102 (2.2701) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [138] [310/312] eta: 0:00:01 lr: 0.003291 min_lr: 0.003291 loss: 2.2014 (2.2677) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [138] [311/312] eta: 0:00:00 lr: 0.003291 min_lr: 0.003291 loss: 2.1911 (2.2674) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [138] Total time: 0:03:46 (0.7253 s / it) Averaged stats: lr: 0.003291 min_lr: 0.003291 loss: 2.1911 (2.2440) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.7135 (0.7135) acc1: 83.5938 (83.5938) acc5: 95.3125 (95.3125) time: 4.3338 data: 4.1136 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0361 (0.9520) acc1: 74.7396 (75.6800) acc5: 93.2292 (93.0880) time: 0.6331 data: 0.4572 max mem: 64948 Test: Total time: 0:00:05 (0.6598 s / it) * Acc@1 76.292 Acc@5 93.332 loss 0.923 Accuracy of the model on the 50000 test images: 76.3% Max accuracy: 76.91% Test: [0/9] eta: 0:00:45 loss: 0.6294 (0.6294) acc1: 83.8542 (83.8542) acc5: 95.0521 (95.0521) time: 5.0112 data: 4.7982 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7919 (0.7952) acc1: 78.3854 (78.4960) acc5: 95.0521 (94.4640) time: 0.7081 data: 0.5332 max mem: 64948 Test: Total time: 0:00:06 (0.7196 s / it) * Acc@1 79.034 Acc@5 94.618 loss 0.783 Accuracy of the model EMA on 50000 test images: 79.0% Max EMA accuracy: 79.03% Epoch: [139] [ 0/312] eta: 0:44:49 lr: 0.003291 min_lr: 0.003291 loss: 2.8427 (2.8427) weight_decay: 0.0500 (0.0500) time: 8.6215 data: 7.5295 max mem: 64948 Epoch: [139] [ 10/312] eta: 0:07:31 lr: 0.003290 min_lr: 0.003290 loss: 2.2872 (2.2272) weight_decay: 0.0500 (0.0500) time: 1.4965 data: 0.6849 max mem: 64948 Epoch: [139] [ 20/312] eta: 0:05:25 lr: 0.003290 min_lr: 0.003290 loss: 2.2872 (2.2921) weight_decay: 0.0500 (0.0500) time: 0.7400 data: 0.0004 max mem: 64948 Epoch: [139] [ 30/312] eta: 0:04:36 lr: 0.003290 min_lr: 0.003290 loss: 2.3004 (2.2605) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [139] [ 40/312] eta: 0:04:07 lr: 0.003289 min_lr: 0.003289 loss: 2.2707 (2.2562) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [139] [ 50/312] eta: 0:03:47 lr: 0.003289 min_lr: 0.003289 loss: 2.3287 (2.2552) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0003 max mem: 64948 Epoch: [139] [ 60/312] eta: 0:03:31 lr: 0.003289 min_lr: 0.003289 loss: 2.4201 (2.2794) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [139] [ 70/312] eta: 0:03:18 lr: 0.003288 min_lr: 0.003288 loss: 2.4201 (2.2829) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [139] [ 80/312] eta: 0:03:06 lr: 0.003288 min_lr: 0.003288 loss: 2.2626 (2.2762) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [139] [ 90/312] eta: 0:02:55 lr: 0.003287 min_lr: 0.003287 loss: 2.2436 (2.2742) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [139] [100/312] eta: 0:02:46 lr: 0.003287 min_lr: 0.003287 loss: 2.2682 (2.2631) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [139] [110/312] eta: 0:02:36 lr: 0.003287 min_lr: 0.003287 loss: 2.2682 (2.2674) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [139] [120/312] eta: 0:02:27 lr: 0.003286 min_lr: 0.003286 loss: 2.1916 (2.2501) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [139] [130/312] eta: 0:02:18 lr: 0.003286 min_lr: 0.003286 loss: 2.1006 (2.2373) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [139] [140/312] eta: 0:02:10 lr: 0.003286 min_lr: 0.003286 loss: 2.2087 (2.2412) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [139] [150/312] eta: 0:02:02 lr: 0.003285 min_lr: 0.003285 loss: 2.1905 (2.2338) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [139] [160/312] eta: 0:01:53 lr: 0.003285 min_lr: 0.003285 loss: 2.2066 (2.2370) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [139] [170/312] eta: 0:01:46 lr: 0.003285 min_lr: 0.003285 loss: 2.3911 (2.2371) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [139] [180/312] eta: 0:01:38 lr: 0.003284 min_lr: 0.003284 loss: 2.2818 (2.2411) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [139] [190/312] eta: 0:01:30 lr: 0.003284 min_lr: 0.003284 loss: 2.3792 (2.2438) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [139] [200/312] eta: 0:01:22 lr: 0.003284 min_lr: 0.003284 loss: 2.2627 (2.2390) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [139] [210/312] eta: 0:01:15 lr: 0.003283 min_lr: 0.003283 loss: 2.0585 (2.2313) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [139] [220/312] eta: 0:01:07 lr: 0.003283 min_lr: 0.003283 loss: 2.2601 (2.2319) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [139] [230/312] eta: 0:01:00 lr: 0.003282 min_lr: 0.003282 loss: 2.3670 (2.2377) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [139] [240/312] eta: 0:00:52 lr: 0.003282 min_lr: 0.003282 loss: 2.3565 (2.2394) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [139] [250/312] eta: 0:00:45 lr: 0.003282 min_lr: 0.003282 loss: 2.2200 (2.2372) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [139] [260/312] eta: 0:00:37 lr: 0.003281 min_lr: 0.003281 loss: 2.0827 (2.2338) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [139] [270/312] eta: 0:00:30 lr: 0.003281 min_lr: 0.003281 loss: 2.0990 (2.2299) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [139] [280/312] eta: 0:00:23 lr: 0.003281 min_lr: 0.003281 loss: 2.3927 (2.2362) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0010 max mem: 64948 Epoch: [139] [290/312] eta: 0:00:15 lr: 0.003280 min_lr: 0.003280 loss: 2.3514 (2.2347) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [139] [300/312] eta: 0:00:08 lr: 0.003280 min_lr: 0.003280 loss: 2.2470 (2.2370) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [139] [310/312] eta: 0:00:01 lr: 0.003280 min_lr: 0.003280 loss: 2.3017 (2.2349) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [139] [311/312] eta: 0:00:00 lr: 0.003280 min_lr: 0.003280 loss: 2.3017 (2.2354) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [139] Total time: 0:03:46 (0.7254 s / it) Averaged stats: lr: 0.003280 min_lr: 0.003280 loss: 2.3017 (2.2485) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6772 (0.6772) acc1: 83.3333 (83.3333) acc5: 95.3125 (95.3125) time: 4.4559 data: 4.2478 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9932 (0.9263) acc1: 76.3021 (76.4480) acc5: 93.7500 (93.5680) time: 0.6490 data: 0.4744 max mem: 64948 Test: Total time: 0:00:06 (0.6755 s / it) * Acc@1 76.820 Acc@5 93.286 loss 0.916 Accuracy of the model on the 50000 test images: 76.8% Max accuracy: 76.91% Test: [0/9] eta: 0:00:43 loss: 0.6275 (0.6275) acc1: 83.8542 (83.8542) acc5: 95.0521 (95.0521) time: 4.8777 data: 4.6712 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7901 (0.7925) acc1: 78.3854 (78.5280) acc5: 95.0521 (94.4960) time: 0.6932 data: 0.5191 max mem: 64948 Test: Total time: 0:00:06 (0.7017 s / it) * Acc@1 79.086 Acc@5 94.648 loss 0.780 Accuracy of the model EMA on 50000 test images: 79.1% Max EMA accuracy: 79.09% Epoch: [140] [ 0/312] eta: 0:47:48 lr: 0.003280 min_lr: 0.003280 loss: 1.7352 (1.7352) weight_decay: 0.0500 (0.0500) time: 9.1940 data: 7.4012 max mem: 64948 Epoch: [140] [ 10/312] eta: 0:07:42 lr: 0.003279 min_lr: 0.003279 loss: 2.4507 (2.3709) weight_decay: 0.0500 (0.0500) time: 1.5329 data: 0.6733 max mem: 64948 Epoch: [140] [ 20/312] eta: 0:05:31 lr: 0.003279 min_lr: 0.003279 loss: 2.4507 (2.3477) weight_decay: 0.0500 (0.0500) time: 0.7310 data: 0.0004 max mem: 64948 Epoch: [140] [ 30/312] eta: 0:04:39 lr: 0.003278 min_lr: 0.003278 loss: 2.3382 (2.3098) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [140] [ 40/312] eta: 0:04:10 lr: 0.003278 min_lr: 0.003278 loss: 2.3262 (2.3201) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [140] [ 50/312] eta: 0:03:49 lr: 0.003278 min_lr: 0.003278 loss: 2.2479 (2.2726) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [140] [ 60/312] eta: 0:03:33 lr: 0.003277 min_lr: 0.003277 loss: 2.2927 (2.2565) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [140] [ 70/312] eta: 0:03:19 lr: 0.003277 min_lr: 0.003277 loss: 2.3146 (2.2503) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [140] [ 80/312] eta: 0:03:07 lr: 0.003277 min_lr: 0.003277 loss: 2.3154 (2.2565) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [140] [ 90/312] eta: 0:02:56 lr: 0.003276 min_lr: 0.003276 loss: 2.2202 (2.2262) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [140] [100/312] eta: 0:02:46 lr: 0.003276 min_lr: 0.003276 loss: 2.1102 (2.2220) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [140] [110/312] eta: 0:02:37 lr: 0.003276 min_lr: 0.003276 loss: 2.3009 (2.2273) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [140] [120/312] eta: 0:02:28 lr: 0.003275 min_lr: 0.003275 loss: 2.2837 (2.2054) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [140] [130/312] eta: 0:02:19 lr: 0.003275 min_lr: 0.003275 loss: 1.9735 (2.2038) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [140] [140/312] eta: 0:02:10 lr: 0.003274 min_lr: 0.003274 loss: 2.4059 (2.2149) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [140] [150/312] eta: 0:02:02 lr: 0.003274 min_lr: 0.003274 loss: 2.4059 (2.2225) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [140] [160/312] eta: 0:01:54 lr: 0.003274 min_lr: 0.003274 loss: 2.2886 (2.2182) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [140] [170/312] eta: 0:01:46 lr: 0.003273 min_lr: 0.003273 loss: 2.2886 (2.2253) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [140] [180/312] eta: 0:01:38 lr: 0.003273 min_lr: 0.003273 loss: 2.3699 (2.2252) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [140] [190/312] eta: 0:01:30 lr: 0.003273 min_lr: 0.003273 loss: 2.2038 (2.2179) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [140] [200/312] eta: 0:01:23 lr: 0.003272 min_lr: 0.003272 loss: 2.1349 (2.2181) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [140] [210/312] eta: 0:01:15 lr: 0.003272 min_lr: 0.003272 loss: 2.2479 (2.2152) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [140] [220/312] eta: 0:01:07 lr: 0.003272 min_lr: 0.003272 loss: 2.2224 (2.2118) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [140] [230/312] eta: 0:01:00 lr: 0.003271 min_lr: 0.003271 loss: 2.2844 (2.2156) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [140] [240/312] eta: 0:00:52 lr: 0.003271 min_lr: 0.003271 loss: 2.2784 (2.2203) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [140] [250/312] eta: 0:00:45 lr: 0.003271 min_lr: 0.003271 loss: 2.3388 (2.2265) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0004 max mem: 64948 Epoch: [140] [260/312] eta: 0:00:37 lr: 0.003270 min_lr: 0.003270 loss: 2.3388 (2.2244) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [140] [270/312] eta: 0:00:30 lr: 0.003270 min_lr: 0.003270 loss: 2.2510 (2.2249) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [140] [280/312] eta: 0:00:23 lr: 0.003269 min_lr: 0.003269 loss: 2.3160 (2.2214) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0010 max mem: 64948 Epoch: [140] [290/312] eta: 0:00:15 lr: 0.003269 min_lr: 0.003269 loss: 2.3378 (2.2250) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [140] [300/312] eta: 0:00:08 lr: 0.003269 min_lr: 0.003269 loss: 2.4178 (2.2313) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [140] [310/312] eta: 0:00:01 lr: 0.003268 min_lr: 0.003268 loss: 2.3169 (2.2297) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [140] [311/312] eta: 0:00:00 lr: 0.003268 min_lr: 0.003268 loss: 2.3463 (2.2305) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [140] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.003268 min_lr: 0.003268 loss: 2.3463 (2.2428) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6425 (0.6425) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.5034 data: 4.2814 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9488 (0.8922) acc1: 75.0000 (76.5120) acc5: 94.2708 (93.7280) time: 0.6522 data: 0.4758 max mem: 64948 Test: Total time: 0:00:06 (0.6776 s / it) * Acc@1 76.968 Acc@5 93.612 loss 0.890 Accuracy of the model on the 50000 test images: 77.0% Max accuracy: 76.97% Test: [0/9] eta: 0:00:41 loss: 0.6255 (0.6255) acc1: 83.8542 (83.8542) acc5: 95.0521 (95.0521) time: 4.6225 data: 4.4048 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7881 (0.7902) acc1: 78.6458 (78.6560) acc5: 95.0521 (94.4960) time: 0.6649 data: 0.4895 max mem: 64948 Test: Total time: 0:00:06 (0.6734 s / it) * Acc@1 79.190 Acc@5 94.666 loss 0.777 Accuracy of the model EMA on 50000 test images: 79.2% Max EMA accuracy: 79.19% Epoch: [141] [ 0/312] eta: 0:50:53 lr: 0.003268 min_lr: 0.003268 loss: 2.4579 (2.4579) weight_decay: 0.0500 (0.0500) time: 9.7884 data: 8.9355 max mem: 64948 Epoch: [141] [ 10/312] eta: 0:07:45 lr: 0.003268 min_lr: 0.003268 loss: 2.3506 (2.3107) weight_decay: 0.0500 (0.0500) time: 1.5400 data: 0.8154 max mem: 64948 Epoch: [141] [ 20/312] eta: 0:05:32 lr: 0.003268 min_lr: 0.003268 loss: 2.2310 (2.2517) weight_decay: 0.0500 (0.0500) time: 0.7059 data: 0.0019 max mem: 64948 Epoch: [141] [ 30/312] eta: 0:04:40 lr: 0.003267 min_lr: 0.003267 loss: 2.0459 (2.1834) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [141] [ 40/312] eta: 0:04:10 lr: 0.003267 min_lr: 0.003267 loss: 2.1578 (2.2312) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [141] [ 50/312] eta: 0:03:49 lr: 0.003266 min_lr: 0.003266 loss: 2.3926 (2.2643) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [141] [ 60/312] eta: 0:03:33 lr: 0.003266 min_lr: 0.003266 loss: 2.3663 (2.2583) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [141] [ 70/312] eta: 0:03:20 lr: 0.003266 min_lr: 0.003266 loss: 2.2594 (2.2433) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [141] [ 80/312] eta: 0:03:08 lr: 0.003265 min_lr: 0.003265 loss: 2.1553 (2.2453) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [141] [ 90/312] eta: 0:02:57 lr: 0.003265 min_lr: 0.003265 loss: 2.3724 (2.2446) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [141] [100/312] eta: 0:02:47 lr: 0.003265 min_lr: 0.003265 loss: 2.3724 (2.2378) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [141] [110/312] eta: 0:02:37 lr: 0.003264 min_lr: 0.003264 loss: 2.3475 (2.2429) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [141] [120/312] eta: 0:02:28 lr: 0.003264 min_lr: 0.003264 loss: 2.4770 (2.2568) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [141] [130/312] eta: 0:02:19 lr: 0.003264 min_lr: 0.003264 loss: 2.3432 (2.2653) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [141] [140/312] eta: 0:02:11 lr: 0.003263 min_lr: 0.003263 loss: 2.2914 (2.2616) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [141] [150/312] eta: 0:02:02 lr: 0.003263 min_lr: 0.003263 loss: 2.3188 (2.2570) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [141] [160/312] eta: 0:01:54 lr: 0.003262 min_lr: 0.003262 loss: 2.3188 (2.2587) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [141] [170/312] eta: 0:01:46 lr: 0.003262 min_lr: 0.003262 loss: 2.4116 (2.2688) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [141] [180/312] eta: 0:01:38 lr: 0.003262 min_lr: 0.003262 loss: 2.4358 (2.2721) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [141] [190/312] eta: 0:01:30 lr: 0.003261 min_lr: 0.003261 loss: 2.4544 (2.2739) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [141] [200/312] eta: 0:01:23 lr: 0.003261 min_lr: 0.003261 loss: 2.4649 (2.2794) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [141] [210/312] eta: 0:01:15 lr: 0.003261 min_lr: 0.003261 loss: 2.4509 (2.2879) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [141] [220/312] eta: 0:01:07 lr: 0.003260 min_lr: 0.003260 loss: 2.3003 (2.2802) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [141] [230/312] eta: 0:01:00 lr: 0.003260 min_lr: 0.003260 loss: 2.2081 (2.2742) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [141] [240/312] eta: 0:00:52 lr: 0.003260 min_lr: 0.003260 loss: 2.2228 (2.2721) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [141] [250/312] eta: 0:00:45 lr: 0.003259 min_lr: 0.003259 loss: 2.3612 (2.2763) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [141] [260/312] eta: 0:00:38 lr: 0.003259 min_lr: 0.003259 loss: 2.4206 (2.2778) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [141] [270/312] eta: 0:00:30 lr: 0.003258 min_lr: 0.003258 loss: 2.2227 (2.2749) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [141] [280/312] eta: 0:00:23 lr: 0.003258 min_lr: 0.003258 loss: 2.2227 (2.2771) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [141] [290/312] eta: 0:00:15 lr: 0.003258 min_lr: 0.003258 loss: 2.2844 (2.2710) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [141] [300/312] eta: 0:00:08 lr: 0.003257 min_lr: 0.003257 loss: 2.1625 (2.2662) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [141] [310/312] eta: 0:00:01 lr: 0.003257 min_lr: 0.003257 loss: 2.2636 (2.2690) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [141] [311/312] eta: 0:00:00 lr: 0.003257 min_lr: 0.003257 loss: 2.3123 (2.2697) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [141] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.003257 min_lr: 0.003257 loss: 2.3123 (2.2449) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6955 (0.6955) acc1: 80.9896 (80.9896) acc5: 95.8333 (95.8333) time: 4.7111 data: 4.4919 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9735 (0.9148) acc1: 77.0833 (75.6480) acc5: 94.5312 (93.7280) time: 0.6748 data: 0.4992 max mem: 64948 Test: Total time: 0:00:06 (0.6975 s / it) * Acc@1 76.812 Acc@5 93.594 loss 0.900 Accuracy of the model on the 50000 test images: 76.8% Max accuracy: 76.97% Test: [0/9] eta: 0:00:44 loss: 0.6237 (0.6237) acc1: 83.8542 (83.8542) acc5: 95.3125 (95.3125) time: 4.8906 data: 4.6725 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7863 (0.7880) acc1: 78.9062 (78.7840) acc5: 95.0521 (94.5280) time: 0.6954 data: 0.5193 max mem: 64948 Test: Total time: 0:00:06 (0.7044 s / it) * Acc@1 79.252 Acc@5 94.696 loss 0.775 Accuracy of the model EMA on 50000 test images: 79.3% Max EMA accuracy: 79.25% Epoch: [142] [ 0/312] eta: 0:44:02 lr: 0.003257 min_lr: 0.003257 loss: 1.5083 (1.5083) weight_decay: 0.0500 (0.0500) time: 8.4699 data: 7.1702 max mem: 64948 Epoch: [142] [ 10/312] eta: 0:07:29 lr: 0.003257 min_lr: 0.003257 loss: 2.3717 (2.2477) weight_decay: 0.0500 (0.0500) time: 1.4888 data: 0.6640 max mem: 64948 Epoch: [142] [ 20/312] eta: 0:05:23 lr: 0.003256 min_lr: 0.003256 loss: 2.2111 (2.1816) weight_decay: 0.0500 (0.0500) time: 0.7414 data: 0.0069 max mem: 64948 Epoch: [142] [ 30/312] eta: 0:04:35 lr: 0.003256 min_lr: 0.003256 loss: 2.2111 (2.2339) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [142] [ 40/312] eta: 0:04:06 lr: 0.003255 min_lr: 0.003255 loss: 2.2867 (2.2212) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [142] [ 50/312] eta: 0:03:46 lr: 0.003255 min_lr: 0.003255 loss: 2.3413 (2.2688) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [142] [ 60/312] eta: 0:03:31 lr: 0.003255 min_lr: 0.003255 loss: 2.4278 (2.2774) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [142] [ 70/312] eta: 0:03:17 lr: 0.003254 min_lr: 0.003254 loss: 2.3524 (2.2672) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [142] [ 80/312] eta: 0:03:06 lr: 0.003254 min_lr: 0.003254 loss: 2.3013 (2.2771) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [142] [ 90/312] eta: 0:02:55 lr: 0.003254 min_lr: 0.003254 loss: 2.2961 (2.2698) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [142] [100/312] eta: 0:02:45 lr: 0.003253 min_lr: 0.003253 loss: 2.2961 (2.2672) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [142] [110/312] eta: 0:02:36 lr: 0.003253 min_lr: 0.003253 loss: 2.2003 (2.2467) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [142] [120/312] eta: 0:02:27 lr: 0.003253 min_lr: 0.003253 loss: 2.1492 (2.2372) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [142] [130/312] eta: 0:02:18 lr: 0.003252 min_lr: 0.003252 loss: 2.2689 (2.2389) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [142] [140/312] eta: 0:02:10 lr: 0.003252 min_lr: 0.003252 loss: 2.2519 (2.2332) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [142] [150/312] eta: 0:02:01 lr: 0.003251 min_lr: 0.003251 loss: 2.1299 (2.2258) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [142] [160/312] eta: 0:01:53 lr: 0.003251 min_lr: 0.003251 loss: 2.2988 (2.2305) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [142] [170/312] eta: 0:01:45 lr: 0.003251 min_lr: 0.003251 loss: 2.3378 (2.2272) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [142] [180/312] eta: 0:01:38 lr: 0.003250 min_lr: 0.003250 loss: 2.2971 (2.2260) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [142] [190/312] eta: 0:01:30 lr: 0.003250 min_lr: 0.003250 loss: 2.3704 (2.2326) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [142] [200/312] eta: 0:01:22 lr: 0.003250 min_lr: 0.003250 loss: 2.3336 (2.2282) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [142] [210/312] eta: 0:01:15 lr: 0.003249 min_lr: 0.003249 loss: 2.2168 (2.2251) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [142] [220/312] eta: 0:01:07 lr: 0.003249 min_lr: 0.003249 loss: 2.0402 (2.2195) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [142] [230/312] eta: 0:01:00 lr: 0.003249 min_lr: 0.003249 loss: 1.9836 (2.2112) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [142] [240/312] eta: 0:00:52 lr: 0.003248 min_lr: 0.003248 loss: 2.2918 (2.2200) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [142] [250/312] eta: 0:00:45 lr: 0.003248 min_lr: 0.003248 loss: 2.3608 (2.2181) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [142] [260/312] eta: 0:00:37 lr: 0.003247 min_lr: 0.003247 loss: 2.1836 (2.2151) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [142] [270/312] eta: 0:00:30 lr: 0.003247 min_lr: 0.003247 loss: 2.2680 (2.2137) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [142] [280/312] eta: 0:00:23 lr: 0.003247 min_lr: 0.003247 loss: 2.3332 (2.2166) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [142] [290/312] eta: 0:00:15 lr: 0.003246 min_lr: 0.003246 loss: 2.2735 (2.2205) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [142] [300/312] eta: 0:00:08 lr: 0.003246 min_lr: 0.003246 loss: 2.2602 (2.2191) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [142] [310/312] eta: 0:00:01 lr: 0.003246 min_lr: 0.003246 loss: 2.2853 (2.2220) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [142] [311/312] eta: 0:00:00 lr: 0.003246 min_lr: 0.003246 loss: 2.2853 (2.2210) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [142] Total time: 0:03:46 (0.7251 s / it) Averaged stats: lr: 0.003246 min_lr: 0.003246 loss: 2.2853 (2.2397) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7074 (0.7074) acc1: 83.3333 (83.3333) acc5: 95.5729 (95.5729) time: 4.4916 data: 4.2721 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9589 (0.9475) acc1: 76.5625 (75.9360) acc5: 94.5312 (93.1520) time: 0.6508 data: 0.4748 max mem: 64948 Test: Total time: 0:00:06 (0.6744 s / it) * Acc@1 76.994 Acc@5 93.374 loss 0.923 Accuracy of the model on the 50000 test images: 77.0% Max accuracy: 76.99% Test: [0/9] eta: 0:00:41 loss: 0.6215 (0.6215) acc1: 83.8542 (83.8542) acc5: 95.3125 (95.3125) time: 4.5728 data: 4.3537 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7854 (0.7861) acc1: 79.1667 (78.8800) acc5: 95.3125 (94.5920) time: 0.6594 data: 0.4838 max mem: 64948 Test: Total time: 0:00:06 (0.6692 s / it) * Acc@1 79.290 Acc@5 94.724 loss 0.772 Accuracy of the model EMA on 50000 test images: 79.3% Max EMA accuracy: 79.29% Epoch: [143] [ 0/312] eta: 0:48:19 lr: 0.003246 min_lr: 0.003246 loss: 1.9652 (1.9652) weight_decay: 0.0500 (0.0500) time: 9.2923 data: 7.4184 max mem: 64948 Epoch: [143] [ 10/312] eta: 0:07:33 lr: 0.003245 min_lr: 0.003245 loss: 2.5072 (2.2953) weight_decay: 0.0500 (0.0500) time: 1.5018 data: 0.6748 max mem: 64948 Epoch: [143] [ 20/312] eta: 0:05:26 lr: 0.003245 min_lr: 0.003245 loss: 2.4019 (2.2252) weight_decay: 0.0500 (0.0500) time: 0.7096 data: 0.0004 max mem: 64948 Epoch: [143] [ 30/312] eta: 0:04:37 lr: 0.003244 min_lr: 0.003244 loss: 2.3567 (2.2658) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [143] [ 40/312] eta: 0:04:08 lr: 0.003244 min_lr: 0.003244 loss: 2.3567 (2.2520) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [143] [ 50/312] eta: 0:03:47 lr: 0.003244 min_lr: 0.003244 loss: 2.2307 (2.2553) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [143] [ 60/312] eta: 0:03:31 lr: 0.003243 min_lr: 0.003243 loss: 2.1642 (2.2191) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [143] [ 70/312] eta: 0:03:18 lr: 0.003243 min_lr: 0.003243 loss: 2.1142 (2.2354) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [143] [ 80/312] eta: 0:03:06 lr: 0.003243 min_lr: 0.003243 loss: 2.2245 (2.2214) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [143] [ 90/312] eta: 0:02:56 lr: 0.003242 min_lr: 0.003242 loss: 2.2245 (2.2296) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [143] [100/312] eta: 0:02:46 lr: 0.003242 min_lr: 0.003242 loss: 2.3741 (2.2462) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [143] [110/312] eta: 0:02:36 lr: 0.003242 min_lr: 0.003242 loss: 2.4103 (2.2588) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [143] [120/312] eta: 0:02:27 lr: 0.003241 min_lr: 0.003241 loss: 2.4103 (2.2632) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [143] [130/312] eta: 0:02:18 lr: 0.003241 min_lr: 0.003241 loss: 2.3058 (2.2615) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [143] [140/312] eta: 0:02:10 lr: 0.003240 min_lr: 0.003240 loss: 2.2898 (2.2603) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [143] [150/312] eta: 0:02:02 lr: 0.003240 min_lr: 0.003240 loss: 2.2700 (2.2536) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [143] [160/312] eta: 0:01:54 lr: 0.003240 min_lr: 0.003240 loss: 2.2050 (2.2508) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [143] [170/312] eta: 0:01:46 lr: 0.003239 min_lr: 0.003239 loss: 2.2774 (2.2520) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [143] [180/312] eta: 0:01:38 lr: 0.003239 min_lr: 0.003239 loss: 2.3012 (2.2495) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [143] [190/312] eta: 0:01:30 lr: 0.003239 min_lr: 0.003239 loss: 2.3012 (2.2510) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [143] [200/312] eta: 0:01:22 lr: 0.003238 min_lr: 0.003238 loss: 2.2660 (2.2441) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [143] [210/312] eta: 0:01:15 lr: 0.003238 min_lr: 0.003238 loss: 2.2228 (2.2398) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [143] [220/312] eta: 0:01:07 lr: 0.003237 min_lr: 0.003237 loss: 2.2681 (2.2388) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [143] [230/312] eta: 0:01:00 lr: 0.003237 min_lr: 0.003237 loss: 2.1917 (2.2334) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [143] [240/312] eta: 0:00:52 lr: 0.003237 min_lr: 0.003237 loss: 2.1740 (2.2310) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [143] [250/312] eta: 0:00:45 lr: 0.003236 min_lr: 0.003236 loss: 2.1116 (2.2224) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [143] [260/312] eta: 0:00:37 lr: 0.003236 min_lr: 0.003236 loss: 2.1429 (2.2233) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [143] [270/312] eta: 0:00:30 lr: 0.003236 min_lr: 0.003236 loss: 2.3108 (2.2243) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [143] [280/312] eta: 0:00:23 lr: 0.003235 min_lr: 0.003235 loss: 2.2894 (2.2233) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [143] [290/312] eta: 0:00:15 lr: 0.003235 min_lr: 0.003235 loss: 2.1195 (2.2204) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [143] [300/312] eta: 0:00:08 lr: 0.003235 min_lr: 0.003235 loss: 2.3662 (2.2224) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [143] [310/312] eta: 0:00:01 lr: 0.003234 min_lr: 0.003234 loss: 2.1564 (2.2192) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [143] [311/312] eta: 0:00:00 lr: 0.003234 min_lr: 0.003234 loss: 2.1617 (2.2190) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [143] Total time: 0:03:46 (0.7262 s / it) Averaged stats: lr: 0.003234 min_lr: 0.003234 loss: 2.1617 (2.2247) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6619 (0.6619) acc1: 82.5521 (82.5521) acc5: 96.8750 (96.8750) time: 4.5700 data: 4.3503 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9694 (0.9264) acc1: 76.5625 (75.1360) acc5: 93.4896 (93.6960) time: 0.6591 data: 0.4835 max mem: 64948 Test: Total time: 0:00:06 (0.6810 s / it) * Acc@1 76.838 Acc@5 93.562 loss 0.909 Accuracy of the model on the 50000 test images: 76.8% Max accuracy: 76.99% Test: [0/9] eta: 0:00:42 loss: 0.6201 (0.6201) acc1: 83.8542 (83.8542) acc5: 95.5729 (95.5729) time: 4.7611 data: 4.5460 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7835 (0.7840) acc1: 79.4271 (78.7520) acc5: 95.3125 (94.6560) time: 0.6804 data: 0.5052 max mem: 64948 Test: Total time: 0:00:06 (0.6891 s / it) * Acc@1 79.358 Acc@5 94.758 loss 0.770 Accuracy of the model EMA on 50000 test images: 79.4% Max EMA accuracy: 79.36% Epoch: [144] [ 0/312] eta: 0:44:54 lr: 0.003234 min_lr: 0.003234 loss: 2.3122 (2.3122) weight_decay: 0.0500 (0.0500) time: 8.6370 data: 7.2149 max mem: 64948 Epoch: [144] [ 10/312] eta: 0:07:25 lr: 0.003234 min_lr: 0.003234 loss: 2.4296 (2.1739) weight_decay: 0.0500 (0.0500) time: 1.4750 data: 0.6563 max mem: 64948 Epoch: [144] [ 20/312] eta: 0:05:22 lr: 0.003233 min_lr: 0.003233 loss: 2.4391 (2.2827) weight_decay: 0.0500 (0.0500) time: 0.7265 data: 0.0004 max mem: 64948 Epoch: [144] [ 30/312] eta: 0:04:34 lr: 0.003233 min_lr: 0.003233 loss: 2.4521 (2.3062) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [144] [ 40/312] eta: 0:04:05 lr: 0.003233 min_lr: 0.003233 loss: 2.2952 (2.2783) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [144] [ 50/312] eta: 0:03:46 lr: 0.003232 min_lr: 0.003232 loss: 2.3467 (2.2678) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [144] [ 60/312] eta: 0:03:30 lr: 0.003232 min_lr: 0.003232 loss: 1.9949 (2.2331) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [144] [ 70/312] eta: 0:03:17 lr: 0.003231 min_lr: 0.003231 loss: 2.3409 (2.2573) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [144] [ 80/312] eta: 0:03:06 lr: 0.003231 min_lr: 0.003231 loss: 2.3716 (2.2656) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [144] [ 90/312] eta: 0:02:55 lr: 0.003231 min_lr: 0.003231 loss: 2.2669 (2.2558) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [144] [100/312] eta: 0:02:45 lr: 0.003230 min_lr: 0.003230 loss: 2.2669 (2.2501) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [144] [110/312] eta: 0:02:36 lr: 0.003230 min_lr: 0.003230 loss: 2.3721 (2.2589) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [144] [120/312] eta: 0:02:27 lr: 0.003230 min_lr: 0.003230 loss: 2.3721 (2.2600) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [144] [130/312] eta: 0:02:18 lr: 0.003229 min_lr: 0.003229 loss: 2.2754 (2.2503) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [144] [140/312] eta: 0:02:10 lr: 0.003229 min_lr: 0.003229 loss: 2.1109 (2.2491) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [144] [150/312] eta: 0:02:01 lr: 0.003229 min_lr: 0.003229 loss: 2.3365 (2.2575) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [144] [160/312] eta: 0:01:53 lr: 0.003228 min_lr: 0.003228 loss: 2.3629 (2.2641) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [144] [170/312] eta: 0:01:45 lr: 0.003228 min_lr: 0.003228 loss: 2.3415 (2.2701) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [144] [180/312] eta: 0:01:38 lr: 0.003227 min_lr: 0.003227 loss: 2.2553 (2.2645) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [144] [190/312] eta: 0:01:30 lr: 0.003227 min_lr: 0.003227 loss: 2.3006 (2.2666) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [144] [200/312] eta: 0:01:22 lr: 0.003227 min_lr: 0.003227 loss: 2.3215 (2.2631) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [144] [210/312] eta: 0:01:15 lr: 0.003226 min_lr: 0.003226 loss: 2.3215 (2.2644) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [144] [220/312] eta: 0:01:07 lr: 0.003226 min_lr: 0.003226 loss: 2.3994 (2.2718) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [144] [230/312] eta: 0:01:00 lr: 0.003226 min_lr: 0.003226 loss: 2.3096 (2.2635) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [144] [240/312] eta: 0:00:52 lr: 0.003225 min_lr: 0.003225 loss: 2.1360 (2.2618) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [144] [250/312] eta: 0:00:45 lr: 0.003225 min_lr: 0.003225 loss: 2.3026 (2.2645) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [144] [260/312] eta: 0:00:37 lr: 0.003224 min_lr: 0.003224 loss: 2.2914 (2.2637) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [144] [270/312] eta: 0:00:30 lr: 0.003224 min_lr: 0.003224 loss: 2.2049 (2.2575) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [144] [280/312] eta: 0:00:23 lr: 0.003224 min_lr: 0.003224 loss: 2.2710 (2.2590) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [144] [290/312] eta: 0:00:15 lr: 0.003223 min_lr: 0.003223 loss: 2.3165 (2.2574) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [144] [300/312] eta: 0:00:08 lr: 0.003223 min_lr: 0.003223 loss: 2.2190 (2.2565) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [144] [310/312] eta: 0:00:01 lr: 0.003223 min_lr: 0.003223 loss: 2.4337 (2.2574) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [144] [311/312] eta: 0:00:00 lr: 0.003223 min_lr: 0.003223 loss: 2.3327 (2.2577) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [144] Total time: 0:03:46 (0.7254 s / it) Averaged stats: lr: 0.003223 min_lr: 0.003223 loss: 2.3327 (2.2312) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7291 (0.7291) acc1: 81.7708 (81.7708) acc5: 95.3125 (95.3125) time: 4.6631 data: 4.4578 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0090 (0.9995) acc1: 75.0000 (74.1440) acc5: 93.4896 (92.6720) time: 0.6696 data: 0.4954 max mem: 64948 Test: Total time: 0:00:06 (0.6939 s / it) * Acc@1 75.530 Acc@5 92.728 loss 0.967 Accuracy of the model on the 50000 test images: 75.5% Max accuracy: 76.99% Test: [0/9] eta: 0:00:41 loss: 0.6181 (0.6181) acc1: 84.1146 (84.1146) acc5: 95.8333 (95.8333) time: 4.5810 data: 4.3628 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7825 (0.7821) acc1: 79.9479 (79.0080) acc5: 95.3125 (94.7520) time: 0.6724 data: 0.4969 max mem: 64948 Test: Total time: 0:00:06 (0.6801 s / it) * Acc@1 79.420 Acc@5 94.796 loss 0.768 Accuracy of the model EMA on 50000 test images: 79.4% Max EMA accuracy: 79.42% Epoch: [145] [ 0/312] eta: 0:49:14 lr: 0.003223 min_lr: 0.003223 loss: 1.6290 (1.6290) weight_decay: 0.0500 (0.0500) time: 9.4707 data: 7.4722 max mem: 64948 Epoch: [145] [ 10/312] eta: 0:07:38 lr: 0.003222 min_lr: 0.003222 loss: 1.8529 (1.9636) weight_decay: 0.0500 (0.0500) time: 1.5198 data: 0.6797 max mem: 64948 Epoch: [145] [ 20/312] eta: 0:05:29 lr: 0.003222 min_lr: 0.003222 loss: 2.2261 (2.1293) weight_decay: 0.0500 (0.0500) time: 0.7096 data: 0.0004 max mem: 64948 Epoch: [145] [ 30/312] eta: 0:04:39 lr: 0.003221 min_lr: 0.003221 loss: 2.3260 (2.1777) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [145] [ 40/312] eta: 0:04:09 lr: 0.003221 min_lr: 0.003221 loss: 2.3568 (2.1868) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [145] [ 50/312] eta: 0:03:49 lr: 0.003221 min_lr: 0.003221 loss: 2.3568 (2.2069) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [145] [ 60/312] eta: 0:03:33 lr: 0.003220 min_lr: 0.003220 loss: 2.3887 (2.2303) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [145] [ 70/312] eta: 0:03:19 lr: 0.003220 min_lr: 0.003220 loss: 2.4064 (2.2595) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [145] [ 80/312] eta: 0:03:07 lr: 0.003220 min_lr: 0.003220 loss: 2.4177 (2.2556) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [145] [ 90/312] eta: 0:02:56 lr: 0.003219 min_lr: 0.003219 loss: 2.1265 (2.2264) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [145] [100/312] eta: 0:02:46 lr: 0.003219 min_lr: 0.003219 loss: 2.1002 (2.2202) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [145] [110/312] eta: 0:02:37 lr: 0.003218 min_lr: 0.003218 loss: 2.2653 (2.2126) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [145] [120/312] eta: 0:02:28 lr: 0.003218 min_lr: 0.003218 loss: 2.2132 (2.2080) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [145] [130/312] eta: 0:02:19 lr: 0.003218 min_lr: 0.003218 loss: 2.0950 (2.2010) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [145] [140/312] eta: 0:02:10 lr: 0.003217 min_lr: 0.003217 loss: 1.9803 (2.1829) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [145] [150/312] eta: 0:02:02 lr: 0.003217 min_lr: 0.003217 loss: 2.2018 (2.2035) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [145] [160/312] eta: 0:01:54 lr: 0.003217 min_lr: 0.003217 loss: 2.5405 (2.2109) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [145] [170/312] eta: 0:01:46 lr: 0.003216 min_lr: 0.003216 loss: 2.3368 (2.2076) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [145] [180/312] eta: 0:01:38 lr: 0.003216 min_lr: 0.003216 loss: 2.1489 (2.2076) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [145] [190/312] eta: 0:01:30 lr: 0.003215 min_lr: 0.003215 loss: 2.3175 (2.2123) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [145] [200/312] eta: 0:01:22 lr: 0.003215 min_lr: 0.003215 loss: 2.3322 (2.2091) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [145] [210/312] eta: 0:01:15 lr: 0.003215 min_lr: 0.003215 loss: 2.2546 (2.2045) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [145] [220/312] eta: 0:01:07 lr: 0.003214 min_lr: 0.003214 loss: 2.2186 (2.2040) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [145] [230/312] eta: 0:01:00 lr: 0.003214 min_lr: 0.003214 loss: 2.2186 (2.2047) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [145] [240/312] eta: 0:00:52 lr: 0.003214 min_lr: 0.003214 loss: 2.2566 (2.2086) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [145] [250/312] eta: 0:00:45 lr: 0.003213 min_lr: 0.003213 loss: 2.3647 (2.2154) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [145] [260/312] eta: 0:00:37 lr: 0.003213 min_lr: 0.003213 loss: 2.3086 (2.2114) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [145] [270/312] eta: 0:00:30 lr: 0.003213 min_lr: 0.003213 loss: 2.2968 (2.2126) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [145] [280/312] eta: 0:00:23 lr: 0.003212 min_lr: 0.003212 loss: 2.2400 (2.2093) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0009 max mem: 64948 Epoch: [145] [290/312] eta: 0:00:15 lr: 0.003212 min_lr: 0.003212 loss: 1.8826 (2.2051) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0008 max mem: 64948 Epoch: [145] [300/312] eta: 0:00:08 lr: 0.003211 min_lr: 0.003211 loss: 1.9438 (2.1983) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [145] [310/312] eta: 0:00:01 lr: 0.003211 min_lr: 0.003211 loss: 2.1200 (2.2004) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [145] [311/312] eta: 0:00:00 lr: 0.003211 min_lr: 0.003211 loss: 2.1456 (2.2008) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [145] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.003211 min_lr: 0.003211 loss: 2.1456 (2.2320) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6819 (0.6819) acc1: 83.0729 (83.0729) acc5: 96.0938 (96.0938) time: 4.5551 data: 4.3303 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0071 (0.9393) acc1: 75.5208 (75.9360) acc5: 93.4896 (92.9920) time: 0.6579 data: 0.4812 max mem: 64948 Test: Total time: 0:00:06 (0.6817 s / it) * Acc@1 77.294 Acc@5 93.742 loss 0.895 Accuracy of the model on the 50000 test images: 77.3% Max accuracy: 77.29% Test: [0/9] eta: 0:00:40 loss: 0.6162 (0.6162) acc1: 83.8542 (83.8542) acc5: 95.8333 (95.8333) time: 4.4799 data: 4.2753 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7813 (0.7801) acc1: 80.2083 (78.9440) acc5: 95.3125 (94.7200) time: 0.6491 data: 0.4751 max mem: 64948 Test: Total time: 0:00:05 (0.6569 s / it) * Acc@1 79.478 Acc@5 94.822 loss 0.765 Accuracy of the model EMA on 50000 test images: 79.5% Max EMA accuracy: 79.48% Epoch: [146] [ 0/312] eta: 0:49:25 lr: 0.003211 min_lr: 0.003211 loss: 2.8202 (2.8202) weight_decay: 0.0500 (0.0500) time: 9.5058 data: 8.6990 max mem: 64948 Epoch: [146] [ 10/312] eta: 0:07:38 lr: 0.003211 min_lr: 0.003211 loss: 2.3559 (2.3264) weight_decay: 0.0500 (0.0500) time: 1.5182 data: 0.7911 max mem: 64948 Epoch: [146] [ 20/312] eta: 0:05:29 lr: 0.003210 min_lr: 0.003210 loss: 2.1345 (2.1611) weight_decay: 0.0500 (0.0500) time: 0.7097 data: 0.0003 max mem: 64948 Epoch: [146] [ 30/312] eta: 0:04:38 lr: 0.003210 min_lr: 0.003210 loss: 2.1089 (2.2124) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [146] [ 40/312] eta: 0:04:09 lr: 0.003209 min_lr: 0.003209 loss: 2.3746 (2.2409) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [146] [ 50/312] eta: 0:03:49 lr: 0.003209 min_lr: 0.003209 loss: 2.3103 (2.2120) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0003 max mem: 64948 Epoch: [146] [ 60/312] eta: 0:03:33 lr: 0.003209 min_lr: 0.003209 loss: 2.1787 (2.2295) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [146] [ 70/312] eta: 0:03:19 lr: 0.003208 min_lr: 0.003208 loss: 2.2207 (2.2195) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [146] [ 80/312] eta: 0:03:07 lr: 0.003208 min_lr: 0.003208 loss: 2.2207 (2.2164) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [146] [ 90/312] eta: 0:02:56 lr: 0.003208 min_lr: 0.003208 loss: 2.2362 (2.2100) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [146] [100/312] eta: 0:02:46 lr: 0.003207 min_lr: 0.003207 loss: 2.1308 (2.1906) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [146] [110/312] eta: 0:02:37 lr: 0.003207 min_lr: 0.003207 loss: 2.1476 (2.2047) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [146] [120/312] eta: 0:02:28 lr: 0.003206 min_lr: 0.003206 loss: 2.4633 (2.2227) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [146] [130/312] eta: 0:02:19 lr: 0.003206 min_lr: 0.003206 loss: 2.3772 (2.2283) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [146] [140/312] eta: 0:02:10 lr: 0.003206 min_lr: 0.003206 loss: 2.2977 (2.2266) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [146] [150/312] eta: 0:02:02 lr: 0.003205 min_lr: 0.003205 loss: 2.3956 (2.2424) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [146] [160/312] eta: 0:01:54 lr: 0.003205 min_lr: 0.003205 loss: 2.3183 (2.2290) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [146] [170/312] eta: 0:01:46 lr: 0.003205 min_lr: 0.003205 loss: 2.2384 (2.2390) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [146] [180/312] eta: 0:01:38 lr: 0.003204 min_lr: 0.003204 loss: 2.4206 (2.2365) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [146] [190/312] eta: 0:01:30 lr: 0.003204 min_lr: 0.003204 loss: 2.3423 (2.2345) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [146] [200/312] eta: 0:01:22 lr: 0.003203 min_lr: 0.003203 loss: 2.1954 (2.2283) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [146] [210/312] eta: 0:01:15 lr: 0.003203 min_lr: 0.003203 loss: 2.1887 (2.2275) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [146] [220/312] eta: 0:01:07 lr: 0.003203 min_lr: 0.003203 loss: 2.2363 (2.2310) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [146] [230/312] eta: 0:01:00 lr: 0.003202 min_lr: 0.003202 loss: 2.2363 (2.2332) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [146] [240/312] eta: 0:00:52 lr: 0.003202 min_lr: 0.003202 loss: 2.1333 (2.2301) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [146] [250/312] eta: 0:00:45 lr: 0.003202 min_lr: 0.003202 loss: 2.1059 (2.2260) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [146] [260/312] eta: 0:00:37 lr: 0.003201 min_lr: 0.003201 loss: 1.9138 (2.2187) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [146] [270/312] eta: 0:00:30 lr: 0.003201 min_lr: 0.003201 loss: 2.2504 (2.2185) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [146] [280/312] eta: 0:00:23 lr: 0.003200 min_lr: 0.003200 loss: 2.3411 (2.2206) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0009 max mem: 64948 Epoch: [146] [290/312] eta: 0:00:15 lr: 0.003200 min_lr: 0.003200 loss: 2.3870 (2.2181) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0008 max mem: 64948 Epoch: [146] [300/312] eta: 0:00:08 lr: 0.003200 min_lr: 0.003200 loss: 2.1690 (2.2201) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [146] [310/312] eta: 0:00:01 lr: 0.003199 min_lr: 0.003199 loss: 2.1690 (2.2152) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [146] [311/312] eta: 0:00:00 lr: 0.003199 min_lr: 0.003199 loss: 2.1690 (2.2125) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [146] Total time: 0:03:46 (0.7266 s / it) Averaged stats: lr: 0.003199 min_lr: 0.003199 loss: 2.1690 (2.2255) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6749 (0.6749) acc1: 83.5938 (83.5938) acc5: 95.3125 (95.3125) time: 4.7447 data: 4.5282 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9601 (0.9435) acc1: 77.6042 (75.5520) acc5: 94.5312 (93.5360) time: 0.6785 data: 0.5032 max mem: 64948 Test: Total time: 0:00:06 (0.7128 s / it) * Acc@1 76.496 Acc@5 93.346 loss 0.909 Accuracy of the model on the 50000 test images: 76.5% Max accuracy: 77.29% Test: [0/9] eta: 0:00:45 loss: 0.6144 (0.6144) acc1: 83.8542 (83.8542) acc5: 96.0938 (96.0938) time: 5.1088 data: 4.8911 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7797 (0.7781) acc1: 80.4688 (78.9760) acc5: 95.3125 (94.6560) time: 0.7237 data: 0.5435 max mem: 64948 Test: Total time: 0:00:06 (0.7322 s / it) * Acc@1 79.522 Acc@5 94.858 loss 0.763 Accuracy of the model EMA on 50000 test images: 79.5% Max EMA accuracy: 79.52% Epoch: [147] [ 0/312] eta: 0:48:54 lr: 0.003199 min_lr: 0.003199 loss: 1.6950 (1.6950) weight_decay: 0.0500 (0.0500) time: 9.4064 data: 8.6174 max mem: 64948 Epoch: [147] [ 10/312] eta: 0:07:39 lr: 0.003199 min_lr: 0.003199 loss: 1.9132 (1.9635) weight_decay: 0.0500 (0.0500) time: 1.5201 data: 0.7838 max mem: 64948 Epoch: [147] [ 20/312] eta: 0:05:29 lr: 0.003199 min_lr: 0.003199 loss: 2.0755 (2.1132) weight_decay: 0.0500 (0.0500) time: 0.7132 data: 0.0004 max mem: 64948 Epoch: [147] [ 30/312] eta: 0:04:38 lr: 0.003198 min_lr: 0.003198 loss: 2.2352 (2.1197) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [147] [ 40/312] eta: 0:04:09 lr: 0.003198 min_lr: 0.003198 loss: 2.2081 (2.1172) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [147] [ 50/312] eta: 0:03:48 lr: 0.003197 min_lr: 0.003197 loss: 2.1447 (2.1339) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [147] [ 60/312] eta: 0:03:32 lr: 0.003197 min_lr: 0.003197 loss: 2.1782 (2.1701) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [147] [ 70/312] eta: 0:03:19 lr: 0.003197 min_lr: 0.003197 loss: 2.3981 (2.1898) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [147] [ 80/312] eta: 0:03:07 lr: 0.003196 min_lr: 0.003196 loss: 2.4133 (2.2045) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [147] [ 90/312] eta: 0:02:56 lr: 0.003196 min_lr: 0.003196 loss: 2.3787 (2.2077) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [147] [100/312] eta: 0:02:46 lr: 0.003196 min_lr: 0.003196 loss: 2.3132 (2.2089) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [147] [110/312] eta: 0:02:37 lr: 0.003195 min_lr: 0.003195 loss: 2.3687 (2.2212) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [147] [120/312] eta: 0:02:28 lr: 0.003195 min_lr: 0.003195 loss: 2.3067 (2.2125) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [147] [130/312] eta: 0:02:19 lr: 0.003194 min_lr: 0.003194 loss: 2.2851 (2.2185) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [147] [140/312] eta: 0:02:10 lr: 0.003194 min_lr: 0.003194 loss: 2.3663 (2.2208) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [147] [150/312] eta: 0:02:02 lr: 0.003194 min_lr: 0.003194 loss: 2.3663 (2.2223) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [147] [160/312] eta: 0:01:54 lr: 0.003193 min_lr: 0.003193 loss: 2.4448 (2.2279) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [147] [170/312] eta: 0:01:46 lr: 0.003193 min_lr: 0.003193 loss: 2.3587 (2.2358) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [147] [180/312] eta: 0:01:38 lr: 0.003193 min_lr: 0.003193 loss: 2.1231 (2.2230) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [147] [190/312] eta: 0:01:30 lr: 0.003192 min_lr: 0.003192 loss: 2.1091 (2.2245) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [147] [200/312] eta: 0:01:22 lr: 0.003192 min_lr: 0.003192 loss: 2.1744 (2.2225) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [147] [210/312] eta: 0:01:15 lr: 0.003191 min_lr: 0.003191 loss: 2.1628 (2.2198) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [147] [220/312] eta: 0:01:07 lr: 0.003191 min_lr: 0.003191 loss: 2.1683 (2.2196) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [147] [230/312] eta: 0:01:00 lr: 0.003191 min_lr: 0.003191 loss: 2.2903 (2.2189) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [147] [240/312] eta: 0:00:52 lr: 0.003190 min_lr: 0.003190 loss: 2.2903 (2.2173) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [147] [250/312] eta: 0:00:45 lr: 0.003190 min_lr: 0.003190 loss: 2.3388 (2.2216) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [147] [260/312] eta: 0:00:37 lr: 0.003190 min_lr: 0.003190 loss: 2.4075 (2.2249) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [147] [270/312] eta: 0:00:30 lr: 0.003189 min_lr: 0.003189 loss: 2.3312 (2.2235) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [147] [280/312] eta: 0:00:23 lr: 0.003189 min_lr: 0.003189 loss: 2.3126 (2.2297) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [147] [290/312] eta: 0:00:15 lr: 0.003188 min_lr: 0.003188 loss: 2.3169 (2.2316) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [147] [300/312] eta: 0:00:08 lr: 0.003188 min_lr: 0.003188 loss: 2.0822 (2.2296) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [147] [310/312] eta: 0:00:01 lr: 0.003188 min_lr: 0.003188 loss: 2.4177 (2.2354) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [147] [311/312] eta: 0:00:00 lr: 0.003188 min_lr: 0.003188 loss: 2.4177 (2.2335) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [147] Total time: 0:03:46 (0.7264 s / it) Averaged stats: lr: 0.003188 min_lr: 0.003188 loss: 2.4177 (2.2175) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6625 (0.6625) acc1: 84.6354 (84.6354) acc5: 95.3125 (95.3125) time: 4.7136 data: 4.4880 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0428 (0.9674) acc1: 76.0417 (75.9360) acc5: 92.7083 (93.3760) time: 0.6750 data: 0.4988 max mem: 64948 Test: Total time: 0:00:06 (0.6969 s / it) * Acc@1 76.528 Acc@5 93.242 loss 0.934 Accuracy of the model on the 50000 test images: 76.5% Max accuracy: 77.29% Test: [0/9] eta: 0:00:42 loss: 0.6128 (0.6128) acc1: 84.3750 (84.3750) acc5: 96.3542 (96.3542) time: 4.7209 data: 4.5028 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7779 (0.7764) acc1: 80.7292 (79.1680) acc5: 95.5729 (94.7520) time: 0.6758 data: 0.5004 max mem: 64948 Test: Total time: 0:00:06 (0.6852 s / it) * Acc@1 79.580 Acc@5 94.888 loss 0.761 Accuracy of the model EMA on 50000 test images: 79.6% Max EMA accuracy: 79.58% Epoch: [148] [ 0/312] eta: 0:49:55 lr: 0.003188 min_lr: 0.003188 loss: 2.4068 (2.4068) weight_decay: 0.0500 (0.0500) time: 9.6000 data: 7.1216 max mem: 64948 Epoch: [148] [ 10/312] eta: 0:07:40 lr: 0.003187 min_lr: 0.003187 loss: 2.2888 (2.2669) weight_decay: 0.0500 (0.0500) time: 1.5249 data: 0.6478 max mem: 64948 Epoch: [148] [ 20/312] eta: 0:05:29 lr: 0.003187 min_lr: 0.003187 loss: 2.3715 (2.3565) weight_decay: 0.0500 (0.0500) time: 0.7052 data: 0.0004 max mem: 64948 Epoch: [148] [ 30/312] eta: 0:04:38 lr: 0.003186 min_lr: 0.003186 loss: 2.2196 (2.2703) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [148] [ 40/312] eta: 0:04:09 lr: 0.003186 min_lr: 0.003186 loss: 2.1303 (2.2537) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [148] [ 50/312] eta: 0:03:49 lr: 0.003186 min_lr: 0.003186 loss: 2.3660 (2.2587) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [148] [ 60/312] eta: 0:03:32 lr: 0.003185 min_lr: 0.003185 loss: 2.3176 (2.2379) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [148] [ 70/312] eta: 0:03:19 lr: 0.003185 min_lr: 0.003185 loss: 2.1986 (2.2169) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [148] [ 80/312] eta: 0:03:07 lr: 0.003185 min_lr: 0.003185 loss: 2.1489 (2.2271) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [148] [ 90/312] eta: 0:02:56 lr: 0.003184 min_lr: 0.003184 loss: 2.2291 (2.2310) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [148] [100/312] eta: 0:02:46 lr: 0.003184 min_lr: 0.003184 loss: 2.2460 (2.2229) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [148] [110/312] eta: 0:02:36 lr: 0.003183 min_lr: 0.003183 loss: 2.3266 (2.2342) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [148] [120/312] eta: 0:02:27 lr: 0.003183 min_lr: 0.003183 loss: 2.3279 (2.2436) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [148] [130/312] eta: 0:02:19 lr: 0.003183 min_lr: 0.003183 loss: 2.1200 (2.2285) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [148] [140/312] eta: 0:02:10 lr: 0.003182 min_lr: 0.003182 loss: 2.3029 (2.2426) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [148] [150/312] eta: 0:02:02 lr: 0.003182 min_lr: 0.003182 loss: 2.3703 (2.2433) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [148] [160/312] eta: 0:01:54 lr: 0.003182 min_lr: 0.003182 loss: 2.4007 (2.2499) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [148] [170/312] eta: 0:01:46 lr: 0.003181 min_lr: 0.003181 loss: 2.3119 (2.2471) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [148] [180/312] eta: 0:01:38 lr: 0.003181 min_lr: 0.003181 loss: 2.2891 (2.2438) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [148] [190/312] eta: 0:01:30 lr: 0.003180 min_lr: 0.003180 loss: 2.3464 (2.2517) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [148] [200/312] eta: 0:01:22 lr: 0.003180 min_lr: 0.003180 loss: 2.3464 (2.2502) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [148] [210/312] eta: 0:01:15 lr: 0.003180 min_lr: 0.003180 loss: 2.1868 (2.2467) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [148] [220/312] eta: 0:01:07 lr: 0.003179 min_lr: 0.003179 loss: 2.1868 (2.2483) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [148] [230/312] eta: 0:01:00 lr: 0.003179 min_lr: 0.003179 loss: 2.3684 (2.2502) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [148] [240/312] eta: 0:00:52 lr: 0.003179 min_lr: 0.003179 loss: 2.3684 (2.2523) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [148] [250/312] eta: 0:00:45 lr: 0.003178 min_lr: 0.003178 loss: 2.2821 (2.2465) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [148] [260/312] eta: 0:00:37 lr: 0.003178 min_lr: 0.003178 loss: 2.1506 (2.2402) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [148] [270/312] eta: 0:00:30 lr: 0.003177 min_lr: 0.003177 loss: 2.1506 (2.2418) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [148] [280/312] eta: 0:00:23 lr: 0.003177 min_lr: 0.003177 loss: 2.3241 (2.2483) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [148] [290/312] eta: 0:00:15 lr: 0.003177 min_lr: 0.003177 loss: 2.3241 (2.2498) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [148] [300/312] eta: 0:00:08 lr: 0.003176 min_lr: 0.003176 loss: 2.1960 (2.2469) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [148] [310/312] eta: 0:00:01 lr: 0.003176 min_lr: 0.003176 loss: 2.1489 (2.2453) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [148] [311/312] eta: 0:00:00 lr: 0.003176 min_lr: 0.003176 loss: 2.2796 (2.2463) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [148] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.003176 min_lr: 0.003176 loss: 2.2796 (2.2255) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6848 (0.6848) acc1: 83.3333 (83.3333) acc5: 94.7917 (94.7917) time: 4.6993 data: 4.4874 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9130 (0.9163) acc1: 77.3438 (76.1600) acc5: 94.5312 (93.6000) time: 0.6734 data: 0.4987 max mem: 64948 Test: Total time: 0:00:06 (0.7020 s / it) * Acc@1 77.048 Acc@5 93.618 loss 0.900 Accuracy of the model on the 50000 test images: 77.0% Max accuracy: 77.29% Test: [0/9] eta: 0:00:42 loss: 0.6115 (0.6115) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.7453 data: 4.5392 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7761 (0.7745) acc1: 80.9896 (79.1360) acc5: 95.8333 (94.8800) time: 0.6886 data: 0.5136 max mem: 64948 Test: Total time: 0:00:06 (0.6968 s / it) * Acc@1 79.642 Acc@5 94.908 loss 0.759 Accuracy of the model EMA on 50000 test images: 79.6% Max EMA accuracy: 79.64% Epoch: [149] [ 0/312] eta: 0:54:41 lr: 0.003176 min_lr: 0.003176 loss: 2.4737 (2.4737) weight_decay: 0.0500 (0.0500) time: 10.5175 data: 9.7715 max mem: 64948 Epoch: [149] [ 10/312] eta: 0:08:04 lr: 0.003175 min_lr: 0.003175 loss: 1.9654 (2.0382) weight_decay: 0.0500 (0.0500) time: 1.6052 data: 0.8886 max mem: 64948 Epoch: [149] [ 20/312] eta: 0:05:42 lr: 0.003175 min_lr: 0.003175 loss: 1.9404 (2.0816) weight_decay: 0.0500 (0.0500) time: 0.7065 data: 0.0003 max mem: 64948 Epoch: [149] [ 30/312] eta: 0:04:47 lr: 0.003175 min_lr: 0.003175 loss: 2.2947 (2.1183) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [149] [ 40/312] eta: 0:04:16 lr: 0.003174 min_lr: 0.003174 loss: 2.3601 (2.1395) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [149] [ 50/312] eta: 0:03:54 lr: 0.003174 min_lr: 0.003174 loss: 2.3593 (2.1607) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [149] [ 60/312] eta: 0:03:37 lr: 0.003174 min_lr: 0.003174 loss: 2.3362 (2.1787) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [149] [ 70/312] eta: 0:03:22 lr: 0.003173 min_lr: 0.003173 loss: 2.2726 (2.1662) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [149] [ 80/312] eta: 0:03:10 lr: 0.003173 min_lr: 0.003173 loss: 2.3805 (2.1905) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [149] [ 90/312] eta: 0:02:59 lr: 0.003172 min_lr: 0.003172 loss: 2.4684 (2.2055) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [149] [100/312] eta: 0:02:48 lr: 0.003172 min_lr: 0.003172 loss: 2.2762 (2.2007) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [149] [110/312] eta: 0:02:38 lr: 0.003172 min_lr: 0.003172 loss: 2.2762 (2.2054) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [149] [120/312] eta: 0:02:29 lr: 0.003171 min_lr: 0.003171 loss: 2.3828 (2.2129) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [149] [130/312] eta: 0:02:20 lr: 0.003171 min_lr: 0.003171 loss: 2.2769 (2.1968) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [149] [140/312] eta: 0:02:11 lr: 0.003170 min_lr: 0.003170 loss: 2.0671 (2.1915) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [149] [150/312] eta: 0:02:03 lr: 0.003170 min_lr: 0.003170 loss: 2.2878 (2.2080) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [149] [160/312] eta: 0:01:55 lr: 0.003170 min_lr: 0.003170 loss: 2.3563 (2.2113) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [149] [170/312] eta: 0:01:47 lr: 0.003169 min_lr: 0.003169 loss: 2.2493 (2.2032) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [149] [180/312] eta: 0:01:39 lr: 0.003169 min_lr: 0.003169 loss: 2.1695 (2.2050) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [149] [190/312] eta: 0:01:31 lr: 0.003169 min_lr: 0.003169 loss: 2.2648 (2.2062) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [149] [200/312] eta: 0:01:23 lr: 0.003168 min_lr: 0.003168 loss: 2.2093 (2.1991) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [149] [210/312] eta: 0:01:15 lr: 0.003168 min_lr: 0.003168 loss: 2.0848 (2.1985) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [149] [220/312] eta: 0:01:08 lr: 0.003167 min_lr: 0.003167 loss: 2.0084 (2.1883) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [149] [230/312] eta: 0:01:00 lr: 0.003167 min_lr: 0.003167 loss: 2.2903 (2.1988) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [149] [240/312] eta: 0:00:53 lr: 0.003167 min_lr: 0.003167 loss: 2.4204 (2.2013) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [149] [250/312] eta: 0:00:45 lr: 0.003166 min_lr: 0.003166 loss: 2.1195 (2.1959) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [149] [260/312] eta: 0:00:38 lr: 0.003166 min_lr: 0.003166 loss: 1.8717 (2.1876) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [149] [270/312] eta: 0:00:30 lr: 0.003166 min_lr: 0.003166 loss: 2.0544 (2.1856) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [149] [280/312] eta: 0:00:23 lr: 0.003165 min_lr: 0.003165 loss: 2.2919 (2.1914) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [149] [290/312] eta: 0:00:16 lr: 0.003165 min_lr: 0.003165 loss: 2.4010 (2.2007) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [149] [300/312] eta: 0:00:08 lr: 0.003164 min_lr: 0.003164 loss: 2.4010 (2.2076) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [149] [310/312] eta: 0:00:01 lr: 0.003164 min_lr: 0.003164 loss: 2.4907 (2.2133) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [149] [311/312] eta: 0:00:00 lr: 0.003164 min_lr: 0.003164 loss: 2.4403 (2.2134) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [149] Total time: 0:03:47 (0.7297 s / it) Averaged stats: lr: 0.003164 min_lr: 0.003164 loss: 2.4403 (2.2253) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.6316 (0.6316) acc1: 84.6354 (84.6354) acc5: 96.0938 (96.0938) time: 4.8796 data: 4.6722 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9440 (0.9190) acc1: 77.0833 (75.8720) acc5: 92.4479 (93.6640) time: 0.6934 data: 0.5192 max mem: 64948 Test: Total time: 0:00:06 (0.7184 s / it) * Acc@1 77.232 Acc@5 93.732 loss 0.894 Accuracy of the model on the 50000 test images: 77.2% Max accuracy: 77.29% Test: [0/9] eta: 0:00:43 loss: 0.6093 (0.6093) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.8705 data: 4.6562 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7746 (0.7729) acc1: 80.9896 (79.1680) acc5: 95.8333 (95.0400) time: 0.6925 data: 0.5175 max mem: 64948 Test: Total time: 0:00:06 (0.7082 s / it) * Acc@1 79.674 Acc@5 94.948 loss 0.757 Accuracy of the model EMA on 50000 test images: 79.7% Max EMA accuracy: 79.67% Epoch: [150] [ 0/312] eta: 0:48:18 lr: 0.003164 min_lr: 0.003164 loss: 1.5054 (1.5054) weight_decay: 0.0500 (0.0500) time: 9.2892 data: 6.9285 max mem: 64948 Epoch: [150] [ 10/312] eta: 0:07:42 lr: 0.003164 min_lr: 0.003164 loss: 2.3349 (2.1344) weight_decay: 0.0500 (0.0500) time: 1.5307 data: 0.6303 max mem: 64948 Epoch: [150] [ 20/312] eta: 0:05:31 lr: 0.003163 min_lr: 0.003163 loss: 2.1878 (2.2231) weight_decay: 0.0500 (0.0500) time: 0.7289 data: 0.0005 max mem: 64948 Epoch: [150] [ 30/312] eta: 0:04:40 lr: 0.003163 min_lr: 0.003163 loss: 2.3072 (2.2975) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [150] [ 40/312] eta: 0:04:10 lr: 0.003162 min_lr: 0.003162 loss: 2.3811 (2.2876) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [150] [ 50/312] eta: 0:03:49 lr: 0.003162 min_lr: 0.003162 loss: 2.1771 (2.2482) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [150] [ 60/312] eta: 0:03:33 lr: 0.003162 min_lr: 0.003162 loss: 2.2683 (2.2514) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [150] [ 70/312] eta: 0:03:19 lr: 0.003161 min_lr: 0.003161 loss: 2.2683 (2.2284) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [150] [ 80/312] eta: 0:03:07 lr: 0.003161 min_lr: 0.003161 loss: 2.2307 (2.2192) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [150] [ 90/312] eta: 0:02:56 lr: 0.003160 min_lr: 0.003160 loss: 2.3749 (2.2425) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [150] [100/312] eta: 0:02:46 lr: 0.003160 min_lr: 0.003160 loss: 2.3112 (2.2355) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [150] [110/312] eta: 0:02:37 lr: 0.003160 min_lr: 0.003160 loss: 2.1009 (2.2258) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [150] [120/312] eta: 0:02:28 lr: 0.003159 min_lr: 0.003159 loss: 2.1009 (2.2158) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [150] [130/312] eta: 0:02:19 lr: 0.003159 min_lr: 0.003159 loss: 2.2610 (2.2162) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [150] [140/312] eta: 0:02:10 lr: 0.003159 min_lr: 0.003159 loss: 2.3874 (2.2311) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [150] [150/312] eta: 0:02:02 lr: 0.003158 min_lr: 0.003158 loss: 2.3911 (2.2355) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [150] [160/312] eta: 0:01:54 lr: 0.003158 min_lr: 0.003158 loss: 2.3859 (2.2451) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [150] [170/312] eta: 0:01:46 lr: 0.003157 min_lr: 0.003157 loss: 2.3554 (2.2446) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [150] [180/312] eta: 0:01:38 lr: 0.003157 min_lr: 0.003157 loss: 2.1324 (2.2357) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [150] [190/312] eta: 0:01:30 lr: 0.003157 min_lr: 0.003157 loss: 1.9873 (2.2300) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0003 max mem: 64948 Epoch: [150] [200/312] eta: 0:01:23 lr: 0.003156 min_lr: 0.003156 loss: 2.4231 (2.2360) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [150] [210/312] eta: 0:01:15 lr: 0.003156 min_lr: 0.003156 loss: 2.4295 (2.2438) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [150] [220/312] eta: 0:01:07 lr: 0.003156 min_lr: 0.003156 loss: 2.4053 (2.2441) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [150] [230/312] eta: 0:01:00 lr: 0.003155 min_lr: 0.003155 loss: 2.2662 (2.2408) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [150] [240/312] eta: 0:00:52 lr: 0.003155 min_lr: 0.003155 loss: 2.2961 (2.2491) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [150] [250/312] eta: 0:00:45 lr: 0.003154 min_lr: 0.003154 loss: 2.2441 (2.2474) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [150] [260/312] eta: 0:00:37 lr: 0.003154 min_lr: 0.003154 loss: 2.1059 (2.2389) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [150] [270/312] eta: 0:00:30 lr: 0.003154 min_lr: 0.003154 loss: 1.9696 (2.2336) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [150] [280/312] eta: 0:00:23 lr: 0.003153 min_lr: 0.003153 loss: 2.2892 (2.2354) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0009 max mem: 64948 Epoch: [150] [290/312] eta: 0:00:15 lr: 0.003153 min_lr: 0.003153 loss: 2.2892 (2.2370) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [150] [300/312] eta: 0:00:08 lr: 0.003152 min_lr: 0.003152 loss: 2.2558 (2.2368) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [150] [310/312] eta: 0:00:01 lr: 0.003152 min_lr: 0.003152 loss: 2.2654 (2.2365) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [150] [311/312] eta: 0:00:00 lr: 0.003152 min_lr: 0.003152 loss: 2.2728 (2.2374) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [150] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.003152 min_lr: 0.003152 loss: 2.2728 (2.2178) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6013 (0.6013) acc1: 84.6354 (84.6354) acc5: 95.0521 (95.0521) time: 4.6921 data: 4.4837 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9696 (0.8879) acc1: 76.3021 (76.9280) acc5: 94.0104 (93.4400) time: 0.6726 data: 0.4983 max mem: 64948 Test: Total time: 0:00:06 (0.6990 s / it) * Acc@1 77.050 Acc@5 93.482 loss 0.884 Accuracy of the model on the 50000 test images: 77.1% Max accuracy: 77.29% Test: [0/9] eta: 0:00:42 loss: 0.6067 (0.6067) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.7560 data: 4.5426 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7727 (0.7710) acc1: 80.9896 (79.1680) acc5: 95.8333 (95.0080) time: 0.6798 data: 0.5049 max mem: 64948 Test: Total time: 0:00:06 (0.6895 s / it) * Acc@1 79.724 Acc@5 94.978 loss 0.755 Accuracy of the model EMA on 50000 test images: 79.7% Max EMA accuracy: 79.72% Epoch: [151] [ 0/312] eta: 0:49:51 lr: 0.003152 min_lr: 0.003152 loss: 1.8126 (1.8126) weight_decay: 0.0500 (0.0500) time: 9.5894 data: 7.7986 max mem: 64948 Epoch: [151] [ 10/312] eta: 0:07:54 lr: 0.003152 min_lr: 0.003152 loss: 1.9376 (1.9937) weight_decay: 0.0500 (0.0500) time: 1.5702 data: 0.7094 max mem: 64948 Epoch: [151] [ 20/312] eta: 0:05:36 lr: 0.003151 min_lr: 0.003151 loss: 1.9376 (2.0123) weight_decay: 0.0500 (0.0500) time: 0.7306 data: 0.0004 max mem: 64948 Epoch: [151] [ 30/312] eta: 0:04:43 lr: 0.003151 min_lr: 0.003151 loss: 2.2944 (2.1468) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [151] [ 40/312] eta: 0:04:12 lr: 0.003150 min_lr: 0.003150 loss: 2.1825 (2.1456) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [151] [ 50/312] eta: 0:03:51 lr: 0.003150 min_lr: 0.003150 loss: 2.2252 (2.1876) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [151] [ 60/312] eta: 0:03:34 lr: 0.003150 min_lr: 0.003150 loss: 2.2466 (2.1753) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [151] [ 70/312] eta: 0:03:20 lr: 0.003149 min_lr: 0.003149 loss: 2.2317 (2.1988) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [151] [ 80/312] eta: 0:03:08 lr: 0.003149 min_lr: 0.003149 loss: 2.0989 (2.1616) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [151] [ 90/312] eta: 0:02:57 lr: 0.003149 min_lr: 0.003149 loss: 2.0699 (2.1636) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [151] [100/312] eta: 0:02:47 lr: 0.003148 min_lr: 0.003148 loss: 2.1922 (2.1634) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [151] [110/312] eta: 0:02:37 lr: 0.003148 min_lr: 0.003148 loss: 2.3201 (2.1818) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [151] [120/312] eta: 0:02:28 lr: 0.003147 min_lr: 0.003147 loss: 2.3201 (2.1847) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [151] [130/312] eta: 0:02:19 lr: 0.003147 min_lr: 0.003147 loss: 2.2726 (2.1959) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [151] [140/312] eta: 0:02:11 lr: 0.003147 min_lr: 0.003147 loss: 2.2726 (2.1963) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [151] [150/312] eta: 0:02:02 lr: 0.003146 min_lr: 0.003146 loss: 2.2627 (2.2042) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [151] [160/312] eta: 0:01:54 lr: 0.003146 min_lr: 0.003146 loss: 2.2928 (2.2095) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [151] [170/312] eta: 0:01:46 lr: 0.003145 min_lr: 0.003145 loss: 2.2464 (2.2073) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [151] [180/312] eta: 0:01:38 lr: 0.003145 min_lr: 0.003145 loss: 2.2011 (2.2132) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [151] [190/312] eta: 0:01:30 lr: 0.003145 min_lr: 0.003145 loss: 2.2023 (2.2029) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [151] [200/312] eta: 0:01:23 lr: 0.003144 min_lr: 0.003144 loss: 2.2023 (2.2056) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [151] [210/312] eta: 0:01:15 lr: 0.003144 min_lr: 0.003144 loss: 2.2096 (2.2063) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [151] [220/312] eta: 0:01:07 lr: 0.003144 min_lr: 0.003144 loss: 2.2096 (2.2120) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [151] [230/312] eta: 0:01:00 lr: 0.003143 min_lr: 0.003143 loss: 2.3174 (2.2126) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [151] [240/312] eta: 0:00:52 lr: 0.003143 min_lr: 0.003143 loss: 2.3062 (2.2151) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [151] [250/312] eta: 0:00:45 lr: 0.003142 min_lr: 0.003142 loss: 2.3612 (2.2185) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [151] [260/312] eta: 0:00:38 lr: 0.003142 min_lr: 0.003142 loss: 2.3612 (2.2158) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [151] [270/312] eta: 0:00:30 lr: 0.003142 min_lr: 0.003142 loss: 2.1263 (2.2149) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [151] [280/312] eta: 0:00:23 lr: 0.003141 min_lr: 0.003141 loss: 2.1263 (2.2144) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0006 max mem: 64948 Epoch: [151] [290/312] eta: 0:00:16 lr: 0.003141 min_lr: 0.003141 loss: 2.1702 (2.2189) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0005 max mem: 64948 Epoch: [151] [300/312] eta: 0:00:08 lr: 0.003141 min_lr: 0.003141 loss: 2.3000 (2.2205) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [151] [310/312] eta: 0:00:01 lr: 0.003140 min_lr: 0.003140 loss: 2.2623 (2.2237) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [151] [311/312] eta: 0:00:00 lr: 0.003140 min_lr: 0.003140 loss: 2.2623 (2.2240) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [151] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.003140 min_lr: 0.003140 loss: 2.2623 (2.2128) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7213 (0.7213) acc1: 81.7708 (81.7708) acc5: 95.5729 (95.5729) time: 4.4903 data: 4.2749 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9729 (0.9198) acc1: 76.3021 (76.4160) acc5: 94.2708 (93.4080) time: 0.6502 data: 0.4751 max mem: 64948 Test: Total time: 0:00:06 (0.6710 s / it) * Acc@1 77.194 Acc@5 93.672 loss 0.899 Accuracy of the model on the 50000 test images: 77.2% Max accuracy: 77.29% Test: [0/9] eta: 0:00:44 loss: 0.6044 (0.6044) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.9006 data: 4.6866 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7708 (0.7693) acc1: 80.9896 (79.1680) acc5: 95.8333 (95.0720) time: 0.6958 data: 0.5208 max mem: 64948 Test: Total time: 0:00:06 (0.7047 s / it) * Acc@1 79.772 Acc@5 95.014 loss 0.753 Accuracy of the model EMA on 50000 test images: 79.8% Max EMA accuracy: 79.77% Epoch: [152] [ 0/312] eta: 0:50:58 lr: 0.003140 min_lr: 0.003140 loss: 2.5291 (2.5291) weight_decay: 0.0500 (0.0500) time: 9.8033 data: 9.0304 max mem: 64948 Epoch: [152] [ 10/312] eta: 0:07:53 lr: 0.003140 min_lr: 0.003140 loss: 2.2863 (2.3356) weight_decay: 0.0500 (0.0500) time: 1.5679 data: 0.8213 max mem: 64948 Epoch: [152] [ 20/312] eta: 0:05:36 lr: 0.003139 min_lr: 0.003139 loss: 2.2634 (2.2835) weight_decay: 0.0500 (0.0500) time: 0.7183 data: 0.0003 max mem: 64948 Epoch: [152] [ 30/312] eta: 0:04:43 lr: 0.003139 min_lr: 0.003139 loss: 2.1720 (2.2130) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [152] [ 40/312] eta: 0:04:12 lr: 0.003139 min_lr: 0.003139 loss: 2.1720 (2.1912) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [152] [ 50/312] eta: 0:03:51 lr: 0.003138 min_lr: 0.003138 loss: 2.3691 (2.2352) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [152] [ 60/312] eta: 0:03:34 lr: 0.003138 min_lr: 0.003138 loss: 2.4015 (2.2400) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [152] [ 70/312] eta: 0:03:21 lr: 0.003137 min_lr: 0.003137 loss: 2.2604 (2.2391) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [152] [ 80/312] eta: 0:03:08 lr: 0.003137 min_lr: 0.003137 loss: 2.1696 (2.2191) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [152] [ 90/312] eta: 0:02:58 lr: 0.003137 min_lr: 0.003137 loss: 2.0627 (2.2018) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [152] [100/312] eta: 0:02:47 lr: 0.003136 min_lr: 0.003136 loss: 2.3086 (2.2043) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [152] [110/312] eta: 0:02:38 lr: 0.003136 min_lr: 0.003136 loss: 2.3907 (2.2172) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [152] [120/312] eta: 0:02:29 lr: 0.003135 min_lr: 0.003135 loss: 2.3252 (2.2237) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [152] [130/312] eta: 0:02:20 lr: 0.003135 min_lr: 0.003135 loss: 2.2930 (2.2203) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [152] [140/312] eta: 0:02:11 lr: 0.003135 min_lr: 0.003135 loss: 2.2542 (2.2204) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [152] [150/312] eta: 0:02:03 lr: 0.003134 min_lr: 0.003134 loss: 2.2542 (2.2142) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [152] [160/312] eta: 0:01:54 lr: 0.003134 min_lr: 0.003134 loss: 2.3132 (2.2167) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [152] [170/312] eta: 0:01:46 lr: 0.003133 min_lr: 0.003133 loss: 2.3132 (2.2206) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [152] [180/312] eta: 0:01:38 lr: 0.003133 min_lr: 0.003133 loss: 2.3094 (2.2249) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [152] [190/312] eta: 0:01:30 lr: 0.003133 min_lr: 0.003133 loss: 2.3094 (2.2227) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [152] [200/312] eta: 0:01:23 lr: 0.003132 min_lr: 0.003132 loss: 2.3489 (2.2236) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [152] [210/312] eta: 0:01:15 lr: 0.003132 min_lr: 0.003132 loss: 2.3438 (2.2246) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [152] [220/312] eta: 0:01:07 lr: 0.003132 min_lr: 0.003132 loss: 2.3438 (2.2300) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [152] [230/312] eta: 0:01:00 lr: 0.003131 min_lr: 0.003131 loss: 2.3005 (2.2272) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [152] [240/312] eta: 0:00:52 lr: 0.003131 min_lr: 0.003131 loss: 2.3005 (2.2304) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [152] [250/312] eta: 0:00:45 lr: 0.003130 min_lr: 0.003130 loss: 2.3599 (2.2335) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [152] [260/312] eta: 0:00:38 lr: 0.003130 min_lr: 0.003130 loss: 2.2369 (2.2291) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [152] [270/312] eta: 0:00:30 lr: 0.003130 min_lr: 0.003130 loss: 2.3141 (2.2326) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [152] [280/312] eta: 0:00:23 lr: 0.003129 min_lr: 0.003129 loss: 2.2901 (2.2326) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [152] [290/312] eta: 0:00:16 lr: 0.003129 min_lr: 0.003129 loss: 2.3187 (2.2369) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [152] [300/312] eta: 0:00:08 lr: 0.003128 min_lr: 0.003128 loss: 2.4302 (2.2391) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [152] [310/312] eta: 0:00:01 lr: 0.003128 min_lr: 0.003128 loss: 2.3491 (2.2370) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [152] [311/312] eta: 0:00:00 lr: 0.003128 min_lr: 0.003128 loss: 2.3649 (2.2390) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [152] Total time: 0:03:47 (0.7286 s / it) Averaged stats: lr: 0.003128 min_lr: 0.003128 loss: 2.3649 (2.2226) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7562 (0.7562) acc1: 81.2500 (81.2500) acc5: 95.5729 (95.5729) time: 4.5376 data: 4.3184 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0030 (0.9363) acc1: 78.1250 (75.4560) acc5: 93.7500 (93.0560) time: 0.6562 data: 0.4799 max mem: 64948 Test: Total time: 0:00:06 (0.6729 s / it) * Acc@1 76.522 Acc@5 93.300 loss 0.918 Accuracy of the model on the 50000 test images: 76.5% Max accuracy: 77.29% Test: [0/9] eta: 0:00:45 loss: 0.6029 (0.6029) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 5.0155 data: 4.7923 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7696 (0.7680) acc1: 80.9896 (79.1680) acc5: 95.8333 (95.1040) time: 0.7086 data: 0.5326 max mem: 64948 Test: Total time: 0:00:06 (0.7215 s / it) * Acc@1 79.834 Acc@5 95.022 loss 0.751 Accuracy of the model EMA on 50000 test images: 79.8% Max EMA accuracy: 79.83% Epoch: [153] [ 0/312] eta: 0:55:48 lr: 0.003128 min_lr: 0.003128 loss: 2.2323 (2.2323) weight_decay: 0.0500 (0.0500) time: 10.7317 data: 9.9964 max mem: 64948 Epoch: [153] [ 10/312] eta: 0:08:12 lr: 0.003128 min_lr: 0.003128 loss: 2.2504 (2.2987) weight_decay: 0.0500 (0.0500) time: 1.6318 data: 0.9091 max mem: 64948 Epoch: [153] [ 20/312] eta: 0:05:46 lr: 0.003127 min_lr: 0.003127 loss: 2.3060 (2.3355) weight_decay: 0.0500 (0.0500) time: 0.7096 data: 0.0004 max mem: 64948 Epoch: [153] [ 30/312] eta: 0:04:50 lr: 0.003127 min_lr: 0.003127 loss: 2.3280 (2.3176) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [153] [ 40/312] eta: 0:04:17 lr: 0.003126 min_lr: 0.003126 loss: 2.3280 (2.3295) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [153] [ 50/312] eta: 0:03:55 lr: 0.003126 min_lr: 0.003126 loss: 2.3319 (2.3194) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [153] [ 60/312] eta: 0:03:37 lr: 0.003126 min_lr: 0.003126 loss: 2.4181 (2.3436) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [153] [ 70/312] eta: 0:03:23 lr: 0.003125 min_lr: 0.003125 loss: 2.3132 (2.3092) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [153] [ 80/312] eta: 0:03:10 lr: 0.003125 min_lr: 0.003125 loss: 2.0872 (2.2834) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [153] [ 90/312] eta: 0:02:59 lr: 0.003125 min_lr: 0.003125 loss: 2.1397 (2.2892) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [153] [100/312] eta: 0:02:49 lr: 0.003124 min_lr: 0.003124 loss: 2.2681 (2.2790) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [153] [110/312] eta: 0:02:39 lr: 0.003124 min_lr: 0.003124 loss: 2.2388 (2.2718) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [153] [120/312] eta: 0:02:29 lr: 0.003123 min_lr: 0.003123 loss: 2.2911 (2.2729) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [153] [130/312] eta: 0:02:20 lr: 0.003123 min_lr: 0.003123 loss: 2.3252 (2.2725) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [153] [140/312] eta: 0:02:12 lr: 0.003123 min_lr: 0.003123 loss: 2.3775 (2.2777) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [153] [150/312] eta: 0:02:03 lr: 0.003122 min_lr: 0.003122 loss: 2.4086 (2.2760) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [153] [160/312] eta: 0:01:55 lr: 0.003122 min_lr: 0.003122 loss: 2.2384 (2.2680) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [153] [170/312] eta: 0:01:47 lr: 0.003121 min_lr: 0.003121 loss: 2.3139 (2.2718) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [153] [180/312] eta: 0:01:39 lr: 0.003121 min_lr: 0.003121 loss: 2.2912 (2.2717) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [153] [190/312] eta: 0:01:31 lr: 0.003121 min_lr: 0.003121 loss: 2.1520 (2.2619) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [153] [200/312] eta: 0:01:23 lr: 0.003120 min_lr: 0.003120 loss: 2.1575 (2.2627) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [153] [210/312] eta: 0:01:15 lr: 0.003120 min_lr: 0.003120 loss: 2.2555 (2.2592) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [153] [220/312] eta: 0:01:08 lr: 0.003119 min_lr: 0.003119 loss: 2.1782 (2.2558) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [153] [230/312] eta: 0:01:00 lr: 0.003119 min_lr: 0.003119 loss: 2.2152 (2.2535) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [153] [240/312] eta: 0:00:53 lr: 0.003119 min_lr: 0.003119 loss: 2.1069 (2.2472) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [153] [250/312] eta: 0:00:45 lr: 0.003118 min_lr: 0.003118 loss: 2.1407 (2.2467) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [153] [260/312] eta: 0:00:38 lr: 0.003118 min_lr: 0.003118 loss: 2.3802 (2.2514) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [153] [270/312] eta: 0:00:30 lr: 0.003118 min_lr: 0.003118 loss: 2.3802 (2.2542) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [153] [280/312] eta: 0:00:23 lr: 0.003117 min_lr: 0.003117 loss: 2.3977 (2.2579) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [153] [290/312] eta: 0:00:16 lr: 0.003117 min_lr: 0.003117 loss: 2.3599 (2.2553) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [153] [300/312] eta: 0:00:08 lr: 0.003116 min_lr: 0.003116 loss: 2.2480 (2.2582) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [153] [310/312] eta: 0:00:01 lr: 0.003116 min_lr: 0.003116 loss: 2.3490 (2.2584) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [153] [311/312] eta: 0:00:00 lr: 0.003116 min_lr: 0.003116 loss: 2.3625 (2.2592) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [153] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.003116 min_lr: 0.003116 loss: 2.3625 (2.2163) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6889 (0.6889) acc1: 81.5104 (81.5104) acc5: 94.5312 (94.5312) time: 4.6354 data: 4.4140 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9800 (0.9305) acc1: 76.3021 (75.6480) acc5: 93.2292 (93.1520) time: 0.6670 data: 0.4905 max mem: 64948 Test: Total time: 0:00:06 (0.6883 s / it) * Acc@1 77.066 Acc@5 93.640 loss 0.885 Accuracy of the model on the 50000 test images: 77.1% Max accuracy: 77.29% Test: [0/9] eta: 0:00:41 loss: 0.6010 (0.6010) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.6373 data: 4.4253 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7686 (0.7664) acc1: 80.9896 (79.1360) acc5: 95.8333 (95.1040) time: 0.6698 data: 0.4918 max mem: 64948 Test: Total time: 0:00:06 (0.6784 s / it) * Acc@1 79.876 Acc@5 95.048 loss 0.749 Accuracy of the model EMA on 50000 test images: 79.9% Max EMA accuracy: 79.88% Epoch: [154] [ 0/312] eta: 0:47:21 lr: 0.003116 min_lr: 0.003116 loss: 2.3676 (2.3676) weight_decay: 0.0500 (0.0500) time: 9.1073 data: 8.1084 max mem: 64948 Epoch: [154] [ 10/312] eta: 0:07:39 lr: 0.003116 min_lr: 0.003116 loss: 2.3676 (2.2029) weight_decay: 0.0500 (0.0500) time: 1.5223 data: 0.7376 max mem: 64948 Epoch: [154] [ 20/312] eta: 0:05:29 lr: 0.003115 min_lr: 0.003115 loss: 2.2915 (2.1807) weight_decay: 0.0500 (0.0500) time: 0.7283 data: 0.0005 max mem: 64948 Epoch: [154] [ 30/312] eta: 0:04:38 lr: 0.003115 min_lr: 0.003115 loss: 2.1400 (2.1488) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [154] [ 40/312] eta: 0:04:09 lr: 0.003114 min_lr: 0.003114 loss: 2.1139 (2.1461) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [154] [ 50/312] eta: 0:03:48 lr: 0.003114 min_lr: 0.003114 loss: 1.8902 (2.0870) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [154] [ 60/312] eta: 0:03:32 lr: 0.003114 min_lr: 0.003114 loss: 1.8856 (2.1138) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [154] [ 70/312] eta: 0:03:19 lr: 0.003113 min_lr: 0.003113 loss: 2.0664 (2.1154) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [154] [ 80/312] eta: 0:03:07 lr: 0.003113 min_lr: 0.003113 loss: 2.1454 (2.1457) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [154] [ 90/312] eta: 0:02:56 lr: 0.003112 min_lr: 0.003112 loss: 2.2724 (2.1449) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [154] [100/312] eta: 0:02:46 lr: 0.003112 min_lr: 0.003112 loss: 2.1586 (2.1443) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [154] [110/312] eta: 0:02:36 lr: 0.003112 min_lr: 0.003112 loss: 2.3083 (2.1702) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [154] [120/312] eta: 0:02:27 lr: 0.003111 min_lr: 0.003111 loss: 2.3888 (2.1811) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [154] [130/312] eta: 0:02:18 lr: 0.003111 min_lr: 0.003111 loss: 2.2162 (2.1689) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [154] [140/312] eta: 0:02:10 lr: 0.003110 min_lr: 0.003110 loss: 2.1799 (2.1799) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [154] [150/312] eta: 0:02:02 lr: 0.003110 min_lr: 0.003110 loss: 2.2900 (2.1640) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [154] [160/312] eta: 0:01:54 lr: 0.003110 min_lr: 0.003110 loss: 2.1663 (2.1710) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [154] [170/312] eta: 0:01:46 lr: 0.003109 min_lr: 0.003109 loss: 2.2630 (2.1815) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [154] [180/312] eta: 0:01:38 lr: 0.003109 min_lr: 0.003109 loss: 2.2830 (2.1850) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [154] [190/312] eta: 0:01:30 lr: 0.003109 min_lr: 0.003109 loss: 2.3726 (2.1977) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [154] [200/312] eta: 0:01:22 lr: 0.003108 min_lr: 0.003108 loss: 2.3728 (2.2047) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [154] [210/312] eta: 0:01:15 lr: 0.003108 min_lr: 0.003108 loss: 2.3428 (2.2043) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [154] [220/312] eta: 0:01:07 lr: 0.003107 min_lr: 0.003107 loss: 2.3428 (2.2016) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [154] [230/312] eta: 0:01:00 lr: 0.003107 min_lr: 0.003107 loss: 2.1863 (2.2019) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [154] [240/312] eta: 0:00:52 lr: 0.003107 min_lr: 0.003107 loss: 2.4345 (2.2163) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [154] [250/312] eta: 0:00:45 lr: 0.003106 min_lr: 0.003106 loss: 2.3302 (2.2116) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [154] [260/312] eta: 0:00:37 lr: 0.003106 min_lr: 0.003106 loss: 2.0278 (2.2048) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [154] [270/312] eta: 0:00:30 lr: 0.003105 min_lr: 0.003105 loss: 1.9440 (2.1983) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [154] [280/312] eta: 0:00:23 lr: 0.003105 min_lr: 0.003105 loss: 2.2140 (2.1970) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0009 max mem: 64948 Epoch: [154] [290/312] eta: 0:00:15 lr: 0.003105 min_lr: 0.003105 loss: 2.3325 (2.2040) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [154] [300/312] eta: 0:00:08 lr: 0.003104 min_lr: 0.003104 loss: 2.2867 (2.2033) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [154] [310/312] eta: 0:00:01 lr: 0.003104 min_lr: 0.003104 loss: 2.1920 (2.2016) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [154] [311/312] eta: 0:00:00 lr: 0.003104 min_lr: 0.003104 loss: 2.2331 (2.2028) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [154] Total time: 0:03:46 (0.7265 s / it) Averaged stats: lr: 0.003104 min_lr: 0.003104 loss: 2.2331 (2.2116) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7235 (0.7235) acc1: 83.5938 (83.5938) acc5: 95.3125 (95.3125) time: 4.6649 data: 4.4580 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9523 (0.9212) acc1: 77.0833 (76.2560) acc5: 94.2708 (93.1840) time: 0.6699 data: 0.4954 max mem: 64948 Test: Total time: 0:00:06 (0.6937 s / it) * Acc@1 77.076 Acc@5 93.436 loss 0.903 Accuracy of the model on the 50000 test images: 77.1% Max accuracy: 77.29% Test: [0/9] eta: 0:00:45 loss: 0.5993 (0.5993) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 5.0393 data: 4.8237 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7680 (0.7651) acc1: 80.9896 (79.1040) acc5: 95.8333 (95.0400) time: 0.7171 data: 0.5361 max mem: 64948 Test: Total time: 0:00:06 (0.7263 s / it) * Acc@1 79.918 Acc@5 95.064 loss 0.748 Accuracy of the model EMA on 50000 test images: 79.9% Max EMA accuracy: 79.92% Epoch: [155] [ 0/312] eta: 0:49:49 lr: 0.003104 min_lr: 0.003104 loss: 2.2661 (2.2661) weight_decay: 0.0500 (0.0500) time: 9.5825 data: 8.7877 max mem: 64948 Epoch: [155] [ 10/312] eta: 0:07:55 lr: 0.003103 min_lr: 0.003103 loss: 2.4113 (2.3139) weight_decay: 0.0500 (0.0500) time: 1.5740 data: 0.8384 max mem: 64948 Epoch: [155] [ 20/312] eta: 0:05:37 lr: 0.003103 min_lr: 0.003103 loss: 2.4113 (2.2871) weight_decay: 0.0500 (0.0500) time: 0.7342 data: 0.0219 max mem: 64948 Epoch: [155] [ 30/312] eta: 0:04:44 lr: 0.003103 min_lr: 0.003103 loss: 2.4058 (2.2859) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [155] [ 40/312] eta: 0:04:13 lr: 0.003102 min_lr: 0.003102 loss: 2.4095 (2.3205) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [155] [ 50/312] eta: 0:03:51 lr: 0.003102 min_lr: 0.003102 loss: 2.3742 (2.2956) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [155] [ 60/312] eta: 0:03:35 lr: 0.003101 min_lr: 0.003101 loss: 2.2730 (2.2940) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [155] [ 70/312] eta: 0:03:21 lr: 0.003101 min_lr: 0.003101 loss: 2.2237 (2.2703) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [155] [ 80/312] eta: 0:03:09 lr: 0.003101 min_lr: 0.003101 loss: 2.4254 (2.3048) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [155] [ 90/312] eta: 0:02:58 lr: 0.003100 min_lr: 0.003100 loss: 2.4404 (2.3042) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [155] [100/312] eta: 0:02:47 lr: 0.003100 min_lr: 0.003100 loss: 2.2494 (2.2920) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [155] [110/312] eta: 0:02:38 lr: 0.003099 min_lr: 0.003099 loss: 2.2166 (2.2683) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [155] [120/312] eta: 0:02:28 lr: 0.003099 min_lr: 0.003099 loss: 2.3151 (2.2805) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [155] [130/312] eta: 0:02:20 lr: 0.003099 min_lr: 0.003099 loss: 2.3950 (2.2805) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [155] [140/312] eta: 0:02:11 lr: 0.003098 min_lr: 0.003098 loss: 2.1028 (2.2750) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [155] [150/312] eta: 0:02:03 lr: 0.003098 min_lr: 0.003098 loss: 2.1202 (2.2752) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [155] [160/312] eta: 0:01:54 lr: 0.003098 min_lr: 0.003098 loss: 2.1334 (2.2636) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [155] [170/312] eta: 0:01:46 lr: 0.003097 min_lr: 0.003097 loss: 2.0897 (2.2575) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [155] [180/312] eta: 0:01:38 lr: 0.003097 min_lr: 0.003097 loss: 2.1029 (2.2478) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [155] [190/312] eta: 0:01:31 lr: 0.003096 min_lr: 0.003096 loss: 1.9997 (2.2393) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [155] [200/312] eta: 0:01:23 lr: 0.003096 min_lr: 0.003096 loss: 2.3219 (2.2490) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [155] [210/312] eta: 0:01:15 lr: 0.003096 min_lr: 0.003096 loss: 2.3371 (2.2485) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [155] [220/312] eta: 0:01:07 lr: 0.003095 min_lr: 0.003095 loss: 2.1574 (2.2458) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [155] [230/312] eta: 0:01:00 lr: 0.003095 min_lr: 0.003095 loss: 2.1728 (2.2436) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [155] [240/312] eta: 0:00:52 lr: 0.003094 min_lr: 0.003094 loss: 2.2435 (2.2381) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [155] [250/312] eta: 0:00:45 lr: 0.003094 min_lr: 0.003094 loss: 2.2809 (2.2401) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [155] [260/312] eta: 0:00:38 lr: 0.003094 min_lr: 0.003094 loss: 2.4218 (2.2434) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [155] [270/312] eta: 0:00:30 lr: 0.003093 min_lr: 0.003093 loss: 2.3883 (2.2463) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [155] [280/312] eta: 0:00:23 lr: 0.003093 min_lr: 0.003093 loss: 2.2598 (2.2407) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [155] [290/312] eta: 0:00:16 lr: 0.003092 min_lr: 0.003092 loss: 2.1707 (2.2431) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [155] [300/312] eta: 0:00:08 lr: 0.003092 min_lr: 0.003092 loss: 2.2719 (2.2387) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [155] [310/312] eta: 0:00:01 lr: 0.003092 min_lr: 0.003092 loss: 2.2719 (2.2327) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [155] [311/312] eta: 0:00:00 lr: 0.003092 min_lr: 0.003092 loss: 1.9032 (2.2316) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [155] Total time: 0:03:47 (0.7288 s / it) Averaged stats: lr: 0.003092 min_lr: 0.003092 loss: 1.9032 (2.2165) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6809 (0.6809) acc1: 84.1146 (84.1146) acc5: 95.5729 (95.5729) time: 4.5056 data: 4.2892 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9700 (0.9207) acc1: 76.5625 (75.8720) acc5: 93.2292 (93.2160) time: 0.6519 data: 0.4767 max mem: 64948 Test: Total time: 0:00:06 (0.6771 s / it) * Acc@1 76.898 Acc@5 93.478 loss 0.905 Accuracy of the model on the 50000 test images: 76.9% Max accuracy: 77.29% Test: [0/9] eta: 0:00:44 loss: 0.5977 (0.5977) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.9799 data: 4.7700 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7682 (0.7637) acc1: 80.9896 (79.1360) acc5: 95.8333 (95.0720) time: 0.7046 data: 0.5301 max mem: 64948 Test: Total time: 0:00:06 (0.7157 s / it) * Acc@1 79.956 Acc@5 95.102 loss 0.746 Accuracy of the model EMA on 50000 test images: 80.0% Max EMA accuracy: 79.96% Epoch: [156] [ 0/312] eta: 0:47:51 lr: 0.003092 min_lr: 0.003092 loss: 2.5090 (2.5090) weight_decay: 0.0500 (0.0500) time: 9.2037 data: 7.5409 max mem: 64948 Epoch: [156] [ 10/312] eta: 0:07:37 lr: 0.003091 min_lr: 0.003091 loss: 2.4009 (2.4292) weight_decay: 0.0500 (0.0500) time: 1.5164 data: 0.7126 max mem: 64948 Epoch: [156] [ 20/312] eta: 0:05:28 lr: 0.003091 min_lr: 0.003091 loss: 2.2538 (2.2804) weight_decay: 0.0500 (0.0500) time: 0.7203 data: 0.0151 max mem: 64948 Epoch: [156] [ 30/312] eta: 0:04:38 lr: 0.003090 min_lr: 0.003090 loss: 1.9925 (2.2331) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [156] [ 40/312] eta: 0:04:08 lr: 0.003090 min_lr: 0.003090 loss: 2.0572 (2.2159) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [156] [ 50/312] eta: 0:03:48 lr: 0.003090 min_lr: 0.003090 loss: 2.0959 (2.1992) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [156] [ 60/312] eta: 0:03:32 lr: 0.003089 min_lr: 0.003089 loss: 2.0959 (2.2031) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [156] [ 70/312] eta: 0:03:19 lr: 0.003089 min_lr: 0.003089 loss: 2.1829 (2.2038) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [156] [ 80/312] eta: 0:03:07 lr: 0.003088 min_lr: 0.003088 loss: 2.3359 (2.2106) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [156] [ 90/312] eta: 0:02:56 lr: 0.003088 min_lr: 0.003088 loss: 2.4018 (2.2363) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [156] [100/312] eta: 0:02:46 lr: 0.003088 min_lr: 0.003088 loss: 2.1492 (2.2233) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [156] [110/312] eta: 0:02:36 lr: 0.003087 min_lr: 0.003087 loss: 2.2072 (2.2253) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [156] [120/312] eta: 0:02:27 lr: 0.003087 min_lr: 0.003087 loss: 2.2115 (2.2100) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [156] [130/312] eta: 0:02:19 lr: 0.003086 min_lr: 0.003086 loss: 2.2115 (2.2124) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [156] [140/312] eta: 0:02:10 lr: 0.003086 min_lr: 0.003086 loss: 2.3092 (2.2224) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [156] [150/312] eta: 0:02:02 lr: 0.003086 min_lr: 0.003086 loss: 2.2446 (2.2179) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [156] [160/312] eta: 0:01:54 lr: 0.003085 min_lr: 0.003085 loss: 2.1774 (2.2077) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [156] [170/312] eta: 0:01:46 lr: 0.003085 min_lr: 0.003085 loss: 2.1256 (2.1977) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [156] [180/312] eta: 0:01:38 lr: 0.003084 min_lr: 0.003084 loss: 2.1135 (2.1937) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [156] [190/312] eta: 0:01:30 lr: 0.003084 min_lr: 0.003084 loss: 2.1666 (2.1966) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [156] [200/312] eta: 0:01:22 lr: 0.003084 min_lr: 0.003084 loss: 2.1395 (2.1951) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [156] [210/312] eta: 0:01:15 lr: 0.003083 min_lr: 0.003083 loss: 2.3118 (2.2043) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [156] [220/312] eta: 0:01:07 lr: 0.003083 min_lr: 0.003083 loss: 2.4040 (2.2110) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [156] [230/312] eta: 0:01:00 lr: 0.003083 min_lr: 0.003083 loss: 2.4278 (2.2161) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [156] [240/312] eta: 0:00:52 lr: 0.003082 min_lr: 0.003082 loss: 2.5369 (2.2231) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [156] [250/312] eta: 0:00:45 lr: 0.003082 min_lr: 0.003082 loss: 2.3960 (2.2300) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [156] [260/312] eta: 0:00:37 lr: 0.003081 min_lr: 0.003081 loss: 2.2999 (2.2260) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [156] [270/312] eta: 0:00:30 lr: 0.003081 min_lr: 0.003081 loss: 2.2999 (2.2306) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [156] [280/312] eta: 0:00:23 lr: 0.003081 min_lr: 0.003081 loss: 2.2972 (2.2201) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [156] [290/312] eta: 0:00:15 lr: 0.003080 min_lr: 0.003080 loss: 1.9959 (2.2218) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [156] [300/312] eta: 0:00:08 lr: 0.003080 min_lr: 0.003080 loss: 2.2902 (2.2227) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [156] [310/312] eta: 0:00:01 lr: 0.003079 min_lr: 0.003079 loss: 2.3474 (2.2273) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [156] [311/312] eta: 0:00:00 lr: 0.003079 min_lr: 0.003079 loss: 2.3631 (2.2294) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [156] Total time: 0:03:46 (0.7264 s / it) Averaged stats: lr: 0.003079 min_lr: 0.003079 loss: 2.3631 (2.2080) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6427 (0.6427) acc1: 82.2917 (82.2917) acc5: 95.5729 (95.5729) time: 4.5492 data: 4.3441 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0380 (0.9481) acc1: 75.0000 (75.5200) acc5: 92.4479 (92.8640) time: 0.6568 data: 0.4828 max mem: 64948 Test: Total time: 0:00:06 (0.7002 s / it) * Acc@1 76.820 Acc@5 93.472 loss 0.909 Accuracy of the model on the 50000 test images: 76.8% Max accuracy: 77.29% Test: [0/9] eta: 0:00:42 loss: 0.5962 (0.5962) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.6761 data: 4.4700 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7672 (0.7625) acc1: 81.2500 (79.1360) acc5: 95.8333 (95.1040) time: 0.6819 data: 0.5078 max mem: 64948 Test: Total time: 0:00:06 (0.6916 s / it) * Acc@1 79.990 Acc@5 95.130 loss 0.744 Accuracy of the model EMA on 50000 test images: 80.0% Max EMA accuracy: 79.99% Epoch: [157] [ 0/312] eta: 0:49:56 lr: 0.003079 min_lr: 0.003079 loss: 2.5424 (2.5424) weight_decay: 0.0500 (0.0500) time: 9.6029 data: 8.7986 max mem: 64948 Epoch: [157] [ 10/312] eta: 0:07:50 lr: 0.003079 min_lr: 0.003079 loss: 2.2949 (2.1375) weight_decay: 0.0500 (0.0500) time: 1.5567 data: 0.8003 max mem: 64948 Epoch: [157] [ 20/312] eta: 0:05:34 lr: 0.003078 min_lr: 0.003078 loss: 2.1028 (2.1465) weight_decay: 0.0500 (0.0500) time: 0.7221 data: 0.0004 max mem: 64948 Epoch: [157] [ 30/312] eta: 0:04:41 lr: 0.003078 min_lr: 0.003078 loss: 2.3024 (2.2421) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [157] [ 40/312] eta: 0:04:12 lr: 0.003078 min_lr: 0.003078 loss: 2.3285 (2.2245) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [157] [ 50/312] eta: 0:03:50 lr: 0.003077 min_lr: 0.003077 loss: 2.1712 (2.2010) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [157] [ 60/312] eta: 0:03:34 lr: 0.003077 min_lr: 0.003077 loss: 2.1479 (2.1874) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [157] [ 70/312] eta: 0:03:20 lr: 0.003077 min_lr: 0.003077 loss: 2.3339 (2.2065) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [157] [ 80/312] eta: 0:03:08 lr: 0.003076 min_lr: 0.003076 loss: 2.3659 (2.2064) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [157] [ 90/312] eta: 0:02:57 lr: 0.003076 min_lr: 0.003076 loss: 2.0713 (2.1850) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [157] [100/312] eta: 0:02:47 lr: 0.003075 min_lr: 0.003075 loss: 2.0713 (2.1916) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [157] [110/312] eta: 0:02:37 lr: 0.003075 min_lr: 0.003075 loss: 2.2160 (2.2081) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [157] [120/312] eta: 0:02:28 lr: 0.003075 min_lr: 0.003075 loss: 2.2900 (2.2110) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [157] [130/312] eta: 0:02:19 lr: 0.003074 min_lr: 0.003074 loss: 2.3316 (2.2239) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [157] [140/312] eta: 0:02:11 lr: 0.003074 min_lr: 0.003074 loss: 2.2648 (2.2105) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [157] [150/312] eta: 0:02:02 lr: 0.003073 min_lr: 0.003073 loss: 2.1660 (2.2090) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [157] [160/312] eta: 0:01:54 lr: 0.003073 min_lr: 0.003073 loss: 2.3190 (2.2100) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [157] [170/312] eta: 0:01:46 lr: 0.003073 min_lr: 0.003073 loss: 2.3686 (2.2110) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [157] [180/312] eta: 0:01:38 lr: 0.003072 min_lr: 0.003072 loss: 2.4026 (2.2177) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [157] [190/312] eta: 0:01:30 lr: 0.003072 min_lr: 0.003072 loss: 2.3097 (2.2183) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [157] [200/312] eta: 0:01:23 lr: 0.003071 min_lr: 0.003071 loss: 2.3203 (2.2224) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [157] [210/312] eta: 0:01:15 lr: 0.003071 min_lr: 0.003071 loss: 2.4072 (2.2297) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [157] [220/312] eta: 0:01:07 lr: 0.003071 min_lr: 0.003071 loss: 2.2865 (2.2253) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [157] [230/312] eta: 0:01:00 lr: 0.003070 min_lr: 0.003070 loss: 2.3417 (2.2317) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [157] [240/312] eta: 0:00:52 lr: 0.003070 min_lr: 0.003070 loss: 2.3638 (2.2338) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [157] [250/312] eta: 0:00:45 lr: 0.003069 min_lr: 0.003069 loss: 2.3133 (2.2299) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [157] [260/312] eta: 0:00:38 lr: 0.003069 min_lr: 0.003069 loss: 2.1342 (2.2219) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [157] [270/312] eta: 0:00:30 lr: 0.003069 min_lr: 0.003069 loss: 2.1352 (2.2186) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [157] [280/312] eta: 0:00:23 lr: 0.003068 min_lr: 0.003068 loss: 2.2735 (2.2238) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [157] [290/312] eta: 0:00:16 lr: 0.003068 min_lr: 0.003068 loss: 2.3878 (2.2289) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0008 max mem: 64948 Epoch: [157] [300/312] eta: 0:00:08 lr: 0.003067 min_lr: 0.003067 loss: 2.3956 (2.2244) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [157] [310/312] eta: 0:00:01 lr: 0.003067 min_lr: 0.003067 loss: 2.3408 (2.2254) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [157] [311/312] eta: 0:00:00 lr: 0.003067 min_lr: 0.003067 loss: 2.3125 (2.2250) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [157] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.003067 min_lr: 0.003067 loss: 2.3125 (2.2049) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7640 (0.7640) acc1: 81.5104 (81.5104) acc5: 94.0104 (94.0104) time: 4.6198 data: 4.4152 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9298 (0.9318) acc1: 78.1250 (76.7360) acc5: 93.4896 (93.2160) time: 0.6646 data: 0.4907 max mem: 64948 Test: Total time: 0:00:06 (0.6873 s / it) * Acc@1 77.260 Acc@5 93.862 loss 0.897 Accuracy of the model on the 50000 test images: 77.3% Max accuracy: 77.29% Test: [0/9] eta: 0:00:40 loss: 0.5949 (0.5949) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.5127 data: 4.3071 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7669 (0.7612) acc1: 81.2500 (79.2640) acc5: 95.8333 (95.1360) time: 0.6527 data: 0.4787 max mem: 64948 Test: Total time: 0:00:05 (0.6608 s / it) * Acc@1 80.010 Acc@5 95.128 loss 0.743 Accuracy of the model EMA on 50000 test images: 80.0% Max EMA accuracy: 80.01% Epoch: [158] [ 0/312] eta: 0:54:36 lr: 0.003067 min_lr: 0.003067 loss: 2.6231 (2.6231) weight_decay: 0.0500 (0.0500) time: 10.5013 data: 9.7098 max mem: 64948 Epoch: [158] [ 10/312] eta: 0:08:02 lr: 0.003067 min_lr: 0.003067 loss: 2.3186 (2.3636) weight_decay: 0.0500 (0.0500) time: 1.5993 data: 0.8830 max mem: 64948 Epoch: [158] [ 20/312] eta: 0:05:41 lr: 0.003066 min_lr: 0.003066 loss: 2.0351 (2.1829) weight_decay: 0.0500 (0.0500) time: 0.7026 data: 0.0003 max mem: 64948 Epoch: [158] [ 30/312] eta: 0:04:46 lr: 0.003066 min_lr: 0.003066 loss: 1.9685 (2.1478) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [158] [ 40/312] eta: 0:04:15 lr: 0.003065 min_lr: 0.003065 loss: 2.1675 (2.1574) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [158] [ 50/312] eta: 0:03:53 lr: 0.003065 min_lr: 0.003065 loss: 2.2845 (2.1937) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [158] [ 60/312] eta: 0:03:36 lr: 0.003065 min_lr: 0.003065 loss: 2.3288 (2.2166) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [158] [ 70/312] eta: 0:03:21 lr: 0.003064 min_lr: 0.003064 loss: 2.2878 (2.2214) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [158] [ 80/312] eta: 0:03:09 lr: 0.003064 min_lr: 0.003064 loss: 2.3087 (2.2379) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [158] [ 90/312] eta: 0:02:58 lr: 0.003063 min_lr: 0.003063 loss: 2.3749 (2.2436) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [158] [100/312] eta: 0:02:48 lr: 0.003063 min_lr: 0.003063 loss: 2.3749 (2.2604) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [158] [110/312] eta: 0:02:38 lr: 0.003063 min_lr: 0.003063 loss: 2.3836 (2.2583) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [158] [120/312] eta: 0:02:29 lr: 0.003062 min_lr: 0.003062 loss: 2.2718 (2.2507) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [158] [130/312] eta: 0:02:20 lr: 0.003062 min_lr: 0.003062 loss: 2.2160 (2.2468) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [158] [140/312] eta: 0:02:11 lr: 0.003061 min_lr: 0.003061 loss: 2.1664 (2.2408) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [158] [150/312] eta: 0:02:03 lr: 0.003061 min_lr: 0.003061 loss: 2.2429 (2.2382) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [158] [160/312] eta: 0:01:54 lr: 0.003061 min_lr: 0.003061 loss: 2.3128 (2.2454) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [158] [170/312] eta: 0:01:46 lr: 0.003060 min_lr: 0.003060 loss: 2.2638 (2.2420) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [158] [180/312] eta: 0:01:38 lr: 0.003060 min_lr: 0.003060 loss: 2.1652 (2.2350) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [158] [190/312] eta: 0:01:31 lr: 0.003059 min_lr: 0.003059 loss: 2.2732 (2.2272) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [158] [200/312] eta: 0:01:23 lr: 0.003059 min_lr: 0.003059 loss: 2.1550 (2.2234) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [158] [210/312] eta: 0:01:15 lr: 0.003059 min_lr: 0.003059 loss: 2.1404 (2.2233) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [158] [220/312] eta: 0:01:08 lr: 0.003058 min_lr: 0.003058 loss: 2.3831 (2.2293) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [158] [230/312] eta: 0:01:00 lr: 0.003058 min_lr: 0.003058 loss: 2.3683 (2.2236) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [158] [240/312] eta: 0:00:52 lr: 0.003057 min_lr: 0.003057 loss: 2.0535 (2.2165) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [158] [250/312] eta: 0:00:45 lr: 0.003057 min_lr: 0.003057 loss: 2.1994 (2.2186) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [158] [260/312] eta: 0:00:38 lr: 0.003057 min_lr: 0.003057 loss: 2.2880 (2.2144) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [158] [270/312] eta: 0:00:30 lr: 0.003056 min_lr: 0.003056 loss: 2.0697 (2.2081) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [158] [280/312] eta: 0:00:23 lr: 0.003056 min_lr: 0.003056 loss: 2.2249 (2.2088) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [158] [290/312] eta: 0:00:16 lr: 0.003055 min_lr: 0.003055 loss: 2.1274 (2.2041) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [158] [300/312] eta: 0:00:08 lr: 0.003055 min_lr: 0.003055 loss: 2.1317 (2.2069) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [158] [310/312] eta: 0:00:01 lr: 0.003055 min_lr: 0.003055 loss: 2.1317 (2.1998) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [158] [311/312] eta: 0:00:00 lr: 0.003055 min_lr: 0.003055 loss: 2.1317 (2.2006) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [158] Total time: 0:03:47 (0.7290 s / it) Averaged stats: lr: 0.003055 min_lr: 0.003055 loss: 2.1317 (2.2011) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7115 (0.7115) acc1: 84.1146 (84.1146) acc5: 93.7500 (93.7500) time: 4.6333 data: 4.4292 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8947 (0.9322) acc1: 78.9062 (75.8720) acc5: 93.2292 (93.0880) time: 0.6661 data: 0.4922 max mem: 64948 Test: Total time: 0:00:06 (0.6875 s / it) * Acc@1 77.122 Acc@5 93.642 loss 0.891 Accuracy of the model on the 50000 test images: 77.1% Max accuracy: 77.29% Test: [0/9] eta: 0:00:43 loss: 0.5944 (0.5944) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.8589 data: 4.6441 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7660 (0.7599) acc1: 80.9896 (79.2000) acc5: 95.8333 (95.1040) time: 0.6916 data: 0.5161 max mem: 64948 Test: Total time: 0:00:06 (0.7022 s / it) * Acc@1 80.048 Acc@5 95.140 loss 0.741 Accuracy of the model EMA on 50000 test images: 80.0% Max EMA accuracy: 80.05% Epoch: [159] [ 0/312] eta: 0:50:49 lr: 0.003055 min_lr: 0.003055 loss: 2.7306 (2.7306) weight_decay: 0.0500 (0.0500) time: 9.7746 data: 8.9744 max mem: 64948 Epoch: [159] [ 10/312] eta: 0:07:55 lr: 0.003054 min_lr: 0.003054 loss: 2.3664 (2.2338) weight_decay: 0.0500 (0.0500) time: 1.5738 data: 0.8162 max mem: 64948 Epoch: [159] [ 20/312] eta: 0:05:37 lr: 0.003054 min_lr: 0.003054 loss: 2.1013 (2.1748) weight_decay: 0.0500 (0.0500) time: 0.7234 data: 0.0003 max mem: 64948 Epoch: [159] [ 30/312] eta: 0:04:44 lr: 0.003053 min_lr: 0.003053 loss: 2.1538 (2.1534) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [159] [ 40/312] eta: 0:04:13 lr: 0.003053 min_lr: 0.003053 loss: 2.2810 (2.1708) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [159] [ 50/312] eta: 0:03:51 lr: 0.003053 min_lr: 0.003053 loss: 2.3349 (2.1691) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [159] [ 60/312] eta: 0:03:35 lr: 0.003052 min_lr: 0.003052 loss: 2.2421 (2.1490) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [159] [ 70/312] eta: 0:03:21 lr: 0.003052 min_lr: 0.003052 loss: 2.2312 (2.1633) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [159] [ 80/312] eta: 0:03:08 lr: 0.003051 min_lr: 0.003051 loss: 2.2834 (2.1858) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [159] [ 90/312] eta: 0:02:57 lr: 0.003051 min_lr: 0.003051 loss: 2.2582 (2.1826) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [159] [100/312] eta: 0:02:47 lr: 0.003051 min_lr: 0.003051 loss: 2.0758 (2.1690) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [159] [110/312] eta: 0:02:37 lr: 0.003050 min_lr: 0.003050 loss: 1.9818 (2.1622) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [159] [120/312] eta: 0:02:28 lr: 0.003050 min_lr: 0.003050 loss: 2.0770 (2.1626) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [159] [130/312] eta: 0:02:19 lr: 0.003049 min_lr: 0.003049 loss: 2.3238 (2.1818) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [159] [140/312] eta: 0:02:11 lr: 0.003049 min_lr: 0.003049 loss: 2.3208 (2.1821) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [159] [150/312] eta: 0:02:03 lr: 0.003049 min_lr: 0.003049 loss: 2.1355 (2.1747) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [159] [160/312] eta: 0:01:54 lr: 0.003048 min_lr: 0.003048 loss: 2.1567 (2.1854) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [159] [170/312] eta: 0:01:46 lr: 0.003048 min_lr: 0.003048 loss: 2.3434 (2.1770) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [159] [180/312] eta: 0:01:38 lr: 0.003047 min_lr: 0.003047 loss: 2.2788 (2.1812) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [159] [190/312] eta: 0:01:31 lr: 0.003047 min_lr: 0.003047 loss: 2.2327 (2.1820) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [159] [200/312] eta: 0:01:23 lr: 0.003047 min_lr: 0.003047 loss: 2.2327 (2.1841) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [159] [210/312] eta: 0:01:15 lr: 0.003046 min_lr: 0.003046 loss: 2.3253 (2.1931) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [159] [220/312] eta: 0:01:07 lr: 0.003046 min_lr: 0.003046 loss: 2.3961 (2.1983) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [159] [230/312] eta: 0:01:00 lr: 0.003045 min_lr: 0.003045 loss: 2.2379 (2.1994) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [159] [240/312] eta: 0:00:52 lr: 0.003045 min_lr: 0.003045 loss: 2.3115 (2.2014) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [159] [250/312] eta: 0:00:45 lr: 0.003045 min_lr: 0.003045 loss: 2.0750 (2.1929) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [159] [260/312] eta: 0:00:38 lr: 0.003044 min_lr: 0.003044 loss: 2.1562 (2.1981) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [159] [270/312] eta: 0:00:30 lr: 0.003044 min_lr: 0.003044 loss: 2.3605 (2.2057) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [159] [280/312] eta: 0:00:23 lr: 0.003043 min_lr: 0.003043 loss: 2.3502 (2.2070) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0009 max mem: 64948 Epoch: [159] [290/312] eta: 0:00:16 lr: 0.003043 min_lr: 0.003043 loss: 2.3644 (2.2118) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0008 max mem: 64948 Epoch: [159] [300/312] eta: 0:00:08 lr: 0.003043 min_lr: 0.003043 loss: 2.3475 (2.2124) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [159] [310/312] eta: 0:00:01 lr: 0.003042 min_lr: 0.003042 loss: 2.3052 (2.2181) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [159] [311/312] eta: 0:00:00 lr: 0.003042 min_lr: 0.003042 loss: 2.2815 (2.2166) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [159] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.003042 min_lr: 0.003042 loss: 2.2815 (2.2064) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.6884 (0.6884) acc1: 82.8125 (82.8125) acc5: 94.7917 (94.7917) time: 4.4354 data: 4.2146 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9372 (0.9238) acc1: 77.6042 (76.6720) acc5: 94.7917 (93.8240) time: 0.6448 data: 0.4684 max mem: 64948 Test: Total time: 0:00:05 (0.6537 s / it) * Acc@1 77.342 Acc@5 93.720 loss 0.898 Accuracy of the model on the 50000 test images: 77.3% Max accuracy: 77.34% Test: [0/9] eta: 0:00:40 loss: 0.5937 (0.5937) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.4867 data: 4.2792 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7650 (0.7589) acc1: 80.9896 (79.1680) acc5: 95.8333 (95.0720) time: 0.6499 data: 0.4756 max mem: 64948 Test: Total time: 0:00:05 (0.6572 s / it) * Acc@1 80.080 Acc@5 95.146 loss 0.740 Accuracy of the model EMA on 50000 test images: 80.1% Max EMA accuracy: 80.08% Epoch: [160] [ 0/312] eta: 0:55:43 lr: 0.003042 min_lr: 0.003042 loss: 1.8975 (1.8975) weight_decay: 0.0500 (0.0500) time: 10.7166 data: 9.9833 max mem: 64948 Epoch: [160] [ 10/312] eta: 0:08:07 lr: 0.003042 min_lr: 0.003042 loss: 2.2359 (2.3106) weight_decay: 0.0500 (0.0500) time: 1.6145 data: 0.9079 max mem: 64948 Epoch: [160] [ 20/312] eta: 0:05:43 lr: 0.003041 min_lr: 0.003041 loss: 2.4070 (2.3876) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0003 max mem: 64948 Epoch: [160] [ 30/312] eta: 0:04:48 lr: 0.003041 min_lr: 0.003041 loss: 2.4553 (2.3710) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [160] [ 40/312] eta: 0:04:16 lr: 0.003041 min_lr: 0.003041 loss: 2.3580 (2.3124) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [160] [ 50/312] eta: 0:03:54 lr: 0.003040 min_lr: 0.003040 loss: 2.1971 (2.2719) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [160] [ 60/312] eta: 0:03:37 lr: 0.003040 min_lr: 0.003040 loss: 2.3270 (2.2931) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [160] [ 70/312] eta: 0:03:23 lr: 0.003039 min_lr: 0.003039 loss: 2.4310 (2.3078) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [160] [ 80/312] eta: 0:03:10 lr: 0.003039 min_lr: 0.003039 loss: 2.2458 (2.2747) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [160] [ 90/312] eta: 0:02:59 lr: 0.003039 min_lr: 0.003039 loss: 1.9728 (2.2539) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [160] [100/312] eta: 0:02:48 lr: 0.003038 min_lr: 0.003038 loss: 2.0049 (2.2366) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [160] [110/312] eta: 0:02:39 lr: 0.003038 min_lr: 0.003038 loss: 2.2465 (2.2396) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [160] [120/312] eta: 0:02:29 lr: 0.003037 min_lr: 0.003037 loss: 2.1442 (2.2128) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [160] [130/312] eta: 0:02:20 lr: 0.003037 min_lr: 0.003037 loss: 2.1442 (2.2149) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [160] [140/312] eta: 0:02:12 lr: 0.003037 min_lr: 0.003037 loss: 2.3401 (2.2244) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [160] [150/312] eta: 0:02:03 lr: 0.003036 min_lr: 0.003036 loss: 2.4014 (2.2357) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [160] [160/312] eta: 0:01:55 lr: 0.003036 min_lr: 0.003036 loss: 2.3389 (2.2359) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [160] [170/312] eta: 0:01:47 lr: 0.003035 min_lr: 0.003035 loss: 2.2166 (2.2306) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [160] [180/312] eta: 0:01:39 lr: 0.003035 min_lr: 0.003035 loss: 2.1681 (2.2266) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [160] [190/312] eta: 0:01:31 lr: 0.003035 min_lr: 0.003035 loss: 2.0653 (2.2090) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [160] [200/312] eta: 0:01:23 lr: 0.003034 min_lr: 0.003034 loss: 2.1941 (2.2120) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [160] [210/312] eta: 0:01:15 lr: 0.003034 min_lr: 0.003034 loss: 2.2069 (2.2015) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [160] [220/312] eta: 0:01:08 lr: 0.003033 min_lr: 0.003033 loss: 2.0204 (2.2021) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [160] [230/312] eta: 0:01:00 lr: 0.003033 min_lr: 0.003033 loss: 2.2323 (2.2015) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [160] [240/312] eta: 0:00:53 lr: 0.003033 min_lr: 0.003033 loss: 2.2928 (2.2039) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [160] [250/312] eta: 0:00:45 lr: 0.003032 min_lr: 0.003032 loss: 2.2964 (2.2027) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [160] [260/312] eta: 0:00:38 lr: 0.003032 min_lr: 0.003032 loss: 2.2007 (2.2015) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [160] [270/312] eta: 0:00:30 lr: 0.003031 min_lr: 0.003031 loss: 2.2160 (2.2045) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [160] [280/312] eta: 0:00:23 lr: 0.003031 min_lr: 0.003031 loss: 2.2547 (2.2012) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [160] [290/312] eta: 0:00:16 lr: 0.003031 min_lr: 0.003031 loss: 2.2494 (2.2039) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [160] [300/312] eta: 0:00:08 lr: 0.003030 min_lr: 0.003030 loss: 2.2466 (2.2031) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [160] [310/312] eta: 0:00:01 lr: 0.003030 min_lr: 0.003030 loss: 2.1695 (2.2009) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [160] [311/312] eta: 0:00:00 lr: 0.003030 min_lr: 0.003030 loss: 2.1695 (2.2017) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [160] Total time: 0:03:47 (0.7301 s / it) Averaged stats: lr: 0.003030 min_lr: 0.003030 loss: 2.1695 (2.1942) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5963 (0.5963) acc1: 85.6771 (85.6771) acc5: 97.1354 (97.1354) time: 4.6795 data: 4.4606 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9523 (0.8759) acc1: 78.6458 (77.5680) acc5: 94.5312 (93.9200) time: 0.6712 data: 0.4957 max mem: 64948 Test: Total time: 0:00:06 (0.6939 s / it) * Acc@1 77.262 Acc@5 93.696 loss 0.874 Accuracy of the model on the 50000 test images: 77.3% Max accuracy: 77.34% Test: [0/9] eta: 0:00:41 loss: 0.5931 (0.5931) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.6097 data: 4.4012 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7639 (0.7579) acc1: 80.7292 (79.1360) acc5: 95.8333 (95.1040) time: 0.6635 data: 0.4891 max mem: 64948 Test: Total time: 0:00:06 (0.6711 s / it) * Acc@1 80.136 Acc@5 95.166 loss 0.738 Accuracy of the model EMA on 50000 test images: 80.1% Max EMA accuracy: 80.14% Epoch: [161] [ 0/312] eta: 0:52:25 lr: 0.003030 min_lr: 0.003030 loss: 1.8760 (1.8760) weight_decay: 0.0500 (0.0500) time: 10.0827 data: 9.3125 max mem: 64948 Epoch: [161] [ 10/312] eta: 0:07:53 lr: 0.003029 min_lr: 0.003029 loss: 2.2914 (2.1341) weight_decay: 0.0500 (0.0500) time: 1.5682 data: 0.8470 max mem: 64948 Epoch: [161] [ 20/312] eta: 0:05:36 lr: 0.003029 min_lr: 0.003029 loss: 2.2914 (2.1550) weight_decay: 0.0500 (0.0500) time: 0.7050 data: 0.0004 max mem: 64948 Epoch: [161] [ 30/312] eta: 0:04:43 lr: 0.003028 min_lr: 0.003028 loss: 2.1600 (2.1493) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [161] [ 40/312] eta: 0:04:12 lr: 0.003028 min_lr: 0.003028 loss: 2.2540 (2.1905) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [161] [ 50/312] eta: 0:03:51 lr: 0.003028 min_lr: 0.003028 loss: 2.3023 (2.1875) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [161] [ 60/312] eta: 0:03:34 lr: 0.003027 min_lr: 0.003027 loss: 2.3126 (2.2150) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [161] [ 70/312] eta: 0:03:20 lr: 0.003027 min_lr: 0.003027 loss: 2.3126 (2.1960) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [161] [ 80/312] eta: 0:03:08 lr: 0.003026 min_lr: 0.003026 loss: 1.9371 (2.1775) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [161] [ 90/312] eta: 0:02:57 lr: 0.003026 min_lr: 0.003026 loss: 1.9141 (2.1758) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [161] [100/312] eta: 0:02:47 lr: 0.003026 min_lr: 0.003026 loss: 2.1296 (2.1939) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [161] [110/312] eta: 0:02:37 lr: 0.003025 min_lr: 0.003025 loss: 2.3964 (2.2033) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [161] [120/312] eta: 0:02:28 lr: 0.003025 min_lr: 0.003025 loss: 2.2523 (2.2127) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [161] [130/312] eta: 0:02:19 lr: 0.003024 min_lr: 0.003024 loss: 2.2369 (2.2016) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [161] [140/312] eta: 0:02:11 lr: 0.003024 min_lr: 0.003024 loss: 2.2369 (2.2067) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [161] [150/312] eta: 0:02:02 lr: 0.003024 min_lr: 0.003024 loss: 2.4055 (2.2118) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [161] [160/312] eta: 0:01:54 lr: 0.003023 min_lr: 0.003023 loss: 2.2964 (2.2100) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [161] [170/312] eta: 0:01:46 lr: 0.003023 min_lr: 0.003023 loss: 2.3877 (2.2266) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [161] [180/312] eta: 0:01:38 lr: 0.003022 min_lr: 0.003022 loss: 2.4584 (2.2366) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [161] [190/312] eta: 0:01:30 lr: 0.003022 min_lr: 0.003022 loss: 2.3686 (2.2399) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [161] [200/312] eta: 0:01:23 lr: 0.003022 min_lr: 0.003022 loss: 2.3079 (2.2451) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [161] [210/312] eta: 0:01:15 lr: 0.003021 min_lr: 0.003021 loss: 2.3041 (2.2464) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [161] [220/312] eta: 0:01:07 lr: 0.003021 min_lr: 0.003021 loss: 2.2906 (2.2478) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [161] [230/312] eta: 0:01:00 lr: 0.003020 min_lr: 0.003020 loss: 2.4015 (2.2531) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [161] [240/312] eta: 0:00:52 lr: 0.003020 min_lr: 0.003020 loss: 2.4015 (2.2527) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [161] [250/312] eta: 0:00:45 lr: 0.003020 min_lr: 0.003020 loss: 2.2488 (2.2465) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [161] [260/312] eta: 0:00:38 lr: 0.003019 min_lr: 0.003019 loss: 2.1913 (2.2422) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [161] [270/312] eta: 0:00:30 lr: 0.003019 min_lr: 0.003019 loss: 2.2920 (2.2445) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [161] [280/312] eta: 0:00:23 lr: 0.003018 min_lr: 0.003018 loss: 2.3980 (2.2432) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [161] [290/312] eta: 0:00:16 lr: 0.003018 min_lr: 0.003018 loss: 2.1340 (2.2397) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0008 max mem: 64948 Epoch: [161] [300/312] eta: 0:00:08 lr: 0.003018 min_lr: 0.003018 loss: 2.1600 (2.2378) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [161] [310/312] eta: 0:00:01 lr: 0.003017 min_lr: 0.003017 loss: 2.3177 (2.2431) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [161] [311/312] eta: 0:00:00 lr: 0.003017 min_lr: 0.003017 loss: 2.4340 (2.2443) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [161] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.003017 min_lr: 0.003017 loss: 2.4340 (2.1978) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6449 (0.6449) acc1: 83.3333 (83.3333) acc5: 96.6146 (96.6146) time: 4.7458 data: 4.5299 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9677 (0.9130) acc1: 78.3854 (76.8320) acc5: 95.0521 (93.3440) time: 0.6785 data: 0.5034 max mem: 64948 Test: Total time: 0:00:06 (0.7032 s / it) * Acc@1 77.630 Acc@5 93.852 loss 0.891 Accuracy of the model on the 50000 test images: 77.6% Max accuracy: 77.63% Test: [0/9] eta: 0:00:41 loss: 0.5922 (0.5922) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.6324 data: 4.4283 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7629 (0.7569) acc1: 80.9896 (79.2320) acc5: 95.8333 (95.1040) time: 0.6660 data: 0.4921 max mem: 64948 Test: Total time: 0:00:06 (0.6731 s / it) * Acc@1 80.182 Acc@5 95.180 loss 0.737 Accuracy of the model EMA on 50000 test images: 80.2% Max EMA accuracy: 80.18% Epoch: [162] [ 0/312] eta: 0:44:29 lr: 0.003017 min_lr: 0.003017 loss: 2.3291 (2.3291) weight_decay: 0.0500 (0.0500) time: 8.5556 data: 7.6971 max mem: 64948 Epoch: [162] [ 10/312] eta: 0:07:36 lr: 0.003017 min_lr: 0.003017 loss: 2.3640 (2.3275) weight_decay: 0.0500 (0.0500) time: 1.5125 data: 0.7850 max mem: 64948 Epoch: [162] [ 20/312] eta: 0:05:27 lr: 0.003016 min_lr: 0.003016 loss: 2.3628 (2.2450) weight_decay: 0.0500 (0.0500) time: 0.7508 data: 0.0471 max mem: 64948 Epoch: [162] [ 30/312] eta: 0:04:37 lr: 0.003016 min_lr: 0.003016 loss: 2.3552 (2.2786) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [162] [ 40/312] eta: 0:04:08 lr: 0.003015 min_lr: 0.003015 loss: 2.3552 (2.2604) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [162] [ 50/312] eta: 0:03:48 lr: 0.003015 min_lr: 0.003015 loss: 2.3451 (2.2558) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [162] [ 60/312] eta: 0:03:32 lr: 0.003015 min_lr: 0.003015 loss: 2.1623 (2.2260) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [162] [ 70/312] eta: 0:03:18 lr: 0.003014 min_lr: 0.003014 loss: 2.1178 (2.2090) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [162] [ 80/312] eta: 0:03:07 lr: 0.003014 min_lr: 0.003014 loss: 2.1178 (2.1867) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [162] [ 90/312] eta: 0:02:56 lr: 0.003013 min_lr: 0.003013 loss: 1.9553 (2.1673) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [162] [100/312] eta: 0:02:46 lr: 0.003013 min_lr: 0.003013 loss: 2.2758 (2.1892) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [162] [110/312] eta: 0:02:36 lr: 0.003013 min_lr: 0.003013 loss: 1.9478 (2.1617) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [162] [120/312] eta: 0:02:27 lr: 0.003012 min_lr: 0.003012 loss: 1.9478 (2.1723) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [162] [130/312] eta: 0:02:19 lr: 0.003012 min_lr: 0.003012 loss: 2.3913 (2.1798) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [162] [140/312] eta: 0:02:10 lr: 0.003011 min_lr: 0.003011 loss: 2.3862 (2.1847) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [162] [150/312] eta: 0:02:02 lr: 0.003011 min_lr: 0.003011 loss: 2.4074 (2.1983) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [162] [160/312] eta: 0:01:54 lr: 0.003011 min_lr: 0.003011 loss: 2.4074 (2.2002) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [162] [170/312] eta: 0:01:46 lr: 0.003010 min_lr: 0.003010 loss: 2.2741 (2.2023) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [162] [180/312] eta: 0:01:38 lr: 0.003010 min_lr: 0.003010 loss: 2.3186 (2.2107) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [162] [190/312] eta: 0:01:30 lr: 0.003009 min_lr: 0.003009 loss: 2.3423 (2.2139) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [162] [200/312] eta: 0:01:22 lr: 0.003009 min_lr: 0.003009 loss: 2.3423 (2.2185) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [162] [210/312] eta: 0:01:15 lr: 0.003009 min_lr: 0.003009 loss: 2.2897 (2.2107) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [162] [220/312] eta: 0:01:07 lr: 0.003008 min_lr: 0.003008 loss: 2.0334 (2.2035) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [162] [230/312] eta: 0:01:00 lr: 0.003008 min_lr: 0.003008 loss: 2.1559 (2.2011) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [162] [240/312] eta: 0:00:52 lr: 0.003007 min_lr: 0.003007 loss: 2.1350 (2.1910) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [162] [250/312] eta: 0:00:45 lr: 0.003007 min_lr: 0.003007 loss: 2.1254 (2.1908) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [162] [260/312] eta: 0:00:37 lr: 0.003007 min_lr: 0.003007 loss: 2.0825 (2.1895) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [162] [270/312] eta: 0:00:30 lr: 0.003006 min_lr: 0.003006 loss: 2.3013 (2.1933) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [162] [280/312] eta: 0:00:23 lr: 0.003006 min_lr: 0.003006 loss: 2.2486 (2.1898) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [162] [290/312] eta: 0:00:15 lr: 0.003005 min_lr: 0.003005 loss: 2.3177 (2.1962) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [162] [300/312] eta: 0:00:08 lr: 0.003005 min_lr: 0.003005 loss: 2.3874 (2.2000) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [162] [310/312] eta: 0:00:01 lr: 0.003005 min_lr: 0.003005 loss: 2.3736 (2.2046) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [162] [311/312] eta: 0:00:00 lr: 0.003005 min_lr: 0.003005 loss: 2.3448 (2.2026) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [162] Total time: 0:03:46 (0.7268 s / it) Averaged stats: lr: 0.003005 min_lr: 0.003005 loss: 2.3448 (2.1974) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6680 (0.6680) acc1: 85.6771 (85.6771) acc5: 95.8333 (95.8333) time: 4.7050 data: 4.5012 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0251 (0.9037) acc1: 76.0417 (77.3760) acc5: 93.2292 (93.7920) time: 0.6741 data: 0.5002 max mem: 64948 Test: Total time: 0:00:06 (0.6989 s / it) * Acc@1 77.358 Acc@5 93.796 loss 0.883 Accuracy of the model on the 50000 test images: 77.4% Max accuracy: 77.63% Test: [0/9] eta: 0:00:42 loss: 0.5906 (0.5906) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.6930 data: 4.4888 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7622 (0.7561) acc1: 80.7292 (79.2640) acc5: 95.8333 (95.1680) time: 0.6729 data: 0.4989 max mem: 64948 Test: Total time: 0:00:06 (0.6857 s / it) * Acc@1 80.214 Acc@5 95.200 loss 0.735 Accuracy of the model EMA on 50000 test images: 80.2% Max EMA accuracy: 80.21% Epoch: [163] [ 0/312] eta: 0:50:32 lr: 0.003004 min_lr: 0.003004 loss: 1.8558 (1.8558) weight_decay: 0.0500 (0.0500) time: 9.7182 data: 8.9200 max mem: 64948 Epoch: [163] [ 10/312] eta: 0:07:42 lr: 0.003004 min_lr: 0.003004 loss: 2.5096 (2.3893) weight_decay: 0.0500 (0.0500) time: 1.5325 data: 0.8112 max mem: 64948 Epoch: [163] [ 20/312] eta: 0:05:31 lr: 0.003004 min_lr: 0.003004 loss: 2.4832 (2.3549) weight_decay: 0.0500 (0.0500) time: 0.7056 data: 0.0004 max mem: 64948 Epoch: [163] [ 30/312] eta: 0:04:40 lr: 0.003003 min_lr: 0.003003 loss: 2.3597 (2.3228) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [163] [ 40/312] eta: 0:04:10 lr: 0.003003 min_lr: 0.003003 loss: 2.3799 (2.3396) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [163] [ 50/312] eta: 0:03:49 lr: 0.003002 min_lr: 0.003002 loss: 2.3799 (2.3164) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [163] [ 60/312] eta: 0:03:33 lr: 0.003002 min_lr: 0.003002 loss: 2.3396 (2.3243) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [163] [ 70/312] eta: 0:03:19 lr: 0.003002 min_lr: 0.003002 loss: 2.3396 (2.3283) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [163] [ 80/312] eta: 0:03:07 lr: 0.003001 min_lr: 0.003001 loss: 2.1809 (2.3045) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [163] [ 90/312] eta: 0:02:57 lr: 0.003001 min_lr: 0.003001 loss: 2.1776 (2.2947) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [163] [100/312] eta: 0:02:46 lr: 0.003000 min_lr: 0.003000 loss: 2.3639 (2.3039) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [163] [110/312] eta: 0:02:37 lr: 0.003000 min_lr: 0.003000 loss: 2.3232 (2.2873) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [163] [120/312] eta: 0:02:28 lr: 0.003000 min_lr: 0.003000 loss: 2.1414 (2.2914) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [163] [130/312] eta: 0:02:19 lr: 0.002999 min_lr: 0.002999 loss: 2.1270 (2.2690) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [163] [140/312] eta: 0:02:10 lr: 0.002999 min_lr: 0.002999 loss: 1.9807 (2.2578) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [163] [150/312] eta: 0:02:02 lr: 0.002998 min_lr: 0.002998 loss: 2.2069 (2.2565) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [163] [160/312] eta: 0:01:54 lr: 0.002998 min_lr: 0.002998 loss: 2.2069 (2.2472) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [163] [170/312] eta: 0:01:46 lr: 0.002998 min_lr: 0.002998 loss: 2.2725 (2.2432) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [163] [180/312] eta: 0:01:38 lr: 0.002997 min_lr: 0.002997 loss: 2.2884 (2.2466) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [163] [190/312] eta: 0:01:30 lr: 0.002997 min_lr: 0.002997 loss: 2.2719 (2.2429) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [163] [200/312] eta: 0:01:22 lr: 0.002996 min_lr: 0.002996 loss: 2.2920 (2.2419) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [163] [210/312] eta: 0:01:15 lr: 0.002996 min_lr: 0.002996 loss: 2.2420 (2.2419) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [163] [220/312] eta: 0:01:07 lr: 0.002996 min_lr: 0.002996 loss: 2.0083 (2.2341) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [163] [230/312] eta: 0:01:00 lr: 0.002995 min_lr: 0.002995 loss: 1.9913 (2.2348) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [163] [240/312] eta: 0:00:52 lr: 0.002995 min_lr: 0.002995 loss: 2.2128 (2.2366) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [163] [250/312] eta: 0:00:45 lr: 0.002994 min_lr: 0.002994 loss: 2.3747 (2.2390) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [163] [260/312] eta: 0:00:37 lr: 0.002994 min_lr: 0.002994 loss: 2.3197 (2.2371) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [163] [270/312] eta: 0:00:30 lr: 0.002994 min_lr: 0.002994 loss: 2.2923 (2.2343) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [163] [280/312] eta: 0:00:23 lr: 0.002993 min_lr: 0.002993 loss: 2.2271 (2.2237) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [163] [290/312] eta: 0:00:15 lr: 0.002993 min_lr: 0.002993 loss: 2.0580 (2.2219) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [163] [300/312] eta: 0:00:08 lr: 0.002992 min_lr: 0.002992 loss: 2.3379 (2.2246) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [163] [310/312] eta: 0:00:01 lr: 0.002992 min_lr: 0.002992 loss: 2.3823 (2.2262) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [163] [311/312] eta: 0:00:00 lr: 0.002992 min_lr: 0.002992 loss: 2.3192 (2.2247) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [163] Total time: 0:03:46 (0.7272 s / it) Averaged stats: lr: 0.002992 min_lr: 0.002992 loss: 2.3192 (2.1918) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7353 (0.7353) acc1: 81.7708 (81.7708) acc5: 94.0104 (94.0104) time: 4.4772 data: 4.2683 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9268 (0.8989) acc1: 76.0417 (76.1600) acc5: 94.0104 (93.8240) time: 0.6487 data: 0.4743 max mem: 64948 Test: Total time: 0:00:06 (0.6706 s / it) * Acc@1 77.628 Acc@5 93.776 loss 0.871 Accuracy of the model on the 50000 test images: 77.6% Max accuracy: 77.63% Test: [0/9] eta: 0:00:40 loss: 0.5890 (0.5890) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.5334 data: 4.3286 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7613 (0.7548) acc1: 80.7292 (79.2960) acc5: 95.8333 (95.2000) time: 0.6550 data: 0.4811 max mem: 64948 Test: Total time: 0:00:05 (0.6657 s / it) * Acc@1 80.250 Acc@5 95.218 loss 0.734 Accuracy of the model EMA on 50000 test images: 80.3% Max EMA accuracy: 80.25% Epoch: [164] [ 0/312] eta: 0:49:42 lr: 0.002992 min_lr: 0.002992 loss: 2.3868 (2.3868) weight_decay: 0.0500 (0.0500) time: 9.5588 data: 8.6462 max mem: 64948 Epoch: [164] [ 10/312] eta: 0:07:38 lr: 0.002991 min_lr: 0.002991 loss: 2.1048 (2.0398) weight_decay: 0.0500 (0.0500) time: 1.5197 data: 0.7864 max mem: 64948 Epoch: [164] [ 20/312] eta: 0:05:29 lr: 0.002991 min_lr: 0.002991 loss: 2.1542 (2.1875) weight_decay: 0.0500 (0.0500) time: 0.7060 data: 0.0004 max mem: 64948 Epoch: [164] [ 30/312] eta: 0:04:38 lr: 0.002991 min_lr: 0.002991 loss: 2.2925 (2.1931) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [164] [ 40/312] eta: 0:04:09 lr: 0.002990 min_lr: 0.002990 loss: 2.1879 (2.1475) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [164] [ 50/312] eta: 0:03:48 lr: 0.002990 min_lr: 0.002990 loss: 2.0452 (2.1205) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [164] [ 60/312] eta: 0:03:33 lr: 0.002989 min_lr: 0.002989 loss: 1.8990 (2.0986) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [164] [ 70/312] eta: 0:03:19 lr: 0.002989 min_lr: 0.002989 loss: 2.0040 (2.0796) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [164] [ 80/312] eta: 0:03:07 lr: 0.002989 min_lr: 0.002989 loss: 1.8532 (2.0625) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [164] [ 90/312] eta: 0:02:56 lr: 0.002988 min_lr: 0.002988 loss: 2.2485 (2.0775) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [164] [100/312] eta: 0:02:46 lr: 0.002988 min_lr: 0.002988 loss: 2.3439 (2.0918) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [164] [110/312] eta: 0:02:37 lr: 0.002987 min_lr: 0.002987 loss: 2.2913 (2.0954) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [164] [120/312] eta: 0:02:28 lr: 0.002987 min_lr: 0.002987 loss: 2.2843 (2.1037) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [164] [130/312] eta: 0:02:19 lr: 0.002987 min_lr: 0.002987 loss: 2.3005 (2.1301) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [164] [140/312] eta: 0:02:10 lr: 0.002986 min_lr: 0.002986 loss: 2.3005 (2.1261) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [164] [150/312] eta: 0:02:02 lr: 0.002986 min_lr: 0.002986 loss: 1.9221 (2.1158) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [164] [160/312] eta: 0:01:54 lr: 0.002985 min_lr: 0.002985 loss: 2.2453 (2.1245) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [164] [170/312] eta: 0:01:46 lr: 0.002985 min_lr: 0.002985 loss: 2.3431 (2.1289) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [164] [180/312] eta: 0:01:38 lr: 0.002984 min_lr: 0.002984 loss: 2.3431 (2.1339) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [164] [190/312] eta: 0:01:30 lr: 0.002984 min_lr: 0.002984 loss: 2.3649 (2.1428) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [164] [200/312] eta: 0:01:22 lr: 0.002984 min_lr: 0.002984 loss: 2.2551 (2.1461) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [164] [210/312] eta: 0:01:15 lr: 0.002983 min_lr: 0.002983 loss: 2.2551 (2.1405) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [164] [220/312] eta: 0:01:07 lr: 0.002983 min_lr: 0.002983 loss: 2.2937 (2.1477) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [164] [230/312] eta: 0:01:00 lr: 0.002982 min_lr: 0.002982 loss: 2.2974 (2.1504) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [164] [240/312] eta: 0:00:52 lr: 0.002982 min_lr: 0.002982 loss: 2.1787 (2.1510) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [164] [250/312] eta: 0:00:45 lr: 0.002982 min_lr: 0.002982 loss: 2.3290 (2.1602) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [164] [260/312] eta: 0:00:37 lr: 0.002981 min_lr: 0.002981 loss: 2.4073 (2.1677) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [164] [270/312] eta: 0:00:30 lr: 0.002981 min_lr: 0.002981 loss: 2.2803 (2.1685) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [164] [280/312] eta: 0:00:23 lr: 0.002980 min_lr: 0.002980 loss: 2.2231 (2.1721) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [164] [290/312] eta: 0:00:15 lr: 0.002980 min_lr: 0.002980 loss: 2.2231 (2.1721) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [164] [300/312] eta: 0:00:08 lr: 0.002980 min_lr: 0.002980 loss: 2.2374 (2.1762) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [164] [310/312] eta: 0:00:01 lr: 0.002979 min_lr: 0.002979 loss: 2.2268 (2.1717) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [164] [311/312] eta: 0:00:00 lr: 0.002979 min_lr: 0.002979 loss: 2.2268 (2.1696) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [164] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.002979 min_lr: 0.002979 loss: 2.2268 (2.1982) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6118 (0.6118) acc1: 83.5938 (83.5938) acc5: 95.5729 (95.5729) time: 4.5471 data: 4.3312 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9838 (0.9095) acc1: 75.2604 (76.3520) acc5: 93.4896 (93.7280) time: 0.6565 data: 0.4813 max mem: 64948 Test: Total time: 0:00:06 (0.6743 s / it) * Acc@1 77.370 Acc@5 93.788 loss 0.880 Accuracy of the model on the 50000 test images: 77.4% Max accuracy: 77.63% Test: [0/9] eta: 0:00:44 loss: 0.5885 (0.5885) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.8900 data: 4.6636 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7602 (0.7538) acc1: 80.4688 (79.2960) acc5: 95.8333 (95.2000) time: 0.6947 data: 0.5183 max mem: 64948 Test: Total time: 0:00:06 (0.7024 s / it) * Acc@1 80.274 Acc@5 95.246 loss 0.732 Accuracy of the model EMA on 50000 test images: 80.3% Max EMA accuracy: 80.27% Epoch: [165] [ 0/312] eta: 0:50:28 lr: 0.002979 min_lr: 0.002979 loss: 1.8937 (1.8937) weight_decay: 0.0500 (0.0500) time: 9.7082 data: 8.9578 max mem: 64948 Epoch: [165] [ 10/312] eta: 0:07:53 lr: 0.002979 min_lr: 0.002979 loss: 1.9148 (1.9867) weight_decay: 0.0500 (0.0500) time: 1.5669 data: 0.8147 max mem: 64948 Epoch: [165] [ 20/312] eta: 0:05:36 lr: 0.002978 min_lr: 0.002978 loss: 2.2062 (2.1286) weight_decay: 0.0500 (0.0500) time: 0.7234 data: 0.0004 max mem: 64948 Epoch: [165] [ 30/312] eta: 0:04:43 lr: 0.002978 min_lr: 0.002978 loss: 2.2910 (2.1488) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [165] [ 40/312] eta: 0:04:13 lr: 0.002977 min_lr: 0.002977 loss: 2.2358 (2.1697) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [165] [ 50/312] eta: 0:03:51 lr: 0.002977 min_lr: 0.002977 loss: 2.3342 (2.2126) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [165] [ 60/312] eta: 0:03:35 lr: 0.002977 min_lr: 0.002977 loss: 2.3458 (2.2323) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [165] [ 70/312] eta: 0:03:21 lr: 0.002976 min_lr: 0.002976 loss: 2.0794 (2.2032) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [165] [ 80/312] eta: 0:03:08 lr: 0.002976 min_lr: 0.002976 loss: 2.2537 (2.2164) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [165] [ 90/312] eta: 0:02:57 lr: 0.002975 min_lr: 0.002975 loss: 2.2537 (2.1929) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [165] [100/312] eta: 0:02:47 lr: 0.002975 min_lr: 0.002975 loss: 2.0934 (2.1943) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [165] [110/312] eta: 0:02:37 lr: 0.002975 min_lr: 0.002975 loss: 2.2977 (2.1953) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [165] [120/312] eta: 0:02:28 lr: 0.002974 min_lr: 0.002974 loss: 2.3164 (2.2094) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [165] [130/312] eta: 0:02:19 lr: 0.002974 min_lr: 0.002974 loss: 2.3747 (2.2184) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [165] [140/312] eta: 0:02:11 lr: 0.002973 min_lr: 0.002973 loss: 2.3747 (2.2259) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [165] [150/312] eta: 0:02:02 lr: 0.002973 min_lr: 0.002973 loss: 2.1719 (2.2145) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [165] [160/312] eta: 0:01:54 lr: 0.002973 min_lr: 0.002973 loss: 2.3515 (2.2329) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [165] [170/312] eta: 0:01:46 lr: 0.002972 min_lr: 0.002972 loss: 2.4010 (2.2282) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [165] [180/312] eta: 0:01:38 lr: 0.002972 min_lr: 0.002972 loss: 1.9943 (2.2174) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [165] [190/312] eta: 0:01:30 lr: 0.002971 min_lr: 0.002971 loss: 2.1889 (2.2213) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [165] [200/312] eta: 0:01:23 lr: 0.002971 min_lr: 0.002971 loss: 2.3096 (2.2203) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [165] [210/312] eta: 0:01:15 lr: 0.002971 min_lr: 0.002971 loss: 2.1846 (2.2198) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [165] [220/312] eta: 0:01:07 lr: 0.002970 min_lr: 0.002970 loss: 2.0953 (2.2118) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [165] [230/312] eta: 0:01:00 lr: 0.002970 min_lr: 0.002970 loss: 2.0702 (2.2121) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [165] [240/312] eta: 0:00:52 lr: 0.002969 min_lr: 0.002969 loss: 2.0966 (2.2097) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [165] [250/312] eta: 0:00:45 lr: 0.002969 min_lr: 0.002969 loss: 2.0047 (2.2037) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [165] [260/312] eta: 0:00:38 lr: 0.002968 min_lr: 0.002968 loss: 2.2240 (2.2092) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [165] [270/312] eta: 0:00:30 lr: 0.002968 min_lr: 0.002968 loss: 2.3687 (2.2114) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [165] [280/312] eta: 0:00:23 lr: 0.002968 min_lr: 0.002968 loss: 2.2921 (2.2089) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [165] [290/312] eta: 0:00:16 lr: 0.002967 min_lr: 0.002967 loss: 1.8678 (2.1991) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [165] [300/312] eta: 0:00:08 lr: 0.002967 min_lr: 0.002967 loss: 1.9696 (2.1993) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [165] [310/312] eta: 0:00:01 lr: 0.002966 min_lr: 0.002966 loss: 2.2852 (2.2021) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [165] [311/312] eta: 0:00:00 lr: 0.002966 min_lr: 0.002966 loss: 2.2478 (2.1995) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [165] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.002966 min_lr: 0.002966 loss: 2.2478 (2.1894) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6962 (0.6962) acc1: 81.7708 (81.7708) acc5: 93.7500 (93.7500) time: 4.6543 data: 4.4427 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9690 (0.8940) acc1: 77.6042 (76.7360) acc5: 93.7500 (93.3440) time: 0.6684 data: 0.4937 max mem: 64948 Test: Total time: 0:00:06 (0.6799 s / it) * Acc@1 77.328 Acc@5 93.594 loss 0.878 Accuracy of the model on the 50000 test images: 77.3% Max accuracy: 77.63% Test: [0/9] eta: 0:00:39 loss: 0.5878 (0.5878) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.3395 data: 4.1338 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7597 (0.7528) acc1: 80.7292 (79.3600) acc5: 95.8333 (95.2640) time: 0.6359 data: 0.4618 max mem: 64948 Test: Total time: 0:00:05 (0.6453 s / it) * Acc@1 80.288 Acc@5 95.274 loss 0.731 Accuracy of the model EMA on 50000 test images: 80.3% Max EMA accuracy: 80.29% Epoch: [166] [ 0/312] eta: 0:48:39 lr: 0.002966 min_lr: 0.002966 loss: 2.2100 (2.2100) weight_decay: 0.0500 (0.0500) time: 9.3575 data: 7.7826 max mem: 64948 Epoch: [166] [ 10/312] eta: 0:07:44 lr: 0.002966 min_lr: 0.002966 loss: 2.2739 (2.1881) weight_decay: 0.0500 (0.0500) time: 1.5383 data: 0.7079 max mem: 64948 Epoch: [166] [ 20/312] eta: 0:05:31 lr: 0.002966 min_lr: 0.002966 loss: 2.2913 (2.2342) weight_decay: 0.0500 (0.0500) time: 0.7256 data: 0.0004 max mem: 64948 Epoch: [166] [ 30/312] eta: 0:04:40 lr: 0.002965 min_lr: 0.002965 loss: 2.2913 (2.2108) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [166] [ 40/312] eta: 0:04:10 lr: 0.002965 min_lr: 0.002965 loss: 2.0761 (2.1829) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [166] [ 50/312] eta: 0:03:49 lr: 0.002964 min_lr: 0.002964 loss: 2.2337 (2.1779) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [166] [ 60/312] eta: 0:03:33 lr: 0.002964 min_lr: 0.002964 loss: 2.2435 (2.1732) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [166] [ 70/312] eta: 0:03:19 lr: 0.002963 min_lr: 0.002963 loss: 2.1601 (2.1672) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [166] [ 80/312] eta: 0:03:07 lr: 0.002963 min_lr: 0.002963 loss: 2.1601 (2.1733) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [166] [ 90/312] eta: 0:02:56 lr: 0.002963 min_lr: 0.002963 loss: 2.1274 (2.1677) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [166] [100/312] eta: 0:02:46 lr: 0.002962 min_lr: 0.002962 loss: 2.4222 (2.2029) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [166] [110/312] eta: 0:02:37 lr: 0.002962 min_lr: 0.002962 loss: 2.3698 (2.2073) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [166] [120/312] eta: 0:02:28 lr: 0.002961 min_lr: 0.002961 loss: 2.3596 (2.2170) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [166] [130/312] eta: 0:02:19 lr: 0.002961 min_lr: 0.002961 loss: 2.3972 (2.2176) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [166] [140/312] eta: 0:02:10 lr: 0.002961 min_lr: 0.002961 loss: 2.3477 (2.2288) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [166] [150/312] eta: 0:02:02 lr: 0.002960 min_lr: 0.002960 loss: 2.2293 (2.2267) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [166] [160/312] eta: 0:01:54 lr: 0.002960 min_lr: 0.002960 loss: 2.1920 (2.2266) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [166] [170/312] eta: 0:01:46 lr: 0.002959 min_lr: 0.002959 loss: 2.2789 (2.2286) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [166] [180/312] eta: 0:01:38 lr: 0.002959 min_lr: 0.002959 loss: 2.2789 (2.2276) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [166] [190/312] eta: 0:01:30 lr: 0.002959 min_lr: 0.002959 loss: 2.2447 (2.2248) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [166] [200/312] eta: 0:01:23 lr: 0.002958 min_lr: 0.002958 loss: 2.0538 (2.2230) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [166] [210/312] eta: 0:01:15 lr: 0.002958 min_lr: 0.002958 loss: 2.0701 (2.2236) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [166] [220/312] eta: 0:01:07 lr: 0.002957 min_lr: 0.002957 loss: 2.1811 (2.2213) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [166] [230/312] eta: 0:01:00 lr: 0.002957 min_lr: 0.002957 loss: 2.3051 (2.2210) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [166] [240/312] eta: 0:00:52 lr: 0.002956 min_lr: 0.002956 loss: 2.2283 (2.2134) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [166] [250/312] eta: 0:00:45 lr: 0.002956 min_lr: 0.002956 loss: 2.2357 (2.2188) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [166] [260/312] eta: 0:00:37 lr: 0.002956 min_lr: 0.002956 loss: 2.2653 (2.2091) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [166] [270/312] eta: 0:00:30 lr: 0.002955 min_lr: 0.002955 loss: 2.1292 (2.2098) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [166] [280/312] eta: 0:00:23 lr: 0.002955 min_lr: 0.002955 loss: 2.1771 (2.2071) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [166] [290/312] eta: 0:00:15 lr: 0.002954 min_lr: 0.002954 loss: 2.3453 (2.2104) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [166] [300/312] eta: 0:00:08 lr: 0.002954 min_lr: 0.002954 loss: 2.3749 (2.2085) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [166] [310/312] eta: 0:00:01 lr: 0.002954 min_lr: 0.002954 loss: 2.2450 (2.2090) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [166] [311/312] eta: 0:00:00 lr: 0.002954 min_lr: 0.002954 loss: 2.2450 (2.2075) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [166] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.002954 min_lr: 0.002954 loss: 2.2450 (2.1883) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6872 (0.6872) acc1: 83.3333 (83.3333) acc5: 94.5312 (94.5312) time: 4.7161 data: 4.5041 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8377 (0.8926) acc1: 77.6042 (77.0240) acc5: 93.7500 (93.7600) time: 0.6754 data: 0.5005 max mem: 64948 Test: Total time: 0:00:06 (0.6965 s / it) * Acc@1 77.300 Acc@5 93.778 loss 0.881 Accuracy of the model on the 50000 test images: 77.3% Max accuracy: 77.63% Test: [0/9] eta: 0:00:42 loss: 0.5869 (0.5869) acc1: 84.3750 (84.3750) acc5: 96.6146 (96.6146) time: 4.7027 data: 4.4896 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7589 (0.7519) acc1: 80.7292 (79.3600) acc5: 95.8333 (95.2320) time: 0.6747 data: 0.4989 max mem: 64948 Test: Total time: 0:00:06 (0.6863 s / it) * Acc@1 80.304 Acc@5 95.300 loss 0.730 Accuracy of the model EMA on 50000 test images: 80.3% Max EMA accuracy: 80.30% Epoch: [167] [ 0/312] eta: 0:48:54 lr: 0.002954 min_lr: 0.002954 loss: 2.1602 (2.1602) weight_decay: 0.0500 (0.0500) time: 9.4067 data: 8.5955 max mem: 64948 Epoch: [167] [ 10/312] eta: 0:07:36 lr: 0.002953 min_lr: 0.002953 loss: 2.1865 (2.2577) weight_decay: 0.0500 (0.0500) time: 1.5109 data: 0.7817 max mem: 64948 Epoch: [167] [ 20/312] eta: 0:05:27 lr: 0.002953 min_lr: 0.002953 loss: 2.3124 (2.3054) weight_decay: 0.0500 (0.0500) time: 0.7075 data: 0.0003 max mem: 64948 Epoch: [167] [ 30/312] eta: 0:04:37 lr: 0.002952 min_lr: 0.002952 loss: 2.3974 (2.3423) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [167] [ 40/312] eta: 0:04:08 lr: 0.002952 min_lr: 0.002952 loss: 2.3438 (2.3283) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [167] [ 50/312] eta: 0:03:48 lr: 0.002951 min_lr: 0.002951 loss: 2.2823 (2.3163) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [167] [ 60/312] eta: 0:03:32 lr: 0.002951 min_lr: 0.002951 loss: 2.2823 (2.3251) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [167] [ 70/312] eta: 0:03:19 lr: 0.002951 min_lr: 0.002951 loss: 2.2817 (2.3083) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [167] [ 80/312] eta: 0:03:07 lr: 0.002950 min_lr: 0.002950 loss: 2.1224 (2.2780) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [167] [ 90/312] eta: 0:02:56 lr: 0.002950 min_lr: 0.002950 loss: 2.1611 (2.2606) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [167] [100/312] eta: 0:02:46 lr: 0.002949 min_lr: 0.002949 loss: 2.1667 (2.2531) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [167] [110/312] eta: 0:02:37 lr: 0.002949 min_lr: 0.002949 loss: 2.2711 (2.2487) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [167] [120/312] eta: 0:02:27 lr: 0.002949 min_lr: 0.002949 loss: 2.3322 (2.2527) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [167] [130/312] eta: 0:02:19 lr: 0.002948 min_lr: 0.002948 loss: 2.3445 (2.2592) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [167] [140/312] eta: 0:02:10 lr: 0.002948 min_lr: 0.002948 loss: 2.3873 (2.2605) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [167] [150/312] eta: 0:02:02 lr: 0.002947 min_lr: 0.002947 loss: 2.3873 (2.2704) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [167] [160/312] eta: 0:01:54 lr: 0.002947 min_lr: 0.002947 loss: 2.2772 (2.2542) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [167] [170/312] eta: 0:01:46 lr: 0.002947 min_lr: 0.002947 loss: 2.2289 (2.2498) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [167] [180/312] eta: 0:01:38 lr: 0.002946 min_lr: 0.002946 loss: 2.2289 (2.2477) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [167] [190/312] eta: 0:01:30 lr: 0.002946 min_lr: 0.002946 loss: 2.0934 (2.2398) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [167] [200/312] eta: 0:01:22 lr: 0.002945 min_lr: 0.002945 loss: 2.1802 (2.2391) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [167] [210/312] eta: 0:01:15 lr: 0.002945 min_lr: 0.002945 loss: 2.3013 (2.2395) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [167] [220/312] eta: 0:01:07 lr: 0.002944 min_lr: 0.002944 loss: 2.2546 (2.2395) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [167] [230/312] eta: 0:01:00 lr: 0.002944 min_lr: 0.002944 loss: 2.2677 (2.2428) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [167] [240/312] eta: 0:00:52 lr: 0.002944 min_lr: 0.002944 loss: 2.2353 (2.2313) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [167] [250/312] eta: 0:00:45 lr: 0.002943 min_lr: 0.002943 loss: 1.8752 (2.2230) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [167] [260/312] eta: 0:00:37 lr: 0.002943 min_lr: 0.002943 loss: 2.2408 (2.2243) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [167] [270/312] eta: 0:00:30 lr: 0.002942 min_lr: 0.002942 loss: 2.2790 (2.2269) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [167] [280/312] eta: 0:00:23 lr: 0.002942 min_lr: 0.002942 loss: 2.2012 (2.2317) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [167] [290/312] eta: 0:00:15 lr: 0.002942 min_lr: 0.002942 loss: 2.3798 (2.2330) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [167] [300/312] eta: 0:00:08 lr: 0.002941 min_lr: 0.002941 loss: 2.2496 (2.2302) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [167] [310/312] eta: 0:00:01 lr: 0.002941 min_lr: 0.002941 loss: 2.2455 (2.2361) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [167] [311/312] eta: 0:00:00 lr: 0.002941 min_lr: 0.002941 loss: 2.2455 (2.2361) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [167] Total time: 0:03:46 (0.7267 s / it) Averaged stats: lr: 0.002941 min_lr: 0.002941 loss: 2.2455 (2.1936) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6618 (0.6618) acc1: 83.8542 (83.8542) acc5: 96.0938 (96.0938) time: 4.5926 data: 4.3767 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8971 (0.9091) acc1: 76.8229 (76.3520) acc5: 93.7500 (93.5360) time: 0.6615 data: 0.4864 max mem: 64948 Test: Total time: 0:00:06 (0.6849 s / it) * Acc@1 77.222 Acc@5 93.798 loss 0.881 Accuracy of the model on the 50000 test images: 77.2% Max accuracy: 77.63% Test: [0/9] eta: 0:00:43 loss: 0.5858 (0.5858) acc1: 84.6354 (84.6354) acc5: 96.6146 (96.6146) time: 4.7899 data: 4.5840 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7580 (0.7506) acc1: 80.7292 (79.4880) acc5: 95.8333 (95.2640) time: 0.6835 data: 0.5094 max mem: 64948 Test: Total time: 0:00:06 (0.6943 s / it) * Acc@1 80.358 Acc@5 95.298 loss 0.729 Accuracy of the model EMA on 50000 test images: 80.4% Max EMA accuracy: 80.36% Epoch: [168] [ 0/312] eta: 0:52:20 lr: 0.002941 min_lr: 0.002941 loss: 2.3454 (2.3454) weight_decay: 0.0500 (0.0500) time: 10.0644 data: 9.2874 max mem: 64948 Epoch: [168] [ 10/312] eta: 0:07:53 lr: 0.002940 min_lr: 0.002940 loss: 2.2997 (2.1999) weight_decay: 0.0500 (0.0500) time: 1.5668 data: 0.8446 max mem: 64948 Epoch: [168] [ 20/312] eta: 0:05:36 lr: 0.002940 min_lr: 0.002940 loss: 2.2191 (2.1427) weight_decay: 0.0500 (0.0500) time: 0.7057 data: 0.0004 max mem: 64948 Epoch: [168] [ 30/312] eta: 0:04:43 lr: 0.002939 min_lr: 0.002939 loss: 2.1933 (2.1615) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [168] [ 40/312] eta: 0:04:12 lr: 0.002939 min_lr: 0.002939 loss: 2.2406 (2.1694) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [168] [ 50/312] eta: 0:03:51 lr: 0.002939 min_lr: 0.002939 loss: 2.3334 (2.2047) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [168] [ 60/312] eta: 0:03:34 lr: 0.002938 min_lr: 0.002938 loss: 2.3334 (2.1991) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [168] [ 70/312] eta: 0:03:21 lr: 0.002938 min_lr: 0.002938 loss: 2.1575 (2.1730) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [168] [ 80/312] eta: 0:03:08 lr: 0.002937 min_lr: 0.002937 loss: 2.1450 (2.1723) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [168] [ 90/312] eta: 0:02:57 lr: 0.002937 min_lr: 0.002937 loss: 2.0760 (2.1625) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [168] [100/312] eta: 0:02:47 lr: 0.002937 min_lr: 0.002937 loss: 2.2194 (2.1657) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [168] [110/312] eta: 0:02:38 lr: 0.002936 min_lr: 0.002936 loss: 2.2377 (2.1639) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [168] [120/312] eta: 0:02:28 lr: 0.002936 min_lr: 0.002936 loss: 2.3178 (2.1716) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0004 max mem: 64948 Epoch: [168] [130/312] eta: 0:02:20 lr: 0.002935 min_lr: 0.002935 loss: 2.3306 (2.1872) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [168] [140/312] eta: 0:02:11 lr: 0.002935 min_lr: 0.002935 loss: 2.3642 (2.1974) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [168] [150/312] eta: 0:02:03 lr: 0.002934 min_lr: 0.002934 loss: 2.3913 (2.1999) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [168] [160/312] eta: 0:01:54 lr: 0.002934 min_lr: 0.002934 loss: 2.2972 (2.2089) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [168] [170/312] eta: 0:01:46 lr: 0.002934 min_lr: 0.002934 loss: 2.2955 (2.2130) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [168] [180/312] eta: 0:01:38 lr: 0.002933 min_lr: 0.002933 loss: 2.2955 (2.2091) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [168] [190/312] eta: 0:01:31 lr: 0.002933 min_lr: 0.002933 loss: 2.2613 (2.2099) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [168] [200/312] eta: 0:01:23 lr: 0.002932 min_lr: 0.002932 loss: 2.2805 (2.2176) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [168] [210/312] eta: 0:01:15 lr: 0.002932 min_lr: 0.002932 loss: 2.2099 (2.2099) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [168] [220/312] eta: 0:01:07 lr: 0.002932 min_lr: 0.002932 loss: 2.2511 (2.2130) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [168] [230/312] eta: 0:01:00 lr: 0.002931 min_lr: 0.002931 loss: 2.1064 (2.2075) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [168] [240/312] eta: 0:00:52 lr: 0.002931 min_lr: 0.002931 loss: 2.0727 (2.2008) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [168] [250/312] eta: 0:00:45 lr: 0.002930 min_lr: 0.002930 loss: 1.9042 (2.1915) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [168] [260/312] eta: 0:00:38 lr: 0.002930 min_lr: 0.002930 loss: 1.9943 (2.1841) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [168] [270/312] eta: 0:00:30 lr: 0.002929 min_lr: 0.002929 loss: 2.0810 (2.1835) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [168] [280/312] eta: 0:00:23 lr: 0.002929 min_lr: 0.002929 loss: 2.2655 (2.1910) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0006 max mem: 64948 Epoch: [168] [290/312] eta: 0:00:16 lr: 0.002929 min_lr: 0.002929 loss: 2.2655 (2.1886) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0005 max mem: 64948 Epoch: [168] [300/312] eta: 0:00:08 lr: 0.002928 min_lr: 0.002928 loss: 2.2558 (2.1909) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [168] [310/312] eta: 0:00:01 lr: 0.002928 min_lr: 0.002928 loss: 2.3145 (2.1932) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [168] [311/312] eta: 0:00:00 lr: 0.002928 min_lr: 0.002928 loss: 2.3145 (2.1930) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [168] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.002928 min_lr: 0.002928 loss: 2.3145 (2.1779) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.7292 (0.7292) acc1: 83.0729 (83.0729) acc5: 93.7500 (93.7500) time: 4.7505 data: 4.5366 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9265 (0.8972) acc1: 78.1250 (76.8000) acc5: 94.0104 (94.1760) time: 0.6791 data: 0.5041 max mem: 64948 Test: Total time: 0:00:06 (0.7017 s / it) * Acc@1 77.470 Acc@5 93.710 loss 0.896 Accuracy of the model on the 50000 test images: 77.5% Max accuracy: 77.63% Test: [0/9] eta: 0:00:43 loss: 0.5851 (0.5851) acc1: 84.8958 (84.8958) acc5: 96.8750 (96.8750) time: 4.7873 data: 4.5802 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7575 (0.7497) acc1: 80.7292 (79.4240) acc5: 95.8333 (95.2960) time: 0.6833 data: 0.5090 max mem: 64948 Test: Total time: 0:00:06 (0.6971 s / it) * Acc@1 80.378 Acc@5 95.312 loss 0.727 Accuracy of the model EMA on 50000 test images: 80.4% Max EMA accuracy: 80.38% Epoch: [169] [ 0/312] eta: 0:49:23 lr: 0.002928 min_lr: 0.002928 loss: 2.6089 (2.6089) weight_decay: 0.0500 (0.0500) time: 9.4995 data: 8.5359 max mem: 64948 Epoch: [169] [ 10/312] eta: 0:07:38 lr: 0.002927 min_lr: 0.002927 loss: 2.2055 (2.1477) weight_decay: 0.0500 (0.0500) time: 1.5198 data: 0.7764 max mem: 64948 Epoch: [169] [ 20/312] eta: 0:05:29 lr: 0.002927 min_lr: 0.002927 loss: 2.2055 (2.1505) weight_decay: 0.0500 (0.0500) time: 0.7105 data: 0.0004 max mem: 64948 Epoch: [169] [ 30/312] eta: 0:04:38 lr: 0.002926 min_lr: 0.002926 loss: 2.3325 (2.2196) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [169] [ 40/312] eta: 0:04:09 lr: 0.002926 min_lr: 0.002926 loss: 2.2096 (2.1977) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [169] [ 50/312] eta: 0:03:49 lr: 0.002926 min_lr: 0.002926 loss: 2.0629 (2.1441) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [169] [ 60/312] eta: 0:03:32 lr: 0.002925 min_lr: 0.002925 loss: 2.0045 (2.1400) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [169] [ 70/312] eta: 0:03:19 lr: 0.002925 min_lr: 0.002925 loss: 2.2021 (2.1628) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [169] [ 80/312] eta: 0:03:07 lr: 0.002924 min_lr: 0.002924 loss: 2.2021 (2.1559) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [169] [ 90/312] eta: 0:02:56 lr: 0.002924 min_lr: 0.002924 loss: 2.0875 (2.1459) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [169] [100/312] eta: 0:02:46 lr: 0.002924 min_lr: 0.002924 loss: 2.0770 (2.1514) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [169] [110/312] eta: 0:02:36 lr: 0.002923 min_lr: 0.002923 loss: 2.0511 (2.1475) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [169] [120/312] eta: 0:02:27 lr: 0.002923 min_lr: 0.002923 loss: 2.0301 (2.1337) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [169] [130/312] eta: 0:02:19 lr: 0.002922 min_lr: 0.002922 loss: 2.2500 (2.1502) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [169] [140/312] eta: 0:02:10 lr: 0.002922 min_lr: 0.002922 loss: 2.3310 (2.1549) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [169] [150/312] eta: 0:02:02 lr: 0.002921 min_lr: 0.002921 loss: 2.3143 (2.1602) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [169] [160/312] eta: 0:01:54 lr: 0.002921 min_lr: 0.002921 loss: 2.3383 (2.1653) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [169] [170/312] eta: 0:01:46 lr: 0.002921 min_lr: 0.002921 loss: 2.1806 (2.1565) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [169] [180/312] eta: 0:01:38 lr: 0.002920 min_lr: 0.002920 loss: 2.0094 (2.1460) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [169] [190/312] eta: 0:01:30 lr: 0.002920 min_lr: 0.002920 loss: 2.0677 (2.1511) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [169] [200/312] eta: 0:01:22 lr: 0.002919 min_lr: 0.002919 loss: 2.3521 (2.1574) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [169] [210/312] eta: 0:01:15 lr: 0.002919 min_lr: 0.002919 loss: 2.3197 (2.1591) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [169] [220/312] eta: 0:01:07 lr: 0.002919 min_lr: 0.002919 loss: 2.2191 (2.1645) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [169] [230/312] eta: 0:01:00 lr: 0.002918 min_lr: 0.002918 loss: 2.1253 (2.1561) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [169] [240/312] eta: 0:00:52 lr: 0.002918 min_lr: 0.002918 loss: 2.0643 (2.1554) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [169] [250/312] eta: 0:00:45 lr: 0.002917 min_lr: 0.002917 loss: 2.2642 (2.1561) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [169] [260/312] eta: 0:00:37 lr: 0.002917 min_lr: 0.002917 loss: 2.2898 (2.1589) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [169] [270/312] eta: 0:00:30 lr: 0.002917 min_lr: 0.002917 loss: 2.3101 (2.1616) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [169] [280/312] eta: 0:00:23 lr: 0.002916 min_lr: 0.002916 loss: 2.2610 (2.1638) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [169] [290/312] eta: 0:00:15 lr: 0.002916 min_lr: 0.002916 loss: 2.2670 (2.1648) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [169] [300/312] eta: 0:00:08 lr: 0.002915 min_lr: 0.002915 loss: 2.2585 (2.1677) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [169] [310/312] eta: 0:00:01 lr: 0.002915 min_lr: 0.002915 loss: 2.2111 (2.1723) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [169] [311/312] eta: 0:00:00 lr: 0.002915 min_lr: 0.002915 loss: 2.2585 (2.1733) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [169] Total time: 0:03:46 (0.7264 s / it) Averaged stats: lr: 0.002915 min_lr: 0.002915 loss: 2.2585 (2.1859) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6488 (0.6488) acc1: 83.0729 (83.0729) acc5: 95.8333 (95.8333) time: 4.5917 data: 4.3821 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9111 (0.8941) acc1: 78.1250 (76.3520) acc5: 94.2708 (93.6640) time: 0.6614 data: 0.4870 max mem: 64948 Test: Total time: 0:00:06 (0.6845 s / it) * Acc@1 77.414 Acc@5 93.940 loss 0.873 Accuracy of the model on the 50000 test images: 77.4% Max accuracy: 77.63% Test: [0/9] eta: 0:00:47 loss: 0.5834 (0.5834) acc1: 84.8958 (84.8958) acc5: 96.8750 (96.8750) time: 5.3031 data: 5.0912 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7564 (0.7483) acc1: 80.4688 (79.3600) acc5: 95.8333 (95.3280) time: 0.7406 data: 0.5658 max mem: 64948 Test: Total time: 0:00:06 (0.7499 s / it) * Acc@1 80.420 Acc@5 95.332 loss 0.726 Accuracy of the model EMA on 50000 test images: 80.4% Max EMA accuracy: 80.42% Epoch: [170] [ 0/312] eta: 0:59:53 lr: 0.002915 min_lr: 0.002915 loss: 1.5643 (1.5643) weight_decay: 0.0500 (0.0500) time: 11.5191 data: 10.7878 max mem: 64948 Epoch: [170] [ 10/312] eta: 0:08:28 lr: 0.002914 min_lr: 0.002914 loss: 2.2391 (2.0907) weight_decay: 0.0500 (0.0500) time: 1.6851 data: 0.9810 max mem: 64948 Epoch: [170] [ 20/312] eta: 0:05:54 lr: 0.002914 min_lr: 0.002914 loss: 2.2964 (2.1519) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [170] [ 30/312] eta: 0:04:54 lr: 0.002914 min_lr: 0.002914 loss: 2.1522 (2.1171) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [170] [ 40/312] eta: 0:04:21 lr: 0.002913 min_lr: 0.002913 loss: 2.1467 (2.1316) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [170] [ 50/312] eta: 0:03:58 lr: 0.002913 min_lr: 0.002913 loss: 2.1467 (2.1280) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [170] [ 60/312] eta: 0:03:40 lr: 0.002912 min_lr: 0.002912 loss: 2.3723 (2.1515) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [170] [ 70/312] eta: 0:03:25 lr: 0.002912 min_lr: 0.002912 loss: 2.3166 (2.1447) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [170] [ 80/312] eta: 0:03:13 lr: 0.002911 min_lr: 0.002911 loss: 2.2788 (2.1714) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [170] [ 90/312] eta: 0:03:01 lr: 0.002911 min_lr: 0.002911 loss: 2.3069 (2.1649) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [170] [100/312] eta: 0:02:50 lr: 0.002911 min_lr: 0.002911 loss: 2.1019 (2.1714) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [170] [110/312] eta: 0:02:40 lr: 0.002910 min_lr: 0.002910 loss: 2.1331 (2.1748) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [170] [120/312] eta: 0:02:30 lr: 0.002910 min_lr: 0.002910 loss: 2.2613 (2.1873) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [170] [130/312] eta: 0:02:21 lr: 0.002909 min_lr: 0.002909 loss: 2.2877 (2.1793) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [170] [140/312] eta: 0:02:13 lr: 0.002909 min_lr: 0.002909 loss: 2.0088 (2.1694) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [170] [150/312] eta: 0:02:04 lr: 0.002909 min_lr: 0.002909 loss: 2.2449 (2.1739) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0003 max mem: 64948 Epoch: [170] [160/312] eta: 0:01:56 lr: 0.002908 min_lr: 0.002908 loss: 2.2339 (2.1658) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [170] [170/312] eta: 0:01:47 lr: 0.002908 min_lr: 0.002908 loss: 2.0342 (2.1597) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [170] [180/312] eta: 0:01:39 lr: 0.002907 min_lr: 0.002907 loss: 2.2330 (2.1604) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [170] [190/312] eta: 0:01:31 lr: 0.002907 min_lr: 0.002907 loss: 2.3228 (2.1643) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0005 max mem: 64948 Epoch: [170] [200/312] eta: 0:01:24 lr: 0.002906 min_lr: 0.002906 loss: 2.1717 (2.1594) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0005 max mem: 64948 Epoch: [170] [210/312] eta: 0:01:16 lr: 0.002906 min_lr: 0.002906 loss: 2.1141 (2.1599) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [170] [220/312] eta: 0:01:08 lr: 0.002906 min_lr: 0.002906 loss: 2.3199 (2.1611) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [170] [230/312] eta: 0:01:00 lr: 0.002905 min_lr: 0.002905 loss: 2.1025 (2.1543) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [170] [240/312] eta: 0:00:53 lr: 0.002905 min_lr: 0.002905 loss: 2.2454 (2.1625) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [170] [250/312] eta: 0:00:45 lr: 0.002904 min_lr: 0.002904 loss: 2.3028 (2.1612) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [170] [260/312] eta: 0:00:38 lr: 0.002904 min_lr: 0.002904 loss: 2.1238 (2.1605) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [170] [270/312] eta: 0:00:30 lr: 0.002903 min_lr: 0.002903 loss: 2.1238 (2.1608) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [170] [280/312] eta: 0:00:23 lr: 0.002903 min_lr: 0.002903 loss: 2.2438 (2.1580) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0009 max mem: 64948 Epoch: [170] [290/312] eta: 0:00:16 lr: 0.002903 min_lr: 0.002903 loss: 2.2822 (2.1597) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0008 max mem: 64948 Epoch: [170] [300/312] eta: 0:00:08 lr: 0.002902 min_lr: 0.002902 loss: 2.2537 (2.1617) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [170] [310/312] eta: 0:00:01 lr: 0.002902 min_lr: 0.002902 loss: 2.1415 (2.1614) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [170] [311/312] eta: 0:00:00 lr: 0.002902 min_lr: 0.002902 loss: 2.1415 (2.1596) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [170] Total time: 0:03:48 (0.7332 s / it) Averaged stats: lr: 0.002902 min_lr: 0.002902 loss: 2.1415 (2.1861) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7090 (0.7090) acc1: 81.7708 (81.7708) acc5: 94.5312 (94.5312) time: 4.6127 data: 4.4042 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9790 (0.8852) acc1: 77.3438 (76.5440) acc5: 94.0104 (93.7600) time: 0.6638 data: 0.4895 max mem: 64948 Test: Total time: 0:00:06 (0.6875 s / it) * Acc@1 77.636 Acc@5 93.838 loss 0.868 Accuracy of the model on the 50000 test images: 77.6% Max accuracy: 77.64% Test: [0/9] eta: 0:00:41 loss: 0.5818 (0.5818) acc1: 84.8958 (84.8958) acc5: 96.8750 (96.8750) time: 4.6562 data: 4.4384 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7559 (0.7471) acc1: 80.4688 (79.4560) acc5: 95.8333 (95.3280) time: 0.6693 data: 0.4933 max mem: 64948 Test: Total time: 0:00:06 (0.6790 s / it) * Acc@1 80.438 Acc@5 95.340 loss 0.725 Accuracy of the model EMA on 50000 test images: 80.4% Max EMA accuracy: 80.44% Epoch: [171] [ 0/312] eta: 0:46:59 lr: 0.002902 min_lr: 0.002902 loss: 2.4404 (2.4404) weight_decay: 0.0500 (0.0500) time: 9.0361 data: 7.6592 max mem: 64948 Epoch: [171] [ 10/312] eta: 0:07:31 lr: 0.002901 min_lr: 0.002901 loss: 2.4282 (2.3122) weight_decay: 0.0500 (0.0500) time: 1.4944 data: 0.6967 max mem: 64948 Epoch: [171] [ 20/312] eta: 0:05:25 lr: 0.002901 min_lr: 0.002901 loss: 2.3454 (2.2815) weight_decay: 0.0500 (0.0500) time: 0.7179 data: 0.0004 max mem: 64948 Epoch: [171] [ 30/312] eta: 0:04:35 lr: 0.002900 min_lr: 0.002900 loss: 2.0857 (2.1711) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [171] [ 40/312] eta: 0:04:07 lr: 0.002900 min_lr: 0.002900 loss: 2.0282 (2.1426) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [171] [ 50/312] eta: 0:03:47 lr: 0.002900 min_lr: 0.002900 loss: 2.1224 (2.1376) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [171] [ 60/312] eta: 0:03:31 lr: 0.002899 min_lr: 0.002899 loss: 2.2482 (2.1532) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [171] [ 70/312] eta: 0:03:18 lr: 0.002899 min_lr: 0.002899 loss: 2.2568 (2.1549) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [171] [ 80/312] eta: 0:03:06 lr: 0.002898 min_lr: 0.002898 loss: 2.1805 (2.1479) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [171] [ 90/312] eta: 0:02:56 lr: 0.002898 min_lr: 0.002898 loss: 2.1805 (2.1490) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0003 max mem: 64948 Epoch: [171] [100/312] eta: 0:02:46 lr: 0.002898 min_lr: 0.002898 loss: 2.2232 (2.1530) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [171] [110/312] eta: 0:02:36 lr: 0.002897 min_lr: 0.002897 loss: 2.0525 (2.1370) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [171] [120/312] eta: 0:02:27 lr: 0.002897 min_lr: 0.002897 loss: 2.0525 (2.1433) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [171] [130/312] eta: 0:02:18 lr: 0.002896 min_lr: 0.002896 loss: 2.0985 (2.1318) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [171] [140/312] eta: 0:02:10 lr: 0.002896 min_lr: 0.002896 loss: 2.2258 (2.1458) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [171] [150/312] eta: 0:02:02 lr: 0.002895 min_lr: 0.002895 loss: 2.3260 (2.1478) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [171] [160/312] eta: 0:01:54 lr: 0.002895 min_lr: 0.002895 loss: 2.2458 (2.1488) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [171] [170/312] eta: 0:01:46 lr: 0.002895 min_lr: 0.002895 loss: 2.2392 (2.1531) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [171] [180/312] eta: 0:01:38 lr: 0.002894 min_lr: 0.002894 loss: 2.2856 (2.1608) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [171] [190/312] eta: 0:01:30 lr: 0.002894 min_lr: 0.002894 loss: 2.2816 (2.1594) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [171] [200/312] eta: 0:01:22 lr: 0.002893 min_lr: 0.002893 loss: 2.2642 (2.1636) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [171] [210/312] eta: 0:01:15 lr: 0.002893 min_lr: 0.002893 loss: 2.2642 (2.1642) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [171] [220/312] eta: 0:01:07 lr: 0.002893 min_lr: 0.002893 loss: 2.3487 (2.1677) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [171] [230/312] eta: 0:01:00 lr: 0.002892 min_lr: 0.002892 loss: 2.2889 (2.1687) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [171] [240/312] eta: 0:00:52 lr: 0.002892 min_lr: 0.002892 loss: 2.2932 (2.1785) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [171] [250/312] eta: 0:00:45 lr: 0.002891 min_lr: 0.002891 loss: 2.3805 (2.1819) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [171] [260/312] eta: 0:00:37 lr: 0.002891 min_lr: 0.002891 loss: 2.2681 (2.1829) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [171] [270/312] eta: 0:00:30 lr: 0.002890 min_lr: 0.002890 loss: 2.2536 (2.1793) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [171] [280/312] eta: 0:00:23 lr: 0.002890 min_lr: 0.002890 loss: 2.2190 (2.1775) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [171] [290/312] eta: 0:00:15 lr: 0.002890 min_lr: 0.002890 loss: 2.0080 (2.1707) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [171] [300/312] eta: 0:00:08 lr: 0.002889 min_lr: 0.002889 loss: 2.1137 (2.1727) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [171] [310/312] eta: 0:00:01 lr: 0.002889 min_lr: 0.002889 loss: 2.1137 (2.1651) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [171] [311/312] eta: 0:00:00 lr: 0.002889 min_lr: 0.002889 loss: 2.1055 (2.1646) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [171] Total time: 0:03:46 (0.7255 s / it) Averaged stats: lr: 0.002889 min_lr: 0.002889 loss: 2.1055 (2.1877) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6796 (0.6796) acc1: 83.5938 (83.5938) acc5: 94.7917 (94.7917) time: 4.7106 data: 4.4913 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 1.0046 (0.9097) acc1: 76.3021 (76.5760) acc5: 92.9688 (93.6000) time: 0.6747 data: 0.4991 max mem: 64948 Test: Total time: 0:00:06 (0.6981 s / it) * Acc@1 77.286 Acc@5 93.452 loss 0.888 Accuracy of the model on the 50000 test images: 77.3% Max accuracy: 77.64% Test: [0/9] eta: 0:00:44 loss: 0.5809 (0.5809) acc1: 84.8958 (84.8958) acc5: 96.8750 (96.8750) time: 4.9254 data: 4.7072 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7552 (0.7460) acc1: 80.4688 (79.4880) acc5: 95.8333 (95.3280) time: 0.6986 data: 0.5231 max mem: 64948 Test: Total time: 0:00:06 (0.7080 s / it) * Acc@1 80.460 Acc@5 95.350 loss 0.724 Accuracy of the model EMA on 50000 test images: 80.5% Max EMA accuracy: 80.46% Epoch: [172] [ 0/312] eta: 0:50:58 lr: 0.002889 min_lr: 0.002889 loss: 2.2764 (2.2764) weight_decay: 0.0500 (0.0500) time: 9.8029 data: 9.0034 max mem: 64948 Epoch: [172] [ 10/312] eta: 0:07:44 lr: 0.002888 min_lr: 0.002888 loss: 2.1574 (2.1021) weight_decay: 0.0500 (0.0500) time: 1.5394 data: 0.8189 max mem: 64948 Epoch: [172] [ 20/312] eta: 0:05:33 lr: 0.002888 min_lr: 0.002888 loss: 2.1798 (2.1563) weight_decay: 0.0500 (0.0500) time: 0.7079 data: 0.0004 max mem: 64948 Epoch: [172] [ 30/312] eta: 0:04:41 lr: 0.002887 min_lr: 0.002887 loss: 2.3096 (2.1558) weight_decay: 0.0500 (0.0500) time: 0.7020 data: 0.0004 max mem: 64948 Epoch: [172] [ 40/312] eta: 0:04:11 lr: 0.002887 min_lr: 0.002887 loss: 2.3452 (2.2185) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [172] [ 50/312] eta: 0:03:50 lr: 0.002887 min_lr: 0.002887 loss: 2.2729 (2.1993) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [172] [ 60/312] eta: 0:03:34 lr: 0.002886 min_lr: 0.002886 loss: 2.2705 (2.2056) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [172] [ 70/312] eta: 0:03:20 lr: 0.002886 min_lr: 0.002886 loss: 2.2458 (2.1900) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [172] [ 80/312] eta: 0:03:08 lr: 0.002885 min_lr: 0.002885 loss: 2.2458 (2.2042) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [172] [ 90/312] eta: 0:02:57 lr: 0.002885 min_lr: 0.002885 loss: 2.1959 (2.1922) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [172] [100/312] eta: 0:02:47 lr: 0.002884 min_lr: 0.002884 loss: 2.1751 (2.1888) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [172] [110/312] eta: 0:02:37 lr: 0.002884 min_lr: 0.002884 loss: 2.1751 (2.1794) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [172] [120/312] eta: 0:02:28 lr: 0.002884 min_lr: 0.002884 loss: 2.2335 (2.1854) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [172] [130/312] eta: 0:02:19 lr: 0.002883 min_lr: 0.002883 loss: 2.1864 (2.1780) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [172] [140/312] eta: 0:02:11 lr: 0.002883 min_lr: 0.002883 loss: 2.1745 (2.1843) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [172] [150/312] eta: 0:02:02 lr: 0.002882 min_lr: 0.002882 loss: 2.3052 (2.1842) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [172] [160/312] eta: 0:01:54 lr: 0.002882 min_lr: 0.002882 loss: 2.3057 (2.1958) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [172] [170/312] eta: 0:01:46 lr: 0.002882 min_lr: 0.002882 loss: 2.4192 (2.1990) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [172] [180/312] eta: 0:01:38 lr: 0.002881 min_lr: 0.002881 loss: 2.4024 (2.2046) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [172] [190/312] eta: 0:01:30 lr: 0.002881 min_lr: 0.002881 loss: 2.3161 (2.2041) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [172] [200/312] eta: 0:01:23 lr: 0.002880 min_lr: 0.002880 loss: 2.2429 (2.1991) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [172] [210/312] eta: 0:01:15 lr: 0.002880 min_lr: 0.002880 loss: 2.2429 (2.1991) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [172] [220/312] eta: 0:01:07 lr: 0.002879 min_lr: 0.002879 loss: 2.4426 (2.2067) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0005 max mem: 64948 Epoch: [172] [230/312] eta: 0:01:00 lr: 0.002879 min_lr: 0.002879 loss: 2.3171 (2.2063) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0005 max mem: 64948 Epoch: [172] [240/312] eta: 0:00:52 lr: 0.002879 min_lr: 0.002879 loss: 2.2839 (2.2113) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [172] [250/312] eta: 0:00:45 lr: 0.002878 min_lr: 0.002878 loss: 2.1896 (2.2046) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [172] [260/312] eta: 0:00:38 lr: 0.002878 min_lr: 0.002878 loss: 2.0873 (2.2012) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [172] [270/312] eta: 0:00:30 lr: 0.002877 min_lr: 0.002877 loss: 2.1816 (2.1982) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [172] [280/312] eta: 0:00:23 lr: 0.002877 min_lr: 0.002877 loss: 2.2285 (2.2046) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [172] [290/312] eta: 0:00:15 lr: 0.002876 min_lr: 0.002876 loss: 2.3007 (2.2057) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [172] [300/312] eta: 0:00:08 lr: 0.002876 min_lr: 0.002876 loss: 2.1759 (2.2050) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [172] [310/312] eta: 0:00:01 lr: 0.002876 min_lr: 0.002876 loss: 2.1046 (2.2001) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [172] [311/312] eta: 0:00:00 lr: 0.002876 min_lr: 0.002876 loss: 2.1046 (2.1997) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [172] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.002876 min_lr: 0.002876 loss: 2.1046 (2.1816) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.6060 (0.6060) acc1: 84.1146 (84.1146) acc5: 95.8333 (95.8333) time: 4.3744 data: 4.1670 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9411 (0.8793) acc1: 74.7396 (75.7760) acc5: 95.8333 (94.3680) time: 0.6374 data: 0.4631 max mem: 64948 Test: Total time: 0:00:05 (0.6599 s / it) * Acc@1 77.628 Acc@5 93.904 loss 0.864 Accuracy of the model on the 50000 test images: 77.6% Max accuracy: 77.64% Test: [0/9] eta: 0:00:44 loss: 0.5792 (0.5792) acc1: 84.8958 (84.8958) acc5: 96.8750 (96.8750) time: 4.9945 data: 4.7767 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7544 (0.7446) acc1: 80.2083 (79.3600) acc5: 95.8333 (95.3600) time: 0.7097 data: 0.5308 max mem: 64948 Test: Total time: 0:00:06 (0.7166 s / it) * Acc@1 80.500 Acc@5 95.374 loss 0.722 Accuracy of the model EMA on 50000 test images: 80.5% Max EMA accuracy: 80.50% Epoch: [173] [ 0/312] eta: 0:49:44 lr: 0.002876 min_lr: 0.002876 loss: 2.6779 (2.6779) weight_decay: 0.0500 (0.0500) time: 9.5654 data: 8.4645 max mem: 64948 Epoch: [173] [ 10/312] eta: 0:07:40 lr: 0.002875 min_lr: 0.002875 loss: 2.1331 (2.0894) weight_decay: 0.0500 (0.0500) time: 1.5245 data: 0.7699 max mem: 64948 Epoch: [173] [ 20/312] eta: 0:05:29 lr: 0.002875 min_lr: 0.002875 loss: 2.2189 (2.0983) weight_decay: 0.0500 (0.0500) time: 0.7074 data: 0.0004 max mem: 64948 Epoch: [173] [ 30/312] eta: 0:04:40 lr: 0.002874 min_lr: 0.002874 loss: 2.2561 (2.0935) weight_decay: 0.0500 (0.0500) time: 0.7024 data: 0.0003 max mem: 64948 Epoch: [173] [ 40/312] eta: 0:04:10 lr: 0.002874 min_lr: 0.002874 loss: 2.1706 (2.0991) weight_decay: 0.0500 (0.0500) time: 0.7042 data: 0.0004 max mem: 64948 Epoch: [173] [ 50/312] eta: 0:03:49 lr: 0.002873 min_lr: 0.002873 loss: 2.2938 (2.1210) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [173] [ 60/312] eta: 0:03:33 lr: 0.002873 min_lr: 0.002873 loss: 2.3588 (2.1593) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [173] [ 70/312] eta: 0:03:19 lr: 0.002873 min_lr: 0.002873 loss: 2.2658 (2.1458) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [173] [ 80/312] eta: 0:03:07 lr: 0.002872 min_lr: 0.002872 loss: 2.1710 (2.1653) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [173] [ 90/312] eta: 0:02:56 lr: 0.002872 min_lr: 0.002872 loss: 2.3033 (2.1804) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [173] [100/312] eta: 0:02:46 lr: 0.002871 min_lr: 0.002871 loss: 2.2569 (2.1818) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [173] [110/312] eta: 0:02:37 lr: 0.002871 min_lr: 0.002871 loss: 2.2755 (2.1950) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [173] [120/312] eta: 0:02:28 lr: 0.002871 min_lr: 0.002871 loss: 2.1672 (2.1875) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [173] [130/312] eta: 0:02:19 lr: 0.002870 min_lr: 0.002870 loss: 2.1091 (2.1669) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [173] [140/312] eta: 0:02:10 lr: 0.002870 min_lr: 0.002870 loss: 2.2499 (2.1760) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [173] [150/312] eta: 0:02:02 lr: 0.002869 min_lr: 0.002869 loss: 2.0738 (2.1598) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [173] [160/312] eta: 0:01:54 lr: 0.002869 min_lr: 0.002869 loss: 2.1217 (2.1596) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [173] [170/312] eta: 0:01:46 lr: 0.002868 min_lr: 0.002868 loss: 2.1747 (2.1567) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [173] [180/312] eta: 0:01:38 lr: 0.002868 min_lr: 0.002868 loss: 2.1747 (2.1601) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [173] [190/312] eta: 0:01:30 lr: 0.002868 min_lr: 0.002868 loss: 2.1154 (2.1499) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [173] [200/312] eta: 0:01:22 lr: 0.002867 min_lr: 0.002867 loss: 2.2022 (2.1607) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [173] [210/312] eta: 0:01:15 lr: 0.002867 min_lr: 0.002867 loss: 2.3889 (2.1632) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [173] [220/312] eta: 0:01:07 lr: 0.002866 min_lr: 0.002866 loss: 2.2927 (2.1673) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [173] [230/312] eta: 0:01:00 lr: 0.002866 min_lr: 0.002866 loss: 2.3743 (2.1732) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [173] [240/312] eta: 0:00:52 lr: 0.002865 min_lr: 0.002865 loss: 2.2909 (2.1735) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [173] [250/312] eta: 0:00:45 lr: 0.002865 min_lr: 0.002865 loss: 2.1749 (2.1725) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [173] [260/312] eta: 0:00:37 lr: 0.002865 min_lr: 0.002865 loss: 2.3806 (2.1766) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [173] [270/312] eta: 0:00:30 lr: 0.002864 min_lr: 0.002864 loss: 2.3292 (2.1765) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [173] [280/312] eta: 0:00:23 lr: 0.002864 min_lr: 0.002864 loss: 2.2859 (2.1800) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [173] [290/312] eta: 0:00:15 lr: 0.002863 min_lr: 0.002863 loss: 2.3925 (2.1854) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [173] [300/312] eta: 0:00:08 lr: 0.002863 min_lr: 0.002863 loss: 2.2750 (2.1866) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [173] [310/312] eta: 0:00:01 lr: 0.002862 min_lr: 0.002862 loss: 2.2466 (2.1849) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [173] [311/312] eta: 0:00:00 lr: 0.002862 min_lr: 0.002862 loss: 2.2466 (2.1848) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [173] Total time: 0:03:46 (0.7272 s / it) Averaged stats: lr: 0.002862 min_lr: 0.002862 loss: 2.2466 (2.1858) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6608 (0.6608) acc1: 82.8125 (82.8125) acc5: 95.0521 (95.0521) time: 4.5564 data: 4.3517 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9536 (0.8778) acc1: 76.5625 (77.5680) acc5: 93.4896 (93.6320) time: 0.6575 data: 0.4836 max mem: 64948 Test: Total time: 0:00:06 (0.6808 s / it) * Acc@1 77.622 Acc@5 93.878 loss 0.868 Accuracy of the model on the 50000 test images: 77.6% Max accuracy: 77.64% Test: [0/9] eta: 0:00:43 loss: 0.5774 (0.5774) acc1: 84.8958 (84.8958) acc5: 96.8750 (96.8750) time: 4.8802 data: 4.6759 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7536 (0.7434) acc1: 80.2083 (79.3280) acc5: 95.8333 (95.4240) time: 0.6935 data: 0.5196 max mem: 64948 Test: Total time: 0:00:06 (0.7013 s / it) * Acc@1 80.504 Acc@5 95.406 loss 0.721 Accuracy of the model EMA on 50000 test images: 80.5% Max EMA accuracy: 80.50% Epoch: [174] [ 0/312] eta: 0:50:58 lr: 0.002862 min_lr: 0.002862 loss: 2.2014 (2.2014) weight_decay: 0.0500 (0.0500) time: 9.8041 data: 9.0195 max mem: 64948 Epoch: [174] [ 10/312] eta: 0:07:56 lr: 0.002862 min_lr: 0.002862 loss: 2.2014 (2.1888) weight_decay: 0.0500 (0.0500) time: 1.5782 data: 0.8204 max mem: 64948 Epoch: [174] [ 20/312] eta: 0:05:38 lr: 0.002862 min_lr: 0.002862 loss: 2.1769 (2.1736) weight_decay: 0.0500 (0.0500) time: 0.7260 data: 0.0004 max mem: 64948 Epoch: [174] [ 30/312] eta: 0:04:44 lr: 0.002861 min_lr: 0.002861 loss: 2.2715 (2.1696) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [174] [ 40/312] eta: 0:04:14 lr: 0.002861 min_lr: 0.002861 loss: 2.2214 (2.1818) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [174] [ 50/312] eta: 0:03:52 lr: 0.002860 min_lr: 0.002860 loss: 2.2160 (2.1793) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [174] [ 60/312] eta: 0:03:35 lr: 0.002860 min_lr: 0.002860 loss: 1.9848 (2.1608) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [174] [ 70/312] eta: 0:03:21 lr: 0.002859 min_lr: 0.002859 loss: 1.9848 (2.1281) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [174] [ 80/312] eta: 0:03:09 lr: 0.002859 min_lr: 0.002859 loss: 2.0158 (2.1147) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [174] [ 90/312] eta: 0:02:58 lr: 0.002859 min_lr: 0.002859 loss: 2.1312 (2.1252) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [174] [100/312] eta: 0:02:48 lr: 0.002858 min_lr: 0.002858 loss: 2.3859 (2.1500) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [174] [110/312] eta: 0:02:38 lr: 0.002858 min_lr: 0.002858 loss: 2.2392 (2.1494) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [174] [120/312] eta: 0:02:29 lr: 0.002857 min_lr: 0.002857 loss: 2.0847 (2.1212) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [174] [130/312] eta: 0:02:20 lr: 0.002857 min_lr: 0.002857 loss: 1.9144 (2.1231) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [174] [140/312] eta: 0:02:11 lr: 0.002856 min_lr: 0.002856 loss: 2.1194 (2.1235) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [174] [150/312] eta: 0:02:03 lr: 0.002856 min_lr: 0.002856 loss: 2.1829 (2.1356) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [174] [160/312] eta: 0:01:54 lr: 0.002856 min_lr: 0.002856 loss: 2.0701 (2.1259) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [174] [170/312] eta: 0:01:46 lr: 0.002855 min_lr: 0.002855 loss: 2.0332 (2.1285) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [174] [180/312] eta: 0:01:38 lr: 0.002855 min_lr: 0.002855 loss: 2.2227 (2.1278) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [174] [190/312] eta: 0:01:31 lr: 0.002854 min_lr: 0.002854 loss: 2.3195 (2.1412) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [174] [200/312] eta: 0:01:23 lr: 0.002854 min_lr: 0.002854 loss: 2.2942 (2.1375) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [174] [210/312] eta: 0:01:15 lr: 0.002854 min_lr: 0.002854 loss: 2.0494 (2.1428) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [174] [220/312] eta: 0:01:07 lr: 0.002853 min_lr: 0.002853 loss: 2.3580 (2.1494) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [174] [230/312] eta: 0:01:00 lr: 0.002853 min_lr: 0.002853 loss: 2.3599 (2.1505) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [174] [240/312] eta: 0:00:52 lr: 0.002852 min_lr: 0.002852 loss: 2.1800 (2.1511) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [174] [250/312] eta: 0:00:45 lr: 0.002852 min_lr: 0.002852 loss: 2.1250 (2.1461) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [174] [260/312] eta: 0:00:38 lr: 0.002851 min_lr: 0.002851 loss: 2.1850 (2.1456) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [174] [270/312] eta: 0:00:30 lr: 0.002851 min_lr: 0.002851 loss: 2.3198 (2.1507) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [174] [280/312] eta: 0:00:23 lr: 0.002851 min_lr: 0.002851 loss: 2.3237 (2.1576) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [174] [290/312] eta: 0:00:16 lr: 0.002850 min_lr: 0.002850 loss: 2.3811 (2.1630) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [174] [300/312] eta: 0:00:08 lr: 0.002850 min_lr: 0.002850 loss: 2.2742 (2.1665) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0002 max mem: 64948 Epoch: [174] [310/312] eta: 0:00:01 lr: 0.002849 min_lr: 0.002849 loss: 2.2952 (2.1745) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [174] [311/312] eta: 0:00:00 lr: 0.002849 min_lr: 0.002849 loss: 2.2952 (2.1764) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [174] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.002849 min_lr: 0.002849 loss: 2.2952 (2.1782) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6791 (0.6791) acc1: 82.5521 (82.5521) acc5: 94.7917 (94.7917) time: 4.5211 data: 4.3108 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9135 (0.8829) acc1: 79.1667 (76.8320) acc5: 94.0104 (93.8240) time: 0.6536 data: 0.4791 max mem: 64948 Test: Total time: 0:00:06 (0.6777 s / it) * Acc@1 77.646 Acc@5 93.842 loss 0.867 Accuracy of the model on the 50000 test images: 77.6% Max accuracy: 77.65% Test: [0/9] eta: 0:00:41 loss: 0.5756 (0.5756) acc1: 84.8958 (84.8958) acc5: 96.8750 (96.8750) time: 4.5906 data: 4.3809 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7529 (0.7423) acc1: 80.2083 (79.3920) acc5: 96.0938 (95.4880) time: 0.6613 data: 0.4869 max mem: 64948 Test: Total time: 0:00:06 (0.6696 s / it) * Acc@1 80.534 Acc@5 95.416 loss 0.720 Accuracy of the model EMA on 50000 test images: 80.5% Max EMA accuracy: 80.53% Epoch: [175] [ 0/312] eta: 0:46:09 lr: 0.002849 min_lr: 0.002849 loss: 2.6511 (2.6511) weight_decay: 0.0500 (0.0500) time: 8.8768 data: 8.0747 max mem: 64948 Epoch: [175] [ 10/312] eta: 0:07:34 lr: 0.002849 min_lr: 0.002849 loss: 2.2405 (2.0431) weight_decay: 0.0500 (0.0500) time: 1.5056 data: 0.7344 max mem: 64948 Epoch: [175] [ 20/312] eta: 0:05:27 lr: 0.002848 min_lr: 0.002848 loss: 2.2052 (2.1073) weight_decay: 0.0500 (0.0500) time: 0.7333 data: 0.0004 max mem: 64948 Epoch: [175] [ 30/312] eta: 0:04:37 lr: 0.002848 min_lr: 0.002848 loss: 2.2587 (2.1590) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [175] [ 40/312] eta: 0:04:08 lr: 0.002848 min_lr: 0.002848 loss: 2.1885 (2.1350) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [175] [ 50/312] eta: 0:03:48 lr: 0.002847 min_lr: 0.002847 loss: 2.0584 (2.1216) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [175] [ 60/312] eta: 0:03:32 lr: 0.002847 min_lr: 0.002847 loss: 2.0563 (2.1291) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [175] [ 70/312] eta: 0:03:18 lr: 0.002846 min_lr: 0.002846 loss: 2.1387 (2.1171) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [175] [ 80/312] eta: 0:03:06 lr: 0.002846 min_lr: 0.002846 loss: 2.1985 (2.1393) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [175] [ 90/312] eta: 0:02:56 lr: 0.002845 min_lr: 0.002845 loss: 2.1970 (2.1300) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [175] [100/312] eta: 0:02:46 lr: 0.002845 min_lr: 0.002845 loss: 1.8739 (2.1132) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [175] [110/312] eta: 0:02:36 lr: 0.002845 min_lr: 0.002845 loss: 1.8712 (2.1022) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [175] [120/312] eta: 0:02:27 lr: 0.002844 min_lr: 0.002844 loss: 1.9826 (2.1102) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [175] [130/312] eta: 0:02:19 lr: 0.002844 min_lr: 0.002844 loss: 2.0222 (2.0993) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [175] [140/312] eta: 0:02:10 lr: 0.002843 min_lr: 0.002843 loss: 1.9816 (2.0949) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [175] [150/312] eta: 0:02:02 lr: 0.002843 min_lr: 0.002843 loss: 2.0330 (2.1006) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [175] [160/312] eta: 0:01:54 lr: 0.002842 min_lr: 0.002842 loss: 2.2166 (2.1063) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [175] [170/312] eta: 0:01:46 lr: 0.002842 min_lr: 0.002842 loss: 2.1038 (2.1047) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [175] [180/312] eta: 0:01:38 lr: 0.002842 min_lr: 0.002842 loss: 2.1038 (2.1125) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [175] [190/312] eta: 0:01:30 lr: 0.002841 min_lr: 0.002841 loss: 2.3138 (2.1135) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [175] [200/312] eta: 0:01:22 lr: 0.002841 min_lr: 0.002841 loss: 2.3079 (2.1238) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [175] [210/312] eta: 0:01:15 lr: 0.002840 min_lr: 0.002840 loss: 2.3000 (2.1213) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [175] [220/312] eta: 0:01:07 lr: 0.002840 min_lr: 0.002840 loss: 2.1226 (2.1232) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [175] [230/312] eta: 0:01:00 lr: 0.002839 min_lr: 0.002839 loss: 2.2675 (2.1277) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [175] [240/312] eta: 0:00:52 lr: 0.002839 min_lr: 0.002839 loss: 2.1103 (2.1227) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [175] [250/312] eta: 0:00:45 lr: 0.002839 min_lr: 0.002839 loss: 2.2715 (2.1311) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [175] [260/312] eta: 0:00:37 lr: 0.002838 min_lr: 0.002838 loss: 2.2715 (2.1271) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [175] [270/312] eta: 0:00:30 lr: 0.002838 min_lr: 0.002838 loss: 2.2283 (2.1331) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [175] [280/312] eta: 0:00:23 lr: 0.002837 min_lr: 0.002837 loss: 2.3084 (2.1393) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0009 max mem: 64948 Epoch: [175] [290/312] eta: 0:00:15 lr: 0.002837 min_lr: 0.002837 loss: 2.2757 (2.1387) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [175] [300/312] eta: 0:00:08 lr: 0.002836 min_lr: 0.002836 loss: 2.3792 (2.1485) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [175] [310/312] eta: 0:00:01 lr: 0.002836 min_lr: 0.002836 loss: 2.3675 (2.1497) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [175] [311/312] eta: 0:00:00 lr: 0.002836 min_lr: 0.002836 loss: 2.3409 (2.1490) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [175] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.002836 min_lr: 0.002836 loss: 2.3409 (2.1746) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6935 (0.6935) acc1: 81.2500 (81.2500) acc5: 94.5312 (94.5312) time: 4.6699 data: 4.4516 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9374 (0.9036) acc1: 76.5625 (76.6720) acc5: 94.0104 (93.4720) time: 0.6701 data: 0.4947 max mem: 64948 Test: Total time: 0:00:06 (0.6932 s / it) * Acc@1 77.472 Acc@5 93.962 loss 0.865 Accuracy of the model on the 50000 test images: 77.5% Max accuracy: 77.65% Test: [0/9] eta: 0:00:46 loss: 0.5743 (0.5743) acc1: 85.1562 (85.1562) acc5: 96.8750 (96.8750) time: 5.1140 data: 4.8961 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7522 (0.7411) acc1: 80.2083 (79.5200) acc5: 96.0938 (95.5200) time: 0.7196 data: 0.5441 max mem: 64948 Test: Total time: 0:00:06 (0.7316 s / it) * Acc@1 80.568 Acc@5 95.428 loss 0.719 Accuracy of the model EMA on 50000 test images: 80.6% Max EMA accuracy: 80.57% Epoch: [176] [ 0/312] eta: 0:46:18 lr: 0.002836 min_lr: 0.002836 loss: 2.6239 (2.6239) weight_decay: 0.0500 (0.0500) time: 8.9066 data: 7.1222 max mem: 64948 Epoch: [176] [ 10/312] eta: 0:07:31 lr: 0.002836 min_lr: 0.002836 loss: 2.2574 (2.1557) weight_decay: 0.0500 (0.0500) time: 1.4946 data: 0.6480 max mem: 64948 Epoch: [176] [ 20/312] eta: 0:05:25 lr: 0.002835 min_lr: 0.002835 loss: 2.0139 (2.1472) weight_decay: 0.0500 (0.0500) time: 0.7237 data: 0.0004 max mem: 64948 Epoch: [176] [ 30/312] eta: 0:04:35 lr: 0.002835 min_lr: 0.002835 loss: 2.2328 (2.1780) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [176] [ 40/312] eta: 0:04:07 lr: 0.002834 min_lr: 0.002834 loss: 2.2328 (2.1417) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [176] [ 50/312] eta: 0:03:47 lr: 0.002834 min_lr: 0.002834 loss: 2.2600 (2.1554) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [176] [ 60/312] eta: 0:03:31 lr: 0.002833 min_lr: 0.002833 loss: 2.2797 (2.1225) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [176] [ 70/312] eta: 0:03:18 lr: 0.002833 min_lr: 0.002833 loss: 2.1874 (2.1299) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [176] [ 80/312] eta: 0:03:06 lr: 0.002833 min_lr: 0.002833 loss: 2.1975 (2.1201) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [176] [ 90/312] eta: 0:02:56 lr: 0.002832 min_lr: 0.002832 loss: 2.1975 (2.1212) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [176] [100/312] eta: 0:02:45 lr: 0.002832 min_lr: 0.002832 loss: 2.1922 (2.1188) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [176] [110/312] eta: 0:02:36 lr: 0.002831 min_lr: 0.002831 loss: 2.0994 (2.1362) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [176] [120/312] eta: 0:02:27 lr: 0.002831 min_lr: 0.002831 loss: 2.4229 (2.1451) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [176] [130/312] eta: 0:02:18 lr: 0.002830 min_lr: 0.002830 loss: 2.2323 (2.1386) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [176] [140/312] eta: 0:02:10 lr: 0.002830 min_lr: 0.002830 loss: 2.2310 (2.1392) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [176] [150/312] eta: 0:02:02 lr: 0.002830 min_lr: 0.002830 loss: 2.1899 (2.1418) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [176] [160/312] eta: 0:01:53 lr: 0.002829 min_lr: 0.002829 loss: 2.2110 (2.1521) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [176] [170/312] eta: 0:01:45 lr: 0.002829 min_lr: 0.002829 loss: 2.2856 (2.1491) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [176] [180/312] eta: 0:01:38 lr: 0.002828 min_lr: 0.002828 loss: 2.2100 (2.1559) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [176] [190/312] eta: 0:01:30 lr: 0.002828 min_lr: 0.002828 loss: 2.2100 (2.1573) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [176] [200/312] eta: 0:01:22 lr: 0.002827 min_lr: 0.002827 loss: 2.1569 (2.1595) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [176] [210/312] eta: 0:01:15 lr: 0.002827 min_lr: 0.002827 loss: 2.2474 (2.1648) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [176] [220/312] eta: 0:01:07 lr: 0.002827 min_lr: 0.002827 loss: 2.2799 (2.1680) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [176] [230/312] eta: 0:01:00 lr: 0.002826 min_lr: 0.002826 loss: 2.2345 (2.1672) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [176] [240/312] eta: 0:00:52 lr: 0.002826 min_lr: 0.002826 loss: 2.1024 (2.1626) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [176] [250/312] eta: 0:00:45 lr: 0.002825 min_lr: 0.002825 loss: 2.1389 (2.1611) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [176] [260/312] eta: 0:00:37 lr: 0.002825 min_lr: 0.002825 loss: 2.2920 (2.1631) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [176] [270/312] eta: 0:00:30 lr: 0.002824 min_lr: 0.002824 loss: 2.2920 (2.1630) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [176] [280/312] eta: 0:00:23 lr: 0.002824 min_lr: 0.002824 loss: 2.2527 (2.1631) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [176] [290/312] eta: 0:00:15 lr: 0.002824 min_lr: 0.002824 loss: 2.3069 (2.1663) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [176] [300/312] eta: 0:00:08 lr: 0.002823 min_lr: 0.002823 loss: 2.2777 (2.1671) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [176] [310/312] eta: 0:00:01 lr: 0.002823 min_lr: 0.002823 loss: 2.2205 (2.1705) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [176] [311/312] eta: 0:00:00 lr: 0.002823 min_lr: 0.002823 loss: 2.1857 (2.1694) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [176] Total time: 0:03:46 (0.7258 s / it) Averaged stats: lr: 0.002823 min_lr: 0.002823 loss: 2.1857 (2.1722) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.6685 (0.6685) acc1: 83.5938 (83.5938) acc5: 96.3542 (96.3542) time: 4.3949 data: 4.1751 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9475 (0.8902) acc1: 76.5625 (77.2480) acc5: 94.5312 (93.8880) time: 0.6396 data: 0.4640 max mem: 64948 Test: Total time: 0:00:05 (0.6631 s / it) * Acc@1 78.040 Acc@5 93.864 loss 0.865 Accuracy of the model on the 50000 test images: 78.0% Max accuracy: 78.04% Test: [0/9] eta: 0:00:38 loss: 0.5729 (0.5729) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 4.3311 data: 4.1174 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7510 (0.7396) acc1: 80.2083 (79.4240) acc5: 96.0938 (95.5840) time: 0.6326 data: 0.4576 max mem: 64948 Test: Total time: 0:00:05 (0.6403 s / it) * Acc@1 80.604 Acc@5 95.440 loss 0.717 Accuracy of the model EMA on 50000 test images: 80.6% Max EMA accuracy: 80.60% Epoch: [177] [ 0/312] eta: 0:46:14 lr: 0.002823 min_lr: 0.002823 loss: 2.1581 (2.1581) weight_decay: 0.0500 (0.0500) time: 8.8911 data: 8.1071 max mem: 64948 Epoch: [177] [ 10/312] eta: 0:07:59 lr: 0.002822 min_lr: 0.002822 loss: 2.1583 (2.1895) weight_decay: 0.0500 (0.0500) time: 1.5862 data: 0.8739 max mem: 64948 Epoch: [177] [ 20/312] eta: 0:05:39 lr: 0.002822 min_lr: 0.002822 loss: 2.1073 (2.0855) weight_decay: 0.0500 (0.0500) time: 0.7759 data: 0.0755 max mem: 64948 Epoch: [177] [ 30/312] eta: 0:04:45 lr: 0.002821 min_lr: 0.002821 loss: 2.0987 (2.1353) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [177] [ 40/312] eta: 0:04:14 lr: 0.002821 min_lr: 0.002821 loss: 2.1791 (2.1005) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [177] [ 50/312] eta: 0:03:52 lr: 0.002821 min_lr: 0.002821 loss: 2.1591 (2.1157) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [177] [ 60/312] eta: 0:03:35 lr: 0.002820 min_lr: 0.002820 loss: 2.2680 (2.1243) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [177] [ 70/312] eta: 0:03:21 lr: 0.002820 min_lr: 0.002820 loss: 1.9759 (2.1120) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [177] [ 80/312] eta: 0:03:09 lr: 0.002819 min_lr: 0.002819 loss: 2.0656 (2.1272) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [177] [ 90/312] eta: 0:02:58 lr: 0.002819 min_lr: 0.002819 loss: 2.2202 (2.1317) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [177] [100/312] eta: 0:02:48 lr: 0.002818 min_lr: 0.002818 loss: 2.1626 (2.1338) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [177] [110/312] eta: 0:02:38 lr: 0.002818 min_lr: 0.002818 loss: 2.3047 (2.1466) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [177] [120/312] eta: 0:02:29 lr: 0.002818 min_lr: 0.002818 loss: 2.2724 (2.1511) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [177] [130/312] eta: 0:02:20 lr: 0.002817 min_lr: 0.002817 loss: 2.2465 (2.1530) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [177] [140/312] eta: 0:02:11 lr: 0.002817 min_lr: 0.002817 loss: 2.3164 (2.1573) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [177] [150/312] eta: 0:02:03 lr: 0.002816 min_lr: 0.002816 loss: 2.3180 (2.1719) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [177] [160/312] eta: 0:01:54 lr: 0.002816 min_lr: 0.002816 loss: 2.2708 (2.1661) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [177] [170/312] eta: 0:01:46 lr: 0.002815 min_lr: 0.002815 loss: 2.0445 (2.1704) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [177] [180/312] eta: 0:01:38 lr: 0.002815 min_lr: 0.002815 loss: 2.1440 (2.1668) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [177] [190/312] eta: 0:01:31 lr: 0.002815 min_lr: 0.002815 loss: 2.3060 (2.1747) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [177] [200/312] eta: 0:01:23 lr: 0.002814 min_lr: 0.002814 loss: 2.3750 (2.1786) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [177] [210/312] eta: 0:01:15 lr: 0.002814 min_lr: 0.002814 loss: 2.3095 (2.1770) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [177] [220/312] eta: 0:01:08 lr: 0.002813 min_lr: 0.002813 loss: 2.3075 (2.1811) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [177] [230/312] eta: 0:01:00 lr: 0.002813 min_lr: 0.002813 loss: 2.3301 (2.1801) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [177] [240/312] eta: 0:00:52 lr: 0.002812 min_lr: 0.002812 loss: 2.2142 (2.1774) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [177] [250/312] eta: 0:00:45 lr: 0.002812 min_lr: 0.002812 loss: 2.2142 (2.1777) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [177] [260/312] eta: 0:00:38 lr: 0.002812 min_lr: 0.002812 loss: 2.3295 (2.1842) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [177] [270/312] eta: 0:00:30 lr: 0.002811 min_lr: 0.002811 loss: 2.2908 (2.1851) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [177] [280/312] eta: 0:00:23 lr: 0.002811 min_lr: 0.002811 loss: 2.1511 (2.1792) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0010 max mem: 64948 Epoch: [177] [290/312] eta: 0:00:16 lr: 0.002810 min_lr: 0.002810 loss: 2.1849 (2.1833) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0008 max mem: 64948 Epoch: [177] [300/312] eta: 0:00:08 lr: 0.002810 min_lr: 0.002810 loss: 2.2671 (2.1854) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [177] [310/312] eta: 0:00:01 lr: 0.002809 min_lr: 0.002809 loss: 2.3145 (2.1869) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [177] [311/312] eta: 0:00:00 lr: 0.002809 min_lr: 0.002809 loss: 2.3851 (2.1886) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [177] Total time: 0:03:47 (0.7293 s / it) Averaged stats: lr: 0.002809 min_lr: 0.002809 loss: 2.3851 (2.1802) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.7087 (0.7087) acc1: 82.0312 (82.0312) acc5: 95.0521 (95.0521) time: 4.5429 data: 4.3237 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8971 (0.9015) acc1: 77.0833 (76.5120) acc5: 95.0521 (93.3760) time: 0.6565 data: 0.4805 max mem: 64948 Test: Total time: 0:00:06 (0.6801 s / it) * Acc@1 77.734 Acc@5 93.812 loss 0.871 Accuracy of the model on the 50000 test images: 77.7% Max accuracy: 78.04% Test: [0/9] eta: 0:00:46 loss: 0.5714 (0.5714) acc1: 85.1562 (85.1562) acc5: 96.8750 (96.8750) time: 5.2141 data: 4.9971 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7497 (0.7380) acc1: 80.2083 (79.5200) acc5: 96.0938 (95.5840) time: 0.7306 data: 0.5553 max mem: 64948 Test: Total time: 0:00:06 (0.7396 s / it) * Acc@1 80.644 Acc@5 95.474 loss 0.716 Accuracy of the model EMA on 50000 test images: 80.6% Max EMA accuracy: 80.64% Epoch: [178] [ 0/312] eta: 0:44:51 lr: 0.002809 min_lr: 0.002809 loss: 2.3016 (2.3016) weight_decay: 0.0500 (0.0500) time: 8.6269 data: 7.8352 max mem: 64948 Epoch: [178] [ 10/312] eta: 0:07:33 lr: 0.002809 min_lr: 0.002809 loss: 2.1190 (2.0194) weight_decay: 0.0500 (0.0500) time: 1.5007 data: 0.7665 max mem: 64948 Epoch: [178] [ 20/312] eta: 0:05:26 lr: 0.002808 min_lr: 0.002808 loss: 2.2211 (2.1307) weight_decay: 0.0500 (0.0500) time: 0.7418 data: 0.0300 max mem: 64948 Epoch: [178] [ 30/312] eta: 0:04:36 lr: 0.002808 min_lr: 0.002808 loss: 2.2915 (2.1889) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0003 max mem: 64948 Epoch: [178] [ 40/312] eta: 0:04:08 lr: 0.002808 min_lr: 0.002808 loss: 2.2391 (2.1412) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [178] [ 50/312] eta: 0:03:47 lr: 0.002807 min_lr: 0.002807 loss: 2.1822 (2.1625) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [178] [ 60/312] eta: 0:03:32 lr: 0.002807 min_lr: 0.002807 loss: 2.2445 (2.1645) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [178] [ 70/312] eta: 0:03:18 lr: 0.002806 min_lr: 0.002806 loss: 2.2686 (2.1797) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [178] [ 80/312] eta: 0:03:06 lr: 0.002806 min_lr: 0.002806 loss: 2.2382 (2.1747) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [178] [ 90/312] eta: 0:02:56 lr: 0.002805 min_lr: 0.002805 loss: 2.2309 (2.1899) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [178] [100/312] eta: 0:02:46 lr: 0.002805 min_lr: 0.002805 loss: 2.3184 (2.1920) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [178] [110/312] eta: 0:02:36 lr: 0.002805 min_lr: 0.002805 loss: 2.1778 (2.1767) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [178] [120/312] eta: 0:02:27 lr: 0.002804 min_lr: 0.002804 loss: 1.9482 (2.1629) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [178] [130/312] eta: 0:02:18 lr: 0.002804 min_lr: 0.002804 loss: 2.0308 (2.1658) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [178] [140/312] eta: 0:02:10 lr: 0.002803 min_lr: 0.002803 loss: 2.2702 (2.1631) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [178] [150/312] eta: 0:02:02 lr: 0.002803 min_lr: 0.002803 loss: 2.3217 (2.1694) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [178] [160/312] eta: 0:01:54 lr: 0.002802 min_lr: 0.002802 loss: 2.4388 (2.1811) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [178] [170/312] eta: 0:01:46 lr: 0.002802 min_lr: 0.002802 loss: 2.2315 (2.1777) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [178] [180/312] eta: 0:01:38 lr: 0.002802 min_lr: 0.002802 loss: 2.1516 (2.1687) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [178] [190/312] eta: 0:01:30 lr: 0.002801 min_lr: 0.002801 loss: 2.1468 (2.1680) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [178] [200/312] eta: 0:01:22 lr: 0.002801 min_lr: 0.002801 loss: 2.1730 (2.1624) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [178] [210/312] eta: 0:01:15 lr: 0.002800 min_lr: 0.002800 loss: 2.1649 (2.1620) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [178] [220/312] eta: 0:01:07 lr: 0.002800 min_lr: 0.002800 loss: 2.2489 (2.1656) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [178] [230/312] eta: 0:01:00 lr: 0.002799 min_lr: 0.002799 loss: 2.2958 (2.1633) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [178] [240/312] eta: 0:00:52 lr: 0.002799 min_lr: 0.002799 loss: 2.1404 (2.1605) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [178] [250/312] eta: 0:00:45 lr: 0.002799 min_lr: 0.002799 loss: 2.1404 (2.1609) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [178] [260/312] eta: 0:00:37 lr: 0.002798 min_lr: 0.002798 loss: 2.2069 (2.1594) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0003 max mem: 64948 Epoch: [178] [270/312] eta: 0:00:30 lr: 0.002798 min_lr: 0.002798 loss: 2.1770 (2.1623) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [178] [280/312] eta: 0:00:23 lr: 0.002797 min_lr: 0.002797 loss: 2.1746 (2.1605) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [178] [290/312] eta: 0:00:15 lr: 0.002797 min_lr: 0.002797 loss: 1.8783 (2.1488) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [178] [300/312] eta: 0:00:08 lr: 0.002796 min_lr: 0.002796 loss: 1.8742 (2.1482) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [178] [310/312] eta: 0:00:01 lr: 0.002796 min_lr: 0.002796 loss: 2.0390 (2.1441) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [178] [311/312] eta: 0:00:00 lr: 0.002796 min_lr: 0.002796 loss: 2.0992 (2.1440) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0001 max mem: 64948 Epoch: [178] Total time: 0:03:46 (0.7264 s / it) Averaged stats: lr: 0.002796 min_lr: 0.002796 loss: 2.0992 (2.1576) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6155 (0.6155) acc1: 85.1562 (85.1562) acc5: 95.8333 (95.8333) time: 4.6140 data: 4.3949 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8897 (0.8667) acc1: 77.8646 (77.2160) acc5: 93.7500 (93.7600) time: 0.6639 data: 0.4884 max mem: 64948 Test: Total time: 0:00:06 (0.6911 s / it) * Acc@1 77.844 Acc@5 93.902 loss 0.858 Accuracy of the model on the 50000 test images: 77.8% Max accuracy: 78.04% Test: [0/9] eta: 0:00:45 loss: 0.5697 (0.5697) acc1: 85.1562 (85.1562) acc5: 96.8750 (96.8750) time: 5.0019 data: 4.7936 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7484 (0.7366) acc1: 80.2083 (79.6160) acc5: 96.0938 (95.6160) time: 0.7071 data: 0.5327 max mem: 64948 Test: Total time: 0:00:06 (0.7202 s / it) * Acc@1 80.668 Acc@5 95.496 loss 0.715 Accuracy of the model EMA on 50000 test images: 80.7% Max EMA accuracy: 80.67% Epoch: [179] [ 0/312] eta: 0:51:56 lr: 0.002796 min_lr: 0.002796 loss: 2.3617 (2.3617) weight_decay: 0.0500 (0.0500) time: 9.9894 data: 9.1332 max mem: 64948 Epoch: [179] [ 10/312] eta: 0:07:48 lr: 0.002796 min_lr: 0.002796 loss: 2.3617 (2.2254) weight_decay: 0.0500 (0.0500) time: 1.5510 data: 0.8307 max mem: 64948 Epoch: [179] [ 20/312] eta: 0:05:33 lr: 0.002795 min_lr: 0.002795 loss: 2.2884 (2.2651) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0004 max mem: 64948 Epoch: [179] [ 30/312] eta: 0:04:41 lr: 0.002795 min_lr: 0.002795 loss: 2.2884 (2.2337) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [179] [ 40/312] eta: 0:04:11 lr: 0.002794 min_lr: 0.002794 loss: 2.1814 (2.2102) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [179] [ 50/312] eta: 0:03:50 lr: 0.002794 min_lr: 0.002794 loss: 2.2840 (2.2340) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [179] [ 60/312] eta: 0:03:34 lr: 0.002793 min_lr: 0.002793 loss: 2.1221 (2.1760) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [179] [ 70/312] eta: 0:03:20 lr: 0.002793 min_lr: 0.002793 loss: 2.0759 (2.1841) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [179] [ 80/312] eta: 0:03:08 lr: 0.002792 min_lr: 0.002792 loss: 2.2216 (2.2004) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [179] [ 90/312] eta: 0:02:57 lr: 0.002792 min_lr: 0.002792 loss: 2.2209 (2.1980) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [179] [100/312] eta: 0:02:47 lr: 0.002792 min_lr: 0.002792 loss: 2.2209 (2.2007) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [179] [110/312] eta: 0:02:37 lr: 0.002791 min_lr: 0.002791 loss: 2.2140 (2.2038) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [179] [120/312] eta: 0:02:28 lr: 0.002791 min_lr: 0.002791 loss: 2.2140 (2.1990) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [179] [130/312] eta: 0:02:19 lr: 0.002790 min_lr: 0.002790 loss: 2.3982 (2.2163) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [179] [140/312] eta: 0:02:11 lr: 0.002790 min_lr: 0.002790 loss: 2.3982 (2.2113) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [179] [150/312] eta: 0:02:02 lr: 0.002789 min_lr: 0.002789 loss: 2.3211 (2.2118) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [179] [160/312] eta: 0:01:54 lr: 0.002789 min_lr: 0.002789 loss: 2.2759 (2.2098) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [179] [170/312] eta: 0:01:46 lr: 0.002789 min_lr: 0.002789 loss: 2.2759 (2.2113) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [179] [180/312] eta: 0:01:38 lr: 0.002788 min_lr: 0.002788 loss: 2.1032 (2.1983) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [179] [190/312] eta: 0:01:30 lr: 0.002788 min_lr: 0.002788 loss: 2.0651 (2.1965) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [179] [200/312] eta: 0:01:23 lr: 0.002787 min_lr: 0.002787 loss: 2.2772 (2.1988) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [179] [210/312] eta: 0:01:15 lr: 0.002787 min_lr: 0.002787 loss: 2.2772 (2.2003) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [179] [220/312] eta: 0:01:07 lr: 0.002786 min_lr: 0.002786 loss: 2.2908 (2.1997) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [179] [230/312] eta: 0:01:00 lr: 0.002786 min_lr: 0.002786 loss: 2.0814 (2.1929) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [179] [240/312] eta: 0:00:52 lr: 0.002786 min_lr: 0.002786 loss: 1.9774 (2.1942) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [179] [250/312] eta: 0:00:45 lr: 0.002785 min_lr: 0.002785 loss: 1.9774 (2.1884) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [179] [260/312] eta: 0:00:38 lr: 0.002785 min_lr: 0.002785 loss: 2.2038 (2.1906) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [179] [270/312] eta: 0:00:30 lr: 0.002784 min_lr: 0.002784 loss: 2.2038 (2.1880) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [179] [280/312] eta: 0:00:23 lr: 0.002784 min_lr: 0.002784 loss: 2.0572 (2.1834) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [179] [290/312] eta: 0:00:15 lr: 0.002783 min_lr: 0.002783 loss: 2.0892 (2.1856) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [179] [300/312] eta: 0:00:08 lr: 0.002783 min_lr: 0.002783 loss: 2.0892 (2.1821) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [179] [310/312] eta: 0:00:01 lr: 0.002783 min_lr: 0.002783 loss: 1.9286 (2.1787) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [179] [311/312] eta: 0:00:00 lr: 0.002783 min_lr: 0.002783 loss: 1.9246 (2.1771) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [179] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.002783 min_lr: 0.002783 loss: 1.9246 (2.1630) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.6456 (0.6456) acc1: 83.8542 (83.8542) acc5: 96.6146 (96.6146) time: 4.7885 data: 4.5689 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9156 (0.8479) acc1: 78.9062 (77.6960) acc5: 94.5312 (94.2720) time: 0.6838 data: 0.5077 max mem: 64948 Test: Total time: 0:00:06 (0.6928 s / it) * Acc@1 78.254 Acc@5 94.122 loss 0.846 Accuracy of the model on the 50000 test images: 78.3% Max accuracy: 78.25% Test: [0/9] eta: 0:00:42 loss: 0.5683 (0.5683) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 4.6922 data: 4.4745 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7470 (0.7352) acc1: 80.2083 (79.6160) acc5: 96.0938 (95.6480) time: 0.6730 data: 0.4973 max mem: 64948 Test: Total time: 0:00:06 (0.6825 s / it) * Acc@1 80.696 Acc@5 95.520 loss 0.713 Accuracy of the model EMA on 50000 test images: 80.7% Max EMA accuracy: 80.70% Epoch: [180] [ 0/312] eta: 0:53:33 lr: 0.002783 min_lr: 0.002783 loss: 2.3747 (2.3747) weight_decay: 0.0500 (0.0500) time: 10.2982 data: 9.4962 max mem: 64948 Epoch: [180] [ 10/312] eta: 0:07:58 lr: 0.002782 min_lr: 0.002782 loss: 2.1833 (2.2096) weight_decay: 0.0500 (0.0500) time: 1.5835 data: 0.8637 max mem: 64948 Epoch: [180] [ 20/312] eta: 0:05:39 lr: 0.002782 min_lr: 0.002782 loss: 2.1383 (2.1458) weight_decay: 0.0500 (0.0500) time: 0.7043 data: 0.0004 max mem: 64948 Epoch: [180] [ 30/312] eta: 0:04:45 lr: 0.002781 min_lr: 0.002781 loss: 2.2360 (2.2168) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0003 max mem: 64948 Epoch: [180] [ 40/312] eta: 0:04:14 lr: 0.002781 min_lr: 0.002781 loss: 2.3094 (2.1995) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [180] [ 50/312] eta: 0:03:52 lr: 0.002780 min_lr: 0.002780 loss: 2.1571 (2.1641) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [180] [ 60/312] eta: 0:03:35 lr: 0.002780 min_lr: 0.002780 loss: 2.2570 (2.1747) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [180] [ 70/312] eta: 0:03:21 lr: 0.002779 min_lr: 0.002779 loss: 2.2570 (2.1481) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [180] [ 80/312] eta: 0:03:09 lr: 0.002779 min_lr: 0.002779 loss: 2.2313 (2.1831) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [180] [ 90/312] eta: 0:02:58 lr: 0.002779 min_lr: 0.002779 loss: 2.4233 (2.2003) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [180] [100/312] eta: 0:02:47 lr: 0.002778 min_lr: 0.002778 loss: 2.0931 (2.1801) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [180] [110/312] eta: 0:02:38 lr: 0.002778 min_lr: 0.002778 loss: 2.0603 (2.1829) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [180] [120/312] eta: 0:02:29 lr: 0.002777 min_lr: 0.002777 loss: 2.1688 (2.1847) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [180] [130/312] eta: 0:02:20 lr: 0.002777 min_lr: 0.002777 loss: 2.1682 (2.1780) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [180] [140/312] eta: 0:02:11 lr: 0.002776 min_lr: 0.002776 loss: 2.1599 (2.1820) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [180] [150/312] eta: 0:02:03 lr: 0.002776 min_lr: 0.002776 loss: 2.1599 (2.1852) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [180] [160/312] eta: 0:01:54 lr: 0.002776 min_lr: 0.002776 loss: 2.4250 (2.2045) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [180] [170/312] eta: 0:01:46 lr: 0.002775 min_lr: 0.002775 loss: 2.3075 (2.2035) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [180] [180/312] eta: 0:01:38 lr: 0.002775 min_lr: 0.002775 loss: 2.2935 (2.2056) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [180] [190/312] eta: 0:01:31 lr: 0.002774 min_lr: 0.002774 loss: 2.1495 (2.1955) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [180] [200/312] eta: 0:01:23 lr: 0.002774 min_lr: 0.002774 loss: 2.2634 (2.2016) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [180] [210/312] eta: 0:01:15 lr: 0.002773 min_lr: 0.002773 loss: 2.2952 (2.2003) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [180] [220/312] eta: 0:01:08 lr: 0.002773 min_lr: 0.002773 loss: 2.2952 (2.2081) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [180] [230/312] eta: 0:01:00 lr: 0.002773 min_lr: 0.002773 loss: 2.3937 (2.2092) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [180] [240/312] eta: 0:00:52 lr: 0.002772 min_lr: 0.002772 loss: 2.3650 (2.2105) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [180] [250/312] eta: 0:00:45 lr: 0.002772 min_lr: 0.002772 loss: 2.2959 (2.2098) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [180] [260/312] eta: 0:00:38 lr: 0.002771 min_lr: 0.002771 loss: 2.2865 (2.2113) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [180] [270/312] eta: 0:00:30 lr: 0.002771 min_lr: 0.002771 loss: 2.1262 (2.2082) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [180] [280/312] eta: 0:00:23 lr: 0.002770 min_lr: 0.002770 loss: 2.0432 (2.2039) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [180] [290/312] eta: 0:00:16 lr: 0.002770 min_lr: 0.002770 loss: 2.2914 (2.2064) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [180] [300/312] eta: 0:00:08 lr: 0.002770 min_lr: 0.002770 loss: 2.2055 (2.1987) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [180] [310/312] eta: 0:00:01 lr: 0.002769 min_lr: 0.002769 loss: 2.1745 (2.1987) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [180] [311/312] eta: 0:00:00 lr: 0.002769 min_lr: 0.002769 loss: 2.1745 (2.1964) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [180] Total time: 0:03:47 (0.7290 s / it) Averaged stats: lr: 0.002769 min_lr: 0.002769 loss: 2.1745 (2.1657) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6357 (0.6357) acc1: 83.5938 (83.5938) acc5: 96.3542 (96.3542) time: 4.5904 data: 4.3824 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8914 (0.8399) acc1: 77.6042 (77.4720) acc5: 94.5312 (94.4640) time: 0.6615 data: 0.4871 max mem: 64948 Test: Total time: 0:00:06 (0.6827 s / it) * Acc@1 78.296 Acc@5 94.370 loss 0.826 Accuracy of the model on the 50000 test images: 78.3% Max accuracy: 78.30% Test: [0/9] eta: 0:00:40 loss: 0.5666 (0.5666) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 4.4531 data: 4.2352 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7455 (0.7338) acc1: 80.2083 (79.6480) acc5: 96.0938 (95.5520) time: 0.6468 data: 0.4707 max mem: 64948 Test: Total time: 0:00:05 (0.6550 s / it) * Acc@1 80.754 Acc@5 95.516 loss 0.712 Accuracy of the model EMA on 50000 test images: 80.8% Max EMA accuracy: 80.75% Epoch: [181] [ 0/312] eta: 0:50:15 lr: 0.002769 min_lr: 0.002769 loss: 2.4271 (2.4271) weight_decay: 0.0500 (0.0500) time: 9.6658 data: 8.8478 max mem: 64948 Epoch: [181] [ 10/312] eta: 0:07:54 lr: 0.002769 min_lr: 0.002769 loss: 2.2654 (2.2240) weight_decay: 0.0500 (0.0500) time: 1.5705 data: 0.8047 max mem: 64948 Epoch: [181] [ 20/312] eta: 0:05:36 lr: 0.002768 min_lr: 0.002768 loss: 2.2654 (2.2161) weight_decay: 0.0500 (0.0500) time: 0.7268 data: 0.0004 max mem: 64948 Epoch: [181] [ 30/312] eta: 0:04:43 lr: 0.002768 min_lr: 0.002768 loss: 2.3254 (2.1963) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [181] [ 40/312] eta: 0:04:13 lr: 0.002767 min_lr: 0.002767 loss: 2.2775 (2.2115) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [181] [ 50/312] eta: 0:03:51 lr: 0.002767 min_lr: 0.002767 loss: 2.2689 (2.2133) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [181] [ 60/312] eta: 0:03:35 lr: 0.002766 min_lr: 0.002766 loss: 2.2302 (2.1854) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [181] [ 70/312] eta: 0:03:21 lr: 0.002766 min_lr: 0.002766 loss: 2.1879 (2.1809) weight_decay: 0.0500 (0.0500) time: 0.7009 data: 0.0004 max mem: 64948 Epoch: [181] [ 80/312] eta: 0:03:09 lr: 0.002766 min_lr: 0.002766 loss: 2.2276 (2.1896) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [181] [ 90/312] eta: 0:02:58 lr: 0.002765 min_lr: 0.002765 loss: 2.1668 (2.1860) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [181] [100/312] eta: 0:02:47 lr: 0.002765 min_lr: 0.002765 loss: 1.9131 (2.1606) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [181] [110/312] eta: 0:02:38 lr: 0.002764 min_lr: 0.002764 loss: 1.9680 (2.1599) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [181] [120/312] eta: 0:02:28 lr: 0.002764 min_lr: 0.002764 loss: 2.1648 (2.1580) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [181] [130/312] eta: 0:02:19 lr: 0.002763 min_lr: 0.002763 loss: 2.1829 (2.1636) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [181] [140/312] eta: 0:02:11 lr: 0.002763 min_lr: 0.002763 loss: 2.2933 (2.1662) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [181] [150/312] eta: 0:02:02 lr: 0.002763 min_lr: 0.002763 loss: 2.2905 (2.1724) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [181] [160/312] eta: 0:01:54 lr: 0.002762 min_lr: 0.002762 loss: 2.2905 (2.1729) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [181] [170/312] eta: 0:01:46 lr: 0.002762 min_lr: 0.002762 loss: 2.2477 (2.1709) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [181] [180/312] eta: 0:01:38 lr: 0.002761 min_lr: 0.002761 loss: 2.2477 (2.1774) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [181] [190/312] eta: 0:01:30 lr: 0.002761 min_lr: 0.002761 loss: 2.2859 (2.1745) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [181] [200/312] eta: 0:01:23 lr: 0.002760 min_lr: 0.002760 loss: 2.2783 (2.1782) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [181] [210/312] eta: 0:01:15 lr: 0.002760 min_lr: 0.002760 loss: 2.3060 (2.1831) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [181] [220/312] eta: 0:01:07 lr: 0.002760 min_lr: 0.002760 loss: 2.2911 (2.1816) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [181] [230/312] eta: 0:01:00 lr: 0.002759 min_lr: 0.002759 loss: 2.2459 (2.1794) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [181] [240/312] eta: 0:00:52 lr: 0.002759 min_lr: 0.002759 loss: 2.0911 (2.1768) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [181] [250/312] eta: 0:00:45 lr: 0.002758 min_lr: 0.002758 loss: 2.1295 (2.1775) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [181] [260/312] eta: 0:00:38 lr: 0.002758 min_lr: 0.002758 loss: 2.2780 (2.1773) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [181] [270/312] eta: 0:00:30 lr: 0.002757 min_lr: 0.002757 loss: 2.2258 (2.1729) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [181] [280/312] eta: 0:00:23 lr: 0.002757 min_lr: 0.002757 loss: 2.1297 (2.1715) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [181] [290/312] eta: 0:00:15 lr: 0.002756 min_lr: 0.002756 loss: 2.1297 (2.1679) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [181] [300/312] eta: 0:00:08 lr: 0.002756 min_lr: 0.002756 loss: 2.2265 (2.1703) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [181] [310/312] eta: 0:00:01 lr: 0.002756 min_lr: 0.002756 loss: 2.2388 (2.1728) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [181] [311/312] eta: 0:00:00 lr: 0.002756 min_lr: 0.002756 loss: 2.2388 (2.1716) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [181] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.002756 min_lr: 0.002756 loss: 2.2388 (2.1713) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7096 (0.7096) acc1: 81.7708 (81.7708) acc5: 95.5729 (95.5729) time: 4.5984 data: 4.3872 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9864 (0.9257) acc1: 76.5625 (75.7760) acc5: 94.7917 (93.4400) time: 0.6622 data: 0.4875 max mem: 64948 Test: Total time: 0:00:06 (0.6733 s / it) * Acc@1 77.112 Acc@5 93.690 loss 0.894 Accuracy of the model on the 50000 test images: 77.1% Max accuracy: 78.30% Test: [0/9] eta: 0:00:47 loss: 0.5652 (0.5652) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 5.3191 data: 5.1093 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7445 (0.7325) acc1: 80.2083 (79.6480) acc5: 96.0938 (95.6480) time: 0.7423 data: 0.5678 max mem: 64948 Test: Total time: 0:00:06 (0.7540 s / it) * Acc@1 80.744 Acc@5 95.526 loss 0.711 Accuracy of the model EMA on 50000 test images: 80.7% Epoch: [182] [ 0/312] eta: 0:53:34 lr: 0.002756 min_lr: 0.002756 loss: 2.5281 (2.5281) weight_decay: 0.0500 (0.0500) time: 10.3044 data: 8.0256 max mem: 64948 Epoch: [182] [ 10/312] eta: 0:08:13 lr: 0.002755 min_lr: 0.002755 loss: 2.2902 (2.2625) weight_decay: 0.0500 (0.0500) time: 1.6348 data: 0.7300 max mem: 64948 Epoch: [182] [ 20/312] eta: 0:05:46 lr: 0.002755 min_lr: 0.002755 loss: 2.2221 (2.1756) weight_decay: 0.0500 (0.0500) time: 0.7313 data: 0.0004 max mem: 64948 Epoch: [182] [ 30/312] eta: 0:04:49 lr: 0.002754 min_lr: 0.002754 loss: 2.1601 (2.1512) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [182] [ 40/312] eta: 0:04:17 lr: 0.002754 min_lr: 0.002754 loss: 2.2726 (2.1936) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [182] [ 50/312] eta: 0:03:55 lr: 0.002753 min_lr: 0.002753 loss: 2.2863 (2.1973) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [182] [ 60/312] eta: 0:03:37 lr: 0.002753 min_lr: 0.002753 loss: 2.2885 (2.2220) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [182] [ 70/312] eta: 0:03:23 lr: 0.002753 min_lr: 0.002753 loss: 2.3940 (2.2204) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [182] [ 80/312] eta: 0:03:10 lr: 0.002752 min_lr: 0.002752 loss: 2.3017 (2.2109) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [182] [ 90/312] eta: 0:02:59 lr: 0.002752 min_lr: 0.002752 loss: 2.0237 (2.1840) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [182] [100/312] eta: 0:02:48 lr: 0.002751 min_lr: 0.002751 loss: 2.1519 (2.1890) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [182] [110/312] eta: 0:02:39 lr: 0.002751 min_lr: 0.002751 loss: 2.1218 (2.1722) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [182] [120/312] eta: 0:02:29 lr: 0.002750 min_lr: 0.002750 loss: 2.1236 (2.1825) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [182] [130/312] eta: 0:02:20 lr: 0.002750 min_lr: 0.002750 loss: 2.2238 (2.1839) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0003 max mem: 64948 Epoch: [182] [140/312] eta: 0:02:11 lr: 0.002749 min_lr: 0.002749 loss: 2.0962 (2.1842) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0003 max mem: 64948 Epoch: [182] [150/312] eta: 0:02:03 lr: 0.002749 min_lr: 0.002749 loss: 2.0278 (2.1732) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [182] [160/312] eta: 0:01:55 lr: 0.002749 min_lr: 0.002749 loss: 2.0865 (2.1729) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [182] [170/312] eta: 0:01:47 lr: 0.002748 min_lr: 0.002748 loss: 2.1322 (2.1691) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [182] [180/312] eta: 0:01:39 lr: 0.002748 min_lr: 0.002748 loss: 2.1635 (2.1680) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [182] [190/312] eta: 0:01:31 lr: 0.002747 min_lr: 0.002747 loss: 2.2357 (2.1658) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [182] [200/312] eta: 0:01:23 lr: 0.002747 min_lr: 0.002747 loss: 2.2282 (2.1683) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [182] [210/312] eta: 0:01:15 lr: 0.002746 min_lr: 0.002746 loss: 2.1187 (2.1647) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [182] [220/312] eta: 0:01:08 lr: 0.002746 min_lr: 0.002746 loss: 2.0030 (2.1562) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [182] [230/312] eta: 0:01:00 lr: 0.002746 min_lr: 0.002746 loss: 2.2498 (2.1633) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [182] [240/312] eta: 0:00:53 lr: 0.002745 min_lr: 0.002745 loss: 2.2498 (2.1559) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [182] [250/312] eta: 0:00:45 lr: 0.002745 min_lr: 0.002745 loss: 2.0397 (2.1533) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [182] [260/312] eta: 0:00:38 lr: 0.002744 min_lr: 0.002744 loss: 2.1002 (2.1513) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [182] [270/312] eta: 0:00:30 lr: 0.002744 min_lr: 0.002744 loss: 2.2958 (2.1589) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [182] [280/312] eta: 0:00:23 lr: 0.002743 min_lr: 0.002743 loss: 2.3554 (2.1645) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [182] [290/312] eta: 0:00:16 lr: 0.002743 min_lr: 0.002743 loss: 2.3002 (2.1587) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [182] [300/312] eta: 0:00:08 lr: 0.002743 min_lr: 0.002743 loss: 1.9448 (2.1544) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [182] [310/312] eta: 0:00:01 lr: 0.002742 min_lr: 0.002742 loss: 2.1169 (2.1567) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [182] [311/312] eta: 0:00:00 lr: 0.002742 min_lr: 0.002742 loss: 2.1246 (2.1576) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [182] Total time: 0:03:47 (0.7303 s / it) Averaged stats: lr: 0.002742 min_lr: 0.002742 loss: 2.1246 (2.1591) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6406 (0.6406) acc1: 83.8542 (83.8542) acc5: 95.8333 (95.8333) time: 4.5551 data: 4.3354 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8757 (0.8456) acc1: 78.1250 (78.0480) acc5: 94.7917 (93.9200) time: 0.6574 data: 0.4818 max mem: 64948 Test: Total time: 0:00:06 (0.6853 s / it) * Acc@1 78.068 Acc@5 94.184 loss 0.848 Accuracy of the model on the 50000 test images: 78.1% Max accuracy: 78.30% Test: [0/9] eta: 0:00:40 loss: 0.5642 (0.5642) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 4.4788 data: 4.2607 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7441 (0.7314) acc1: 80.4688 (79.6800) acc5: 96.0938 (95.6800) time: 0.6494 data: 0.4735 max mem: 64948 Test: Total time: 0:00:05 (0.6597 s / it) * Acc@1 80.748 Acc@5 95.544 loss 0.710 Accuracy of the model EMA on 50000 test images: 80.7% Epoch: [183] [ 0/312] eta: 0:55:20 lr: 0.002742 min_lr: 0.002742 loss: 1.7819 (1.7819) weight_decay: 0.0500 (0.0500) time: 10.6413 data: 9.2545 max mem: 64948 Epoch: [183] [ 10/312] eta: 0:08:12 lr: 0.002742 min_lr: 0.002742 loss: 2.3084 (2.2130) weight_decay: 0.0500 (0.0500) time: 1.6302 data: 0.8418 max mem: 64948 Epoch: [183] [ 20/312] eta: 0:05:46 lr: 0.002741 min_lr: 0.002741 loss: 2.2759 (2.1299) weight_decay: 0.0500 (0.0500) time: 0.7139 data: 0.0004 max mem: 64948 Epoch: [183] [ 30/312] eta: 0:04:50 lr: 0.002741 min_lr: 0.002741 loss: 2.3245 (2.2183) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0003 max mem: 64948 Epoch: [183] [ 40/312] eta: 0:04:17 lr: 0.002740 min_lr: 0.002740 loss: 2.2508 (2.1303) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [183] [ 50/312] eta: 0:03:55 lr: 0.002740 min_lr: 0.002740 loss: 2.2288 (2.1655) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [183] [ 60/312] eta: 0:03:37 lr: 0.002739 min_lr: 0.002739 loss: 2.2288 (2.1602) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [183] [ 70/312] eta: 0:03:23 lr: 0.002739 min_lr: 0.002739 loss: 2.1343 (2.1656) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [183] [ 80/312] eta: 0:03:10 lr: 0.002739 min_lr: 0.002739 loss: 2.1343 (2.1479) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [183] [ 90/312] eta: 0:02:59 lr: 0.002738 min_lr: 0.002738 loss: 1.9008 (2.1275) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [183] [100/312] eta: 0:02:49 lr: 0.002738 min_lr: 0.002738 loss: 2.0187 (2.1278) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [183] [110/312] eta: 0:02:39 lr: 0.002737 min_lr: 0.002737 loss: 2.1793 (2.1331) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [183] [120/312] eta: 0:02:29 lr: 0.002737 min_lr: 0.002737 loss: 2.1793 (2.1292) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [183] [130/312] eta: 0:02:20 lr: 0.002736 min_lr: 0.002736 loss: 1.9178 (2.1152) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [183] [140/312] eta: 0:02:12 lr: 0.002736 min_lr: 0.002736 loss: 1.9004 (2.1114) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [183] [150/312] eta: 0:02:03 lr: 0.002735 min_lr: 0.002735 loss: 2.1769 (2.1045) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [183] [160/312] eta: 0:01:55 lr: 0.002735 min_lr: 0.002735 loss: 2.1815 (2.1068) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [183] [170/312] eta: 0:01:47 lr: 0.002735 min_lr: 0.002735 loss: 2.2694 (2.1121) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [183] [180/312] eta: 0:01:39 lr: 0.002734 min_lr: 0.002734 loss: 2.2760 (2.1162) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [183] [190/312] eta: 0:01:31 lr: 0.002734 min_lr: 0.002734 loss: 2.2484 (2.1186) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [183] [200/312] eta: 0:01:23 lr: 0.002733 min_lr: 0.002733 loss: 2.1472 (2.1145) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [183] [210/312] eta: 0:01:15 lr: 0.002733 min_lr: 0.002733 loss: 2.2804 (2.1244) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [183] [220/312] eta: 0:01:08 lr: 0.002732 min_lr: 0.002732 loss: 2.3188 (2.1308) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [183] [230/312] eta: 0:01:00 lr: 0.002732 min_lr: 0.002732 loss: 2.3188 (2.1398) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [183] [240/312] eta: 0:00:53 lr: 0.002732 min_lr: 0.002732 loss: 2.2991 (2.1370) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [183] [250/312] eta: 0:00:45 lr: 0.002731 min_lr: 0.002731 loss: 2.2224 (2.1392) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [183] [260/312] eta: 0:00:38 lr: 0.002731 min_lr: 0.002731 loss: 2.1399 (2.1362) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [183] [270/312] eta: 0:00:30 lr: 0.002730 min_lr: 0.002730 loss: 2.0841 (2.1359) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [183] [280/312] eta: 0:00:23 lr: 0.002730 min_lr: 0.002730 loss: 2.0453 (2.1354) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [183] [290/312] eta: 0:00:16 lr: 0.002729 min_lr: 0.002729 loss: 2.2733 (2.1416) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [183] [300/312] eta: 0:00:08 lr: 0.002729 min_lr: 0.002729 loss: 2.2519 (2.1355) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [183] [310/312] eta: 0:00:01 lr: 0.002728 min_lr: 0.002728 loss: 2.1528 (2.1357) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [183] [311/312] eta: 0:00:00 lr: 0.002728 min_lr: 0.002728 loss: 2.1528 (2.1370) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [183] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.002728 min_lr: 0.002728 loss: 2.1528 (2.1595) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6488 (0.6488) acc1: 84.3750 (84.3750) acc5: 95.5729 (95.5729) time: 4.4863 data: 4.2700 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8648 (0.8691) acc1: 77.3438 (77.0240) acc5: 94.5312 (93.7920) time: 0.6498 data: 0.4745 max mem: 64948 Test: Total time: 0:00:06 (0.6702 s / it) * Acc@1 77.634 Acc@5 93.966 loss 0.859 Accuracy of the model on the 50000 test images: 77.6% Max accuracy: 78.30% Test: [0/9] eta: 0:00:43 loss: 0.5629 (0.5629) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 4.7896 data: 4.5776 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7425 (0.7301) acc1: 80.4688 (79.8080) acc5: 96.0938 (95.7120) time: 0.6843 data: 0.5087 max mem: 64948 Test: Total time: 0:00:06 (0.6935 s / it) * Acc@1 80.784 Acc@5 95.556 loss 0.709 Accuracy of the model EMA on 50000 test images: 80.8% Max EMA accuracy: 80.78% Epoch: [184] [ 0/312] eta: 0:56:29 lr: 0.002728 min_lr: 0.002728 loss: 2.5653 (2.5653) weight_decay: 0.0500 (0.0500) time: 10.8641 data: 10.1332 max mem: 64948 Epoch: [184] [ 10/312] eta: 0:08:12 lr: 0.002728 min_lr: 0.002728 loss: 2.3458 (2.2964) weight_decay: 0.0500 (0.0500) time: 1.6297 data: 0.9215 max mem: 64948 Epoch: [184] [ 20/312] eta: 0:05:46 lr: 0.002728 min_lr: 0.002728 loss: 2.0889 (2.1399) weight_decay: 0.0500 (0.0500) time: 0.7012 data: 0.0003 max mem: 64948 Epoch: [184] [ 30/312] eta: 0:04:49 lr: 0.002727 min_lr: 0.002727 loss: 2.0417 (2.1371) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [184] [ 40/312] eta: 0:04:17 lr: 0.002727 min_lr: 0.002727 loss: 2.0660 (2.1505) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [184] [ 50/312] eta: 0:03:54 lr: 0.002726 min_lr: 0.002726 loss: 2.2172 (2.1454) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [184] [ 60/312] eta: 0:03:37 lr: 0.002726 min_lr: 0.002726 loss: 2.0894 (2.1014) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [184] [ 70/312] eta: 0:03:23 lr: 0.002725 min_lr: 0.002725 loss: 2.0430 (2.0999) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [184] [ 80/312] eta: 0:03:10 lr: 0.002725 min_lr: 0.002725 loss: 2.0939 (2.0978) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [184] [ 90/312] eta: 0:02:59 lr: 0.002724 min_lr: 0.002724 loss: 1.8869 (2.0913) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [184] [100/312] eta: 0:02:49 lr: 0.002724 min_lr: 0.002724 loss: 2.0398 (2.0941) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [184] [110/312] eta: 0:02:39 lr: 0.002724 min_lr: 0.002724 loss: 2.2481 (2.1088) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [184] [120/312] eta: 0:02:29 lr: 0.002723 min_lr: 0.002723 loss: 2.2665 (2.1154) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [184] [130/312] eta: 0:02:20 lr: 0.002723 min_lr: 0.002723 loss: 2.1779 (2.1202) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [184] [140/312] eta: 0:02:12 lr: 0.002722 min_lr: 0.002722 loss: 2.1779 (2.1128) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [184] [150/312] eta: 0:02:03 lr: 0.002722 min_lr: 0.002722 loss: 2.1667 (2.1167) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [184] [160/312] eta: 0:01:55 lr: 0.002721 min_lr: 0.002721 loss: 2.2041 (2.1209) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [184] [170/312] eta: 0:01:47 lr: 0.002721 min_lr: 0.002721 loss: 2.1919 (2.1238) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [184] [180/312] eta: 0:01:39 lr: 0.002721 min_lr: 0.002721 loss: 2.3134 (2.1383) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [184] [190/312] eta: 0:01:31 lr: 0.002720 min_lr: 0.002720 loss: 2.3318 (2.1419) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [184] [200/312] eta: 0:01:23 lr: 0.002720 min_lr: 0.002720 loss: 2.2812 (2.1520) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [184] [210/312] eta: 0:01:15 lr: 0.002719 min_lr: 0.002719 loss: 2.2868 (2.1582) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [184] [220/312] eta: 0:01:08 lr: 0.002719 min_lr: 0.002719 loss: 2.2143 (2.1579) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [184] [230/312] eta: 0:01:00 lr: 0.002718 min_lr: 0.002718 loss: 2.2544 (2.1632) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [184] [240/312] eta: 0:00:53 lr: 0.002718 min_lr: 0.002718 loss: 2.2608 (2.1627) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [184] [250/312] eta: 0:00:45 lr: 0.002717 min_lr: 0.002717 loss: 2.2608 (2.1617) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [184] [260/312] eta: 0:00:38 lr: 0.002717 min_lr: 0.002717 loss: 2.2285 (2.1600) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [184] [270/312] eta: 0:00:30 lr: 0.002717 min_lr: 0.002717 loss: 2.2375 (2.1617) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [184] [280/312] eta: 0:00:23 lr: 0.002716 min_lr: 0.002716 loss: 2.2697 (2.1671) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0009 max mem: 64948 Epoch: [184] [290/312] eta: 0:00:16 lr: 0.002716 min_lr: 0.002716 loss: 2.3123 (2.1694) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [184] [300/312] eta: 0:00:08 lr: 0.002715 min_lr: 0.002715 loss: 2.1336 (2.1638) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [184] [310/312] eta: 0:00:01 lr: 0.002715 min_lr: 0.002715 loss: 2.0037 (2.1585) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [184] [311/312] eta: 0:00:00 lr: 0.002715 min_lr: 0.002715 loss: 2.0037 (2.1596) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [184] Total time: 0:03:47 (0.7302 s / it) Averaged stats: lr: 0.002715 min_lr: 0.002715 loss: 2.0037 (2.1526) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6671 (0.6671) acc1: 82.8125 (82.8125) acc5: 94.7917 (94.7917) time: 4.6171 data: 4.4005 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8880 (0.8722) acc1: 79.1667 (77.1520) acc5: 94.7917 (93.7920) time: 0.6643 data: 0.4890 max mem: 64948 Test: Total time: 0:00:06 (0.6904 s / it) * Acc@1 77.968 Acc@5 94.010 loss 0.850 Accuracy of the model on the 50000 test images: 78.0% Max accuracy: 78.30% Test: [0/9] eta: 0:00:43 loss: 0.5619 (0.5619) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 4.8837 data: 4.6595 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7414 (0.7289) acc1: 80.7292 (79.8720) acc5: 96.0938 (95.7120) time: 0.6942 data: 0.5178 max mem: 64948 Test: Total time: 0:00:06 (0.7058 s / it) * Acc@1 80.832 Acc@5 95.562 loss 0.708 Accuracy of the model EMA on 50000 test images: 80.8% Max EMA accuracy: 80.83% Epoch: [185] [ 0/312] eta: 0:48:08 lr: 0.002715 min_lr: 0.002715 loss: 1.7018 (1.7018) weight_decay: 0.0500 (0.0500) time: 9.2590 data: 7.1845 max mem: 64948 Epoch: [185] [ 10/312] eta: 0:07:33 lr: 0.002714 min_lr: 0.002714 loss: 2.2076 (2.1455) weight_decay: 0.0500 (0.0500) time: 1.5021 data: 0.6539 max mem: 64948 Epoch: [185] [ 20/312] eta: 0:05:26 lr: 0.002714 min_lr: 0.002714 loss: 2.1295 (2.0583) weight_decay: 0.0500 (0.0500) time: 0.7104 data: 0.0006 max mem: 64948 Epoch: [185] [ 30/312] eta: 0:04:36 lr: 0.002713 min_lr: 0.002713 loss: 2.1295 (2.1116) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [185] [ 40/312] eta: 0:04:08 lr: 0.002713 min_lr: 0.002713 loss: 2.1987 (2.1103) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [185] [ 50/312] eta: 0:03:47 lr: 0.002713 min_lr: 0.002713 loss: 2.1769 (2.1414) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [185] [ 60/312] eta: 0:03:32 lr: 0.002712 min_lr: 0.002712 loss: 2.1769 (2.1408) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [185] [ 70/312] eta: 0:03:18 lr: 0.002712 min_lr: 0.002712 loss: 2.2545 (2.1591) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [185] [ 80/312] eta: 0:03:06 lr: 0.002711 min_lr: 0.002711 loss: 2.2912 (2.1664) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [185] [ 90/312] eta: 0:02:56 lr: 0.002711 min_lr: 0.002711 loss: 2.2413 (2.1577) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [185] [100/312] eta: 0:02:46 lr: 0.002710 min_lr: 0.002710 loss: 2.1322 (2.1536) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [185] [110/312] eta: 0:02:36 lr: 0.002710 min_lr: 0.002710 loss: 2.0456 (2.1485) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [185] [120/312] eta: 0:02:27 lr: 0.002710 min_lr: 0.002710 loss: 2.1567 (2.1393) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [185] [130/312] eta: 0:02:18 lr: 0.002709 min_lr: 0.002709 loss: 2.2196 (2.1509) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [185] [140/312] eta: 0:02:10 lr: 0.002709 min_lr: 0.002709 loss: 2.2343 (2.1391) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [185] [150/312] eta: 0:02:02 lr: 0.002708 min_lr: 0.002708 loss: 1.9977 (2.1268) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [185] [160/312] eta: 0:01:54 lr: 0.002708 min_lr: 0.002708 loss: 1.8753 (2.1121) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [185] [170/312] eta: 0:01:46 lr: 0.002707 min_lr: 0.002707 loss: 2.0328 (2.1224) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [185] [180/312] eta: 0:01:38 lr: 0.002707 min_lr: 0.002707 loss: 2.2298 (2.1207) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [185] [190/312] eta: 0:01:30 lr: 0.002706 min_lr: 0.002706 loss: 2.1388 (2.1193) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [185] [200/312] eta: 0:01:22 lr: 0.002706 min_lr: 0.002706 loss: 2.1160 (2.1164) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [185] [210/312] eta: 0:01:15 lr: 0.002706 min_lr: 0.002706 loss: 2.1160 (2.1126) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [185] [220/312] eta: 0:01:07 lr: 0.002705 min_lr: 0.002705 loss: 2.1242 (2.1141) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [185] [230/312] eta: 0:01:00 lr: 0.002705 min_lr: 0.002705 loss: 2.2306 (2.1227) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [185] [240/312] eta: 0:00:52 lr: 0.002704 min_lr: 0.002704 loss: 2.2882 (2.1260) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [185] [250/312] eta: 0:00:45 lr: 0.002704 min_lr: 0.002704 loss: 2.2372 (2.1284) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [185] [260/312] eta: 0:00:37 lr: 0.002703 min_lr: 0.002703 loss: 2.2372 (2.1285) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [185] [270/312] eta: 0:00:30 lr: 0.002703 min_lr: 0.002703 loss: 2.1531 (2.1289) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [185] [280/312] eta: 0:00:23 lr: 0.002703 min_lr: 0.002703 loss: 2.1164 (2.1271) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [185] [290/312] eta: 0:00:15 lr: 0.002702 min_lr: 0.002702 loss: 2.1755 (2.1299) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [185] [300/312] eta: 0:00:08 lr: 0.002702 min_lr: 0.002702 loss: 2.1861 (2.1300) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [185] [310/312] eta: 0:00:01 lr: 0.002701 min_lr: 0.002701 loss: 2.1692 (2.1304) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [185] [311/312] eta: 0:00:00 lr: 0.002701 min_lr: 0.002701 loss: 2.1692 (2.1309) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [185] Total time: 0:03:46 (0.7262 s / it) Averaged stats: lr: 0.002701 min_lr: 0.002701 loss: 2.1692 (2.1540) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6279 (0.6279) acc1: 83.5938 (83.5938) acc5: 97.3958 (97.3958) time: 4.6510 data: 4.4405 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9077 (0.8723) acc1: 78.1250 (77.1520) acc5: 94.2708 (94.1760) time: 0.6686 data: 0.4935 max mem: 64948 Test: Total time: 0:00:06 (0.6942 s / it) * Acc@1 78.132 Acc@5 94.132 loss 0.850 Accuracy of the model on the 50000 test images: 78.1% Max accuracy: 78.30% Test: [0/9] eta: 0:00:44 loss: 0.5606 (0.5606) acc1: 85.1562 (85.1562) acc5: 96.8750 (96.8750) time: 4.9904 data: 4.7723 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7397 (0.7277) acc1: 80.7292 (79.9040) acc5: 96.0938 (95.6800) time: 0.7062 data: 0.5304 max mem: 64948 Test: Total time: 0:00:06 (0.7141 s / it) * Acc@1 80.850 Acc@5 95.592 loss 0.707 Accuracy of the model EMA on 50000 test images: 80.9% Max EMA accuracy: 80.85% Epoch: [186] [ 0/312] eta: 0:49:21 lr: 0.002701 min_lr: 0.002701 loss: 2.6376 (2.6376) weight_decay: 0.0500 (0.0500) time: 9.4920 data: 7.4246 max mem: 64948 Epoch: [186] [ 10/312] eta: 0:07:38 lr: 0.002701 min_lr: 0.002701 loss: 2.2366 (2.3617) weight_decay: 0.0500 (0.0500) time: 1.5172 data: 0.6753 max mem: 64948 Epoch: [186] [ 20/312] eta: 0:05:28 lr: 0.002700 min_lr: 0.002700 loss: 2.3578 (2.3574) weight_decay: 0.0500 (0.0500) time: 0.7068 data: 0.0004 max mem: 64948 Epoch: [186] [ 30/312] eta: 0:04:37 lr: 0.002700 min_lr: 0.002700 loss: 2.3291 (2.3216) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [186] [ 40/312] eta: 0:04:09 lr: 0.002699 min_lr: 0.002699 loss: 2.2641 (2.3104) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [186] [ 50/312] eta: 0:03:48 lr: 0.002699 min_lr: 0.002699 loss: 2.0805 (2.2579) weight_decay: 0.0500 (0.0500) time: 0.7009 data: 0.0003 max mem: 64948 Epoch: [186] [ 60/312] eta: 0:03:32 lr: 0.002698 min_lr: 0.002698 loss: 2.0401 (2.2261) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [186] [ 70/312] eta: 0:03:19 lr: 0.002698 min_lr: 0.002698 loss: 2.1574 (2.2390) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [186] [ 80/312] eta: 0:03:07 lr: 0.002698 min_lr: 0.002698 loss: 2.1871 (2.2429) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [186] [ 90/312] eta: 0:02:56 lr: 0.002697 min_lr: 0.002697 loss: 2.0736 (2.2105) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [186] [100/312] eta: 0:02:46 lr: 0.002697 min_lr: 0.002697 loss: 2.1400 (2.2167) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [186] [110/312] eta: 0:02:36 lr: 0.002696 min_lr: 0.002696 loss: 2.2023 (2.2074) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [186] [120/312] eta: 0:02:27 lr: 0.002696 min_lr: 0.002696 loss: 1.8240 (2.1712) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [186] [130/312] eta: 0:02:19 lr: 0.002695 min_lr: 0.002695 loss: 1.9853 (2.1711) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [186] [140/312] eta: 0:02:10 lr: 0.002695 min_lr: 0.002695 loss: 2.0764 (2.1634) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [186] [150/312] eta: 0:02:02 lr: 0.002695 min_lr: 0.002695 loss: 2.0764 (2.1658) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [186] [160/312] eta: 0:01:54 lr: 0.002694 min_lr: 0.002694 loss: 2.3032 (2.1684) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [186] [170/312] eta: 0:01:46 lr: 0.002694 min_lr: 0.002694 loss: 2.3691 (2.1792) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [186] [180/312] eta: 0:01:38 lr: 0.002693 min_lr: 0.002693 loss: 2.2338 (2.1711) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [186] [190/312] eta: 0:01:30 lr: 0.002693 min_lr: 0.002693 loss: 2.0756 (2.1674) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [186] [200/312] eta: 0:01:22 lr: 0.002692 min_lr: 0.002692 loss: 2.2318 (2.1660) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [186] [210/312] eta: 0:01:15 lr: 0.002692 min_lr: 0.002692 loss: 2.2875 (2.1680) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [186] [220/312] eta: 0:01:07 lr: 0.002691 min_lr: 0.002691 loss: 2.3190 (2.1688) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [186] [230/312] eta: 0:01:00 lr: 0.002691 min_lr: 0.002691 loss: 2.2599 (2.1706) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [186] [240/312] eta: 0:00:52 lr: 0.002691 min_lr: 0.002691 loss: 2.3382 (2.1744) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [186] [250/312] eta: 0:00:45 lr: 0.002690 min_lr: 0.002690 loss: 2.4334 (2.1767) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [186] [260/312] eta: 0:00:37 lr: 0.002690 min_lr: 0.002690 loss: 1.9980 (2.1637) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [186] [270/312] eta: 0:00:30 lr: 0.002689 min_lr: 0.002689 loss: 2.1315 (2.1668) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [186] [280/312] eta: 0:00:23 lr: 0.002689 min_lr: 0.002689 loss: 2.2919 (2.1686) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [186] [290/312] eta: 0:00:15 lr: 0.002688 min_lr: 0.002688 loss: 2.3348 (2.1732) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [186] [300/312] eta: 0:00:08 lr: 0.002688 min_lr: 0.002688 loss: 2.1664 (2.1722) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [186] [310/312] eta: 0:00:01 lr: 0.002688 min_lr: 0.002688 loss: 2.1627 (2.1729) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [186] [311/312] eta: 0:00:00 lr: 0.002687 min_lr: 0.002687 loss: 2.1627 (2.1724) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [186] Total time: 0:03:46 (0.7260 s / it) Averaged stats: lr: 0.002687 min_lr: 0.002687 loss: 2.1627 (2.1537) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6531 (0.6531) acc1: 84.6354 (84.6354) acc5: 95.5729 (95.5729) time: 4.5195 data: 4.2995 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9135 (0.8820) acc1: 78.6458 (77.0240) acc5: 94.7917 (93.7920) time: 0.6534 data: 0.4778 max mem: 64948 Test: Total time: 0:00:06 (0.6775 s / it) * Acc@1 78.148 Acc@5 94.106 loss 0.859 Accuracy of the model on the 50000 test images: 78.1% Max accuracy: 78.30% Test: [0/9] eta: 0:00:43 loss: 0.5591 (0.5591) acc1: 85.1562 (85.1562) acc5: 96.8750 (96.8750) time: 4.7923 data: 4.5814 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7385 (0.7265) acc1: 80.9896 (79.9360) acc5: 96.0938 (95.7120) time: 0.6838 data: 0.5092 max mem: 64948 Test: Total time: 0:00:06 (0.6912 s / it) * Acc@1 80.870 Acc@5 95.604 loss 0.706 Accuracy of the model EMA on 50000 test images: 80.9% Max EMA accuracy: 80.87% Epoch: [187] [ 0/312] eta: 0:47:39 lr: 0.002687 min_lr: 0.002687 loss: 2.0888 (2.0888) weight_decay: 0.0500 (0.0500) time: 9.1663 data: 7.9925 max mem: 64948 Epoch: [187] [ 10/312] eta: 0:07:31 lr: 0.002687 min_lr: 0.002687 loss: 2.0888 (2.0641) weight_decay: 0.0500 (0.0500) time: 1.4935 data: 0.7270 max mem: 64948 Epoch: [187] [ 20/312] eta: 0:05:25 lr: 0.002687 min_lr: 0.002687 loss: 1.9651 (2.0478) weight_decay: 0.0500 (0.0500) time: 0.7108 data: 0.0004 max mem: 64948 Epoch: [187] [ 30/312] eta: 0:04:35 lr: 0.002686 min_lr: 0.002686 loss: 2.0049 (2.0423) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [187] [ 40/312] eta: 0:04:07 lr: 0.002686 min_lr: 0.002686 loss: 2.1040 (2.0715) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [187] [ 50/312] eta: 0:03:47 lr: 0.002685 min_lr: 0.002685 loss: 2.0015 (2.0612) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [187] [ 60/312] eta: 0:03:31 lr: 0.002685 min_lr: 0.002685 loss: 1.9927 (2.0624) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [187] [ 70/312] eta: 0:03:18 lr: 0.002684 min_lr: 0.002684 loss: 2.0712 (2.0692) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [187] [ 80/312] eta: 0:03:06 lr: 0.002684 min_lr: 0.002684 loss: 2.0432 (2.0613) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [187] [ 90/312] eta: 0:02:55 lr: 0.002683 min_lr: 0.002683 loss: 2.2107 (2.1047) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [187] [100/312] eta: 0:02:45 lr: 0.002683 min_lr: 0.002683 loss: 2.3729 (2.1158) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [187] [110/312] eta: 0:02:36 lr: 0.002683 min_lr: 0.002683 loss: 2.1260 (2.1059) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [187] [120/312] eta: 0:02:27 lr: 0.002682 min_lr: 0.002682 loss: 2.2170 (2.1251) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [187] [130/312] eta: 0:02:18 lr: 0.002682 min_lr: 0.002682 loss: 2.3203 (2.1378) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [187] [140/312] eta: 0:02:10 lr: 0.002681 min_lr: 0.002681 loss: 2.2766 (2.1322) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [187] [150/312] eta: 0:02:02 lr: 0.002681 min_lr: 0.002681 loss: 2.2561 (2.1433) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [187] [160/312] eta: 0:01:54 lr: 0.002680 min_lr: 0.002680 loss: 2.2561 (2.1453) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [187] [170/312] eta: 0:01:46 lr: 0.002680 min_lr: 0.002680 loss: 2.1872 (2.1482) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [187] [180/312] eta: 0:01:38 lr: 0.002679 min_lr: 0.002679 loss: 2.1406 (2.1427) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [187] [190/312] eta: 0:01:30 lr: 0.002679 min_lr: 0.002679 loss: 2.1086 (2.1396) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [187] [200/312] eta: 0:01:22 lr: 0.002679 min_lr: 0.002679 loss: 2.1041 (2.1382) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [187] [210/312] eta: 0:01:15 lr: 0.002678 min_lr: 0.002678 loss: 2.0724 (2.1289) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [187] [220/312] eta: 0:01:07 lr: 0.002678 min_lr: 0.002678 loss: 2.0002 (2.1256) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [187] [230/312] eta: 0:01:00 lr: 0.002677 min_lr: 0.002677 loss: 2.0653 (2.1263) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [187] [240/312] eta: 0:00:52 lr: 0.002677 min_lr: 0.002677 loss: 2.1855 (2.1269) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [187] [250/312] eta: 0:00:45 lr: 0.002676 min_lr: 0.002676 loss: 2.1798 (2.1228) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [187] [260/312] eta: 0:00:37 lr: 0.002676 min_lr: 0.002676 loss: 1.9038 (2.1192) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [187] [270/312] eta: 0:00:30 lr: 0.002676 min_lr: 0.002676 loss: 2.1962 (2.1245) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [187] [280/312] eta: 0:00:23 lr: 0.002675 min_lr: 0.002675 loss: 2.1962 (2.1150) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [187] [290/312] eta: 0:00:15 lr: 0.002675 min_lr: 0.002675 loss: 1.8172 (2.1171) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0008 max mem: 64948 Epoch: [187] [300/312] eta: 0:00:08 lr: 0.002674 min_lr: 0.002674 loss: 2.1174 (2.1177) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [187] [310/312] eta: 0:00:01 lr: 0.002674 min_lr: 0.002674 loss: 2.0940 (2.1176) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [187] [311/312] eta: 0:00:00 lr: 0.002674 min_lr: 0.002674 loss: 2.0940 (2.1185) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [187] Total time: 0:03:46 (0.7254 s / it) Averaged stats: lr: 0.002674 min_lr: 0.002674 loss: 2.0940 (2.1522) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6882 (0.6882) acc1: 83.5938 (83.5938) acc5: 95.0521 (95.0521) time: 4.6000 data: 4.3922 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9384 (0.8980) acc1: 76.8229 (76.3200) acc5: 94.7917 (93.9200) time: 0.6624 data: 0.4881 max mem: 64948 Test: Total time: 0:00:06 (0.6869 s / it) * Acc@1 77.702 Acc@5 93.910 loss 0.864 Accuracy of the model on the 50000 test images: 77.7% Max accuracy: 78.30% Test: [0/9] eta: 0:00:42 loss: 0.5575 (0.5575) acc1: 85.1562 (85.1562) acc5: 96.8750 (96.8750) time: 4.6860 data: 4.4748 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7365 (0.7254) acc1: 80.7292 (80.0000) acc5: 96.0938 (95.7760) time: 0.6720 data: 0.4973 max mem: 64948 Test: Total time: 0:00:06 (0.6820 s / it) * Acc@1 80.878 Acc@5 95.618 loss 0.705 Accuracy of the model EMA on 50000 test images: 80.9% Max EMA accuracy: 80.88% Epoch: [188] [ 0/312] eta: 0:52:40 lr: 0.002674 min_lr: 0.002674 loss: 2.2907 (2.2907) weight_decay: 0.0500 (0.0500) time: 10.1289 data: 9.3709 max mem: 64948 Epoch: [188] [ 10/312] eta: 0:08:04 lr: 0.002673 min_lr: 0.002673 loss: 2.2373 (2.1636) weight_decay: 0.0500 (0.0500) time: 1.6032 data: 0.8523 max mem: 64948 Epoch: [188] [ 20/312] eta: 0:05:41 lr: 0.002673 min_lr: 0.002673 loss: 2.3013 (2.2549) weight_decay: 0.0500 (0.0500) time: 0.7228 data: 0.0004 max mem: 64948 Epoch: [188] [ 30/312] eta: 0:04:47 lr: 0.002672 min_lr: 0.002672 loss: 2.3013 (2.2255) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0003 max mem: 64948 Epoch: [188] [ 40/312] eta: 0:04:16 lr: 0.002672 min_lr: 0.002672 loss: 2.2465 (2.1984) weight_decay: 0.0500 (0.0500) time: 0.7006 data: 0.0004 max mem: 64948 Epoch: [188] [ 50/312] eta: 0:03:54 lr: 0.002671 min_lr: 0.002671 loss: 2.2465 (2.2031) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [188] [ 60/312] eta: 0:03:36 lr: 0.002671 min_lr: 0.002671 loss: 2.1753 (2.1932) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [188] [ 70/312] eta: 0:03:22 lr: 0.002671 min_lr: 0.002671 loss: 2.1753 (2.1766) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [188] [ 80/312] eta: 0:03:10 lr: 0.002670 min_lr: 0.002670 loss: 2.0535 (2.1553) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [188] [ 90/312] eta: 0:02:58 lr: 0.002670 min_lr: 0.002670 loss: 2.0535 (2.1594) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [188] [100/312] eta: 0:02:48 lr: 0.002669 min_lr: 0.002669 loss: 2.2896 (2.1716) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [188] [110/312] eta: 0:02:38 lr: 0.002669 min_lr: 0.002669 loss: 2.2841 (2.1684) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [188] [120/312] eta: 0:02:29 lr: 0.002668 min_lr: 0.002668 loss: 2.1418 (2.1697) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [188] [130/312] eta: 0:02:20 lr: 0.002668 min_lr: 0.002668 loss: 2.2325 (2.1754) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [188] [140/312] eta: 0:02:11 lr: 0.002668 min_lr: 0.002668 loss: 2.3162 (2.1822) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [188] [150/312] eta: 0:02:03 lr: 0.002667 min_lr: 0.002667 loss: 2.3450 (2.1820) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [188] [160/312] eta: 0:01:55 lr: 0.002667 min_lr: 0.002667 loss: 1.9970 (2.1676) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [188] [170/312] eta: 0:01:47 lr: 0.002666 min_lr: 0.002666 loss: 2.0016 (2.1667) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [188] [180/312] eta: 0:01:39 lr: 0.002666 min_lr: 0.002666 loss: 2.0696 (2.1561) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [188] [190/312] eta: 0:01:31 lr: 0.002665 min_lr: 0.002665 loss: 2.0757 (2.1536) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [188] [200/312] eta: 0:01:23 lr: 0.002665 min_lr: 0.002665 loss: 2.1078 (2.1530) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [188] [210/312] eta: 0:01:15 lr: 0.002664 min_lr: 0.002664 loss: 2.2104 (2.1597) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [188] [220/312] eta: 0:01:08 lr: 0.002664 min_lr: 0.002664 loss: 2.2115 (2.1649) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [188] [230/312] eta: 0:01:00 lr: 0.002664 min_lr: 0.002664 loss: 2.2428 (2.1679) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [188] [240/312] eta: 0:00:53 lr: 0.002663 min_lr: 0.002663 loss: 2.2194 (2.1636) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [188] [250/312] eta: 0:00:45 lr: 0.002663 min_lr: 0.002663 loss: 1.9687 (2.1589) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [188] [260/312] eta: 0:00:38 lr: 0.002662 min_lr: 0.002662 loss: 1.9687 (2.1519) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [188] [270/312] eta: 0:00:30 lr: 0.002662 min_lr: 0.002662 loss: 2.1103 (2.1577) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [188] [280/312] eta: 0:00:23 lr: 0.002661 min_lr: 0.002661 loss: 2.2328 (2.1556) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [188] [290/312] eta: 0:00:16 lr: 0.002661 min_lr: 0.002661 loss: 1.9931 (2.1489) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [188] [300/312] eta: 0:00:08 lr: 0.002660 min_lr: 0.002660 loss: 2.2058 (2.1555) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [188] [310/312] eta: 0:00:01 lr: 0.002660 min_lr: 0.002660 loss: 2.3346 (2.1605) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [188] [311/312] eta: 0:00:00 lr: 0.002660 min_lr: 0.002660 loss: 2.3631 (2.1611) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [188] Total time: 0:03:47 (0.7295 s / it) Averaged stats: lr: 0.002660 min_lr: 0.002660 loss: 2.3631 (2.1535) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6794 (0.6794) acc1: 81.5104 (81.5104) acc5: 96.8750 (96.8750) time: 4.5972 data: 4.3860 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9080 (0.8904) acc1: 77.8646 (76.2560) acc5: 95.0521 (94.4000) time: 0.6620 data: 0.4874 max mem: 64948 Test: Total time: 0:00:06 (0.6850 s / it) * Acc@1 77.904 Acc@5 94.110 loss 0.854 Accuracy of the model on the 50000 test images: 77.9% Max accuracy: 78.30% Test: [0/9] eta: 0:00:44 loss: 0.5561 (0.5561) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 4.9617 data: 4.7522 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7349 (0.7243) acc1: 80.7292 (80.0000) acc5: 96.0938 (95.8080) time: 0.7026 data: 0.5281 max mem: 64948 Test: Total time: 0:00:06 (0.7161 s / it) * Acc@1 80.934 Acc@5 95.632 loss 0.704 Accuracy of the model EMA on 50000 test images: 80.9% Max EMA accuracy: 80.93% Epoch: [189] [ 0/312] eta: 0:45:47 lr: 0.002660 min_lr: 0.002660 loss: 1.8420 (1.8420) weight_decay: 0.0500 (0.0500) time: 8.8073 data: 7.7298 max mem: 64948 Epoch: [189] [ 10/312] eta: 0:07:31 lr: 0.002659 min_lr: 0.002659 loss: 1.8420 (2.0302) weight_decay: 0.0500 (0.0500) time: 1.4943 data: 0.7409 max mem: 64948 Epoch: [189] [ 20/312] eta: 0:05:25 lr: 0.002659 min_lr: 0.002659 loss: 2.1078 (2.1423) weight_decay: 0.0500 (0.0500) time: 0.7301 data: 0.0212 max mem: 64948 Epoch: [189] [ 30/312] eta: 0:04:36 lr: 0.002659 min_lr: 0.002659 loss: 2.1459 (2.1177) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [189] [ 40/312] eta: 0:04:07 lr: 0.002658 min_lr: 0.002658 loss: 2.1627 (2.1205) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [189] [ 50/312] eta: 0:03:47 lr: 0.002658 min_lr: 0.002658 loss: 2.1435 (2.1180) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [189] [ 60/312] eta: 0:03:31 lr: 0.002657 min_lr: 0.002657 loss: 2.0839 (2.1056) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [189] [ 70/312] eta: 0:03:18 lr: 0.002657 min_lr: 0.002657 loss: 2.1673 (2.1098) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [189] [ 80/312] eta: 0:03:06 lr: 0.002656 min_lr: 0.002656 loss: 2.1782 (2.1216) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [189] [ 90/312] eta: 0:02:55 lr: 0.002656 min_lr: 0.002656 loss: 2.2293 (2.1029) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [189] [100/312] eta: 0:02:46 lr: 0.002655 min_lr: 0.002655 loss: 2.2020 (2.1115) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [189] [110/312] eta: 0:02:36 lr: 0.002655 min_lr: 0.002655 loss: 2.2897 (2.1286) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [189] [120/312] eta: 0:02:27 lr: 0.002655 min_lr: 0.002655 loss: 2.3044 (2.1272) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [189] [130/312] eta: 0:02:18 lr: 0.002654 min_lr: 0.002654 loss: 2.2521 (2.1229) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [189] [140/312] eta: 0:02:10 lr: 0.002654 min_lr: 0.002654 loss: 2.1057 (2.1251) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [189] [150/312] eta: 0:02:02 lr: 0.002653 min_lr: 0.002653 loss: 2.2301 (2.1377) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [189] [160/312] eta: 0:01:54 lr: 0.002653 min_lr: 0.002653 loss: 2.3241 (2.1438) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [189] [170/312] eta: 0:01:46 lr: 0.002652 min_lr: 0.002652 loss: 2.1840 (2.1384) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [189] [180/312] eta: 0:01:38 lr: 0.002652 min_lr: 0.002652 loss: 1.9537 (2.1326) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [189] [190/312] eta: 0:01:30 lr: 0.002652 min_lr: 0.002652 loss: 2.2204 (2.1397) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [189] [200/312] eta: 0:01:22 lr: 0.002651 min_lr: 0.002651 loss: 2.2342 (2.1398) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [189] [210/312] eta: 0:01:15 lr: 0.002651 min_lr: 0.002651 loss: 2.1513 (2.1429) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [189] [220/312] eta: 0:01:07 lr: 0.002650 min_lr: 0.002650 loss: 2.0923 (2.1362) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [189] [230/312] eta: 0:01:00 lr: 0.002650 min_lr: 0.002650 loss: 2.1752 (2.1413) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [189] [240/312] eta: 0:00:52 lr: 0.002649 min_lr: 0.002649 loss: 2.3769 (2.1475) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [189] [250/312] eta: 0:00:45 lr: 0.002649 min_lr: 0.002649 loss: 2.3389 (2.1525) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [189] [260/312] eta: 0:00:37 lr: 0.002648 min_lr: 0.002648 loss: 2.2760 (2.1550) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [189] [270/312] eta: 0:00:30 lr: 0.002648 min_lr: 0.002648 loss: 2.0813 (2.1569) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [189] [280/312] eta: 0:00:23 lr: 0.002648 min_lr: 0.002648 loss: 2.0994 (2.1529) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [189] [290/312] eta: 0:00:15 lr: 0.002647 min_lr: 0.002647 loss: 2.0926 (2.1496) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [189] [300/312] eta: 0:00:08 lr: 0.002647 min_lr: 0.002647 loss: 2.1108 (2.1501) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [189] [310/312] eta: 0:00:01 lr: 0.002646 min_lr: 0.002646 loss: 2.1082 (2.1445) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [189] [311/312] eta: 0:00:00 lr: 0.002646 min_lr: 0.002646 loss: 2.1082 (2.1449) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [189] Total time: 0:03:46 (0.7258 s / it) Averaged stats: lr: 0.002646 min_lr: 0.002646 loss: 2.1082 (2.1487) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6275 (0.6275) acc1: 83.0729 (83.0729) acc5: 96.6146 (96.6146) time: 4.5080 data: 4.2938 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9321 (0.8597) acc1: 77.0833 (76.7360) acc5: 94.7917 (94.1440) time: 0.6521 data: 0.4772 max mem: 64948 Test: Total time: 0:00:06 (0.6806 s / it) * Acc@1 77.886 Acc@5 94.174 loss 0.844 Accuracy of the model on the 50000 test images: 77.9% Max accuracy: 78.30% Test: [0/9] eta: 0:00:45 loss: 0.5546 (0.5546) acc1: 85.1562 (85.1562) acc5: 97.3958 (97.3958) time: 5.0779 data: 4.8602 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7335 (0.7232) acc1: 80.7292 (79.9680) acc5: 96.0938 (95.8400) time: 0.7157 data: 0.5401 max mem: 64948 Test: Total time: 0:00:06 (0.7240 s / it) * Acc@1 80.966 Acc@5 95.638 loss 0.703 Accuracy of the model EMA on 50000 test images: 81.0% Max EMA accuracy: 80.97% Epoch: [190] [ 0/312] eta: 0:49:25 lr: 0.002646 min_lr: 0.002646 loss: 2.2537 (2.2537) weight_decay: 0.0500 (0.0500) time: 9.5050 data: 8.7261 max mem: 64948 Epoch: [190] [ 10/312] eta: 0:07:41 lr: 0.002646 min_lr: 0.002646 loss: 2.2537 (2.0504) weight_decay: 0.0500 (0.0500) time: 1.5269 data: 0.7936 max mem: 64948 Epoch: [190] [ 20/312] eta: 0:05:30 lr: 0.002645 min_lr: 0.002645 loss: 2.0372 (2.0324) weight_decay: 0.0500 (0.0500) time: 0.7138 data: 0.0003 max mem: 64948 Epoch: [190] [ 30/312] eta: 0:04:39 lr: 0.002645 min_lr: 0.002645 loss: 2.1328 (2.0972) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [190] [ 40/312] eta: 0:04:10 lr: 0.002644 min_lr: 0.002644 loss: 2.2101 (2.1283) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [190] [ 50/312] eta: 0:03:49 lr: 0.002644 min_lr: 0.002644 loss: 2.2204 (2.1332) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [190] [ 60/312] eta: 0:03:33 lr: 0.002643 min_lr: 0.002643 loss: 2.0961 (2.1287) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [190] [ 70/312] eta: 0:03:19 lr: 0.002643 min_lr: 0.002643 loss: 2.3485 (2.1602) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [190] [ 80/312] eta: 0:03:07 lr: 0.002643 min_lr: 0.002643 loss: 2.3636 (2.1715) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [190] [ 90/312] eta: 0:02:56 lr: 0.002642 min_lr: 0.002642 loss: 2.3509 (2.1733) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [190] [100/312] eta: 0:02:46 lr: 0.002642 min_lr: 0.002642 loss: 2.2985 (2.1790) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [190] [110/312] eta: 0:02:37 lr: 0.002641 min_lr: 0.002641 loss: 2.2892 (2.1851) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [190] [120/312] eta: 0:02:28 lr: 0.002641 min_lr: 0.002641 loss: 2.2794 (2.1809) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [190] [130/312] eta: 0:02:19 lr: 0.002640 min_lr: 0.002640 loss: 2.1569 (2.1755) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [190] [140/312] eta: 0:02:10 lr: 0.002640 min_lr: 0.002640 loss: 2.1679 (2.1785) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [190] [150/312] eta: 0:02:02 lr: 0.002639 min_lr: 0.002639 loss: 2.2164 (2.1791) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [190] [160/312] eta: 0:01:54 lr: 0.002639 min_lr: 0.002639 loss: 2.1463 (2.1768) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [190] [170/312] eta: 0:01:46 lr: 0.002639 min_lr: 0.002639 loss: 2.1050 (2.1799) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [190] [180/312] eta: 0:01:38 lr: 0.002638 min_lr: 0.002638 loss: 2.0753 (2.1669) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [190] [190/312] eta: 0:01:30 lr: 0.002638 min_lr: 0.002638 loss: 1.8869 (2.1562) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [190] [200/312] eta: 0:01:22 lr: 0.002637 min_lr: 0.002637 loss: 2.1245 (2.1531) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [190] [210/312] eta: 0:01:15 lr: 0.002637 min_lr: 0.002637 loss: 2.2376 (2.1585) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [190] [220/312] eta: 0:01:07 lr: 0.002636 min_lr: 0.002636 loss: 2.2376 (2.1604) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [190] [230/312] eta: 0:01:00 lr: 0.002636 min_lr: 0.002636 loss: 2.1783 (2.1606) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [190] [240/312] eta: 0:00:52 lr: 0.002635 min_lr: 0.002635 loss: 2.1582 (2.1657) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [190] [250/312] eta: 0:00:45 lr: 0.002635 min_lr: 0.002635 loss: 2.4358 (2.1739) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [190] [260/312] eta: 0:00:37 lr: 0.002635 min_lr: 0.002635 loss: 2.1683 (2.1676) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [190] [270/312] eta: 0:00:30 lr: 0.002634 min_lr: 0.002634 loss: 2.0624 (2.1666) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [190] [280/312] eta: 0:00:23 lr: 0.002634 min_lr: 0.002634 loss: 2.3017 (2.1718) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [190] [290/312] eta: 0:00:15 lr: 0.002633 min_lr: 0.002633 loss: 2.2390 (2.1683) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [190] [300/312] eta: 0:00:08 lr: 0.002633 min_lr: 0.002633 loss: 2.1118 (2.1691) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [190] [310/312] eta: 0:00:01 lr: 0.002632 min_lr: 0.002632 loss: 2.1118 (2.1684) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [190] [311/312] eta: 0:00:00 lr: 0.002632 min_lr: 0.002632 loss: 2.1118 (2.1689) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [190] Total time: 0:03:46 (0.7266 s / it) Averaged stats: lr: 0.002632 min_lr: 0.002632 loss: 2.1118 (2.1382) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.6393 (0.6393) acc1: 82.8125 (82.8125) acc5: 95.8333 (95.8333) time: 4.4433 data: 4.2252 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9017 (0.8579) acc1: 77.3438 (76.7040) acc5: 95.3125 (94.1760) time: 0.6451 data: 0.4696 max mem: 64948 Test: Total time: 0:00:06 (0.6690 s / it) * Acc@1 78.130 Acc@5 94.068 loss 0.843 Accuracy of the model on the 50000 test images: 78.1% Max accuracy: 78.30% Test: [0/9] eta: 0:00:43 loss: 0.5532 (0.5532) acc1: 85.1562 (85.1562) acc5: 97.3958 (97.3958) time: 4.7888 data: 4.5810 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7323 (0.7222) acc1: 80.4688 (79.8720) acc5: 96.3542 (95.8080) time: 0.6842 data: 0.5091 max mem: 64948 Test: Total time: 0:00:06 (0.7028 s / it) * Acc@1 80.958 Acc@5 95.638 loss 0.702 Accuracy of the model EMA on 50000 test images: 81.0% Epoch: [191] [ 0/312] eta: 0:58:15 lr: 0.002632 min_lr: 0.002632 loss: 2.1775 (2.1775) weight_decay: 0.0500 (0.0500) time: 11.2036 data: 7.8012 max mem: 64948 Epoch: [191] [ 10/312] eta: 0:08:24 lr: 0.002632 min_lr: 0.002632 loss: 2.2250 (2.1859) weight_decay: 0.0500 (0.0500) time: 1.6695 data: 0.7096 max mem: 64948 Epoch: [191] [ 20/312] eta: 0:05:51 lr: 0.002631 min_lr: 0.002631 loss: 2.2174 (2.0810) weight_decay: 0.0500 (0.0500) time: 0.7047 data: 0.0004 max mem: 64948 Epoch: [191] [ 30/312] eta: 0:04:53 lr: 0.002631 min_lr: 0.002631 loss: 2.1965 (2.1383) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [191] [ 40/312] eta: 0:04:19 lr: 0.002630 min_lr: 0.002630 loss: 2.3687 (2.1802) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [191] [ 50/312] eta: 0:03:57 lr: 0.002630 min_lr: 0.002630 loss: 2.0492 (2.1426) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [191] [ 60/312] eta: 0:03:39 lr: 0.002630 min_lr: 0.002630 loss: 2.0064 (2.1206) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [191] [ 70/312] eta: 0:03:24 lr: 0.002629 min_lr: 0.002629 loss: 1.9539 (2.0950) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [191] [ 80/312] eta: 0:03:11 lr: 0.002629 min_lr: 0.002629 loss: 1.9640 (2.1014) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [191] [ 90/312] eta: 0:03:00 lr: 0.002628 min_lr: 0.002628 loss: 2.2224 (2.1089) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [191] [100/312] eta: 0:02:50 lr: 0.002628 min_lr: 0.002628 loss: 1.9439 (2.0891) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [191] [110/312] eta: 0:02:40 lr: 0.002627 min_lr: 0.002627 loss: 1.9244 (2.0866) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [191] [120/312] eta: 0:02:30 lr: 0.002627 min_lr: 0.002627 loss: 2.0433 (2.0873) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [191] [130/312] eta: 0:02:21 lr: 0.002626 min_lr: 0.002626 loss: 2.1339 (2.0894) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [191] [140/312] eta: 0:02:12 lr: 0.002626 min_lr: 0.002626 loss: 2.1339 (2.0793) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [191] [150/312] eta: 0:02:04 lr: 0.002626 min_lr: 0.002626 loss: 2.1805 (2.0891) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [191] [160/312] eta: 0:01:55 lr: 0.002625 min_lr: 0.002625 loss: 2.3279 (2.1013) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [191] [170/312] eta: 0:01:47 lr: 0.002625 min_lr: 0.002625 loss: 2.1931 (2.1002) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [191] [180/312] eta: 0:01:39 lr: 0.002624 min_lr: 0.002624 loss: 2.0584 (2.0995) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [191] [190/312] eta: 0:01:31 lr: 0.002624 min_lr: 0.002624 loss: 2.0584 (2.1021) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [191] [200/312] eta: 0:01:23 lr: 0.002623 min_lr: 0.002623 loss: 2.1558 (2.1003) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [191] [210/312] eta: 0:01:16 lr: 0.002623 min_lr: 0.002623 loss: 2.2760 (2.1083) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [191] [220/312] eta: 0:01:08 lr: 0.002622 min_lr: 0.002622 loss: 2.4131 (2.1170) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [191] [230/312] eta: 0:01:00 lr: 0.002622 min_lr: 0.002622 loss: 2.2040 (2.1173) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [191] [240/312] eta: 0:00:53 lr: 0.002622 min_lr: 0.002622 loss: 2.2040 (2.1192) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [191] [250/312] eta: 0:00:45 lr: 0.002621 min_lr: 0.002621 loss: 2.1868 (2.1178) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [191] [260/312] eta: 0:00:38 lr: 0.002621 min_lr: 0.002621 loss: 2.1270 (2.1134) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [191] [270/312] eta: 0:00:30 lr: 0.002620 min_lr: 0.002620 loss: 2.2003 (2.1206) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [191] [280/312] eta: 0:00:23 lr: 0.002620 min_lr: 0.002620 loss: 2.3080 (2.1258) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [191] [290/312] eta: 0:00:16 lr: 0.002619 min_lr: 0.002619 loss: 2.1800 (2.1231) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [191] [300/312] eta: 0:00:08 lr: 0.002619 min_lr: 0.002619 loss: 2.1496 (2.1234) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [191] [310/312] eta: 0:00:01 lr: 0.002618 min_lr: 0.002618 loss: 2.2082 (2.1257) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [191] [311/312] eta: 0:00:00 lr: 0.002618 min_lr: 0.002618 loss: 2.2166 (2.1273) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [191] Total time: 0:03:48 (0.7324 s / it) Averaged stats: lr: 0.002618 min_lr: 0.002618 loss: 2.2166 (2.1463) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6777 (0.6777) acc1: 82.8125 (82.8125) acc5: 94.7917 (94.7917) time: 4.6329 data: 4.4192 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8326 (0.8681) acc1: 78.3854 (77.3440) acc5: 94.2708 (93.7600) time: 0.6661 data: 0.4911 max mem: 64948 Test: Total time: 0:00:06 (0.6914 s / it) * Acc@1 78.126 Acc@5 94.166 loss 0.843 Accuracy of the model on the 50000 test images: 78.1% Max accuracy: 78.30% Test: [0/9] eta: 0:00:45 loss: 0.5516 (0.5516) acc1: 85.1562 (85.1562) acc5: 97.3958 (97.3958) time: 5.0357 data: 4.8176 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7317 (0.7212) acc1: 80.7292 (79.8720) acc5: 96.3542 (95.8080) time: 0.7116 data: 0.5354 max mem: 64948 Test: Total time: 0:00:06 (0.7249 s / it) * Acc@1 80.976 Acc@5 95.634 loss 0.701 Accuracy of the model EMA on 50000 test images: 81.0% Max EMA accuracy: 80.98% Epoch: [192] [ 0/312] eta: 0:47:18 lr: 0.002618 min_lr: 0.002618 loss: 1.5576 (1.5576) weight_decay: 0.0500 (0.0500) time: 9.0966 data: 7.9074 max mem: 64948 Epoch: [192] [ 10/312] eta: 0:07:33 lr: 0.002618 min_lr: 0.002618 loss: 2.0167 (1.9849) weight_decay: 0.0500 (0.0500) time: 1.5013 data: 0.7227 max mem: 64948 Epoch: [192] [ 20/312] eta: 0:05:26 lr: 0.002617 min_lr: 0.002617 loss: 2.1303 (2.0550) weight_decay: 0.0500 (0.0500) time: 0.7183 data: 0.0023 max mem: 64948 Epoch: [192] [ 30/312] eta: 0:04:36 lr: 0.002617 min_lr: 0.002617 loss: 2.1148 (2.0420) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [192] [ 40/312] eta: 0:04:07 lr: 0.002617 min_lr: 0.002617 loss: 2.1788 (2.1111) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [192] [ 50/312] eta: 0:03:47 lr: 0.002616 min_lr: 0.002616 loss: 2.2673 (2.1145) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [192] [ 60/312] eta: 0:03:32 lr: 0.002616 min_lr: 0.002616 loss: 2.2673 (2.1284) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [192] [ 70/312] eta: 0:03:18 lr: 0.002615 min_lr: 0.002615 loss: 2.2391 (2.1387) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [192] [ 80/312] eta: 0:03:06 lr: 0.002615 min_lr: 0.002615 loss: 2.2391 (2.1323) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [192] [ 90/312] eta: 0:02:56 lr: 0.002614 min_lr: 0.002614 loss: 2.2963 (2.1465) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [192] [100/312] eta: 0:02:46 lr: 0.002614 min_lr: 0.002614 loss: 2.2991 (2.1516) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [192] [110/312] eta: 0:02:36 lr: 0.002613 min_lr: 0.002613 loss: 2.2486 (2.1498) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [192] [120/312] eta: 0:02:27 lr: 0.002613 min_lr: 0.002613 loss: 2.1640 (2.1441) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [192] [130/312] eta: 0:02:18 lr: 0.002613 min_lr: 0.002613 loss: 2.1975 (2.1453) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [192] [140/312] eta: 0:02:10 lr: 0.002612 min_lr: 0.002612 loss: 2.2093 (2.1532) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [192] [150/312] eta: 0:02:02 lr: 0.002612 min_lr: 0.002612 loss: 2.2984 (2.1664) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [192] [160/312] eta: 0:01:54 lr: 0.002611 min_lr: 0.002611 loss: 2.2824 (2.1675) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [192] [170/312] eta: 0:01:46 lr: 0.002611 min_lr: 0.002611 loss: 2.0062 (2.1578) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [192] [180/312] eta: 0:01:38 lr: 0.002610 min_lr: 0.002610 loss: 2.1014 (2.1611) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [192] [190/312] eta: 0:01:30 lr: 0.002610 min_lr: 0.002610 loss: 2.3549 (2.1679) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [192] [200/312] eta: 0:01:22 lr: 0.002609 min_lr: 0.002609 loss: 2.3275 (2.1642) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [192] [210/312] eta: 0:01:15 lr: 0.002609 min_lr: 0.002609 loss: 2.1700 (2.1665) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [192] [220/312] eta: 0:01:07 lr: 0.002609 min_lr: 0.002609 loss: 2.2824 (2.1699) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [192] [230/312] eta: 0:01:00 lr: 0.002608 min_lr: 0.002608 loss: 2.2971 (2.1697) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [192] [240/312] eta: 0:00:52 lr: 0.002608 min_lr: 0.002608 loss: 2.2200 (2.1710) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [192] [250/312] eta: 0:00:45 lr: 0.002607 min_lr: 0.002607 loss: 2.2200 (2.1717) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0003 max mem: 64948 Epoch: [192] [260/312] eta: 0:00:37 lr: 0.002607 min_lr: 0.002607 loss: 2.0925 (2.1702) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [192] [270/312] eta: 0:00:30 lr: 0.002606 min_lr: 0.002606 loss: 2.0925 (2.1655) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [192] [280/312] eta: 0:00:23 lr: 0.002606 min_lr: 0.002606 loss: 2.2607 (2.1655) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0010 max mem: 64948 Epoch: [192] [290/312] eta: 0:00:15 lr: 0.002605 min_lr: 0.002605 loss: 2.2010 (2.1622) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0008 max mem: 64948 Epoch: [192] [300/312] eta: 0:00:08 lr: 0.002605 min_lr: 0.002605 loss: 2.0868 (2.1632) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [192] [310/312] eta: 0:00:01 lr: 0.002605 min_lr: 0.002605 loss: 2.3715 (2.1722) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [192] [311/312] eta: 0:00:00 lr: 0.002605 min_lr: 0.002605 loss: 2.3715 (2.1718) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [192] Total time: 0:03:46 (0.7257 s / it) Averaged stats: lr: 0.002605 min_lr: 0.002605 loss: 2.3715 (2.1498) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6522 (0.6522) acc1: 84.6354 (84.6354) acc5: 95.5729 (95.5729) time: 4.5137 data: 4.2946 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9222 (0.8977) acc1: 76.0417 (76.7680) acc5: 94.5312 (93.5680) time: 0.6531 data: 0.4773 max mem: 64948 Test: Total time: 0:00:06 (0.6777 s / it) * Acc@1 77.714 Acc@5 93.870 loss 0.857 Accuracy of the model on the 50000 test images: 77.7% Max accuracy: 78.30% Test: [0/9] eta: 0:00:44 loss: 0.5497 (0.5497) acc1: 85.1562 (85.1562) acc5: 97.3958 (97.3958) time: 4.9053 data: 4.6998 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7309 (0.7204) acc1: 80.7292 (79.8720) acc5: 96.3542 (95.8400) time: 0.6968 data: 0.5223 max mem: 64948 Test: Total time: 0:00:06 (0.7078 s / it) * Acc@1 80.988 Acc@5 95.650 loss 0.700 Accuracy of the model EMA on 50000 test images: 81.0% Max EMA accuracy: 80.99% Epoch: [193] [ 0/312] eta: 0:48:26 lr: 0.002604 min_lr: 0.002604 loss: 1.7805 (1.7805) weight_decay: 0.0500 (0.0500) time: 9.3142 data: 7.7943 max mem: 64948 Epoch: [193] [ 10/312] eta: 0:07:34 lr: 0.002604 min_lr: 0.002604 loss: 1.8502 (1.9544) weight_decay: 0.0500 (0.0500) time: 1.5064 data: 0.7090 max mem: 64948 Epoch: [193] [ 20/312] eta: 0:05:26 lr: 0.002604 min_lr: 0.002604 loss: 1.9761 (2.0949) weight_decay: 0.0500 (0.0500) time: 0.7095 data: 0.0004 max mem: 64948 Epoch: [193] [ 30/312] eta: 0:04:37 lr: 0.002603 min_lr: 0.002603 loss: 2.0450 (2.0874) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [193] [ 40/312] eta: 0:04:08 lr: 0.002603 min_lr: 0.002603 loss: 2.1005 (2.0967) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0003 max mem: 64948 Epoch: [193] [ 50/312] eta: 0:03:48 lr: 0.002602 min_lr: 0.002602 loss: 2.1033 (2.0976) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [193] [ 60/312] eta: 0:03:31 lr: 0.002602 min_lr: 0.002602 loss: 2.2080 (2.0797) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [193] [ 70/312] eta: 0:03:18 lr: 0.002601 min_lr: 0.002601 loss: 2.0557 (2.0659) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [193] [ 80/312] eta: 0:03:06 lr: 0.002601 min_lr: 0.002601 loss: 2.1651 (2.0964) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [193] [ 90/312] eta: 0:02:56 lr: 0.002600 min_lr: 0.002600 loss: 2.2928 (2.0986) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [193] [100/312] eta: 0:02:46 lr: 0.002600 min_lr: 0.002600 loss: 2.1247 (2.0912) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [193] [110/312] eta: 0:02:36 lr: 0.002600 min_lr: 0.002600 loss: 2.0905 (2.0907) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [193] [120/312] eta: 0:02:27 lr: 0.002599 min_lr: 0.002599 loss: 2.2114 (2.0875) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [193] [130/312] eta: 0:02:18 lr: 0.002599 min_lr: 0.002599 loss: 2.2523 (2.0959) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [193] [140/312] eta: 0:02:10 lr: 0.002598 min_lr: 0.002598 loss: 2.2770 (2.1112) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [193] [150/312] eta: 0:02:02 lr: 0.002598 min_lr: 0.002598 loss: 2.2885 (2.1140) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [193] [160/312] eta: 0:01:54 lr: 0.002597 min_lr: 0.002597 loss: 1.9602 (2.1042) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [193] [170/312] eta: 0:01:46 lr: 0.002597 min_lr: 0.002597 loss: 2.0583 (2.1114) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [193] [180/312] eta: 0:01:38 lr: 0.002596 min_lr: 0.002596 loss: 2.3048 (2.1133) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [193] [190/312] eta: 0:01:30 lr: 0.002596 min_lr: 0.002596 loss: 2.2241 (2.1162) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [193] [200/312] eta: 0:01:22 lr: 0.002596 min_lr: 0.002596 loss: 2.0993 (2.1115) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [193] [210/312] eta: 0:01:15 lr: 0.002595 min_lr: 0.002595 loss: 1.9250 (2.1086) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [193] [220/312] eta: 0:01:07 lr: 0.002595 min_lr: 0.002595 loss: 2.3991 (2.1229) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [193] [230/312] eta: 0:01:00 lr: 0.002594 min_lr: 0.002594 loss: 2.3919 (2.1246) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [193] [240/312] eta: 0:00:52 lr: 0.002594 min_lr: 0.002594 loss: 2.1392 (2.1222) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [193] [250/312] eta: 0:00:45 lr: 0.002593 min_lr: 0.002593 loss: 2.2802 (2.1252) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [193] [260/312] eta: 0:00:37 lr: 0.002593 min_lr: 0.002593 loss: 2.2647 (2.1228) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [193] [270/312] eta: 0:00:30 lr: 0.002592 min_lr: 0.002592 loss: 2.1450 (2.1211) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [193] [280/312] eta: 0:00:23 lr: 0.002592 min_lr: 0.002592 loss: 1.9797 (2.1189) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0010 max mem: 64948 Epoch: [193] [290/312] eta: 0:00:15 lr: 0.002592 min_lr: 0.002592 loss: 1.9705 (2.1128) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [193] [300/312] eta: 0:00:08 lr: 0.002591 min_lr: 0.002591 loss: 2.1826 (2.1202) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [193] [310/312] eta: 0:00:01 lr: 0.002591 min_lr: 0.002591 loss: 2.3301 (2.1231) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [193] [311/312] eta: 0:00:00 lr: 0.002591 min_lr: 0.002591 loss: 2.3180 (2.1237) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [193] Total time: 0:03:46 (0.7261 s / it) Averaged stats: lr: 0.002591 min_lr: 0.002591 loss: 2.3180 (2.1523) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5860 (0.5860) acc1: 83.8542 (83.8542) acc5: 95.0521 (95.0521) time: 4.5746 data: 4.3501 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9145 (0.8526) acc1: 78.1250 (76.8640) acc5: 95.0521 (94.5280) time: 0.6596 data: 0.4834 max mem: 64948 Test: Total time: 0:00:06 (0.6792 s / it) * Acc@1 78.256 Acc@5 94.132 loss 0.847 Accuracy of the model on the 50000 test images: 78.3% Max accuracy: 78.30% Test: [0/9] eta: 0:00:45 loss: 0.5481 (0.5481) acc1: 85.1562 (85.1562) acc5: 97.1354 (97.1354) time: 5.1046 data: 4.8868 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7304 (0.7196) acc1: 80.9896 (79.9040) acc5: 96.3542 (95.8080) time: 0.7191 data: 0.5431 max mem: 64948 Test: Total time: 0:00:06 (0.7285 s / it) * Acc@1 81.010 Acc@5 95.674 loss 0.699 Accuracy of the model EMA on 50000 test images: 81.0% Max EMA accuracy: 81.01% Epoch: [194] [ 0/312] eta: 0:47:04 lr: 0.002591 min_lr: 0.002591 loss: 2.2457 (2.2457) weight_decay: 0.0500 (0.0500) time: 9.0530 data: 7.7491 max mem: 64948 Epoch: [194] [ 10/312] eta: 0:07:41 lr: 0.002590 min_lr: 0.002590 loss: 2.2457 (2.2273) weight_decay: 0.0500 (0.0500) time: 1.5279 data: 0.7319 max mem: 64948 Epoch: [194] [ 20/312] eta: 0:05:30 lr: 0.002590 min_lr: 0.002590 loss: 2.3187 (2.3381) weight_decay: 0.0500 (0.0500) time: 0.7344 data: 0.0153 max mem: 64948 Epoch: [194] [ 30/312] eta: 0:04:38 lr: 0.002589 min_lr: 0.002589 loss: 2.2850 (2.2859) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [194] [ 40/312] eta: 0:04:09 lr: 0.002589 min_lr: 0.002589 loss: 2.0475 (2.1911) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [194] [ 50/312] eta: 0:03:49 lr: 0.002588 min_lr: 0.002588 loss: 2.0052 (2.1538) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [194] [ 60/312] eta: 0:03:33 lr: 0.002588 min_lr: 0.002588 loss: 2.2417 (2.1871) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [194] [ 70/312] eta: 0:03:19 lr: 0.002587 min_lr: 0.002587 loss: 2.3531 (2.1868) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [194] [ 80/312] eta: 0:03:07 lr: 0.002587 min_lr: 0.002587 loss: 2.0202 (2.1760) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [194] [ 90/312] eta: 0:02:56 lr: 0.002587 min_lr: 0.002587 loss: 2.1062 (2.1777) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [194] [100/312] eta: 0:02:46 lr: 0.002586 min_lr: 0.002586 loss: 2.1062 (2.1588) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [194] [110/312] eta: 0:02:37 lr: 0.002586 min_lr: 0.002586 loss: 2.1651 (2.1671) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [194] [120/312] eta: 0:02:27 lr: 0.002585 min_lr: 0.002585 loss: 2.3215 (2.1717) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [194] [130/312] eta: 0:02:19 lr: 0.002585 min_lr: 0.002585 loss: 2.1679 (2.1719) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [194] [140/312] eta: 0:02:10 lr: 0.002584 min_lr: 0.002584 loss: 2.1272 (2.1716) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [194] [150/312] eta: 0:02:02 lr: 0.002584 min_lr: 0.002584 loss: 2.1951 (2.1725) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [194] [160/312] eta: 0:01:54 lr: 0.002583 min_lr: 0.002583 loss: 2.2518 (2.1812) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [194] [170/312] eta: 0:01:46 lr: 0.002583 min_lr: 0.002583 loss: 2.2747 (2.1799) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [194] [180/312] eta: 0:01:38 lr: 0.002582 min_lr: 0.002582 loss: 2.1860 (2.1715) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [194] [190/312] eta: 0:01:30 lr: 0.002582 min_lr: 0.002582 loss: 2.1860 (2.1724) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [194] [200/312] eta: 0:01:22 lr: 0.002582 min_lr: 0.002582 loss: 2.2004 (2.1680) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [194] [210/312] eta: 0:01:15 lr: 0.002581 min_lr: 0.002581 loss: 2.2413 (2.1732) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0005 max mem: 64948 Epoch: [194] [220/312] eta: 0:01:07 lr: 0.002581 min_lr: 0.002581 loss: 2.3136 (2.1772) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0005 max mem: 64948 Epoch: [194] [230/312] eta: 0:01:00 lr: 0.002580 min_lr: 0.002580 loss: 2.3191 (2.1836) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [194] [240/312] eta: 0:00:52 lr: 0.002580 min_lr: 0.002580 loss: 2.1581 (2.1711) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [194] [250/312] eta: 0:00:45 lr: 0.002579 min_lr: 0.002579 loss: 1.8188 (2.1676) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [194] [260/312] eta: 0:00:37 lr: 0.002579 min_lr: 0.002579 loss: 2.1969 (2.1665) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [194] [270/312] eta: 0:00:30 lr: 0.002578 min_lr: 0.002578 loss: 2.2230 (2.1677) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [194] [280/312] eta: 0:00:23 lr: 0.002578 min_lr: 0.002578 loss: 2.3493 (2.1706) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0013 max mem: 64948 Epoch: [194] [290/312] eta: 0:00:15 lr: 0.002578 min_lr: 0.002578 loss: 2.2799 (2.1680) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0012 max mem: 64948 Epoch: [194] [300/312] eta: 0:00:08 lr: 0.002577 min_lr: 0.002577 loss: 2.2011 (2.1709) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [194] [310/312] eta: 0:00:01 lr: 0.002577 min_lr: 0.002577 loss: 2.3063 (2.1774) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [194] [311/312] eta: 0:00:00 lr: 0.002577 min_lr: 0.002577 loss: 2.3276 (2.1790) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [194] Total time: 0:03:46 (0.7266 s / it) Averaged stats: lr: 0.002577 min_lr: 0.002577 loss: 2.3276 (2.1432) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6167 (0.6167) acc1: 84.3750 (84.3750) acc5: 95.8333 (95.8333) time: 4.6062 data: 4.3869 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9034 (0.8557) acc1: 77.3438 (77.4080) acc5: 93.7500 (94.1120) time: 0.6633 data: 0.4875 max mem: 64948 Test: Total time: 0:00:06 (0.6873 s / it) * Acc@1 78.430 Acc@5 94.182 loss 0.835 Accuracy of the model on the 50000 test images: 78.4% Max accuracy: 78.43% Test: [0/9] eta: 0:00:41 loss: 0.5463 (0.5463) acc1: 85.1562 (85.1562) acc5: 97.3958 (97.3958) time: 4.6180 data: 4.4087 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7294 (0.7187) acc1: 80.4688 (79.8400) acc5: 96.3542 (95.9040) time: 0.6643 data: 0.4899 max mem: 64948 Test: Total time: 0:00:06 (0.6737 s / it) * Acc@1 81.024 Acc@5 95.692 loss 0.698 Accuracy of the model EMA on 50000 test images: 81.0% Max EMA accuracy: 81.02% Epoch: [195] [ 0/312] eta: 0:48:40 lr: 0.002577 min_lr: 0.002577 loss: 1.2058 (1.2058) weight_decay: 0.0500 (0.0500) time: 9.3604 data: 7.3679 max mem: 64948 Epoch: [195] [ 10/312] eta: 0:07:35 lr: 0.002576 min_lr: 0.002576 loss: 2.2505 (2.1487) weight_decay: 0.0500 (0.0500) time: 1.5089 data: 0.6702 max mem: 64948 Epoch: [195] [ 20/312] eta: 0:05:27 lr: 0.002576 min_lr: 0.002576 loss: 2.0923 (2.0038) weight_decay: 0.0500 (0.0500) time: 0.7105 data: 0.0004 max mem: 64948 Epoch: [195] [ 30/312] eta: 0:04:37 lr: 0.002575 min_lr: 0.002575 loss: 2.0167 (2.0199) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [195] [ 40/312] eta: 0:04:08 lr: 0.002575 min_lr: 0.002575 loss: 2.1209 (2.0373) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [195] [ 50/312] eta: 0:03:48 lr: 0.002574 min_lr: 0.002574 loss: 2.0775 (2.0538) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [195] [ 60/312] eta: 0:03:32 lr: 0.002574 min_lr: 0.002574 loss: 2.0902 (2.0479) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [195] [ 70/312] eta: 0:03:18 lr: 0.002573 min_lr: 0.002573 loss: 2.1939 (2.0627) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [195] [ 80/312] eta: 0:03:06 lr: 0.002573 min_lr: 0.002573 loss: 2.2148 (2.0736) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [195] [ 90/312] eta: 0:02:56 lr: 0.002573 min_lr: 0.002573 loss: 2.1733 (2.0908) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [195] [100/312] eta: 0:02:46 lr: 0.002572 min_lr: 0.002572 loss: 2.1118 (2.0799) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [195] [110/312] eta: 0:02:36 lr: 0.002572 min_lr: 0.002572 loss: 2.1410 (2.0883) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [195] [120/312] eta: 0:02:27 lr: 0.002571 min_lr: 0.002571 loss: 2.1410 (2.0882) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [195] [130/312] eta: 0:02:18 lr: 0.002571 min_lr: 0.002571 loss: 2.0177 (2.0894) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [195] [140/312] eta: 0:02:10 lr: 0.002570 min_lr: 0.002570 loss: 2.2162 (2.1017) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [195] [150/312] eta: 0:02:02 lr: 0.002570 min_lr: 0.002570 loss: 2.1717 (2.1096) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [195] [160/312] eta: 0:01:54 lr: 0.002569 min_lr: 0.002569 loss: 2.1094 (2.1074) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [195] [170/312] eta: 0:01:46 lr: 0.002569 min_lr: 0.002569 loss: 1.9167 (2.1030) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [195] [180/312] eta: 0:01:38 lr: 0.002568 min_lr: 0.002568 loss: 2.1286 (2.1047) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [195] [190/312] eta: 0:01:30 lr: 0.002568 min_lr: 0.002568 loss: 2.1633 (2.1039) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [195] [200/312] eta: 0:01:22 lr: 0.002568 min_lr: 0.002568 loss: 2.3278 (2.1192) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [195] [210/312] eta: 0:01:15 lr: 0.002567 min_lr: 0.002567 loss: 2.2949 (2.1132) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [195] [220/312] eta: 0:01:07 lr: 0.002567 min_lr: 0.002567 loss: 2.1084 (2.1171) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [195] [230/312] eta: 0:01:00 lr: 0.002566 min_lr: 0.002566 loss: 2.1101 (2.1119) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [195] [240/312] eta: 0:00:52 lr: 0.002566 min_lr: 0.002566 loss: 2.0157 (2.1096) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [195] [250/312] eta: 0:00:45 lr: 0.002565 min_lr: 0.002565 loss: 2.0157 (2.1065) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [195] [260/312] eta: 0:00:37 lr: 0.002565 min_lr: 0.002565 loss: 1.9777 (2.1045) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [195] [270/312] eta: 0:00:30 lr: 0.002564 min_lr: 0.002564 loss: 2.1280 (2.1072) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [195] [280/312] eta: 0:00:23 lr: 0.002564 min_lr: 0.002564 loss: 2.1972 (2.1155) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [195] [290/312] eta: 0:00:15 lr: 0.002564 min_lr: 0.002564 loss: 2.1972 (2.1152) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [195] [300/312] eta: 0:00:08 lr: 0.002563 min_lr: 0.002563 loss: 2.0425 (2.1116) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [195] [310/312] eta: 0:00:01 lr: 0.002563 min_lr: 0.002563 loss: 2.1975 (2.1128) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [195] [311/312] eta: 0:00:00 lr: 0.002563 min_lr: 0.002563 loss: 2.2634 (2.1137) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [195] Total time: 0:03:46 (0.7261 s / it) Averaged stats: lr: 0.002563 min_lr: 0.002563 loss: 2.2634 (2.1413) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6123 (0.6123) acc1: 84.8958 (84.8958) acc5: 96.3542 (96.3542) time: 4.5644 data: 4.3447 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8985 (0.8663) acc1: 77.3438 (77.8240) acc5: 94.5312 (94.1440) time: 0.6585 data: 0.4828 max mem: 64948 Test: Total time: 0:00:06 (0.6829 s / it) * Acc@1 78.324 Acc@5 94.250 loss 0.844 Accuracy of the model on the 50000 test images: 78.3% Max accuracy: 78.43% Test: [0/9] eta: 0:00:44 loss: 0.5442 (0.5442) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.9554 data: 4.7429 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7283 (0.7178) acc1: 80.7292 (79.8400) acc5: 96.3542 (95.9360) time: 0.7018 data: 0.5271 max mem: 64948 Test: Total time: 0:00:06 (0.7091 s / it) * Acc@1 81.056 Acc@5 95.692 loss 0.697 Accuracy of the model EMA on 50000 test images: 81.1% Max EMA accuracy: 81.06% Epoch: [196] [ 0/312] eta: 0:46:50 lr: 0.002563 min_lr: 0.002563 loss: 1.7233 (1.7233) weight_decay: 0.0500 (0.0500) time: 9.0082 data: 7.8975 max mem: 64948 Epoch: [196] [ 10/312] eta: 0:07:43 lr: 0.002562 min_lr: 0.002562 loss: 2.2864 (2.1523) weight_decay: 0.0500 (0.0500) time: 1.5343 data: 0.7211 max mem: 64948 Epoch: [196] [ 20/312] eta: 0:05:31 lr: 0.002562 min_lr: 0.002562 loss: 2.0693 (2.0084) weight_decay: 0.0500 (0.0500) time: 0.7409 data: 0.0019 max mem: 64948 Epoch: [196] [ 30/312] eta: 0:04:40 lr: 0.002561 min_lr: 0.002561 loss: 2.1374 (2.0842) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [196] [ 40/312] eta: 0:04:10 lr: 0.002561 min_lr: 0.002561 loss: 2.1817 (2.0530) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [196] [ 50/312] eta: 0:03:49 lr: 0.002560 min_lr: 0.002560 loss: 1.9389 (2.0355) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [196] [ 60/312] eta: 0:03:33 lr: 0.002560 min_lr: 0.002560 loss: 2.0279 (2.0430) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [196] [ 70/312] eta: 0:03:19 lr: 0.002559 min_lr: 0.002559 loss: 2.3226 (2.0784) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [196] [ 80/312] eta: 0:03:07 lr: 0.002559 min_lr: 0.002559 loss: 2.3226 (2.1005) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [196] [ 90/312] eta: 0:02:56 lr: 0.002559 min_lr: 0.002559 loss: 2.1827 (2.1088) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [196] [100/312] eta: 0:02:46 lr: 0.002558 min_lr: 0.002558 loss: 2.0725 (2.1117) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [196] [110/312] eta: 0:02:37 lr: 0.002558 min_lr: 0.002558 loss: 2.1818 (2.1213) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [196] [120/312] eta: 0:02:28 lr: 0.002557 min_lr: 0.002557 loss: 2.2261 (2.1302) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [196] [130/312] eta: 0:02:19 lr: 0.002557 min_lr: 0.002557 loss: 2.1207 (2.1263) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [196] [140/312] eta: 0:02:10 lr: 0.002556 min_lr: 0.002556 loss: 1.9952 (2.1159) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [196] [150/312] eta: 0:02:02 lr: 0.002556 min_lr: 0.002556 loss: 2.0715 (2.1190) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [196] [160/312] eta: 0:01:54 lr: 0.002555 min_lr: 0.002555 loss: 2.2516 (2.1279) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [196] [170/312] eta: 0:01:46 lr: 0.002555 min_lr: 0.002555 loss: 2.2399 (2.1250) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [196] [180/312] eta: 0:01:38 lr: 0.002554 min_lr: 0.002554 loss: 2.0076 (2.1203) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [196] [190/312] eta: 0:01:30 lr: 0.002554 min_lr: 0.002554 loss: 2.1140 (2.1230) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [196] [200/312] eta: 0:01:22 lr: 0.002554 min_lr: 0.002554 loss: 2.2248 (2.1245) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [196] [210/312] eta: 0:01:15 lr: 0.002553 min_lr: 0.002553 loss: 2.1702 (2.1201) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [196] [220/312] eta: 0:01:07 lr: 0.002553 min_lr: 0.002553 loss: 2.0341 (2.1181) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [196] [230/312] eta: 0:01:00 lr: 0.002552 min_lr: 0.002552 loss: 2.1413 (2.1229) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [196] [240/312] eta: 0:00:52 lr: 0.002552 min_lr: 0.002552 loss: 2.2035 (2.1193) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [196] [250/312] eta: 0:00:45 lr: 0.002551 min_lr: 0.002551 loss: 2.1373 (2.1145) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0004 max mem: 64948 Epoch: [196] [260/312] eta: 0:00:37 lr: 0.002551 min_lr: 0.002551 loss: 2.0795 (2.1137) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [196] [270/312] eta: 0:00:30 lr: 0.002550 min_lr: 0.002550 loss: 2.0364 (2.1089) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [196] [280/312] eta: 0:00:23 lr: 0.002550 min_lr: 0.002550 loss: 2.0400 (2.1120) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [196] [290/312] eta: 0:00:15 lr: 0.002550 min_lr: 0.002550 loss: 2.2002 (2.1142) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [196] [300/312] eta: 0:00:08 lr: 0.002549 min_lr: 0.002549 loss: 2.1479 (2.1148) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [196] [310/312] eta: 0:00:01 lr: 0.002549 min_lr: 0.002549 loss: 2.1479 (2.1118) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [196] [311/312] eta: 0:00:00 lr: 0.002549 min_lr: 0.002549 loss: 2.1479 (2.1128) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [196] Total time: 0:03:46 (0.7267 s / it) Averaged stats: lr: 0.002549 min_lr: 0.002549 loss: 2.1479 (2.1241) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6231 (0.6231) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.5856 data: 4.3672 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9268 (0.8575) acc1: 76.5625 (77.6000) acc5: 94.5312 (94.0800) time: 0.6609 data: 0.4854 max mem: 64948 Test: Total time: 0:00:06 (0.6822 s / it) * Acc@1 78.162 Acc@5 94.026 loss 0.846 Accuracy of the model on the 50000 test images: 78.2% Max accuracy: 78.43% Test: [0/9] eta: 0:00:46 loss: 0.5424 (0.5424) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 5.1488 data: 4.9343 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7270 (0.7168) acc1: 80.7292 (79.8720) acc5: 96.6146 (95.9680) time: 0.7234 data: 0.5484 max mem: 64948 Test: Total time: 0:00:06 (0.7307 s / it) * Acc@1 81.090 Acc@5 95.698 loss 0.696 Accuracy of the model EMA on 50000 test images: 81.1% Max EMA accuracy: 81.09% Epoch: [197] [ 0/312] eta: 0:46:32 lr: 0.002549 min_lr: 0.002549 loss: 1.9889 (1.9889) weight_decay: 0.0500 (0.0500) time: 8.9496 data: 7.3644 max mem: 64948 Epoch: [197] [ 10/312] eta: 0:07:36 lr: 0.002548 min_lr: 0.002548 loss: 2.0814 (2.0607) weight_decay: 0.0500 (0.0500) time: 1.5122 data: 0.7177 max mem: 64948 Epoch: [197] [ 20/312] eta: 0:05:28 lr: 0.002548 min_lr: 0.002548 loss: 2.2197 (2.1261) weight_decay: 0.0500 (0.0500) time: 0.7326 data: 0.0267 max mem: 64948 Epoch: [197] [ 30/312] eta: 0:04:38 lr: 0.002547 min_lr: 0.002547 loss: 1.9986 (2.0663) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [197] [ 40/312] eta: 0:04:09 lr: 0.002547 min_lr: 0.002547 loss: 1.8919 (2.0368) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [197] [ 50/312] eta: 0:03:49 lr: 0.002546 min_lr: 0.002546 loss: 2.1790 (2.0553) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [197] [ 60/312] eta: 0:03:32 lr: 0.002546 min_lr: 0.002546 loss: 2.1790 (2.0630) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [197] [ 70/312] eta: 0:03:19 lr: 0.002545 min_lr: 0.002545 loss: 2.1234 (2.0774) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [197] [ 80/312] eta: 0:03:07 lr: 0.002545 min_lr: 0.002545 loss: 2.1030 (2.0941) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [197] [ 90/312] eta: 0:02:56 lr: 0.002544 min_lr: 0.002544 loss: 2.0733 (2.0919) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [197] [100/312] eta: 0:02:46 lr: 0.002544 min_lr: 0.002544 loss: 2.1665 (2.0964) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [197] [110/312] eta: 0:02:36 lr: 0.002544 min_lr: 0.002544 loss: 2.2129 (2.0982) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [197] [120/312] eta: 0:02:27 lr: 0.002543 min_lr: 0.002543 loss: 2.2129 (2.0994) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [197] [130/312] eta: 0:02:19 lr: 0.002543 min_lr: 0.002543 loss: 2.2286 (2.1011) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [197] [140/312] eta: 0:02:10 lr: 0.002542 min_lr: 0.002542 loss: 2.1269 (2.0974) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [197] [150/312] eta: 0:02:02 lr: 0.002542 min_lr: 0.002542 loss: 2.0693 (2.1021) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [197] [160/312] eta: 0:01:54 lr: 0.002541 min_lr: 0.002541 loss: 2.1479 (2.1018) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [197] [170/312] eta: 0:01:46 lr: 0.002541 min_lr: 0.002541 loss: 2.1316 (2.0984) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [197] [180/312] eta: 0:01:38 lr: 0.002540 min_lr: 0.002540 loss: 2.1316 (2.0992) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [197] [190/312] eta: 0:01:30 lr: 0.002540 min_lr: 0.002540 loss: 2.1912 (2.1035) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [197] [200/312] eta: 0:01:22 lr: 0.002540 min_lr: 0.002540 loss: 2.1844 (2.1092) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [197] [210/312] eta: 0:01:15 lr: 0.002539 min_lr: 0.002539 loss: 2.1195 (2.1042) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [197] [220/312] eta: 0:01:07 lr: 0.002539 min_lr: 0.002539 loss: 2.1993 (2.1138) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [197] [230/312] eta: 0:01:00 lr: 0.002538 min_lr: 0.002538 loss: 2.2232 (2.1137) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [197] [240/312] eta: 0:00:52 lr: 0.002538 min_lr: 0.002538 loss: 2.1836 (2.1157) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [197] [250/312] eta: 0:00:45 lr: 0.002537 min_lr: 0.002537 loss: 2.1962 (2.1177) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [197] [260/312] eta: 0:00:37 lr: 0.002537 min_lr: 0.002537 loss: 2.1962 (2.1152) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [197] [270/312] eta: 0:00:30 lr: 0.002536 min_lr: 0.002536 loss: 2.2420 (2.1191) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [197] [280/312] eta: 0:00:23 lr: 0.002536 min_lr: 0.002536 loss: 2.2665 (2.1225) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0010 max mem: 64948 Epoch: [197] [290/312] eta: 0:00:15 lr: 0.002535 min_lr: 0.002535 loss: 2.2730 (2.1253) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0008 max mem: 64948 Epoch: [197] [300/312] eta: 0:00:08 lr: 0.002535 min_lr: 0.002535 loss: 2.2189 (2.1228) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [197] [310/312] eta: 0:00:01 lr: 0.002535 min_lr: 0.002535 loss: 1.9925 (2.1154) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [197] [311/312] eta: 0:00:00 lr: 0.002534 min_lr: 0.002534 loss: 1.9925 (2.1162) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [197] Total time: 0:03:46 (0.7261 s / it) Averaged stats: lr: 0.002534 min_lr: 0.002534 loss: 1.9925 (2.1297) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6201 (0.6201) acc1: 84.6354 (84.6354) acc5: 95.3125 (95.3125) time: 4.5008 data: 4.2866 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8604 (0.8519) acc1: 79.1667 (78.3040) acc5: 95.3125 (94.3040) time: 0.6514 data: 0.4764 max mem: 64948 Test: Total time: 0:00:06 (0.6760 s / it) * Acc@1 78.546 Acc@5 94.370 loss 0.822 Accuracy of the model on the 50000 test images: 78.5% Max accuracy: 78.55% Test: [0/9] eta: 0:00:41 loss: 0.5406 (0.5406) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.5807 data: 4.3756 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7260 (0.7161) acc1: 80.9896 (79.8400) acc5: 96.6146 (96.0000) time: 0.6603 data: 0.4863 max mem: 64948 Test: Total time: 0:00:06 (0.6679 s / it) * Acc@1 81.108 Acc@5 95.712 loss 0.696 Accuracy of the model EMA on 50000 test images: 81.1% Max EMA accuracy: 81.11% Epoch: [198] [ 0/312] eta: 0:50:20 lr: 0.002534 min_lr: 0.002534 loss: 2.4424 (2.4424) weight_decay: 0.0500 (0.0500) time: 9.6808 data: 8.8851 max mem: 64948 Epoch: [198] [ 10/312] eta: 0:07:42 lr: 0.002534 min_lr: 0.002534 loss: 2.3273 (2.3484) weight_decay: 0.0500 (0.0500) time: 1.5324 data: 0.8082 max mem: 64948 Epoch: [198] [ 20/312] eta: 0:05:30 lr: 0.002534 min_lr: 0.002534 loss: 2.2417 (2.1637) weight_decay: 0.0500 (0.0500) time: 0.7060 data: 0.0004 max mem: 64948 Epoch: [198] [ 30/312] eta: 0:04:39 lr: 0.002533 min_lr: 0.002533 loss: 2.1942 (2.1803) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [198] [ 40/312] eta: 0:04:10 lr: 0.002533 min_lr: 0.002533 loss: 2.1942 (2.1530) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [198] [ 50/312] eta: 0:03:49 lr: 0.002532 min_lr: 0.002532 loss: 2.1177 (2.1604) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [198] [ 60/312] eta: 0:03:33 lr: 0.002532 min_lr: 0.002532 loss: 2.2280 (2.1324) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [198] [ 70/312] eta: 0:03:19 lr: 0.002531 min_lr: 0.002531 loss: 1.9798 (2.1032) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [198] [ 80/312] eta: 0:03:07 lr: 0.002531 min_lr: 0.002531 loss: 1.9798 (2.0930) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [198] [ 90/312] eta: 0:02:56 lr: 0.002530 min_lr: 0.002530 loss: 2.0819 (2.1028) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [198] [100/312] eta: 0:02:46 lr: 0.002530 min_lr: 0.002530 loss: 2.2415 (2.1080) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [198] [110/312] eta: 0:02:37 lr: 0.002529 min_lr: 0.002529 loss: 2.2619 (2.1193) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [198] [120/312] eta: 0:02:28 lr: 0.002529 min_lr: 0.002529 loss: 2.2880 (2.1246) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [198] [130/312] eta: 0:02:19 lr: 0.002529 min_lr: 0.002529 loss: 2.1209 (2.1189) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [198] [140/312] eta: 0:02:10 lr: 0.002528 min_lr: 0.002528 loss: 2.1528 (2.1308) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [198] [150/312] eta: 0:02:02 lr: 0.002528 min_lr: 0.002528 loss: 2.2866 (2.1398) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [198] [160/312] eta: 0:01:54 lr: 0.002527 min_lr: 0.002527 loss: 2.2803 (2.1478) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [198] [170/312] eta: 0:01:46 lr: 0.002527 min_lr: 0.002527 loss: 2.2644 (2.1497) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [198] [180/312] eta: 0:01:38 lr: 0.002526 min_lr: 0.002526 loss: 2.2403 (2.1516) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [198] [190/312] eta: 0:01:30 lr: 0.002526 min_lr: 0.002526 loss: 2.1943 (2.1498) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [198] [200/312] eta: 0:01:22 lr: 0.002525 min_lr: 0.002525 loss: 2.1399 (2.1480) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [198] [210/312] eta: 0:01:15 lr: 0.002525 min_lr: 0.002525 loss: 2.1093 (2.1398) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [198] [220/312] eta: 0:01:07 lr: 0.002525 min_lr: 0.002525 loss: 2.1954 (2.1404) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [198] [230/312] eta: 0:01:00 lr: 0.002524 min_lr: 0.002524 loss: 2.2147 (2.1433) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [198] [240/312] eta: 0:00:52 lr: 0.002524 min_lr: 0.002524 loss: 2.2646 (2.1472) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [198] [250/312] eta: 0:00:45 lr: 0.002523 min_lr: 0.002523 loss: 2.1452 (2.1445) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [198] [260/312] eta: 0:00:37 lr: 0.002523 min_lr: 0.002523 loss: 2.0828 (2.1445) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [198] [270/312] eta: 0:00:30 lr: 0.002522 min_lr: 0.002522 loss: 2.1831 (2.1427) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [198] [280/312] eta: 0:00:23 lr: 0.002522 min_lr: 0.002522 loss: 2.2564 (2.1448) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0009 max mem: 64948 Epoch: [198] [290/312] eta: 0:00:15 lr: 0.002521 min_lr: 0.002521 loss: 2.2070 (2.1407) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0008 max mem: 64948 Epoch: [198] [300/312] eta: 0:00:08 lr: 0.002521 min_lr: 0.002521 loss: 2.1355 (2.1397) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [198] [310/312] eta: 0:00:01 lr: 0.002520 min_lr: 0.002520 loss: 2.0922 (2.1374) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [198] [311/312] eta: 0:00:00 lr: 0.002520 min_lr: 0.002520 loss: 2.0623 (2.1361) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [198] Total time: 0:03:46 (0.7274 s / it) Averaged stats: lr: 0.002520 min_lr: 0.002520 loss: 2.0623 (2.1327) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6569 (0.6569) acc1: 84.6354 (84.6354) acc5: 95.5729 (95.5729) time: 4.7047 data: 4.4932 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8926 (0.8609) acc1: 79.6875 (77.8240) acc5: 95.5729 (94.4320) time: 0.6740 data: 0.4993 max mem: 64948 Test: Total time: 0:00:06 (0.6958 s / it) * Acc@1 78.530 Acc@5 94.284 loss 0.830 Accuracy of the model on the 50000 test images: 78.5% Max accuracy: 78.55% Test: [0/9] eta: 0:00:44 loss: 0.5393 (0.5393) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.9961 data: 4.7700 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7251 (0.7152) acc1: 80.9896 (79.9680) acc5: 96.6146 (96.0000) time: 0.7070 data: 0.5301 max mem: 64948 Test: Total time: 0:00:06 (0.7164 s / it) * Acc@1 81.140 Acc@5 95.730 loss 0.695 Accuracy of the model EMA on 50000 test images: 81.1% Max EMA accuracy: 81.14% Epoch: [199] [ 0/312] eta: 0:51:20 lr: 0.002520 min_lr: 0.002520 loss: 2.3795 (2.3795) weight_decay: 0.0500 (0.0500) time: 9.8741 data: 9.0999 max mem: 64948 Epoch: [199] [ 10/312] eta: 0:07:58 lr: 0.002520 min_lr: 0.002520 loss: 2.0482 (1.9938) weight_decay: 0.0500 (0.0500) time: 1.5852 data: 0.8276 max mem: 64948 Epoch: [199] [ 20/312] eta: 0:05:39 lr: 0.002519 min_lr: 0.002519 loss: 1.9947 (1.9697) weight_decay: 0.0500 (0.0500) time: 0.7267 data: 0.0004 max mem: 64948 Epoch: [199] [ 30/312] eta: 0:04:45 lr: 0.002519 min_lr: 0.002519 loss: 1.9947 (2.0000) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [199] [ 40/312] eta: 0:04:13 lr: 0.002519 min_lr: 0.002519 loss: 2.1258 (2.0060) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [199] [ 50/312] eta: 0:03:52 lr: 0.002518 min_lr: 0.002518 loss: 2.0383 (2.0123) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [199] [ 60/312] eta: 0:03:35 lr: 0.002518 min_lr: 0.002518 loss: 2.1224 (2.0382) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [199] [ 70/312] eta: 0:03:21 lr: 0.002517 min_lr: 0.002517 loss: 2.1105 (2.0533) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [199] [ 80/312] eta: 0:03:09 lr: 0.002517 min_lr: 0.002517 loss: 2.2256 (2.0671) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [199] [ 90/312] eta: 0:02:58 lr: 0.002516 min_lr: 0.002516 loss: 2.2611 (2.0922) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [199] [100/312] eta: 0:02:47 lr: 0.002516 min_lr: 0.002516 loss: 2.2318 (2.1028) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [199] [110/312] eta: 0:02:38 lr: 0.002515 min_lr: 0.002515 loss: 2.1178 (2.1076) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [199] [120/312] eta: 0:02:28 lr: 0.002515 min_lr: 0.002515 loss: 2.1178 (2.0984) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [199] [130/312] eta: 0:02:20 lr: 0.002514 min_lr: 0.002514 loss: 2.0967 (2.0907) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [199] [140/312] eta: 0:02:11 lr: 0.002514 min_lr: 0.002514 loss: 2.1280 (2.0921) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [199] [150/312] eta: 0:02:03 lr: 0.002514 min_lr: 0.002514 loss: 2.1617 (2.0909) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [199] [160/312] eta: 0:01:54 lr: 0.002513 min_lr: 0.002513 loss: 2.0740 (2.0974) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [199] [170/312] eta: 0:01:46 lr: 0.002513 min_lr: 0.002513 loss: 2.0190 (2.0898) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [199] [180/312] eta: 0:01:38 lr: 0.002512 min_lr: 0.002512 loss: 2.0147 (2.0918) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [199] [190/312] eta: 0:01:31 lr: 0.002512 min_lr: 0.002512 loss: 1.9858 (2.0858) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [199] [200/312] eta: 0:01:23 lr: 0.002511 min_lr: 0.002511 loss: 2.1419 (2.0920) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [199] [210/312] eta: 0:01:15 lr: 0.002511 min_lr: 0.002511 loss: 2.2555 (2.0972) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [199] [220/312] eta: 0:01:07 lr: 0.002510 min_lr: 0.002510 loss: 2.1494 (2.0994) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [199] [230/312] eta: 0:01:00 lr: 0.002510 min_lr: 0.002510 loss: 2.2774 (2.1022) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [199] [240/312] eta: 0:00:52 lr: 0.002510 min_lr: 0.002510 loss: 2.2538 (2.1022) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [199] [250/312] eta: 0:00:45 lr: 0.002509 min_lr: 0.002509 loss: 2.1923 (2.1047) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [199] [260/312] eta: 0:00:38 lr: 0.002509 min_lr: 0.002509 loss: 2.2611 (2.1042) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [199] [270/312] eta: 0:00:30 lr: 0.002508 min_lr: 0.002508 loss: 2.2364 (2.1066) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [199] [280/312] eta: 0:00:23 lr: 0.002508 min_lr: 0.002508 loss: 2.2530 (2.1146) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [199] [290/312] eta: 0:00:16 lr: 0.002507 min_lr: 0.002507 loss: 2.1366 (2.1066) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [199] [300/312] eta: 0:00:08 lr: 0.002507 min_lr: 0.002507 loss: 2.0086 (2.1073) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [199] [310/312] eta: 0:00:01 lr: 0.002506 min_lr: 0.002506 loss: 2.1587 (2.1096) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [199] [311/312] eta: 0:00:00 lr: 0.002506 min_lr: 0.002506 loss: 2.1726 (2.1099) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [199] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.002506 min_lr: 0.002506 loss: 2.1726 (2.1363) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6456 (0.6456) acc1: 84.1146 (84.1146) acc5: 95.0521 (95.0521) time: 4.6107 data: 4.3911 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9523 (0.8658) acc1: 77.0833 (77.0880) acc5: 94.7917 (93.7600) time: 0.6640 data: 0.4880 max mem: 64948 Test: Total time: 0:00:06 (0.6845 s / it) * Acc@1 78.538 Acc@5 94.192 loss 0.835 Accuracy of the model on the 50000 test images: 78.5% Max accuracy: 78.55% Test: [0/9] eta: 0:00:44 loss: 0.5382 (0.5382) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.9869 data: 4.7691 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7236 (0.7141) acc1: 80.9896 (80.0000) acc5: 96.6146 (96.0000) time: 0.7055 data: 0.5300 max mem: 64948 Test: Total time: 0:00:06 (0.7168 s / it) * Acc@1 81.160 Acc@5 95.736 loss 0.694 Accuracy of the model EMA on 50000 test images: 81.2% Max EMA accuracy: 81.16% Epoch: [200] [ 0/312] eta: 0:49:18 lr: 0.002506 min_lr: 0.002506 loss: 2.5753 (2.5753) weight_decay: 0.0500 (0.0500) time: 9.4829 data: 7.4056 max mem: 64948 Epoch: [200] [ 10/312] eta: 0:07:49 lr: 0.002506 min_lr: 0.002506 loss: 2.0461 (2.0607) weight_decay: 0.0500 (0.0500) time: 1.5531 data: 0.6736 max mem: 64948 Epoch: [200] [ 20/312] eta: 0:05:34 lr: 0.002505 min_lr: 0.002505 loss: 2.0461 (2.0612) weight_decay: 0.0500 (0.0500) time: 0.7278 data: 0.0003 max mem: 64948 Epoch: [200] [ 30/312] eta: 0:04:42 lr: 0.002505 min_lr: 0.002505 loss: 2.1428 (2.0586) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [200] [ 40/312] eta: 0:04:11 lr: 0.002504 min_lr: 0.002504 loss: 2.0960 (2.0698) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [200] [ 50/312] eta: 0:03:50 lr: 0.002504 min_lr: 0.002504 loss: 2.2385 (2.1181) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [200] [ 60/312] eta: 0:03:34 lr: 0.002504 min_lr: 0.002504 loss: 2.2863 (2.1192) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [200] [ 70/312] eta: 0:03:20 lr: 0.002503 min_lr: 0.002503 loss: 2.2153 (2.1165) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [200] [ 80/312] eta: 0:03:08 lr: 0.002503 min_lr: 0.002503 loss: 2.2167 (2.1085) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [200] [ 90/312] eta: 0:02:57 lr: 0.002502 min_lr: 0.002502 loss: 2.0894 (2.0986) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [200] [100/312] eta: 0:02:47 lr: 0.002502 min_lr: 0.002502 loss: 2.0515 (2.1001) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [200] [110/312] eta: 0:02:37 lr: 0.002501 min_lr: 0.002501 loss: 2.1786 (2.1106) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [200] [120/312] eta: 0:02:28 lr: 0.002501 min_lr: 0.002501 loss: 1.9905 (2.0813) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [200] [130/312] eta: 0:02:19 lr: 0.002500 min_lr: 0.002500 loss: 1.9240 (2.0845) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [200] [140/312] eta: 0:02:11 lr: 0.002500 min_lr: 0.002500 loss: 2.3129 (2.1043) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [200] [150/312] eta: 0:02:02 lr: 0.002499 min_lr: 0.002499 loss: 2.3150 (2.1086) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [200] [160/312] eta: 0:01:54 lr: 0.002499 min_lr: 0.002499 loss: 2.0545 (2.1032) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [200] [170/312] eta: 0:01:46 lr: 0.002499 min_lr: 0.002499 loss: 2.0364 (2.1037) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [200] [180/312] eta: 0:01:38 lr: 0.002498 min_lr: 0.002498 loss: 2.1514 (2.1094) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [200] [190/312] eta: 0:01:30 lr: 0.002498 min_lr: 0.002498 loss: 2.0043 (2.1044) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [200] [200/312] eta: 0:01:23 lr: 0.002497 min_lr: 0.002497 loss: 1.9978 (2.0995) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [200] [210/312] eta: 0:01:15 lr: 0.002497 min_lr: 0.002497 loss: 2.2429 (2.1114) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [200] [220/312] eta: 0:01:07 lr: 0.002496 min_lr: 0.002496 loss: 2.2257 (2.1071) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [200] [230/312] eta: 0:01:00 lr: 0.002496 min_lr: 0.002496 loss: 1.9979 (2.1030) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [200] [240/312] eta: 0:00:52 lr: 0.002495 min_lr: 0.002495 loss: 2.1616 (2.1097) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [200] [250/312] eta: 0:00:45 lr: 0.002495 min_lr: 0.002495 loss: 2.2564 (2.1045) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [200] [260/312] eta: 0:00:38 lr: 0.002494 min_lr: 0.002494 loss: 2.2564 (2.1108) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [200] [270/312] eta: 0:00:30 lr: 0.002494 min_lr: 0.002494 loss: 2.3006 (2.1179) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [200] [280/312] eta: 0:00:23 lr: 0.002494 min_lr: 0.002494 loss: 2.2751 (2.1145) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [200] [290/312] eta: 0:00:15 lr: 0.002493 min_lr: 0.002493 loss: 2.1238 (2.1222) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [200] [300/312] eta: 0:00:08 lr: 0.002493 min_lr: 0.002493 loss: 2.3185 (2.1255) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [200] [310/312] eta: 0:00:01 lr: 0.002492 min_lr: 0.002492 loss: 2.1751 (2.1241) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [200] [311/312] eta: 0:00:00 lr: 0.002492 min_lr: 0.002492 loss: 2.1751 (2.1255) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [200] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.002492 min_lr: 0.002492 loss: 2.1751 (2.1312) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6488 (0.6488) acc1: 84.6354 (84.6354) acc5: 95.8333 (95.8333) time: 4.4451 data: 4.2253 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9250 (0.8647) acc1: 77.0833 (77.3760) acc5: 93.7500 (93.9200) time: 0.6459 data: 0.4696 max mem: 64948 Test: Total time: 0:00:06 (0.6679 s / it) * Acc@1 78.386 Acc@5 94.258 loss 0.839 Accuracy of the model on the 50000 test images: 78.4% Max accuracy: 78.55% Test: [0/9] eta: 0:00:41 loss: 0.5375 (0.5375) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.6322 data: 4.4207 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7230 (0.7133) acc1: 81.2500 (79.9680) acc5: 96.6146 (96.0000) time: 0.6661 data: 0.4913 max mem: 64948 Test: Total time: 0:00:06 (0.6755 s / it) * Acc@1 81.196 Acc@5 95.736 loss 0.693 Accuracy of the model EMA on 50000 test images: 81.2% Max EMA accuracy: 81.20% Epoch: [201] [ 0/312] eta: 0:51:55 lr: 0.002492 min_lr: 0.002492 loss: 2.2348 (2.2348) weight_decay: 0.0500 (0.0500) time: 9.9869 data: 9.1380 max mem: 64948 Epoch: [201] [ 10/312] eta: 0:07:50 lr: 0.002492 min_lr: 0.002492 loss: 2.2119 (2.1140) weight_decay: 0.0500 (0.0500) time: 1.5577 data: 0.8311 max mem: 64948 Epoch: [201] [ 20/312] eta: 0:05:34 lr: 0.002491 min_lr: 0.002491 loss: 2.0779 (2.1354) weight_decay: 0.0500 (0.0500) time: 0.7048 data: 0.0004 max mem: 64948 Epoch: [201] [ 30/312] eta: 0:04:42 lr: 0.002491 min_lr: 0.002491 loss: 2.2174 (2.1360) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [201] [ 40/312] eta: 0:04:12 lr: 0.002490 min_lr: 0.002490 loss: 2.1940 (2.1113) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [201] [ 50/312] eta: 0:03:50 lr: 0.002490 min_lr: 0.002490 loss: 2.0462 (2.0853) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [201] [ 60/312] eta: 0:03:34 lr: 0.002489 min_lr: 0.002489 loss: 2.0570 (2.0969) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [201] [ 70/312] eta: 0:03:20 lr: 0.002489 min_lr: 0.002489 loss: 2.2044 (2.1187) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [201] [ 80/312] eta: 0:03:08 lr: 0.002488 min_lr: 0.002488 loss: 2.2384 (2.1241) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [201] [ 90/312] eta: 0:02:57 lr: 0.002488 min_lr: 0.002488 loss: 2.0475 (2.1122) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [201] [100/312] eta: 0:02:47 lr: 0.002488 min_lr: 0.002488 loss: 2.1814 (2.1227) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [201] [110/312] eta: 0:02:37 lr: 0.002487 min_lr: 0.002487 loss: 2.4018 (2.1364) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [201] [120/312] eta: 0:02:28 lr: 0.002487 min_lr: 0.002487 loss: 2.2829 (2.1306) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [201] [130/312] eta: 0:02:19 lr: 0.002486 min_lr: 0.002486 loss: 2.0510 (2.1228) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [201] [140/312] eta: 0:02:11 lr: 0.002486 min_lr: 0.002486 loss: 1.9959 (2.1106) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [201] [150/312] eta: 0:02:02 lr: 0.002485 min_lr: 0.002485 loss: 2.1539 (2.1264) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [201] [160/312] eta: 0:01:54 lr: 0.002485 min_lr: 0.002485 loss: 2.4080 (2.1324) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [201] [170/312] eta: 0:01:46 lr: 0.002484 min_lr: 0.002484 loss: 2.2177 (2.1365) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [201] [180/312] eta: 0:01:38 lr: 0.002484 min_lr: 0.002484 loss: 2.2177 (2.1375) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [201] [190/312] eta: 0:01:30 lr: 0.002483 min_lr: 0.002483 loss: 2.2898 (2.1482) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [201] [200/312] eta: 0:01:23 lr: 0.002483 min_lr: 0.002483 loss: 2.2914 (2.1502) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [201] [210/312] eta: 0:01:15 lr: 0.002483 min_lr: 0.002483 loss: 2.2270 (2.1478) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [201] [220/312] eta: 0:01:07 lr: 0.002482 min_lr: 0.002482 loss: 2.1526 (2.1491) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [201] [230/312] eta: 0:01:00 lr: 0.002482 min_lr: 0.002482 loss: 2.1999 (2.1502) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [201] [240/312] eta: 0:00:52 lr: 0.002481 min_lr: 0.002481 loss: 2.1999 (2.1507) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [201] [250/312] eta: 0:00:45 lr: 0.002481 min_lr: 0.002481 loss: 2.0754 (2.1463) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [201] [260/312] eta: 0:00:38 lr: 0.002480 min_lr: 0.002480 loss: 2.0914 (2.1485) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [201] [270/312] eta: 0:00:30 lr: 0.002480 min_lr: 0.002480 loss: 2.2856 (2.1532) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [201] [280/312] eta: 0:00:23 lr: 0.002479 min_lr: 0.002479 loss: 2.2463 (2.1506) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [201] [290/312] eta: 0:00:16 lr: 0.002479 min_lr: 0.002479 loss: 2.1236 (2.1480) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [201] [300/312] eta: 0:00:08 lr: 0.002478 min_lr: 0.002478 loss: 2.2823 (2.1509) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [201] [310/312] eta: 0:00:01 lr: 0.002478 min_lr: 0.002478 loss: 2.1890 (2.1468) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [201] [311/312] eta: 0:00:00 lr: 0.002478 min_lr: 0.002478 loss: 2.1890 (2.1444) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [201] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.002478 min_lr: 0.002478 loss: 2.1890 (2.1176) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6410 (0.6410) acc1: 84.6354 (84.6354) acc5: 95.5729 (95.5729) time: 4.6930 data: 4.4752 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9065 (0.8623) acc1: 79.1667 (77.9520) acc5: 94.2708 (94.0800) time: 0.6727 data: 0.4973 max mem: 64948 Test: Total time: 0:00:06 (0.7003 s / it) * Acc@1 78.394 Acc@5 94.262 loss 0.833 Accuracy of the model on the 50000 test images: 78.4% Max accuracy: 78.55% Test: [0/9] eta: 0:00:44 loss: 0.5360 (0.5360) acc1: 85.4167 (85.4167) acc5: 97.6562 (97.6562) time: 4.9469 data: 4.7328 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7223 (0.7125) acc1: 81.2500 (80.0320) acc5: 96.6146 (96.0320) time: 0.7052 data: 0.5260 max mem: 64948 Test: Total time: 0:00:06 (0.7154 s / it) * Acc@1 81.204 Acc@5 95.754 loss 0.692 Accuracy of the model EMA on 50000 test images: 81.2% Max EMA accuracy: 81.20% Epoch: [202] [ 0/312] eta: 0:54:53 lr: 0.002478 min_lr: 0.002478 loss: 2.5467 (2.5467) weight_decay: 0.0500 (0.0500) time: 10.5549 data: 9.8263 max mem: 64948 Epoch: [202] [ 10/312] eta: 0:08:04 lr: 0.002477 min_lr: 0.002477 loss: 2.3138 (2.0868) weight_decay: 0.0500 (0.0500) time: 1.6058 data: 0.8936 max mem: 64948 Epoch: [202] [ 20/312] eta: 0:05:42 lr: 0.002477 min_lr: 0.002477 loss: 2.0163 (2.1177) weight_decay: 0.0500 (0.0500) time: 0.7039 data: 0.0003 max mem: 64948 Epoch: [202] [ 30/312] eta: 0:04:47 lr: 0.002477 min_lr: 0.002477 loss: 2.0257 (2.1125) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [202] [ 40/312] eta: 0:04:15 lr: 0.002476 min_lr: 0.002476 loss: 2.1137 (2.0963) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [202] [ 50/312] eta: 0:03:53 lr: 0.002476 min_lr: 0.002476 loss: 2.1653 (2.1002) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [202] [ 60/312] eta: 0:03:36 lr: 0.002475 min_lr: 0.002475 loss: 2.1653 (2.1107) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [202] [ 70/312] eta: 0:03:22 lr: 0.002475 min_lr: 0.002475 loss: 2.2974 (2.1401) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [202] [ 80/312] eta: 0:03:10 lr: 0.002474 min_lr: 0.002474 loss: 2.2974 (2.1526) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [202] [ 90/312] eta: 0:02:59 lr: 0.002474 min_lr: 0.002474 loss: 2.2262 (2.1409) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [202] [100/312] eta: 0:02:48 lr: 0.002473 min_lr: 0.002473 loss: 2.2145 (2.1561) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [202] [110/312] eta: 0:02:38 lr: 0.002473 min_lr: 0.002473 loss: 2.2145 (2.1486) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0003 max mem: 64948 Epoch: [202] [120/312] eta: 0:02:29 lr: 0.002472 min_lr: 0.002472 loss: 2.0170 (2.1354) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [202] [130/312] eta: 0:02:20 lr: 0.002472 min_lr: 0.002472 loss: 1.9468 (2.1229) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [202] [140/312] eta: 0:02:11 lr: 0.002472 min_lr: 0.002472 loss: 1.9468 (2.1145) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [202] [150/312] eta: 0:02:03 lr: 0.002471 min_lr: 0.002471 loss: 2.1353 (2.1180) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [202] [160/312] eta: 0:01:55 lr: 0.002471 min_lr: 0.002471 loss: 2.2166 (2.1222) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [202] [170/312] eta: 0:01:47 lr: 0.002470 min_lr: 0.002470 loss: 2.1733 (2.1189) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [202] [180/312] eta: 0:01:39 lr: 0.002470 min_lr: 0.002470 loss: 2.1303 (2.1169) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [202] [190/312] eta: 0:01:31 lr: 0.002469 min_lr: 0.002469 loss: 2.1349 (2.1185) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [202] [200/312] eta: 0:01:23 lr: 0.002469 min_lr: 0.002469 loss: 2.2164 (2.1234) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [202] [210/312] eta: 0:01:15 lr: 0.002468 min_lr: 0.002468 loss: 2.2164 (2.1207) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [202] [220/312] eta: 0:01:08 lr: 0.002468 min_lr: 0.002468 loss: 2.1296 (2.1209) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [202] [230/312] eta: 0:01:00 lr: 0.002467 min_lr: 0.002467 loss: 2.0714 (2.1205) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [202] [240/312] eta: 0:00:53 lr: 0.002467 min_lr: 0.002467 loss: 2.2111 (2.1256) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [202] [250/312] eta: 0:00:45 lr: 0.002467 min_lr: 0.002467 loss: 2.2724 (2.1289) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [202] [260/312] eta: 0:00:38 lr: 0.002466 min_lr: 0.002466 loss: 2.1998 (2.1295) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [202] [270/312] eta: 0:00:30 lr: 0.002466 min_lr: 0.002466 loss: 2.0895 (2.1271) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [202] [280/312] eta: 0:00:23 lr: 0.002465 min_lr: 0.002465 loss: 2.2307 (2.1289) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0010 max mem: 64948 Epoch: [202] [290/312] eta: 0:00:16 lr: 0.002465 min_lr: 0.002465 loss: 2.2559 (2.1273) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0009 max mem: 64948 Epoch: [202] [300/312] eta: 0:00:08 lr: 0.002464 min_lr: 0.002464 loss: 2.2460 (2.1317) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [202] [310/312] eta: 0:00:01 lr: 0.002464 min_lr: 0.002464 loss: 2.2730 (2.1364) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [202] [311/312] eta: 0:00:00 lr: 0.002464 min_lr: 0.002464 loss: 2.2730 (2.1362) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [202] Total time: 0:03:47 (0.7299 s / it) Averaged stats: lr: 0.002464 min_lr: 0.002464 loss: 2.2730 (2.1274) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6892 (0.6892) acc1: 83.0729 (83.0729) acc5: 94.5312 (94.5312) time: 4.6061 data: 4.3876 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9106 (0.8631) acc1: 77.6042 (76.9920) acc5: 93.4896 (94.0160) time: 0.6631 data: 0.4876 max mem: 64948 Test: Total time: 0:00:06 (0.6863 s / it) * Acc@1 78.206 Acc@5 94.222 loss 0.852 Accuracy of the model on the 50000 test images: 78.2% Max accuracy: 78.55% Test: [0/9] eta: 0:00:41 loss: 0.5349 (0.5349) acc1: 85.4167 (85.4167) acc5: 97.6562 (97.6562) time: 4.6595 data: 4.4414 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7220 (0.7114) acc1: 81.2500 (80.0640) acc5: 96.3542 (96.0960) time: 0.6757 data: 0.4998 max mem: 64948 Test: Total time: 0:00:06 (0.6885 s / it) * Acc@1 81.230 Acc@5 95.764 loss 0.691 Accuracy of the model EMA on 50000 test images: 81.2% Max EMA accuracy: 81.23% Epoch: [203] [ 0/312] eta: 0:45:42 lr: 0.002464 min_lr: 0.002464 loss: 2.5492 (2.5492) weight_decay: 0.0500 (0.0500) time: 8.7906 data: 7.1200 max mem: 64948 Epoch: [203] [ 10/312] eta: 0:07:23 lr: 0.002463 min_lr: 0.002463 loss: 2.1953 (2.2528) weight_decay: 0.0500 (0.0500) time: 1.4687 data: 0.6478 max mem: 64948 Epoch: [203] [ 20/312] eta: 0:05:21 lr: 0.002463 min_lr: 0.002463 loss: 2.1873 (2.1538) weight_decay: 0.0500 (0.0500) time: 0.7152 data: 0.0004 max mem: 64948 Epoch: [203] [ 30/312] eta: 0:04:33 lr: 0.002462 min_lr: 0.002462 loss: 2.1582 (2.1449) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [203] [ 40/312] eta: 0:04:05 lr: 0.002462 min_lr: 0.002462 loss: 2.2712 (2.1833) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [203] [ 50/312] eta: 0:03:45 lr: 0.002461 min_lr: 0.002461 loss: 2.2401 (2.1716) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [203] [ 60/312] eta: 0:03:30 lr: 0.002461 min_lr: 0.002461 loss: 2.1636 (2.1750) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [203] [ 70/312] eta: 0:03:17 lr: 0.002461 min_lr: 0.002461 loss: 2.1213 (2.1510) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [203] [ 80/312] eta: 0:03:05 lr: 0.002460 min_lr: 0.002460 loss: 2.2776 (2.1490) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [203] [ 90/312] eta: 0:02:55 lr: 0.002460 min_lr: 0.002460 loss: 2.2723 (2.1538) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0003 max mem: 64948 Epoch: [203] [100/312] eta: 0:02:45 lr: 0.002459 min_lr: 0.002459 loss: 2.1799 (2.1477) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [203] [110/312] eta: 0:02:36 lr: 0.002459 min_lr: 0.002459 loss: 2.2059 (2.1496) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [203] [120/312] eta: 0:02:27 lr: 0.002458 min_lr: 0.002458 loss: 2.3105 (2.1570) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [203] [130/312] eta: 0:02:18 lr: 0.002458 min_lr: 0.002458 loss: 2.1720 (2.1492) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [203] [140/312] eta: 0:02:10 lr: 0.002457 min_lr: 0.002457 loss: 2.1291 (2.1519) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [203] [150/312] eta: 0:02:01 lr: 0.002457 min_lr: 0.002457 loss: 2.1291 (2.1418) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [203] [160/312] eta: 0:01:53 lr: 0.002456 min_lr: 0.002456 loss: 1.9987 (2.1405) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [203] [170/312] eta: 0:01:45 lr: 0.002456 min_lr: 0.002456 loss: 2.2405 (2.1488) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [203] [180/312] eta: 0:01:37 lr: 0.002456 min_lr: 0.002456 loss: 2.2405 (2.1423) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [203] [190/312] eta: 0:01:30 lr: 0.002455 min_lr: 0.002455 loss: 2.1040 (2.1394) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [203] [200/312] eta: 0:01:22 lr: 0.002455 min_lr: 0.002455 loss: 2.1040 (2.1390) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [203] [210/312] eta: 0:01:15 lr: 0.002454 min_lr: 0.002454 loss: 2.2559 (2.1414) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [203] [220/312] eta: 0:01:07 lr: 0.002454 min_lr: 0.002454 loss: 2.2294 (2.1354) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [203] [230/312] eta: 0:01:00 lr: 0.002453 min_lr: 0.002453 loss: 2.1041 (2.1332) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [203] [240/312] eta: 0:00:52 lr: 0.002453 min_lr: 0.002453 loss: 2.1468 (2.1333) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [203] [250/312] eta: 0:00:45 lr: 0.002452 min_lr: 0.002452 loss: 2.1413 (2.1303) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [203] [260/312] eta: 0:00:37 lr: 0.002452 min_lr: 0.002452 loss: 2.0787 (2.1260) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [203] [270/312] eta: 0:00:30 lr: 0.002451 min_lr: 0.002451 loss: 2.0809 (2.1225) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [203] [280/312] eta: 0:00:23 lr: 0.002451 min_lr: 0.002451 loss: 2.2130 (2.1234) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [203] [290/312] eta: 0:00:15 lr: 0.002451 min_lr: 0.002451 loss: 2.1780 (2.1182) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [203] [300/312] eta: 0:00:08 lr: 0.002450 min_lr: 0.002450 loss: 2.1780 (2.1207) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [203] [310/312] eta: 0:00:01 lr: 0.002450 min_lr: 0.002450 loss: 2.1942 (2.1180) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [203] [311/312] eta: 0:00:00 lr: 0.002450 min_lr: 0.002450 loss: 2.1879 (2.1183) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [203] Total time: 0:03:46 (0.7246 s / it) Averaged stats: lr: 0.002450 min_lr: 0.002450 loss: 2.1879 (2.1220) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5992 (0.5992) acc1: 85.9375 (85.9375) acc5: 96.8750 (96.8750) time: 4.4489 data: 4.2439 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9219 (0.8481) acc1: 77.3438 (77.3120) acc5: 95.0521 (94.1760) time: 0.6458 data: 0.4716 max mem: 64948 Test: Total time: 0:00:06 (0.6689 s / it) * Acc@1 78.570 Acc@5 94.356 loss 0.825 Accuracy of the model on the 50000 test images: 78.6% Max accuracy: 78.57% Test: [0/9] eta: 0:00:39 loss: 0.5334 (0.5334) acc1: 85.4167 (85.4167) acc5: 97.6562 (97.6562) time: 4.3858 data: 4.1686 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7211 (0.7104) acc1: 81.2500 (80.0640) acc5: 96.6146 (96.0960) time: 0.6395 data: 0.4633 max mem: 64948 Test: Total time: 0:00:05 (0.6480 s / it) * Acc@1 81.246 Acc@5 95.780 loss 0.690 Accuracy of the model EMA on 50000 test images: 81.2% Max EMA accuracy: 81.25% Epoch: [204] [ 0/312] eta: 0:54:03 lr: 0.002449 min_lr: 0.002449 loss: 1.6591 (1.6591) weight_decay: 0.0500 (0.0500) time: 10.3963 data: 9.6225 max mem: 64948 Epoch: [204] [ 10/312] eta: 0:08:01 lr: 0.002449 min_lr: 0.002449 loss: 1.9037 (2.0337) weight_decay: 0.0500 (0.0500) time: 1.5943 data: 0.8751 max mem: 64948 Epoch: [204] [ 20/312] eta: 0:05:40 lr: 0.002449 min_lr: 0.002449 loss: 2.2785 (2.1981) weight_decay: 0.0500 (0.0500) time: 0.7061 data: 0.0004 max mem: 64948 Epoch: [204] [ 30/312] eta: 0:04:46 lr: 0.002448 min_lr: 0.002448 loss: 2.2897 (2.1859) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [204] [ 40/312] eta: 0:04:14 lr: 0.002448 min_lr: 0.002448 loss: 2.1520 (2.1587) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [204] [ 50/312] eta: 0:03:53 lr: 0.002447 min_lr: 0.002447 loss: 2.1301 (2.1363) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [204] [ 60/312] eta: 0:03:36 lr: 0.002447 min_lr: 0.002447 loss: 2.1023 (2.1274) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [204] [ 70/312] eta: 0:03:22 lr: 0.002446 min_lr: 0.002446 loss: 2.2219 (2.1311) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [204] [ 80/312] eta: 0:03:09 lr: 0.002446 min_lr: 0.002446 loss: 2.2201 (2.1321) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [204] [ 90/312] eta: 0:02:58 lr: 0.002445 min_lr: 0.002445 loss: 2.1975 (2.1413) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [204] [100/312] eta: 0:02:48 lr: 0.002445 min_lr: 0.002445 loss: 2.1218 (2.1238) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [204] [110/312] eta: 0:02:38 lr: 0.002444 min_lr: 0.002444 loss: 2.1182 (2.1361) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [204] [120/312] eta: 0:02:29 lr: 0.002444 min_lr: 0.002444 loss: 2.2116 (2.1346) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [204] [130/312] eta: 0:02:20 lr: 0.002444 min_lr: 0.002444 loss: 2.2116 (2.1345) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [204] [140/312] eta: 0:02:11 lr: 0.002443 min_lr: 0.002443 loss: 2.2448 (2.1434) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [204] [150/312] eta: 0:02:03 lr: 0.002443 min_lr: 0.002443 loss: 2.1956 (2.1445) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [204] [160/312] eta: 0:01:55 lr: 0.002442 min_lr: 0.002442 loss: 2.0978 (2.1407) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [204] [170/312] eta: 0:01:46 lr: 0.002442 min_lr: 0.002442 loss: 2.2682 (2.1454) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [204] [180/312] eta: 0:01:38 lr: 0.002441 min_lr: 0.002441 loss: 2.2076 (2.1418) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [204] [190/312] eta: 0:01:31 lr: 0.002441 min_lr: 0.002441 loss: 2.0898 (2.1398) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [204] [200/312] eta: 0:01:23 lr: 0.002440 min_lr: 0.002440 loss: 2.0977 (2.1435) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [204] [210/312] eta: 0:01:15 lr: 0.002440 min_lr: 0.002440 loss: 2.0604 (2.1335) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [204] [220/312] eta: 0:01:08 lr: 0.002439 min_lr: 0.002439 loss: 2.2297 (2.1407) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [204] [230/312] eta: 0:01:00 lr: 0.002439 min_lr: 0.002439 loss: 2.3213 (2.1448) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [204] [240/312] eta: 0:00:52 lr: 0.002439 min_lr: 0.002439 loss: 2.2026 (2.1436) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [204] [250/312] eta: 0:00:45 lr: 0.002438 min_lr: 0.002438 loss: 2.0927 (2.1476) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [204] [260/312] eta: 0:00:38 lr: 0.002438 min_lr: 0.002438 loss: 2.1486 (2.1473) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [204] [270/312] eta: 0:00:30 lr: 0.002437 min_lr: 0.002437 loss: 2.1945 (2.1486) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [204] [280/312] eta: 0:00:23 lr: 0.002437 min_lr: 0.002437 loss: 2.2456 (2.1468) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [204] [290/312] eta: 0:00:16 lr: 0.002436 min_lr: 0.002436 loss: 2.0913 (2.1432) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [204] [300/312] eta: 0:00:08 lr: 0.002436 min_lr: 0.002436 loss: 2.0583 (2.1414) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [204] [310/312] eta: 0:00:01 lr: 0.002435 min_lr: 0.002435 loss: 2.0949 (2.1430) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [204] [311/312] eta: 0:00:00 lr: 0.002435 min_lr: 0.002435 loss: 2.1206 (2.1446) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [204] Total time: 0:03:47 (0.7293 s / it) Averaged stats: lr: 0.002435 min_lr: 0.002435 loss: 2.1206 (2.1326) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5924 (0.5924) acc1: 85.1562 (85.1562) acc5: 96.8750 (96.8750) time: 4.6587 data: 4.4439 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9101 (0.8390) acc1: 78.3854 (78.0160) acc5: 95.3125 (94.9440) time: 0.6689 data: 0.4939 max mem: 64948 Test: Total time: 0:00:06 (0.6938 s / it) * Acc@1 78.676 Acc@5 94.468 loss 0.823 Accuracy of the model on the 50000 test images: 78.7% Max accuracy: 78.68% Test: [0/9] eta: 0:00:40 loss: 0.5325 (0.5325) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.5050 data: 4.2870 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7209 (0.7097) acc1: 81.2500 (80.0640) acc5: 96.6146 (96.0640) time: 0.6519 data: 0.4764 max mem: 64948 Test: Total time: 0:00:05 (0.6616 s / it) * Acc@1 81.262 Acc@5 95.788 loss 0.689 Accuracy of the model EMA on 50000 test images: 81.3% Max EMA accuracy: 81.26% Epoch: [205] [ 0/312] eta: 0:52:49 lr: 0.002435 min_lr: 0.002435 loss: 2.3236 (2.3236) weight_decay: 0.0500 (0.0500) time: 10.1581 data: 9.3721 max mem: 64948 Epoch: [205] [ 10/312] eta: 0:07:54 lr: 0.002435 min_lr: 0.002435 loss: 2.2344 (2.0784) weight_decay: 0.0500 (0.0500) time: 1.5703 data: 0.8523 max mem: 64948 Epoch: [205] [ 20/312] eta: 0:05:36 lr: 0.002434 min_lr: 0.002434 loss: 2.2481 (2.1158) weight_decay: 0.0500 (0.0500) time: 0.7019 data: 0.0003 max mem: 64948 Epoch: [205] [ 30/312] eta: 0:04:43 lr: 0.002434 min_lr: 0.002434 loss: 2.2481 (2.1333) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [205] [ 40/312] eta: 0:04:13 lr: 0.002433 min_lr: 0.002433 loss: 2.0530 (2.1026) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [205] [ 50/312] eta: 0:03:51 lr: 0.002433 min_lr: 0.002433 loss: 2.1602 (2.1272) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [205] [ 60/312] eta: 0:03:35 lr: 0.002433 min_lr: 0.002433 loss: 2.1980 (2.1293) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [205] [ 70/312] eta: 0:03:21 lr: 0.002432 min_lr: 0.002432 loss: 2.1888 (2.1236) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [205] [ 80/312] eta: 0:03:09 lr: 0.002432 min_lr: 0.002432 loss: 2.0682 (2.1009) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [205] [ 90/312] eta: 0:02:58 lr: 0.002431 min_lr: 0.002431 loss: 2.0527 (2.1091) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [205] [100/312] eta: 0:02:47 lr: 0.002431 min_lr: 0.002431 loss: 2.2115 (2.1121) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [205] [110/312] eta: 0:02:38 lr: 0.002430 min_lr: 0.002430 loss: 2.0993 (2.1151) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [205] [120/312] eta: 0:02:29 lr: 0.002430 min_lr: 0.002430 loss: 2.0938 (2.1133) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [205] [130/312] eta: 0:02:20 lr: 0.002429 min_lr: 0.002429 loss: 2.1285 (2.1104) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [205] [140/312] eta: 0:02:11 lr: 0.002429 min_lr: 0.002429 loss: 2.1931 (2.1143) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [205] [150/312] eta: 0:02:03 lr: 0.002428 min_lr: 0.002428 loss: 2.2234 (2.1157) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [205] [160/312] eta: 0:01:54 lr: 0.002428 min_lr: 0.002428 loss: 2.1767 (2.1047) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [205] [170/312] eta: 0:01:46 lr: 0.002427 min_lr: 0.002427 loss: 1.9428 (2.1014) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [205] [180/312] eta: 0:01:38 lr: 0.002427 min_lr: 0.002427 loss: 1.9647 (2.0990) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [205] [190/312] eta: 0:01:30 lr: 0.002427 min_lr: 0.002427 loss: 2.1511 (2.1037) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [205] [200/312] eta: 0:01:23 lr: 0.002426 min_lr: 0.002426 loss: 1.9518 (2.0965) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [205] [210/312] eta: 0:01:15 lr: 0.002426 min_lr: 0.002426 loss: 1.9233 (2.0976) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [205] [220/312] eta: 0:01:07 lr: 0.002425 min_lr: 0.002425 loss: 2.1968 (2.1071) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [205] [230/312] eta: 0:01:00 lr: 0.002425 min_lr: 0.002425 loss: 2.1968 (2.0999) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [205] [240/312] eta: 0:00:52 lr: 0.002424 min_lr: 0.002424 loss: 2.0401 (2.0947) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [205] [250/312] eta: 0:00:45 lr: 0.002424 min_lr: 0.002424 loss: 2.0774 (2.0958) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [205] [260/312] eta: 0:00:38 lr: 0.002423 min_lr: 0.002423 loss: 2.2026 (2.0993) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [205] [270/312] eta: 0:00:30 lr: 0.002423 min_lr: 0.002423 loss: 2.2026 (2.1010) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [205] [280/312] eta: 0:00:23 lr: 0.002422 min_lr: 0.002422 loss: 2.1581 (2.1004) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [205] [290/312] eta: 0:00:16 lr: 0.002422 min_lr: 0.002422 loss: 2.2237 (2.1038) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [205] [300/312] eta: 0:00:08 lr: 0.002422 min_lr: 0.002422 loss: 2.2122 (2.1013) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [205] [310/312] eta: 0:00:01 lr: 0.002421 min_lr: 0.002421 loss: 2.0272 (2.0990) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [205] [311/312] eta: 0:00:00 lr: 0.002421 min_lr: 0.002421 loss: 2.0172 (2.0982) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [205] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.002421 min_lr: 0.002421 loss: 2.0172 (2.1121) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6452 (0.6452) acc1: 84.3750 (84.3750) acc5: 96.0938 (96.0938) time: 4.6636 data: 4.4485 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9253 (0.8595) acc1: 78.3854 (77.6640) acc5: 94.7917 (94.1760) time: 0.6695 data: 0.4944 max mem: 64948 Test: Total time: 0:00:06 (0.6802 s / it) * Acc@1 78.540 Acc@5 94.400 loss 0.833 Accuracy of the model on the 50000 test images: 78.5% Max accuracy: 78.68% Test: [0/9] eta: 0:00:44 loss: 0.5313 (0.5313) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.9854 data: 4.7790 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7208 (0.7090) acc1: 81.2500 (79.9680) acc5: 96.6146 (96.0640) time: 0.7052 data: 0.5311 max mem: 64948 Test: Total time: 0:00:06 (0.7174 s / it) * Acc@1 81.302 Acc@5 95.816 loss 0.688 Accuracy of the model EMA on 50000 test images: 81.3% Max EMA accuracy: 81.30% Epoch: [206] [ 0/312] eta: 0:52:32 lr: 0.002421 min_lr: 0.002421 loss: 2.1356 (2.1356) weight_decay: 0.0500 (0.0500) time: 10.1043 data: 9.3195 max mem: 64948 Epoch: [206] [ 10/312] eta: 0:07:52 lr: 0.002421 min_lr: 0.002421 loss: 2.0618 (2.0390) weight_decay: 0.0500 (0.0500) time: 1.5647 data: 0.8476 max mem: 64948 Epoch: [206] [ 20/312] eta: 0:05:35 lr: 0.002420 min_lr: 0.002420 loss: 2.0618 (2.1126) weight_decay: 0.0500 (0.0500) time: 0.7023 data: 0.0003 max mem: 64948 Epoch: [206] [ 30/312] eta: 0:04:42 lr: 0.002420 min_lr: 0.002420 loss: 2.2546 (2.1695) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [206] [ 40/312] eta: 0:04:12 lr: 0.002419 min_lr: 0.002419 loss: 2.3646 (2.1817) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [206] [ 50/312] eta: 0:03:51 lr: 0.002419 min_lr: 0.002419 loss: 2.0969 (2.1384) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [206] [ 60/312] eta: 0:03:34 lr: 0.002418 min_lr: 0.002418 loss: 2.1089 (2.1626) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [206] [ 70/312] eta: 0:03:20 lr: 0.002418 min_lr: 0.002418 loss: 2.1776 (2.1600) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [206] [ 80/312] eta: 0:03:08 lr: 0.002417 min_lr: 0.002417 loss: 2.0831 (2.1502) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [206] [ 90/312] eta: 0:02:57 lr: 0.002417 min_lr: 0.002417 loss: 2.1346 (2.1426) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [206] [100/312] eta: 0:02:47 lr: 0.002416 min_lr: 0.002416 loss: 2.1758 (2.1397) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [206] [110/312] eta: 0:02:37 lr: 0.002416 min_lr: 0.002416 loss: 2.1796 (2.1426) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [206] [120/312] eta: 0:02:28 lr: 0.002415 min_lr: 0.002415 loss: 2.1567 (2.1346) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [206] [130/312] eta: 0:02:19 lr: 0.002415 min_lr: 0.002415 loss: 2.0974 (2.1288) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [206] [140/312] eta: 0:02:11 lr: 0.002415 min_lr: 0.002415 loss: 2.1771 (2.1380) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [206] [150/312] eta: 0:02:02 lr: 0.002414 min_lr: 0.002414 loss: 2.3468 (2.1452) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [206] [160/312] eta: 0:01:54 lr: 0.002414 min_lr: 0.002414 loss: 2.2552 (2.1269) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [206] [170/312] eta: 0:01:46 lr: 0.002413 min_lr: 0.002413 loss: 1.9409 (2.1215) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [206] [180/312] eta: 0:01:38 lr: 0.002413 min_lr: 0.002413 loss: 2.1754 (2.1242) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [206] [190/312] eta: 0:01:30 lr: 0.002412 min_lr: 0.002412 loss: 2.1754 (2.1263) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [206] [200/312] eta: 0:01:23 lr: 0.002412 min_lr: 0.002412 loss: 2.1933 (2.1294) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [206] [210/312] eta: 0:01:15 lr: 0.002411 min_lr: 0.002411 loss: 2.2907 (2.1319) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [206] [220/312] eta: 0:01:07 lr: 0.002411 min_lr: 0.002411 loss: 2.2588 (2.1308) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [206] [230/312] eta: 0:01:00 lr: 0.002410 min_lr: 0.002410 loss: 2.1188 (2.1266) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [206] [240/312] eta: 0:00:52 lr: 0.002410 min_lr: 0.002410 loss: 2.0056 (2.1193) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [206] [250/312] eta: 0:00:45 lr: 0.002410 min_lr: 0.002410 loss: 2.0717 (2.1225) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [206] [260/312] eta: 0:00:38 lr: 0.002409 min_lr: 0.002409 loss: 2.1698 (2.1143) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [206] [270/312] eta: 0:00:30 lr: 0.002409 min_lr: 0.002409 loss: 2.1066 (2.1163) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [206] [280/312] eta: 0:00:23 lr: 0.002408 min_lr: 0.002408 loss: 2.1048 (2.1125) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [206] [290/312] eta: 0:00:16 lr: 0.002408 min_lr: 0.002408 loss: 1.9937 (2.1151) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [206] [300/312] eta: 0:00:08 lr: 0.002407 min_lr: 0.002407 loss: 2.2495 (2.1153) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [206] [310/312] eta: 0:00:01 lr: 0.002407 min_lr: 0.002407 loss: 2.1915 (2.1197) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [206] [311/312] eta: 0:00:00 lr: 0.002407 min_lr: 0.002407 loss: 2.1915 (2.1197) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [206] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.002407 min_lr: 0.002407 loss: 2.1915 (2.1099) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6130 (0.6130) acc1: 86.1979 (86.1979) acc5: 96.8750 (96.8750) time: 4.4708 data: 4.2516 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9355 (0.8559) acc1: 77.6042 (77.6320) acc5: 94.7917 (94.3360) time: 0.6480 data: 0.4725 max mem: 64948 Test: Total time: 0:00:06 (0.6732 s / it) * Acc@1 78.530 Acc@5 94.290 loss 0.830 Accuracy of the model on the 50000 test images: 78.5% Max accuracy: 78.68% Test: [0/9] eta: 0:00:43 loss: 0.5303 (0.5303) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.8768 data: 4.6600 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7208 (0.7083) acc1: 81.2500 (80.0000) acc5: 96.6146 (96.0640) time: 0.6939 data: 0.5179 max mem: 64948 Test: Total time: 0:00:06 (0.7018 s / it) * Acc@1 81.316 Acc@5 95.830 loss 0.687 Accuracy of the model EMA on 50000 test images: 81.3% Max EMA accuracy: 81.32% Epoch: [207] [ 0/312] eta: 0:51:30 lr: 0.002407 min_lr: 0.002407 loss: 1.7753 (1.7753) weight_decay: 0.0500 (0.0500) time: 9.9062 data: 8.6102 max mem: 64948 Epoch: [207] [ 10/312] eta: 0:07:59 lr: 0.002406 min_lr: 0.002406 loss: 1.9679 (1.9469) weight_decay: 0.0500 (0.0500) time: 1.5861 data: 0.7830 max mem: 64948 Epoch: [207] [ 20/312] eta: 0:05:39 lr: 0.002406 min_lr: 0.002406 loss: 2.0804 (2.1348) weight_decay: 0.0500 (0.0500) time: 0.7255 data: 0.0003 max mem: 64948 Epoch: [207] [ 30/312] eta: 0:04:45 lr: 0.002405 min_lr: 0.002405 loss: 2.3246 (2.2199) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [207] [ 40/312] eta: 0:04:14 lr: 0.002405 min_lr: 0.002405 loss: 2.2847 (2.1790) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [207] [ 50/312] eta: 0:03:53 lr: 0.002404 min_lr: 0.002404 loss: 2.1272 (2.1634) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [207] [ 60/312] eta: 0:03:36 lr: 0.002404 min_lr: 0.002404 loss: 2.1272 (2.1460) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [207] [ 70/312] eta: 0:03:21 lr: 0.002403 min_lr: 0.002403 loss: 2.1549 (2.1353) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [207] [ 80/312] eta: 0:03:09 lr: 0.002403 min_lr: 0.002403 loss: 2.1074 (2.1025) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [207] [ 90/312] eta: 0:02:58 lr: 0.002403 min_lr: 0.002403 loss: 2.0095 (2.1129) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [207] [100/312] eta: 0:02:48 lr: 0.002402 min_lr: 0.002402 loss: 2.2214 (2.1208) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [207] [110/312] eta: 0:02:38 lr: 0.002402 min_lr: 0.002402 loss: 2.1969 (2.1178) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [207] [120/312] eta: 0:02:29 lr: 0.002401 min_lr: 0.002401 loss: 2.1878 (2.1226) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [207] [130/312] eta: 0:02:20 lr: 0.002401 min_lr: 0.002401 loss: 2.2099 (2.1237) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [207] [140/312] eta: 0:02:11 lr: 0.002400 min_lr: 0.002400 loss: 2.1499 (2.1166) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0003 max mem: 64948 Epoch: [207] [150/312] eta: 0:02:03 lr: 0.002400 min_lr: 0.002400 loss: 2.0375 (2.1087) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [207] [160/312] eta: 0:01:54 lr: 0.002399 min_lr: 0.002399 loss: 2.1684 (2.1283) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [207] [170/312] eta: 0:01:46 lr: 0.002399 min_lr: 0.002399 loss: 2.2935 (2.1243) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [207] [180/312] eta: 0:01:38 lr: 0.002398 min_lr: 0.002398 loss: 2.0537 (2.1165) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [207] [190/312] eta: 0:01:31 lr: 0.002398 min_lr: 0.002398 loss: 1.9668 (2.1140) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [207] [200/312] eta: 0:01:23 lr: 0.002398 min_lr: 0.002398 loss: 2.1573 (2.1158) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [207] [210/312] eta: 0:01:15 lr: 0.002397 min_lr: 0.002397 loss: 2.2302 (2.1168) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [207] [220/312] eta: 0:01:08 lr: 0.002397 min_lr: 0.002397 loss: 2.0347 (2.1101) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [207] [230/312] eta: 0:01:00 lr: 0.002396 min_lr: 0.002396 loss: 2.0426 (2.1083) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [207] [240/312] eta: 0:00:52 lr: 0.002396 min_lr: 0.002396 loss: 2.0971 (2.1034) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [207] [250/312] eta: 0:00:45 lr: 0.002395 min_lr: 0.002395 loss: 2.0917 (2.1002) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [207] [260/312] eta: 0:00:38 lr: 0.002395 min_lr: 0.002395 loss: 2.0917 (2.1016) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [207] [270/312] eta: 0:00:30 lr: 0.002394 min_lr: 0.002394 loss: 2.1573 (2.1010) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [207] [280/312] eta: 0:00:23 lr: 0.002394 min_lr: 0.002394 loss: 2.1937 (2.1077) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [207] [290/312] eta: 0:00:16 lr: 0.002393 min_lr: 0.002393 loss: 2.2926 (2.1118) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [207] [300/312] eta: 0:00:08 lr: 0.002393 min_lr: 0.002393 loss: 2.2360 (2.1116) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [207] [310/312] eta: 0:00:01 lr: 0.002392 min_lr: 0.002392 loss: 2.2360 (2.1144) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [207] [311/312] eta: 0:00:00 lr: 0.002392 min_lr: 0.002392 loss: 2.2751 (2.1156) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [207] Total time: 0:03:47 (0.7287 s / it) Averaged stats: lr: 0.002392 min_lr: 0.002392 loss: 2.2751 (2.1158) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6416 (0.6416) acc1: 84.6354 (84.6354) acc5: 95.0521 (95.0521) time: 4.7414 data: 4.5222 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9065 (0.8871) acc1: 78.6458 (77.6320) acc5: 94.3396 (94.0480) time: 0.6788 data: 0.5026 max mem: 64948 Test: Total time: 0:00:06 (0.7008 s / it) * Acc@1 78.214 Acc@5 94.138 loss 0.846 Accuracy of the model on the 50000 test images: 78.2% Max accuracy: 78.68% Test: [0/9] eta: 0:00:42 loss: 0.5295 (0.5295) acc1: 84.8958 (84.8958) acc5: 97.6562 (97.6562) time: 4.6699 data: 4.4661 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7208 (0.7078) acc1: 81.2500 (79.9360) acc5: 96.6146 (96.0640) time: 0.6701 data: 0.4963 max mem: 64948 Test: Total time: 0:00:06 (0.6794 s / it) * Acc@1 81.310 Acc@5 95.830 loss 0.686 Accuracy of the model EMA on 50000 test images: 81.3% Epoch: [208] [ 0/312] eta: 0:53:09 lr: 0.002392 min_lr: 0.002392 loss: 1.9538 (1.9538) weight_decay: 0.0500 (0.0500) time: 10.2231 data: 7.4519 max mem: 64948 Epoch: [208] [ 10/312] eta: 0:08:01 lr: 0.002392 min_lr: 0.002392 loss: 2.2087 (2.1767) weight_decay: 0.0500 (0.0500) time: 1.5929 data: 0.6779 max mem: 64948 Epoch: [208] [ 20/312] eta: 0:05:40 lr: 0.002391 min_lr: 0.002391 loss: 2.2087 (2.1510) weight_decay: 0.0500 (0.0500) time: 0.7129 data: 0.0004 max mem: 64948 Epoch: [208] [ 30/312] eta: 0:04:45 lr: 0.002391 min_lr: 0.002391 loss: 2.1584 (2.1282) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [208] [ 40/312] eta: 0:04:14 lr: 0.002391 min_lr: 0.002391 loss: 2.2335 (2.1518) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [208] [ 50/312] eta: 0:03:52 lr: 0.002390 min_lr: 0.002390 loss: 2.2335 (2.1240) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [208] [ 60/312] eta: 0:03:35 lr: 0.002390 min_lr: 0.002390 loss: 2.1290 (2.1423) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [208] [ 70/312] eta: 0:03:21 lr: 0.002389 min_lr: 0.002389 loss: 2.1290 (2.1199) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [208] [ 80/312] eta: 0:03:09 lr: 0.002389 min_lr: 0.002389 loss: 1.9720 (2.1231) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0003 max mem: 64948 Epoch: [208] [ 90/312] eta: 0:02:58 lr: 0.002388 min_lr: 0.002388 loss: 2.2089 (2.1219) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [208] [100/312] eta: 0:02:48 lr: 0.002388 min_lr: 0.002388 loss: 2.1290 (2.1257) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [208] [110/312] eta: 0:02:38 lr: 0.002387 min_lr: 0.002387 loss: 2.1725 (2.1231) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [208] [120/312] eta: 0:02:29 lr: 0.002387 min_lr: 0.002387 loss: 1.9665 (2.1080) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [208] [130/312] eta: 0:02:20 lr: 0.002386 min_lr: 0.002386 loss: 2.0839 (2.1171) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [208] [140/312] eta: 0:02:11 lr: 0.002386 min_lr: 0.002386 loss: 2.2029 (2.1071) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [208] [150/312] eta: 0:02:03 lr: 0.002385 min_lr: 0.002385 loss: 2.0673 (2.1090) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [208] [160/312] eta: 0:01:55 lr: 0.002385 min_lr: 0.002385 loss: 2.0933 (2.1102) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [208] [170/312] eta: 0:01:46 lr: 0.002385 min_lr: 0.002385 loss: 2.0933 (2.1170) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [208] [180/312] eta: 0:01:38 lr: 0.002384 min_lr: 0.002384 loss: 1.9944 (2.1073) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [208] [190/312] eta: 0:01:31 lr: 0.002384 min_lr: 0.002384 loss: 2.1672 (2.1155) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [208] [200/312] eta: 0:01:23 lr: 0.002383 min_lr: 0.002383 loss: 2.2534 (2.1127) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [208] [210/312] eta: 0:01:15 lr: 0.002383 min_lr: 0.002383 loss: 2.2910 (2.1177) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [208] [220/312] eta: 0:01:08 lr: 0.002382 min_lr: 0.002382 loss: 2.3052 (2.1225) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [208] [230/312] eta: 0:01:00 lr: 0.002382 min_lr: 0.002382 loss: 2.1428 (2.1140) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [208] [240/312] eta: 0:00:53 lr: 0.002381 min_lr: 0.002381 loss: 1.7632 (2.1093) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [208] [250/312] eta: 0:00:45 lr: 0.002381 min_lr: 0.002381 loss: 2.0933 (2.1083) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [208] [260/312] eta: 0:00:38 lr: 0.002380 min_lr: 0.002380 loss: 2.1108 (2.1084) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [208] [270/312] eta: 0:00:30 lr: 0.002380 min_lr: 0.002380 loss: 2.2029 (2.1105) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [208] [280/312] eta: 0:00:23 lr: 0.002380 min_lr: 0.002380 loss: 2.2050 (2.1150) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [208] [290/312] eta: 0:00:16 lr: 0.002379 min_lr: 0.002379 loss: 2.2884 (2.1232) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [208] [300/312] eta: 0:00:08 lr: 0.002379 min_lr: 0.002379 loss: 2.2666 (2.1251) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [208] [310/312] eta: 0:00:01 lr: 0.002378 min_lr: 0.002378 loss: 2.2666 (2.1329) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [208] [311/312] eta: 0:00:00 lr: 0.002378 min_lr: 0.002378 loss: 2.2666 (2.1329) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [208] Total time: 0:03:47 (0.7297 s / it) Averaged stats: lr: 0.002378 min_lr: 0.002378 loss: 2.2666 (2.1123) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6000 (0.6000) acc1: 84.1146 (84.1146) acc5: 97.3958 (97.3958) time: 4.6754 data: 4.4710 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9448 (0.8518) acc1: 78.1250 (78.3040) acc5: 95.0521 (94.4960) time: 0.6707 data: 0.4969 max mem: 64948 Test: Total time: 0:00:06 (0.6816 s / it) * Acc@1 78.720 Acc@5 94.470 loss 0.828 Accuracy of the model on the 50000 test images: 78.7% Max accuracy: 78.72% Test: [0/9] eta: 0:00:39 loss: 0.5290 (0.5290) acc1: 84.8958 (84.8958) acc5: 97.6562 (97.6562) time: 4.3918 data: 4.1861 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7213 (0.7075) acc1: 80.9896 (79.9040) acc5: 96.6146 (96.0320) time: 0.6393 data: 0.4652 max mem: 64948 Test: Total time: 0:00:05 (0.6467 s / it) * Acc@1 81.342 Acc@5 95.820 loss 0.686 Accuracy of the model EMA on 50000 test images: 81.3% Max EMA accuracy: 81.34% Epoch: [209] [ 0/312] eta: 0:49:14 lr: 0.002378 min_lr: 0.002378 loss: 2.6319 (2.6319) weight_decay: 0.0500 (0.0500) time: 9.4688 data: 8.6110 max mem: 64948 Epoch: [209] [ 10/312] eta: 0:07:38 lr: 0.002378 min_lr: 0.002378 loss: 1.9766 (2.0670) weight_decay: 0.0500 (0.0500) time: 1.5181 data: 0.7832 max mem: 64948 Epoch: [209] [ 20/312] eta: 0:05:28 lr: 0.002377 min_lr: 0.002377 loss: 2.2179 (2.1653) weight_decay: 0.0500 (0.0500) time: 0.7090 data: 0.0004 max mem: 64948 Epoch: [209] [ 30/312] eta: 0:04:38 lr: 0.002377 min_lr: 0.002377 loss: 2.2103 (2.1183) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [209] [ 40/312] eta: 0:04:09 lr: 0.002376 min_lr: 0.002376 loss: 1.9690 (2.1068) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0003 max mem: 64948 Epoch: [209] [ 50/312] eta: 0:03:49 lr: 0.002376 min_lr: 0.002376 loss: 2.0950 (2.1204) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [209] [ 60/312] eta: 0:03:32 lr: 0.002375 min_lr: 0.002375 loss: 2.1624 (2.1345) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [209] [ 70/312] eta: 0:03:19 lr: 0.002375 min_lr: 0.002375 loss: 2.2176 (2.1375) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [209] [ 80/312] eta: 0:03:07 lr: 0.002374 min_lr: 0.002374 loss: 2.2176 (2.1306) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [209] [ 90/312] eta: 0:02:56 lr: 0.002374 min_lr: 0.002374 loss: 2.1396 (2.1282) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [209] [100/312] eta: 0:02:46 lr: 0.002373 min_lr: 0.002373 loss: 2.1199 (2.1136) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [209] [110/312] eta: 0:02:36 lr: 0.002373 min_lr: 0.002373 loss: 2.1250 (2.1152) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [209] [120/312] eta: 0:02:27 lr: 0.002373 min_lr: 0.002373 loss: 2.1097 (2.1047) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [209] [130/312] eta: 0:02:19 lr: 0.002372 min_lr: 0.002372 loss: 2.0695 (2.0925) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [209] [140/312] eta: 0:02:10 lr: 0.002372 min_lr: 0.002372 loss: 2.2378 (2.1019) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [209] [150/312] eta: 0:02:02 lr: 0.002371 min_lr: 0.002371 loss: 2.2378 (2.0988) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [209] [160/312] eta: 0:01:54 lr: 0.002371 min_lr: 0.002371 loss: 1.9743 (2.0864) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [209] [170/312] eta: 0:01:46 lr: 0.002370 min_lr: 0.002370 loss: 2.2667 (2.0970) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [209] [180/312] eta: 0:01:38 lr: 0.002370 min_lr: 0.002370 loss: 2.2537 (2.0940) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [209] [190/312] eta: 0:01:30 lr: 0.002369 min_lr: 0.002369 loss: 2.0496 (2.0973) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [209] [200/312] eta: 0:01:22 lr: 0.002369 min_lr: 0.002369 loss: 2.1434 (2.1041) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [209] [210/312] eta: 0:01:15 lr: 0.002368 min_lr: 0.002368 loss: 2.3194 (2.1131) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [209] [220/312] eta: 0:01:07 lr: 0.002368 min_lr: 0.002368 loss: 2.3496 (2.1213) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [209] [230/312] eta: 0:01:00 lr: 0.002367 min_lr: 0.002367 loss: 2.3259 (2.1280) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [209] [240/312] eta: 0:00:52 lr: 0.002367 min_lr: 0.002367 loss: 2.1371 (2.1280) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [209] [250/312] eta: 0:00:45 lr: 0.002367 min_lr: 0.002367 loss: 2.2345 (2.1328) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [209] [260/312] eta: 0:00:37 lr: 0.002366 min_lr: 0.002366 loss: 2.3367 (2.1392) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [209] [270/312] eta: 0:00:30 lr: 0.002366 min_lr: 0.002366 loss: 2.1872 (2.1356) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [209] [280/312] eta: 0:00:23 lr: 0.002365 min_lr: 0.002365 loss: 2.2487 (2.1391) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [209] [290/312] eta: 0:00:15 lr: 0.002365 min_lr: 0.002365 loss: 2.2451 (2.1396) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [209] [300/312] eta: 0:00:08 lr: 0.002364 min_lr: 0.002364 loss: 2.2451 (2.1441) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [209] [310/312] eta: 0:00:01 lr: 0.002364 min_lr: 0.002364 loss: 2.2886 (2.1464) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [209] [311/312] eta: 0:00:00 lr: 0.002364 min_lr: 0.002364 loss: 2.2821 (2.1452) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [209] Total time: 0:03:46 (0.7267 s / it) Averaged stats: lr: 0.002364 min_lr: 0.002364 loss: 2.2821 (2.1200) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6350 (0.6350) acc1: 83.8542 (83.8542) acc5: 96.8750 (96.8750) time: 4.4507 data: 4.2464 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9010 (0.8564) acc1: 78.6458 (77.5360) acc5: 94.2708 (94.3040) time: 0.6458 data: 0.4719 max mem: 64948 Test: Total time: 0:00:06 (0.6674 s / it) * Acc@1 78.654 Acc@5 94.464 loss 0.818 Accuracy of the model on the 50000 test images: 78.7% Max accuracy: 78.72% Test: [0/9] eta: 0:00:41 loss: 0.5279 (0.5279) acc1: 84.8958 (84.8958) acc5: 97.6562 (97.6562) time: 4.5898 data: 4.3716 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7214 (0.7067) acc1: 80.9896 (79.9360) acc5: 96.6146 (96.0320) time: 0.6733 data: 0.4978 max mem: 64948 Test: Total time: 0:00:06 (0.6829 s / it) * Acc@1 81.382 Acc@5 95.810 loss 0.685 Accuracy of the model EMA on 50000 test images: 81.4% Max EMA accuracy: 81.38% Epoch: [210] [ 0/312] eta: 0:48:07 lr: 0.002364 min_lr: 0.002364 loss: 2.5553 (2.5553) weight_decay: 0.0500 (0.0500) time: 9.2551 data: 8.3904 max mem: 64948 Epoch: [210] [ 10/312] eta: 0:07:45 lr: 0.002363 min_lr: 0.002363 loss: 2.2431 (2.1580) weight_decay: 0.0500 (0.0500) time: 1.5427 data: 0.7631 max mem: 64948 Epoch: [210] [ 20/312] eta: 0:05:32 lr: 0.002363 min_lr: 0.002363 loss: 2.2717 (2.2446) weight_decay: 0.0500 (0.0500) time: 0.7325 data: 0.0004 max mem: 64948 Epoch: [210] [ 30/312] eta: 0:04:40 lr: 0.002362 min_lr: 0.002362 loss: 2.2564 (2.1944) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [210] [ 40/312] eta: 0:04:10 lr: 0.002362 min_lr: 0.002362 loss: 2.1652 (2.1829) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [210] [ 50/312] eta: 0:03:50 lr: 0.002361 min_lr: 0.002361 loss: 2.1858 (2.1793) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [210] [ 60/312] eta: 0:03:34 lr: 0.002361 min_lr: 0.002361 loss: 2.1968 (2.1541) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [210] [ 70/312] eta: 0:03:20 lr: 0.002360 min_lr: 0.002360 loss: 2.1446 (2.1326) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [210] [ 80/312] eta: 0:03:08 lr: 0.002360 min_lr: 0.002360 loss: 1.9669 (2.1083) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [210] [ 90/312] eta: 0:02:57 lr: 0.002360 min_lr: 0.002360 loss: 2.1160 (2.1224) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [210] [100/312] eta: 0:02:47 lr: 0.002359 min_lr: 0.002359 loss: 2.1971 (2.1130) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [210] [110/312] eta: 0:02:37 lr: 0.002359 min_lr: 0.002359 loss: 2.0706 (2.1044) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [210] [120/312] eta: 0:02:28 lr: 0.002358 min_lr: 0.002358 loss: 2.0340 (2.0972) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [210] [130/312] eta: 0:02:19 lr: 0.002358 min_lr: 0.002358 loss: 1.9996 (2.0867) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [210] [140/312] eta: 0:02:11 lr: 0.002357 min_lr: 0.002357 loss: 1.9996 (2.0765) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [210] [150/312] eta: 0:02:02 lr: 0.002357 min_lr: 0.002357 loss: 2.2412 (2.0893) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [210] [160/312] eta: 0:01:54 lr: 0.002356 min_lr: 0.002356 loss: 2.1916 (2.0894) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [210] [170/312] eta: 0:01:46 lr: 0.002356 min_lr: 0.002356 loss: 2.1101 (2.0892) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [210] [180/312] eta: 0:01:38 lr: 0.002355 min_lr: 0.002355 loss: 2.1051 (2.0843) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [210] [190/312] eta: 0:01:30 lr: 0.002355 min_lr: 0.002355 loss: 2.1051 (2.0858) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [210] [200/312] eta: 0:01:23 lr: 0.002354 min_lr: 0.002354 loss: 2.1915 (2.0874) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [210] [210/312] eta: 0:01:15 lr: 0.002354 min_lr: 0.002354 loss: 2.1774 (2.0896) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [210] [220/312] eta: 0:01:07 lr: 0.002354 min_lr: 0.002354 loss: 1.9471 (2.0820) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [210] [230/312] eta: 0:01:00 lr: 0.002353 min_lr: 0.002353 loss: 1.9000 (2.0844) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [210] [240/312] eta: 0:00:52 lr: 0.002353 min_lr: 0.002353 loss: 2.3066 (2.0917) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [210] [250/312] eta: 0:00:45 lr: 0.002352 min_lr: 0.002352 loss: 2.3278 (2.1005) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [210] [260/312] eta: 0:00:38 lr: 0.002352 min_lr: 0.002352 loss: 2.1250 (2.0871) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [210] [270/312] eta: 0:00:30 lr: 0.002351 min_lr: 0.002351 loss: 2.2426 (2.0958) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [210] [280/312] eta: 0:00:23 lr: 0.002351 min_lr: 0.002351 loss: 2.2676 (2.0994) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [210] [290/312] eta: 0:00:15 lr: 0.002350 min_lr: 0.002350 loss: 2.1952 (2.0995) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [210] [300/312] eta: 0:00:08 lr: 0.002350 min_lr: 0.002350 loss: 2.0028 (2.0963) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [210] [310/312] eta: 0:00:01 lr: 0.002349 min_lr: 0.002349 loss: 1.8215 (2.0901) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [210] [311/312] eta: 0:00:00 lr: 0.002349 min_lr: 0.002349 loss: 1.8175 (2.0880) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [210] Total time: 0:03:46 (0.7275 s / it) Averaged stats: lr: 0.002349 min_lr: 0.002349 loss: 1.8175 (2.1051) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6010 (0.6010) acc1: 84.1146 (84.1146) acc5: 96.8750 (96.8750) time: 4.5009 data: 4.2852 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9021 (0.8222) acc1: 77.6042 (78.1760) acc5: 94.2708 (94.3040) time: 0.6514 data: 0.4762 max mem: 64948 Test: Total time: 0:00:06 (0.6748 s / it) * Acc@1 79.000 Acc@5 94.596 loss 0.809 Accuracy of the model on the 50000 test images: 79.0% Max accuracy: 79.00% Test: [0/9] eta: 0:00:40 loss: 0.5272 (0.5272) acc1: 84.8958 (84.8958) acc5: 97.6562 (97.6562) time: 4.5368 data: 4.3188 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7215 (0.7062) acc1: 80.7292 (79.9680) acc5: 96.6146 (96.0320) time: 0.6555 data: 0.4800 max mem: 64948 Test: Total time: 0:00:05 (0.6644 s / it) * Acc@1 81.398 Acc@5 95.814 loss 0.684 Accuracy of the model EMA on 50000 test images: 81.4% Max EMA accuracy: 81.40% Epoch: [211] [ 0/312] eta: 0:55:15 lr: 0.002349 min_lr: 0.002349 loss: 1.6447 (1.6447) weight_decay: 0.0500 (0.0500) time: 10.6280 data: 9.8316 max mem: 64948 Epoch: [211] [ 10/312] eta: 0:08:12 lr: 0.002349 min_lr: 0.002349 loss: 1.9842 (2.0546) weight_decay: 0.0500 (0.0500) time: 1.6320 data: 0.8941 max mem: 64948 Epoch: [211] [ 20/312] eta: 0:05:46 lr: 0.002348 min_lr: 0.002348 loss: 2.1005 (2.1333) weight_decay: 0.0500 (0.0500) time: 0.7131 data: 0.0003 max mem: 64948 Epoch: [211] [ 30/312] eta: 0:04:49 lr: 0.002348 min_lr: 0.002348 loss: 2.1748 (2.1215) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [211] [ 40/312] eta: 0:04:17 lr: 0.002347 min_lr: 0.002347 loss: 2.1819 (2.1243) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [211] [ 50/312] eta: 0:03:54 lr: 0.002347 min_lr: 0.002347 loss: 2.2129 (2.1366) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [211] [ 60/312] eta: 0:03:37 lr: 0.002347 min_lr: 0.002347 loss: 2.1130 (2.1224) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [211] [ 70/312] eta: 0:03:23 lr: 0.002346 min_lr: 0.002346 loss: 1.9311 (2.1264) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [211] [ 80/312] eta: 0:03:10 lr: 0.002346 min_lr: 0.002346 loss: 2.2684 (2.1488) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [211] [ 90/312] eta: 0:02:59 lr: 0.002345 min_lr: 0.002345 loss: 2.2883 (2.1527) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [211] [100/312] eta: 0:02:49 lr: 0.002345 min_lr: 0.002345 loss: 2.1844 (2.1431) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [211] [110/312] eta: 0:02:39 lr: 0.002344 min_lr: 0.002344 loss: 2.1844 (2.1520) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [211] [120/312] eta: 0:02:29 lr: 0.002344 min_lr: 0.002344 loss: 2.1576 (2.1413) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [211] [130/312] eta: 0:02:20 lr: 0.002343 min_lr: 0.002343 loss: 1.9566 (2.1307) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [211] [140/312] eta: 0:02:12 lr: 0.002343 min_lr: 0.002343 loss: 1.9566 (2.1213) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [211] [150/312] eta: 0:02:03 lr: 0.002342 min_lr: 0.002342 loss: 1.9174 (2.1096) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [211] [160/312] eta: 0:01:55 lr: 0.002342 min_lr: 0.002342 loss: 2.1673 (2.1213) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [211] [170/312] eta: 0:01:47 lr: 0.002341 min_lr: 0.002341 loss: 2.1909 (2.1195) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [211] [180/312] eta: 0:01:39 lr: 0.002341 min_lr: 0.002341 loss: 2.1557 (2.1219) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [211] [190/312] eta: 0:01:31 lr: 0.002341 min_lr: 0.002341 loss: 2.1600 (2.1168) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [211] [200/312] eta: 0:01:23 lr: 0.002340 min_lr: 0.002340 loss: 2.1286 (2.1150) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [211] [210/312] eta: 0:01:15 lr: 0.002340 min_lr: 0.002340 loss: 2.2578 (2.1164) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [211] [220/312] eta: 0:01:08 lr: 0.002339 min_lr: 0.002339 loss: 2.2659 (2.1170) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [211] [230/312] eta: 0:01:00 lr: 0.002339 min_lr: 0.002339 loss: 2.2325 (2.1133) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [211] [240/312] eta: 0:00:53 lr: 0.002338 min_lr: 0.002338 loss: 2.0817 (2.1112) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [211] [250/312] eta: 0:00:45 lr: 0.002338 min_lr: 0.002338 loss: 2.1456 (2.1199) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [211] [260/312] eta: 0:00:38 lr: 0.002337 min_lr: 0.002337 loss: 2.1951 (2.1217) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [211] [270/312] eta: 0:00:30 lr: 0.002337 min_lr: 0.002337 loss: 2.1953 (2.1237) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [211] [280/312] eta: 0:00:23 lr: 0.002336 min_lr: 0.002336 loss: 2.2037 (2.1285) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [211] [290/312] eta: 0:00:16 lr: 0.002336 min_lr: 0.002336 loss: 2.0765 (2.1215) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [211] [300/312] eta: 0:00:08 lr: 0.002335 min_lr: 0.002335 loss: 2.0394 (2.1231) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [211] [310/312] eta: 0:00:01 lr: 0.002335 min_lr: 0.002335 loss: 2.1895 (2.1248) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [211] [311/312] eta: 0:00:00 lr: 0.002335 min_lr: 0.002335 loss: 2.1895 (2.1257) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [211] Total time: 0:03:47 (0.7306 s / it) Averaged stats: lr: 0.002335 min_lr: 0.002335 loss: 2.1895 (2.1135) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6216 (0.6216) acc1: 84.1146 (84.1146) acc5: 96.8750 (96.8750) time: 4.4769 data: 4.2670 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8860 (0.8444) acc1: 79.6875 (78.4000) acc5: 94.7917 (94.5920) time: 0.6487 data: 0.4742 max mem: 64948 Test: Total time: 0:00:06 (0.6699 s / it) * Acc@1 78.760 Acc@5 94.478 loss 0.822 Accuracy of the model on the 50000 test images: 78.8% Max accuracy: 79.00% Test: [0/9] eta: 0:00:46 loss: 0.5266 (0.5266) acc1: 84.8958 (84.8958) acc5: 97.6562 (97.6562) time: 5.1424 data: 4.9363 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7207 (0.7055) acc1: 80.7292 (79.9360) acc5: 96.8750 (96.0960) time: 0.7227 data: 0.5486 max mem: 64948 Test: Total time: 0:00:06 (0.7338 s / it) * Acc@1 81.418 Acc@5 95.828 loss 0.683 Accuracy of the model EMA on 50000 test images: 81.4% Max EMA accuracy: 81.42% Epoch: [212] [ 0/312] eta: 0:51:00 lr: 0.002335 min_lr: 0.002335 loss: 2.3223 (2.3223) weight_decay: 0.0500 (0.0500) time: 9.8085 data: 7.0420 max mem: 64948 Epoch: [212] [ 10/312] eta: 0:07:45 lr: 0.002334 min_lr: 0.002334 loss: 2.1785 (2.1198) weight_decay: 0.0500 (0.0500) time: 1.5414 data: 0.6406 max mem: 64948 Epoch: [212] [ 20/312] eta: 0:05:32 lr: 0.002334 min_lr: 0.002334 loss: 2.1785 (2.1709) weight_decay: 0.0500 (0.0500) time: 0.7061 data: 0.0004 max mem: 64948 Epoch: [212] [ 30/312] eta: 0:04:41 lr: 0.002334 min_lr: 0.002334 loss: 2.2624 (2.1639) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [212] [ 40/312] eta: 0:04:11 lr: 0.002333 min_lr: 0.002333 loss: 2.1922 (2.1215) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [212] [ 50/312] eta: 0:03:50 lr: 0.002333 min_lr: 0.002333 loss: 2.1712 (2.1266) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [212] [ 60/312] eta: 0:03:33 lr: 0.002332 min_lr: 0.002332 loss: 2.1544 (2.1246) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [212] [ 70/312] eta: 0:03:20 lr: 0.002332 min_lr: 0.002332 loss: 2.1475 (2.1205) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [212] [ 80/312] eta: 0:03:08 lr: 0.002331 min_lr: 0.002331 loss: 2.1240 (2.0992) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [212] [ 90/312] eta: 0:02:57 lr: 0.002331 min_lr: 0.002331 loss: 2.1498 (2.1180) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [212] [100/312] eta: 0:02:47 lr: 0.002330 min_lr: 0.002330 loss: 2.1991 (2.1264) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [212] [110/312] eta: 0:02:37 lr: 0.002330 min_lr: 0.002330 loss: 2.1674 (2.1176) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [212] [120/312] eta: 0:02:28 lr: 0.002329 min_lr: 0.002329 loss: 2.1674 (2.1105) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [212] [130/312] eta: 0:02:19 lr: 0.002329 min_lr: 0.002329 loss: 2.1851 (2.1047) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [212] [140/312] eta: 0:02:10 lr: 0.002328 min_lr: 0.002328 loss: 2.0110 (2.1022) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [212] [150/312] eta: 0:02:02 lr: 0.002328 min_lr: 0.002328 loss: 1.9509 (2.0994) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [212] [160/312] eta: 0:01:54 lr: 0.002328 min_lr: 0.002328 loss: 2.2386 (2.1060) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [212] [170/312] eta: 0:01:46 lr: 0.002327 min_lr: 0.002327 loss: 2.0599 (2.0897) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [212] [180/312] eta: 0:01:38 lr: 0.002327 min_lr: 0.002327 loss: 1.8028 (2.0819) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [212] [190/312] eta: 0:01:30 lr: 0.002326 min_lr: 0.002326 loss: 1.9510 (2.0810) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [212] [200/312] eta: 0:01:23 lr: 0.002326 min_lr: 0.002326 loss: 2.1655 (2.0858) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [212] [210/312] eta: 0:01:15 lr: 0.002325 min_lr: 0.002325 loss: 2.1605 (2.0865) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [212] [220/312] eta: 0:01:07 lr: 0.002325 min_lr: 0.002325 loss: 2.2394 (2.0952) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [212] [230/312] eta: 0:01:00 lr: 0.002324 min_lr: 0.002324 loss: 2.3494 (2.0997) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [212] [240/312] eta: 0:00:52 lr: 0.002324 min_lr: 0.002324 loss: 2.2824 (2.0970) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [212] [250/312] eta: 0:00:45 lr: 0.002323 min_lr: 0.002323 loss: 2.2285 (2.1033) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [212] [260/312] eta: 0:00:37 lr: 0.002323 min_lr: 0.002323 loss: 2.1301 (2.1026) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [212] [270/312] eta: 0:00:30 lr: 0.002322 min_lr: 0.002322 loss: 2.1150 (2.1062) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [212] [280/312] eta: 0:00:23 lr: 0.002322 min_lr: 0.002322 loss: 2.1150 (2.1012) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [212] [290/312] eta: 0:00:15 lr: 0.002322 min_lr: 0.002322 loss: 2.1380 (2.1069) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [212] [300/312] eta: 0:00:08 lr: 0.002321 min_lr: 0.002321 loss: 2.1638 (2.1065) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [212] [310/312] eta: 0:00:01 lr: 0.002321 min_lr: 0.002321 loss: 2.0897 (2.1010) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0002 max mem: 64948 Epoch: [212] [311/312] eta: 0:00:00 lr: 0.002321 min_lr: 0.002321 loss: 2.0897 (2.1015) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0002 max mem: 64948 Epoch: [212] Total time: 0:03:46 (0.7272 s / it) Averaged stats: lr: 0.002321 min_lr: 0.002321 loss: 2.0897 (2.1021) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.6021 (0.6021) acc1: 83.8542 (83.8542) acc5: 96.6146 (96.6146) time: 4.2843 data: 4.0761 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9134 (0.8443) acc1: 77.3438 (77.8560) acc5: 94.7917 (94.1440) time: 0.6325 data: 0.4581 max mem: 64948 Test: Total time: 0:00:05 (0.6583 s / it) * Acc@1 78.480 Acc@5 94.390 loss 0.820 Accuracy of the model on the 50000 test images: 78.5% Max accuracy: 79.00% Test: [0/9] eta: 0:00:46 loss: 0.5257 (0.5257) acc1: 84.8958 (84.8958) acc5: 97.6562 (97.6562) time: 5.2081 data: 4.9909 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7202 (0.7050) acc1: 80.7292 (80.0000) acc5: 96.8750 (96.0960) time: 0.7300 data: 0.5546 max mem: 64948 Test: Total time: 0:00:06 (0.7375 s / it) * Acc@1 81.440 Acc@5 95.814 loss 0.682 Accuracy of the model EMA on 50000 test images: 81.4% Max EMA accuracy: 81.44% Epoch: [213] [ 0/312] eta: 0:48:37 lr: 0.002321 min_lr: 0.002321 loss: 2.5820 (2.5820) weight_decay: 0.0500 (0.0500) time: 9.3515 data: 8.5966 max mem: 64948 Epoch: [213] [ 10/312] eta: 0:07:45 lr: 0.002320 min_lr: 0.002320 loss: 2.1469 (2.1016) weight_decay: 0.0500 (0.0500) time: 1.5402 data: 0.7819 max mem: 64948 Epoch: [213] [ 20/312] eta: 0:05:32 lr: 0.002320 min_lr: 0.002320 loss: 2.0373 (2.0256) weight_decay: 0.0500 (0.0500) time: 0.7263 data: 0.0004 max mem: 64948 Epoch: [213] [ 30/312] eta: 0:04:40 lr: 0.002319 min_lr: 0.002319 loss: 2.1185 (2.0598) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [213] [ 40/312] eta: 0:04:10 lr: 0.002319 min_lr: 0.002319 loss: 2.0639 (2.0303) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [213] [ 50/312] eta: 0:03:49 lr: 0.002318 min_lr: 0.002318 loss: 2.0152 (2.0605) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [213] [ 60/312] eta: 0:03:33 lr: 0.002318 min_lr: 0.002318 loss: 2.1966 (2.0710) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [213] [ 70/312] eta: 0:03:20 lr: 0.002317 min_lr: 0.002317 loss: 2.2127 (2.0919) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [213] [ 80/312] eta: 0:03:08 lr: 0.002317 min_lr: 0.002317 loss: 2.2127 (2.0868) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [213] [ 90/312] eta: 0:02:57 lr: 0.002316 min_lr: 0.002316 loss: 2.0747 (2.0923) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [213] [100/312] eta: 0:02:47 lr: 0.002316 min_lr: 0.002316 loss: 2.1360 (2.0967) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [213] [110/312] eta: 0:02:37 lr: 0.002315 min_lr: 0.002315 loss: 2.1451 (2.0970) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [213] [120/312] eta: 0:02:28 lr: 0.002315 min_lr: 0.002315 loss: 2.1451 (2.1020) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [213] [130/312] eta: 0:02:19 lr: 0.002314 min_lr: 0.002314 loss: 2.2462 (2.1049) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [213] [140/312] eta: 0:02:10 lr: 0.002314 min_lr: 0.002314 loss: 2.1900 (2.0916) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [213] [150/312] eta: 0:02:02 lr: 0.002314 min_lr: 0.002314 loss: 1.9311 (2.0865) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [213] [160/312] eta: 0:01:54 lr: 0.002313 min_lr: 0.002313 loss: 2.0595 (2.0894) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [213] [170/312] eta: 0:01:46 lr: 0.002313 min_lr: 0.002313 loss: 1.8744 (2.0771) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [213] [180/312] eta: 0:01:38 lr: 0.002312 min_lr: 0.002312 loss: 2.0374 (2.0766) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0005 max mem: 64948 Epoch: [213] [190/312] eta: 0:01:30 lr: 0.002312 min_lr: 0.002312 loss: 2.1398 (2.0771) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [213] [200/312] eta: 0:01:22 lr: 0.002311 min_lr: 0.002311 loss: 2.1787 (2.0838) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [213] [210/312] eta: 0:01:15 lr: 0.002311 min_lr: 0.002311 loss: 2.1555 (2.0829) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [213] [220/312] eta: 0:01:07 lr: 0.002310 min_lr: 0.002310 loss: 2.1388 (2.0885) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [213] [230/312] eta: 0:01:00 lr: 0.002310 min_lr: 0.002310 loss: 2.1388 (2.0923) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [213] [240/312] eta: 0:00:52 lr: 0.002309 min_lr: 0.002309 loss: 2.1204 (2.0865) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [213] [250/312] eta: 0:00:45 lr: 0.002309 min_lr: 0.002309 loss: 2.1340 (2.0935) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [213] [260/312] eta: 0:00:37 lr: 0.002308 min_lr: 0.002308 loss: 2.2264 (2.0962) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [213] [270/312] eta: 0:00:30 lr: 0.002308 min_lr: 0.002308 loss: 2.2239 (2.0971) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [213] [280/312] eta: 0:00:23 lr: 0.002308 min_lr: 0.002308 loss: 2.1164 (2.0975) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0010 max mem: 64948 Epoch: [213] [290/312] eta: 0:00:15 lr: 0.002307 min_lr: 0.002307 loss: 2.1195 (2.0974) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0009 max mem: 64948 Epoch: [213] [300/312] eta: 0:00:08 lr: 0.002307 min_lr: 0.002307 loss: 2.1167 (2.0956) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [213] [310/312] eta: 0:00:01 lr: 0.002306 min_lr: 0.002306 loss: 2.1837 (2.0981) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [213] [311/312] eta: 0:00:00 lr: 0.002306 min_lr: 0.002306 loss: 2.1837 (2.0979) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [213] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.002306 min_lr: 0.002306 loss: 2.1837 (2.1000) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5387 (0.5387) acc1: 86.7188 (86.7188) acc5: 97.1354 (97.1354) time: 4.6339 data: 4.4146 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8708 (0.8198) acc1: 77.3438 (77.8240) acc5: 94.7917 (94.5920) time: 0.6663 data: 0.4906 max mem: 64948 Test: Total time: 0:00:06 (0.6896 s / it) * Acc@1 79.116 Acc@5 94.592 loss 0.800 Accuracy of the model on the 50000 test images: 79.1% Max accuracy: 79.12% Test: [0/9] eta: 0:00:42 loss: 0.5246 (0.5246) acc1: 85.4167 (85.4167) acc5: 97.6562 (97.6562) time: 4.7728 data: 4.5571 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7196 (0.7045) acc1: 80.9896 (80.1280) acc5: 96.8750 (96.0960) time: 0.6816 data: 0.5064 max mem: 64948 Test: Total time: 0:00:06 (0.6890 s / it) * Acc@1 81.462 Acc@5 95.804 loss 0.682 Accuracy of the model EMA on 50000 test images: 81.5% Max EMA accuracy: 81.46% Epoch: [214] [ 0/312] eta: 0:59:05 lr: 0.002306 min_lr: 0.002306 loss: 2.2934 (2.2934) weight_decay: 0.0500 (0.0500) time: 11.3640 data: 10.6324 max mem: 64948 Epoch: [214] [ 10/312] eta: 0:08:23 lr: 0.002306 min_lr: 0.002306 loss: 2.1928 (2.0801) weight_decay: 0.0500 (0.0500) time: 1.6684 data: 0.9669 max mem: 64948 Epoch: [214] [ 20/312] eta: 0:05:51 lr: 0.002305 min_lr: 0.002305 loss: 2.1597 (2.1015) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [214] [ 30/312] eta: 0:04:53 lr: 0.002305 min_lr: 0.002305 loss: 2.1230 (2.0979) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [214] [ 40/312] eta: 0:04:20 lr: 0.002304 min_lr: 0.002304 loss: 2.1293 (2.1244) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [214] [ 50/312] eta: 0:03:57 lr: 0.002304 min_lr: 0.002304 loss: 2.1652 (2.1083) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [214] [ 60/312] eta: 0:03:39 lr: 0.002303 min_lr: 0.002303 loss: 2.2886 (2.1442) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [214] [ 70/312] eta: 0:03:24 lr: 0.002303 min_lr: 0.002303 loss: 2.3304 (2.1561) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [214] [ 80/312] eta: 0:03:11 lr: 0.002302 min_lr: 0.002302 loss: 2.3499 (2.1736) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [214] [ 90/312] eta: 0:03:00 lr: 0.002302 min_lr: 0.002302 loss: 2.3754 (2.1848) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [214] [100/312] eta: 0:02:49 lr: 0.002301 min_lr: 0.002301 loss: 2.1654 (2.1752) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [214] [110/312] eta: 0:02:39 lr: 0.002301 min_lr: 0.002301 loss: 2.0918 (2.1634) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [214] [120/312] eta: 0:02:30 lr: 0.002301 min_lr: 0.002301 loss: 2.1944 (2.1575) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [214] [130/312] eta: 0:02:21 lr: 0.002300 min_lr: 0.002300 loss: 2.1538 (2.1414) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [214] [140/312] eta: 0:02:12 lr: 0.002300 min_lr: 0.002300 loss: 1.9500 (2.1273) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [214] [150/312] eta: 0:02:04 lr: 0.002299 min_lr: 0.002299 loss: 2.1118 (2.1290) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [214] [160/312] eta: 0:01:55 lr: 0.002299 min_lr: 0.002299 loss: 2.2877 (2.1392) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [214] [170/312] eta: 0:01:47 lr: 0.002298 min_lr: 0.002298 loss: 2.2877 (2.1380) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [214] [180/312] eta: 0:01:39 lr: 0.002298 min_lr: 0.002298 loss: 2.0360 (2.1403) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [214] [190/312] eta: 0:01:31 lr: 0.002297 min_lr: 0.002297 loss: 2.1028 (2.1341) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [214] [200/312] eta: 0:01:23 lr: 0.002297 min_lr: 0.002297 loss: 2.2294 (2.1346) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [214] [210/312] eta: 0:01:16 lr: 0.002296 min_lr: 0.002296 loss: 2.2069 (2.1312) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [214] [220/312] eta: 0:01:08 lr: 0.002296 min_lr: 0.002296 loss: 2.0844 (2.1338) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [214] [230/312] eta: 0:01:00 lr: 0.002295 min_lr: 0.002295 loss: 2.0431 (2.1241) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [214] [240/312] eta: 0:00:53 lr: 0.002295 min_lr: 0.002295 loss: 1.9461 (2.1198) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [214] [250/312] eta: 0:00:45 lr: 0.002295 min_lr: 0.002295 loss: 2.0856 (2.1242) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [214] [260/312] eta: 0:00:38 lr: 0.002294 min_lr: 0.002294 loss: 2.0856 (2.1183) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [214] [270/312] eta: 0:00:30 lr: 0.002294 min_lr: 0.002294 loss: 2.0492 (2.1163) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [214] [280/312] eta: 0:00:23 lr: 0.002293 min_lr: 0.002293 loss: 2.1545 (2.1156) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0006 max mem: 64948 Epoch: [214] [290/312] eta: 0:00:16 lr: 0.002293 min_lr: 0.002293 loss: 2.1551 (2.1152) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0005 max mem: 64948 Epoch: [214] [300/312] eta: 0:00:08 lr: 0.002292 min_lr: 0.002292 loss: 2.2007 (2.1194) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [214] [310/312] eta: 0:00:01 lr: 0.002292 min_lr: 0.002292 loss: 2.2007 (2.1182) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [214] [311/312] eta: 0:00:00 lr: 0.002292 min_lr: 0.002292 loss: 2.1626 (2.1180) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [214] Total time: 0:03:48 (0.7319 s / it) Averaged stats: lr: 0.002292 min_lr: 0.002292 loss: 2.1626 (2.1002) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.6213 (0.6213) acc1: 84.8958 (84.8958) acc5: 96.6146 (96.6146) time: 4.7965 data: 4.5822 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9183 (0.8596) acc1: 78.3854 (78.3360) acc5: 94.5312 (94.4960) time: 0.6842 data: 0.5092 max mem: 64948 Test: Total time: 0:00:06 (0.7086 s / it) * Acc@1 78.604 Acc@5 94.516 loss 0.838 Accuracy of the model on the 50000 test images: 78.6% Max accuracy: 79.12% Test: [0/9] eta: 0:00:42 loss: 0.5228 (0.5228) acc1: 85.4167 (85.4167) acc5: 97.6562 (97.6562) time: 4.7635 data: 4.5593 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7186 (0.7039) acc1: 80.9896 (80.0960) acc5: 96.8750 (96.1280) time: 0.6864 data: 0.5125 max mem: 64948 Test: Total time: 0:00:06 (0.6966 s / it) * Acc@1 81.462 Acc@5 95.822 loss 0.681 Accuracy of the model EMA on 50000 test images: 81.5% Epoch: [215] [ 0/312] eta: 0:51:36 lr: 0.002292 min_lr: 0.002292 loss: 1.6503 (1.6503) weight_decay: 0.0500 (0.0500) time: 9.9235 data: 8.2689 max mem: 64948 Epoch: [215] [ 10/312] eta: 0:08:04 lr: 0.002291 min_lr: 0.002291 loss: 1.9768 (1.9151) weight_decay: 0.0500 (0.0500) time: 1.6053 data: 0.7522 max mem: 64948 Epoch: [215] [ 20/312] eta: 0:05:41 lr: 0.002291 min_lr: 0.002291 loss: 2.0426 (2.0398) weight_decay: 0.0500 (0.0500) time: 0.7333 data: 0.0004 max mem: 64948 Epoch: [215] [ 30/312] eta: 0:04:46 lr: 0.002290 min_lr: 0.002290 loss: 1.8385 (1.9664) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [215] [ 40/312] eta: 0:04:15 lr: 0.002290 min_lr: 0.002290 loss: 1.8385 (1.9844) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [215] [ 50/312] eta: 0:03:53 lr: 0.002289 min_lr: 0.002289 loss: 2.0843 (2.0189) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [215] [ 60/312] eta: 0:03:36 lr: 0.002289 min_lr: 0.002289 loss: 2.1169 (2.0144) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [215] [ 70/312] eta: 0:03:22 lr: 0.002288 min_lr: 0.002288 loss: 2.1169 (2.0393) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [215] [ 80/312] eta: 0:03:09 lr: 0.002288 min_lr: 0.002288 loss: 1.9780 (2.0422) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [215] [ 90/312] eta: 0:02:58 lr: 0.002287 min_lr: 0.002287 loss: 1.9780 (2.0534) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [215] [100/312] eta: 0:02:48 lr: 0.002287 min_lr: 0.002287 loss: 2.2126 (2.0692) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [215] [110/312] eta: 0:02:38 lr: 0.002287 min_lr: 0.002287 loss: 2.2846 (2.0880) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [215] [120/312] eta: 0:02:29 lr: 0.002286 min_lr: 0.002286 loss: 2.0056 (2.0701) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [215] [130/312] eta: 0:02:20 lr: 0.002286 min_lr: 0.002286 loss: 1.8822 (2.0703) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [215] [140/312] eta: 0:02:11 lr: 0.002285 min_lr: 0.002285 loss: 2.0675 (2.0797) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [215] [150/312] eta: 0:02:03 lr: 0.002285 min_lr: 0.002285 loss: 1.9795 (2.0759) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [215] [160/312] eta: 0:01:55 lr: 0.002284 min_lr: 0.002284 loss: 2.0918 (2.0754) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [215] [170/312] eta: 0:01:46 lr: 0.002284 min_lr: 0.002284 loss: 2.1035 (2.0694) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [215] [180/312] eta: 0:01:39 lr: 0.002283 min_lr: 0.002283 loss: 2.0124 (2.0663) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [215] [190/312] eta: 0:01:31 lr: 0.002283 min_lr: 0.002283 loss: 2.0320 (2.0654) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [215] [200/312] eta: 0:01:23 lr: 0.002282 min_lr: 0.002282 loss: 2.1084 (2.0692) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [215] [210/312] eta: 0:01:15 lr: 0.002282 min_lr: 0.002282 loss: 2.1922 (2.0800) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [215] [220/312] eta: 0:01:08 lr: 0.002281 min_lr: 0.002281 loss: 2.2400 (2.0777) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [215] [230/312] eta: 0:01:00 lr: 0.002281 min_lr: 0.002281 loss: 2.0648 (2.0775) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [215] [240/312] eta: 0:00:53 lr: 0.002281 min_lr: 0.002281 loss: 1.9179 (2.0674) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [215] [250/312] eta: 0:00:45 lr: 0.002280 min_lr: 0.002280 loss: 1.9008 (2.0666) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [215] [260/312] eta: 0:00:38 lr: 0.002280 min_lr: 0.002280 loss: 1.9714 (2.0667) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [215] [270/312] eta: 0:00:30 lr: 0.002279 min_lr: 0.002279 loss: 2.2102 (2.0733) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [215] [280/312] eta: 0:00:23 lr: 0.002279 min_lr: 0.002279 loss: 2.2575 (2.0767) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0011 max mem: 64948 Epoch: [215] [290/312] eta: 0:00:16 lr: 0.002278 min_lr: 0.002278 loss: 2.1345 (2.0756) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0010 max mem: 64948 Epoch: [215] [300/312] eta: 0:00:08 lr: 0.002278 min_lr: 0.002278 loss: 1.9940 (2.0708) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [215] [310/312] eta: 0:00:01 lr: 0.002277 min_lr: 0.002277 loss: 2.0393 (2.0748) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [215] [311/312] eta: 0:00:00 lr: 0.002277 min_lr: 0.002277 loss: 2.0393 (2.0751) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [215] Total time: 0:03:47 (0.7297 s / it) Averaged stats: lr: 0.002277 min_lr: 0.002277 loss: 2.0393 (2.1029) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5794 (0.5794) acc1: 85.6771 (85.6771) acc5: 96.8750 (96.8750) time: 4.6192 data: 4.4045 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9015 (0.8315) acc1: 78.3854 (78.4640) acc5: 95.5729 (94.6560) time: 0.6645 data: 0.4895 max mem: 64948 Test: Total time: 0:00:06 (0.6883 s / it) * Acc@1 78.936 Acc@5 94.708 loss 0.816 Accuracy of the model on the 50000 test images: 78.9% Max accuracy: 79.12% Test: [0/9] eta: 0:00:40 loss: 0.5214 (0.5214) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.5111 data: 4.2989 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7181 (0.7032) acc1: 80.7292 (79.9680) acc5: 96.8750 (96.1280) time: 0.6525 data: 0.4778 max mem: 64948 Test: Total time: 0:00:05 (0.6659 s / it) * Acc@1 81.464 Acc@5 95.834 loss 0.680 Accuracy of the model EMA on 50000 test images: 81.5% Max EMA accuracy: 81.46% Epoch: [216] [ 0/312] eta: 0:58:19 lr: 0.002277 min_lr: 0.002277 loss: 2.6304 (2.6304) weight_decay: 0.0500 (0.0500) time: 11.2152 data: 10.4895 max mem: 64948 Epoch: [216] [ 10/312] eta: 0:08:21 lr: 0.002277 min_lr: 0.002277 loss: 2.0803 (2.0168) weight_decay: 0.0500 (0.0500) time: 1.6596 data: 0.9539 max mem: 64948 Epoch: [216] [ 20/312] eta: 0:05:51 lr: 0.002276 min_lr: 0.002276 loss: 2.1256 (2.1082) weight_decay: 0.0500 (0.0500) time: 0.7029 data: 0.0003 max mem: 64948 Epoch: [216] [ 30/312] eta: 0:04:53 lr: 0.002276 min_lr: 0.002276 loss: 2.2091 (2.0773) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0003 max mem: 64948 Epoch: [216] [ 40/312] eta: 0:04:19 lr: 0.002275 min_lr: 0.002275 loss: 2.0647 (2.0648) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [216] [ 50/312] eta: 0:03:56 lr: 0.002275 min_lr: 0.002275 loss: 2.1216 (2.0862) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [216] [ 60/312] eta: 0:03:39 lr: 0.002274 min_lr: 0.002274 loss: 2.1776 (2.0620) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [216] [ 70/312] eta: 0:03:24 lr: 0.002274 min_lr: 0.002274 loss: 1.8835 (2.0617) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [216] [ 80/312] eta: 0:03:11 lr: 0.002273 min_lr: 0.002273 loss: 2.1718 (2.0655) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [216] [ 90/312] eta: 0:03:00 lr: 0.002273 min_lr: 0.002273 loss: 2.0395 (2.0563) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [216] [100/312] eta: 0:02:49 lr: 0.002273 min_lr: 0.002273 loss: 2.0081 (2.0484) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [216] [110/312] eta: 0:02:39 lr: 0.002272 min_lr: 0.002272 loss: 2.1567 (2.0626) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [216] [120/312] eta: 0:02:30 lr: 0.002272 min_lr: 0.002272 loss: 2.1159 (2.0589) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [216] [130/312] eta: 0:02:21 lr: 0.002271 min_lr: 0.002271 loss: 2.0855 (2.0585) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [216] [140/312] eta: 0:02:12 lr: 0.002271 min_lr: 0.002271 loss: 1.9659 (2.0441) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [216] [150/312] eta: 0:02:04 lr: 0.002270 min_lr: 0.002270 loss: 1.8179 (2.0445) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [216] [160/312] eta: 0:01:55 lr: 0.002270 min_lr: 0.002270 loss: 1.9097 (2.0305) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [216] [170/312] eta: 0:01:47 lr: 0.002269 min_lr: 0.002269 loss: 2.0005 (2.0395) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [216] [180/312] eta: 0:01:39 lr: 0.002269 min_lr: 0.002269 loss: 2.2034 (2.0507) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [216] [190/312] eta: 0:01:31 lr: 0.002268 min_lr: 0.002268 loss: 2.1815 (2.0544) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [216] [200/312] eta: 0:01:23 lr: 0.002268 min_lr: 0.002268 loss: 2.1579 (2.0579) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [216] [210/312] eta: 0:01:16 lr: 0.002267 min_lr: 0.002267 loss: 2.2686 (2.0577) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [216] [220/312] eta: 0:01:08 lr: 0.002267 min_lr: 0.002267 loss: 2.1192 (2.0610) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [216] [230/312] eta: 0:01:00 lr: 0.002266 min_lr: 0.002266 loss: 2.1606 (2.0636) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [216] [240/312] eta: 0:00:53 lr: 0.002266 min_lr: 0.002266 loss: 2.2377 (2.0683) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [216] [250/312] eta: 0:00:45 lr: 0.002266 min_lr: 0.002266 loss: 2.2020 (2.0710) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [216] [260/312] eta: 0:00:38 lr: 0.002265 min_lr: 0.002265 loss: 2.2474 (2.0714) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [216] [270/312] eta: 0:00:30 lr: 0.002265 min_lr: 0.002265 loss: 2.2655 (2.0749) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [216] [280/312] eta: 0:00:23 lr: 0.002264 min_lr: 0.002264 loss: 2.2792 (2.0821) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0009 max mem: 64948 Epoch: [216] [290/312] eta: 0:00:16 lr: 0.002264 min_lr: 0.002264 loss: 2.2283 (2.0846) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [216] [300/312] eta: 0:00:08 lr: 0.002263 min_lr: 0.002263 loss: 2.0954 (2.0813) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [216] [310/312] eta: 0:00:01 lr: 0.002263 min_lr: 0.002263 loss: 2.1637 (2.0863) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [216] [311/312] eta: 0:00:00 lr: 0.002263 min_lr: 0.002263 loss: 2.2060 (2.0867) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [216] Total time: 0:03:48 (0.7315 s / it) Averaged stats: lr: 0.002263 min_lr: 0.002263 loss: 2.2060 (2.0986) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5818 (0.5818) acc1: 86.1979 (86.1979) acc5: 96.6146 (96.6146) time: 4.6211 data: 4.4103 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8867 (0.8211) acc1: 78.9062 (78.1120) acc5: 94.2708 (94.6880) time: 0.6653 data: 0.4901 max mem: 64948 Test: Total time: 0:00:06 (0.6923 s / it) * Acc@1 78.816 Acc@5 94.598 loss 0.810 Accuracy of the model on the 50000 test images: 78.8% Max accuracy: 79.12% Test: [0/9] eta: 0:00:43 loss: 0.5199 (0.5199) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.8235 data: 4.6066 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7172 (0.7024) acc1: 80.9896 (80.0640) acc5: 96.8750 (96.1920) time: 0.6873 data: 0.5120 max mem: 64948 Test: Total time: 0:00:06 (0.7018 s / it) * Acc@1 81.494 Acc@5 95.860 loss 0.679 Accuracy of the model EMA on 50000 test images: 81.5% Max EMA accuracy: 81.49% Epoch: [217] [ 0/312] eta: 0:48:17 lr: 0.002263 min_lr: 0.002263 loss: 2.3992 (2.3992) weight_decay: 0.0500 (0.0500) time: 9.2884 data: 8.1576 max mem: 64948 Epoch: [217] [ 10/312] eta: 0:07:36 lr: 0.002262 min_lr: 0.002262 loss: 2.2022 (2.1146) weight_decay: 0.0500 (0.0500) time: 1.5123 data: 0.7420 max mem: 64948 Epoch: [217] [ 20/312] eta: 0:05:28 lr: 0.002262 min_lr: 0.002262 loss: 2.0765 (1.9971) weight_decay: 0.0500 (0.0500) time: 0.7161 data: 0.0004 max mem: 64948 Epoch: [217] [ 30/312] eta: 0:04:38 lr: 0.002261 min_lr: 0.002261 loss: 2.0765 (2.0623) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [217] [ 40/312] eta: 0:04:09 lr: 0.002261 min_lr: 0.002261 loss: 2.1563 (2.0354) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [217] [ 50/312] eta: 0:03:48 lr: 0.002260 min_lr: 0.002260 loss: 2.0875 (2.0697) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [217] [ 60/312] eta: 0:03:32 lr: 0.002260 min_lr: 0.002260 loss: 2.2665 (2.0758) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [217] [ 70/312] eta: 0:03:19 lr: 0.002259 min_lr: 0.002259 loss: 1.9445 (2.0515) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [217] [ 80/312] eta: 0:03:07 lr: 0.002259 min_lr: 0.002259 loss: 1.8665 (2.0465) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [217] [ 90/312] eta: 0:02:56 lr: 0.002259 min_lr: 0.002259 loss: 2.0329 (2.0515) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [217] [100/312] eta: 0:02:46 lr: 0.002258 min_lr: 0.002258 loss: 2.2270 (2.0840) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [217] [110/312] eta: 0:02:36 lr: 0.002258 min_lr: 0.002258 loss: 2.2962 (2.0879) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [217] [120/312] eta: 0:02:27 lr: 0.002257 min_lr: 0.002257 loss: 2.1278 (2.0845) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [217] [130/312] eta: 0:02:19 lr: 0.002257 min_lr: 0.002257 loss: 2.1139 (2.0813) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [217] [140/312] eta: 0:02:10 lr: 0.002256 min_lr: 0.002256 loss: 1.9194 (2.0762) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [217] [150/312] eta: 0:02:02 lr: 0.002256 min_lr: 0.002256 loss: 1.8466 (2.0592) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [217] [160/312] eta: 0:01:54 lr: 0.002255 min_lr: 0.002255 loss: 1.9213 (2.0550) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [217] [170/312] eta: 0:01:46 lr: 0.002255 min_lr: 0.002255 loss: 2.0398 (2.0557) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [217] [180/312] eta: 0:01:38 lr: 0.002254 min_lr: 0.002254 loss: 2.1602 (2.0636) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [217] [190/312] eta: 0:01:30 lr: 0.002254 min_lr: 0.002254 loss: 2.2995 (2.0678) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [217] [200/312] eta: 0:01:22 lr: 0.002253 min_lr: 0.002253 loss: 2.2476 (2.0749) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [217] [210/312] eta: 0:01:15 lr: 0.002253 min_lr: 0.002253 loss: 2.2359 (2.0748) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [217] [220/312] eta: 0:01:07 lr: 0.002252 min_lr: 0.002252 loss: 2.1705 (2.0757) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [217] [230/312] eta: 0:01:00 lr: 0.002252 min_lr: 0.002252 loss: 1.9987 (2.0658) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [217] [240/312] eta: 0:00:52 lr: 0.002252 min_lr: 0.002252 loss: 1.9987 (2.0653) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [217] [250/312] eta: 0:00:45 lr: 0.002251 min_lr: 0.002251 loss: 2.1163 (2.0685) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [217] [260/312] eta: 0:00:37 lr: 0.002251 min_lr: 0.002251 loss: 2.0158 (2.0686) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [217] [270/312] eta: 0:00:30 lr: 0.002250 min_lr: 0.002250 loss: 1.9614 (2.0652) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [217] [280/312] eta: 0:00:23 lr: 0.002250 min_lr: 0.002250 loss: 2.2882 (2.0720) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0010 max mem: 64948 Epoch: [217] [290/312] eta: 0:00:15 lr: 0.002249 min_lr: 0.002249 loss: 2.2374 (2.0669) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [217] [300/312] eta: 0:00:08 lr: 0.002249 min_lr: 0.002249 loss: 2.0151 (2.0701) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [217] [310/312] eta: 0:00:01 lr: 0.002248 min_lr: 0.002248 loss: 2.1991 (2.0711) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [217] [311/312] eta: 0:00:00 lr: 0.002248 min_lr: 0.002248 loss: 2.1991 (2.0696) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [217] Total time: 0:03:46 (0.7264 s / it) Averaged stats: lr: 0.002248 min_lr: 0.002248 loss: 2.1991 (2.0906) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6210 (0.6210) acc1: 84.8958 (84.8958) acc5: 96.3542 (96.3542) time: 4.5900 data: 4.3743 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8863 (0.8420) acc1: 79.1667 (78.1120) acc5: 95.8333 (94.5920) time: 0.6613 data: 0.4861 max mem: 64948 Test: Total time: 0:00:06 (0.6942 s / it) * Acc@1 78.894 Acc@5 94.498 loss 0.815 Accuracy of the model on the 50000 test images: 78.9% Max accuracy: 79.12% Test: [0/9] eta: 0:00:42 loss: 0.5185 (0.5185) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.7115 data: 4.4966 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7162 (0.7017) acc1: 80.9896 (80.1280) acc5: 96.8750 (96.1920) time: 0.6831 data: 0.5080 max mem: 64948 Test: Total time: 0:00:06 (0.6913 s / it) * Acc@1 81.528 Acc@5 95.868 loss 0.678 Accuracy of the model EMA on 50000 test images: 81.5% Max EMA accuracy: 81.53% Epoch: [218] [ 0/312] eta: 0:50:12 lr: 0.002248 min_lr: 0.002248 loss: 1.6787 (1.6787) weight_decay: 0.0500 (0.0500) time: 9.6556 data: 8.5879 max mem: 64948 Epoch: [218] [ 10/312] eta: 0:07:42 lr: 0.002248 min_lr: 0.002248 loss: 2.2021 (2.0731) weight_decay: 0.0500 (0.0500) time: 1.5329 data: 0.7811 max mem: 64948 Epoch: [218] [ 20/312] eta: 0:05:31 lr: 0.002247 min_lr: 0.002247 loss: 2.2021 (2.1330) weight_decay: 0.0500 (0.0500) time: 0.7093 data: 0.0003 max mem: 64948 Epoch: [218] [ 30/312] eta: 0:04:40 lr: 0.002247 min_lr: 0.002247 loss: 2.0382 (2.0689) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [218] [ 40/312] eta: 0:04:10 lr: 0.002246 min_lr: 0.002246 loss: 2.0382 (2.0609) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [218] [ 50/312] eta: 0:03:50 lr: 0.002246 min_lr: 0.002246 loss: 2.0837 (2.0594) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [218] [ 60/312] eta: 0:03:33 lr: 0.002245 min_lr: 0.002245 loss: 2.0058 (2.0536) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [218] [ 70/312] eta: 0:03:19 lr: 0.002245 min_lr: 0.002245 loss: 2.0264 (2.0523) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [218] [ 80/312] eta: 0:03:07 lr: 0.002244 min_lr: 0.002244 loss: 2.1287 (2.0633) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [218] [ 90/312] eta: 0:02:56 lr: 0.002244 min_lr: 0.002244 loss: 2.2083 (2.0820) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [218] [100/312] eta: 0:02:46 lr: 0.002244 min_lr: 0.002244 loss: 2.1995 (2.0856) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [218] [110/312] eta: 0:02:37 lr: 0.002243 min_lr: 0.002243 loss: 2.0109 (2.0712) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [218] [120/312] eta: 0:02:28 lr: 0.002243 min_lr: 0.002243 loss: 2.1847 (2.0976) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [218] [130/312] eta: 0:02:19 lr: 0.002242 min_lr: 0.002242 loss: 2.2047 (2.0851) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [218] [140/312] eta: 0:02:10 lr: 0.002242 min_lr: 0.002242 loss: 2.1134 (2.0867) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [218] [150/312] eta: 0:02:02 lr: 0.002241 min_lr: 0.002241 loss: 2.0529 (2.0820) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [218] [160/312] eta: 0:01:54 lr: 0.002241 min_lr: 0.002241 loss: 2.1849 (2.0976) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [218] [170/312] eta: 0:01:46 lr: 0.002240 min_lr: 0.002240 loss: 2.3270 (2.1065) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [218] [180/312] eta: 0:01:38 lr: 0.002240 min_lr: 0.002240 loss: 2.2316 (2.1118) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [218] [190/312] eta: 0:01:30 lr: 0.002239 min_lr: 0.002239 loss: 2.1415 (2.1071) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [218] [200/312] eta: 0:01:22 lr: 0.002239 min_lr: 0.002239 loss: 2.1415 (2.1079) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [218] [210/312] eta: 0:01:15 lr: 0.002238 min_lr: 0.002238 loss: 2.2353 (2.1104) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [218] [220/312] eta: 0:01:07 lr: 0.002238 min_lr: 0.002238 loss: 2.2353 (2.1130) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [218] [230/312] eta: 0:01:00 lr: 0.002238 min_lr: 0.002238 loss: 2.1993 (2.1145) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [218] [240/312] eta: 0:00:52 lr: 0.002237 min_lr: 0.002237 loss: 2.2547 (2.1189) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [218] [250/312] eta: 0:00:45 lr: 0.002237 min_lr: 0.002237 loss: 2.2547 (2.1206) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [218] [260/312] eta: 0:00:37 lr: 0.002236 min_lr: 0.002236 loss: 2.1757 (2.1219) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [218] [270/312] eta: 0:00:30 lr: 0.002236 min_lr: 0.002236 loss: 1.9810 (2.1169) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [218] [280/312] eta: 0:00:23 lr: 0.002235 min_lr: 0.002235 loss: 2.0295 (2.1156) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [218] [290/312] eta: 0:00:15 lr: 0.002235 min_lr: 0.002235 loss: 2.2411 (2.1185) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [218] [300/312] eta: 0:00:08 lr: 0.002234 min_lr: 0.002234 loss: 2.3037 (2.1225) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [218] [310/312] eta: 0:00:01 lr: 0.002234 min_lr: 0.002234 loss: 2.2667 (2.1253) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [218] [311/312] eta: 0:00:00 lr: 0.002234 min_lr: 0.002234 loss: 2.2457 (2.1251) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [218] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.002234 min_lr: 0.002234 loss: 2.2457 (2.0953) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6110 (0.6110) acc1: 84.3750 (84.3750) acc5: 96.3542 (96.3542) time: 4.6916 data: 4.4752 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9149 (0.8490) acc1: 78.3854 (77.8880) acc5: 94.2708 (94.5600) time: 0.6726 data: 0.4973 max mem: 64948 Test: Total time: 0:00:06 (0.6949 s / it) * Acc@1 78.776 Acc@5 94.464 loss 0.822 Accuracy of the model on the 50000 test images: 78.8% Max accuracy: 79.12% Test: [0/9] eta: 0:00:42 loss: 0.5169 (0.5169) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.7558 data: 4.5419 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7155 (0.7011) acc1: 80.9896 (80.1600) acc5: 96.8750 (96.1600) time: 0.6798 data: 0.5048 max mem: 64948 Test: Total time: 0:00:06 (0.6910 s / it) * Acc@1 81.568 Acc@5 95.884 loss 0.677 Accuracy of the model EMA on 50000 test images: 81.6% Max EMA accuracy: 81.57% Epoch: [219] [ 0/312] eta: 0:53:33 lr: 0.002234 min_lr: 0.002234 loss: 2.3411 (2.3411) weight_decay: 0.0500 (0.0500) time: 10.3009 data: 9.4886 max mem: 64948 Epoch: [219] [ 10/312] eta: 0:07:56 lr: 0.002233 min_lr: 0.002233 loss: 2.1109 (2.1279) weight_decay: 0.0500 (0.0500) time: 1.5770 data: 0.8630 max mem: 64948 Epoch: [219] [ 20/312] eta: 0:05:38 lr: 0.002233 min_lr: 0.002233 loss: 2.1437 (2.2033) weight_decay: 0.0500 (0.0500) time: 0.7026 data: 0.0004 max mem: 64948 Epoch: [219] [ 30/312] eta: 0:04:44 lr: 0.002232 min_lr: 0.002232 loss: 2.2267 (2.1875) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [219] [ 40/312] eta: 0:04:13 lr: 0.002232 min_lr: 0.002232 loss: 2.2095 (2.1911) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [219] [ 50/312] eta: 0:03:52 lr: 0.002231 min_lr: 0.002231 loss: 2.2222 (2.1974) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [219] [ 60/312] eta: 0:03:35 lr: 0.002231 min_lr: 0.002231 loss: 2.2130 (2.1885) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [219] [ 70/312] eta: 0:03:21 lr: 0.002230 min_lr: 0.002230 loss: 2.1565 (2.1852) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [219] [ 80/312] eta: 0:03:09 lr: 0.002230 min_lr: 0.002230 loss: 2.2093 (2.1871) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [219] [ 90/312] eta: 0:02:58 lr: 0.002230 min_lr: 0.002230 loss: 2.2388 (2.1758) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [219] [100/312] eta: 0:02:47 lr: 0.002229 min_lr: 0.002229 loss: 1.9515 (2.1669) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [219] [110/312] eta: 0:02:38 lr: 0.002229 min_lr: 0.002229 loss: 2.2866 (2.1884) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [219] [120/312] eta: 0:02:29 lr: 0.002228 min_lr: 0.002228 loss: 2.2866 (2.1840) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [219] [130/312] eta: 0:02:20 lr: 0.002228 min_lr: 0.002228 loss: 2.1594 (2.1692) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [219] [140/312] eta: 0:02:11 lr: 0.002227 min_lr: 0.002227 loss: 2.1967 (2.1780) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [219] [150/312] eta: 0:02:03 lr: 0.002227 min_lr: 0.002227 loss: 2.3127 (2.1717) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [219] [160/312] eta: 0:01:54 lr: 0.002226 min_lr: 0.002226 loss: 2.2286 (2.1647) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [219] [170/312] eta: 0:01:46 lr: 0.002226 min_lr: 0.002226 loss: 1.9976 (2.1479) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [219] [180/312] eta: 0:01:38 lr: 0.002225 min_lr: 0.002225 loss: 2.1318 (2.1432) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [219] [190/312] eta: 0:01:31 lr: 0.002225 min_lr: 0.002225 loss: 2.0896 (2.1354) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [219] [200/312] eta: 0:01:23 lr: 0.002224 min_lr: 0.002224 loss: 2.1533 (2.1402) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [219] [210/312] eta: 0:01:15 lr: 0.002224 min_lr: 0.002224 loss: 2.1633 (2.1348) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [219] [220/312] eta: 0:01:07 lr: 0.002223 min_lr: 0.002223 loss: 2.1275 (2.1384) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [219] [230/312] eta: 0:01:00 lr: 0.002223 min_lr: 0.002223 loss: 2.2252 (2.1380) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [219] [240/312] eta: 0:00:52 lr: 0.002223 min_lr: 0.002223 loss: 2.3286 (2.1471) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [219] [250/312] eta: 0:00:45 lr: 0.002222 min_lr: 0.002222 loss: 2.0800 (2.1358) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [219] [260/312] eta: 0:00:38 lr: 0.002222 min_lr: 0.002222 loss: 1.9076 (2.1338) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [219] [270/312] eta: 0:00:30 lr: 0.002221 min_lr: 0.002221 loss: 2.2326 (2.1349) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [219] [280/312] eta: 0:00:23 lr: 0.002221 min_lr: 0.002221 loss: 2.2241 (2.1306) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [219] [290/312] eta: 0:00:16 lr: 0.002220 min_lr: 0.002220 loss: 2.1123 (2.1304) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0008 max mem: 64948 Epoch: [219] [300/312] eta: 0:00:08 lr: 0.002220 min_lr: 0.002220 loss: 2.2152 (2.1321) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [219] [310/312] eta: 0:00:01 lr: 0.002219 min_lr: 0.002219 loss: 2.0648 (2.1269) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [219] [311/312] eta: 0:00:00 lr: 0.002219 min_lr: 0.002219 loss: 1.9470 (2.1263) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [219] Total time: 0:03:47 (0.7288 s / it) Averaged stats: lr: 0.002219 min_lr: 0.002219 loss: 1.9470 (2.0921) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5849 (0.5849) acc1: 84.6354 (84.6354) acc5: 96.3542 (96.3542) time: 4.4953 data: 4.2756 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8309 (0.8216) acc1: 77.8646 (78.0800) acc5: 95.5729 (94.5280) time: 0.6514 data: 0.4752 max mem: 64948 Test: Total time: 0:00:05 (0.6620 s / it) * Acc@1 78.976 Acc@5 94.642 loss 0.800 Accuracy of the model on the 50000 test images: 79.0% Max accuracy: 79.12% Test: [0/9] eta: 0:00:43 loss: 0.5152 (0.5152) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.8557 data: 4.6471 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7151 (0.7005) acc1: 80.9896 (80.1600) acc5: 96.8750 (96.1920) time: 0.6908 data: 0.5164 max mem: 64948 Test: Total time: 0:00:06 (0.7050 s / it) * Acc@1 81.602 Acc@5 95.894 loss 0.677 Accuracy of the model EMA on 50000 test images: 81.6% Max EMA accuracy: 81.60% Epoch: [220] [ 0/312] eta: 0:50:52 lr: 0.002219 min_lr: 0.002219 loss: 2.3329 (2.3329) weight_decay: 0.0500 (0.0500) time: 9.7822 data: 9.0067 max mem: 64948 Epoch: [220] [ 10/312] eta: 0:07:45 lr: 0.002219 min_lr: 0.002219 loss: 2.2800 (2.2058) weight_decay: 0.0500 (0.0500) time: 1.5427 data: 0.8191 max mem: 64948 Epoch: [220] [ 20/312] eta: 0:05:32 lr: 0.002218 min_lr: 0.002218 loss: 2.2492 (2.2242) weight_decay: 0.0500 (0.0500) time: 0.7069 data: 0.0003 max mem: 64948 Epoch: [220] [ 30/312] eta: 0:04:40 lr: 0.002218 min_lr: 0.002218 loss: 2.2166 (2.1873) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [220] [ 40/312] eta: 0:04:11 lr: 0.002217 min_lr: 0.002217 loss: 2.1079 (2.0998) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [220] [ 50/312] eta: 0:03:50 lr: 0.002217 min_lr: 0.002217 loss: 2.1079 (2.0996) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [220] [ 60/312] eta: 0:03:33 lr: 0.002216 min_lr: 0.002216 loss: 2.1678 (2.0817) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [220] [ 70/312] eta: 0:03:20 lr: 0.002216 min_lr: 0.002216 loss: 2.0591 (2.0744) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [220] [ 80/312] eta: 0:03:08 lr: 0.002215 min_lr: 0.002215 loss: 2.2440 (2.1052) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [220] [ 90/312] eta: 0:02:57 lr: 0.002215 min_lr: 0.002215 loss: 2.2722 (2.1121) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [220] [100/312] eta: 0:02:47 lr: 0.002215 min_lr: 0.002215 loss: 2.1257 (2.0924) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [220] [110/312] eta: 0:02:37 lr: 0.002214 min_lr: 0.002214 loss: 2.0792 (2.0958) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [220] [120/312] eta: 0:02:28 lr: 0.002214 min_lr: 0.002214 loss: 2.0887 (2.0869) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [220] [130/312] eta: 0:02:19 lr: 0.002213 min_lr: 0.002213 loss: 2.0770 (2.0783) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [220] [140/312] eta: 0:02:10 lr: 0.002213 min_lr: 0.002213 loss: 1.9596 (2.0726) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [220] [150/312] eta: 0:02:02 lr: 0.002212 min_lr: 0.002212 loss: 2.0160 (2.0729) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [220] [160/312] eta: 0:01:54 lr: 0.002212 min_lr: 0.002212 loss: 2.0878 (2.0718) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [220] [170/312] eta: 0:01:46 lr: 0.002211 min_lr: 0.002211 loss: 2.0878 (2.0699) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [220] [180/312] eta: 0:01:38 lr: 0.002211 min_lr: 0.002211 loss: 2.1487 (2.0741) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [220] [190/312] eta: 0:01:30 lr: 0.002210 min_lr: 0.002210 loss: 2.2911 (2.0859) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [220] [200/312] eta: 0:01:23 lr: 0.002210 min_lr: 0.002210 loss: 2.2512 (2.0870) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [220] [210/312] eta: 0:01:15 lr: 0.002209 min_lr: 0.002209 loss: 2.1420 (2.0814) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [220] [220/312] eta: 0:01:07 lr: 0.002209 min_lr: 0.002209 loss: 2.1263 (2.0805) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [220] [230/312] eta: 0:01:00 lr: 0.002208 min_lr: 0.002208 loss: 2.1263 (2.0817) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [220] [240/312] eta: 0:00:52 lr: 0.002208 min_lr: 0.002208 loss: 2.0211 (2.0727) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [220] [250/312] eta: 0:00:45 lr: 0.002208 min_lr: 0.002208 loss: 2.1367 (2.0810) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [220] [260/312] eta: 0:00:37 lr: 0.002207 min_lr: 0.002207 loss: 2.2055 (2.0826) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [220] [270/312] eta: 0:00:30 lr: 0.002207 min_lr: 0.002207 loss: 2.1695 (2.0783) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [220] [280/312] eta: 0:00:23 lr: 0.002206 min_lr: 0.002206 loss: 2.1695 (2.0821) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [220] [290/312] eta: 0:00:15 lr: 0.002206 min_lr: 0.002206 loss: 2.2828 (2.0862) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [220] [300/312] eta: 0:00:08 lr: 0.002205 min_lr: 0.002205 loss: 2.2831 (2.0919) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [220] [310/312] eta: 0:00:01 lr: 0.002205 min_lr: 0.002205 loss: 2.1568 (2.0890) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [220] [311/312] eta: 0:00:00 lr: 0.002205 min_lr: 0.002205 loss: 2.1934 (2.0899) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [220] Total time: 0:03:46 (0.7272 s / it) Averaged stats: lr: 0.002205 min_lr: 0.002205 loss: 2.1934 (2.0833) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6076 (0.6076) acc1: 85.4167 (85.4167) acc5: 96.0938 (96.0938) time: 4.5597 data: 4.3418 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8621 (0.8364) acc1: 77.3438 (77.6960) acc5: 94.0104 (94.4320) time: 0.6579 data: 0.4825 max mem: 64948 Test: Total time: 0:00:06 (0.6805 s / it) * Acc@1 78.906 Acc@5 94.582 loss 0.803 Accuracy of the model on the 50000 test images: 78.9% Max accuracy: 79.12% Test: [0/9] eta: 0:00:40 loss: 0.5142 (0.5142) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.4964 data: 4.2782 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7143 (0.6999) acc1: 81.2500 (80.1920) acc5: 96.8750 (96.2240) time: 0.6885 data: 0.5131 max mem: 64948 Test: Total time: 0:00:06 (0.6961 s / it) * Acc@1 81.612 Acc@5 95.918 loss 0.676 Accuracy of the model EMA on 50000 test images: 81.6% Max EMA accuracy: 81.61% Epoch: [221] [ 0/312] eta: 0:48:58 lr: 0.002205 min_lr: 0.002205 loss: 1.5149 (1.5149) weight_decay: 0.0500 (0.0500) time: 9.4171 data: 8.6340 max mem: 64948 Epoch: [221] [ 10/312] eta: 0:07:36 lr: 0.002204 min_lr: 0.002204 loss: 2.2240 (2.1627) weight_decay: 0.0500 (0.0500) time: 1.5123 data: 0.7853 max mem: 64948 Epoch: [221] [ 20/312] eta: 0:05:27 lr: 0.002204 min_lr: 0.002204 loss: 2.2640 (2.1931) weight_decay: 0.0500 (0.0500) time: 0.7075 data: 0.0004 max mem: 64948 Epoch: [221] [ 30/312] eta: 0:04:38 lr: 0.002203 min_lr: 0.002203 loss: 2.1898 (2.1493) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [221] [ 40/312] eta: 0:04:08 lr: 0.002203 min_lr: 0.002203 loss: 2.0650 (2.1392) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [221] [ 50/312] eta: 0:03:48 lr: 0.002202 min_lr: 0.002202 loss: 2.0650 (2.1403) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [221] [ 60/312] eta: 0:03:32 lr: 0.002202 min_lr: 0.002202 loss: 2.1470 (2.1352) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [221] [ 70/312] eta: 0:03:18 lr: 0.002201 min_lr: 0.002201 loss: 2.1261 (2.1186) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [221] [ 80/312] eta: 0:03:07 lr: 0.002201 min_lr: 0.002201 loss: 2.0805 (2.1145) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [221] [ 90/312] eta: 0:02:56 lr: 0.002200 min_lr: 0.002200 loss: 1.9197 (2.0815) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [221] [100/312] eta: 0:02:46 lr: 0.002200 min_lr: 0.002200 loss: 2.0535 (2.0918) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [221] [110/312] eta: 0:02:36 lr: 0.002200 min_lr: 0.002200 loss: 2.2392 (2.1115) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [221] [120/312] eta: 0:02:27 lr: 0.002199 min_lr: 0.002199 loss: 2.2036 (2.0949) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [221] [130/312] eta: 0:02:19 lr: 0.002199 min_lr: 0.002199 loss: 1.8654 (2.0831) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [221] [140/312] eta: 0:02:10 lr: 0.002198 min_lr: 0.002198 loss: 1.8600 (2.0721) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [221] [150/312] eta: 0:02:02 lr: 0.002198 min_lr: 0.002198 loss: 1.9894 (2.0699) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [221] [160/312] eta: 0:01:54 lr: 0.002197 min_lr: 0.002197 loss: 2.0410 (2.0706) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [221] [170/312] eta: 0:01:46 lr: 0.002197 min_lr: 0.002197 loss: 1.9515 (2.0644) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [221] [180/312] eta: 0:01:38 lr: 0.002196 min_lr: 0.002196 loss: 1.9954 (2.0644) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [221] [190/312] eta: 0:01:30 lr: 0.002196 min_lr: 0.002196 loss: 2.0769 (2.0614) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [221] [200/312] eta: 0:01:22 lr: 0.002195 min_lr: 0.002195 loss: 2.0769 (2.0649) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [221] [210/312] eta: 0:01:15 lr: 0.002195 min_lr: 0.002195 loss: 2.2041 (2.0660) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [221] [220/312] eta: 0:01:07 lr: 0.002194 min_lr: 0.002194 loss: 2.1306 (2.0673) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [221] [230/312] eta: 0:01:00 lr: 0.002194 min_lr: 0.002194 loss: 2.1306 (2.0694) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [221] [240/312] eta: 0:00:52 lr: 0.002193 min_lr: 0.002193 loss: 2.0689 (2.0668) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [221] [250/312] eta: 0:00:45 lr: 0.002193 min_lr: 0.002193 loss: 2.1122 (2.0696) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [221] [260/312] eta: 0:00:37 lr: 0.002193 min_lr: 0.002193 loss: 2.1122 (2.0684) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [221] [270/312] eta: 0:00:30 lr: 0.002192 min_lr: 0.002192 loss: 2.2291 (2.0743) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [221] [280/312] eta: 0:00:23 lr: 0.002192 min_lr: 0.002192 loss: 2.2291 (2.0741) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0010 max mem: 64948 Epoch: [221] [290/312] eta: 0:00:15 lr: 0.002191 min_lr: 0.002191 loss: 2.1439 (2.0674) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0008 max mem: 64948 Epoch: [221] [300/312] eta: 0:00:08 lr: 0.002191 min_lr: 0.002191 loss: 2.0743 (2.0688) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [221] [310/312] eta: 0:00:01 lr: 0.002190 min_lr: 0.002190 loss: 2.1112 (2.0688) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [221] [311/312] eta: 0:00:00 lr: 0.002190 min_lr: 0.002190 loss: 2.1112 (2.0701) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [221] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.002190 min_lr: 0.002190 loss: 2.1112 (2.0883) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6198 (0.6198) acc1: 85.6771 (85.6771) acc5: 95.8333 (95.8333) time: 4.4583 data: 4.2498 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9520 (0.8697) acc1: 75.7812 (77.1840) acc5: 93.7500 (94.3360) time: 0.6466 data: 0.4723 max mem: 64948 Test: Total time: 0:00:06 (0.6676 s / it) * Acc@1 78.480 Acc@5 94.252 loss 0.836 Accuracy of the model on the 50000 test images: 78.5% Max accuracy: 79.12% Test: [0/9] eta: 0:00:46 loss: 0.5134 (0.5134) acc1: 85.4167 (85.4167) acc5: 97.9167 (97.9167) time: 5.1597 data: 4.9541 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7140 (0.6993) acc1: 80.9896 (80.2560) acc5: 96.8750 (96.2240) time: 0.7246 data: 0.5505 max mem: 64948 Test: Total time: 0:00:06 (0.7329 s / it) * Acc@1 81.628 Acc@5 95.906 loss 0.675 Accuracy of the model EMA on 50000 test images: 81.6% Max EMA accuracy: 81.63% Epoch: [222] [ 0/312] eta: 0:51:29 lr: 0.002190 min_lr: 0.002190 loss: 2.1781 (2.1781) weight_decay: 0.0500 (0.0500) time: 9.9028 data: 9.1102 max mem: 64948 Epoch: [222] [ 10/312] eta: 0:07:50 lr: 0.002190 min_lr: 0.002190 loss: 2.1974 (2.0873) weight_decay: 0.0500 (0.0500) time: 1.5574 data: 0.8286 max mem: 64948 Epoch: [222] [ 20/312] eta: 0:05:34 lr: 0.002189 min_lr: 0.002189 loss: 2.0430 (2.0417) weight_decay: 0.0500 (0.0500) time: 0.7084 data: 0.0004 max mem: 64948 Epoch: [222] [ 30/312] eta: 0:04:42 lr: 0.002189 min_lr: 0.002189 loss: 2.1547 (2.0928) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [222] [ 40/312] eta: 0:04:12 lr: 0.002188 min_lr: 0.002188 loss: 2.2040 (2.0877) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [222] [ 50/312] eta: 0:03:51 lr: 0.002188 min_lr: 0.002188 loss: 2.0546 (2.0648) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [222] [ 60/312] eta: 0:03:34 lr: 0.002187 min_lr: 0.002187 loss: 2.1039 (2.0787) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [222] [ 70/312] eta: 0:03:20 lr: 0.002187 min_lr: 0.002187 loss: 2.1688 (2.0790) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [222] [ 80/312] eta: 0:03:08 lr: 0.002186 min_lr: 0.002186 loss: 2.1949 (2.0889) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [222] [ 90/312] eta: 0:02:57 lr: 0.002186 min_lr: 0.002186 loss: 2.1222 (2.0929) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [222] [100/312] eta: 0:02:47 lr: 0.002185 min_lr: 0.002185 loss: 2.2466 (2.1008) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [222] [110/312] eta: 0:02:37 lr: 0.002185 min_lr: 0.002185 loss: 2.2364 (2.0947) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [222] [120/312] eta: 0:02:28 lr: 0.002185 min_lr: 0.002185 loss: 2.0974 (2.0953) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [222] [130/312] eta: 0:02:19 lr: 0.002184 min_lr: 0.002184 loss: 2.0974 (2.0931) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [222] [140/312] eta: 0:02:11 lr: 0.002184 min_lr: 0.002184 loss: 2.0367 (2.0811) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [222] [150/312] eta: 0:02:02 lr: 0.002183 min_lr: 0.002183 loss: 1.9651 (2.0755) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [222] [160/312] eta: 0:01:54 lr: 0.002183 min_lr: 0.002183 loss: 2.2335 (2.0908) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [222] [170/312] eta: 0:01:46 lr: 0.002182 min_lr: 0.002182 loss: 2.2825 (2.0999) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [222] [180/312] eta: 0:01:38 lr: 0.002182 min_lr: 0.002182 loss: 2.2292 (2.1032) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [222] [190/312] eta: 0:01:30 lr: 0.002181 min_lr: 0.002181 loss: 2.2513 (2.1030) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [222] [200/312] eta: 0:01:23 lr: 0.002181 min_lr: 0.002181 loss: 2.1240 (2.1040) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [222] [210/312] eta: 0:01:15 lr: 0.002180 min_lr: 0.002180 loss: 2.1240 (2.0976) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [222] [220/312] eta: 0:01:07 lr: 0.002180 min_lr: 0.002180 loss: 2.0656 (2.0976) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [222] [230/312] eta: 0:01:00 lr: 0.002179 min_lr: 0.002179 loss: 2.0608 (2.0949) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [222] [240/312] eta: 0:00:52 lr: 0.002179 min_lr: 0.002179 loss: 2.0608 (2.0889) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [222] [250/312] eta: 0:00:45 lr: 0.002178 min_lr: 0.002178 loss: 2.1119 (2.0890) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [222] [260/312] eta: 0:00:38 lr: 0.002178 min_lr: 0.002178 loss: 2.1084 (2.0878) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [222] [270/312] eta: 0:00:30 lr: 0.002178 min_lr: 0.002178 loss: 2.1006 (2.0886) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [222] [280/312] eta: 0:00:23 lr: 0.002177 min_lr: 0.002177 loss: 2.1752 (2.0898) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [222] [290/312] eta: 0:00:16 lr: 0.002177 min_lr: 0.002177 loss: 2.2273 (2.0904) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [222] [300/312] eta: 0:00:08 lr: 0.002176 min_lr: 0.002176 loss: 2.0635 (2.0858) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [222] [310/312] eta: 0:00:01 lr: 0.002176 min_lr: 0.002176 loss: 1.8857 (2.0806) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [222] [311/312] eta: 0:00:00 lr: 0.002176 min_lr: 0.002176 loss: 1.9050 (2.0807) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [222] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.002176 min_lr: 0.002176 loss: 1.9050 (2.0906) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6339 (0.6339) acc1: 83.5938 (83.5938) acc5: 97.3958 (97.3958) time: 4.6409 data: 4.4280 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8737 (0.8170) acc1: 78.6458 (77.9200) acc5: 95.3125 (95.1680) time: 0.6669 data: 0.4921 max mem: 64948 Test: Total time: 0:00:06 (0.6930 s / it) * Acc@1 79.318 Acc@5 94.842 loss 0.789 Accuracy of the model on the 50000 test images: 79.3% Max accuracy: 79.32% Test: [0/9] eta: 0:00:40 loss: 0.5127 (0.5127) acc1: 85.4167 (85.4167) acc5: 97.9167 (97.9167) time: 4.5356 data: 4.3178 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7135 (0.6987) acc1: 81.5104 (80.3520) acc5: 96.8750 (96.2560) time: 0.6553 data: 0.4798 max mem: 64948 Test: Total time: 0:00:05 (0.6633 s / it) * Acc@1 81.678 Acc@5 95.900 loss 0.675 Accuracy of the model EMA on 50000 test images: 81.7% Max EMA accuracy: 81.68% Epoch: [223] [ 0/312] eta: 0:50:21 lr: 0.002176 min_lr: 0.002176 loss: 1.8746 (1.8746) weight_decay: 0.0500 (0.0500) time: 9.6828 data: 7.2084 max mem: 64948 Epoch: [223] [ 10/312] eta: 0:07:50 lr: 0.002175 min_lr: 0.002175 loss: 1.8746 (1.9156) weight_decay: 0.0500 (0.0500) time: 1.5585 data: 0.6837 max mem: 64948 Epoch: [223] [ 20/312] eta: 0:05:35 lr: 0.002175 min_lr: 0.002175 loss: 2.0576 (2.0022) weight_decay: 0.0500 (0.0500) time: 0.7227 data: 0.0158 max mem: 64948 Epoch: [223] [ 30/312] eta: 0:04:42 lr: 0.002174 min_lr: 0.002174 loss: 2.0375 (2.0036) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [223] [ 40/312] eta: 0:04:12 lr: 0.002174 min_lr: 0.002174 loss: 2.1637 (2.0396) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [223] [ 50/312] eta: 0:03:51 lr: 0.002173 min_lr: 0.002173 loss: 2.0892 (1.9979) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [223] [ 60/312] eta: 0:03:34 lr: 0.002173 min_lr: 0.002173 loss: 1.8747 (2.0056) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [223] [ 70/312] eta: 0:03:20 lr: 0.002172 min_lr: 0.002172 loss: 1.9675 (2.0009) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [223] [ 80/312] eta: 0:03:08 lr: 0.002172 min_lr: 0.002172 loss: 1.9675 (1.9988) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0003 max mem: 64948 Epoch: [223] [ 90/312] eta: 0:02:57 lr: 0.002171 min_lr: 0.002171 loss: 2.0665 (2.0094) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [223] [100/312] eta: 0:02:47 lr: 0.002171 min_lr: 0.002171 loss: 2.0726 (2.0035) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [223] [110/312] eta: 0:02:37 lr: 0.002170 min_lr: 0.002170 loss: 2.1238 (2.0138) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [223] [120/312] eta: 0:02:28 lr: 0.002170 min_lr: 0.002170 loss: 2.1274 (2.0192) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [223] [130/312] eta: 0:02:19 lr: 0.002170 min_lr: 0.002170 loss: 2.1440 (2.0278) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [223] [140/312] eta: 0:02:11 lr: 0.002169 min_lr: 0.002169 loss: 2.1560 (2.0344) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [223] [150/312] eta: 0:02:02 lr: 0.002169 min_lr: 0.002169 loss: 2.0464 (2.0286) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [223] [160/312] eta: 0:01:54 lr: 0.002168 min_lr: 0.002168 loss: 2.1022 (2.0443) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [223] [170/312] eta: 0:01:46 lr: 0.002168 min_lr: 0.002168 loss: 2.3403 (2.0504) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [223] [180/312] eta: 0:01:38 lr: 0.002167 min_lr: 0.002167 loss: 2.2055 (2.0525) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [223] [190/312] eta: 0:01:30 lr: 0.002167 min_lr: 0.002167 loss: 2.2382 (2.0684) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [223] [200/312] eta: 0:01:23 lr: 0.002166 min_lr: 0.002166 loss: 2.1938 (2.0643) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [223] [210/312] eta: 0:01:15 lr: 0.002166 min_lr: 0.002166 loss: 2.0552 (2.0566) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [223] [220/312] eta: 0:01:07 lr: 0.002165 min_lr: 0.002165 loss: 2.1105 (2.0612) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [223] [230/312] eta: 0:01:00 lr: 0.002165 min_lr: 0.002165 loss: 2.2394 (2.0657) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [223] [240/312] eta: 0:00:52 lr: 0.002164 min_lr: 0.002164 loss: 2.2295 (2.0596) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [223] [250/312] eta: 0:00:45 lr: 0.002164 min_lr: 0.002164 loss: 1.9917 (2.0605) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [223] [260/312] eta: 0:00:38 lr: 0.002163 min_lr: 0.002163 loss: 2.2020 (2.0631) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [223] [270/312] eta: 0:00:30 lr: 0.002163 min_lr: 0.002163 loss: 2.2020 (2.0694) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [223] [280/312] eta: 0:00:23 lr: 0.002163 min_lr: 0.002163 loss: 2.1235 (2.0637) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0006 max mem: 64948 Epoch: [223] [290/312] eta: 0:00:16 lr: 0.002162 min_lr: 0.002162 loss: 1.9442 (2.0620) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0005 max mem: 64948 Epoch: [223] [300/312] eta: 0:00:08 lr: 0.002162 min_lr: 0.002162 loss: 1.8155 (2.0499) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [223] [310/312] eta: 0:00:01 lr: 0.002161 min_lr: 0.002161 loss: 1.8155 (2.0513) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [223] [311/312] eta: 0:00:00 lr: 0.002161 min_lr: 0.002161 loss: 1.7985 (2.0504) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [223] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.002161 min_lr: 0.002161 loss: 1.7985 (2.0896) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6339 (0.6339) acc1: 83.3333 (83.3333) acc5: 95.8333 (95.8333) time: 4.6005 data: 4.3851 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8590 (0.8163) acc1: 79.4271 (78.4640) acc5: 95.3125 (94.4960) time: 0.6625 data: 0.4873 max mem: 64948 Test: Total time: 0:00:06 (0.6855 s / it) * Acc@1 79.294 Acc@5 94.598 loss 0.796 Accuracy of the model on the 50000 test images: 79.3% Max accuracy: 79.32% Test: [0/9] eta: 0:00:41 loss: 0.5129 (0.5129) acc1: 85.4167 (85.4167) acc5: 97.9167 (97.9167) time: 4.6591 data: 4.4410 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7133 (0.6983) acc1: 81.5104 (80.3200) acc5: 96.8750 (96.2560) time: 0.6694 data: 0.4936 max mem: 64948 Test: Total time: 0:00:06 (0.6795 s / it) * Acc@1 81.696 Acc@5 95.916 loss 0.674 Accuracy of the model EMA on 50000 test images: 81.7% Max EMA accuracy: 81.70% Epoch: [224] [ 0/312] eta: 0:48:58 lr: 0.002161 min_lr: 0.002161 loss: 2.3034 (2.3034) weight_decay: 0.0500 (0.0500) time: 9.4168 data: 8.6254 max mem: 64948 Epoch: [224] [ 10/312] eta: 0:07:48 lr: 0.002161 min_lr: 0.002161 loss: 2.1337 (2.0599) weight_decay: 0.0500 (0.0500) time: 1.5500 data: 0.7846 max mem: 64948 Epoch: [224] [ 20/312] eta: 0:05:33 lr: 0.002160 min_lr: 0.002160 loss: 2.1337 (2.1038) weight_decay: 0.0500 (0.0500) time: 0.7280 data: 0.0004 max mem: 64948 Epoch: [224] [ 30/312] eta: 0:04:41 lr: 0.002160 min_lr: 0.002160 loss: 2.1661 (2.0708) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [224] [ 40/312] eta: 0:04:11 lr: 0.002159 min_lr: 0.002159 loss: 2.1742 (2.0910) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0003 max mem: 64948 Epoch: [224] [ 50/312] eta: 0:03:50 lr: 0.002159 min_lr: 0.002159 loss: 2.2478 (2.1229) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [224] [ 60/312] eta: 0:03:34 lr: 0.002158 min_lr: 0.002158 loss: 2.0768 (2.0895) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [224] [ 70/312] eta: 0:03:20 lr: 0.002158 min_lr: 0.002158 loss: 1.9984 (2.0886) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [224] [ 80/312] eta: 0:03:08 lr: 0.002157 min_lr: 0.002157 loss: 1.9984 (2.0758) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [224] [ 90/312] eta: 0:02:57 lr: 0.002157 min_lr: 0.002157 loss: 1.9387 (2.0684) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [224] [100/312] eta: 0:02:47 lr: 0.002156 min_lr: 0.002156 loss: 1.8606 (2.0562) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [224] [110/312] eta: 0:02:37 lr: 0.002156 min_lr: 0.002156 loss: 1.9476 (2.0509) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [224] [120/312] eta: 0:02:28 lr: 0.002155 min_lr: 0.002155 loss: 1.9476 (2.0381) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [224] [130/312] eta: 0:02:19 lr: 0.002155 min_lr: 0.002155 loss: 2.1025 (2.0516) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [224] [140/312] eta: 0:02:11 lr: 0.002154 min_lr: 0.002154 loss: 2.1166 (2.0405) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [224] [150/312] eta: 0:02:02 lr: 0.002154 min_lr: 0.002154 loss: 2.0141 (2.0330) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [224] [160/312] eta: 0:01:54 lr: 0.002154 min_lr: 0.002154 loss: 2.1347 (2.0424) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [224] [170/312] eta: 0:01:46 lr: 0.002153 min_lr: 0.002153 loss: 2.1379 (2.0395) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [224] [180/312] eta: 0:01:38 lr: 0.002153 min_lr: 0.002153 loss: 1.9686 (2.0412) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [224] [190/312] eta: 0:01:30 lr: 0.002152 min_lr: 0.002152 loss: 2.2290 (2.0534) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [224] [200/312] eta: 0:01:23 lr: 0.002152 min_lr: 0.002152 loss: 2.2569 (2.0541) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [224] [210/312] eta: 0:01:15 lr: 0.002151 min_lr: 0.002151 loss: 1.9399 (2.0400) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [224] [220/312] eta: 0:01:07 lr: 0.002151 min_lr: 0.002151 loss: 1.8950 (2.0389) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [224] [230/312] eta: 0:01:00 lr: 0.002150 min_lr: 0.002150 loss: 1.9533 (2.0411) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [224] [240/312] eta: 0:00:52 lr: 0.002150 min_lr: 0.002150 loss: 2.0931 (2.0418) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [224] [250/312] eta: 0:00:45 lr: 0.002149 min_lr: 0.002149 loss: 1.9254 (2.0375) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [224] [260/312] eta: 0:00:38 lr: 0.002149 min_lr: 0.002149 loss: 2.0100 (2.0427) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [224] [270/312] eta: 0:00:30 lr: 0.002148 min_lr: 0.002148 loss: 2.2575 (2.0491) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [224] [280/312] eta: 0:00:23 lr: 0.002148 min_lr: 0.002148 loss: 2.1578 (2.0491) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [224] [290/312] eta: 0:00:15 lr: 0.002147 min_lr: 0.002147 loss: 1.9784 (2.0429) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [224] [300/312] eta: 0:00:08 lr: 0.002147 min_lr: 0.002147 loss: 2.0948 (2.0477) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [224] [310/312] eta: 0:00:01 lr: 0.002147 min_lr: 0.002147 loss: 2.1930 (2.0510) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [224] [311/312] eta: 0:00:00 lr: 0.002147 min_lr: 0.002147 loss: 2.1930 (2.0527) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [224] Total time: 0:03:47 (0.7278 s / it) Averaged stats: lr: 0.002147 min_lr: 0.002147 loss: 2.1930 (2.0804) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6035 (0.6035) acc1: 85.4167 (85.4167) acc5: 95.8333 (95.8333) time: 4.4721 data: 4.2585 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8392 (0.8156) acc1: 78.6458 (78.1120) acc5: 95.3125 (94.5280) time: 0.6487 data: 0.4733 max mem: 64948 Test: Total time: 0:00:06 (0.6719 s / it) * Acc@1 79.436 Acc@5 94.738 loss 0.798 Accuracy of the model on the 50000 test images: 79.4% Max accuracy: 79.44% Test: [0/9] eta: 0:00:40 loss: 0.5125 (0.5125) acc1: 85.6771 (85.6771) acc5: 97.9167 (97.9167) time: 4.4963 data: 4.2872 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7131 (0.6978) acc1: 81.5104 (80.3520) acc5: 96.8750 (96.2240) time: 0.6508 data: 0.4765 max mem: 64948 Test: Total time: 0:00:05 (0.6611 s / it) * Acc@1 81.720 Acc@5 95.924 loss 0.673 Accuracy of the model EMA on 50000 test images: 81.7% Max EMA accuracy: 81.72% Epoch: [225] [ 0/312] eta: 0:47:23 lr: 0.002146 min_lr: 0.002146 loss: 2.0354 (2.0354) weight_decay: 0.0500 (0.0500) time: 9.1123 data: 8.2397 max mem: 64948 Epoch: [225] [ 10/312] eta: 0:07:28 lr: 0.002146 min_lr: 0.002146 loss: 2.2231 (2.0785) weight_decay: 0.0500 (0.0500) time: 1.4838 data: 0.7495 max mem: 64948 Epoch: [225] [ 20/312] eta: 0:05:24 lr: 0.002146 min_lr: 0.002146 loss: 2.1345 (2.0266) weight_decay: 0.0500 (0.0500) time: 0.7107 data: 0.0004 max mem: 64948 Epoch: [225] [ 30/312] eta: 0:04:35 lr: 0.002145 min_lr: 0.002145 loss: 1.8496 (1.9668) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [225] [ 40/312] eta: 0:04:06 lr: 0.002145 min_lr: 0.002145 loss: 1.9660 (2.0116) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [225] [ 50/312] eta: 0:03:46 lr: 0.002144 min_lr: 0.002144 loss: 2.1413 (2.0116) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [225] [ 60/312] eta: 0:03:31 lr: 0.002144 min_lr: 0.002144 loss: 1.9552 (1.9999) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [225] [ 70/312] eta: 0:03:17 lr: 0.002143 min_lr: 0.002143 loss: 2.1703 (2.0360) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [225] [ 80/312] eta: 0:03:06 lr: 0.002143 min_lr: 0.002143 loss: 2.1703 (2.0212) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [225] [ 90/312] eta: 0:02:55 lr: 0.002142 min_lr: 0.002142 loss: 2.1390 (2.0404) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [225] [100/312] eta: 0:02:45 lr: 0.002142 min_lr: 0.002142 loss: 2.3454 (2.0580) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [225] [110/312] eta: 0:02:36 lr: 0.002141 min_lr: 0.002141 loss: 2.1344 (2.0466) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [225] [120/312] eta: 0:02:27 lr: 0.002141 min_lr: 0.002141 loss: 1.8750 (2.0457) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [225] [130/312] eta: 0:02:18 lr: 0.002140 min_lr: 0.002140 loss: 2.2184 (2.0521) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [225] [140/312] eta: 0:02:10 lr: 0.002140 min_lr: 0.002140 loss: 2.2184 (2.0627) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [225] [150/312] eta: 0:02:01 lr: 0.002139 min_lr: 0.002139 loss: 2.2225 (2.0734) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [225] [160/312] eta: 0:01:53 lr: 0.002139 min_lr: 0.002139 loss: 2.2792 (2.0762) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [225] [170/312] eta: 0:01:45 lr: 0.002139 min_lr: 0.002139 loss: 2.1548 (2.0732) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [225] [180/312] eta: 0:01:38 lr: 0.002138 min_lr: 0.002138 loss: 2.1992 (2.0858) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [225] [190/312] eta: 0:01:30 lr: 0.002138 min_lr: 0.002138 loss: 2.2573 (2.0917) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [225] [200/312] eta: 0:01:22 lr: 0.002137 min_lr: 0.002137 loss: 2.2366 (2.0919) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [225] [210/312] eta: 0:01:15 lr: 0.002137 min_lr: 0.002137 loss: 2.2032 (2.0981) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [225] [220/312] eta: 0:01:07 lr: 0.002136 min_lr: 0.002136 loss: 2.2271 (2.1006) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [225] [230/312] eta: 0:01:00 lr: 0.002136 min_lr: 0.002136 loss: 2.1044 (2.0964) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [225] [240/312] eta: 0:00:52 lr: 0.002135 min_lr: 0.002135 loss: 2.0458 (2.0929) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [225] [250/312] eta: 0:00:45 lr: 0.002135 min_lr: 0.002135 loss: 2.1531 (2.0944) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [225] [260/312] eta: 0:00:37 lr: 0.002134 min_lr: 0.002134 loss: 2.2203 (2.0959) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [225] [270/312] eta: 0:00:30 lr: 0.002134 min_lr: 0.002134 loss: 2.2348 (2.0942) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [225] [280/312] eta: 0:00:23 lr: 0.002133 min_lr: 0.002133 loss: 2.1956 (2.0941) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [225] [290/312] eta: 0:00:15 lr: 0.002133 min_lr: 0.002133 loss: 2.1123 (2.0898) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [225] [300/312] eta: 0:00:08 lr: 0.002132 min_lr: 0.002132 loss: 1.9851 (2.0875) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [225] [310/312] eta: 0:00:01 lr: 0.002132 min_lr: 0.002132 loss: 2.1059 (2.0885) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [225] [311/312] eta: 0:00:00 lr: 0.002132 min_lr: 0.002132 loss: 2.1059 (2.0875) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [225] Total time: 0:03:46 (0.7249 s / it) Averaged stats: lr: 0.002132 min_lr: 0.002132 loss: 2.1059 (2.0895) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.6489 (0.6489) acc1: 83.8542 (83.8542) acc5: 95.8333 (95.8333) time: 4.2606 data: 4.0435 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.9238 (0.8791) acc1: 78.1250 (77.2800) acc5: 95.0521 (94.1440) time: 0.6352 data: 0.4599 max mem: 64948 Test: Total time: 0:00:05 (0.6528 s / it) * Acc@1 78.514 Acc@5 94.504 loss 0.837 Accuracy of the model on the 50000 test images: 78.5% Max accuracy: 79.44% Test: [0/9] eta: 0:00:44 loss: 0.5125 (0.5125) acc1: 85.9375 (85.9375) acc5: 97.9167 (97.9167) time: 4.9581 data: 4.7504 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7125 (0.6974) acc1: 81.2500 (80.3520) acc5: 96.6146 (96.3200) time: 0.7054 data: 0.5279 max mem: 64948 Test: Total time: 0:00:06 (0.7149 s / it) * Acc@1 81.746 Acc@5 95.942 loss 0.672 Accuracy of the model EMA on 50000 test images: 81.7% Max EMA accuracy: 81.75% Epoch: [226] [ 0/312] eta: 0:50:39 lr: 0.002132 min_lr: 0.002132 loss: 2.0807 (2.0807) weight_decay: 0.0500 (0.0500) time: 9.7436 data: 8.9515 max mem: 64948 Epoch: [226] [ 10/312] eta: 0:07:47 lr: 0.002131 min_lr: 0.002131 loss: 2.0807 (2.1497) weight_decay: 0.0500 (0.0500) time: 1.5492 data: 0.8142 max mem: 64948 Epoch: [226] [ 20/312] eta: 0:05:33 lr: 0.002131 min_lr: 0.002131 loss: 2.1006 (2.1504) weight_decay: 0.0500 (0.0500) time: 0.7112 data: 0.0004 max mem: 64948 Epoch: [226] [ 30/312] eta: 0:04:41 lr: 0.002130 min_lr: 0.002130 loss: 2.2546 (2.1606) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [226] [ 40/312] eta: 0:04:11 lr: 0.002130 min_lr: 0.002130 loss: 2.1151 (2.1042) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [226] [ 50/312] eta: 0:03:50 lr: 0.002130 min_lr: 0.002130 loss: 2.1106 (2.1106) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [226] [ 60/312] eta: 0:03:34 lr: 0.002129 min_lr: 0.002129 loss: 2.2335 (2.1217) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [226] [ 70/312] eta: 0:03:20 lr: 0.002129 min_lr: 0.002129 loss: 2.0510 (2.0963) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [226] [ 80/312] eta: 0:03:08 lr: 0.002128 min_lr: 0.002128 loss: 2.0244 (2.1000) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [226] [ 90/312] eta: 0:02:57 lr: 0.002128 min_lr: 0.002128 loss: 2.1426 (2.1199) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [226] [100/312] eta: 0:02:47 lr: 0.002127 min_lr: 0.002127 loss: 2.2255 (2.1165) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [226] [110/312] eta: 0:02:37 lr: 0.002127 min_lr: 0.002127 loss: 2.1546 (2.1106) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [226] [120/312] eta: 0:02:28 lr: 0.002126 min_lr: 0.002126 loss: 2.1305 (2.1049) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [226] [130/312] eta: 0:02:19 lr: 0.002126 min_lr: 0.002126 loss: 2.1243 (2.1078) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [226] [140/312] eta: 0:02:11 lr: 0.002125 min_lr: 0.002125 loss: 2.1825 (2.1067) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [226] [150/312] eta: 0:02:02 lr: 0.002125 min_lr: 0.002125 loss: 2.1562 (2.1125) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [226] [160/312] eta: 0:01:54 lr: 0.002124 min_lr: 0.002124 loss: 2.1372 (2.1147) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [226] [170/312] eta: 0:01:46 lr: 0.002124 min_lr: 0.002124 loss: 2.1571 (2.1162) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [226] [180/312] eta: 0:01:38 lr: 0.002123 min_lr: 0.002123 loss: 2.1571 (2.1176) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [226] [190/312] eta: 0:01:30 lr: 0.002123 min_lr: 0.002123 loss: 2.2891 (2.1191) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [226] [200/312] eta: 0:01:23 lr: 0.002123 min_lr: 0.002123 loss: 2.2891 (2.1234) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [226] [210/312] eta: 0:01:15 lr: 0.002122 min_lr: 0.002122 loss: 2.2187 (2.1232) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [226] [220/312] eta: 0:01:07 lr: 0.002122 min_lr: 0.002122 loss: 2.2187 (2.1261) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [226] [230/312] eta: 0:01:00 lr: 0.002121 min_lr: 0.002121 loss: 2.2133 (2.1218) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [226] [240/312] eta: 0:00:52 lr: 0.002121 min_lr: 0.002121 loss: 1.8914 (2.1142) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [226] [250/312] eta: 0:00:45 lr: 0.002120 min_lr: 0.002120 loss: 2.0516 (2.1126) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [226] [260/312] eta: 0:00:37 lr: 0.002120 min_lr: 0.002120 loss: 2.1324 (2.1086) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [226] [270/312] eta: 0:00:30 lr: 0.002119 min_lr: 0.002119 loss: 2.0809 (2.1061) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [226] [280/312] eta: 0:00:23 lr: 0.002119 min_lr: 0.002119 loss: 2.0746 (2.0975) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [226] [290/312] eta: 0:00:15 lr: 0.002118 min_lr: 0.002118 loss: 1.7491 (2.0926) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [226] [300/312] eta: 0:00:08 lr: 0.002118 min_lr: 0.002118 loss: 2.0412 (2.0928) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [226] [310/312] eta: 0:00:01 lr: 0.002117 min_lr: 0.002117 loss: 2.0412 (2.0906) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [226] [311/312] eta: 0:00:00 lr: 0.002117 min_lr: 0.002117 loss: 2.0412 (2.0909) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [226] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.002117 min_lr: 0.002117 loss: 2.0412 (2.0776) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6353 (0.6353) acc1: 84.1146 (84.1146) acc5: 96.0938 (96.0938) time: 4.7242 data: 4.5153 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8375 (0.8269) acc1: 77.8646 (78.5280) acc5: 95.0521 (94.7840) time: 0.6762 data: 0.5018 max mem: 64948 Test: Total time: 0:00:06 (0.6960 s / it) * Acc@1 78.900 Acc@5 94.620 loss 0.804 Accuracy of the model on the 50000 test images: 78.9% Max accuracy: 79.44% Test: [0/9] eta: 0:00:44 loss: 0.5125 (0.5125) acc1: 85.9375 (85.9375) acc5: 97.9167 (97.9167) time: 4.8895 data: 4.6777 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7118 (0.6972) acc1: 81.2500 (80.3200) acc5: 96.8750 (96.3200) time: 0.6946 data: 0.5199 max mem: 64948 Test: Total time: 0:00:06 (0.7033 s / it) * Acc@1 81.748 Acc@5 95.960 loss 0.672 Accuracy of the model EMA on 50000 test images: 81.7% Max EMA accuracy: 81.75% Epoch: [227] [ 0/312] eta: 0:52:25 lr: 0.002117 min_lr: 0.002117 loss: 2.5513 (2.5513) weight_decay: 0.0500 (0.0500) time: 10.0814 data: 9.3048 max mem: 64948 Epoch: [227] [ 10/312] eta: 0:07:53 lr: 0.002117 min_lr: 0.002117 loss: 2.1071 (2.1563) weight_decay: 0.0500 (0.0500) time: 1.5686 data: 0.8462 max mem: 64948 Epoch: [227] [ 20/312] eta: 0:05:36 lr: 0.002116 min_lr: 0.002116 loss: 2.0235 (2.0572) weight_decay: 0.0500 (0.0500) time: 0.7053 data: 0.0004 max mem: 64948 Epoch: [227] [ 30/312] eta: 0:04:43 lr: 0.002116 min_lr: 0.002116 loss: 2.1557 (2.1044) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [227] [ 40/312] eta: 0:04:12 lr: 0.002115 min_lr: 0.002115 loss: 2.2162 (2.1239) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [227] [ 50/312] eta: 0:03:51 lr: 0.002115 min_lr: 0.002115 loss: 2.1195 (2.0942) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [227] [ 60/312] eta: 0:03:35 lr: 0.002114 min_lr: 0.002114 loss: 1.9454 (2.0585) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [227] [ 70/312] eta: 0:03:21 lr: 0.002114 min_lr: 0.002114 loss: 1.9502 (2.0720) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [227] [ 80/312] eta: 0:03:08 lr: 0.002114 min_lr: 0.002114 loss: 2.2987 (2.0674) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [227] [ 90/312] eta: 0:02:57 lr: 0.002113 min_lr: 0.002113 loss: 2.2859 (2.0829) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [227] [100/312] eta: 0:02:47 lr: 0.002113 min_lr: 0.002113 loss: 2.0935 (2.0746) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [227] [110/312] eta: 0:02:37 lr: 0.002112 min_lr: 0.002112 loss: 1.9426 (2.0635) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [227] [120/312] eta: 0:02:28 lr: 0.002112 min_lr: 0.002112 loss: 2.0955 (2.0716) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [227] [130/312] eta: 0:02:19 lr: 0.002111 min_lr: 0.002111 loss: 2.2626 (2.0857) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [227] [140/312] eta: 0:02:11 lr: 0.002111 min_lr: 0.002111 loss: 2.2626 (2.0970) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [227] [150/312] eta: 0:02:02 lr: 0.002110 min_lr: 0.002110 loss: 2.2477 (2.0998) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [227] [160/312] eta: 0:01:54 lr: 0.002110 min_lr: 0.002110 loss: 2.1569 (2.0978) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [227] [170/312] eta: 0:01:46 lr: 0.002109 min_lr: 0.002109 loss: 2.0221 (2.0912) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [227] [180/312] eta: 0:01:38 lr: 0.002109 min_lr: 0.002109 loss: 2.0155 (2.0869) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [227] [190/312] eta: 0:01:30 lr: 0.002108 min_lr: 0.002108 loss: 2.1353 (2.0913) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [227] [200/312] eta: 0:01:23 lr: 0.002108 min_lr: 0.002108 loss: 2.2482 (2.0986) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [227] [210/312] eta: 0:01:15 lr: 0.002107 min_lr: 0.002107 loss: 2.2537 (2.1014) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [227] [220/312] eta: 0:01:07 lr: 0.002107 min_lr: 0.002107 loss: 2.2298 (2.1070) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [227] [230/312] eta: 0:01:00 lr: 0.002107 min_lr: 0.002107 loss: 2.2679 (2.1125) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [227] [240/312] eta: 0:00:52 lr: 0.002106 min_lr: 0.002106 loss: 2.1225 (2.1061) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [227] [250/312] eta: 0:00:45 lr: 0.002106 min_lr: 0.002106 loss: 1.9796 (2.0985) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [227] [260/312] eta: 0:00:38 lr: 0.002105 min_lr: 0.002105 loss: 2.0413 (2.0955) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [227] [270/312] eta: 0:00:30 lr: 0.002105 min_lr: 0.002105 loss: 1.9184 (2.0891) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [227] [280/312] eta: 0:00:23 lr: 0.002104 min_lr: 0.002104 loss: 1.9184 (2.0850) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [227] [290/312] eta: 0:00:16 lr: 0.002104 min_lr: 0.002104 loss: 1.9909 (2.0830) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [227] [300/312] eta: 0:00:08 lr: 0.002103 min_lr: 0.002103 loss: 2.1062 (2.0879) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [227] [310/312] eta: 0:00:01 lr: 0.002103 min_lr: 0.002103 loss: 2.2302 (2.0858) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [227] [311/312] eta: 0:00:00 lr: 0.002103 min_lr: 0.002103 loss: 2.2302 (2.0853) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [227] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.002103 min_lr: 0.002103 loss: 2.2302 (2.0802) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5725 (0.5725) acc1: 85.1562 (85.1562) acc5: 95.0521 (95.0521) time: 4.4349 data: 4.2153 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8082 (0.8011) acc1: 79.4271 (78.5600) acc5: 95.0521 (94.5920) time: 0.6447 data: 0.4684 max mem: 64948 Test: Total time: 0:00:06 (0.6679 s / it) * Acc@1 79.186 Acc@5 94.740 loss 0.793 Accuracy of the model on the 50000 test images: 79.2% Max accuracy: 79.44% Test: [0/9] eta: 0:00:45 loss: 0.5120 (0.5120) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 5.0747 data: 4.8632 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7115 (0.6967) acc1: 81.2500 (80.3840) acc5: 96.8750 (96.3200) time: 0.7152 data: 0.5405 max mem: 64948 Test: Total time: 0:00:06 (0.7245 s / it) * Acc@1 81.776 Acc@5 95.974 loss 0.671 Accuracy of the model EMA on 50000 test images: 81.8% Max EMA accuracy: 81.78% Epoch: [228] [ 0/312] eta: 0:47:03 lr: 0.002103 min_lr: 0.002103 loss: 2.4038 (2.4038) weight_decay: 0.0500 (0.0500) time: 9.0491 data: 7.5663 max mem: 64948 Epoch: [228] [ 10/312] eta: 0:07:54 lr: 0.002102 min_lr: 0.002102 loss: 2.0204 (1.9907) weight_decay: 0.0500 (0.0500) time: 1.5698 data: 0.7787 max mem: 64948 Epoch: [228] [ 20/312] eta: 0:05:36 lr: 0.002102 min_lr: 0.002102 loss: 2.1201 (2.0694) weight_decay: 0.0500 (0.0500) time: 0.7587 data: 0.0502 max mem: 64948 Epoch: [228] [ 30/312] eta: 0:04:43 lr: 0.002101 min_lr: 0.002101 loss: 2.1201 (2.0695) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [228] [ 40/312] eta: 0:04:13 lr: 0.002101 min_lr: 0.002101 loss: 2.0326 (2.0796) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [228] [ 50/312] eta: 0:03:51 lr: 0.002100 min_lr: 0.002100 loss: 2.1244 (2.0987) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [228] [ 60/312] eta: 0:03:35 lr: 0.002100 min_lr: 0.002100 loss: 2.0779 (2.0592) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [228] [ 70/312] eta: 0:03:21 lr: 0.002099 min_lr: 0.002099 loss: 1.8632 (2.0560) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [228] [ 80/312] eta: 0:03:09 lr: 0.002099 min_lr: 0.002099 loss: 2.0841 (2.0498) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [228] [ 90/312] eta: 0:02:58 lr: 0.002099 min_lr: 0.002099 loss: 2.0841 (2.0472) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [228] [100/312] eta: 0:02:47 lr: 0.002098 min_lr: 0.002098 loss: 2.0678 (2.0558) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [228] [110/312] eta: 0:02:38 lr: 0.002098 min_lr: 0.002098 loss: 2.2371 (2.0816) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [228] [120/312] eta: 0:02:28 lr: 0.002097 min_lr: 0.002097 loss: 2.2392 (2.0726) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [228] [130/312] eta: 0:02:20 lr: 0.002097 min_lr: 0.002097 loss: 2.2437 (2.0766) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [228] [140/312] eta: 0:02:11 lr: 0.002096 min_lr: 0.002096 loss: 2.2442 (2.0847) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [228] [150/312] eta: 0:02:03 lr: 0.002096 min_lr: 0.002096 loss: 2.2176 (2.0737) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [228] [160/312] eta: 0:01:54 lr: 0.002095 min_lr: 0.002095 loss: 1.8732 (2.0672) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [228] [170/312] eta: 0:01:46 lr: 0.002095 min_lr: 0.002095 loss: 2.1243 (2.0706) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [228] [180/312] eta: 0:01:38 lr: 0.002094 min_lr: 0.002094 loss: 2.3589 (2.0831) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [228] [190/312] eta: 0:01:30 lr: 0.002094 min_lr: 0.002094 loss: 1.9975 (2.0667) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [228] [200/312] eta: 0:01:23 lr: 0.002093 min_lr: 0.002093 loss: 1.8342 (2.0610) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [228] [210/312] eta: 0:01:15 lr: 0.002093 min_lr: 0.002093 loss: 1.9327 (2.0546) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [228] [220/312] eta: 0:01:07 lr: 0.002092 min_lr: 0.002092 loss: 1.9439 (2.0506) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [228] [230/312] eta: 0:01:00 lr: 0.002092 min_lr: 0.002092 loss: 2.1389 (2.0486) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [228] [240/312] eta: 0:00:52 lr: 0.002091 min_lr: 0.002091 loss: 2.0290 (2.0438) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [228] [250/312] eta: 0:00:45 lr: 0.002091 min_lr: 0.002091 loss: 2.1283 (2.0478) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [228] [260/312] eta: 0:00:38 lr: 0.002091 min_lr: 0.002091 loss: 2.1283 (2.0515) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0003 max mem: 64948 Epoch: [228] [270/312] eta: 0:00:30 lr: 0.002090 min_lr: 0.002090 loss: 2.1351 (2.0553) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [228] [280/312] eta: 0:00:23 lr: 0.002090 min_lr: 0.002090 loss: 2.2791 (2.0616) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0009 max mem: 64948 Epoch: [228] [290/312] eta: 0:00:16 lr: 0.002089 min_lr: 0.002089 loss: 2.0743 (2.0540) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [228] [300/312] eta: 0:00:08 lr: 0.002089 min_lr: 0.002089 loss: 2.0440 (2.0558) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [228] [310/312] eta: 0:00:01 lr: 0.002088 min_lr: 0.002088 loss: 2.2229 (2.0666) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [228] [311/312] eta: 0:00:00 lr: 0.002088 min_lr: 0.002088 loss: 2.2130 (2.0660) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [228] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.002088 min_lr: 0.002088 loss: 2.2130 (2.0732) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6301 (0.6301) acc1: 82.5521 (82.5521) acc5: 95.0521 (95.0521) time: 4.7074 data: 4.4890 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8445 (0.8404) acc1: 80.2083 (78.3680) acc5: 94.0104 (94.3360) time: 0.6743 data: 0.4989 max mem: 64948 Test: Total time: 0:00:06 (0.6968 s / it) * Acc@1 78.706 Acc@5 94.420 loss 0.820 Accuracy of the model on the 50000 test images: 78.7% Max accuracy: 79.44% Test: [0/9] eta: 0:00:43 loss: 0.5112 (0.5112) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 4.7821 data: 4.5746 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7122 (0.6963) acc1: 81.2500 (80.3840) acc5: 96.8750 (96.3520) time: 0.6826 data: 0.5084 max mem: 64948 Test: Total time: 0:00:06 (0.6927 s / it) * Acc@1 81.796 Acc@5 95.996 loss 0.670 Accuracy of the model EMA on 50000 test images: 81.8% Max EMA accuracy: 81.80% Epoch: [229] [ 0/312] eta: 0:48:54 lr: 0.002088 min_lr: 0.002088 loss: 1.7727 (1.7727) weight_decay: 0.0500 (0.0500) time: 9.4057 data: 7.5350 max mem: 64948 Epoch: [229] [ 10/312] eta: 0:07:51 lr: 0.002088 min_lr: 0.002088 loss: 2.1456 (2.1291) weight_decay: 0.0500 (0.0500) time: 1.5620 data: 0.6854 max mem: 64948 Epoch: [229] [ 20/312] eta: 0:05:35 lr: 0.002087 min_lr: 0.002087 loss: 2.1892 (2.1025) weight_decay: 0.0500 (0.0500) time: 0.7354 data: 0.0004 max mem: 64948 Epoch: [229] [ 30/312] eta: 0:04:42 lr: 0.002087 min_lr: 0.002087 loss: 2.1892 (2.1033) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [229] [ 40/312] eta: 0:04:12 lr: 0.002086 min_lr: 0.002086 loss: 2.2173 (2.1125) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [229] [ 50/312] eta: 0:03:51 lr: 0.002086 min_lr: 0.002086 loss: 2.2226 (2.1108) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [229] [ 60/312] eta: 0:03:34 lr: 0.002085 min_lr: 0.002085 loss: 2.2039 (2.0902) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [229] [ 70/312] eta: 0:03:20 lr: 0.002085 min_lr: 0.002085 loss: 1.8412 (2.0570) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [229] [ 80/312] eta: 0:03:08 lr: 0.002084 min_lr: 0.002084 loss: 1.8801 (2.0504) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [229] [ 90/312] eta: 0:02:57 lr: 0.002084 min_lr: 0.002084 loss: 1.9195 (2.0420) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [229] [100/312] eta: 0:02:47 lr: 0.002083 min_lr: 0.002083 loss: 1.9622 (2.0419) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [229] [110/312] eta: 0:02:37 lr: 0.002083 min_lr: 0.002083 loss: 2.0278 (2.0487) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [229] [120/312] eta: 0:02:28 lr: 0.002083 min_lr: 0.002083 loss: 2.0785 (2.0523) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [229] [130/312] eta: 0:02:19 lr: 0.002082 min_lr: 0.002082 loss: 2.0125 (2.0503) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [229] [140/312] eta: 0:02:11 lr: 0.002082 min_lr: 0.002082 loss: 2.1084 (2.0520) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [229] [150/312] eta: 0:02:02 lr: 0.002081 min_lr: 0.002081 loss: 2.1094 (2.0503) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [229] [160/312] eta: 0:01:54 lr: 0.002081 min_lr: 0.002081 loss: 2.1740 (2.0526) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [229] [170/312] eta: 0:01:46 lr: 0.002080 min_lr: 0.002080 loss: 2.2193 (2.0632) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [229] [180/312] eta: 0:01:38 lr: 0.002080 min_lr: 0.002080 loss: 2.2761 (2.0662) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [229] [190/312] eta: 0:01:30 lr: 0.002079 min_lr: 0.002079 loss: 2.1146 (2.0653) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [229] [200/312] eta: 0:01:23 lr: 0.002079 min_lr: 0.002079 loss: 2.0175 (2.0549) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [229] [210/312] eta: 0:01:15 lr: 0.002078 min_lr: 0.002078 loss: 1.9300 (2.0557) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [229] [220/312] eta: 0:01:07 lr: 0.002078 min_lr: 0.002078 loss: 2.1115 (2.0556) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [229] [230/312] eta: 0:01:00 lr: 0.002077 min_lr: 0.002077 loss: 2.2126 (2.0606) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [229] [240/312] eta: 0:00:52 lr: 0.002077 min_lr: 0.002077 loss: 2.2327 (2.0626) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [229] [250/312] eta: 0:00:45 lr: 0.002076 min_lr: 0.002076 loss: 2.2156 (2.0687) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [229] [260/312] eta: 0:00:38 lr: 0.002076 min_lr: 0.002076 loss: 2.2156 (2.0726) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [229] [270/312] eta: 0:00:30 lr: 0.002075 min_lr: 0.002075 loss: 2.0920 (2.0692) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [229] [280/312] eta: 0:00:23 lr: 0.002075 min_lr: 0.002075 loss: 2.0977 (2.0724) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [229] [290/312] eta: 0:00:15 lr: 0.002075 min_lr: 0.002075 loss: 2.2483 (2.0712) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [229] [300/312] eta: 0:00:08 lr: 0.002074 min_lr: 0.002074 loss: 2.2574 (2.0736) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [229] [310/312] eta: 0:00:01 lr: 0.002074 min_lr: 0.002074 loss: 2.2353 (2.0774) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [229] [311/312] eta: 0:00:00 lr: 0.002074 min_lr: 0.002074 loss: 2.2245 (2.0771) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [229] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.002074 min_lr: 0.002074 loss: 2.2245 (2.0702) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.6134 (0.6134) acc1: 84.6354 (84.6354) acc5: 96.3542 (96.3542) time: 4.8550 data: 4.6338 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7788 (0.7960) acc1: 79.9479 (78.8160) acc5: 96.0938 (95.1040) time: 0.6908 data: 0.5149 max mem: 64948 Test: Total time: 0:00:06 (0.7141 s / it) * Acc@1 79.470 Acc@5 94.852 loss 0.788 Accuracy of the model on the 50000 test images: 79.5% Max accuracy: 79.47% Test: [0/9] eta: 0:00:39 loss: 0.5101 (0.5101) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.3652 data: 4.1497 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7128 (0.6957) acc1: 81.2500 (80.3840) acc5: 96.8750 (96.3520) time: 0.6363 data: 0.4612 max mem: 64948 Test: Total time: 0:00:05 (0.6454 s / it) * Acc@1 81.808 Acc@5 95.994 loss 0.670 Accuracy of the model EMA on 50000 test images: 81.8% Max EMA accuracy: 81.81% Epoch: [230] [ 0/312] eta: 0:52:48 lr: 0.002074 min_lr: 0.002074 loss: 1.6462 (1.6462) weight_decay: 0.0500 (0.0500) time: 10.1548 data: 9.3934 max mem: 64948 Epoch: [230] [ 10/312] eta: 0:08:04 lr: 0.002073 min_lr: 0.002073 loss: 2.2076 (2.1126) weight_decay: 0.0500 (0.0500) time: 1.6032 data: 0.8543 max mem: 64948 Epoch: [230] [ 20/312] eta: 0:05:41 lr: 0.002073 min_lr: 0.002073 loss: 2.2563 (2.0983) weight_decay: 0.0500 (0.0500) time: 0.7212 data: 0.0004 max mem: 64948 Epoch: [230] [ 30/312] eta: 0:04:46 lr: 0.002072 min_lr: 0.002072 loss: 1.9179 (2.0104) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [230] [ 40/312] eta: 0:04:15 lr: 0.002072 min_lr: 0.002072 loss: 1.9413 (2.0108) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [230] [ 50/312] eta: 0:03:53 lr: 0.002071 min_lr: 0.002071 loss: 2.1302 (2.0328) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [230] [ 60/312] eta: 0:03:36 lr: 0.002071 min_lr: 0.002071 loss: 2.0615 (2.0152) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [230] [ 70/312] eta: 0:03:22 lr: 0.002070 min_lr: 0.002070 loss: 2.1284 (2.0264) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [230] [ 80/312] eta: 0:03:09 lr: 0.002070 min_lr: 0.002070 loss: 2.1284 (2.0299) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [230] [ 90/312] eta: 0:02:58 lr: 0.002069 min_lr: 0.002069 loss: 2.1227 (2.0402) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [230] [100/312] eta: 0:02:48 lr: 0.002069 min_lr: 0.002069 loss: 2.1281 (2.0332) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [230] [110/312] eta: 0:02:38 lr: 0.002068 min_lr: 0.002068 loss: 2.1414 (2.0361) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [230] [120/312] eta: 0:02:29 lr: 0.002068 min_lr: 0.002068 loss: 2.1414 (2.0351) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [230] [130/312] eta: 0:02:20 lr: 0.002067 min_lr: 0.002067 loss: 1.9803 (2.0205) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [230] [140/312] eta: 0:02:11 lr: 0.002067 min_lr: 0.002067 loss: 1.8912 (2.0225) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [230] [150/312] eta: 0:02:03 lr: 0.002067 min_lr: 0.002067 loss: 1.9552 (2.0186) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [230] [160/312] eta: 0:01:55 lr: 0.002066 min_lr: 0.002066 loss: 2.1556 (2.0176) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [230] [170/312] eta: 0:01:46 lr: 0.002066 min_lr: 0.002066 loss: 2.1556 (2.0146) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [230] [180/312] eta: 0:01:38 lr: 0.002065 min_lr: 0.002065 loss: 2.1105 (2.0165) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [230] [190/312] eta: 0:01:31 lr: 0.002065 min_lr: 0.002065 loss: 2.1802 (2.0226) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [230] [200/312] eta: 0:01:23 lr: 0.002064 min_lr: 0.002064 loss: 2.1931 (2.0188) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [230] [210/312] eta: 0:01:15 lr: 0.002064 min_lr: 0.002064 loss: 2.0558 (2.0315) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [230] [220/312] eta: 0:01:08 lr: 0.002063 min_lr: 0.002063 loss: 2.3034 (2.0359) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [230] [230/312] eta: 0:01:00 lr: 0.002063 min_lr: 0.002063 loss: 2.2135 (2.0428) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [230] [240/312] eta: 0:00:52 lr: 0.002062 min_lr: 0.002062 loss: 2.2687 (2.0458) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [230] [250/312] eta: 0:00:45 lr: 0.002062 min_lr: 0.002062 loss: 2.2423 (2.0486) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [230] [260/312] eta: 0:00:38 lr: 0.002061 min_lr: 0.002061 loss: 2.1598 (2.0526) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [230] [270/312] eta: 0:00:30 lr: 0.002061 min_lr: 0.002061 loss: 2.1407 (2.0515) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [230] [280/312] eta: 0:00:23 lr: 0.002060 min_lr: 0.002060 loss: 2.1407 (2.0481) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0010 max mem: 64948 Epoch: [230] [290/312] eta: 0:00:16 lr: 0.002060 min_lr: 0.002060 loss: 2.1685 (2.0489) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0008 max mem: 64948 Epoch: [230] [300/312] eta: 0:00:08 lr: 0.002059 min_lr: 0.002059 loss: 1.9383 (2.0454) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [230] [310/312] eta: 0:00:01 lr: 0.002059 min_lr: 0.002059 loss: 1.9896 (2.0473) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [230] [311/312] eta: 0:00:00 lr: 0.002059 min_lr: 0.002059 loss: 2.0573 (2.0478) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [230] Total time: 0:03:47 (0.7290 s / it) Averaged stats: lr: 0.002059 min_lr: 0.002059 loss: 2.0573 (2.0687) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.7010 (0.7010) acc1: 83.0729 (83.0729) acc5: 95.0521 (95.0521) time: 4.6461 data: 4.4319 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8645 (0.8471) acc1: 78.6458 (78.2080) acc5: 95.0521 (94.2400) time: 0.6675 data: 0.4925 max mem: 64948 Test: Total time: 0:00:06 (0.6919 s / it) * Acc@1 78.924 Acc@5 94.492 loss 0.810 Accuracy of the model on the 50000 test images: 78.9% Max accuracy: 79.47% Test: [0/9] eta: 0:00:42 loss: 0.5096 (0.5096) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.7268 data: 4.5126 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7129 (0.6950) acc1: 80.9896 (80.2560) acc5: 96.8750 (96.3520) time: 0.6945 data: 0.5195 max mem: 64948 Test: Total time: 0:00:06 (0.7032 s / it) * Acc@1 81.836 Acc@5 95.988 loss 0.669 Accuracy of the model EMA on 50000 test images: 81.8% Max EMA accuracy: 81.84% Epoch: [231] [ 0/312] eta: 0:48:39 lr: 0.002059 min_lr: 0.002059 loss: 2.2564 (2.2564) weight_decay: 0.0500 (0.0500) time: 9.3558 data: 7.6950 max mem: 64948 Epoch: [231] [ 10/312] eta: 0:07:40 lr: 0.002058 min_lr: 0.002058 loss: 2.2564 (2.1532) weight_decay: 0.0500 (0.0500) time: 1.5261 data: 0.7184 max mem: 64948 Epoch: [231] [ 20/312] eta: 0:05:30 lr: 0.002058 min_lr: 0.002058 loss: 2.1727 (2.1477) weight_decay: 0.0500 (0.0500) time: 0.7195 data: 0.0105 max mem: 64948 Epoch: [231] [ 30/312] eta: 0:04:38 lr: 0.002058 min_lr: 0.002058 loss: 2.1553 (2.1267) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [231] [ 40/312] eta: 0:04:09 lr: 0.002057 min_lr: 0.002057 loss: 2.1713 (2.1621) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [231] [ 50/312] eta: 0:03:49 lr: 0.002057 min_lr: 0.002057 loss: 2.2863 (2.1531) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [231] [ 60/312] eta: 0:03:33 lr: 0.002056 min_lr: 0.002056 loss: 2.1023 (2.1320) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [231] [ 70/312] eta: 0:03:19 lr: 0.002056 min_lr: 0.002056 loss: 2.0927 (2.1055) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [231] [ 80/312] eta: 0:03:07 lr: 0.002055 min_lr: 0.002055 loss: 2.0551 (2.1042) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [231] [ 90/312] eta: 0:02:56 lr: 0.002055 min_lr: 0.002055 loss: 2.0868 (2.0965) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [231] [100/312] eta: 0:02:46 lr: 0.002054 min_lr: 0.002054 loss: 2.0116 (2.0769) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [231] [110/312] eta: 0:02:37 lr: 0.002054 min_lr: 0.002054 loss: 1.9480 (2.0702) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [231] [120/312] eta: 0:02:28 lr: 0.002053 min_lr: 0.002053 loss: 1.8806 (2.0496) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [231] [130/312] eta: 0:02:19 lr: 0.002053 min_lr: 0.002053 loss: 1.9266 (2.0566) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [231] [140/312] eta: 0:02:10 lr: 0.002052 min_lr: 0.002052 loss: 2.1783 (2.0641) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [231] [150/312] eta: 0:02:02 lr: 0.002052 min_lr: 0.002052 loss: 2.1377 (2.0693) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [231] [160/312] eta: 0:01:54 lr: 0.002051 min_lr: 0.002051 loss: 2.1377 (2.0813) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [231] [170/312] eta: 0:01:46 lr: 0.002051 min_lr: 0.002051 loss: 2.2405 (2.0805) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [231] [180/312] eta: 0:01:38 lr: 0.002051 min_lr: 0.002051 loss: 2.0329 (2.0714) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [231] [190/312] eta: 0:01:30 lr: 0.002050 min_lr: 0.002050 loss: 2.1231 (2.0786) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [231] [200/312] eta: 0:01:22 lr: 0.002050 min_lr: 0.002050 loss: 2.2188 (2.0837) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [231] [210/312] eta: 0:01:15 lr: 0.002049 min_lr: 0.002049 loss: 2.2091 (2.0908) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [231] [220/312] eta: 0:01:07 lr: 0.002049 min_lr: 0.002049 loss: 2.1938 (2.0959) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [231] [230/312] eta: 0:01:00 lr: 0.002048 min_lr: 0.002048 loss: 2.1921 (2.0951) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [231] [240/312] eta: 0:00:52 lr: 0.002048 min_lr: 0.002048 loss: 2.1852 (2.0969) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [231] [250/312] eta: 0:00:45 lr: 0.002047 min_lr: 0.002047 loss: 2.0790 (2.0949) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [231] [260/312] eta: 0:00:37 lr: 0.002047 min_lr: 0.002047 loss: 1.8429 (2.0882) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [231] [270/312] eta: 0:00:30 lr: 0.002046 min_lr: 0.002046 loss: 1.9250 (2.0881) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [231] [280/312] eta: 0:00:23 lr: 0.002046 min_lr: 0.002046 loss: 2.0844 (2.0880) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0010 max mem: 64948 Epoch: [231] [290/312] eta: 0:00:15 lr: 0.002045 min_lr: 0.002045 loss: 2.0658 (2.0847) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [231] [300/312] eta: 0:00:08 lr: 0.002045 min_lr: 0.002045 loss: 1.9999 (2.0810) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [231] [310/312] eta: 0:00:01 lr: 0.002044 min_lr: 0.002044 loss: 2.1037 (2.0779) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [231] [311/312] eta: 0:00:00 lr: 0.002044 min_lr: 0.002044 loss: 2.1059 (2.0783) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [231] Total time: 0:03:46 (0.7267 s / it) Averaged stats: lr: 0.002044 min_lr: 0.002044 loss: 2.1059 (2.0648) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6277 (0.6277) acc1: 83.5938 (83.5938) acc5: 96.3542 (96.3542) time: 4.4558 data: 4.2362 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8532 (0.8205) acc1: 79.9479 (78.0800) acc5: 95.0521 (94.5920) time: 0.6467 data: 0.4708 max mem: 64948 Test: Total time: 0:00:06 (0.6711 s / it) * Acc@1 79.046 Acc@5 94.752 loss 0.798 Accuracy of the model on the 50000 test images: 79.0% Max accuracy: 79.47% Test: [0/9] eta: 0:00:45 loss: 0.5098 (0.5098) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 5.0488 data: 4.8415 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7132 (0.6944) acc1: 81.2500 (80.3840) acc5: 96.8750 (96.3840) time: 0.7123 data: 0.5380 max mem: 64948 Test: Total time: 0:00:06 (0.7219 s / it) * Acc@1 81.848 Acc@5 96.010 loss 0.668 Accuracy of the model EMA on 50000 test images: 81.8% Max EMA accuracy: 81.85% Epoch: [232] [ 0/312] eta: 0:44:48 lr: 0.002044 min_lr: 0.002044 loss: 2.1819 (2.1819) weight_decay: 0.0500 (0.0500) time: 8.6177 data: 7.6016 max mem: 64948 Epoch: [232] [ 10/312] eta: 0:07:37 lr: 0.002044 min_lr: 0.002044 loss: 2.2946 (2.2463) weight_decay: 0.0500 (0.0500) time: 1.5140 data: 0.6915 max mem: 64948 Epoch: [232] [ 20/312] eta: 0:05:28 lr: 0.002043 min_lr: 0.002043 loss: 2.1402 (2.0867) weight_decay: 0.0500 (0.0500) time: 0.7492 data: 0.0004 max mem: 64948 Epoch: [232] [ 30/312] eta: 0:04:38 lr: 0.002043 min_lr: 0.002043 loss: 1.8952 (2.0773) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0003 max mem: 64948 Epoch: [232] [ 40/312] eta: 0:04:09 lr: 0.002042 min_lr: 0.002042 loss: 1.8952 (2.0492) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0003 max mem: 64948 Epoch: [232] [ 50/312] eta: 0:03:48 lr: 0.002042 min_lr: 0.002042 loss: 2.0972 (2.0488) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [232] [ 60/312] eta: 0:03:32 lr: 0.002042 min_lr: 0.002042 loss: 2.0837 (2.0416) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [232] [ 70/312] eta: 0:03:19 lr: 0.002041 min_lr: 0.002041 loss: 2.1908 (2.0608) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [232] [ 80/312] eta: 0:03:07 lr: 0.002041 min_lr: 0.002041 loss: 2.2230 (2.0664) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [232] [ 90/312] eta: 0:02:56 lr: 0.002040 min_lr: 0.002040 loss: 2.0674 (2.0678) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [232] [100/312] eta: 0:02:46 lr: 0.002040 min_lr: 0.002040 loss: 2.0674 (2.0510) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [232] [110/312] eta: 0:02:37 lr: 0.002039 min_lr: 0.002039 loss: 2.0713 (2.0514) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [232] [120/312] eta: 0:02:28 lr: 0.002039 min_lr: 0.002039 loss: 2.1177 (2.0532) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [232] [130/312] eta: 0:02:19 lr: 0.002038 min_lr: 0.002038 loss: 1.9582 (2.0339) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [232] [140/312] eta: 0:02:10 lr: 0.002038 min_lr: 0.002038 loss: 2.0019 (2.0343) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [232] [150/312] eta: 0:02:02 lr: 0.002037 min_lr: 0.002037 loss: 2.0636 (2.0244) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [232] [160/312] eta: 0:01:54 lr: 0.002037 min_lr: 0.002037 loss: 2.0693 (2.0256) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [232] [170/312] eta: 0:01:46 lr: 0.002036 min_lr: 0.002036 loss: 2.1168 (2.0292) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [232] [180/312] eta: 0:01:38 lr: 0.002036 min_lr: 0.002036 loss: 1.9952 (2.0160) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [232] [190/312] eta: 0:01:30 lr: 0.002035 min_lr: 0.002035 loss: 1.8792 (2.0159) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [232] [200/312] eta: 0:01:22 lr: 0.002035 min_lr: 0.002035 loss: 2.1292 (2.0167) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [232] [210/312] eta: 0:01:15 lr: 0.002034 min_lr: 0.002034 loss: 2.1625 (2.0216) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [232] [220/312] eta: 0:01:07 lr: 0.002034 min_lr: 0.002034 loss: 2.0507 (2.0143) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [232] [230/312] eta: 0:01:00 lr: 0.002034 min_lr: 0.002034 loss: 1.8498 (2.0101) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [232] [240/312] eta: 0:00:52 lr: 0.002033 min_lr: 0.002033 loss: 1.9691 (2.0115) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [232] [250/312] eta: 0:00:45 lr: 0.002033 min_lr: 0.002033 loss: 1.9691 (2.0132) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [232] [260/312] eta: 0:00:37 lr: 0.002032 min_lr: 0.002032 loss: 2.2674 (2.0220) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [232] [270/312] eta: 0:00:30 lr: 0.002032 min_lr: 0.002032 loss: 2.2871 (2.0261) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [232] [280/312] eta: 0:00:23 lr: 0.002031 min_lr: 0.002031 loss: 2.1944 (2.0303) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0010 max mem: 64948 Epoch: [232] [290/312] eta: 0:00:15 lr: 0.002031 min_lr: 0.002031 loss: 2.1126 (2.0238) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0009 max mem: 64948 Epoch: [232] [300/312] eta: 0:00:08 lr: 0.002030 min_lr: 0.002030 loss: 2.1357 (2.0273) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [232] [310/312] eta: 0:00:01 lr: 0.002030 min_lr: 0.002030 loss: 2.1357 (2.0245) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [232] [311/312] eta: 0:00:00 lr: 0.002030 min_lr: 0.002030 loss: 2.1357 (2.0254) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [232] Total time: 0:03:46 (0.7267 s / it) Averaged stats: lr: 0.002030 min_lr: 0.002030 loss: 2.1357 (2.0591) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6138 (0.6138) acc1: 83.5938 (83.5938) acc5: 95.8333 (95.8333) time: 4.5548 data: 4.3454 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8013 (0.8068) acc1: 78.9062 (78.2720) acc5: 95.0521 (94.6880) time: 0.6573 data: 0.4829 max mem: 64948 Test: Total time: 0:00:06 (0.6785 s / it) * Acc@1 79.298 Acc@5 94.734 loss 0.800 Accuracy of the model on the 50000 test images: 79.3% Max accuracy: 79.47% Test: [0/9] eta: 0:00:44 loss: 0.5092 (0.5092) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 4.9280 data: 4.7187 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7137 (0.6935) acc1: 81.2500 (80.4800) acc5: 96.8750 (96.4160) time: 0.6989 data: 0.5244 max mem: 64948 Test: Total time: 0:00:06 (0.7068 s / it) * Acc@1 81.866 Acc@5 96.024 loss 0.667 Accuracy of the model EMA on 50000 test images: 81.9% Max EMA accuracy: 81.87% Epoch: [233] [ 0/312] eta: 0:47:04 lr: 0.002030 min_lr: 0.002030 loss: 2.5073 (2.5073) weight_decay: 0.0500 (0.0500) time: 9.0517 data: 7.6748 max mem: 64948 Epoch: [233] [ 10/312] eta: 0:07:47 lr: 0.002029 min_lr: 0.002029 loss: 2.2444 (2.2702) weight_decay: 0.0500 (0.0500) time: 1.5474 data: 0.6981 max mem: 64948 Epoch: [233] [ 20/312] eta: 0:05:33 lr: 0.002029 min_lr: 0.002029 loss: 2.2109 (2.2144) weight_decay: 0.0500 (0.0500) time: 0.7482 data: 0.0004 max mem: 64948 Epoch: [233] [ 30/312] eta: 0:04:42 lr: 0.002028 min_lr: 0.002028 loss: 2.1301 (2.1812) weight_decay: 0.0500 (0.0500) time: 0.7010 data: 0.0004 max mem: 64948 Epoch: [233] [ 40/312] eta: 0:04:12 lr: 0.002028 min_lr: 0.002028 loss: 2.1906 (2.1667) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [233] [ 50/312] eta: 0:03:50 lr: 0.002027 min_lr: 0.002027 loss: 2.1930 (2.1407) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [233] [ 60/312] eta: 0:03:34 lr: 0.002027 min_lr: 0.002027 loss: 2.1723 (2.1406) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [233] [ 70/312] eta: 0:03:20 lr: 0.002026 min_lr: 0.002026 loss: 2.1481 (2.1249) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [233] [ 80/312] eta: 0:03:08 lr: 0.002026 min_lr: 0.002026 loss: 1.9073 (2.1035) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [233] [ 90/312] eta: 0:02:57 lr: 0.002026 min_lr: 0.002026 loss: 1.8446 (2.0840) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [233] [100/312] eta: 0:02:47 lr: 0.002025 min_lr: 0.002025 loss: 2.1132 (2.0929) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [233] [110/312] eta: 0:02:37 lr: 0.002025 min_lr: 0.002025 loss: 1.9751 (2.0779) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [233] [120/312] eta: 0:02:28 lr: 0.002024 min_lr: 0.002024 loss: 1.9163 (2.0775) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [233] [130/312] eta: 0:02:19 lr: 0.002024 min_lr: 0.002024 loss: 2.1930 (2.0764) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [233] [140/312] eta: 0:02:11 lr: 0.002023 min_lr: 0.002023 loss: 2.2032 (2.0782) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [233] [150/312] eta: 0:02:02 lr: 0.002023 min_lr: 0.002023 loss: 2.2182 (2.0863) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [233] [160/312] eta: 0:01:54 lr: 0.002022 min_lr: 0.002022 loss: 2.1402 (2.0771) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [233] [170/312] eta: 0:01:46 lr: 0.002022 min_lr: 0.002022 loss: 2.0341 (2.0775) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [233] [180/312] eta: 0:01:38 lr: 0.002021 min_lr: 0.002021 loss: 2.1834 (2.0775) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [233] [190/312] eta: 0:01:30 lr: 0.002021 min_lr: 0.002021 loss: 2.1983 (2.0867) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [233] [200/312] eta: 0:01:23 lr: 0.002020 min_lr: 0.002020 loss: 2.3642 (2.0998) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [233] [210/312] eta: 0:01:15 lr: 0.002020 min_lr: 0.002020 loss: 2.2439 (2.0916) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [233] [220/312] eta: 0:01:07 lr: 0.002019 min_lr: 0.002019 loss: 2.1925 (2.1008) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [233] [230/312] eta: 0:01:00 lr: 0.002019 min_lr: 0.002019 loss: 2.2326 (2.0941) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [233] [240/312] eta: 0:00:52 lr: 0.002018 min_lr: 0.002018 loss: 2.1124 (2.0918) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [233] [250/312] eta: 0:00:45 lr: 0.002018 min_lr: 0.002018 loss: 2.0682 (2.0921) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [233] [260/312] eta: 0:00:37 lr: 0.002018 min_lr: 0.002018 loss: 2.2018 (2.0921) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [233] [270/312] eta: 0:00:30 lr: 0.002017 min_lr: 0.002017 loss: 2.0244 (2.0901) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [233] [280/312] eta: 0:00:23 lr: 0.002017 min_lr: 0.002017 loss: 2.0244 (2.0936) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0011 max mem: 64948 Epoch: [233] [290/312] eta: 0:00:15 lr: 0.002016 min_lr: 0.002016 loss: 2.2903 (2.1011) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0010 max mem: 64948 Epoch: [233] [300/312] eta: 0:00:08 lr: 0.002016 min_lr: 0.002016 loss: 2.3302 (2.1047) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [233] [310/312] eta: 0:00:01 lr: 0.002015 min_lr: 0.002015 loss: 2.1742 (2.1053) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [233] [311/312] eta: 0:00:00 lr: 0.002015 min_lr: 0.002015 loss: 2.1742 (2.1043) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [233] Total time: 0:03:46 (0.7274 s / it) Averaged stats: lr: 0.002015 min_lr: 0.002015 loss: 2.1742 (2.0705) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5906 (0.5906) acc1: 84.6354 (84.6354) acc5: 95.8333 (95.8333) time: 4.4866 data: 4.2667 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8302 (0.8110) acc1: 79.4271 (78.7520) acc5: 95.3125 (94.9440) time: 0.6503 data: 0.4742 max mem: 64948 Test: Total time: 0:00:06 (0.6719 s / it) * Acc@1 79.328 Acc@5 94.706 loss 0.794 Accuracy of the model on the 50000 test images: 79.3% Max accuracy: 79.47% Test: [0/9] eta: 0:00:45 loss: 0.5083 (0.5083) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 5.0754 data: 4.8723 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7142 (0.6926) acc1: 81.2500 (80.4480) acc5: 96.8750 (96.4160) time: 0.7204 data: 0.5415 max mem: 64948 Test: Total time: 0:00:06 (0.7283 s / it) * Acc@1 81.880 Acc@5 96.018 loss 0.667 Accuracy of the model EMA on 50000 test images: 81.9% Max EMA accuracy: 81.88% Epoch: [234] [ 0/312] eta: 0:52:02 lr: 0.002015 min_lr: 0.002015 loss: 1.5162 (1.5162) weight_decay: 0.0500 (0.0500) time: 10.0075 data: 8.5012 max mem: 64948 Epoch: [234] [ 10/312] eta: 0:07:51 lr: 0.002015 min_lr: 0.002015 loss: 1.9290 (1.9192) weight_decay: 0.0500 (0.0500) time: 1.5607 data: 0.7752 max mem: 64948 Epoch: [234] [ 20/312] eta: 0:05:35 lr: 0.002014 min_lr: 0.002014 loss: 2.0273 (1.9991) weight_decay: 0.0500 (0.0500) time: 0.7056 data: 0.0015 max mem: 64948 Epoch: [234] [ 30/312] eta: 0:04:42 lr: 0.002014 min_lr: 0.002014 loss: 2.1518 (2.0599) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [234] [ 40/312] eta: 0:04:12 lr: 0.002013 min_lr: 0.002013 loss: 2.0656 (2.0416) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [234] [ 50/312] eta: 0:03:51 lr: 0.002013 min_lr: 0.002013 loss: 1.9840 (2.0142) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [234] [ 60/312] eta: 0:03:34 lr: 0.002012 min_lr: 0.002012 loss: 2.1937 (2.0609) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [234] [ 70/312] eta: 0:03:21 lr: 0.002012 min_lr: 0.002012 loss: 2.1984 (2.0689) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [234] [ 80/312] eta: 0:03:09 lr: 0.002011 min_lr: 0.002011 loss: 2.0684 (2.0703) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [234] [ 90/312] eta: 0:02:57 lr: 0.002011 min_lr: 0.002011 loss: 2.2421 (2.0927) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [234] [100/312] eta: 0:02:47 lr: 0.002010 min_lr: 0.002010 loss: 2.0677 (2.0775) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [234] [110/312] eta: 0:02:38 lr: 0.002010 min_lr: 0.002010 loss: 2.0741 (2.0782) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [234] [120/312] eta: 0:02:28 lr: 0.002009 min_lr: 0.002009 loss: 2.1391 (2.0862) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [234] [130/312] eta: 0:02:20 lr: 0.002009 min_lr: 0.002009 loss: 2.2109 (2.0976) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [234] [140/312] eta: 0:02:11 lr: 0.002009 min_lr: 0.002009 loss: 2.0722 (2.0814) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0003 max mem: 64948 Epoch: [234] [150/312] eta: 0:02:03 lr: 0.002008 min_lr: 0.002008 loss: 1.9757 (2.0851) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [234] [160/312] eta: 0:01:54 lr: 0.002008 min_lr: 0.002008 loss: 2.0729 (2.0818) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [234] [170/312] eta: 0:01:46 lr: 0.002007 min_lr: 0.002007 loss: 2.0077 (2.0809) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [234] [180/312] eta: 0:01:38 lr: 0.002007 min_lr: 0.002007 loss: 2.0462 (2.0758) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [234] [190/312] eta: 0:01:31 lr: 0.002006 min_lr: 0.002006 loss: 2.1566 (2.0843) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [234] [200/312] eta: 0:01:23 lr: 0.002006 min_lr: 0.002006 loss: 2.2243 (2.0879) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [234] [210/312] eta: 0:01:15 lr: 0.002005 min_lr: 0.002005 loss: 2.2681 (2.0920) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [234] [220/312] eta: 0:01:07 lr: 0.002005 min_lr: 0.002005 loss: 2.2134 (2.0929) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [234] [230/312] eta: 0:01:00 lr: 0.002004 min_lr: 0.002004 loss: 2.0727 (2.0894) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [234] [240/312] eta: 0:00:52 lr: 0.002004 min_lr: 0.002004 loss: 2.0488 (2.0814) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [234] [250/312] eta: 0:00:45 lr: 0.002003 min_lr: 0.002003 loss: 2.0989 (2.0832) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [234] [260/312] eta: 0:00:38 lr: 0.002003 min_lr: 0.002003 loss: 2.1526 (2.0865) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [234] [270/312] eta: 0:00:30 lr: 0.002002 min_lr: 0.002002 loss: 2.1541 (2.0883) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [234] [280/312] eta: 0:00:23 lr: 0.002002 min_lr: 0.002002 loss: 2.0810 (2.0886) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [234] [290/312] eta: 0:00:16 lr: 0.002002 min_lr: 0.002002 loss: 1.9514 (2.0854) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [234] [300/312] eta: 0:00:08 lr: 0.002001 min_lr: 0.002001 loss: 2.0865 (2.0882) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [234] [310/312] eta: 0:00:01 lr: 0.002001 min_lr: 0.002001 loss: 2.0557 (2.0843) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [234] [311/312] eta: 0:00:00 lr: 0.002001 min_lr: 0.002001 loss: 2.0557 (2.0848) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [234] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.002001 min_lr: 0.002001 loss: 2.0557 (2.0643) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5998 (0.5998) acc1: 85.1562 (85.1562) acc5: 96.6146 (96.6146) time: 4.4881 data: 4.2682 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8180 (0.8170) acc1: 78.9062 (78.8480) acc5: 94.5312 (94.7840) time: 0.6506 data: 0.4743 max mem: 64948 Test: Total time: 0:00:06 (0.6769 s / it) * Acc@1 79.146 Acc@5 94.654 loss 0.805 Accuracy of the model on the 50000 test images: 79.1% Max accuracy: 79.47% Test: [0/9] eta: 0:00:42 loss: 0.5077 (0.5077) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.7094 data: 4.4973 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7144 (0.6919) acc1: 81.2500 (80.4160) acc5: 96.8750 (96.4480) time: 0.6761 data: 0.4998 max mem: 64948 Test: Total time: 0:00:06 (0.6829 s / it) * Acc@1 81.894 Acc@5 96.028 loss 0.666 Accuracy of the model EMA on 50000 test images: 81.9% Max EMA accuracy: 81.89% Epoch: [235] [ 0/312] eta: 0:48:39 lr: 0.002001 min_lr: 0.002001 loss: 2.2791 (2.2791) weight_decay: 0.0500 (0.0500) time: 9.3562 data: 8.2689 max mem: 64948 Epoch: [235] [ 10/312] eta: 0:07:44 lr: 0.002000 min_lr: 0.002000 loss: 2.0127 (2.0146) weight_decay: 0.0500 (0.0500) time: 1.5368 data: 0.7820 max mem: 64948 Epoch: [235] [ 20/312] eta: 0:05:32 lr: 0.002000 min_lr: 0.002000 loss: 2.0127 (2.0150) weight_decay: 0.0500 (0.0500) time: 0.7281 data: 0.0168 max mem: 64948 Epoch: [235] [ 30/312] eta: 0:04:40 lr: 0.001999 min_lr: 0.001999 loss: 2.2076 (2.0359) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0003 max mem: 64948 Epoch: [235] [ 40/312] eta: 0:04:10 lr: 0.001999 min_lr: 0.001999 loss: 2.1973 (2.0486) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [235] [ 50/312] eta: 0:03:49 lr: 0.001998 min_lr: 0.001998 loss: 2.1973 (2.0573) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [235] [ 60/312] eta: 0:03:33 lr: 0.001998 min_lr: 0.001998 loss: 2.0772 (2.0564) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [235] [ 70/312] eta: 0:03:19 lr: 0.001997 min_lr: 0.001997 loss: 2.0265 (2.0574) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [235] [ 80/312] eta: 0:03:07 lr: 0.001997 min_lr: 0.001997 loss: 2.0265 (2.0585) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [235] [ 90/312] eta: 0:02:57 lr: 0.001996 min_lr: 0.001996 loss: 2.0505 (2.0553) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [235] [100/312] eta: 0:02:46 lr: 0.001996 min_lr: 0.001996 loss: 2.1949 (2.0635) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [235] [110/312] eta: 0:02:37 lr: 0.001995 min_lr: 0.001995 loss: 2.1391 (2.0620) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [235] [120/312] eta: 0:02:28 lr: 0.001995 min_lr: 0.001995 loss: 2.0593 (2.0691) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [235] [130/312] eta: 0:02:19 lr: 0.001994 min_lr: 0.001994 loss: 2.0305 (2.0595) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [235] [140/312] eta: 0:02:10 lr: 0.001994 min_lr: 0.001994 loss: 2.0785 (2.0658) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [235] [150/312] eta: 0:02:02 lr: 0.001993 min_lr: 0.001993 loss: 2.0759 (2.0534) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [235] [160/312] eta: 0:01:54 lr: 0.001993 min_lr: 0.001993 loss: 2.0229 (2.0608) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [235] [170/312] eta: 0:01:46 lr: 0.001993 min_lr: 0.001993 loss: 2.1932 (2.0672) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [235] [180/312] eta: 0:01:38 lr: 0.001992 min_lr: 0.001992 loss: 2.1035 (2.0586) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [235] [190/312] eta: 0:01:30 lr: 0.001992 min_lr: 0.001992 loss: 2.0513 (2.0620) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [235] [200/312] eta: 0:01:22 lr: 0.001991 min_lr: 0.001991 loss: 2.1861 (2.0597) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [235] [210/312] eta: 0:01:15 lr: 0.001991 min_lr: 0.001991 loss: 2.0893 (2.0614) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [235] [220/312] eta: 0:01:07 lr: 0.001990 min_lr: 0.001990 loss: 2.0893 (2.0617) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [235] [230/312] eta: 0:01:00 lr: 0.001990 min_lr: 0.001990 loss: 1.9645 (2.0583) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [235] [240/312] eta: 0:00:52 lr: 0.001989 min_lr: 0.001989 loss: 2.0065 (2.0525) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [235] [250/312] eta: 0:00:45 lr: 0.001989 min_lr: 0.001989 loss: 2.0592 (2.0557) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [235] [260/312] eta: 0:00:37 lr: 0.001988 min_lr: 0.001988 loss: 2.1646 (2.0548) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [235] [270/312] eta: 0:00:30 lr: 0.001988 min_lr: 0.001988 loss: 2.1210 (2.0554) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [235] [280/312] eta: 0:00:23 lr: 0.001987 min_lr: 0.001987 loss: 1.9900 (2.0498) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0009 max mem: 64948 Epoch: [235] [290/312] eta: 0:00:15 lr: 0.001987 min_lr: 0.001987 loss: 2.1330 (2.0562) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0008 max mem: 64948 Epoch: [235] [300/312] eta: 0:00:08 lr: 0.001986 min_lr: 0.001986 loss: 2.1436 (2.0551) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [235] [310/312] eta: 0:00:01 lr: 0.001986 min_lr: 0.001986 loss: 2.1271 (2.0582) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [235] [311/312] eta: 0:00:00 lr: 0.001986 min_lr: 0.001986 loss: 2.1578 (2.0585) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [235] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.001986 min_lr: 0.001986 loss: 2.1578 (2.0675) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5911 (0.5911) acc1: 86.1979 (86.1979) acc5: 95.5729 (95.5729) time: 4.4243 data: 4.2076 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7641 (0.7863) acc1: 80.7292 (79.1360) acc5: 95.3125 (94.9440) time: 0.6429 data: 0.4676 max mem: 64948 Test: Total time: 0:00:05 (0.6526 s / it) * Acc@1 79.626 Acc@5 94.904 loss 0.778 Accuracy of the model on the 50000 test images: 79.6% Max accuracy: 79.63% Test: [0/9] eta: 0:00:41 loss: 0.5070 (0.5070) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.5724 data: 4.3634 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7138 (0.6911) acc1: 81.2500 (80.3520) acc5: 96.8750 (96.4800) time: 0.6593 data: 0.4849 max mem: 64948 Test: Total time: 0:00:06 (0.6671 s / it) * Acc@1 81.870 Acc@5 96.036 loss 0.665 Accuracy of the model EMA on 50000 test images: 81.9% Epoch: [236] [ 0/312] eta: 0:56:46 lr: 0.001986 min_lr: 0.001986 loss: 1.9377 (1.9377) weight_decay: 0.0500 (0.0500) time: 10.9172 data: 7.9180 max mem: 64948 Epoch: [236] [ 10/312] eta: 0:08:19 lr: 0.001985 min_lr: 0.001985 loss: 2.0406 (1.9743) weight_decay: 0.0500 (0.0500) time: 1.6530 data: 0.7202 max mem: 64948 Epoch: [236] [ 20/312] eta: 0:05:49 lr: 0.001985 min_lr: 0.001985 loss: 2.0520 (2.0083) weight_decay: 0.0500 (0.0500) time: 0.7101 data: 0.0004 max mem: 64948 Epoch: [236] [ 30/312] eta: 0:04:51 lr: 0.001984 min_lr: 0.001984 loss: 1.9457 (1.9464) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [236] [ 40/312] eta: 0:04:19 lr: 0.001984 min_lr: 0.001984 loss: 1.9623 (2.0018) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [236] [ 50/312] eta: 0:03:56 lr: 0.001984 min_lr: 0.001984 loss: 2.2007 (2.0033) weight_decay: 0.0500 (0.0500) time: 0.7008 data: 0.0004 max mem: 64948 Epoch: [236] [ 60/312] eta: 0:03:39 lr: 0.001983 min_lr: 0.001983 loss: 2.1398 (2.0120) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [236] [ 70/312] eta: 0:03:24 lr: 0.001983 min_lr: 0.001983 loss: 2.0688 (1.9919) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [236] [ 80/312] eta: 0:03:11 lr: 0.001982 min_lr: 0.001982 loss: 1.8896 (1.9861) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [236] [ 90/312] eta: 0:03:00 lr: 0.001982 min_lr: 0.001982 loss: 2.0053 (1.9826) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [236] [100/312] eta: 0:02:49 lr: 0.001981 min_lr: 0.001981 loss: 2.1204 (1.9989) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [236] [110/312] eta: 0:02:39 lr: 0.001981 min_lr: 0.001981 loss: 2.1925 (2.0065) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [236] [120/312] eta: 0:02:30 lr: 0.001980 min_lr: 0.001980 loss: 2.1371 (2.0058) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [236] [130/312] eta: 0:02:21 lr: 0.001980 min_lr: 0.001980 loss: 2.1844 (2.0205) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [236] [140/312] eta: 0:02:12 lr: 0.001979 min_lr: 0.001979 loss: 2.1016 (2.0183) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [236] [150/312] eta: 0:02:03 lr: 0.001979 min_lr: 0.001979 loss: 2.0012 (2.0223) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [236] [160/312] eta: 0:01:55 lr: 0.001978 min_lr: 0.001978 loss: 2.0816 (2.0214) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [236] [170/312] eta: 0:01:47 lr: 0.001978 min_lr: 0.001978 loss: 2.1497 (2.0315) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [236] [180/312] eta: 0:01:39 lr: 0.001977 min_lr: 0.001977 loss: 2.1497 (2.0289) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [236] [190/312] eta: 0:01:31 lr: 0.001977 min_lr: 0.001977 loss: 2.1009 (2.0334) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [236] [200/312] eta: 0:01:23 lr: 0.001977 min_lr: 0.001977 loss: 2.0871 (2.0294) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [236] [210/312] eta: 0:01:15 lr: 0.001976 min_lr: 0.001976 loss: 2.1738 (2.0382) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [236] [220/312] eta: 0:01:08 lr: 0.001976 min_lr: 0.001976 loss: 2.1738 (2.0347) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [236] [230/312] eta: 0:01:00 lr: 0.001975 min_lr: 0.001975 loss: 2.1229 (2.0393) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [236] [240/312] eta: 0:00:53 lr: 0.001975 min_lr: 0.001975 loss: 2.0726 (2.0336) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [236] [250/312] eta: 0:00:45 lr: 0.001974 min_lr: 0.001974 loss: 1.8795 (2.0315) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [236] [260/312] eta: 0:00:38 lr: 0.001974 min_lr: 0.001974 loss: 1.9761 (2.0287) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [236] [270/312] eta: 0:00:30 lr: 0.001973 min_lr: 0.001973 loss: 2.0197 (2.0304) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [236] [280/312] eta: 0:00:23 lr: 0.001973 min_lr: 0.001973 loss: 2.1624 (2.0336) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0010 max mem: 64948 Epoch: [236] [290/312] eta: 0:00:16 lr: 0.001972 min_lr: 0.001972 loss: 2.1267 (2.0326) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [236] [300/312] eta: 0:00:08 lr: 0.001972 min_lr: 0.001972 loss: 2.0020 (2.0307) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [236] [310/312] eta: 0:00:01 lr: 0.001971 min_lr: 0.001971 loss: 2.0608 (2.0329) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [236] [311/312] eta: 0:00:00 lr: 0.001971 min_lr: 0.001971 loss: 2.0020 (2.0311) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [236] Total time: 0:03:48 (0.7311 s / it) Averaged stats: lr: 0.001971 min_lr: 0.001971 loss: 2.0020 (2.0619) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6258 (0.6258) acc1: 83.8542 (83.8542) acc5: 96.0938 (96.0938) time: 4.6163 data: 4.3966 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7612 (0.8105) acc1: 80.4688 (78.2720) acc5: 94.7917 (94.7520) time: 0.6642 data: 0.4886 max mem: 64948 Test: Total time: 0:00:06 (0.6746 s / it) * Acc@1 79.366 Acc@5 94.736 loss 0.793 Accuracy of the model on the 50000 test images: 79.4% Max accuracy: 79.63% Test: [0/9] eta: 0:00:43 loss: 0.5066 (0.5066) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 4.8784 data: 4.6605 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7134 (0.6904) acc1: 81.5104 (80.3520) acc5: 96.8750 (96.4800) time: 0.6936 data: 0.5179 max mem: 64948 Test: Total time: 0:00:06 (0.7027 s / it) * Acc@1 81.914 Acc@5 96.048 loss 0.665 Accuracy of the model EMA on 50000 test images: 81.9% Max EMA accuracy: 81.91% Epoch: [237] [ 0/312] eta: 0:50:22 lr: 0.001971 min_lr: 0.001971 loss: 1.5946 (1.5946) weight_decay: 0.0500 (0.0500) time: 9.6864 data: 8.5822 max mem: 64948 Epoch: [237] [ 10/312] eta: 0:07:50 lr: 0.001971 min_lr: 0.001971 loss: 1.9737 (1.8621) weight_decay: 0.0500 (0.0500) time: 1.5589 data: 0.7806 max mem: 64948 Epoch: [237] [ 20/312] eta: 0:05:35 lr: 0.001970 min_lr: 0.001970 loss: 2.0470 (1.9555) weight_decay: 0.0500 (0.0500) time: 0.7230 data: 0.0003 max mem: 64948 Epoch: [237] [ 30/312] eta: 0:04:43 lr: 0.001970 min_lr: 0.001970 loss: 2.0522 (1.9668) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [237] [ 40/312] eta: 0:04:12 lr: 0.001969 min_lr: 0.001969 loss: 1.9960 (1.9733) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [237] [ 50/312] eta: 0:03:51 lr: 0.001969 min_lr: 0.001969 loss: 2.1610 (2.0179) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [237] [ 60/312] eta: 0:03:34 lr: 0.001968 min_lr: 0.001968 loss: 2.2129 (2.0245) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [237] [ 70/312] eta: 0:03:20 lr: 0.001968 min_lr: 0.001968 loss: 2.0682 (2.0195) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [237] [ 80/312] eta: 0:03:08 lr: 0.001968 min_lr: 0.001968 loss: 2.0839 (2.0403) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [237] [ 90/312] eta: 0:02:57 lr: 0.001967 min_lr: 0.001967 loss: 2.2100 (2.0476) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [237] [100/312] eta: 0:02:47 lr: 0.001967 min_lr: 0.001967 loss: 1.9766 (2.0365) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [237] [110/312] eta: 0:02:37 lr: 0.001966 min_lr: 0.001966 loss: 1.8841 (2.0365) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [237] [120/312] eta: 0:02:28 lr: 0.001966 min_lr: 0.001966 loss: 2.1288 (2.0409) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [237] [130/312] eta: 0:02:19 lr: 0.001965 min_lr: 0.001965 loss: 2.0913 (2.0385) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [237] [140/312] eta: 0:02:11 lr: 0.001965 min_lr: 0.001965 loss: 2.0866 (2.0414) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [237] [150/312] eta: 0:02:02 lr: 0.001964 min_lr: 0.001964 loss: 2.2161 (2.0456) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [237] [160/312] eta: 0:01:54 lr: 0.001964 min_lr: 0.001964 loss: 2.0509 (2.0420) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0004 max mem: 64948 Epoch: [237] [170/312] eta: 0:01:46 lr: 0.001963 min_lr: 0.001963 loss: 2.1929 (2.0521) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [237] [180/312] eta: 0:01:38 lr: 0.001963 min_lr: 0.001963 loss: 2.2181 (2.0666) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [237] [190/312] eta: 0:01:30 lr: 0.001962 min_lr: 0.001962 loss: 2.2181 (2.0659) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [237] [200/312] eta: 0:01:23 lr: 0.001962 min_lr: 0.001962 loss: 1.9969 (2.0621) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [237] [210/312] eta: 0:01:15 lr: 0.001961 min_lr: 0.001961 loss: 2.0235 (2.0603) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [237] [220/312] eta: 0:01:07 lr: 0.001961 min_lr: 0.001961 loss: 2.1601 (2.0638) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [237] [230/312] eta: 0:01:00 lr: 0.001961 min_lr: 0.001961 loss: 2.0528 (2.0520) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [237] [240/312] eta: 0:00:52 lr: 0.001960 min_lr: 0.001960 loss: 2.1005 (2.0599) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [237] [250/312] eta: 0:00:45 lr: 0.001960 min_lr: 0.001960 loss: 2.1026 (2.0587) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [237] [260/312] eta: 0:00:38 lr: 0.001959 min_lr: 0.001959 loss: 2.0396 (2.0583) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [237] [270/312] eta: 0:00:30 lr: 0.001959 min_lr: 0.001959 loss: 1.9482 (2.0510) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [237] [280/312] eta: 0:00:23 lr: 0.001958 min_lr: 0.001958 loss: 1.7455 (2.0455) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [237] [290/312] eta: 0:00:16 lr: 0.001958 min_lr: 0.001958 loss: 1.9116 (2.0448) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [237] [300/312] eta: 0:00:08 lr: 0.001957 min_lr: 0.001957 loss: 1.9116 (2.0411) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [237] [310/312] eta: 0:00:01 lr: 0.001957 min_lr: 0.001957 loss: 1.9841 (2.0449) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [237] [311/312] eta: 0:00:00 lr: 0.001957 min_lr: 0.001957 loss: 1.9841 (2.0452) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [237] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.001957 min_lr: 0.001957 loss: 1.9841 (2.0479) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5362 (0.5362) acc1: 85.1562 (85.1562) acc5: 98.4375 (98.4375) time: 4.4055 data: 4.1846 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8035 (0.8224) acc1: 79.1667 (78.4320) acc5: 95.0521 (94.7520) time: 0.6415 data: 0.4650 max mem: 64948 Test: Total time: 0:00:05 (0.6504 s / it) * Acc@1 79.562 Acc@5 94.860 loss 0.791 Accuracy of the model on the 50000 test images: 79.6% Max accuracy: 79.63% Test: [0/9] eta: 0:00:45 loss: 0.5063 (0.5063) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 5.0718 data: 4.8652 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7122 (0.6898) acc1: 81.2500 (80.3200) acc5: 96.8750 (96.4800) time: 0.7196 data: 0.5407 max mem: 64948 Test: Total time: 0:00:06 (0.7262 s / it) * Acc@1 81.920 Acc@5 96.058 loss 0.664 Accuracy of the model EMA on 50000 test images: 81.9% Max EMA accuracy: 81.92% Epoch: [238] [ 0/312] eta: 0:52:22 lr: 0.001957 min_lr: 0.001957 loss: 1.9954 (1.9954) weight_decay: 0.0500 (0.0500) time: 10.0708 data: 7.7467 max mem: 64948 Epoch: [238] [ 10/312] eta: 0:07:53 lr: 0.001956 min_lr: 0.001956 loss: 2.0507 (2.0531) weight_decay: 0.0500 (0.0500) time: 1.5692 data: 0.7047 max mem: 64948 Epoch: [238] [ 20/312] eta: 0:05:36 lr: 0.001956 min_lr: 0.001956 loss: 2.1071 (2.0573) weight_decay: 0.0500 (0.0500) time: 0.7056 data: 0.0004 max mem: 64948 Epoch: [238] [ 30/312] eta: 0:04:43 lr: 0.001955 min_lr: 0.001955 loss: 2.1071 (2.0378) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [238] [ 40/312] eta: 0:04:13 lr: 0.001955 min_lr: 0.001955 loss: 2.1688 (2.0681) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0003 max mem: 64948 Epoch: [238] [ 50/312] eta: 0:03:51 lr: 0.001954 min_lr: 0.001954 loss: 2.1899 (2.0759) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [238] [ 60/312] eta: 0:03:34 lr: 0.001954 min_lr: 0.001954 loss: 2.1665 (2.0689) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [238] [ 70/312] eta: 0:03:21 lr: 0.001953 min_lr: 0.001953 loss: 2.2317 (2.0926) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [238] [ 80/312] eta: 0:03:08 lr: 0.001953 min_lr: 0.001953 loss: 2.2375 (2.0987) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [238] [ 90/312] eta: 0:02:57 lr: 0.001952 min_lr: 0.001952 loss: 2.0840 (2.0887) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [238] [100/312] eta: 0:02:47 lr: 0.001952 min_lr: 0.001952 loss: 2.2106 (2.1059) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [238] [110/312] eta: 0:02:37 lr: 0.001952 min_lr: 0.001952 loss: 2.1510 (2.0993) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [238] [120/312] eta: 0:02:28 lr: 0.001951 min_lr: 0.001951 loss: 2.0384 (2.0924) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [238] [130/312] eta: 0:02:19 lr: 0.001951 min_lr: 0.001951 loss: 2.2258 (2.1015) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [238] [140/312] eta: 0:02:11 lr: 0.001950 min_lr: 0.001950 loss: 2.1737 (2.1076) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [238] [150/312] eta: 0:02:02 lr: 0.001950 min_lr: 0.001950 loss: 2.0868 (2.0968) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [238] [160/312] eta: 0:01:54 lr: 0.001949 min_lr: 0.001949 loss: 2.0145 (2.0872) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [238] [170/312] eta: 0:01:46 lr: 0.001949 min_lr: 0.001949 loss: 2.1110 (2.0882) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [238] [180/312] eta: 0:01:38 lr: 0.001948 min_lr: 0.001948 loss: 2.0828 (2.0793) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [238] [190/312] eta: 0:01:30 lr: 0.001948 min_lr: 0.001948 loss: 1.9530 (2.0714) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [238] [200/312] eta: 0:01:23 lr: 0.001947 min_lr: 0.001947 loss: 2.0162 (2.0726) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [238] [210/312] eta: 0:01:15 lr: 0.001947 min_lr: 0.001947 loss: 2.1958 (2.0774) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [238] [220/312] eta: 0:01:07 lr: 0.001946 min_lr: 0.001946 loss: 2.1958 (2.0844) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [238] [230/312] eta: 0:01:00 lr: 0.001946 min_lr: 0.001946 loss: 2.1817 (2.0842) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [238] [240/312] eta: 0:00:52 lr: 0.001945 min_lr: 0.001945 loss: 2.0529 (2.0821) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [238] [250/312] eta: 0:00:45 lr: 0.001945 min_lr: 0.001945 loss: 2.0529 (2.0757) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [238] [260/312] eta: 0:00:38 lr: 0.001945 min_lr: 0.001945 loss: 2.0281 (2.0746) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [238] [270/312] eta: 0:00:30 lr: 0.001944 min_lr: 0.001944 loss: 2.0295 (2.0756) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [238] [280/312] eta: 0:00:23 lr: 0.001944 min_lr: 0.001944 loss: 2.1467 (2.0762) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [238] [290/312] eta: 0:00:15 lr: 0.001943 min_lr: 0.001943 loss: 1.9253 (2.0689) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [238] [300/312] eta: 0:00:08 lr: 0.001943 min_lr: 0.001943 loss: 1.8300 (2.0636) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [238] [310/312] eta: 0:00:01 lr: 0.001942 min_lr: 0.001942 loss: 2.0897 (2.0650) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [238] [311/312] eta: 0:00:00 lr: 0.001942 min_lr: 0.001942 loss: 2.1537 (2.0654) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [238] Total time: 0:03:46 (0.7276 s / it) Averaged stats: lr: 0.001942 min_lr: 0.001942 loss: 2.1537 (2.0526) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6120 (0.6120) acc1: 84.6354 (84.6354) acc5: 96.0938 (96.0938) time: 4.6533 data: 4.4497 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8502 (0.8253) acc1: 80.4688 (78.5280) acc5: 95.0521 (94.8160) time: 0.6683 data: 0.4945 max mem: 64948 Test: Total time: 0:00:06 (0.6889 s / it) * Acc@1 79.356 Acc@5 94.662 loss 0.801 Accuracy of the model on the 50000 test images: 79.4% Max accuracy: 79.63% Test: [0/9] eta: 0:00:41 loss: 0.5054 (0.5054) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 4.6511 data: 4.4343 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7110 (0.6892) acc1: 81.5104 (80.3520) acc5: 96.8750 (96.4480) time: 0.6681 data: 0.4928 max mem: 64948 Test: Total time: 0:00:06 (0.6793 s / it) * Acc@1 81.942 Acc@5 96.054 loss 0.663 Accuracy of the model EMA on 50000 test images: 81.9% Max EMA accuracy: 81.94% Epoch: [239] [ 0/312] eta: 0:48:58 lr: 0.001942 min_lr: 0.001942 loss: 2.2642 (2.2642) weight_decay: 0.0500 (0.0500) time: 9.4190 data: 8.1197 max mem: 64948 Epoch: [239] [ 10/312] eta: 0:07:46 lr: 0.001942 min_lr: 0.001942 loss: 2.1543 (2.0300) weight_decay: 0.0500 (0.0500) time: 1.5462 data: 0.7386 max mem: 64948 Epoch: [239] [ 20/312] eta: 0:05:33 lr: 0.001941 min_lr: 0.001941 loss: 2.0479 (2.0422) weight_decay: 0.0500 (0.0500) time: 0.7274 data: 0.0004 max mem: 64948 Epoch: [239] [ 30/312] eta: 0:04:41 lr: 0.001941 min_lr: 0.001941 loss: 2.0479 (2.0457) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [239] [ 40/312] eta: 0:04:11 lr: 0.001940 min_lr: 0.001940 loss: 2.1899 (2.1086) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [239] [ 50/312] eta: 0:03:50 lr: 0.001940 min_lr: 0.001940 loss: 2.1035 (2.0622) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [239] [ 60/312] eta: 0:03:33 lr: 0.001939 min_lr: 0.001939 loss: 1.7882 (2.0189) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [239] [ 70/312] eta: 0:03:20 lr: 0.001939 min_lr: 0.001939 loss: 1.8863 (2.0539) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [239] [ 80/312] eta: 0:03:08 lr: 0.001938 min_lr: 0.001938 loss: 2.2328 (2.0579) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [239] [ 90/312] eta: 0:02:57 lr: 0.001938 min_lr: 0.001938 loss: 2.2328 (2.0706) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [239] [100/312] eta: 0:02:47 lr: 0.001937 min_lr: 0.001937 loss: 2.1549 (2.0694) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [239] [110/312] eta: 0:02:37 lr: 0.001937 min_lr: 0.001937 loss: 2.0866 (2.0747) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [239] [120/312] eta: 0:02:28 lr: 0.001936 min_lr: 0.001936 loss: 2.1030 (2.0813) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [239] [130/312] eta: 0:02:19 lr: 0.001936 min_lr: 0.001936 loss: 2.1973 (2.0801) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [239] [140/312] eta: 0:02:10 lr: 0.001936 min_lr: 0.001936 loss: 1.8839 (2.0719) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [239] [150/312] eta: 0:02:02 lr: 0.001935 min_lr: 0.001935 loss: 2.1197 (2.0787) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [239] [160/312] eta: 0:01:54 lr: 0.001935 min_lr: 0.001935 loss: 2.1982 (2.0852) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [239] [170/312] eta: 0:01:46 lr: 0.001934 min_lr: 0.001934 loss: 2.2345 (2.0960) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [239] [180/312] eta: 0:01:38 lr: 0.001934 min_lr: 0.001934 loss: 2.2463 (2.0991) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [239] [190/312] eta: 0:01:30 lr: 0.001933 min_lr: 0.001933 loss: 2.1834 (2.0966) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [239] [200/312] eta: 0:01:23 lr: 0.001933 min_lr: 0.001933 loss: 2.1234 (2.0910) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [239] [210/312] eta: 0:01:15 lr: 0.001932 min_lr: 0.001932 loss: 2.0605 (2.0865) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [239] [220/312] eta: 0:01:07 lr: 0.001932 min_lr: 0.001932 loss: 2.1083 (2.0873) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [239] [230/312] eta: 0:01:00 lr: 0.001931 min_lr: 0.001931 loss: 2.1109 (2.0891) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [239] [240/312] eta: 0:00:52 lr: 0.001931 min_lr: 0.001931 loss: 2.1687 (2.0943) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [239] [250/312] eta: 0:00:45 lr: 0.001930 min_lr: 0.001930 loss: 2.1502 (2.0895) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [239] [260/312] eta: 0:00:37 lr: 0.001930 min_lr: 0.001930 loss: 1.9354 (2.0805) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [239] [270/312] eta: 0:00:30 lr: 0.001929 min_lr: 0.001929 loss: 1.9540 (2.0776) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [239] [280/312] eta: 0:00:23 lr: 0.001929 min_lr: 0.001929 loss: 2.0668 (2.0788) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [239] [290/312] eta: 0:00:15 lr: 0.001929 min_lr: 0.001929 loss: 2.2459 (2.0838) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [239] [300/312] eta: 0:00:08 lr: 0.001928 min_lr: 0.001928 loss: 2.2064 (2.0840) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [239] [310/312] eta: 0:00:01 lr: 0.001928 min_lr: 0.001928 loss: 2.1084 (2.0809) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [239] [311/312] eta: 0:00:00 lr: 0.001928 min_lr: 0.001928 loss: 2.1084 (2.0809) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [239] Total time: 0:03:46 (0.7274 s / it) Averaged stats: lr: 0.001928 min_lr: 0.001928 loss: 2.1084 (2.0543) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6158 (0.6158) acc1: 84.8958 (84.8958) acc5: 95.5729 (95.5729) time: 4.6443 data: 4.4251 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8134 (0.8315) acc1: 78.3854 (78.2400) acc5: 94.2708 (93.9840) time: 0.6679 data: 0.4918 max mem: 64948 Test: Total time: 0:00:06 (0.6922 s / it) * Acc@1 79.114 Acc@5 94.598 loss 0.803 Accuracy of the model on the 50000 test images: 79.1% Max accuracy: 79.63% Test: [0/9] eta: 0:00:46 loss: 0.5047 (0.5047) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 5.1889 data: 4.9707 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7092 (0.6886) acc1: 81.2500 (80.3520) acc5: 96.8750 (96.4480) time: 0.7279 data: 0.5524 max mem: 64948 Test: Total time: 0:00:06 (0.7370 s / it) * Acc@1 81.956 Acc@5 96.076 loss 0.663 Accuracy of the model EMA on 50000 test images: 82.0% Max EMA accuracy: 81.96% Epoch: [240] [ 0/312] eta: 0:47:16 lr: 0.001927 min_lr: 0.001927 loss: 1.6945 (1.6945) weight_decay: 0.0500 (0.0500) time: 9.0928 data: 8.2900 max mem: 64948 Epoch: [240] [ 10/312] eta: 0:08:07 lr: 0.001927 min_lr: 0.001927 loss: 2.1546 (2.1202) weight_decay: 0.0500 (0.0500) time: 1.6139 data: 0.8499 max mem: 64948 Epoch: [240] [ 20/312] eta: 0:05:43 lr: 0.001927 min_lr: 0.001927 loss: 2.1729 (2.1033) weight_decay: 0.0500 (0.0500) time: 0.7814 data: 0.0531 max mem: 64948 Epoch: [240] [ 30/312] eta: 0:04:47 lr: 0.001926 min_lr: 0.001926 loss: 2.1525 (2.1068) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [240] [ 40/312] eta: 0:04:15 lr: 0.001926 min_lr: 0.001926 loss: 2.1126 (2.0941) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [240] [ 50/312] eta: 0:03:54 lr: 0.001925 min_lr: 0.001925 loss: 2.1179 (2.0870) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [240] [ 60/312] eta: 0:03:37 lr: 0.001925 min_lr: 0.001925 loss: 2.1348 (2.0888) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [240] [ 70/312] eta: 0:03:22 lr: 0.001924 min_lr: 0.001924 loss: 2.1143 (2.0836) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [240] [ 80/312] eta: 0:03:10 lr: 0.001924 min_lr: 0.001924 loss: 2.0488 (2.0724) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [240] [ 90/312] eta: 0:02:58 lr: 0.001923 min_lr: 0.001923 loss: 1.9700 (2.0605) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [240] [100/312] eta: 0:02:48 lr: 0.001923 min_lr: 0.001923 loss: 1.9654 (2.0533) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [240] [110/312] eta: 0:02:38 lr: 0.001922 min_lr: 0.001922 loss: 2.1378 (2.0535) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [240] [120/312] eta: 0:02:29 lr: 0.001922 min_lr: 0.001922 loss: 2.1378 (2.0521) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [240] [130/312] eta: 0:02:20 lr: 0.001921 min_lr: 0.001921 loss: 2.0106 (2.0427) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [240] [140/312] eta: 0:02:11 lr: 0.001921 min_lr: 0.001921 loss: 1.9075 (2.0355) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [240] [150/312] eta: 0:02:03 lr: 0.001920 min_lr: 0.001920 loss: 1.9890 (2.0413) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [240] [160/312] eta: 0:01:55 lr: 0.001920 min_lr: 0.001920 loss: 2.1996 (2.0425) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [240] [170/312] eta: 0:01:47 lr: 0.001920 min_lr: 0.001920 loss: 2.2123 (2.0489) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [240] [180/312] eta: 0:01:39 lr: 0.001919 min_lr: 0.001919 loss: 2.2123 (2.0516) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [240] [190/312] eta: 0:01:31 lr: 0.001919 min_lr: 0.001919 loss: 2.1187 (2.0494) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [240] [200/312] eta: 0:01:23 lr: 0.001918 min_lr: 0.001918 loss: 1.9216 (2.0407) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [240] [210/312] eta: 0:01:15 lr: 0.001918 min_lr: 0.001918 loss: 1.9461 (2.0474) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [240] [220/312] eta: 0:01:08 lr: 0.001917 min_lr: 0.001917 loss: 2.0511 (2.0448) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [240] [230/312] eta: 0:01:00 lr: 0.001917 min_lr: 0.001917 loss: 2.0813 (2.0514) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [240] [240/312] eta: 0:00:53 lr: 0.001916 min_lr: 0.001916 loss: 2.0926 (2.0512) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [240] [250/312] eta: 0:00:45 lr: 0.001916 min_lr: 0.001916 loss: 2.0856 (2.0518) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [240] [260/312] eta: 0:00:38 lr: 0.001915 min_lr: 0.001915 loss: 2.0139 (2.0508) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [240] [270/312] eta: 0:00:30 lr: 0.001915 min_lr: 0.001915 loss: 2.0213 (2.0463) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [240] [280/312] eta: 0:00:23 lr: 0.001914 min_lr: 0.001914 loss: 2.0213 (2.0445) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0011 max mem: 64948 Epoch: [240] [290/312] eta: 0:00:16 lr: 0.001914 min_lr: 0.001914 loss: 2.0827 (2.0474) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [240] [300/312] eta: 0:00:08 lr: 0.001913 min_lr: 0.001913 loss: 2.0889 (2.0481) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [240] [310/312] eta: 0:00:01 lr: 0.001913 min_lr: 0.001913 loss: 2.0889 (2.0493) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [240] [311/312] eta: 0:00:00 lr: 0.001913 min_lr: 0.001913 loss: 2.0940 (2.0496) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [240] Total time: 0:03:47 (0.7303 s / it) Averaged stats: lr: 0.001913 min_lr: 0.001913 loss: 2.0940 (2.0514) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.6110 (0.6110) acc1: 83.8542 (83.8542) acc5: 95.8333 (95.8333) time: 4.8154 data: 4.6111 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8423 (0.8336) acc1: 76.3021 (77.3760) acc5: 95.5729 (94.1760) time: 0.6863 data: 0.5124 max mem: 64948 Test: Total time: 0:00:06 (0.7076 s / it) * Acc@1 78.978 Acc@5 94.560 loss 0.803 Accuracy of the model on the 50000 test images: 79.0% Max accuracy: 79.63% Test: [0/9] eta: 0:00:45 loss: 0.5043 (0.5043) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 5.0352 data: 4.8173 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7076 (0.6879) acc1: 81.2500 (80.4160) acc5: 96.8750 (96.4480) time: 0.7109 data: 0.5354 max mem: 64948 Test: Total time: 0:00:06 (0.7198 s / it) * Acc@1 81.974 Acc@5 96.084 loss 0.662 Accuracy of the model EMA on 50000 test images: 82.0% Max EMA accuracy: 81.97% Epoch: [241] [ 0/312] eta: 0:54:24 lr: 0.001913 min_lr: 0.001913 loss: 1.4628 (1.4628) weight_decay: 0.0500 (0.0500) time: 10.4630 data: 9.6718 max mem: 64948 Epoch: [241] [ 10/312] eta: 0:08:00 lr: 0.001912 min_lr: 0.001912 loss: 2.2737 (2.1744) weight_decay: 0.0500 (0.0500) time: 1.5903 data: 0.8795 max mem: 64948 Epoch: [241] [ 20/312] eta: 0:05:39 lr: 0.001912 min_lr: 0.001912 loss: 2.2547 (2.0350) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0003 max mem: 64948 Epoch: [241] [ 30/312] eta: 0:04:45 lr: 0.001911 min_lr: 0.001911 loss: 1.9704 (2.0793) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [241] [ 40/312] eta: 0:04:14 lr: 0.001911 min_lr: 0.001911 loss: 2.3020 (2.1165) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [241] [ 50/312] eta: 0:03:53 lr: 0.001911 min_lr: 0.001911 loss: 2.1537 (2.0766) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [241] [ 60/312] eta: 0:03:36 lr: 0.001910 min_lr: 0.001910 loss: 2.1187 (2.0785) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [241] [ 70/312] eta: 0:03:21 lr: 0.001910 min_lr: 0.001910 loss: 1.8930 (2.0567) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [241] [ 80/312] eta: 0:03:09 lr: 0.001909 min_lr: 0.001909 loss: 1.8930 (2.0579) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [241] [ 90/312] eta: 0:02:58 lr: 0.001909 min_lr: 0.001909 loss: 2.0624 (2.0480) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [241] [100/312] eta: 0:02:48 lr: 0.001908 min_lr: 0.001908 loss: 2.1591 (2.0556) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [241] [110/312] eta: 0:02:38 lr: 0.001908 min_lr: 0.001908 loss: 2.2199 (2.0626) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [241] [120/312] eta: 0:02:29 lr: 0.001907 min_lr: 0.001907 loss: 2.1997 (2.0673) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [241] [130/312] eta: 0:02:20 lr: 0.001907 min_lr: 0.001907 loss: 2.2018 (2.0681) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [241] [140/312] eta: 0:02:11 lr: 0.001906 min_lr: 0.001906 loss: 2.1821 (2.0695) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [241] [150/312] eta: 0:02:03 lr: 0.001906 min_lr: 0.001906 loss: 2.2121 (2.0725) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [241] [160/312] eta: 0:01:54 lr: 0.001905 min_lr: 0.001905 loss: 2.2121 (2.0675) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [241] [170/312] eta: 0:01:46 lr: 0.001905 min_lr: 0.001905 loss: 2.1033 (2.0617) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [241] [180/312] eta: 0:01:38 lr: 0.001904 min_lr: 0.001904 loss: 2.1533 (2.0646) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [241] [190/312] eta: 0:01:31 lr: 0.001904 min_lr: 0.001904 loss: 2.1533 (2.0561) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [241] [200/312] eta: 0:01:23 lr: 0.001904 min_lr: 0.001904 loss: 1.9424 (2.0522) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [241] [210/312] eta: 0:01:15 lr: 0.001903 min_lr: 0.001903 loss: 2.1365 (2.0576) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [241] [220/312] eta: 0:01:08 lr: 0.001903 min_lr: 0.001903 loss: 2.1839 (2.0596) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [241] [230/312] eta: 0:01:00 lr: 0.001902 min_lr: 0.001902 loss: 2.1753 (2.0638) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [241] [240/312] eta: 0:00:53 lr: 0.001902 min_lr: 0.001902 loss: 2.1505 (2.0655) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [241] [250/312] eta: 0:00:45 lr: 0.001901 min_lr: 0.001901 loss: 2.0439 (2.0682) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [241] [260/312] eta: 0:00:38 lr: 0.001901 min_lr: 0.001901 loss: 2.1694 (2.0726) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [241] [270/312] eta: 0:00:30 lr: 0.001900 min_lr: 0.001900 loss: 2.2073 (2.0725) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [241] [280/312] eta: 0:00:23 lr: 0.001900 min_lr: 0.001900 loss: 2.1284 (2.0727) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0010 max mem: 64948 Epoch: [241] [290/312] eta: 0:00:16 lr: 0.001899 min_lr: 0.001899 loss: 1.9646 (2.0663) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [241] [300/312] eta: 0:00:08 lr: 0.001899 min_lr: 0.001899 loss: 1.9672 (2.0659) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [241] [310/312] eta: 0:00:01 lr: 0.001898 min_lr: 0.001898 loss: 2.0718 (2.0676) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [241] [311/312] eta: 0:00:00 lr: 0.001898 min_lr: 0.001898 loss: 2.0718 (2.0670) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [241] Total time: 0:03:47 (0.7292 s / it) Averaged stats: lr: 0.001898 min_lr: 0.001898 loss: 2.0718 (2.0558) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5842 (0.5842) acc1: 84.1146 (84.1146) acc5: 96.6146 (96.6146) time: 4.7737 data: 4.5676 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7756 (0.8087) acc1: 78.6458 (78.1760) acc5: 94.5312 (94.7200) time: 0.6819 data: 0.5076 max mem: 64948 Test: Total time: 0:00:06 (0.7071 s / it) * Acc@1 79.568 Acc@5 94.678 loss 0.785 Accuracy of the model on the 50000 test images: 79.6% Max accuracy: 79.63% Test: [0/9] eta: 0:00:43 loss: 0.5038 (0.5038) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 4.8375 data: 4.6153 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7059 (0.6873) acc1: 81.5104 (80.4480) acc5: 96.8750 (96.4480) time: 0.6888 data: 0.5129 max mem: 64948 Test: Total time: 0:00:06 (0.7016 s / it) * Acc@1 81.994 Acc@5 96.094 loss 0.662 Accuracy of the model EMA on 50000 test images: 82.0% Max EMA accuracy: 81.99% Epoch: [242] [ 0/312] eta: 0:50:37 lr: 0.001898 min_lr: 0.001898 loss: 1.6409 (1.6409) weight_decay: 0.0500 (0.0500) time: 9.7355 data: 8.9296 max mem: 64948 Epoch: [242] [ 10/312] eta: 0:07:44 lr: 0.001898 min_lr: 0.001898 loss: 2.1809 (2.1699) weight_decay: 0.0500 (0.0500) time: 1.5379 data: 0.8122 max mem: 64948 Epoch: [242] [ 20/312] eta: 0:05:31 lr: 0.001897 min_lr: 0.001897 loss: 2.1869 (2.1218) weight_decay: 0.0500 (0.0500) time: 0.7061 data: 0.0004 max mem: 64948 Epoch: [242] [ 30/312] eta: 0:04:40 lr: 0.001897 min_lr: 0.001897 loss: 2.1869 (2.0882) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [242] [ 40/312] eta: 0:04:10 lr: 0.001896 min_lr: 0.001896 loss: 2.1461 (2.0718) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [242] [ 50/312] eta: 0:03:49 lr: 0.001896 min_lr: 0.001896 loss: 2.1461 (2.0525) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [242] [ 60/312] eta: 0:03:33 lr: 0.001895 min_lr: 0.001895 loss: 2.1463 (2.0636) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [242] [ 70/312] eta: 0:03:19 lr: 0.001895 min_lr: 0.001895 loss: 2.0760 (2.0593) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [242] [ 80/312] eta: 0:03:08 lr: 0.001895 min_lr: 0.001895 loss: 2.0760 (2.0693) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [242] [ 90/312] eta: 0:02:57 lr: 0.001894 min_lr: 0.001894 loss: 2.1485 (2.0876) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [242] [100/312] eta: 0:02:47 lr: 0.001894 min_lr: 0.001894 loss: 2.2269 (2.0854) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [242] [110/312] eta: 0:02:37 lr: 0.001893 min_lr: 0.001893 loss: 2.1037 (2.0882) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [242] [120/312] eta: 0:02:28 lr: 0.001893 min_lr: 0.001893 loss: 2.0579 (2.0877) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [242] [130/312] eta: 0:02:19 lr: 0.001892 min_lr: 0.001892 loss: 2.1500 (2.0936) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [242] [140/312] eta: 0:02:10 lr: 0.001892 min_lr: 0.001892 loss: 2.1500 (2.0956) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [242] [150/312] eta: 0:02:02 lr: 0.001891 min_lr: 0.001891 loss: 2.0572 (2.0889) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [242] [160/312] eta: 0:01:54 lr: 0.001891 min_lr: 0.001891 loss: 2.1148 (2.0932) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [242] [170/312] eta: 0:01:46 lr: 0.001890 min_lr: 0.001890 loss: 2.1640 (2.0962) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [242] [180/312] eta: 0:01:38 lr: 0.001890 min_lr: 0.001890 loss: 2.2033 (2.0938) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [242] [190/312] eta: 0:01:30 lr: 0.001889 min_lr: 0.001889 loss: 2.0136 (2.0870) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [242] [200/312] eta: 0:01:23 lr: 0.001889 min_lr: 0.001889 loss: 2.0136 (2.0921) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [242] [210/312] eta: 0:01:15 lr: 0.001888 min_lr: 0.001888 loss: 2.1014 (2.0880) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [242] [220/312] eta: 0:01:07 lr: 0.001888 min_lr: 0.001888 loss: 2.0954 (2.0886) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [242] [230/312] eta: 0:01:00 lr: 0.001888 min_lr: 0.001888 loss: 2.1628 (2.0878) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [242] [240/312] eta: 0:00:52 lr: 0.001887 min_lr: 0.001887 loss: 2.2236 (2.0917) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [242] [250/312] eta: 0:00:45 lr: 0.001887 min_lr: 0.001887 loss: 2.1583 (2.0888) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [242] [260/312] eta: 0:00:37 lr: 0.001886 min_lr: 0.001886 loss: 2.0950 (2.0923) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [242] [270/312] eta: 0:00:30 lr: 0.001886 min_lr: 0.001886 loss: 2.0916 (2.0881) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [242] [280/312] eta: 0:00:23 lr: 0.001885 min_lr: 0.001885 loss: 2.0916 (2.0857) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0010 max mem: 64948 Epoch: [242] [290/312] eta: 0:00:15 lr: 0.001885 min_lr: 0.001885 loss: 2.1232 (2.0839) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [242] [300/312] eta: 0:00:08 lr: 0.001884 min_lr: 0.001884 loss: 2.1278 (2.0857) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [242] [310/312] eta: 0:00:01 lr: 0.001884 min_lr: 0.001884 loss: 2.1278 (2.0792) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [242] [311/312] eta: 0:00:00 lr: 0.001884 min_lr: 0.001884 loss: 1.9756 (2.0788) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [242] Total time: 0:03:46 (0.7274 s / it) Averaged stats: lr: 0.001884 min_lr: 0.001884 loss: 1.9756 (2.0431) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5412 (0.5412) acc1: 84.8958 (84.8958) acc5: 96.0938 (96.0938) time: 4.5010 data: 4.2809 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7872 (0.7938) acc1: 79.6875 (78.4000) acc5: 95.3125 (95.1360) time: 0.6517 data: 0.4757 max mem: 64948 Test: Total time: 0:00:06 (0.6765 s / it) * Acc@1 79.534 Acc@5 94.956 loss 0.779 Accuracy of the model on the 50000 test images: 79.5% Max accuracy: 79.63% Test: [0/9] eta: 0:00:44 loss: 0.5030 (0.5030) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 4.9887 data: 4.7765 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7039 (0.6866) acc1: 81.5104 (80.5440) acc5: 96.8750 (96.4160) time: 0.7056 data: 0.5308 max mem: 64948 Test: Total time: 0:00:06 (0.7143 s / it) * Acc@1 81.986 Acc@5 96.102 loss 0.661 Accuracy of the model EMA on 50000 test images: 82.0% Epoch: [243] [ 0/312] eta: 0:55:20 lr: 0.001884 min_lr: 0.001884 loss: 1.3277 (1.3277) weight_decay: 0.0500 (0.0500) time: 10.6431 data: 7.4726 max mem: 64948 Epoch: [243] [ 10/312] eta: 0:08:19 lr: 0.001883 min_lr: 0.001883 loss: 2.1178 (1.9531) weight_decay: 0.0500 (0.0500) time: 1.6530 data: 0.6797 max mem: 64948 Epoch: [243] [ 20/312] eta: 0:05:49 lr: 0.001883 min_lr: 0.001883 loss: 2.1229 (2.0250) weight_decay: 0.0500 (0.0500) time: 0.7242 data: 0.0004 max mem: 64948 Epoch: [243] [ 30/312] eta: 0:04:51 lr: 0.001882 min_lr: 0.001882 loss: 2.1229 (1.9943) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [243] [ 40/312] eta: 0:04:18 lr: 0.001882 min_lr: 0.001882 loss: 2.1755 (2.0313) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [243] [ 50/312] eta: 0:03:56 lr: 0.001881 min_lr: 0.001881 loss: 2.1517 (2.0300) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [243] [ 60/312] eta: 0:03:38 lr: 0.001881 min_lr: 0.001881 loss: 2.1075 (2.0358) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [243] [ 70/312] eta: 0:03:24 lr: 0.001880 min_lr: 0.001880 loss: 2.0895 (2.0189) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [243] [ 80/312] eta: 0:03:11 lr: 0.001880 min_lr: 0.001880 loss: 1.9729 (2.0159) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [243] [ 90/312] eta: 0:02:59 lr: 0.001879 min_lr: 0.001879 loss: 1.9836 (2.0193) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [243] [100/312] eta: 0:02:49 lr: 0.001879 min_lr: 0.001879 loss: 2.0314 (2.0222) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [243] [110/312] eta: 0:02:39 lr: 0.001879 min_lr: 0.001879 loss: 2.0314 (2.0143) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [243] [120/312] eta: 0:02:30 lr: 0.001878 min_lr: 0.001878 loss: 1.9249 (2.0119) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [243] [130/312] eta: 0:02:21 lr: 0.001878 min_lr: 0.001878 loss: 1.9989 (2.0144) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [243] [140/312] eta: 0:02:12 lr: 0.001877 min_lr: 0.001877 loss: 2.0955 (2.0131) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [243] [150/312] eta: 0:02:03 lr: 0.001877 min_lr: 0.001877 loss: 1.8695 (2.0030) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [243] [160/312] eta: 0:01:55 lr: 0.001876 min_lr: 0.001876 loss: 2.0364 (2.0062) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [243] [170/312] eta: 0:01:47 lr: 0.001876 min_lr: 0.001876 loss: 2.0953 (2.0058) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [243] [180/312] eta: 0:01:39 lr: 0.001875 min_lr: 0.001875 loss: 1.8899 (2.0004) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [243] [190/312] eta: 0:01:31 lr: 0.001875 min_lr: 0.001875 loss: 1.8750 (1.9971) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [243] [200/312] eta: 0:01:23 lr: 0.001874 min_lr: 0.001874 loss: 1.8616 (1.9950) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [243] [210/312] eta: 0:01:15 lr: 0.001874 min_lr: 0.001874 loss: 2.0605 (1.9994) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [243] [220/312] eta: 0:01:08 lr: 0.001873 min_lr: 0.001873 loss: 2.1667 (2.0041) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [243] [230/312] eta: 0:01:00 lr: 0.001873 min_lr: 0.001873 loss: 2.0923 (2.0009) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [243] [240/312] eta: 0:00:53 lr: 0.001872 min_lr: 0.001872 loss: 2.0373 (1.9987) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [243] [250/312] eta: 0:00:45 lr: 0.001872 min_lr: 0.001872 loss: 2.0113 (1.9975) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [243] [260/312] eta: 0:00:38 lr: 0.001872 min_lr: 0.001872 loss: 2.1132 (2.0047) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [243] [270/312] eta: 0:00:30 lr: 0.001871 min_lr: 0.001871 loss: 2.2330 (2.0113) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [243] [280/312] eta: 0:00:23 lr: 0.001871 min_lr: 0.001871 loss: 2.1389 (2.0121) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [243] [290/312] eta: 0:00:16 lr: 0.001870 min_lr: 0.001870 loss: 2.1051 (2.0139) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [243] [300/312] eta: 0:00:08 lr: 0.001870 min_lr: 0.001870 loss: 2.1245 (2.0202) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [243] [310/312] eta: 0:00:01 lr: 0.001869 min_lr: 0.001869 loss: 2.2664 (2.0251) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [243] [311/312] eta: 0:00:00 lr: 0.001869 min_lr: 0.001869 loss: 2.1997 (2.0253) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [243] Total time: 0:03:48 (0.7308 s / it) Averaged stats: lr: 0.001869 min_lr: 0.001869 loss: 2.1997 (2.0358) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6234 (0.6234) acc1: 85.4167 (85.4167) acc5: 96.3542 (96.3542) time: 4.6315 data: 4.4166 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8373 (0.7873) acc1: 78.3854 (78.7840) acc5: 95.5729 (94.9120) time: 0.6659 data: 0.4908 max mem: 64948 Test: Total time: 0:00:06 (0.6864 s / it) * Acc@1 79.400 Acc@5 94.836 loss 0.783 Accuracy of the model on the 50000 test images: 79.4% Max accuracy: 79.63% Test: [0/9] eta: 0:00:42 loss: 0.5023 (0.5023) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 4.7355 data: 4.5223 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7016 (0.6857) acc1: 81.7708 (80.6080) acc5: 96.8750 (96.4480) time: 0.6774 data: 0.5026 max mem: 64948 Test: Total time: 0:00:06 (0.6851 s / it) * Acc@1 82.020 Acc@5 96.114 loss 0.660 Accuracy of the model EMA on 50000 test images: 82.0% Max EMA accuracy: 82.02% Epoch: [244] [ 0/312] eta: 0:50:59 lr: 0.001869 min_lr: 0.001869 loss: 2.5167 (2.5167) weight_decay: 0.0500 (0.0500) time: 9.8053 data: 8.9974 max mem: 64948 Epoch: [244] [ 10/312] eta: 0:07:50 lr: 0.001869 min_lr: 0.001869 loss: 2.0432 (2.1002) weight_decay: 0.0500 (0.0500) time: 1.5591 data: 0.8303 max mem: 64948 Epoch: [244] [ 20/312] eta: 0:05:35 lr: 0.001868 min_lr: 0.001868 loss: 2.0238 (2.0969) weight_decay: 0.0500 (0.0500) time: 0.7148 data: 0.0069 max mem: 64948 Epoch: [244] [ 30/312] eta: 0:04:42 lr: 0.001868 min_lr: 0.001868 loss: 1.9924 (2.0134) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [244] [ 40/312] eta: 0:04:12 lr: 0.001867 min_lr: 0.001867 loss: 1.9497 (2.0047) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [244] [ 50/312] eta: 0:03:51 lr: 0.001867 min_lr: 0.001867 loss: 2.0115 (2.0156) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [244] [ 60/312] eta: 0:03:34 lr: 0.001866 min_lr: 0.001866 loss: 2.1128 (2.0198) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [244] [ 70/312] eta: 0:03:20 lr: 0.001866 min_lr: 0.001866 loss: 2.1031 (2.0212) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [244] [ 80/312] eta: 0:03:08 lr: 0.001865 min_lr: 0.001865 loss: 2.0967 (2.0008) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [244] [ 90/312] eta: 0:02:57 lr: 0.001865 min_lr: 0.001865 loss: 2.0386 (2.0079) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [244] [100/312] eta: 0:02:47 lr: 0.001864 min_lr: 0.001864 loss: 2.2782 (2.0295) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [244] [110/312] eta: 0:02:37 lr: 0.001864 min_lr: 0.001864 loss: 2.1817 (2.0327) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [244] [120/312] eta: 0:02:28 lr: 0.001864 min_lr: 0.001864 loss: 1.9930 (2.0287) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [244] [130/312] eta: 0:02:19 lr: 0.001863 min_lr: 0.001863 loss: 1.9822 (2.0339) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [244] [140/312] eta: 0:02:11 lr: 0.001863 min_lr: 0.001863 loss: 1.9968 (2.0343) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [244] [150/312] eta: 0:02:02 lr: 0.001862 min_lr: 0.001862 loss: 1.9659 (2.0412) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [244] [160/312] eta: 0:01:54 lr: 0.001862 min_lr: 0.001862 loss: 1.9659 (2.0387) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [244] [170/312] eta: 0:01:46 lr: 0.001861 min_lr: 0.001861 loss: 1.9643 (2.0328) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [244] [180/312] eta: 0:01:38 lr: 0.001861 min_lr: 0.001861 loss: 2.0327 (2.0313) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [244] [190/312] eta: 0:01:30 lr: 0.001860 min_lr: 0.001860 loss: 2.0327 (2.0265) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [244] [200/312] eta: 0:01:23 lr: 0.001860 min_lr: 0.001860 loss: 1.9545 (2.0264) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [244] [210/312] eta: 0:01:15 lr: 0.001859 min_lr: 0.001859 loss: 1.9545 (2.0260) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [244] [220/312] eta: 0:01:07 lr: 0.001859 min_lr: 0.001859 loss: 2.1211 (2.0288) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [244] [230/312] eta: 0:01:00 lr: 0.001858 min_lr: 0.001858 loss: 2.0274 (2.0276) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [244] [240/312] eta: 0:00:52 lr: 0.001858 min_lr: 0.001858 loss: 1.9258 (2.0223) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [244] [250/312] eta: 0:00:45 lr: 0.001857 min_lr: 0.001857 loss: 1.9665 (2.0224) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [244] [260/312] eta: 0:00:38 lr: 0.001857 min_lr: 0.001857 loss: 1.9665 (2.0172) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [244] [270/312] eta: 0:00:30 lr: 0.001857 min_lr: 0.001857 loss: 2.0633 (2.0226) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [244] [280/312] eta: 0:00:23 lr: 0.001856 min_lr: 0.001856 loss: 2.3155 (2.0317) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0009 max mem: 64948 Epoch: [244] [290/312] eta: 0:00:16 lr: 0.001856 min_lr: 0.001856 loss: 2.3039 (2.0305) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [244] [300/312] eta: 0:00:08 lr: 0.001855 min_lr: 0.001855 loss: 1.9967 (2.0285) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [244] [310/312] eta: 0:00:01 lr: 0.001855 min_lr: 0.001855 loss: 1.9967 (2.0300) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [244] [311/312] eta: 0:00:00 lr: 0.001855 min_lr: 0.001855 loss: 2.0593 (2.0304) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [244] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.001855 min_lr: 0.001855 loss: 2.0593 (2.0326) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6216 (0.6216) acc1: 85.1562 (85.1562) acc5: 96.0938 (96.0938) time: 4.6924 data: 4.4838 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8036 (0.8198) acc1: 79.1667 (78.4000) acc5: 95.5729 (94.8160) time: 0.6727 data: 0.4983 max mem: 64948 Test: Total time: 0:00:06 (0.6847 s / it) * Acc@1 79.604 Acc@5 94.988 loss 0.780 Accuracy of the model on the 50000 test images: 79.6% Max accuracy: 79.63% Test: [0/9] eta: 0:00:46 loss: 0.5023 (0.5023) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 5.1280 data: 4.9252 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6993 (0.6849) acc1: 81.7708 (80.5440) acc5: 96.8750 (96.5120) time: 0.7210 data: 0.5473 max mem: 64948 Test: Total time: 0:00:06 (0.7294 s / it) * Acc@1 82.044 Acc@5 96.104 loss 0.660 Accuracy of the model EMA on 50000 test images: 82.0% Max EMA accuracy: 82.04% Epoch: [245] [ 0/312] eta: 0:43:36 lr: 0.001855 min_lr: 0.001855 loss: 1.6762 (1.6762) weight_decay: 0.0500 (0.0500) time: 8.3863 data: 6.7690 max mem: 64948 Epoch: [245] [ 10/312] eta: 0:07:19 lr: 0.001854 min_lr: 0.001854 loss: 2.1357 (2.0922) weight_decay: 0.0500 (0.0500) time: 1.4550 data: 0.6158 max mem: 64948 Epoch: [245] [ 20/312] eta: 0:05:19 lr: 0.001854 min_lr: 0.001854 loss: 2.1252 (2.0323) weight_decay: 0.0500 (0.0500) time: 0.7278 data: 0.0004 max mem: 64948 Epoch: [245] [ 30/312] eta: 0:04:31 lr: 0.001853 min_lr: 0.001853 loss: 1.8973 (1.9866) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [245] [ 40/312] eta: 0:04:04 lr: 0.001853 min_lr: 0.001853 loss: 2.0054 (1.9889) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [245] [ 50/312] eta: 0:03:44 lr: 0.001852 min_lr: 0.001852 loss: 2.1463 (2.0022) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [245] [ 60/312] eta: 0:03:29 lr: 0.001852 min_lr: 0.001852 loss: 2.0410 (2.0158) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [245] [ 70/312] eta: 0:03:16 lr: 0.001851 min_lr: 0.001851 loss: 2.0241 (2.0178) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [245] [ 80/312] eta: 0:03:05 lr: 0.001851 min_lr: 0.001851 loss: 1.9565 (2.0171) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [245] [ 90/312] eta: 0:02:54 lr: 0.001850 min_lr: 0.001850 loss: 1.9565 (2.0032) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [245] [100/312] eta: 0:02:44 lr: 0.001850 min_lr: 0.001850 loss: 2.1146 (2.0310) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [245] [110/312] eta: 0:02:35 lr: 0.001849 min_lr: 0.001849 loss: 2.1146 (2.0290) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [245] [120/312] eta: 0:02:26 lr: 0.001849 min_lr: 0.001849 loss: 2.0698 (2.0261) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [245] [130/312] eta: 0:02:18 lr: 0.001848 min_lr: 0.001848 loss: 2.0884 (2.0211) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [245] [140/312] eta: 0:02:09 lr: 0.001848 min_lr: 0.001848 loss: 2.0234 (2.0259) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [245] [150/312] eta: 0:02:01 lr: 0.001848 min_lr: 0.001848 loss: 2.1386 (2.0361) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [245] [160/312] eta: 0:01:53 lr: 0.001847 min_lr: 0.001847 loss: 2.1816 (2.0366) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [245] [170/312] eta: 0:01:45 lr: 0.001847 min_lr: 0.001847 loss: 2.0195 (2.0342) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [245] [180/312] eta: 0:01:37 lr: 0.001846 min_lr: 0.001846 loss: 2.1578 (2.0409) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [245] [190/312] eta: 0:01:30 lr: 0.001846 min_lr: 0.001846 loss: 2.1762 (2.0432) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [245] [200/312] eta: 0:01:22 lr: 0.001845 min_lr: 0.001845 loss: 2.0909 (2.0439) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [245] [210/312] eta: 0:01:14 lr: 0.001845 min_lr: 0.001845 loss: 1.9208 (2.0340) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [245] [220/312] eta: 0:01:07 lr: 0.001844 min_lr: 0.001844 loss: 2.0233 (2.0346) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [245] [230/312] eta: 0:00:59 lr: 0.001844 min_lr: 0.001844 loss: 2.1159 (2.0338) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [245] [240/312] eta: 0:00:52 lr: 0.001843 min_lr: 0.001843 loss: 2.1487 (2.0413) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [245] [250/312] eta: 0:00:45 lr: 0.001843 min_lr: 0.001843 loss: 2.2262 (2.0463) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [245] [260/312] eta: 0:00:37 lr: 0.001842 min_lr: 0.001842 loss: 2.2262 (2.0462) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [245] [270/312] eta: 0:00:30 lr: 0.001842 min_lr: 0.001842 loss: 2.1580 (2.0439) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [245] [280/312] eta: 0:00:23 lr: 0.001841 min_lr: 0.001841 loss: 2.0690 (2.0442) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [245] [290/312] eta: 0:00:15 lr: 0.001841 min_lr: 0.001841 loss: 2.1065 (2.0457) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [245] [300/312] eta: 0:00:08 lr: 0.001841 min_lr: 0.001841 loss: 2.1488 (2.0420) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [245] [310/312] eta: 0:00:01 lr: 0.001840 min_lr: 0.001840 loss: 2.1164 (2.0429) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [245] [311/312] eta: 0:00:00 lr: 0.001840 min_lr: 0.001840 loss: 2.1164 (2.0426) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [245] Total time: 0:03:45 (0.7243 s / it) Averaged stats: lr: 0.001840 min_lr: 0.001840 loss: 2.1164 (2.0457) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6262 (0.6262) acc1: 84.8958 (84.8958) acc5: 95.8333 (95.8333) time: 4.7004 data: 4.4824 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8814 (0.8222) acc1: 78.1250 (78.3680) acc5: 95.5729 (94.8160) time: 0.6742 data: 0.4981 max mem: 64948 Test: Total time: 0:00:06 (0.6975 s / it) * Acc@1 79.426 Acc@5 94.782 loss 0.792 Accuracy of the model on the 50000 test images: 79.4% Max accuracy: 79.63% Test: [0/9] eta: 0:00:42 loss: 0.5023 (0.5023) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 4.7597 data: 4.5416 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6981 (0.6841) acc1: 81.7708 (80.5120) acc5: 96.8750 (96.4480) time: 0.6808 data: 0.5047 max mem: 64948 Test: Total time: 0:00:06 (0.6935 s / it) * Acc@1 82.070 Acc@5 96.102 loss 0.659 Accuracy of the model EMA on 50000 test images: 82.1% Max EMA accuracy: 82.07% Epoch: [246] [ 0/312] eta: 0:51:52 lr: 0.001840 min_lr: 0.001840 loss: 2.3133 (2.3133) weight_decay: 0.0500 (0.0500) time: 9.9760 data: 9.1678 max mem: 64948 Epoch: [246] [ 10/312] eta: 0:07:50 lr: 0.001840 min_lr: 0.001840 loss: 1.9287 (1.9691) weight_decay: 0.0500 (0.0500) time: 1.5577 data: 0.8338 max mem: 64948 Epoch: [246] [ 20/312] eta: 0:05:35 lr: 0.001839 min_lr: 0.001839 loss: 2.0565 (2.0551) weight_decay: 0.0500 (0.0500) time: 0.7084 data: 0.0005 max mem: 64948 Epoch: [246] [ 30/312] eta: 0:04:42 lr: 0.001839 min_lr: 0.001839 loss: 2.0973 (2.0350) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [246] [ 40/312] eta: 0:04:12 lr: 0.001838 min_lr: 0.001838 loss: 2.0973 (2.0189) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [246] [ 50/312] eta: 0:03:51 lr: 0.001838 min_lr: 0.001838 loss: 2.1686 (2.0439) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [246] [ 60/312] eta: 0:03:34 lr: 0.001837 min_lr: 0.001837 loss: 2.2028 (2.0626) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [246] [ 70/312] eta: 0:03:20 lr: 0.001837 min_lr: 0.001837 loss: 2.1104 (2.0738) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [246] [ 80/312] eta: 0:03:08 lr: 0.001836 min_lr: 0.001836 loss: 2.1432 (2.0737) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [246] [ 90/312] eta: 0:02:57 lr: 0.001836 min_lr: 0.001836 loss: 2.1432 (2.0794) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [246] [100/312] eta: 0:02:47 lr: 0.001835 min_lr: 0.001835 loss: 2.0526 (2.0650) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [246] [110/312] eta: 0:02:37 lr: 0.001835 min_lr: 0.001835 loss: 2.0519 (2.0768) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [246] [120/312] eta: 0:02:28 lr: 0.001834 min_lr: 0.001834 loss: 2.1956 (2.0818) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [246] [130/312] eta: 0:02:19 lr: 0.001834 min_lr: 0.001834 loss: 2.1335 (2.0794) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [246] [140/312] eta: 0:02:11 lr: 0.001833 min_lr: 0.001833 loss: 2.0723 (2.0828) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [246] [150/312] eta: 0:02:02 lr: 0.001833 min_lr: 0.001833 loss: 2.1201 (2.0763) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [246] [160/312] eta: 0:01:54 lr: 0.001833 min_lr: 0.001833 loss: 1.9082 (2.0708) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [246] [170/312] eta: 0:01:46 lr: 0.001832 min_lr: 0.001832 loss: 2.1211 (2.0752) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [246] [180/312] eta: 0:01:38 lr: 0.001832 min_lr: 0.001832 loss: 2.2206 (2.0736) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [246] [190/312] eta: 0:01:30 lr: 0.001831 min_lr: 0.001831 loss: 2.0092 (2.0738) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [246] [200/312] eta: 0:01:23 lr: 0.001831 min_lr: 0.001831 loss: 2.0117 (2.0713) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [246] [210/312] eta: 0:01:15 lr: 0.001830 min_lr: 0.001830 loss: 2.0117 (2.0678) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [246] [220/312] eta: 0:01:07 lr: 0.001830 min_lr: 0.001830 loss: 2.0357 (2.0654) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [246] [230/312] eta: 0:01:00 lr: 0.001829 min_lr: 0.001829 loss: 2.0972 (2.0579) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [246] [240/312] eta: 0:00:52 lr: 0.001829 min_lr: 0.001829 loss: 1.9355 (2.0567) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [246] [250/312] eta: 0:00:45 lr: 0.001828 min_lr: 0.001828 loss: 2.0087 (2.0559) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [246] [260/312] eta: 0:00:38 lr: 0.001828 min_lr: 0.001828 loss: 2.0795 (2.0562) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [246] [270/312] eta: 0:00:30 lr: 0.001827 min_lr: 0.001827 loss: 2.1369 (2.0556) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [246] [280/312] eta: 0:00:23 lr: 0.001827 min_lr: 0.001827 loss: 2.0975 (2.0572) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0010 max mem: 64948 Epoch: [246] [290/312] eta: 0:00:16 lr: 0.001826 min_lr: 0.001826 loss: 1.9593 (2.0536) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [246] [300/312] eta: 0:00:08 lr: 0.001826 min_lr: 0.001826 loss: 1.9455 (2.0560) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [246] [310/312] eta: 0:00:01 lr: 0.001826 min_lr: 0.001826 loss: 2.1601 (2.0540) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [246] [311/312] eta: 0:00:00 lr: 0.001825 min_lr: 0.001825 loss: 2.0007 (2.0538) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [246] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.001825 min_lr: 0.001825 loss: 2.0007 (2.0462) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6075 (0.6075) acc1: 86.4583 (86.4583) acc5: 96.8750 (96.8750) time: 4.6811 data: 4.4750 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8103 (0.7972) acc1: 79.1667 (78.8160) acc5: 95.0521 (95.0400) time: 0.6714 data: 0.4973 max mem: 64948 Test: Total time: 0:00:06 (0.6969 s / it) * Acc@1 79.870 Acc@5 95.072 loss 0.769 Accuracy of the model on the 50000 test images: 79.9% Max accuracy: 79.87% Test: [0/9] eta: 0:00:41 loss: 0.5020 (0.5020) acc1: 85.9375 (85.9375) acc5: 97.3958 (97.3958) time: 4.6607 data: 4.4430 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6968 (0.6835) acc1: 81.5104 (80.4800) acc5: 96.8750 (96.3840) time: 0.6692 data: 0.4938 max mem: 64948 Test: Total time: 0:00:06 (0.6777 s / it) * Acc@1 82.108 Acc@5 96.092 loss 0.658 Accuracy of the model EMA on 50000 test images: 82.1% Max EMA accuracy: 82.11% Epoch: [247] [ 0/312] eta: 0:53:03 lr: 0.001825 min_lr: 0.001825 loss: 2.2953 (2.2953) weight_decay: 0.0500 (0.0500) time: 10.2032 data: 9.3950 max mem: 64948 Epoch: [247] [ 10/312] eta: 0:07:55 lr: 0.001825 min_lr: 0.001825 loss: 2.2953 (2.1094) weight_decay: 0.0500 (0.0500) time: 1.5761 data: 0.8544 max mem: 64948 Epoch: [247] [ 20/312] eta: 0:05:37 lr: 0.001824 min_lr: 0.001824 loss: 2.1814 (2.1165) weight_decay: 0.0500 (0.0500) time: 0.7032 data: 0.0004 max mem: 64948 Epoch: [247] [ 30/312] eta: 0:04:44 lr: 0.001824 min_lr: 0.001824 loss: 2.1782 (2.1135) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [247] [ 40/312] eta: 0:04:13 lr: 0.001824 min_lr: 0.001824 loss: 2.0863 (2.0960) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [247] [ 50/312] eta: 0:03:52 lr: 0.001823 min_lr: 0.001823 loss: 2.0863 (2.0906) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [247] [ 60/312] eta: 0:03:35 lr: 0.001823 min_lr: 0.001823 loss: 2.1196 (2.0711) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [247] [ 70/312] eta: 0:03:21 lr: 0.001822 min_lr: 0.001822 loss: 1.9487 (2.0325) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [247] [ 80/312] eta: 0:03:09 lr: 0.001822 min_lr: 0.001822 loss: 2.1238 (2.0485) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [247] [ 90/312] eta: 0:02:58 lr: 0.001821 min_lr: 0.001821 loss: 2.1369 (2.0523) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [247] [100/312] eta: 0:02:47 lr: 0.001821 min_lr: 0.001821 loss: 2.1157 (2.0607) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [247] [110/312] eta: 0:02:38 lr: 0.001820 min_lr: 0.001820 loss: 2.1526 (2.0690) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [247] [120/312] eta: 0:02:28 lr: 0.001820 min_lr: 0.001820 loss: 2.1743 (2.0733) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [247] [130/312] eta: 0:02:19 lr: 0.001819 min_lr: 0.001819 loss: 2.1743 (2.0760) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [247] [140/312] eta: 0:02:11 lr: 0.001819 min_lr: 0.001819 loss: 2.1143 (2.0769) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [247] [150/312] eta: 0:02:02 lr: 0.001818 min_lr: 0.001818 loss: 2.2595 (2.0855) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [247] [160/312] eta: 0:01:54 lr: 0.001818 min_lr: 0.001818 loss: 2.1975 (2.0760) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [247] [170/312] eta: 0:01:46 lr: 0.001817 min_lr: 0.001817 loss: 2.1975 (2.0842) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [247] [180/312] eta: 0:01:38 lr: 0.001817 min_lr: 0.001817 loss: 2.2460 (2.0850) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [247] [190/312] eta: 0:01:30 lr: 0.001817 min_lr: 0.001817 loss: 2.0464 (2.0795) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [247] [200/312] eta: 0:01:23 lr: 0.001816 min_lr: 0.001816 loss: 2.0464 (2.0768) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [247] [210/312] eta: 0:01:15 lr: 0.001816 min_lr: 0.001816 loss: 2.0220 (2.0706) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [247] [220/312] eta: 0:01:07 lr: 0.001815 min_lr: 0.001815 loss: 2.0234 (2.0700) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [247] [230/312] eta: 0:01:00 lr: 0.001815 min_lr: 0.001815 loss: 2.1572 (2.0667) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [247] [240/312] eta: 0:00:52 lr: 0.001814 min_lr: 0.001814 loss: 2.0535 (2.0674) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [247] [250/312] eta: 0:00:45 lr: 0.001814 min_lr: 0.001814 loss: 2.0160 (2.0629) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [247] [260/312] eta: 0:00:38 lr: 0.001813 min_lr: 0.001813 loss: 2.1749 (2.0675) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [247] [270/312] eta: 0:00:30 lr: 0.001813 min_lr: 0.001813 loss: 2.1749 (2.0661) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [247] [280/312] eta: 0:00:23 lr: 0.001812 min_lr: 0.001812 loss: 2.1547 (2.0704) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [247] [290/312] eta: 0:00:16 lr: 0.001812 min_lr: 0.001812 loss: 2.2426 (2.0746) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [247] [300/312] eta: 0:00:08 lr: 0.001811 min_lr: 0.001811 loss: 2.1064 (2.0710) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [247] [310/312] eta: 0:00:01 lr: 0.001811 min_lr: 0.001811 loss: 2.0271 (2.0701) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [247] [311/312] eta: 0:00:00 lr: 0.001811 min_lr: 0.001811 loss: 2.0111 (2.0685) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [247] Total time: 0:03:47 (0.7287 s / it) Averaged stats: lr: 0.001811 min_lr: 0.001811 loss: 2.0111 (2.0306) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5983 (0.5983) acc1: 85.4167 (85.4167) acc5: 97.1354 (97.1354) time: 4.4045 data: 4.1930 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8313 (0.8034) acc1: 79.4271 (78.5280) acc5: 95.5729 (95.1040) time: 0.6407 data: 0.4660 max mem: 64948 Test: Total time: 0:00:05 (0.6655 s / it) * Acc@1 79.780 Acc@5 94.996 loss 0.773 Accuracy of the model on the 50000 test images: 79.8% Max accuracy: 79.87% Test: [0/9] eta: 0:00:45 loss: 0.5014 (0.5014) acc1: 86.1979 (86.1979) acc5: 97.3958 (97.3958) time: 5.0135 data: 4.7954 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6956 (0.6830) acc1: 81.7708 (80.6400) acc5: 96.8750 (96.3840) time: 0.7090 data: 0.5329 max mem: 64948 Test: Total time: 0:00:06 (0.7167 s / it) * Acc@1 82.126 Acc@5 96.098 loss 0.658 Accuracy of the model EMA on 50000 test images: 82.1% Max EMA accuracy: 82.13% Epoch: [248] [ 0/312] eta: 0:46:47 lr: 0.001811 min_lr: 0.001811 loss: 1.7493 (1.7493) weight_decay: 0.0500 (0.0500) time: 8.9977 data: 8.1926 max mem: 64948 Epoch: [248] [ 10/312] eta: 0:07:28 lr: 0.001810 min_lr: 0.001810 loss: 2.1003 (2.1333) weight_decay: 0.0500 (0.0500) time: 1.4850 data: 0.7480 max mem: 64948 Epoch: [248] [ 20/312] eta: 0:05:23 lr: 0.001810 min_lr: 0.001810 loss: 2.1066 (2.1356) weight_decay: 0.0500 (0.0500) time: 0.7124 data: 0.0019 max mem: 64948 Epoch: [248] [ 30/312] eta: 0:04:34 lr: 0.001809 min_lr: 0.001809 loss: 2.1549 (2.1319) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [248] [ 40/312] eta: 0:04:06 lr: 0.001809 min_lr: 0.001809 loss: 2.1584 (2.1382) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [248] [ 50/312] eta: 0:03:46 lr: 0.001809 min_lr: 0.001809 loss: 2.2327 (2.1500) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [248] [ 60/312] eta: 0:03:31 lr: 0.001808 min_lr: 0.001808 loss: 2.0168 (2.0972) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [248] [ 70/312] eta: 0:03:17 lr: 0.001808 min_lr: 0.001808 loss: 1.7815 (2.0675) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [248] [ 80/312] eta: 0:03:06 lr: 0.001807 min_lr: 0.001807 loss: 2.0241 (2.0575) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [248] [ 90/312] eta: 0:02:55 lr: 0.001807 min_lr: 0.001807 loss: 2.0241 (2.0502) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [248] [100/312] eta: 0:02:45 lr: 0.001806 min_lr: 0.001806 loss: 2.0035 (2.0641) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [248] [110/312] eta: 0:02:36 lr: 0.001806 min_lr: 0.001806 loss: 2.0707 (2.0596) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [248] [120/312] eta: 0:02:27 lr: 0.001805 min_lr: 0.001805 loss: 2.0118 (2.0516) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [248] [130/312] eta: 0:02:18 lr: 0.001805 min_lr: 0.001805 loss: 2.0901 (2.0559) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [248] [140/312] eta: 0:02:10 lr: 0.001804 min_lr: 0.001804 loss: 2.1098 (2.0522) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [248] [150/312] eta: 0:02:01 lr: 0.001804 min_lr: 0.001804 loss: 1.9930 (2.0519) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [248] [160/312] eta: 0:01:53 lr: 0.001803 min_lr: 0.001803 loss: 1.7272 (2.0324) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [248] [170/312] eta: 0:01:45 lr: 0.001803 min_lr: 0.001803 loss: 1.8447 (2.0376) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [248] [180/312] eta: 0:01:38 lr: 0.001802 min_lr: 0.001802 loss: 2.0409 (2.0366) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [248] [190/312] eta: 0:01:30 lr: 0.001802 min_lr: 0.001802 loss: 2.0409 (2.0377) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [248] [200/312] eta: 0:01:22 lr: 0.001802 min_lr: 0.001802 loss: 2.0029 (2.0338) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [248] [210/312] eta: 0:01:15 lr: 0.001801 min_lr: 0.001801 loss: 2.2771 (2.0391) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [248] [220/312] eta: 0:01:07 lr: 0.001801 min_lr: 0.001801 loss: 2.1372 (2.0364) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [248] [230/312] eta: 0:01:00 lr: 0.001800 min_lr: 0.001800 loss: 2.0721 (2.0391) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [248] [240/312] eta: 0:00:52 lr: 0.001800 min_lr: 0.001800 loss: 2.2842 (2.0443) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [248] [250/312] eta: 0:00:45 lr: 0.001799 min_lr: 0.001799 loss: 2.1591 (2.0379) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [248] [260/312] eta: 0:00:37 lr: 0.001799 min_lr: 0.001799 loss: 2.1218 (2.0375) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [248] [270/312] eta: 0:00:30 lr: 0.001798 min_lr: 0.001798 loss: 2.1449 (2.0336) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [248] [280/312] eta: 0:00:23 lr: 0.001798 min_lr: 0.001798 loss: 2.0331 (2.0287) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0010 max mem: 64948 Epoch: [248] [290/312] eta: 0:00:15 lr: 0.001797 min_lr: 0.001797 loss: 2.0360 (2.0311) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [248] [300/312] eta: 0:00:08 lr: 0.001797 min_lr: 0.001797 loss: 2.0460 (2.0311) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [248] [310/312] eta: 0:00:01 lr: 0.001796 min_lr: 0.001796 loss: 1.8371 (2.0278) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [248] [311/312] eta: 0:00:00 lr: 0.001796 min_lr: 0.001796 loss: 1.9658 (2.0283) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [248] Total time: 0:03:46 (0.7252 s / it) Averaged stats: lr: 0.001796 min_lr: 0.001796 loss: 1.9658 (2.0333) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5965 (0.5965) acc1: 86.7188 (86.7188) acc5: 96.0938 (96.0938) time: 4.5999 data: 4.3853 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8174 (0.8107) acc1: 79.1667 (79.0400) acc5: 94.7917 (94.8800) time: 0.6630 data: 0.4873 max mem: 64948 Test: Total time: 0:00:06 (0.6883 s / it) * Acc@1 79.620 Acc@5 94.982 loss 0.779 Accuracy of the model on the 50000 test images: 79.6% Max accuracy: 79.87% Test: [0/9] eta: 0:00:43 loss: 0.5006 (0.5006) acc1: 86.1979 (86.1979) acc5: 97.3958 (97.3958) time: 4.8225 data: 4.6083 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6938 (0.6823) acc1: 81.7708 (80.6400) acc5: 96.8750 (96.3520) time: 0.6873 data: 0.5123 max mem: 64948 Test: Total time: 0:00:06 (0.6954 s / it) * Acc@1 82.140 Acc@5 96.118 loss 0.657 Accuracy of the model EMA on 50000 test images: 82.1% Max EMA accuracy: 82.14% Epoch: [249] [ 0/312] eta: 0:51:27 lr: 0.001796 min_lr: 0.001796 loss: 1.3030 (1.3030) weight_decay: 0.0500 (0.0500) time: 9.8945 data: 7.1839 max mem: 64948 Epoch: [249] [ 10/312] eta: 0:07:48 lr: 0.001796 min_lr: 0.001796 loss: 1.6300 (1.7406) weight_decay: 0.0500 (0.0500) time: 1.5514 data: 0.6535 max mem: 64948 Epoch: [249] [ 20/312] eta: 0:05:33 lr: 0.001795 min_lr: 0.001795 loss: 1.7746 (1.8397) weight_decay: 0.0500 (0.0500) time: 0.7042 data: 0.0004 max mem: 64948 Epoch: [249] [ 30/312] eta: 0:04:41 lr: 0.001795 min_lr: 0.001795 loss: 1.8178 (1.8528) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [249] [ 40/312] eta: 0:04:11 lr: 0.001794 min_lr: 0.001794 loss: 1.9706 (1.8751) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [249] [ 50/312] eta: 0:03:50 lr: 0.001794 min_lr: 0.001794 loss: 1.9706 (1.8777) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [249] [ 60/312] eta: 0:03:34 lr: 0.001794 min_lr: 0.001794 loss: 1.8348 (1.8963) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [249] [ 70/312] eta: 0:03:20 lr: 0.001793 min_lr: 0.001793 loss: 2.0293 (1.9123) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [249] [ 80/312] eta: 0:03:08 lr: 0.001793 min_lr: 0.001793 loss: 2.1352 (1.9498) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [249] [ 90/312] eta: 0:02:57 lr: 0.001792 min_lr: 0.001792 loss: 2.2166 (1.9760) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [249] [100/312] eta: 0:02:47 lr: 0.001792 min_lr: 0.001792 loss: 2.1205 (1.9666) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [249] [110/312] eta: 0:02:37 lr: 0.001791 min_lr: 0.001791 loss: 2.0988 (1.9861) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [249] [120/312] eta: 0:02:28 lr: 0.001791 min_lr: 0.001791 loss: 2.0759 (1.9885) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [249] [130/312] eta: 0:02:19 lr: 0.001790 min_lr: 0.001790 loss: 1.9203 (1.9881) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [249] [140/312] eta: 0:02:11 lr: 0.001790 min_lr: 0.001790 loss: 1.8995 (1.9835) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [249] [150/312] eta: 0:02:02 lr: 0.001789 min_lr: 0.001789 loss: 1.9094 (1.9840) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [249] [160/312] eta: 0:01:54 lr: 0.001789 min_lr: 0.001789 loss: 1.8613 (1.9827) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [249] [170/312] eta: 0:01:46 lr: 0.001788 min_lr: 0.001788 loss: 2.0035 (1.9902) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [249] [180/312] eta: 0:01:38 lr: 0.001788 min_lr: 0.001788 loss: 2.1552 (2.0014) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [249] [190/312] eta: 0:01:30 lr: 0.001787 min_lr: 0.001787 loss: 2.0685 (2.0001) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [249] [200/312] eta: 0:01:23 lr: 0.001787 min_lr: 0.001787 loss: 2.0203 (2.0013) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0003 max mem: 64948 Epoch: [249] [210/312] eta: 0:01:15 lr: 0.001787 min_lr: 0.001787 loss: 1.9418 (1.9969) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [249] [220/312] eta: 0:01:07 lr: 0.001786 min_lr: 0.001786 loss: 2.0083 (1.9976) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [249] [230/312] eta: 0:01:00 lr: 0.001786 min_lr: 0.001786 loss: 2.2311 (2.0068) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [249] [240/312] eta: 0:00:52 lr: 0.001785 min_lr: 0.001785 loss: 2.2785 (2.0148) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [249] [250/312] eta: 0:00:45 lr: 0.001785 min_lr: 0.001785 loss: 2.2271 (2.0146) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [249] [260/312] eta: 0:00:38 lr: 0.001784 min_lr: 0.001784 loss: 2.1319 (2.0188) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [249] [270/312] eta: 0:00:30 lr: 0.001784 min_lr: 0.001784 loss: 2.1368 (2.0186) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [249] [280/312] eta: 0:00:23 lr: 0.001783 min_lr: 0.001783 loss: 2.0993 (2.0168) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [249] [290/312] eta: 0:00:15 lr: 0.001783 min_lr: 0.001783 loss: 2.0993 (2.0212) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [249] [300/312] eta: 0:00:08 lr: 0.001782 min_lr: 0.001782 loss: 2.0652 (2.0215) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [249] [310/312] eta: 0:00:01 lr: 0.001782 min_lr: 0.001782 loss: 2.0742 (2.0201) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [249] [311/312] eta: 0:00:00 lr: 0.001782 min_lr: 0.001782 loss: 2.0452 (2.0191) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [249] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.001782 min_lr: 0.001782 loss: 2.0452 (2.0326) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.5478 (0.5478) acc1: 85.6771 (85.6771) acc5: 96.8750 (96.8750) time: 4.7821 data: 4.5752 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8243 (0.8006) acc1: 80.7292 (78.8160) acc5: 94.7917 (94.7840) time: 0.6826 data: 0.5084 max mem: 64948 Test: Total time: 0:00:06 (0.7009 s / it) * Acc@1 79.642 Acc@5 94.854 loss 0.775 Accuracy of the model on the 50000 test images: 79.6% Max accuracy: 79.87% Test: [0/9] eta: 0:00:45 loss: 0.4999 (0.4999) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 5.0168 data: 4.8011 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6924 (0.6818) acc1: 82.0312 (80.6080) acc5: 97.1354 (96.4160) time: 0.7087 data: 0.5335 max mem: 64948 Test: Total time: 0:00:06 (0.7162 s / it) * Acc@1 82.152 Acc@5 96.120 loss 0.657 Accuracy of the model EMA on 50000 test images: 82.2% Max EMA accuracy: 82.15% Epoch: [250] [ 0/312] eta: 0:47:53 lr: 0.001782 min_lr: 0.001782 loss: 1.7528 (1.7528) weight_decay: 0.0500 (0.0500) time: 9.2093 data: 7.1293 max mem: 64948 Epoch: [250] [ 10/312] eta: 0:07:42 lr: 0.001781 min_lr: 0.001781 loss: 2.0072 (1.9520) weight_decay: 0.0500 (0.0500) time: 1.5318 data: 0.6485 max mem: 64948 Epoch: [250] [ 20/312] eta: 0:05:31 lr: 0.001781 min_lr: 0.001781 loss: 2.0283 (1.9692) weight_decay: 0.0500 (0.0500) time: 0.7301 data: 0.0004 max mem: 64948 Epoch: [250] [ 30/312] eta: 0:04:39 lr: 0.001780 min_lr: 0.001780 loss: 2.1913 (1.9665) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [250] [ 40/312] eta: 0:04:10 lr: 0.001780 min_lr: 0.001780 loss: 2.1913 (2.0253) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [250] [ 50/312] eta: 0:03:49 lr: 0.001779 min_lr: 0.001779 loss: 2.2001 (2.0256) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [250] [ 60/312] eta: 0:03:33 lr: 0.001779 min_lr: 0.001779 loss: 2.1967 (2.0540) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [250] [ 70/312] eta: 0:03:19 lr: 0.001779 min_lr: 0.001779 loss: 2.1295 (2.0354) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [250] [ 80/312] eta: 0:03:07 lr: 0.001778 min_lr: 0.001778 loss: 2.0147 (2.0238) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [250] [ 90/312] eta: 0:02:56 lr: 0.001778 min_lr: 0.001778 loss: 2.0988 (2.0287) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [250] [100/312] eta: 0:02:46 lr: 0.001777 min_lr: 0.001777 loss: 2.1361 (2.0426) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [250] [110/312] eta: 0:02:37 lr: 0.001777 min_lr: 0.001777 loss: 2.1214 (2.0450) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [250] [120/312] eta: 0:02:28 lr: 0.001776 min_lr: 0.001776 loss: 2.0716 (2.0372) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [250] [130/312] eta: 0:02:19 lr: 0.001776 min_lr: 0.001776 loss: 2.1636 (2.0435) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [250] [140/312] eta: 0:02:10 lr: 0.001775 min_lr: 0.001775 loss: 2.1714 (2.0487) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [250] [150/312] eta: 0:02:02 lr: 0.001775 min_lr: 0.001775 loss: 2.0258 (2.0420) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [250] [160/312] eta: 0:01:54 lr: 0.001774 min_lr: 0.001774 loss: 2.1084 (2.0553) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [250] [170/312] eta: 0:01:46 lr: 0.001774 min_lr: 0.001774 loss: 2.1084 (2.0561) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [250] [180/312] eta: 0:01:38 lr: 0.001773 min_lr: 0.001773 loss: 2.0547 (2.0585) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [250] [190/312] eta: 0:01:30 lr: 0.001773 min_lr: 0.001773 loss: 2.1267 (2.0607) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [250] [200/312] eta: 0:01:22 lr: 0.001773 min_lr: 0.001773 loss: 2.1377 (2.0560) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [250] [210/312] eta: 0:01:15 lr: 0.001772 min_lr: 0.001772 loss: 1.8365 (2.0462) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [250] [220/312] eta: 0:01:07 lr: 0.001772 min_lr: 0.001772 loss: 2.0392 (2.0527) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [250] [230/312] eta: 0:01:00 lr: 0.001771 min_lr: 0.001771 loss: 2.1757 (2.0540) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [250] [240/312] eta: 0:00:52 lr: 0.001771 min_lr: 0.001771 loss: 2.0685 (2.0516) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [250] [250/312] eta: 0:00:45 lr: 0.001770 min_lr: 0.001770 loss: 1.8066 (2.0411) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [250] [260/312] eta: 0:00:37 lr: 0.001770 min_lr: 0.001770 loss: 1.9334 (2.0423) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [250] [270/312] eta: 0:00:30 lr: 0.001769 min_lr: 0.001769 loss: 2.1685 (2.0403) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [250] [280/312] eta: 0:00:23 lr: 0.001769 min_lr: 0.001769 loss: 1.9624 (2.0343) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [250] [290/312] eta: 0:00:15 lr: 0.001768 min_lr: 0.001768 loss: 1.9114 (2.0345) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [250] [300/312] eta: 0:00:08 lr: 0.001768 min_lr: 0.001768 loss: 2.1844 (2.0399) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [250] [310/312] eta: 0:00:01 lr: 0.001767 min_lr: 0.001767 loss: 2.1844 (2.0368) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [250] [311/312] eta: 0:00:00 lr: 0.001767 min_lr: 0.001767 loss: 2.1988 (2.0373) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [250] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.001767 min_lr: 0.001767 loss: 2.1988 (2.0317) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5923 (0.5923) acc1: 85.9375 (85.9375) acc5: 96.8750 (96.8750) time: 4.5940 data: 4.3880 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7930 (0.7932) acc1: 78.3854 (78.9120) acc5: 95.5729 (95.0720) time: 0.6617 data: 0.4876 max mem: 64948 Test: Total time: 0:00:06 (0.6882 s / it) * Acc@1 79.816 Acc@5 95.014 loss 0.775 Accuracy of the model on the 50000 test images: 79.8% Max accuracy: 79.87% Test: [0/9] eta: 0:00:44 loss: 0.4993 (0.4993) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 4.9524 data: 4.7461 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6908 (0.6814) acc1: 82.0312 (80.7040) acc5: 97.1354 (96.4800) time: 0.7021 data: 0.5274 max mem: 64948 Test: Total time: 0:00:06 (0.7095 s / it) * Acc@1 82.158 Acc@5 96.142 loss 0.656 Accuracy of the model EMA on 50000 test images: 82.2% Max EMA accuracy: 82.16% Epoch: [251] [ 0/312] eta: 0:56:04 lr: 0.001767 min_lr: 0.001767 loss: 2.3979 (2.3979) weight_decay: 0.0500 (0.0500) time: 10.7824 data: 10.0127 max mem: 64948 Epoch: [251] [ 10/312] eta: 0:08:08 lr: 0.001767 min_lr: 0.001767 loss: 2.1034 (2.1258) weight_decay: 0.0500 (0.0500) time: 1.6168 data: 0.9106 max mem: 64948 Epoch: [251] [ 20/312] eta: 0:05:44 lr: 0.001766 min_lr: 0.001766 loss: 2.1268 (2.0961) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0004 max mem: 64948 Epoch: [251] [ 30/312] eta: 0:04:49 lr: 0.001766 min_lr: 0.001766 loss: 2.0578 (2.0357) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [251] [ 40/312] eta: 0:04:16 lr: 0.001765 min_lr: 0.001765 loss: 2.0046 (2.0064) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [251] [ 50/312] eta: 0:03:54 lr: 0.001765 min_lr: 0.001765 loss: 2.0046 (2.0067) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [251] [ 60/312] eta: 0:03:37 lr: 0.001765 min_lr: 0.001765 loss: 2.0858 (2.0298) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [251] [ 70/312] eta: 0:03:22 lr: 0.001764 min_lr: 0.001764 loss: 2.0950 (2.0392) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [251] [ 80/312] eta: 0:03:10 lr: 0.001764 min_lr: 0.001764 loss: 1.9972 (2.0263) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [251] [ 90/312] eta: 0:02:59 lr: 0.001763 min_lr: 0.001763 loss: 1.8445 (2.0100) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [251] [100/312] eta: 0:02:48 lr: 0.001763 min_lr: 0.001763 loss: 1.9696 (2.0113) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [251] [110/312] eta: 0:02:39 lr: 0.001762 min_lr: 0.001762 loss: 2.0437 (2.0121) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [251] [120/312] eta: 0:02:29 lr: 0.001762 min_lr: 0.001762 loss: 1.8015 (1.9856) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [251] [130/312] eta: 0:02:20 lr: 0.001761 min_lr: 0.001761 loss: 1.8015 (1.9851) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [251] [140/312] eta: 0:02:12 lr: 0.001761 min_lr: 0.001761 loss: 2.1174 (1.9963) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [251] [150/312] eta: 0:02:03 lr: 0.001760 min_lr: 0.001760 loss: 2.1075 (1.9944) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [251] [160/312] eta: 0:01:55 lr: 0.001760 min_lr: 0.001760 loss: 2.0668 (1.9994) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [251] [170/312] eta: 0:01:47 lr: 0.001759 min_lr: 0.001759 loss: 2.1123 (2.0007) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [251] [180/312] eta: 0:01:39 lr: 0.001759 min_lr: 0.001759 loss: 2.1123 (1.9989) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [251] [190/312] eta: 0:01:31 lr: 0.001758 min_lr: 0.001758 loss: 1.9094 (1.9958) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [251] [200/312] eta: 0:01:23 lr: 0.001758 min_lr: 0.001758 loss: 2.0670 (2.0027) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [251] [210/312] eta: 0:01:15 lr: 0.001758 min_lr: 0.001758 loss: 2.0959 (2.0004) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [251] [220/312] eta: 0:01:08 lr: 0.001757 min_lr: 0.001757 loss: 2.1104 (2.0070) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [251] [230/312] eta: 0:01:00 lr: 0.001757 min_lr: 0.001757 loss: 2.1104 (2.0055) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [251] [240/312] eta: 0:00:53 lr: 0.001756 min_lr: 0.001756 loss: 1.9886 (2.0061) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [251] [250/312] eta: 0:00:45 lr: 0.001756 min_lr: 0.001756 loss: 1.9478 (2.0008) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [251] [260/312] eta: 0:00:38 lr: 0.001755 min_lr: 0.001755 loss: 1.9478 (2.0006) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [251] [270/312] eta: 0:00:30 lr: 0.001755 min_lr: 0.001755 loss: 2.1284 (2.0075) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [251] [280/312] eta: 0:00:23 lr: 0.001754 min_lr: 0.001754 loss: 2.2384 (2.0147) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [251] [290/312] eta: 0:00:16 lr: 0.001754 min_lr: 0.001754 loss: 2.1339 (2.0163) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [251] [300/312] eta: 0:00:08 lr: 0.001753 min_lr: 0.001753 loss: 2.0932 (2.0146) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [251] [310/312] eta: 0:00:01 lr: 0.001753 min_lr: 0.001753 loss: 2.0338 (2.0149) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [251] [311/312] eta: 0:00:00 lr: 0.001753 min_lr: 0.001753 loss: 2.0218 (2.0133) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [251] Total time: 0:03:47 (0.7301 s / it) Averaged stats: lr: 0.001753 min_lr: 0.001753 loss: 2.0218 (2.0202) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5754 (0.5754) acc1: 85.1562 (85.1562) acc5: 97.3958 (97.3958) time: 4.5958 data: 4.3725 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8100 (0.7953) acc1: 78.9062 (78.3360) acc5: 95.0521 (95.2320) time: 0.6619 data: 0.4859 max mem: 64948 Test: Total time: 0:00:06 (0.6840 s / it) * Acc@1 79.564 Acc@5 94.962 loss 0.782 Accuracy of the model on the 50000 test images: 79.6% Max accuracy: 79.87% Test: [0/9] eta: 0:00:41 loss: 0.4986 (0.4986) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.6449 data: 4.4407 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6894 (0.6810) acc1: 81.7708 (80.6080) acc5: 97.1354 (96.4480) time: 0.6727 data: 0.4943 max mem: 64948 Test: Total time: 0:00:06 (0.6822 s / it) * Acc@1 82.172 Acc@5 96.138 loss 0.656 Accuracy of the model EMA on 50000 test images: 82.2% Max EMA accuracy: 82.17% Epoch: [252] [ 0/312] eta: 0:47:50 lr: 0.001753 min_lr: 0.001753 loss: 1.6767 (1.6767) weight_decay: 0.0500 (0.0500) time: 9.1993 data: 8.4168 max mem: 64948 Epoch: [252] [ 10/312] eta: 0:07:41 lr: 0.001752 min_lr: 0.001752 loss: 2.1397 (2.0604) weight_decay: 0.0500 (0.0500) time: 1.5280 data: 0.7656 max mem: 64948 Epoch: [252] [ 20/312] eta: 0:05:30 lr: 0.001752 min_lr: 0.001752 loss: 2.0810 (2.0093) weight_decay: 0.0500 (0.0500) time: 0.7268 data: 0.0004 max mem: 64948 Epoch: [252] [ 30/312] eta: 0:04:38 lr: 0.001751 min_lr: 0.001751 loss: 2.0747 (2.0620) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [252] [ 40/312] eta: 0:04:09 lr: 0.001751 min_lr: 0.001751 loss: 2.1977 (2.0835) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [252] [ 50/312] eta: 0:03:49 lr: 0.001750 min_lr: 0.001750 loss: 2.1133 (2.0653) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [252] [ 60/312] eta: 0:03:33 lr: 0.001750 min_lr: 0.001750 loss: 1.9234 (2.0699) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [252] [ 70/312] eta: 0:03:19 lr: 0.001750 min_lr: 0.001750 loss: 2.0320 (2.0595) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [252] [ 80/312] eta: 0:03:07 lr: 0.001749 min_lr: 0.001749 loss: 2.0720 (2.0714) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [252] [ 90/312] eta: 0:02:56 lr: 0.001749 min_lr: 0.001749 loss: 2.0720 (2.0587) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [252] [100/312] eta: 0:02:46 lr: 0.001748 min_lr: 0.001748 loss: 1.9792 (2.0562) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [252] [110/312] eta: 0:02:37 lr: 0.001748 min_lr: 0.001748 loss: 2.1413 (2.0568) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [252] [120/312] eta: 0:02:28 lr: 0.001747 min_lr: 0.001747 loss: 2.1413 (2.0572) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [252] [130/312] eta: 0:02:19 lr: 0.001747 min_lr: 0.001747 loss: 2.1321 (2.0580) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [252] [140/312] eta: 0:02:10 lr: 0.001746 min_lr: 0.001746 loss: 2.1483 (2.0592) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [252] [150/312] eta: 0:02:02 lr: 0.001746 min_lr: 0.001746 loss: 2.1273 (2.0497) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [252] [160/312] eta: 0:01:54 lr: 0.001745 min_lr: 0.001745 loss: 1.9399 (2.0439) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [252] [170/312] eta: 0:01:46 lr: 0.001745 min_lr: 0.001745 loss: 1.8919 (2.0349) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [252] [180/312] eta: 0:01:38 lr: 0.001744 min_lr: 0.001744 loss: 1.8030 (2.0231) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [252] [190/312] eta: 0:01:30 lr: 0.001744 min_lr: 0.001744 loss: 1.9894 (2.0265) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [252] [200/312] eta: 0:01:22 lr: 0.001744 min_lr: 0.001744 loss: 2.0559 (2.0225) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [252] [210/312] eta: 0:01:15 lr: 0.001743 min_lr: 0.001743 loss: 1.9321 (2.0200) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [252] [220/312] eta: 0:01:07 lr: 0.001743 min_lr: 0.001743 loss: 2.1112 (2.0215) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [252] [230/312] eta: 0:01:00 lr: 0.001742 min_lr: 0.001742 loss: 2.1566 (2.0227) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [252] [240/312] eta: 0:00:52 lr: 0.001742 min_lr: 0.001742 loss: 2.1566 (2.0281) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [252] [250/312] eta: 0:00:45 lr: 0.001741 min_lr: 0.001741 loss: 2.0731 (2.0253) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [252] [260/312] eta: 0:00:37 lr: 0.001741 min_lr: 0.001741 loss: 1.8877 (2.0229) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [252] [270/312] eta: 0:00:30 lr: 0.001740 min_lr: 0.001740 loss: 1.8613 (2.0162) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [252] [280/312] eta: 0:00:23 lr: 0.001740 min_lr: 0.001740 loss: 2.0133 (2.0225) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0009 max mem: 64948 Epoch: [252] [290/312] eta: 0:00:15 lr: 0.001739 min_lr: 0.001739 loss: 2.1679 (2.0231) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [252] [300/312] eta: 0:00:08 lr: 0.001739 min_lr: 0.001739 loss: 2.1706 (2.0294) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [252] [310/312] eta: 0:00:01 lr: 0.001738 min_lr: 0.001738 loss: 2.1190 (2.0253) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [252] [311/312] eta: 0:00:00 lr: 0.001738 min_lr: 0.001738 loss: 2.0913 (2.0247) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [252] Total time: 0:03:46 (0.7266 s / it) Averaged stats: lr: 0.001738 min_lr: 0.001738 loss: 2.0913 (2.0212) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6187 (0.6187) acc1: 84.6354 (84.6354) acc5: 95.8333 (95.8333) time: 4.5699 data: 4.3621 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7989 (0.8010) acc1: 79.9479 (78.4000) acc5: 94.7917 (94.6880) time: 0.6590 data: 0.4848 max mem: 64948 Test: Total time: 0:00:06 (0.6812 s / it) * Acc@1 79.618 Acc@5 94.884 loss 0.777 Accuracy of the model on the 50000 test images: 79.6% Max accuracy: 79.87% Test: [0/9] eta: 0:00:45 loss: 0.4982 (0.4982) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 5.0518 data: 4.8390 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6880 (0.6805) acc1: 81.7708 (80.6400) acc5: 97.1354 (96.4160) time: 0.7126 data: 0.5378 max mem: 64948 Test: Total time: 0:00:06 (0.7208 s / it) * Acc@1 82.162 Acc@5 96.152 loss 0.655 Accuracy of the model EMA on 50000 test images: 82.2% Epoch: [253] [ 0/312] eta: 0:54:58 lr: 0.001738 min_lr: 0.001738 loss: 2.2445 (2.2445) weight_decay: 0.0500 (0.0500) time: 10.5727 data: 9.0954 max mem: 64948 Epoch: [253] [ 10/312] eta: 0:08:10 lr: 0.001738 min_lr: 0.001738 loss: 1.9229 (1.9642) weight_decay: 0.0500 (0.0500) time: 1.6227 data: 0.8272 max mem: 64948 Epoch: [253] [ 20/312] eta: 0:05:45 lr: 0.001737 min_lr: 0.001737 loss: 2.0439 (2.0397) weight_decay: 0.0500 (0.0500) time: 0.7120 data: 0.0004 max mem: 64948 Epoch: [253] [ 30/312] eta: 0:04:48 lr: 0.001737 min_lr: 0.001737 loss: 2.0740 (1.9882) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [253] [ 40/312] eta: 0:04:16 lr: 0.001736 min_lr: 0.001736 loss: 1.7345 (1.9190) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [253] [ 50/312] eta: 0:03:54 lr: 0.001736 min_lr: 0.001736 loss: 1.7950 (1.9513) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [253] [ 60/312] eta: 0:03:37 lr: 0.001736 min_lr: 0.001736 loss: 2.0908 (1.9545) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [253] [ 70/312] eta: 0:03:23 lr: 0.001735 min_lr: 0.001735 loss: 2.0394 (1.9696) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [253] [ 80/312] eta: 0:03:10 lr: 0.001735 min_lr: 0.001735 loss: 2.0303 (1.9707) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [253] [ 90/312] eta: 0:02:59 lr: 0.001734 min_lr: 0.001734 loss: 2.0473 (1.9821) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [253] [100/312] eta: 0:02:48 lr: 0.001734 min_lr: 0.001734 loss: 2.0473 (1.9865) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [253] [110/312] eta: 0:02:39 lr: 0.001733 min_lr: 0.001733 loss: 2.1525 (2.0038) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [253] [120/312] eta: 0:02:29 lr: 0.001733 min_lr: 0.001733 loss: 2.0201 (1.9920) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [253] [130/312] eta: 0:02:20 lr: 0.001732 min_lr: 0.001732 loss: 1.9900 (2.0024) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [253] [140/312] eta: 0:02:12 lr: 0.001732 min_lr: 0.001732 loss: 2.1101 (2.0040) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [253] [150/312] eta: 0:02:03 lr: 0.001731 min_lr: 0.001731 loss: 2.0178 (1.9955) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [253] [160/312] eta: 0:01:55 lr: 0.001731 min_lr: 0.001731 loss: 1.8258 (1.9896) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [253] [170/312] eta: 0:01:47 lr: 0.001730 min_lr: 0.001730 loss: 1.7996 (1.9842) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [253] [180/312] eta: 0:01:39 lr: 0.001730 min_lr: 0.001730 loss: 1.7324 (1.9806) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [253] [190/312] eta: 0:01:31 lr: 0.001729 min_lr: 0.001729 loss: 2.0498 (1.9881) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [253] [200/312] eta: 0:01:23 lr: 0.001729 min_lr: 0.001729 loss: 2.1470 (1.9969) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [253] [210/312] eta: 0:01:15 lr: 0.001729 min_lr: 0.001729 loss: 2.1040 (1.9934) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [253] [220/312] eta: 0:01:08 lr: 0.001728 min_lr: 0.001728 loss: 2.0007 (1.9956) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [253] [230/312] eta: 0:01:00 lr: 0.001728 min_lr: 0.001728 loss: 2.1668 (2.0041) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [253] [240/312] eta: 0:00:53 lr: 0.001727 min_lr: 0.001727 loss: 2.1106 (2.0037) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [253] [250/312] eta: 0:00:45 lr: 0.001727 min_lr: 0.001727 loss: 2.0214 (2.0022) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [253] [260/312] eta: 0:00:38 lr: 0.001726 min_lr: 0.001726 loss: 2.0988 (2.0054) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [253] [270/312] eta: 0:00:30 lr: 0.001726 min_lr: 0.001726 loss: 2.1103 (2.0058) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [253] [280/312] eta: 0:00:23 lr: 0.001725 min_lr: 0.001725 loss: 2.0353 (2.0067) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [253] [290/312] eta: 0:00:16 lr: 0.001725 min_lr: 0.001725 loss: 2.1042 (2.0140) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [253] [300/312] eta: 0:00:08 lr: 0.001724 min_lr: 0.001724 loss: 2.2059 (2.0185) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [253] [310/312] eta: 0:00:01 lr: 0.001724 min_lr: 0.001724 loss: 2.3296 (2.0265) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [253] [311/312] eta: 0:00:00 lr: 0.001724 min_lr: 0.001724 loss: 2.3490 (2.0280) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [253] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.001724 min_lr: 0.001724 loss: 2.3490 (2.0243) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.6442 (0.6442) acc1: 83.5938 (83.5938) acc5: 95.8333 (95.8333) time: 4.7679 data: 4.5488 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8527 (0.8313) acc1: 79.1667 (77.7280) acc5: 95.0521 (94.7840) time: 0.6813 data: 0.5055 max mem: 64948 Test: Total time: 0:00:06 (0.7187 s / it) * Acc@1 79.430 Acc@5 94.664 loss 0.801 Accuracy of the model on the 50000 test images: 79.4% Max accuracy: 79.87% Test: [0/9] eta: 0:00:44 loss: 0.4970 (0.4970) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.8918 data: 4.6763 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6871 (0.6801) acc1: 81.7708 (80.6080) acc5: 97.1354 (96.4160) time: 0.6956 data: 0.5197 max mem: 64948 Test: Total time: 0:00:06 (0.7071 s / it) * Acc@1 82.200 Acc@5 96.166 loss 0.655 Accuracy of the model EMA on 50000 test images: 82.2% Max EMA accuracy: 82.20% Epoch: [254] [ 0/312] eta: 0:48:06 lr: 0.001724 min_lr: 0.001724 loss: 2.5597 (2.5597) weight_decay: 0.0500 (0.0500) time: 9.2502 data: 7.9088 max mem: 64948 Epoch: [254] [ 10/312] eta: 0:07:45 lr: 0.001723 min_lr: 0.001723 loss: 2.2134 (2.1571) weight_decay: 0.0500 (0.0500) time: 1.5413 data: 0.7314 max mem: 64948 Epoch: [254] [ 20/312] eta: 0:05:32 lr: 0.001723 min_lr: 0.001723 loss: 2.0763 (2.0623) weight_decay: 0.0500 (0.0500) time: 0.7322 data: 0.0070 max mem: 64948 Epoch: [254] [ 30/312] eta: 0:04:41 lr: 0.001722 min_lr: 0.001722 loss: 2.1115 (2.0598) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [254] [ 40/312] eta: 0:04:11 lr: 0.001722 min_lr: 0.001722 loss: 2.0895 (2.0391) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [254] [ 50/312] eta: 0:03:50 lr: 0.001722 min_lr: 0.001722 loss: 1.9460 (2.0038) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [254] [ 60/312] eta: 0:03:33 lr: 0.001721 min_lr: 0.001721 loss: 2.0057 (2.0197) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [254] [ 70/312] eta: 0:03:20 lr: 0.001721 min_lr: 0.001721 loss: 2.1611 (2.0256) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [254] [ 80/312] eta: 0:03:08 lr: 0.001720 min_lr: 0.001720 loss: 2.1245 (2.0399) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [254] [ 90/312] eta: 0:02:57 lr: 0.001720 min_lr: 0.001720 loss: 2.1097 (2.0339) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [254] [100/312] eta: 0:02:47 lr: 0.001719 min_lr: 0.001719 loss: 2.0348 (2.0327) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [254] [110/312] eta: 0:02:37 lr: 0.001719 min_lr: 0.001719 loss: 2.1558 (2.0438) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [254] [120/312] eta: 0:02:28 lr: 0.001718 min_lr: 0.001718 loss: 2.1636 (2.0557) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [254] [130/312] eta: 0:02:19 lr: 0.001718 min_lr: 0.001718 loss: 2.1636 (2.0607) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [254] [140/312] eta: 0:02:11 lr: 0.001717 min_lr: 0.001717 loss: 1.9912 (2.0446) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [254] [150/312] eta: 0:02:02 lr: 0.001717 min_lr: 0.001717 loss: 1.7860 (2.0405) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [254] [160/312] eta: 0:01:54 lr: 0.001716 min_lr: 0.001716 loss: 1.9425 (2.0299) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [254] [170/312] eta: 0:01:46 lr: 0.001716 min_lr: 0.001716 loss: 2.0413 (2.0382) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [254] [180/312] eta: 0:01:38 lr: 0.001715 min_lr: 0.001715 loss: 2.1385 (2.0432) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [254] [190/312] eta: 0:01:30 lr: 0.001715 min_lr: 0.001715 loss: 2.1385 (2.0425) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [254] [200/312] eta: 0:01:23 lr: 0.001715 min_lr: 0.001715 loss: 2.1790 (2.0502) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [254] [210/312] eta: 0:01:15 lr: 0.001714 min_lr: 0.001714 loss: 2.1753 (2.0573) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [254] [220/312] eta: 0:01:07 lr: 0.001714 min_lr: 0.001714 loss: 2.1372 (2.0553) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [254] [230/312] eta: 0:01:00 lr: 0.001713 min_lr: 0.001713 loss: 2.0836 (2.0552) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [254] [240/312] eta: 0:00:52 lr: 0.001713 min_lr: 0.001713 loss: 2.0938 (2.0571) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [254] [250/312] eta: 0:00:45 lr: 0.001712 min_lr: 0.001712 loss: 2.0938 (2.0530) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [254] [260/312] eta: 0:00:38 lr: 0.001712 min_lr: 0.001712 loss: 2.0386 (2.0518) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [254] [270/312] eta: 0:00:30 lr: 0.001711 min_lr: 0.001711 loss: 2.1610 (2.0569) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [254] [280/312] eta: 0:00:23 lr: 0.001711 min_lr: 0.001711 loss: 2.1256 (2.0531) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0009 max mem: 64948 Epoch: [254] [290/312] eta: 0:00:15 lr: 0.001710 min_lr: 0.001710 loss: 1.9023 (2.0466) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0008 max mem: 64948 Epoch: [254] [300/312] eta: 0:00:08 lr: 0.001710 min_lr: 0.001710 loss: 1.9092 (2.0431) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0001 max mem: 64948 Epoch: [254] [310/312] eta: 0:00:01 lr: 0.001709 min_lr: 0.001709 loss: 1.9092 (2.0377) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [254] [311/312] eta: 0:00:00 lr: 0.001709 min_lr: 0.001709 loss: 1.9292 (2.0396) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0001 max mem: 64948 Epoch: [254] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.001709 min_lr: 0.001709 loss: 1.9292 (2.0236) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.6297 (0.6297) acc1: 83.5938 (83.5938) acc5: 95.5729 (95.5729) time: 4.3562 data: 4.1360 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8183 (0.8075) acc1: 80.2083 (78.8480) acc5: 95.5729 (94.8800) time: 0.6359 data: 0.4596 max mem: 64948 Test: Total time: 0:00:05 (0.6546 s / it) * Acc@1 79.918 Acc@5 94.868 loss 0.774 Accuracy of the model on the 50000 test images: 79.9% Max accuracy: 79.92% Test: [0/9] eta: 0:00:41 loss: 0.4966 (0.4966) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.6148 data: 4.3968 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6854 (0.6795) acc1: 81.7708 (80.6720) acc5: 97.3958 (96.4160) time: 0.6648 data: 0.4886 max mem: 64948 Test: Total time: 0:00:06 (0.6724 s / it) * Acc@1 82.238 Acc@5 96.174 loss 0.654 Accuracy of the model EMA on 50000 test images: 82.2% Max EMA accuracy: 82.24% Epoch: [255] [ 0/312] eta: 0:48:56 lr: 0.001709 min_lr: 0.001709 loss: 2.4471 (2.4471) weight_decay: 0.0500 (0.0500) time: 9.4105 data: 8.6245 max mem: 64948 Epoch: [255] [ 10/312] eta: 0:07:49 lr: 0.001709 min_lr: 0.001709 loss: 2.1442 (2.1282) weight_decay: 0.0500 (0.0500) time: 1.5561 data: 0.8380 max mem: 64948 Epoch: [255] [ 20/312] eta: 0:05:34 lr: 0.001708 min_lr: 0.001708 loss: 2.0837 (2.0500) weight_decay: 0.0500 (0.0500) time: 0.7327 data: 0.0299 max mem: 64948 Epoch: [255] [ 30/312] eta: 0:04:42 lr: 0.001708 min_lr: 0.001708 loss: 1.9780 (2.0153) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [255] [ 40/312] eta: 0:04:12 lr: 0.001708 min_lr: 0.001708 loss: 1.8937 (1.9964) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [255] [ 50/312] eta: 0:03:50 lr: 0.001707 min_lr: 0.001707 loss: 1.9953 (1.9764) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [255] [ 60/312] eta: 0:03:34 lr: 0.001707 min_lr: 0.001707 loss: 2.0494 (1.9812) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [255] [ 70/312] eta: 0:03:20 lr: 0.001706 min_lr: 0.001706 loss: 2.0671 (2.0001) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [255] [ 80/312] eta: 0:03:08 lr: 0.001706 min_lr: 0.001706 loss: 1.9275 (1.9835) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [255] [ 90/312] eta: 0:02:57 lr: 0.001705 min_lr: 0.001705 loss: 1.8185 (1.9925) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0003 max mem: 64948 Epoch: [255] [100/312] eta: 0:02:47 lr: 0.001705 min_lr: 0.001705 loss: 2.1037 (1.9834) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [255] [110/312] eta: 0:02:37 lr: 0.001704 min_lr: 0.001704 loss: 2.1008 (1.9753) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [255] [120/312] eta: 0:02:28 lr: 0.001704 min_lr: 0.001704 loss: 1.9598 (1.9720) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [255] [130/312] eta: 0:02:19 lr: 0.001703 min_lr: 0.001703 loss: 2.0173 (1.9779) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [255] [140/312] eta: 0:02:11 lr: 0.001703 min_lr: 0.001703 loss: 1.9918 (1.9670) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [255] [150/312] eta: 0:02:02 lr: 0.001702 min_lr: 0.001702 loss: 1.8636 (1.9699) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [255] [160/312] eta: 0:01:54 lr: 0.001702 min_lr: 0.001702 loss: 1.9850 (1.9739) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [255] [170/312] eta: 0:01:46 lr: 0.001701 min_lr: 0.001701 loss: 1.9293 (1.9697) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [255] [180/312] eta: 0:01:38 lr: 0.001701 min_lr: 0.001701 loss: 2.0618 (1.9825) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [255] [190/312] eta: 0:01:30 lr: 0.001701 min_lr: 0.001701 loss: 2.0342 (1.9850) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [255] [200/312] eta: 0:01:23 lr: 0.001700 min_lr: 0.001700 loss: 1.9914 (1.9878) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [255] [210/312] eta: 0:01:15 lr: 0.001700 min_lr: 0.001700 loss: 1.8533 (1.9781) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [255] [220/312] eta: 0:01:07 lr: 0.001699 min_lr: 0.001699 loss: 1.9296 (1.9771) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [255] [230/312] eta: 0:01:00 lr: 0.001699 min_lr: 0.001699 loss: 1.9687 (1.9763) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [255] [240/312] eta: 0:00:52 lr: 0.001698 min_lr: 0.001698 loss: 2.0535 (1.9767) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [255] [250/312] eta: 0:00:45 lr: 0.001698 min_lr: 0.001698 loss: 2.2620 (1.9939) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [255] [260/312] eta: 0:00:38 lr: 0.001697 min_lr: 0.001697 loss: 2.2365 (1.9922) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [255] [270/312] eta: 0:00:30 lr: 0.001697 min_lr: 0.001697 loss: 2.0350 (1.9896) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [255] [280/312] eta: 0:00:23 lr: 0.001696 min_lr: 0.001696 loss: 1.9165 (1.9849) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0014 max mem: 64948 Epoch: [255] [290/312] eta: 0:00:16 lr: 0.001696 min_lr: 0.001696 loss: 1.9806 (1.9892) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0013 max mem: 64948 Epoch: [255] [300/312] eta: 0:00:08 lr: 0.001695 min_lr: 0.001695 loss: 2.1238 (1.9919) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [255] [310/312] eta: 0:00:01 lr: 0.001695 min_lr: 0.001695 loss: 2.1133 (1.9901) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [255] [311/312] eta: 0:00:00 lr: 0.001695 min_lr: 0.001695 loss: 2.1238 (1.9907) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [255] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.001695 min_lr: 0.001695 loss: 2.1238 (2.0089) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.5757 (0.5757) acc1: 86.7188 (86.7188) acc5: 96.6146 (96.6146) time: 4.3204 data: 4.1003 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8259 (0.7782) acc1: 78.6458 (79.5200) acc5: 95.3125 (94.8480) time: 0.6320 data: 0.4557 max mem: 64948 Test: Total time: 0:00:05 (0.6546 s / it) * Acc@1 79.836 Acc@5 94.888 loss 0.773 Accuracy of the model on the 50000 test images: 79.8% Max accuracy: 79.92% Test: [0/9] eta: 0:00:43 loss: 0.4960 (0.4960) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.8053 data: 4.5929 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6839 (0.6790) acc1: 81.7708 (80.6720) acc5: 97.3958 (96.4480) time: 0.6852 data: 0.5104 max mem: 64948 Test: Total time: 0:00:06 (0.6952 s / it) * Acc@1 82.252 Acc@5 96.188 loss 0.654 Accuracy of the model EMA on 50000 test images: 82.3% Max EMA accuracy: 82.25% Epoch: [256] [ 0/312] eta: 0:51:55 lr: 0.001695 min_lr: 0.001695 loss: 2.0226 (2.0226) weight_decay: 0.0500 (0.0500) time: 9.9859 data: 9.2151 max mem: 64948 Epoch: [256] [ 10/312] eta: 0:07:51 lr: 0.001694 min_lr: 0.001694 loss: 2.0097 (1.9325) weight_decay: 0.0500 (0.0500) time: 1.5623 data: 0.8381 max mem: 64948 Epoch: [256] [ 20/312] eta: 0:05:35 lr: 0.001694 min_lr: 0.001694 loss: 1.9821 (1.9504) weight_decay: 0.0500 (0.0500) time: 0.7087 data: 0.0004 max mem: 64948 Epoch: [256] [ 30/312] eta: 0:04:42 lr: 0.001694 min_lr: 0.001694 loss: 1.9663 (1.9592) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [256] [ 40/312] eta: 0:04:12 lr: 0.001693 min_lr: 0.001693 loss: 1.9542 (1.9424) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [256] [ 50/312] eta: 0:03:51 lr: 0.001693 min_lr: 0.001693 loss: 1.9693 (1.9586) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [256] [ 60/312] eta: 0:03:34 lr: 0.001692 min_lr: 0.001692 loss: 2.1401 (1.9749) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [256] [ 70/312] eta: 0:03:20 lr: 0.001692 min_lr: 0.001692 loss: 2.1401 (1.9811) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [256] [ 80/312] eta: 0:03:08 lr: 0.001691 min_lr: 0.001691 loss: 2.1364 (2.0022) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [256] [ 90/312] eta: 0:02:57 lr: 0.001691 min_lr: 0.001691 loss: 2.1094 (2.0027) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [256] [100/312] eta: 0:02:47 lr: 0.001690 min_lr: 0.001690 loss: 2.1628 (2.0180) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [256] [110/312] eta: 0:02:37 lr: 0.001690 min_lr: 0.001690 loss: 2.1668 (2.0221) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [256] [120/312] eta: 0:02:28 lr: 0.001689 min_lr: 0.001689 loss: 2.0091 (2.0139) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [256] [130/312] eta: 0:02:19 lr: 0.001689 min_lr: 0.001689 loss: 2.0091 (2.0191) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [256] [140/312] eta: 0:02:11 lr: 0.001688 min_lr: 0.001688 loss: 2.0241 (2.0161) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [256] [150/312] eta: 0:02:02 lr: 0.001688 min_lr: 0.001688 loss: 2.0981 (2.0240) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [256] [160/312] eta: 0:01:54 lr: 0.001688 min_lr: 0.001688 loss: 1.9912 (2.0161) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [256] [170/312] eta: 0:01:46 lr: 0.001687 min_lr: 0.001687 loss: 1.8573 (2.0121) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [256] [180/312] eta: 0:01:38 lr: 0.001687 min_lr: 0.001687 loss: 2.0666 (2.0170) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [256] [190/312] eta: 0:01:30 lr: 0.001686 min_lr: 0.001686 loss: 1.9277 (2.0063) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [256] [200/312] eta: 0:01:23 lr: 0.001686 min_lr: 0.001686 loss: 1.9681 (2.0066) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [256] [210/312] eta: 0:01:15 lr: 0.001685 min_lr: 0.001685 loss: 2.0863 (2.0049) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [256] [220/312] eta: 0:01:07 lr: 0.001685 min_lr: 0.001685 loss: 2.0863 (2.0116) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [256] [230/312] eta: 0:01:00 lr: 0.001684 min_lr: 0.001684 loss: 2.0057 (1.9980) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [256] [240/312] eta: 0:00:52 lr: 0.001684 min_lr: 0.001684 loss: 1.7831 (2.0070) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [256] [250/312] eta: 0:00:45 lr: 0.001683 min_lr: 0.001683 loss: 2.1116 (2.0061) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [256] [260/312] eta: 0:00:38 lr: 0.001683 min_lr: 0.001683 loss: 1.9812 (2.0082) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [256] [270/312] eta: 0:00:30 lr: 0.001682 min_lr: 0.001682 loss: 2.0017 (2.0055) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [256] [280/312] eta: 0:00:23 lr: 0.001682 min_lr: 0.001682 loss: 2.0460 (2.0029) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [256] [290/312] eta: 0:00:16 lr: 0.001682 min_lr: 0.001682 loss: 2.0831 (2.0079) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [256] [300/312] eta: 0:00:08 lr: 0.001681 min_lr: 0.001681 loss: 2.0910 (2.0028) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [256] [310/312] eta: 0:00:01 lr: 0.001681 min_lr: 0.001681 loss: 1.7377 (1.9944) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [256] [311/312] eta: 0:00:00 lr: 0.001681 min_lr: 0.001681 loss: 1.8043 (1.9956) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [256] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.001681 min_lr: 0.001681 loss: 1.8043 (2.0189) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5919 (0.5919) acc1: 83.5938 (83.5938) acc5: 96.0938 (96.0938) time: 4.3651 data: 4.1536 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7502 (0.7792) acc1: 80.4688 (78.8160) acc5: 95.8333 (94.7200) time: 0.6363 data: 0.4616 max mem: 64948 Test: Total time: 0:00:05 (0.6551 s / it) * Acc@1 80.220 Acc@5 95.170 loss 0.749 Accuracy of the model on the 50000 test images: 80.2% Max accuracy: 80.22% Test: [0/9] eta: 0:00:39 loss: 0.4954 (0.4954) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.3948 data: 4.1881 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6831 (0.6786) acc1: 81.7708 (80.5440) acc5: 97.3958 (96.4480) time: 0.6404 data: 0.4655 max mem: 64948 Test: Total time: 0:00:05 (0.6500 s / it) * Acc@1 82.254 Acc@5 96.188 loss 0.653 Accuracy of the model EMA on 50000 test images: 82.3% Max EMA accuracy: 82.25% Epoch: [257] [ 0/312] eta: 0:48:45 lr: 0.001680 min_lr: 0.001680 loss: 2.4095 (2.4095) weight_decay: 0.0500 (0.0500) time: 9.3762 data: 8.5828 max mem: 64948 Epoch: [257] [ 10/312] eta: 0:07:34 lr: 0.001680 min_lr: 0.001680 loss: 1.9247 (1.9597) weight_decay: 0.0500 (0.0500) time: 1.5045 data: 0.7806 max mem: 64948 Epoch: [257] [ 20/312] eta: 0:05:26 lr: 0.001680 min_lr: 0.001680 loss: 1.9247 (1.9994) weight_decay: 0.0500 (0.0500) time: 0.7057 data: 0.0004 max mem: 64948 Epoch: [257] [ 30/312] eta: 0:04:36 lr: 0.001679 min_lr: 0.001679 loss: 2.0666 (2.0022) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [257] [ 40/312] eta: 0:04:08 lr: 0.001679 min_lr: 0.001679 loss: 2.1294 (2.0238) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [257] [ 50/312] eta: 0:03:48 lr: 0.001678 min_lr: 0.001678 loss: 2.1295 (2.0282) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [257] [ 60/312] eta: 0:03:32 lr: 0.001678 min_lr: 0.001678 loss: 2.1295 (2.0229) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [257] [ 70/312] eta: 0:03:18 lr: 0.001677 min_lr: 0.001677 loss: 2.1639 (2.0388) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [257] [ 80/312] eta: 0:03:06 lr: 0.001677 min_lr: 0.001677 loss: 2.1261 (2.0560) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [257] [ 90/312] eta: 0:02:56 lr: 0.001676 min_lr: 0.001676 loss: 2.0048 (2.0441) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [257] [100/312] eta: 0:02:46 lr: 0.001676 min_lr: 0.001676 loss: 2.0376 (2.0424) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [257] [110/312] eta: 0:02:36 lr: 0.001675 min_lr: 0.001675 loss: 2.0376 (2.0320) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [257] [120/312] eta: 0:02:27 lr: 0.001675 min_lr: 0.001675 loss: 2.0616 (2.0424) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [257] [130/312] eta: 0:02:19 lr: 0.001674 min_lr: 0.001674 loss: 2.0939 (2.0330) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [257] [140/312] eta: 0:02:10 lr: 0.001674 min_lr: 0.001674 loss: 2.0760 (2.0385) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [257] [150/312] eta: 0:02:02 lr: 0.001674 min_lr: 0.001674 loss: 2.1624 (2.0540) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [257] [160/312] eta: 0:01:54 lr: 0.001673 min_lr: 0.001673 loss: 2.1924 (2.0505) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [257] [170/312] eta: 0:01:46 lr: 0.001673 min_lr: 0.001673 loss: 2.0831 (2.0419) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [257] [180/312] eta: 0:01:38 lr: 0.001672 min_lr: 0.001672 loss: 1.9228 (2.0318) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [257] [190/312] eta: 0:01:30 lr: 0.001672 min_lr: 0.001672 loss: 1.8759 (2.0288) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [257] [200/312] eta: 0:01:22 lr: 0.001671 min_lr: 0.001671 loss: 1.8759 (2.0294) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [257] [210/312] eta: 0:01:15 lr: 0.001671 min_lr: 0.001671 loss: 2.1253 (2.0371) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [257] [220/312] eta: 0:01:07 lr: 0.001670 min_lr: 0.001670 loss: 2.0674 (2.0305) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [257] [230/312] eta: 0:01:00 lr: 0.001670 min_lr: 0.001670 loss: 2.0177 (2.0335) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [257] [240/312] eta: 0:00:52 lr: 0.001669 min_lr: 0.001669 loss: 2.0499 (2.0298) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [257] [250/312] eta: 0:00:45 lr: 0.001669 min_lr: 0.001669 loss: 2.0000 (2.0272) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [257] [260/312] eta: 0:00:37 lr: 0.001668 min_lr: 0.001668 loss: 2.0491 (2.0247) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [257] [270/312] eta: 0:00:30 lr: 0.001668 min_lr: 0.001668 loss: 2.0890 (2.0214) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [257] [280/312] eta: 0:00:23 lr: 0.001668 min_lr: 0.001668 loss: 2.0346 (2.0184) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [257] [290/312] eta: 0:00:15 lr: 0.001667 min_lr: 0.001667 loss: 2.0688 (2.0204) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [257] [300/312] eta: 0:00:08 lr: 0.001667 min_lr: 0.001667 loss: 1.9800 (2.0154) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [257] [310/312] eta: 0:00:01 lr: 0.001666 min_lr: 0.001666 loss: 1.9795 (2.0149) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [257] [311/312] eta: 0:00:00 lr: 0.001666 min_lr: 0.001666 loss: 1.9921 (2.0152) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [257] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.001666 min_lr: 0.001666 loss: 1.9921 (2.0115) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.5839 (0.5839) acc1: 85.4167 (85.4167) acc5: 96.3542 (96.3542) time: 4.2936 data: 4.0848 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7827 (0.7902) acc1: 80.2083 (79.6160) acc5: 94.0104 (94.2400) time: 0.6284 data: 0.4540 max mem: 64948 Test: Total time: 0:00:05 (0.6488 s / it) * Acc@1 79.676 Acc@5 94.914 loss 0.776 Accuracy of the model on the 50000 test images: 79.7% Max accuracy: 80.22% Test: [0/9] eta: 0:00:45 loss: 0.4948 (0.4948) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 5.0300 data: 4.8122 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6822 (0.6782) acc1: 81.7708 (80.5440) acc5: 97.3958 (96.4480) time: 0.7107 data: 0.5348 max mem: 64948 Test: Total time: 0:00:06 (0.7185 s / it) * Acc@1 82.294 Acc@5 96.196 loss 0.653 Accuracy of the model EMA on 50000 test images: 82.3% Max EMA accuracy: 82.29% Epoch: [258] [ 0/312] eta: 0:49:15 lr: 0.001666 min_lr: 0.001666 loss: 2.1848 (2.1848) weight_decay: 0.0500 (0.0500) time: 9.4741 data: 8.4242 max mem: 64948 Epoch: [258] [ 10/312] eta: 0:07:40 lr: 0.001666 min_lr: 0.001666 loss: 2.1809 (1.9727) weight_decay: 0.0500 (0.0500) time: 1.5237 data: 0.7662 max mem: 64948 Epoch: [258] [ 20/312] eta: 0:05:29 lr: 0.001665 min_lr: 0.001665 loss: 2.1391 (2.0607) weight_decay: 0.0500 (0.0500) time: 0.7127 data: 0.0004 max mem: 64948 Epoch: [258] [ 30/312] eta: 0:04:38 lr: 0.001665 min_lr: 0.001665 loss: 2.1391 (2.0586) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [258] [ 40/312] eta: 0:04:09 lr: 0.001664 min_lr: 0.001664 loss: 2.0894 (2.0576) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [258] [ 50/312] eta: 0:03:48 lr: 0.001664 min_lr: 0.001664 loss: 1.9286 (2.0220) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [258] [ 60/312] eta: 0:03:32 lr: 0.001663 min_lr: 0.001663 loss: 1.8399 (2.0150) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [258] [ 70/312] eta: 0:03:19 lr: 0.001663 min_lr: 0.001663 loss: 2.0905 (2.0318) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [258] [ 80/312] eta: 0:03:07 lr: 0.001662 min_lr: 0.001662 loss: 2.0668 (2.0247) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [258] [ 90/312] eta: 0:02:56 lr: 0.001662 min_lr: 0.001662 loss: 2.1085 (2.0401) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0003 max mem: 64948 Epoch: [258] [100/312] eta: 0:02:46 lr: 0.001661 min_lr: 0.001661 loss: 2.1086 (2.0325) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [258] [110/312] eta: 0:02:37 lr: 0.001661 min_lr: 0.001661 loss: 2.0089 (2.0354) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [258] [120/312] eta: 0:02:27 lr: 0.001661 min_lr: 0.001661 loss: 2.1661 (2.0430) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [258] [130/312] eta: 0:02:19 lr: 0.001660 min_lr: 0.001660 loss: 2.0281 (2.0355) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [258] [140/312] eta: 0:02:10 lr: 0.001660 min_lr: 0.001660 loss: 1.9573 (2.0241) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [258] [150/312] eta: 0:02:02 lr: 0.001659 min_lr: 0.001659 loss: 2.0301 (2.0237) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [258] [160/312] eta: 0:01:54 lr: 0.001659 min_lr: 0.001659 loss: 2.2875 (2.0384) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [258] [170/312] eta: 0:01:46 lr: 0.001658 min_lr: 0.001658 loss: 2.2584 (2.0282) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [258] [180/312] eta: 0:01:38 lr: 0.001658 min_lr: 0.001658 loss: 1.8522 (2.0267) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [258] [190/312] eta: 0:01:30 lr: 0.001657 min_lr: 0.001657 loss: 1.9930 (2.0260) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [258] [200/312] eta: 0:01:22 lr: 0.001657 min_lr: 0.001657 loss: 2.0578 (2.0270) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [258] [210/312] eta: 0:01:15 lr: 0.001656 min_lr: 0.001656 loss: 1.9024 (2.0174) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [258] [220/312] eta: 0:01:07 lr: 0.001656 min_lr: 0.001656 loss: 1.8605 (2.0142) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [258] [230/312] eta: 0:01:00 lr: 0.001655 min_lr: 0.001655 loss: 1.9959 (2.0142) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [258] [240/312] eta: 0:00:52 lr: 0.001655 min_lr: 0.001655 loss: 1.9477 (2.0144) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [258] [250/312] eta: 0:00:45 lr: 0.001655 min_lr: 0.001655 loss: 1.9916 (2.0197) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [258] [260/312] eta: 0:00:37 lr: 0.001654 min_lr: 0.001654 loss: 2.1552 (2.0196) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [258] [270/312] eta: 0:00:30 lr: 0.001654 min_lr: 0.001654 loss: 1.9630 (2.0137) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [258] [280/312] eta: 0:00:23 lr: 0.001653 min_lr: 0.001653 loss: 1.8499 (2.0126) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [258] [290/312] eta: 0:00:15 lr: 0.001653 min_lr: 0.001653 loss: 2.0842 (2.0120) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0008 max mem: 64948 Epoch: [258] [300/312] eta: 0:00:08 lr: 0.001652 min_lr: 0.001652 loss: 2.0842 (2.0112) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [258] [310/312] eta: 0:00:01 lr: 0.001652 min_lr: 0.001652 loss: 2.0319 (2.0129) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [258] [311/312] eta: 0:00:00 lr: 0.001652 min_lr: 0.001652 loss: 2.1916 (2.0139) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [258] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.001652 min_lr: 0.001652 loss: 2.1916 (2.0153) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.6043 (0.6043) acc1: 84.6354 (84.6354) acc5: 96.6146 (96.6146) time: 4.4366 data: 4.2204 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8087 (0.7770) acc1: 78.6458 (78.6880) acc5: 96.0938 (94.9120) time: 0.6442 data: 0.4690 max mem: 64948 Test: Total time: 0:00:05 (0.6636 s / it) * Acc@1 79.968 Acc@5 94.944 loss 0.760 Accuracy of the model on the 50000 test images: 80.0% Max accuracy: 80.22% Test: [0/9] eta: 0:00:43 loss: 0.4946 (0.4946) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.8865 data: 4.6688 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6812 (0.6775) acc1: 81.7708 (80.5120) acc5: 97.1354 (96.4160) time: 0.6942 data: 0.5189 max mem: 64948 Test: Total time: 0:00:06 (0.7050 s / it) * Acc@1 82.296 Acc@5 96.200 loss 0.652 Accuracy of the model EMA on 50000 test images: 82.3% Max EMA accuracy: 82.30% Epoch: [259] [ 0/312] eta: 0:48:22 lr: 0.001652 min_lr: 0.001652 loss: 2.3365 (2.3365) weight_decay: 0.0500 (0.0500) time: 9.3031 data: 8.4918 max mem: 64948 Epoch: [259] [ 10/312] eta: 0:07:34 lr: 0.001651 min_lr: 0.001651 loss: 1.9760 (2.0462) weight_decay: 0.0500 (0.0500) time: 1.5056 data: 0.7723 max mem: 64948 Epoch: [259] [ 20/312] eta: 0:05:27 lr: 0.001651 min_lr: 0.001651 loss: 1.9760 (2.0531) weight_decay: 0.0500 (0.0500) time: 0.7110 data: 0.0004 max mem: 64948 Epoch: [259] [ 30/312] eta: 0:04:37 lr: 0.001650 min_lr: 0.001650 loss: 1.9754 (2.0367) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [259] [ 40/312] eta: 0:04:08 lr: 0.001650 min_lr: 0.001650 loss: 1.9416 (2.0304) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0003 max mem: 64948 Epoch: [259] [ 50/312] eta: 0:03:48 lr: 0.001649 min_lr: 0.001649 loss: 2.0198 (2.0221) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [259] [ 60/312] eta: 0:03:32 lr: 0.001649 min_lr: 0.001649 loss: 2.1440 (2.0392) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [259] [ 70/312] eta: 0:03:18 lr: 0.001648 min_lr: 0.001648 loss: 2.0216 (2.0118) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [259] [ 80/312] eta: 0:03:07 lr: 0.001648 min_lr: 0.001648 loss: 2.0216 (2.0330) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [259] [ 90/312] eta: 0:02:56 lr: 0.001648 min_lr: 0.001648 loss: 2.0898 (2.0337) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [259] [100/312] eta: 0:02:46 lr: 0.001647 min_lr: 0.001647 loss: 2.1434 (2.0429) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [259] [110/312] eta: 0:02:36 lr: 0.001647 min_lr: 0.001647 loss: 2.1874 (2.0526) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [259] [120/312] eta: 0:02:27 lr: 0.001646 min_lr: 0.001646 loss: 2.1002 (2.0545) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [259] [130/312] eta: 0:02:19 lr: 0.001646 min_lr: 0.001646 loss: 2.0921 (2.0502) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [259] [140/312] eta: 0:02:10 lr: 0.001645 min_lr: 0.001645 loss: 2.0756 (2.0541) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [259] [150/312] eta: 0:02:02 lr: 0.001645 min_lr: 0.001645 loss: 2.2238 (2.0637) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [259] [160/312] eta: 0:01:54 lr: 0.001644 min_lr: 0.001644 loss: 2.2238 (2.0665) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [259] [170/312] eta: 0:01:46 lr: 0.001644 min_lr: 0.001644 loss: 2.1558 (2.0680) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [259] [180/312] eta: 0:01:38 lr: 0.001643 min_lr: 0.001643 loss: 2.0715 (2.0659) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [259] [190/312] eta: 0:01:30 lr: 0.001643 min_lr: 0.001643 loss: 2.0715 (2.0679) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [259] [200/312] eta: 0:01:22 lr: 0.001642 min_lr: 0.001642 loss: 2.2480 (2.0690) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [259] [210/312] eta: 0:01:15 lr: 0.001642 min_lr: 0.001642 loss: 2.1677 (2.0733) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [259] [220/312] eta: 0:01:07 lr: 0.001642 min_lr: 0.001642 loss: 2.0808 (2.0622) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [259] [230/312] eta: 0:01:00 lr: 0.001641 min_lr: 0.001641 loss: 2.0808 (2.0651) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [259] [240/312] eta: 0:00:52 lr: 0.001641 min_lr: 0.001641 loss: 2.0650 (2.0598) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [259] [250/312] eta: 0:00:45 lr: 0.001640 min_lr: 0.001640 loss: 1.8511 (2.0495) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [259] [260/312] eta: 0:00:37 lr: 0.001640 min_lr: 0.001640 loss: 1.8956 (2.0476) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [259] [270/312] eta: 0:00:30 lr: 0.001639 min_lr: 0.001639 loss: 2.1365 (2.0529) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [259] [280/312] eta: 0:00:23 lr: 0.001639 min_lr: 0.001639 loss: 2.0856 (2.0446) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [259] [290/312] eta: 0:00:15 lr: 0.001638 min_lr: 0.001638 loss: 1.9139 (2.0434) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [259] [300/312] eta: 0:00:08 lr: 0.001638 min_lr: 0.001638 loss: 2.0363 (2.0405) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [259] [310/312] eta: 0:00:01 lr: 0.001637 min_lr: 0.001637 loss: 1.9107 (2.0365) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [259] [311/312] eta: 0:00:00 lr: 0.001637 min_lr: 0.001637 loss: 1.8984 (2.0337) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [259] Total time: 0:03:46 (0.7268 s / it) Averaged stats: lr: 0.001637 min_lr: 0.001637 loss: 1.8984 (2.0161) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5690 (0.5690) acc1: 85.1562 (85.1562) acc5: 96.8750 (96.8750) time: 4.4891 data: 4.2699 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7852 (0.7621) acc1: 79.4271 (79.3920) acc5: 95.3125 (95.2320) time: 0.6503 data: 0.4745 max mem: 64948 Test: Total time: 0:00:06 (0.6714 s / it) * Acc@1 79.982 Acc@5 95.076 loss 0.761 Accuracy of the model on the 50000 test images: 80.0% Max accuracy: 80.22% Test: [0/9] eta: 0:00:42 loss: 0.4947 (0.4947) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.7486 data: 4.5305 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6803 (0.6769) acc1: 81.7708 (80.4480) acc5: 97.1354 (96.4160) time: 0.6790 data: 0.5035 max mem: 64948 Test: Total time: 0:00:06 (0.6870 s / it) * Acc@1 82.290 Acc@5 96.204 loss 0.652 Accuracy of the model EMA on 50000 test images: 82.3% Epoch: [260] [ 0/312] eta: 0:53:16 lr: 0.001637 min_lr: 0.001637 loss: 2.1153 (2.1153) weight_decay: 0.0500 (0.0500) time: 10.2461 data: 6.1655 max mem: 64948 Epoch: [260] [ 10/312] eta: 0:08:04 lr: 0.001637 min_lr: 0.001637 loss: 2.0728 (2.0569) weight_decay: 0.0500 (0.0500) time: 1.6032 data: 0.5610 max mem: 64948 Epoch: [260] [ 20/312] eta: 0:05:41 lr: 0.001636 min_lr: 0.001636 loss: 2.0700 (2.0366) weight_decay: 0.0500 (0.0500) time: 0.7164 data: 0.0005 max mem: 64948 Epoch: [260] [ 30/312] eta: 0:04:47 lr: 0.001636 min_lr: 0.001636 loss: 2.1092 (2.0380) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [260] [ 40/312] eta: 0:04:15 lr: 0.001635 min_lr: 0.001635 loss: 2.0909 (2.0369) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [260] [ 50/312] eta: 0:03:53 lr: 0.001635 min_lr: 0.001635 loss: 2.0585 (2.0129) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [260] [ 60/312] eta: 0:03:36 lr: 0.001635 min_lr: 0.001635 loss: 2.0708 (2.0234) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [260] [ 70/312] eta: 0:03:22 lr: 0.001634 min_lr: 0.001634 loss: 2.1036 (2.0278) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [260] [ 80/312] eta: 0:03:10 lr: 0.001634 min_lr: 0.001634 loss: 1.9348 (1.9952) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [260] [ 90/312] eta: 0:02:58 lr: 0.001633 min_lr: 0.001633 loss: 1.9348 (1.9934) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [260] [100/312] eta: 0:02:48 lr: 0.001633 min_lr: 0.001633 loss: 1.9589 (1.9798) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [260] [110/312] eta: 0:02:38 lr: 0.001632 min_lr: 0.001632 loss: 1.9035 (1.9731) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [260] [120/312] eta: 0:02:29 lr: 0.001632 min_lr: 0.001632 loss: 2.0246 (1.9867) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [260] [130/312] eta: 0:02:20 lr: 0.001631 min_lr: 0.001631 loss: 2.0031 (1.9762) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [260] [140/312] eta: 0:02:11 lr: 0.001631 min_lr: 0.001631 loss: 2.0031 (1.9837) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [260] [150/312] eta: 0:02:03 lr: 0.001630 min_lr: 0.001630 loss: 2.1405 (1.9825) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [260] [160/312] eta: 0:01:55 lr: 0.001630 min_lr: 0.001630 loss: 1.9868 (1.9799) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [260] [170/312] eta: 0:01:46 lr: 0.001629 min_lr: 0.001629 loss: 2.0417 (1.9828) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [260] [180/312] eta: 0:01:39 lr: 0.001629 min_lr: 0.001629 loss: 2.0871 (1.9823) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [260] [190/312] eta: 0:01:31 lr: 0.001629 min_lr: 0.001629 loss: 2.0397 (1.9859) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [260] [200/312] eta: 0:01:23 lr: 0.001628 min_lr: 0.001628 loss: 1.9121 (1.9833) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [260] [210/312] eta: 0:01:15 lr: 0.001628 min_lr: 0.001628 loss: 1.8798 (1.9799) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [260] [220/312] eta: 0:01:08 lr: 0.001627 min_lr: 0.001627 loss: 2.0072 (1.9850) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [260] [230/312] eta: 0:01:00 lr: 0.001627 min_lr: 0.001627 loss: 1.9647 (1.9827) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [260] [240/312] eta: 0:00:53 lr: 0.001626 min_lr: 0.001626 loss: 2.0446 (1.9875) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [260] [250/312] eta: 0:00:45 lr: 0.001626 min_lr: 0.001626 loss: 2.0900 (1.9869) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [260] [260/312] eta: 0:00:38 lr: 0.001625 min_lr: 0.001625 loss: 2.0484 (1.9880) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [260] [270/312] eta: 0:00:30 lr: 0.001625 min_lr: 0.001625 loss: 1.9467 (1.9823) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [260] [280/312] eta: 0:00:23 lr: 0.001624 min_lr: 0.001624 loss: 1.9717 (1.9821) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0009 max mem: 64948 Epoch: [260] [290/312] eta: 0:00:16 lr: 0.001624 min_lr: 0.001624 loss: 2.0221 (1.9794) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [260] [300/312] eta: 0:00:08 lr: 0.001624 min_lr: 0.001624 loss: 1.9085 (1.9786) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [260] [310/312] eta: 0:00:01 lr: 0.001623 min_lr: 0.001623 loss: 2.0107 (1.9806) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [260] [311/312] eta: 0:00:00 lr: 0.001623 min_lr: 0.001623 loss: 2.0179 (1.9813) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [260] Total time: 0:03:47 (0.7295 s / it) Averaged stats: lr: 0.001623 min_lr: 0.001623 loss: 2.0179 (2.0071) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5942 (0.5942) acc1: 84.6354 (84.6354) acc5: 95.8333 (95.8333) time: 4.4348 data: 4.2259 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7632 (0.7812) acc1: 81.7708 (79.5840) acc5: 94.5312 (94.7840) time: 0.6440 data: 0.4696 max mem: 64948 Test: Total time: 0:00:06 (0.6676 s / it) * Acc@1 80.094 Acc@5 95.050 loss 0.764 Accuracy of the model on the 50000 test images: 80.1% Max accuracy: 80.22% Test: [0/9] eta: 0:00:44 loss: 0.4937 (0.4937) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.9042 data: 4.6965 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6792 (0.6762) acc1: 81.7708 (80.4800) acc5: 97.1354 (96.4160) time: 0.6962 data: 0.5219 max mem: 64948 Test: Total time: 0:00:06 (0.7054 s / it) * Acc@1 82.296 Acc@5 96.212 loss 0.651 Accuracy of the model EMA on 50000 test images: 82.3% Max EMA accuracy: 82.30% Epoch: [261] [ 0/312] eta: 0:49:54 lr: 0.001623 min_lr: 0.001623 loss: 2.2370 (2.2370) weight_decay: 0.0500 (0.0500) time: 9.5977 data: 8.2032 max mem: 64948 Epoch: [261] [ 10/312] eta: 0:07:45 lr: 0.001623 min_lr: 0.001623 loss: 1.9859 (1.9780) weight_decay: 0.0500 (0.0500) time: 1.5416 data: 0.7461 max mem: 64948 Epoch: [261] [ 20/312] eta: 0:05:31 lr: 0.001622 min_lr: 0.001622 loss: 2.0732 (2.0660) weight_decay: 0.0500 (0.0500) time: 0.7139 data: 0.0004 max mem: 64948 Epoch: [261] [ 30/312] eta: 0:04:40 lr: 0.001622 min_lr: 0.001622 loss: 2.1565 (2.0717) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [261] [ 40/312] eta: 0:04:10 lr: 0.001621 min_lr: 0.001621 loss: 2.1477 (2.0421) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [261] [ 50/312] eta: 0:03:49 lr: 0.001621 min_lr: 0.001621 loss: 2.1124 (2.0308) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [261] [ 60/312] eta: 0:03:33 lr: 0.001620 min_lr: 0.001620 loss: 2.1124 (2.0188) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [261] [ 70/312] eta: 0:03:20 lr: 0.001620 min_lr: 0.001620 loss: 1.8118 (1.9838) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [261] [ 80/312] eta: 0:03:08 lr: 0.001619 min_lr: 0.001619 loss: 1.7894 (1.9698) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [261] [ 90/312] eta: 0:02:57 lr: 0.001619 min_lr: 0.001619 loss: 2.0060 (1.9826) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [261] [100/312] eta: 0:02:47 lr: 0.001618 min_lr: 0.001618 loss: 2.0880 (1.9818) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [261] [110/312] eta: 0:02:37 lr: 0.001618 min_lr: 0.001618 loss: 1.8650 (1.9674) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [261] [120/312] eta: 0:02:28 lr: 0.001617 min_lr: 0.001617 loss: 1.8304 (1.9668) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [261] [130/312] eta: 0:02:19 lr: 0.001617 min_lr: 0.001617 loss: 2.0978 (1.9802) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [261] [140/312] eta: 0:02:11 lr: 0.001617 min_lr: 0.001617 loss: 2.1557 (1.9813) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [261] [150/312] eta: 0:02:02 lr: 0.001616 min_lr: 0.001616 loss: 2.0189 (1.9798) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [261] [160/312] eta: 0:01:54 lr: 0.001616 min_lr: 0.001616 loss: 1.8822 (1.9802) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [261] [170/312] eta: 0:01:46 lr: 0.001615 min_lr: 0.001615 loss: 2.0483 (1.9807) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [261] [180/312] eta: 0:01:38 lr: 0.001615 min_lr: 0.001615 loss: 2.0483 (1.9797) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [261] [190/312] eta: 0:01:30 lr: 0.001614 min_lr: 0.001614 loss: 2.0087 (1.9794) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [261] [200/312] eta: 0:01:23 lr: 0.001614 min_lr: 0.001614 loss: 2.1106 (1.9893) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [261] [210/312] eta: 0:01:15 lr: 0.001613 min_lr: 0.001613 loss: 2.1022 (1.9852) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [261] [220/312] eta: 0:01:07 lr: 0.001613 min_lr: 0.001613 loss: 1.8011 (1.9832) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [261] [230/312] eta: 0:01:00 lr: 0.001612 min_lr: 0.001612 loss: 2.1661 (1.9947) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [261] [240/312] eta: 0:00:52 lr: 0.001612 min_lr: 0.001612 loss: 2.1811 (1.9961) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [261] [250/312] eta: 0:00:45 lr: 0.001611 min_lr: 0.001611 loss: 2.1085 (1.9987) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [261] [260/312] eta: 0:00:38 lr: 0.001611 min_lr: 0.001611 loss: 2.1140 (2.0006) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [261] [270/312] eta: 0:00:30 lr: 0.001611 min_lr: 0.001611 loss: 1.9996 (1.9986) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [261] [280/312] eta: 0:00:23 lr: 0.001610 min_lr: 0.001610 loss: 1.9797 (1.9990) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0010 max mem: 64948 Epoch: [261] [290/312] eta: 0:00:16 lr: 0.001610 min_lr: 0.001610 loss: 2.0287 (1.9982) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [261] [300/312] eta: 0:00:08 lr: 0.001609 min_lr: 0.001609 loss: 2.0001 (1.9994) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [261] [310/312] eta: 0:00:01 lr: 0.001609 min_lr: 0.001609 loss: 2.0820 (2.0017) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [261] [311/312] eta: 0:00:00 lr: 0.001609 min_lr: 0.001609 loss: 2.0273 (2.0000) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [261] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.001609 min_lr: 0.001609 loss: 2.0273 (2.0088) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.6028 (0.6028) acc1: 85.9375 (85.9375) acc5: 96.8750 (96.8750) time: 4.6570 data: 4.4453 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7813 (0.8009) acc1: 79.9479 (78.7840) acc5: 95.8333 (94.9760) time: 0.6688 data: 0.4940 max mem: 64948 Test: Total time: 0:00:06 (0.6968 s / it) * Acc@1 80.036 Acc@5 95.034 loss 0.770 Accuracy of the model on the 50000 test images: 80.0% Max accuracy: 80.22% Test: [0/9] eta: 0:00:41 loss: 0.4929 (0.4929) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.5954 data: 4.3750 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6785 (0.6756) acc1: 81.7708 (80.5120) acc5: 97.1354 (96.3520) time: 0.6625 data: 0.4862 max mem: 64948 Test: Total time: 0:00:06 (0.6746 s / it) * Acc@1 82.318 Acc@5 96.222 loss 0.651 Accuracy of the model EMA on 50000 test images: 82.3% Max EMA accuracy: 82.32% Epoch: [262] [ 0/312] eta: 0:51:26 lr: 0.001609 min_lr: 0.001609 loss: 1.8039 (1.8039) weight_decay: 0.0500 (0.0500) time: 9.8935 data: 8.1955 max mem: 64948 Epoch: [262] [ 10/312] eta: 0:08:02 lr: 0.001608 min_lr: 0.001608 loss: 2.0422 (1.8941) weight_decay: 0.0500 (0.0500) time: 1.5969 data: 0.7930 max mem: 64948 Epoch: [262] [ 20/312] eta: 0:05:41 lr: 0.001608 min_lr: 0.001608 loss: 2.0057 (1.9087) weight_decay: 0.0500 (0.0500) time: 0.7320 data: 0.0265 max mem: 64948 Epoch: [262] [ 30/312] eta: 0:04:46 lr: 0.001607 min_lr: 0.001607 loss: 2.0057 (1.9401) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [262] [ 40/312] eta: 0:04:15 lr: 0.001607 min_lr: 0.001607 loss: 2.0532 (1.9473) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0004 max mem: 64948 Epoch: [262] [ 50/312] eta: 0:03:53 lr: 0.001606 min_lr: 0.001606 loss: 2.0800 (1.9657) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [262] [ 60/312] eta: 0:03:36 lr: 0.001606 min_lr: 0.001606 loss: 2.1592 (1.9910) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [262] [ 70/312] eta: 0:03:22 lr: 0.001605 min_lr: 0.001605 loss: 2.2010 (2.0131) weight_decay: 0.0500 (0.0500) time: 0.7030 data: 0.0004 max mem: 64948 Epoch: [262] [ 80/312] eta: 0:03:10 lr: 0.001605 min_lr: 0.001605 loss: 2.1432 (2.0085) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [262] [ 90/312] eta: 0:02:59 lr: 0.001604 min_lr: 0.001604 loss: 2.0165 (2.0120) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [262] [100/312] eta: 0:02:48 lr: 0.001604 min_lr: 0.001604 loss: 2.0332 (2.0169) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [262] [110/312] eta: 0:02:38 lr: 0.001604 min_lr: 0.001604 loss: 2.0607 (2.0243) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [262] [120/312] eta: 0:02:29 lr: 0.001603 min_lr: 0.001603 loss: 2.0607 (2.0262) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [262] [130/312] eta: 0:02:20 lr: 0.001603 min_lr: 0.001603 loss: 1.9788 (2.0309) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [262] [140/312] eta: 0:02:11 lr: 0.001602 min_lr: 0.001602 loss: 1.9705 (2.0210) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [262] [150/312] eta: 0:02:03 lr: 0.001602 min_lr: 0.001602 loss: 1.8858 (2.0150) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [262] [160/312] eta: 0:01:55 lr: 0.001601 min_lr: 0.001601 loss: 2.1444 (2.0278) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [262] [170/312] eta: 0:01:47 lr: 0.001601 min_lr: 0.001601 loss: 2.1444 (2.0235) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [262] [180/312] eta: 0:01:39 lr: 0.001600 min_lr: 0.001600 loss: 1.9092 (2.0152) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [262] [190/312] eta: 0:01:31 lr: 0.001600 min_lr: 0.001600 loss: 2.0665 (2.0192) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [262] [200/312] eta: 0:01:23 lr: 0.001599 min_lr: 0.001599 loss: 2.0916 (2.0158) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [262] [210/312] eta: 0:01:15 lr: 0.001599 min_lr: 0.001599 loss: 2.0000 (2.0180) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [262] [220/312] eta: 0:01:08 lr: 0.001599 min_lr: 0.001599 loss: 2.0993 (2.0195) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [262] [230/312] eta: 0:01:00 lr: 0.001598 min_lr: 0.001598 loss: 2.1175 (2.0243) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [262] [240/312] eta: 0:00:53 lr: 0.001598 min_lr: 0.001598 loss: 2.1598 (2.0273) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [262] [250/312] eta: 0:00:45 lr: 0.001597 min_lr: 0.001597 loss: 2.1633 (2.0348) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [262] [260/312] eta: 0:00:38 lr: 0.001597 min_lr: 0.001597 loss: 2.1532 (2.0353) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [262] [270/312] eta: 0:00:30 lr: 0.001596 min_lr: 0.001596 loss: 2.0656 (2.0336) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [262] [280/312] eta: 0:00:23 lr: 0.001596 min_lr: 0.001596 loss: 2.1843 (2.0393) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [262] [290/312] eta: 0:00:16 lr: 0.001595 min_lr: 0.001595 loss: 2.1065 (2.0384) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [262] [300/312] eta: 0:00:08 lr: 0.001595 min_lr: 0.001595 loss: 2.0473 (2.0401) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [262] [310/312] eta: 0:00:01 lr: 0.001594 min_lr: 0.001594 loss: 2.1664 (2.0442) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [262] [311/312] eta: 0:00:00 lr: 0.001594 min_lr: 0.001594 loss: 2.1664 (2.0442) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [262] Total time: 0:03:47 (0.7296 s / it) Averaged stats: lr: 0.001594 min_lr: 0.001594 loss: 2.1664 (2.0089) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5464 (0.5464) acc1: 85.4167 (85.4167) acc5: 97.3958 (97.3958) time: 4.3503 data: 4.1337 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8020 (0.7677) acc1: 80.2083 (79.2640) acc5: 96.0938 (95.4560) time: 0.6347 data: 0.4594 max mem: 64948 Test: Total time: 0:00:05 (0.6595 s / it) * Acc@1 80.252 Acc@5 95.208 loss 0.756 Accuracy of the model on the 50000 test images: 80.3% Max accuracy: 80.25% Test: [0/9] eta: 0:00:40 loss: 0.4928 (0.4928) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.5267 data: 4.3088 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6782 (0.6751) acc1: 81.7708 (80.4480) acc5: 97.1354 (96.3520) time: 0.6545 data: 0.4789 max mem: 64948 Test: Total time: 0:00:05 (0.6621 s / it) * Acc@1 82.314 Acc@5 96.238 loss 0.650 Accuracy of the model EMA on 50000 test images: 82.3% Epoch: [263] [ 0/312] eta: 0:55:08 lr: 0.001594 min_lr: 0.001594 loss: 1.7673 (1.7673) weight_decay: 0.0500 (0.0500) time: 10.6039 data: 7.0947 max mem: 64948 Epoch: [263] [ 10/312] eta: 0:08:08 lr: 0.001594 min_lr: 0.001594 loss: 2.0941 (2.0379) weight_decay: 0.0500 (0.0500) time: 1.6187 data: 0.6454 max mem: 64948 Epoch: [263] [ 20/312] eta: 0:05:44 lr: 0.001593 min_lr: 0.001593 loss: 2.0032 (2.0435) weight_decay: 0.0500 (0.0500) time: 0.7082 data: 0.0004 max mem: 64948 Epoch: [263] [ 30/312] eta: 0:04:49 lr: 0.001593 min_lr: 0.001593 loss: 2.1241 (2.0998) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [263] [ 40/312] eta: 0:04:16 lr: 0.001592 min_lr: 0.001592 loss: 2.2134 (2.0756) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [263] [ 50/312] eta: 0:03:54 lr: 0.001592 min_lr: 0.001592 loss: 2.0246 (2.0792) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [263] [ 60/312] eta: 0:03:37 lr: 0.001592 min_lr: 0.001592 loss: 1.9453 (2.0396) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [263] [ 70/312] eta: 0:03:23 lr: 0.001591 min_lr: 0.001591 loss: 2.0578 (2.0657) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [263] [ 80/312] eta: 0:03:10 lr: 0.001591 min_lr: 0.001591 loss: 2.1662 (2.0382) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [263] [ 90/312] eta: 0:02:59 lr: 0.001590 min_lr: 0.001590 loss: 2.0476 (2.0422) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [263] [100/312] eta: 0:02:48 lr: 0.001590 min_lr: 0.001590 loss: 1.8946 (2.0169) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [263] [110/312] eta: 0:02:39 lr: 0.001589 min_lr: 0.001589 loss: 1.8652 (2.0146) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [263] [120/312] eta: 0:02:29 lr: 0.001589 min_lr: 0.001589 loss: 2.0184 (2.0164) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [263] [130/312] eta: 0:02:20 lr: 0.001588 min_lr: 0.001588 loss: 1.9627 (2.0089) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [263] [140/312] eta: 0:02:11 lr: 0.001588 min_lr: 0.001588 loss: 2.0386 (2.0140) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [263] [150/312] eta: 0:02:03 lr: 0.001587 min_lr: 0.001587 loss: 2.0642 (2.0118) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [263] [160/312] eta: 0:01:55 lr: 0.001587 min_lr: 0.001587 loss: 1.9131 (2.0072) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [263] [170/312] eta: 0:01:47 lr: 0.001587 min_lr: 0.001587 loss: 1.9984 (2.0070) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [263] [180/312] eta: 0:01:39 lr: 0.001586 min_lr: 0.001586 loss: 2.1358 (2.0118) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [263] [190/312] eta: 0:01:31 lr: 0.001586 min_lr: 0.001586 loss: 2.0630 (2.0074) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [263] [200/312] eta: 0:01:23 lr: 0.001585 min_lr: 0.001585 loss: 1.9409 (2.0034) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [263] [210/312] eta: 0:01:15 lr: 0.001585 min_lr: 0.001585 loss: 2.0236 (2.0038) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [263] [220/312] eta: 0:01:08 lr: 0.001584 min_lr: 0.001584 loss: 2.0236 (1.9972) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [263] [230/312] eta: 0:01:00 lr: 0.001584 min_lr: 0.001584 loss: 1.9632 (1.9938) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [263] [240/312] eta: 0:00:53 lr: 0.001583 min_lr: 0.001583 loss: 2.1677 (2.0037) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [263] [250/312] eta: 0:00:45 lr: 0.001583 min_lr: 0.001583 loss: 2.2123 (2.0047) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [263] [260/312] eta: 0:00:38 lr: 0.001582 min_lr: 0.001582 loss: 2.0558 (2.0035) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [263] [270/312] eta: 0:00:30 lr: 0.001582 min_lr: 0.001582 loss: 2.0814 (2.0014) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [263] [280/312] eta: 0:00:23 lr: 0.001581 min_lr: 0.001581 loss: 1.8779 (1.9992) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [263] [290/312] eta: 0:00:16 lr: 0.001581 min_lr: 0.001581 loss: 1.8706 (1.9971) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [263] [300/312] eta: 0:00:08 lr: 0.001581 min_lr: 0.001581 loss: 1.9193 (1.9959) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [263] [310/312] eta: 0:00:01 lr: 0.001580 min_lr: 0.001580 loss: 2.1470 (1.9981) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [263] [311/312] eta: 0:00:00 lr: 0.001580 min_lr: 0.001580 loss: 2.0471 (1.9970) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [263] Total time: 0:03:47 (0.7300 s / it) Averaged stats: lr: 0.001580 min_lr: 0.001580 loss: 2.0471 (2.0121) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5563 (0.5563) acc1: 86.7188 (86.7188) acc5: 96.6146 (96.6146) time: 4.4945 data: 4.2904 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7914 (0.7770) acc1: 81.2500 (79.8400) acc5: 95.5729 (95.3600) time: 0.6507 data: 0.4768 max mem: 64948 Test: Total time: 0:00:06 (0.6708 s / it) * Acc@1 80.256 Acc@5 95.096 loss 0.759 Accuracy of the model on the 50000 test images: 80.3% Max accuracy: 80.26% Test: [0/9] eta: 0:00:42 loss: 0.4922 (0.4922) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.7637 data: 4.5459 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6782 (0.6747) acc1: 81.7708 (80.4480) acc5: 97.1354 (96.3840) time: 0.6813 data: 0.5052 max mem: 64948 Test: Total time: 0:00:06 (0.6901 s / it) * Acc@1 82.324 Acc@5 96.254 loss 0.650 Accuracy of the model EMA on 50000 test images: 82.3% Max EMA accuracy: 82.32% Epoch: [264] [ 0/312] eta: 0:47:16 lr: 0.001580 min_lr: 0.001580 loss: 1.8080 (1.8080) weight_decay: 0.0500 (0.0500) time: 9.0921 data: 7.2210 max mem: 64948 Epoch: [264] [ 10/312] eta: 0:07:39 lr: 0.001580 min_lr: 0.001580 loss: 1.9030 (1.9295) weight_decay: 0.0500 (0.0500) time: 1.5203 data: 0.6570 max mem: 64948 Epoch: [264] [ 20/312] eta: 0:05:29 lr: 0.001579 min_lr: 0.001579 loss: 1.9851 (1.9723) weight_decay: 0.0500 (0.0500) time: 0.7289 data: 0.0004 max mem: 64948 Epoch: [264] [ 30/312] eta: 0:04:38 lr: 0.001579 min_lr: 0.001579 loss: 1.9851 (1.8843) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [264] [ 40/312] eta: 0:04:09 lr: 0.001578 min_lr: 0.001578 loss: 1.9293 (1.9028) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [264] [ 50/312] eta: 0:03:48 lr: 0.001578 min_lr: 0.001578 loss: 2.0652 (1.9628) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [264] [ 60/312] eta: 0:03:32 lr: 0.001577 min_lr: 0.001577 loss: 2.2893 (1.9852) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [264] [ 70/312] eta: 0:03:19 lr: 0.001577 min_lr: 0.001577 loss: 2.1330 (1.9870) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [264] [ 80/312] eta: 0:03:07 lr: 0.001576 min_lr: 0.001576 loss: 2.1072 (1.9961) weight_decay: 0.0500 (0.0500) time: 0.7006 data: 0.0004 max mem: 64948 Epoch: [264] [ 90/312] eta: 0:02:56 lr: 0.001576 min_lr: 0.001576 loss: 2.1866 (2.0116) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [264] [100/312] eta: 0:02:46 lr: 0.001575 min_lr: 0.001575 loss: 2.1536 (2.0059) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [264] [110/312] eta: 0:02:37 lr: 0.001575 min_lr: 0.001575 loss: 2.0778 (2.0091) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [264] [120/312] eta: 0:02:28 lr: 0.001575 min_lr: 0.001575 loss: 1.8759 (2.0008) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [264] [130/312] eta: 0:02:19 lr: 0.001574 min_lr: 0.001574 loss: 1.8759 (2.0087) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [264] [140/312] eta: 0:02:10 lr: 0.001574 min_lr: 0.001574 loss: 2.2003 (2.0220) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [264] [150/312] eta: 0:02:02 lr: 0.001573 min_lr: 0.001573 loss: 2.0672 (2.0205) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [264] [160/312] eta: 0:01:54 lr: 0.001573 min_lr: 0.001573 loss: 2.0106 (2.0176) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [264] [170/312] eta: 0:01:46 lr: 0.001572 min_lr: 0.001572 loss: 2.1613 (2.0210) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [264] [180/312] eta: 0:01:38 lr: 0.001572 min_lr: 0.001572 loss: 2.2302 (2.0323) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [264] [190/312] eta: 0:01:30 lr: 0.001571 min_lr: 0.001571 loss: 2.2007 (2.0304) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [264] [200/312] eta: 0:01:22 lr: 0.001571 min_lr: 0.001571 loss: 1.8899 (2.0254) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [264] [210/312] eta: 0:01:15 lr: 0.001570 min_lr: 0.001570 loss: 2.1405 (2.0288) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [264] [220/312] eta: 0:01:07 lr: 0.001570 min_lr: 0.001570 loss: 2.1405 (2.0296) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [264] [230/312] eta: 0:01:00 lr: 0.001569 min_lr: 0.001569 loss: 2.1060 (2.0321) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [264] [240/312] eta: 0:00:52 lr: 0.001569 min_lr: 0.001569 loss: 2.1137 (2.0337) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [264] [250/312] eta: 0:00:45 lr: 0.001569 min_lr: 0.001569 loss: 2.2037 (2.0397) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [264] [260/312] eta: 0:00:37 lr: 0.001568 min_lr: 0.001568 loss: 2.2037 (2.0391) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [264] [270/312] eta: 0:00:30 lr: 0.001568 min_lr: 0.001568 loss: 2.0031 (2.0321) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [264] [280/312] eta: 0:00:23 lr: 0.001567 min_lr: 0.001567 loss: 2.0176 (2.0326) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0009 max mem: 64948 Epoch: [264] [290/312] eta: 0:00:15 lr: 0.001567 min_lr: 0.001567 loss: 2.1214 (2.0294) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [264] [300/312] eta: 0:00:08 lr: 0.001566 min_lr: 0.001566 loss: 1.9140 (2.0181) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [264] [310/312] eta: 0:00:01 lr: 0.001566 min_lr: 0.001566 loss: 2.0318 (2.0231) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [264] [311/312] eta: 0:00:00 lr: 0.001566 min_lr: 0.001566 loss: 2.0346 (2.0232) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [264] Total time: 0:03:46 (0.7266 s / it) Averaged stats: lr: 0.001566 min_lr: 0.001566 loss: 2.0346 (2.0024) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5579 (0.5579) acc1: 85.1562 (85.1562) acc5: 96.6146 (96.6146) time: 4.6599 data: 4.4473 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8201 (0.7971) acc1: 79.6875 (78.8160) acc5: 95.8333 (94.8800) time: 0.6690 data: 0.4942 max mem: 64948 Test: Total time: 0:00:06 (0.6899 s / it) * Acc@1 80.200 Acc@5 95.246 loss 0.748 Accuracy of the model on the 50000 test images: 80.2% Max accuracy: 80.26% Test: [0/9] eta: 0:00:43 loss: 0.4914 (0.4914) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.8422 data: 4.6242 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6780 (0.6745) acc1: 81.7708 (80.4800) acc5: 97.1354 (96.3840) time: 0.6897 data: 0.5139 max mem: 64948 Test: Total time: 0:00:06 (0.7012 s / it) * Acc@1 82.342 Acc@5 96.252 loss 0.649 Accuracy of the model EMA on 50000 test images: 82.3% Max EMA accuracy: 82.34% Epoch: [265] [ 0/312] eta: 0:54:51 lr: 0.001566 min_lr: 0.001566 loss: 2.4700 (2.4700) weight_decay: 0.0500 (0.0500) time: 10.5505 data: 9.7616 max mem: 64948 Epoch: [265] [ 10/312] eta: 0:08:05 lr: 0.001565 min_lr: 0.001565 loss: 2.1790 (1.9875) weight_decay: 0.0500 (0.0500) time: 1.6076 data: 0.8878 max mem: 64948 Epoch: [265] [ 20/312] eta: 0:05:43 lr: 0.001565 min_lr: 0.001565 loss: 2.1790 (2.0249) weight_decay: 0.0500 (0.0500) time: 0.7064 data: 0.0004 max mem: 64948 Epoch: [265] [ 30/312] eta: 0:04:47 lr: 0.001564 min_lr: 0.001564 loss: 2.0143 (2.0307) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0003 max mem: 64948 Epoch: [265] [ 40/312] eta: 0:04:16 lr: 0.001564 min_lr: 0.001564 loss: 2.0143 (2.0257) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [265] [ 50/312] eta: 0:03:53 lr: 0.001563 min_lr: 0.001563 loss: 1.9843 (2.0102) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [265] [ 60/312] eta: 0:03:36 lr: 0.001563 min_lr: 0.001563 loss: 1.9000 (1.9960) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [265] [ 70/312] eta: 0:03:22 lr: 0.001563 min_lr: 0.001563 loss: 1.9044 (1.9883) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [265] [ 80/312] eta: 0:03:10 lr: 0.001562 min_lr: 0.001562 loss: 2.0491 (2.0247) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [265] [ 90/312] eta: 0:02:58 lr: 0.001562 min_lr: 0.001562 loss: 2.1951 (2.0212) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [265] [100/312] eta: 0:02:48 lr: 0.001561 min_lr: 0.001561 loss: 2.0402 (2.0124) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [265] [110/312] eta: 0:02:38 lr: 0.001561 min_lr: 0.001561 loss: 2.1066 (2.0166) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [265] [120/312] eta: 0:02:29 lr: 0.001560 min_lr: 0.001560 loss: 2.0891 (2.0085) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [265] [130/312] eta: 0:02:20 lr: 0.001560 min_lr: 0.001560 loss: 2.0821 (2.0073) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [265] [140/312] eta: 0:02:11 lr: 0.001559 min_lr: 0.001559 loss: 2.0189 (1.9931) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [265] [150/312] eta: 0:02:03 lr: 0.001559 min_lr: 0.001559 loss: 2.0335 (2.0026) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [265] [160/312] eta: 0:01:55 lr: 0.001558 min_lr: 0.001558 loss: 2.0980 (2.0019) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [265] [170/312] eta: 0:01:47 lr: 0.001558 min_lr: 0.001558 loss: 2.0980 (2.0095) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [265] [180/312] eta: 0:01:39 lr: 0.001558 min_lr: 0.001558 loss: 2.1191 (2.0136) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [265] [190/312] eta: 0:01:31 lr: 0.001557 min_lr: 0.001557 loss: 2.0509 (2.0075) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [265] [200/312] eta: 0:01:23 lr: 0.001557 min_lr: 0.001557 loss: 2.1116 (2.0112) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [265] [210/312] eta: 0:01:15 lr: 0.001556 min_lr: 0.001556 loss: 2.0254 (2.0049) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [265] [220/312] eta: 0:01:08 lr: 0.001556 min_lr: 0.001556 loss: 2.0254 (2.0113) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [265] [230/312] eta: 0:01:00 lr: 0.001555 min_lr: 0.001555 loss: 2.1412 (2.0156) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [265] [240/312] eta: 0:00:53 lr: 0.001555 min_lr: 0.001555 loss: 2.0354 (2.0089) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [265] [250/312] eta: 0:00:45 lr: 0.001554 min_lr: 0.001554 loss: 2.0986 (2.0153) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [265] [260/312] eta: 0:00:38 lr: 0.001554 min_lr: 0.001554 loss: 2.1234 (2.0178) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [265] [270/312] eta: 0:00:30 lr: 0.001553 min_lr: 0.001553 loss: 2.0613 (2.0167) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [265] [280/312] eta: 0:00:23 lr: 0.001553 min_lr: 0.001553 loss: 2.0051 (2.0145) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0009 max mem: 64948 Epoch: [265] [290/312] eta: 0:00:16 lr: 0.001553 min_lr: 0.001553 loss: 2.0151 (2.0144) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0008 max mem: 64948 Epoch: [265] [300/312] eta: 0:00:08 lr: 0.001552 min_lr: 0.001552 loss: 2.0151 (2.0117) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [265] [310/312] eta: 0:00:01 lr: 0.001552 min_lr: 0.001552 loss: 2.0034 (2.0088) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [265] [311/312] eta: 0:00:00 lr: 0.001552 min_lr: 0.001552 loss: 2.0005 (2.0081) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [265] Total time: 0:03:47 (0.7301 s / it) Averaged stats: lr: 0.001552 min_lr: 0.001552 loss: 2.0005 (1.9998) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.5358 (0.5358) acc1: 85.9375 (85.9375) acc5: 96.8750 (96.8750) time: 4.3001 data: 4.0801 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8248 (0.7793) acc1: 79.4271 (79.1360) acc5: 95.8333 (94.9440) time: 0.6297 data: 0.4534 max mem: 64948 Test: Total time: 0:00:05 (0.6514 s / it) * Acc@1 79.928 Acc@5 95.050 loss 0.772 Accuracy of the model on the 50000 test images: 79.9% Max accuracy: 80.26% Test: [0/9] eta: 0:00:41 loss: 0.4910 (0.4910) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.6000 data: 4.3796 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6776 (0.6743) acc1: 82.0312 (80.5440) acc5: 97.1354 (96.4160) time: 0.6677 data: 0.4867 max mem: 64948 Test: Total time: 0:00:06 (0.6800 s / it) * Acc@1 82.366 Acc@5 96.260 loss 0.649 Accuracy of the model EMA on 50000 test images: 82.4% Max EMA accuracy: 82.37% Epoch: [266] [ 0/312] eta: 0:49:56 lr: 0.001552 min_lr: 0.001552 loss: 2.0878 (2.0878) weight_decay: 0.0500 (0.0500) time: 9.6030 data: 8.4955 max mem: 64948 Epoch: [266] [ 10/312] eta: 0:07:49 lr: 0.001551 min_lr: 0.001551 loss: 1.9873 (1.9872) weight_decay: 0.0500 (0.0500) time: 1.5559 data: 0.7727 max mem: 64948 Epoch: [266] [ 20/312] eta: 0:05:34 lr: 0.001551 min_lr: 0.001551 loss: 1.9507 (1.9147) weight_decay: 0.0500 (0.0500) time: 0.7216 data: 0.0004 max mem: 64948 Epoch: [266] [ 30/312] eta: 0:04:41 lr: 0.001550 min_lr: 0.001550 loss: 1.9507 (1.9302) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [266] [ 40/312] eta: 0:04:11 lr: 0.001550 min_lr: 0.001550 loss: 2.0616 (1.9418) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [266] [ 50/312] eta: 0:03:50 lr: 0.001549 min_lr: 0.001549 loss: 2.0616 (1.9499) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [266] [ 60/312] eta: 0:03:34 lr: 0.001549 min_lr: 0.001549 loss: 1.8711 (1.9324) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [266] [ 70/312] eta: 0:03:20 lr: 0.001548 min_lr: 0.001548 loss: 1.7925 (1.9084) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [266] [ 80/312] eta: 0:03:08 lr: 0.001548 min_lr: 0.001548 loss: 2.0736 (1.9370) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [266] [ 90/312] eta: 0:02:57 lr: 0.001547 min_lr: 0.001547 loss: 2.1288 (1.9501) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [266] [100/312] eta: 0:02:47 lr: 0.001547 min_lr: 0.001547 loss: 2.0655 (1.9614) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [266] [110/312] eta: 0:02:37 lr: 0.001546 min_lr: 0.001546 loss: 2.0383 (1.9528) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [266] [120/312] eta: 0:02:28 lr: 0.001546 min_lr: 0.001546 loss: 2.0339 (1.9530) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [266] [130/312] eta: 0:02:19 lr: 0.001546 min_lr: 0.001546 loss: 2.0544 (1.9659) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [266] [140/312] eta: 0:02:11 lr: 0.001545 min_lr: 0.001545 loss: 2.1618 (1.9704) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [266] [150/312] eta: 0:02:02 lr: 0.001545 min_lr: 0.001545 loss: 2.1136 (1.9760) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [266] [160/312] eta: 0:01:54 lr: 0.001544 min_lr: 0.001544 loss: 2.1046 (1.9803) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [266] [170/312] eta: 0:01:46 lr: 0.001544 min_lr: 0.001544 loss: 2.1046 (1.9822) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [266] [180/312] eta: 0:01:38 lr: 0.001543 min_lr: 0.001543 loss: 2.1501 (1.9903) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [266] [190/312] eta: 0:01:30 lr: 0.001543 min_lr: 0.001543 loss: 1.9275 (1.9788) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [266] [200/312] eta: 0:01:23 lr: 0.001542 min_lr: 0.001542 loss: 1.9275 (1.9836) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [266] [210/312] eta: 0:01:15 lr: 0.001542 min_lr: 0.001542 loss: 2.1284 (1.9857) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [266] [220/312] eta: 0:01:07 lr: 0.001541 min_lr: 0.001541 loss: 2.1064 (1.9923) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [266] [230/312] eta: 0:01:00 lr: 0.001541 min_lr: 0.001541 loss: 2.0824 (1.9975) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [266] [240/312] eta: 0:00:52 lr: 0.001541 min_lr: 0.001541 loss: 2.0602 (1.9977) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [266] [250/312] eta: 0:00:45 lr: 0.001540 min_lr: 0.001540 loss: 2.1051 (1.9978) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [266] [260/312] eta: 0:00:37 lr: 0.001540 min_lr: 0.001540 loss: 1.7952 (1.9875) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [266] [270/312] eta: 0:00:30 lr: 0.001539 min_lr: 0.001539 loss: 1.9628 (1.9937) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [266] [280/312] eta: 0:00:23 lr: 0.001539 min_lr: 0.001539 loss: 2.1690 (1.9964) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0011 max mem: 64948 Epoch: [266] [290/312] eta: 0:00:15 lr: 0.001538 min_lr: 0.001538 loss: 2.0932 (1.9995) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0010 max mem: 64948 Epoch: [266] [300/312] eta: 0:00:08 lr: 0.001538 min_lr: 0.001538 loss: 1.9393 (1.9948) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [266] [310/312] eta: 0:00:01 lr: 0.001537 min_lr: 0.001537 loss: 1.8815 (1.9933) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [266] [311/312] eta: 0:00:00 lr: 0.001537 min_lr: 0.001537 loss: 1.8815 (1.9929) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [266] Total time: 0:03:46 (0.7274 s / it) Averaged stats: lr: 0.001537 min_lr: 0.001537 loss: 1.8815 (1.9935) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5309 (0.5309) acc1: 86.4583 (86.4583) acc5: 96.6146 (96.6146) time: 4.6908 data: 4.4716 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8148 (0.7796) acc1: 78.9062 (79.0080) acc5: 95.8333 (94.7840) time: 0.6724 data: 0.4969 max mem: 64948 Test: Total time: 0:00:06 (0.6986 s / it) * Acc@1 80.138 Acc@5 95.132 loss 0.758 Accuracy of the model on the 50000 test images: 80.1% Max accuracy: 80.26% Test: [0/9] eta: 0:00:44 loss: 0.4908 (0.4908) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.9943 data: 4.7856 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6774 (0.6739) acc1: 82.0312 (80.6400) acc5: 97.1354 (96.4160) time: 0.7062 data: 0.5318 max mem: 64948 Test: Total time: 0:00:06 (0.7217 s / it) * Acc@1 82.392 Acc@5 96.254 loss 0.648 Accuracy of the model EMA on 50000 test images: 82.4% Max EMA accuracy: 82.39% Epoch: [267] [ 0/312] eta: 0:46:44 lr: 0.001537 min_lr: 0.001537 loss: 2.2937 (2.2937) weight_decay: 0.0500 (0.0500) time: 8.9887 data: 7.5766 max mem: 64948 Epoch: [267] [ 10/312] eta: 0:07:36 lr: 0.001537 min_lr: 0.001537 loss: 2.1530 (2.0854) weight_decay: 0.0500 (0.0500) time: 1.5112 data: 0.6892 max mem: 64948 Epoch: [267] [ 20/312] eta: 0:05:27 lr: 0.001536 min_lr: 0.001536 loss: 2.1530 (2.0669) weight_decay: 0.0500 (0.0500) time: 0.7288 data: 0.0004 max mem: 64948 Epoch: [267] [ 30/312] eta: 0:04:37 lr: 0.001536 min_lr: 0.001536 loss: 1.9116 (1.9673) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [267] [ 40/312] eta: 0:04:08 lr: 0.001535 min_lr: 0.001535 loss: 1.9293 (1.9885) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [267] [ 50/312] eta: 0:03:48 lr: 0.001535 min_lr: 0.001535 loss: 1.9855 (1.9813) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [267] [ 60/312] eta: 0:03:32 lr: 0.001535 min_lr: 0.001535 loss: 1.8615 (1.9618) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [267] [ 70/312] eta: 0:03:19 lr: 0.001534 min_lr: 0.001534 loss: 1.9956 (1.9615) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [267] [ 80/312] eta: 0:03:07 lr: 0.001534 min_lr: 0.001534 loss: 1.9240 (1.9420) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [267] [ 90/312] eta: 0:02:56 lr: 0.001533 min_lr: 0.001533 loss: 2.0276 (1.9512) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [267] [100/312] eta: 0:02:46 lr: 0.001533 min_lr: 0.001533 loss: 2.0276 (1.9481) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [267] [110/312] eta: 0:02:36 lr: 0.001532 min_lr: 0.001532 loss: 1.9555 (1.9651) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [267] [120/312] eta: 0:02:27 lr: 0.001532 min_lr: 0.001532 loss: 2.0786 (1.9667) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [267] [130/312] eta: 0:02:19 lr: 0.001531 min_lr: 0.001531 loss: 1.9178 (1.9609) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [267] [140/312] eta: 0:02:10 lr: 0.001531 min_lr: 0.001531 loss: 1.7810 (1.9595) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [267] [150/312] eta: 0:02:02 lr: 0.001530 min_lr: 0.001530 loss: 1.9183 (1.9653) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [267] [160/312] eta: 0:01:54 lr: 0.001530 min_lr: 0.001530 loss: 2.1103 (1.9821) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [267] [170/312] eta: 0:01:46 lr: 0.001530 min_lr: 0.001530 loss: 2.1103 (1.9813) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [267] [180/312] eta: 0:01:38 lr: 0.001529 min_lr: 0.001529 loss: 2.0648 (1.9866) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [267] [190/312] eta: 0:01:30 lr: 0.001529 min_lr: 0.001529 loss: 2.0442 (1.9835) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [267] [200/312] eta: 0:01:22 lr: 0.001528 min_lr: 0.001528 loss: 2.1462 (1.9938) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [267] [210/312] eta: 0:01:15 lr: 0.001528 min_lr: 0.001528 loss: 2.1389 (1.9885) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [267] [220/312] eta: 0:01:07 lr: 0.001527 min_lr: 0.001527 loss: 1.8296 (1.9841) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [267] [230/312] eta: 0:01:00 lr: 0.001527 min_lr: 0.001527 loss: 2.0372 (1.9887) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [267] [240/312] eta: 0:00:52 lr: 0.001526 min_lr: 0.001526 loss: 2.0372 (1.9841) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [267] [250/312] eta: 0:00:45 lr: 0.001526 min_lr: 0.001526 loss: 2.0033 (1.9867) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [267] [260/312] eta: 0:00:37 lr: 0.001525 min_lr: 0.001525 loss: 2.1094 (1.9922) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [267] [270/312] eta: 0:00:30 lr: 0.001525 min_lr: 0.001525 loss: 2.1663 (1.9932) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [267] [280/312] eta: 0:00:23 lr: 0.001525 min_lr: 0.001525 loss: 2.1644 (1.9968) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0010 max mem: 64948 Epoch: [267] [290/312] eta: 0:00:15 lr: 0.001524 min_lr: 0.001524 loss: 2.1619 (2.0053) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [267] [300/312] eta: 0:00:08 lr: 0.001524 min_lr: 0.001524 loss: 2.0869 (2.0040) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [267] [310/312] eta: 0:00:01 lr: 0.001523 min_lr: 0.001523 loss: 2.1226 (2.0079) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [267] [311/312] eta: 0:00:00 lr: 0.001523 min_lr: 0.001523 loss: 2.1377 (2.0089) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [267] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.001523 min_lr: 0.001523 loss: 2.1377 (1.9920) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.6004 (0.6004) acc1: 85.1562 (85.1562) acc5: 96.3542 (96.3542) time: 4.5394 data: 4.3345 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7522 (0.7863) acc1: 80.7292 (78.8800) acc5: 95.3125 (95.1360) time: 0.6558 data: 0.4817 max mem: 64948 Test: Total time: 0:00:06 (0.6787 s / it) * Acc@1 80.018 Acc@5 95.108 loss 0.765 Accuracy of the model on the 50000 test images: 80.0% Max accuracy: 80.26% Test: [0/9] eta: 0:00:42 loss: 0.4905 (0.4905) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.7621 data: 4.5441 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6768 (0.6736) acc1: 82.0312 (80.6720) acc5: 97.1354 (96.3520) time: 0.6809 data: 0.5050 max mem: 64948 Test: Total time: 0:00:06 (0.6897 s / it) * Acc@1 82.426 Acc@5 96.254 loss 0.648 Accuracy of the model EMA on 50000 test images: 82.4% Max EMA accuracy: 82.43% Epoch: [268] [ 0/312] eta: 0:52:04 lr: 0.001523 min_lr: 0.001523 loss: 1.7131 (1.7131) weight_decay: 0.0500 (0.0500) time: 10.0160 data: 9.2328 max mem: 64948 Epoch: [268] [ 10/312] eta: 0:07:51 lr: 0.001523 min_lr: 0.001523 loss: 2.1331 (2.0687) weight_decay: 0.0500 (0.0500) time: 1.5598 data: 0.8398 max mem: 64948 Epoch: [268] [ 20/312] eta: 0:05:35 lr: 0.001522 min_lr: 0.001522 loss: 2.0353 (1.9926) weight_decay: 0.0500 (0.0500) time: 0.7041 data: 0.0005 max mem: 64948 Epoch: [268] [ 30/312] eta: 0:04:42 lr: 0.001522 min_lr: 0.001522 loss: 1.8699 (1.9694) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [268] [ 40/312] eta: 0:04:11 lr: 0.001521 min_lr: 0.001521 loss: 1.8699 (1.9705) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [268] [ 50/312] eta: 0:03:50 lr: 0.001521 min_lr: 0.001521 loss: 2.1395 (2.0011) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [268] [ 60/312] eta: 0:03:34 lr: 0.001520 min_lr: 0.001520 loss: 2.1395 (2.0281) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [268] [ 70/312] eta: 0:03:20 lr: 0.001520 min_lr: 0.001520 loss: 2.0924 (2.0315) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [268] [ 80/312] eta: 0:03:08 lr: 0.001519 min_lr: 0.001519 loss: 2.1524 (2.0437) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [268] [ 90/312] eta: 0:02:57 lr: 0.001519 min_lr: 0.001519 loss: 2.1730 (2.0280) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [268] [100/312] eta: 0:02:47 lr: 0.001519 min_lr: 0.001519 loss: 1.9824 (2.0195) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [268] [110/312] eta: 0:02:37 lr: 0.001518 min_lr: 0.001518 loss: 1.9021 (2.0031) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [268] [120/312] eta: 0:02:28 lr: 0.001518 min_lr: 0.001518 loss: 1.8976 (1.9993) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [268] [130/312] eta: 0:02:19 lr: 0.001517 min_lr: 0.001517 loss: 2.0240 (2.0073) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [268] [140/312] eta: 0:02:11 lr: 0.001517 min_lr: 0.001517 loss: 2.1537 (2.0129) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [268] [150/312] eta: 0:02:02 lr: 0.001516 min_lr: 0.001516 loss: 2.1537 (2.0118) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [268] [160/312] eta: 0:01:54 lr: 0.001516 min_lr: 0.001516 loss: 2.0489 (2.0143) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [268] [170/312] eta: 0:01:46 lr: 0.001515 min_lr: 0.001515 loss: 2.0489 (2.0169) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [268] [180/312] eta: 0:01:38 lr: 0.001515 min_lr: 0.001515 loss: 1.9978 (2.0123) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [268] [190/312] eta: 0:01:30 lr: 0.001514 min_lr: 0.001514 loss: 2.1191 (2.0152) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [268] [200/312] eta: 0:01:23 lr: 0.001514 min_lr: 0.001514 loss: 2.2211 (2.0242) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [268] [210/312] eta: 0:01:15 lr: 0.001514 min_lr: 0.001514 loss: 2.1449 (2.0258) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [268] [220/312] eta: 0:01:07 lr: 0.001513 min_lr: 0.001513 loss: 1.9325 (2.0165) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [268] [230/312] eta: 0:01:00 lr: 0.001513 min_lr: 0.001513 loss: 1.9266 (2.0211) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [268] [240/312] eta: 0:00:52 lr: 0.001512 min_lr: 0.001512 loss: 2.0586 (2.0159) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [268] [250/312] eta: 0:00:45 lr: 0.001512 min_lr: 0.001512 loss: 2.0406 (2.0117) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [268] [260/312] eta: 0:00:38 lr: 0.001511 min_lr: 0.001511 loss: 2.1454 (2.0176) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [268] [270/312] eta: 0:00:30 lr: 0.001511 min_lr: 0.001511 loss: 2.1143 (2.0190) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [268] [280/312] eta: 0:00:23 lr: 0.001510 min_lr: 0.001510 loss: 2.0921 (2.0181) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0010 max mem: 64948 Epoch: [268] [290/312] eta: 0:00:15 lr: 0.001510 min_lr: 0.001510 loss: 2.0258 (2.0124) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0008 max mem: 64948 Epoch: [268] [300/312] eta: 0:00:08 lr: 0.001509 min_lr: 0.001509 loss: 1.9420 (2.0072) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [268] [310/312] eta: 0:00:01 lr: 0.001509 min_lr: 0.001509 loss: 1.9985 (2.0058) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [268] [311/312] eta: 0:00:00 lr: 0.001509 min_lr: 0.001509 loss: 1.9985 (2.0059) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [268] Total time: 0:03:47 (0.7278 s / it) Averaged stats: lr: 0.001509 min_lr: 0.001509 loss: 1.9985 (1.9920) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5628 (0.5628) acc1: 86.4583 (86.4583) acc5: 96.3542 (96.3542) time: 4.5509 data: 4.3365 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7736 (0.7762) acc1: 80.2083 (79.3600) acc5: 95.5729 (95.2640) time: 0.6569 data: 0.4819 max mem: 64948 Test: Total time: 0:00:06 (0.6810 s / it) * Acc@1 80.390 Acc@5 95.134 loss 0.753 Accuracy of the model on the 50000 test images: 80.4% Max accuracy: 80.39% Test: [0/9] eta: 0:00:42 loss: 0.4901 (0.4901) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 4.6764 data: 4.4586 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6758 (0.6733) acc1: 82.0312 (80.6080) acc5: 97.1354 (96.3520) time: 0.6714 data: 0.4955 max mem: 64948 Test: Total time: 0:00:06 (0.6790 s / it) * Acc@1 82.432 Acc@5 96.266 loss 0.648 Accuracy of the model EMA on 50000 test images: 82.4% Max EMA accuracy: 82.43% Epoch: [269] [ 0/312] eta: 0:54:03 lr: 0.001509 min_lr: 0.001509 loss: 2.1047 (2.1047) weight_decay: 0.0500 (0.0500) time: 10.3943 data: 9.6061 max mem: 64948 Epoch: [269] [ 10/312] eta: 0:08:04 lr: 0.001508 min_lr: 0.001508 loss: 2.1783 (2.1206) weight_decay: 0.0500 (0.0500) time: 1.6043 data: 0.8736 max mem: 64948 Epoch: [269] [ 20/312] eta: 0:05:42 lr: 0.001508 min_lr: 0.001508 loss: 2.0395 (1.9885) weight_decay: 0.0500 (0.0500) time: 0.7109 data: 0.0004 max mem: 64948 Epoch: [269] [ 30/312] eta: 0:04:47 lr: 0.001508 min_lr: 0.001508 loss: 1.8385 (1.9486) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [269] [ 40/312] eta: 0:04:15 lr: 0.001507 min_lr: 0.001507 loss: 1.9852 (1.9334) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [269] [ 50/312] eta: 0:03:53 lr: 0.001507 min_lr: 0.001507 loss: 1.9847 (1.9249) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [269] [ 60/312] eta: 0:03:36 lr: 0.001506 min_lr: 0.001506 loss: 1.9847 (1.9345) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [269] [ 70/312] eta: 0:03:22 lr: 0.001506 min_lr: 0.001506 loss: 2.0022 (1.9506) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [269] [ 80/312] eta: 0:03:09 lr: 0.001505 min_lr: 0.001505 loss: 2.0486 (1.9570) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [269] [ 90/312] eta: 0:02:58 lr: 0.001505 min_lr: 0.001505 loss: 2.1795 (1.9835) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [269] [100/312] eta: 0:02:48 lr: 0.001504 min_lr: 0.001504 loss: 2.1356 (1.9718) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [269] [110/312] eta: 0:02:38 lr: 0.001504 min_lr: 0.001504 loss: 2.0126 (1.9773) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [269] [120/312] eta: 0:02:29 lr: 0.001503 min_lr: 0.001503 loss: 2.0943 (1.9864) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [269] [130/312] eta: 0:02:20 lr: 0.001503 min_lr: 0.001503 loss: 2.0407 (1.9903) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [269] [140/312] eta: 0:02:11 lr: 0.001503 min_lr: 0.001503 loss: 2.0407 (1.9840) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [269] [150/312] eta: 0:02:03 lr: 0.001502 min_lr: 0.001502 loss: 2.0075 (1.9865) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [269] [160/312] eta: 0:01:55 lr: 0.001502 min_lr: 0.001502 loss: 2.1478 (1.9928) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [269] [170/312] eta: 0:01:46 lr: 0.001501 min_lr: 0.001501 loss: 2.1478 (2.0036) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [269] [180/312] eta: 0:01:39 lr: 0.001501 min_lr: 0.001501 loss: 1.9821 (1.9996) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [269] [190/312] eta: 0:01:31 lr: 0.001500 min_lr: 0.001500 loss: 1.9527 (1.9963) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [269] [200/312] eta: 0:01:23 lr: 0.001500 min_lr: 0.001500 loss: 2.0978 (2.0000) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [269] [210/312] eta: 0:01:15 lr: 0.001499 min_lr: 0.001499 loss: 2.0577 (1.9966) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [269] [220/312] eta: 0:01:08 lr: 0.001499 min_lr: 0.001499 loss: 2.0415 (1.9985) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [269] [230/312] eta: 0:01:00 lr: 0.001498 min_lr: 0.001498 loss: 2.0456 (2.0004) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [269] [240/312] eta: 0:00:53 lr: 0.001498 min_lr: 0.001498 loss: 2.0456 (1.9976) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [269] [250/312] eta: 0:00:45 lr: 0.001498 min_lr: 0.001498 loss: 2.0778 (2.0026) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [269] [260/312] eta: 0:00:38 lr: 0.001497 min_lr: 0.001497 loss: 2.0958 (2.0027) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [269] [270/312] eta: 0:00:30 lr: 0.001497 min_lr: 0.001497 loss: 1.9594 (1.9985) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [269] [280/312] eta: 0:00:23 lr: 0.001496 min_lr: 0.001496 loss: 2.0618 (2.0019) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [269] [290/312] eta: 0:00:16 lr: 0.001496 min_lr: 0.001496 loss: 1.9924 (1.9948) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [269] [300/312] eta: 0:00:08 lr: 0.001495 min_lr: 0.001495 loss: 1.9288 (1.9993) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [269] [310/312] eta: 0:00:01 lr: 0.001495 min_lr: 0.001495 loss: 2.0889 (2.0008) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [269] [311/312] eta: 0:00:00 lr: 0.001495 min_lr: 0.001495 loss: 2.0889 (2.0020) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [269] Total time: 0:03:47 (0.7295 s / it) Averaged stats: lr: 0.001495 min_lr: 0.001495 loss: 2.0889 (1.9887) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5798 (0.5798) acc1: 84.1146 (84.1146) acc5: 97.1354 (97.1354) time: 4.5814 data: 4.3569 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7873 (0.7788) acc1: 80.7292 (79.7760) acc5: 95.8333 (95.4560) time: 0.6605 data: 0.4842 max mem: 64948 Test: Total time: 0:00:06 (0.6879 s / it) * Acc@1 80.304 Acc@5 95.178 loss 0.752 Accuracy of the model on the 50000 test images: 80.3% Max accuracy: 80.39% Test: [0/9] eta: 0:00:44 loss: 0.4898 (0.4898) acc1: 86.7188 (86.7188) acc5: 97.6562 (97.6562) time: 4.9662 data: 4.7540 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6745 (0.6728) acc1: 82.0312 (80.6720) acc5: 97.1354 (96.3520) time: 0.7031 data: 0.5283 max mem: 64948 Test: Total time: 0:00:06 (0.7117 s / it) * Acc@1 82.468 Acc@5 96.268 loss 0.647 Accuracy of the model EMA on 50000 test images: 82.5% Max EMA accuracy: 82.47% Epoch: [270] [ 0/312] eta: 0:55:06 lr: 0.001495 min_lr: 0.001495 loss: 2.0980 (2.0980) weight_decay: 0.0500 (0.0500) time: 10.5975 data: 9.8054 max mem: 64948 Epoch: [270] [ 10/312] eta: 0:08:06 lr: 0.001494 min_lr: 0.001494 loss: 1.8491 (1.9523) weight_decay: 0.0500 (0.0500) time: 1.6100 data: 0.8917 max mem: 64948 Epoch: [270] [ 20/312] eta: 0:05:42 lr: 0.001494 min_lr: 0.001494 loss: 1.9053 (1.9216) weight_decay: 0.0500 (0.0500) time: 0.7027 data: 0.0004 max mem: 64948 Epoch: [270] [ 30/312] eta: 0:04:47 lr: 0.001493 min_lr: 0.001493 loss: 1.9814 (1.9555) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [270] [ 40/312] eta: 0:04:15 lr: 0.001493 min_lr: 0.001493 loss: 2.1181 (1.9772) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [270] [ 50/312] eta: 0:03:54 lr: 0.001492 min_lr: 0.001492 loss: 2.1823 (1.9789) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0003 max mem: 64948 Epoch: [270] [ 60/312] eta: 0:03:36 lr: 0.001492 min_lr: 0.001492 loss: 2.0495 (1.9727) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [270] [ 70/312] eta: 0:03:22 lr: 0.001492 min_lr: 0.001492 loss: 2.0439 (1.9869) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [270] [ 80/312] eta: 0:03:10 lr: 0.001491 min_lr: 0.001491 loss: 2.0425 (1.9864) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [270] [ 90/312] eta: 0:02:58 lr: 0.001491 min_lr: 0.001491 loss: 2.1442 (2.0156) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [270] [100/312] eta: 0:02:48 lr: 0.001490 min_lr: 0.001490 loss: 2.1577 (2.0085) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [270] [110/312] eta: 0:02:38 lr: 0.001490 min_lr: 0.001490 loss: 1.9121 (1.9952) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [270] [120/312] eta: 0:02:29 lr: 0.001489 min_lr: 0.001489 loss: 1.8884 (1.9975) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [270] [130/312] eta: 0:02:20 lr: 0.001489 min_lr: 0.001489 loss: 2.0183 (2.0003) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [270] [140/312] eta: 0:02:11 lr: 0.001488 min_lr: 0.001488 loss: 2.0183 (1.9905) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [270] [150/312] eta: 0:02:03 lr: 0.001488 min_lr: 0.001488 loss: 2.0526 (1.9932) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [270] [160/312] eta: 0:01:55 lr: 0.001488 min_lr: 0.001488 loss: 2.0070 (1.9875) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [270] [170/312] eta: 0:01:47 lr: 0.001487 min_lr: 0.001487 loss: 1.9138 (1.9863) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [270] [180/312] eta: 0:01:39 lr: 0.001487 min_lr: 0.001487 loss: 2.1644 (1.9953) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [270] [190/312] eta: 0:01:31 lr: 0.001486 min_lr: 0.001486 loss: 2.0856 (1.9856) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [270] [200/312] eta: 0:01:23 lr: 0.001486 min_lr: 0.001486 loss: 1.9236 (1.9806) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [270] [210/312] eta: 0:01:15 lr: 0.001485 min_lr: 0.001485 loss: 1.9522 (1.9769) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [270] [220/312] eta: 0:01:08 lr: 0.001485 min_lr: 0.001485 loss: 1.9987 (1.9779) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [270] [230/312] eta: 0:01:00 lr: 0.001484 min_lr: 0.001484 loss: 2.0860 (1.9794) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [270] [240/312] eta: 0:00:53 lr: 0.001484 min_lr: 0.001484 loss: 2.1045 (1.9817) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [270] [250/312] eta: 0:00:45 lr: 0.001483 min_lr: 0.001483 loss: 2.0942 (1.9833) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [270] [260/312] eta: 0:00:38 lr: 0.001483 min_lr: 0.001483 loss: 2.1394 (1.9929) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [270] [270/312] eta: 0:00:30 lr: 0.001483 min_lr: 0.001483 loss: 2.1534 (1.9961) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [270] [280/312] eta: 0:00:23 lr: 0.001482 min_lr: 0.001482 loss: 1.9045 (1.9846) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0010 max mem: 64948 Epoch: [270] [290/312] eta: 0:00:16 lr: 0.001482 min_lr: 0.001482 loss: 1.8330 (1.9879) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [270] [300/312] eta: 0:00:08 lr: 0.001481 min_lr: 0.001481 loss: 2.0119 (1.9844) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [270] [310/312] eta: 0:00:01 lr: 0.001481 min_lr: 0.001481 loss: 1.9192 (1.9838) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [270] [311/312] eta: 0:00:00 lr: 0.001481 min_lr: 0.001481 loss: 1.9192 (1.9831) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [270] Total time: 0:03:47 (0.7298 s / it) Averaged stats: lr: 0.001481 min_lr: 0.001481 loss: 1.9192 (1.9838) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5431 (0.5431) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.5234 data: 4.3079 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7338 (0.7639) acc1: 79.9479 (79.5200) acc5: 95.5729 (95.4560) time: 0.6546 data: 0.4787 max mem: 64948 Test: Total time: 0:00:06 (0.6776 s / it) * Acc@1 80.578 Acc@5 95.278 loss 0.747 Accuracy of the model on the 50000 test images: 80.6% Max accuracy: 80.58% Test: [0/9] eta: 0:00:39 loss: 0.4890 (0.4890) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 4.3866 data: 4.1712 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6733 (0.6723) acc1: 82.0312 (80.6720) acc5: 97.3958 (96.3840) time: 0.6387 data: 0.4636 max mem: 64948 Test: Total time: 0:00:05 (0.6476 s / it) * Acc@1 82.468 Acc@5 96.278 loss 0.647 Accuracy of the model EMA on 50000 test images: 82.5% Epoch: [271] [ 0/312] eta: 0:51:39 lr: 0.001481 min_lr: 0.001481 loss: 1.5783 (1.5783) weight_decay: 0.0500 (0.0500) time: 9.9331 data: 7.2618 max mem: 64948 Epoch: [271] [ 10/312] eta: 0:08:00 lr: 0.001480 min_lr: 0.001480 loss: 1.9804 (1.9634) weight_decay: 0.0500 (0.0500) time: 1.5922 data: 0.6607 max mem: 64948 Epoch: [271] [ 20/312] eta: 0:05:40 lr: 0.001480 min_lr: 0.001480 loss: 2.1347 (1.9962) weight_decay: 0.0500 (0.0500) time: 0.7283 data: 0.0004 max mem: 64948 Epoch: [271] [ 30/312] eta: 0:04:46 lr: 0.001479 min_lr: 0.001479 loss: 2.1719 (1.9949) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0003 max mem: 64948 Epoch: [271] [ 40/312] eta: 0:04:15 lr: 0.001479 min_lr: 0.001479 loss: 2.1176 (1.9803) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [271] [ 50/312] eta: 0:03:53 lr: 0.001478 min_lr: 0.001478 loss: 2.1244 (2.0025) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [271] [ 60/312] eta: 0:03:36 lr: 0.001478 min_lr: 0.001478 loss: 2.0233 (1.9981) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [271] [ 70/312] eta: 0:03:22 lr: 0.001477 min_lr: 0.001477 loss: 1.9303 (1.9759) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [271] [ 80/312] eta: 0:03:10 lr: 0.001477 min_lr: 0.001477 loss: 2.1133 (1.9892) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [271] [ 90/312] eta: 0:02:58 lr: 0.001477 min_lr: 0.001477 loss: 2.1740 (1.9983) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [271] [100/312] eta: 0:02:48 lr: 0.001476 min_lr: 0.001476 loss: 2.0253 (1.9835) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [271] [110/312] eta: 0:02:38 lr: 0.001476 min_lr: 0.001476 loss: 2.0193 (1.9917) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [271] [120/312] eta: 0:02:29 lr: 0.001475 min_lr: 0.001475 loss: 2.0769 (2.0002) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [271] [130/312] eta: 0:02:20 lr: 0.001475 min_lr: 0.001475 loss: 2.1911 (2.0228) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [271] [140/312] eta: 0:02:11 lr: 0.001474 min_lr: 0.001474 loss: 2.1911 (2.0274) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [271] [150/312] eta: 0:02:03 lr: 0.001474 min_lr: 0.001474 loss: 2.1416 (2.0272) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [271] [160/312] eta: 0:01:55 lr: 0.001473 min_lr: 0.001473 loss: 2.1416 (2.0278) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [271] [170/312] eta: 0:01:47 lr: 0.001473 min_lr: 0.001473 loss: 2.2196 (2.0340) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [271] [180/312] eta: 0:01:39 lr: 0.001473 min_lr: 0.001473 loss: 1.9244 (2.0192) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [271] [190/312] eta: 0:01:31 lr: 0.001472 min_lr: 0.001472 loss: 1.9113 (2.0202) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [271] [200/312] eta: 0:01:23 lr: 0.001472 min_lr: 0.001472 loss: 2.0382 (2.0170) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [271] [210/312] eta: 0:01:15 lr: 0.001471 min_lr: 0.001471 loss: 1.9964 (2.0124) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [271] [220/312] eta: 0:01:08 lr: 0.001471 min_lr: 0.001471 loss: 1.8569 (2.0083) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [271] [230/312] eta: 0:01:00 lr: 0.001470 min_lr: 0.001470 loss: 2.0684 (2.0049) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [271] [240/312] eta: 0:00:53 lr: 0.001470 min_lr: 0.001470 loss: 2.0601 (2.0057) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [271] [250/312] eta: 0:00:45 lr: 0.001469 min_lr: 0.001469 loss: 2.0601 (2.0066) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [271] [260/312] eta: 0:00:38 lr: 0.001469 min_lr: 0.001469 loss: 2.1545 (2.0107) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [271] [270/312] eta: 0:00:30 lr: 0.001468 min_lr: 0.001468 loss: 2.1545 (2.0097) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [271] [280/312] eta: 0:00:23 lr: 0.001468 min_lr: 0.001468 loss: 1.9653 (2.0069) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [271] [290/312] eta: 0:00:16 lr: 0.001468 min_lr: 0.001468 loss: 1.9660 (2.0111) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [271] [300/312] eta: 0:00:08 lr: 0.001467 min_lr: 0.001467 loss: 2.0899 (2.0130) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [271] [310/312] eta: 0:00:01 lr: 0.001467 min_lr: 0.001467 loss: 2.0775 (2.0148) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [271] [311/312] eta: 0:00:00 lr: 0.001467 min_lr: 0.001467 loss: 2.0775 (2.0151) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [271] Total time: 0:03:47 (0.7294 s / it) Averaged stats: lr: 0.001467 min_lr: 0.001467 loss: 2.0775 (1.9897) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5858 (0.5858) acc1: 85.1562 (85.1562) acc5: 96.0938 (96.0938) time: 4.6491 data: 4.4312 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7960 (0.7836) acc1: 79.9479 (79.3280) acc5: 94.7917 (94.5280) time: 0.6679 data: 0.4924 max mem: 64948 Test: Total time: 0:00:06 (0.6912 s / it) * Acc@1 80.280 Acc@5 95.060 loss 0.758 Accuracy of the model on the 50000 test images: 80.3% Max accuracy: 80.58% Test: [0/9] eta: 0:00:45 loss: 0.4885 (0.4885) acc1: 86.7188 (86.7188) acc5: 97.6562 (97.6562) time: 5.0802 data: 4.8623 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6721 (0.6719) acc1: 81.7708 (80.6720) acc5: 97.3958 (96.4480) time: 0.7164 data: 0.5403 max mem: 64948 Test: Total time: 0:00:06 (0.7261 s / it) * Acc@1 82.490 Acc@5 96.294 loss 0.646 Accuracy of the model EMA on 50000 test images: 82.5% Max EMA accuracy: 82.49% Epoch: [272] [ 0/312] eta: 0:53:32 lr: 0.001467 min_lr: 0.001467 loss: 2.0061 (2.0061) weight_decay: 0.0500 (0.0500) time: 10.2960 data: 9.5036 max mem: 64948 Epoch: [272] [ 10/312] eta: 0:07:58 lr: 0.001466 min_lr: 0.001466 loss: 1.6650 (1.7392) weight_decay: 0.0500 (0.0500) time: 1.5851 data: 0.8643 max mem: 64948 Epoch: [272] [ 20/312] eta: 0:05:39 lr: 0.001466 min_lr: 0.001466 loss: 1.8256 (1.8843) weight_decay: 0.0500 (0.0500) time: 0.7050 data: 0.0004 max mem: 64948 Epoch: [272] [ 30/312] eta: 0:04:45 lr: 0.001465 min_lr: 0.001465 loss: 1.9692 (1.9096) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [272] [ 40/312] eta: 0:04:14 lr: 0.001465 min_lr: 0.001465 loss: 1.9318 (1.8971) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [272] [ 50/312] eta: 0:03:52 lr: 0.001464 min_lr: 0.001464 loss: 2.0092 (1.9481) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [272] [ 60/312] eta: 0:03:35 lr: 0.001464 min_lr: 0.001464 loss: 2.1410 (1.9684) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [272] [ 70/312] eta: 0:03:21 lr: 0.001463 min_lr: 0.001463 loss: 2.1027 (1.9732) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0005 max mem: 64948 Epoch: [272] [ 80/312] eta: 0:03:09 lr: 0.001463 min_lr: 0.001463 loss: 2.0921 (1.9819) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [272] [ 90/312] eta: 0:02:58 lr: 0.001462 min_lr: 0.001462 loss: 2.1160 (1.9871) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [272] [100/312] eta: 0:02:47 lr: 0.001462 min_lr: 0.001462 loss: 2.1045 (1.9808) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [272] [110/312] eta: 0:02:38 lr: 0.001462 min_lr: 0.001462 loss: 2.0613 (1.9874) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [272] [120/312] eta: 0:02:28 lr: 0.001461 min_lr: 0.001461 loss: 2.0537 (1.9896) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [272] [130/312] eta: 0:02:20 lr: 0.001461 min_lr: 0.001461 loss: 1.8983 (1.9763) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [272] [140/312] eta: 0:02:11 lr: 0.001460 min_lr: 0.001460 loss: 1.9489 (1.9832) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [272] [150/312] eta: 0:02:03 lr: 0.001460 min_lr: 0.001460 loss: 2.0243 (1.9697) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [272] [160/312] eta: 0:01:54 lr: 0.001459 min_lr: 0.001459 loss: 1.8865 (1.9726) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [272] [170/312] eta: 0:01:46 lr: 0.001459 min_lr: 0.001459 loss: 2.0742 (1.9731) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [272] [180/312] eta: 0:01:38 lr: 0.001458 min_lr: 0.001458 loss: 2.0365 (1.9761) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [272] [190/312] eta: 0:01:31 lr: 0.001458 min_lr: 0.001458 loss: 1.9577 (1.9689) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [272] [200/312] eta: 0:01:23 lr: 0.001458 min_lr: 0.001458 loss: 1.8132 (1.9626) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [272] [210/312] eta: 0:01:15 lr: 0.001457 min_lr: 0.001457 loss: 2.0495 (1.9654) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [272] [220/312] eta: 0:01:08 lr: 0.001457 min_lr: 0.001457 loss: 2.0559 (1.9601) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [272] [230/312] eta: 0:01:00 lr: 0.001456 min_lr: 0.001456 loss: 1.8942 (1.9581) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [272] [240/312] eta: 0:00:52 lr: 0.001456 min_lr: 0.001456 loss: 2.1363 (1.9714) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [272] [250/312] eta: 0:00:45 lr: 0.001455 min_lr: 0.001455 loss: 2.2010 (1.9748) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [272] [260/312] eta: 0:00:38 lr: 0.001455 min_lr: 0.001455 loss: 1.8660 (1.9661) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [272] [270/312] eta: 0:00:30 lr: 0.001454 min_lr: 0.001454 loss: 1.8599 (1.9701) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [272] [280/312] eta: 0:00:23 lr: 0.001454 min_lr: 0.001454 loss: 2.0787 (1.9775) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [272] [290/312] eta: 0:00:16 lr: 0.001453 min_lr: 0.001453 loss: 2.0643 (1.9746) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [272] [300/312] eta: 0:00:08 lr: 0.001453 min_lr: 0.001453 loss: 2.0527 (1.9807) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [272] [310/312] eta: 0:00:01 lr: 0.001453 min_lr: 0.001453 loss: 2.1474 (1.9839) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [272] [311/312] eta: 0:00:00 lr: 0.001453 min_lr: 0.001453 loss: 2.1474 (1.9854) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [272] Total time: 0:03:47 (0.7288 s / it) Averaged stats: lr: 0.001453 min_lr: 0.001453 loss: 2.1474 (1.9834) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.5359 (0.5359) acc1: 86.9792 (86.9792) acc5: 96.3542 (96.3542) time: 4.8232 data: 4.6141 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7959 (0.7533) acc1: 79.1667 (79.4240) acc5: 96.0938 (95.4880) time: 0.6872 data: 0.5128 max mem: 64948 Test: Total time: 0:00:06 (0.7125 s / it) * Acc@1 80.476 Acc@5 95.178 loss 0.746 Accuracy of the model on the 50000 test images: 80.5% Max accuracy: 80.58% Test: [0/9] eta: 0:00:43 loss: 0.4884 (0.4884) acc1: 86.7188 (86.7188) acc5: 97.9167 (97.9167) time: 4.8476 data: 4.6295 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6713 (0.6716) acc1: 81.7708 (80.7360) acc5: 97.3958 (96.5120) time: 0.6901 data: 0.5145 max mem: 64948 Test: Total time: 0:00:06 (0.6981 s / it) * Acc@1 82.512 Acc@5 96.310 loss 0.646 Accuracy of the model EMA on 50000 test images: 82.5% Max EMA accuracy: 82.51% Epoch: [273] [ 0/312] eta: 0:46:56 lr: 0.001452 min_lr: 0.001452 loss: 2.3307 (2.3307) weight_decay: 0.0500 (0.0500) time: 9.0270 data: 8.2486 max mem: 64948 Epoch: [273] [ 10/312] eta: 0:07:39 lr: 0.001452 min_lr: 0.001452 loss: 1.9675 (1.9832) weight_decay: 0.0500 (0.0500) time: 1.5204 data: 0.7541 max mem: 64948 Epoch: [273] [ 20/312] eta: 0:05:29 lr: 0.001452 min_lr: 0.001452 loss: 1.8846 (1.8815) weight_decay: 0.0500 (0.0500) time: 0.7344 data: 0.0025 max mem: 64948 Epoch: [273] [ 30/312] eta: 0:04:38 lr: 0.001451 min_lr: 0.001451 loss: 1.8997 (1.9059) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [273] [ 40/312] eta: 0:04:09 lr: 0.001451 min_lr: 0.001451 loss: 2.0237 (1.9290) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [273] [ 50/312] eta: 0:03:49 lr: 0.001450 min_lr: 0.001450 loss: 2.0947 (1.9639) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [273] [ 60/312] eta: 0:03:32 lr: 0.001450 min_lr: 0.001450 loss: 2.0146 (1.9627) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [273] [ 70/312] eta: 0:03:19 lr: 0.001449 min_lr: 0.001449 loss: 1.9049 (1.9695) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [273] [ 80/312] eta: 0:03:07 lr: 0.001449 min_lr: 0.001449 loss: 1.8733 (1.9604) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [273] [ 90/312] eta: 0:02:56 lr: 0.001448 min_lr: 0.001448 loss: 1.9273 (1.9580) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [273] [100/312] eta: 0:02:46 lr: 0.001448 min_lr: 0.001448 loss: 1.9800 (1.9489) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [273] [110/312] eta: 0:02:37 lr: 0.001448 min_lr: 0.001448 loss: 1.9800 (1.9438) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [273] [120/312] eta: 0:02:28 lr: 0.001447 min_lr: 0.001447 loss: 1.9733 (1.9477) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [273] [130/312] eta: 0:02:19 lr: 0.001447 min_lr: 0.001447 loss: 1.9202 (1.9361) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [273] [140/312] eta: 0:02:10 lr: 0.001446 min_lr: 0.001446 loss: 1.8389 (1.9439) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [273] [150/312] eta: 0:02:02 lr: 0.001446 min_lr: 0.001446 loss: 2.0130 (1.9458) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [273] [160/312] eta: 0:01:54 lr: 0.001445 min_lr: 0.001445 loss: 2.0088 (1.9504) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [273] [170/312] eta: 0:01:46 lr: 0.001445 min_lr: 0.001445 loss: 1.9986 (1.9504) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [273] [180/312] eta: 0:01:38 lr: 0.001444 min_lr: 0.001444 loss: 2.1515 (1.9605) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [273] [190/312] eta: 0:01:30 lr: 0.001444 min_lr: 0.001444 loss: 2.1638 (1.9688) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [273] [200/312] eta: 0:01:23 lr: 0.001443 min_lr: 0.001443 loss: 2.0466 (1.9630) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [273] [210/312] eta: 0:01:15 lr: 0.001443 min_lr: 0.001443 loss: 1.8693 (1.9654) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [273] [220/312] eta: 0:01:07 lr: 0.001443 min_lr: 0.001443 loss: 2.0189 (1.9646) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [273] [230/312] eta: 0:01:00 lr: 0.001442 min_lr: 0.001442 loss: 2.0189 (1.9670) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [273] [240/312] eta: 0:00:52 lr: 0.001442 min_lr: 0.001442 loss: 1.9785 (1.9625) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [273] [250/312] eta: 0:00:45 lr: 0.001441 min_lr: 0.001441 loss: 1.9785 (1.9645) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [273] [260/312] eta: 0:00:37 lr: 0.001441 min_lr: 0.001441 loss: 2.2356 (1.9724) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [273] [270/312] eta: 0:00:30 lr: 0.001440 min_lr: 0.001440 loss: 2.2828 (1.9765) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [273] [280/312] eta: 0:00:23 lr: 0.001440 min_lr: 0.001440 loss: 2.0907 (1.9791) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [273] [290/312] eta: 0:00:15 lr: 0.001439 min_lr: 0.001439 loss: 2.0907 (1.9811) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [273] [300/312] eta: 0:00:08 lr: 0.001439 min_lr: 0.001439 loss: 2.0356 (1.9810) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [273] [310/312] eta: 0:00:01 lr: 0.001439 min_lr: 0.001439 loss: 1.9658 (1.9787) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [273] [311/312] eta: 0:00:00 lr: 0.001438 min_lr: 0.001438 loss: 1.9674 (1.9788) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [273] Total time: 0:03:46 (0.7271 s / it) Averaged stats: lr: 0.001438 min_lr: 0.001438 loss: 1.9674 (1.9849) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5586 (0.5586) acc1: 85.9375 (85.9375) acc5: 97.1354 (97.1354) time: 4.7362 data: 4.5239 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7480 (0.7545) acc1: 79.6875 (79.3600) acc5: 95.3125 (95.2320) time: 0.6780 data: 0.5027 max mem: 64948 Test: Total time: 0:00:06 (0.7003 s / it) * Acc@1 80.254 Acc@5 95.292 loss 0.750 Accuracy of the model on the 50000 test images: 80.3% Max accuracy: 80.58% Test: [0/9] eta: 0:00:43 loss: 0.4879 (0.4879) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 4.8539 data: 4.6433 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6704 (0.6711) acc1: 81.7708 (80.7680) acc5: 97.3958 (96.5440) time: 0.6906 data: 0.5160 max mem: 64948 Test: Total time: 0:00:06 (0.7035 s / it) * Acc@1 82.530 Acc@5 96.298 loss 0.646 Accuracy of the model EMA on 50000 test images: 82.5% Max EMA accuracy: 82.53% Epoch: [274] [ 0/312] eta: 0:51:17 lr: 0.001438 min_lr: 0.001438 loss: 2.2722 (2.2722) weight_decay: 0.0500 (0.0500) time: 9.8634 data: 7.9961 max mem: 64948 Epoch: [274] [ 10/312] eta: 0:07:47 lr: 0.001438 min_lr: 0.001438 loss: 2.0666 (1.9969) weight_decay: 0.0500 (0.0500) time: 1.5496 data: 0.7273 max mem: 64948 Epoch: [274] [ 20/312] eta: 0:05:34 lr: 0.001438 min_lr: 0.001438 loss: 1.9838 (1.8851) weight_decay: 0.0500 (0.0500) time: 0.7087 data: 0.0004 max mem: 64948 Epoch: [274] [ 30/312] eta: 0:04:42 lr: 0.001437 min_lr: 0.001437 loss: 1.8286 (1.8923) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0003 max mem: 64948 Epoch: [274] [ 40/312] eta: 0:04:12 lr: 0.001437 min_lr: 0.001437 loss: 2.0386 (1.9296) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [274] [ 50/312] eta: 0:03:50 lr: 0.001436 min_lr: 0.001436 loss: 2.0386 (1.9328) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [274] [ 60/312] eta: 0:03:34 lr: 0.001436 min_lr: 0.001436 loss: 2.0034 (1.9384) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [274] [ 70/312] eta: 0:03:20 lr: 0.001435 min_lr: 0.001435 loss: 2.0758 (1.9588) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [274] [ 80/312] eta: 0:03:08 lr: 0.001435 min_lr: 0.001435 loss: 2.1878 (1.9852) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [274] [ 90/312] eta: 0:02:57 lr: 0.001434 min_lr: 0.001434 loss: 2.1958 (1.9902) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [274] [100/312] eta: 0:02:47 lr: 0.001434 min_lr: 0.001434 loss: 1.9860 (1.9911) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [274] [110/312] eta: 0:02:37 lr: 0.001434 min_lr: 0.001434 loss: 1.9860 (1.9958) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [274] [120/312] eta: 0:02:28 lr: 0.001433 min_lr: 0.001433 loss: 2.0618 (1.9905) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [274] [130/312] eta: 0:02:19 lr: 0.001433 min_lr: 0.001433 loss: 2.1107 (1.9820) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [274] [140/312] eta: 0:02:11 lr: 0.001432 min_lr: 0.001432 loss: 2.1574 (1.9916) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [274] [150/312] eta: 0:02:02 lr: 0.001432 min_lr: 0.001432 loss: 2.0493 (1.9854) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [274] [160/312] eta: 0:01:54 lr: 0.001431 min_lr: 0.001431 loss: 1.9885 (1.9889) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [274] [170/312] eta: 0:01:46 lr: 0.001431 min_lr: 0.001431 loss: 2.0451 (1.9909) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [274] [180/312] eta: 0:01:38 lr: 0.001430 min_lr: 0.001430 loss: 2.0614 (1.9846) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [274] [190/312] eta: 0:01:30 lr: 0.001430 min_lr: 0.001430 loss: 1.9128 (1.9793) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [274] [200/312] eta: 0:01:23 lr: 0.001429 min_lr: 0.001429 loss: 1.9292 (1.9810) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [274] [210/312] eta: 0:01:15 lr: 0.001429 min_lr: 0.001429 loss: 2.0593 (1.9852) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [274] [220/312] eta: 0:01:07 lr: 0.001429 min_lr: 0.001429 loss: 2.0694 (1.9844) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [274] [230/312] eta: 0:01:00 lr: 0.001428 min_lr: 0.001428 loss: 2.1244 (1.9898) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [274] [240/312] eta: 0:00:52 lr: 0.001428 min_lr: 0.001428 loss: 2.1244 (1.9901) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [274] [250/312] eta: 0:00:45 lr: 0.001427 min_lr: 0.001427 loss: 1.9772 (1.9871) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [274] [260/312] eta: 0:00:38 lr: 0.001427 min_lr: 0.001427 loss: 2.1017 (1.9893) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [274] [270/312] eta: 0:00:30 lr: 0.001426 min_lr: 0.001426 loss: 2.1177 (1.9926) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [274] [280/312] eta: 0:00:23 lr: 0.001426 min_lr: 0.001426 loss: 2.0208 (1.9925) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [274] [290/312] eta: 0:00:16 lr: 0.001425 min_lr: 0.001425 loss: 2.0409 (1.9938) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [274] [300/312] eta: 0:00:08 lr: 0.001425 min_lr: 0.001425 loss: 2.1022 (1.9933) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [274] [310/312] eta: 0:00:01 lr: 0.001425 min_lr: 0.001425 loss: 1.8725 (1.9894) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [274] [311/312] eta: 0:00:00 lr: 0.001424 min_lr: 0.001424 loss: 1.8725 (1.9895) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [274] Total time: 0:03:47 (0.7278 s / it) Averaged stats: lr: 0.001424 min_lr: 0.001424 loss: 1.8725 (1.9791) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5677 (0.5677) acc1: 85.4167 (85.4167) acc5: 96.0938 (96.0938) time: 4.3845 data: 4.1647 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8058 (0.7469) acc1: 80.9896 (79.8400) acc5: 95.8333 (95.2640) time: 0.6385 data: 0.4628 max mem: 64948 Test: Total time: 0:00:05 (0.6637 s / it) * Acc@1 80.354 Acc@5 95.268 loss 0.745 Accuracy of the model on the 50000 test images: 80.4% Max accuracy: 80.58% Test: [0/9] eta: 0:00:44 loss: 0.4872 (0.4872) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 4.9592 data: 4.7441 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6698 (0.6707) acc1: 81.5104 (80.6720) acc5: 97.3958 (96.5440) time: 0.7024 data: 0.5272 max mem: 64948 Test: Total time: 0:00:06 (0.7131 s / it) * Acc@1 82.560 Acc@5 96.286 loss 0.645 Accuracy of the model EMA on 50000 test images: 82.6% Max EMA accuracy: 82.56% Epoch: [275] [ 0/312] eta: 0:50:56 lr: 0.001424 min_lr: 0.001424 loss: 1.3945 (1.3945) weight_decay: 0.0500 (0.0500) time: 9.7964 data: 7.0046 max mem: 64948 Epoch: [275] [ 10/312] eta: 0:07:54 lr: 0.001424 min_lr: 0.001424 loss: 1.6892 (1.7685) weight_decay: 0.0500 (0.0500) time: 1.5718 data: 0.6372 max mem: 64948 Epoch: [275] [ 20/312] eta: 0:05:36 lr: 0.001424 min_lr: 0.001424 loss: 2.0651 (1.9430) weight_decay: 0.0500 (0.0500) time: 0.7206 data: 0.0004 max mem: 64948 Epoch: [275] [ 30/312] eta: 0:04:43 lr: 0.001423 min_lr: 0.001423 loss: 2.0821 (1.9656) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0004 max mem: 64948 Epoch: [275] [ 40/312] eta: 0:04:12 lr: 0.001423 min_lr: 0.001423 loss: 1.9598 (1.9409) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [275] [ 50/312] eta: 0:03:51 lr: 0.001422 min_lr: 0.001422 loss: 1.9929 (1.9948) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [275] [ 60/312] eta: 0:03:34 lr: 0.001422 min_lr: 0.001422 loss: 2.2080 (2.0142) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [275] [ 70/312] eta: 0:03:20 lr: 0.001421 min_lr: 0.001421 loss: 2.0878 (2.0092) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [275] [ 80/312] eta: 0:03:08 lr: 0.001421 min_lr: 0.001421 loss: 2.0616 (2.0005) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [275] [ 90/312] eta: 0:02:57 lr: 0.001420 min_lr: 0.001420 loss: 1.8048 (1.9784) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [275] [100/312] eta: 0:02:47 lr: 0.001420 min_lr: 0.001420 loss: 1.8022 (1.9650) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [275] [110/312] eta: 0:02:37 lr: 0.001420 min_lr: 0.001420 loss: 1.8700 (1.9707) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [275] [120/312] eta: 0:02:28 lr: 0.001419 min_lr: 0.001419 loss: 1.9921 (1.9691) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [275] [130/312] eta: 0:02:19 lr: 0.001419 min_lr: 0.001419 loss: 1.8784 (1.9566) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [275] [140/312] eta: 0:02:11 lr: 0.001418 min_lr: 0.001418 loss: 1.9316 (1.9568) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [275] [150/312] eta: 0:02:02 lr: 0.001418 min_lr: 0.001418 loss: 2.0520 (1.9623) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [275] [160/312] eta: 0:01:54 lr: 0.001417 min_lr: 0.001417 loss: 2.0994 (1.9674) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [275] [170/312] eta: 0:01:46 lr: 0.001417 min_lr: 0.001417 loss: 1.9127 (1.9653) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [275] [180/312] eta: 0:01:38 lr: 0.001416 min_lr: 0.001416 loss: 2.0442 (1.9721) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [275] [190/312] eta: 0:01:30 lr: 0.001416 min_lr: 0.001416 loss: 2.1471 (1.9759) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [275] [200/312] eta: 0:01:23 lr: 0.001415 min_lr: 0.001415 loss: 2.1183 (1.9749) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [275] [210/312] eta: 0:01:15 lr: 0.001415 min_lr: 0.001415 loss: 2.1348 (1.9804) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [275] [220/312] eta: 0:01:07 lr: 0.001415 min_lr: 0.001415 loss: 2.0398 (1.9829) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [275] [230/312] eta: 0:01:00 lr: 0.001414 min_lr: 0.001414 loss: 1.9588 (1.9764) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [275] [240/312] eta: 0:00:52 lr: 0.001414 min_lr: 0.001414 loss: 1.9588 (1.9748) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [275] [250/312] eta: 0:00:45 lr: 0.001413 min_lr: 0.001413 loss: 1.8390 (1.9643) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [275] [260/312] eta: 0:00:38 lr: 0.001413 min_lr: 0.001413 loss: 1.8390 (1.9632) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [275] [270/312] eta: 0:00:30 lr: 0.001412 min_lr: 0.001412 loss: 2.0508 (1.9650) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [275] [280/312] eta: 0:00:23 lr: 0.001412 min_lr: 0.001412 loss: 1.9592 (1.9615) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [275] [290/312] eta: 0:00:16 lr: 0.001411 min_lr: 0.001411 loss: 1.9551 (1.9618) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [275] [300/312] eta: 0:00:08 lr: 0.001411 min_lr: 0.001411 loss: 1.9769 (1.9618) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [275] [310/312] eta: 0:00:01 lr: 0.001411 min_lr: 0.001411 loss: 2.0141 (1.9609) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [275] [311/312] eta: 0:00:00 lr: 0.001411 min_lr: 0.001411 loss: 1.9964 (1.9591) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [275] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.001411 min_lr: 0.001411 loss: 1.9964 (1.9737) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5320 (0.5320) acc1: 86.7188 (86.7188) acc5: 96.6146 (96.6146) time: 4.6540 data: 4.4347 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7992 (0.7586) acc1: 80.9896 (80.1280) acc5: 95.8333 (95.2960) time: 0.6689 data: 0.4928 max mem: 64948 Test: Total time: 0:00:06 (0.6889 s / it) * Acc@1 80.442 Acc@5 95.342 loss 0.744 Accuracy of the model on the 50000 test images: 80.4% Max accuracy: 80.58% Test: [0/9] eta: 0:00:45 loss: 0.4856 (0.4856) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 5.0244 data: 4.8156 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6688 (0.6702) acc1: 81.7708 (80.7360) acc5: 97.3958 (96.4800) time: 0.7095 data: 0.5352 max mem: 64948 Test: Total time: 0:00:06 (0.7180 s / it) * Acc@1 82.558 Acc@5 96.284 loss 0.645 Accuracy of the model EMA on 50000 test images: 82.6% Epoch: [276] [ 0/312] eta: 0:55:18 lr: 0.001410 min_lr: 0.001410 loss: 2.3524 (2.3524) weight_decay: 0.0500 (0.0500) time: 10.6362 data: 7.9940 max mem: 64948 Epoch: [276] [ 10/312] eta: 0:08:12 lr: 0.001410 min_lr: 0.001410 loss: 2.0115 (2.0146) weight_decay: 0.0500 (0.0500) time: 1.6306 data: 0.7272 max mem: 64948 Epoch: [276] [ 20/312] eta: 0:05:45 lr: 0.001410 min_lr: 0.001410 loss: 2.0064 (1.9990) weight_decay: 0.0500 (0.0500) time: 0.7121 data: 0.0004 max mem: 64948 Epoch: [276] [ 30/312] eta: 0:04:49 lr: 0.001409 min_lr: 0.001409 loss: 1.7894 (1.9626) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [276] [ 40/312] eta: 0:04:17 lr: 0.001409 min_lr: 0.001409 loss: 1.8766 (1.9565) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [276] [ 50/312] eta: 0:03:55 lr: 0.001408 min_lr: 0.001408 loss: 2.0799 (1.9595) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [276] [ 60/312] eta: 0:03:37 lr: 0.001408 min_lr: 0.001408 loss: 1.8266 (1.9298) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [276] [ 70/312] eta: 0:03:23 lr: 0.001407 min_lr: 0.001407 loss: 1.8806 (1.9494) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [276] [ 80/312] eta: 0:03:11 lr: 0.001407 min_lr: 0.001407 loss: 2.0410 (1.9408) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0005 max mem: 64948 Epoch: [276] [ 90/312] eta: 0:02:59 lr: 0.001406 min_lr: 0.001406 loss: 2.0086 (1.9606) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0005 max mem: 64948 Epoch: [276] [100/312] eta: 0:02:49 lr: 0.001406 min_lr: 0.001406 loss: 1.9863 (1.9523) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [276] [110/312] eta: 0:02:39 lr: 0.001406 min_lr: 0.001406 loss: 1.9131 (1.9432) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [276] [120/312] eta: 0:02:29 lr: 0.001405 min_lr: 0.001405 loss: 1.7724 (1.9357) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [276] [130/312] eta: 0:02:20 lr: 0.001405 min_lr: 0.001405 loss: 1.8480 (1.9339) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [276] [140/312] eta: 0:02:12 lr: 0.001404 min_lr: 0.001404 loss: 1.8480 (1.9276) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [276] [150/312] eta: 0:02:03 lr: 0.001404 min_lr: 0.001404 loss: 2.0864 (1.9455) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [276] [160/312] eta: 0:01:55 lr: 0.001403 min_lr: 0.001403 loss: 2.0864 (1.9312) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [276] [170/312] eta: 0:01:47 lr: 0.001403 min_lr: 0.001403 loss: 1.7016 (1.9258) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [276] [180/312] eta: 0:01:39 lr: 0.001402 min_lr: 0.001402 loss: 2.0041 (1.9261) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [276] [190/312] eta: 0:01:31 lr: 0.001402 min_lr: 0.001402 loss: 2.0063 (1.9336) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [276] [200/312] eta: 0:01:23 lr: 0.001402 min_lr: 0.001402 loss: 2.1094 (1.9462) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [276] [210/312] eta: 0:01:15 lr: 0.001401 min_lr: 0.001401 loss: 2.1711 (1.9504) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [276] [220/312] eta: 0:01:08 lr: 0.001401 min_lr: 0.001401 loss: 2.1362 (1.9542) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [276] [230/312] eta: 0:01:00 lr: 0.001400 min_lr: 0.001400 loss: 2.1059 (1.9581) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [276] [240/312] eta: 0:00:53 lr: 0.001400 min_lr: 0.001400 loss: 2.0643 (1.9622) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [276] [250/312] eta: 0:00:45 lr: 0.001399 min_lr: 0.001399 loss: 1.8780 (1.9534) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [276] [260/312] eta: 0:00:38 lr: 0.001399 min_lr: 0.001399 loss: 1.8165 (1.9574) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [276] [270/312] eta: 0:00:30 lr: 0.001398 min_lr: 0.001398 loss: 2.1170 (1.9585) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [276] [280/312] eta: 0:00:23 lr: 0.001398 min_lr: 0.001398 loss: 2.1104 (1.9628) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0010 max mem: 64948 Epoch: [276] [290/312] eta: 0:00:16 lr: 0.001398 min_lr: 0.001398 loss: 2.0463 (1.9614) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [276] [300/312] eta: 0:00:08 lr: 0.001397 min_lr: 0.001397 loss: 1.8914 (1.9587) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [276] [310/312] eta: 0:00:01 lr: 0.001397 min_lr: 0.001397 loss: 1.9820 (1.9650) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [276] [311/312] eta: 0:00:00 lr: 0.001397 min_lr: 0.001397 loss: 2.1274 (1.9659) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [276] Total time: 0:03:48 (0.7315 s / it) Averaged stats: lr: 0.001397 min_lr: 0.001397 loss: 2.1274 (1.9822) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5297 (0.5297) acc1: 86.4583 (86.4583) acc5: 96.8750 (96.8750) time: 4.5415 data: 4.3214 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7662 (0.7658) acc1: 79.1667 (79.4240) acc5: 95.5729 (95.2320) time: 0.6566 data: 0.4803 max mem: 64948 Test: Total time: 0:00:06 (0.6807 s / it) * Acc@1 80.636 Acc@5 95.264 loss 0.744 Accuracy of the model on the 50000 test images: 80.6% Max accuracy: 80.64% Test: [0/9] eta: 0:00:43 loss: 0.4845 (0.4845) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 4.7889 data: 4.5712 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6681 (0.6698) acc1: 81.7708 (80.8000) acc5: 97.3958 (96.4800) time: 0.6839 data: 0.5080 max mem: 64948 Test: Total time: 0:00:06 (0.6916 s / it) * Acc@1 82.574 Acc@5 96.296 loss 0.645 Accuracy of the model EMA on 50000 test images: 82.6% Max EMA accuracy: 82.57% Epoch: [277] [ 0/312] eta: 0:51:45 lr: 0.001397 min_lr: 0.001397 loss: 2.1439 (2.1439) weight_decay: 0.0500 (0.0500) time: 9.9520 data: 9.1757 max mem: 64948 Epoch: [277] [ 10/312] eta: 0:07:50 lr: 0.001396 min_lr: 0.001396 loss: 1.8751 (1.9398) weight_decay: 0.0500 (0.0500) time: 1.5567 data: 0.8345 max mem: 64948 Epoch: [277] [ 20/312] eta: 0:05:34 lr: 0.001396 min_lr: 0.001396 loss: 2.0544 (2.0610) weight_decay: 0.0500 (0.0500) time: 0.7065 data: 0.0004 max mem: 64948 Epoch: [277] [ 30/312] eta: 0:04:42 lr: 0.001395 min_lr: 0.001395 loss: 2.0831 (2.0100) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [277] [ 40/312] eta: 0:04:12 lr: 0.001395 min_lr: 0.001395 loss: 1.9852 (2.0087) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [277] [ 50/312] eta: 0:03:51 lr: 0.001394 min_lr: 0.001394 loss: 1.9742 (1.9833) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [277] [ 60/312] eta: 0:03:34 lr: 0.001394 min_lr: 0.001394 loss: 1.9619 (1.9818) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [277] [ 70/312] eta: 0:03:20 lr: 0.001393 min_lr: 0.001393 loss: 1.8772 (1.9463) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [277] [ 80/312] eta: 0:03:08 lr: 0.001393 min_lr: 0.001393 loss: 1.6556 (1.9236) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [277] [ 90/312] eta: 0:02:57 lr: 0.001393 min_lr: 0.001393 loss: 1.8930 (1.9304) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [277] [100/312] eta: 0:02:47 lr: 0.001392 min_lr: 0.001392 loss: 2.0923 (1.9327) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [277] [110/312] eta: 0:02:37 lr: 0.001392 min_lr: 0.001392 loss: 1.9206 (1.9288) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [277] [120/312] eta: 0:02:28 lr: 0.001391 min_lr: 0.001391 loss: 2.0749 (1.9430) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [277] [130/312] eta: 0:02:19 lr: 0.001391 min_lr: 0.001391 loss: 2.0812 (1.9308) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [277] [140/312] eta: 0:02:11 lr: 0.001390 min_lr: 0.001390 loss: 1.8650 (1.9286) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [277] [150/312] eta: 0:02:02 lr: 0.001390 min_lr: 0.001390 loss: 1.9470 (1.9291) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [277] [160/312] eta: 0:01:54 lr: 0.001389 min_lr: 0.001389 loss: 2.0223 (1.9372) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [277] [170/312] eta: 0:01:46 lr: 0.001389 min_lr: 0.001389 loss: 1.9720 (1.9308) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [277] [180/312] eta: 0:01:38 lr: 0.001389 min_lr: 0.001389 loss: 1.9720 (1.9366) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [277] [190/312] eta: 0:01:31 lr: 0.001388 min_lr: 0.001388 loss: 2.2131 (1.9421) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [277] [200/312] eta: 0:01:23 lr: 0.001388 min_lr: 0.001388 loss: 2.0347 (1.9440) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [277] [210/312] eta: 0:01:15 lr: 0.001387 min_lr: 0.001387 loss: 1.9589 (1.9377) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [277] [220/312] eta: 0:01:07 lr: 0.001387 min_lr: 0.001387 loss: 2.1701 (1.9511) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [277] [230/312] eta: 0:01:00 lr: 0.001386 min_lr: 0.001386 loss: 2.1560 (1.9425) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [277] [240/312] eta: 0:00:52 lr: 0.001386 min_lr: 0.001386 loss: 1.8858 (1.9441) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [277] [250/312] eta: 0:00:45 lr: 0.001385 min_lr: 0.001385 loss: 2.0510 (1.9431) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [277] [260/312] eta: 0:00:38 lr: 0.001385 min_lr: 0.001385 loss: 2.0663 (1.9465) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [277] [270/312] eta: 0:00:30 lr: 0.001384 min_lr: 0.001384 loss: 1.8425 (1.9412) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [277] [280/312] eta: 0:00:23 lr: 0.001384 min_lr: 0.001384 loss: 1.8306 (1.9388) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [277] [290/312] eta: 0:00:16 lr: 0.001384 min_lr: 0.001384 loss: 1.9640 (1.9429) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [277] [300/312] eta: 0:00:08 lr: 0.001383 min_lr: 0.001383 loss: 2.1231 (1.9422) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [277] [310/312] eta: 0:00:01 lr: 0.001383 min_lr: 0.001383 loss: 2.0977 (1.9454) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [277] [311/312] eta: 0:00:00 lr: 0.001383 min_lr: 0.001383 loss: 1.9276 (1.9438) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [277] Total time: 0:03:47 (0.7284 s / it) Averaged stats: lr: 0.001383 min_lr: 0.001383 loss: 1.9276 (1.9728) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5653 (0.5653) acc1: 84.8958 (84.8958) acc5: 97.1354 (97.1354) time: 4.5119 data: 4.2897 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7964 (0.7715) acc1: 81.2500 (79.6160) acc5: 95.3125 (95.0400) time: 0.6531 data: 0.4767 max mem: 64948 Test: Total time: 0:00:06 (0.6756 s / it) * Acc@1 80.612 Acc@5 95.256 loss 0.745 Accuracy of the model on the 50000 test images: 80.6% Max accuracy: 80.64% Test: [0/9] eta: 0:00:45 loss: 0.4836 (0.4836) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 5.0840 data: 4.8778 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6674 (0.6695) acc1: 81.7708 (80.8320) acc5: 97.1354 (96.4160) time: 0.7162 data: 0.5421 max mem: 64948 Test: Total time: 0:00:06 (0.7245 s / it) * Acc@1 82.590 Acc@5 96.304 loss 0.644 Accuracy of the model EMA on 50000 test images: 82.6% Max EMA accuracy: 82.59% Epoch: [278] [ 0/312] eta: 0:48:37 lr: 0.001383 min_lr: 0.001383 loss: 1.2126 (1.2126) weight_decay: 0.0500 (0.0500) time: 9.3495 data: 8.5681 max mem: 64948 Epoch: [278] [ 10/312] eta: 0:07:37 lr: 0.001382 min_lr: 0.001382 loss: 2.0791 (1.9157) weight_decay: 0.0500 (0.0500) time: 1.5142 data: 0.7793 max mem: 64948 Epoch: [278] [ 20/312] eta: 0:05:28 lr: 0.001382 min_lr: 0.001382 loss: 2.0724 (1.9187) weight_decay: 0.0500 (0.0500) time: 0.7125 data: 0.0004 max mem: 64948 Epoch: [278] [ 30/312] eta: 0:04:37 lr: 0.001381 min_lr: 0.001381 loss: 1.9391 (1.9143) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [278] [ 40/312] eta: 0:04:08 lr: 0.001381 min_lr: 0.001381 loss: 1.7854 (1.8844) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [278] [ 50/312] eta: 0:03:48 lr: 0.001380 min_lr: 0.001380 loss: 1.9732 (1.9153) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [278] [ 60/312] eta: 0:03:32 lr: 0.001380 min_lr: 0.001380 loss: 2.0647 (1.9327) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [278] [ 70/312] eta: 0:03:19 lr: 0.001380 min_lr: 0.001380 loss: 1.8469 (1.8960) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [278] [ 80/312] eta: 0:03:07 lr: 0.001379 min_lr: 0.001379 loss: 1.8581 (1.9194) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [278] [ 90/312] eta: 0:02:56 lr: 0.001379 min_lr: 0.001379 loss: 2.0872 (1.9288) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [278] [100/312] eta: 0:02:46 lr: 0.001378 min_lr: 0.001378 loss: 2.0888 (1.9410) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [278] [110/312] eta: 0:02:37 lr: 0.001378 min_lr: 0.001378 loss: 2.0888 (1.9533) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [278] [120/312] eta: 0:02:27 lr: 0.001377 min_lr: 0.001377 loss: 1.9114 (1.9361) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [278] [130/312] eta: 0:02:19 lr: 0.001377 min_lr: 0.001377 loss: 1.7551 (1.9401) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [278] [140/312] eta: 0:02:10 lr: 0.001376 min_lr: 0.001376 loss: 2.0059 (1.9424) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [278] [150/312] eta: 0:02:02 lr: 0.001376 min_lr: 0.001376 loss: 1.9452 (1.9463) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [278] [160/312] eta: 0:01:54 lr: 0.001376 min_lr: 0.001376 loss: 2.1221 (1.9560) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [278] [170/312] eta: 0:01:46 lr: 0.001375 min_lr: 0.001375 loss: 2.1487 (1.9669) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [278] [180/312] eta: 0:01:38 lr: 0.001375 min_lr: 0.001375 loss: 2.1348 (1.9705) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [278] [190/312] eta: 0:01:30 lr: 0.001374 min_lr: 0.001374 loss: 2.1605 (1.9779) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [278] [200/312] eta: 0:01:22 lr: 0.001374 min_lr: 0.001374 loss: 2.1615 (1.9850) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [278] [210/312] eta: 0:01:15 lr: 0.001373 min_lr: 0.001373 loss: 2.0674 (1.9835) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [278] [220/312] eta: 0:01:07 lr: 0.001373 min_lr: 0.001373 loss: 1.8024 (1.9760) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [278] [230/312] eta: 0:01:00 lr: 0.001372 min_lr: 0.001372 loss: 1.8103 (1.9736) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [278] [240/312] eta: 0:00:52 lr: 0.001372 min_lr: 0.001372 loss: 2.0935 (1.9796) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [278] [250/312] eta: 0:00:45 lr: 0.001371 min_lr: 0.001371 loss: 2.1397 (1.9848) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [278] [260/312] eta: 0:00:37 lr: 0.001371 min_lr: 0.001371 loss: 2.1678 (1.9854) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [278] [270/312] eta: 0:00:30 lr: 0.001371 min_lr: 0.001371 loss: 1.9577 (1.9833) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [278] [280/312] eta: 0:00:23 lr: 0.001370 min_lr: 0.001370 loss: 1.8046 (1.9749) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [278] [290/312] eta: 0:00:15 lr: 0.001370 min_lr: 0.001370 loss: 1.8573 (1.9721) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [278] [300/312] eta: 0:00:08 lr: 0.001369 min_lr: 0.001369 loss: 1.9960 (1.9765) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [278] [310/312] eta: 0:00:01 lr: 0.001369 min_lr: 0.001369 loss: 2.0656 (1.9779) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [278] [311/312] eta: 0:00:00 lr: 0.001369 min_lr: 0.001369 loss: 2.0656 (1.9790) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [278] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.001369 min_lr: 0.001369 loss: 2.0656 (1.9719) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5220 (0.5220) acc1: 85.9375 (85.9375) acc5: 97.6562 (97.6562) time: 4.7540 data: 4.5456 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8177 (0.7618) acc1: 80.2083 (79.4560) acc5: 96.0938 (95.4560) time: 0.6797 data: 0.5051 max mem: 64948 Test: Total time: 0:00:06 (0.7055 s / it) * Acc@1 80.522 Acc@5 95.226 loss 0.750 Accuracy of the model on the 50000 test images: 80.5% Max accuracy: 80.64% Test: [0/9] eta: 0:00:42 loss: 0.4828 (0.4828) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 4.7009 data: 4.4987 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6666 (0.6692) acc1: 81.7708 (80.8960) acc5: 97.1354 (96.4480) time: 0.6737 data: 0.5000 max mem: 64948 Test: Total time: 0:00:06 (0.6856 s / it) * Acc@1 82.600 Acc@5 96.320 loss 0.644 Accuracy of the model EMA on 50000 test images: 82.6% Max EMA accuracy: 82.60% Epoch: [279] [ 0/312] eta: 0:51:14 lr: 0.001369 min_lr: 0.001369 loss: 1.7899 (1.7899) weight_decay: 0.0500 (0.0500) time: 9.8543 data: 9.0649 max mem: 64948 Epoch: [279] [ 10/312] eta: 0:07:48 lr: 0.001368 min_lr: 0.001368 loss: 1.6627 (1.7385) weight_decay: 0.0500 (0.0500) time: 1.5521 data: 0.8244 max mem: 64948 Epoch: [279] [ 20/312] eta: 0:05:33 lr: 0.001368 min_lr: 0.001368 loss: 1.9248 (1.8844) weight_decay: 0.0500 (0.0500) time: 0.7070 data: 0.0003 max mem: 64948 Epoch: [279] [ 30/312] eta: 0:04:42 lr: 0.001367 min_lr: 0.001367 loss: 1.9429 (1.8814) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [279] [ 40/312] eta: 0:04:11 lr: 0.001367 min_lr: 0.001367 loss: 1.9360 (1.8915) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0003 max mem: 64948 Epoch: [279] [ 50/312] eta: 0:03:50 lr: 0.001367 min_lr: 0.001367 loss: 1.9360 (1.9023) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [279] [ 60/312] eta: 0:03:34 lr: 0.001366 min_lr: 0.001366 loss: 1.9059 (1.9051) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [279] [ 70/312] eta: 0:03:20 lr: 0.001366 min_lr: 0.001366 loss: 2.0426 (1.9308) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [279] [ 80/312] eta: 0:03:08 lr: 0.001365 min_lr: 0.001365 loss: 2.1358 (1.9373) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [279] [ 90/312] eta: 0:02:57 lr: 0.001365 min_lr: 0.001365 loss: 2.0521 (1.9248) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [279] [100/312] eta: 0:02:47 lr: 0.001364 min_lr: 0.001364 loss: 1.8803 (1.9259) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [279] [110/312] eta: 0:02:37 lr: 0.001364 min_lr: 0.001364 loss: 1.9147 (1.9277) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0004 max mem: 64948 Epoch: [279] [120/312] eta: 0:02:28 lr: 0.001363 min_lr: 0.001363 loss: 1.9694 (1.9239) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [279] [130/312] eta: 0:02:19 lr: 0.001363 min_lr: 0.001363 loss: 2.0927 (1.9339) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [279] [140/312] eta: 0:02:11 lr: 0.001363 min_lr: 0.001363 loss: 2.1259 (1.9461) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [279] [150/312] eta: 0:02:02 lr: 0.001362 min_lr: 0.001362 loss: 2.1609 (1.9527) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [279] [160/312] eta: 0:01:54 lr: 0.001362 min_lr: 0.001362 loss: 1.9400 (1.9414) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [279] [170/312] eta: 0:01:46 lr: 0.001361 min_lr: 0.001361 loss: 1.7294 (1.9381) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [279] [180/312] eta: 0:01:38 lr: 0.001361 min_lr: 0.001361 loss: 1.9085 (1.9395) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [279] [190/312] eta: 0:01:30 lr: 0.001360 min_lr: 0.001360 loss: 1.9926 (1.9461) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [279] [200/312] eta: 0:01:23 lr: 0.001360 min_lr: 0.001360 loss: 2.1794 (1.9521) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [279] [210/312] eta: 0:01:15 lr: 0.001359 min_lr: 0.001359 loss: 2.1282 (1.9604) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [279] [220/312] eta: 0:01:07 lr: 0.001359 min_lr: 0.001359 loss: 2.1282 (1.9682) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [279] [230/312] eta: 0:01:00 lr: 0.001359 min_lr: 0.001359 loss: 2.0285 (1.9621) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [279] [240/312] eta: 0:00:52 lr: 0.001358 min_lr: 0.001358 loss: 2.0410 (1.9639) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [279] [250/312] eta: 0:00:45 lr: 0.001358 min_lr: 0.001358 loss: 2.0353 (1.9617) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [279] [260/312] eta: 0:00:38 lr: 0.001357 min_lr: 0.001357 loss: 2.0221 (1.9652) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [279] [270/312] eta: 0:00:30 lr: 0.001357 min_lr: 0.001357 loss: 2.1245 (1.9676) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [279] [280/312] eta: 0:00:23 lr: 0.001356 min_lr: 0.001356 loss: 2.0282 (1.9704) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0012 max mem: 64948 Epoch: [279] [290/312] eta: 0:00:16 lr: 0.001356 min_lr: 0.001356 loss: 2.0397 (1.9726) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0011 max mem: 64948 Epoch: [279] [300/312] eta: 0:00:08 lr: 0.001355 min_lr: 0.001355 loss: 2.0025 (1.9706) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [279] [310/312] eta: 0:00:01 lr: 0.001355 min_lr: 0.001355 loss: 2.0025 (1.9751) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [279] [311/312] eta: 0:00:00 lr: 0.001355 min_lr: 0.001355 loss: 2.0025 (1.9739) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [279] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.001355 min_lr: 0.001355 loss: 2.0025 (1.9711) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.5358 (0.5358) acc1: 85.6771 (85.6771) acc5: 96.8750 (96.8750) time: 4.8900 data: 4.6823 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.8008 (0.7622) acc1: 80.4688 (79.8720) acc5: 95.5729 (95.0720) time: 0.6949 data: 0.5204 max mem: 64948 Test: Total time: 0:00:06 (0.7209 s / it) * Acc@1 80.556 Acc@5 95.114 loss 0.753 Accuracy of the model on the 50000 test images: 80.6% Max accuracy: 80.64% Test: [0/9] eta: 0:00:39 loss: 0.4822 (0.4822) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 4.4427 data: 4.2317 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6660 (0.6688) acc1: 81.7708 (80.8960) acc5: 97.1354 (96.4160) time: 0.6533 data: 0.4786 max mem: 64948 Test: Total time: 0:00:05 (0.6662 s / it) * Acc@1 82.624 Acc@5 96.328 loss 0.644 Accuracy of the model EMA on 50000 test images: 82.6% Max EMA accuracy: 82.62% Epoch: [280] [ 0/312] eta: 0:49:57 lr: 0.001355 min_lr: 0.001355 loss: 2.4178 (2.4178) weight_decay: 0.0500 (0.0500) time: 9.6085 data: 8.7904 max mem: 64948 Epoch: [280] [ 10/312] eta: 0:07:40 lr: 0.001354 min_lr: 0.001354 loss: 2.0718 (2.0825) weight_decay: 0.0500 (0.0500) time: 1.5254 data: 0.7995 max mem: 64948 Epoch: [280] [ 20/312] eta: 0:05:29 lr: 0.001354 min_lr: 0.001354 loss: 1.9914 (2.0023) weight_decay: 0.0500 (0.0500) time: 0.7051 data: 0.0004 max mem: 64948 Epoch: [280] [ 30/312] eta: 0:04:39 lr: 0.001354 min_lr: 0.001354 loss: 1.9502 (1.9943) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [280] [ 40/312] eta: 0:04:09 lr: 0.001353 min_lr: 0.001353 loss: 1.9502 (1.9803) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [280] [ 50/312] eta: 0:03:48 lr: 0.001353 min_lr: 0.001353 loss: 1.9553 (1.9593) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [280] [ 60/312] eta: 0:03:32 lr: 0.001352 min_lr: 0.001352 loss: 1.9318 (1.9603) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [280] [ 70/312] eta: 0:03:19 lr: 0.001352 min_lr: 0.001352 loss: 2.0199 (1.9623) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [280] [ 80/312] eta: 0:03:07 lr: 0.001351 min_lr: 0.001351 loss: 2.0297 (1.9425) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [280] [ 90/312] eta: 0:02:56 lr: 0.001351 min_lr: 0.001351 loss: 1.9727 (1.9636) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [280] [100/312] eta: 0:02:46 lr: 0.001350 min_lr: 0.001350 loss: 2.0779 (1.9668) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [280] [110/312] eta: 0:02:37 lr: 0.001350 min_lr: 0.001350 loss: 1.9326 (1.9552) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [280] [120/312] eta: 0:02:28 lr: 0.001350 min_lr: 0.001350 loss: 1.9048 (1.9502) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [280] [130/312] eta: 0:02:19 lr: 0.001349 min_lr: 0.001349 loss: 2.0869 (1.9574) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [280] [140/312] eta: 0:02:10 lr: 0.001349 min_lr: 0.001349 loss: 2.1073 (1.9639) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [280] [150/312] eta: 0:02:02 lr: 0.001348 min_lr: 0.001348 loss: 1.9075 (1.9466) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [280] [160/312] eta: 0:01:54 lr: 0.001348 min_lr: 0.001348 loss: 1.7454 (1.9452) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0003 max mem: 64948 Epoch: [280] [170/312] eta: 0:01:46 lr: 0.001347 min_lr: 0.001347 loss: 2.0629 (1.9552) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [280] [180/312] eta: 0:01:38 lr: 0.001347 min_lr: 0.001347 loss: 2.0454 (1.9540) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [280] [190/312] eta: 0:01:30 lr: 0.001346 min_lr: 0.001346 loss: 1.9631 (1.9538) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [280] [200/312] eta: 0:01:22 lr: 0.001346 min_lr: 0.001346 loss: 1.9629 (1.9540) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [280] [210/312] eta: 0:01:15 lr: 0.001346 min_lr: 0.001346 loss: 2.0199 (1.9606) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [280] [220/312] eta: 0:01:07 lr: 0.001345 min_lr: 0.001345 loss: 2.0906 (1.9655) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [280] [230/312] eta: 0:01:00 lr: 0.001345 min_lr: 0.001345 loss: 2.0906 (1.9646) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [280] [240/312] eta: 0:00:52 lr: 0.001344 min_lr: 0.001344 loss: 1.9767 (1.9626) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [280] [250/312] eta: 0:00:45 lr: 0.001344 min_lr: 0.001344 loss: 2.0957 (1.9678) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [280] [260/312] eta: 0:00:37 lr: 0.001343 min_lr: 0.001343 loss: 2.0927 (1.9629) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [280] [270/312] eta: 0:00:30 lr: 0.001343 min_lr: 0.001343 loss: 2.0506 (1.9693) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [280] [280/312] eta: 0:00:23 lr: 0.001343 min_lr: 0.001343 loss: 2.0209 (1.9652) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [280] [290/312] eta: 0:00:15 lr: 0.001342 min_lr: 0.001342 loss: 1.8781 (1.9628) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [280] [300/312] eta: 0:00:08 lr: 0.001342 min_lr: 0.001342 loss: 2.0013 (1.9628) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [280] [310/312] eta: 0:00:01 lr: 0.001341 min_lr: 0.001341 loss: 2.0009 (1.9600) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [280] [311/312] eta: 0:00:00 lr: 0.001341 min_lr: 0.001341 loss: 2.0009 (1.9607) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [280] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.001341 min_lr: 0.001341 loss: 2.0009 (1.9667) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.5399 (0.5399) acc1: 85.4167 (85.4167) acc5: 97.9167 (97.9167) time: 4.7808 data: 4.5613 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7830 (0.7400) acc1: 79.9479 (79.6800) acc5: 96.0938 (95.7440) time: 0.6828 data: 0.5069 max mem: 64948 Test: Total time: 0:00:06 (0.7038 s / it) * Acc@1 80.840 Acc@5 95.462 loss 0.731 Accuracy of the model on the 50000 test images: 80.8% Max accuracy: 80.84% Test: [0/9] eta: 0:00:40 loss: 0.4819 (0.4819) acc1: 86.1979 (86.1979) acc5: 97.9167 (97.9167) time: 4.5290 data: 4.3275 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6659 (0.6684) acc1: 82.0312 (80.8960) acc5: 97.1354 (96.3840) time: 0.6545 data: 0.4809 max mem: 64948 Test: Total time: 0:00:05 (0.6636 s / it) * Acc@1 82.636 Acc@5 96.312 loss 0.643 Accuracy of the model EMA on 50000 test images: 82.6% Max EMA accuracy: 82.64% Epoch: [281] [ 0/312] eta: 0:47:32 lr: 0.001341 min_lr: 0.001341 loss: 1.8700 (1.8700) weight_decay: 0.0500 (0.0500) time: 9.1421 data: 8.2961 max mem: 64948 Epoch: [281] [ 10/312] eta: 0:07:31 lr: 0.001341 min_lr: 0.001341 loss: 1.8700 (1.8440) weight_decay: 0.0500 (0.0500) time: 1.4941 data: 0.7546 max mem: 64948 Epoch: [281] [ 20/312] eta: 0:05:25 lr: 0.001340 min_lr: 0.001340 loss: 1.8763 (1.8925) weight_decay: 0.0500 (0.0500) time: 0.7119 data: 0.0004 max mem: 64948 Epoch: [281] [ 30/312] eta: 0:04:35 lr: 0.001340 min_lr: 0.001340 loss: 2.1061 (1.9426) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [281] [ 40/312] eta: 0:04:07 lr: 0.001339 min_lr: 0.001339 loss: 2.0353 (1.9305) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [281] [ 50/312] eta: 0:03:47 lr: 0.001339 min_lr: 0.001339 loss: 2.0141 (1.9433) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [281] [ 60/312] eta: 0:03:31 lr: 0.001338 min_lr: 0.001338 loss: 2.0119 (1.9471) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [281] [ 70/312] eta: 0:03:18 lr: 0.001338 min_lr: 0.001338 loss: 1.8632 (1.9331) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [281] [ 80/312] eta: 0:03:06 lr: 0.001338 min_lr: 0.001338 loss: 1.9208 (1.9469) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [281] [ 90/312] eta: 0:02:55 lr: 0.001337 min_lr: 0.001337 loss: 2.1819 (1.9639) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [281] [100/312] eta: 0:02:45 lr: 0.001337 min_lr: 0.001337 loss: 2.1480 (1.9665) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [281] [110/312] eta: 0:02:36 lr: 0.001336 min_lr: 0.001336 loss: 2.0691 (1.9646) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [281] [120/312] eta: 0:02:27 lr: 0.001336 min_lr: 0.001336 loss: 2.1348 (1.9673) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [281] [130/312] eta: 0:02:18 lr: 0.001335 min_lr: 0.001335 loss: 2.1083 (1.9718) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [281] [140/312] eta: 0:02:10 lr: 0.001335 min_lr: 0.001335 loss: 2.0340 (1.9706) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [281] [150/312] eta: 0:02:02 lr: 0.001334 min_lr: 0.001334 loss: 1.8697 (1.9636) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [281] [160/312] eta: 0:01:54 lr: 0.001334 min_lr: 0.001334 loss: 1.9193 (1.9656) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [281] [170/312] eta: 0:01:46 lr: 0.001334 min_lr: 0.001334 loss: 1.9699 (1.9615) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [281] [180/312] eta: 0:01:38 lr: 0.001333 min_lr: 0.001333 loss: 2.0111 (1.9661) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [281] [190/312] eta: 0:01:30 lr: 0.001333 min_lr: 0.001333 loss: 1.9630 (1.9649) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [281] [200/312] eta: 0:01:22 lr: 0.001332 min_lr: 0.001332 loss: 1.9644 (1.9667) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [281] [210/312] eta: 0:01:15 lr: 0.001332 min_lr: 0.001332 loss: 2.1206 (1.9712) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [281] [220/312] eta: 0:01:07 lr: 0.001331 min_lr: 0.001331 loss: 2.1331 (1.9748) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [281] [230/312] eta: 0:01:00 lr: 0.001331 min_lr: 0.001331 loss: 2.1125 (1.9722) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [281] [240/312] eta: 0:00:52 lr: 0.001330 min_lr: 0.001330 loss: 1.7052 (1.9616) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [281] [250/312] eta: 0:00:45 lr: 0.001330 min_lr: 0.001330 loss: 1.8715 (1.9623) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [281] [260/312] eta: 0:00:37 lr: 0.001330 min_lr: 0.001330 loss: 2.0197 (1.9638) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [281] [270/312] eta: 0:00:30 lr: 0.001329 min_lr: 0.001329 loss: 1.9969 (1.9594) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [281] [280/312] eta: 0:00:23 lr: 0.001329 min_lr: 0.001329 loss: 1.9238 (1.9607) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [281] [290/312] eta: 0:00:15 lr: 0.001328 min_lr: 0.001328 loss: 2.0899 (1.9667) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [281] [300/312] eta: 0:00:08 lr: 0.001328 min_lr: 0.001328 loss: 2.0393 (1.9606) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [281] [310/312] eta: 0:00:01 lr: 0.001327 min_lr: 0.001327 loss: 2.1111 (1.9693) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [281] [311/312] eta: 0:00:00 lr: 0.001327 min_lr: 0.001327 loss: 2.1111 (1.9697) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [281] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.001327 min_lr: 0.001327 loss: 2.1111 (1.9664) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4860 (0.4860) acc1: 85.6771 (85.6771) acc5: 97.6562 (97.6562) time: 4.6610 data: 4.4496 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7641 (0.7575) acc1: 81.2500 (79.7760) acc5: 96.3542 (95.4560) time: 0.6691 data: 0.4945 max mem: 64948 Test: Total time: 0:00:06 (0.6890 s / it) * Acc@1 80.392 Acc@5 95.248 loss 0.749 Accuracy of the model on the 50000 test images: 80.4% Max accuracy: 80.84% Test: [0/9] eta: 0:00:40 loss: 0.4815 (0.4815) acc1: 86.4583 (86.4583) acc5: 97.9167 (97.9167) time: 4.4560 data: 4.2379 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6657 (0.6679) acc1: 82.0312 (81.0240) acc5: 97.1354 (96.3840) time: 0.6534 data: 0.4778 max mem: 64948 Test: Total time: 0:00:06 (0.6704 s / it) * Acc@1 82.640 Acc@5 96.318 loss 0.643 Accuracy of the model EMA on 50000 test images: 82.6% Max EMA accuracy: 82.64% Epoch: [282] [ 0/312] eta: 0:48:22 lr: 0.001327 min_lr: 0.001327 loss: 2.2074 (2.2074) weight_decay: 0.0500 (0.0500) time: 9.3043 data: 8.3032 max mem: 64948 Epoch: [282] [ 10/312] eta: 0:07:42 lr: 0.001327 min_lr: 0.001327 loss: 2.1387 (2.0723) weight_decay: 0.0500 (0.0500) time: 1.5320 data: 0.7552 max mem: 64948 Epoch: [282] [ 20/312] eta: 0:05:31 lr: 0.001326 min_lr: 0.001326 loss: 2.0504 (1.9657) weight_decay: 0.0500 (0.0500) time: 0.7265 data: 0.0003 max mem: 64948 Epoch: [282] [ 30/312] eta: 0:04:40 lr: 0.001326 min_lr: 0.001326 loss: 1.9032 (1.9371) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [282] [ 40/312] eta: 0:04:10 lr: 0.001326 min_lr: 0.001326 loss: 1.8726 (1.8921) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [282] [ 50/312] eta: 0:03:49 lr: 0.001325 min_lr: 0.001325 loss: 1.7759 (1.8741) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [282] [ 60/312] eta: 0:03:33 lr: 0.001325 min_lr: 0.001325 loss: 1.9521 (1.8945) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [282] [ 70/312] eta: 0:03:19 lr: 0.001324 min_lr: 0.001324 loss: 1.9521 (1.8892) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [282] [ 80/312] eta: 0:03:07 lr: 0.001324 min_lr: 0.001324 loss: 1.8909 (1.8895) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [282] [ 90/312] eta: 0:02:57 lr: 0.001323 min_lr: 0.001323 loss: 2.0814 (1.9159) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [282] [100/312] eta: 0:02:46 lr: 0.001323 min_lr: 0.001323 loss: 2.1120 (1.9041) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [282] [110/312] eta: 0:02:37 lr: 0.001322 min_lr: 0.001322 loss: 1.8271 (1.9117) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [282] [120/312] eta: 0:02:28 lr: 0.001322 min_lr: 0.001322 loss: 1.9291 (1.9123) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [282] [130/312] eta: 0:02:19 lr: 0.001322 min_lr: 0.001322 loss: 2.0268 (1.9195) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [282] [140/312] eta: 0:02:10 lr: 0.001321 min_lr: 0.001321 loss: 2.0268 (1.9186) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [282] [150/312] eta: 0:02:02 lr: 0.001321 min_lr: 0.001321 loss: 1.8306 (1.9100) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [282] [160/312] eta: 0:01:54 lr: 0.001320 min_lr: 0.001320 loss: 1.8306 (1.9048) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [282] [170/312] eta: 0:01:46 lr: 0.001320 min_lr: 0.001320 loss: 2.0277 (1.9155) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [282] [180/312] eta: 0:01:38 lr: 0.001319 min_lr: 0.001319 loss: 2.0041 (1.9102) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [282] [190/312] eta: 0:01:30 lr: 0.001319 min_lr: 0.001319 loss: 2.0041 (1.9208) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [282] [200/312] eta: 0:01:22 lr: 0.001319 min_lr: 0.001319 loss: 2.0961 (1.9226) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [282] [210/312] eta: 0:01:15 lr: 0.001318 min_lr: 0.001318 loss: 2.0788 (1.9321) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [282] [220/312] eta: 0:01:07 lr: 0.001318 min_lr: 0.001318 loss: 2.0788 (1.9334) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [282] [230/312] eta: 0:01:00 lr: 0.001317 min_lr: 0.001317 loss: 1.8583 (1.9300) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [282] [240/312] eta: 0:00:52 lr: 0.001317 min_lr: 0.001317 loss: 1.8583 (1.9290) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [282] [250/312] eta: 0:00:45 lr: 0.001316 min_lr: 0.001316 loss: 1.8505 (1.9244) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [282] [260/312] eta: 0:00:37 lr: 0.001316 min_lr: 0.001316 loss: 1.8505 (1.9285) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [282] [270/312] eta: 0:00:30 lr: 0.001315 min_lr: 0.001315 loss: 1.8388 (1.9276) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [282] [280/312] eta: 0:00:23 lr: 0.001315 min_lr: 0.001315 loss: 2.0792 (1.9310) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0009 max mem: 64948 Epoch: [282] [290/312] eta: 0:00:15 lr: 0.001315 min_lr: 0.001315 loss: 2.0835 (1.9363) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [282] [300/312] eta: 0:00:08 lr: 0.001314 min_lr: 0.001314 loss: 2.0863 (1.9403) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [282] [310/312] eta: 0:00:01 lr: 0.001314 min_lr: 0.001314 loss: 2.0408 (1.9402) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [282] [311/312] eta: 0:00:00 lr: 0.001314 min_lr: 0.001314 loss: 2.0337 (1.9380) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [282] Total time: 0:03:46 (0.7267 s / it) Averaged stats: lr: 0.001314 min_lr: 0.001314 loss: 2.0337 (1.9570) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5810 (0.5810) acc1: 86.1979 (86.1979) acc5: 96.8750 (96.8750) time: 4.6289 data: 4.4069 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7462 (0.7719) acc1: 80.4688 (79.6480) acc5: 95.0521 (95.2960) time: 0.6662 data: 0.4897 max mem: 64948 Test: Total time: 0:00:06 (0.6900 s / it) * Acc@1 80.398 Acc@5 95.172 loss 0.759 Accuracy of the model on the 50000 test images: 80.4% Max accuracy: 80.84% Test: [0/9] eta: 0:00:45 loss: 0.4813 (0.4813) acc1: 86.1979 (86.1979) acc5: 97.9167 (97.9167) time: 5.0638 data: 4.8604 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6644 (0.6675) acc1: 81.7708 (80.9280) acc5: 97.1354 (96.4160) time: 0.7139 data: 0.5401 max mem: 64948 Test: Total time: 0:00:06 (0.7221 s / it) * Acc@1 82.644 Acc@5 96.320 loss 0.643 Accuracy of the model EMA on 50000 test images: 82.6% Max EMA accuracy: 82.64% Epoch: [283] [ 0/312] eta: 0:42:07 lr: 0.001314 min_lr: 0.001314 loss: 1.3586 (1.3586) weight_decay: 0.0500 (0.0500) time: 8.1006 data: 7.2499 max mem: 64948 Epoch: [283] [ 10/312] eta: 0:07:32 lr: 0.001313 min_lr: 0.001313 loss: 2.0414 (1.9328) weight_decay: 0.0500 (0.0500) time: 1.4992 data: 0.7137 max mem: 64948 Epoch: [283] [ 20/312] eta: 0:05:26 lr: 0.001313 min_lr: 0.001313 loss: 2.0414 (1.9225) weight_decay: 0.0500 (0.0500) time: 0.7685 data: 0.0302 max mem: 64948 Epoch: [283] [ 30/312] eta: 0:04:36 lr: 0.001312 min_lr: 0.001312 loss: 2.0710 (1.9723) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [283] [ 40/312] eta: 0:04:08 lr: 0.001312 min_lr: 0.001312 loss: 1.9725 (1.9161) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [283] [ 50/312] eta: 0:03:48 lr: 0.001311 min_lr: 0.001311 loss: 1.8172 (1.9022) weight_decay: 0.0500 (0.0500) time: 0.7013 data: 0.0004 max mem: 64948 Epoch: [283] [ 60/312] eta: 0:03:32 lr: 0.001311 min_lr: 0.001311 loss: 1.9704 (1.9223) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [283] [ 70/312] eta: 0:03:18 lr: 0.001311 min_lr: 0.001311 loss: 2.0095 (1.9400) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [283] [ 80/312] eta: 0:03:06 lr: 0.001310 min_lr: 0.001310 loss: 2.0095 (1.9528) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [283] [ 90/312] eta: 0:02:56 lr: 0.001310 min_lr: 0.001310 loss: 2.0883 (1.9742) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [283] [100/312] eta: 0:02:46 lr: 0.001309 min_lr: 0.001309 loss: 2.1117 (1.9754) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [283] [110/312] eta: 0:02:36 lr: 0.001309 min_lr: 0.001309 loss: 2.1410 (1.9853) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [283] [120/312] eta: 0:02:27 lr: 0.001308 min_lr: 0.001308 loss: 1.9801 (1.9744) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [283] [130/312] eta: 0:02:18 lr: 0.001308 min_lr: 0.001308 loss: 1.8915 (1.9725) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [283] [140/312] eta: 0:02:10 lr: 0.001307 min_lr: 0.001307 loss: 2.0512 (1.9784) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [283] [150/312] eta: 0:02:02 lr: 0.001307 min_lr: 0.001307 loss: 1.9972 (1.9746) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [283] [160/312] eta: 0:01:54 lr: 0.001307 min_lr: 0.001307 loss: 2.0683 (1.9848) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [283] [170/312] eta: 0:01:46 lr: 0.001306 min_lr: 0.001306 loss: 2.0550 (1.9802) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [283] [180/312] eta: 0:01:38 lr: 0.001306 min_lr: 0.001306 loss: 2.0275 (1.9897) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [283] [190/312] eta: 0:01:30 lr: 0.001305 min_lr: 0.001305 loss: 2.0332 (1.9887) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [283] [200/312] eta: 0:01:22 lr: 0.001305 min_lr: 0.001305 loss: 1.9274 (1.9827) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [283] [210/312] eta: 0:01:15 lr: 0.001304 min_lr: 0.001304 loss: 1.9528 (1.9849) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [283] [220/312] eta: 0:01:07 lr: 0.001304 min_lr: 0.001304 loss: 1.9791 (1.9857) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [283] [230/312] eta: 0:01:00 lr: 0.001303 min_lr: 0.001303 loss: 1.9435 (1.9860) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [283] [240/312] eta: 0:00:52 lr: 0.001303 min_lr: 0.001303 loss: 1.9590 (1.9805) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [283] [250/312] eta: 0:00:45 lr: 0.001303 min_lr: 0.001303 loss: 2.0860 (1.9841) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [283] [260/312] eta: 0:00:37 lr: 0.001302 min_lr: 0.001302 loss: 2.1050 (1.9850) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [283] [270/312] eta: 0:00:30 lr: 0.001302 min_lr: 0.001302 loss: 1.9233 (1.9783) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [283] [280/312] eta: 0:00:23 lr: 0.001301 min_lr: 0.001301 loss: 1.9344 (1.9824) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0010 max mem: 64948 Epoch: [283] [290/312] eta: 0:00:15 lr: 0.001301 min_lr: 0.001301 loss: 1.9453 (1.9810) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0008 max mem: 64948 Epoch: [283] [300/312] eta: 0:00:08 lr: 0.001300 min_lr: 0.001300 loss: 2.1099 (1.9854) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [283] [310/312] eta: 0:00:01 lr: 0.001300 min_lr: 0.001300 loss: 2.0973 (1.9858) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [283] [311/312] eta: 0:00:00 lr: 0.001300 min_lr: 0.001300 loss: 2.0973 (1.9859) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [283] Total time: 0:03:46 (0.7261 s / it) Averaged stats: lr: 0.001300 min_lr: 0.001300 loss: 2.0973 (1.9552) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4964 (0.4964) acc1: 86.4583 (86.4583) acc5: 97.3958 (97.3958) time: 4.7810 data: 4.5743 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7555 (0.7533) acc1: 80.9896 (79.8720) acc5: 95.5729 (95.5520) time: 0.6826 data: 0.5084 max mem: 64948 Test: Total time: 0:00:06 (0.7081 s / it) * Acc@1 80.892 Acc@5 95.420 loss 0.729 Accuracy of the model on the 50000 test images: 80.9% Max accuracy: 80.89% Test: [0/9] eta: 0:00:40 loss: 0.4806 (0.4806) acc1: 86.1979 (86.1979) acc5: 98.1771 (98.1771) time: 4.5163 data: 4.3123 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6637 (0.6672) acc1: 81.7708 (80.8960) acc5: 97.1354 (96.4800) time: 0.6531 data: 0.4793 max mem: 64948 Test: Total time: 0:00:05 (0.6609 s / it) * Acc@1 82.634 Acc@5 96.332 loss 0.642 Accuracy of the model EMA on 50000 test images: 82.6% Epoch: [284] [ 0/312] eta: 0:55:58 lr: 0.001300 min_lr: 0.001300 loss: 1.9629 (1.9629) weight_decay: 0.0500 (0.0500) time: 10.7655 data: 7.5946 max mem: 64948 Epoch: [284] [ 10/312] eta: 0:08:23 lr: 0.001299 min_lr: 0.001299 loss: 2.1011 (2.0924) weight_decay: 0.0500 (0.0500) time: 1.6658 data: 0.6908 max mem: 64948 Epoch: [284] [ 20/312] eta: 0:05:51 lr: 0.001299 min_lr: 0.001299 loss: 2.0660 (2.0582) weight_decay: 0.0500 (0.0500) time: 0.7257 data: 0.0004 max mem: 64948 Epoch: [284] [ 30/312] eta: 0:04:53 lr: 0.001299 min_lr: 0.001299 loss: 2.0442 (2.0394) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [284] [ 40/312] eta: 0:04:19 lr: 0.001298 min_lr: 0.001298 loss: 2.0714 (2.0223) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [284] [ 50/312] eta: 0:03:56 lr: 0.001298 min_lr: 0.001298 loss: 2.1026 (2.0285) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [284] [ 60/312] eta: 0:03:39 lr: 0.001297 min_lr: 0.001297 loss: 2.0281 (2.0131) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [284] [ 70/312] eta: 0:03:24 lr: 0.001297 min_lr: 0.001297 loss: 2.0256 (2.0148) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [284] [ 80/312] eta: 0:03:12 lr: 0.001296 min_lr: 0.001296 loss: 2.0213 (2.0046) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [284] [ 90/312] eta: 0:03:00 lr: 0.001296 min_lr: 0.001296 loss: 2.0936 (2.0233) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [284] [100/312] eta: 0:02:50 lr: 0.001295 min_lr: 0.001295 loss: 2.0365 (1.9984) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [284] [110/312] eta: 0:02:40 lr: 0.001295 min_lr: 0.001295 loss: 1.9754 (1.9969) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [284] [120/312] eta: 0:02:30 lr: 0.001295 min_lr: 0.001295 loss: 2.0779 (1.9966) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [284] [130/312] eta: 0:02:21 lr: 0.001294 min_lr: 0.001294 loss: 2.0519 (1.9837) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [284] [140/312] eta: 0:02:12 lr: 0.001294 min_lr: 0.001294 loss: 1.9275 (1.9788) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [284] [150/312] eta: 0:02:04 lr: 0.001293 min_lr: 0.001293 loss: 2.0065 (1.9764) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [284] [160/312] eta: 0:01:55 lr: 0.001293 min_lr: 0.001293 loss: 1.9816 (1.9786) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [284] [170/312] eta: 0:01:47 lr: 0.001292 min_lr: 0.001292 loss: 1.8358 (1.9654) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [284] [180/312] eta: 0:01:39 lr: 0.001292 min_lr: 0.001292 loss: 1.8154 (1.9629) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [284] [190/312] eta: 0:01:31 lr: 0.001292 min_lr: 0.001292 loss: 1.9079 (1.9597) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [284] [200/312] eta: 0:01:23 lr: 0.001291 min_lr: 0.001291 loss: 1.8916 (1.9572) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [284] [210/312] eta: 0:01:16 lr: 0.001291 min_lr: 0.001291 loss: 1.8974 (1.9535) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [284] [220/312] eta: 0:01:08 lr: 0.001290 min_lr: 0.001290 loss: 1.9869 (1.9558) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [284] [230/312] eta: 0:01:00 lr: 0.001290 min_lr: 0.001290 loss: 2.0084 (1.9611) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [284] [240/312] eta: 0:00:53 lr: 0.001289 min_lr: 0.001289 loss: 1.8549 (1.9528) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [284] [250/312] eta: 0:00:45 lr: 0.001289 min_lr: 0.001289 loss: 1.8510 (1.9556) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [284] [260/312] eta: 0:00:38 lr: 0.001288 min_lr: 0.001288 loss: 1.9639 (1.9566) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [284] [270/312] eta: 0:00:30 lr: 0.001288 min_lr: 0.001288 loss: 1.9042 (1.9590) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [284] [280/312] eta: 0:00:23 lr: 0.001288 min_lr: 0.001288 loss: 1.9184 (1.9598) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0010 max mem: 64948 Epoch: [284] [290/312] eta: 0:00:16 lr: 0.001287 min_lr: 0.001287 loss: 1.9083 (1.9573) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [284] [300/312] eta: 0:00:08 lr: 0.001287 min_lr: 0.001287 loss: 2.0117 (1.9620) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [284] [310/312] eta: 0:00:01 lr: 0.001286 min_lr: 0.001286 loss: 2.0417 (1.9621) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [284] [311/312] eta: 0:00:00 lr: 0.001286 min_lr: 0.001286 loss: 2.0221 (1.9596) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [284] Total time: 0:03:48 (0.7321 s / it) Averaged stats: lr: 0.001286 min_lr: 0.001286 loss: 2.0221 (1.9544) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5482 (0.5482) acc1: 86.7188 (86.7188) acc5: 97.1354 (97.1354) time: 4.6626 data: 4.4468 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7321 (0.7400) acc1: 80.9896 (79.8720) acc5: 95.8333 (95.2960) time: 0.6694 data: 0.4942 max mem: 64948 Test: Total time: 0:00:06 (0.6907 s / it) * Acc@1 80.530 Acc@5 95.534 loss 0.741 Accuracy of the model on the 50000 test images: 80.5% Max accuracy: 80.89% Test: [0/9] eta: 0:00:46 loss: 0.4797 (0.4797) acc1: 86.1979 (86.1979) acc5: 97.9167 (97.9167) time: 5.1160 data: 4.9007 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6632 (0.6669) acc1: 81.7708 (80.9280) acc5: 97.1354 (96.4480) time: 0.7198 data: 0.5446 max mem: 64948 Test: Total time: 0:00:06 (0.7289 s / it) * Acc@1 82.638 Acc@5 96.316 loss 0.642 Accuracy of the model EMA on 50000 test images: 82.6% Epoch: [285] [ 0/312] eta: 0:53:29 lr: 0.001286 min_lr: 0.001286 loss: 1.5788 (1.5788) weight_decay: 0.0500 (0.0500) time: 10.2853 data: 5.8942 max mem: 64948 Epoch: [285] [ 10/312] eta: 0:08:02 lr: 0.001286 min_lr: 0.001286 loss: 2.0252 (1.9750) weight_decay: 0.0500 (0.0500) time: 1.5975 data: 0.5363 max mem: 64948 Epoch: [285] [ 20/312] eta: 0:05:40 lr: 0.001285 min_lr: 0.001285 loss: 2.0252 (1.9851) weight_decay: 0.0500 (0.0500) time: 0.7116 data: 0.0005 max mem: 64948 Epoch: [285] [ 30/312] eta: 0:04:46 lr: 0.001285 min_lr: 0.001285 loss: 1.9871 (1.9437) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [285] [ 40/312] eta: 0:04:15 lr: 0.001284 min_lr: 0.001284 loss: 1.9871 (1.9445) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [285] [ 50/312] eta: 0:03:53 lr: 0.001284 min_lr: 0.001284 loss: 1.9830 (1.9542) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [285] [ 60/312] eta: 0:03:36 lr: 0.001284 min_lr: 0.001284 loss: 2.0299 (1.9650) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [285] [ 70/312] eta: 0:03:22 lr: 0.001283 min_lr: 0.001283 loss: 2.0385 (1.9635) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [285] [ 80/312] eta: 0:03:09 lr: 0.001283 min_lr: 0.001283 loss: 2.0594 (1.9826) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [285] [ 90/312] eta: 0:02:58 lr: 0.001282 min_lr: 0.001282 loss: 2.1671 (1.9679) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [285] [100/312] eta: 0:02:48 lr: 0.001282 min_lr: 0.001282 loss: 1.9185 (1.9367) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [285] [110/312] eta: 0:02:38 lr: 0.001281 min_lr: 0.001281 loss: 1.9245 (1.9467) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [285] [120/312] eta: 0:02:29 lr: 0.001281 min_lr: 0.001281 loss: 2.0055 (1.9424) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [285] [130/312] eta: 0:02:20 lr: 0.001281 min_lr: 0.001281 loss: 1.9584 (1.9323) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [285] [140/312] eta: 0:02:11 lr: 0.001280 min_lr: 0.001280 loss: 1.9761 (1.9436) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [285] [150/312] eta: 0:02:03 lr: 0.001280 min_lr: 0.001280 loss: 1.9751 (1.9296) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [285] [160/312] eta: 0:01:55 lr: 0.001279 min_lr: 0.001279 loss: 1.8342 (1.9306) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [285] [170/312] eta: 0:01:47 lr: 0.001279 min_lr: 0.001279 loss: 2.0511 (1.9296) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [285] [180/312] eta: 0:01:39 lr: 0.001278 min_lr: 0.001278 loss: 1.6973 (1.9212) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [285] [190/312] eta: 0:01:31 lr: 0.001278 min_lr: 0.001278 loss: 1.6973 (1.9210) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [285] [200/312] eta: 0:01:23 lr: 0.001277 min_lr: 0.001277 loss: 1.9613 (1.9207) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [285] [210/312] eta: 0:01:15 lr: 0.001277 min_lr: 0.001277 loss: 2.0942 (1.9281) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [285] [220/312] eta: 0:01:08 lr: 0.001277 min_lr: 0.001277 loss: 2.2130 (1.9316) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [285] [230/312] eta: 0:01:00 lr: 0.001276 min_lr: 0.001276 loss: 1.9409 (1.9320) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [285] [240/312] eta: 0:00:52 lr: 0.001276 min_lr: 0.001276 loss: 1.9198 (1.9348) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [285] [250/312] eta: 0:00:45 lr: 0.001275 min_lr: 0.001275 loss: 1.8060 (1.9217) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [285] [260/312] eta: 0:00:38 lr: 0.001275 min_lr: 0.001275 loss: 1.7110 (1.9218) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [285] [270/312] eta: 0:00:30 lr: 0.001274 min_lr: 0.001274 loss: 2.0153 (1.9234) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [285] [280/312] eta: 0:00:23 lr: 0.001274 min_lr: 0.001274 loss: 2.0153 (1.9244) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [285] [290/312] eta: 0:00:16 lr: 0.001274 min_lr: 0.001274 loss: 2.0684 (1.9272) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [285] [300/312] eta: 0:00:08 lr: 0.001273 min_lr: 0.001273 loss: 1.9619 (1.9252) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [285] [310/312] eta: 0:00:01 lr: 0.001273 min_lr: 0.001273 loss: 1.7947 (1.9231) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [285] [311/312] eta: 0:00:00 lr: 0.001273 min_lr: 0.001273 loss: 1.7911 (1.9221) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [285] Total time: 0:03:47 (0.7292 s / it) Averaged stats: lr: 0.001273 min_lr: 0.001273 loss: 1.7911 (1.9539) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4810 (0.4810) acc1: 87.5000 (87.5000) acc5: 97.6562 (97.6562) time: 4.6923 data: 4.4889 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7343 (0.7355) acc1: 80.4688 (80.0320) acc5: 96.3542 (95.8080) time: 0.6726 data: 0.4989 max mem: 64948 Test: Total time: 0:00:06 (0.6986 s / it) * Acc@1 80.850 Acc@5 95.488 loss 0.728 Accuracy of the model on the 50000 test images: 80.9% Max accuracy: 80.89% Test: [0/9] eta: 0:00:42 loss: 0.4790 (0.4790) acc1: 86.4583 (86.4583) acc5: 97.9167 (97.9167) time: 4.6721 data: 4.4582 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6622 (0.6666) acc1: 82.0312 (81.0240) acc5: 97.1354 (96.4800) time: 0.6706 data: 0.4955 max mem: 64948 Test: Total time: 0:00:06 (0.6798 s / it) * Acc@1 82.650 Acc@5 96.318 loss 0.642 Accuracy of the model EMA on 50000 test images: 82.7% Max EMA accuracy: 82.65% Epoch: [286] [ 0/312] eta: 0:47:42 lr: 0.001273 min_lr: 0.001273 loss: 1.4309 (1.4309) weight_decay: 0.0500 (0.0500) time: 9.1749 data: 8.2754 max mem: 64948 Epoch: [286] [ 10/312] eta: 0:07:40 lr: 0.001272 min_lr: 0.001272 loss: 1.9697 (1.8637) weight_decay: 0.0500 (0.0500) time: 1.5264 data: 0.7527 max mem: 64948 Epoch: [286] [ 20/312] eta: 0:05:30 lr: 0.001272 min_lr: 0.001272 loss: 1.9464 (1.8588) weight_decay: 0.0500 (0.0500) time: 0.7301 data: 0.0004 max mem: 64948 Epoch: [286] [ 30/312] eta: 0:04:39 lr: 0.001271 min_lr: 0.001271 loss: 1.8995 (1.8862) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [286] [ 40/312] eta: 0:04:09 lr: 0.001271 min_lr: 0.001271 loss: 2.0130 (1.9116) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [286] [ 50/312] eta: 0:03:49 lr: 0.001270 min_lr: 0.001270 loss: 2.0391 (1.9405) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [286] [ 60/312] eta: 0:03:33 lr: 0.001270 min_lr: 0.001270 loss: 2.0391 (1.9508) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [286] [ 70/312] eta: 0:03:19 lr: 0.001270 min_lr: 0.001270 loss: 2.0026 (1.9497) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [286] [ 80/312] eta: 0:03:07 lr: 0.001269 min_lr: 0.001269 loss: 1.9575 (1.9537) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [286] [ 90/312] eta: 0:02:56 lr: 0.001269 min_lr: 0.001269 loss: 1.9409 (1.9415) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [286] [100/312] eta: 0:02:46 lr: 0.001268 min_lr: 0.001268 loss: 1.8194 (1.9367) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [286] [110/312] eta: 0:02:37 lr: 0.001268 min_lr: 0.001268 loss: 2.0141 (1.9550) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [286] [120/312] eta: 0:02:28 lr: 0.001267 min_lr: 0.001267 loss: 1.9869 (1.9403) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [286] [130/312] eta: 0:02:19 lr: 0.001267 min_lr: 0.001267 loss: 1.7830 (1.9363) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [286] [140/312] eta: 0:02:10 lr: 0.001266 min_lr: 0.001266 loss: 1.9309 (1.9329) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [286] [150/312] eta: 0:02:02 lr: 0.001266 min_lr: 0.001266 loss: 1.9360 (1.9352) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [286] [160/312] eta: 0:01:54 lr: 0.001266 min_lr: 0.001266 loss: 2.0302 (1.9412) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [286] [170/312] eta: 0:01:46 lr: 0.001265 min_lr: 0.001265 loss: 2.0004 (1.9300) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [286] [180/312] eta: 0:01:38 lr: 0.001265 min_lr: 0.001265 loss: 2.0004 (1.9374) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [286] [190/312] eta: 0:01:30 lr: 0.001264 min_lr: 0.001264 loss: 2.0695 (1.9373) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [286] [200/312] eta: 0:01:22 lr: 0.001264 min_lr: 0.001264 loss: 2.1640 (1.9498) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [286] [210/312] eta: 0:01:15 lr: 0.001263 min_lr: 0.001263 loss: 2.1640 (1.9524) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [286] [220/312] eta: 0:01:07 lr: 0.001263 min_lr: 0.001263 loss: 2.0361 (1.9558) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [286] [230/312] eta: 0:01:00 lr: 0.001263 min_lr: 0.001263 loss: 2.0126 (1.9529) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [286] [240/312] eta: 0:00:52 lr: 0.001262 min_lr: 0.001262 loss: 2.0126 (1.9558) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [286] [250/312] eta: 0:00:45 lr: 0.001262 min_lr: 0.001262 loss: 2.1768 (1.9576) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [286] [260/312] eta: 0:00:37 lr: 0.001261 min_lr: 0.001261 loss: 1.9949 (1.9580) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [286] [270/312] eta: 0:00:30 lr: 0.001261 min_lr: 0.001261 loss: 1.8859 (1.9616) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [286] [280/312] eta: 0:00:23 lr: 0.001260 min_lr: 0.001260 loss: 2.0224 (1.9658) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0010 max mem: 64948 Epoch: [286] [290/312] eta: 0:00:15 lr: 0.001260 min_lr: 0.001260 loss: 1.9886 (1.9636) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [286] [300/312] eta: 0:00:08 lr: 0.001260 min_lr: 0.001260 loss: 1.7634 (1.9581) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [286] [310/312] eta: 0:00:01 lr: 0.001259 min_lr: 0.001259 loss: 1.9318 (1.9584) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [286] [311/312] eta: 0:00:00 lr: 0.001259 min_lr: 0.001259 loss: 1.9318 (1.9583) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [286] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.001259 min_lr: 0.001259 loss: 1.9318 (1.9548) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5153 (0.5153) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.5650 data: 4.3453 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7323 (0.7535) acc1: 81.5104 (80.1280) acc5: 95.5729 (95.7120) time: 0.6588 data: 0.4829 max mem: 64948 Test: Total time: 0:00:06 (0.6804 s / it) * Acc@1 80.924 Acc@5 95.358 loss 0.730 Accuracy of the model on the 50000 test images: 80.9% Max accuracy: 80.92% Test: [0/9] eta: 0:00:41 loss: 0.4782 (0.4782) acc1: 86.4583 (86.4583) acc5: 97.9167 (97.9167) time: 4.6618 data: 4.4545 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6613 (0.6662) acc1: 81.7708 (81.0240) acc5: 97.1354 (96.5120) time: 0.6693 data: 0.4951 max mem: 64948 Test: Total time: 0:00:06 (0.6770 s / it) * Acc@1 82.666 Acc@5 96.326 loss 0.641 Accuracy of the model EMA on 50000 test images: 82.7% Max EMA accuracy: 82.67% Epoch: [287] [ 0/312] eta: 0:50:27 lr: 0.001259 min_lr: 0.001259 loss: 1.7852 (1.7852) weight_decay: 0.0500 (0.0500) time: 9.7042 data: 8.8915 max mem: 64948 Epoch: [287] [ 10/312] eta: 0:07:44 lr: 0.001259 min_lr: 0.001259 loss: 2.0741 (2.0048) weight_decay: 0.0500 (0.0500) time: 1.5375 data: 0.8087 max mem: 64948 Epoch: [287] [ 20/312] eta: 0:05:31 lr: 0.001258 min_lr: 0.001258 loss: 2.0410 (1.9718) weight_decay: 0.0500 (0.0500) time: 0.7070 data: 0.0004 max mem: 64948 Epoch: [287] [ 30/312] eta: 0:04:40 lr: 0.001258 min_lr: 0.001258 loss: 1.9387 (1.9480) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [287] [ 40/312] eta: 0:04:10 lr: 0.001257 min_lr: 0.001257 loss: 2.0096 (1.9589) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [287] [ 50/312] eta: 0:03:49 lr: 0.001257 min_lr: 0.001257 loss: 2.1007 (1.9963) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [287] [ 60/312] eta: 0:03:33 lr: 0.001256 min_lr: 0.001256 loss: 2.0627 (1.9954) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [287] [ 70/312] eta: 0:03:20 lr: 0.001256 min_lr: 0.001256 loss: 2.0654 (2.0084) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [287] [ 80/312] eta: 0:03:08 lr: 0.001256 min_lr: 0.001256 loss: 2.1295 (2.0155) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [287] [ 90/312] eta: 0:02:57 lr: 0.001255 min_lr: 0.001255 loss: 2.0042 (1.9799) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [287] [100/312] eta: 0:02:46 lr: 0.001255 min_lr: 0.001255 loss: 1.7290 (1.9839) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [287] [110/312] eta: 0:02:37 lr: 0.001254 min_lr: 0.001254 loss: 2.1593 (1.9954) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [287] [120/312] eta: 0:02:28 lr: 0.001254 min_lr: 0.001254 loss: 2.0400 (1.9928) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [287] [130/312] eta: 0:02:19 lr: 0.001253 min_lr: 0.001253 loss: 1.9583 (1.9902) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [287] [140/312] eta: 0:02:10 lr: 0.001253 min_lr: 0.001253 loss: 1.9807 (1.9883) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [287] [150/312] eta: 0:02:02 lr: 0.001252 min_lr: 0.001252 loss: 1.9728 (1.9748) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [287] [160/312] eta: 0:01:54 lr: 0.001252 min_lr: 0.001252 loss: 1.9728 (1.9745) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [287] [170/312] eta: 0:01:46 lr: 0.001252 min_lr: 0.001252 loss: 1.9143 (1.9594) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [287] [180/312] eta: 0:01:38 lr: 0.001251 min_lr: 0.001251 loss: 2.0083 (1.9612) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [287] [190/312] eta: 0:01:30 lr: 0.001251 min_lr: 0.001251 loss: 1.9715 (1.9456) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [287] [200/312] eta: 0:01:23 lr: 0.001250 min_lr: 0.001250 loss: 1.8806 (1.9489) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [287] [210/312] eta: 0:01:15 lr: 0.001250 min_lr: 0.001250 loss: 1.8806 (1.9460) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [287] [220/312] eta: 0:01:07 lr: 0.001249 min_lr: 0.001249 loss: 2.0930 (1.9514) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [287] [230/312] eta: 0:01:00 lr: 0.001249 min_lr: 0.001249 loss: 2.1331 (1.9532) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [287] [240/312] eta: 0:00:52 lr: 0.001249 min_lr: 0.001249 loss: 2.0699 (1.9572) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [287] [250/312] eta: 0:00:45 lr: 0.001248 min_lr: 0.001248 loss: 2.1275 (1.9583) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [287] [260/312] eta: 0:00:37 lr: 0.001248 min_lr: 0.001248 loss: 2.1419 (1.9601) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [287] [270/312] eta: 0:00:30 lr: 0.001247 min_lr: 0.001247 loss: 2.0341 (1.9624) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [287] [280/312] eta: 0:00:23 lr: 0.001247 min_lr: 0.001247 loss: 1.9447 (1.9617) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [287] [290/312] eta: 0:00:15 lr: 0.001246 min_lr: 0.001246 loss: 2.0106 (1.9633) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [287] [300/312] eta: 0:00:08 lr: 0.001246 min_lr: 0.001246 loss: 2.1070 (1.9649) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [287] [310/312] eta: 0:00:01 lr: 0.001246 min_lr: 0.001246 loss: 1.9337 (1.9615) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [287] [311/312] eta: 0:00:00 lr: 0.001246 min_lr: 0.001246 loss: 1.9337 (1.9618) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [287] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.001246 min_lr: 0.001246 loss: 1.9337 (1.9473) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5399 (0.5399) acc1: 85.6771 (85.6771) acc5: 96.3542 (96.3542) time: 4.5133 data: 4.2967 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7572 (0.7465) acc1: 79.9479 (79.9680) acc5: 96.3542 (95.5840) time: 0.6527 data: 0.4775 max mem: 64948 Test: Total time: 0:00:06 (0.6762 s / it) * Acc@1 80.848 Acc@5 95.422 loss 0.731 Accuracy of the model on the 50000 test images: 80.8% Max accuracy: 80.92% Test: [0/9] eta: 0:00:40 loss: 0.4772 (0.4772) acc1: 86.4583 (86.4583) acc5: 97.9167 (97.9167) time: 4.5260 data: 4.3119 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6599 (0.6660) acc1: 81.7708 (80.9920) acc5: 97.1354 (96.5440) time: 0.6543 data: 0.4792 max mem: 64948 Test: Total time: 0:00:05 (0.6622 s / it) * Acc@1 82.674 Acc@5 96.334 loss 0.641 Accuracy of the model EMA on 50000 test images: 82.7% Max EMA accuracy: 82.67% Epoch: [288] [ 0/312] eta: 0:48:55 lr: 0.001245 min_lr: 0.001245 loss: 1.1474 (1.1474) weight_decay: 0.0500 (0.0500) time: 9.4075 data: 6.9855 max mem: 64948 Epoch: [288] [ 10/312] eta: 0:07:37 lr: 0.001245 min_lr: 0.001245 loss: 1.8702 (1.7738) weight_decay: 0.0500 (0.0500) time: 1.5151 data: 0.6392 max mem: 64948 Epoch: [288] [ 20/312] eta: 0:05:28 lr: 0.001245 min_lr: 0.001245 loss: 1.9134 (1.8605) weight_decay: 0.0500 (0.0500) time: 0.7102 data: 0.0025 max mem: 64948 Epoch: [288] [ 30/312] eta: 0:04:38 lr: 0.001244 min_lr: 0.001244 loss: 1.9134 (1.8492) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [288] [ 40/312] eta: 0:04:09 lr: 0.001244 min_lr: 0.001244 loss: 1.9239 (1.8922) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [288] [ 50/312] eta: 0:03:48 lr: 0.001243 min_lr: 0.001243 loss: 2.0596 (1.9174) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [288] [ 60/312] eta: 0:03:32 lr: 0.001243 min_lr: 0.001243 loss: 2.1366 (1.9345) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [288] [ 70/312] eta: 0:03:19 lr: 0.001242 min_lr: 0.001242 loss: 2.0836 (1.9352) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [288] [ 80/312] eta: 0:03:07 lr: 0.001242 min_lr: 0.001242 loss: 1.8965 (1.9134) weight_decay: 0.0500 (0.0500) time: 0.7013 data: 0.0004 max mem: 64948 Epoch: [288] [ 90/312] eta: 0:02:56 lr: 0.001242 min_lr: 0.001242 loss: 1.8965 (1.9265) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [288] [100/312] eta: 0:02:46 lr: 0.001241 min_lr: 0.001241 loss: 2.0656 (1.9232) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [288] [110/312] eta: 0:02:37 lr: 0.001241 min_lr: 0.001241 loss: 1.9356 (1.9353) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [288] [120/312] eta: 0:02:28 lr: 0.001240 min_lr: 0.001240 loss: 2.1774 (1.9524) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [288] [130/312] eta: 0:02:19 lr: 0.001240 min_lr: 0.001240 loss: 2.0269 (1.9356) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [288] [140/312] eta: 0:02:10 lr: 0.001239 min_lr: 0.001239 loss: 1.7032 (1.9274) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [288] [150/312] eta: 0:02:02 lr: 0.001239 min_lr: 0.001239 loss: 1.7620 (1.9210) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [288] [160/312] eta: 0:01:54 lr: 0.001239 min_lr: 0.001239 loss: 1.8572 (1.9116) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [288] [170/312] eta: 0:01:46 lr: 0.001238 min_lr: 0.001238 loss: 2.0144 (1.9213) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [288] [180/312] eta: 0:01:38 lr: 0.001238 min_lr: 0.001238 loss: 2.0344 (1.9134) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [288] [190/312] eta: 0:01:30 lr: 0.001237 min_lr: 0.001237 loss: 2.0344 (1.9183) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [288] [200/312] eta: 0:01:23 lr: 0.001237 min_lr: 0.001237 loss: 2.0758 (1.9178) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [288] [210/312] eta: 0:01:15 lr: 0.001236 min_lr: 0.001236 loss: 2.0085 (1.9209) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [288] [220/312] eta: 0:01:07 lr: 0.001236 min_lr: 0.001236 loss: 1.9552 (1.9191) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [288] [230/312] eta: 0:01:00 lr: 0.001235 min_lr: 0.001235 loss: 1.9347 (1.9228) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [288] [240/312] eta: 0:00:52 lr: 0.001235 min_lr: 0.001235 loss: 2.0298 (1.9202) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [288] [250/312] eta: 0:00:45 lr: 0.001235 min_lr: 0.001235 loss: 2.0950 (1.9301) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [288] [260/312] eta: 0:00:37 lr: 0.001234 min_lr: 0.001234 loss: 2.1097 (1.9352) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [288] [270/312] eta: 0:00:30 lr: 0.001234 min_lr: 0.001234 loss: 2.0300 (1.9353) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [288] [280/312] eta: 0:00:23 lr: 0.001233 min_lr: 0.001233 loss: 1.9944 (1.9390) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0006 max mem: 64948 Epoch: [288] [290/312] eta: 0:00:15 lr: 0.001233 min_lr: 0.001233 loss: 1.9651 (1.9341) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0005 max mem: 64948 Epoch: [288] [300/312] eta: 0:00:08 lr: 0.001232 min_lr: 0.001232 loss: 1.8901 (1.9306) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [288] [310/312] eta: 0:00:01 lr: 0.001232 min_lr: 0.001232 loss: 1.8475 (1.9257) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [288] [311/312] eta: 0:00:00 lr: 0.001232 min_lr: 0.001232 loss: 1.8429 (1.9255) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [288] Total time: 0:03:46 (0.7275 s / it) Averaged stats: lr: 0.001232 min_lr: 0.001232 loss: 1.8429 (1.9494) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5635 (0.5635) acc1: 84.3750 (84.3750) acc5: 98.4375 (98.4375) time: 4.7005 data: 4.4899 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7316 (0.7418) acc1: 81.2500 (79.8080) acc5: 95.5729 (95.7440) time: 0.6735 data: 0.4990 max mem: 64948 Test: Total time: 0:00:06 (0.6980 s / it) * Acc@1 80.782 Acc@5 95.306 loss 0.739 Accuracy of the model on the 50000 test images: 80.8% Max accuracy: 80.92% Test: [0/9] eta: 0:00:44 loss: 0.4763 (0.4763) acc1: 86.7188 (86.7188) acc5: 97.9167 (97.9167) time: 4.9956 data: 4.7849 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6595 (0.6658) acc1: 81.7708 (81.0560) acc5: 97.1354 (96.5760) time: 0.7131 data: 0.5318 max mem: 64948 Test: Total time: 0:00:06 (0.7246 s / it) * Acc@1 82.706 Acc@5 96.342 loss 0.641 Accuracy of the model EMA on 50000 test images: 82.7% Max EMA accuracy: 82.71% Epoch: [289] [ 0/312] eta: 0:54:53 lr: 0.001232 min_lr: 0.001232 loss: 2.2387 (2.2387) weight_decay: 0.0500 (0.0500) time: 10.5567 data: 9.8223 max mem: 64948 Epoch: [289] [ 10/312] eta: 0:08:05 lr: 0.001232 min_lr: 0.001232 loss: 2.1090 (2.0365) weight_decay: 0.0500 (0.0500) time: 1.6082 data: 0.8932 max mem: 64948 Epoch: [289] [ 20/312] eta: 0:05:42 lr: 0.001231 min_lr: 0.001231 loss: 1.9908 (1.9837) weight_decay: 0.0500 (0.0500) time: 0.7030 data: 0.0003 max mem: 64948 Epoch: [289] [ 30/312] eta: 0:04:47 lr: 0.001231 min_lr: 0.001231 loss: 2.1355 (2.0078) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [289] [ 40/312] eta: 0:04:15 lr: 0.001230 min_lr: 0.001230 loss: 2.0236 (1.9717) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [289] [ 50/312] eta: 0:03:53 lr: 0.001230 min_lr: 0.001230 loss: 1.8602 (1.9459) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [289] [ 60/312] eta: 0:03:36 lr: 0.001229 min_lr: 0.001229 loss: 1.8142 (1.9386) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [289] [ 70/312] eta: 0:03:22 lr: 0.001229 min_lr: 0.001229 loss: 1.9728 (1.9386) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [289] [ 80/312] eta: 0:03:09 lr: 0.001228 min_lr: 0.001228 loss: 2.0750 (1.9444) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [289] [ 90/312] eta: 0:02:58 lr: 0.001228 min_lr: 0.001228 loss: 2.0849 (1.9454) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [289] [100/312] eta: 0:02:48 lr: 0.001228 min_lr: 0.001228 loss: 1.9996 (1.9386) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [289] [110/312] eta: 0:02:38 lr: 0.001227 min_lr: 0.001227 loss: 1.9936 (1.9477) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [289] [120/312] eta: 0:02:29 lr: 0.001227 min_lr: 0.001227 loss: 2.0651 (1.9520) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [289] [130/312] eta: 0:02:20 lr: 0.001226 min_lr: 0.001226 loss: 1.9079 (1.9413) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [289] [140/312] eta: 0:02:11 lr: 0.001226 min_lr: 0.001226 loss: 1.8511 (1.9311) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [289] [150/312] eta: 0:02:03 lr: 0.001225 min_lr: 0.001225 loss: 2.1099 (1.9416) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [289] [160/312] eta: 0:01:55 lr: 0.001225 min_lr: 0.001225 loss: 2.0244 (1.9313) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [289] [170/312] eta: 0:01:46 lr: 0.001225 min_lr: 0.001225 loss: 1.6879 (1.9245) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [289] [180/312] eta: 0:01:39 lr: 0.001224 min_lr: 0.001224 loss: 2.1061 (1.9349) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [289] [190/312] eta: 0:01:31 lr: 0.001224 min_lr: 0.001224 loss: 2.0950 (1.9278) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [289] [200/312] eta: 0:01:23 lr: 0.001223 min_lr: 0.001223 loss: 1.8246 (1.9260) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [289] [210/312] eta: 0:01:15 lr: 0.001223 min_lr: 0.001223 loss: 1.9726 (1.9296) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [289] [220/312] eta: 0:01:08 lr: 0.001222 min_lr: 0.001222 loss: 2.0361 (1.9295) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [289] [230/312] eta: 0:01:00 lr: 0.001222 min_lr: 0.001222 loss: 1.9608 (1.9245) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [289] [240/312] eta: 0:00:52 lr: 0.001222 min_lr: 0.001222 loss: 2.0383 (1.9331) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [289] [250/312] eta: 0:00:45 lr: 0.001221 min_lr: 0.001221 loss: 2.0775 (1.9364) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [289] [260/312] eta: 0:00:38 lr: 0.001221 min_lr: 0.001221 loss: 2.0096 (1.9391) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [289] [270/312] eta: 0:00:30 lr: 0.001220 min_lr: 0.001220 loss: 2.0640 (1.9379) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [289] [280/312] eta: 0:00:23 lr: 0.001220 min_lr: 0.001220 loss: 2.0818 (1.9435) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [289] [290/312] eta: 0:00:16 lr: 0.001219 min_lr: 0.001219 loss: 1.8168 (1.9360) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [289] [300/312] eta: 0:00:08 lr: 0.001219 min_lr: 0.001219 loss: 1.8482 (1.9382) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [289] [310/312] eta: 0:00:01 lr: 0.001219 min_lr: 0.001219 loss: 1.9969 (1.9402) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [289] [311/312] eta: 0:00:00 lr: 0.001219 min_lr: 0.001219 loss: 2.0220 (1.9413) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [289] Total time: 0:03:47 (0.7292 s / it) Averaged stats: lr: 0.001219 min_lr: 0.001219 loss: 2.0220 (1.9413) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5090 (0.5090) acc1: 86.7188 (86.7188) acc5: 97.1354 (97.1354) time: 4.6866 data: 4.4700 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7192 (0.7451) acc1: 81.2500 (80.1920) acc5: 95.8333 (95.2640) time: 0.6720 data: 0.4967 max mem: 64948 Test: Total time: 0:00:06 (0.6962 s / it) * Acc@1 80.926 Acc@5 95.448 loss 0.725 Accuracy of the model on the 50000 test images: 80.9% Max accuracy: 80.93% Test: [0/9] eta: 0:00:40 loss: 0.4761 (0.4761) acc1: 86.7188 (86.7188) acc5: 97.9167 (97.9167) time: 4.4747 data: 4.2568 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6584 (0.6654) acc1: 82.0312 (81.0560) acc5: 97.1354 (96.5760) time: 0.6485 data: 0.4731 max mem: 64948 Test: Total time: 0:00:05 (0.6572 s / it) * Acc@1 82.704 Acc@5 96.348 loss 0.640 Accuracy of the model EMA on 50000 test images: 82.7% Epoch: [290] [ 0/312] eta: 0:55:08 lr: 0.001218 min_lr: 0.001218 loss: 2.3249 (2.3249) weight_decay: 0.0500 (0.0500) time: 10.6043 data: 9.6392 max mem: 64948 Epoch: [290] [ 10/312] eta: 0:08:11 lr: 0.001218 min_lr: 0.001218 loss: 2.0832 (1.9311) weight_decay: 0.0500 (0.0500) time: 1.6260 data: 0.8767 max mem: 64948 Epoch: [290] [ 20/312] eta: 0:05:44 lr: 0.001218 min_lr: 0.001218 loss: 2.0031 (1.9364) weight_decay: 0.0500 (0.0500) time: 0.7103 data: 0.0004 max mem: 64948 Epoch: [290] [ 30/312] eta: 0:04:49 lr: 0.001217 min_lr: 0.001217 loss: 2.0031 (1.9137) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [290] [ 40/312] eta: 0:04:16 lr: 0.001217 min_lr: 0.001217 loss: 1.9806 (1.8963) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [290] [ 50/312] eta: 0:03:54 lr: 0.001216 min_lr: 0.001216 loss: 2.0129 (1.9254) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [290] [ 60/312] eta: 0:03:37 lr: 0.001216 min_lr: 0.001216 loss: 2.0827 (1.9342) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [290] [ 70/312] eta: 0:03:23 lr: 0.001215 min_lr: 0.001215 loss: 2.0943 (1.9546) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [290] [ 80/312] eta: 0:03:10 lr: 0.001215 min_lr: 0.001215 loss: 2.1170 (1.9586) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [290] [ 90/312] eta: 0:02:59 lr: 0.001215 min_lr: 0.001215 loss: 1.8997 (1.9347) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [290] [100/312] eta: 0:02:48 lr: 0.001214 min_lr: 0.001214 loss: 1.6074 (1.9137) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [290] [110/312] eta: 0:02:38 lr: 0.001214 min_lr: 0.001214 loss: 1.9514 (1.9123) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [290] [120/312] eta: 0:02:29 lr: 0.001213 min_lr: 0.001213 loss: 1.9735 (1.9156) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [290] [130/312] eta: 0:02:20 lr: 0.001213 min_lr: 0.001213 loss: 1.9735 (1.9145) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [290] [140/312] eta: 0:02:11 lr: 0.001212 min_lr: 0.001212 loss: 1.8947 (1.9111) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [290] [150/312] eta: 0:02:03 lr: 0.001212 min_lr: 0.001212 loss: 2.0193 (1.9162) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [290] [160/312] eta: 0:01:55 lr: 0.001212 min_lr: 0.001212 loss: 1.9636 (1.9111) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [290] [170/312] eta: 0:01:47 lr: 0.001211 min_lr: 0.001211 loss: 1.9039 (1.9159) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [290] [180/312] eta: 0:01:39 lr: 0.001211 min_lr: 0.001211 loss: 1.9515 (1.9117) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [290] [190/312] eta: 0:01:31 lr: 0.001210 min_lr: 0.001210 loss: 1.8156 (1.8982) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [290] [200/312] eta: 0:01:23 lr: 0.001210 min_lr: 0.001210 loss: 1.8156 (1.8988) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [290] [210/312] eta: 0:01:15 lr: 0.001209 min_lr: 0.001209 loss: 1.9105 (1.9009) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [290] [220/312] eta: 0:01:08 lr: 0.001209 min_lr: 0.001209 loss: 2.1431 (1.9137) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [290] [230/312] eta: 0:01:00 lr: 0.001209 min_lr: 0.001209 loss: 2.1603 (1.9200) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [290] [240/312] eta: 0:00:53 lr: 0.001208 min_lr: 0.001208 loss: 2.0796 (1.9262) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [290] [250/312] eta: 0:00:45 lr: 0.001208 min_lr: 0.001208 loss: 2.0800 (1.9297) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [290] [260/312] eta: 0:00:38 lr: 0.001207 min_lr: 0.001207 loss: 1.9770 (1.9246) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [290] [270/312] eta: 0:00:30 lr: 0.001207 min_lr: 0.001207 loss: 1.9291 (1.9260) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [290] [280/312] eta: 0:00:23 lr: 0.001206 min_lr: 0.001206 loss: 1.9469 (1.9251) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [290] [290/312] eta: 0:00:16 lr: 0.001206 min_lr: 0.001206 loss: 1.9538 (1.9252) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [290] [300/312] eta: 0:00:08 lr: 0.001206 min_lr: 0.001206 loss: 1.9612 (1.9275) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [290] [310/312] eta: 0:00:01 lr: 0.001205 min_lr: 0.001205 loss: 2.1038 (1.9311) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [290] [311/312] eta: 0:00:00 lr: 0.001205 min_lr: 0.001205 loss: 2.1652 (1.9320) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [290] Total time: 0:03:47 (0.7303 s / it) Averaged stats: lr: 0.001205 min_lr: 0.001205 loss: 2.1652 (1.9411) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5735 (0.5735) acc1: 86.1979 (86.1979) acc5: 95.8333 (95.8333) time: 4.6926 data: 4.4749 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7616 (0.7588) acc1: 80.9896 (80.0960) acc5: 95.8333 (95.2640) time: 0.6727 data: 0.4973 max mem: 64948 Test: Total time: 0:00:06 (0.7050 s / it) * Acc@1 80.964 Acc@5 95.376 loss 0.731 Accuracy of the model on the 50000 test images: 81.0% Max accuracy: 80.96% Test: [0/9] eta: 0:00:40 loss: 0.4757 (0.4757) acc1: 86.7188 (86.7188) acc5: 97.9167 (97.9167) time: 4.4931 data: 4.2752 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6576 (0.6650) acc1: 82.0312 (81.1200) acc5: 97.1354 (96.5440) time: 0.6523 data: 0.4751 max mem: 64948 Test: Total time: 0:00:05 (0.6615 s / it) * Acc@1 82.702 Acc@5 96.354 loss 0.640 Accuracy of the model EMA on 50000 test images: 82.7% Epoch: [291] [ 0/312] eta: 0:58:33 lr: 0.001205 min_lr: 0.001205 loss: 1.7872 (1.7872) weight_decay: 0.0500 (0.0500) time: 11.2617 data: 7.3522 max mem: 64948 Epoch: [291] [ 10/312] eta: 0:08:26 lr: 0.001205 min_lr: 0.001205 loss: 1.8030 (1.7979) weight_decay: 0.0500 (0.0500) time: 1.6778 data: 0.6688 max mem: 64948 Epoch: [291] [ 20/312] eta: 0:05:53 lr: 0.001204 min_lr: 0.001204 loss: 1.8833 (1.8331) weight_decay: 0.0500 (0.0500) time: 0.7081 data: 0.0004 max mem: 64948 Epoch: [291] [ 30/312] eta: 0:04:55 lr: 0.001204 min_lr: 0.001204 loss: 1.9692 (1.8453) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [291] [ 40/312] eta: 0:04:21 lr: 0.001203 min_lr: 0.001203 loss: 2.0129 (1.8987) weight_decay: 0.0500 (0.0500) time: 0.7022 data: 0.0004 max mem: 64948 Epoch: [291] [ 50/312] eta: 0:03:58 lr: 0.001203 min_lr: 0.001203 loss: 2.0693 (1.8914) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [291] [ 60/312] eta: 0:03:40 lr: 0.001202 min_lr: 0.001202 loss: 2.0476 (1.9011) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [291] [ 70/312] eta: 0:03:25 lr: 0.001202 min_lr: 0.001202 loss: 2.1206 (1.9440) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [291] [ 80/312] eta: 0:03:12 lr: 0.001202 min_lr: 0.001202 loss: 2.1206 (1.9531) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [291] [ 90/312] eta: 0:03:01 lr: 0.001201 min_lr: 0.001201 loss: 2.1108 (1.9689) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [291] [100/312] eta: 0:02:50 lr: 0.001201 min_lr: 0.001201 loss: 2.1385 (1.9704) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [291] [110/312] eta: 0:02:40 lr: 0.001200 min_lr: 0.001200 loss: 1.9622 (1.9702) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [291] [120/312] eta: 0:02:31 lr: 0.001200 min_lr: 0.001200 loss: 1.8664 (1.9542) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [291] [130/312] eta: 0:02:21 lr: 0.001199 min_lr: 0.001199 loss: 2.0140 (1.9576) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [291] [140/312] eta: 0:02:13 lr: 0.001199 min_lr: 0.001199 loss: 2.1195 (1.9632) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [291] [150/312] eta: 0:02:04 lr: 0.001199 min_lr: 0.001199 loss: 1.9538 (1.9613) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [291] [160/312] eta: 0:01:56 lr: 0.001198 min_lr: 0.001198 loss: 1.9538 (1.9564) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [291] [170/312] eta: 0:01:47 lr: 0.001198 min_lr: 0.001198 loss: 2.0003 (1.9585) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [291] [180/312] eta: 0:01:39 lr: 0.001197 min_lr: 0.001197 loss: 2.0207 (1.9578) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [291] [190/312] eta: 0:01:31 lr: 0.001197 min_lr: 0.001197 loss: 2.0052 (1.9543) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [291] [200/312] eta: 0:01:24 lr: 0.001196 min_lr: 0.001196 loss: 2.0316 (1.9558) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [291] [210/312] eta: 0:01:16 lr: 0.001196 min_lr: 0.001196 loss: 2.0532 (1.9500) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [291] [220/312] eta: 0:01:08 lr: 0.001196 min_lr: 0.001196 loss: 1.8047 (1.9449) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [291] [230/312] eta: 0:01:00 lr: 0.001195 min_lr: 0.001195 loss: 1.7472 (1.9376) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [291] [240/312] eta: 0:00:53 lr: 0.001195 min_lr: 0.001195 loss: 1.7821 (1.9369) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [291] [250/312] eta: 0:00:45 lr: 0.001194 min_lr: 0.001194 loss: 2.0203 (1.9383) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [291] [260/312] eta: 0:00:38 lr: 0.001194 min_lr: 0.001194 loss: 2.0587 (1.9430) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0004 max mem: 64948 Epoch: [291] [270/312] eta: 0:00:30 lr: 0.001193 min_lr: 0.001193 loss: 1.9873 (1.9435) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [291] [280/312] eta: 0:00:23 lr: 0.001193 min_lr: 0.001193 loss: 1.9807 (1.9440) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0009 max mem: 64948 Epoch: [291] [290/312] eta: 0:00:16 lr: 0.001193 min_lr: 0.001193 loss: 2.0548 (1.9476) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0008 max mem: 64948 Epoch: [291] [300/312] eta: 0:00:08 lr: 0.001192 min_lr: 0.001192 loss: 2.0557 (1.9499) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [291] [310/312] eta: 0:00:01 lr: 0.001192 min_lr: 0.001192 loss: 1.9377 (1.9456) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [291] [311/312] eta: 0:00:00 lr: 0.001192 min_lr: 0.001192 loss: 1.9377 (1.9464) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [291] Total time: 0:03:48 (0.7327 s / it) Averaged stats: lr: 0.001192 min_lr: 0.001192 loss: 1.9377 (1.9409) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5234 (0.5234) acc1: 86.7188 (86.7188) acc5: 96.8750 (96.8750) time: 4.5648 data: 4.3451 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7345 (0.7322) acc1: 78.9062 (80.1600) acc5: 96.0938 (95.6160) time: 0.6585 data: 0.4829 max mem: 64948 Test: Total time: 0:00:06 (0.6803 s / it) * Acc@1 80.956 Acc@5 95.480 loss 0.726 Accuracy of the model on the 50000 test images: 81.0% Max accuracy: 80.96% Test: [0/9] eta: 0:00:42 loss: 0.4749 (0.4749) acc1: 86.7188 (86.7188) acc5: 97.9167 (97.9167) time: 4.7372 data: 4.5191 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6563 (0.6644) acc1: 82.0312 (81.1840) acc5: 97.1354 (96.5440) time: 0.6780 data: 0.5022 max mem: 64948 Test: Total time: 0:00:06 (0.6883 s / it) * Acc@1 82.718 Acc@5 96.358 loss 0.640 Accuracy of the model EMA on 50000 test images: 82.7% Max EMA accuracy: 82.72% Epoch: [292] [ 0/312] eta: 0:46:37 lr: 0.001192 min_lr: 0.001192 loss: 2.5233 (2.5233) weight_decay: 0.0500 (0.0500) time: 8.9666 data: 8.1731 max mem: 64948 Epoch: [292] [ 10/312] eta: 0:07:24 lr: 0.001191 min_lr: 0.001191 loss: 1.8049 (1.8667) weight_decay: 0.0500 (0.0500) time: 1.4710 data: 0.7434 max mem: 64948 Epoch: [292] [ 20/312] eta: 0:05:21 lr: 0.001191 min_lr: 0.001191 loss: 1.8049 (1.8439) weight_decay: 0.0500 (0.0500) time: 0.7072 data: 0.0004 max mem: 64948 Epoch: [292] [ 30/312] eta: 0:04:33 lr: 0.001190 min_lr: 0.001190 loss: 2.0096 (1.9123) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [292] [ 40/312] eta: 0:04:05 lr: 0.001190 min_lr: 0.001190 loss: 2.1439 (1.9648) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [292] [ 50/312] eta: 0:03:46 lr: 0.001190 min_lr: 0.001190 loss: 2.0601 (1.9717) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [292] [ 60/312] eta: 0:03:30 lr: 0.001189 min_lr: 0.001189 loss: 2.0300 (1.9734) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [292] [ 70/312] eta: 0:03:17 lr: 0.001189 min_lr: 0.001189 loss: 2.0631 (1.9856) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [292] [ 80/312] eta: 0:03:05 lr: 0.001188 min_lr: 0.001188 loss: 2.0203 (1.9621) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [292] [ 90/312] eta: 0:02:55 lr: 0.001188 min_lr: 0.001188 loss: 1.7772 (1.9450) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [292] [100/312] eta: 0:02:45 lr: 0.001187 min_lr: 0.001187 loss: 1.8375 (1.9556) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [292] [110/312] eta: 0:02:36 lr: 0.001187 min_lr: 0.001187 loss: 1.9801 (1.9464) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [292] [120/312] eta: 0:02:27 lr: 0.001187 min_lr: 0.001187 loss: 1.8665 (1.9307) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [292] [130/312] eta: 0:02:18 lr: 0.001186 min_lr: 0.001186 loss: 1.9170 (1.9296) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [292] [140/312] eta: 0:02:10 lr: 0.001186 min_lr: 0.001186 loss: 1.9593 (1.9285) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [292] [150/312] eta: 0:02:01 lr: 0.001185 min_lr: 0.001185 loss: 2.0128 (1.9301) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [292] [160/312] eta: 0:01:53 lr: 0.001185 min_lr: 0.001185 loss: 2.0116 (1.9357) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [292] [170/312] eta: 0:01:45 lr: 0.001184 min_lr: 0.001184 loss: 1.9883 (1.9420) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [292] [180/312] eta: 0:01:38 lr: 0.001184 min_lr: 0.001184 loss: 1.9637 (1.9353) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [292] [190/312] eta: 0:01:30 lr: 0.001184 min_lr: 0.001184 loss: 1.9637 (1.9328) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [292] [200/312] eta: 0:01:22 lr: 0.001183 min_lr: 0.001183 loss: 2.0486 (1.9435) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [292] [210/312] eta: 0:01:15 lr: 0.001183 min_lr: 0.001183 loss: 2.0837 (1.9421) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [292] [220/312] eta: 0:01:07 lr: 0.001182 min_lr: 0.001182 loss: 2.0837 (1.9452) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [292] [230/312] eta: 0:01:00 lr: 0.001182 min_lr: 0.001182 loss: 1.8082 (1.9335) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [292] [240/312] eta: 0:00:52 lr: 0.001181 min_lr: 0.001181 loss: 1.8082 (1.9363) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [292] [250/312] eta: 0:00:45 lr: 0.001181 min_lr: 0.001181 loss: 1.7333 (1.9241) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [292] [260/312] eta: 0:00:37 lr: 0.001181 min_lr: 0.001181 loss: 1.6558 (1.9184) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [292] [270/312] eta: 0:00:30 lr: 0.001180 min_lr: 0.001180 loss: 1.9331 (1.9212) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [292] [280/312] eta: 0:00:23 lr: 0.001180 min_lr: 0.001180 loss: 1.9578 (1.9218) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [292] [290/312] eta: 0:00:15 lr: 0.001179 min_lr: 0.001179 loss: 1.8978 (1.9188) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [292] [300/312] eta: 0:00:08 lr: 0.001179 min_lr: 0.001179 loss: 1.9650 (1.9206) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [292] [310/312] eta: 0:00:01 lr: 0.001178 min_lr: 0.001178 loss: 1.9571 (1.9195) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [292] [311/312] eta: 0:00:00 lr: 0.001178 min_lr: 0.001178 loss: 1.9650 (1.9197) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [292] Total time: 0:03:46 (0.7249 s / it) Averaged stats: lr: 0.001178 min_lr: 0.001178 loss: 1.9650 (1.9433) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5256 (0.5256) acc1: 85.9375 (85.9375) acc5: 98.4375 (98.4375) time: 4.6189 data: 4.3977 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7460 (0.7547) acc1: 80.9896 (79.4880) acc5: 95.8333 (95.3600) time: 0.6647 data: 0.4887 max mem: 64948 Test: Total time: 0:00:06 (0.6880 s / it) * Acc@1 80.664 Acc@5 95.478 loss 0.732 Accuracy of the model on the 50000 test images: 80.7% Max accuracy: 80.96% Test: [0/9] eta: 0:00:42 loss: 0.4747 (0.4747) acc1: 86.7188 (86.7188) acc5: 97.9167 (97.9167) time: 4.7650 data: 4.5467 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6553 (0.6641) acc1: 82.2917 (81.1520) acc5: 97.1354 (96.5120) time: 0.6809 data: 0.5053 max mem: 64948 Test: Total time: 0:00:06 (0.6932 s / it) * Acc@1 82.700 Acc@5 96.366 loss 0.639 Accuracy of the model EMA on 50000 test images: 82.7% Epoch: [293] [ 0/312] eta: 0:55:18 lr: 0.001178 min_lr: 0.001178 loss: 1.5703 (1.5703) weight_decay: 0.0500 (0.0500) time: 10.6366 data: 9.0312 max mem: 64948 Epoch: [293] [ 10/312] eta: 0:08:16 lr: 0.001178 min_lr: 0.001178 loss: 1.7678 (1.8441) weight_decay: 0.0500 (0.0500) time: 1.6427 data: 0.8214 max mem: 64948 Epoch: [293] [ 20/312] eta: 0:05:48 lr: 0.001177 min_lr: 0.001177 loss: 1.8335 (1.8974) weight_decay: 0.0500 (0.0500) time: 0.7204 data: 0.0004 max mem: 64948 Epoch: [293] [ 30/312] eta: 0:04:51 lr: 0.001177 min_lr: 0.001177 loss: 1.9491 (1.9066) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [293] [ 40/312] eta: 0:04:18 lr: 0.001177 min_lr: 0.001177 loss: 2.0425 (1.9376) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [293] [ 50/312] eta: 0:03:55 lr: 0.001176 min_lr: 0.001176 loss: 2.0984 (1.9498) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [293] [ 60/312] eta: 0:03:38 lr: 0.001176 min_lr: 0.001176 loss: 2.0865 (1.9497) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [293] [ 70/312] eta: 0:03:23 lr: 0.001175 min_lr: 0.001175 loss: 1.9920 (1.9443) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [293] [ 80/312] eta: 0:03:11 lr: 0.001175 min_lr: 0.001175 loss: 1.8670 (1.9439) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [293] [ 90/312] eta: 0:02:59 lr: 0.001175 min_lr: 0.001175 loss: 1.8210 (1.9253) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [293] [100/312] eta: 0:02:49 lr: 0.001174 min_lr: 0.001174 loss: 1.8482 (1.9279) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [293] [110/312] eta: 0:02:39 lr: 0.001174 min_lr: 0.001174 loss: 2.1022 (1.9368) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [293] [120/312] eta: 0:02:30 lr: 0.001173 min_lr: 0.001173 loss: 2.0712 (1.9408) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [293] [130/312] eta: 0:02:21 lr: 0.001173 min_lr: 0.001173 loss: 1.9970 (1.9500) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [293] [140/312] eta: 0:02:12 lr: 0.001172 min_lr: 0.001172 loss: 1.9902 (1.9404) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [293] [150/312] eta: 0:02:03 lr: 0.001172 min_lr: 0.001172 loss: 1.9992 (1.9474) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [293] [160/312] eta: 0:01:55 lr: 0.001172 min_lr: 0.001172 loss: 2.1154 (1.9543) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [293] [170/312] eta: 0:01:47 lr: 0.001171 min_lr: 0.001171 loss: 2.1103 (1.9496) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [293] [180/312] eta: 0:01:39 lr: 0.001171 min_lr: 0.001171 loss: 2.1197 (1.9554) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [293] [190/312] eta: 0:01:31 lr: 0.001170 min_lr: 0.001170 loss: 2.0077 (1.9459) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [293] [200/312] eta: 0:01:23 lr: 0.001170 min_lr: 0.001170 loss: 2.0061 (1.9511) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [293] [210/312] eta: 0:01:15 lr: 0.001169 min_lr: 0.001169 loss: 1.8855 (1.9462) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [293] [220/312] eta: 0:01:08 lr: 0.001169 min_lr: 0.001169 loss: 1.8241 (1.9467) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [293] [230/312] eta: 0:01:00 lr: 0.001169 min_lr: 0.001169 loss: 1.8241 (1.9408) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [293] [240/312] eta: 0:00:53 lr: 0.001168 min_lr: 0.001168 loss: 2.0498 (1.9446) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [293] [250/312] eta: 0:00:45 lr: 0.001168 min_lr: 0.001168 loss: 2.0869 (1.9450) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [293] [260/312] eta: 0:00:38 lr: 0.001167 min_lr: 0.001167 loss: 1.9583 (1.9431) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [293] [270/312] eta: 0:00:30 lr: 0.001167 min_lr: 0.001167 loss: 1.9785 (1.9450) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [293] [280/312] eta: 0:00:23 lr: 0.001166 min_lr: 0.001166 loss: 2.0644 (1.9403) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0009 max mem: 64948 Epoch: [293] [290/312] eta: 0:00:16 lr: 0.001166 min_lr: 0.001166 loss: 1.9420 (1.9411) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [293] [300/312] eta: 0:00:08 lr: 0.001166 min_lr: 0.001166 loss: 1.9904 (1.9429) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [293] [310/312] eta: 0:00:01 lr: 0.001165 min_lr: 0.001165 loss: 1.7877 (1.9363) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [293] [311/312] eta: 0:00:00 lr: 0.001165 min_lr: 0.001165 loss: 1.7877 (1.9376) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [293] Total time: 0:03:48 (0.7308 s / it) Averaged stats: lr: 0.001165 min_lr: 0.001165 loss: 1.7877 (1.9282) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5101 (0.5101) acc1: 86.7188 (86.7188) acc5: 96.8750 (96.8750) time: 4.5439 data: 4.3267 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7388 (0.7392) acc1: 80.4688 (80.2560) acc5: 95.8333 (95.4560) time: 0.6562 data: 0.4808 max mem: 64948 Test: Total time: 0:00:06 (0.6776 s / it) * Acc@1 80.996 Acc@5 95.528 loss 0.722 Accuracy of the model on the 50000 test images: 81.0% Max accuracy: 81.00% Test: [0/9] eta: 0:00:40 loss: 0.4744 (0.4744) acc1: 86.7188 (86.7188) acc5: 97.9167 (97.9167) time: 4.5304 data: 4.3127 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6536 (0.6636) acc1: 82.2917 (81.1200) acc5: 97.1354 (96.5120) time: 0.6554 data: 0.4793 max mem: 64948 Test: Total time: 0:00:05 (0.6644 s / it) * Acc@1 82.716 Acc@5 96.378 loss 0.639 Accuracy of the model EMA on 50000 test images: 82.7% Epoch: [294] [ 0/312] eta: 0:54:33 lr: 0.001165 min_lr: 0.001165 loss: 2.0062 (2.0062) weight_decay: 0.0500 (0.0500) time: 10.4934 data: 9.5210 max mem: 64948 Epoch: [294] [ 10/312] eta: 0:08:07 lr: 0.001165 min_lr: 0.001165 loss: 2.0062 (1.9145) weight_decay: 0.0500 (0.0500) time: 1.6134 data: 0.8659 max mem: 64948 Epoch: [294] [ 20/312] eta: 0:05:44 lr: 0.001164 min_lr: 0.001164 loss: 1.7186 (1.8383) weight_decay: 0.0500 (0.0500) time: 0.7126 data: 0.0004 max mem: 64948 Epoch: [294] [ 30/312] eta: 0:04:48 lr: 0.001164 min_lr: 0.001164 loss: 1.9870 (1.8864) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0003 max mem: 64948 Epoch: [294] [ 40/312] eta: 0:04:16 lr: 0.001163 min_lr: 0.001163 loss: 1.8799 (1.8634) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [294] [ 50/312] eta: 0:03:54 lr: 0.001163 min_lr: 0.001163 loss: 1.8122 (1.8691) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [294] [ 60/312] eta: 0:03:37 lr: 0.001162 min_lr: 0.001162 loss: 1.8645 (1.8610) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [294] [ 70/312] eta: 0:03:23 lr: 0.001162 min_lr: 0.001162 loss: 1.9652 (1.8607) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [294] [ 80/312] eta: 0:03:10 lr: 0.001162 min_lr: 0.001162 loss: 1.9609 (1.8567) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [294] [ 90/312] eta: 0:02:59 lr: 0.001161 min_lr: 0.001161 loss: 1.7930 (1.8568) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [294] [100/312] eta: 0:02:48 lr: 0.001161 min_lr: 0.001161 loss: 1.8787 (1.8588) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [294] [110/312] eta: 0:02:39 lr: 0.001160 min_lr: 0.001160 loss: 2.0370 (1.8710) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [294] [120/312] eta: 0:02:29 lr: 0.001160 min_lr: 0.001160 loss: 2.0370 (1.8637) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [294] [130/312] eta: 0:02:20 lr: 0.001160 min_lr: 0.001160 loss: 1.6291 (1.8525) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [294] [140/312] eta: 0:02:12 lr: 0.001159 min_lr: 0.001159 loss: 1.7459 (1.8519) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [294] [150/312] eta: 0:02:03 lr: 0.001159 min_lr: 0.001159 loss: 1.8730 (1.8504) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [294] [160/312] eta: 0:01:55 lr: 0.001158 min_lr: 0.001158 loss: 1.9899 (1.8550) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [294] [170/312] eta: 0:01:47 lr: 0.001158 min_lr: 0.001158 loss: 2.0189 (1.8622) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [294] [180/312] eta: 0:01:39 lr: 0.001157 min_lr: 0.001157 loss: 1.9774 (1.8635) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [294] [190/312] eta: 0:01:31 lr: 0.001157 min_lr: 0.001157 loss: 1.8955 (1.8668) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [294] [200/312] eta: 0:01:23 lr: 0.001157 min_lr: 0.001157 loss: 1.7545 (1.8669) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [294] [210/312] eta: 0:01:15 lr: 0.001156 min_lr: 0.001156 loss: 1.8344 (1.8724) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [294] [220/312] eta: 0:01:08 lr: 0.001156 min_lr: 0.001156 loss: 1.9812 (1.8745) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [294] [230/312] eta: 0:01:00 lr: 0.001155 min_lr: 0.001155 loss: 1.9037 (1.8753) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [294] [240/312] eta: 0:00:53 lr: 0.001155 min_lr: 0.001155 loss: 1.9970 (1.8769) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [294] [250/312] eta: 0:00:45 lr: 0.001154 min_lr: 0.001154 loss: 1.9970 (1.8812) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [294] [260/312] eta: 0:00:38 lr: 0.001154 min_lr: 0.001154 loss: 2.0594 (1.8862) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [294] [270/312] eta: 0:00:30 lr: 0.001154 min_lr: 0.001154 loss: 2.1218 (1.8888) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [294] [280/312] eta: 0:00:23 lr: 0.001153 min_lr: 0.001153 loss: 2.1320 (1.8927) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [294] [290/312] eta: 0:00:16 lr: 0.001153 min_lr: 0.001153 loss: 2.1038 (1.8969) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [294] [300/312] eta: 0:00:08 lr: 0.001152 min_lr: 0.001152 loss: 1.7736 (1.8873) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [294] [310/312] eta: 0:00:01 lr: 0.001152 min_lr: 0.001152 loss: 1.8337 (1.8852) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [294] [311/312] eta: 0:00:00 lr: 0.001152 min_lr: 0.001152 loss: 1.8337 (1.8875) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [294] Total time: 0:03:47 (0.7300 s / it) Averaged stats: lr: 0.001152 min_lr: 0.001152 loss: 1.8337 (1.9256) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4941 (0.4941) acc1: 86.4583 (86.4583) acc5: 97.9167 (97.9167) time: 4.7064 data: 4.4874 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7057 (0.7369) acc1: 78.9062 (79.9360) acc5: 96.0938 (95.7440) time: 0.6742 data: 0.4987 max mem: 64948 Test: Total time: 0:00:06 (0.6972 s / it) * Acc@1 81.224 Acc@5 95.566 loss 0.719 Accuracy of the model on the 50000 test images: 81.2% Max accuracy: 81.22% Test: [0/9] eta: 0:00:40 loss: 0.4741 (0.4741) acc1: 86.7188 (86.7188) acc5: 97.9167 (97.9167) time: 4.4977 data: 4.2798 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6527 (0.6631) acc1: 82.2917 (81.1840) acc5: 97.1354 (96.5120) time: 0.6519 data: 0.4757 max mem: 64948 Test: Total time: 0:00:05 (0.6594 s / it) * Acc@1 82.740 Acc@5 96.392 loss 0.639 Accuracy of the model EMA on 50000 test images: 82.7% Max EMA accuracy: 82.74% Epoch: [295] [ 0/312] eta: 0:52:02 lr: 0.001152 min_lr: 0.001152 loss: 1.9231 (1.9231) weight_decay: 0.0500 (0.0500) time: 10.0083 data: 9.2183 max mem: 64948 Epoch: [295] [ 10/312] eta: 0:07:50 lr: 0.001151 min_lr: 0.001151 loss: 1.9231 (1.9303) weight_decay: 0.0500 (0.0500) time: 1.5587 data: 0.8384 max mem: 64948 Epoch: [295] [ 20/312] eta: 0:05:34 lr: 0.001151 min_lr: 0.001151 loss: 1.8720 (1.9147) weight_decay: 0.0500 (0.0500) time: 0.7032 data: 0.0004 max mem: 64948 Epoch: [295] [ 30/312] eta: 0:04:42 lr: 0.001151 min_lr: 0.001151 loss: 1.9126 (1.9443) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [295] [ 40/312] eta: 0:04:12 lr: 0.001150 min_lr: 0.001150 loss: 1.9148 (1.9318) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [295] [ 50/312] eta: 0:03:50 lr: 0.001150 min_lr: 0.001150 loss: 1.9380 (1.9480) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [295] [ 60/312] eta: 0:03:34 lr: 0.001149 min_lr: 0.001149 loss: 2.0364 (1.9312) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [295] [ 70/312] eta: 0:03:20 lr: 0.001149 min_lr: 0.001149 loss: 1.8533 (1.9024) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [295] [ 80/312] eta: 0:03:08 lr: 0.001148 min_lr: 0.001148 loss: 1.8533 (1.9073) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0004 max mem: 64948 Epoch: [295] [ 90/312] eta: 0:02:57 lr: 0.001148 min_lr: 0.001148 loss: 1.9565 (1.9047) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [295] [100/312] eta: 0:02:47 lr: 0.001148 min_lr: 0.001148 loss: 1.9749 (1.9079) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [295] [110/312] eta: 0:02:37 lr: 0.001147 min_lr: 0.001147 loss: 2.0844 (1.9098) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [295] [120/312] eta: 0:02:28 lr: 0.001147 min_lr: 0.001147 loss: 2.0326 (1.9144) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [295] [130/312] eta: 0:02:19 lr: 0.001146 min_lr: 0.001146 loss: 2.0963 (1.9255) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [295] [140/312] eta: 0:02:11 lr: 0.001146 min_lr: 0.001146 loss: 2.0674 (1.9230) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [295] [150/312] eta: 0:02:02 lr: 0.001145 min_lr: 0.001145 loss: 1.9047 (1.9165) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [295] [160/312] eta: 0:01:54 lr: 0.001145 min_lr: 0.001145 loss: 1.9753 (1.9228) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [295] [170/312] eta: 0:01:46 lr: 0.001145 min_lr: 0.001145 loss: 2.0805 (1.9241) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [295] [180/312] eta: 0:01:38 lr: 0.001144 min_lr: 0.001144 loss: 1.9135 (1.9252) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [295] [190/312] eta: 0:01:30 lr: 0.001144 min_lr: 0.001144 loss: 1.9584 (1.9246) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [295] [200/312] eta: 0:01:23 lr: 0.001143 min_lr: 0.001143 loss: 2.0537 (1.9258) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [295] [210/312] eta: 0:01:15 lr: 0.001143 min_lr: 0.001143 loss: 1.9784 (1.9216) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [295] [220/312] eta: 0:01:07 lr: 0.001142 min_lr: 0.001142 loss: 1.9467 (1.9252) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [295] [230/312] eta: 0:01:00 lr: 0.001142 min_lr: 0.001142 loss: 2.0497 (1.9331) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [295] [240/312] eta: 0:00:52 lr: 0.001142 min_lr: 0.001142 loss: 1.9864 (1.9315) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [295] [250/312] eta: 0:00:45 lr: 0.001141 min_lr: 0.001141 loss: 1.7587 (1.9199) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [295] [260/312] eta: 0:00:38 lr: 0.001141 min_lr: 0.001141 loss: 1.6909 (1.9138) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [295] [270/312] eta: 0:00:30 lr: 0.001140 min_lr: 0.001140 loss: 1.9159 (1.9159) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [295] [280/312] eta: 0:00:23 lr: 0.001140 min_lr: 0.001140 loss: 2.0309 (1.9200) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [295] [290/312] eta: 0:00:16 lr: 0.001140 min_lr: 0.001140 loss: 1.9758 (1.9194) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [295] [300/312] eta: 0:00:08 lr: 0.001139 min_lr: 0.001139 loss: 1.9150 (1.9188) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [295] [310/312] eta: 0:00:01 lr: 0.001139 min_lr: 0.001139 loss: 1.9589 (1.9207) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [295] [311/312] eta: 0:00:00 lr: 0.001139 min_lr: 0.001139 loss: 1.9589 (1.9216) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [295] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.001139 min_lr: 0.001139 loss: 1.9589 (1.9273) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.5236 (0.5236) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.8006 data: 4.5809 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7348 (0.7490) acc1: 80.2083 (79.7760) acc5: 96.3542 (95.8080) time: 0.6852 data: 0.5091 max mem: 64948 Test: Total time: 0:00:06 (0.7093 s / it) * Acc@1 81.052 Acc@5 95.462 loss 0.735 Accuracy of the model on the 50000 test images: 81.1% Max accuracy: 81.22% Test: [0/9] eta: 0:00:43 loss: 0.4733 (0.4733) acc1: 86.9792 (86.9792) acc5: 97.9167 (97.9167) time: 4.8724 data: 4.6609 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6516 (0.6626) acc1: 82.2917 (81.1840) acc5: 97.1354 (96.4800) time: 0.6927 data: 0.5180 max mem: 64948 Test: Total time: 0:00:06 (0.7045 s / it) * Acc@1 82.752 Acc@5 96.396 loss 0.638 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.75% Epoch: [296] [ 0/312] eta: 0:47:03 lr: 0.001139 min_lr: 0.001139 loss: 1.4047 (1.4047) weight_decay: 0.0500 (0.0500) time: 9.0488 data: 8.2763 max mem: 64948 Epoch: [296] [ 10/312] eta: 0:07:28 lr: 0.001138 min_lr: 0.001138 loss: 1.9662 (1.9255) weight_decay: 0.0500 (0.0500) time: 1.4840 data: 0.7565 max mem: 64948 Epoch: [296] [ 20/312] eta: 0:05:23 lr: 0.001138 min_lr: 0.001138 loss: 1.9115 (1.8082) weight_decay: 0.0500 (0.0500) time: 0.7110 data: 0.0024 max mem: 64948 Epoch: [296] [ 30/312] eta: 0:04:34 lr: 0.001137 min_lr: 0.001137 loss: 1.7710 (1.8605) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [296] [ 40/312] eta: 0:04:06 lr: 0.001137 min_lr: 0.001137 loss: 2.0178 (1.8699) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [296] [ 50/312] eta: 0:03:46 lr: 0.001136 min_lr: 0.001136 loss: 1.8644 (1.8684) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [296] [ 60/312] eta: 0:03:30 lr: 0.001136 min_lr: 0.001136 loss: 1.8636 (1.8829) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [296] [ 70/312] eta: 0:03:17 lr: 0.001136 min_lr: 0.001136 loss: 1.6711 (1.8396) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [296] [ 80/312] eta: 0:03:06 lr: 0.001135 min_lr: 0.001135 loss: 1.5526 (1.8332) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [296] [ 90/312] eta: 0:02:55 lr: 0.001135 min_lr: 0.001135 loss: 1.8863 (1.8408) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [296] [100/312] eta: 0:02:45 lr: 0.001134 min_lr: 0.001134 loss: 1.9180 (1.8482) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [296] [110/312] eta: 0:02:36 lr: 0.001134 min_lr: 0.001134 loss: 2.1104 (1.8750) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [296] [120/312] eta: 0:02:27 lr: 0.001134 min_lr: 0.001134 loss: 2.0367 (1.8698) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [296] [130/312] eta: 0:02:18 lr: 0.001133 min_lr: 0.001133 loss: 1.8153 (1.8650) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [296] [140/312] eta: 0:02:10 lr: 0.001133 min_lr: 0.001133 loss: 1.9462 (1.8638) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [296] [150/312] eta: 0:02:01 lr: 0.001132 min_lr: 0.001132 loss: 1.9479 (1.8664) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [296] [160/312] eta: 0:01:53 lr: 0.001132 min_lr: 0.001132 loss: 1.9264 (1.8698) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [296] [170/312] eta: 0:01:45 lr: 0.001131 min_lr: 0.001131 loss: 1.9916 (1.8683) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [296] [180/312] eta: 0:01:38 lr: 0.001131 min_lr: 0.001131 loss: 1.9653 (1.8721) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [296] [190/312] eta: 0:01:30 lr: 0.001131 min_lr: 0.001131 loss: 1.9153 (1.8669) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [296] [200/312] eta: 0:01:22 lr: 0.001130 min_lr: 0.001130 loss: 1.9406 (1.8740) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [296] [210/312] eta: 0:01:15 lr: 0.001130 min_lr: 0.001130 loss: 1.9887 (1.8773) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [296] [220/312] eta: 0:01:07 lr: 0.001129 min_lr: 0.001129 loss: 1.9779 (1.8773) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [296] [230/312] eta: 0:01:00 lr: 0.001129 min_lr: 0.001129 loss: 1.9779 (1.8804) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [296] [240/312] eta: 0:00:52 lr: 0.001128 min_lr: 0.001128 loss: 1.8936 (1.8763) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [296] [250/312] eta: 0:00:45 lr: 0.001128 min_lr: 0.001128 loss: 1.5991 (1.8672) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [296] [260/312] eta: 0:00:37 lr: 0.001128 min_lr: 0.001128 loss: 1.6869 (1.8685) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [296] [270/312] eta: 0:00:30 lr: 0.001127 min_lr: 0.001127 loss: 2.0106 (1.8744) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [296] [280/312] eta: 0:00:23 lr: 0.001127 min_lr: 0.001127 loss: 2.0107 (1.8760) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [296] [290/312] eta: 0:00:15 lr: 0.001126 min_lr: 0.001126 loss: 2.0206 (1.8816) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [296] [300/312] eta: 0:00:08 lr: 0.001126 min_lr: 0.001126 loss: 2.0300 (1.8801) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [296] [310/312] eta: 0:00:01 lr: 0.001126 min_lr: 0.001126 loss: 2.0538 (1.8844) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [296] [311/312] eta: 0:00:00 lr: 0.001125 min_lr: 0.001125 loss: 2.0707 (1.8852) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [296] Total time: 0:03:46 (0.7254 s / it) Averaged stats: lr: 0.001125 min_lr: 0.001125 loss: 2.0707 (1.9321) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5047 (0.5047) acc1: 86.9792 (86.9792) acc5: 97.3958 (97.3958) time: 4.5471 data: 4.3360 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7729 (0.7318) acc1: 80.9896 (80.1600) acc5: 96.0938 (95.5840) time: 0.6565 data: 0.4819 max mem: 64948 Test: Total time: 0:00:06 (0.6778 s / it) * Acc@1 81.096 Acc@5 95.550 loss 0.719 Accuracy of the model on the 50000 test images: 81.1% Max accuracy: 81.22% Test: [0/9] eta: 0:00:44 loss: 0.4729 (0.4729) acc1: 86.9792 (86.9792) acc5: 97.9167 (97.9167) time: 4.9377 data: 4.7325 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6509 (0.6621) acc1: 82.2917 (81.1520) acc5: 97.1354 (96.4800) time: 0.6999 data: 0.5259 max mem: 64948 Test: Total time: 0:00:06 (0.7113 s / it) * Acc@1 82.756 Acc@5 96.410 loss 0.638 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.76% Epoch: [297] [ 0/312] eta: 0:47:12 lr: 0.001125 min_lr: 0.001125 loss: 2.4577 (2.4577) weight_decay: 0.0500 (0.0500) time: 9.0798 data: 7.2044 max mem: 64948 Epoch: [297] [ 10/312] eta: 0:07:27 lr: 0.001125 min_lr: 0.001125 loss: 2.1094 (2.1556) weight_decay: 0.0500 (0.0500) time: 1.4817 data: 0.6554 max mem: 64948 Epoch: [297] [ 20/312] eta: 0:05:23 lr: 0.001125 min_lr: 0.001125 loss: 2.0316 (1.9859) weight_decay: 0.0500 (0.0500) time: 0.7093 data: 0.0004 max mem: 64948 Epoch: [297] [ 30/312] eta: 0:04:35 lr: 0.001124 min_lr: 0.001124 loss: 1.8799 (1.9824) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [297] [ 40/312] eta: 0:04:06 lr: 0.001124 min_lr: 0.001124 loss: 1.9215 (1.9666) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [297] [ 50/312] eta: 0:03:47 lr: 0.001123 min_lr: 0.001123 loss: 1.9215 (1.9625) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [297] [ 60/312] eta: 0:03:31 lr: 0.001123 min_lr: 0.001123 loss: 2.0831 (1.9739) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [297] [ 70/312] eta: 0:03:18 lr: 0.001122 min_lr: 0.001122 loss: 2.0831 (1.9642) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [297] [ 80/312] eta: 0:03:06 lr: 0.001122 min_lr: 0.001122 loss: 2.0393 (1.9484) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [297] [ 90/312] eta: 0:02:55 lr: 0.001122 min_lr: 0.001122 loss: 1.7906 (1.9409) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [297] [100/312] eta: 0:02:45 lr: 0.001121 min_lr: 0.001121 loss: 1.8332 (1.9356) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [297] [110/312] eta: 0:02:36 lr: 0.001121 min_lr: 0.001121 loss: 1.9884 (1.9330) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [297] [120/312] eta: 0:02:27 lr: 0.001120 min_lr: 0.001120 loss: 1.9716 (1.9233) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [297] [130/312] eta: 0:02:18 lr: 0.001120 min_lr: 0.001120 loss: 1.8051 (1.9105) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [297] [140/312] eta: 0:02:10 lr: 0.001120 min_lr: 0.001120 loss: 1.9581 (1.9150) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [297] [150/312] eta: 0:02:01 lr: 0.001119 min_lr: 0.001119 loss: 1.9292 (1.9069) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [297] [160/312] eta: 0:01:53 lr: 0.001119 min_lr: 0.001119 loss: 1.9018 (1.9006) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [297] [170/312] eta: 0:01:45 lr: 0.001118 min_lr: 0.001118 loss: 2.0169 (1.9074) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [297] [180/312] eta: 0:01:38 lr: 0.001118 min_lr: 0.001118 loss: 1.9525 (1.9069) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [297] [190/312] eta: 0:01:30 lr: 0.001117 min_lr: 0.001117 loss: 1.9281 (1.9043) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [297] [200/312] eta: 0:01:22 lr: 0.001117 min_lr: 0.001117 loss: 1.9697 (1.9048) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [297] [210/312] eta: 0:01:15 lr: 0.001117 min_lr: 0.001117 loss: 1.9697 (1.9008) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [297] [220/312] eta: 0:01:07 lr: 0.001116 min_lr: 0.001116 loss: 1.8253 (1.8999) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [297] [230/312] eta: 0:01:00 lr: 0.001116 min_lr: 0.001116 loss: 2.0182 (1.9031) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [297] [240/312] eta: 0:00:52 lr: 0.001115 min_lr: 0.001115 loss: 2.0261 (1.9020) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [297] [250/312] eta: 0:00:45 lr: 0.001115 min_lr: 0.001115 loss: 1.8812 (1.8997) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [297] [260/312] eta: 0:00:37 lr: 0.001115 min_lr: 0.001115 loss: 1.9203 (1.9026) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [297] [270/312] eta: 0:00:30 lr: 0.001114 min_lr: 0.001114 loss: 1.9203 (1.9020) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [297] [280/312] eta: 0:00:23 lr: 0.001114 min_lr: 0.001114 loss: 2.0919 (1.9110) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [297] [290/312] eta: 0:00:15 lr: 0.001113 min_lr: 0.001113 loss: 2.1085 (1.9157) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [297] [300/312] eta: 0:00:08 lr: 0.001113 min_lr: 0.001113 loss: 2.0809 (1.9223) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [297] [310/312] eta: 0:00:01 lr: 0.001112 min_lr: 0.001112 loss: 2.0119 (1.9193) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [297] [311/312] eta: 0:00:00 lr: 0.001112 min_lr: 0.001112 loss: 2.0148 (1.9200) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [297] Total time: 0:03:46 (0.7253 s / it) Averaged stats: lr: 0.001112 min_lr: 0.001112 loss: 2.0148 (1.9224) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5305 (0.5305) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 4.6035 data: 4.3978 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7449 (0.7402) acc1: 81.5104 (80.1600) acc5: 96.8750 (95.8400) time: 0.6628 data: 0.4887 max mem: 64948 Test: Total time: 0:00:06 (0.6884 s / it) * Acc@1 81.228 Acc@5 95.660 loss 0.721 Accuracy of the model on the 50000 test images: 81.2% Max accuracy: 81.23% Test: [0/9] eta: 0:00:40 loss: 0.4727 (0.4727) acc1: 86.9792 (86.9792) acc5: 97.9167 (97.9167) time: 4.5225 data: 4.3159 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6501 (0.6617) acc1: 82.2917 (81.1200) acc5: 97.1354 (96.5440) time: 0.6538 data: 0.4796 max mem: 64948 Test: Total time: 0:00:05 (0.6613 s / it) * Acc@1 82.762 Acc@5 96.416 loss 0.638 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.76% Epoch: [298] [ 0/312] eta: 0:46:26 lr: 0.001112 min_lr: 0.001112 loss: 2.0846 (2.0846) weight_decay: 0.0500 (0.0500) time: 8.9300 data: 8.2039 max mem: 64948 Epoch: [298] [ 10/312] eta: 0:07:46 lr: 0.001112 min_lr: 0.001112 loss: 2.0521 (2.0188) weight_decay: 0.0500 (0.0500) time: 1.5431 data: 0.7969 max mem: 64948 Epoch: [298] [ 20/312] eta: 0:05:32 lr: 0.001111 min_lr: 0.001111 loss: 1.9708 (1.9302) weight_decay: 0.0500 (0.0500) time: 0.7483 data: 0.0283 max mem: 64948 Epoch: [298] [ 30/312] eta: 0:04:40 lr: 0.001111 min_lr: 0.001111 loss: 1.9458 (1.9373) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [298] [ 40/312] eta: 0:04:10 lr: 0.001111 min_lr: 0.001111 loss: 1.9991 (1.9538) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [298] [ 50/312] eta: 0:03:50 lr: 0.001110 min_lr: 0.001110 loss: 2.0963 (1.9805) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [298] [ 60/312] eta: 0:03:34 lr: 0.001110 min_lr: 0.001110 loss: 2.0785 (1.9589) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0004 max mem: 64948 Epoch: [298] [ 70/312] eta: 0:03:20 lr: 0.001109 min_lr: 0.001109 loss: 1.9997 (1.9489) weight_decay: 0.0500 (0.0500) time: 0.7021 data: 0.0003 max mem: 64948 Epoch: [298] [ 80/312] eta: 0:03:08 lr: 0.001109 min_lr: 0.001109 loss: 1.7659 (1.9367) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0003 max mem: 64948 Epoch: [298] [ 90/312] eta: 0:02:57 lr: 0.001109 min_lr: 0.001109 loss: 1.7659 (1.9256) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [298] [100/312] eta: 0:02:47 lr: 0.001108 min_lr: 0.001108 loss: 1.7971 (1.9182) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [298] [110/312] eta: 0:02:37 lr: 0.001108 min_lr: 0.001108 loss: 1.7971 (1.9153) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [298] [120/312] eta: 0:02:28 lr: 0.001107 min_lr: 0.001107 loss: 1.9658 (1.9207) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [298] [130/312] eta: 0:02:19 lr: 0.001107 min_lr: 0.001107 loss: 2.0788 (1.9299) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [298] [140/312] eta: 0:02:11 lr: 0.001106 min_lr: 0.001106 loss: 2.0826 (1.9384) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [298] [150/312] eta: 0:02:02 lr: 0.001106 min_lr: 0.001106 loss: 2.0348 (1.9448) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [298] [160/312] eta: 0:01:54 lr: 0.001106 min_lr: 0.001106 loss: 1.9631 (1.9460) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [298] [170/312] eta: 0:01:46 lr: 0.001105 min_lr: 0.001105 loss: 1.9631 (1.9437) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [298] [180/312] eta: 0:01:38 lr: 0.001105 min_lr: 0.001105 loss: 1.9795 (1.9486) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [298] [190/312] eta: 0:01:30 lr: 0.001104 min_lr: 0.001104 loss: 1.9843 (1.9383) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [298] [200/312] eta: 0:01:23 lr: 0.001104 min_lr: 0.001104 loss: 1.9979 (1.9346) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [298] [210/312] eta: 0:01:15 lr: 0.001104 min_lr: 0.001104 loss: 1.9798 (1.9342) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [298] [220/312] eta: 0:01:07 lr: 0.001103 min_lr: 0.001103 loss: 1.8242 (1.9266) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [298] [230/312] eta: 0:01:00 lr: 0.001103 min_lr: 0.001103 loss: 1.9192 (1.9273) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [298] [240/312] eta: 0:00:52 lr: 0.001102 min_lr: 0.001102 loss: 1.9470 (1.9277) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [298] [250/312] eta: 0:00:45 lr: 0.001102 min_lr: 0.001102 loss: 1.7726 (1.9190) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [298] [260/312] eta: 0:00:38 lr: 0.001101 min_lr: 0.001101 loss: 1.7726 (1.9166) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [298] [270/312] eta: 0:00:30 lr: 0.001101 min_lr: 0.001101 loss: 1.9560 (1.9116) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [298] [280/312] eta: 0:00:23 lr: 0.001101 min_lr: 0.001101 loss: 2.0131 (1.9164) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [298] [290/312] eta: 0:00:16 lr: 0.001100 min_lr: 0.001100 loss: 2.1233 (1.9176) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [298] [300/312] eta: 0:00:08 lr: 0.001100 min_lr: 0.001100 loss: 1.9928 (1.9172) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [298] [310/312] eta: 0:00:01 lr: 0.001099 min_lr: 0.001099 loss: 1.8527 (1.9169) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [298] [311/312] eta: 0:00:00 lr: 0.001099 min_lr: 0.001099 loss: 1.8527 (1.9181) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [298] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.001099 min_lr: 0.001099 loss: 1.8527 (1.9295) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5192 (0.5192) acc1: 86.9792 (86.9792) acc5: 97.1354 (97.1354) time: 4.6392 data: 4.4281 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7744 (0.7652) acc1: 79.6875 (79.9680) acc5: 95.0521 (94.8800) time: 0.6668 data: 0.4921 max mem: 64948 Test: Total time: 0:00:06 (0.6939 s / it) * Acc@1 81.090 Acc@5 95.442 loss 0.728 Accuracy of the model on the 50000 test images: 81.1% Max accuracy: 81.23% Test: [0/9] eta: 0:00:44 loss: 0.4723 (0.4723) acc1: 86.9792 (86.9792) acc5: 97.9167 (97.9167) time: 4.9553 data: 4.7478 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6495 (0.6614) acc1: 82.5521 (81.1520) acc5: 97.1354 (96.5440) time: 0.7019 data: 0.5276 max mem: 64948 Test: Total time: 0:00:06 (0.7095 s / it) * Acc@1 82.776 Acc@5 96.418 loss 0.637 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.78% Epoch: [299] [ 0/312] eta: 0:53:14 lr: 0.001099 min_lr: 0.001099 loss: 2.4028 (2.4028) weight_decay: 0.0500 (0.0500) time: 10.2382 data: 9.4544 max mem: 64948 Epoch: [299] [ 10/312] eta: 0:08:07 lr: 0.001099 min_lr: 0.001099 loss: 1.7519 (1.7315) weight_decay: 0.0500 (0.0500) time: 1.6158 data: 0.8598 max mem: 64948 Epoch: [299] [ 20/312] eta: 0:05:44 lr: 0.001098 min_lr: 0.001098 loss: 1.7519 (1.7709) weight_decay: 0.0500 (0.0500) time: 0.7282 data: 0.0004 max mem: 64948 Epoch: [299] [ 30/312] eta: 0:04:49 lr: 0.001098 min_lr: 0.001098 loss: 1.9755 (1.8152) weight_decay: 0.0500 (0.0500) time: 0.7011 data: 0.0003 max mem: 64948 Epoch: [299] [ 40/312] eta: 0:04:17 lr: 0.001098 min_lr: 0.001098 loss: 1.9548 (1.8396) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [299] [ 50/312] eta: 0:03:55 lr: 0.001097 min_lr: 0.001097 loss: 1.9047 (1.8771) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [299] [ 60/312] eta: 0:03:37 lr: 0.001097 min_lr: 0.001097 loss: 1.9262 (1.8790) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [299] [ 70/312] eta: 0:03:23 lr: 0.001096 min_lr: 0.001096 loss: 1.9106 (1.8857) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [299] [ 80/312] eta: 0:03:10 lr: 0.001096 min_lr: 0.001096 loss: 1.9161 (1.8854) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [299] [ 90/312] eta: 0:02:59 lr: 0.001096 min_lr: 0.001096 loss: 1.9117 (1.8803) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [299] [100/312] eta: 0:02:49 lr: 0.001095 min_lr: 0.001095 loss: 1.9064 (1.8834) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [299] [110/312] eta: 0:02:39 lr: 0.001095 min_lr: 0.001095 loss: 1.8310 (1.8711) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [299] [120/312] eta: 0:02:29 lr: 0.001094 min_lr: 0.001094 loss: 1.7850 (1.8653) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [299] [130/312] eta: 0:02:20 lr: 0.001094 min_lr: 0.001094 loss: 1.9347 (1.8749) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [299] [140/312] eta: 0:02:12 lr: 0.001093 min_lr: 0.001093 loss: 1.9480 (1.8727) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [299] [150/312] eta: 0:02:03 lr: 0.001093 min_lr: 0.001093 loss: 1.9975 (1.8876) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [299] [160/312] eta: 0:01:55 lr: 0.001093 min_lr: 0.001093 loss: 2.0563 (1.8988) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [299] [170/312] eta: 0:01:47 lr: 0.001092 min_lr: 0.001092 loss: 2.0930 (1.9139) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [299] [180/312] eta: 0:01:39 lr: 0.001092 min_lr: 0.001092 loss: 2.0348 (1.9133) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [299] [190/312] eta: 0:01:31 lr: 0.001091 min_lr: 0.001091 loss: 2.0348 (1.9203) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [299] [200/312] eta: 0:01:23 lr: 0.001091 min_lr: 0.001091 loss: 2.1788 (1.9239) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [299] [210/312] eta: 0:01:15 lr: 0.001090 min_lr: 0.001090 loss: 2.0049 (1.9253) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [299] [220/312] eta: 0:01:08 lr: 0.001090 min_lr: 0.001090 loss: 1.8829 (1.9247) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [299] [230/312] eta: 0:01:00 lr: 0.001090 min_lr: 0.001090 loss: 1.8872 (1.9228) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [299] [240/312] eta: 0:00:53 lr: 0.001089 min_lr: 0.001089 loss: 1.9227 (1.9229) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [299] [250/312] eta: 0:00:45 lr: 0.001089 min_lr: 0.001089 loss: 1.9227 (1.9237) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [299] [260/312] eta: 0:00:38 lr: 0.001088 min_lr: 0.001088 loss: 1.8374 (1.9203) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [299] [270/312] eta: 0:00:30 lr: 0.001088 min_lr: 0.001088 loss: 1.7645 (1.9156) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [299] [280/312] eta: 0:00:23 lr: 0.001088 min_lr: 0.001088 loss: 1.8267 (1.9163) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [299] [290/312] eta: 0:00:16 lr: 0.001087 min_lr: 0.001087 loss: 1.8718 (1.9155) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [299] [300/312] eta: 0:00:08 lr: 0.001087 min_lr: 0.001087 loss: 1.8591 (1.9123) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [299] [310/312] eta: 0:00:01 lr: 0.001086 min_lr: 0.001086 loss: 1.8591 (1.9099) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [299] [311/312] eta: 0:00:00 lr: 0.001086 min_lr: 0.001086 loss: 1.8591 (1.9105) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [299] Total time: 0:03:47 (0.7301 s / it) Averaged stats: lr: 0.001086 min_lr: 0.001086 loss: 1.8591 (1.9144) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5446 (0.5446) acc1: 86.4583 (86.4583) acc5: 96.3542 (96.3542) time: 4.4987 data: 4.2827 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7068 (0.7272) acc1: 81.7708 (80.5120) acc5: 96.3542 (95.4560) time: 0.6511 data: 0.4759 max mem: 64948 Test: Total time: 0:00:06 (0.6732 s / it) * Acc@1 81.438 Acc@5 95.584 loss 0.711 Accuracy of the model on the 50000 test images: 81.4% Max accuracy: 81.44% Test: [0/9] eta: 0:00:42 loss: 0.4721 (0.4721) acc1: 86.9792 (86.9792) acc5: 97.9167 (97.9167) time: 4.7723 data: 4.5545 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6490 (0.6611) acc1: 82.5521 (81.1840) acc5: 97.1354 (96.5440) time: 0.6822 data: 0.5062 max mem: 64948 Test: Total time: 0:00:06 (0.6919 s / it) * Acc@1 82.784 Acc@5 96.422 loss 0.637 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.78% Epoch: [300] [ 0/312] eta: 0:50:27 lr: 0.001086 min_lr: 0.001086 loss: 1.7718 (1.7718) weight_decay: 0.0500 (0.0500) time: 9.7045 data: 8.9318 max mem: 64948 Epoch: [300] [ 10/312] eta: 0:07:55 lr: 0.001086 min_lr: 0.001086 loss: 1.7718 (1.8400) weight_decay: 0.0500 (0.0500) time: 1.5736 data: 0.8124 max mem: 64948 Epoch: [300] [ 20/312] eta: 0:05:37 lr: 0.001085 min_lr: 0.001085 loss: 1.9807 (1.8868) weight_decay: 0.0500 (0.0500) time: 0.7290 data: 0.0004 max mem: 64948 Epoch: [300] [ 30/312] eta: 0:04:44 lr: 0.001085 min_lr: 0.001085 loss: 1.9807 (1.8959) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [300] [ 40/312] eta: 0:04:13 lr: 0.001085 min_lr: 0.001085 loss: 1.9601 (1.8869) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [300] [ 50/312] eta: 0:03:52 lr: 0.001084 min_lr: 0.001084 loss: 1.9908 (1.9217) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [300] [ 60/312] eta: 0:03:35 lr: 0.001084 min_lr: 0.001084 loss: 2.0772 (1.9389) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [300] [ 70/312] eta: 0:03:21 lr: 0.001083 min_lr: 0.001083 loss: 2.0348 (1.9386) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [300] [ 80/312] eta: 0:03:09 lr: 0.001083 min_lr: 0.001083 loss: 2.0271 (1.9249) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [300] [ 90/312] eta: 0:02:58 lr: 0.001082 min_lr: 0.001082 loss: 2.0123 (1.9202) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [300] [100/312] eta: 0:02:47 lr: 0.001082 min_lr: 0.001082 loss: 1.8423 (1.9200) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [300] [110/312] eta: 0:02:38 lr: 0.001082 min_lr: 0.001082 loss: 1.9947 (1.9326) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [300] [120/312] eta: 0:02:28 lr: 0.001081 min_lr: 0.001081 loss: 1.9950 (1.9265) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [300] [130/312] eta: 0:02:19 lr: 0.001081 min_lr: 0.001081 loss: 2.0130 (1.9289) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [300] [140/312] eta: 0:02:11 lr: 0.001080 min_lr: 0.001080 loss: 2.0130 (1.9287) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [300] [150/312] eta: 0:02:03 lr: 0.001080 min_lr: 0.001080 loss: 1.9547 (1.9329) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [300] [160/312] eta: 0:01:54 lr: 0.001080 min_lr: 0.001080 loss: 1.8812 (1.9161) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [300] [170/312] eta: 0:01:46 lr: 0.001079 min_lr: 0.001079 loss: 1.7153 (1.9150) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [300] [180/312] eta: 0:01:38 lr: 0.001079 min_lr: 0.001079 loss: 1.8979 (1.9094) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [300] [190/312] eta: 0:01:30 lr: 0.001078 min_lr: 0.001078 loss: 2.0569 (1.9205) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [300] [200/312] eta: 0:01:23 lr: 0.001078 min_lr: 0.001078 loss: 2.0236 (1.9244) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [300] [210/312] eta: 0:01:15 lr: 0.001078 min_lr: 0.001078 loss: 1.9622 (1.9219) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [300] [220/312] eta: 0:01:07 lr: 0.001077 min_lr: 0.001077 loss: 2.0119 (1.9266) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [300] [230/312] eta: 0:01:00 lr: 0.001077 min_lr: 0.001077 loss: 2.0177 (1.9244) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [300] [240/312] eta: 0:00:52 lr: 0.001076 min_lr: 0.001076 loss: 2.0532 (1.9234) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [300] [250/312] eta: 0:00:45 lr: 0.001076 min_lr: 0.001076 loss: 2.0458 (1.9280) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [300] [260/312] eta: 0:00:38 lr: 0.001075 min_lr: 0.001075 loss: 2.0496 (1.9302) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [300] [270/312] eta: 0:00:30 lr: 0.001075 min_lr: 0.001075 loss: 2.1000 (1.9339) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [300] [280/312] eta: 0:00:23 lr: 0.001075 min_lr: 0.001075 loss: 2.0542 (1.9335) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [300] [290/312] eta: 0:00:16 lr: 0.001074 min_lr: 0.001074 loss: 2.0027 (1.9330) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [300] [300/312] eta: 0:00:08 lr: 0.001074 min_lr: 0.001074 loss: 1.8863 (1.9219) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [300] [310/312] eta: 0:00:01 lr: 0.001073 min_lr: 0.001073 loss: 1.8817 (1.9235) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [300] [311/312] eta: 0:00:00 lr: 0.001073 min_lr: 0.001073 loss: 1.8863 (1.9244) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [300] Total time: 0:03:47 (0.7284 s / it) Averaged stats: lr: 0.001073 min_lr: 0.001073 loss: 1.8863 (1.9117) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5245 (0.5245) acc1: 86.9792 (86.9792) acc5: 96.0938 (96.0938) time: 4.6767 data: 4.4573 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7109 (0.7420) acc1: 81.2500 (80.0320) acc5: 95.8333 (95.5840) time: 0.6710 data: 0.4954 max mem: 64948 Test: Total time: 0:00:06 (0.6957 s / it) * Acc@1 81.084 Acc@5 95.482 loss 0.725 Accuracy of the model on the 50000 test images: 81.1% Max accuracy: 81.44% Test: [0/9] eta: 0:00:41 loss: 0.4718 (0.4718) acc1: 86.9792 (86.9792) acc5: 97.9167 (97.9167) time: 4.5656 data: 4.3533 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6478 (0.6606) acc1: 82.5521 (81.1520) acc5: 97.3958 (96.5760) time: 0.6587 data: 0.4838 max mem: 64948 Test: Total time: 0:00:06 (0.6672 s / it) * Acc@1 82.790 Acc@5 96.440 loss 0.636 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.79% Epoch: [301] [ 0/312] eta: 0:51:12 lr: 0.001073 min_lr: 0.001073 loss: 1.7724 (1.7724) weight_decay: 0.0500 (0.0500) time: 9.8471 data: 9.0375 max mem: 64948 Epoch: [301] [ 10/312] eta: 0:07:46 lr: 0.001073 min_lr: 0.001073 loss: 1.7724 (1.8542) weight_decay: 0.0500 (0.0500) time: 1.5446 data: 0.8219 max mem: 64948 Epoch: [301] [ 20/312] eta: 0:05:32 lr: 0.001072 min_lr: 0.001072 loss: 1.9562 (1.9206) weight_decay: 0.0500 (0.0500) time: 0.7040 data: 0.0004 max mem: 64948 Epoch: [301] [ 30/312] eta: 0:04:40 lr: 0.001072 min_lr: 0.001072 loss: 1.9597 (1.8878) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [301] [ 40/312] eta: 0:04:10 lr: 0.001072 min_lr: 0.001072 loss: 1.9709 (1.9006) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [301] [ 50/312] eta: 0:03:49 lr: 0.001071 min_lr: 0.001071 loss: 2.0244 (1.9128) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [301] [ 60/312] eta: 0:03:33 lr: 0.001071 min_lr: 0.001071 loss: 2.0232 (1.9313) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [301] [ 70/312] eta: 0:03:20 lr: 0.001070 min_lr: 0.001070 loss: 1.9902 (1.9278) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [301] [ 80/312] eta: 0:03:08 lr: 0.001070 min_lr: 0.001070 loss: 1.9456 (1.9399) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [301] [ 90/312] eta: 0:02:57 lr: 0.001070 min_lr: 0.001070 loss: 2.0052 (1.9370) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [301] [100/312] eta: 0:02:47 lr: 0.001069 min_lr: 0.001069 loss: 1.8141 (1.9210) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [301] [110/312] eta: 0:02:37 lr: 0.001069 min_lr: 0.001069 loss: 1.9961 (1.9333) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [301] [120/312] eta: 0:02:28 lr: 0.001068 min_lr: 0.001068 loss: 2.0086 (1.9221) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [301] [130/312] eta: 0:02:19 lr: 0.001068 min_lr: 0.001068 loss: 1.9615 (1.9311) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [301] [140/312] eta: 0:02:11 lr: 0.001067 min_lr: 0.001067 loss: 1.9514 (1.9225) weight_decay: 0.0500 (0.0500) time: 0.7015 data: 0.0004 max mem: 64948 Epoch: [301] [150/312] eta: 0:02:02 lr: 0.001067 min_lr: 0.001067 loss: 1.9524 (1.9259) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [301] [160/312] eta: 0:01:54 lr: 0.001067 min_lr: 0.001067 loss: 1.9963 (1.9217) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [301] [170/312] eta: 0:01:46 lr: 0.001066 min_lr: 0.001066 loss: 1.9860 (1.9242) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [301] [180/312] eta: 0:01:38 lr: 0.001066 min_lr: 0.001066 loss: 1.9755 (1.9195) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0003 max mem: 64948 Epoch: [301] [190/312] eta: 0:01:30 lr: 0.001065 min_lr: 0.001065 loss: 1.8801 (1.9215) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [301] [200/312] eta: 0:01:23 lr: 0.001065 min_lr: 0.001065 loss: 2.0449 (1.9315) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [301] [210/312] eta: 0:01:15 lr: 0.001065 min_lr: 0.001065 loss: 2.0010 (1.9333) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [301] [220/312] eta: 0:01:07 lr: 0.001064 min_lr: 0.001064 loss: 1.8896 (1.9261) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [301] [230/312] eta: 0:01:00 lr: 0.001064 min_lr: 0.001064 loss: 1.8476 (1.9254) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [301] [240/312] eta: 0:00:52 lr: 0.001063 min_lr: 0.001063 loss: 1.9587 (1.9231) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [301] [250/312] eta: 0:00:45 lr: 0.001063 min_lr: 0.001063 loss: 2.0219 (1.9279) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [301] [260/312] eta: 0:00:38 lr: 0.001063 min_lr: 0.001063 loss: 2.0353 (1.9284) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [301] [270/312] eta: 0:00:30 lr: 0.001062 min_lr: 0.001062 loss: 2.0166 (1.9311) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [301] [280/312] eta: 0:00:23 lr: 0.001062 min_lr: 0.001062 loss: 2.0166 (1.9312) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0006 max mem: 64948 Epoch: [301] [290/312] eta: 0:00:15 lr: 0.001061 min_lr: 0.001061 loss: 1.8304 (1.9243) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0005 max mem: 64948 Epoch: [301] [300/312] eta: 0:00:08 lr: 0.001061 min_lr: 0.001061 loss: 1.8781 (1.9261) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [301] [310/312] eta: 0:00:01 lr: 0.001060 min_lr: 0.001060 loss: 2.0219 (1.9269) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [301] [311/312] eta: 0:00:00 lr: 0.001060 min_lr: 0.001060 loss: 2.0219 (1.9280) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [301] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.001060 min_lr: 0.001060 loss: 2.0219 (1.9224) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5431 (0.5431) acc1: 85.4167 (85.4167) acc5: 96.8750 (96.8750) time: 4.6749 data: 4.4559 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7048 (0.7174) acc1: 82.0312 (80.4800) acc5: 96.8750 (96.0640) time: 0.6707 data: 0.4952 max mem: 64948 Test: Total time: 0:00:06 (0.6973 s / it) * Acc@1 81.456 Acc@5 95.740 loss 0.701 Accuracy of the model on the 50000 test images: 81.5% Max accuracy: 81.46% Test: [0/9] eta: 0:00:40 loss: 0.4718 (0.4718) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.4956 data: 4.2915 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6465 (0.6602) acc1: 82.5521 (81.1200) acc5: 97.3958 (96.5440) time: 0.6508 data: 0.4769 max mem: 64948 Test: Total time: 0:00:05 (0.6584 s / it) * Acc@1 82.788 Acc@5 96.446 loss 0.636 Accuracy of the model EMA on 50000 test images: 82.8% Epoch: [302] [ 0/312] eta: 0:53:40 lr: 0.001060 min_lr: 0.001060 loss: 1.9196 (1.9196) weight_decay: 0.0500 (0.0500) time: 10.3207 data: 8.0344 max mem: 64948 Epoch: [302] [ 10/312] eta: 0:08:06 lr: 0.001060 min_lr: 0.001060 loss: 2.0313 (1.9808) weight_decay: 0.0500 (0.0500) time: 1.6093 data: 0.7308 max mem: 64948 Epoch: [302] [ 20/312] eta: 0:05:42 lr: 0.001060 min_lr: 0.001060 loss: 1.8642 (1.8390) weight_decay: 0.0500 (0.0500) time: 0.7164 data: 0.0004 max mem: 64948 Epoch: [302] [ 30/312] eta: 0:04:47 lr: 0.001059 min_lr: 0.001059 loss: 1.8172 (1.9013) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [302] [ 40/312] eta: 0:04:16 lr: 0.001059 min_lr: 0.001059 loss: 2.0285 (1.9014) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [302] [ 50/312] eta: 0:03:54 lr: 0.001058 min_lr: 0.001058 loss: 1.9388 (1.9162) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [302] [ 60/312] eta: 0:03:36 lr: 0.001058 min_lr: 0.001058 loss: 2.0277 (1.9330) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [302] [ 70/312] eta: 0:03:22 lr: 0.001057 min_lr: 0.001057 loss: 1.9870 (1.9221) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [302] [ 80/312] eta: 0:03:10 lr: 0.001057 min_lr: 0.001057 loss: 1.9468 (1.9207) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [302] [ 90/312] eta: 0:02:58 lr: 0.001057 min_lr: 0.001057 loss: 1.9806 (1.9086) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [302] [100/312] eta: 0:02:48 lr: 0.001056 min_lr: 0.001056 loss: 1.9648 (1.9179) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [302] [110/312] eta: 0:02:38 lr: 0.001056 min_lr: 0.001056 loss: 1.8547 (1.9071) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [302] [120/312] eta: 0:02:29 lr: 0.001055 min_lr: 0.001055 loss: 1.8993 (1.9148) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [302] [130/312] eta: 0:02:20 lr: 0.001055 min_lr: 0.001055 loss: 2.0962 (1.9219) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [302] [140/312] eta: 0:02:11 lr: 0.001055 min_lr: 0.001055 loss: 2.1166 (1.9322) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [302] [150/312] eta: 0:02:03 lr: 0.001054 min_lr: 0.001054 loss: 1.9336 (1.9251) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [302] [160/312] eta: 0:01:55 lr: 0.001054 min_lr: 0.001054 loss: 1.8402 (1.9165) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [302] [170/312] eta: 0:01:47 lr: 0.001053 min_lr: 0.001053 loss: 1.9294 (1.9236) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [302] [180/312] eta: 0:01:39 lr: 0.001053 min_lr: 0.001053 loss: 2.0530 (1.9299) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [302] [190/312] eta: 0:01:31 lr: 0.001053 min_lr: 0.001053 loss: 2.0530 (1.9351) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [302] [200/312] eta: 0:01:23 lr: 0.001052 min_lr: 0.001052 loss: 2.0552 (1.9363) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [302] [210/312] eta: 0:01:15 lr: 0.001052 min_lr: 0.001052 loss: 2.0636 (1.9413) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [302] [220/312] eta: 0:01:08 lr: 0.001051 min_lr: 0.001051 loss: 2.0645 (1.9404) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [302] [230/312] eta: 0:01:00 lr: 0.001051 min_lr: 0.001051 loss: 1.9934 (1.9429) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [302] [240/312] eta: 0:00:53 lr: 0.001050 min_lr: 0.001050 loss: 1.9934 (1.9424) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [302] [250/312] eta: 0:00:45 lr: 0.001050 min_lr: 0.001050 loss: 1.9639 (1.9428) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [302] [260/312] eta: 0:00:38 lr: 0.001050 min_lr: 0.001050 loss: 1.9639 (1.9443) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [302] [270/312] eta: 0:00:30 lr: 0.001049 min_lr: 0.001049 loss: 2.0078 (1.9446) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [302] [280/312] eta: 0:00:23 lr: 0.001049 min_lr: 0.001049 loss: 2.0408 (1.9443) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0010 max mem: 64948 Epoch: [302] [290/312] eta: 0:00:16 lr: 0.001048 min_lr: 0.001048 loss: 1.8929 (1.9426) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [302] [300/312] eta: 0:00:08 lr: 0.001048 min_lr: 0.001048 loss: 1.9802 (1.9442) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [302] [310/312] eta: 0:00:01 lr: 0.001048 min_lr: 0.001048 loss: 2.0379 (1.9420) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [302] [311/312] eta: 0:00:00 lr: 0.001048 min_lr: 0.001048 loss: 1.9802 (1.9404) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [302] Total time: 0:03:47 (0.7300 s / it) Averaged stats: lr: 0.001048 min_lr: 0.001048 loss: 1.9802 (1.9107) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5137 (0.5137) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.7255 data: 4.5111 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7137 (0.7443) acc1: 81.5104 (80.3200) acc5: 96.3542 (95.8080) time: 0.6763 data: 0.5013 max mem: 64948 Test: Total time: 0:00:06 (0.7015 s / it) * Acc@1 81.332 Acc@5 95.720 loss 0.710 Accuracy of the model on the 50000 test images: 81.3% Max accuracy: 81.46% Test: [0/9] eta: 0:00:42 loss: 0.4716 (0.4716) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.7400 data: 4.5320 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6455 (0.6599) acc1: 82.5521 (81.0880) acc5: 97.3958 (96.5440) time: 0.6832 data: 0.5037 max mem: 64948 Test: Total time: 0:00:06 (0.6936 s / it) * Acc@1 82.794 Acc@5 96.432 loss 0.636 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.79% Epoch: [303] [ 0/312] eta: 0:51:22 lr: 0.001047 min_lr: 0.001047 loss: 2.1622 (2.1622) weight_decay: 0.0500 (0.0500) time: 9.8796 data: 9.0644 max mem: 64948 Epoch: [303] [ 10/312] eta: 0:07:48 lr: 0.001047 min_lr: 0.001047 loss: 1.9353 (1.9290) weight_decay: 0.0500 (0.0500) time: 1.5529 data: 0.8248 max mem: 64948 Epoch: [303] [ 20/312] eta: 0:05:34 lr: 0.001047 min_lr: 0.001047 loss: 1.9268 (1.9281) weight_decay: 0.0500 (0.0500) time: 0.7087 data: 0.0006 max mem: 64948 Epoch: [303] [ 30/312] eta: 0:04:41 lr: 0.001046 min_lr: 0.001046 loss: 1.9268 (1.8861) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [303] [ 40/312] eta: 0:04:11 lr: 0.001046 min_lr: 0.001046 loss: 1.5630 (1.8346) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [303] [ 50/312] eta: 0:03:50 lr: 0.001045 min_lr: 0.001045 loss: 1.9816 (1.8480) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [303] [ 60/312] eta: 0:03:34 lr: 0.001045 min_lr: 0.001045 loss: 2.0461 (1.8728) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [303] [ 70/312] eta: 0:03:20 lr: 0.001045 min_lr: 0.001045 loss: 2.0960 (1.9141) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0005 max mem: 64948 Epoch: [303] [ 80/312] eta: 0:03:08 lr: 0.001044 min_lr: 0.001044 loss: 2.0963 (1.9325) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [303] [ 90/312] eta: 0:02:57 lr: 0.001044 min_lr: 0.001044 loss: 2.0642 (1.9313) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [303] [100/312] eta: 0:02:47 lr: 0.001043 min_lr: 0.001043 loss: 1.8412 (1.9209) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [303] [110/312] eta: 0:02:37 lr: 0.001043 min_lr: 0.001043 loss: 1.9089 (1.9225) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [303] [120/312] eta: 0:02:28 lr: 0.001043 min_lr: 0.001043 loss: 1.9914 (1.9186) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [303] [130/312] eta: 0:02:19 lr: 0.001042 min_lr: 0.001042 loss: 1.9345 (1.9220) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [303] [140/312] eta: 0:02:11 lr: 0.001042 min_lr: 0.001042 loss: 1.9345 (1.9181) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [303] [150/312] eta: 0:02:02 lr: 0.001041 min_lr: 0.001041 loss: 1.8107 (1.9069) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [303] [160/312] eta: 0:01:54 lr: 0.001041 min_lr: 0.001041 loss: 1.8651 (1.9110) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [303] [170/312] eta: 0:01:46 lr: 0.001041 min_lr: 0.001041 loss: 2.0235 (1.9161) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [303] [180/312] eta: 0:01:38 lr: 0.001040 min_lr: 0.001040 loss: 2.0747 (1.9254) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [303] [190/312] eta: 0:01:30 lr: 0.001040 min_lr: 0.001040 loss: 1.9322 (1.9153) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [303] [200/312] eta: 0:01:23 lr: 0.001039 min_lr: 0.001039 loss: 1.7432 (1.9128) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [303] [210/312] eta: 0:01:15 lr: 0.001039 min_lr: 0.001039 loss: 1.8783 (1.9115) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [303] [220/312] eta: 0:01:07 lr: 0.001038 min_lr: 0.001038 loss: 1.9751 (1.9139) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [303] [230/312] eta: 0:01:00 lr: 0.001038 min_lr: 0.001038 loss: 1.9347 (1.9102) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [303] [240/312] eta: 0:00:52 lr: 0.001038 min_lr: 0.001038 loss: 2.0719 (1.9182) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [303] [250/312] eta: 0:00:45 lr: 0.001037 min_lr: 0.001037 loss: 2.1122 (1.9240) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [303] [260/312] eta: 0:00:38 lr: 0.001037 min_lr: 0.001037 loss: 2.1121 (1.9240) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [303] [270/312] eta: 0:00:30 lr: 0.001036 min_lr: 0.001036 loss: 1.8933 (1.9222) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [303] [280/312] eta: 0:00:23 lr: 0.001036 min_lr: 0.001036 loss: 1.8933 (1.9207) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [303] [290/312] eta: 0:00:15 lr: 0.001036 min_lr: 0.001036 loss: 2.0071 (1.9230) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [303] [300/312] eta: 0:00:08 lr: 0.001035 min_lr: 0.001035 loss: 2.1214 (1.9264) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [303] [310/312] eta: 0:00:01 lr: 0.001035 min_lr: 0.001035 loss: 2.1214 (1.9318) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [303] [311/312] eta: 0:00:00 lr: 0.001035 min_lr: 0.001035 loss: 2.1675 (1.9329) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [303] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.001035 min_lr: 0.001035 loss: 2.1675 (1.9079) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5499 (0.5499) acc1: 86.7188 (86.7188) acc5: 97.3958 (97.3958) time: 4.7042 data: 4.4849 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7127 (0.7422) acc1: 82.8125 (80.2880) acc5: 96.3542 (95.9360) time: 0.6741 data: 0.4984 max mem: 64948 Test: Total time: 0:00:06 (0.6993 s / it) * Acc@1 81.198 Acc@5 95.694 loss 0.719 Accuracy of the model on the 50000 test images: 81.2% Max accuracy: 81.46% Test: [0/9] eta: 0:00:41 loss: 0.4716 (0.4716) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.5980 data: 4.3922 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6447 (0.6596) acc1: 82.5521 (81.0880) acc5: 97.3958 (96.5760) time: 0.6622 data: 0.4881 max mem: 64948 Test: Total time: 0:00:06 (0.6776 s / it) * Acc@1 82.790 Acc@5 96.434 loss 0.635 Accuracy of the model EMA on 50000 test images: 82.8% Epoch: [304] [ 0/312] eta: 0:52:53 lr: 0.001035 min_lr: 0.001035 loss: 1.9839 (1.9839) weight_decay: 0.0500 (0.0500) time: 10.1716 data: 9.3187 max mem: 64948 Epoch: [304] [ 10/312] eta: 0:07:59 lr: 0.001034 min_lr: 0.001034 loss: 1.9319 (1.9429) weight_decay: 0.0500 (0.0500) time: 1.5884 data: 0.8476 max mem: 64948 Epoch: [304] [ 20/312] eta: 0:05:39 lr: 0.001034 min_lr: 0.001034 loss: 1.9668 (1.9811) weight_decay: 0.0500 (0.0500) time: 0.7110 data: 0.0004 max mem: 64948 Epoch: [304] [ 30/312] eta: 0:04:45 lr: 0.001033 min_lr: 0.001033 loss: 2.0751 (1.9835) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [304] [ 40/312] eta: 0:04:14 lr: 0.001033 min_lr: 0.001033 loss: 2.0139 (1.9526) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [304] [ 50/312] eta: 0:03:53 lr: 0.001033 min_lr: 0.001033 loss: 1.9399 (1.9546) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [304] [ 60/312] eta: 0:03:36 lr: 0.001032 min_lr: 0.001032 loss: 1.9708 (1.9276) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [304] [ 70/312] eta: 0:03:22 lr: 0.001032 min_lr: 0.001032 loss: 1.7334 (1.9033) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [304] [ 80/312] eta: 0:03:09 lr: 0.001031 min_lr: 0.001031 loss: 1.8404 (1.8972) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [304] [ 90/312] eta: 0:02:58 lr: 0.001031 min_lr: 0.001031 loss: 1.8493 (1.8881) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [304] [100/312] eta: 0:02:48 lr: 0.001031 min_lr: 0.001031 loss: 1.8493 (1.8878) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [304] [110/312] eta: 0:02:38 lr: 0.001030 min_lr: 0.001030 loss: 1.9834 (1.8992) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [304] [120/312] eta: 0:02:29 lr: 0.001030 min_lr: 0.001030 loss: 1.8747 (1.8763) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [304] [130/312] eta: 0:02:20 lr: 0.001029 min_lr: 0.001029 loss: 1.7641 (1.8846) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [304] [140/312] eta: 0:02:11 lr: 0.001029 min_lr: 0.001029 loss: 1.8747 (1.8836) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [304] [150/312] eta: 0:02:03 lr: 0.001029 min_lr: 0.001029 loss: 1.9285 (1.8826) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [304] [160/312] eta: 0:01:54 lr: 0.001028 min_lr: 0.001028 loss: 1.9381 (1.8895) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [304] [170/312] eta: 0:01:46 lr: 0.001028 min_lr: 0.001028 loss: 1.9052 (1.8833) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [304] [180/312] eta: 0:01:38 lr: 0.001027 min_lr: 0.001027 loss: 1.6954 (1.8808) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [304] [190/312] eta: 0:01:31 lr: 0.001027 min_lr: 0.001027 loss: 1.7744 (1.8701) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [304] [200/312] eta: 0:01:23 lr: 0.001026 min_lr: 0.001026 loss: 1.9461 (1.8787) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [304] [210/312] eta: 0:01:15 lr: 0.001026 min_lr: 0.001026 loss: 1.9498 (1.8789) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [304] [220/312] eta: 0:01:08 lr: 0.001026 min_lr: 0.001026 loss: 1.9094 (1.8786) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [304] [230/312] eta: 0:01:00 lr: 0.001025 min_lr: 0.001025 loss: 1.9859 (1.8773) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [304] [240/312] eta: 0:00:52 lr: 0.001025 min_lr: 0.001025 loss: 2.0450 (1.8819) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [304] [250/312] eta: 0:00:45 lr: 0.001024 min_lr: 0.001024 loss: 1.9507 (1.8795) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [304] [260/312] eta: 0:00:38 lr: 0.001024 min_lr: 0.001024 loss: 2.0713 (1.8911) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [304] [270/312] eta: 0:00:30 lr: 0.001024 min_lr: 0.001024 loss: 2.0555 (1.8891) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [304] [280/312] eta: 0:00:23 lr: 0.001023 min_lr: 0.001023 loss: 1.9739 (1.8889) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [304] [290/312] eta: 0:00:16 lr: 0.001023 min_lr: 0.001023 loss: 2.0544 (1.8926) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0008 max mem: 64948 Epoch: [304] [300/312] eta: 0:00:08 lr: 0.001022 min_lr: 0.001022 loss: 1.9901 (1.8931) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [304] [310/312] eta: 0:00:01 lr: 0.001022 min_lr: 0.001022 loss: 1.8744 (1.8911) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [304] [311/312] eta: 0:00:00 lr: 0.001022 min_lr: 0.001022 loss: 1.8744 (1.8924) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [304] Total time: 0:03:47 (0.7296 s / it) Averaged stats: lr: 0.001022 min_lr: 0.001022 loss: 1.8744 (1.9120) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4955 (0.4955) acc1: 86.4583 (86.4583) acc5: 96.8750 (96.8750) time: 4.8468 data: 4.6233 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6885 (0.7106) acc1: 81.7708 (80.7040) acc5: 96.8750 (96.0000) time: 0.6898 data: 0.5138 max mem: 64948 Test: Total time: 0:00:06 (0.7088 s / it) * Acc@1 81.402 Acc@5 95.728 loss 0.710 Accuracy of the model on the 50000 test images: 81.4% Max accuracy: 81.46% Test: [0/9] eta: 0:00:44 loss: 0.4715 (0.4715) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.9469 data: 4.7364 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6439 (0.6591) acc1: 82.5521 (81.1840) acc5: 97.3958 (96.5760) time: 0.7056 data: 0.5264 max mem: 64948 Test: Total time: 0:00:06 (0.7156 s / it) * Acc@1 82.796 Acc@5 96.432 loss 0.635 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.80% Epoch: [305] [ 0/312] eta: 0:55:08 lr: 0.001022 min_lr: 0.001022 loss: 2.1120 (2.1120) weight_decay: 0.0500 (0.0500) time: 10.6043 data: 9.7886 max mem: 64948 Epoch: [305] [ 10/312] eta: 0:08:07 lr: 0.001022 min_lr: 0.001022 loss: 2.0997 (2.0039) weight_decay: 0.0500 (0.0500) time: 1.6140 data: 0.8902 max mem: 64948 Epoch: [305] [ 20/312] eta: 0:05:43 lr: 0.001021 min_lr: 0.001021 loss: 2.1035 (2.0533) weight_decay: 0.0500 (0.0500) time: 0.7054 data: 0.0003 max mem: 64948 Epoch: [305] [ 30/312] eta: 0:04:47 lr: 0.001021 min_lr: 0.001021 loss: 2.0690 (1.9981) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [305] [ 40/312] eta: 0:04:16 lr: 0.001020 min_lr: 0.001020 loss: 1.9207 (2.0181) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [305] [ 50/312] eta: 0:03:54 lr: 0.001020 min_lr: 0.001020 loss: 2.0916 (2.0214) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [305] [ 60/312] eta: 0:03:37 lr: 0.001019 min_lr: 0.001019 loss: 1.9236 (1.9908) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [305] [ 70/312] eta: 0:03:22 lr: 0.001019 min_lr: 0.001019 loss: 1.8455 (1.9866) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [305] [ 80/312] eta: 0:03:10 lr: 0.001019 min_lr: 0.001019 loss: 1.9600 (1.9720) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [305] [ 90/312] eta: 0:02:59 lr: 0.001018 min_lr: 0.001018 loss: 1.8269 (1.9441) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [305] [100/312] eta: 0:02:48 lr: 0.001018 min_lr: 0.001018 loss: 1.8269 (1.9398) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [305] [110/312] eta: 0:02:39 lr: 0.001017 min_lr: 0.001017 loss: 1.9182 (1.9435) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [305] [120/312] eta: 0:02:29 lr: 0.001017 min_lr: 0.001017 loss: 1.9234 (1.9386) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [305] [130/312] eta: 0:02:20 lr: 0.001017 min_lr: 0.001017 loss: 1.8715 (1.9317) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [305] [140/312] eta: 0:02:11 lr: 0.001016 min_lr: 0.001016 loss: 1.8715 (1.9210) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [305] [150/312] eta: 0:02:03 lr: 0.001016 min_lr: 0.001016 loss: 1.9521 (1.9214) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [305] [160/312] eta: 0:01:55 lr: 0.001015 min_lr: 0.001015 loss: 2.0880 (1.9273) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [305] [170/312] eta: 0:01:47 lr: 0.001015 min_lr: 0.001015 loss: 1.6221 (1.9053) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [305] [180/312] eta: 0:01:39 lr: 0.001015 min_lr: 0.001015 loss: 1.6492 (1.9036) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [305] [190/312] eta: 0:01:31 lr: 0.001014 min_lr: 0.001014 loss: 2.0038 (1.9110) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0004 max mem: 64948 Epoch: [305] [200/312] eta: 0:01:23 lr: 0.001014 min_lr: 0.001014 loss: 2.0986 (1.9128) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0004 max mem: 64948 Epoch: [305] [210/312] eta: 0:01:15 lr: 0.001013 min_lr: 0.001013 loss: 2.0142 (1.9135) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0004 max mem: 64948 Epoch: [305] [220/312] eta: 0:01:08 lr: 0.001013 min_lr: 0.001013 loss: 1.7893 (1.9071) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [305] [230/312] eta: 0:01:00 lr: 0.001013 min_lr: 0.001013 loss: 1.9287 (1.9111) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [305] [240/312] eta: 0:00:53 lr: 0.001012 min_lr: 0.001012 loss: 2.0390 (1.9142) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [305] [250/312] eta: 0:00:45 lr: 0.001012 min_lr: 0.001012 loss: 1.9017 (1.9120) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [305] [260/312] eta: 0:00:38 lr: 0.001011 min_lr: 0.001011 loss: 1.9017 (1.9101) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [305] [270/312] eta: 0:00:30 lr: 0.001011 min_lr: 0.001011 loss: 1.9878 (1.9140) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [305] [280/312] eta: 0:00:23 lr: 0.001010 min_lr: 0.001010 loss: 2.0263 (1.9136) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [305] [290/312] eta: 0:00:16 lr: 0.001010 min_lr: 0.001010 loss: 1.8959 (1.9071) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [305] [300/312] eta: 0:00:08 lr: 0.001010 min_lr: 0.001010 loss: 1.6302 (1.9016) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [305] [310/312] eta: 0:00:01 lr: 0.001009 min_lr: 0.001009 loss: 1.7807 (1.8973) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [305] [311/312] eta: 0:00:00 lr: 0.001009 min_lr: 0.001009 loss: 1.8186 (1.8976) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [305] Total time: 0:03:47 (0.7295 s / it) Averaged stats: lr: 0.001009 min_lr: 0.001009 loss: 1.8186 (1.9019) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5115 (0.5115) acc1: 86.1979 (86.1979) acc5: 96.8750 (96.8750) time: 4.5567 data: 4.3457 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7258 (0.7244) acc1: 80.7292 (80.3520) acc5: 96.3542 (95.8720) time: 0.6575 data: 0.4829 max mem: 64948 Test: Total time: 0:00:06 (0.6828 s / it) * Acc@1 81.350 Acc@5 95.610 loss 0.717 Accuracy of the model on the 50000 test images: 81.4% Max accuracy: 81.46% Test: [0/9] eta: 0:00:41 loss: 0.4713 (0.4713) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.5700 data: 4.3671 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6433 (0.6586) acc1: 82.8125 (81.2480) acc5: 97.3958 (96.6080) time: 0.6591 data: 0.4853 max mem: 64948 Test: Total time: 0:00:06 (0.6748 s / it) * Acc@1 82.796 Acc@5 96.440 loss 0.635 Accuracy of the model EMA on 50000 test images: 82.8% Epoch: [306] [ 0/312] eta: 0:59:50 lr: 0.001009 min_lr: 0.001009 loss: 1.8900 (1.8900) weight_decay: 0.0500 (0.0500) time: 11.5077 data: 6.9124 max mem: 64948 Epoch: [306] [ 10/312] eta: 0:08:34 lr: 0.001009 min_lr: 0.001009 loss: 1.9382 (1.8980) weight_decay: 0.0500 (0.0500) time: 1.7020 data: 0.6288 max mem: 64948 Epoch: [306] [ 20/312] eta: 0:05:57 lr: 0.001008 min_lr: 0.001008 loss: 1.9382 (1.8836) weight_decay: 0.0500 (0.0500) time: 0.7085 data: 0.0004 max mem: 64948 Epoch: [306] [ 30/312] eta: 0:04:56 lr: 0.001008 min_lr: 0.001008 loss: 1.9247 (1.9092) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [306] [ 40/312] eta: 0:04:22 lr: 0.001008 min_lr: 0.001008 loss: 2.0964 (1.9633) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [306] [ 50/312] eta: 0:03:58 lr: 0.001007 min_lr: 0.001007 loss: 2.0888 (1.9418) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [306] [ 60/312] eta: 0:03:40 lr: 0.001007 min_lr: 0.001007 loss: 1.9574 (1.9468) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [306] [ 70/312] eta: 0:03:26 lr: 0.001006 min_lr: 0.001006 loss: 1.9265 (1.9239) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [306] [ 80/312] eta: 0:03:13 lr: 0.001006 min_lr: 0.001006 loss: 1.6692 (1.9080) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [306] [ 90/312] eta: 0:03:01 lr: 0.001006 min_lr: 0.001006 loss: 1.9309 (1.9142) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [306] [100/312] eta: 0:02:50 lr: 0.001005 min_lr: 0.001005 loss: 2.0470 (1.9179) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [306] [110/312] eta: 0:02:40 lr: 0.001005 min_lr: 0.001005 loss: 1.8463 (1.9052) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [306] [120/312] eta: 0:02:31 lr: 0.001004 min_lr: 0.001004 loss: 1.9318 (1.9057) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [306] [130/312] eta: 0:02:22 lr: 0.001004 min_lr: 0.001004 loss: 1.9802 (1.9081) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [306] [140/312] eta: 0:02:13 lr: 0.001004 min_lr: 0.001004 loss: 2.0815 (1.9166) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [306] [150/312] eta: 0:02:04 lr: 0.001003 min_lr: 0.001003 loss: 2.0121 (1.9134) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [306] [160/312] eta: 0:01:56 lr: 0.001003 min_lr: 0.001003 loss: 1.9720 (1.9146) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [306] [170/312] eta: 0:01:47 lr: 0.001002 min_lr: 0.001002 loss: 1.8813 (1.9093) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [306] [180/312] eta: 0:01:39 lr: 0.001002 min_lr: 0.001002 loss: 1.8669 (1.9043) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [306] [190/312] eta: 0:01:31 lr: 0.001001 min_lr: 0.001001 loss: 1.9350 (1.9034) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [306] [200/312] eta: 0:01:24 lr: 0.001001 min_lr: 0.001001 loss: 1.8430 (1.8985) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [306] [210/312] eta: 0:01:16 lr: 0.001001 min_lr: 0.001001 loss: 1.8724 (1.9026) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [306] [220/312] eta: 0:01:08 lr: 0.001000 min_lr: 0.001000 loss: 2.0297 (1.8965) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [306] [230/312] eta: 0:01:00 lr: 0.001000 min_lr: 0.001000 loss: 2.0418 (1.9011) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [306] [240/312] eta: 0:00:53 lr: 0.000999 min_lr: 0.000999 loss: 2.0229 (1.9005) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [306] [250/312] eta: 0:00:45 lr: 0.000999 min_lr: 0.000999 loss: 1.8885 (1.9033) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [306] [260/312] eta: 0:00:38 lr: 0.000999 min_lr: 0.000999 loss: 2.1152 (1.9122) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [306] [270/312] eta: 0:00:30 lr: 0.000998 min_lr: 0.000998 loss: 2.0409 (1.9123) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [306] [280/312] eta: 0:00:23 lr: 0.000998 min_lr: 0.000998 loss: 1.9953 (1.9174) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0009 max mem: 64948 Epoch: [306] [290/312] eta: 0:00:16 lr: 0.000997 min_lr: 0.000997 loss: 1.9741 (1.9113) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0008 max mem: 64948 Epoch: [306] [300/312] eta: 0:00:08 lr: 0.000997 min_lr: 0.000997 loss: 1.9825 (1.9146) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [306] [310/312] eta: 0:00:01 lr: 0.000997 min_lr: 0.000997 loss: 1.8536 (1.9130) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [306] [311/312] eta: 0:00:00 lr: 0.000997 min_lr: 0.000997 loss: 1.8536 (1.9137) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [306] Total time: 0:03:48 (0.7331 s / it) Averaged stats: lr: 0.000997 min_lr: 0.000997 loss: 1.8536 (1.9073) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5009 (0.5009) acc1: 87.2396 (87.2396) acc5: 97.1354 (97.1354) time: 4.6707 data: 4.4483 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7221 (0.7256) acc1: 81.7708 (80.8000) acc5: 96.0938 (95.8080) time: 0.6705 data: 0.4944 max mem: 64948 Test: Total time: 0:00:06 (0.6964 s / it) * Acc@1 81.498 Acc@5 95.720 loss 0.707 Accuracy of the model on the 50000 test images: 81.5% Max accuracy: 81.50% Test: [0/9] eta: 0:00:41 loss: 0.4711 (0.4711) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.5659 data: 4.3518 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6420 (0.6580) acc1: 82.8125 (81.2160) acc5: 97.3958 (96.6400) time: 0.6586 data: 0.4836 max mem: 64948 Test: Total time: 0:00:06 (0.6668 s / it) * Acc@1 82.816 Acc@5 96.446 loss 0.634 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.82% Epoch: [307] [ 0/312] eta: 0:57:49 lr: 0.000997 min_lr: 0.000997 loss: 2.0638 (2.0638) weight_decay: 0.0500 (0.0500) time: 11.1210 data: 10.4134 max mem: 64948 Epoch: [307] [ 10/312] eta: 0:08:17 lr: 0.000996 min_lr: 0.000996 loss: 1.9880 (2.0305) weight_decay: 0.0500 (0.0500) time: 1.6481 data: 0.9470 max mem: 64948 Epoch: [307] [ 20/312] eta: 0:05:49 lr: 0.000996 min_lr: 0.000996 loss: 1.8909 (1.9268) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0003 max mem: 64948 Epoch: [307] [ 30/312] eta: 0:04:51 lr: 0.000995 min_lr: 0.000995 loss: 1.8925 (1.9340) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [307] [ 40/312] eta: 0:04:19 lr: 0.000995 min_lr: 0.000995 loss: 1.9542 (1.9128) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [307] [ 50/312] eta: 0:03:56 lr: 0.000995 min_lr: 0.000995 loss: 1.9542 (1.9269) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [307] [ 60/312] eta: 0:03:38 lr: 0.000994 min_lr: 0.000994 loss: 1.9310 (1.9265) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [307] [ 70/312] eta: 0:03:24 lr: 0.000994 min_lr: 0.000994 loss: 1.9600 (1.9374) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [307] [ 80/312] eta: 0:03:11 lr: 0.000993 min_lr: 0.000993 loss: 1.8538 (1.8947) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [307] [ 90/312] eta: 0:02:59 lr: 0.000993 min_lr: 0.000993 loss: 1.8069 (1.9032) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [307] [100/312] eta: 0:02:49 lr: 0.000992 min_lr: 0.000992 loss: 2.0287 (1.9099) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [307] [110/312] eta: 0:02:39 lr: 0.000992 min_lr: 0.000992 loss: 1.9178 (1.9019) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [307] [120/312] eta: 0:02:30 lr: 0.000992 min_lr: 0.000992 loss: 1.8629 (1.9043) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [307] [130/312] eta: 0:02:21 lr: 0.000991 min_lr: 0.000991 loss: 2.0038 (1.9033) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [307] [140/312] eta: 0:02:12 lr: 0.000991 min_lr: 0.000991 loss: 2.0411 (1.9100) weight_decay: 0.0500 (0.0500) time: 0.7033 data: 0.0004 max mem: 64948 Epoch: [307] [150/312] eta: 0:02:04 lr: 0.000990 min_lr: 0.000990 loss: 1.9112 (1.9068) weight_decay: 0.0500 (0.0500) time: 0.7028 data: 0.0004 max mem: 64948 Epoch: [307] [160/312] eta: 0:01:55 lr: 0.000990 min_lr: 0.000990 loss: 1.8947 (1.9029) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [307] [170/312] eta: 0:01:47 lr: 0.000990 min_lr: 0.000990 loss: 1.7362 (1.8942) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [307] [180/312] eta: 0:01:39 lr: 0.000989 min_lr: 0.000989 loss: 1.9262 (1.8989) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [307] [190/312] eta: 0:01:31 lr: 0.000989 min_lr: 0.000989 loss: 2.0343 (1.9022) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [307] [200/312] eta: 0:01:23 lr: 0.000988 min_lr: 0.000988 loss: 1.9795 (1.8956) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [307] [210/312] eta: 0:01:16 lr: 0.000988 min_lr: 0.000988 loss: 1.8555 (1.8868) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [307] [220/312] eta: 0:01:08 lr: 0.000988 min_lr: 0.000988 loss: 1.8644 (1.8879) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [307] [230/312] eta: 0:01:00 lr: 0.000987 min_lr: 0.000987 loss: 1.8798 (1.8811) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [307] [240/312] eta: 0:00:53 lr: 0.000987 min_lr: 0.000987 loss: 1.8989 (1.8793) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [307] [250/312] eta: 0:00:45 lr: 0.000986 min_lr: 0.000986 loss: 1.9395 (1.8816) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [307] [260/312] eta: 0:00:38 lr: 0.000986 min_lr: 0.000986 loss: 1.9740 (1.8870) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [307] [270/312] eta: 0:00:30 lr: 0.000986 min_lr: 0.000986 loss: 1.9379 (1.8889) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [307] [280/312] eta: 0:00:23 lr: 0.000985 min_lr: 0.000985 loss: 1.8053 (1.8864) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [307] [290/312] eta: 0:00:16 lr: 0.000985 min_lr: 0.000985 loss: 1.8053 (1.8870) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [307] [300/312] eta: 0:00:08 lr: 0.000984 min_lr: 0.000984 loss: 1.9961 (1.8867) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [307] [310/312] eta: 0:00:01 lr: 0.000984 min_lr: 0.000984 loss: 1.9585 (1.8864) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [307] [311/312] eta: 0:00:00 lr: 0.000984 min_lr: 0.000984 loss: 1.9930 (1.8870) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [307] Total time: 0:03:48 (0.7318 s / it) Averaged stats: lr: 0.000984 min_lr: 0.000984 loss: 1.9930 (1.8970) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.5336 (0.5336) acc1: 86.7188 (86.7188) acc5: 97.1354 (97.1354) time: 4.2584 data: 4.0513 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7120 (0.7275) acc1: 81.2500 (80.4160) acc5: 96.0938 (95.8080) time: 0.6278 data: 0.4536 max mem: 64948 Test: Total time: 0:00:05 (0.6455 s / it) * Acc@1 81.302 Acc@5 95.612 loss 0.713 Accuracy of the model on the 50000 test images: 81.3% Max accuracy: 81.50% Test: [0/9] eta: 0:00:45 loss: 0.4706 (0.4706) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 5.0318 data: 4.8205 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6415 (0.6575) acc1: 83.0729 (81.2800) acc5: 97.3958 (96.6720) time: 0.7104 data: 0.5357 max mem: 64948 Test: Total time: 0:00:06 (0.7195 s / it) * Acc@1 82.808 Acc@5 96.460 loss 0.634 Accuracy of the model EMA on 50000 test images: 82.8% Epoch: [308] [ 0/312] eta: 0:57:53 lr: 0.000984 min_lr: 0.000984 loss: 1.7291 (1.7291) weight_decay: 0.0500 (0.0500) time: 11.1331 data: 7.7799 max mem: 64948 Epoch: [308] [ 10/312] eta: 0:08:23 lr: 0.000984 min_lr: 0.000984 loss: 1.7422 (1.8488) weight_decay: 0.0500 (0.0500) time: 1.6682 data: 0.7076 max mem: 64948 Epoch: [308] [ 20/312] eta: 0:05:52 lr: 0.000983 min_lr: 0.000983 loss: 1.9857 (1.9086) weight_decay: 0.0500 (0.0500) time: 0.7109 data: 0.0004 max mem: 64948 Epoch: [308] [ 30/312] eta: 0:04:53 lr: 0.000983 min_lr: 0.000983 loss: 2.0277 (1.9576) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [308] [ 40/312] eta: 0:04:20 lr: 0.000982 min_lr: 0.000982 loss: 2.0265 (1.9251) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [308] [ 50/312] eta: 0:03:57 lr: 0.000982 min_lr: 0.000982 loss: 1.8781 (1.9181) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [308] [ 60/312] eta: 0:03:39 lr: 0.000982 min_lr: 0.000982 loss: 2.0047 (1.9261) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [308] [ 70/312] eta: 0:03:24 lr: 0.000981 min_lr: 0.000981 loss: 2.0518 (1.9348) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [308] [ 80/312] eta: 0:03:12 lr: 0.000981 min_lr: 0.000981 loss: 1.8836 (1.9166) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [308] [ 90/312] eta: 0:03:00 lr: 0.000980 min_lr: 0.000980 loss: 1.8447 (1.9119) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [308] [100/312] eta: 0:02:50 lr: 0.000980 min_lr: 0.000980 loss: 1.9718 (1.9116) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [308] [110/312] eta: 0:02:40 lr: 0.000979 min_lr: 0.000979 loss: 1.6533 (1.8862) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [308] [120/312] eta: 0:02:30 lr: 0.000979 min_lr: 0.000979 loss: 1.6252 (1.8812) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [308] [130/312] eta: 0:02:21 lr: 0.000979 min_lr: 0.000979 loss: 1.6847 (1.8728) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [308] [140/312] eta: 0:02:12 lr: 0.000978 min_lr: 0.000978 loss: 1.9555 (1.8800) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [308] [150/312] eta: 0:02:04 lr: 0.000978 min_lr: 0.000978 loss: 1.8485 (1.8727) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [308] [160/312] eta: 0:01:55 lr: 0.000977 min_lr: 0.000977 loss: 1.7882 (1.8671) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [308] [170/312] eta: 0:01:47 lr: 0.000977 min_lr: 0.000977 loss: 1.9224 (1.8640) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [308] [180/312] eta: 0:01:39 lr: 0.000977 min_lr: 0.000977 loss: 1.9514 (1.8654) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [308] [190/312] eta: 0:01:31 lr: 0.000976 min_lr: 0.000976 loss: 1.8672 (1.8680) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [308] [200/312] eta: 0:01:23 lr: 0.000976 min_lr: 0.000976 loss: 1.7864 (1.8621) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [308] [210/312] eta: 0:01:16 lr: 0.000975 min_lr: 0.000975 loss: 1.7543 (1.8696) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [308] [220/312] eta: 0:01:08 lr: 0.000975 min_lr: 0.000975 loss: 1.8631 (1.8667) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [308] [230/312] eta: 0:01:00 lr: 0.000975 min_lr: 0.000975 loss: 1.7151 (1.8617) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [308] [240/312] eta: 0:00:53 lr: 0.000974 min_lr: 0.000974 loss: 1.9154 (1.8634) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [308] [250/312] eta: 0:00:45 lr: 0.000974 min_lr: 0.000974 loss: 1.8842 (1.8614) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [308] [260/312] eta: 0:00:38 lr: 0.000973 min_lr: 0.000973 loss: 1.8531 (1.8530) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [308] [270/312] eta: 0:00:30 lr: 0.000973 min_lr: 0.000973 loss: 1.9497 (1.8599) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [308] [280/312] eta: 0:00:23 lr: 0.000973 min_lr: 0.000973 loss: 2.0267 (1.8625) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0010 max mem: 64948 Epoch: [308] [290/312] eta: 0:00:16 lr: 0.000972 min_lr: 0.000972 loss: 2.0223 (1.8638) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [308] [300/312] eta: 0:00:08 lr: 0.000972 min_lr: 0.000972 loss: 2.0239 (1.8678) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [308] [310/312] eta: 0:00:01 lr: 0.000971 min_lr: 0.000971 loss: 2.0646 (1.8703) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [308] [311/312] eta: 0:00:00 lr: 0.000971 min_lr: 0.000971 loss: 2.0646 (1.8696) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [308] Total time: 0:03:48 (0.7323 s / it) Averaged stats: lr: 0.000971 min_lr: 0.000971 loss: 2.0646 (1.8992) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4801 (0.4801) acc1: 86.9792 (86.9792) acc5: 97.6562 (97.6562) time: 4.5174 data: 4.3097 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6862 (0.7094) acc1: 82.8125 (81.0880) acc5: 95.8333 (95.9360) time: 0.6532 data: 0.4789 max mem: 64948 Test: Total time: 0:00:06 (0.6737 s / it) * Acc@1 81.630 Acc@5 95.790 loss 0.700 Accuracy of the model on the 50000 test images: 81.6% Max accuracy: 81.63% Test: [0/9] eta: 0:00:40 loss: 0.4699 (0.4699) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.5024 data: 4.2993 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6410 (0.6571) acc1: 83.0729 (81.2800) acc5: 97.3958 (96.7040) time: 0.6516 data: 0.4778 max mem: 64948 Test: Total time: 0:00:05 (0.6614 s / it) * Acc@1 82.814 Acc@5 96.466 loss 0.634 Accuracy of the model EMA on 50000 test images: 82.8% Epoch: [309] [ 0/312] eta: 1:01:20 lr: 0.000971 min_lr: 0.000971 loss: 2.2865 (2.2865) weight_decay: 0.0500 (0.0500) time: 11.7959 data: 9.1036 max mem: 64948 Epoch: [309] [ 10/312] eta: 0:08:40 lr: 0.000971 min_lr: 0.000971 loss: 2.0704 (2.0657) weight_decay: 0.0500 (0.0500) time: 1.7221 data: 0.8280 max mem: 64948 Epoch: [309] [ 20/312] eta: 0:05:59 lr: 0.000971 min_lr: 0.000971 loss: 1.9997 (2.0374) weight_decay: 0.0500 (0.0500) time: 0.7032 data: 0.0004 max mem: 64948 Epoch: [309] [ 30/312] eta: 0:04:58 lr: 0.000970 min_lr: 0.000970 loss: 1.8902 (1.9727) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [309] [ 40/312] eta: 0:04:23 lr: 0.000970 min_lr: 0.000970 loss: 1.9121 (1.9749) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [309] [ 50/312] eta: 0:04:00 lr: 0.000969 min_lr: 0.000969 loss: 1.9332 (1.9677) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [309] [ 60/312] eta: 0:03:42 lr: 0.000969 min_lr: 0.000969 loss: 1.9386 (1.9764) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [309] [ 70/312] eta: 0:03:27 lr: 0.000969 min_lr: 0.000969 loss: 1.8928 (1.9658) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0004 max mem: 64948 Epoch: [309] [ 80/312] eta: 0:03:13 lr: 0.000968 min_lr: 0.000968 loss: 1.8928 (1.9532) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [309] [ 90/312] eta: 0:03:02 lr: 0.000968 min_lr: 0.000968 loss: 1.9861 (1.9581) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [309] [100/312] eta: 0:02:51 lr: 0.000967 min_lr: 0.000967 loss: 2.1198 (1.9551) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [309] [110/312] eta: 0:02:41 lr: 0.000967 min_lr: 0.000967 loss: 2.1605 (1.9628) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [309] [120/312] eta: 0:02:31 lr: 0.000967 min_lr: 0.000967 loss: 2.0389 (1.9492) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [309] [130/312] eta: 0:02:22 lr: 0.000966 min_lr: 0.000966 loss: 1.9102 (1.9511) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [309] [140/312] eta: 0:02:13 lr: 0.000966 min_lr: 0.000966 loss: 1.9841 (1.9398) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [309] [150/312] eta: 0:02:04 lr: 0.000965 min_lr: 0.000965 loss: 1.9990 (1.9467) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [309] [160/312] eta: 0:01:56 lr: 0.000965 min_lr: 0.000965 loss: 1.9783 (1.9485) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [309] [170/312] eta: 0:01:48 lr: 0.000965 min_lr: 0.000965 loss: 1.9690 (1.9398) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [309] [180/312] eta: 0:01:40 lr: 0.000964 min_lr: 0.000964 loss: 1.9324 (1.9487) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [309] [190/312] eta: 0:01:32 lr: 0.000964 min_lr: 0.000964 loss: 1.9949 (1.9513) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [309] [200/312] eta: 0:01:24 lr: 0.000963 min_lr: 0.000963 loss: 1.9949 (1.9537) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [309] [210/312] eta: 0:01:16 lr: 0.000963 min_lr: 0.000963 loss: 1.9357 (1.9463) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [309] [220/312] eta: 0:01:08 lr: 0.000963 min_lr: 0.000963 loss: 1.8897 (1.9477) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [309] [230/312] eta: 0:01:01 lr: 0.000962 min_lr: 0.000962 loss: 2.0065 (1.9452) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [309] [240/312] eta: 0:00:53 lr: 0.000962 min_lr: 0.000962 loss: 1.9233 (1.9449) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [309] [250/312] eta: 0:00:45 lr: 0.000961 min_lr: 0.000961 loss: 1.9222 (1.9422) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [309] [260/312] eta: 0:00:38 lr: 0.000961 min_lr: 0.000961 loss: 2.0126 (1.9490) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [309] [270/312] eta: 0:00:30 lr: 0.000961 min_lr: 0.000961 loss: 2.0126 (1.9428) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [309] [280/312] eta: 0:00:23 lr: 0.000960 min_lr: 0.000960 loss: 1.9013 (1.9432) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [309] [290/312] eta: 0:00:16 lr: 0.000960 min_lr: 0.000960 loss: 1.9104 (1.9427) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [309] [300/312] eta: 0:00:08 lr: 0.000959 min_lr: 0.000959 loss: 2.0511 (1.9469) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [309] [310/312] eta: 0:00:01 lr: 0.000959 min_lr: 0.000959 loss: 2.0511 (1.9445) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [309] [311/312] eta: 0:00:00 lr: 0.000959 min_lr: 0.000959 loss: 2.0329 (1.9432) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [309] Total time: 0:03:49 (0.7346 s / it) Averaged stats: lr: 0.000959 min_lr: 0.000959 loss: 2.0329 (1.8978) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4970 (0.4970) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.6832 data: 4.4741 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7351 (0.7233) acc1: 81.5104 (80.2560) acc5: 97.1354 (96.2560) time: 0.6716 data: 0.4972 max mem: 64948 Test: Total time: 0:00:06 (0.6911 s / it) * Acc@1 81.408 Acc@5 95.666 loss 0.708 Accuracy of the model on the 50000 test images: 81.4% Max accuracy: 81.63% Test: [0/9] eta: 0:00:43 loss: 0.4689 (0.4689) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.8408 data: 4.6250 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6408 (0.6567) acc1: 83.0729 (81.2800) acc5: 97.3958 (96.7040) time: 0.6892 data: 0.5140 max mem: 64948 Test: Total time: 0:00:06 (0.6994 s / it) * Acc@1 82.820 Acc@5 96.464 loss 0.634 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.82% Epoch: [310] [ 0/312] eta: 0:54:55 lr: 0.000959 min_lr: 0.000959 loss: 1.4220 (1.4220) weight_decay: 0.0500 (0.0500) time: 10.5628 data: 8.2631 max mem: 64948 Epoch: [310] [ 10/312] eta: 0:08:06 lr: 0.000958 min_lr: 0.000958 loss: 1.8953 (1.8677) weight_decay: 0.0500 (0.0500) time: 1.6096 data: 0.7515 max mem: 64948 Epoch: [310] [ 20/312] eta: 0:05:42 lr: 0.000958 min_lr: 0.000958 loss: 1.9006 (1.9219) weight_decay: 0.0500 (0.0500) time: 0.7047 data: 0.0003 max mem: 64948 Epoch: [310] [ 30/312] eta: 0:04:47 lr: 0.000958 min_lr: 0.000958 loss: 1.9645 (1.9120) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [310] [ 40/312] eta: 0:04:15 lr: 0.000957 min_lr: 0.000957 loss: 1.9048 (1.9090) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [310] [ 50/312] eta: 0:03:54 lr: 0.000957 min_lr: 0.000957 loss: 2.0866 (1.9293) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [310] [ 60/312] eta: 0:03:36 lr: 0.000956 min_lr: 0.000956 loss: 2.0667 (1.9151) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [310] [ 70/312] eta: 0:03:22 lr: 0.000956 min_lr: 0.000956 loss: 2.0192 (1.9268) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [310] [ 80/312] eta: 0:03:10 lr: 0.000956 min_lr: 0.000956 loss: 2.0405 (1.9228) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [310] [ 90/312] eta: 0:02:59 lr: 0.000955 min_lr: 0.000955 loss: 1.8764 (1.9192) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [310] [100/312] eta: 0:02:48 lr: 0.000955 min_lr: 0.000955 loss: 1.9581 (1.9244) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [310] [110/312] eta: 0:02:38 lr: 0.000954 min_lr: 0.000954 loss: 1.9231 (1.9184) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [310] [120/312] eta: 0:02:29 lr: 0.000954 min_lr: 0.000954 loss: 1.9161 (1.9110) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [310] [130/312] eta: 0:02:20 lr: 0.000954 min_lr: 0.000954 loss: 1.7566 (1.9091) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [310] [140/312] eta: 0:02:11 lr: 0.000953 min_lr: 0.000953 loss: 1.7566 (1.8994) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [310] [150/312] eta: 0:02:03 lr: 0.000953 min_lr: 0.000953 loss: 1.9241 (1.8957) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [310] [160/312] eta: 0:01:55 lr: 0.000952 min_lr: 0.000952 loss: 2.0133 (1.8991) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [310] [170/312] eta: 0:01:47 lr: 0.000952 min_lr: 0.000952 loss: 2.0012 (1.9023) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [310] [180/312] eta: 0:01:39 lr: 0.000952 min_lr: 0.000952 loss: 1.9086 (1.8958) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [310] [190/312] eta: 0:01:31 lr: 0.000951 min_lr: 0.000951 loss: 1.9766 (1.8983) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [310] [200/312] eta: 0:01:23 lr: 0.000951 min_lr: 0.000951 loss: 1.9917 (1.8956) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [310] [210/312] eta: 0:01:15 lr: 0.000950 min_lr: 0.000950 loss: 1.9306 (1.8970) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [310] [220/312] eta: 0:01:08 lr: 0.000950 min_lr: 0.000950 loss: 1.9769 (1.9007) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [310] [230/312] eta: 0:01:00 lr: 0.000950 min_lr: 0.000950 loss: 1.9834 (1.9017) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [310] [240/312] eta: 0:00:53 lr: 0.000949 min_lr: 0.000949 loss: 1.9061 (1.8969) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [310] [250/312] eta: 0:00:45 lr: 0.000949 min_lr: 0.000949 loss: 1.8788 (1.8961) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [310] [260/312] eta: 0:00:38 lr: 0.000949 min_lr: 0.000949 loss: 1.9453 (1.8954) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [310] [270/312] eta: 0:00:30 lr: 0.000948 min_lr: 0.000948 loss: 1.8117 (1.8897) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [310] [280/312] eta: 0:00:23 lr: 0.000948 min_lr: 0.000948 loss: 1.7503 (1.8919) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0012 max mem: 64948 Epoch: [310] [290/312] eta: 0:00:16 lr: 0.000947 min_lr: 0.000947 loss: 1.9837 (1.8949) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0011 max mem: 64948 Epoch: [310] [300/312] eta: 0:00:08 lr: 0.000947 min_lr: 0.000947 loss: 1.9922 (1.8947) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [310] [310/312] eta: 0:00:01 lr: 0.000947 min_lr: 0.000947 loss: 1.9922 (1.8978) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [310] [311/312] eta: 0:00:00 lr: 0.000946 min_lr: 0.000946 loss: 1.9977 (1.8990) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [310] Total time: 0:03:47 (0.7302 s / it) Averaged stats: lr: 0.000946 min_lr: 0.000946 loss: 1.9977 (1.8990) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4823 (0.4823) acc1: 86.9792 (86.9792) acc5: 97.9167 (97.9167) time: 4.5644 data: 4.3552 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7386 (0.7388) acc1: 82.2917 (80.4480) acc5: 96.0938 (95.5840) time: 0.6584 data: 0.4840 max mem: 64948 Test: Total time: 0:00:06 (0.6715 s / it) * Acc@1 81.332 Acc@5 95.638 loss 0.714 Accuracy of the model on the 50000 test images: 81.3% Max accuracy: 81.63% Test: [0/9] eta: 0:00:43 loss: 0.4683 (0.4683) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.8865 data: 4.6792 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6404 (0.6564) acc1: 83.0729 (81.2800) acc5: 97.3958 (96.7360) time: 0.6943 data: 0.5200 max mem: 64948 Test: Total time: 0:00:06 (0.7024 s / it) * Acc@1 82.844 Acc@5 96.470 loss 0.633 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.84% Epoch: [311] [ 0/312] eta: 0:52:23 lr: 0.000946 min_lr: 0.000946 loss: 1.7418 (1.7418) weight_decay: 0.0500 (0.0500) time: 10.0744 data: 9.2896 max mem: 64948 Epoch: [311] [ 10/312] eta: 0:07:53 lr: 0.000946 min_lr: 0.000946 loss: 1.9375 (1.8344) weight_decay: 0.0500 (0.0500) time: 1.5673 data: 0.8449 max mem: 64948 Epoch: [311] [ 20/312] eta: 0:05:36 lr: 0.000946 min_lr: 0.000946 loss: 1.8410 (1.7861) weight_decay: 0.0500 (0.0500) time: 0.7075 data: 0.0004 max mem: 64948 Epoch: [311] [ 30/312] eta: 0:04:43 lr: 0.000945 min_lr: 0.000945 loss: 1.8689 (1.8474) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [311] [ 40/312] eta: 0:04:12 lr: 0.000945 min_lr: 0.000945 loss: 1.9282 (1.8546) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [311] [ 50/312] eta: 0:03:51 lr: 0.000944 min_lr: 0.000944 loss: 1.8473 (1.8488) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [311] [ 60/312] eta: 0:03:34 lr: 0.000944 min_lr: 0.000944 loss: 1.9280 (1.8510) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [311] [ 70/312] eta: 0:03:21 lr: 0.000944 min_lr: 0.000944 loss: 1.9781 (1.8629) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [311] [ 80/312] eta: 0:03:08 lr: 0.000943 min_lr: 0.000943 loss: 2.0232 (1.8901) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [311] [ 90/312] eta: 0:02:57 lr: 0.000943 min_lr: 0.000943 loss: 1.9621 (1.8766) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [311] [100/312] eta: 0:02:47 lr: 0.000942 min_lr: 0.000942 loss: 1.8812 (1.8711) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [311] [110/312] eta: 0:02:38 lr: 0.000942 min_lr: 0.000942 loss: 1.9188 (1.8742) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [311] [120/312] eta: 0:02:28 lr: 0.000942 min_lr: 0.000942 loss: 1.9665 (1.8820) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [311] [130/312] eta: 0:02:19 lr: 0.000941 min_lr: 0.000941 loss: 1.9111 (1.8786) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [311] [140/312] eta: 0:02:11 lr: 0.000941 min_lr: 0.000941 loss: 1.9016 (1.8819) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [311] [150/312] eta: 0:02:02 lr: 0.000940 min_lr: 0.000940 loss: 2.0178 (1.8826) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [311] [160/312] eta: 0:01:54 lr: 0.000940 min_lr: 0.000940 loss: 2.0178 (1.8786) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [311] [170/312] eta: 0:01:46 lr: 0.000940 min_lr: 0.000940 loss: 1.9056 (1.8774) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [311] [180/312] eta: 0:01:38 lr: 0.000939 min_lr: 0.000939 loss: 1.9056 (1.8756) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [311] [190/312] eta: 0:01:30 lr: 0.000939 min_lr: 0.000939 loss: 1.8714 (1.8772) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [311] [200/312] eta: 0:01:23 lr: 0.000938 min_lr: 0.000938 loss: 1.9801 (1.8820) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [311] [210/312] eta: 0:01:15 lr: 0.000938 min_lr: 0.000938 loss: 2.0442 (1.8853) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [311] [220/312] eta: 0:01:07 lr: 0.000938 min_lr: 0.000938 loss: 1.9673 (1.8834) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [311] [230/312] eta: 0:01:00 lr: 0.000937 min_lr: 0.000937 loss: 1.8378 (1.8804) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [311] [240/312] eta: 0:00:52 lr: 0.000937 min_lr: 0.000937 loss: 1.9411 (1.8805) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [311] [250/312] eta: 0:00:45 lr: 0.000937 min_lr: 0.000937 loss: 1.9840 (1.8808) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [311] [260/312] eta: 0:00:38 lr: 0.000936 min_lr: 0.000936 loss: 1.9532 (1.8838) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [311] [270/312] eta: 0:00:30 lr: 0.000936 min_lr: 0.000936 loss: 1.9832 (1.8879) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [311] [280/312] eta: 0:00:23 lr: 0.000935 min_lr: 0.000935 loss: 2.0375 (1.8896) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0010 max mem: 64948 Epoch: [311] [290/312] eta: 0:00:16 lr: 0.000935 min_lr: 0.000935 loss: 1.8733 (1.8843) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0009 max mem: 64948 Epoch: [311] [300/312] eta: 0:00:08 lr: 0.000935 min_lr: 0.000935 loss: 1.8733 (1.8870) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [311] [310/312] eta: 0:00:01 lr: 0.000934 min_lr: 0.000934 loss: 1.9972 (1.8912) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [311] [311/312] eta: 0:00:00 lr: 0.000934 min_lr: 0.000934 loss: 2.0009 (1.8918) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [311] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.000934 min_lr: 0.000934 loss: 2.0009 (1.8964) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5090 (0.5090) acc1: 86.4583 (86.4583) acc5: 97.1354 (97.1354) time: 4.6347 data: 4.4190 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7240 (0.7236) acc1: 81.5104 (80.2880) acc5: 96.6146 (95.8080) time: 0.6663 data: 0.4911 max mem: 64948 Test: Total time: 0:00:06 (0.6882 s / it) * Acc@1 81.578 Acc@5 95.756 loss 0.704 Accuracy of the model on the 50000 test images: 81.6% Max accuracy: 81.63% Test: [0/9] eta: 0:00:43 loss: 0.4673 (0.4673) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.8313 data: 4.6133 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6402 (0.6560) acc1: 82.8125 (81.2480) acc5: 97.3958 (96.7360) time: 0.6883 data: 0.5127 max mem: 64948 Test: Total time: 0:00:06 (0.7024 s / it) * Acc@1 82.832 Acc@5 96.464 loss 0.633 Accuracy of the model EMA on 50000 test images: 82.8% Epoch: [312] [ 0/312] eta: 0:59:28 lr: 0.000934 min_lr: 0.000934 loss: 2.0075 (2.0075) weight_decay: 0.0500 (0.0500) time: 11.4388 data: 8.8285 max mem: 64948 Epoch: [312] [ 10/312] eta: 0:08:32 lr: 0.000934 min_lr: 0.000934 loss: 2.2249 (2.1268) weight_decay: 0.0500 (0.0500) time: 1.6968 data: 0.8030 max mem: 64948 Epoch: [312] [ 20/312] eta: 0:05:56 lr: 0.000933 min_lr: 0.000933 loss: 2.0383 (2.0316) weight_decay: 0.0500 (0.0500) time: 0.7107 data: 0.0004 max mem: 64948 Epoch: [312] [ 30/312] eta: 0:04:56 lr: 0.000933 min_lr: 0.000933 loss: 1.9932 (1.9551) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [312] [ 40/312] eta: 0:04:22 lr: 0.000932 min_lr: 0.000932 loss: 1.9126 (1.9334) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [312] [ 50/312] eta: 0:03:58 lr: 0.000932 min_lr: 0.000932 loss: 2.0694 (1.9589) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [312] [ 60/312] eta: 0:03:40 lr: 0.000932 min_lr: 0.000932 loss: 2.0021 (1.9349) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [312] [ 70/312] eta: 0:03:25 lr: 0.000931 min_lr: 0.000931 loss: 1.8470 (1.9256) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [312] [ 80/312] eta: 0:03:12 lr: 0.000931 min_lr: 0.000931 loss: 1.8470 (1.9069) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [312] [ 90/312] eta: 0:03:01 lr: 0.000930 min_lr: 0.000930 loss: 1.9075 (1.9183) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [312] [100/312] eta: 0:02:50 lr: 0.000930 min_lr: 0.000930 loss: 1.9953 (1.9279) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [312] [110/312] eta: 0:02:40 lr: 0.000930 min_lr: 0.000930 loss: 1.9425 (1.9141) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [312] [120/312] eta: 0:02:31 lr: 0.000929 min_lr: 0.000929 loss: 1.7030 (1.9006) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [312] [130/312] eta: 0:02:21 lr: 0.000929 min_lr: 0.000929 loss: 1.8216 (1.9015) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [312] [140/312] eta: 0:02:13 lr: 0.000929 min_lr: 0.000929 loss: 1.7879 (1.8822) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [312] [150/312] eta: 0:02:04 lr: 0.000928 min_lr: 0.000928 loss: 1.6104 (1.8720) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [312] [160/312] eta: 0:01:56 lr: 0.000928 min_lr: 0.000928 loss: 1.8005 (1.8686) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [312] [170/312] eta: 0:01:48 lr: 0.000927 min_lr: 0.000927 loss: 1.8005 (1.8619) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [312] [180/312] eta: 0:01:39 lr: 0.000927 min_lr: 0.000927 loss: 1.7285 (1.8549) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [312] [190/312] eta: 0:01:31 lr: 0.000927 min_lr: 0.000927 loss: 1.8262 (1.8580) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [312] [200/312] eta: 0:01:24 lr: 0.000926 min_lr: 0.000926 loss: 2.0199 (1.8651) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [312] [210/312] eta: 0:01:16 lr: 0.000926 min_lr: 0.000926 loss: 2.0184 (1.8710) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [312] [220/312] eta: 0:01:08 lr: 0.000925 min_lr: 0.000925 loss: 1.9340 (1.8652) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [312] [230/312] eta: 0:01:00 lr: 0.000925 min_lr: 0.000925 loss: 1.8956 (1.8731) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [312] [240/312] eta: 0:00:53 lr: 0.000925 min_lr: 0.000925 loss: 2.0496 (1.8751) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [312] [250/312] eta: 0:00:45 lr: 0.000924 min_lr: 0.000924 loss: 2.0655 (1.8849) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [312] [260/312] eta: 0:00:38 lr: 0.000924 min_lr: 0.000924 loss: 2.0579 (1.8859) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [312] [270/312] eta: 0:00:30 lr: 0.000923 min_lr: 0.000923 loss: 1.9421 (1.8907) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [312] [280/312] eta: 0:00:23 lr: 0.000923 min_lr: 0.000923 loss: 2.0010 (1.8908) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [312] [290/312] eta: 0:00:16 lr: 0.000923 min_lr: 0.000923 loss: 1.8474 (1.8785) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [312] [300/312] eta: 0:00:08 lr: 0.000922 min_lr: 0.000922 loss: 1.5668 (1.8735) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [312] [310/312] eta: 0:00:01 lr: 0.000922 min_lr: 0.000922 loss: 1.8757 (1.8751) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [312] [311/312] eta: 0:00:00 lr: 0.000922 min_lr: 0.000922 loss: 1.8757 (1.8765) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [312] Total time: 0:03:48 (0.7329 s / it) Averaged stats: lr: 0.000922 min_lr: 0.000922 loss: 1.8757 (1.8838) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.5144 (0.5144) acc1: 86.4583 (86.4583) acc5: 97.1354 (97.1354) time: 4.4071 data: 4.1877 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7217 (0.7266) acc1: 82.2917 (80.5120) acc5: 96.0938 (95.7120) time: 0.6412 data: 0.4654 max mem: 64948 Test: Total time: 0:00:05 (0.6611 s / it) * Acc@1 81.752 Acc@5 95.754 loss 0.699 Accuracy of the model on the 50000 test images: 81.8% Max accuracy: 81.75% Test: [0/9] eta: 0:00:39 loss: 0.4666 (0.4666) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.4273 data: 4.2094 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6402 (0.6558) acc1: 82.5521 (81.2480) acc5: 97.3958 (96.7360) time: 0.6434 data: 0.4678 max mem: 64948 Test: Total time: 0:00:05 (0.6544 s / it) * Acc@1 82.848 Acc@5 96.464 loss 0.633 Accuracy of the model EMA on 50000 test images: 82.8% Max EMA accuracy: 82.85% Epoch: [313] [ 0/312] eta: 0:48:42 lr: 0.000922 min_lr: 0.000922 loss: 2.3453 (2.3453) weight_decay: 0.0500 (0.0500) time: 9.3662 data: 7.9587 max mem: 64948 Epoch: [313] [ 10/312] eta: 0:07:47 lr: 0.000921 min_lr: 0.000921 loss: 1.9908 (1.9074) weight_decay: 0.0500 (0.0500) time: 1.5476 data: 0.7540 max mem: 64948 Epoch: [313] [ 20/312] eta: 0:05:32 lr: 0.000921 min_lr: 0.000921 loss: 1.9998 (1.9775) weight_decay: 0.0500 (0.0500) time: 0.7288 data: 0.0169 max mem: 64948 Epoch: [313] [ 30/312] eta: 0:04:41 lr: 0.000921 min_lr: 0.000921 loss: 2.0545 (1.9538) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0003 max mem: 64948 Epoch: [313] [ 40/312] eta: 0:04:11 lr: 0.000920 min_lr: 0.000920 loss: 2.0014 (1.9487) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0003 max mem: 64948 Epoch: [313] [ 50/312] eta: 0:03:50 lr: 0.000920 min_lr: 0.000920 loss: 2.0243 (1.9461) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [313] [ 60/312] eta: 0:03:34 lr: 0.000919 min_lr: 0.000919 loss: 1.9834 (1.9399) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [313] [ 70/312] eta: 0:03:20 lr: 0.000919 min_lr: 0.000919 loss: 1.8803 (1.9165) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [313] [ 80/312] eta: 0:03:08 lr: 0.000919 min_lr: 0.000919 loss: 1.9351 (1.9294) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [313] [ 90/312] eta: 0:02:57 lr: 0.000918 min_lr: 0.000918 loss: 1.9674 (1.9085) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [313] [100/312] eta: 0:02:47 lr: 0.000918 min_lr: 0.000918 loss: 1.8312 (1.9001) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [313] [110/312] eta: 0:02:37 lr: 0.000917 min_lr: 0.000917 loss: 1.7366 (1.8736) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [313] [120/312] eta: 0:02:28 lr: 0.000917 min_lr: 0.000917 loss: 1.7557 (1.8695) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [313] [130/312] eta: 0:02:19 lr: 0.000917 min_lr: 0.000917 loss: 2.0217 (1.8737) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [313] [140/312] eta: 0:02:11 lr: 0.000916 min_lr: 0.000916 loss: 2.0643 (1.8736) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [313] [150/312] eta: 0:02:02 lr: 0.000916 min_lr: 0.000916 loss: 2.0178 (1.8754) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [313] [160/312] eta: 0:01:54 lr: 0.000915 min_lr: 0.000915 loss: 1.8633 (1.8774) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [313] [170/312] eta: 0:01:46 lr: 0.000915 min_lr: 0.000915 loss: 1.8290 (1.8685) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [313] [180/312] eta: 0:01:38 lr: 0.000915 min_lr: 0.000915 loss: 1.5653 (1.8537) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [313] [190/312] eta: 0:01:30 lr: 0.000914 min_lr: 0.000914 loss: 1.8327 (1.8634) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [313] [200/312] eta: 0:01:23 lr: 0.000914 min_lr: 0.000914 loss: 2.0363 (1.8646) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [313] [210/312] eta: 0:01:15 lr: 0.000913 min_lr: 0.000913 loss: 1.9732 (1.8651) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [313] [220/312] eta: 0:01:07 lr: 0.000913 min_lr: 0.000913 loss: 2.0259 (1.8715) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [313] [230/312] eta: 0:01:00 lr: 0.000913 min_lr: 0.000913 loss: 1.8894 (1.8692) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [313] [240/312] eta: 0:00:52 lr: 0.000912 min_lr: 0.000912 loss: 1.8303 (1.8693) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [313] [250/312] eta: 0:00:45 lr: 0.000912 min_lr: 0.000912 loss: 1.6995 (1.8623) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [313] [260/312] eta: 0:00:38 lr: 0.000911 min_lr: 0.000911 loss: 1.7405 (1.8613) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [313] [270/312] eta: 0:00:30 lr: 0.000911 min_lr: 0.000911 loss: 1.9807 (1.8645) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [313] [280/312] eta: 0:00:23 lr: 0.000911 min_lr: 0.000911 loss: 1.9903 (1.8645) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0009 max mem: 64948 Epoch: [313] [290/312] eta: 0:00:16 lr: 0.000910 min_lr: 0.000910 loss: 1.9863 (1.8706) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [313] [300/312] eta: 0:00:08 lr: 0.000910 min_lr: 0.000910 loss: 2.0751 (1.8744) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [313] [310/312] eta: 0:00:01 lr: 0.000910 min_lr: 0.000910 loss: 1.9980 (1.8739) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [313] [311/312] eta: 0:00:00 lr: 0.000909 min_lr: 0.000909 loss: 1.9949 (1.8743) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [313] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.000909 min_lr: 0.000909 loss: 1.9949 (1.8763) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4901 (0.4901) acc1: 85.4167 (85.4167) acc5: 97.3958 (97.3958) time: 4.4568 data: 4.2529 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7198 (0.7187) acc1: 82.5521 (80.6080) acc5: 96.6146 (95.9680) time: 0.6465 data: 0.4726 max mem: 64948 Test: Total time: 0:00:05 (0.6610 s / it) * Acc@1 81.658 Acc@5 95.764 loss 0.701 Accuracy of the model on the 50000 test images: 81.7% Max accuracy: 81.75% Test: [0/9] eta: 0:00:42 loss: 0.4660 (0.4660) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.6775 data: 4.4662 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6403 (0.6556) acc1: 82.5521 (81.2480) acc5: 97.3958 (96.6720) time: 0.6749 data: 0.4964 max mem: 64948 Test: Total time: 0:00:06 (0.6831 s / it) * Acc@1 82.848 Acc@5 96.464 loss 0.633 Accuracy of the model EMA on 50000 test images: 82.8% Epoch: [314] [ 0/312] eta: 0:55:19 lr: 0.000909 min_lr: 0.000909 loss: 1.4644 (1.4644) weight_decay: 0.0500 (0.0500) time: 10.6384 data: 8.3534 max mem: 64948 Epoch: [314] [ 10/312] eta: 0:08:13 lr: 0.000909 min_lr: 0.000909 loss: 1.8847 (1.8318) weight_decay: 0.0500 (0.0500) time: 1.6339 data: 0.7598 max mem: 64948 Epoch: [314] [ 20/312] eta: 0:05:46 lr: 0.000909 min_lr: 0.000909 loss: 1.9205 (1.8629) weight_decay: 0.0500 (0.0500) time: 0.7146 data: 0.0004 max mem: 64948 Epoch: [314] [ 30/312] eta: 0:04:50 lr: 0.000908 min_lr: 0.000908 loss: 2.0762 (1.9243) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0003 max mem: 64948 Epoch: [314] [ 40/312] eta: 0:04:17 lr: 0.000908 min_lr: 0.000908 loss: 1.8610 (1.8834) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [314] [ 50/312] eta: 0:03:55 lr: 0.000907 min_lr: 0.000907 loss: 1.8063 (1.8899) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [314] [ 60/312] eta: 0:03:38 lr: 0.000907 min_lr: 0.000907 loss: 1.9970 (1.8874) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [314] [ 70/312] eta: 0:03:23 lr: 0.000907 min_lr: 0.000907 loss: 1.9676 (1.8902) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [314] [ 80/312] eta: 0:03:11 lr: 0.000906 min_lr: 0.000906 loss: 1.8463 (1.8803) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [314] [ 90/312] eta: 0:02:59 lr: 0.000906 min_lr: 0.000906 loss: 1.8678 (1.8804) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [314] [100/312] eta: 0:02:49 lr: 0.000906 min_lr: 0.000906 loss: 1.9146 (1.8884) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [314] [110/312] eta: 0:02:39 lr: 0.000905 min_lr: 0.000905 loss: 2.0598 (1.8997) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [314] [120/312] eta: 0:02:29 lr: 0.000905 min_lr: 0.000905 loss: 1.9714 (1.9000) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [314] [130/312] eta: 0:02:21 lr: 0.000904 min_lr: 0.000904 loss: 1.8493 (1.8940) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [314] [140/312] eta: 0:02:12 lr: 0.000904 min_lr: 0.000904 loss: 1.7783 (1.8895) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [314] [150/312] eta: 0:02:03 lr: 0.000904 min_lr: 0.000904 loss: 1.8409 (1.8970) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [314] [160/312] eta: 0:01:55 lr: 0.000903 min_lr: 0.000903 loss: 1.8368 (1.8783) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [314] [170/312] eta: 0:01:47 lr: 0.000903 min_lr: 0.000903 loss: 1.7610 (1.8764) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [314] [180/312] eta: 0:01:39 lr: 0.000902 min_lr: 0.000902 loss: 1.8693 (1.8717) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [314] [190/312] eta: 0:01:31 lr: 0.000902 min_lr: 0.000902 loss: 1.9091 (1.8773) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [314] [200/312] eta: 0:01:23 lr: 0.000902 min_lr: 0.000902 loss: 2.0304 (1.8790) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [314] [210/312] eta: 0:01:15 lr: 0.000901 min_lr: 0.000901 loss: 1.7972 (1.8665) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [314] [220/312] eta: 0:01:08 lr: 0.000901 min_lr: 0.000901 loss: 1.7315 (1.8643) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [314] [230/312] eta: 0:01:00 lr: 0.000900 min_lr: 0.000900 loss: 1.9502 (1.8702) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [314] [240/312] eta: 0:00:53 lr: 0.000900 min_lr: 0.000900 loss: 1.9502 (1.8673) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [314] [250/312] eta: 0:00:45 lr: 0.000900 min_lr: 0.000900 loss: 1.7110 (1.8631) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [314] [260/312] eta: 0:00:38 lr: 0.000899 min_lr: 0.000899 loss: 1.9513 (1.8684) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [314] [270/312] eta: 0:00:30 lr: 0.000899 min_lr: 0.000899 loss: 2.0212 (1.8657) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [314] [280/312] eta: 0:00:23 lr: 0.000898 min_lr: 0.000898 loss: 2.0212 (1.8702) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0010 max mem: 64948 Epoch: [314] [290/312] eta: 0:00:16 lr: 0.000898 min_lr: 0.000898 loss: 2.0477 (1.8748) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [314] [300/312] eta: 0:00:08 lr: 0.000898 min_lr: 0.000898 loss: 1.9370 (1.8759) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [314] [310/312] eta: 0:00:01 lr: 0.000897 min_lr: 0.000897 loss: 1.9865 (1.8750) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [314] [311/312] eta: 0:00:00 lr: 0.000897 min_lr: 0.000897 loss: 1.9874 (1.8758) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [314] Total time: 0:03:47 (0.7307 s / it) Averaged stats: lr: 0.000897 min_lr: 0.000897 loss: 1.9874 (1.8799) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4679 (0.4679) acc1: 88.2812 (88.2812) acc5: 96.8750 (96.8750) time: 4.5424 data: 4.3227 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7103 (0.7225) acc1: 82.5521 (80.8000) acc5: 96.0938 (95.6800) time: 0.6565 data: 0.4804 max mem: 64948 Test: Total time: 0:00:06 (0.6821 s / it) * Acc@1 81.422 Acc@5 95.716 loss 0.713 Accuracy of the model on the 50000 test images: 81.4% Max accuracy: 81.75% Test: [0/9] eta: 0:00:45 loss: 0.4653 (0.4653) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 5.0169 data: 4.8047 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6401 (0.6554) acc1: 82.5521 (81.2160) acc5: 97.3958 (96.6720) time: 0.7087 data: 0.5340 max mem: 64948 Test: Total time: 0:00:06 (0.7213 s / it) * Acc@1 82.870 Acc@5 96.468 loss 0.632 Accuracy of the model EMA on 50000 test images: 82.9% Max EMA accuracy: 82.87% Epoch: [315] [ 0/312] eta: 0:49:08 lr: 0.000897 min_lr: 0.000897 loss: 2.2392 (2.2392) weight_decay: 0.0500 (0.0500) time: 9.4515 data: 8.6424 max mem: 64948 Epoch: [315] [ 10/312] eta: 0:07:40 lr: 0.000897 min_lr: 0.000897 loss: 2.1939 (2.0122) weight_decay: 0.0500 (0.0500) time: 1.5246 data: 0.7861 max mem: 64948 Epoch: [315] [ 20/312] eta: 0:05:30 lr: 0.000896 min_lr: 0.000896 loss: 1.9832 (1.9649) weight_decay: 0.0500 (0.0500) time: 0.7166 data: 0.0004 max mem: 64948 Epoch: [315] [ 30/312] eta: 0:04:39 lr: 0.000896 min_lr: 0.000896 loss: 1.9619 (1.9419) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [315] [ 40/312] eta: 0:04:10 lr: 0.000896 min_lr: 0.000896 loss: 2.0419 (1.9340) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [315] [ 50/312] eta: 0:03:49 lr: 0.000895 min_lr: 0.000895 loss: 2.0157 (1.9297) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [315] [ 60/312] eta: 0:03:33 lr: 0.000895 min_lr: 0.000895 loss: 1.9388 (1.9237) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [315] [ 70/312] eta: 0:03:19 lr: 0.000895 min_lr: 0.000895 loss: 1.9163 (1.9249) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [315] [ 80/312] eta: 0:03:08 lr: 0.000894 min_lr: 0.000894 loss: 1.9037 (1.9178) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [315] [ 90/312] eta: 0:02:57 lr: 0.000894 min_lr: 0.000894 loss: 1.9967 (1.9252) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [315] [100/312] eta: 0:02:46 lr: 0.000893 min_lr: 0.000893 loss: 2.0816 (1.9376) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [315] [110/312] eta: 0:02:37 lr: 0.000893 min_lr: 0.000893 loss: 1.9990 (1.9289) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [315] [120/312] eta: 0:02:28 lr: 0.000893 min_lr: 0.000893 loss: 1.8770 (1.9199) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [315] [130/312] eta: 0:02:19 lr: 0.000892 min_lr: 0.000892 loss: 1.9561 (1.9281) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [315] [140/312] eta: 0:02:10 lr: 0.000892 min_lr: 0.000892 loss: 1.7577 (1.9102) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [315] [150/312] eta: 0:02:02 lr: 0.000891 min_lr: 0.000891 loss: 1.9203 (1.9106) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [315] [160/312] eta: 0:01:54 lr: 0.000891 min_lr: 0.000891 loss: 1.9992 (1.9222) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [315] [170/312] eta: 0:01:46 lr: 0.000891 min_lr: 0.000891 loss: 2.0205 (1.9203) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [315] [180/312] eta: 0:01:38 lr: 0.000890 min_lr: 0.000890 loss: 1.9780 (1.9196) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [315] [190/312] eta: 0:01:30 lr: 0.000890 min_lr: 0.000890 loss: 1.9780 (1.9194) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [315] [200/312] eta: 0:01:23 lr: 0.000889 min_lr: 0.000889 loss: 1.9122 (1.9185) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [315] [210/312] eta: 0:01:15 lr: 0.000889 min_lr: 0.000889 loss: 1.9721 (1.9229) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [315] [220/312] eta: 0:01:07 lr: 0.000889 min_lr: 0.000889 loss: 2.0016 (1.9236) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [315] [230/312] eta: 0:01:00 lr: 0.000888 min_lr: 0.000888 loss: 1.9554 (1.9178) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [315] [240/312] eta: 0:00:52 lr: 0.000888 min_lr: 0.000888 loss: 1.9900 (1.9214) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [315] [250/312] eta: 0:00:45 lr: 0.000887 min_lr: 0.000887 loss: 2.0176 (1.9221) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [315] [260/312] eta: 0:00:38 lr: 0.000887 min_lr: 0.000887 loss: 1.9846 (1.9231) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [315] [270/312] eta: 0:00:30 lr: 0.000887 min_lr: 0.000887 loss: 1.9846 (1.9242) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [315] [280/312] eta: 0:00:23 lr: 0.000886 min_lr: 0.000886 loss: 2.0179 (1.9258) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [315] [290/312] eta: 0:00:15 lr: 0.000886 min_lr: 0.000886 loss: 2.0529 (1.9301) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [315] [300/312] eta: 0:00:08 lr: 0.000886 min_lr: 0.000886 loss: 1.8948 (1.9243) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [315] [310/312] eta: 0:00:01 lr: 0.000885 min_lr: 0.000885 loss: 1.7345 (1.9193) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [315] [311/312] eta: 0:00:00 lr: 0.000885 min_lr: 0.000885 loss: 1.7345 (1.9205) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [315] Total time: 0:03:47 (0.7279 s / it) Averaged stats: lr: 0.000885 min_lr: 0.000885 loss: 1.7345 (1.8809) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4682 (0.4682) acc1: 87.2396 (87.2396) acc5: 98.1771 (98.1771) time: 4.5087 data: 4.2904 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7239 (0.6972) acc1: 82.8125 (81.2160) acc5: 95.8333 (95.7760) time: 0.6522 data: 0.4768 max mem: 64948 Test: Total time: 0:00:06 (0.6744 s / it) * Acc@1 81.730 Acc@5 95.764 loss 0.693 Accuracy of the model on the 50000 test images: 81.7% Max accuracy: 81.75% Test: [0/9] eta: 0:00:43 loss: 0.4644 (0.4644) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.8500 data: 4.6321 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6400 (0.6551) acc1: 82.5521 (81.2480) acc5: 97.3958 (96.6720) time: 0.6910 data: 0.5148 max mem: 64948 Test: Total time: 0:00:06 (0.6986 s / it) * Acc@1 82.906 Acc@5 96.478 loss 0.632 Accuracy of the model EMA on 50000 test images: 82.9% Max EMA accuracy: 82.91% Epoch: [316] [ 0/312] eta: 0:49:24 lr: 0.000885 min_lr: 0.000885 loss: 1.6611 (1.6611) weight_decay: 0.0500 (0.0500) time: 9.5027 data: 7.3847 max mem: 64948 Epoch: [316] [ 10/312] eta: 0:07:46 lr: 0.000885 min_lr: 0.000885 loss: 2.0508 (1.8766) weight_decay: 0.0500 (0.0500) time: 1.5439 data: 0.6734 max mem: 64948 Epoch: [316] [ 20/312] eta: 0:05:32 lr: 0.000884 min_lr: 0.000884 loss: 2.0508 (1.8937) weight_decay: 0.0500 (0.0500) time: 0.7205 data: 0.0013 max mem: 64948 Epoch: [316] [ 30/312] eta: 0:04:40 lr: 0.000884 min_lr: 0.000884 loss: 1.9148 (1.9214) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [316] [ 40/312] eta: 0:04:10 lr: 0.000884 min_lr: 0.000884 loss: 1.9148 (1.9468) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [316] [ 50/312] eta: 0:03:50 lr: 0.000883 min_lr: 0.000883 loss: 2.0062 (1.9615) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [316] [ 60/312] eta: 0:03:33 lr: 0.000883 min_lr: 0.000883 loss: 1.9606 (1.9387) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [316] [ 70/312] eta: 0:03:19 lr: 0.000882 min_lr: 0.000882 loss: 1.9054 (1.9308) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [316] [ 80/312] eta: 0:03:08 lr: 0.000882 min_lr: 0.000882 loss: 1.8054 (1.9100) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [316] [ 90/312] eta: 0:02:57 lr: 0.000882 min_lr: 0.000882 loss: 1.8488 (1.9085) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [316] [100/312] eta: 0:02:47 lr: 0.000881 min_lr: 0.000881 loss: 2.0799 (1.9150) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [316] [110/312] eta: 0:02:37 lr: 0.000881 min_lr: 0.000881 loss: 1.9831 (1.9205) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [316] [120/312] eta: 0:02:28 lr: 0.000880 min_lr: 0.000880 loss: 1.9113 (1.9091) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [316] [130/312] eta: 0:02:19 lr: 0.000880 min_lr: 0.000880 loss: 1.8481 (1.9028) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [316] [140/312] eta: 0:02:10 lr: 0.000880 min_lr: 0.000880 loss: 1.8481 (1.8993) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [316] [150/312] eta: 0:02:02 lr: 0.000879 min_lr: 0.000879 loss: 1.9539 (1.9034) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [316] [160/312] eta: 0:01:54 lr: 0.000879 min_lr: 0.000879 loss: 1.9540 (1.9057) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [316] [170/312] eta: 0:01:46 lr: 0.000878 min_lr: 0.000878 loss: 1.9526 (1.9015) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [316] [180/312] eta: 0:01:38 lr: 0.000878 min_lr: 0.000878 loss: 1.9396 (1.8988) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [316] [190/312] eta: 0:01:30 lr: 0.000878 min_lr: 0.000878 loss: 2.0463 (1.9007) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [316] [200/312] eta: 0:01:23 lr: 0.000877 min_lr: 0.000877 loss: 1.8579 (1.8946) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [316] [210/312] eta: 0:01:15 lr: 0.000877 min_lr: 0.000877 loss: 1.8579 (1.8961) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [316] [220/312] eta: 0:01:07 lr: 0.000877 min_lr: 0.000877 loss: 1.9491 (1.8961) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [316] [230/312] eta: 0:01:00 lr: 0.000876 min_lr: 0.000876 loss: 1.9491 (1.8926) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [316] [240/312] eta: 0:00:52 lr: 0.000876 min_lr: 0.000876 loss: 1.7993 (1.8866) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [316] [250/312] eta: 0:00:45 lr: 0.000875 min_lr: 0.000875 loss: 1.8335 (1.8890) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [316] [260/312] eta: 0:00:37 lr: 0.000875 min_lr: 0.000875 loss: 1.9683 (1.8899) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [316] [270/312] eta: 0:00:30 lr: 0.000875 min_lr: 0.000875 loss: 1.9205 (1.8922) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [316] [280/312] eta: 0:00:23 lr: 0.000874 min_lr: 0.000874 loss: 1.9205 (1.8870) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0009 max mem: 64948 Epoch: [316] [290/312] eta: 0:00:15 lr: 0.000874 min_lr: 0.000874 loss: 1.9583 (1.8861) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [316] [300/312] eta: 0:00:08 lr: 0.000873 min_lr: 0.000873 loss: 1.9519 (1.8852) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [316] [310/312] eta: 0:00:01 lr: 0.000873 min_lr: 0.000873 loss: 1.7973 (1.8820) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [316] [311/312] eta: 0:00:00 lr: 0.000873 min_lr: 0.000873 loss: 1.7973 (1.8825) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [316] Total time: 0:03:46 (0.7274 s / it) Averaged stats: lr: 0.000873 min_lr: 0.000873 loss: 1.7973 (1.8811) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5042 (0.5042) acc1: 87.2396 (87.2396) acc5: 97.1354 (97.1354) time: 4.5042 data: 4.2943 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7098 (0.7173) acc1: 82.5521 (80.9280) acc5: 96.3542 (95.7760) time: 0.6518 data: 0.4772 max mem: 64948 Test: Total time: 0:00:06 (0.6720 s / it) * Acc@1 81.588 Acc@5 95.830 loss 0.699 Accuracy of the model on the 50000 test images: 81.6% Max accuracy: 81.75% Test: [0/9] eta: 0:00:43 loss: 0.4637 (0.4637) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.8297 data: 4.6117 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6399 (0.6548) acc1: 82.5521 (81.2480) acc5: 97.3958 (96.6400) time: 0.6881 data: 0.5125 max mem: 64948 Test: Total time: 0:00:06 (0.7030 s / it) * Acc@1 82.924 Acc@5 96.478 loss 0.632 Accuracy of the model EMA on 50000 test images: 82.9% Max EMA accuracy: 82.92% Epoch: [317] [ 0/312] eta: 0:53:32 lr: 0.000873 min_lr: 0.000873 loss: 1.3836 (1.3836) weight_decay: 0.0500 (0.0500) time: 10.2956 data: 9.4857 max mem: 64948 Epoch: [317] [ 10/312] eta: 0:08:00 lr: 0.000873 min_lr: 0.000873 loss: 1.8910 (1.8392) weight_decay: 0.0500 (0.0500) time: 1.5920 data: 0.8626 max mem: 64948 Epoch: [317] [ 20/312] eta: 0:05:40 lr: 0.000872 min_lr: 0.000872 loss: 1.8910 (1.8356) weight_decay: 0.0500 (0.0500) time: 0.7104 data: 0.0003 max mem: 64948 Epoch: [317] [ 30/312] eta: 0:04:46 lr: 0.000872 min_lr: 0.000872 loss: 1.8142 (1.8271) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0003 max mem: 64948 Epoch: [317] [ 40/312] eta: 0:04:14 lr: 0.000871 min_lr: 0.000871 loss: 1.8534 (1.8557) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [317] [ 50/312] eta: 0:03:53 lr: 0.000871 min_lr: 0.000871 loss: 1.9026 (1.8735) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [317] [ 60/312] eta: 0:03:36 lr: 0.000871 min_lr: 0.000871 loss: 1.8879 (1.8662) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [317] [ 70/312] eta: 0:03:22 lr: 0.000870 min_lr: 0.000870 loss: 1.9569 (1.8843) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [317] [ 80/312] eta: 0:03:09 lr: 0.000870 min_lr: 0.000870 loss: 1.9430 (1.8789) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [317] [ 90/312] eta: 0:02:58 lr: 0.000870 min_lr: 0.000870 loss: 1.9247 (1.8852) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [317] [100/312] eta: 0:02:48 lr: 0.000869 min_lr: 0.000869 loss: 1.9484 (1.8850) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [317] [110/312] eta: 0:02:38 lr: 0.000869 min_lr: 0.000869 loss: 1.8486 (1.8811) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [317] [120/312] eta: 0:02:29 lr: 0.000868 min_lr: 0.000868 loss: 1.8735 (1.8811) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [317] [130/312] eta: 0:02:20 lr: 0.000868 min_lr: 0.000868 loss: 1.9909 (1.8803) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [317] [140/312] eta: 0:02:11 lr: 0.000868 min_lr: 0.000868 loss: 1.9513 (1.8807) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [317] [150/312] eta: 0:02:03 lr: 0.000867 min_lr: 0.000867 loss: 2.0467 (1.8790) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [317] [160/312] eta: 0:01:55 lr: 0.000867 min_lr: 0.000867 loss: 1.7186 (1.8761) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [317] [170/312] eta: 0:01:46 lr: 0.000866 min_lr: 0.000866 loss: 1.7898 (1.8749) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [317] [180/312] eta: 0:01:38 lr: 0.000866 min_lr: 0.000866 loss: 1.7200 (1.8630) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [317] [190/312] eta: 0:01:31 lr: 0.000866 min_lr: 0.000866 loss: 1.8326 (1.8681) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [317] [200/312] eta: 0:01:23 lr: 0.000865 min_lr: 0.000865 loss: 1.9497 (1.8649) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [317] [210/312] eta: 0:01:15 lr: 0.000865 min_lr: 0.000865 loss: 1.8631 (1.8623) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [317] [220/312] eta: 0:01:08 lr: 0.000865 min_lr: 0.000865 loss: 1.9207 (1.8687) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [317] [230/312] eta: 0:01:00 lr: 0.000864 min_lr: 0.000864 loss: 2.0503 (1.8709) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [317] [240/312] eta: 0:00:52 lr: 0.000864 min_lr: 0.000864 loss: 1.9438 (1.8718) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [317] [250/312] eta: 0:00:45 lr: 0.000863 min_lr: 0.000863 loss: 1.9438 (1.8683) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [317] [260/312] eta: 0:00:38 lr: 0.000863 min_lr: 0.000863 loss: 1.8822 (1.8697) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [317] [270/312] eta: 0:00:30 lr: 0.000863 min_lr: 0.000863 loss: 1.8792 (1.8696) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [317] [280/312] eta: 0:00:23 lr: 0.000862 min_lr: 0.000862 loss: 1.8792 (1.8719) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [317] [290/312] eta: 0:00:16 lr: 0.000862 min_lr: 0.000862 loss: 2.0087 (1.8778) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [317] [300/312] eta: 0:00:08 lr: 0.000861 min_lr: 0.000861 loss: 1.9198 (1.8720) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [317] [310/312] eta: 0:00:01 lr: 0.000861 min_lr: 0.000861 loss: 1.7862 (1.8729) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [317] [311/312] eta: 0:00:00 lr: 0.000861 min_lr: 0.000861 loss: 1.7906 (1.8741) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [317] Total time: 0:03:47 (0.7293 s / it) Averaged stats: lr: 0.000861 min_lr: 0.000861 loss: 1.7906 (1.8739) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.5508 (0.5508) acc1: 85.1562 (85.1562) acc5: 97.6562 (97.6562) time: 4.8846 data: 4.6653 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6588 (0.7129) acc1: 82.8125 (80.8640) acc5: 97.3958 (96.3200) time: 0.6941 data: 0.5185 max mem: 64948 Test: Total time: 0:00:06 (0.7190 s / it) * Acc@1 81.696 Acc@5 95.758 loss 0.694 Accuracy of the model on the 50000 test images: 81.7% Max accuracy: 81.75% Test: [0/9] eta: 0:00:44 loss: 0.4632 (0.4632) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.9985 data: 4.7911 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6400 (0.6547) acc1: 82.5521 (81.2480) acc5: 97.3958 (96.6720) time: 0.7067 data: 0.5325 max mem: 64948 Test: Total time: 0:00:06 (0.7167 s / it) * Acc@1 82.902 Acc@5 96.478 loss 0.632 Accuracy of the model EMA on 50000 test images: 82.9% Epoch: [318] [ 0/312] eta: 0:56:37 lr: 0.000861 min_lr: 0.000861 loss: 2.0989 (2.0989) weight_decay: 0.0500 (0.0500) time: 10.8891 data: 7.4796 max mem: 64948 Epoch: [318] [ 10/312] eta: 0:08:17 lr: 0.000861 min_lr: 0.000861 loss: 1.8248 (1.8211) weight_decay: 0.0500 (0.0500) time: 1.6479 data: 0.6803 max mem: 64948 Epoch: [318] [ 20/312] eta: 0:05:49 lr: 0.000860 min_lr: 0.000860 loss: 1.8248 (1.8348) weight_decay: 0.0500 (0.0500) time: 0.7118 data: 0.0004 max mem: 64948 Epoch: [318] [ 30/312] eta: 0:04:51 lr: 0.000860 min_lr: 0.000860 loss: 2.0113 (1.9023) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [318] [ 40/312] eta: 0:04:18 lr: 0.000859 min_lr: 0.000859 loss: 2.0619 (1.9461) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [318] [ 50/312] eta: 0:03:56 lr: 0.000859 min_lr: 0.000859 loss: 2.0589 (1.9577) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [318] [ 60/312] eta: 0:03:38 lr: 0.000859 min_lr: 0.000859 loss: 2.0304 (1.9266) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [318] [ 70/312] eta: 0:03:24 lr: 0.000858 min_lr: 0.000858 loss: 1.9299 (1.9326) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [318] [ 80/312] eta: 0:03:11 lr: 0.000858 min_lr: 0.000858 loss: 1.8942 (1.9278) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [318] [ 90/312] eta: 0:03:00 lr: 0.000857 min_lr: 0.000857 loss: 1.8801 (1.9236) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [318] [100/312] eta: 0:02:49 lr: 0.000857 min_lr: 0.000857 loss: 1.9394 (1.9155) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [318] [110/312] eta: 0:02:39 lr: 0.000857 min_lr: 0.000857 loss: 1.9801 (1.9149) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0003 max mem: 64948 Epoch: [318] [120/312] eta: 0:02:30 lr: 0.000856 min_lr: 0.000856 loss: 1.7490 (1.8973) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [318] [130/312] eta: 0:02:21 lr: 0.000856 min_lr: 0.000856 loss: 1.7490 (1.8952) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [318] [140/312] eta: 0:02:12 lr: 0.000856 min_lr: 0.000856 loss: 1.8538 (1.8947) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [318] [150/312] eta: 0:02:04 lr: 0.000855 min_lr: 0.000855 loss: 1.8538 (1.8949) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [318] [160/312] eta: 0:01:55 lr: 0.000855 min_lr: 0.000855 loss: 2.0368 (1.9051) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [318] [170/312] eta: 0:01:47 lr: 0.000854 min_lr: 0.000854 loss: 2.0384 (1.9084) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [318] [180/312] eta: 0:01:39 lr: 0.000854 min_lr: 0.000854 loss: 1.8688 (1.9003) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [318] [190/312] eta: 0:01:31 lr: 0.000854 min_lr: 0.000854 loss: 1.8683 (1.9008) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [318] [200/312] eta: 0:01:23 lr: 0.000853 min_lr: 0.000853 loss: 2.0152 (1.9018) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [318] [210/312] eta: 0:01:16 lr: 0.000853 min_lr: 0.000853 loss: 2.0593 (1.9106) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [318] [220/312] eta: 0:01:08 lr: 0.000853 min_lr: 0.000853 loss: 1.9259 (1.9102) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [318] [230/312] eta: 0:01:00 lr: 0.000852 min_lr: 0.000852 loss: 1.9507 (1.9154) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [318] [240/312] eta: 0:00:53 lr: 0.000852 min_lr: 0.000852 loss: 2.0977 (1.9253) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [318] [250/312] eta: 0:00:45 lr: 0.000851 min_lr: 0.000851 loss: 2.0814 (1.9282) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0003 max mem: 64948 Epoch: [318] [260/312] eta: 0:00:38 lr: 0.000851 min_lr: 0.000851 loss: 1.9752 (1.9238) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [318] [270/312] eta: 0:00:30 lr: 0.000851 min_lr: 0.000851 loss: 1.7730 (1.9153) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [318] [280/312] eta: 0:00:23 lr: 0.000850 min_lr: 0.000850 loss: 1.9239 (1.9177) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [318] [290/312] eta: 0:00:16 lr: 0.000850 min_lr: 0.000850 loss: 1.9911 (1.9172) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [318] [300/312] eta: 0:00:08 lr: 0.000849 min_lr: 0.000849 loss: 2.0337 (1.9205) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0001 max mem: 64948 Epoch: [318] [310/312] eta: 0:00:01 lr: 0.000849 min_lr: 0.000849 loss: 1.9997 (1.9173) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0001 max mem: 64948 Epoch: [318] [311/312] eta: 0:00:00 lr: 0.000849 min_lr: 0.000849 loss: 1.8572 (1.9163) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0001 max mem: 64948 Epoch: [318] Total time: 0:03:48 (0.7321 s / it) Averaged stats: lr: 0.000849 min_lr: 0.000849 loss: 1.8572 (1.8802) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4825 (0.4825) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.8379 data: 4.6177 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7190 (0.7016) acc1: 81.5104 (81.0880) acc5: 96.3542 (95.8080) time: 0.6895 data: 0.5132 max mem: 64948 Test: Total time: 0:00:06 (0.7150 s / it) * Acc@1 81.756 Acc@5 95.770 loss 0.690 Accuracy of the model on the 50000 test images: 81.8% Max accuracy: 81.76% Test: [0/9] eta: 0:00:41 loss: 0.4626 (0.4626) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.5823 data: 4.3622 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6396 (0.6546) acc1: 82.8125 (81.3120) acc5: 97.3958 (96.6080) time: 0.6608 data: 0.4848 max mem: 64948 Test: Total time: 0:00:06 (0.6698 s / it) * Acc@1 82.904 Acc@5 96.474 loss 0.632 Accuracy of the model EMA on 50000 test images: 82.9% Epoch: [319] [ 0/312] eta: 0:54:01 lr: 0.000849 min_lr: 0.000849 loss: 1.3924 (1.3924) weight_decay: 0.0500 (0.0500) time: 10.3897 data: 6.4305 max mem: 64948 Epoch: [319] [ 10/312] eta: 0:08:14 lr: 0.000849 min_lr: 0.000849 loss: 1.9519 (1.8101) weight_decay: 0.0500 (0.0500) time: 1.6386 data: 0.5851 max mem: 64948 Epoch: [319] [ 20/312] eta: 0:05:47 lr: 0.000848 min_lr: 0.000848 loss: 1.9519 (1.8774) weight_decay: 0.0500 (0.0500) time: 0.7295 data: 0.0005 max mem: 64948 Epoch: [319] [ 30/312] eta: 0:04:50 lr: 0.000848 min_lr: 0.000848 loss: 1.9131 (1.8467) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [319] [ 40/312] eta: 0:04:17 lr: 0.000847 min_lr: 0.000847 loss: 1.9131 (1.8572) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [319] [ 50/312] eta: 0:03:55 lr: 0.000847 min_lr: 0.000847 loss: 1.9809 (1.8691) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [319] [ 60/312] eta: 0:03:37 lr: 0.000847 min_lr: 0.000847 loss: 2.0667 (1.8950) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [319] [ 70/312] eta: 0:03:23 lr: 0.000846 min_lr: 0.000846 loss: 1.9010 (1.8587) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [319] [ 80/312] eta: 0:03:10 lr: 0.000846 min_lr: 0.000846 loss: 1.5873 (1.8519) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [319] [ 90/312] eta: 0:02:59 lr: 0.000846 min_lr: 0.000846 loss: 1.9461 (1.8564) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [319] [100/312] eta: 0:02:49 lr: 0.000845 min_lr: 0.000845 loss: 1.9061 (1.8482) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [319] [110/312] eta: 0:02:39 lr: 0.000845 min_lr: 0.000845 loss: 1.9919 (1.8516) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [319] [120/312] eta: 0:02:29 lr: 0.000844 min_lr: 0.000844 loss: 2.0012 (1.8620) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [319] [130/312] eta: 0:02:20 lr: 0.000844 min_lr: 0.000844 loss: 2.0156 (1.8634) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [319] [140/312] eta: 0:02:12 lr: 0.000844 min_lr: 0.000844 loss: 2.0465 (1.8725) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [319] [150/312] eta: 0:02:03 lr: 0.000843 min_lr: 0.000843 loss: 2.0465 (1.8744) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [319] [160/312] eta: 0:01:55 lr: 0.000843 min_lr: 0.000843 loss: 1.9290 (1.8719) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [319] [170/312] eta: 0:01:47 lr: 0.000842 min_lr: 0.000842 loss: 1.9100 (1.8685) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [319] [180/312] eta: 0:01:39 lr: 0.000842 min_lr: 0.000842 loss: 1.8017 (1.8591) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [319] [190/312] eta: 0:01:31 lr: 0.000842 min_lr: 0.000842 loss: 1.8447 (1.8540) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [319] [200/312] eta: 0:01:23 lr: 0.000841 min_lr: 0.000841 loss: 1.9176 (1.8569) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [319] [210/312] eta: 0:01:15 lr: 0.000841 min_lr: 0.000841 loss: 1.9472 (1.8589) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [319] [220/312] eta: 0:01:08 lr: 0.000841 min_lr: 0.000841 loss: 1.8623 (1.8560) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [319] [230/312] eta: 0:01:00 lr: 0.000840 min_lr: 0.000840 loss: 1.7128 (1.8533) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [319] [240/312] eta: 0:00:53 lr: 0.000840 min_lr: 0.000840 loss: 1.6825 (1.8469) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [319] [250/312] eta: 0:00:45 lr: 0.000839 min_lr: 0.000839 loss: 1.7596 (1.8458) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [319] [260/312] eta: 0:00:38 lr: 0.000839 min_lr: 0.000839 loss: 1.9757 (1.8497) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [319] [270/312] eta: 0:00:30 lr: 0.000839 min_lr: 0.000839 loss: 1.9757 (1.8457) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [319] [280/312] eta: 0:00:23 lr: 0.000838 min_lr: 0.000838 loss: 1.8016 (1.8477) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0009 max mem: 64948 Epoch: [319] [290/312] eta: 0:00:16 lr: 0.000838 min_lr: 0.000838 loss: 1.9166 (1.8508) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [319] [300/312] eta: 0:00:08 lr: 0.000838 min_lr: 0.000838 loss: 1.9962 (1.8549) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [319] [310/312] eta: 0:00:01 lr: 0.000837 min_lr: 0.000837 loss: 2.0186 (1.8592) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [319] [311/312] eta: 0:00:00 lr: 0.000837 min_lr: 0.000837 loss: 2.0078 (1.8589) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [319] Total time: 0:03:47 (0.7306 s / it) Averaged stats: lr: 0.000837 min_lr: 0.000837 loss: 2.0078 (1.8781) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4860 (0.4860) acc1: 87.5000 (87.5000) acc5: 98.4375 (98.4375) time: 4.3729 data: 4.1533 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6816 (0.7035) acc1: 82.8125 (80.9600) acc5: 96.6146 (96.4160) time: 0.6462 data: 0.4706 max mem: 64948 Test: Total time: 0:00:05 (0.6590 s / it) * Acc@1 81.828 Acc@5 96.018 loss 0.688 Accuracy of the model on the 50000 test images: 81.8% Max accuracy: 81.83% Test: [0/9] eta: 0:00:40 loss: 0.4624 (0.4624) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.4754 data: 4.2683 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6392 (0.6546) acc1: 82.8125 (81.3760) acc5: 97.3958 (96.6400) time: 0.6486 data: 0.4744 max mem: 64948 Test: Total time: 0:00:05 (0.6559 s / it) * Acc@1 82.918 Acc@5 96.480 loss 0.631 Accuracy of the model EMA on 50000 test images: 82.9% Epoch: [320] [ 0/312] eta: 0:55:29 lr: 0.000837 min_lr: 0.000837 loss: 1.9944 (1.9944) weight_decay: 0.0500 (0.0500) time: 10.6711 data: 6.2964 max mem: 64948 Epoch: [320] [ 10/312] eta: 0:08:11 lr: 0.000837 min_lr: 0.000837 loss: 1.9097 (1.7367) weight_decay: 0.0500 (0.0500) time: 1.6279 data: 0.5728 max mem: 64948 Epoch: [320] [ 20/312] eta: 0:05:45 lr: 0.000836 min_lr: 0.000836 loss: 1.9097 (1.7987) weight_decay: 0.0500 (0.0500) time: 0.7084 data: 0.0004 max mem: 64948 Epoch: [320] [ 30/312] eta: 0:04:49 lr: 0.000836 min_lr: 0.000836 loss: 1.9249 (1.8075) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [320] [ 40/312] eta: 0:04:17 lr: 0.000836 min_lr: 0.000836 loss: 1.9673 (1.8513) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [320] [ 50/312] eta: 0:03:54 lr: 0.000835 min_lr: 0.000835 loss: 2.0254 (1.8795) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [320] [ 60/312] eta: 0:03:37 lr: 0.000835 min_lr: 0.000835 loss: 1.9454 (1.8609) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [320] [ 70/312] eta: 0:03:23 lr: 0.000834 min_lr: 0.000834 loss: 2.0083 (1.8872) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [320] [ 80/312] eta: 0:03:10 lr: 0.000834 min_lr: 0.000834 loss: 2.0083 (1.8778) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [320] [ 90/312] eta: 0:02:59 lr: 0.000834 min_lr: 0.000834 loss: 1.7527 (1.8774) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [320] [100/312] eta: 0:02:49 lr: 0.000833 min_lr: 0.000833 loss: 1.9046 (1.8733) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [320] [110/312] eta: 0:02:39 lr: 0.000833 min_lr: 0.000833 loss: 1.9046 (1.8712) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [320] [120/312] eta: 0:02:29 lr: 0.000833 min_lr: 0.000833 loss: 1.9252 (1.8675) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [320] [130/312] eta: 0:02:20 lr: 0.000832 min_lr: 0.000832 loss: 1.9275 (1.8746) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [320] [140/312] eta: 0:02:12 lr: 0.000832 min_lr: 0.000832 loss: 2.0763 (1.8879) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [320] [150/312] eta: 0:02:03 lr: 0.000831 min_lr: 0.000831 loss: 2.0215 (1.8878) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [320] [160/312] eta: 0:01:55 lr: 0.000831 min_lr: 0.000831 loss: 1.9136 (1.8936) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [320] [170/312] eta: 0:01:47 lr: 0.000831 min_lr: 0.000831 loss: 1.9297 (1.8862) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [320] [180/312] eta: 0:01:39 lr: 0.000830 min_lr: 0.000830 loss: 1.9058 (1.8850) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [320] [190/312] eta: 0:01:31 lr: 0.000830 min_lr: 0.000830 loss: 1.8252 (1.8800) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [320] [200/312] eta: 0:01:23 lr: 0.000829 min_lr: 0.000829 loss: 1.8125 (1.8756) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [320] [210/312] eta: 0:01:15 lr: 0.000829 min_lr: 0.000829 loss: 1.9311 (1.8743) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [320] [220/312] eta: 0:01:08 lr: 0.000829 min_lr: 0.000829 loss: 1.6878 (1.8638) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [320] [230/312] eta: 0:01:00 lr: 0.000828 min_lr: 0.000828 loss: 1.7110 (1.8621) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [320] [240/312] eta: 0:00:53 lr: 0.000828 min_lr: 0.000828 loss: 1.9262 (1.8630) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [320] [250/312] eta: 0:00:45 lr: 0.000828 min_lr: 0.000828 loss: 1.8865 (1.8603) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [320] [260/312] eta: 0:00:38 lr: 0.000827 min_lr: 0.000827 loss: 1.6285 (1.8577) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [320] [270/312] eta: 0:00:30 lr: 0.000827 min_lr: 0.000827 loss: 1.6424 (1.8576) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [320] [280/312] eta: 0:00:23 lr: 0.000826 min_lr: 0.000826 loss: 1.9783 (1.8608) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [320] [290/312] eta: 0:00:16 lr: 0.000826 min_lr: 0.000826 loss: 1.9269 (1.8566) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [320] [300/312] eta: 0:00:08 lr: 0.000826 min_lr: 0.000826 loss: 1.7527 (1.8569) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [320] [310/312] eta: 0:00:01 lr: 0.000825 min_lr: 0.000825 loss: 1.7555 (1.8516) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [320] [311/312] eta: 0:00:00 lr: 0.000825 min_lr: 0.000825 loss: 1.7577 (1.8524) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [320] Total time: 0:03:48 (0.7308 s / it) Averaged stats: lr: 0.000825 min_lr: 0.000825 loss: 1.7577 (1.8653) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5158 (0.5158) acc1: 86.7188 (86.7188) acc5: 96.6146 (96.6146) time: 4.6744 data: 4.4549 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7143 (0.7313) acc1: 82.5521 (80.4480) acc5: 96.3542 (95.8400) time: 0.6707 data: 0.4951 max mem: 64948 Test: Total time: 0:00:06 (0.6948 s / it) * Acc@1 81.380 Acc@5 95.788 loss 0.710 Accuracy of the model on the 50000 test images: 81.4% Max accuracy: 81.83% Test: [0/9] eta: 0:00:45 loss: 0.4616 (0.4616) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.0094 data: 4.8024 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6385 (0.6542) acc1: 82.8125 (81.4400) acc5: 97.3958 (96.6720) time: 0.7079 data: 0.5337 max mem: 64948 Test: Total time: 0:00:06 (0.7190 s / it) * Acc@1 82.948 Acc@5 96.474 loss 0.631 Accuracy of the model EMA on 50000 test images: 82.9% Max EMA accuracy: 82.95% Epoch: [321] [ 0/312] eta: 0:49:22 lr: 0.000825 min_lr: 0.000825 loss: 1.9797 (1.9797) weight_decay: 0.0500 (0.0500) time: 9.4939 data: 8.6919 max mem: 64948 Epoch: [321] [ 10/312] eta: 0:07:40 lr: 0.000825 min_lr: 0.000825 loss: 1.9498 (1.9669) weight_decay: 0.0500 (0.0500) time: 1.5261 data: 0.7905 max mem: 64948 Epoch: [321] [ 20/312] eta: 0:05:29 lr: 0.000824 min_lr: 0.000824 loss: 1.9789 (1.9928) weight_decay: 0.0500 (0.0500) time: 0.7117 data: 0.0004 max mem: 64948 Epoch: [321] [ 30/312] eta: 0:04:39 lr: 0.000824 min_lr: 0.000824 loss: 1.9771 (1.9600) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [321] [ 40/312] eta: 0:04:10 lr: 0.000824 min_lr: 0.000824 loss: 1.8807 (1.9190) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [321] [ 50/312] eta: 0:03:49 lr: 0.000823 min_lr: 0.000823 loss: 2.0449 (1.9488) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [321] [ 60/312] eta: 0:03:33 lr: 0.000823 min_lr: 0.000823 loss: 2.0569 (1.9277) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [321] [ 70/312] eta: 0:03:19 lr: 0.000823 min_lr: 0.000823 loss: 1.9275 (1.9319) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [321] [ 80/312] eta: 0:03:07 lr: 0.000822 min_lr: 0.000822 loss: 1.8525 (1.9124) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [321] [ 90/312] eta: 0:02:56 lr: 0.000822 min_lr: 0.000822 loss: 1.7934 (1.9075) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [321] [100/312] eta: 0:02:46 lr: 0.000821 min_lr: 0.000821 loss: 2.0660 (1.9149) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [321] [110/312] eta: 0:02:37 lr: 0.000821 min_lr: 0.000821 loss: 1.9768 (1.9075) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [321] [120/312] eta: 0:02:28 lr: 0.000821 min_lr: 0.000821 loss: 1.9567 (1.9062) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [321] [130/312] eta: 0:02:19 lr: 0.000820 min_lr: 0.000820 loss: 2.0483 (1.9167) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [321] [140/312] eta: 0:02:10 lr: 0.000820 min_lr: 0.000820 loss: 1.8884 (1.8980) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [321] [150/312] eta: 0:02:02 lr: 0.000820 min_lr: 0.000820 loss: 1.6764 (1.8925) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [321] [160/312] eta: 0:01:54 lr: 0.000819 min_lr: 0.000819 loss: 1.9674 (1.8986) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [321] [170/312] eta: 0:01:46 lr: 0.000819 min_lr: 0.000819 loss: 1.9903 (1.9096) weight_decay: 0.0500 (0.0500) time: 0.7015 data: 0.0004 max mem: 64948 Epoch: [321] [180/312] eta: 0:01:38 lr: 0.000818 min_lr: 0.000818 loss: 1.9601 (1.9060) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [321] [190/312] eta: 0:01:30 lr: 0.000818 min_lr: 0.000818 loss: 1.8520 (1.9002) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [321] [200/312] eta: 0:01:23 lr: 0.000818 min_lr: 0.000818 loss: 1.9647 (1.8982) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [321] [210/312] eta: 0:01:15 lr: 0.000817 min_lr: 0.000817 loss: 1.9978 (1.9050) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [321] [220/312] eta: 0:01:07 lr: 0.000817 min_lr: 0.000817 loss: 2.0348 (1.9038) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [321] [230/312] eta: 0:01:00 lr: 0.000817 min_lr: 0.000817 loss: 1.9309 (1.8980) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [321] [240/312] eta: 0:00:52 lr: 0.000816 min_lr: 0.000816 loss: 1.8306 (1.8920) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [321] [250/312] eta: 0:00:45 lr: 0.000816 min_lr: 0.000816 loss: 1.8606 (1.8926) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [321] [260/312] eta: 0:00:38 lr: 0.000815 min_lr: 0.000815 loss: 1.9090 (1.8952) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [321] [270/312] eta: 0:00:30 lr: 0.000815 min_lr: 0.000815 loss: 1.7871 (1.8871) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [321] [280/312] eta: 0:00:23 lr: 0.000815 min_lr: 0.000815 loss: 1.7739 (1.8843) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0010 max mem: 64948 Epoch: [321] [290/312] eta: 0:00:15 lr: 0.000814 min_lr: 0.000814 loss: 1.8672 (1.8838) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [321] [300/312] eta: 0:00:08 lr: 0.000814 min_lr: 0.000814 loss: 1.6978 (1.8764) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [321] [310/312] eta: 0:00:01 lr: 0.000814 min_lr: 0.000814 loss: 1.6641 (1.8766) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [321] [311/312] eta: 0:00:00 lr: 0.000813 min_lr: 0.000813 loss: 1.7347 (1.8763) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [321] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.000813 min_lr: 0.000813 loss: 1.7347 (1.8660) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4758 (0.4758) acc1: 87.2396 (87.2396) acc5: 97.3958 (97.3958) time: 4.4276 data: 4.2080 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7507 (0.7278) acc1: 81.2500 (80.6080) acc5: 96.3542 (95.5520) time: 0.6439 data: 0.4677 max mem: 64948 Test: Total time: 0:00:06 (0.6670 s / it) * Acc@1 81.708 Acc@5 95.772 loss 0.698 Accuracy of the model on the 50000 test images: 81.7% Max accuracy: 81.83% Test: [0/9] eta: 0:00:44 loss: 0.4611 (0.4611) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.8985 data: 4.6934 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6384 (0.6541) acc1: 83.0729 (81.3440) acc5: 97.1354 (96.6400) time: 0.6956 data: 0.5216 max mem: 64948 Test: Total time: 0:00:06 (0.7060 s / it) * Acc@1 82.966 Acc@5 96.482 loss 0.631 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 82.97% Epoch: [322] [ 0/312] eta: 0:48:21 lr: 0.000813 min_lr: 0.000813 loss: 1.2538 (1.2538) weight_decay: 0.0500 (0.0500) time: 9.3005 data: 7.1412 max mem: 64948 Epoch: [322] [ 10/312] eta: 0:07:36 lr: 0.000813 min_lr: 0.000813 loss: 1.8404 (1.8177) weight_decay: 0.0500 (0.0500) time: 1.5121 data: 0.6496 max mem: 64948 Epoch: [322] [ 20/312] eta: 0:05:28 lr: 0.000813 min_lr: 0.000813 loss: 1.8404 (1.7543) weight_decay: 0.0500 (0.0500) time: 0.7154 data: 0.0004 max mem: 64948 Epoch: [322] [ 30/312] eta: 0:04:37 lr: 0.000812 min_lr: 0.000812 loss: 1.8818 (1.7927) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [322] [ 40/312] eta: 0:04:09 lr: 0.000812 min_lr: 0.000812 loss: 1.9863 (1.8009) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [322] [ 50/312] eta: 0:03:48 lr: 0.000812 min_lr: 0.000812 loss: 1.7457 (1.7968) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [322] [ 60/312] eta: 0:03:32 lr: 0.000811 min_lr: 0.000811 loss: 1.9287 (1.8436) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [322] [ 70/312] eta: 0:03:18 lr: 0.000811 min_lr: 0.000811 loss: 1.9474 (1.8550) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [322] [ 80/312] eta: 0:03:06 lr: 0.000810 min_lr: 0.000810 loss: 1.9434 (1.8417) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [322] [ 90/312] eta: 0:02:56 lr: 0.000810 min_lr: 0.000810 loss: 1.9455 (1.8460) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [322] [100/312] eta: 0:02:46 lr: 0.000810 min_lr: 0.000810 loss: 1.8256 (1.8398) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [322] [110/312] eta: 0:02:36 lr: 0.000809 min_lr: 0.000809 loss: 1.8066 (1.8418) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [322] [120/312] eta: 0:02:27 lr: 0.000809 min_lr: 0.000809 loss: 2.0118 (1.8466) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [322] [130/312] eta: 0:02:19 lr: 0.000809 min_lr: 0.000809 loss: 1.9656 (1.8506) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [322] [140/312] eta: 0:02:10 lr: 0.000808 min_lr: 0.000808 loss: 1.8671 (1.8452) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [322] [150/312] eta: 0:02:02 lr: 0.000808 min_lr: 0.000808 loss: 1.7756 (1.8445) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [322] [160/312] eta: 0:01:54 lr: 0.000807 min_lr: 0.000807 loss: 1.7804 (1.8383) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [322] [170/312] eta: 0:01:46 lr: 0.000807 min_lr: 0.000807 loss: 1.8637 (1.8456) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [322] [180/312] eta: 0:01:38 lr: 0.000807 min_lr: 0.000807 loss: 1.9447 (1.8505) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [322] [190/312] eta: 0:01:30 lr: 0.000806 min_lr: 0.000806 loss: 1.8811 (1.8467) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [322] [200/312] eta: 0:01:22 lr: 0.000806 min_lr: 0.000806 loss: 1.8811 (1.8464) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [322] [210/312] eta: 0:01:15 lr: 0.000806 min_lr: 0.000806 loss: 2.0380 (1.8507) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [322] [220/312] eta: 0:01:07 lr: 0.000805 min_lr: 0.000805 loss: 1.9068 (1.8459) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [322] [230/312] eta: 0:01:00 lr: 0.000805 min_lr: 0.000805 loss: 1.8729 (1.8438) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [322] [240/312] eta: 0:00:52 lr: 0.000804 min_lr: 0.000804 loss: 1.8729 (1.8435) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [322] [250/312] eta: 0:00:45 lr: 0.000804 min_lr: 0.000804 loss: 1.9554 (1.8483) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [322] [260/312] eta: 0:00:37 lr: 0.000804 min_lr: 0.000804 loss: 1.9819 (1.8478) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [322] [270/312] eta: 0:00:30 lr: 0.000803 min_lr: 0.000803 loss: 1.9377 (1.8522) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [322] [280/312] eta: 0:00:23 lr: 0.000803 min_lr: 0.000803 loss: 1.9377 (1.8521) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [322] [290/312] eta: 0:00:15 lr: 0.000803 min_lr: 0.000803 loss: 1.8358 (1.8495) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [322] [300/312] eta: 0:00:08 lr: 0.000802 min_lr: 0.000802 loss: 1.9784 (1.8545) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0002 max mem: 64948 Epoch: [322] [310/312] eta: 0:00:01 lr: 0.000802 min_lr: 0.000802 loss: 1.9719 (1.8556) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [322] [311/312] eta: 0:00:00 lr: 0.000802 min_lr: 0.000802 loss: 1.9569 (1.8559) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [322] Total time: 0:03:46 (0.7263 s / it) Averaged stats: lr: 0.000802 min_lr: 0.000802 loss: 1.9569 (1.8616) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4867 (0.4867) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 4.6409 data: 4.4281 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7256 (0.7135) acc1: 82.5521 (81.0880) acc5: 96.3542 (95.7120) time: 0.6669 data: 0.4921 max mem: 64948 Test: Total time: 0:00:06 (0.6922 s / it) * Acc@1 81.726 Acc@5 95.884 loss 0.694 Accuracy of the model on the 50000 test images: 81.7% Max accuracy: 81.83% Test: [0/9] eta: 0:00:42 loss: 0.4613 (0.4613) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.6989 data: 4.4847 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6382 (0.6540) acc1: 83.3333 (81.3760) acc5: 97.1354 (96.6400) time: 0.6774 data: 0.4984 max mem: 64948 Test: Total time: 0:00:06 (0.6919 s / it) * Acc@1 82.964 Acc@5 96.480 loss 0.631 Accuracy of the model EMA on 50000 test images: 83.0% Epoch: [323] [ 0/312] eta: 0:55:07 lr: 0.000802 min_lr: 0.000802 loss: 1.8419 (1.8419) weight_decay: 0.0500 (0.0500) time: 10.6017 data: 7.7085 max mem: 64948 Epoch: [323] [ 10/312] eta: 0:08:09 lr: 0.000801 min_lr: 0.000801 loss: 1.8419 (1.8488) weight_decay: 0.0500 (0.0500) time: 1.6211 data: 0.7012 max mem: 64948 Epoch: [323] [ 20/312] eta: 0:05:44 lr: 0.000801 min_lr: 0.000801 loss: 1.9080 (1.8712) weight_decay: 0.0500 (0.0500) time: 0.7096 data: 0.0004 max mem: 64948 Epoch: [323] [ 30/312] eta: 0:04:48 lr: 0.000801 min_lr: 0.000801 loss: 1.9688 (1.8749) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [323] [ 40/312] eta: 0:04:16 lr: 0.000800 min_lr: 0.000800 loss: 1.8161 (1.8489) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [323] [ 50/312] eta: 0:03:54 lr: 0.000800 min_lr: 0.000800 loss: 1.9319 (1.8771) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0003 max mem: 64948 Epoch: [323] [ 60/312] eta: 0:03:37 lr: 0.000799 min_lr: 0.000799 loss: 1.9319 (1.8547) weight_decay: 0.0500 (0.0500) time: 0.7016 data: 0.0004 max mem: 64948 Epoch: [323] [ 70/312] eta: 0:03:23 lr: 0.000799 min_lr: 0.000799 loss: 1.8384 (1.8689) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [323] [ 80/312] eta: 0:03:10 lr: 0.000799 min_lr: 0.000799 loss: 2.0286 (1.8779) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [323] [ 90/312] eta: 0:02:59 lr: 0.000798 min_lr: 0.000798 loss: 1.9696 (1.8719) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [323] [100/312] eta: 0:02:48 lr: 0.000798 min_lr: 0.000798 loss: 1.9696 (1.8787) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [323] [110/312] eta: 0:02:39 lr: 0.000798 min_lr: 0.000798 loss: 1.8853 (1.8651) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [323] [120/312] eta: 0:02:29 lr: 0.000797 min_lr: 0.000797 loss: 1.7065 (1.8564) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [323] [130/312] eta: 0:02:20 lr: 0.000797 min_lr: 0.000797 loss: 1.8271 (1.8538) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [323] [140/312] eta: 0:02:12 lr: 0.000796 min_lr: 0.000796 loss: 1.8616 (1.8455) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [323] [150/312] eta: 0:02:03 lr: 0.000796 min_lr: 0.000796 loss: 1.8561 (1.8439) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [323] [160/312] eta: 0:01:55 lr: 0.000796 min_lr: 0.000796 loss: 1.7444 (1.8380) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [323] [170/312] eta: 0:01:47 lr: 0.000795 min_lr: 0.000795 loss: 1.6740 (1.8318) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [323] [180/312] eta: 0:01:39 lr: 0.000795 min_lr: 0.000795 loss: 1.8266 (1.8322) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [323] [190/312] eta: 0:01:31 lr: 0.000795 min_lr: 0.000795 loss: 1.9062 (1.8356) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [323] [200/312] eta: 0:01:23 lr: 0.000794 min_lr: 0.000794 loss: 1.8453 (1.8381) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [323] [210/312] eta: 0:01:15 lr: 0.000794 min_lr: 0.000794 loss: 1.8924 (1.8396) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [323] [220/312] eta: 0:01:08 lr: 0.000793 min_lr: 0.000793 loss: 1.9523 (1.8439) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [323] [230/312] eta: 0:01:00 lr: 0.000793 min_lr: 0.000793 loss: 1.9606 (1.8462) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [323] [240/312] eta: 0:00:53 lr: 0.000793 min_lr: 0.000793 loss: 2.0256 (1.8567) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [323] [250/312] eta: 0:00:45 lr: 0.000792 min_lr: 0.000792 loss: 1.9833 (1.8512) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [323] [260/312] eta: 0:00:38 lr: 0.000792 min_lr: 0.000792 loss: 1.7958 (1.8483) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [323] [270/312] eta: 0:00:30 lr: 0.000792 min_lr: 0.000792 loss: 1.7990 (1.8461) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [323] [280/312] eta: 0:00:23 lr: 0.000791 min_lr: 0.000791 loss: 1.8779 (1.8483) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0010 max mem: 64948 Epoch: [323] [290/312] eta: 0:00:16 lr: 0.000791 min_lr: 0.000791 loss: 1.9166 (1.8524) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [323] [300/312] eta: 0:00:08 lr: 0.000791 min_lr: 0.000791 loss: 1.9035 (1.8497) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [323] [310/312] eta: 0:00:01 lr: 0.000790 min_lr: 0.000790 loss: 1.6222 (1.8441) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [323] [311/312] eta: 0:00:00 lr: 0.000790 min_lr: 0.000790 loss: 1.6904 (1.8445) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [323] Total time: 0:03:47 (0.7301 s / it) Averaged stats: lr: 0.000790 min_lr: 0.000790 loss: 1.6904 (1.8612) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4972 (0.4972) acc1: 86.7188 (86.7188) acc5: 97.3958 (97.3958) time: 4.5025 data: 4.2832 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7101 (0.7042) acc1: 82.2917 (80.8640) acc5: 96.6146 (96.0000) time: 0.6521 data: 0.4760 max mem: 64948 Test: Total time: 0:00:06 (0.6752 s / it) * Acc@1 81.858 Acc@5 95.908 loss 0.689 Accuracy of the model on the 50000 test images: 81.9% Max accuracy: 81.86% Test: [0/9] eta: 0:00:41 loss: 0.4613 (0.4613) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.6632 data: 4.4552 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6381 (0.6538) acc1: 83.3333 (81.4080) acc5: 97.1354 (96.6400) time: 0.6694 data: 0.4951 max mem: 64948 Test: Total time: 0:00:06 (0.6771 s / it) * Acc@1 82.970 Acc@5 96.480 loss 0.631 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 82.97% Epoch: [324] [ 0/312] eta: 0:50:59 lr: 0.000790 min_lr: 0.000790 loss: 2.0045 (2.0045) weight_decay: 0.0500 (0.0500) time: 9.8057 data: 9.0125 max mem: 64948 Epoch: [324] [ 10/312] eta: 0:07:46 lr: 0.000790 min_lr: 0.000790 loss: 2.0285 (1.9931) weight_decay: 0.0500 (0.0500) time: 1.5437 data: 0.8197 max mem: 64948 Epoch: [324] [ 20/312] eta: 0:05:32 lr: 0.000789 min_lr: 0.000789 loss: 1.9810 (1.9524) weight_decay: 0.0500 (0.0500) time: 0.7052 data: 0.0004 max mem: 64948 Epoch: [324] [ 30/312] eta: 0:04:40 lr: 0.000789 min_lr: 0.000789 loss: 1.8890 (1.9172) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [324] [ 40/312] eta: 0:04:11 lr: 0.000789 min_lr: 0.000789 loss: 1.6267 (1.8282) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0003 max mem: 64948 Epoch: [324] [ 50/312] eta: 0:03:50 lr: 0.000788 min_lr: 0.000788 loss: 1.6544 (1.8249) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0003 max mem: 64948 Epoch: [324] [ 60/312] eta: 0:03:33 lr: 0.000788 min_lr: 0.000788 loss: 1.8869 (1.8489) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [324] [ 70/312] eta: 0:03:20 lr: 0.000787 min_lr: 0.000787 loss: 2.0312 (1.8617) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [324] [ 80/312] eta: 0:03:08 lr: 0.000787 min_lr: 0.000787 loss: 2.0608 (1.8814) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [324] [ 90/312] eta: 0:02:57 lr: 0.000787 min_lr: 0.000787 loss: 2.0802 (1.8908) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [324] [100/312] eta: 0:02:47 lr: 0.000786 min_lr: 0.000786 loss: 1.7779 (1.8606) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [324] [110/312] eta: 0:02:37 lr: 0.000786 min_lr: 0.000786 loss: 1.5218 (1.8404) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [324] [120/312] eta: 0:02:28 lr: 0.000786 min_lr: 0.000786 loss: 1.6737 (1.8375) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [324] [130/312] eta: 0:02:19 lr: 0.000785 min_lr: 0.000785 loss: 1.8325 (1.8447) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [324] [140/312] eta: 0:02:11 lr: 0.000785 min_lr: 0.000785 loss: 1.9274 (1.8431) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [324] [150/312] eta: 0:02:02 lr: 0.000784 min_lr: 0.000784 loss: 1.9274 (1.8405) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [324] [160/312] eta: 0:01:54 lr: 0.000784 min_lr: 0.000784 loss: 1.9410 (1.8498) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [324] [170/312] eta: 0:01:46 lr: 0.000784 min_lr: 0.000784 loss: 1.9416 (1.8517) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [324] [180/312] eta: 0:01:38 lr: 0.000783 min_lr: 0.000783 loss: 1.8895 (1.8532) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [324] [190/312] eta: 0:01:30 lr: 0.000783 min_lr: 0.000783 loss: 1.8895 (1.8558) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [324] [200/312] eta: 0:01:23 lr: 0.000783 min_lr: 0.000783 loss: 2.0153 (1.8591) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [324] [210/312] eta: 0:01:15 lr: 0.000782 min_lr: 0.000782 loss: 2.0026 (1.8616) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [324] [220/312] eta: 0:01:07 lr: 0.000782 min_lr: 0.000782 loss: 2.0262 (1.8688) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [324] [230/312] eta: 0:01:00 lr: 0.000781 min_lr: 0.000781 loss: 2.0564 (1.8722) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [324] [240/312] eta: 0:00:52 lr: 0.000781 min_lr: 0.000781 loss: 2.0437 (1.8823) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [324] [250/312] eta: 0:00:45 lr: 0.000781 min_lr: 0.000781 loss: 1.9671 (1.8787) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [324] [260/312] eta: 0:00:37 lr: 0.000780 min_lr: 0.000780 loss: 1.8161 (1.8762) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [324] [270/312] eta: 0:00:30 lr: 0.000780 min_lr: 0.000780 loss: 1.8322 (1.8745) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [324] [280/312] eta: 0:00:23 lr: 0.000780 min_lr: 0.000780 loss: 1.7933 (1.8721) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [324] [290/312] eta: 0:00:15 lr: 0.000779 min_lr: 0.000779 loss: 1.8848 (1.8757) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [324] [300/312] eta: 0:00:08 lr: 0.000779 min_lr: 0.000779 loss: 1.8848 (1.8724) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [324] [310/312] eta: 0:00:01 lr: 0.000779 min_lr: 0.000779 loss: 1.8455 (1.8706) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [324] [311/312] eta: 0:00:00 lr: 0.000778 min_lr: 0.000778 loss: 1.8585 (1.8719) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [324] Total time: 0:03:46 (0.7275 s / it) Averaged stats: lr: 0.000778 min_lr: 0.000778 loss: 1.8585 (1.8571) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:45 loss: 0.5069 (0.5069) acc1: 86.4583 (86.4583) acc5: 97.1354 (97.1354) time: 5.0671 data: 4.8553 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7372 (0.7188) acc1: 81.7708 (80.7040) acc5: 96.3542 (95.9680) time: 0.7143 data: 0.5396 max mem: 64948 Test: Total time: 0:00:06 (0.7401 s / it) * Acc@1 81.770 Acc@5 95.868 loss 0.693 Accuracy of the model on the 50000 test images: 81.8% Max accuracy: 81.86% Test: [0/9] eta: 0:00:45 loss: 0.4615 (0.4615) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 5.1011 data: 4.8877 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6380 (0.6535) acc1: 83.3333 (81.4720) acc5: 97.1354 (96.6720) time: 0.7180 data: 0.5432 max mem: 64948 Test: Total time: 0:00:06 (0.7274 s / it) * Acc@1 82.968 Acc@5 96.482 loss 0.630 Accuracy of the model EMA on 50000 test images: 83.0% Epoch: [325] [ 0/312] eta: 0:53:15 lr: 0.000778 min_lr: 0.000778 loss: 2.3090 (2.3090) weight_decay: 0.0500 (0.0500) time: 10.2426 data: 7.8392 max mem: 64948 Epoch: [325] [ 10/312] eta: 0:08:00 lr: 0.000778 min_lr: 0.000778 loss: 2.1387 (2.0492) weight_decay: 0.0500 (0.0500) time: 1.5916 data: 0.7131 max mem: 64948 Epoch: [325] [ 20/312] eta: 0:05:40 lr: 0.000778 min_lr: 0.000778 loss: 1.9834 (1.9755) weight_decay: 0.0500 (0.0500) time: 0.7134 data: 0.0004 max mem: 64948 Epoch: [325] [ 30/312] eta: 0:04:46 lr: 0.000777 min_lr: 0.000777 loss: 1.9069 (1.9213) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [325] [ 40/312] eta: 0:04:14 lr: 0.000777 min_lr: 0.000777 loss: 1.7148 (1.8541) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [325] [ 50/312] eta: 0:03:52 lr: 0.000777 min_lr: 0.000777 loss: 1.8372 (1.8530) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [325] [ 60/312] eta: 0:03:36 lr: 0.000776 min_lr: 0.000776 loss: 1.9556 (1.8512) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [325] [ 70/312] eta: 0:03:21 lr: 0.000776 min_lr: 0.000776 loss: 1.9135 (1.8388) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [325] [ 80/312] eta: 0:03:09 lr: 0.000775 min_lr: 0.000775 loss: 1.9210 (1.8623) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [325] [ 90/312] eta: 0:02:58 lr: 0.000775 min_lr: 0.000775 loss: 1.9618 (1.8708) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [325] [100/312] eta: 0:02:48 lr: 0.000775 min_lr: 0.000775 loss: 1.9618 (1.8795) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [325] [110/312] eta: 0:02:38 lr: 0.000774 min_lr: 0.000774 loss: 1.9934 (1.8784) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [325] [120/312] eta: 0:02:29 lr: 0.000774 min_lr: 0.000774 loss: 1.8813 (1.8701) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [325] [130/312] eta: 0:02:20 lr: 0.000774 min_lr: 0.000774 loss: 1.8813 (1.8680) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [325] [140/312] eta: 0:02:11 lr: 0.000773 min_lr: 0.000773 loss: 1.9410 (1.8682) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [325] [150/312] eta: 0:02:03 lr: 0.000773 min_lr: 0.000773 loss: 1.9523 (1.8677) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [325] [160/312] eta: 0:01:54 lr: 0.000773 min_lr: 0.000773 loss: 1.8158 (1.8577) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [325] [170/312] eta: 0:01:46 lr: 0.000772 min_lr: 0.000772 loss: 1.8158 (1.8639) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [325] [180/312] eta: 0:01:38 lr: 0.000772 min_lr: 0.000772 loss: 1.9184 (1.8595) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0004 max mem: 64948 Epoch: [325] [190/312] eta: 0:01:31 lr: 0.000771 min_lr: 0.000771 loss: 1.8490 (1.8557) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [325] [200/312] eta: 0:01:23 lr: 0.000771 min_lr: 0.000771 loss: 1.9334 (1.8601) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [325] [210/312] eta: 0:01:15 lr: 0.000771 min_lr: 0.000771 loss: 1.9093 (1.8536) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [325] [220/312] eta: 0:01:08 lr: 0.000770 min_lr: 0.000770 loss: 1.8253 (1.8558) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [325] [230/312] eta: 0:01:00 lr: 0.000770 min_lr: 0.000770 loss: 2.0072 (1.8567) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [325] [240/312] eta: 0:00:52 lr: 0.000770 min_lr: 0.000770 loss: 1.9818 (1.8580) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [325] [250/312] eta: 0:00:45 lr: 0.000769 min_lr: 0.000769 loss: 1.9289 (1.8567) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [325] [260/312] eta: 0:00:38 lr: 0.000769 min_lr: 0.000769 loss: 1.8963 (1.8573) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [325] [270/312] eta: 0:00:30 lr: 0.000768 min_lr: 0.000768 loss: 1.9294 (1.8523) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [325] [280/312] eta: 0:00:23 lr: 0.000768 min_lr: 0.000768 loss: 1.9652 (1.8519) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [325] [290/312] eta: 0:00:16 lr: 0.000768 min_lr: 0.000768 loss: 1.9656 (1.8532) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [325] [300/312] eta: 0:00:08 lr: 0.000767 min_lr: 0.000767 loss: 1.9656 (1.8533) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [325] [310/312] eta: 0:00:01 lr: 0.000767 min_lr: 0.000767 loss: 1.9407 (1.8532) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [325] [311/312] eta: 0:00:00 lr: 0.000767 min_lr: 0.000767 loss: 1.9218 (1.8510) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [325] Total time: 0:03:47 (0.7289 s / it) Averaged stats: lr: 0.000767 min_lr: 0.000767 loss: 1.9218 (1.8520) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4681 (0.4681) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.3864 data: 4.1824 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7195 (0.7039) acc1: 82.5521 (81.2160) acc5: 96.8750 (96.0960) time: 0.6386 data: 0.4648 max mem: 64948 Test: Total time: 0:00:05 (0.6620 s / it) * Acc@1 82.078 Acc@5 95.912 loss 0.686 Accuracy of the model on the 50000 test images: 82.1% Max accuracy: 82.08% Test: [0/9] eta: 0:00:40 loss: 0.4616 (0.4616) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 4.5282 data: 4.3216 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6379 (0.6534) acc1: 83.3333 (81.5040) acc5: 97.1354 (96.7040) time: 0.6552 data: 0.4803 max mem: 64948 Test: Total time: 0:00:05 (0.6647 s / it) * Acc@1 82.980 Acc@5 96.494 loss 0.630 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 82.98% Epoch: [326] [ 0/312] eta: 0:54:19 lr: 0.000767 min_lr: 0.000767 loss: 1.9423 (1.9423) weight_decay: 0.0500 (0.0500) time: 10.4461 data: 9.7164 max mem: 64948 Epoch: [326] [ 10/312] eta: 0:08:00 lr: 0.000767 min_lr: 0.000767 loss: 1.9478 (1.9615) weight_decay: 0.0500 (0.0500) time: 1.5917 data: 0.8836 max mem: 64948 Epoch: [326] [ 20/312] eta: 0:05:40 lr: 0.000766 min_lr: 0.000766 loss: 2.0629 (1.9879) weight_decay: 0.0500 (0.0500) time: 0.7006 data: 0.0004 max mem: 64948 Epoch: [326] [ 30/312] eta: 0:04:45 lr: 0.000766 min_lr: 0.000766 loss: 2.0323 (1.9224) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [326] [ 40/312] eta: 0:04:14 lr: 0.000765 min_lr: 0.000765 loss: 2.0323 (1.9368) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [326] [ 50/312] eta: 0:03:52 lr: 0.000765 min_lr: 0.000765 loss: 1.9818 (1.9107) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [326] [ 60/312] eta: 0:03:35 lr: 0.000765 min_lr: 0.000765 loss: 1.8715 (1.9091) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [326] [ 70/312] eta: 0:03:21 lr: 0.000764 min_lr: 0.000764 loss: 1.9624 (1.9016) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [326] [ 80/312] eta: 0:03:09 lr: 0.000764 min_lr: 0.000764 loss: 1.9777 (1.8984) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [326] [ 90/312] eta: 0:02:58 lr: 0.000764 min_lr: 0.000764 loss: 1.9506 (1.8968) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [326] [100/312] eta: 0:02:47 lr: 0.000763 min_lr: 0.000763 loss: 1.9447 (1.8878) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [326] [110/312] eta: 0:02:38 lr: 0.000763 min_lr: 0.000763 loss: 1.9241 (1.8846) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [326] [120/312] eta: 0:02:29 lr: 0.000763 min_lr: 0.000763 loss: 1.9063 (1.8908) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [326] [130/312] eta: 0:02:20 lr: 0.000762 min_lr: 0.000762 loss: 1.7824 (1.8749) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [326] [140/312] eta: 0:02:11 lr: 0.000762 min_lr: 0.000762 loss: 1.6653 (1.8688) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [326] [150/312] eta: 0:02:03 lr: 0.000761 min_lr: 0.000761 loss: 1.7976 (1.8621) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [326] [160/312] eta: 0:01:55 lr: 0.000761 min_lr: 0.000761 loss: 1.8033 (1.8602) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [326] [170/312] eta: 0:01:46 lr: 0.000761 min_lr: 0.000761 loss: 1.9532 (1.8707) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [326] [180/312] eta: 0:01:38 lr: 0.000760 min_lr: 0.000760 loss: 1.9640 (1.8694) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [326] [190/312] eta: 0:01:31 lr: 0.000760 min_lr: 0.000760 loss: 1.9591 (1.8697) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [326] [200/312] eta: 0:01:23 lr: 0.000760 min_lr: 0.000760 loss: 1.9365 (1.8803) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [326] [210/312] eta: 0:01:15 lr: 0.000759 min_lr: 0.000759 loss: 1.9138 (1.8791) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [326] [220/312] eta: 0:01:08 lr: 0.000759 min_lr: 0.000759 loss: 2.0310 (1.8826) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [326] [230/312] eta: 0:01:00 lr: 0.000758 min_lr: 0.000758 loss: 2.0310 (1.8849) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [326] [240/312] eta: 0:00:52 lr: 0.000758 min_lr: 0.000758 loss: 1.9822 (1.8847) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [326] [250/312] eta: 0:00:45 lr: 0.000758 min_lr: 0.000758 loss: 2.0030 (1.8879) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [326] [260/312] eta: 0:00:38 lr: 0.000757 min_lr: 0.000757 loss: 1.9636 (1.8935) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [326] [270/312] eta: 0:00:30 lr: 0.000757 min_lr: 0.000757 loss: 1.9002 (1.8925) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [326] [280/312] eta: 0:00:23 lr: 0.000757 min_lr: 0.000757 loss: 1.8843 (1.8943) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [326] [290/312] eta: 0:00:16 lr: 0.000756 min_lr: 0.000756 loss: 1.9725 (1.8940) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [326] [300/312] eta: 0:00:08 lr: 0.000756 min_lr: 0.000756 loss: 1.9350 (1.8951) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [326] [310/312] eta: 0:00:01 lr: 0.000756 min_lr: 0.000756 loss: 1.8726 (1.8918) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [326] [311/312] eta: 0:00:00 lr: 0.000756 min_lr: 0.000756 loss: 1.7962 (1.8898) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [326] Total time: 0:03:47 (0.7293 s / it) Averaged stats: lr: 0.000756 min_lr: 0.000756 loss: 1.7962 (1.8458) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.4782 (0.4782) acc1: 87.5000 (87.5000) acc5: 98.1771 (98.1771) time: 4.2608 data: 4.0476 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6833 (0.6869) acc1: 83.3333 (81.6960) acc5: 96.6146 (96.2880) time: 0.6350 data: 0.4602 max mem: 64948 Test: Total time: 0:00:05 (0.6523 s / it) * Acc@1 81.938 Acc@5 96.096 loss 0.680 Accuracy of the model on the 50000 test images: 81.9% Max accuracy: 82.08% Test: [0/9] eta: 0:00:43 loss: 0.4616 (0.4616) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 4.8720 data: 4.6541 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6379 (0.6533) acc1: 83.3333 (81.5360) acc5: 97.1354 (96.6720) time: 0.6928 data: 0.5172 max mem: 64948 Test: Total time: 0:00:06 (0.7038 s / it) * Acc@1 82.984 Acc@5 96.488 loss 0.630 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 82.98% Epoch: [327] [ 0/312] eta: 0:47:32 lr: 0.000755 min_lr: 0.000755 loss: 1.2238 (1.2238) weight_decay: 0.0500 (0.0500) time: 9.1420 data: 7.1473 max mem: 64948 Epoch: [327] [ 10/312] eta: 0:07:36 lr: 0.000755 min_lr: 0.000755 loss: 1.9002 (1.7567) weight_decay: 0.0500 (0.0500) time: 1.5129 data: 0.6503 max mem: 64948 Epoch: [327] [ 20/312] eta: 0:05:27 lr: 0.000755 min_lr: 0.000755 loss: 1.8398 (1.7537) weight_decay: 0.0500 (0.0500) time: 0.7221 data: 0.0004 max mem: 64948 Epoch: [327] [ 30/312] eta: 0:04:38 lr: 0.000754 min_lr: 0.000754 loss: 1.8391 (1.7967) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [327] [ 40/312] eta: 0:04:09 lr: 0.000754 min_lr: 0.000754 loss: 1.8391 (1.8038) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [327] [ 50/312] eta: 0:03:48 lr: 0.000754 min_lr: 0.000754 loss: 1.6956 (1.7700) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0003 max mem: 64948 Epoch: [327] [ 60/312] eta: 0:03:32 lr: 0.000753 min_lr: 0.000753 loss: 1.6956 (1.7899) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [327] [ 70/312] eta: 0:03:19 lr: 0.000753 min_lr: 0.000753 loss: 1.9472 (1.8128) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [327] [ 80/312] eta: 0:03:07 lr: 0.000753 min_lr: 0.000753 loss: 1.9472 (1.8102) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [327] [ 90/312] eta: 0:02:56 lr: 0.000752 min_lr: 0.000752 loss: 1.6755 (1.7894) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [327] [100/312] eta: 0:02:46 lr: 0.000752 min_lr: 0.000752 loss: 1.7635 (1.7972) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [327] [110/312] eta: 0:02:37 lr: 0.000751 min_lr: 0.000751 loss: 1.7795 (1.7960) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [327] [120/312] eta: 0:02:28 lr: 0.000751 min_lr: 0.000751 loss: 1.9820 (1.8217) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [327] [130/312] eta: 0:02:19 lr: 0.000751 min_lr: 0.000751 loss: 2.0214 (1.8290) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [327] [140/312] eta: 0:02:10 lr: 0.000750 min_lr: 0.000750 loss: 1.9132 (1.8272) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [327] [150/312] eta: 0:02:02 lr: 0.000750 min_lr: 0.000750 loss: 2.0326 (1.8512) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [327] [160/312] eta: 0:01:54 lr: 0.000750 min_lr: 0.000750 loss: 2.0518 (1.8458) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [327] [170/312] eta: 0:01:46 lr: 0.000749 min_lr: 0.000749 loss: 1.9433 (1.8493) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [327] [180/312] eta: 0:01:38 lr: 0.000749 min_lr: 0.000749 loss: 1.8353 (1.8438) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [327] [190/312] eta: 0:01:30 lr: 0.000749 min_lr: 0.000749 loss: 1.7380 (1.8393) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [327] [200/312] eta: 0:01:22 lr: 0.000748 min_lr: 0.000748 loss: 1.8358 (1.8427) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [327] [210/312] eta: 0:01:15 lr: 0.000748 min_lr: 0.000748 loss: 1.8808 (1.8366) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [327] [220/312] eta: 0:01:07 lr: 0.000747 min_lr: 0.000747 loss: 1.8417 (1.8384) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [327] [230/312] eta: 0:01:00 lr: 0.000747 min_lr: 0.000747 loss: 1.8297 (1.8347) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [327] [240/312] eta: 0:00:52 lr: 0.000747 min_lr: 0.000747 loss: 1.8783 (1.8387) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [327] [250/312] eta: 0:00:45 lr: 0.000746 min_lr: 0.000746 loss: 1.8885 (1.8361) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [327] [260/312] eta: 0:00:37 lr: 0.000746 min_lr: 0.000746 loss: 1.9088 (1.8398) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [327] [270/312] eta: 0:00:30 lr: 0.000746 min_lr: 0.000746 loss: 1.9088 (1.8397) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [327] [280/312] eta: 0:00:23 lr: 0.000745 min_lr: 0.000745 loss: 1.9908 (1.8445) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [327] [290/312] eta: 0:00:15 lr: 0.000745 min_lr: 0.000745 loss: 2.0214 (1.8433) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [327] [300/312] eta: 0:00:08 lr: 0.000745 min_lr: 0.000745 loss: 1.8366 (1.8433) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [327] [310/312] eta: 0:00:01 lr: 0.000744 min_lr: 0.000744 loss: 1.8761 (1.8413) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [327] [311/312] eta: 0:00:00 lr: 0.000744 min_lr: 0.000744 loss: 1.8761 (1.8430) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [327] Total time: 0:03:46 (0.7265 s / it) Averaged stats: lr: 0.000744 min_lr: 0.000744 loss: 1.8761 (1.8559) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.5242 (0.5242) acc1: 86.1979 (86.1979) acc5: 97.9167 (97.9167) time: 4.5770 data: 4.3682 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7005 (0.7062) acc1: 82.0312 (81.4720) acc5: 96.0938 (96.0320) time: 0.6598 data: 0.4854 max mem: 64948 Test: Total time: 0:00:06 (0.6820 s / it) * Acc@1 82.018 Acc@5 95.930 loss 0.687 Accuracy of the model on the 50000 test images: 82.0% Max accuracy: 82.08% Test: [0/9] eta: 0:00:44 loss: 0.4622 (0.4622) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 4.9588 data: 4.7558 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6378 (0.6532) acc1: 83.3333 (81.4720) acc5: 97.1354 (96.6720) time: 0.7023 data: 0.5285 max mem: 64948 Test: Total time: 0:00:06 (0.7094 s / it) * Acc@1 82.988 Acc@5 96.498 loss 0.630 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 82.99% Epoch: [328] [ 0/312] eta: 0:48:52 lr: 0.000744 min_lr: 0.000744 loss: 1.7311 (1.7311) weight_decay: 0.0500 (0.0500) time: 9.3992 data: 8.6130 max mem: 64948 Epoch: [328] [ 10/312] eta: 0:07:37 lr: 0.000744 min_lr: 0.000744 loss: 1.9862 (1.9385) weight_decay: 0.0500 (0.0500) time: 1.5163 data: 0.7835 max mem: 64948 Epoch: [328] [ 20/312] eta: 0:05:28 lr: 0.000743 min_lr: 0.000743 loss: 1.7792 (1.8661) weight_decay: 0.0500 (0.0500) time: 0.7108 data: 0.0004 max mem: 64948 Epoch: [328] [ 30/312] eta: 0:04:38 lr: 0.000743 min_lr: 0.000743 loss: 1.8629 (1.8779) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [328] [ 40/312] eta: 0:04:08 lr: 0.000743 min_lr: 0.000743 loss: 1.8601 (1.8441) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [328] [ 50/312] eta: 0:03:48 lr: 0.000742 min_lr: 0.000742 loss: 1.8216 (1.8427) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [328] [ 60/312] eta: 0:03:32 lr: 0.000742 min_lr: 0.000742 loss: 1.7937 (1.8171) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [328] [ 70/312] eta: 0:03:19 lr: 0.000742 min_lr: 0.000742 loss: 1.5867 (1.7788) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [328] [ 80/312] eta: 0:03:07 lr: 0.000741 min_lr: 0.000741 loss: 1.6868 (1.8001) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [328] [ 90/312] eta: 0:02:56 lr: 0.000741 min_lr: 0.000741 loss: 1.9407 (1.8037) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [328] [100/312] eta: 0:02:46 lr: 0.000740 min_lr: 0.000740 loss: 1.7059 (1.7916) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [328] [110/312] eta: 0:02:36 lr: 0.000740 min_lr: 0.000740 loss: 1.7499 (1.7885) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [328] [120/312] eta: 0:02:27 lr: 0.000740 min_lr: 0.000740 loss: 1.7551 (1.7806) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [328] [130/312] eta: 0:02:19 lr: 0.000739 min_lr: 0.000739 loss: 1.7771 (1.7879) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [328] [140/312] eta: 0:02:10 lr: 0.000739 min_lr: 0.000739 loss: 1.8584 (1.7892) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [328] [150/312] eta: 0:02:02 lr: 0.000739 min_lr: 0.000739 loss: 1.8869 (1.7949) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [328] [160/312] eta: 0:01:54 lr: 0.000738 min_lr: 0.000738 loss: 1.9295 (1.7951) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [328] [170/312] eta: 0:01:46 lr: 0.000738 min_lr: 0.000738 loss: 1.9329 (1.8096) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [328] [180/312] eta: 0:01:38 lr: 0.000738 min_lr: 0.000738 loss: 2.0270 (1.8091) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [328] [190/312] eta: 0:01:30 lr: 0.000737 min_lr: 0.000737 loss: 1.9086 (1.8132) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [328] [200/312] eta: 0:01:22 lr: 0.000737 min_lr: 0.000737 loss: 1.9086 (1.8141) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [328] [210/312] eta: 0:01:15 lr: 0.000736 min_lr: 0.000736 loss: 1.7836 (1.8149) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [328] [220/312] eta: 0:01:07 lr: 0.000736 min_lr: 0.000736 loss: 1.8172 (1.8159) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [328] [230/312] eta: 0:01:00 lr: 0.000736 min_lr: 0.000736 loss: 1.9069 (1.8230) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [328] [240/312] eta: 0:00:52 lr: 0.000735 min_lr: 0.000735 loss: 2.0304 (1.8270) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [328] [250/312] eta: 0:00:45 lr: 0.000735 min_lr: 0.000735 loss: 2.0031 (1.8312) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [328] [260/312] eta: 0:00:37 lr: 0.000735 min_lr: 0.000735 loss: 1.9358 (1.8289) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [328] [270/312] eta: 0:00:30 lr: 0.000734 min_lr: 0.000734 loss: 1.7946 (1.8300) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [328] [280/312] eta: 0:00:23 lr: 0.000734 min_lr: 0.000734 loss: 1.9344 (1.8280) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [328] [290/312] eta: 0:00:15 lr: 0.000734 min_lr: 0.000734 loss: 1.9649 (1.8362) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [328] [300/312] eta: 0:00:08 lr: 0.000733 min_lr: 0.000733 loss: 1.9001 (1.8341) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [328] [310/312] eta: 0:00:01 lr: 0.000733 min_lr: 0.000733 loss: 1.8879 (1.8365) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [328] [311/312] eta: 0:00:00 lr: 0.000733 min_lr: 0.000733 loss: 1.8745 (1.8359) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [328] Total time: 0:03:46 (0.7265 s / it) Averaged stats: lr: 0.000733 min_lr: 0.000733 loss: 1.8745 (1.8486) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4860 (0.4860) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.6139 data: 4.4076 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6751 (0.6970) acc1: 82.5521 (81.0240) acc5: 96.6146 (96.0000) time: 0.6640 data: 0.4898 max mem: 64948 Test: Total time: 0:00:06 (0.6878 s / it) * Acc@1 82.056 Acc@5 95.952 loss 0.686 Accuracy of the model on the 50000 test images: 82.1% Max accuracy: 82.08% Test: [0/9] eta: 0:00:42 loss: 0.4627 (0.4627) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 4.7203 data: 4.5044 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6377 (0.6530) acc1: 83.3333 (81.4400) acc5: 97.1354 (96.6720) time: 0.6758 data: 0.5006 max mem: 64948 Test: Total time: 0:00:06 (0.6835 s / it) * Acc@1 82.986 Acc@5 96.486 loss 0.630 Accuracy of the model EMA on 50000 test images: 83.0% Epoch: [329] [ 0/312] eta: 0:53:25 lr: 0.000733 min_lr: 0.000733 loss: 1.7400 (1.7400) weight_decay: 0.0500 (0.0500) time: 10.2744 data: 8.1586 max mem: 64948 Epoch: [329] [ 10/312] eta: 0:08:10 lr: 0.000732 min_lr: 0.000732 loss: 1.7400 (1.7652) weight_decay: 0.0500 (0.0500) time: 1.6249 data: 0.7421 max mem: 64948 Epoch: [329] [ 20/312] eta: 0:05:45 lr: 0.000732 min_lr: 0.000732 loss: 1.7677 (1.7455) weight_decay: 0.0500 (0.0500) time: 0.7280 data: 0.0004 max mem: 64948 Epoch: [329] [ 30/312] eta: 0:04:49 lr: 0.000732 min_lr: 0.000732 loss: 1.8585 (1.7988) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0003 max mem: 64948 Epoch: [329] [ 40/312] eta: 0:04:17 lr: 0.000731 min_lr: 0.000731 loss: 1.9161 (1.8098) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0003 max mem: 64948 Epoch: [329] [ 50/312] eta: 0:03:54 lr: 0.000731 min_lr: 0.000731 loss: 1.9248 (1.8287) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [329] [ 60/312] eta: 0:03:37 lr: 0.000731 min_lr: 0.000731 loss: 1.8797 (1.8227) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [329] [ 70/312] eta: 0:03:23 lr: 0.000730 min_lr: 0.000730 loss: 1.9513 (1.8442) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [329] [ 80/312] eta: 0:03:10 lr: 0.000730 min_lr: 0.000730 loss: 1.9513 (1.8526) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [329] [ 90/312] eta: 0:02:59 lr: 0.000729 min_lr: 0.000729 loss: 1.9556 (1.8611) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [329] [100/312] eta: 0:02:48 lr: 0.000729 min_lr: 0.000729 loss: 1.9562 (1.8609) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [329] [110/312] eta: 0:02:39 lr: 0.000729 min_lr: 0.000729 loss: 2.0008 (1.8579) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [329] [120/312] eta: 0:02:29 lr: 0.000728 min_lr: 0.000728 loss: 1.9844 (1.8649) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [329] [130/312] eta: 0:02:20 lr: 0.000728 min_lr: 0.000728 loss: 1.9017 (1.8694) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [329] [140/312] eta: 0:02:12 lr: 0.000728 min_lr: 0.000728 loss: 1.8662 (1.8655) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [329] [150/312] eta: 0:02:03 lr: 0.000727 min_lr: 0.000727 loss: 1.6972 (1.8531) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [329] [160/312] eta: 0:01:55 lr: 0.000727 min_lr: 0.000727 loss: 1.7919 (1.8576) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [329] [170/312] eta: 0:01:47 lr: 0.000727 min_lr: 0.000727 loss: 1.9102 (1.8583) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [329] [180/312] eta: 0:01:39 lr: 0.000726 min_lr: 0.000726 loss: 1.8723 (1.8579) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [329] [190/312] eta: 0:01:31 lr: 0.000726 min_lr: 0.000726 loss: 1.8865 (1.8585) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [329] [200/312] eta: 0:01:23 lr: 0.000726 min_lr: 0.000726 loss: 2.0277 (1.8656) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [329] [210/312] eta: 0:01:15 lr: 0.000725 min_lr: 0.000725 loss: 1.9220 (1.8595) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [329] [220/312] eta: 0:01:08 lr: 0.000725 min_lr: 0.000725 loss: 1.8701 (1.8649) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [329] [230/312] eta: 0:01:00 lr: 0.000724 min_lr: 0.000724 loss: 2.0161 (1.8660) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [329] [240/312] eta: 0:00:53 lr: 0.000724 min_lr: 0.000724 loss: 1.8547 (1.8593) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [329] [250/312] eta: 0:00:45 lr: 0.000724 min_lr: 0.000724 loss: 1.7637 (1.8554) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [329] [260/312] eta: 0:00:38 lr: 0.000723 min_lr: 0.000723 loss: 1.8539 (1.8531) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [329] [270/312] eta: 0:00:30 lr: 0.000723 min_lr: 0.000723 loss: 2.0191 (1.8563) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [329] [280/312] eta: 0:00:23 lr: 0.000723 min_lr: 0.000723 loss: 1.9947 (1.8549) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [329] [290/312] eta: 0:00:16 lr: 0.000722 min_lr: 0.000722 loss: 1.8805 (1.8592) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0008 max mem: 64948 Epoch: [329] [300/312] eta: 0:00:08 lr: 0.000722 min_lr: 0.000722 loss: 1.9850 (1.8622) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [329] [310/312] eta: 0:00:01 lr: 0.000722 min_lr: 0.000722 loss: 1.9010 (1.8610) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [329] [311/312] eta: 0:00:00 lr: 0.000722 min_lr: 0.000722 loss: 1.7923 (1.8598) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [329] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.000722 min_lr: 0.000722 loss: 1.7923 (1.8431) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.5269 (0.5269) acc1: 85.6771 (85.6771) acc5: 97.6562 (97.6562) time: 4.4625 data: 4.2423 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6872 (0.6960) acc1: 82.8125 (81.2480) acc5: 96.8750 (95.8720) time: 0.6472 data: 0.4715 max mem: 64948 Test: Total time: 0:00:06 (0.6692 s / it) * Acc@1 82.092 Acc@5 95.864 loss 0.682 Accuracy of the model on the 50000 test images: 82.1% Max accuracy: 82.09% Test: [0/9] eta: 0:00:41 loss: 0.4627 (0.4627) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 4.5767 data: 4.3589 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6375 (0.6527) acc1: 83.0729 (81.4080) acc5: 97.1354 (96.7040) time: 0.6599 data: 0.4844 max mem: 64948 Test: Total time: 0:00:06 (0.6677 s / it) * Acc@1 82.978 Acc@5 96.492 loss 0.629 Accuracy of the model EMA on 50000 test images: 83.0% Epoch: [330] [ 0/312] eta: 0:52:52 lr: 0.000721 min_lr: 0.000721 loss: 1.3253 (1.3253) weight_decay: 0.0500 (0.0500) time: 10.1690 data: 6.9631 max mem: 64948 Epoch: [330] [ 10/312] eta: 0:08:01 lr: 0.000721 min_lr: 0.000721 loss: 1.8882 (1.7557) weight_decay: 0.0500 (0.0500) time: 1.5957 data: 0.6334 max mem: 64948 Epoch: [330] [ 20/312] eta: 0:05:40 lr: 0.000721 min_lr: 0.000721 loss: 1.9157 (1.8619) weight_decay: 0.0500 (0.0500) time: 0.7173 data: 0.0004 max mem: 64948 Epoch: [330] [ 30/312] eta: 0:04:46 lr: 0.000720 min_lr: 0.000720 loss: 1.9676 (1.8383) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [330] [ 40/312] eta: 0:04:14 lr: 0.000720 min_lr: 0.000720 loss: 1.8776 (1.8423) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [330] [ 50/312] eta: 0:03:53 lr: 0.000720 min_lr: 0.000720 loss: 1.9696 (1.8331) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [330] [ 60/312] eta: 0:03:36 lr: 0.000719 min_lr: 0.000719 loss: 1.8804 (1.8177) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [330] [ 70/312] eta: 0:03:22 lr: 0.000719 min_lr: 0.000719 loss: 1.8804 (1.8211) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [330] [ 80/312] eta: 0:03:09 lr: 0.000719 min_lr: 0.000719 loss: 1.9251 (1.8262) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [330] [ 90/312] eta: 0:02:58 lr: 0.000718 min_lr: 0.000718 loss: 1.9117 (1.8213) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [330] [100/312] eta: 0:02:48 lr: 0.000718 min_lr: 0.000718 loss: 1.7341 (1.8109) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [330] [110/312] eta: 0:02:38 lr: 0.000718 min_lr: 0.000718 loss: 1.8386 (1.8275) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [330] [120/312] eta: 0:02:29 lr: 0.000717 min_lr: 0.000717 loss: 1.8705 (1.8203) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [330] [130/312] eta: 0:02:20 lr: 0.000717 min_lr: 0.000717 loss: 1.8672 (1.8220) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [330] [140/312] eta: 0:02:11 lr: 0.000716 min_lr: 0.000716 loss: 1.9629 (1.8269) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [330] [150/312] eta: 0:02:03 lr: 0.000716 min_lr: 0.000716 loss: 2.0015 (1.8348) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [330] [160/312] eta: 0:01:55 lr: 0.000716 min_lr: 0.000716 loss: 1.9834 (1.8348) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [330] [170/312] eta: 0:01:47 lr: 0.000715 min_lr: 0.000715 loss: 1.9616 (1.8365) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [330] [180/312] eta: 0:01:39 lr: 0.000715 min_lr: 0.000715 loss: 1.8284 (1.8350) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [330] [190/312] eta: 0:01:31 lr: 0.000715 min_lr: 0.000715 loss: 1.7290 (1.8282) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [330] [200/312] eta: 0:01:23 lr: 0.000714 min_lr: 0.000714 loss: 1.8981 (1.8380) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [330] [210/312] eta: 0:01:15 lr: 0.000714 min_lr: 0.000714 loss: 1.8759 (1.8300) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [330] [220/312] eta: 0:01:08 lr: 0.000714 min_lr: 0.000714 loss: 1.8334 (1.8275) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [330] [230/312] eta: 0:01:00 lr: 0.000713 min_lr: 0.000713 loss: 1.9190 (1.8322) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [330] [240/312] eta: 0:00:53 lr: 0.000713 min_lr: 0.000713 loss: 1.9630 (1.8363) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [330] [250/312] eta: 0:00:45 lr: 0.000712 min_lr: 0.000712 loss: 1.9630 (1.8374) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [330] [260/312] eta: 0:00:38 lr: 0.000712 min_lr: 0.000712 loss: 1.8217 (1.8361) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [330] [270/312] eta: 0:00:30 lr: 0.000712 min_lr: 0.000712 loss: 1.9841 (1.8421) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [330] [280/312] eta: 0:00:23 lr: 0.000711 min_lr: 0.000711 loss: 1.9841 (1.8476) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [330] [290/312] eta: 0:00:16 lr: 0.000711 min_lr: 0.000711 loss: 1.8982 (1.8399) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [330] [300/312] eta: 0:00:08 lr: 0.000711 min_lr: 0.000711 loss: 1.6045 (1.8387) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [330] [310/312] eta: 0:00:01 lr: 0.000710 min_lr: 0.000710 loss: 1.8656 (1.8397) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [330] [311/312] eta: 0:00:00 lr: 0.000710 min_lr: 0.000710 loss: 1.8647 (1.8394) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [330] Total time: 0:03:47 (0.7298 s / it) Averaged stats: lr: 0.000710 min_lr: 0.000710 loss: 1.8647 (1.8410) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5018 (0.5018) acc1: 87.2396 (87.2396) acc5: 98.4375 (98.4375) time: 4.7306 data: 4.5265 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7152 (0.7216) acc1: 81.7708 (81.1840) acc5: 96.0938 (96.0640) time: 0.6768 data: 0.5030 max mem: 64948 Test: Total time: 0:00:06 (0.7003 s / it) * Acc@1 81.964 Acc@5 95.930 loss 0.692 Accuracy of the model on the 50000 test images: 82.0% Max accuracy: 82.09% Test: [0/9] eta: 0:00:41 loss: 0.4625 (0.4625) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.6217 data: 4.4062 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6371 (0.6525) acc1: 83.0729 (81.4720) acc5: 97.1354 (96.7040) time: 0.6648 data: 0.4897 max mem: 64948 Test: Total time: 0:00:06 (0.6778 s / it) * Acc@1 82.996 Acc@5 96.486 loss 0.629 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 83.00% Epoch: [331] [ 0/312] eta: 0:49:58 lr: 0.000710 min_lr: 0.000710 loss: 2.3387 (2.3387) weight_decay: 0.0500 (0.0500) time: 9.6102 data: 8.8290 max mem: 64948 Epoch: [331] [ 10/312] eta: 0:07:50 lr: 0.000710 min_lr: 0.000710 loss: 2.1753 (2.0143) weight_decay: 0.0500 (0.0500) time: 1.5585 data: 0.8030 max mem: 64948 Epoch: [331] [ 20/312] eta: 0:05:35 lr: 0.000710 min_lr: 0.000710 loss: 1.9305 (1.8975) weight_decay: 0.0500 (0.0500) time: 0.7261 data: 0.0004 max mem: 64948 Epoch: [331] [ 30/312] eta: 0:04:42 lr: 0.000709 min_lr: 0.000709 loss: 1.9305 (1.8732) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [331] [ 40/312] eta: 0:04:12 lr: 0.000709 min_lr: 0.000709 loss: 1.8314 (1.8470) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [331] [ 50/312] eta: 0:03:51 lr: 0.000708 min_lr: 0.000708 loss: 1.7578 (1.8387) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [331] [ 60/312] eta: 0:03:34 lr: 0.000708 min_lr: 0.000708 loss: 1.8873 (1.8428) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [331] [ 70/312] eta: 0:03:20 lr: 0.000708 min_lr: 0.000708 loss: 1.9032 (1.8502) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [331] [ 80/312] eta: 0:03:08 lr: 0.000707 min_lr: 0.000707 loss: 1.8944 (1.8296) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [331] [ 90/312] eta: 0:02:57 lr: 0.000707 min_lr: 0.000707 loss: 1.8391 (1.8212) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [331] [100/312] eta: 0:02:47 lr: 0.000707 min_lr: 0.000707 loss: 1.9328 (1.8399) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [331] [110/312] eta: 0:02:37 lr: 0.000706 min_lr: 0.000706 loss: 1.9297 (1.8412) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [331] [120/312] eta: 0:02:28 lr: 0.000706 min_lr: 0.000706 loss: 1.9132 (1.8515) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [331] [130/312] eta: 0:02:19 lr: 0.000706 min_lr: 0.000706 loss: 1.9080 (1.8576) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [331] [140/312] eta: 0:02:11 lr: 0.000705 min_lr: 0.000705 loss: 1.8905 (1.8596) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [331] [150/312] eta: 0:02:02 lr: 0.000705 min_lr: 0.000705 loss: 1.9564 (1.8661) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [331] [160/312] eta: 0:01:54 lr: 0.000705 min_lr: 0.000705 loss: 2.0308 (1.8774) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [331] [170/312] eta: 0:01:46 lr: 0.000704 min_lr: 0.000704 loss: 1.9846 (1.8673) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [331] [180/312] eta: 0:01:38 lr: 0.000704 min_lr: 0.000704 loss: 1.6809 (1.8572) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [331] [190/312] eta: 0:01:30 lr: 0.000703 min_lr: 0.000703 loss: 1.7195 (1.8559) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [331] [200/312] eta: 0:01:23 lr: 0.000703 min_lr: 0.000703 loss: 1.8369 (1.8543) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [331] [210/312] eta: 0:01:15 lr: 0.000703 min_lr: 0.000703 loss: 1.8652 (1.8526) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [331] [220/312] eta: 0:01:07 lr: 0.000702 min_lr: 0.000702 loss: 1.8109 (1.8499) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [331] [230/312] eta: 0:01:00 lr: 0.000702 min_lr: 0.000702 loss: 1.8730 (1.8541) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [331] [240/312] eta: 0:00:52 lr: 0.000702 min_lr: 0.000702 loss: 1.8856 (1.8530) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [331] [250/312] eta: 0:00:45 lr: 0.000701 min_lr: 0.000701 loss: 1.8801 (1.8518) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [331] [260/312] eta: 0:00:38 lr: 0.000701 min_lr: 0.000701 loss: 1.8927 (1.8536) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [331] [270/312] eta: 0:00:30 lr: 0.000701 min_lr: 0.000701 loss: 1.9335 (1.8563) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [331] [280/312] eta: 0:00:23 lr: 0.000700 min_lr: 0.000700 loss: 1.8252 (1.8466) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [331] [290/312] eta: 0:00:16 lr: 0.000700 min_lr: 0.000700 loss: 1.9446 (1.8538) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [331] [300/312] eta: 0:00:08 lr: 0.000700 min_lr: 0.000700 loss: 1.9931 (1.8579) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [331] [310/312] eta: 0:00:01 lr: 0.000699 min_lr: 0.000699 loss: 1.9715 (1.8568) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [331] [311/312] eta: 0:00:00 lr: 0.000699 min_lr: 0.000699 loss: 1.9779 (1.8573) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [331] Total time: 0:03:47 (0.7282 s / it) Averaged stats: lr: 0.000699 min_lr: 0.000699 loss: 1.9779 (1.8506) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4898 (0.4898) acc1: 86.7188 (86.7188) acc5: 97.1354 (97.1354) time: 4.6138 data: 4.4077 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6884 (0.7153) acc1: 82.0312 (80.8640) acc5: 96.0938 (95.7440) time: 0.6641 data: 0.4898 max mem: 64948 Test: Total time: 0:00:06 (0.6904 s / it) * Acc@1 81.982 Acc@5 95.882 loss 0.691 Accuracy of the model on the 50000 test images: 82.0% Max accuracy: 82.09% Test: [0/9] eta: 0:00:47 loss: 0.4623 (0.4623) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 5.2762 data: 5.0634 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6365 (0.6523) acc1: 83.0729 (81.4720) acc5: 97.1354 (96.7040) time: 0.7375 data: 0.5627 max mem: 64948 Test: Total time: 0:00:06 (0.7445 s / it) * Acc@1 82.992 Acc@5 96.482 loss 0.629 Accuracy of the model EMA on 50000 test images: 83.0% Epoch: [332] [ 0/312] eta: 1:00:41 lr: 0.000699 min_lr: 0.000699 loss: 1.6396 (1.6396) weight_decay: 0.0500 (0.0500) time: 11.6707 data: 7.9112 max mem: 64948 Epoch: [332] [ 10/312] eta: 0:08:45 lr: 0.000699 min_lr: 0.000699 loss: 1.7584 (1.7591) weight_decay: 0.0500 (0.0500) time: 1.7390 data: 0.7197 max mem: 64948 Epoch: [332] [ 20/312] eta: 0:06:02 lr: 0.000698 min_lr: 0.000698 loss: 1.7584 (1.7361) weight_decay: 0.0500 (0.0500) time: 0.7209 data: 0.0005 max mem: 64948 Epoch: [332] [ 30/312] eta: 0:05:00 lr: 0.000698 min_lr: 0.000698 loss: 1.7319 (1.7582) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [332] [ 40/312] eta: 0:04:25 lr: 0.000698 min_lr: 0.000698 loss: 1.9269 (1.7783) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [332] [ 50/312] eta: 0:04:01 lr: 0.000697 min_lr: 0.000697 loss: 1.7728 (1.7681) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [332] [ 60/312] eta: 0:03:42 lr: 0.000697 min_lr: 0.000697 loss: 1.8295 (1.7772) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [332] [ 70/312] eta: 0:03:27 lr: 0.000697 min_lr: 0.000697 loss: 1.9326 (1.8048) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [332] [ 80/312] eta: 0:03:14 lr: 0.000696 min_lr: 0.000696 loss: 1.8261 (1.7923) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [332] [ 90/312] eta: 0:03:02 lr: 0.000696 min_lr: 0.000696 loss: 1.7975 (1.8032) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [332] [100/312] eta: 0:02:51 lr: 0.000696 min_lr: 0.000696 loss: 1.9491 (1.7996) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [332] [110/312] eta: 0:02:41 lr: 0.000695 min_lr: 0.000695 loss: 1.8847 (1.8091) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [332] [120/312] eta: 0:02:31 lr: 0.000695 min_lr: 0.000695 loss: 1.8847 (1.8093) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [332] [130/312] eta: 0:02:22 lr: 0.000695 min_lr: 0.000695 loss: 1.9246 (1.8149) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [332] [140/312] eta: 0:02:13 lr: 0.000694 min_lr: 0.000694 loss: 1.8575 (1.8126) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [332] [150/312] eta: 0:02:05 lr: 0.000694 min_lr: 0.000694 loss: 1.7963 (1.8118) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [332] [160/312] eta: 0:01:56 lr: 0.000693 min_lr: 0.000693 loss: 1.7246 (1.7996) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [332] [170/312] eta: 0:01:48 lr: 0.000693 min_lr: 0.000693 loss: 1.7246 (1.8009) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [332] [180/312] eta: 0:01:40 lr: 0.000693 min_lr: 0.000693 loss: 1.8430 (1.7940) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [332] [190/312] eta: 0:01:32 lr: 0.000692 min_lr: 0.000692 loss: 1.8577 (1.7977) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [332] [200/312] eta: 0:01:24 lr: 0.000692 min_lr: 0.000692 loss: 1.9319 (1.7959) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [332] [210/312] eta: 0:01:16 lr: 0.000692 min_lr: 0.000692 loss: 1.8344 (1.7989) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [332] [220/312] eta: 0:01:08 lr: 0.000691 min_lr: 0.000691 loss: 1.8746 (1.7961) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [332] [230/312] eta: 0:01:01 lr: 0.000691 min_lr: 0.000691 loss: 1.9733 (1.8059) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [332] [240/312] eta: 0:00:53 lr: 0.000691 min_lr: 0.000691 loss: 2.0456 (1.8058) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [332] [250/312] eta: 0:00:45 lr: 0.000690 min_lr: 0.000690 loss: 1.7829 (1.8048) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [332] [260/312] eta: 0:00:38 lr: 0.000690 min_lr: 0.000690 loss: 1.6335 (1.8016) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [332] [270/312] eta: 0:00:30 lr: 0.000690 min_lr: 0.000690 loss: 1.7795 (1.8060) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [332] [280/312] eta: 0:00:23 lr: 0.000689 min_lr: 0.000689 loss: 1.9039 (1.8087) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0009 max mem: 64948 Epoch: [332] [290/312] eta: 0:00:16 lr: 0.000689 min_lr: 0.000689 loss: 1.8466 (1.8089) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [332] [300/312] eta: 0:00:08 lr: 0.000689 min_lr: 0.000689 loss: 1.8466 (1.8146) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [332] [310/312] eta: 0:00:01 lr: 0.000688 min_lr: 0.000688 loss: 1.9636 (1.8193) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [332] [311/312] eta: 0:00:00 lr: 0.000688 min_lr: 0.000688 loss: 1.9013 (1.8196) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [332] Total time: 0:03:49 (0.7350 s / it) Averaged stats: lr: 0.000688 min_lr: 0.000688 loss: 1.9013 (1.8364) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.4843 (0.4843) acc1: 87.7604 (87.7604) acc5: 96.8750 (96.8750) time: 4.8907 data: 4.6678 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7258 (0.7021) acc1: 82.5521 (81.6320) acc5: 96.3542 (96.0320) time: 0.6947 data: 0.5187 max mem: 64948 Test: Total time: 0:00:06 (0.7223 s / it) * Acc@1 82.062 Acc@5 95.978 loss 0.681 Accuracy of the model on the 50000 test images: 82.1% Max accuracy: 82.09% Test: [0/9] eta: 0:00:48 loss: 0.4624 (0.4624) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 5.3944 data: 5.1800 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6365 (0.6523) acc1: 83.0729 (81.5040) acc5: 97.1354 (96.6720) time: 0.7507 data: 0.5757 max mem: 64948 Test: Total time: 0:00:06 (0.7667 s / it) * Acc@1 83.012 Acc@5 96.476 loss 0.629 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 83.01% Epoch: [333] [ 0/312] eta: 0:48:12 lr: 0.000688 min_lr: 0.000688 loss: 1.4511 (1.4511) weight_decay: 0.0500 (0.0500) time: 9.2719 data: 8.3583 max mem: 64948 Epoch: [333] [ 10/312] eta: 0:07:34 lr: 0.000688 min_lr: 0.000688 loss: 1.9359 (1.8287) weight_decay: 0.0500 (0.0500) time: 1.5038 data: 0.7603 max mem: 64948 Epoch: [333] [ 20/312] eta: 0:05:26 lr: 0.000687 min_lr: 0.000687 loss: 1.9359 (1.8546) weight_decay: 0.0500 (0.0500) time: 0.7111 data: 0.0005 max mem: 64948 Epoch: [333] [ 30/312] eta: 0:04:36 lr: 0.000687 min_lr: 0.000687 loss: 1.9221 (1.8555) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [333] [ 40/312] eta: 0:04:08 lr: 0.000687 min_lr: 0.000687 loss: 1.9812 (1.8955) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0003 max mem: 64948 Epoch: [333] [ 50/312] eta: 0:03:48 lr: 0.000686 min_lr: 0.000686 loss: 2.0034 (1.9016) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [333] [ 60/312] eta: 0:03:32 lr: 0.000686 min_lr: 0.000686 loss: 1.9126 (1.8923) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [333] [ 70/312] eta: 0:03:18 lr: 0.000686 min_lr: 0.000686 loss: 1.8475 (1.8577) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [333] [ 80/312] eta: 0:03:07 lr: 0.000685 min_lr: 0.000685 loss: 1.8658 (1.8677) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [333] [ 90/312] eta: 0:02:56 lr: 0.000685 min_lr: 0.000685 loss: 1.9713 (1.8667) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [333] [100/312] eta: 0:02:46 lr: 0.000685 min_lr: 0.000685 loss: 1.9187 (1.8534) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [333] [110/312] eta: 0:02:36 lr: 0.000684 min_lr: 0.000684 loss: 1.7524 (1.8486) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [333] [120/312] eta: 0:02:27 lr: 0.000684 min_lr: 0.000684 loss: 1.7764 (1.8434) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [333] [130/312] eta: 0:02:19 lr: 0.000684 min_lr: 0.000684 loss: 1.8317 (1.8395) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [333] [140/312] eta: 0:02:10 lr: 0.000683 min_lr: 0.000683 loss: 1.8317 (1.8398) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [333] [150/312] eta: 0:02:02 lr: 0.000683 min_lr: 0.000683 loss: 1.9621 (1.8447) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [333] [160/312] eta: 0:01:54 lr: 0.000682 min_lr: 0.000682 loss: 1.9868 (1.8489) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [333] [170/312] eta: 0:01:46 lr: 0.000682 min_lr: 0.000682 loss: 2.0526 (1.8545) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [333] [180/312] eta: 0:01:38 lr: 0.000682 min_lr: 0.000682 loss: 1.9483 (1.8494) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [333] [190/312] eta: 0:01:30 lr: 0.000681 min_lr: 0.000681 loss: 1.6130 (1.8435) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [333] [200/312] eta: 0:01:22 lr: 0.000681 min_lr: 0.000681 loss: 1.7121 (1.8399) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [333] [210/312] eta: 0:01:15 lr: 0.000681 min_lr: 0.000681 loss: 1.9124 (1.8390) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [333] [220/312] eta: 0:01:07 lr: 0.000680 min_lr: 0.000680 loss: 1.8533 (1.8353) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [333] [230/312] eta: 0:01:00 lr: 0.000680 min_lr: 0.000680 loss: 1.8909 (1.8386) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [333] [240/312] eta: 0:00:52 lr: 0.000680 min_lr: 0.000680 loss: 1.9641 (1.8370) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [333] [250/312] eta: 0:00:45 lr: 0.000679 min_lr: 0.000679 loss: 1.8842 (1.8333) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [333] [260/312] eta: 0:00:37 lr: 0.000679 min_lr: 0.000679 loss: 1.6735 (1.8273) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [333] [270/312] eta: 0:00:30 lr: 0.000679 min_lr: 0.000679 loss: 1.7552 (1.8293) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [333] [280/312] eta: 0:00:23 lr: 0.000678 min_lr: 0.000678 loss: 1.9039 (1.8315) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [333] [290/312] eta: 0:00:15 lr: 0.000678 min_lr: 0.000678 loss: 1.9284 (1.8355) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [333] [300/312] eta: 0:00:08 lr: 0.000678 min_lr: 0.000678 loss: 1.8186 (1.8329) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [333] [310/312] eta: 0:00:01 lr: 0.000677 min_lr: 0.000677 loss: 1.7484 (1.8332) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [333] [311/312] eta: 0:00:00 lr: 0.000677 min_lr: 0.000677 loss: 1.7604 (1.8330) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [333] Total time: 0:03:46 (0.7266 s / it) Averaged stats: lr: 0.000677 min_lr: 0.000677 loss: 1.7604 (1.8337) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4881 (0.4881) acc1: 87.2396 (87.2396) acc5: 98.1771 (98.1771) time: 4.8180 data: 4.6075 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7255 (0.6951) acc1: 81.7708 (81.2800) acc5: 96.8750 (96.0960) time: 0.6866 data: 0.5120 max mem: 64948 Test: Total time: 0:00:06 (0.7131 s / it) * Acc@1 82.056 Acc@5 96.012 loss 0.679 Accuracy of the model on the 50000 test images: 82.1% Max accuracy: 82.09% Test: [0/9] eta: 0:00:46 loss: 0.4622 (0.4622) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 5.1995 data: 4.9783 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6363 (0.6521) acc1: 83.3333 (81.5360) acc5: 97.3958 (96.7040) time: 0.7298 data: 0.5532 max mem: 64948 Test: Total time: 0:00:06 (0.7553 s / it) * Acc@1 83.020 Acc@5 96.492 loss 0.629 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 83.02% Epoch: [334] [ 0/312] eta: 0:47:08 lr: 0.000677 min_lr: 0.000677 loss: 1.2789 (1.2789) weight_decay: 0.0500 (0.0500) time: 9.0662 data: 8.0026 max mem: 64948 Epoch: [334] [ 10/312] eta: 0:07:38 lr: 0.000677 min_lr: 0.000677 loss: 1.9675 (1.8481) weight_decay: 0.0500 (0.0500) time: 1.5170 data: 0.7644 max mem: 64948 Epoch: [334] [ 20/312] eta: 0:05:28 lr: 0.000676 min_lr: 0.000676 loss: 1.9675 (1.8468) weight_decay: 0.0500 (0.0500) time: 0.7293 data: 0.0205 max mem: 64948 Epoch: [334] [ 30/312] eta: 0:04:39 lr: 0.000676 min_lr: 0.000676 loss: 1.7124 (1.8643) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0003 max mem: 64948 Epoch: [334] [ 40/312] eta: 0:04:09 lr: 0.000676 min_lr: 0.000676 loss: 1.8115 (1.8391) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0004 max mem: 64948 Epoch: [334] [ 50/312] eta: 0:03:49 lr: 0.000675 min_lr: 0.000675 loss: 1.8115 (1.8381) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [334] [ 60/312] eta: 0:03:33 lr: 0.000675 min_lr: 0.000675 loss: 1.8105 (1.8195) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [334] [ 70/312] eta: 0:03:19 lr: 0.000675 min_lr: 0.000675 loss: 1.6746 (1.8095) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [334] [ 80/312] eta: 0:03:07 lr: 0.000674 min_lr: 0.000674 loss: 1.6962 (1.8077) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [334] [ 90/312] eta: 0:02:56 lr: 0.000674 min_lr: 0.000674 loss: 1.7509 (1.8023) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [334] [100/312] eta: 0:02:46 lr: 0.000674 min_lr: 0.000674 loss: 1.7509 (1.7981) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [334] [110/312] eta: 0:02:37 lr: 0.000673 min_lr: 0.000673 loss: 1.8219 (1.8034) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [334] [120/312] eta: 0:02:27 lr: 0.000673 min_lr: 0.000673 loss: 1.8643 (1.8045) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [334] [130/312] eta: 0:02:19 lr: 0.000673 min_lr: 0.000673 loss: 1.8199 (1.8009) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [334] [140/312] eta: 0:02:10 lr: 0.000672 min_lr: 0.000672 loss: 1.8199 (1.8167) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [334] [150/312] eta: 0:02:02 lr: 0.000672 min_lr: 0.000672 loss: 1.7743 (1.8076) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [334] [160/312] eta: 0:01:54 lr: 0.000672 min_lr: 0.000672 loss: 1.6551 (1.8022) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [334] [170/312] eta: 0:01:46 lr: 0.000671 min_lr: 0.000671 loss: 1.9067 (1.8068) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [334] [180/312] eta: 0:01:38 lr: 0.000671 min_lr: 0.000671 loss: 1.9067 (1.8090) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [334] [190/312] eta: 0:01:30 lr: 0.000670 min_lr: 0.000670 loss: 1.9747 (1.8177) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [334] [200/312] eta: 0:01:22 lr: 0.000670 min_lr: 0.000670 loss: 2.0318 (1.8214) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [334] [210/312] eta: 0:01:15 lr: 0.000670 min_lr: 0.000670 loss: 1.8937 (1.8168) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0003 max mem: 64948 Epoch: [334] [220/312] eta: 0:01:07 lr: 0.000669 min_lr: 0.000669 loss: 1.5994 (1.8094) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [334] [230/312] eta: 0:01:00 lr: 0.000669 min_lr: 0.000669 loss: 1.8654 (1.8125) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [334] [240/312] eta: 0:00:52 lr: 0.000669 min_lr: 0.000669 loss: 2.0266 (1.8208) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [334] [250/312] eta: 0:00:45 lr: 0.000668 min_lr: 0.000668 loss: 2.0181 (1.8158) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [334] [260/312] eta: 0:00:37 lr: 0.000668 min_lr: 0.000668 loss: 1.8410 (1.8223) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [334] [270/312] eta: 0:00:30 lr: 0.000668 min_lr: 0.000668 loss: 1.9640 (1.8239) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [334] [280/312] eta: 0:00:23 lr: 0.000667 min_lr: 0.000667 loss: 1.9460 (1.8263) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0010 max mem: 64948 Epoch: [334] [290/312] eta: 0:00:15 lr: 0.000667 min_lr: 0.000667 loss: 1.9460 (1.8267) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0009 max mem: 64948 Epoch: [334] [300/312] eta: 0:00:08 lr: 0.000667 min_lr: 0.000667 loss: 1.8441 (1.8255) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [334] [310/312] eta: 0:00:01 lr: 0.000666 min_lr: 0.000666 loss: 1.8441 (1.8218) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [334] [311/312] eta: 0:00:00 lr: 0.000666 min_lr: 0.000666 loss: 1.9288 (1.8231) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [334] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.000666 min_lr: 0.000666 loss: 1.9288 (1.8324) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4538 (0.4538) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.7061 data: 4.4897 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7341 (0.6945) acc1: 82.8125 (81.8560) acc5: 96.6146 (96.2560) time: 0.6742 data: 0.4989 max mem: 64948 Test: Total time: 0:00:06 (0.7023 s / it) * Acc@1 82.212 Acc@5 96.030 loss 0.676 Accuracy of the model on the 50000 test images: 82.2% Max accuracy: 82.21% Test: [0/9] eta: 0:00:44 loss: 0.4624 (0.4624) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.9057 data: 4.6821 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6365 (0.6519) acc1: 83.5938 (81.6320) acc5: 97.3958 (96.6720) time: 0.6964 data: 0.5203 max mem: 64948 Test: Total time: 0:00:06 (0.7042 s / it) * Acc@1 83.020 Acc@5 96.488 loss 0.629 Accuracy of the model EMA on 50000 test images: 83.0% Epoch: [335] [ 0/312] eta: 0:53:21 lr: 0.000666 min_lr: 0.000666 loss: 2.0687 (2.0687) weight_decay: 0.0500 (0.0500) time: 10.2626 data: 6.4459 max mem: 64948 Epoch: [335] [ 10/312] eta: 0:08:11 lr: 0.000666 min_lr: 0.000666 loss: 1.8202 (1.7738) weight_decay: 0.0500 (0.0500) time: 1.6288 data: 0.5865 max mem: 64948 Epoch: [335] [ 20/312] eta: 0:05:45 lr: 0.000665 min_lr: 0.000665 loss: 1.7999 (1.8037) weight_decay: 0.0500 (0.0500) time: 0.7305 data: 0.0005 max mem: 64948 Epoch: [335] [ 30/312] eta: 0:04:49 lr: 0.000665 min_lr: 0.000665 loss: 1.8023 (1.8006) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [335] [ 40/312] eta: 0:04:17 lr: 0.000665 min_lr: 0.000665 loss: 1.7550 (1.7671) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [335] [ 50/312] eta: 0:03:54 lr: 0.000664 min_lr: 0.000664 loss: 1.7101 (1.7749) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [335] [ 60/312] eta: 0:03:37 lr: 0.000664 min_lr: 0.000664 loss: 1.8241 (1.7788) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [335] [ 70/312] eta: 0:03:23 lr: 0.000664 min_lr: 0.000664 loss: 1.9192 (1.7881) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [335] [ 80/312] eta: 0:03:11 lr: 0.000663 min_lr: 0.000663 loss: 1.8576 (1.7809) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [335] [ 90/312] eta: 0:02:59 lr: 0.000663 min_lr: 0.000663 loss: 1.8576 (1.7833) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [335] [100/312] eta: 0:02:49 lr: 0.000663 min_lr: 0.000663 loss: 1.8514 (1.7924) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [335] [110/312] eta: 0:02:39 lr: 0.000662 min_lr: 0.000662 loss: 1.7455 (1.7873) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [335] [120/312] eta: 0:02:30 lr: 0.000662 min_lr: 0.000662 loss: 1.8302 (1.7838) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [335] [130/312] eta: 0:02:20 lr: 0.000662 min_lr: 0.000662 loss: 1.5611 (1.7705) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [335] [140/312] eta: 0:02:12 lr: 0.000661 min_lr: 0.000661 loss: 1.6520 (1.7702) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [335] [150/312] eta: 0:02:03 lr: 0.000661 min_lr: 0.000661 loss: 1.7575 (1.7732) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [335] [160/312] eta: 0:01:55 lr: 0.000661 min_lr: 0.000661 loss: 1.9343 (1.7767) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [335] [170/312] eta: 0:01:47 lr: 0.000660 min_lr: 0.000660 loss: 1.8507 (1.7772) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [335] [180/312] eta: 0:01:39 lr: 0.000660 min_lr: 0.000660 loss: 2.0042 (1.7894) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [335] [190/312] eta: 0:01:31 lr: 0.000660 min_lr: 0.000660 loss: 2.0190 (1.8002) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [335] [200/312] eta: 0:01:23 lr: 0.000659 min_lr: 0.000659 loss: 1.9086 (1.8021) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [335] [210/312] eta: 0:01:15 lr: 0.000659 min_lr: 0.000659 loss: 1.6641 (1.8011) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [335] [220/312] eta: 0:01:08 lr: 0.000659 min_lr: 0.000659 loss: 1.7786 (1.8062) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [335] [230/312] eta: 0:01:00 lr: 0.000658 min_lr: 0.000658 loss: 1.8963 (1.8055) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [335] [240/312] eta: 0:00:53 lr: 0.000658 min_lr: 0.000658 loss: 1.8802 (1.8043) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [335] [250/312] eta: 0:00:45 lr: 0.000658 min_lr: 0.000658 loss: 1.8802 (1.8095) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [335] [260/312] eta: 0:00:38 lr: 0.000657 min_lr: 0.000657 loss: 1.8362 (1.8095) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [335] [270/312] eta: 0:00:30 lr: 0.000657 min_lr: 0.000657 loss: 1.8776 (1.8119) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [335] [280/312] eta: 0:00:23 lr: 0.000656 min_lr: 0.000656 loss: 1.9205 (1.8165) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [335] [290/312] eta: 0:00:16 lr: 0.000656 min_lr: 0.000656 loss: 1.9057 (1.8164) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [335] [300/312] eta: 0:00:08 lr: 0.000656 min_lr: 0.000656 loss: 1.9563 (1.8203) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [335] [310/312] eta: 0:00:01 lr: 0.000655 min_lr: 0.000655 loss: 1.8072 (1.8131) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [335] [311/312] eta: 0:00:00 lr: 0.000655 min_lr: 0.000655 loss: 1.8072 (1.8139) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [335] Total time: 0:03:48 (0.7312 s / it) Averaged stats: lr: 0.000655 min_lr: 0.000655 loss: 1.8072 (1.8317) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4778 (0.4778) acc1: 86.9792 (86.9792) acc5: 98.1771 (98.1771) time: 4.7501 data: 4.5304 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7271 (0.7157) acc1: 82.2917 (81.5040) acc5: 96.3542 (95.9040) time: 0.6798 data: 0.5035 max mem: 64948 Test: Total time: 0:00:06 (0.7071 s / it) * Acc@1 82.122 Acc@5 95.932 loss 0.679 Accuracy of the model on the 50000 test images: 82.1% Max accuracy: 82.21% Test: [0/9] eta: 0:00:49 loss: 0.4627 (0.4627) acc1: 87.7604 (87.7604) acc5: 98.4375 (98.4375) time: 5.4888 data: 5.2838 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6370 (0.6518) acc1: 83.5938 (81.7280) acc5: 97.3958 (96.8000) time: 0.7611 data: 0.5872 max mem: 64948 Test: Total time: 0:00:06 (0.7710 s / it) * Acc@1 83.028 Acc@5 96.498 loss 0.628 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 83.03% Epoch: [336] [ 0/312] eta: 0:45:15 lr: 0.000655 min_lr: 0.000655 loss: 2.0112 (2.0112) weight_decay: 0.0500 (0.0500) time: 8.7047 data: 7.5080 max mem: 64948 Epoch: [336] [ 10/312] eta: 0:07:31 lr: 0.000655 min_lr: 0.000655 loss: 1.7433 (1.7149) weight_decay: 0.0500 (0.0500) time: 1.4953 data: 0.6856 max mem: 64948 Epoch: [336] [ 20/312] eta: 0:05:25 lr: 0.000655 min_lr: 0.000655 loss: 1.7878 (1.7915) weight_decay: 0.0500 (0.0500) time: 0.7337 data: 0.0019 max mem: 64948 Epoch: [336] [ 30/312] eta: 0:04:35 lr: 0.000654 min_lr: 0.000654 loss: 1.7878 (1.7367) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [336] [ 40/312] eta: 0:04:07 lr: 0.000654 min_lr: 0.000654 loss: 1.7533 (1.7417) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [336] [ 50/312] eta: 0:03:47 lr: 0.000654 min_lr: 0.000654 loss: 1.8497 (1.7364) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [336] [ 60/312] eta: 0:03:31 lr: 0.000653 min_lr: 0.000653 loss: 1.8837 (1.7648) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [336] [ 70/312] eta: 0:03:18 lr: 0.000653 min_lr: 0.000653 loss: 1.9351 (1.7876) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [336] [ 80/312] eta: 0:03:06 lr: 0.000653 min_lr: 0.000653 loss: 1.9259 (1.7807) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [336] [ 90/312] eta: 0:02:55 lr: 0.000652 min_lr: 0.000652 loss: 1.8181 (1.7799) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [336] [100/312] eta: 0:02:45 lr: 0.000652 min_lr: 0.000652 loss: 1.8608 (1.7889) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [336] [110/312] eta: 0:02:36 lr: 0.000652 min_lr: 0.000652 loss: 1.8624 (1.7927) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [336] [120/312] eta: 0:02:27 lr: 0.000651 min_lr: 0.000651 loss: 1.8624 (1.7866) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [336] [130/312] eta: 0:02:18 lr: 0.000651 min_lr: 0.000651 loss: 1.8669 (1.7889) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [336] [140/312] eta: 0:02:10 lr: 0.000651 min_lr: 0.000651 loss: 1.8941 (1.7859) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [336] [150/312] eta: 0:02:02 lr: 0.000650 min_lr: 0.000650 loss: 2.0015 (1.8038) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [336] [160/312] eta: 0:01:54 lr: 0.000650 min_lr: 0.000650 loss: 1.9262 (1.7999) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [336] [170/312] eta: 0:01:46 lr: 0.000649 min_lr: 0.000649 loss: 1.9054 (1.8146) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [336] [180/312] eta: 0:01:38 lr: 0.000649 min_lr: 0.000649 loss: 1.9978 (1.8133) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [336] [190/312] eta: 0:01:30 lr: 0.000649 min_lr: 0.000649 loss: 1.8262 (1.8123) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [336] [200/312] eta: 0:01:22 lr: 0.000648 min_lr: 0.000648 loss: 1.8407 (1.8140) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [336] [210/312] eta: 0:01:15 lr: 0.000648 min_lr: 0.000648 loss: 1.8421 (1.8148) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [336] [220/312] eta: 0:01:07 lr: 0.000648 min_lr: 0.000648 loss: 1.8367 (1.8140) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [336] [230/312] eta: 0:01:00 lr: 0.000647 min_lr: 0.000647 loss: 1.8303 (1.8169) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [336] [240/312] eta: 0:00:52 lr: 0.000647 min_lr: 0.000647 loss: 1.8216 (1.8182) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [336] [250/312] eta: 0:00:45 lr: 0.000647 min_lr: 0.000647 loss: 1.7603 (1.8128) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [336] [260/312] eta: 0:00:37 lr: 0.000646 min_lr: 0.000646 loss: 1.6945 (1.8102) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [336] [270/312] eta: 0:00:30 lr: 0.000646 min_lr: 0.000646 loss: 1.9071 (1.8152) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [336] [280/312] eta: 0:00:23 lr: 0.000646 min_lr: 0.000646 loss: 1.9706 (1.8163) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0006 max mem: 64948 Epoch: [336] [290/312] eta: 0:00:15 lr: 0.000645 min_lr: 0.000645 loss: 1.8633 (1.8180) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0005 max mem: 64948 Epoch: [336] [300/312] eta: 0:00:08 lr: 0.000645 min_lr: 0.000645 loss: 1.9506 (1.8218) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [336] [310/312] eta: 0:00:01 lr: 0.000645 min_lr: 0.000645 loss: 1.9287 (1.8231) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [336] [311/312] eta: 0:00:00 lr: 0.000645 min_lr: 0.000645 loss: 1.9506 (1.8236) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [336] Total time: 0:03:46 (0.7261 s / it) Averaged stats: lr: 0.000645 min_lr: 0.000645 loss: 1.9506 (1.8339) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.5288 (0.5288) acc1: 85.4167 (85.4167) acc5: 97.3958 (97.3958) time: 4.8419 data: 4.6176 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7209 (0.6891) acc1: 82.8125 (81.3760) acc5: 96.8750 (96.0320) time: 0.6900 data: 0.5132 max mem: 64948 Test: Total time: 0:00:06 (0.7199 s / it) * Acc@1 82.178 Acc@5 95.952 loss 0.676 Accuracy of the model on the 50000 test images: 82.2% Max accuracy: 82.21% Test: [0/9] eta: 0:00:46 loss: 0.4630 (0.4630) acc1: 87.7604 (87.7604) acc5: 98.4375 (98.4375) time: 5.1719 data: 4.9538 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6373 (0.6516) acc1: 83.5938 (81.7600) acc5: 97.3958 (96.8000) time: 0.7262 data: 0.5505 max mem: 64948 Test: Total time: 0:00:06 (0.7487 s / it) * Acc@1 83.022 Acc@5 96.490 loss 0.628 Accuracy of the model EMA on 50000 test images: 83.0% Epoch: [337] [ 0/312] eta: 0:55:26 lr: 0.000645 min_lr: 0.000645 loss: 1.9477 (1.9477) weight_decay: 0.0500 (0.0500) time: 10.6631 data: 9.5525 max mem: 64948 Epoch: [337] [ 10/312] eta: 0:08:20 lr: 0.000644 min_lr: 0.000644 loss: 1.8773 (1.7617) weight_decay: 0.0500 (0.0500) time: 1.6574 data: 0.8688 max mem: 64948 Epoch: [337] [ 20/312] eta: 0:05:50 lr: 0.000644 min_lr: 0.000644 loss: 1.8773 (1.7583) weight_decay: 0.0500 (0.0500) time: 0.7257 data: 0.0004 max mem: 64948 Epoch: [337] [ 30/312] eta: 0:04:52 lr: 0.000644 min_lr: 0.000644 loss: 1.8843 (1.7973) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [337] [ 40/312] eta: 0:04:19 lr: 0.000643 min_lr: 0.000643 loss: 1.7778 (1.7874) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [337] [ 50/312] eta: 0:03:56 lr: 0.000643 min_lr: 0.000643 loss: 1.7176 (1.7930) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [337] [ 60/312] eta: 0:03:39 lr: 0.000643 min_lr: 0.000643 loss: 1.7114 (1.7909) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [337] [ 70/312] eta: 0:03:24 lr: 0.000642 min_lr: 0.000642 loss: 1.9539 (1.8208) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [337] [ 80/312] eta: 0:03:11 lr: 0.000642 min_lr: 0.000642 loss: 1.9591 (1.8194) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [337] [ 90/312] eta: 0:03:00 lr: 0.000641 min_lr: 0.000641 loss: 1.9162 (1.8210) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [337] [100/312] eta: 0:02:49 lr: 0.000641 min_lr: 0.000641 loss: 1.8906 (1.8191) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [337] [110/312] eta: 0:02:39 lr: 0.000641 min_lr: 0.000641 loss: 1.7800 (1.8078) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [337] [120/312] eta: 0:02:30 lr: 0.000640 min_lr: 0.000640 loss: 1.7800 (1.8097) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [337] [130/312] eta: 0:02:21 lr: 0.000640 min_lr: 0.000640 loss: 1.8019 (1.8124) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [337] [140/312] eta: 0:02:12 lr: 0.000640 min_lr: 0.000640 loss: 1.9050 (1.8127) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [337] [150/312] eta: 0:02:04 lr: 0.000639 min_lr: 0.000639 loss: 1.8083 (1.8011) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [337] [160/312] eta: 0:01:55 lr: 0.000639 min_lr: 0.000639 loss: 1.6823 (1.7980) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [337] [170/312] eta: 0:01:47 lr: 0.000639 min_lr: 0.000639 loss: 1.6901 (1.7873) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [337] [180/312] eta: 0:01:39 lr: 0.000638 min_lr: 0.000638 loss: 1.7676 (1.7933) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0003 max mem: 64948 Epoch: [337] [190/312] eta: 0:01:31 lr: 0.000638 min_lr: 0.000638 loss: 1.8312 (1.7912) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [337] [200/312] eta: 0:01:23 lr: 0.000638 min_lr: 0.000638 loss: 1.8902 (1.7960) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [337] [210/312] eta: 0:01:16 lr: 0.000637 min_lr: 0.000637 loss: 1.9508 (1.8004) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [337] [220/312] eta: 0:01:08 lr: 0.000637 min_lr: 0.000637 loss: 1.8346 (1.7939) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [337] [230/312] eta: 0:01:00 lr: 0.000637 min_lr: 0.000637 loss: 1.8869 (1.7978) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [337] [240/312] eta: 0:00:53 lr: 0.000636 min_lr: 0.000636 loss: 1.8624 (1.7975) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [337] [250/312] eta: 0:00:45 lr: 0.000636 min_lr: 0.000636 loss: 1.8624 (1.8068) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [337] [260/312] eta: 0:00:38 lr: 0.000636 min_lr: 0.000636 loss: 1.8422 (1.8051) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [337] [270/312] eta: 0:00:30 lr: 0.000635 min_lr: 0.000635 loss: 1.8422 (1.8078) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [337] [280/312] eta: 0:00:23 lr: 0.000635 min_lr: 0.000635 loss: 1.9466 (1.8046) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0010 max mem: 64948 Epoch: [337] [290/312] eta: 0:00:16 lr: 0.000635 min_lr: 0.000635 loss: 1.8201 (1.8057) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [337] [300/312] eta: 0:00:08 lr: 0.000634 min_lr: 0.000634 loss: 1.8963 (1.8112) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [337] [310/312] eta: 0:00:01 lr: 0.000634 min_lr: 0.000634 loss: 1.8204 (1.8104) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [337] [311/312] eta: 0:00:00 lr: 0.000634 min_lr: 0.000634 loss: 1.8025 (1.8101) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [337] Total time: 0:03:48 (0.7322 s / it) Averaged stats: lr: 0.000634 min_lr: 0.000634 loss: 1.8025 (1.8262) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4873 (0.4873) acc1: 87.2396 (87.2396) acc5: 98.1771 (98.1771) time: 4.7349 data: 4.5241 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6851 (0.6950) acc1: 84.6354 (81.6320) acc5: 96.3542 (96.0640) time: 0.6774 data: 0.5028 max mem: 64948 Test: Total time: 0:00:06 (0.7057 s / it) * Acc@1 82.326 Acc@5 96.048 loss 0.680 Accuracy of the model on the 50000 test images: 82.3% Max accuracy: 82.33% Test: [0/9] eta: 0:00:42 loss: 0.4631 (0.4631) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.7359 data: 4.5340 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6373 (0.6514) acc1: 83.5938 (81.8240) acc5: 97.3958 (96.7680) time: 0.6775 data: 0.5039 max mem: 64948 Test: Total time: 0:00:06 (0.6857 s / it) * Acc@1 83.036 Acc@5 96.492 loss 0.628 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 83.04% Epoch: [338] [ 0/312] eta: 0:55:53 lr: 0.000634 min_lr: 0.000634 loss: 2.0047 (2.0047) weight_decay: 0.0500 (0.0500) time: 10.7494 data: 10.0246 max mem: 64948 Epoch: [338] [ 10/312] eta: 0:08:09 lr: 0.000634 min_lr: 0.000634 loss: 2.0907 (1.9363) weight_decay: 0.0500 (0.0500) time: 1.6212 data: 0.9117 max mem: 64948 Epoch: [338] [ 20/312] eta: 0:05:46 lr: 0.000633 min_lr: 0.000633 loss: 1.9014 (1.8324) weight_decay: 0.0500 (0.0500) time: 0.7079 data: 0.0004 max mem: 64948 Epoch: [338] [ 30/312] eta: 0:04:49 lr: 0.000633 min_lr: 0.000633 loss: 1.9017 (1.8676) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [338] [ 40/312] eta: 0:04:17 lr: 0.000633 min_lr: 0.000633 loss: 1.9017 (1.8178) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [338] [ 50/312] eta: 0:03:55 lr: 0.000632 min_lr: 0.000632 loss: 1.6410 (1.8137) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [338] [ 60/312] eta: 0:03:37 lr: 0.000632 min_lr: 0.000632 loss: 1.8990 (1.8529) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [338] [ 70/312] eta: 0:03:23 lr: 0.000631 min_lr: 0.000631 loss: 1.9787 (1.8577) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [338] [ 80/312] eta: 0:03:10 lr: 0.000631 min_lr: 0.000631 loss: 1.9119 (1.8558) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [338] [ 90/312] eta: 0:02:59 lr: 0.000631 min_lr: 0.000631 loss: 1.9119 (1.8591) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [338] [100/312] eta: 0:02:48 lr: 0.000630 min_lr: 0.000630 loss: 1.9277 (1.8722) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [338] [110/312] eta: 0:02:39 lr: 0.000630 min_lr: 0.000630 loss: 1.9526 (1.8690) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [338] [120/312] eta: 0:02:29 lr: 0.000630 min_lr: 0.000630 loss: 1.9961 (1.8639) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [338] [130/312] eta: 0:02:20 lr: 0.000629 min_lr: 0.000629 loss: 1.8121 (1.8554) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [338] [140/312] eta: 0:02:12 lr: 0.000629 min_lr: 0.000629 loss: 1.7773 (1.8502) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [338] [150/312] eta: 0:02:03 lr: 0.000629 min_lr: 0.000629 loss: 1.9489 (1.8521) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [338] [160/312] eta: 0:01:55 lr: 0.000628 min_lr: 0.000628 loss: 1.8931 (1.8471) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [338] [170/312] eta: 0:01:47 lr: 0.000628 min_lr: 0.000628 loss: 1.7974 (1.8471) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [338] [180/312] eta: 0:01:39 lr: 0.000628 min_lr: 0.000628 loss: 1.7964 (1.8349) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [338] [190/312] eta: 0:01:31 lr: 0.000627 min_lr: 0.000627 loss: 1.7420 (1.8341) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [338] [200/312] eta: 0:01:23 lr: 0.000627 min_lr: 0.000627 loss: 1.7200 (1.8273) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [338] [210/312] eta: 0:01:15 lr: 0.000627 min_lr: 0.000627 loss: 1.7200 (1.8219) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [338] [220/312] eta: 0:01:08 lr: 0.000626 min_lr: 0.000626 loss: 1.9689 (1.8248) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [338] [230/312] eta: 0:01:00 lr: 0.000626 min_lr: 0.000626 loss: 1.9392 (1.8244) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [338] [240/312] eta: 0:00:53 lr: 0.000626 min_lr: 0.000626 loss: 1.7316 (1.8201) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [338] [250/312] eta: 0:00:45 lr: 0.000625 min_lr: 0.000625 loss: 1.8440 (1.8233) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [338] [260/312] eta: 0:00:38 lr: 0.000625 min_lr: 0.000625 loss: 1.8440 (1.8237) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [338] [270/312] eta: 0:00:30 lr: 0.000625 min_lr: 0.000625 loss: 1.8397 (1.8242) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [338] [280/312] eta: 0:00:23 lr: 0.000624 min_lr: 0.000624 loss: 1.9515 (1.8235) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [338] [290/312] eta: 0:00:16 lr: 0.000624 min_lr: 0.000624 loss: 1.9515 (1.8274) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [338] [300/312] eta: 0:00:08 lr: 0.000624 min_lr: 0.000624 loss: 2.0075 (1.8344) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [338] [310/312] eta: 0:00:01 lr: 0.000623 min_lr: 0.000623 loss: 1.9591 (1.8315) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [338] [311/312] eta: 0:00:00 lr: 0.000623 min_lr: 0.000623 loss: 1.9591 (1.8326) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [338] Total time: 0:03:47 (0.7306 s / it) Averaged stats: lr: 0.000623 min_lr: 0.000623 loss: 1.9591 (1.8272) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.4749 (0.4749) acc1: 86.4583 (86.4583) acc5: 96.8750 (96.8750) time: 4.9233 data: 4.7040 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6729 (0.6868) acc1: 82.8125 (81.4080) acc5: 96.6146 (96.1280) time: 0.6983 data: 0.5227 max mem: 64948 Test: Total time: 0:00:06 (0.7229 s / it) * Acc@1 82.146 Acc@5 96.086 loss 0.676 Accuracy of the model on the 50000 test images: 82.1% Max accuracy: 82.33% Test: [0/9] eta: 0:00:48 loss: 0.4629 (0.4629) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 5.3850 data: 5.1747 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6374 (0.6512) acc1: 83.5938 (81.7920) acc5: 97.3958 (96.8000) time: 0.7497 data: 0.5751 max mem: 64948 Test: Total time: 0:00:06 (0.7611 s / it) * Acc@1 83.046 Acc@5 96.498 loss 0.628 Accuracy of the model EMA on 50000 test images: 83.0% Max EMA accuracy: 83.05% Epoch: [339] [ 0/312] eta: 0:44:11 lr: 0.000623 min_lr: 0.000623 loss: 2.2505 (2.2505) weight_decay: 0.0500 (0.0500) time: 8.4983 data: 7.3145 max mem: 64948 Epoch: [339] [ 10/312] eta: 0:07:58 lr: 0.000623 min_lr: 0.000623 loss: 1.9304 (1.8853) weight_decay: 0.0500 (0.0500) time: 1.5849 data: 0.8100 max mem: 64948 Epoch: [339] [ 20/312] eta: 0:05:38 lr: 0.000623 min_lr: 0.000623 loss: 1.9213 (1.8892) weight_decay: 0.0500 (0.0500) time: 0.7932 data: 0.0800 max mem: 64948 Epoch: [339] [ 30/312] eta: 0:04:45 lr: 0.000622 min_lr: 0.000622 loss: 1.9039 (1.8480) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [339] [ 40/312] eta: 0:04:14 lr: 0.000622 min_lr: 0.000622 loss: 1.8289 (1.8486) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [339] [ 50/312] eta: 0:03:52 lr: 0.000622 min_lr: 0.000622 loss: 1.8697 (1.8407) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [339] [ 60/312] eta: 0:03:35 lr: 0.000621 min_lr: 0.000621 loss: 2.0057 (1.8780) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [339] [ 70/312] eta: 0:03:21 lr: 0.000621 min_lr: 0.000621 loss: 1.9853 (1.8696) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [339] [ 80/312] eta: 0:03:09 lr: 0.000621 min_lr: 0.000621 loss: 1.8006 (1.8498) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [339] [ 90/312] eta: 0:02:58 lr: 0.000620 min_lr: 0.000620 loss: 1.7763 (1.8497) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [339] [100/312] eta: 0:02:48 lr: 0.000620 min_lr: 0.000620 loss: 1.8323 (1.8453) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [339] [110/312] eta: 0:02:38 lr: 0.000620 min_lr: 0.000620 loss: 1.8981 (1.8456) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [339] [120/312] eta: 0:02:29 lr: 0.000619 min_lr: 0.000619 loss: 1.9165 (1.8434) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [339] [130/312] eta: 0:02:20 lr: 0.000619 min_lr: 0.000619 loss: 1.9165 (1.8478) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [339] [140/312] eta: 0:02:11 lr: 0.000619 min_lr: 0.000619 loss: 1.9256 (1.8512) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [339] [150/312] eta: 0:02:03 lr: 0.000618 min_lr: 0.000618 loss: 1.8715 (1.8476) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [339] [160/312] eta: 0:01:54 lr: 0.000618 min_lr: 0.000618 loss: 1.8936 (1.8494) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [339] [170/312] eta: 0:01:46 lr: 0.000617 min_lr: 0.000617 loss: 1.9305 (1.8478) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [339] [180/312] eta: 0:01:38 lr: 0.000617 min_lr: 0.000617 loss: 1.9360 (1.8525) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [339] [190/312] eta: 0:01:31 lr: 0.000617 min_lr: 0.000617 loss: 1.7977 (1.8496) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [339] [200/312] eta: 0:01:23 lr: 0.000616 min_lr: 0.000616 loss: 1.7299 (1.8421) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [339] [210/312] eta: 0:01:15 lr: 0.000616 min_lr: 0.000616 loss: 1.7896 (1.8415) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [339] [220/312] eta: 0:01:08 lr: 0.000616 min_lr: 0.000616 loss: 1.8303 (1.8394) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [339] [230/312] eta: 0:01:00 lr: 0.000615 min_lr: 0.000615 loss: 2.0176 (1.8451) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [339] [240/312] eta: 0:00:53 lr: 0.000615 min_lr: 0.000615 loss: 1.9351 (1.8449) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [339] [250/312] eta: 0:00:45 lr: 0.000615 min_lr: 0.000615 loss: 1.8208 (1.8411) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [339] [260/312] eta: 0:00:38 lr: 0.000614 min_lr: 0.000614 loss: 1.8187 (1.8387) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [339] [270/312] eta: 0:00:30 lr: 0.000614 min_lr: 0.000614 loss: 1.8503 (1.8365) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [339] [280/312] eta: 0:00:23 lr: 0.000614 min_lr: 0.000614 loss: 1.8703 (1.8387) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [339] [290/312] eta: 0:00:16 lr: 0.000613 min_lr: 0.000613 loss: 1.8585 (1.8357) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [339] [300/312] eta: 0:00:08 lr: 0.000613 min_lr: 0.000613 loss: 1.8056 (1.8350) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [339] [310/312] eta: 0:00:01 lr: 0.000613 min_lr: 0.000613 loss: 1.8728 (1.8361) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [339] [311/312] eta: 0:00:00 lr: 0.000613 min_lr: 0.000613 loss: 1.8745 (1.8370) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [339] Total time: 0:03:47 (0.7297 s / it) Averaged stats: lr: 0.000613 min_lr: 0.000613 loss: 1.8745 (1.8184) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.5040 (0.5040) acc1: 86.4583 (86.4583) acc5: 97.1354 (97.1354) time: 4.9029 data: 4.6903 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7048 (0.6930) acc1: 82.2917 (80.9920) acc5: 97.1354 (96.3200) time: 0.6966 data: 0.5212 max mem: 64948 Test: Total time: 0:00:06 (0.7227 s / it) * Acc@1 82.148 Acc@5 96.132 loss 0.674 Accuracy of the model on the 50000 test images: 82.1% Max accuracy: 82.33% Test: [0/9] eta: 0:00:48 loss: 0.4630 (0.4630) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 5.3508 data: 5.1329 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6371 (0.6510) acc1: 83.5938 (81.7280) acc5: 97.3958 (96.7680) time: 0.7465 data: 0.5704 max mem: 64948 Test: Total time: 0:00:06 (0.7544 s / it) * Acc@1 83.082 Acc@5 96.494 loss 0.628 Accuracy of the model EMA on 50000 test images: 83.1% Max EMA accuracy: 83.08% Epoch: [340] [ 0/312] eta: 0:58:21 lr: 0.000613 min_lr: 0.000613 loss: 1.5002 (1.5002) weight_decay: 0.0500 (0.0500) time: 11.2241 data: 10.4731 max mem: 64948 Epoch: [340] [ 10/312] eta: 0:08:21 lr: 0.000612 min_lr: 0.000612 loss: 1.5251 (1.6659) weight_decay: 0.0500 (0.0500) time: 1.6613 data: 0.9524 max mem: 64948 Epoch: [340] [ 20/312] eta: 0:05:50 lr: 0.000612 min_lr: 0.000612 loss: 1.8151 (1.8064) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0003 max mem: 64948 Epoch: [340] [ 30/312] eta: 0:04:52 lr: 0.000612 min_lr: 0.000612 loss: 2.0209 (1.8926) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [340] [ 40/312] eta: 0:04:19 lr: 0.000611 min_lr: 0.000611 loss: 1.9466 (1.8281) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [340] [ 50/312] eta: 0:03:56 lr: 0.000611 min_lr: 0.000611 loss: 1.8470 (1.8369) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [340] [ 60/312] eta: 0:03:39 lr: 0.000611 min_lr: 0.000611 loss: 1.7261 (1.8118) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [340] [ 70/312] eta: 0:03:24 lr: 0.000610 min_lr: 0.000610 loss: 1.7754 (1.8196) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [340] [ 80/312] eta: 0:03:12 lr: 0.000610 min_lr: 0.000610 loss: 1.8237 (1.8111) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [340] [ 90/312] eta: 0:03:00 lr: 0.000610 min_lr: 0.000610 loss: 1.7109 (1.8053) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [340] [100/312] eta: 0:02:50 lr: 0.000609 min_lr: 0.000609 loss: 1.7865 (1.8068) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [340] [110/312] eta: 0:02:40 lr: 0.000609 min_lr: 0.000609 loss: 1.8078 (1.8016) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [340] [120/312] eta: 0:02:30 lr: 0.000609 min_lr: 0.000609 loss: 1.7920 (1.8089) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [340] [130/312] eta: 0:02:21 lr: 0.000608 min_lr: 0.000608 loss: 1.7920 (1.7998) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [340] [140/312] eta: 0:02:12 lr: 0.000608 min_lr: 0.000608 loss: 1.8591 (1.8071) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [340] [150/312] eta: 0:02:04 lr: 0.000608 min_lr: 0.000608 loss: 1.9919 (1.8125) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [340] [160/312] eta: 0:01:55 lr: 0.000607 min_lr: 0.000607 loss: 1.8706 (1.8121) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [340] [170/312] eta: 0:01:47 lr: 0.000607 min_lr: 0.000607 loss: 1.8970 (1.8208) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [340] [180/312] eta: 0:01:39 lr: 0.000607 min_lr: 0.000607 loss: 2.0035 (1.8248) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [340] [190/312] eta: 0:01:31 lr: 0.000606 min_lr: 0.000606 loss: 1.8659 (1.8181) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [340] [200/312] eta: 0:01:23 lr: 0.000606 min_lr: 0.000606 loss: 1.6413 (1.8131) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [340] [210/312] eta: 0:01:16 lr: 0.000606 min_lr: 0.000606 loss: 1.6190 (1.8022) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [340] [220/312] eta: 0:01:08 lr: 0.000605 min_lr: 0.000605 loss: 1.7749 (1.8031) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [340] [230/312] eta: 0:01:00 lr: 0.000605 min_lr: 0.000605 loss: 1.8655 (1.8057) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [340] [240/312] eta: 0:00:53 lr: 0.000605 min_lr: 0.000605 loss: 1.8655 (1.8067) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [340] [250/312] eta: 0:00:45 lr: 0.000604 min_lr: 0.000604 loss: 1.8952 (1.8117) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [340] [260/312] eta: 0:00:38 lr: 0.000604 min_lr: 0.000604 loss: 1.8294 (1.7996) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [340] [270/312] eta: 0:00:30 lr: 0.000604 min_lr: 0.000604 loss: 1.8619 (1.8064) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [340] [280/312] eta: 0:00:23 lr: 0.000603 min_lr: 0.000603 loss: 2.0119 (1.8111) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [340] [290/312] eta: 0:00:16 lr: 0.000603 min_lr: 0.000603 loss: 1.9495 (1.8111) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [340] [300/312] eta: 0:00:08 lr: 0.000603 min_lr: 0.000603 loss: 1.9156 (1.8138) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [340] [310/312] eta: 0:00:01 lr: 0.000602 min_lr: 0.000602 loss: 1.8977 (1.8143) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [340] [311/312] eta: 0:00:00 lr: 0.000602 min_lr: 0.000602 loss: 1.8977 (1.8138) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [340] Total time: 0:03:48 (0.7325 s / it) Averaged stats: lr: 0.000602 min_lr: 0.000602 loss: 1.8977 (1.8125) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4854 (0.4854) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.7115 data: 4.5024 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6798 (0.6952) acc1: 83.0729 (81.2480) acc5: 96.2264 (96.2560) time: 0.6749 data: 0.5004 max mem: 64948 Test: Total time: 0:00:06 (0.7032 s / it) * Acc@1 82.246 Acc@5 96.006 loss 0.677 Accuracy of the model on the 50000 test images: 82.2% Max accuracy: 82.33% Test: [0/9] eta: 0:00:48 loss: 0.4628 (0.4628) acc1: 87.5000 (87.5000) acc5: 98.1771 (98.1771) time: 5.4267 data: 5.2088 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6368 (0.6509) acc1: 83.5938 (81.7280) acc5: 97.3958 (96.7680) time: 0.7546 data: 0.5789 max mem: 64948 Test: Total time: 0:00:06 (0.7703 s / it) * Acc@1 83.096 Acc@5 96.500 loss 0.627 Accuracy of the model EMA on 50000 test images: 83.1% Max EMA accuracy: 83.10% Epoch: [341] [ 0/312] eta: 0:54:03 lr: 0.000602 min_lr: 0.000602 loss: 1.7945 (1.7945) weight_decay: 0.0500 (0.0500) time: 10.3946 data: 9.6116 max mem: 64948 Epoch: [341] [ 10/312] eta: 0:08:01 lr: 0.000602 min_lr: 0.000602 loss: 1.8627 (1.8160) weight_decay: 0.0500 (0.0500) time: 1.5928 data: 0.8742 max mem: 64948 Epoch: [341] [ 20/312] eta: 0:05:41 lr: 0.000602 min_lr: 0.000602 loss: 1.9129 (1.8502) weight_decay: 0.0500 (0.0500) time: 0.7082 data: 0.0004 max mem: 64948 Epoch: [341] [ 30/312] eta: 0:04:46 lr: 0.000601 min_lr: 0.000601 loss: 1.7958 (1.8316) weight_decay: 0.0500 (0.0500) time: 0.7008 data: 0.0004 max mem: 64948 Epoch: [341] [ 40/312] eta: 0:04:15 lr: 0.000601 min_lr: 0.000601 loss: 1.7854 (1.8165) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [341] [ 50/312] eta: 0:03:53 lr: 0.000601 min_lr: 0.000601 loss: 1.8332 (1.8088) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [341] [ 60/312] eta: 0:03:36 lr: 0.000600 min_lr: 0.000600 loss: 1.8551 (1.8115) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [341] [ 70/312] eta: 0:03:22 lr: 0.000600 min_lr: 0.000600 loss: 1.9000 (1.8243) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [341] [ 80/312] eta: 0:03:09 lr: 0.000600 min_lr: 0.000600 loss: 1.7457 (1.7802) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [341] [ 90/312] eta: 0:02:58 lr: 0.000599 min_lr: 0.000599 loss: 1.5439 (1.7748) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [341] [100/312] eta: 0:02:48 lr: 0.000599 min_lr: 0.000599 loss: 1.7858 (1.7846) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [341] [110/312] eta: 0:02:38 lr: 0.000599 min_lr: 0.000599 loss: 1.8831 (1.7913) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [341] [120/312] eta: 0:02:29 lr: 0.000598 min_lr: 0.000598 loss: 2.0046 (1.8002) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [341] [130/312] eta: 0:02:20 lr: 0.000598 min_lr: 0.000598 loss: 1.9594 (1.8020) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [341] [140/312] eta: 0:02:11 lr: 0.000598 min_lr: 0.000598 loss: 1.9594 (1.8060) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [341] [150/312] eta: 0:02:03 lr: 0.000597 min_lr: 0.000597 loss: 1.9396 (1.8001) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [341] [160/312] eta: 0:01:55 lr: 0.000597 min_lr: 0.000597 loss: 1.8233 (1.7999) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [341] [170/312] eta: 0:01:47 lr: 0.000597 min_lr: 0.000597 loss: 1.8091 (1.7922) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [341] [180/312] eta: 0:01:39 lr: 0.000596 min_lr: 0.000596 loss: 1.7687 (1.7887) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [341] [190/312] eta: 0:01:31 lr: 0.000596 min_lr: 0.000596 loss: 1.7687 (1.7926) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [341] [200/312] eta: 0:01:23 lr: 0.000596 min_lr: 0.000596 loss: 1.8142 (1.7934) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [341] [210/312] eta: 0:01:15 lr: 0.000595 min_lr: 0.000595 loss: 1.8926 (1.7956) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [341] [220/312] eta: 0:01:08 lr: 0.000595 min_lr: 0.000595 loss: 1.9281 (1.8020) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [341] [230/312] eta: 0:01:00 lr: 0.000595 min_lr: 0.000595 loss: 1.9281 (1.8059) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [341] [240/312] eta: 0:00:53 lr: 0.000594 min_lr: 0.000594 loss: 1.7244 (1.8062) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [341] [250/312] eta: 0:00:45 lr: 0.000594 min_lr: 0.000594 loss: 1.7190 (1.8049) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [341] [260/312] eta: 0:00:38 lr: 0.000594 min_lr: 0.000594 loss: 1.7883 (1.8082) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [341] [270/312] eta: 0:00:30 lr: 0.000593 min_lr: 0.000593 loss: 1.9738 (1.8138) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [341] [280/312] eta: 0:00:23 lr: 0.000593 min_lr: 0.000593 loss: 1.9773 (1.8157) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0006 max mem: 64948 Epoch: [341] [290/312] eta: 0:00:16 lr: 0.000593 min_lr: 0.000593 loss: 1.9067 (1.8156) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0005 max mem: 64948 Epoch: [341] [300/312] eta: 0:00:08 lr: 0.000592 min_lr: 0.000592 loss: 1.7409 (1.8147) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [341] [310/312] eta: 0:00:01 lr: 0.000592 min_lr: 0.000592 loss: 1.6679 (1.8097) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [341] [311/312] eta: 0:00:00 lr: 0.000592 min_lr: 0.000592 loss: 1.6898 (1.8101) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [341] Total time: 0:03:47 (0.7304 s / it) Averaged stats: lr: 0.000592 min_lr: 0.000592 loss: 1.6898 (1.8230) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4702 (0.4702) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.7724 data: 4.5600 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6887 (0.6937) acc1: 82.8125 (81.3760) acc5: 97.1354 (96.2880) time: 0.6815 data: 0.5068 max mem: 64948 Test: Total time: 0:00:06 (0.7056 s / it) * Acc@1 82.362 Acc@5 96.058 loss 0.671 Accuracy of the model on the 50000 test images: 82.4% Max accuracy: 82.36% Test: [0/9] eta: 0:00:40 loss: 0.4628 (0.4628) acc1: 87.5000 (87.5000) acc5: 98.1771 (98.1771) time: 4.5436 data: 4.3379 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6366 (0.6508) acc1: 84.3750 (81.7600) acc5: 97.3958 (96.7680) time: 0.6562 data: 0.4821 max mem: 64948 Test: Total time: 0:00:05 (0.6653 s / it) * Acc@1 83.108 Acc@5 96.506 loss 0.627 Accuracy of the model EMA on 50000 test images: 83.1% Max EMA accuracy: 83.11% Epoch: [342] [ 0/312] eta: 0:57:26 lr: 0.000592 min_lr: 0.000592 loss: 1.7927 (1.7927) weight_decay: 0.0500 (0.0500) time: 11.0476 data: 10.3395 max mem: 64948 Epoch: [342] [ 10/312] eta: 0:08:17 lr: 0.000591 min_lr: 0.000591 loss: 1.8988 (1.8758) weight_decay: 0.0500 (0.0500) time: 1.6486 data: 0.9402 max mem: 64948 Epoch: [342] [ 20/312] eta: 0:05:48 lr: 0.000591 min_lr: 0.000591 loss: 1.9036 (1.8618) weight_decay: 0.0500 (0.0500) time: 0.7010 data: 0.0003 max mem: 64948 Epoch: [342] [ 30/312] eta: 0:04:51 lr: 0.000591 min_lr: 0.000591 loss: 1.9103 (1.8277) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [342] [ 40/312] eta: 0:04:18 lr: 0.000590 min_lr: 0.000590 loss: 1.8011 (1.8286) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [342] [ 50/312] eta: 0:03:56 lr: 0.000590 min_lr: 0.000590 loss: 1.8351 (1.8370) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [342] [ 60/312] eta: 0:03:38 lr: 0.000590 min_lr: 0.000590 loss: 1.8664 (1.8115) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [342] [ 70/312] eta: 0:03:24 lr: 0.000589 min_lr: 0.000589 loss: 1.7766 (1.7959) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [342] [ 80/312] eta: 0:03:11 lr: 0.000589 min_lr: 0.000589 loss: 1.8443 (1.7985) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [342] [ 90/312] eta: 0:03:00 lr: 0.000589 min_lr: 0.000589 loss: 1.9082 (1.7984) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [342] [100/312] eta: 0:02:49 lr: 0.000588 min_lr: 0.000588 loss: 1.9248 (1.7975) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [342] [110/312] eta: 0:02:39 lr: 0.000588 min_lr: 0.000588 loss: 1.8207 (1.7875) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [342] [120/312] eta: 0:02:30 lr: 0.000588 min_lr: 0.000588 loss: 1.7767 (1.7875) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [342] [130/312] eta: 0:02:21 lr: 0.000588 min_lr: 0.000588 loss: 1.8764 (1.7952) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [342] [140/312] eta: 0:02:12 lr: 0.000587 min_lr: 0.000587 loss: 1.8460 (1.7908) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [342] [150/312] eta: 0:02:03 lr: 0.000587 min_lr: 0.000587 loss: 1.8460 (1.7952) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [342] [160/312] eta: 0:01:55 lr: 0.000587 min_lr: 0.000587 loss: 1.8800 (1.7959) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [342] [170/312] eta: 0:01:47 lr: 0.000586 min_lr: 0.000586 loss: 1.8703 (1.7908) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [342] [180/312] eta: 0:01:39 lr: 0.000586 min_lr: 0.000586 loss: 1.7672 (1.7861) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [342] [190/312] eta: 0:01:31 lr: 0.000586 min_lr: 0.000586 loss: 1.7974 (1.7860) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [342] [200/312] eta: 0:01:23 lr: 0.000585 min_lr: 0.000585 loss: 1.9022 (1.7942) weight_decay: 0.0500 (0.0500) time: 0.7023 data: 0.0004 max mem: 64948 Epoch: [342] [210/312] eta: 0:01:16 lr: 0.000585 min_lr: 0.000585 loss: 2.0936 (1.8035) weight_decay: 0.0500 (0.0500) time: 0.7006 data: 0.0004 max mem: 64948 Epoch: [342] [220/312] eta: 0:01:08 lr: 0.000585 min_lr: 0.000585 loss: 2.0115 (1.8047) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [342] [230/312] eta: 0:01:00 lr: 0.000584 min_lr: 0.000584 loss: 1.7060 (1.7993) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [342] [240/312] eta: 0:00:53 lr: 0.000584 min_lr: 0.000584 loss: 1.7060 (1.7985) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [342] [250/312] eta: 0:00:45 lr: 0.000584 min_lr: 0.000584 loss: 1.8306 (1.8020) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [342] [260/312] eta: 0:00:38 lr: 0.000583 min_lr: 0.000583 loss: 1.8516 (1.8013) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [342] [270/312] eta: 0:00:30 lr: 0.000583 min_lr: 0.000583 loss: 1.9365 (1.8080) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [342] [280/312] eta: 0:00:23 lr: 0.000583 min_lr: 0.000583 loss: 1.9126 (1.8039) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [342] [290/312] eta: 0:00:16 lr: 0.000582 min_lr: 0.000582 loss: 1.6400 (1.8013) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [342] [300/312] eta: 0:00:08 lr: 0.000582 min_lr: 0.000582 loss: 1.7105 (1.8009) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [342] [310/312] eta: 0:00:01 lr: 0.000582 min_lr: 0.000582 loss: 1.8258 (1.8003) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [342] [311/312] eta: 0:00:00 lr: 0.000582 min_lr: 0.000582 loss: 1.8258 (1.8013) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [342] Total time: 0:03:48 (0.7316 s / it) Averaged stats: lr: 0.000582 min_lr: 0.000582 loss: 1.8258 (1.8174) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.4516 (0.4516) acc1: 88.8021 (88.8021) acc5: 97.6562 (97.6562) time: 4.8920 data: 4.6723 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6767 (0.6949) acc1: 84.1146 (81.3440) acc5: 96.8750 (95.9680) time: 0.6950 data: 0.5192 max mem: 64948 Test: Total time: 0:00:06 (0.7255 s / it) * Acc@1 82.412 Acc@5 96.038 loss 0.671 Accuracy of the model on the 50000 test images: 82.4% Max accuracy: 82.41% Test: [0/9] eta: 0:00:42 loss: 0.4624 (0.4624) acc1: 87.5000 (87.5000) acc5: 98.1771 (98.1771) time: 4.7663 data: 4.5506 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6363 (0.6506) acc1: 84.1146 (81.7280) acc5: 97.3958 (96.7680) time: 0.6810 data: 0.5057 max mem: 64948 Test: Total time: 0:00:06 (0.6898 s / it) * Acc@1 83.118 Acc@5 96.510 loss 0.627 Accuracy of the model EMA on 50000 test images: 83.1% Max EMA accuracy: 83.12% Epoch: [343] [ 0/312] eta: 0:56:55 lr: 0.000581 min_lr: 0.000581 loss: 2.3038 (2.3038) weight_decay: 0.0500 (0.0500) time: 10.9484 data: 10.2302 max mem: 64948 Epoch: [343] [ 10/312] eta: 0:08:12 lr: 0.000581 min_lr: 0.000581 loss: 1.9761 (1.9413) weight_decay: 0.0500 (0.0500) time: 1.6302 data: 0.9303 max mem: 64948 Epoch: [343] [ 20/312] eta: 0:05:47 lr: 0.000581 min_lr: 0.000581 loss: 1.8374 (1.8350) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0003 max mem: 64948 Epoch: [343] [ 30/312] eta: 0:04:50 lr: 0.000580 min_lr: 0.000580 loss: 1.7814 (1.7878) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [343] [ 40/312] eta: 0:04:18 lr: 0.000580 min_lr: 0.000580 loss: 1.7814 (1.8014) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0004 max mem: 64948 Epoch: [343] [ 50/312] eta: 0:03:56 lr: 0.000580 min_lr: 0.000580 loss: 1.7220 (1.7845) weight_decay: 0.0500 (0.0500) time: 0.7012 data: 0.0004 max mem: 64948 Epoch: [343] [ 60/312] eta: 0:03:38 lr: 0.000580 min_lr: 0.000580 loss: 1.7376 (1.7977) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [343] [ 70/312] eta: 0:03:24 lr: 0.000579 min_lr: 0.000579 loss: 1.9175 (1.8069) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [343] [ 80/312] eta: 0:03:11 lr: 0.000579 min_lr: 0.000579 loss: 1.9179 (1.8196) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [343] [ 90/312] eta: 0:03:00 lr: 0.000579 min_lr: 0.000579 loss: 1.7570 (1.8015) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [343] [100/312] eta: 0:02:49 lr: 0.000578 min_lr: 0.000578 loss: 1.7289 (1.7934) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [343] [110/312] eta: 0:02:39 lr: 0.000578 min_lr: 0.000578 loss: 1.7297 (1.7928) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [343] [120/312] eta: 0:02:30 lr: 0.000578 min_lr: 0.000578 loss: 1.7800 (1.7906) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [343] [130/312] eta: 0:02:21 lr: 0.000577 min_lr: 0.000577 loss: 1.7790 (1.7811) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [343] [140/312] eta: 0:02:12 lr: 0.000577 min_lr: 0.000577 loss: 1.7626 (1.7807) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [343] [150/312] eta: 0:02:03 lr: 0.000577 min_lr: 0.000577 loss: 1.8296 (1.7900) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0003 max mem: 64948 Epoch: [343] [160/312] eta: 0:01:55 lr: 0.000576 min_lr: 0.000576 loss: 1.7292 (1.7849) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0003 max mem: 64948 Epoch: [343] [170/312] eta: 0:01:47 lr: 0.000576 min_lr: 0.000576 loss: 1.7219 (1.7782) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [343] [180/312] eta: 0:01:39 lr: 0.000576 min_lr: 0.000576 loss: 1.7998 (1.7769) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [343] [190/312] eta: 0:01:31 lr: 0.000575 min_lr: 0.000575 loss: 1.6608 (1.7716) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [343] [200/312] eta: 0:01:23 lr: 0.000575 min_lr: 0.000575 loss: 1.6948 (1.7744) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [343] [210/312] eta: 0:01:16 lr: 0.000575 min_lr: 0.000575 loss: 1.9862 (1.7812) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [343] [220/312] eta: 0:01:08 lr: 0.000574 min_lr: 0.000574 loss: 1.9697 (1.7843) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [343] [230/312] eta: 0:01:00 lr: 0.000574 min_lr: 0.000574 loss: 1.9680 (1.7922) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [343] [240/312] eta: 0:00:53 lr: 0.000574 min_lr: 0.000574 loss: 1.9954 (1.7990) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [343] [250/312] eta: 0:00:45 lr: 0.000573 min_lr: 0.000573 loss: 1.9343 (1.7972) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [343] [260/312] eta: 0:00:38 lr: 0.000573 min_lr: 0.000573 loss: 1.7595 (1.7947) weight_decay: 0.0500 (0.0500) time: 0.7015 data: 0.0004 max mem: 64948 Epoch: [343] [270/312] eta: 0:00:30 lr: 0.000573 min_lr: 0.000573 loss: 1.8356 (1.7980) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [343] [280/312] eta: 0:00:23 lr: 0.000572 min_lr: 0.000572 loss: 1.8109 (1.7930) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0010 max mem: 64948 Epoch: [343] [290/312] eta: 0:00:16 lr: 0.000572 min_lr: 0.000572 loss: 1.8941 (1.7967) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0009 max mem: 64948 Epoch: [343] [300/312] eta: 0:00:08 lr: 0.000572 min_lr: 0.000572 loss: 1.9378 (1.8013) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [343] [310/312] eta: 0:00:01 lr: 0.000571 min_lr: 0.000571 loss: 1.8726 (1.8027) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [343] [311/312] eta: 0:00:00 lr: 0.000571 min_lr: 0.000571 loss: 1.8670 (1.8018) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [343] Total time: 0:03:48 (0.7323 s / it) Averaged stats: lr: 0.000571 min_lr: 0.000571 loss: 1.8670 (1.8079) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4903 (0.4903) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.4944 data: 4.2743 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7076 (0.6822) acc1: 80.9896 (80.7680) acc5: 97.1354 (96.2560) time: 0.6507 data: 0.4750 max mem: 64948 Test: Total time: 0:00:06 (0.6822 s / it) * Acc@1 82.356 Acc@5 96.166 loss 0.670 Accuracy of the model on the 50000 test images: 82.4% Max accuracy: 82.41% Test: [0/9] eta: 0:00:48 loss: 0.4619 (0.4619) acc1: 87.5000 (87.5000) acc5: 98.1771 (98.1771) time: 5.3648 data: 5.1533 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6358 (0.6506) acc1: 84.1146 (81.6960) acc5: 97.3958 (96.7680) time: 0.7475 data: 0.5727 max mem: 64948 Test: Total time: 0:00:06 (0.7600 s / it) * Acc@1 83.126 Acc@5 96.528 loss 0.627 Accuracy of the model EMA on 50000 test images: 83.1% Max EMA accuracy: 83.13% Epoch: [344] [ 0/312] eta: 0:58:53 lr: 0.000571 min_lr: 0.000571 loss: 1.9275 (1.9275) weight_decay: 0.0500 (0.0500) time: 11.3238 data: 10.4168 max mem: 64948 Epoch: [344] [ 10/312] eta: 0:08:22 lr: 0.000571 min_lr: 0.000571 loss: 1.8800 (1.7783) weight_decay: 0.0500 (0.0500) time: 1.6623 data: 0.9473 max mem: 64948 Epoch: [344] [ 20/312] eta: 0:05:51 lr: 0.000571 min_lr: 0.000571 loss: 1.8270 (1.7585) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [344] [ 30/312] eta: 0:04:52 lr: 0.000570 min_lr: 0.000570 loss: 1.8854 (1.7947) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [344] [ 40/312] eta: 0:04:19 lr: 0.000570 min_lr: 0.000570 loss: 1.8916 (1.7602) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [344] [ 50/312] eta: 0:03:56 lr: 0.000570 min_lr: 0.000570 loss: 1.8551 (1.7874) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [344] [ 60/312] eta: 0:03:39 lr: 0.000569 min_lr: 0.000569 loss: 1.9738 (1.7993) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [344] [ 70/312] eta: 0:03:24 lr: 0.000569 min_lr: 0.000569 loss: 1.9347 (1.8017) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [344] [ 80/312] eta: 0:03:12 lr: 0.000569 min_lr: 0.000569 loss: 1.8417 (1.7981) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [344] [ 90/312] eta: 0:03:00 lr: 0.000568 min_lr: 0.000568 loss: 1.8676 (1.7951) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [344] [100/312] eta: 0:02:50 lr: 0.000568 min_lr: 0.000568 loss: 1.9168 (1.7946) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [344] [110/312] eta: 0:02:40 lr: 0.000568 min_lr: 0.000568 loss: 1.8445 (1.7898) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [344] [120/312] eta: 0:02:30 lr: 0.000567 min_lr: 0.000567 loss: 1.8790 (1.7980) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [344] [130/312] eta: 0:02:21 lr: 0.000567 min_lr: 0.000567 loss: 1.9928 (1.8078) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [344] [140/312] eta: 0:02:12 lr: 0.000567 min_lr: 0.000567 loss: 1.9499 (1.8109) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [344] [150/312] eta: 0:02:04 lr: 0.000566 min_lr: 0.000566 loss: 1.8832 (1.8087) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [344] [160/312] eta: 0:01:55 lr: 0.000566 min_lr: 0.000566 loss: 1.8775 (1.8100) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [344] [170/312] eta: 0:01:47 lr: 0.000566 min_lr: 0.000566 loss: 1.8551 (1.8051) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [344] [180/312] eta: 0:01:39 lr: 0.000565 min_lr: 0.000565 loss: 1.8075 (1.8065) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [344] [190/312] eta: 0:01:31 lr: 0.000565 min_lr: 0.000565 loss: 1.9176 (1.8083) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [344] [200/312] eta: 0:01:23 lr: 0.000565 min_lr: 0.000565 loss: 1.8530 (1.8081) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [344] [210/312] eta: 0:01:16 lr: 0.000564 min_lr: 0.000564 loss: 1.8529 (1.8063) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [344] [220/312] eta: 0:01:08 lr: 0.000564 min_lr: 0.000564 loss: 1.9034 (1.8035) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [344] [230/312] eta: 0:01:00 lr: 0.000564 min_lr: 0.000564 loss: 1.9313 (1.8039) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [344] [240/312] eta: 0:00:53 lr: 0.000563 min_lr: 0.000563 loss: 1.9313 (1.8070) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [344] [250/312] eta: 0:00:45 lr: 0.000563 min_lr: 0.000563 loss: 1.9852 (1.8088) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [344] [260/312] eta: 0:00:38 lr: 0.000563 min_lr: 0.000563 loss: 1.9997 (1.8131) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [344] [270/312] eta: 0:00:30 lr: 0.000562 min_lr: 0.000562 loss: 1.9997 (1.8162) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [344] [280/312] eta: 0:00:23 lr: 0.000562 min_lr: 0.000562 loss: 1.8762 (1.8154) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0010 max mem: 64948 Epoch: [344] [290/312] eta: 0:00:16 lr: 0.000562 min_lr: 0.000562 loss: 1.8762 (1.8190) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [344] [300/312] eta: 0:00:08 lr: 0.000561 min_lr: 0.000561 loss: 1.9360 (1.8188) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [344] [310/312] eta: 0:00:01 lr: 0.000561 min_lr: 0.000561 loss: 1.9360 (1.8230) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [344] [311/312] eta: 0:00:00 lr: 0.000561 min_lr: 0.000561 loss: 1.9482 (1.8234) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [344] Total time: 0:03:48 (0.7325 s / it) Averaged stats: lr: 0.000561 min_lr: 0.000561 loss: 1.9482 (1.8067) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4913 (0.4913) acc1: 86.7188 (86.7188) acc5: 97.6562 (97.6562) time: 4.7119 data: 4.4929 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7339 (0.6966) acc1: 82.5521 (81.6000) acc5: 96.2264 (95.5520) time: 0.6749 data: 0.4993 max mem: 64948 Test: Total time: 0:00:06 (0.7035 s / it) * Acc@1 82.280 Acc@5 96.006 loss 0.681 Accuracy of the model on the 50000 test images: 82.3% Max accuracy: 82.41% Test: [0/9] eta: 0:00:44 loss: 0.4618 (0.4618) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.8947 data: 4.6790 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6359 (0.6504) acc1: 84.1146 (81.6960) acc5: 97.3958 (96.7360) time: 0.6951 data: 0.5200 max mem: 64948 Test: Total time: 0:00:06 (0.7046 s / it) * Acc@1 83.126 Acc@5 96.516 loss 0.627 Accuracy of the model EMA on 50000 test images: 83.1% Epoch: [345] [ 0/312] eta: 0:58:40 lr: 0.000561 min_lr: 0.000561 loss: 1.5490 (1.5490) weight_decay: 0.0500 (0.0500) time: 11.2841 data: 7.1638 max mem: 64948 Epoch: [345] [ 10/312] eta: 0:08:24 lr: 0.000561 min_lr: 0.000561 loss: 1.7559 (1.6545) weight_decay: 0.0500 (0.0500) time: 1.6718 data: 0.6517 max mem: 64948 Epoch: [345] [ 20/312] eta: 0:05:51 lr: 0.000560 min_lr: 0.000560 loss: 1.8170 (1.7787) weight_decay: 0.0500 (0.0500) time: 0.7014 data: 0.0004 max mem: 64948 Epoch: [345] [ 30/312] eta: 0:04:53 lr: 0.000560 min_lr: 0.000560 loss: 1.9224 (1.8064) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [345] [ 40/312] eta: 0:04:20 lr: 0.000560 min_lr: 0.000560 loss: 1.8605 (1.8022) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [345] [ 50/312] eta: 0:03:57 lr: 0.000559 min_lr: 0.000559 loss: 1.8605 (1.8165) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [345] [ 60/312] eta: 0:03:39 lr: 0.000559 min_lr: 0.000559 loss: 1.9496 (1.8285) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [345] [ 70/312] eta: 0:03:25 lr: 0.000559 min_lr: 0.000559 loss: 1.9190 (1.8292) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [345] [ 80/312] eta: 0:03:12 lr: 0.000558 min_lr: 0.000558 loss: 1.9190 (1.8309) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [345] [ 90/312] eta: 0:03:00 lr: 0.000558 min_lr: 0.000558 loss: 1.9211 (1.8211) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [345] [100/312] eta: 0:02:50 lr: 0.000558 min_lr: 0.000558 loss: 1.6505 (1.8011) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [345] [110/312] eta: 0:02:40 lr: 0.000557 min_lr: 0.000557 loss: 1.8375 (1.8118) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [345] [120/312] eta: 0:02:30 lr: 0.000557 min_lr: 0.000557 loss: 1.9448 (1.8101) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [345] [130/312] eta: 0:02:21 lr: 0.000557 min_lr: 0.000557 loss: 1.9146 (1.8173) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [345] [140/312] eta: 0:02:12 lr: 0.000557 min_lr: 0.000557 loss: 1.9291 (1.8218) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [345] [150/312] eta: 0:02:04 lr: 0.000556 min_lr: 0.000556 loss: 1.9716 (1.8290) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [345] [160/312] eta: 0:01:55 lr: 0.000556 min_lr: 0.000556 loss: 1.8071 (1.8263) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [345] [170/312] eta: 0:01:47 lr: 0.000556 min_lr: 0.000556 loss: 1.8313 (1.8303) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [345] [180/312] eta: 0:01:39 lr: 0.000555 min_lr: 0.000555 loss: 1.8575 (1.8290) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [345] [190/312] eta: 0:01:31 lr: 0.000555 min_lr: 0.000555 loss: 1.8725 (1.8364) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [345] [200/312] eta: 0:01:23 lr: 0.000555 min_lr: 0.000555 loss: 1.9910 (1.8393) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [345] [210/312] eta: 0:01:16 lr: 0.000554 min_lr: 0.000554 loss: 1.9627 (1.8447) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [345] [220/312] eta: 0:01:08 lr: 0.000554 min_lr: 0.000554 loss: 1.9372 (1.8455) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [345] [230/312] eta: 0:01:00 lr: 0.000554 min_lr: 0.000554 loss: 1.9831 (1.8507) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [345] [240/312] eta: 0:00:53 lr: 0.000553 min_lr: 0.000553 loss: 1.9915 (1.8504) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [345] [250/312] eta: 0:00:45 lr: 0.000553 min_lr: 0.000553 loss: 1.9457 (1.8548) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [345] [260/312] eta: 0:00:38 lr: 0.000553 min_lr: 0.000553 loss: 1.8299 (1.8491) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [345] [270/312] eta: 0:00:30 lr: 0.000552 min_lr: 0.000552 loss: 1.8732 (1.8484) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [345] [280/312] eta: 0:00:23 lr: 0.000552 min_lr: 0.000552 loss: 1.9053 (1.8466) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0009 max mem: 64948 Epoch: [345] [290/312] eta: 0:00:16 lr: 0.000552 min_lr: 0.000552 loss: 1.8615 (1.8428) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0008 max mem: 64948 Epoch: [345] [300/312] eta: 0:00:08 lr: 0.000551 min_lr: 0.000551 loss: 1.8615 (1.8420) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [345] [310/312] eta: 0:00:01 lr: 0.000551 min_lr: 0.000551 loss: 1.8240 (1.8406) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [345] [311/312] eta: 0:00:00 lr: 0.000551 min_lr: 0.000551 loss: 1.7743 (1.8392) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [345] Total time: 0:03:48 (0.7328 s / it) Averaged stats: lr: 0.000551 min_lr: 0.000551 loss: 1.7743 (1.8134) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4931 (0.4931) acc1: 86.7188 (86.7188) acc5: 97.6562 (97.6562) time: 4.6885 data: 4.4779 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.7007 (0.6923) acc1: 81.7708 (80.9920) acc5: 96.8750 (96.2880) time: 0.6722 data: 0.4976 max mem: 64948 Test: Total time: 0:00:06 (0.6989 s / it) * Acc@1 82.328 Acc@5 96.158 loss 0.666 Accuracy of the model on the 50000 test images: 82.3% Max accuracy: 82.41% Test: [0/9] eta: 0:00:47 loss: 0.4617 (0.4617) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.2427 data: 5.0265 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6357 (0.6502) acc1: 84.1146 (81.7920) acc5: 97.3958 (96.7360) time: 0.7338 data: 0.5586 max mem: 64948 Test: Total time: 0:00:06 (0.7426 s / it) * Acc@1 83.144 Acc@5 96.520 loss 0.627 Accuracy of the model EMA on 50000 test images: 83.1% Max EMA accuracy: 83.14% Epoch: [346] [ 0/312] eta: 0:53:50 lr: 0.000551 min_lr: 0.000551 loss: 2.2762 (2.2762) weight_decay: 0.0500 (0.0500) time: 10.3549 data: 7.5638 max mem: 64948 Epoch: [346] [ 10/312] eta: 0:07:58 lr: 0.000551 min_lr: 0.000551 loss: 1.8411 (1.7847) weight_decay: 0.0500 (0.0500) time: 1.5860 data: 0.6880 max mem: 64948 Epoch: [346] [ 20/312] eta: 0:05:38 lr: 0.000550 min_lr: 0.000550 loss: 1.8411 (1.8322) weight_decay: 0.0500 (0.0500) time: 0.7011 data: 0.0004 max mem: 64948 Epoch: [346] [ 30/312] eta: 0:04:44 lr: 0.000550 min_lr: 0.000550 loss: 1.9099 (1.8345) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [346] [ 40/312] eta: 0:04:14 lr: 0.000550 min_lr: 0.000550 loss: 1.9279 (1.8492) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [346] [ 50/312] eta: 0:03:52 lr: 0.000549 min_lr: 0.000549 loss: 1.8449 (1.8329) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [346] [ 60/312] eta: 0:03:35 lr: 0.000549 min_lr: 0.000549 loss: 1.7890 (1.8125) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [346] [ 70/312] eta: 0:03:21 lr: 0.000549 min_lr: 0.000549 loss: 1.6638 (1.7910) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [346] [ 80/312] eta: 0:03:09 lr: 0.000548 min_lr: 0.000548 loss: 1.8637 (1.8120) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [346] [ 90/312] eta: 0:02:58 lr: 0.000548 min_lr: 0.000548 loss: 1.8271 (1.8022) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [346] [100/312] eta: 0:02:48 lr: 0.000548 min_lr: 0.000548 loss: 1.6252 (1.7874) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [346] [110/312] eta: 0:02:38 lr: 0.000547 min_lr: 0.000547 loss: 1.6380 (1.7883) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [346] [120/312] eta: 0:02:29 lr: 0.000547 min_lr: 0.000547 loss: 1.8183 (1.7907) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [346] [130/312] eta: 0:02:20 lr: 0.000547 min_lr: 0.000547 loss: 1.9065 (1.7923) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [346] [140/312] eta: 0:02:11 lr: 0.000546 min_lr: 0.000546 loss: 1.8568 (1.7994) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [346] [150/312] eta: 0:02:03 lr: 0.000546 min_lr: 0.000546 loss: 1.8596 (1.8038) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [346] [160/312] eta: 0:01:54 lr: 0.000546 min_lr: 0.000546 loss: 1.8836 (1.8053) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [346] [170/312] eta: 0:01:46 lr: 0.000545 min_lr: 0.000545 loss: 1.8148 (1.7973) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [346] [180/312] eta: 0:01:38 lr: 0.000545 min_lr: 0.000545 loss: 1.8495 (1.8012) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [346] [190/312] eta: 0:01:31 lr: 0.000545 min_lr: 0.000545 loss: 1.8849 (1.8024) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [346] [200/312] eta: 0:01:23 lr: 0.000545 min_lr: 0.000545 loss: 1.7999 (1.7987) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [346] [210/312] eta: 0:01:15 lr: 0.000544 min_lr: 0.000544 loss: 1.8289 (1.8018) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [346] [220/312] eta: 0:01:08 lr: 0.000544 min_lr: 0.000544 loss: 1.8520 (1.7995) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [346] [230/312] eta: 0:01:00 lr: 0.000544 min_lr: 0.000544 loss: 1.6881 (1.7971) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [346] [240/312] eta: 0:00:52 lr: 0.000543 min_lr: 0.000543 loss: 1.8877 (1.8027) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [346] [250/312] eta: 0:00:45 lr: 0.000543 min_lr: 0.000543 loss: 1.9044 (1.8067) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [346] [260/312] eta: 0:00:38 lr: 0.000543 min_lr: 0.000543 loss: 1.9627 (1.8126) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [346] [270/312] eta: 0:00:30 lr: 0.000542 min_lr: 0.000542 loss: 1.9627 (1.8137) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [346] [280/312] eta: 0:00:23 lr: 0.000542 min_lr: 0.000542 loss: 1.9155 (1.8147) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0009 max mem: 64948 Epoch: [346] [290/312] eta: 0:00:16 lr: 0.000542 min_lr: 0.000542 loss: 1.8302 (1.8119) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [346] [300/312] eta: 0:00:08 lr: 0.000541 min_lr: 0.000541 loss: 1.8251 (1.8089) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [346] [310/312] eta: 0:00:01 lr: 0.000541 min_lr: 0.000541 loss: 1.7467 (1.8105) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [346] [311/312] eta: 0:00:00 lr: 0.000541 min_lr: 0.000541 loss: 1.7881 (1.8107) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [346] Total time: 0:03:47 (0.7291 s / it) Averaged stats: lr: 0.000541 min_lr: 0.000541 loss: 1.7881 (1.8012) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.5097 (0.5097) acc1: 86.7188 (86.7188) acc5: 97.1354 (97.1354) time: 4.8327 data: 4.6160 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6542 (0.6801) acc1: 83.3333 (81.1840) acc5: 96.8750 (96.0000) time: 0.6882 data: 0.5130 max mem: 64948 Test: Total time: 0:00:06 (0.7168 s / it) * Acc@1 82.400 Acc@5 96.110 loss 0.667 Accuracy of the model on the 50000 test images: 82.4% Max accuracy: 82.41% Test: [0/9] eta: 0:00:46 loss: 0.4617 (0.4617) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.1604 data: 4.9520 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6352 (0.6500) acc1: 84.1146 (81.7920) acc5: 97.3958 (96.7040) time: 0.7247 data: 0.5503 max mem: 64948 Test: Total time: 0:00:06 (0.7334 s / it) * Acc@1 83.152 Acc@5 96.508 loss 0.626 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.15% Epoch: [347] [ 0/312] eta: 0:52:51 lr: 0.000541 min_lr: 0.000541 loss: 2.1107 (2.1107) weight_decay: 0.0500 (0.0500) time: 10.1643 data: 7.7456 max mem: 64948 Epoch: [347] [ 10/312] eta: 0:07:53 lr: 0.000541 min_lr: 0.000541 loss: 1.9931 (1.8446) weight_decay: 0.0500 (0.0500) time: 1.5673 data: 0.7045 max mem: 64948 Epoch: [347] [ 20/312] eta: 0:05:36 lr: 0.000540 min_lr: 0.000540 loss: 1.7747 (1.7482) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0004 max mem: 64948 Epoch: [347] [ 30/312] eta: 0:04:43 lr: 0.000540 min_lr: 0.000540 loss: 1.5807 (1.7184) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [347] [ 40/312] eta: 0:04:13 lr: 0.000540 min_lr: 0.000540 loss: 1.6957 (1.7496) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [347] [ 50/312] eta: 0:03:51 lr: 0.000539 min_lr: 0.000539 loss: 1.8766 (1.7856) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [347] [ 60/312] eta: 0:03:35 lr: 0.000539 min_lr: 0.000539 loss: 1.8570 (1.7765) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [347] [ 70/312] eta: 0:03:21 lr: 0.000539 min_lr: 0.000539 loss: 1.8056 (1.7687) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [347] [ 80/312] eta: 0:03:09 lr: 0.000538 min_lr: 0.000538 loss: 1.7941 (1.7522) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [347] [ 90/312] eta: 0:02:58 lr: 0.000538 min_lr: 0.000538 loss: 1.8100 (1.7592) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0004 max mem: 64948 Epoch: [347] [100/312] eta: 0:02:48 lr: 0.000538 min_lr: 0.000538 loss: 1.8575 (1.7615) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [347] [110/312] eta: 0:02:38 lr: 0.000537 min_lr: 0.000537 loss: 1.8488 (1.7623) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [347] [120/312] eta: 0:02:29 lr: 0.000537 min_lr: 0.000537 loss: 1.8331 (1.7678) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [347] [130/312] eta: 0:02:20 lr: 0.000537 min_lr: 0.000537 loss: 2.0328 (1.7852) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [347] [140/312] eta: 0:02:11 lr: 0.000536 min_lr: 0.000536 loss: 1.9850 (1.7900) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [347] [150/312] eta: 0:02:03 lr: 0.000536 min_lr: 0.000536 loss: 1.8557 (1.7965) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [347] [160/312] eta: 0:01:54 lr: 0.000536 min_lr: 0.000536 loss: 1.8302 (1.7916) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [347] [170/312] eta: 0:01:46 lr: 0.000536 min_lr: 0.000536 loss: 1.8302 (1.7914) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [347] [180/312] eta: 0:01:38 lr: 0.000535 min_lr: 0.000535 loss: 1.9160 (1.7971) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [347] [190/312] eta: 0:01:31 lr: 0.000535 min_lr: 0.000535 loss: 1.9095 (1.8002) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [347] [200/312] eta: 0:01:23 lr: 0.000535 min_lr: 0.000535 loss: 1.9000 (1.8015) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [347] [210/312] eta: 0:01:15 lr: 0.000534 min_lr: 0.000534 loss: 1.7626 (1.7954) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [347] [220/312] eta: 0:01:08 lr: 0.000534 min_lr: 0.000534 loss: 1.5720 (1.7916) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [347] [230/312] eta: 0:01:00 lr: 0.000534 min_lr: 0.000534 loss: 1.7761 (1.7950) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [347] [240/312] eta: 0:00:52 lr: 0.000533 min_lr: 0.000533 loss: 1.8018 (1.7950) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [347] [250/312] eta: 0:00:45 lr: 0.000533 min_lr: 0.000533 loss: 1.9350 (1.7994) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [347] [260/312] eta: 0:00:38 lr: 0.000533 min_lr: 0.000533 loss: 1.9109 (1.8028) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [347] [270/312] eta: 0:00:30 lr: 0.000532 min_lr: 0.000532 loss: 1.8030 (1.7983) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [347] [280/312] eta: 0:00:23 lr: 0.000532 min_lr: 0.000532 loss: 1.7375 (1.7981) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [347] [290/312] eta: 0:00:16 lr: 0.000532 min_lr: 0.000532 loss: 1.8960 (1.8009) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [347] [300/312] eta: 0:00:08 lr: 0.000531 min_lr: 0.000531 loss: 1.8598 (1.7975) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [347] [310/312] eta: 0:00:01 lr: 0.000531 min_lr: 0.000531 loss: 1.5675 (1.7906) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [347] [311/312] eta: 0:00:00 lr: 0.000531 min_lr: 0.000531 loss: 1.5620 (1.7898) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [347] Total time: 0:03:47 (0.7292 s / it) Averaged stats: lr: 0.000531 min_lr: 0.000531 loss: 1.5620 (1.7986) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.5048 (0.5048) acc1: 86.1979 (86.1979) acc5: 97.6562 (97.6562) time: 4.8562 data: 4.6488 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6712 (0.6876) acc1: 82.5521 (81.2800) acc5: 96.3542 (95.8720) time: 0.6908 data: 0.5166 max mem: 64948 Test: Total time: 0:00:06 (0.7162 s / it) * Acc@1 82.368 Acc@5 96.082 loss 0.676 Accuracy of the model on the 50000 test images: 82.4% Max accuracy: 82.41% Test: [0/9] eta: 0:00:41 loss: 0.4618 (0.4618) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.6152 data: 4.4006 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6345 (0.6498) acc1: 84.3750 (81.8560) acc5: 97.3958 (96.7040) time: 0.6647 data: 0.4891 max mem: 64948 Test: Total time: 0:00:06 (0.6739 s / it) * Acc@1 83.160 Acc@5 96.514 loss 0.626 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.16% Epoch: [348] [ 0/312] eta: 0:49:53 lr: 0.000531 min_lr: 0.000531 loss: 1.5232 (1.5232) weight_decay: 0.0500 (0.0500) time: 9.5945 data: 8.7822 max mem: 64948 Epoch: [348] [ 10/312] eta: 0:07:49 lr: 0.000531 min_lr: 0.000531 loss: 1.4777 (1.5737) weight_decay: 0.0500 (0.0500) time: 1.5547 data: 0.8022 max mem: 64948 Epoch: [348] [ 20/312] eta: 0:05:34 lr: 0.000530 min_lr: 0.000530 loss: 1.6233 (1.6533) weight_decay: 0.0500 (0.0500) time: 0.7231 data: 0.0023 max mem: 64948 Epoch: [348] [ 30/312] eta: 0:04:42 lr: 0.000530 min_lr: 0.000530 loss: 1.7208 (1.6748) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [348] [ 40/312] eta: 0:04:11 lr: 0.000530 min_lr: 0.000530 loss: 1.7208 (1.6966) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [348] [ 50/312] eta: 0:03:51 lr: 0.000529 min_lr: 0.000529 loss: 1.9432 (1.7287) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [348] [ 60/312] eta: 0:03:34 lr: 0.000529 min_lr: 0.000529 loss: 1.8354 (1.7371) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [348] [ 70/312] eta: 0:03:20 lr: 0.000529 min_lr: 0.000529 loss: 1.7872 (1.7500) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [348] [ 80/312] eta: 0:03:08 lr: 0.000528 min_lr: 0.000528 loss: 1.8120 (1.7511) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [348] [ 90/312] eta: 0:02:57 lr: 0.000528 min_lr: 0.000528 loss: 1.8120 (1.7522) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [348] [100/312] eta: 0:02:47 lr: 0.000528 min_lr: 0.000528 loss: 1.7481 (1.7456) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0005 max mem: 64948 Epoch: [348] [110/312] eta: 0:02:37 lr: 0.000527 min_lr: 0.000527 loss: 1.8174 (1.7689) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0005 max mem: 64948 Epoch: [348] [120/312] eta: 0:02:28 lr: 0.000527 min_lr: 0.000527 loss: 1.7881 (1.7622) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0005 max mem: 64948 Epoch: [348] [130/312] eta: 0:02:19 lr: 0.000527 min_lr: 0.000527 loss: 1.7855 (1.7784) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0005 max mem: 64948 Epoch: [348] [140/312] eta: 0:02:11 lr: 0.000527 min_lr: 0.000527 loss: 1.9183 (1.7810) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0005 max mem: 64948 Epoch: [348] [150/312] eta: 0:02:02 lr: 0.000526 min_lr: 0.000526 loss: 1.8286 (1.7832) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0005 max mem: 64948 Epoch: [348] [160/312] eta: 0:01:54 lr: 0.000526 min_lr: 0.000526 loss: 1.8956 (1.7927) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0005 max mem: 64948 Epoch: [348] [170/312] eta: 0:01:46 lr: 0.000526 min_lr: 0.000526 loss: 1.8625 (1.7888) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0005 max mem: 64948 Epoch: [348] [180/312] eta: 0:01:38 lr: 0.000525 min_lr: 0.000525 loss: 1.7300 (1.7800) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0005 max mem: 64948 Epoch: [348] [190/312] eta: 0:01:30 lr: 0.000525 min_lr: 0.000525 loss: 1.5048 (1.7655) weight_decay: 0.0500 (0.0500) time: 0.7009 data: 0.0005 max mem: 64948 Epoch: [348] [200/312] eta: 0:01:23 lr: 0.000525 min_lr: 0.000525 loss: 1.6895 (1.7703) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0005 max mem: 64948 Epoch: [348] [210/312] eta: 0:01:15 lr: 0.000524 min_lr: 0.000524 loss: 1.7644 (1.7666) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0005 max mem: 64948 Epoch: [348] [220/312] eta: 0:01:07 lr: 0.000524 min_lr: 0.000524 loss: 1.6155 (1.7645) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0006 max mem: 64948 Epoch: [348] [230/312] eta: 0:01:00 lr: 0.000524 min_lr: 0.000524 loss: 1.7176 (1.7692) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0006 max mem: 64948 Epoch: [348] [240/312] eta: 0:00:52 lr: 0.000523 min_lr: 0.000523 loss: 1.9012 (1.7755) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0005 max mem: 64948 Epoch: [348] [250/312] eta: 0:00:45 lr: 0.000523 min_lr: 0.000523 loss: 1.9526 (1.7801) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0005 max mem: 64948 Epoch: [348] [260/312] eta: 0:00:38 lr: 0.000523 min_lr: 0.000523 loss: 1.8887 (1.7785) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0006 max mem: 64948 Epoch: [348] [270/312] eta: 0:00:30 lr: 0.000522 min_lr: 0.000522 loss: 1.8050 (1.7798) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0005 max mem: 64948 Epoch: [348] [280/312] eta: 0:00:23 lr: 0.000522 min_lr: 0.000522 loss: 1.8078 (1.7772) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0014 max mem: 64948 Epoch: [348] [290/312] eta: 0:00:16 lr: 0.000522 min_lr: 0.000522 loss: 1.8078 (1.7774) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0012 max mem: 64948 Epoch: [348] [300/312] eta: 0:00:08 lr: 0.000521 min_lr: 0.000521 loss: 1.7488 (1.7742) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [348] [310/312] eta: 0:00:01 lr: 0.000521 min_lr: 0.000521 loss: 1.7819 (1.7774) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [348] [311/312] eta: 0:00:00 lr: 0.000521 min_lr: 0.000521 loss: 1.7956 (1.7775) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [348] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.000521 min_lr: 0.000521 loss: 1.7956 (1.7895) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4743 (0.4743) acc1: 88.8021 (88.8021) acc5: 98.4375 (98.4375) time: 4.8086 data: 4.6037 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6761 (0.6930) acc1: 82.5521 (81.4400) acc5: 96.8750 (96.2560) time: 0.6855 data: 0.5116 max mem: 64948 Test: Total time: 0:00:06 (0.7127 s / it) * Acc@1 82.506 Acc@5 96.276 loss 0.663 Accuracy of the model on the 50000 test images: 82.5% Max accuracy: 82.51% Test: [0/9] eta: 0:00:44 loss: 0.4622 (0.4622) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.9099 data: 4.7076 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6343 (0.6498) acc1: 84.6354 (81.9200) acc5: 97.1354 (96.6720) time: 0.6968 data: 0.5232 max mem: 64948 Test: Total time: 0:00:06 (0.7050 s / it) * Acc@1 83.170 Acc@5 96.528 loss 0.626 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.17% Epoch: [349] [ 0/312] eta: 0:51:28 lr: 0.000521 min_lr: 0.000521 loss: 1.4477 (1.4477) weight_decay: 0.0500 (0.0500) time: 9.8980 data: 8.5302 max mem: 64948 Epoch: [349] [ 10/312] eta: 0:07:49 lr: 0.000521 min_lr: 0.000521 loss: 2.0351 (1.9104) weight_decay: 0.0500 (0.0500) time: 1.5558 data: 0.7760 max mem: 64948 Epoch: [349] [ 20/312] eta: 0:05:35 lr: 0.000520 min_lr: 0.000520 loss: 1.9415 (1.9038) weight_decay: 0.0500 (0.0500) time: 0.7117 data: 0.0005 max mem: 64948 Epoch: [349] [ 30/312] eta: 0:04:43 lr: 0.000520 min_lr: 0.000520 loss: 1.8840 (1.8826) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0004 max mem: 64948 Epoch: [349] [ 40/312] eta: 0:04:12 lr: 0.000520 min_lr: 0.000520 loss: 1.7444 (1.8495) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [349] [ 50/312] eta: 0:03:51 lr: 0.000520 min_lr: 0.000520 loss: 1.8748 (1.8561) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [349] [ 60/312] eta: 0:03:35 lr: 0.000519 min_lr: 0.000519 loss: 1.8748 (1.8404) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [349] [ 70/312] eta: 0:03:21 lr: 0.000519 min_lr: 0.000519 loss: 1.6575 (1.7989) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [349] [ 80/312] eta: 0:03:09 lr: 0.000519 min_lr: 0.000519 loss: 1.5416 (1.7755) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [349] [ 90/312] eta: 0:02:57 lr: 0.000518 min_lr: 0.000518 loss: 1.6206 (1.7697) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [349] [100/312] eta: 0:02:47 lr: 0.000518 min_lr: 0.000518 loss: 1.8230 (1.7691) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [349] [110/312] eta: 0:02:38 lr: 0.000518 min_lr: 0.000518 loss: 1.8404 (1.7728) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [349] [120/312] eta: 0:02:28 lr: 0.000517 min_lr: 0.000517 loss: 1.7956 (1.7625) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [349] [130/312] eta: 0:02:19 lr: 0.000517 min_lr: 0.000517 loss: 1.8083 (1.7603) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [349] [140/312] eta: 0:02:11 lr: 0.000517 min_lr: 0.000517 loss: 1.7418 (1.7587) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [349] [150/312] eta: 0:02:03 lr: 0.000516 min_lr: 0.000516 loss: 1.7150 (1.7551) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [349] [160/312] eta: 0:01:54 lr: 0.000516 min_lr: 0.000516 loss: 1.6638 (1.7526) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [349] [170/312] eta: 0:01:46 lr: 0.000516 min_lr: 0.000516 loss: 1.8269 (1.7580) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [349] [180/312] eta: 0:01:38 lr: 0.000515 min_lr: 0.000515 loss: 1.8581 (1.7628) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [349] [190/312] eta: 0:01:31 lr: 0.000515 min_lr: 0.000515 loss: 1.8541 (1.7637) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [349] [200/312] eta: 0:01:23 lr: 0.000515 min_lr: 0.000515 loss: 1.8442 (1.7656) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [349] [210/312] eta: 0:01:15 lr: 0.000515 min_lr: 0.000515 loss: 1.8021 (1.7547) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [349] [220/312] eta: 0:01:08 lr: 0.000514 min_lr: 0.000514 loss: 1.7841 (1.7584) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [349] [230/312] eta: 0:01:00 lr: 0.000514 min_lr: 0.000514 loss: 1.9217 (1.7627) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [349] [240/312] eta: 0:00:52 lr: 0.000514 min_lr: 0.000514 loss: 1.9248 (1.7688) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [349] [250/312] eta: 0:00:45 lr: 0.000513 min_lr: 0.000513 loss: 1.7805 (1.7662) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [349] [260/312] eta: 0:00:38 lr: 0.000513 min_lr: 0.000513 loss: 1.6612 (1.7620) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [349] [270/312] eta: 0:00:30 lr: 0.000513 min_lr: 0.000513 loss: 1.8308 (1.7654) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [349] [280/312] eta: 0:00:23 lr: 0.000512 min_lr: 0.000512 loss: 1.7675 (1.7635) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [349] [290/312] eta: 0:00:16 lr: 0.000512 min_lr: 0.000512 loss: 1.9497 (1.7667) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [349] [300/312] eta: 0:00:08 lr: 0.000512 min_lr: 0.000512 loss: 1.9497 (1.7652) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [349] [310/312] eta: 0:00:01 lr: 0.000511 min_lr: 0.000511 loss: 1.8342 (1.7662) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [349] [311/312] eta: 0:00:00 lr: 0.000511 min_lr: 0.000511 loss: 1.8342 (1.7668) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [349] Total time: 0:03:47 (0.7289 s / it) Averaged stats: lr: 0.000511 min_lr: 0.000511 loss: 1.8342 (1.7954) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4882 (0.4882) acc1: 86.4583 (86.4583) acc5: 97.6562 (97.6562) time: 4.8206 data: 4.6073 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6409 (0.6830) acc1: 83.0729 (81.7600) acc5: 97.3958 (96.3200) time: 0.6869 data: 0.5120 max mem: 64948 Test: Total time: 0:00:06 (0.7143 s / it) * Acc@1 82.476 Acc@5 96.094 loss 0.671 Accuracy of the model on the 50000 test images: 82.5% Max accuracy: 82.51% Test: [0/9] eta: 0:00:48 loss: 0.4624 (0.4624) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.4216 data: 5.2035 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6339 (0.6498) acc1: 84.6354 (81.9200) acc5: 97.1354 (96.6400) time: 0.7541 data: 0.5783 max mem: 64948 Test: Total time: 0:00:06 (0.7664 s / it) * Acc@1 83.162 Acc@5 96.536 loss 0.626 Accuracy of the model EMA on 50000 test images: 83.2% Epoch: [350] [ 0/312] eta: 1:00:07 lr: 0.000511 min_lr: 0.000511 loss: 1.9254 (1.9254) weight_decay: 0.0500 (0.0500) time: 11.5638 data: 9.5015 max mem: 64948 Epoch: [350] [ 10/312] eta: 0:08:35 lr: 0.000511 min_lr: 0.000511 loss: 1.9254 (1.9009) weight_decay: 0.0500 (0.0500) time: 1.7085 data: 0.8642 max mem: 64948 Epoch: [350] [ 20/312] eta: 0:05:58 lr: 0.000511 min_lr: 0.000511 loss: 1.9371 (1.9202) weight_decay: 0.0500 (0.0500) time: 0.7094 data: 0.0004 max mem: 64948 Epoch: [350] [ 30/312] eta: 0:04:57 lr: 0.000510 min_lr: 0.000510 loss: 1.8810 (1.8884) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [350] [ 40/312] eta: 0:04:23 lr: 0.000510 min_lr: 0.000510 loss: 1.6266 (1.8148) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [350] [ 50/312] eta: 0:03:59 lr: 0.000510 min_lr: 0.000510 loss: 1.4932 (1.7677) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [350] [ 60/312] eta: 0:03:41 lr: 0.000509 min_lr: 0.000509 loss: 1.6978 (1.7540) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [350] [ 70/312] eta: 0:03:26 lr: 0.000509 min_lr: 0.000509 loss: 1.7778 (1.7454) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [350] [ 80/312] eta: 0:03:13 lr: 0.000509 min_lr: 0.000509 loss: 1.7984 (1.7484) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [350] [ 90/312] eta: 0:03:01 lr: 0.000509 min_lr: 0.000509 loss: 1.8575 (1.7568) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [350] [100/312] eta: 0:02:51 lr: 0.000508 min_lr: 0.000508 loss: 1.8830 (1.7631) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [350] [110/312] eta: 0:02:41 lr: 0.000508 min_lr: 0.000508 loss: 1.9040 (1.7768) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [350] [120/312] eta: 0:02:31 lr: 0.000508 min_lr: 0.000508 loss: 1.9855 (1.7823) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [350] [130/312] eta: 0:02:22 lr: 0.000507 min_lr: 0.000507 loss: 1.9433 (1.7922) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [350] [140/312] eta: 0:02:13 lr: 0.000507 min_lr: 0.000507 loss: 1.9260 (1.7952) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [350] [150/312] eta: 0:02:04 lr: 0.000507 min_lr: 0.000507 loss: 1.9081 (1.7955) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [350] [160/312] eta: 0:01:56 lr: 0.000506 min_lr: 0.000506 loss: 1.7614 (1.7929) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [350] [170/312] eta: 0:01:48 lr: 0.000506 min_lr: 0.000506 loss: 1.9060 (1.7948) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [350] [180/312] eta: 0:01:40 lr: 0.000506 min_lr: 0.000506 loss: 1.8023 (1.7902) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [350] [190/312] eta: 0:01:32 lr: 0.000505 min_lr: 0.000505 loss: 1.7481 (1.7858) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [350] [200/312] eta: 0:01:24 lr: 0.000505 min_lr: 0.000505 loss: 1.8571 (1.7917) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [350] [210/312] eta: 0:01:16 lr: 0.000505 min_lr: 0.000505 loss: 1.8651 (1.7920) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [350] [220/312] eta: 0:01:08 lr: 0.000504 min_lr: 0.000504 loss: 1.8998 (1.7947) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [350] [230/312] eta: 0:01:01 lr: 0.000504 min_lr: 0.000504 loss: 1.8844 (1.7943) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [350] [240/312] eta: 0:00:53 lr: 0.000504 min_lr: 0.000504 loss: 1.8646 (1.7977) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [350] [250/312] eta: 0:00:45 lr: 0.000504 min_lr: 0.000504 loss: 1.9192 (1.7975) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [350] [260/312] eta: 0:00:38 lr: 0.000503 min_lr: 0.000503 loss: 1.9211 (1.7982) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [350] [270/312] eta: 0:00:30 lr: 0.000503 min_lr: 0.000503 loss: 1.9211 (1.7996) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [350] [280/312] eta: 0:00:23 lr: 0.000503 min_lr: 0.000503 loss: 1.9013 (1.7993) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [350] [290/312] eta: 0:00:16 lr: 0.000502 min_lr: 0.000502 loss: 1.9726 (1.8056) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [350] [300/312] eta: 0:00:08 lr: 0.000502 min_lr: 0.000502 loss: 1.9220 (1.8028) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [350] [310/312] eta: 0:00:01 lr: 0.000502 min_lr: 0.000502 loss: 1.8019 (1.8027) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [350] [311/312] eta: 0:00:00 lr: 0.000502 min_lr: 0.000502 loss: 1.8019 (1.8010) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [350] Total time: 0:03:49 (0.7342 s / it) Averaged stats: lr: 0.000502 min_lr: 0.000502 loss: 1.8019 (1.7898) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4541 (0.4541) acc1: 87.2396 (87.2396) acc5: 98.4375 (98.4375) time: 4.7529 data: 4.5443 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6819 (0.6865) acc1: 83.5938 (81.8560) acc5: 96.3542 (96.1280) time: 0.6794 data: 0.5050 max mem: 64948 Test: Total time: 0:00:06 (0.7051 s / it) * Acc@1 82.484 Acc@5 96.144 loss 0.667 Accuracy of the model on the 50000 test images: 82.5% Max accuracy: 82.51% Test: [0/9] eta: 0:00:47 loss: 0.4620 (0.4620) acc1: 88.0208 (88.0208) acc5: 97.9167 (97.9167) time: 5.2795 data: 5.0726 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6334 (0.6494) acc1: 84.6354 (82.0160) acc5: 97.1354 (96.6400) time: 0.7379 data: 0.5637 max mem: 64948 Test: Total time: 0:00:06 (0.7525 s / it) * Acc@1 83.170 Acc@5 96.534 loss 0.626 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.17% Epoch: [351] [ 0/312] eta: 0:56:09 lr: 0.000502 min_lr: 0.000502 loss: 1.4652 (1.4652) weight_decay: 0.0500 (0.0500) time: 10.7997 data: 10.0387 max mem: 64948 Epoch: [351] [ 10/312] eta: 0:08:10 lr: 0.000501 min_lr: 0.000501 loss: 1.8736 (1.7603) weight_decay: 0.0500 (0.0500) time: 1.6235 data: 0.9129 max mem: 64948 Epoch: [351] [ 20/312] eta: 0:05:44 lr: 0.000501 min_lr: 0.000501 loss: 1.9605 (1.8054) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [351] [ 30/312] eta: 0:04:48 lr: 0.000501 min_lr: 0.000501 loss: 1.9209 (1.7684) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [351] [ 40/312] eta: 0:04:17 lr: 0.000500 min_lr: 0.000500 loss: 1.8166 (1.8078) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [351] [ 50/312] eta: 0:03:55 lr: 0.000500 min_lr: 0.000500 loss: 1.9274 (1.8081) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [351] [ 60/312] eta: 0:03:37 lr: 0.000500 min_lr: 0.000500 loss: 1.8681 (1.8103) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [351] [ 70/312] eta: 0:03:23 lr: 0.000499 min_lr: 0.000499 loss: 1.8620 (1.8062) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [351] [ 80/312] eta: 0:03:10 lr: 0.000499 min_lr: 0.000499 loss: 1.6774 (1.7906) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [351] [ 90/312] eta: 0:02:59 lr: 0.000499 min_lr: 0.000499 loss: 1.8509 (1.8039) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [351] [100/312] eta: 0:02:49 lr: 0.000499 min_lr: 0.000499 loss: 2.0487 (1.8078) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [351] [110/312] eta: 0:02:39 lr: 0.000498 min_lr: 0.000498 loss: 1.9334 (1.8038) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [351] [120/312] eta: 0:02:30 lr: 0.000498 min_lr: 0.000498 loss: 1.6824 (1.7937) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [351] [130/312] eta: 0:02:21 lr: 0.000498 min_lr: 0.000498 loss: 1.6748 (1.7861) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [351] [140/312] eta: 0:02:12 lr: 0.000497 min_lr: 0.000497 loss: 1.6748 (1.7813) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [351] [150/312] eta: 0:02:03 lr: 0.000497 min_lr: 0.000497 loss: 1.6384 (1.7681) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [351] [160/312] eta: 0:01:55 lr: 0.000497 min_lr: 0.000497 loss: 1.6450 (1.7671) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [351] [170/312] eta: 0:01:47 lr: 0.000496 min_lr: 0.000496 loss: 1.8061 (1.7691) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [351] [180/312] eta: 0:01:39 lr: 0.000496 min_lr: 0.000496 loss: 1.8731 (1.7718) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [351] [190/312] eta: 0:01:31 lr: 0.000496 min_lr: 0.000496 loss: 1.8845 (1.7707) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [351] [200/312] eta: 0:01:23 lr: 0.000495 min_lr: 0.000495 loss: 1.8869 (1.7790) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [351] [210/312] eta: 0:01:15 lr: 0.000495 min_lr: 0.000495 loss: 1.7591 (1.7669) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [351] [220/312] eta: 0:01:08 lr: 0.000495 min_lr: 0.000495 loss: 1.7198 (1.7706) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [351] [230/312] eta: 0:01:00 lr: 0.000495 min_lr: 0.000495 loss: 1.8203 (1.7726) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [351] [240/312] eta: 0:00:53 lr: 0.000494 min_lr: 0.000494 loss: 1.7733 (1.7720) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [351] [250/312] eta: 0:00:45 lr: 0.000494 min_lr: 0.000494 loss: 1.6672 (1.7677) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [351] [260/312] eta: 0:00:38 lr: 0.000494 min_lr: 0.000494 loss: 1.7904 (1.7693) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [351] [270/312] eta: 0:00:30 lr: 0.000493 min_lr: 0.000493 loss: 1.8423 (1.7727) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [351] [280/312] eta: 0:00:23 lr: 0.000493 min_lr: 0.000493 loss: 1.9173 (1.7782) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [351] [290/312] eta: 0:00:16 lr: 0.000493 min_lr: 0.000493 loss: 1.8761 (1.7741) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [351] [300/312] eta: 0:00:08 lr: 0.000492 min_lr: 0.000492 loss: 1.8483 (1.7741) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [351] [310/312] eta: 0:00:01 lr: 0.000492 min_lr: 0.000492 loss: 1.7301 (1.7748) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [351] [311/312] eta: 0:00:00 lr: 0.000492 min_lr: 0.000492 loss: 1.7301 (1.7738) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [351] Total time: 0:03:48 (0.7311 s / it) Averaged stats: lr: 0.000492 min_lr: 0.000492 loss: 1.7301 (1.7828) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.4676 (0.4676) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.9774 data: 4.7581 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6833 (0.6928) acc1: 82.0312 (80.7680) acc5: 96.8750 (96.1280) time: 0.7043 data: 0.5288 max mem: 64948 Test: Total time: 0:00:06 (0.7133 s / it) * Acc@1 82.456 Acc@5 96.240 loss 0.670 Accuracy of the model on the 50000 test images: 82.5% Max accuracy: 82.51% Test: [0/9] eta: 0:00:45 loss: 0.4617 (0.4617) acc1: 88.0208 (88.0208) acc5: 97.9167 (97.9167) time: 5.0676 data: 4.8496 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6331 (0.6492) acc1: 84.3750 (82.0480) acc5: 97.1354 (96.6400) time: 0.7149 data: 0.5389 max mem: 64948 Test: Total time: 0:00:06 (0.7246 s / it) * Acc@1 83.182 Acc@5 96.552 loss 0.626 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.18% Epoch: [352] [ 0/312] eta: 0:56:44 lr: 0.000492 min_lr: 0.000492 loss: 2.1130 (2.1130) weight_decay: 0.0500 (0.0500) time: 10.9110 data: 10.1892 max mem: 64948 Epoch: [352] [ 10/312] eta: 0:08:21 lr: 0.000492 min_lr: 0.000492 loss: 1.7916 (1.7890) weight_decay: 0.0500 (0.0500) time: 1.6612 data: 0.9266 max mem: 64948 Epoch: [352] [ 20/312] eta: 0:05:50 lr: 0.000491 min_lr: 0.000491 loss: 1.8659 (1.7875) weight_decay: 0.0500 (0.0500) time: 0.7144 data: 0.0004 max mem: 64948 Epoch: [352] [ 30/312] eta: 0:04:52 lr: 0.000491 min_lr: 0.000491 loss: 1.9465 (1.8609) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [352] [ 40/312] eta: 0:04:19 lr: 0.000491 min_lr: 0.000491 loss: 1.8951 (1.8249) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [352] [ 50/312] eta: 0:03:56 lr: 0.000490 min_lr: 0.000490 loss: 1.7570 (1.8147) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [352] [ 60/312] eta: 0:03:39 lr: 0.000490 min_lr: 0.000490 loss: 1.9422 (1.8337) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [352] [ 70/312] eta: 0:03:24 lr: 0.000490 min_lr: 0.000490 loss: 1.9701 (1.8281) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [352] [ 80/312] eta: 0:03:11 lr: 0.000490 min_lr: 0.000490 loss: 1.7424 (1.8050) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [352] [ 90/312] eta: 0:03:00 lr: 0.000489 min_lr: 0.000489 loss: 1.7881 (1.8084) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [352] [100/312] eta: 0:02:49 lr: 0.000489 min_lr: 0.000489 loss: 1.8355 (1.8098) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [352] [110/312] eta: 0:02:39 lr: 0.000489 min_lr: 0.000489 loss: 1.8851 (1.8152) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [352] [120/312] eta: 0:02:30 lr: 0.000488 min_lr: 0.000488 loss: 1.9049 (1.8197) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [352] [130/312] eta: 0:02:21 lr: 0.000488 min_lr: 0.000488 loss: 1.8445 (1.8195) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [352] [140/312] eta: 0:02:12 lr: 0.000488 min_lr: 0.000488 loss: 1.9981 (1.8277) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [352] [150/312] eta: 0:02:04 lr: 0.000487 min_lr: 0.000487 loss: 1.8330 (1.8187) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [352] [160/312] eta: 0:01:55 lr: 0.000487 min_lr: 0.000487 loss: 1.7555 (1.8178) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [352] [170/312] eta: 0:01:47 lr: 0.000487 min_lr: 0.000487 loss: 1.7030 (1.8112) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [352] [180/312] eta: 0:01:39 lr: 0.000486 min_lr: 0.000486 loss: 1.7620 (1.8175) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [352] [190/312] eta: 0:01:31 lr: 0.000486 min_lr: 0.000486 loss: 1.9419 (1.8168) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [352] [200/312] eta: 0:01:23 lr: 0.000486 min_lr: 0.000486 loss: 1.8078 (1.8091) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [352] [210/312] eta: 0:01:16 lr: 0.000486 min_lr: 0.000486 loss: 1.7207 (1.8080) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [352] [220/312] eta: 0:01:08 lr: 0.000485 min_lr: 0.000485 loss: 1.7207 (1.8057) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [352] [230/312] eta: 0:01:00 lr: 0.000485 min_lr: 0.000485 loss: 1.8342 (1.8083) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [352] [240/312] eta: 0:00:53 lr: 0.000485 min_lr: 0.000485 loss: 1.9177 (1.8062) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [352] [250/312] eta: 0:00:45 lr: 0.000484 min_lr: 0.000484 loss: 1.9003 (1.8091) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [352] [260/312] eta: 0:00:38 lr: 0.000484 min_lr: 0.000484 loss: 1.8914 (1.8041) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [352] [270/312] eta: 0:00:30 lr: 0.000484 min_lr: 0.000484 loss: 1.4982 (1.8005) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [352] [280/312] eta: 0:00:23 lr: 0.000483 min_lr: 0.000483 loss: 1.6639 (1.7969) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [352] [290/312] eta: 0:00:16 lr: 0.000483 min_lr: 0.000483 loss: 1.6536 (1.7922) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [352] [300/312] eta: 0:00:08 lr: 0.000483 min_lr: 0.000483 loss: 1.7857 (1.7933) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [352] [310/312] eta: 0:00:01 lr: 0.000483 min_lr: 0.000483 loss: 1.9572 (1.7983) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [352] [311/312] eta: 0:00:00 lr: 0.000482 min_lr: 0.000482 loss: 1.9786 (1.7994) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [352] Total time: 0:03:48 (0.7319 s / it) Averaged stats: lr: 0.000482 min_lr: 0.000482 loss: 1.9786 (1.7925) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:45 loss: 0.4619 (0.4619) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 5.0008 data: 4.7806 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6513 (0.6849) acc1: 83.0729 (81.3120) acc5: 96.6146 (96.3840) time: 0.7073 data: 0.5313 max mem: 64948 Test: Total time: 0:00:06 (0.7343 s / it) * Acc@1 82.586 Acc@5 96.310 loss 0.655 Accuracy of the model on the 50000 test images: 82.6% Max accuracy: 82.59% Test: [0/9] eta: 0:00:43 loss: 0.4614 (0.4614) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 4.8557 data: 4.6497 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6332 (0.6492) acc1: 84.3750 (82.0160) acc5: 97.1354 (96.6080) time: 0.6910 data: 0.5168 max mem: 64948 Test: Total time: 0:00:06 (0.6998 s / it) * Acc@1 83.200 Acc@5 96.548 loss 0.626 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.20% Epoch: [353] [ 0/312] eta: 0:54:27 lr: 0.000482 min_lr: 0.000482 loss: 2.1539 (2.1539) weight_decay: 0.0500 (0.0500) time: 10.4723 data: 9.6994 max mem: 64948 Epoch: [353] [ 10/312] eta: 0:08:01 lr: 0.000482 min_lr: 0.000482 loss: 1.8815 (1.7429) weight_decay: 0.0500 (0.0500) time: 1.5952 data: 0.8821 max mem: 64948 Epoch: [353] [ 20/312] eta: 0:05:40 lr: 0.000482 min_lr: 0.000482 loss: 1.8815 (1.7489) weight_decay: 0.0500 (0.0500) time: 0.7009 data: 0.0004 max mem: 64948 Epoch: [353] [ 30/312] eta: 0:04:46 lr: 0.000482 min_lr: 0.000482 loss: 1.5828 (1.6779) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [353] [ 40/312] eta: 0:04:14 lr: 0.000481 min_lr: 0.000481 loss: 1.7289 (1.7439) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [353] [ 50/312] eta: 0:03:53 lr: 0.000481 min_lr: 0.000481 loss: 1.9276 (1.7474) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [353] [ 60/312] eta: 0:03:36 lr: 0.000481 min_lr: 0.000481 loss: 1.8246 (1.7589) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [353] [ 70/312] eta: 0:03:22 lr: 0.000480 min_lr: 0.000480 loss: 1.8799 (1.7738) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [353] [ 80/312] eta: 0:03:09 lr: 0.000480 min_lr: 0.000480 loss: 1.8199 (1.7604) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [353] [ 90/312] eta: 0:02:58 lr: 0.000480 min_lr: 0.000480 loss: 1.8025 (1.7611) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [353] [100/312] eta: 0:02:48 lr: 0.000479 min_lr: 0.000479 loss: 1.8066 (1.7664) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [353] [110/312] eta: 0:02:38 lr: 0.000479 min_lr: 0.000479 loss: 1.8066 (1.7652) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [353] [120/312] eta: 0:02:29 lr: 0.000479 min_lr: 0.000479 loss: 1.6392 (1.7499) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [353] [130/312] eta: 0:02:20 lr: 0.000478 min_lr: 0.000478 loss: 1.6392 (1.7536) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [353] [140/312] eta: 0:02:11 lr: 0.000478 min_lr: 0.000478 loss: 1.8748 (1.7596) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [353] [150/312] eta: 0:02:03 lr: 0.000478 min_lr: 0.000478 loss: 1.8785 (1.7644) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [353] [160/312] eta: 0:01:55 lr: 0.000478 min_lr: 0.000478 loss: 1.8477 (1.7630) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [353] [170/312] eta: 0:01:47 lr: 0.000477 min_lr: 0.000477 loss: 1.8040 (1.7663) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [353] [180/312] eta: 0:01:39 lr: 0.000477 min_lr: 0.000477 loss: 1.8417 (1.7700) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [353] [190/312] eta: 0:01:31 lr: 0.000477 min_lr: 0.000477 loss: 1.7042 (1.7653) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [353] [200/312] eta: 0:01:23 lr: 0.000476 min_lr: 0.000476 loss: 1.7042 (1.7651) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [353] [210/312] eta: 0:01:15 lr: 0.000476 min_lr: 0.000476 loss: 1.8459 (1.7722) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [353] [220/312] eta: 0:01:08 lr: 0.000476 min_lr: 0.000476 loss: 1.8383 (1.7696) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [353] [230/312] eta: 0:01:00 lr: 0.000475 min_lr: 0.000475 loss: 1.8383 (1.7753) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [353] [240/312] eta: 0:00:53 lr: 0.000475 min_lr: 0.000475 loss: 1.7468 (1.7684) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [353] [250/312] eta: 0:00:45 lr: 0.000475 min_lr: 0.000475 loss: 1.6659 (1.7663) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [353] [260/312] eta: 0:00:38 lr: 0.000475 min_lr: 0.000475 loss: 1.6659 (1.7639) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [353] [270/312] eta: 0:00:30 lr: 0.000474 min_lr: 0.000474 loss: 1.6200 (1.7589) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [353] [280/312] eta: 0:00:23 lr: 0.000474 min_lr: 0.000474 loss: 1.7133 (1.7597) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [353] [290/312] eta: 0:00:16 lr: 0.000474 min_lr: 0.000474 loss: 1.7492 (1.7582) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [353] [300/312] eta: 0:00:08 lr: 0.000473 min_lr: 0.000473 loss: 1.8595 (1.7601) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [353] [310/312] eta: 0:00:01 lr: 0.000473 min_lr: 0.000473 loss: 1.7752 (1.7574) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [353] [311/312] eta: 0:00:00 lr: 0.000473 min_lr: 0.000473 loss: 1.7413 (1.7564) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [353] Total time: 0:03:47 (0.7302 s / it) Averaged stats: lr: 0.000473 min_lr: 0.000473 loss: 1.7413 (1.7802) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4443 (0.4443) acc1: 88.8021 (88.8021) acc5: 98.6979 (98.6979) time: 4.7309 data: 4.5254 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6944 (0.6827) acc1: 82.0312 (81.5040) acc5: 96.6146 (96.2880) time: 0.6770 data: 0.5029 max mem: 64948 Test: Total time: 0:00:06 (0.7050 s / it) * Acc@1 82.542 Acc@5 96.234 loss 0.671 Accuracy of the model on the 50000 test images: 82.5% Max accuracy: 82.59% Test: [0/9] eta: 0:00:45 loss: 0.4610 (0.4610) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 5.0552 data: 4.8440 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6331 (0.6490) acc1: 84.3750 (82.0480) acc5: 97.1354 (96.5440) time: 0.7130 data: 0.5383 max mem: 64948 Test: Total time: 0:00:06 (0.7251 s / it) * Acc@1 83.198 Acc@5 96.546 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Epoch: [354] [ 0/312] eta: 0:59:13 lr: 0.000473 min_lr: 0.000473 loss: 2.1523 (2.1523) weight_decay: 0.0500 (0.0500) time: 11.3883 data: 5.9857 max mem: 64948 Epoch: [354] [ 10/312] eta: 0:08:30 lr: 0.000473 min_lr: 0.000473 loss: 1.8735 (1.9078) weight_decay: 0.0500 (0.0500) time: 1.6899 data: 0.5446 max mem: 64948 Epoch: [354] [ 20/312] eta: 0:05:55 lr: 0.000472 min_lr: 0.000472 loss: 1.8835 (1.9039) weight_decay: 0.0500 (0.0500) time: 0.7081 data: 0.0004 max mem: 64948 Epoch: [354] [ 30/312] eta: 0:04:55 lr: 0.000472 min_lr: 0.000472 loss: 1.7962 (1.8342) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [354] [ 40/312] eta: 0:04:21 lr: 0.000472 min_lr: 0.000472 loss: 1.6790 (1.7923) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [354] [ 50/312] eta: 0:03:58 lr: 0.000471 min_lr: 0.000471 loss: 1.7288 (1.8030) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [354] [ 60/312] eta: 0:03:40 lr: 0.000471 min_lr: 0.000471 loss: 1.8275 (1.8068) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [354] [ 70/312] eta: 0:03:25 lr: 0.000471 min_lr: 0.000471 loss: 1.8370 (1.8110) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [354] [ 80/312] eta: 0:03:12 lr: 0.000471 min_lr: 0.000471 loss: 1.8370 (1.8132) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [354] [ 90/312] eta: 0:03:01 lr: 0.000470 min_lr: 0.000470 loss: 1.9530 (1.8245) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [354] [100/312] eta: 0:02:50 lr: 0.000470 min_lr: 0.000470 loss: 1.8552 (1.8053) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [354] [110/312] eta: 0:02:40 lr: 0.000470 min_lr: 0.000470 loss: 1.6306 (1.8083) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [354] [120/312] eta: 0:02:30 lr: 0.000469 min_lr: 0.000469 loss: 1.7180 (1.8077) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [354] [130/312] eta: 0:02:21 lr: 0.000469 min_lr: 0.000469 loss: 1.8201 (1.8099) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [354] [140/312] eta: 0:02:13 lr: 0.000469 min_lr: 0.000469 loss: 1.8201 (1.8022) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [354] [150/312] eta: 0:02:04 lr: 0.000468 min_lr: 0.000468 loss: 1.6758 (1.7939) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [354] [160/312] eta: 0:01:56 lr: 0.000468 min_lr: 0.000468 loss: 1.6621 (1.7887) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [354] [170/312] eta: 0:01:47 lr: 0.000468 min_lr: 0.000468 loss: 1.7999 (1.7913) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [354] [180/312] eta: 0:01:39 lr: 0.000468 min_lr: 0.000468 loss: 1.7514 (1.7884) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [354] [190/312] eta: 0:01:31 lr: 0.000467 min_lr: 0.000467 loss: 1.7514 (1.7880) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [354] [200/312] eta: 0:01:23 lr: 0.000467 min_lr: 0.000467 loss: 1.7594 (1.7872) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [354] [210/312] eta: 0:01:16 lr: 0.000467 min_lr: 0.000467 loss: 1.7532 (1.7814) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [354] [220/312] eta: 0:01:08 lr: 0.000466 min_lr: 0.000466 loss: 1.8880 (1.7874) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [354] [230/312] eta: 0:01:00 lr: 0.000466 min_lr: 0.000466 loss: 1.9118 (1.7926) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [354] [240/312] eta: 0:00:53 lr: 0.000466 min_lr: 0.000466 loss: 1.8846 (1.7904) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [354] [250/312] eta: 0:00:45 lr: 0.000465 min_lr: 0.000465 loss: 1.8013 (1.7880) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [354] [260/312] eta: 0:00:38 lr: 0.000465 min_lr: 0.000465 loss: 1.7050 (1.7840) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [354] [270/312] eta: 0:00:30 lr: 0.000465 min_lr: 0.000465 loss: 1.7050 (1.7822) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [354] [280/312] eta: 0:00:23 lr: 0.000465 min_lr: 0.000465 loss: 1.7333 (1.7792) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [354] [290/312] eta: 0:00:16 lr: 0.000464 min_lr: 0.000464 loss: 1.8069 (1.7788) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0008 max mem: 64948 Epoch: [354] [300/312] eta: 0:00:08 lr: 0.000464 min_lr: 0.000464 loss: 1.7120 (1.7733) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [354] [310/312] eta: 0:00:01 lr: 0.000464 min_lr: 0.000464 loss: 1.7120 (1.7770) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [354] [311/312] eta: 0:00:00 lr: 0.000464 min_lr: 0.000464 loss: 1.7120 (1.7771) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [354] Total time: 0:03:48 (0.7327 s / it) Averaged stats: lr: 0.000464 min_lr: 0.000464 loss: 1.7120 (1.7833) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4888 (0.4888) acc1: 86.7188 (86.7188) acc5: 98.6979 (98.6979) time: 4.5754 data: 4.3555 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6836 (0.6850) acc1: 82.0312 (81.2480) acc5: 96.3542 (96.4480) time: 0.6603 data: 0.4840 max mem: 64948 Test: Total time: 0:00:06 (0.6841 s / it) * Acc@1 82.620 Acc@5 96.212 loss 0.656 Accuracy of the model on the 50000 test images: 82.6% Max accuracy: 82.62% Test: [0/9] eta: 0:00:40 loss: 0.4611 (0.4611) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 4.4860 data: 4.2834 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6332 (0.6489) acc1: 84.3750 (82.0480) acc5: 97.1354 (96.5440) time: 0.6498 data: 0.4760 max mem: 64948 Test: Total time: 0:00:05 (0.6572 s / it) * Acc@1 83.204 Acc@5 96.554 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.20% Epoch: [355] [ 0/312] eta: 0:52:51 lr: 0.000464 min_lr: 0.000464 loss: 2.0124 (2.0124) weight_decay: 0.0500 (0.0500) time: 10.1640 data: 8.6915 max mem: 64948 Epoch: [355] [ 10/312] eta: 0:07:53 lr: 0.000463 min_lr: 0.000463 loss: 1.9930 (1.8895) weight_decay: 0.0500 (0.0500) time: 1.5676 data: 0.7905 max mem: 64948 Epoch: [355] [ 20/312] eta: 0:05:36 lr: 0.000463 min_lr: 0.000463 loss: 1.8369 (1.8382) weight_decay: 0.0500 (0.0500) time: 0.7030 data: 0.0004 max mem: 64948 Epoch: [355] [ 30/312] eta: 0:04:43 lr: 0.000463 min_lr: 0.000463 loss: 1.8369 (1.8269) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [355] [ 40/312] eta: 0:04:12 lr: 0.000462 min_lr: 0.000462 loss: 1.7741 (1.8084) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [355] [ 50/312] eta: 0:03:51 lr: 0.000462 min_lr: 0.000462 loss: 1.8313 (1.8064) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [355] [ 60/312] eta: 0:03:34 lr: 0.000462 min_lr: 0.000462 loss: 1.7070 (1.7876) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [355] [ 70/312] eta: 0:03:21 lr: 0.000461 min_lr: 0.000461 loss: 1.7070 (1.7722) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [355] [ 80/312] eta: 0:03:09 lr: 0.000461 min_lr: 0.000461 loss: 1.8282 (1.7793) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [355] [ 90/312] eta: 0:02:58 lr: 0.000461 min_lr: 0.000461 loss: 1.9195 (1.7804) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [355] [100/312] eta: 0:02:47 lr: 0.000461 min_lr: 0.000461 loss: 1.7413 (1.7786) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [355] [110/312] eta: 0:02:38 lr: 0.000460 min_lr: 0.000460 loss: 1.8823 (1.7922) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [355] [120/312] eta: 0:02:28 lr: 0.000460 min_lr: 0.000460 loss: 1.9312 (1.7893) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [355] [130/312] eta: 0:02:20 lr: 0.000460 min_lr: 0.000460 loss: 1.8788 (1.7929) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [355] [140/312] eta: 0:02:11 lr: 0.000459 min_lr: 0.000459 loss: 1.8421 (1.7954) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [355] [150/312] eta: 0:02:03 lr: 0.000459 min_lr: 0.000459 loss: 1.7881 (1.7976) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [355] [160/312] eta: 0:01:54 lr: 0.000459 min_lr: 0.000459 loss: 1.8515 (1.7961) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [355] [170/312] eta: 0:01:46 lr: 0.000459 min_lr: 0.000459 loss: 1.7812 (1.7929) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [355] [180/312] eta: 0:01:38 lr: 0.000458 min_lr: 0.000458 loss: 1.7937 (1.7971) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [355] [190/312] eta: 0:01:31 lr: 0.000458 min_lr: 0.000458 loss: 1.8203 (1.7983) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [355] [200/312] eta: 0:01:23 lr: 0.000458 min_lr: 0.000458 loss: 1.8272 (1.7991) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [355] [210/312] eta: 0:01:15 lr: 0.000457 min_lr: 0.000457 loss: 1.8351 (1.7980) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [355] [220/312] eta: 0:01:07 lr: 0.000457 min_lr: 0.000457 loss: 1.8771 (1.8024) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [355] [230/312] eta: 0:01:00 lr: 0.000457 min_lr: 0.000457 loss: 1.9408 (1.8056) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [355] [240/312] eta: 0:00:52 lr: 0.000456 min_lr: 0.000456 loss: 1.9619 (1.8112) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [355] [250/312] eta: 0:00:45 lr: 0.000456 min_lr: 0.000456 loss: 1.8384 (1.8098) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [355] [260/312] eta: 0:00:38 lr: 0.000456 min_lr: 0.000456 loss: 1.8201 (1.8089) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [355] [270/312] eta: 0:00:30 lr: 0.000456 min_lr: 0.000456 loss: 1.8857 (1.8078) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [355] [280/312] eta: 0:00:23 lr: 0.000455 min_lr: 0.000455 loss: 1.8857 (1.8060) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [355] [290/312] eta: 0:00:16 lr: 0.000455 min_lr: 0.000455 loss: 1.9007 (1.8101) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [355] [300/312] eta: 0:00:08 lr: 0.000455 min_lr: 0.000455 loss: 1.9601 (1.8104) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [355] [310/312] eta: 0:00:01 lr: 0.000454 min_lr: 0.000454 loss: 1.9794 (1.8158) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [355] [311/312] eta: 0:00:00 lr: 0.000454 min_lr: 0.000454 loss: 1.9794 (1.8165) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [355] Total time: 0:03:47 (0.7286 s / it) Averaged stats: lr: 0.000454 min_lr: 0.000454 loss: 1.9794 (1.7835) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4797 (0.4797) acc1: 86.4583 (86.4583) acc5: 98.6979 (98.6979) time: 4.6455 data: 4.4259 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6803 (0.6784) acc1: 82.0312 (81.4720) acc5: 96.8750 (96.5120) time: 0.6676 data: 0.4918 max mem: 64948 Test: Total time: 0:00:06 (0.6904 s / it) * Acc@1 82.552 Acc@5 96.280 loss 0.658 Accuracy of the model on the 50000 test images: 82.6% Max accuracy: 82.62% Test: [0/9] eta: 0:00:40 loss: 0.4609 (0.4609) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 4.5300 data: 4.3133 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6328 (0.6487) acc1: 84.3750 (82.0480) acc5: 97.1354 (96.5440) time: 0.6754 data: 0.4959 max mem: 64948 Test: Total time: 0:00:06 (0.6895 s / it) * Acc@1 83.196 Acc@5 96.560 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Epoch: [356] [ 0/312] eta: 0:59:01 lr: 0.000454 min_lr: 0.000454 loss: 2.0101 (2.0101) weight_decay: 0.0500 (0.0500) time: 11.3509 data: 8.9753 max mem: 64948 Epoch: [356] [ 10/312] eta: 0:08:26 lr: 0.000454 min_lr: 0.000454 loss: 1.9659 (1.8803) weight_decay: 0.0500 (0.0500) time: 1.6775 data: 0.8163 max mem: 64948 Epoch: [356] [ 20/312] eta: 0:05:53 lr: 0.000454 min_lr: 0.000454 loss: 1.9571 (1.8745) weight_decay: 0.0500 (0.0500) time: 0.7021 data: 0.0003 max mem: 64948 Epoch: [356] [ 30/312] eta: 0:04:54 lr: 0.000453 min_lr: 0.000453 loss: 1.8063 (1.8236) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [356] [ 40/312] eta: 0:04:20 lr: 0.000453 min_lr: 0.000453 loss: 1.7650 (1.8496) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [356] [ 50/312] eta: 0:03:57 lr: 0.000453 min_lr: 0.000453 loss: 1.8181 (1.8284) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [356] [ 60/312] eta: 0:03:39 lr: 0.000453 min_lr: 0.000453 loss: 1.7785 (1.8278) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [356] [ 70/312] eta: 0:03:25 lr: 0.000452 min_lr: 0.000452 loss: 1.8476 (1.8351) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [356] [ 80/312] eta: 0:03:12 lr: 0.000452 min_lr: 0.000452 loss: 1.9295 (1.8420) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [356] [ 90/312] eta: 0:03:00 lr: 0.000452 min_lr: 0.000452 loss: 1.8921 (1.8344) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [356] [100/312] eta: 0:02:50 lr: 0.000451 min_lr: 0.000451 loss: 1.7477 (1.8332) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [356] [110/312] eta: 0:02:40 lr: 0.000451 min_lr: 0.000451 loss: 1.8834 (1.8412) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [356] [120/312] eta: 0:02:30 lr: 0.000451 min_lr: 0.000451 loss: 1.5872 (1.8157) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [356] [130/312] eta: 0:02:21 lr: 0.000450 min_lr: 0.000450 loss: 1.5313 (1.8098) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [356] [140/312] eta: 0:02:12 lr: 0.000450 min_lr: 0.000450 loss: 1.7105 (1.8029) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [356] [150/312] eta: 0:02:04 lr: 0.000450 min_lr: 0.000450 loss: 1.7357 (1.8006) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [356] [160/312] eta: 0:01:55 lr: 0.000450 min_lr: 0.000450 loss: 1.8686 (1.8065) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [356] [170/312] eta: 0:01:47 lr: 0.000449 min_lr: 0.000449 loss: 1.9101 (1.8074) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [356] [180/312] eta: 0:01:39 lr: 0.000449 min_lr: 0.000449 loss: 1.8955 (1.8092) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [356] [190/312] eta: 0:01:31 lr: 0.000449 min_lr: 0.000449 loss: 1.9015 (1.8122) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [356] [200/312] eta: 0:01:23 lr: 0.000448 min_lr: 0.000448 loss: 1.8025 (1.8048) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [356] [210/312] eta: 0:01:16 lr: 0.000448 min_lr: 0.000448 loss: 1.5886 (1.7978) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [356] [220/312] eta: 0:01:08 lr: 0.000448 min_lr: 0.000448 loss: 1.5886 (1.7939) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [356] [230/312] eta: 0:01:00 lr: 0.000447 min_lr: 0.000447 loss: 1.8271 (1.7928) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [356] [240/312] eta: 0:00:53 lr: 0.000447 min_lr: 0.000447 loss: 1.8907 (1.7963) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [356] [250/312] eta: 0:00:45 lr: 0.000447 min_lr: 0.000447 loss: 1.8552 (1.7899) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [356] [260/312] eta: 0:00:38 lr: 0.000447 min_lr: 0.000447 loss: 1.8622 (1.7886) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [356] [270/312] eta: 0:00:30 lr: 0.000446 min_lr: 0.000446 loss: 1.7823 (1.7852) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [356] [280/312] eta: 0:00:23 lr: 0.000446 min_lr: 0.000446 loss: 1.6745 (1.7821) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [356] [290/312] eta: 0:00:16 lr: 0.000446 min_lr: 0.000446 loss: 1.7058 (1.7810) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [356] [300/312] eta: 0:00:08 lr: 0.000445 min_lr: 0.000445 loss: 1.7263 (1.7775) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [356] [310/312] eta: 0:00:01 lr: 0.000445 min_lr: 0.000445 loss: 1.8629 (1.7815) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [356] [311/312] eta: 0:00:00 lr: 0.000445 min_lr: 0.000445 loss: 1.8629 (1.7826) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [356] Total time: 0:03:48 (0.7324 s / it) Averaged stats: lr: 0.000445 min_lr: 0.000445 loss: 1.8629 (1.7761) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4725 (0.4725) acc1: 86.7188 (86.7188) acc5: 99.2188 (99.2188) time: 4.7521 data: 4.5390 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6851 (0.6792) acc1: 83.0729 (81.4080) acc5: 96.8750 (96.4160) time: 0.6799 data: 0.5044 max mem: 64948 Test: Total time: 0:00:06 (0.7023 s / it) * Acc@1 82.534 Acc@5 96.266 loss 0.661 Accuracy of the model on the 50000 test images: 82.5% Max accuracy: 82.62% Test: [0/9] eta: 0:00:44 loss: 0.4606 (0.4606) acc1: 88.2812 (88.2812) acc5: 97.6562 (97.6562) time: 4.9082 data: 4.6976 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6325 (0.6486) acc1: 84.3750 (82.1440) acc5: 97.1354 (96.5760) time: 0.7019 data: 0.5221 max mem: 64948 Test: Total time: 0:00:06 (0.7186 s / it) * Acc@1 83.226 Acc@5 96.562 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.23% Epoch: [357] [ 0/312] eta: 0:51:34 lr: 0.000445 min_lr: 0.000445 loss: 1.3846 (1.3846) weight_decay: 0.0500 (0.0500) time: 9.9187 data: 9.1398 max mem: 64948 Epoch: [357] [ 10/312] eta: 0:07:54 lr: 0.000445 min_lr: 0.000445 loss: 1.6628 (1.6630) weight_decay: 0.0500 (0.0500) time: 1.5716 data: 0.8312 max mem: 64948 Epoch: [357] [ 20/312] eta: 0:05:37 lr: 0.000444 min_lr: 0.000444 loss: 1.6628 (1.6855) weight_decay: 0.0500 (0.0500) time: 0.7161 data: 0.0003 max mem: 64948 Epoch: [357] [ 30/312] eta: 0:04:43 lr: 0.000444 min_lr: 0.000444 loss: 1.6924 (1.6849) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [357] [ 40/312] eta: 0:04:12 lr: 0.000444 min_lr: 0.000444 loss: 1.7391 (1.6898) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [357] [ 50/312] eta: 0:03:51 lr: 0.000444 min_lr: 0.000444 loss: 1.7380 (1.6773) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [357] [ 60/312] eta: 0:03:34 lr: 0.000443 min_lr: 0.000443 loss: 1.8350 (1.6965) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [357] [ 70/312] eta: 0:03:21 lr: 0.000443 min_lr: 0.000443 loss: 1.8128 (1.6881) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0004 max mem: 64948 Epoch: [357] [ 80/312] eta: 0:03:09 lr: 0.000443 min_lr: 0.000443 loss: 1.7893 (1.7116) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0004 max mem: 64948 Epoch: [357] [ 90/312] eta: 0:02:58 lr: 0.000442 min_lr: 0.000442 loss: 1.7893 (1.7127) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [357] [100/312] eta: 0:02:47 lr: 0.000442 min_lr: 0.000442 loss: 1.7930 (1.7173) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [357] [110/312] eta: 0:02:38 lr: 0.000442 min_lr: 0.000442 loss: 1.9103 (1.7338) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [357] [120/312] eta: 0:02:28 lr: 0.000442 min_lr: 0.000442 loss: 1.8911 (1.7378) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [357] [130/312] eta: 0:02:20 lr: 0.000441 min_lr: 0.000441 loss: 1.8041 (1.7318) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [357] [140/312] eta: 0:02:11 lr: 0.000441 min_lr: 0.000441 loss: 1.8116 (1.7419) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [357] [150/312] eta: 0:02:03 lr: 0.000441 min_lr: 0.000441 loss: 1.8959 (1.7525) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [357] [160/312] eta: 0:01:54 lr: 0.000440 min_lr: 0.000440 loss: 1.8713 (1.7528) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [357] [170/312] eta: 0:01:46 lr: 0.000440 min_lr: 0.000440 loss: 1.7713 (1.7565) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [357] [180/312] eta: 0:01:38 lr: 0.000440 min_lr: 0.000440 loss: 1.8786 (1.7652) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [357] [190/312] eta: 0:01:31 lr: 0.000439 min_lr: 0.000439 loss: 1.8903 (1.7646) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [357] [200/312] eta: 0:01:23 lr: 0.000439 min_lr: 0.000439 loss: 1.7757 (1.7623) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [357] [210/312] eta: 0:01:15 lr: 0.000439 min_lr: 0.000439 loss: 1.8491 (1.7688) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [357] [220/312] eta: 0:01:08 lr: 0.000439 min_lr: 0.000439 loss: 1.9456 (1.7690) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [357] [230/312] eta: 0:01:00 lr: 0.000438 min_lr: 0.000438 loss: 1.8693 (1.7664) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [357] [240/312] eta: 0:00:52 lr: 0.000438 min_lr: 0.000438 loss: 1.8920 (1.7678) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [357] [250/312] eta: 0:00:45 lr: 0.000438 min_lr: 0.000438 loss: 1.7503 (1.7690) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [357] [260/312] eta: 0:00:38 lr: 0.000437 min_lr: 0.000437 loss: 1.8769 (1.7756) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [357] [270/312] eta: 0:00:30 lr: 0.000437 min_lr: 0.000437 loss: 1.9814 (1.7783) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [357] [280/312] eta: 0:00:23 lr: 0.000437 min_lr: 0.000437 loss: 1.6007 (1.7696) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0010 max mem: 64948 Epoch: [357] [290/312] eta: 0:00:16 lr: 0.000437 min_lr: 0.000437 loss: 1.5489 (1.7674) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [357] [300/312] eta: 0:00:08 lr: 0.000436 min_lr: 0.000436 loss: 1.7451 (1.7710) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [357] [310/312] eta: 0:00:01 lr: 0.000436 min_lr: 0.000436 loss: 1.8631 (1.7726) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [357] [311/312] eta: 0:00:00 lr: 0.000436 min_lr: 0.000436 loss: 1.8631 (1.7719) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [357] Total time: 0:03:47 (0.7287 s / it) Averaged stats: lr: 0.000436 min_lr: 0.000436 loss: 1.8631 (1.7728) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4709 (0.4709) acc1: 86.9792 (86.9792) acc5: 98.4375 (98.4375) time: 4.6355 data: 4.4297 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6610 (0.6937) acc1: 83.0729 (81.4400) acc5: 96.0938 (95.8720) time: 0.6664 data: 0.4923 max mem: 64948 Test: Total time: 0:00:06 (0.6852 s / it) * Acc@1 82.694 Acc@5 96.148 loss 0.663 Accuracy of the model on the 50000 test images: 82.7% Max accuracy: 82.69% Test: [0/9] eta: 0:00:40 loss: 0.4604 (0.4604) acc1: 88.2812 (88.2812) acc5: 97.6562 (97.6562) time: 4.4939 data: 4.2906 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6325 (0.6485) acc1: 84.3750 (82.1760) acc5: 97.1354 (96.5760) time: 0.6506 data: 0.4769 max mem: 64948 Test: Total time: 0:00:05 (0.6602 s / it) * Acc@1 83.242 Acc@5 96.568 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Max EMA accuracy: 83.24% Epoch: [358] [ 0/312] eta: 0:53:04 lr: 0.000436 min_lr: 0.000436 loss: 2.1210 (2.1210) weight_decay: 0.0500 (0.0500) time: 10.2056 data: 7.0968 max mem: 64948 Epoch: [358] [ 10/312] eta: 0:08:03 lr: 0.000436 min_lr: 0.000436 loss: 1.8507 (1.7607) weight_decay: 0.0500 (0.0500) time: 1.6026 data: 0.6456 max mem: 64948 Epoch: [358] [ 20/312] eta: 0:05:42 lr: 0.000435 min_lr: 0.000435 loss: 1.7887 (1.6939) weight_decay: 0.0500 (0.0500) time: 0.7213 data: 0.0004 max mem: 64948 Epoch: [358] [ 30/312] eta: 0:04:47 lr: 0.000435 min_lr: 0.000435 loss: 1.7576 (1.7051) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [358] [ 40/312] eta: 0:04:15 lr: 0.000435 min_lr: 0.000435 loss: 1.6280 (1.6860) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [358] [ 50/312] eta: 0:03:53 lr: 0.000434 min_lr: 0.000434 loss: 1.6280 (1.6945) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [358] [ 60/312] eta: 0:03:36 lr: 0.000434 min_lr: 0.000434 loss: 1.8588 (1.7146) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [358] [ 70/312] eta: 0:03:22 lr: 0.000434 min_lr: 0.000434 loss: 1.7800 (1.7122) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [358] [ 80/312] eta: 0:03:10 lr: 0.000434 min_lr: 0.000434 loss: 1.7138 (1.7261) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [358] [ 90/312] eta: 0:02:58 lr: 0.000433 min_lr: 0.000433 loss: 1.6939 (1.7210) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [358] [100/312] eta: 0:02:48 lr: 0.000433 min_lr: 0.000433 loss: 1.6977 (1.7282) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [358] [110/312] eta: 0:02:38 lr: 0.000433 min_lr: 0.000433 loss: 1.8238 (1.7354) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [358] [120/312] eta: 0:02:29 lr: 0.000432 min_lr: 0.000432 loss: 1.7853 (1.7272) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [358] [130/312] eta: 0:02:20 lr: 0.000432 min_lr: 0.000432 loss: 1.7853 (1.7375) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [358] [140/312] eta: 0:02:11 lr: 0.000432 min_lr: 0.000432 loss: 1.7628 (1.7251) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [358] [150/312] eta: 0:02:03 lr: 0.000432 min_lr: 0.000432 loss: 1.7628 (1.7290) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [358] [160/312] eta: 0:01:55 lr: 0.000431 min_lr: 0.000431 loss: 1.9244 (1.7330) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [358] [170/312] eta: 0:01:46 lr: 0.000431 min_lr: 0.000431 loss: 1.8417 (1.7360) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [358] [180/312] eta: 0:01:39 lr: 0.000431 min_lr: 0.000431 loss: 1.8889 (1.7437) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [358] [190/312] eta: 0:01:31 lr: 0.000430 min_lr: 0.000430 loss: 1.8531 (1.7398) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [358] [200/312] eta: 0:01:23 lr: 0.000430 min_lr: 0.000430 loss: 1.6807 (1.7429) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [358] [210/312] eta: 0:01:15 lr: 0.000430 min_lr: 0.000430 loss: 1.6807 (1.7426) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [358] [220/312] eta: 0:01:08 lr: 0.000430 min_lr: 0.000430 loss: 1.6601 (1.7371) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [358] [230/312] eta: 0:01:00 lr: 0.000429 min_lr: 0.000429 loss: 1.7235 (1.7391) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [358] [240/312] eta: 0:00:53 lr: 0.000429 min_lr: 0.000429 loss: 1.6850 (1.7397) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [358] [250/312] eta: 0:00:45 lr: 0.000429 min_lr: 0.000429 loss: 1.8628 (1.7469) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [358] [260/312] eta: 0:00:38 lr: 0.000428 min_lr: 0.000428 loss: 1.6973 (1.7458) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [358] [270/312] eta: 0:00:30 lr: 0.000428 min_lr: 0.000428 loss: 1.6462 (1.7466) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [358] [280/312] eta: 0:00:23 lr: 0.000428 min_lr: 0.000428 loss: 1.7829 (1.7432) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [358] [290/312] eta: 0:00:16 lr: 0.000428 min_lr: 0.000428 loss: 1.6871 (1.7452) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [358] [300/312] eta: 0:00:08 lr: 0.000427 min_lr: 0.000427 loss: 1.5141 (1.7401) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0001 max mem: 64948 Epoch: [358] [310/312] eta: 0:00:01 lr: 0.000427 min_lr: 0.000427 loss: 1.9469 (1.7475) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [358] [311/312] eta: 0:00:00 lr: 0.000427 min_lr: 0.000427 loss: 1.9469 (1.7474) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [358] Total time: 0:03:47 (0.7295 s / it) Averaged stats: lr: 0.000427 min_lr: 0.000427 loss: 1.9469 (1.7727) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4933 (0.4933) acc1: 88.0208 (88.0208) acc5: 98.4375 (98.4375) time: 4.6172 data: 4.4071 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6242 (0.6735) acc1: 83.5938 (81.8880) acc5: 96.8750 (96.3520) time: 0.6643 data: 0.4898 max mem: 64948 Test: Total time: 0:00:06 (0.6882 s / it) * Acc@1 82.738 Acc@5 96.258 loss 0.658 Accuracy of the model on the 50000 test images: 82.7% Max accuracy: 82.74% Test: [0/9] eta: 0:00:42 loss: 0.4603 (0.4603) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 4.6931 data: 4.4812 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6318 (0.6486) acc1: 84.3750 (82.1120) acc5: 97.1354 (96.5440) time: 0.6734 data: 0.4980 max mem: 64948 Test: Total time: 0:00:06 (0.6810 s / it) * Acc@1 83.238 Acc@5 96.558 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Epoch: [359] [ 0/312] eta: 1:01:25 lr: 0.000427 min_lr: 0.000427 loss: 1.8860 (1.8860) weight_decay: 0.0500 (0.0500) time: 11.8113 data: 7.3349 max mem: 64948 Epoch: [359] [ 10/312] eta: 0:08:37 lr: 0.000427 min_lr: 0.000427 loss: 1.9254 (1.9197) weight_decay: 0.0500 (0.0500) time: 1.7141 data: 0.6672 max mem: 64948 Epoch: [359] [ 20/312] eta: 0:05:59 lr: 0.000426 min_lr: 0.000426 loss: 1.9169 (1.8894) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0003 max mem: 64948 Epoch: [359] [ 30/312] eta: 0:04:58 lr: 0.000426 min_lr: 0.000426 loss: 1.8435 (1.8601) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [359] [ 40/312] eta: 0:04:23 lr: 0.000426 min_lr: 0.000426 loss: 1.9059 (1.8709) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [359] [ 50/312] eta: 0:03:59 lr: 0.000425 min_lr: 0.000425 loss: 1.8546 (1.8555) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [359] [ 60/312] eta: 0:03:41 lr: 0.000425 min_lr: 0.000425 loss: 1.7554 (1.8542) weight_decay: 0.0500 (0.0500) time: 0.7014 data: 0.0004 max mem: 64948 Epoch: [359] [ 70/312] eta: 0:03:27 lr: 0.000425 min_lr: 0.000425 loss: 1.7554 (1.8364) weight_decay: 0.0500 (0.0500) time: 0.7019 data: 0.0004 max mem: 64948 Epoch: [359] [ 80/312] eta: 0:03:13 lr: 0.000425 min_lr: 0.000425 loss: 1.5910 (1.8006) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0004 max mem: 64948 Epoch: [359] [ 90/312] eta: 0:03:02 lr: 0.000424 min_lr: 0.000424 loss: 1.7321 (1.8030) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [359] [100/312] eta: 0:02:51 lr: 0.000424 min_lr: 0.000424 loss: 1.8686 (1.8161) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [359] [110/312] eta: 0:02:41 lr: 0.000424 min_lr: 0.000424 loss: 1.8866 (1.8140) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [359] [120/312] eta: 0:02:31 lr: 0.000423 min_lr: 0.000423 loss: 1.7645 (1.8056) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [359] [130/312] eta: 0:02:22 lr: 0.000423 min_lr: 0.000423 loss: 1.7257 (1.7998) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [359] [140/312] eta: 0:02:13 lr: 0.000423 min_lr: 0.000423 loss: 1.8774 (1.8105) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [359] [150/312] eta: 0:02:04 lr: 0.000423 min_lr: 0.000423 loss: 1.8774 (1.8031) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [359] [160/312] eta: 0:01:56 lr: 0.000422 min_lr: 0.000422 loss: 1.8253 (1.8055) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [359] [170/312] eta: 0:01:48 lr: 0.000422 min_lr: 0.000422 loss: 1.8002 (1.7948) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [359] [180/312] eta: 0:01:40 lr: 0.000422 min_lr: 0.000422 loss: 1.8161 (1.7953) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [359] [190/312] eta: 0:01:32 lr: 0.000421 min_lr: 0.000421 loss: 1.8873 (1.7888) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [359] [200/312] eta: 0:01:24 lr: 0.000421 min_lr: 0.000421 loss: 1.7489 (1.7903) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [359] [210/312] eta: 0:01:16 lr: 0.000421 min_lr: 0.000421 loss: 1.7736 (1.7852) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [359] [220/312] eta: 0:01:08 lr: 0.000421 min_lr: 0.000421 loss: 1.7736 (1.7828) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [359] [230/312] eta: 0:01:01 lr: 0.000420 min_lr: 0.000420 loss: 1.7825 (1.7881) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [359] [240/312] eta: 0:00:53 lr: 0.000420 min_lr: 0.000420 loss: 1.8583 (1.7885) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [359] [250/312] eta: 0:00:45 lr: 0.000420 min_lr: 0.000420 loss: 1.8583 (1.7925) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [359] [260/312] eta: 0:00:38 lr: 0.000419 min_lr: 0.000419 loss: 1.8796 (1.7943) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [359] [270/312] eta: 0:00:30 lr: 0.000419 min_lr: 0.000419 loss: 1.8559 (1.7960) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [359] [280/312] eta: 0:00:23 lr: 0.000419 min_lr: 0.000419 loss: 1.9143 (1.7967) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0009 max mem: 64948 Epoch: [359] [290/312] eta: 0:00:16 lr: 0.000419 min_lr: 0.000419 loss: 1.8569 (1.7918) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0008 max mem: 64948 Epoch: [359] [300/312] eta: 0:00:08 lr: 0.000418 min_lr: 0.000418 loss: 1.8133 (1.7902) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [359] [310/312] eta: 0:00:01 lr: 0.000418 min_lr: 0.000418 loss: 1.7977 (1.7892) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [359] [311/312] eta: 0:00:00 lr: 0.000418 min_lr: 0.000418 loss: 1.7977 (1.7891) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [359] Total time: 0:03:49 (0.7341 s / it) Averaged stats: lr: 0.000418 min_lr: 0.000418 loss: 1.7977 (1.7748) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4748 (0.4748) acc1: 86.9792 (86.9792) acc5: 97.9167 (97.9167) time: 4.7614 data: 4.5477 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6811 (0.6819) acc1: 83.3333 (81.5040) acc5: 96.8750 (96.3520) time: 0.6803 data: 0.5054 max mem: 64948 Test: Total time: 0:00:06 (0.6994 s / it) * Acc@1 82.616 Acc@5 96.264 loss 0.659 Accuracy of the model on the 50000 test images: 82.6% Max accuracy: 82.74% Test: [0/9] eta: 0:00:43 loss: 0.4601 (0.4601) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 4.8568 data: 4.6350 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6312 (0.6485) acc1: 84.3750 (82.1760) acc5: 97.1354 (96.5440) time: 0.6909 data: 0.5151 max mem: 64948 Test: Total time: 0:00:06 (0.7085 s / it) * Acc@1 83.242 Acc@5 96.556 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Epoch: [360] [ 0/312] eta: 0:54:23 lr: 0.000418 min_lr: 0.000418 loss: 2.1386 (2.1386) weight_decay: 0.0500 (0.0500) time: 10.4600 data: 6.3646 max mem: 64948 Epoch: [360] [ 10/312] eta: 0:08:06 lr: 0.000418 min_lr: 0.000418 loss: 1.8284 (1.7715) weight_decay: 0.0500 (0.0500) time: 1.6122 data: 0.5791 max mem: 64948 Epoch: [360] [ 20/312] eta: 0:05:43 lr: 0.000417 min_lr: 0.000417 loss: 1.7712 (1.6953) weight_decay: 0.0500 (0.0500) time: 0.7104 data: 0.0005 max mem: 64948 Epoch: [360] [ 30/312] eta: 0:04:47 lr: 0.000417 min_lr: 0.000417 loss: 1.7176 (1.7178) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [360] [ 40/312] eta: 0:04:16 lr: 0.000417 min_lr: 0.000417 loss: 1.9064 (1.7579) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [360] [ 50/312] eta: 0:03:53 lr: 0.000416 min_lr: 0.000416 loss: 1.6485 (1.7265) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [360] [ 60/312] eta: 0:03:37 lr: 0.000416 min_lr: 0.000416 loss: 1.6485 (1.7224) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [360] [ 70/312] eta: 0:03:22 lr: 0.000416 min_lr: 0.000416 loss: 1.7367 (1.7249) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0003 max mem: 64948 Epoch: [360] [ 80/312] eta: 0:03:10 lr: 0.000416 min_lr: 0.000416 loss: 1.8362 (1.7424) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [360] [ 90/312] eta: 0:02:59 lr: 0.000415 min_lr: 0.000415 loss: 1.8583 (1.7540) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [360] [100/312] eta: 0:02:48 lr: 0.000415 min_lr: 0.000415 loss: 1.7650 (1.7453) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [360] [110/312] eta: 0:02:39 lr: 0.000415 min_lr: 0.000415 loss: 1.7468 (1.7492) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [360] [120/312] eta: 0:02:29 lr: 0.000414 min_lr: 0.000414 loss: 1.7729 (1.7531) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [360] [130/312] eta: 0:02:20 lr: 0.000414 min_lr: 0.000414 loss: 1.7774 (1.7529) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [360] [140/312] eta: 0:02:11 lr: 0.000414 min_lr: 0.000414 loss: 1.7157 (1.7469) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [360] [150/312] eta: 0:02:03 lr: 0.000414 min_lr: 0.000414 loss: 1.7694 (1.7463) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [360] [160/312] eta: 0:01:55 lr: 0.000413 min_lr: 0.000413 loss: 1.7330 (1.7419) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [360] [170/312] eta: 0:01:47 lr: 0.000413 min_lr: 0.000413 loss: 1.6247 (1.7349) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [360] [180/312] eta: 0:01:39 lr: 0.000413 min_lr: 0.000413 loss: 1.7856 (1.7385) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [360] [190/312] eta: 0:01:31 lr: 0.000412 min_lr: 0.000412 loss: 1.7856 (1.7363) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [360] [200/312] eta: 0:01:23 lr: 0.000412 min_lr: 0.000412 loss: 1.7792 (1.7404) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [360] [210/312] eta: 0:01:15 lr: 0.000412 min_lr: 0.000412 loss: 1.8452 (1.7365) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [360] [220/312] eta: 0:01:08 lr: 0.000412 min_lr: 0.000412 loss: 1.5892 (1.7339) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [360] [230/312] eta: 0:01:00 lr: 0.000411 min_lr: 0.000411 loss: 1.6694 (1.7401) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [360] [240/312] eta: 0:00:53 lr: 0.000411 min_lr: 0.000411 loss: 1.7796 (1.7401) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [360] [250/312] eta: 0:00:45 lr: 0.000411 min_lr: 0.000411 loss: 1.6788 (1.7333) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [360] [260/312] eta: 0:00:38 lr: 0.000410 min_lr: 0.000410 loss: 1.6788 (1.7329) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [360] [270/312] eta: 0:00:30 lr: 0.000410 min_lr: 0.000410 loss: 1.8004 (1.7337) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [360] [280/312] eta: 0:00:23 lr: 0.000410 min_lr: 0.000410 loss: 1.9029 (1.7364) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0009 max mem: 64948 Epoch: [360] [290/312] eta: 0:00:16 lr: 0.000410 min_lr: 0.000410 loss: 1.9029 (1.7430) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0008 max mem: 64948 Epoch: [360] [300/312] eta: 0:00:08 lr: 0.000409 min_lr: 0.000409 loss: 1.8695 (1.7444) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [360] [310/312] eta: 0:00:01 lr: 0.000409 min_lr: 0.000409 loss: 1.8150 (1.7457) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [360] [311/312] eta: 0:00:00 lr: 0.000409 min_lr: 0.000409 loss: 1.8150 (1.7444) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [360] Total time: 0:03:47 (0.7306 s / it) Averaged stats: lr: 0.000409 min_lr: 0.000409 loss: 1.8150 (1.7602) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4784 (0.4784) acc1: 87.7604 (87.7604) acc5: 97.3958 (97.3958) time: 4.6247 data: 4.4158 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6839 (0.6817) acc1: 83.5938 (81.6320) acc5: 96.6146 (96.2880) time: 0.6651 data: 0.4907 max mem: 64948 Test: Total time: 0:00:06 (0.6924 s / it) * Acc@1 82.628 Acc@5 96.314 loss 0.659 Accuracy of the model on the 50000 test images: 82.6% Max accuracy: 82.74% Test: [0/9] eta: 0:00:43 loss: 0.4600 (0.4600) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 4.8382 data: 4.6200 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6309 (0.6483) acc1: 84.6354 (82.2400) acc5: 97.1354 (96.5440) time: 0.6888 data: 0.5134 max mem: 64948 Test: Total time: 0:00:06 (0.7001 s / it) * Acc@1 83.242 Acc@5 96.558 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Epoch: [361] [ 0/312] eta: 0:55:49 lr: 0.000409 min_lr: 0.000409 loss: 1.2903 (1.2903) weight_decay: 0.0500 (0.0500) time: 10.7357 data: 8.6293 max mem: 64948 Epoch: [361] [ 10/312] eta: 0:08:24 lr: 0.000409 min_lr: 0.000409 loss: 1.9296 (1.8172) weight_decay: 0.0500 (0.0500) time: 1.6690 data: 0.7849 max mem: 64948 Epoch: [361] [ 20/312] eta: 0:05:51 lr: 0.000408 min_lr: 0.000408 loss: 1.8106 (1.7997) weight_decay: 0.0500 (0.0500) time: 0.7280 data: 0.0004 max mem: 64948 Epoch: [361] [ 30/312] eta: 0:04:53 lr: 0.000408 min_lr: 0.000408 loss: 1.8187 (1.8151) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0003 max mem: 64948 Epoch: [361] [ 40/312] eta: 0:04:20 lr: 0.000408 min_lr: 0.000408 loss: 1.8408 (1.8056) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [361] [ 50/312] eta: 0:03:57 lr: 0.000408 min_lr: 0.000408 loss: 1.8408 (1.8096) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [361] [ 60/312] eta: 0:03:39 lr: 0.000407 min_lr: 0.000407 loss: 1.6776 (1.7857) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [361] [ 70/312] eta: 0:03:24 lr: 0.000407 min_lr: 0.000407 loss: 1.8808 (1.7888) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [361] [ 80/312] eta: 0:03:12 lr: 0.000407 min_lr: 0.000407 loss: 1.8808 (1.7835) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [361] [ 90/312] eta: 0:03:00 lr: 0.000406 min_lr: 0.000406 loss: 1.6837 (1.7732) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [361] [100/312] eta: 0:02:50 lr: 0.000406 min_lr: 0.000406 loss: 1.7242 (1.7651) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [361] [110/312] eta: 0:02:40 lr: 0.000406 min_lr: 0.000406 loss: 1.7242 (1.7536) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [361] [120/312] eta: 0:02:30 lr: 0.000406 min_lr: 0.000406 loss: 1.7082 (1.7618) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [361] [130/312] eta: 0:02:21 lr: 0.000405 min_lr: 0.000405 loss: 1.8589 (1.7619) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [361] [140/312] eta: 0:02:12 lr: 0.000405 min_lr: 0.000405 loss: 1.6465 (1.7504) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [361] [150/312] eta: 0:02:04 lr: 0.000405 min_lr: 0.000405 loss: 1.6302 (1.7432) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [361] [160/312] eta: 0:01:55 lr: 0.000404 min_lr: 0.000404 loss: 1.7426 (1.7454) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [361] [170/312] eta: 0:01:47 lr: 0.000404 min_lr: 0.000404 loss: 1.8896 (1.7472) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [361] [180/312] eta: 0:01:39 lr: 0.000404 min_lr: 0.000404 loss: 1.8789 (1.7506) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [361] [190/312] eta: 0:01:31 lr: 0.000404 min_lr: 0.000404 loss: 1.7048 (1.7437) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [361] [200/312] eta: 0:01:23 lr: 0.000403 min_lr: 0.000403 loss: 1.7050 (1.7458) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [361] [210/312] eta: 0:01:16 lr: 0.000403 min_lr: 0.000403 loss: 1.8230 (1.7455) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [361] [220/312] eta: 0:01:08 lr: 0.000403 min_lr: 0.000403 loss: 1.8230 (1.7456) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [361] [230/312] eta: 0:01:00 lr: 0.000403 min_lr: 0.000403 loss: 1.8663 (1.7483) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [361] [240/312] eta: 0:00:53 lr: 0.000402 min_lr: 0.000402 loss: 1.8082 (1.7458) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [361] [250/312] eta: 0:00:45 lr: 0.000402 min_lr: 0.000402 loss: 1.9210 (1.7553) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [361] [260/312] eta: 0:00:38 lr: 0.000402 min_lr: 0.000402 loss: 1.9303 (1.7548) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [361] [270/312] eta: 0:00:30 lr: 0.000401 min_lr: 0.000401 loss: 1.9303 (1.7591) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [361] [280/312] eta: 0:00:23 lr: 0.000401 min_lr: 0.000401 loss: 1.9533 (1.7629) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [361] [290/312] eta: 0:00:16 lr: 0.000401 min_lr: 0.000401 loss: 1.8822 (1.7626) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [361] [300/312] eta: 0:00:08 lr: 0.000401 min_lr: 0.000401 loss: 1.6775 (1.7585) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [361] [310/312] eta: 0:00:01 lr: 0.000400 min_lr: 0.000400 loss: 1.6775 (1.7583) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [361] [311/312] eta: 0:00:00 lr: 0.000400 min_lr: 0.000400 loss: 1.6597 (1.7573) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [361] Total time: 0:03:48 (0.7320 s / it) Averaged stats: lr: 0.000400 min_lr: 0.000400 loss: 1.6597 (1.7677) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4742 (0.4742) acc1: 87.5000 (87.5000) acc5: 97.6562 (97.6562) time: 4.4319 data: 4.2161 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6322 (0.6702) acc1: 82.2917 (81.4720) acc5: 97.1354 (96.4480) time: 0.6437 data: 0.4686 max mem: 64948 Test: Total time: 0:00:06 (0.6691 s / it) * Acc@1 82.810 Acc@5 96.296 loss 0.652 Accuracy of the model on the 50000 test images: 82.8% Max accuracy: 82.81% Test: [0/9] eta: 0:00:39 loss: 0.4599 (0.4599) acc1: 88.0208 (88.0208) acc5: 97.9167 (97.9167) time: 4.3715 data: 4.1641 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6303 (0.6483) acc1: 84.6354 (82.2400) acc5: 97.1354 (96.6080) time: 0.6371 data: 0.4628 max mem: 64948 Test: Total time: 0:00:05 (0.6456 s / it) * Acc@1 83.234 Acc@5 96.554 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Epoch: [362] [ 0/312] eta: 0:57:03 lr: 0.000400 min_lr: 0.000400 loss: 2.1383 (2.1383) weight_decay: 0.0500 (0.0500) time: 10.9730 data: 8.6664 max mem: 64948 Epoch: [362] [ 10/312] eta: 0:08:20 lr: 0.000400 min_lr: 0.000400 loss: 1.8829 (1.7226) weight_decay: 0.0500 (0.0500) time: 1.6569 data: 0.7883 max mem: 64948 Epoch: [362] [ 20/312] eta: 0:05:49 lr: 0.000400 min_lr: 0.000400 loss: 1.7837 (1.7011) weight_decay: 0.0500 (0.0500) time: 0.7096 data: 0.0004 max mem: 64948 Epoch: [362] [ 30/312] eta: 0:04:52 lr: 0.000399 min_lr: 0.000399 loss: 1.8167 (1.7719) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [362] [ 40/312] eta: 0:04:19 lr: 0.000399 min_lr: 0.000399 loss: 1.9157 (1.7831) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [362] [ 50/312] eta: 0:03:56 lr: 0.000399 min_lr: 0.000399 loss: 1.8601 (1.7799) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [362] [ 60/312] eta: 0:03:38 lr: 0.000399 min_lr: 0.000399 loss: 1.8312 (1.7694) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [362] [ 70/312] eta: 0:03:24 lr: 0.000398 min_lr: 0.000398 loss: 1.8302 (1.7694) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [362] [ 80/312] eta: 0:03:11 lr: 0.000398 min_lr: 0.000398 loss: 1.8431 (1.7731) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [362] [ 90/312] eta: 0:03:00 lr: 0.000398 min_lr: 0.000398 loss: 1.8882 (1.7908) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [362] [100/312] eta: 0:02:49 lr: 0.000397 min_lr: 0.000397 loss: 1.8882 (1.7937) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [362] [110/312] eta: 0:02:39 lr: 0.000397 min_lr: 0.000397 loss: 1.8146 (1.7908) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [362] [120/312] eta: 0:02:30 lr: 0.000397 min_lr: 0.000397 loss: 1.8791 (1.7955) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [362] [130/312] eta: 0:02:21 lr: 0.000397 min_lr: 0.000397 loss: 1.8181 (1.7871) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [362] [140/312] eta: 0:02:12 lr: 0.000396 min_lr: 0.000396 loss: 1.6414 (1.7744) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [362] [150/312] eta: 0:02:03 lr: 0.000396 min_lr: 0.000396 loss: 1.6475 (1.7734) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [362] [160/312] eta: 0:01:55 lr: 0.000396 min_lr: 0.000396 loss: 1.7393 (1.7766) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [362] [170/312] eta: 0:01:47 lr: 0.000395 min_lr: 0.000395 loss: 1.9023 (1.7816) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [362] [180/312] eta: 0:01:39 lr: 0.000395 min_lr: 0.000395 loss: 1.8494 (1.7774) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [362] [190/312] eta: 0:01:31 lr: 0.000395 min_lr: 0.000395 loss: 1.6604 (1.7643) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [362] [200/312] eta: 0:01:23 lr: 0.000395 min_lr: 0.000395 loss: 1.5875 (1.7597) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [362] [210/312] eta: 0:01:16 lr: 0.000394 min_lr: 0.000394 loss: 1.8302 (1.7594) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [362] [220/312] eta: 0:01:08 lr: 0.000394 min_lr: 0.000394 loss: 1.8856 (1.7569) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [362] [230/312] eta: 0:01:00 lr: 0.000394 min_lr: 0.000394 loss: 1.7307 (1.7571) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [362] [240/312] eta: 0:00:53 lr: 0.000394 min_lr: 0.000394 loss: 1.7359 (1.7582) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [362] [250/312] eta: 0:00:45 lr: 0.000393 min_lr: 0.000393 loss: 1.8553 (1.7624) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [362] [260/312] eta: 0:00:38 lr: 0.000393 min_lr: 0.000393 loss: 1.7989 (1.7642) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [362] [270/312] eta: 0:00:30 lr: 0.000393 min_lr: 0.000393 loss: 1.8478 (1.7694) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [362] [280/312] eta: 0:00:23 lr: 0.000392 min_lr: 0.000392 loss: 1.7925 (1.7676) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0009 max mem: 64948 Epoch: [362] [290/312] eta: 0:00:16 lr: 0.000392 min_lr: 0.000392 loss: 1.6868 (1.7649) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [362] [300/312] eta: 0:00:08 lr: 0.000392 min_lr: 0.000392 loss: 1.8156 (1.7681) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [362] [310/312] eta: 0:00:01 lr: 0.000392 min_lr: 0.000392 loss: 1.9552 (1.7711) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [362] [311/312] eta: 0:00:00 lr: 0.000392 min_lr: 0.000392 loss: 1.8641 (1.7708) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [362] Total time: 0:03:48 (0.7317 s / it) Averaged stats: lr: 0.000392 min_lr: 0.000392 loss: 1.8641 (1.7673) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.5007 (0.5007) acc1: 86.9792 (86.9792) acc5: 98.1771 (98.1771) time: 4.7426 data: 4.5298 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6524 (0.6790) acc1: 84.1146 (81.6640) acc5: 96.6146 (96.3840) time: 0.6782 data: 0.5034 max mem: 64948 Test: Total time: 0:00:06 (0.7023 s / it) * Acc@1 82.796 Acc@5 96.324 loss 0.657 Accuracy of the model on the 50000 test images: 82.8% Max accuracy: 82.81% Test: [0/9] eta: 0:00:45 loss: 0.4599 (0.4599) acc1: 88.0208 (88.0208) acc5: 97.9167 (97.9167) time: 5.0211 data: 4.8031 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6299 (0.6482) acc1: 84.6354 (82.2080) acc5: 97.1354 (96.6080) time: 0.7094 data: 0.5338 max mem: 64948 Test: Total time: 0:00:06 (0.7190 s / it) * Acc@1 83.236 Acc@5 96.556 loss 0.625 Accuracy of the model EMA on 50000 test images: 83.2% Epoch: [363] [ 0/312] eta: 0:54:49 lr: 0.000392 min_lr: 0.000392 loss: 1.2886 (1.2886) weight_decay: 0.0500 (0.0500) time: 10.5428 data: 7.8621 max mem: 64948 Epoch: [363] [ 10/312] eta: 0:08:16 lr: 0.000391 min_lr: 0.000391 loss: 1.8571 (1.7542) weight_decay: 0.0500 (0.0500) time: 1.6438 data: 0.7152 max mem: 64948 Epoch: [363] [ 20/312] eta: 0:05:47 lr: 0.000391 min_lr: 0.000391 loss: 1.8274 (1.6829) weight_decay: 0.0500 (0.0500) time: 0.7241 data: 0.0004 max mem: 64948 Epoch: [363] [ 30/312] eta: 0:04:51 lr: 0.000391 min_lr: 0.000391 loss: 1.7550 (1.6879) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [363] [ 40/312] eta: 0:04:18 lr: 0.000390 min_lr: 0.000390 loss: 1.8404 (1.7417) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [363] [ 50/312] eta: 0:03:55 lr: 0.000390 min_lr: 0.000390 loss: 1.8605 (1.7536) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [363] [ 60/312] eta: 0:03:38 lr: 0.000390 min_lr: 0.000390 loss: 1.6564 (1.7259) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [363] [ 70/312] eta: 0:03:23 lr: 0.000390 min_lr: 0.000390 loss: 1.7405 (1.7506) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [363] [ 80/312] eta: 0:03:11 lr: 0.000389 min_lr: 0.000389 loss: 1.8953 (1.7545) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [363] [ 90/312] eta: 0:03:00 lr: 0.000389 min_lr: 0.000389 loss: 1.7259 (1.7464) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [363] [100/312] eta: 0:02:49 lr: 0.000389 min_lr: 0.000389 loss: 1.8251 (1.7577) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [363] [110/312] eta: 0:02:39 lr: 0.000388 min_lr: 0.000388 loss: 1.8927 (1.7565) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [363] [120/312] eta: 0:02:30 lr: 0.000388 min_lr: 0.000388 loss: 1.7347 (1.7462) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [363] [130/312] eta: 0:02:21 lr: 0.000388 min_lr: 0.000388 loss: 1.7347 (1.7508) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [363] [140/312] eta: 0:02:12 lr: 0.000388 min_lr: 0.000388 loss: 1.8394 (1.7476) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [363] [150/312] eta: 0:02:03 lr: 0.000387 min_lr: 0.000387 loss: 1.8833 (1.7549) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [363] [160/312] eta: 0:01:55 lr: 0.000387 min_lr: 0.000387 loss: 1.8132 (1.7493) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [363] [170/312] eta: 0:01:47 lr: 0.000387 min_lr: 0.000387 loss: 1.7939 (1.7577) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [363] [180/312] eta: 0:01:39 lr: 0.000387 min_lr: 0.000387 loss: 1.7939 (1.7589) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [363] [190/312] eta: 0:01:31 lr: 0.000386 min_lr: 0.000386 loss: 1.6764 (1.7542) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [363] [200/312] eta: 0:01:23 lr: 0.000386 min_lr: 0.000386 loss: 1.5177 (1.7506) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [363] [210/312] eta: 0:01:15 lr: 0.000386 min_lr: 0.000386 loss: 1.7389 (1.7531) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0005 max mem: 64948 Epoch: [363] [220/312] eta: 0:01:08 lr: 0.000385 min_lr: 0.000385 loss: 1.7389 (1.7497) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0005 max mem: 64948 Epoch: [363] [230/312] eta: 0:01:00 lr: 0.000385 min_lr: 0.000385 loss: 1.7866 (1.7547) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [363] [240/312] eta: 0:00:53 lr: 0.000385 min_lr: 0.000385 loss: 1.8260 (1.7536) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [363] [250/312] eta: 0:00:45 lr: 0.000385 min_lr: 0.000385 loss: 1.7157 (1.7535) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [363] [260/312] eta: 0:00:38 lr: 0.000384 min_lr: 0.000384 loss: 1.8330 (1.7573) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [363] [270/312] eta: 0:00:30 lr: 0.000384 min_lr: 0.000384 loss: 1.8674 (1.7599) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [363] [280/312] eta: 0:00:23 lr: 0.000384 min_lr: 0.000384 loss: 1.8609 (1.7597) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0009 max mem: 64948 Epoch: [363] [290/312] eta: 0:00:16 lr: 0.000383 min_lr: 0.000383 loss: 1.6962 (1.7561) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [363] [300/312] eta: 0:00:08 lr: 0.000383 min_lr: 0.000383 loss: 1.7178 (1.7561) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [363] [310/312] eta: 0:00:01 lr: 0.000383 min_lr: 0.000383 loss: 1.7836 (1.7567) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [363] [311/312] eta: 0:00:00 lr: 0.000383 min_lr: 0.000383 loss: 1.7836 (1.7569) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [363] Total time: 0:03:48 (0.7311 s / it) Averaged stats: lr: 0.000383 min_lr: 0.000383 loss: 1.7836 (1.7594) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4904 (0.4904) acc1: 86.1979 (86.1979) acc5: 97.3958 (97.3958) time: 4.5119 data: 4.2932 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6657 (0.6841) acc1: 81.7708 (81.5040) acc5: 97.1354 (96.2880) time: 0.6526 data: 0.4771 max mem: 64948 Test: Total time: 0:00:06 (0.6762 s / it) * Acc@1 82.662 Acc@5 96.282 loss 0.660 Accuracy of the model on the 50000 test images: 82.7% Max accuracy: 82.81% Test: [0/9] eta: 0:00:41 loss: 0.4600 (0.4600) acc1: 88.0208 (88.0208) acc5: 97.9167 (97.9167) time: 4.5620 data: 4.3581 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6294 (0.6480) acc1: 84.6354 (82.2720) acc5: 97.1354 (96.6080) time: 0.6583 data: 0.4844 max mem: 64948 Test: Total time: 0:00:06 (0.6687 s / it) * Acc@1 83.260 Acc@5 96.566 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.26% Epoch: [364] [ 0/312] eta: 0:54:19 lr: 0.000383 min_lr: 0.000383 loss: 1.5794 (1.5794) weight_decay: 0.0500 (0.0500) time: 10.4476 data: 9.7069 max mem: 64948 Epoch: [364] [ 10/312] eta: 0:08:00 lr: 0.000383 min_lr: 0.000383 loss: 1.8640 (1.7788) weight_decay: 0.0500 (0.0500) time: 1.5905 data: 0.8827 max mem: 64948 Epoch: [364] [ 20/312] eta: 0:05:39 lr: 0.000382 min_lr: 0.000382 loss: 1.7061 (1.7335) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0003 max mem: 64948 Epoch: [364] [ 30/312] eta: 0:04:45 lr: 0.000382 min_lr: 0.000382 loss: 1.5557 (1.7223) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [364] [ 40/312] eta: 0:04:14 lr: 0.000382 min_lr: 0.000382 loss: 1.6873 (1.7415) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [364] [ 50/312] eta: 0:03:53 lr: 0.000381 min_lr: 0.000381 loss: 1.7712 (1.7276) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [364] [ 60/312] eta: 0:03:36 lr: 0.000381 min_lr: 0.000381 loss: 1.7536 (1.7269) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [364] [ 70/312] eta: 0:03:22 lr: 0.000381 min_lr: 0.000381 loss: 1.7403 (1.7292) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [364] [ 80/312] eta: 0:03:09 lr: 0.000381 min_lr: 0.000381 loss: 1.8054 (1.7435) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [364] [ 90/312] eta: 0:02:58 lr: 0.000380 min_lr: 0.000380 loss: 1.9148 (1.7430) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [364] [100/312] eta: 0:02:48 lr: 0.000380 min_lr: 0.000380 loss: 1.9218 (1.7526) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [364] [110/312] eta: 0:02:38 lr: 0.000380 min_lr: 0.000380 loss: 1.9218 (1.7566) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [364] [120/312] eta: 0:02:29 lr: 0.000380 min_lr: 0.000380 loss: 1.7864 (1.7545) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [364] [130/312] eta: 0:02:20 lr: 0.000379 min_lr: 0.000379 loss: 1.9101 (1.7610) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [364] [140/312] eta: 0:02:11 lr: 0.000379 min_lr: 0.000379 loss: 1.7560 (1.7539) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [364] [150/312] eta: 0:02:03 lr: 0.000379 min_lr: 0.000379 loss: 1.6016 (1.7344) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [364] [160/312] eta: 0:01:55 lr: 0.000378 min_lr: 0.000378 loss: 1.5060 (1.7328) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [364] [170/312] eta: 0:01:47 lr: 0.000378 min_lr: 0.000378 loss: 1.8336 (1.7384) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [364] [180/312] eta: 0:01:39 lr: 0.000378 min_lr: 0.000378 loss: 1.7997 (1.7383) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [364] [190/312] eta: 0:01:31 lr: 0.000378 min_lr: 0.000378 loss: 1.7157 (1.7404) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [364] [200/312] eta: 0:01:23 lr: 0.000377 min_lr: 0.000377 loss: 1.7203 (1.7381) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [364] [210/312] eta: 0:01:15 lr: 0.000377 min_lr: 0.000377 loss: 1.7892 (1.7386) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [364] [220/312] eta: 0:01:08 lr: 0.000377 min_lr: 0.000377 loss: 1.8537 (1.7403) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [364] [230/312] eta: 0:01:00 lr: 0.000377 min_lr: 0.000377 loss: 1.8858 (1.7418) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [364] [240/312] eta: 0:00:53 lr: 0.000376 min_lr: 0.000376 loss: 1.9390 (1.7394) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [364] [250/312] eta: 0:00:45 lr: 0.000376 min_lr: 0.000376 loss: 1.7010 (1.7360) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [364] [260/312] eta: 0:00:38 lr: 0.000376 min_lr: 0.000376 loss: 1.8503 (1.7371) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [364] [270/312] eta: 0:00:30 lr: 0.000375 min_lr: 0.000375 loss: 1.8503 (1.7421) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [364] [280/312] eta: 0:00:23 lr: 0.000375 min_lr: 0.000375 loss: 1.7932 (1.7398) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0009 max mem: 64948 Epoch: [364] [290/312] eta: 0:00:16 lr: 0.000375 min_lr: 0.000375 loss: 1.7775 (1.7402) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [364] [300/312] eta: 0:00:08 lr: 0.000375 min_lr: 0.000375 loss: 1.8593 (1.7445) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [364] [310/312] eta: 0:00:01 lr: 0.000374 min_lr: 0.000374 loss: 1.8633 (1.7472) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [364] [311/312] eta: 0:00:00 lr: 0.000374 min_lr: 0.000374 loss: 1.9087 (1.7480) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [364] Total time: 0:03:47 (0.7297 s / it) Averaged stats: lr: 0.000374 min_lr: 0.000374 loss: 1.9087 (1.7617) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4798 (0.4798) acc1: 86.7188 (86.7188) acc5: 98.4375 (98.4375) time: 4.5788 data: 4.3625 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6437 (0.6620) acc1: 83.3333 (82.0160) acc5: 97.1354 (96.5120) time: 0.6600 data: 0.4848 max mem: 64948 Test: Total time: 0:00:06 (0.6850 s / it) * Acc@1 82.984 Acc@5 96.410 loss 0.651 Accuracy of the model on the 50000 test images: 83.0% Max accuracy: 82.98% Test: [0/9] eta: 0:00:42 loss: 0.4603 (0.4603) acc1: 88.0208 (88.0208) acc5: 97.9167 (97.9167) time: 4.7444 data: 4.5320 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6289 (0.6480) acc1: 84.6354 (82.4000) acc5: 97.1354 (96.6080) time: 0.6785 data: 0.5037 max mem: 64948 Test: Total time: 0:00:06 (0.6874 s / it) * Acc@1 83.288 Acc@5 96.568 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.29% Epoch: [365] [ 0/312] eta: 0:53:34 lr: 0.000374 min_lr: 0.000374 loss: 1.8945 (1.8945) weight_decay: 0.0500 (0.0500) time: 10.3042 data: 8.8014 max mem: 64948 Epoch: [365] [ 10/312] eta: 0:07:58 lr: 0.000374 min_lr: 0.000374 loss: 1.8326 (1.7666) weight_decay: 0.0500 (0.0500) time: 1.5842 data: 0.8005 max mem: 64948 Epoch: [365] [ 20/312] eta: 0:05:38 lr: 0.000374 min_lr: 0.000374 loss: 1.8869 (1.8671) weight_decay: 0.0500 (0.0500) time: 0.7029 data: 0.0004 max mem: 64948 Epoch: [365] [ 30/312] eta: 0:04:45 lr: 0.000374 min_lr: 0.000374 loss: 1.8976 (1.8325) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [365] [ 40/312] eta: 0:04:14 lr: 0.000373 min_lr: 0.000373 loss: 1.9038 (1.8265) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [365] [ 50/312] eta: 0:03:52 lr: 0.000373 min_lr: 0.000373 loss: 1.6366 (1.7834) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [365] [ 60/312] eta: 0:03:35 lr: 0.000373 min_lr: 0.000373 loss: 1.7469 (1.7884) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [365] [ 70/312] eta: 0:03:21 lr: 0.000372 min_lr: 0.000372 loss: 1.8644 (1.7920) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [365] [ 80/312] eta: 0:03:09 lr: 0.000372 min_lr: 0.000372 loss: 1.8189 (1.7998) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [365] [ 90/312] eta: 0:02:58 lr: 0.000372 min_lr: 0.000372 loss: 1.7981 (1.7859) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [365] [100/312] eta: 0:02:47 lr: 0.000372 min_lr: 0.000372 loss: 1.8615 (1.7898) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [365] [110/312] eta: 0:02:38 lr: 0.000371 min_lr: 0.000371 loss: 1.8615 (1.7785) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [365] [120/312] eta: 0:02:29 lr: 0.000371 min_lr: 0.000371 loss: 1.7013 (1.7761) weight_decay: 0.0500 (0.0500) time: 0.7015 data: 0.0004 max mem: 64948 Epoch: [365] [130/312] eta: 0:02:20 lr: 0.000371 min_lr: 0.000371 loss: 1.7772 (1.7793) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [365] [140/312] eta: 0:02:11 lr: 0.000371 min_lr: 0.000371 loss: 1.8374 (1.7814) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [365] [150/312] eta: 0:02:03 lr: 0.000370 min_lr: 0.000370 loss: 1.7498 (1.7727) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [365] [160/312] eta: 0:01:55 lr: 0.000370 min_lr: 0.000370 loss: 1.8123 (1.7744) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [365] [170/312] eta: 0:01:46 lr: 0.000370 min_lr: 0.000370 loss: 1.9080 (1.7767) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [365] [180/312] eta: 0:01:38 lr: 0.000369 min_lr: 0.000369 loss: 1.9080 (1.7694) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [365] [190/312] eta: 0:01:31 lr: 0.000369 min_lr: 0.000369 loss: 1.5210 (1.7587) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [365] [200/312] eta: 0:01:23 lr: 0.000369 min_lr: 0.000369 loss: 1.7722 (1.7627) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [365] [210/312] eta: 0:01:15 lr: 0.000369 min_lr: 0.000369 loss: 1.7969 (1.7642) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [365] [220/312] eta: 0:01:08 lr: 0.000368 min_lr: 0.000368 loss: 1.7075 (1.7590) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [365] [230/312] eta: 0:01:00 lr: 0.000368 min_lr: 0.000368 loss: 1.7720 (1.7638) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [365] [240/312] eta: 0:00:52 lr: 0.000368 min_lr: 0.000368 loss: 1.7950 (1.7602) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [365] [250/312] eta: 0:00:45 lr: 0.000368 min_lr: 0.000368 loss: 1.7794 (1.7600) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [365] [260/312] eta: 0:00:38 lr: 0.000367 min_lr: 0.000367 loss: 1.8744 (1.7612) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [365] [270/312] eta: 0:00:30 lr: 0.000367 min_lr: 0.000367 loss: 1.9178 (1.7669) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [365] [280/312] eta: 0:00:23 lr: 0.000367 min_lr: 0.000367 loss: 1.8872 (1.7686) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [365] [290/312] eta: 0:00:16 lr: 0.000366 min_lr: 0.000366 loss: 1.7711 (1.7703) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [365] [300/312] eta: 0:00:08 lr: 0.000366 min_lr: 0.000366 loss: 1.7373 (1.7686) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [365] [310/312] eta: 0:00:01 lr: 0.000366 min_lr: 0.000366 loss: 1.7204 (1.7684) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [365] [311/312] eta: 0:00:00 lr: 0.000366 min_lr: 0.000366 loss: 1.7373 (1.7686) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [365] Total time: 0:03:47 (0.7290 s / it) Averaged stats: lr: 0.000366 min_lr: 0.000366 loss: 1.7373 (1.7584) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4716 (0.4716) acc1: 86.9792 (86.9792) acc5: 98.1771 (98.1771) time: 4.7348 data: 4.5235 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6522 (0.6663) acc1: 83.0729 (82.3360) acc5: 97.1354 (96.5440) time: 0.6773 data: 0.5027 max mem: 64948 Test: Total time: 0:00:06 (0.7009 s / it) * Acc@1 83.014 Acc@5 96.388 loss 0.648 Accuracy of the model on the 50000 test images: 83.0% Max accuracy: 83.01% Test: [0/9] eta: 0:00:43 loss: 0.4600 (0.4600) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.8746 data: 4.6568 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6286 (0.6478) acc1: 84.6354 (82.4000) acc5: 97.1354 (96.6080) time: 0.6930 data: 0.5175 max mem: 64948 Test: Total time: 0:00:06 (0.7002 s / it) * Acc@1 83.300 Acc@5 96.574 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.30% Epoch: [366] [ 0/312] eta: 0:49:42 lr: 0.000366 min_lr: 0.000366 loss: 2.0626 (2.0626) weight_decay: 0.0500 (0.0500) time: 9.5578 data: 7.6705 max mem: 64948 Epoch: [366] [ 10/312] eta: 0:07:39 lr: 0.000366 min_lr: 0.000366 loss: 1.7357 (1.7803) weight_decay: 0.0500 (0.0500) time: 1.5216 data: 0.6977 max mem: 64948 Epoch: [366] [ 20/312] eta: 0:05:29 lr: 0.000365 min_lr: 0.000365 loss: 1.7357 (1.7841) weight_decay: 0.0500 (0.0500) time: 0.7074 data: 0.0004 max mem: 64948 Epoch: [366] [ 30/312] eta: 0:04:38 lr: 0.000365 min_lr: 0.000365 loss: 1.8824 (1.8294) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [366] [ 40/312] eta: 0:04:09 lr: 0.000365 min_lr: 0.000365 loss: 1.9026 (1.8170) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [366] [ 50/312] eta: 0:03:48 lr: 0.000365 min_lr: 0.000365 loss: 1.8581 (1.8084) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [366] [ 60/312] eta: 0:03:32 lr: 0.000364 min_lr: 0.000364 loss: 1.7846 (1.7718) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [366] [ 70/312] eta: 0:03:19 lr: 0.000364 min_lr: 0.000364 loss: 1.7788 (1.7611) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [366] [ 80/312] eta: 0:03:07 lr: 0.000364 min_lr: 0.000364 loss: 1.7790 (1.7664) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [366] [ 90/312] eta: 0:02:56 lr: 0.000363 min_lr: 0.000363 loss: 1.7814 (1.7660) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [366] [100/312] eta: 0:02:46 lr: 0.000363 min_lr: 0.000363 loss: 1.7462 (1.7509) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [366] [110/312] eta: 0:02:36 lr: 0.000363 min_lr: 0.000363 loss: 1.8016 (1.7624) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [366] [120/312] eta: 0:02:27 lr: 0.000363 min_lr: 0.000363 loss: 1.9148 (1.7660) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [366] [130/312] eta: 0:02:19 lr: 0.000362 min_lr: 0.000362 loss: 1.9148 (1.7736) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [366] [140/312] eta: 0:02:10 lr: 0.000362 min_lr: 0.000362 loss: 1.8526 (1.7678) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [366] [150/312] eta: 0:02:02 lr: 0.000362 min_lr: 0.000362 loss: 1.8526 (1.7763) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [366] [160/312] eta: 0:01:54 lr: 0.000362 min_lr: 0.000362 loss: 1.8592 (1.7790) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [366] [170/312] eta: 0:01:46 lr: 0.000361 min_lr: 0.000361 loss: 1.8592 (1.7833) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [366] [180/312] eta: 0:01:38 lr: 0.000361 min_lr: 0.000361 loss: 1.8076 (1.7826) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [366] [190/312] eta: 0:01:30 lr: 0.000361 min_lr: 0.000361 loss: 1.8076 (1.7893) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [366] [200/312] eta: 0:01:22 lr: 0.000360 min_lr: 0.000360 loss: 1.8526 (1.7875) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [366] [210/312] eta: 0:01:15 lr: 0.000360 min_lr: 0.000360 loss: 1.6517 (1.7827) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [366] [220/312] eta: 0:01:07 lr: 0.000360 min_lr: 0.000360 loss: 1.6517 (1.7794) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [366] [230/312] eta: 0:01:00 lr: 0.000360 min_lr: 0.000360 loss: 1.7497 (1.7783) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [366] [240/312] eta: 0:00:52 lr: 0.000359 min_lr: 0.000359 loss: 1.8111 (1.7796) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [366] [250/312] eta: 0:00:45 lr: 0.000359 min_lr: 0.000359 loss: 1.7923 (1.7798) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [366] [260/312] eta: 0:00:37 lr: 0.000359 min_lr: 0.000359 loss: 1.7991 (1.7777) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [366] [270/312] eta: 0:00:30 lr: 0.000359 min_lr: 0.000359 loss: 1.7911 (1.7745) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [366] [280/312] eta: 0:00:23 lr: 0.000358 min_lr: 0.000358 loss: 1.7694 (1.7749) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [366] [290/312] eta: 0:00:15 lr: 0.000358 min_lr: 0.000358 loss: 1.8818 (1.7809) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [366] [300/312] eta: 0:00:08 lr: 0.000358 min_lr: 0.000358 loss: 1.8937 (1.7816) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [366] [310/312] eta: 0:00:01 lr: 0.000358 min_lr: 0.000358 loss: 1.8708 (1.7803) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [366] [311/312] eta: 0:00:00 lr: 0.000358 min_lr: 0.000358 loss: 1.8904 (1.7812) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [366] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.000358 min_lr: 0.000358 loss: 1.8904 (1.7613) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.4486 (0.4486) acc1: 88.2812 (88.2812) acc5: 98.1771 (98.1771) time: 4.9014 data: 4.6802 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6461 (0.6606) acc1: 83.8542 (81.8240) acc5: 96.6146 (96.4800) time: 0.6962 data: 0.5201 max mem: 64948 Test: Total time: 0:00:06 (0.7227 s / it) * Acc@1 82.930 Acc@5 96.354 loss 0.644 Accuracy of the model on the 50000 test images: 82.9% Max accuracy: 83.01% Test: [0/9] eta: 0:00:46 loss: 0.4597 (0.4597) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.2116 data: 5.0073 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6278 (0.6476) acc1: 84.6354 (82.3680) acc5: 97.1354 (96.6400) time: 0.7304 data: 0.5565 max mem: 64948 Test: Total time: 0:00:06 (0.7415 s / it) * Acc@1 83.316 Acc@5 96.574 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.32% Epoch: [367] [ 0/312] eta: 0:52:52 lr: 0.000358 min_lr: 0.000358 loss: 1.9610 (1.9610) weight_decay: 0.0500 (0.0500) time: 10.1686 data: 8.4709 max mem: 64948 Epoch: [367] [ 10/312] eta: 0:07:56 lr: 0.000357 min_lr: 0.000357 loss: 1.8193 (1.8233) weight_decay: 0.0500 (0.0500) time: 1.5765 data: 0.7705 max mem: 64948 Epoch: [367] [ 20/312] eta: 0:05:37 lr: 0.000357 min_lr: 0.000357 loss: 1.7795 (1.7825) weight_decay: 0.0500 (0.0500) time: 0.7050 data: 0.0004 max mem: 64948 Epoch: [367] [ 30/312] eta: 0:04:44 lr: 0.000357 min_lr: 0.000357 loss: 1.7986 (1.7891) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [367] [ 40/312] eta: 0:04:13 lr: 0.000356 min_lr: 0.000356 loss: 1.8236 (1.7969) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [367] [ 50/312] eta: 0:03:52 lr: 0.000356 min_lr: 0.000356 loss: 1.8900 (1.8077) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [367] [ 60/312] eta: 0:03:35 lr: 0.000356 min_lr: 0.000356 loss: 1.9736 (1.7802) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [367] [ 70/312] eta: 0:03:21 lr: 0.000356 min_lr: 0.000356 loss: 1.7879 (1.7723) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [367] [ 80/312] eta: 0:03:09 lr: 0.000355 min_lr: 0.000355 loss: 1.8406 (1.7830) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [367] [ 90/312] eta: 0:02:58 lr: 0.000355 min_lr: 0.000355 loss: 1.8184 (1.7738) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [367] [100/312] eta: 0:02:47 lr: 0.000355 min_lr: 0.000355 loss: 1.8184 (1.7809) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [367] [110/312] eta: 0:02:38 lr: 0.000355 min_lr: 0.000355 loss: 1.8581 (1.7950) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [367] [120/312] eta: 0:02:28 lr: 0.000354 min_lr: 0.000354 loss: 1.8649 (1.7971) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [367] [130/312] eta: 0:02:20 lr: 0.000354 min_lr: 0.000354 loss: 1.8256 (1.7883) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [367] [140/312] eta: 0:02:11 lr: 0.000354 min_lr: 0.000354 loss: 1.8816 (1.7938) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [367] [150/312] eta: 0:02:03 lr: 0.000354 min_lr: 0.000354 loss: 1.8816 (1.7927) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [367] [160/312] eta: 0:01:54 lr: 0.000353 min_lr: 0.000353 loss: 1.7434 (1.7926) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [367] [170/312] eta: 0:01:46 lr: 0.000353 min_lr: 0.000353 loss: 1.7559 (1.7917) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [367] [180/312] eta: 0:01:38 lr: 0.000353 min_lr: 0.000353 loss: 1.8427 (1.7952) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [367] [190/312] eta: 0:01:31 lr: 0.000352 min_lr: 0.000352 loss: 1.8552 (1.8002) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [367] [200/312] eta: 0:01:23 lr: 0.000352 min_lr: 0.000352 loss: 1.7477 (1.7966) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [367] [210/312] eta: 0:01:15 lr: 0.000352 min_lr: 0.000352 loss: 1.6973 (1.7899) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [367] [220/312] eta: 0:01:07 lr: 0.000352 min_lr: 0.000352 loss: 1.6996 (1.7932) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [367] [230/312] eta: 0:01:00 lr: 0.000351 min_lr: 0.000351 loss: 1.7908 (1.7970) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [367] [240/312] eta: 0:00:52 lr: 0.000351 min_lr: 0.000351 loss: 1.7753 (1.7930) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [367] [250/312] eta: 0:00:45 lr: 0.000351 min_lr: 0.000351 loss: 1.7406 (1.7944) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [367] [260/312] eta: 0:00:38 lr: 0.000351 min_lr: 0.000351 loss: 1.7595 (1.7903) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [367] [270/312] eta: 0:00:30 lr: 0.000350 min_lr: 0.000350 loss: 1.8365 (1.7929) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [367] [280/312] eta: 0:00:23 lr: 0.000350 min_lr: 0.000350 loss: 1.8587 (1.7881) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0009 max mem: 64948 Epoch: [367] [290/312] eta: 0:00:16 lr: 0.000350 min_lr: 0.000350 loss: 1.8587 (1.7912) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [367] [300/312] eta: 0:00:08 lr: 0.000350 min_lr: 0.000350 loss: 1.8004 (1.7918) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [367] [310/312] eta: 0:00:01 lr: 0.000349 min_lr: 0.000349 loss: 1.8159 (1.7947) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [367] [311/312] eta: 0:00:00 lr: 0.000349 min_lr: 0.000349 loss: 1.8159 (1.7931) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [367] Total time: 0:03:47 (0.7288 s / it) Averaged stats: lr: 0.000349 min_lr: 0.000349 loss: 1.8159 (1.7500) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4611 (0.4611) acc1: 87.5000 (87.5000) acc5: 98.1771 (98.1771) time: 4.8050 data: 4.5938 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6729 (0.6726) acc1: 82.8125 (81.8240) acc5: 96.8750 (96.4160) time: 0.6852 data: 0.5105 max mem: 64948 Test: Total time: 0:00:06 (0.7090 s / it) * Acc@1 82.980 Acc@5 96.356 loss 0.650 Accuracy of the model on the 50000 test images: 83.0% Max accuracy: 83.01% Test: [0/9] eta: 0:00:46 loss: 0.4596 (0.4596) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.1159 data: 4.9089 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6274 (0.6474) acc1: 84.6354 (82.3680) acc5: 97.1354 (96.6400) time: 0.7198 data: 0.5455 max mem: 64948 Test: Total time: 0:00:06 (0.7502 s / it) * Acc@1 83.326 Acc@5 96.582 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.33% Epoch: [368] [ 0/312] eta: 0:45:13 lr: 0.000349 min_lr: 0.000349 loss: 1.3037 (1.3037) weight_decay: 0.0500 (0.0500) time: 8.6979 data: 7.7397 max mem: 64948 Epoch: [368] [ 10/312] eta: 0:07:35 lr: 0.000349 min_lr: 0.000349 loss: 1.8723 (1.7568) weight_decay: 0.0500 (0.0500) time: 1.5070 data: 0.7596 max mem: 64948 Epoch: [368] [ 20/312] eta: 0:05:27 lr: 0.000349 min_lr: 0.000349 loss: 1.8723 (1.8105) weight_decay: 0.0500 (0.0500) time: 0.7425 data: 0.0309 max mem: 64948 Epoch: [368] [ 30/312] eta: 0:04:37 lr: 0.000348 min_lr: 0.000348 loss: 1.9231 (1.8411) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0003 max mem: 64948 Epoch: [368] [ 40/312] eta: 0:04:08 lr: 0.000348 min_lr: 0.000348 loss: 1.7432 (1.8086) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [368] [ 50/312] eta: 0:03:48 lr: 0.000348 min_lr: 0.000348 loss: 1.7432 (1.8198) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [368] [ 60/312] eta: 0:03:32 lr: 0.000348 min_lr: 0.000348 loss: 1.7618 (1.7954) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [368] [ 70/312] eta: 0:03:19 lr: 0.000347 min_lr: 0.000347 loss: 1.7271 (1.7646) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [368] [ 80/312] eta: 0:03:07 lr: 0.000347 min_lr: 0.000347 loss: 1.7308 (1.7773) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [368] [ 90/312] eta: 0:02:56 lr: 0.000347 min_lr: 0.000347 loss: 1.7739 (1.7714) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [368] [100/312] eta: 0:02:46 lr: 0.000347 min_lr: 0.000347 loss: 1.7739 (1.7807) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [368] [110/312] eta: 0:02:37 lr: 0.000346 min_lr: 0.000346 loss: 1.9124 (1.7715) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [368] [120/312] eta: 0:02:27 lr: 0.000346 min_lr: 0.000346 loss: 1.7944 (1.7625) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [368] [130/312] eta: 0:02:19 lr: 0.000346 min_lr: 0.000346 loss: 1.8569 (1.7687) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [368] [140/312] eta: 0:02:10 lr: 0.000346 min_lr: 0.000346 loss: 1.8909 (1.7718) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [368] [150/312] eta: 0:02:02 lr: 0.000345 min_lr: 0.000345 loss: 1.7831 (1.7744) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [368] [160/312] eta: 0:01:54 lr: 0.000345 min_lr: 0.000345 loss: 1.6648 (1.7678) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [368] [170/312] eta: 0:01:46 lr: 0.000345 min_lr: 0.000345 loss: 1.5616 (1.7603) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [368] [180/312] eta: 0:01:38 lr: 0.000344 min_lr: 0.000344 loss: 1.6182 (1.7547) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [368] [190/312] eta: 0:01:30 lr: 0.000344 min_lr: 0.000344 loss: 1.7342 (1.7600) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [368] [200/312] eta: 0:01:22 lr: 0.000344 min_lr: 0.000344 loss: 1.7774 (1.7581) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [368] [210/312] eta: 0:01:15 lr: 0.000344 min_lr: 0.000344 loss: 1.8093 (1.7582) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [368] [220/312] eta: 0:01:07 lr: 0.000343 min_lr: 0.000343 loss: 1.7977 (1.7482) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [368] [230/312] eta: 0:01:00 lr: 0.000343 min_lr: 0.000343 loss: 1.7109 (1.7478) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [368] [240/312] eta: 0:00:52 lr: 0.000343 min_lr: 0.000343 loss: 1.7109 (1.7452) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [368] [250/312] eta: 0:00:45 lr: 0.000343 min_lr: 0.000343 loss: 1.8223 (1.7504) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [368] [260/312] eta: 0:00:37 lr: 0.000342 min_lr: 0.000342 loss: 1.7119 (1.7483) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [368] [270/312] eta: 0:00:30 lr: 0.000342 min_lr: 0.000342 loss: 1.5760 (1.7425) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [368] [280/312] eta: 0:00:23 lr: 0.000342 min_lr: 0.000342 loss: 1.6026 (1.7428) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [368] [290/312] eta: 0:00:15 lr: 0.000342 min_lr: 0.000342 loss: 1.6732 (1.7393) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [368] [300/312] eta: 0:00:08 lr: 0.000341 min_lr: 0.000341 loss: 1.6993 (1.7390) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [368] [310/312] eta: 0:00:01 lr: 0.000341 min_lr: 0.000341 loss: 1.6755 (1.7363) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [368] [311/312] eta: 0:00:00 lr: 0.000341 min_lr: 0.000341 loss: 1.7095 (1.7368) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [368] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.000341 min_lr: 0.000341 loss: 1.7095 (1.7509) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4720 (0.4720) acc1: 88.2812 (88.2812) acc5: 98.4375 (98.4375) time: 4.4511 data: 4.2390 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6588 (0.6571) acc1: 83.3333 (81.7600) acc5: 96.8750 (96.7040) time: 0.6458 data: 0.4711 max mem: 64948 Test: Total time: 0:00:06 (0.6681 s / it) * Acc@1 83.002 Acc@5 96.414 loss 0.649 Accuracy of the model on the 50000 test images: 83.0% Max accuracy: 83.01% Test: [0/9] eta: 0:00:44 loss: 0.4594 (0.4594) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.8981 data: 4.6801 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6270 (0.6473) acc1: 84.6354 (82.3040) acc5: 97.1354 (96.6400) time: 0.6996 data: 0.5201 max mem: 64948 Test: Total time: 0:00:06 (0.7144 s / it) * Acc@1 83.326 Acc@5 96.590 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.33% Epoch: [369] [ 0/312] eta: 0:45:48 lr: 0.000341 min_lr: 0.000341 loss: 1.9321 (1.9321) weight_decay: 0.0500 (0.0500) time: 8.8081 data: 7.9151 max mem: 64948 Epoch: [369] [ 10/312] eta: 0:07:27 lr: 0.000341 min_lr: 0.000341 loss: 1.8648 (1.7901) weight_decay: 0.0500 (0.0500) time: 1.4827 data: 0.7200 max mem: 64948 Epoch: [369] [ 20/312] eta: 0:05:23 lr: 0.000341 min_lr: 0.000341 loss: 1.6475 (1.6824) weight_decay: 0.0500 (0.0500) time: 0.7240 data: 0.0004 max mem: 64948 Epoch: [369] [ 30/312] eta: 0:04:35 lr: 0.000340 min_lr: 0.000340 loss: 1.6787 (1.7333) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [369] [ 40/312] eta: 0:04:07 lr: 0.000340 min_lr: 0.000340 loss: 1.6787 (1.7217) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [369] [ 50/312] eta: 0:03:47 lr: 0.000340 min_lr: 0.000340 loss: 1.7785 (1.7303) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [369] [ 60/312] eta: 0:03:31 lr: 0.000339 min_lr: 0.000339 loss: 1.7748 (1.6980) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [369] [ 70/312] eta: 0:03:18 lr: 0.000339 min_lr: 0.000339 loss: 1.7748 (1.7149) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [369] [ 80/312] eta: 0:03:06 lr: 0.000339 min_lr: 0.000339 loss: 1.8245 (1.7245) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [369] [ 90/312] eta: 0:02:55 lr: 0.000339 min_lr: 0.000339 loss: 1.9361 (1.7544) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [369] [100/312] eta: 0:02:45 lr: 0.000338 min_lr: 0.000338 loss: 1.9361 (1.7609) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [369] [110/312] eta: 0:02:36 lr: 0.000338 min_lr: 0.000338 loss: 1.8207 (1.7520) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [369] [120/312] eta: 0:02:27 lr: 0.000338 min_lr: 0.000338 loss: 1.6620 (1.7565) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [369] [130/312] eta: 0:02:18 lr: 0.000338 min_lr: 0.000338 loss: 1.8794 (1.7631) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [369] [140/312] eta: 0:02:10 lr: 0.000337 min_lr: 0.000337 loss: 1.8587 (1.7678) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [369] [150/312] eta: 0:02:02 lr: 0.000337 min_lr: 0.000337 loss: 1.8131 (1.7742) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [369] [160/312] eta: 0:01:53 lr: 0.000337 min_lr: 0.000337 loss: 1.7714 (1.7559) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [369] [170/312] eta: 0:01:46 lr: 0.000337 min_lr: 0.000337 loss: 1.6216 (1.7577) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [369] [180/312] eta: 0:01:38 lr: 0.000336 min_lr: 0.000336 loss: 1.7564 (1.7566) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [369] [190/312] eta: 0:01:30 lr: 0.000336 min_lr: 0.000336 loss: 1.7271 (1.7598) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [369] [200/312] eta: 0:01:22 lr: 0.000336 min_lr: 0.000336 loss: 1.7263 (1.7556) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [369] [210/312] eta: 0:01:15 lr: 0.000336 min_lr: 0.000336 loss: 1.7154 (1.7578) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [369] [220/312] eta: 0:01:07 lr: 0.000335 min_lr: 0.000335 loss: 1.9082 (1.7610) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [369] [230/312] eta: 0:01:00 lr: 0.000335 min_lr: 0.000335 loss: 1.8629 (1.7606) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [369] [240/312] eta: 0:00:52 lr: 0.000335 min_lr: 0.000335 loss: 1.7003 (1.7579) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [369] [250/312] eta: 0:00:45 lr: 0.000335 min_lr: 0.000335 loss: 1.6673 (1.7579) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [369] [260/312] eta: 0:00:37 lr: 0.000334 min_lr: 0.000334 loss: 1.6266 (1.7532) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [369] [270/312] eta: 0:00:30 lr: 0.000334 min_lr: 0.000334 loss: 1.6874 (1.7524) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [369] [280/312] eta: 0:00:23 lr: 0.000334 min_lr: 0.000334 loss: 1.7827 (1.7562) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [369] [290/312] eta: 0:00:15 lr: 0.000333 min_lr: 0.000333 loss: 1.7394 (1.7529) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [369] [300/312] eta: 0:00:08 lr: 0.000333 min_lr: 0.000333 loss: 1.7394 (1.7544) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [369] [310/312] eta: 0:00:01 lr: 0.000333 min_lr: 0.000333 loss: 1.8542 (1.7588) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [369] [311/312] eta: 0:00:00 lr: 0.000333 min_lr: 0.000333 loss: 1.8542 (1.7592) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [369] Total time: 0:03:46 (0.7257 s / it) Averaged stats: lr: 0.000333 min_lr: 0.000333 loss: 1.8542 (1.7507) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4554 (0.4554) acc1: 87.2396 (87.2396) acc5: 97.3958 (97.3958) time: 4.8575 data: 4.6514 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6462 (0.6648) acc1: 83.3333 (82.2400) acc5: 96.8750 (96.5120) time: 0.6910 data: 0.5169 max mem: 64948 Test: Total time: 0:00:06 (0.7136 s / it) * Acc@1 82.884 Acc@5 96.352 loss 0.647 Accuracy of the model on the 50000 test images: 82.9% Max accuracy: 83.01% Test: [0/9] eta: 0:00:42 loss: 0.4589 (0.4589) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.7147 data: 4.5099 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6268 (0.6471) acc1: 84.6354 (82.2080) acc5: 97.1354 (96.6400) time: 0.6751 data: 0.5012 max mem: 64948 Test: Total time: 0:00:06 (0.6913 s / it) * Acc@1 83.316 Acc@5 96.598 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Epoch: [370] [ 0/312] eta: 0:55:28 lr: 0.000333 min_lr: 0.000333 loss: 1.5719 (1.5719) weight_decay: 0.0500 (0.0500) time: 10.6689 data: 9.2734 max mem: 64948 Epoch: [370] [ 10/312] eta: 0:08:11 lr: 0.000333 min_lr: 0.000333 loss: 1.7902 (1.7862) weight_decay: 0.0500 (0.0500) time: 1.6273 data: 0.8434 max mem: 64948 Epoch: [370] [ 20/312] eta: 0:05:46 lr: 0.000332 min_lr: 0.000332 loss: 1.8260 (1.8383) weight_decay: 0.0500 (0.0500) time: 0.7116 data: 0.0003 max mem: 64948 Epoch: [370] [ 30/312] eta: 0:04:49 lr: 0.000332 min_lr: 0.000332 loss: 1.8260 (1.7861) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [370] [ 40/312] eta: 0:04:17 lr: 0.000332 min_lr: 0.000332 loss: 1.7789 (1.7687) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [370] [ 50/312] eta: 0:03:54 lr: 0.000332 min_lr: 0.000332 loss: 1.8657 (1.7911) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [370] [ 60/312] eta: 0:03:37 lr: 0.000331 min_lr: 0.000331 loss: 1.8571 (1.7788) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [370] [ 70/312] eta: 0:03:23 lr: 0.000331 min_lr: 0.000331 loss: 1.7624 (1.7732) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [370] [ 80/312] eta: 0:03:10 lr: 0.000331 min_lr: 0.000331 loss: 1.8527 (1.7810) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [370] [ 90/312] eta: 0:02:59 lr: 0.000331 min_lr: 0.000331 loss: 1.8527 (1.7804) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [370] [100/312] eta: 0:02:49 lr: 0.000330 min_lr: 0.000330 loss: 1.8388 (1.7792) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [370] [110/312] eta: 0:02:39 lr: 0.000330 min_lr: 0.000330 loss: 1.5428 (1.7562) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [370] [120/312] eta: 0:02:29 lr: 0.000330 min_lr: 0.000330 loss: 1.7136 (1.7711) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [370] [130/312] eta: 0:02:20 lr: 0.000330 min_lr: 0.000330 loss: 1.9020 (1.7727) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [370] [140/312] eta: 0:02:12 lr: 0.000329 min_lr: 0.000329 loss: 1.8685 (1.7796) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [370] [150/312] eta: 0:02:03 lr: 0.000329 min_lr: 0.000329 loss: 1.8429 (1.7853) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [370] [160/312] eta: 0:01:55 lr: 0.000329 min_lr: 0.000329 loss: 1.7444 (1.7758) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [370] [170/312] eta: 0:01:47 lr: 0.000329 min_lr: 0.000329 loss: 1.6936 (1.7707) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [370] [180/312] eta: 0:01:39 lr: 0.000328 min_lr: 0.000328 loss: 1.7952 (1.7727) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [370] [190/312] eta: 0:01:31 lr: 0.000328 min_lr: 0.000328 loss: 1.9134 (1.7787) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [370] [200/312] eta: 0:01:23 lr: 0.000328 min_lr: 0.000328 loss: 1.7791 (1.7707) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [370] [210/312] eta: 0:01:15 lr: 0.000328 min_lr: 0.000328 loss: 1.7748 (1.7742) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [370] [220/312] eta: 0:01:08 lr: 0.000327 min_lr: 0.000327 loss: 1.7748 (1.7680) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [370] [230/312] eta: 0:01:00 lr: 0.000327 min_lr: 0.000327 loss: 1.5605 (1.7599) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [370] [240/312] eta: 0:00:53 lr: 0.000327 min_lr: 0.000327 loss: 1.7140 (1.7628) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [370] [250/312] eta: 0:00:45 lr: 0.000326 min_lr: 0.000326 loss: 1.8546 (1.7622) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [370] [260/312] eta: 0:00:38 lr: 0.000326 min_lr: 0.000326 loss: 1.8571 (1.7687) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [370] [270/312] eta: 0:00:30 lr: 0.000326 min_lr: 0.000326 loss: 1.9408 (1.7664) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [370] [280/312] eta: 0:00:23 lr: 0.000326 min_lr: 0.000326 loss: 1.8001 (1.7651) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [370] [290/312] eta: 0:00:16 lr: 0.000325 min_lr: 0.000325 loss: 1.8196 (1.7653) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0008 max mem: 64948 Epoch: [370] [300/312] eta: 0:00:08 lr: 0.000325 min_lr: 0.000325 loss: 1.8711 (1.7656) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [370] [310/312] eta: 0:00:01 lr: 0.000325 min_lr: 0.000325 loss: 1.8431 (1.7680) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [370] [311/312] eta: 0:00:00 lr: 0.000325 min_lr: 0.000325 loss: 1.8431 (1.7692) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [370] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.000325 min_lr: 0.000325 loss: 1.8431 (1.7442) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4461 (0.4461) acc1: 88.5417 (88.5417) acc5: 98.4375 (98.4375) time: 4.6341 data: 4.4156 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6615 (0.6591) acc1: 82.8125 (81.9200) acc5: 97.1354 (96.7040) time: 0.6662 data: 0.4907 max mem: 64948 Test: Total time: 0:00:06 (0.6902 s / it) * Acc@1 83.036 Acc@5 96.358 loss 0.646 Accuracy of the model on the 50000 test images: 83.0% Max accuracy: 83.04% Test: [0/9] eta: 0:00:44 loss: 0.4584 (0.4584) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.9241 data: 4.7062 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6266 (0.6468) acc1: 84.6354 (82.2080) acc5: 97.1354 (96.6400) time: 0.6988 data: 0.5230 max mem: 64948 Test: Total time: 0:00:06 (0.7091 s / it) * Acc@1 83.316 Acc@5 96.604 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Epoch: [371] [ 0/312] eta: 0:58:08 lr: 0.000325 min_lr: 0.000325 loss: 1.9871 (1.9871) weight_decay: 0.0500 (0.0500) time: 11.1809 data: 7.4321 max mem: 64948 Epoch: [371] [ 10/312] eta: 0:08:25 lr: 0.000325 min_lr: 0.000325 loss: 1.8445 (1.7515) weight_decay: 0.0500 (0.0500) time: 1.6754 data: 0.6761 max mem: 64948 Epoch: [371] [ 20/312] eta: 0:05:52 lr: 0.000324 min_lr: 0.000324 loss: 1.8533 (1.7938) weight_decay: 0.0500 (0.0500) time: 0.7099 data: 0.0004 max mem: 64948 Epoch: [371] [ 30/312] eta: 0:04:54 lr: 0.000324 min_lr: 0.000324 loss: 1.8414 (1.7492) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [371] [ 40/312] eta: 0:04:20 lr: 0.000324 min_lr: 0.000324 loss: 1.6209 (1.7106) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [371] [ 50/312] eta: 0:03:57 lr: 0.000324 min_lr: 0.000324 loss: 1.6880 (1.7507) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [371] [ 60/312] eta: 0:03:39 lr: 0.000323 min_lr: 0.000323 loss: 1.9141 (1.7571) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [371] [ 70/312] eta: 0:03:25 lr: 0.000323 min_lr: 0.000323 loss: 1.7971 (1.7388) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [371] [ 80/312] eta: 0:03:12 lr: 0.000323 min_lr: 0.000323 loss: 1.7971 (1.7543) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [371] [ 90/312] eta: 0:03:00 lr: 0.000323 min_lr: 0.000323 loss: 1.8796 (1.7639) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [371] [100/312] eta: 0:02:50 lr: 0.000322 min_lr: 0.000322 loss: 1.7827 (1.7554) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [371] [110/312] eta: 0:02:40 lr: 0.000322 min_lr: 0.000322 loss: 1.7479 (1.7512) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [371] [120/312] eta: 0:02:30 lr: 0.000322 min_lr: 0.000322 loss: 1.6084 (1.7404) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [371] [130/312] eta: 0:02:21 lr: 0.000322 min_lr: 0.000322 loss: 1.7788 (1.7433) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [371] [140/312] eta: 0:02:12 lr: 0.000321 min_lr: 0.000321 loss: 1.8326 (1.7416) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [371] [150/312] eta: 0:02:04 lr: 0.000321 min_lr: 0.000321 loss: 1.8326 (1.7430) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [371] [160/312] eta: 0:01:55 lr: 0.000321 min_lr: 0.000321 loss: 1.8824 (1.7498) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [371] [170/312] eta: 0:01:47 lr: 0.000321 min_lr: 0.000321 loss: 1.8481 (1.7446) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [371] [180/312] eta: 0:01:39 lr: 0.000320 min_lr: 0.000320 loss: 1.7432 (1.7408) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [371] [190/312] eta: 0:01:31 lr: 0.000320 min_lr: 0.000320 loss: 1.7530 (1.7386) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [371] [200/312] eta: 0:01:23 lr: 0.000320 min_lr: 0.000320 loss: 1.7527 (1.7377) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [371] [210/312] eta: 0:01:16 lr: 0.000320 min_lr: 0.000320 loss: 1.7527 (1.7389) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [371] [220/312] eta: 0:01:08 lr: 0.000319 min_lr: 0.000319 loss: 1.7736 (1.7368) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [371] [230/312] eta: 0:01:00 lr: 0.000319 min_lr: 0.000319 loss: 1.8306 (1.7436) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [371] [240/312] eta: 0:00:53 lr: 0.000319 min_lr: 0.000319 loss: 1.8700 (1.7430) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [371] [250/312] eta: 0:00:45 lr: 0.000319 min_lr: 0.000319 loss: 1.7902 (1.7420) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [371] [260/312] eta: 0:00:38 lr: 0.000318 min_lr: 0.000318 loss: 1.7852 (1.7434) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [371] [270/312] eta: 0:00:30 lr: 0.000318 min_lr: 0.000318 loss: 1.7588 (1.7430) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [371] [280/312] eta: 0:00:23 lr: 0.000318 min_lr: 0.000318 loss: 1.7800 (1.7428) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [371] [290/312] eta: 0:00:16 lr: 0.000318 min_lr: 0.000318 loss: 1.7800 (1.7435) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [371] [300/312] eta: 0:00:08 lr: 0.000317 min_lr: 0.000317 loss: 1.7381 (1.7387) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [371] [310/312] eta: 0:00:01 lr: 0.000317 min_lr: 0.000317 loss: 1.7464 (1.7366) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [371] [311/312] eta: 0:00:00 lr: 0.000317 min_lr: 0.000317 loss: 1.7102 (1.7359) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [371] Total time: 0:03:48 (0.7325 s / it) Averaged stats: lr: 0.000317 min_lr: 0.000317 loss: 1.7102 (1.7357) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4531 (0.4531) acc1: 88.5417 (88.5417) acc5: 98.4375 (98.4375) time: 4.7634 data: 4.5574 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6500 (0.6654) acc1: 82.2917 (82.2720) acc5: 97.1354 (96.7360) time: 0.6806 data: 0.5065 max mem: 64948 Test: Total time: 0:00:06 (0.7057 s / it) * Acc@1 83.086 Acc@5 96.444 loss 0.644 Accuracy of the model on the 50000 test images: 83.1% Max accuracy: 83.09% Test: [0/9] eta: 0:00:41 loss: 0.4581 (0.4581) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.6233 data: 4.4011 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6264 (0.6465) acc1: 84.6354 (82.2080) acc5: 97.1354 (96.6400) time: 0.6655 data: 0.4891 max mem: 64948 Test: Total time: 0:00:06 (0.6745 s / it) * Acc@1 83.328 Acc@5 96.610 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.33% Epoch: [372] [ 0/312] eta: 0:53:01 lr: 0.000317 min_lr: 0.000317 loss: 1.9412 (1.9412) weight_decay: 0.0500 (0.0500) time: 10.1956 data: 9.4051 max mem: 64948 Epoch: [372] [ 10/312] eta: 0:07:56 lr: 0.000317 min_lr: 0.000317 loss: 1.8320 (1.7989) weight_decay: 0.0500 (0.0500) time: 1.5768 data: 0.8553 max mem: 64948 Epoch: [372] [ 20/312] eta: 0:05:37 lr: 0.000316 min_lr: 0.000316 loss: 1.8141 (1.7701) weight_decay: 0.0500 (0.0500) time: 0.7053 data: 0.0004 max mem: 64948 Epoch: [372] [ 30/312] eta: 0:04:44 lr: 0.000316 min_lr: 0.000316 loss: 1.7677 (1.7502) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [372] [ 40/312] eta: 0:04:13 lr: 0.000316 min_lr: 0.000316 loss: 1.7265 (1.7408) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [372] [ 50/312] eta: 0:03:52 lr: 0.000316 min_lr: 0.000316 loss: 1.8260 (1.7402) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [372] [ 60/312] eta: 0:03:35 lr: 0.000315 min_lr: 0.000315 loss: 1.8304 (1.7496) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [372] [ 70/312] eta: 0:03:21 lr: 0.000315 min_lr: 0.000315 loss: 1.8617 (1.7451) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [372] [ 80/312] eta: 0:03:09 lr: 0.000315 min_lr: 0.000315 loss: 1.7436 (1.7280) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [372] [ 90/312] eta: 0:02:58 lr: 0.000315 min_lr: 0.000315 loss: 1.7069 (1.7229) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [372] [100/312] eta: 0:02:47 lr: 0.000314 min_lr: 0.000314 loss: 1.6809 (1.7277) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [372] [110/312] eta: 0:02:38 lr: 0.000314 min_lr: 0.000314 loss: 1.8520 (1.7360) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [372] [120/312] eta: 0:02:29 lr: 0.000314 min_lr: 0.000314 loss: 1.8506 (1.7339) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [372] [130/312] eta: 0:02:20 lr: 0.000314 min_lr: 0.000314 loss: 1.7363 (1.7344) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [372] [140/312] eta: 0:02:11 lr: 0.000313 min_lr: 0.000313 loss: 1.7745 (1.7360) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [372] [150/312] eta: 0:02:03 lr: 0.000313 min_lr: 0.000313 loss: 1.6438 (1.7317) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [372] [160/312] eta: 0:01:54 lr: 0.000313 min_lr: 0.000313 loss: 1.7727 (1.7352) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [372] [170/312] eta: 0:01:46 lr: 0.000313 min_lr: 0.000313 loss: 1.8585 (1.7426) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [372] [180/312] eta: 0:01:38 lr: 0.000312 min_lr: 0.000312 loss: 1.8325 (1.7385) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [372] [190/312] eta: 0:01:31 lr: 0.000312 min_lr: 0.000312 loss: 1.7886 (1.7468) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [372] [200/312] eta: 0:01:23 lr: 0.000312 min_lr: 0.000312 loss: 1.8378 (1.7483) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [372] [210/312] eta: 0:01:15 lr: 0.000312 min_lr: 0.000312 loss: 1.7364 (1.7413) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [372] [220/312] eta: 0:01:08 lr: 0.000311 min_lr: 0.000311 loss: 1.6133 (1.7351) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [372] [230/312] eta: 0:01:00 lr: 0.000311 min_lr: 0.000311 loss: 1.7797 (1.7383) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [372] [240/312] eta: 0:00:52 lr: 0.000311 min_lr: 0.000311 loss: 1.8871 (1.7431) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [372] [250/312] eta: 0:00:45 lr: 0.000311 min_lr: 0.000311 loss: 1.8136 (1.7368) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [372] [260/312] eta: 0:00:38 lr: 0.000310 min_lr: 0.000310 loss: 1.7251 (1.7421) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [372] [270/312] eta: 0:00:30 lr: 0.000310 min_lr: 0.000310 loss: 1.8484 (1.7420) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [372] [280/312] eta: 0:00:23 lr: 0.000310 min_lr: 0.000310 loss: 1.8070 (1.7427) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [372] [290/312] eta: 0:00:16 lr: 0.000310 min_lr: 0.000310 loss: 1.8300 (1.7429) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [372] [300/312] eta: 0:00:08 lr: 0.000309 min_lr: 0.000309 loss: 1.8418 (1.7440) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [372] [310/312] eta: 0:00:01 lr: 0.000309 min_lr: 0.000309 loss: 1.8451 (1.7441) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [372] [311/312] eta: 0:00:00 lr: 0.000309 min_lr: 0.000309 loss: 1.8416 (1.7432) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [372] Total time: 0:03:47 (0.7292 s / it) Averaged stats: lr: 0.000309 min_lr: 0.000309 loss: 1.8416 (1.7373) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4645 (0.4645) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 4.5933 data: 4.3730 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6421 (0.6725) acc1: 83.0729 (82.1440) acc5: 97.1354 (96.5120) time: 0.6623 data: 0.4860 max mem: 64948 Test: Total time: 0:00:06 (0.6859 s / it) * Acc@1 83.246 Acc@5 96.464 loss 0.646 Accuracy of the model on the 50000 test images: 83.2% Max accuracy: 83.25% Test: [0/9] eta: 0:00:41 loss: 0.4577 (0.4577) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.6455 data: 4.4313 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6269 (0.6465) acc1: 84.6354 (82.2080) acc5: 97.1354 (96.6400) time: 0.6675 data: 0.4925 max mem: 64948 Test: Total time: 0:00:06 (0.6747 s / it) * Acc@1 83.338 Acc@5 96.620 loss 0.624 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.34% Epoch: [373] [ 0/312] eta: 0:52:49 lr: 0.000309 min_lr: 0.000309 loss: 2.1809 (2.1809) weight_decay: 0.0500 (0.0500) time: 10.1572 data: 9.0065 max mem: 64948 Epoch: [373] [ 10/312] eta: 0:07:53 lr: 0.000309 min_lr: 0.000309 loss: 1.8969 (1.7832) weight_decay: 0.0500 (0.0500) time: 1.5683 data: 0.8191 max mem: 64948 Epoch: [373] [ 20/312] eta: 0:05:36 lr: 0.000309 min_lr: 0.000309 loss: 1.7270 (1.6913) weight_decay: 0.0500 (0.0500) time: 0.7022 data: 0.0004 max mem: 64948 Epoch: [373] [ 30/312] eta: 0:04:43 lr: 0.000308 min_lr: 0.000308 loss: 1.7607 (1.7587) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [373] [ 40/312] eta: 0:04:12 lr: 0.000308 min_lr: 0.000308 loss: 1.8774 (1.7748) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [373] [ 50/312] eta: 0:03:51 lr: 0.000308 min_lr: 0.000308 loss: 1.8671 (1.7697) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [373] [ 60/312] eta: 0:03:34 lr: 0.000308 min_lr: 0.000308 loss: 2.0132 (1.7934) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [373] [ 70/312] eta: 0:03:20 lr: 0.000307 min_lr: 0.000307 loss: 2.0132 (1.8109) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [373] [ 80/312] eta: 0:03:08 lr: 0.000307 min_lr: 0.000307 loss: 1.9358 (1.7979) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [373] [ 90/312] eta: 0:02:57 lr: 0.000307 min_lr: 0.000307 loss: 1.8233 (1.7946) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [373] [100/312] eta: 0:02:47 lr: 0.000307 min_lr: 0.000307 loss: 1.8192 (1.7910) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [373] [110/312] eta: 0:02:37 lr: 0.000306 min_lr: 0.000306 loss: 1.7091 (1.7839) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [373] [120/312] eta: 0:02:28 lr: 0.000306 min_lr: 0.000306 loss: 1.8335 (1.7885) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [373] [130/312] eta: 0:02:19 lr: 0.000306 min_lr: 0.000306 loss: 1.8148 (1.7799) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [373] [140/312] eta: 0:02:11 lr: 0.000306 min_lr: 0.000306 loss: 1.8824 (1.7824) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [373] [150/312] eta: 0:02:02 lr: 0.000305 min_lr: 0.000305 loss: 1.9580 (1.7894) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [373] [160/312] eta: 0:01:54 lr: 0.000305 min_lr: 0.000305 loss: 1.9580 (1.7846) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [373] [170/312] eta: 0:01:46 lr: 0.000305 min_lr: 0.000305 loss: 1.7276 (1.7789) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [373] [180/312] eta: 0:01:38 lr: 0.000305 min_lr: 0.000305 loss: 1.7662 (1.7761) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [373] [190/312] eta: 0:01:30 lr: 0.000304 min_lr: 0.000304 loss: 1.7662 (1.7686) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [373] [200/312] eta: 0:01:23 lr: 0.000304 min_lr: 0.000304 loss: 1.7798 (1.7661) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [373] [210/312] eta: 0:01:15 lr: 0.000304 min_lr: 0.000304 loss: 1.7644 (1.7603) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [373] [220/312] eta: 0:01:07 lr: 0.000304 min_lr: 0.000304 loss: 1.8248 (1.7673) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [373] [230/312] eta: 0:01:00 lr: 0.000303 min_lr: 0.000303 loss: 1.8764 (1.7676) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [373] [240/312] eta: 0:00:52 lr: 0.000303 min_lr: 0.000303 loss: 1.8764 (1.7757) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [373] [250/312] eta: 0:00:45 lr: 0.000303 min_lr: 0.000303 loss: 1.8396 (1.7747) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [373] [260/312] eta: 0:00:38 lr: 0.000303 min_lr: 0.000303 loss: 1.8150 (1.7747) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [373] [270/312] eta: 0:00:30 lr: 0.000302 min_lr: 0.000302 loss: 1.8340 (1.7766) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [373] [280/312] eta: 0:00:23 lr: 0.000302 min_lr: 0.000302 loss: 1.9452 (1.7832) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [373] [290/312] eta: 0:00:16 lr: 0.000302 min_lr: 0.000302 loss: 1.9137 (1.7761) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [373] [300/312] eta: 0:00:08 lr: 0.000302 min_lr: 0.000302 loss: 1.6608 (1.7725) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [373] [310/312] eta: 0:00:01 lr: 0.000301 min_lr: 0.000301 loss: 1.6370 (1.7686) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [373] [311/312] eta: 0:00:00 lr: 0.000301 min_lr: 0.000301 loss: 1.6472 (1.7682) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [373] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.000301 min_lr: 0.000301 loss: 1.6472 (1.7427) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4693 (0.4693) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 4.5860 data: 4.3693 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6420 (0.6621) acc1: 83.0729 (82.3040) acc5: 97.1354 (96.6400) time: 0.6608 data: 0.4856 max mem: 64948 Test: Total time: 0:00:06 (0.6834 s / it) * Acc@1 83.234 Acc@5 96.502 loss 0.645 Accuracy of the model on the 50000 test images: 83.2% Max accuracy: 83.25% Test: [0/9] eta: 0:00:46 loss: 0.4575 (0.4575) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.1790 data: 4.9744 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6269 (0.6463) acc1: 84.6354 (82.2080) acc5: 97.1354 (96.6400) time: 0.7271 data: 0.5528 max mem: 64948 Test: Total time: 0:00:06 (0.7366 s / it) * Acc@1 83.344 Acc@5 96.612 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.3% Max EMA accuracy: 83.34% Epoch: [374] [ 0/312] eta: 0:45:58 lr: 0.000301 min_lr: 0.000301 loss: 1.2818 (1.2818) weight_decay: 0.0500 (0.0500) time: 8.8411 data: 7.9977 max mem: 64948 Epoch: [374] [ 10/312] eta: 0:07:24 lr: 0.000301 min_lr: 0.000301 loss: 2.0259 (1.8878) weight_decay: 0.0500 (0.0500) time: 1.4718 data: 0.7274 max mem: 64948 Epoch: [374] [ 20/312] eta: 0:05:21 lr: 0.000301 min_lr: 0.000301 loss: 1.9516 (1.9157) weight_decay: 0.0500 (0.0500) time: 0.7144 data: 0.0004 max mem: 64948 Epoch: [374] [ 30/312] eta: 0:04:33 lr: 0.000301 min_lr: 0.000301 loss: 1.8138 (1.7750) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [374] [ 40/312] eta: 0:04:05 lr: 0.000300 min_lr: 0.000300 loss: 1.5507 (1.7477) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [374] [ 50/312] eta: 0:03:45 lr: 0.000300 min_lr: 0.000300 loss: 1.7063 (1.7385) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [374] [ 60/312] eta: 0:03:30 lr: 0.000300 min_lr: 0.000300 loss: 1.8284 (1.7525) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [374] [ 70/312] eta: 0:03:17 lr: 0.000300 min_lr: 0.000300 loss: 1.8116 (1.7491) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [374] [ 80/312] eta: 0:03:05 lr: 0.000299 min_lr: 0.000299 loss: 1.8116 (1.7531) weight_decay: 0.0500 (0.0500) time: 0.7006 data: 0.0004 max mem: 64948 Epoch: [374] [ 90/312] eta: 0:02:55 lr: 0.000299 min_lr: 0.000299 loss: 1.8184 (1.7539) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [374] [100/312] eta: 0:02:45 lr: 0.000299 min_lr: 0.000299 loss: 1.6977 (1.7478) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [374] [110/312] eta: 0:02:36 lr: 0.000299 min_lr: 0.000299 loss: 1.8584 (1.7619) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [374] [120/312] eta: 0:02:27 lr: 0.000298 min_lr: 0.000298 loss: 1.9000 (1.7617) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [374] [130/312] eta: 0:02:18 lr: 0.000298 min_lr: 0.000298 loss: 1.8894 (1.7669) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [374] [140/312] eta: 0:02:10 lr: 0.000298 min_lr: 0.000298 loss: 1.8894 (1.7642) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [374] [150/312] eta: 0:02:01 lr: 0.000298 min_lr: 0.000298 loss: 1.7635 (1.7637) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [374] [160/312] eta: 0:01:53 lr: 0.000297 min_lr: 0.000297 loss: 1.7309 (1.7605) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [374] [170/312] eta: 0:01:45 lr: 0.000297 min_lr: 0.000297 loss: 1.6632 (1.7554) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [374] [180/312] eta: 0:01:38 lr: 0.000297 min_lr: 0.000297 loss: 1.6632 (1.7565) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [374] [190/312] eta: 0:01:30 lr: 0.000297 min_lr: 0.000297 loss: 1.6081 (1.7481) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [374] [200/312] eta: 0:01:22 lr: 0.000296 min_lr: 0.000296 loss: 1.6081 (1.7486) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [374] [210/312] eta: 0:01:15 lr: 0.000296 min_lr: 0.000296 loss: 1.7946 (1.7472) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [374] [220/312] eta: 0:01:07 lr: 0.000296 min_lr: 0.000296 loss: 1.7968 (1.7498) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [374] [230/312] eta: 0:01:00 lr: 0.000296 min_lr: 0.000296 loss: 1.7540 (1.7480) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [374] [240/312] eta: 0:00:52 lr: 0.000295 min_lr: 0.000295 loss: 1.7538 (1.7476) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [374] [250/312] eta: 0:00:45 lr: 0.000295 min_lr: 0.000295 loss: 1.8759 (1.7506) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [374] [260/312] eta: 0:00:37 lr: 0.000295 min_lr: 0.000295 loss: 1.8188 (1.7488) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [374] [270/312] eta: 0:00:30 lr: 0.000295 min_lr: 0.000295 loss: 1.7475 (1.7493) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [374] [280/312] eta: 0:00:23 lr: 0.000295 min_lr: 0.000295 loss: 1.7475 (1.7471) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [374] [290/312] eta: 0:00:15 lr: 0.000294 min_lr: 0.000294 loss: 1.8736 (1.7476) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [374] [300/312] eta: 0:00:08 lr: 0.000294 min_lr: 0.000294 loss: 1.8441 (1.7497) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [374] [310/312] eta: 0:00:01 lr: 0.000294 min_lr: 0.000294 loss: 1.8441 (1.7509) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [374] [311/312] eta: 0:00:00 lr: 0.000294 min_lr: 0.000294 loss: 1.8424 (1.7501) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [374] Total time: 0:03:46 (0.7253 s / it) Averaged stats: lr: 0.000294 min_lr: 0.000294 loss: 1.8424 (1.7367) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4817 (0.4817) acc1: 86.7188 (86.7188) acc5: 98.1771 (98.1771) time: 4.3814 data: 4.1748 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6392 (0.6703) acc1: 84.1146 (81.8560) acc5: 97.1354 (96.6720) time: 0.6380 data: 0.4639 max mem: 64948 Test: Total time: 0:00:05 (0.6635 s / it) * Acc@1 83.028 Acc@5 96.428 loss 0.650 Accuracy of the model on the 50000 test images: 83.0% Max accuracy: 83.25% Test: [0/9] eta: 0:00:45 loss: 0.4575 (0.4575) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.0418 data: 4.8287 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6269 (0.6461) acc1: 84.6354 (82.1760) acc5: 97.1354 (96.6400) time: 0.7115 data: 0.5366 max mem: 64948 Test: Total time: 0:00:06 (0.7194 s / it) * Acc@1 83.370 Acc@5 96.628 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.4% Max EMA accuracy: 83.37% Epoch: [375] [ 0/312] eta: 0:44:21 lr: 0.000294 min_lr: 0.000294 loss: 1.7423 (1.7423) weight_decay: 0.0500 (0.0500) time: 8.5308 data: 7.7325 max mem: 64948 Epoch: [375] [ 10/312] eta: 0:07:22 lr: 0.000293 min_lr: 0.000293 loss: 1.6783 (1.6510) weight_decay: 0.0500 (0.0500) time: 1.4666 data: 0.7033 max mem: 64948 Epoch: [375] [ 20/312] eta: 0:05:21 lr: 0.000293 min_lr: 0.000293 loss: 1.7360 (1.7298) weight_decay: 0.0500 (0.0500) time: 0.7287 data: 0.0004 max mem: 64948 Epoch: [375] [ 30/312] eta: 0:04:33 lr: 0.000293 min_lr: 0.000293 loss: 1.8876 (1.7692) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [375] [ 40/312] eta: 0:04:05 lr: 0.000293 min_lr: 0.000293 loss: 1.8780 (1.7516) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [375] [ 50/312] eta: 0:03:45 lr: 0.000293 min_lr: 0.000293 loss: 1.7179 (1.7465) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [375] [ 60/312] eta: 0:03:30 lr: 0.000292 min_lr: 0.000292 loss: 1.7179 (1.7366) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [375] [ 70/312] eta: 0:03:17 lr: 0.000292 min_lr: 0.000292 loss: 1.7275 (1.7499) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [375] [ 80/312] eta: 0:03:05 lr: 0.000292 min_lr: 0.000292 loss: 1.8463 (1.7426) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [375] [ 90/312] eta: 0:02:55 lr: 0.000292 min_lr: 0.000292 loss: 1.8293 (1.7428) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [375] [100/312] eta: 0:02:45 lr: 0.000291 min_lr: 0.000291 loss: 1.9114 (1.7591) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [375] [110/312] eta: 0:02:35 lr: 0.000291 min_lr: 0.000291 loss: 1.9395 (1.7686) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [375] [120/312] eta: 0:02:27 lr: 0.000291 min_lr: 0.000291 loss: 1.8922 (1.7658) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [375] [130/312] eta: 0:02:18 lr: 0.000291 min_lr: 0.000291 loss: 1.7858 (1.7564) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [375] [140/312] eta: 0:02:09 lr: 0.000290 min_lr: 0.000290 loss: 1.8712 (1.7568) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [375] [150/312] eta: 0:02:01 lr: 0.000290 min_lr: 0.000290 loss: 1.8590 (1.7593) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [375] [160/312] eta: 0:01:53 lr: 0.000290 min_lr: 0.000290 loss: 1.7250 (1.7446) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [375] [170/312] eta: 0:01:45 lr: 0.000290 min_lr: 0.000290 loss: 1.7310 (1.7462) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [375] [180/312] eta: 0:01:37 lr: 0.000289 min_lr: 0.000289 loss: 1.8572 (1.7497) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [375] [190/312] eta: 0:01:30 lr: 0.000289 min_lr: 0.000289 loss: 1.7273 (1.7482) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [375] [200/312] eta: 0:01:22 lr: 0.000289 min_lr: 0.000289 loss: 1.8087 (1.7521) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [375] [210/312] eta: 0:01:14 lr: 0.000289 min_lr: 0.000289 loss: 1.8380 (1.7563) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [375] [220/312] eta: 0:01:07 lr: 0.000288 min_lr: 0.000288 loss: 1.7396 (1.7524) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [375] [230/312] eta: 0:00:59 lr: 0.000288 min_lr: 0.000288 loss: 1.7432 (1.7533) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [375] [240/312] eta: 0:00:52 lr: 0.000288 min_lr: 0.000288 loss: 1.8194 (1.7551) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [375] [250/312] eta: 0:00:45 lr: 0.000288 min_lr: 0.000288 loss: 1.7701 (1.7481) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [375] [260/312] eta: 0:00:37 lr: 0.000287 min_lr: 0.000287 loss: 1.5969 (1.7450) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [375] [270/312] eta: 0:00:30 lr: 0.000287 min_lr: 0.000287 loss: 1.6606 (1.7428) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [375] [280/312] eta: 0:00:23 lr: 0.000287 min_lr: 0.000287 loss: 1.7002 (1.7418) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [375] [290/312] eta: 0:00:15 lr: 0.000287 min_lr: 0.000287 loss: 1.7602 (1.7443) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [375] [300/312] eta: 0:00:08 lr: 0.000286 min_lr: 0.000286 loss: 1.6981 (1.7424) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [375] [310/312] eta: 0:00:01 lr: 0.000286 min_lr: 0.000286 loss: 1.6607 (1.7422) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [375] [311/312] eta: 0:00:00 lr: 0.000286 min_lr: 0.000286 loss: 1.6607 (1.7427) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [375] Total time: 0:03:46 (0.7248 s / it) Averaged stats: lr: 0.000286 min_lr: 0.000286 loss: 1.6607 (1.7311) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4311 (0.4311) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 4.5634 data: 4.3438 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6456 (0.6569) acc1: 83.0729 (82.3680) acc5: 97.1354 (96.6400) time: 0.6589 data: 0.4827 max mem: 64948 Test: Total time: 0:00:06 (0.6838 s / it) * Acc@1 83.126 Acc@5 96.478 loss 0.641 Accuracy of the model on the 50000 test images: 83.1% Max accuracy: 83.25% Test: [0/9] eta: 0:00:44 loss: 0.4574 (0.4574) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.9844 data: 4.7660 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6268 (0.6460) acc1: 84.6354 (82.2400) acc5: 97.1354 (96.6080) time: 0.7053 data: 0.5297 max mem: 64948 Test: Total time: 0:00:06 (0.7143 s / it) * Acc@1 83.390 Acc@5 96.624 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.4% Max EMA accuracy: 83.39% Epoch: [376] [ 0/312] eta: 0:52:37 lr: 0.000286 min_lr: 0.000286 loss: 1.0090 (1.0090) weight_decay: 0.0500 (0.0500) time: 10.1204 data: 9.3179 max mem: 64948 Epoch: [376] [ 10/312] eta: 0:07:52 lr: 0.000286 min_lr: 0.000286 loss: 1.7988 (1.7410) weight_decay: 0.0500 (0.0500) time: 1.5630 data: 0.8474 max mem: 64948 Epoch: [376] [ 20/312] eta: 0:05:35 lr: 0.000286 min_lr: 0.000286 loss: 1.7988 (1.7599) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [376] [ 30/312] eta: 0:04:42 lr: 0.000285 min_lr: 0.000285 loss: 1.6577 (1.7033) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [376] [ 40/312] eta: 0:04:12 lr: 0.000285 min_lr: 0.000285 loss: 1.5678 (1.6855) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [376] [ 50/312] eta: 0:03:51 lr: 0.000285 min_lr: 0.000285 loss: 1.6645 (1.6827) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [376] [ 60/312] eta: 0:03:34 lr: 0.000285 min_lr: 0.000285 loss: 1.5777 (1.6610) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [376] [ 70/312] eta: 0:03:20 lr: 0.000284 min_lr: 0.000284 loss: 1.8030 (1.7075) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [376] [ 80/312] eta: 0:03:08 lr: 0.000284 min_lr: 0.000284 loss: 1.9171 (1.7075) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [376] [ 90/312] eta: 0:02:57 lr: 0.000284 min_lr: 0.000284 loss: 1.7959 (1.7191) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [376] [100/312] eta: 0:02:47 lr: 0.000284 min_lr: 0.000284 loss: 1.8932 (1.7191) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [376] [110/312] eta: 0:02:37 lr: 0.000284 min_lr: 0.000284 loss: 1.8943 (1.7203) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [376] [120/312] eta: 0:02:28 lr: 0.000283 min_lr: 0.000283 loss: 1.8565 (1.7183) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [376] [130/312] eta: 0:02:19 lr: 0.000283 min_lr: 0.000283 loss: 1.8565 (1.7272) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [376] [140/312] eta: 0:02:11 lr: 0.000283 min_lr: 0.000283 loss: 1.7651 (1.7259) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [376] [150/312] eta: 0:02:02 lr: 0.000283 min_lr: 0.000283 loss: 1.6518 (1.7131) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [376] [160/312] eta: 0:01:54 lr: 0.000282 min_lr: 0.000282 loss: 1.7505 (1.7193) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [376] [170/312] eta: 0:01:46 lr: 0.000282 min_lr: 0.000282 loss: 1.7611 (1.7118) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [376] [180/312] eta: 0:01:38 lr: 0.000282 min_lr: 0.000282 loss: 1.7306 (1.7171) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [376] [190/312] eta: 0:01:30 lr: 0.000282 min_lr: 0.000282 loss: 1.7724 (1.7200) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [376] [200/312] eta: 0:01:23 lr: 0.000281 min_lr: 0.000281 loss: 1.8871 (1.7318) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [376] [210/312] eta: 0:01:15 lr: 0.000281 min_lr: 0.000281 loss: 1.9309 (1.7365) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [376] [220/312] eta: 0:01:07 lr: 0.000281 min_lr: 0.000281 loss: 1.8438 (1.7340) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [376] [230/312] eta: 0:01:00 lr: 0.000281 min_lr: 0.000281 loss: 1.6287 (1.7297) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [376] [240/312] eta: 0:00:52 lr: 0.000280 min_lr: 0.000280 loss: 1.6360 (1.7302) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [376] [250/312] eta: 0:00:45 lr: 0.000280 min_lr: 0.000280 loss: 1.8225 (1.7364) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [376] [260/312] eta: 0:00:38 lr: 0.000280 min_lr: 0.000280 loss: 1.8952 (1.7362) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [376] [270/312] eta: 0:00:30 lr: 0.000280 min_lr: 0.000280 loss: 1.8348 (1.7392) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [376] [280/312] eta: 0:00:23 lr: 0.000279 min_lr: 0.000279 loss: 1.8245 (1.7385) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0009 max mem: 64948 Epoch: [376] [290/312] eta: 0:00:16 lr: 0.000279 min_lr: 0.000279 loss: 1.7569 (1.7404) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [376] [300/312] eta: 0:00:08 lr: 0.000279 min_lr: 0.000279 loss: 1.7260 (1.7350) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [376] [310/312] eta: 0:00:01 lr: 0.000279 min_lr: 0.000279 loss: 1.7447 (1.7371) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [376] [311/312] eta: 0:00:00 lr: 0.000279 min_lr: 0.000279 loss: 1.6965 (1.7356) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [376] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.000279 min_lr: 0.000279 loss: 1.6965 (1.7278) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4669 (0.4669) acc1: 86.9792 (86.9792) acc5: 98.4375 (98.4375) time: 4.3608 data: 4.1410 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6442 (0.6617) acc1: 84.3750 (82.0480) acc5: 97.1354 (96.7040) time: 0.6361 data: 0.4602 max mem: 64948 Test: Total time: 0:00:06 (0.6687 s / it) * Acc@1 83.210 Acc@5 96.444 loss 0.645 Accuracy of the model on the 50000 test images: 83.2% Max accuracy: 83.25% Test: [0/9] eta: 0:00:42 loss: 0.4574 (0.4574) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.7701 data: 4.5531 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6268 (0.6459) acc1: 84.6354 (82.2080) acc5: 97.1354 (96.6080) time: 0.6955 data: 0.5202 max mem: 64948 Test: Total time: 0:00:06 (0.7036 s / it) * Acc@1 83.408 Acc@5 96.616 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.4% Max EMA accuracy: 83.41% Epoch: [377] [ 0/312] eta: 0:48:18 lr: 0.000279 min_lr: 0.000279 loss: 2.0323 (2.0323) weight_decay: 0.0500 (0.0500) time: 9.2909 data: 8.3235 max mem: 64948 Epoch: [377] [ 10/312] eta: 0:07:42 lr: 0.000278 min_lr: 0.000278 loss: 1.6927 (1.6777) weight_decay: 0.0500 (0.0500) time: 1.5311 data: 0.7571 max mem: 64948 Epoch: [377] [ 20/312] eta: 0:05:30 lr: 0.000278 min_lr: 0.000278 loss: 1.7808 (1.7148) weight_decay: 0.0500 (0.0500) time: 0.7251 data: 0.0004 max mem: 64948 Epoch: [377] [ 30/312] eta: 0:04:39 lr: 0.000278 min_lr: 0.000278 loss: 1.8018 (1.6928) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [377] [ 40/312] eta: 0:04:10 lr: 0.000278 min_lr: 0.000278 loss: 1.7786 (1.6979) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [377] [ 50/312] eta: 0:03:49 lr: 0.000278 min_lr: 0.000278 loss: 1.7786 (1.7085) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [377] [ 60/312] eta: 0:03:33 lr: 0.000277 min_lr: 0.000277 loss: 1.7361 (1.7117) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [377] [ 70/312] eta: 0:03:19 lr: 0.000277 min_lr: 0.000277 loss: 1.7508 (1.7175) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [377] [ 80/312] eta: 0:03:07 lr: 0.000277 min_lr: 0.000277 loss: 1.7508 (1.7111) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [377] [ 90/312] eta: 0:02:56 lr: 0.000277 min_lr: 0.000277 loss: 1.5995 (1.7033) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [377] [100/312] eta: 0:02:46 lr: 0.000276 min_lr: 0.000276 loss: 1.8803 (1.7275) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [377] [110/312] eta: 0:02:37 lr: 0.000276 min_lr: 0.000276 loss: 1.8843 (1.7322) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [377] [120/312] eta: 0:02:27 lr: 0.000276 min_lr: 0.000276 loss: 1.8448 (1.7410) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [377] [130/312] eta: 0:02:19 lr: 0.000276 min_lr: 0.000276 loss: 1.8064 (1.7420) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [377] [140/312] eta: 0:02:10 lr: 0.000275 min_lr: 0.000275 loss: 1.7680 (1.7449) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [377] [150/312] eta: 0:02:02 lr: 0.000275 min_lr: 0.000275 loss: 1.7544 (1.7381) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [377] [160/312] eta: 0:01:54 lr: 0.000275 min_lr: 0.000275 loss: 1.6826 (1.7431) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [377] [170/312] eta: 0:01:46 lr: 0.000275 min_lr: 0.000275 loss: 1.7452 (1.7362) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [377] [180/312] eta: 0:01:38 lr: 0.000274 min_lr: 0.000274 loss: 1.7863 (1.7400) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [377] [190/312] eta: 0:01:30 lr: 0.000274 min_lr: 0.000274 loss: 1.8181 (1.7463) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [377] [200/312] eta: 0:01:22 lr: 0.000274 min_lr: 0.000274 loss: 1.6923 (1.7321) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [377] [210/312] eta: 0:01:15 lr: 0.000274 min_lr: 0.000274 loss: 1.5771 (1.7264) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [377] [220/312] eta: 0:01:07 lr: 0.000273 min_lr: 0.000273 loss: 1.7805 (1.7284) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [377] [230/312] eta: 0:01:00 lr: 0.000273 min_lr: 0.000273 loss: 1.7819 (1.7280) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [377] [240/312] eta: 0:00:52 lr: 0.000273 min_lr: 0.000273 loss: 1.7373 (1.7183) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [377] [250/312] eta: 0:00:45 lr: 0.000273 min_lr: 0.000273 loss: 1.8066 (1.7265) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [377] [260/312] eta: 0:00:37 lr: 0.000273 min_lr: 0.000273 loss: 1.8867 (1.7250) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [377] [270/312] eta: 0:00:30 lr: 0.000272 min_lr: 0.000272 loss: 1.8293 (1.7291) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [377] [280/312] eta: 0:00:23 lr: 0.000272 min_lr: 0.000272 loss: 1.8546 (1.7323) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [377] [290/312] eta: 0:00:15 lr: 0.000272 min_lr: 0.000272 loss: 1.8161 (1.7331) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [377] [300/312] eta: 0:00:08 lr: 0.000272 min_lr: 0.000272 loss: 1.8161 (1.7383) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [377] [310/312] eta: 0:00:01 lr: 0.000271 min_lr: 0.000271 loss: 1.8627 (1.7396) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [377] [311/312] eta: 0:00:00 lr: 0.000271 min_lr: 0.000271 loss: 1.8627 (1.7378) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [377] Total time: 0:03:46 (0.7266 s / it) Averaged stats: lr: 0.000271 min_lr: 0.000271 loss: 1.8627 (1.7318) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4419 (0.4419) acc1: 87.7604 (87.7604) acc5: 98.4375 (98.4375) time: 4.5329 data: 4.3132 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6382 (0.6538) acc1: 84.3750 (82.5600) acc5: 97.1354 (96.7680) time: 0.6550 data: 0.4793 max mem: 64948 Test: Total time: 0:00:06 (0.6778 s / it) * Acc@1 83.072 Acc@5 96.402 loss 0.646 Accuracy of the model on the 50000 test images: 83.1% Max accuracy: 83.25% Test: [0/9] eta: 0:00:42 loss: 0.4572 (0.4572) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.7734 data: 4.5568 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6260 (0.6457) acc1: 84.6354 (82.3040) acc5: 97.1354 (96.6080) time: 0.6817 data: 0.5064 max mem: 64948 Test: Total time: 0:00:06 (0.6893 s / it) * Acc@1 83.438 Acc@5 96.628 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.4% Max EMA accuracy: 83.44% Epoch: [378] [ 0/312] eta: 0:53:20 lr: 0.000271 min_lr: 0.000271 loss: 1.9445 (1.9445) weight_decay: 0.0500 (0.0500) time: 10.2572 data: 9.4776 max mem: 64948 Epoch: [378] [ 10/312] eta: 0:07:56 lr: 0.000271 min_lr: 0.000271 loss: 1.8704 (1.8455) weight_decay: 0.0500 (0.0500) time: 1.5770 data: 0.8620 max mem: 64948 Epoch: [378] [ 20/312] eta: 0:05:37 lr: 0.000271 min_lr: 0.000271 loss: 1.8602 (1.7978) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0004 max mem: 64948 Epoch: [378] [ 30/312] eta: 0:04:43 lr: 0.000271 min_lr: 0.000271 loss: 1.8663 (1.8223) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [378] [ 40/312] eta: 0:04:13 lr: 0.000270 min_lr: 0.000270 loss: 1.8808 (1.8090) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [378] [ 50/312] eta: 0:03:51 lr: 0.000270 min_lr: 0.000270 loss: 1.8319 (1.8164) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [378] [ 60/312] eta: 0:03:35 lr: 0.000270 min_lr: 0.000270 loss: 1.8105 (1.8073) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [378] [ 70/312] eta: 0:03:21 lr: 0.000270 min_lr: 0.000270 loss: 1.6408 (1.7616) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [378] [ 80/312] eta: 0:03:09 lr: 0.000269 min_lr: 0.000269 loss: 1.5865 (1.7588) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [378] [ 90/312] eta: 0:02:58 lr: 0.000269 min_lr: 0.000269 loss: 1.7790 (1.7616) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [378] [100/312] eta: 0:02:47 lr: 0.000269 min_lr: 0.000269 loss: 1.8668 (1.7735) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [378] [110/312] eta: 0:02:38 lr: 0.000269 min_lr: 0.000269 loss: 1.8483 (1.7645) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [378] [120/312] eta: 0:02:28 lr: 0.000269 min_lr: 0.000269 loss: 1.6683 (1.7596) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [378] [130/312] eta: 0:02:20 lr: 0.000268 min_lr: 0.000268 loss: 1.6683 (1.7461) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [378] [140/312] eta: 0:02:11 lr: 0.000268 min_lr: 0.000268 loss: 1.5464 (1.7352) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [378] [150/312] eta: 0:02:03 lr: 0.000268 min_lr: 0.000268 loss: 1.5464 (1.7293) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [378] [160/312] eta: 0:01:54 lr: 0.000268 min_lr: 0.000268 loss: 1.8621 (1.7324) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [378] [170/312] eta: 0:01:46 lr: 0.000267 min_lr: 0.000267 loss: 1.9442 (1.7417) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [378] [180/312] eta: 0:01:38 lr: 0.000267 min_lr: 0.000267 loss: 1.8734 (1.7439) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [378] [190/312] eta: 0:01:30 lr: 0.000267 min_lr: 0.000267 loss: 1.7532 (1.7420) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [378] [200/312] eta: 0:01:23 lr: 0.000267 min_lr: 0.000267 loss: 1.8706 (1.7451) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [378] [210/312] eta: 0:01:15 lr: 0.000266 min_lr: 0.000266 loss: 1.7178 (1.7359) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [378] [220/312] eta: 0:01:07 lr: 0.000266 min_lr: 0.000266 loss: 1.6282 (1.7358) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [378] [230/312] eta: 0:01:00 lr: 0.000266 min_lr: 0.000266 loss: 1.6282 (1.7306) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [378] [240/312] eta: 0:00:52 lr: 0.000266 min_lr: 0.000266 loss: 1.7426 (1.7316) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [378] [250/312] eta: 0:00:45 lr: 0.000265 min_lr: 0.000265 loss: 1.7358 (1.7261) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [378] [260/312] eta: 0:00:38 lr: 0.000265 min_lr: 0.000265 loss: 1.7518 (1.7254) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [378] [270/312] eta: 0:00:30 lr: 0.000265 min_lr: 0.000265 loss: 1.8394 (1.7299) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [378] [280/312] eta: 0:00:23 lr: 0.000265 min_lr: 0.000265 loss: 1.8981 (1.7290) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0006 max mem: 64948 Epoch: [378] [290/312] eta: 0:00:16 lr: 0.000265 min_lr: 0.000265 loss: 1.7578 (1.7243) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0005 max mem: 64948 Epoch: [378] [300/312] eta: 0:00:08 lr: 0.000264 min_lr: 0.000264 loss: 1.6469 (1.7229) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [378] [310/312] eta: 0:00:01 lr: 0.000264 min_lr: 0.000264 loss: 1.8336 (1.7211) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [378] [311/312] eta: 0:00:00 lr: 0.000264 min_lr: 0.000264 loss: 1.8566 (1.7216) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [378] Total time: 0:03:47 (0.7284 s / it) Averaged stats: lr: 0.000264 min_lr: 0.000264 loss: 1.8566 (1.7223) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4561 (0.4561) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 4.6669 data: 4.4533 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6525 (0.6549) acc1: 84.1146 (82.3360) acc5: 96.6146 (96.6400) time: 0.6698 data: 0.4949 max mem: 64948 Test: Total time: 0:00:06 (0.6966 s / it) * Acc@1 83.212 Acc@5 96.464 loss 0.644 Accuracy of the model on the 50000 test images: 83.2% Max accuracy: 83.25% Test: [0/9] eta: 0:00:40 loss: 0.4570 (0.4570) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.5199 data: 4.3055 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6256 (0.6454) acc1: 84.6354 (82.3360) acc5: 97.1354 (96.6080) time: 0.6535 data: 0.4785 max mem: 64948 Test: Total time: 0:00:05 (0.6624 s / it) * Acc@1 83.450 Acc@5 96.618 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.45% Epoch: [379] [ 0/312] eta: 0:47:34 lr: 0.000264 min_lr: 0.000264 loss: 2.1159 (2.1159) weight_decay: 0.0500 (0.0500) time: 9.1503 data: 8.3266 max mem: 64948 Epoch: [379] [ 10/312] eta: 0:07:42 lr: 0.000264 min_lr: 0.000264 loss: 1.9277 (1.8205) weight_decay: 0.0500 (0.0500) time: 1.5320 data: 0.7573 max mem: 64948 Epoch: [379] [ 20/312] eta: 0:05:30 lr: 0.000264 min_lr: 0.000264 loss: 1.8996 (1.8334) weight_decay: 0.0500 (0.0500) time: 0.7321 data: 0.0004 max mem: 64948 Epoch: [379] [ 30/312] eta: 0:04:39 lr: 0.000263 min_lr: 0.000263 loss: 1.8953 (1.8128) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [379] [ 40/312] eta: 0:04:09 lr: 0.000263 min_lr: 0.000263 loss: 1.8850 (1.8468) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [379] [ 50/312] eta: 0:03:49 lr: 0.000263 min_lr: 0.000263 loss: 1.8787 (1.8307) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [379] [ 60/312] eta: 0:03:33 lr: 0.000263 min_lr: 0.000263 loss: 1.8877 (1.8272) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [379] [ 70/312] eta: 0:03:19 lr: 0.000262 min_lr: 0.000262 loss: 1.9003 (1.8388) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [379] [ 80/312] eta: 0:03:07 lr: 0.000262 min_lr: 0.000262 loss: 1.8832 (1.8240) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [379] [ 90/312] eta: 0:02:56 lr: 0.000262 min_lr: 0.000262 loss: 1.7383 (1.8116) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [379] [100/312] eta: 0:02:46 lr: 0.000262 min_lr: 0.000262 loss: 1.8702 (1.8174) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [379] [110/312] eta: 0:02:37 lr: 0.000261 min_lr: 0.000261 loss: 1.8902 (1.8077) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [379] [120/312] eta: 0:02:28 lr: 0.000261 min_lr: 0.000261 loss: 1.8811 (1.8118) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [379] [130/312] eta: 0:02:19 lr: 0.000261 min_lr: 0.000261 loss: 1.8864 (1.8042) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [379] [140/312] eta: 0:02:10 lr: 0.000261 min_lr: 0.000261 loss: 1.8887 (1.7980) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [379] [150/312] eta: 0:02:02 lr: 0.000261 min_lr: 0.000261 loss: 1.8751 (1.7968) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [379] [160/312] eta: 0:01:54 lr: 0.000260 min_lr: 0.000260 loss: 1.7719 (1.7852) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [379] [170/312] eta: 0:01:46 lr: 0.000260 min_lr: 0.000260 loss: 1.7229 (1.7855) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [379] [180/312] eta: 0:01:38 lr: 0.000260 min_lr: 0.000260 loss: 1.7235 (1.7795) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [379] [190/312] eta: 0:01:30 lr: 0.000260 min_lr: 0.000260 loss: 1.6846 (1.7681) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [379] [200/312] eta: 0:01:23 lr: 0.000259 min_lr: 0.000259 loss: 1.7415 (1.7673) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [379] [210/312] eta: 0:01:15 lr: 0.000259 min_lr: 0.000259 loss: 1.7730 (1.7681) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [379] [220/312] eta: 0:01:07 lr: 0.000259 min_lr: 0.000259 loss: 1.7097 (1.7622) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [379] [230/312] eta: 0:01:00 lr: 0.000259 min_lr: 0.000259 loss: 1.7135 (1.7627) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [379] [240/312] eta: 0:00:52 lr: 0.000258 min_lr: 0.000258 loss: 1.7320 (1.7601) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [379] [250/312] eta: 0:00:45 lr: 0.000258 min_lr: 0.000258 loss: 1.6602 (1.7638) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [379] [260/312] eta: 0:00:37 lr: 0.000258 min_lr: 0.000258 loss: 1.8113 (1.7580) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [379] [270/312] eta: 0:00:30 lr: 0.000258 min_lr: 0.000258 loss: 1.6243 (1.7525) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [379] [280/312] eta: 0:00:23 lr: 0.000258 min_lr: 0.000258 loss: 1.4969 (1.7470) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [379] [290/312] eta: 0:00:15 lr: 0.000257 min_lr: 0.000257 loss: 1.8677 (1.7527) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [379] [300/312] eta: 0:00:08 lr: 0.000257 min_lr: 0.000257 loss: 1.7864 (1.7461) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [379] [310/312] eta: 0:00:01 lr: 0.000257 min_lr: 0.000257 loss: 1.7383 (1.7491) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [379] [311/312] eta: 0:00:00 lr: 0.000257 min_lr: 0.000257 loss: 1.7383 (1.7487) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [379] Total time: 0:03:46 (0.7272 s / it) Averaged stats: lr: 0.000257 min_lr: 0.000257 loss: 1.7383 (1.7266) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4561 (0.4561) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.8356 data: 4.6143 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6452 (0.6418) acc1: 84.6354 (82.5600) acc5: 97.3958 (96.8960) time: 0.6887 data: 0.5128 max mem: 64948 Test: Total time: 0:00:06 (0.7124 s / it) * Acc@1 83.140 Acc@5 96.482 loss 0.640 Accuracy of the model on the 50000 test images: 83.1% Max accuracy: 83.25% Test: [0/9] eta: 0:00:45 loss: 0.4567 (0.4567) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 5.0107 data: 4.7950 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6253 (0.6450) acc1: 84.6354 (82.3040) acc5: 97.1354 (96.6080) time: 0.7080 data: 0.5329 max mem: 64948 Test: Total time: 0:00:06 (0.7174 s / it) * Acc@1 83.454 Acc@5 96.616 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.45% Epoch: [380] [ 0/312] eta: 0:53:23 lr: 0.000257 min_lr: 0.000257 loss: 1.1892 (1.1892) weight_decay: 0.0500 (0.0500) time: 10.2672 data: 9.4853 max mem: 64948 Epoch: [380] [ 10/312] eta: 0:07:56 lr: 0.000257 min_lr: 0.000257 loss: 1.7752 (1.6918) weight_decay: 0.0500 (0.0500) time: 1.5768 data: 0.8626 max mem: 64948 Epoch: [380] [ 20/312] eta: 0:05:37 lr: 0.000256 min_lr: 0.000256 loss: 1.7786 (1.7342) weight_decay: 0.0500 (0.0500) time: 0.7007 data: 0.0003 max mem: 64948 Epoch: [380] [ 30/312] eta: 0:04:44 lr: 0.000256 min_lr: 0.000256 loss: 1.8652 (1.7340) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [380] [ 40/312] eta: 0:04:13 lr: 0.000256 min_lr: 0.000256 loss: 1.8558 (1.7420) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0003 max mem: 64948 Epoch: [380] [ 50/312] eta: 0:03:52 lr: 0.000256 min_lr: 0.000256 loss: 1.7384 (1.7037) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [380] [ 60/312] eta: 0:03:35 lr: 0.000255 min_lr: 0.000255 loss: 1.6215 (1.7046) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [380] [ 70/312] eta: 0:03:21 lr: 0.000255 min_lr: 0.000255 loss: 1.4924 (1.6895) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [380] [ 80/312] eta: 0:03:09 lr: 0.000255 min_lr: 0.000255 loss: 1.9078 (1.7150) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [380] [ 90/312] eta: 0:02:58 lr: 0.000255 min_lr: 0.000255 loss: 1.8801 (1.7226) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [380] [100/312] eta: 0:02:47 lr: 0.000255 min_lr: 0.000255 loss: 1.8055 (1.7186) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [380] [110/312] eta: 0:02:38 lr: 0.000254 min_lr: 0.000254 loss: 1.6132 (1.7137) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [380] [120/312] eta: 0:02:28 lr: 0.000254 min_lr: 0.000254 loss: 1.5862 (1.7013) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [380] [130/312] eta: 0:02:20 lr: 0.000254 min_lr: 0.000254 loss: 1.7017 (1.7105) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [380] [140/312] eta: 0:02:11 lr: 0.000254 min_lr: 0.000254 loss: 1.7982 (1.7171) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [380] [150/312] eta: 0:02:03 lr: 0.000253 min_lr: 0.000253 loss: 1.7007 (1.7111) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [380] [160/312] eta: 0:01:54 lr: 0.000253 min_lr: 0.000253 loss: 1.6218 (1.7060) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [380] [170/312] eta: 0:01:46 lr: 0.000253 min_lr: 0.000253 loss: 1.7871 (1.7077) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [380] [180/312] eta: 0:01:38 lr: 0.000253 min_lr: 0.000253 loss: 1.6933 (1.7067) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [380] [190/312] eta: 0:01:31 lr: 0.000253 min_lr: 0.000253 loss: 1.6931 (1.7044) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [380] [200/312] eta: 0:01:23 lr: 0.000252 min_lr: 0.000252 loss: 1.7477 (1.6978) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [380] [210/312] eta: 0:01:15 lr: 0.000252 min_lr: 0.000252 loss: 1.7548 (1.6997) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [380] [220/312] eta: 0:01:07 lr: 0.000252 min_lr: 0.000252 loss: 1.5780 (1.6920) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [380] [230/312] eta: 0:01:00 lr: 0.000252 min_lr: 0.000252 loss: 1.6656 (1.6988) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [380] [240/312] eta: 0:00:52 lr: 0.000251 min_lr: 0.000251 loss: 1.8661 (1.7021) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [380] [250/312] eta: 0:00:45 lr: 0.000251 min_lr: 0.000251 loss: 1.8661 (1.7007) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [380] [260/312] eta: 0:00:38 lr: 0.000251 min_lr: 0.000251 loss: 1.8408 (1.6948) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [380] [270/312] eta: 0:00:30 lr: 0.000251 min_lr: 0.000251 loss: 1.5121 (1.6894) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [380] [280/312] eta: 0:00:23 lr: 0.000250 min_lr: 0.000250 loss: 1.6939 (1.6925) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0010 max mem: 64948 Epoch: [380] [290/312] eta: 0:00:16 lr: 0.000250 min_lr: 0.000250 loss: 1.7861 (1.6902) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0008 max mem: 64948 Epoch: [380] [300/312] eta: 0:00:08 lr: 0.000250 min_lr: 0.000250 loss: 1.5136 (1.6872) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [380] [310/312] eta: 0:00:01 lr: 0.000250 min_lr: 0.000250 loss: 1.5016 (1.6863) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [380] [311/312] eta: 0:00:00 lr: 0.000250 min_lr: 0.000250 loss: 1.5016 (1.6869) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [380] Total time: 0:03:47 (0.7289 s / it) Averaged stats: lr: 0.000250 min_lr: 0.000250 loss: 1.5016 (1.7186) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4604 (0.4604) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 4.7169 data: 4.5039 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6413 (0.6592) acc1: 83.0729 (82.1760) acc5: 96.8750 (96.7040) time: 0.6753 data: 0.5005 max mem: 64948 Test: Total time: 0:00:06 (0.6959 s / it) * Acc@1 83.308 Acc@5 96.534 loss 0.639 Accuracy of the model on the 50000 test images: 83.3% Max accuracy: 83.31% Test: [0/9] eta: 0:00:40 loss: 0.4566 (0.4566) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.5275 data: 4.3224 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6249 (0.6448) acc1: 84.6354 (82.2720) acc5: 97.3958 (96.6720) time: 0.6544 data: 0.4804 max mem: 64948 Test: Total time: 0:00:05 (0.6622 s / it) * Acc@1 83.456 Acc@5 96.634 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.46% Epoch: [381] [ 0/312] eta: 0:49:03 lr: 0.000250 min_lr: 0.000250 loss: 1.6366 (1.6366) weight_decay: 0.0500 (0.0500) time: 9.4327 data: 8.3319 max mem: 64948 Epoch: [381] [ 10/312] eta: 0:07:57 lr: 0.000250 min_lr: 0.000250 loss: 1.7759 (1.6888) weight_decay: 0.0500 (0.0500) time: 1.5818 data: 0.7578 max mem: 64948 Epoch: [381] [ 20/312] eta: 0:05:38 lr: 0.000249 min_lr: 0.000249 loss: 1.8036 (1.7608) weight_decay: 0.0500 (0.0500) time: 0.7450 data: 0.0004 max mem: 64948 Epoch: [381] [ 30/312] eta: 0:04:44 lr: 0.000249 min_lr: 0.000249 loss: 1.8092 (1.7627) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [381] [ 40/312] eta: 0:04:13 lr: 0.000249 min_lr: 0.000249 loss: 1.8092 (1.7841) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [381] [ 50/312] eta: 0:03:52 lr: 0.000249 min_lr: 0.000249 loss: 1.8204 (1.7709) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [381] [ 60/312] eta: 0:03:35 lr: 0.000248 min_lr: 0.000248 loss: 1.7813 (1.7818) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [381] [ 70/312] eta: 0:03:21 lr: 0.000248 min_lr: 0.000248 loss: 1.9106 (1.7815) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [381] [ 80/312] eta: 0:03:09 lr: 0.000248 min_lr: 0.000248 loss: 1.9652 (1.7934) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [381] [ 90/312] eta: 0:02:58 lr: 0.000248 min_lr: 0.000248 loss: 1.8860 (1.7737) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [381] [100/312] eta: 0:02:48 lr: 0.000247 min_lr: 0.000247 loss: 1.8198 (1.7698) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [381] [110/312] eta: 0:02:38 lr: 0.000247 min_lr: 0.000247 loss: 1.6531 (1.7541) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [381] [120/312] eta: 0:02:29 lr: 0.000247 min_lr: 0.000247 loss: 1.6748 (1.7557) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [381] [130/312] eta: 0:02:20 lr: 0.000247 min_lr: 0.000247 loss: 1.6961 (1.7430) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [381] [140/312] eta: 0:02:11 lr: 0.000247 min_lr: 0.000247 loss: 1.4889 (1.7381) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [381] [150/312] eta: 0:02:03 lr: 0.000246 min_lr: 0.000246 loss: 1.7614 (1.7353) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [381] [160/312] eta: 0:01:54 lr: 0.000246 min_lr: 0.000246 loss: 1.7614 (1.7305) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [381] [170/312] eta: 0:01:46 lr: 0.000246 min_lr: 0.000246 loss: 1.8135 (1.7415) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [381] [180/312] eta: 0:01:38 lr: 0.000246 min_lr: 0.000246 loss: 1.7865 (1.7354) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [381] [190/312] eta: 0:01:31 lr: 0.000245 min_lr: 0.000245 loss: 1.7094 (1.7413) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [381] [200/312] eta: 0:01:23 lr: 0.000245 min_lr: 0.000245 loss: 1.8423 (1.7426) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [381] [210/312] eta: 0:01:15 lr: 0.000245 min_lr: 0.000245 loss: 1.8249 (1.7394) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [381] [220/312] eta: 0:01:08 lr: 0.000245 min_lr: 0.000245 loss: 1.7261 (1.7402) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [381] [230/312] eta: 0:01:00 lr: 0.000245 min_lr: 0.000245 loss: 1.7022 (1.7373) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [381] [240/312] eta: 0:00:52 lr: 0.000244 min_lr: 0.000244 loss: 1.6978 (1.7389) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [381] [250/312] eta: 0:00:45 lr: 0.000244 min_lr: 0.000244 loss: 1.8013 (1.7389) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [381] [260/312] eta: 0:00:38 lr: 0.000244 min_lr: 0.000244 loss: 1.8019 (1.7392) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [381] [270/312] eta: 0:00:30 lr: 0.000244 min_lr: 0.000244 loss: 1.8295 (1.7443) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [381] [280/312] eta: 0:00:23 lr: 0.000243 min_lr: 0.000243 loss: 1.8963 (1.7463) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [381] [290/312] eta: 0:00:16 lr: 0.000243 min_lr: 0.000243 loss: 1.8005 (1.7499) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [381] [300/312] eta: 0:00:08 lr: 0.000243 min_lr: 0.000243 loss: 1.7884 (1.7450) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [381] [310/312] eta: 0:00:01 lr: 0.000243 min_lr: 0.000243 loss: 1.8601 (1.7507) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [381] [311/312] eta: 0:00:00 lr: 0.000243 min_lr: 0.000243 loss: 1.8601 (1.7515) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [381] Total time: 0:03:47 (0.7291 s / it) Averaged stats: lr: 0.000243 min_lr: 0.000243 loss: 1.8601 (1.7234) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4595 (0.4595) acc1: 88.5417 (88.5417) acc5: 98.4375 (98.4375) time: 4.7504 data: 4.5428 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6478 (0.6553) acc1: 84.3750 (82.4320) acc5: 96.6146 (96.4800) time: 0.6791 data: 0.5048 max mem: 64948 Test: Total time: 0:00:06 (0.7012 s / it) * Acc@1 83.190 Acc@5 96.452 loss 0.637 Accuracy of the model on the 50000 test images: 83.2% Max accuracy: 83.31% Test: [0/9] eta: 0:00:46 loss: 0.4567 (0.4567) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 5.1483 data: 4.9431 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6250 (0.6447) acc1: 84.6354 (82.2720) acc5: 97.3958 (96.6720) time: 0.7233 data: 0.5493 max mem: 64948 Test: Total time: 0:00:06 (0.7304 s / it) * Acc@1 83.454 Acc@5 96.644 loss 0.623 Accuracy of the model EMA on 50000 test images: 83.5% Epoch: [382] [ 0/312] eta: 0:56:15 lr: 0.000243 min_lr: 0.000243 loss: 1.8586 (1.8586) weight_decay: 0.0500 (0.0500) time: 10.8174 data: 6.6929 max mem: 64948 Epoch: [382] [ 10/312] eta: 0:08:17 lr: 0.000243 min_lr: 0.000243 loss: 1.7520 (1.7310) weight_decay: 0.0500 (0.0500) time: 1.6474 data: 0.6125 max mem: 64948 Epoch: [382] [ 20/312] eta: 0:05:48 lr: 0.000242 min_lr: 0.000242 loss: 1.7520 (1.7284) weight_decay: 0.0500 (0.0500) time: 0.7124 data: 0.0024 max mem: 64948 Epoch: [382] [ 30/312] eta: 0:04:51 lr: 0.000242 min_lr: 0.000242 loss: 1.8299 (1.7441) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0003 max mem: 64948 Epoch: [382] [ 40/312] eta: 0:04:18 lr: 0.000242 min_lr: 0.000242 loss: 1.7651 (1.7238) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [382] [ 50/312] eta: 0:03:56 lr: 0.000242 min_lr: 0.000242 loss: 1.7901 (1.7361) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [382] [ 60/312] eta: 0:03:39 lr: 0.000241 min_lr: 0.000241 loss: 1.8033 (1.7008) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [382] [ 70/312] eta: 0:03:24 lr: 0.000241 min_lr: 0.000241 loss: 1.6868 (1.7150) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [382] [ 80/312] eta: 0:03:11 lr: 0.000241 min_lr: 0.000241 loss: 1.6252 (1.6880) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [382] [ 90/312] eta: 0:03:00 lr: 0.000241 min_lr: 0.000241 loss: 1.6252 (1.6919) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [382] [100/312] eta: 0:02:49 lr: 0.000240 min_lr: 0.000240 loss: 1.8349 (1.7039) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [382] [110/312] eta: 0:02:39 lr: 0.000240 min_lr: 0.000240 loss: 1.8136 (1.6949) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [382] [120/312] eta: 0:02:30 lr: 0.000240 min_lr: 0.000240 loss: 1.8220 (1.7050) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [382] [130/312] eta: 0:02:21 lr: 0.000240 min_lr: 0.000240 loss: 1.8316 (1.7093) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [382] [140/312] eta: 0:02:12 lr: 0.000240 min_lr: 0.000240 loss: 1.7703 (1.7074) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [382] [150/312] eta: 0:02:04 lr: 0.000239 min_lr: 0.000239 loss: 1.7402 (1.7050) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [382] [160/312] eta: 0:01:55 lr: 0.000239 min_lr: 0.000239 loss: 1.7508 (1.7062) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [382] [170/312] eta: 0:01:47 lr: 0.000239 min_lr: 0.000239 loss: 1.7544 (1.7039) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [382] [180/312] eta: 0:01:39 lr: 0.000239 min_lr: 0.000239 loss: 1.6918 (1.7087) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [382] [190/312] eta: 0:01:31 lr: 0.000239 min_lr: 0.000239 loss: 1.8125 (1.7082) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [382] [200/312] eta: 0:01:23 lr: 0.000238 min_lr: 0.000238 loss: 1.8762 (1.7137) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [382] [210/312] eta: 0:01:16 lr: 0.000238 min_lr: 0.000238 loss: 1.8817 (1.7197) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [382] [220/312] eta: 0:01:08 lr: 0.000238 min_lr: 0.000238 loss: 1.7174 (1.7157) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [382] [230/312] eta: 0:01:00 lr: 0.000238 min_lr: 0.000238 loss: 1.7903 (1.7209) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [382] [240/312] eta: 0:00:53 lr: 0.000237 min_lr: 0.000237 loss: 1.9812 (1.7296) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [382] [250/312] eta: 0:00:45 lr: 0.000237 min_lr: 0.000237 loss: 1.8637 (1.7262) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [382] [260/312] eta: 0:00:38 lr: 0.000237 min_lr: 0.000237 loss: 1.6296 (1.7204) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [382] [270/312] eta: 0:00:30 lr: 0.000237 min_lr: 0.000237 loss: 1.6741 (1.7165) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [382] [280/312] eta: 0:00:23 lr: 0.000237 min_lr: 0.000237 loss: 1.6693 (1.7129) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0009 max mem: 64948 Epoch: [382] [290/312] eta: 0:00:16 lr: 0.000236 min_lr: 0.000236 loss: 1.8064 (1.7126) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0008 max mem: 64948 Epoch: [382] [300/312] eta: 0:00:08 lr: 0.000236 min_lr: 0.000236 loss: 1.8142 (1.7160) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [382] [310/312] eta: 0:00:01 lr: 0.000236 min_lr: 0.000236 loss: 1.7177 (1.7116) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [382] [311/312] eta: 0:00:00 lr: 0.000236 min_lr: 0.000236 loss: 1.7177 (1.7133) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [382] Total time: 0:03:48 (0.7315 s / it) Averaged stats: lr: 0.000236 min_lr: 0.000236 loss: 1.7177 (1.7238) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4596 (0.4596) acc1: 88.2812 (88.2812) acc5: 98.4375 (98.4375) time: 4.4956 data: 4.2859 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6783 (0.6632) acc1: 83.8542 (82.5600) acc5: 96.8750 (96.4800) time: 0.6508 data: 0.4763 max mem: 64948 Test: Total time: 0:00:06 (0.6712 s / it) * Acc@1 83.108 Acc@5 96.452 loss 0.642 Accuracy of the model on the 50000 test images: 83.1% Max accuracy: 83.31% Test: [0/9] eta: 0:00:44 loss: 0.4567 (0.4567) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.9262 data: 4.7211 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6252 (0.6446) acc1: 84.6354 (82.2400) acc5: 97.3958 (96.6720) time: 0.6986 data: 0.5247 max mem: 64948 Test: Total time: 0:00:06 (0.7073 s / it) * Acc@1 83.440 Acc@5 96.650 loss 0.622 Accuracy of the model EMA on 50000 test images: 83.4% Epoch: [383] [ 0/312] eta: 0:56:30 lr: 0.000236 min_lr: 0.000236 loss: 1.9029 (1.9029) weight_decay: 0.0500 (0.0500) time: 10.8670 data: 6.6812 max mem: 64948 Epoch: [383] [ 10/312] eta: 0:08:17 lr: 0.000236 min_lr: 0.000236 loss: 1.9109 (1.8707) weight_decay: 0.0500 (0.0500) time: 1.6479 data: 0.6078 max mem: 64948 Epoch: [383] [ 20/312] eta: 0:05:48 lr: 0.000235 min_lr: 0.000235 loss: 1.8250 (1.7522) weight_decay: 0.0500 (0.0500) time: 0.7094 data: 0.0004 max mem: 64948 Epoch: [383] [ 30/312] eta: 0:04:51 lr: 0.000235 min_lr: 0.000235 loss: 1.6027 (1.6998) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [383] [ 40/312] eta: 0:04:18 lr: 0.000235 min_lr: 0.000235 loss: 1.5594 (1.6550) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [383] [ 50/312] eta: 0:03:55 lr: 0.000235 min_lr: 0.000235 loss: 1.7346 (1.6928) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [383] [ 60/312] eta: 0:03:38 lr: 0.000234 min_lr: 0.000234 loss: 1.7494 (1.6785) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [383] [ 70/312] eta: 0:03:23 lr: 0.000234 min_lr: 0.000234 loss: 1.6152 (1.6663) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [383] [ 80/312] eta: 0:03:11 lr: 0.000234 min_lr: 0.000234 loss: 1.5970 (1.6614) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [383] [ 90/312] eta: 0:02:59 lr: 0.000234 min_lr: 0.000234 loss: 1.8487 (1.6756) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [383] [100/312] eta: 0:02:49 lr: 0.000234 min_lr: 0.000234 loss: 1.7993 (1.6749) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [383] [110/312] eta: 0:02:39 lr: 0.000233 min_lr: 0.000233 loss: 1.7349 (1.6660) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [383] [120/312] eta: 0:02:30 lr: 0.000233 min_lr: 0.000233 loss: 1.7805 (1.6745) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [383] [130/312] eta: 0:02:21 lr: 0.000233 min_lr: 0.000233 loss: 1.7805 (1.6730) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [383] [140/312] eta: 0:02:12 lr: 0.000233 min_lr: 0.000233 loss: 1.7900 (1.6766) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [383] [150/312] eta: 0:02:03 lr: 0.000233 min_lr: 0.000233 loss: 1.7540 (1.6826) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [383] [160/312] eta: 0:01:55 lr: 0.000232 min_lr: 0.000232 loss: 1.7540 (1.6816) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [383] [170/312] eta: 0:01:47 lr: 0.000232 min_lr: 0.000232 loss: 1.7674 (1.6792) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [383] [180/312] eta: 0:01:39 lr: 0.000232 min_lr: 0.000232 loss: 1.8708 (1.6861) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [383] [190/312] eta: 0:01:31 lr: 0.000232 min_lr: 0.000232 loss: 1.8844 (1.6958) weight_decay: 0.0500 (0.0500) time: 0.6985 data: 0.0004 max mem: 64948 Epoch: [383] [200/312] eta: 0:01:23 lr: 0.000231 min_lr: 0.000231 loss: 1.8509 (1.6980) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [383] [210/312] eta: 0:01:15 lr: 0.000231 min_lr: 0.000231 loss: 1.7982 (1.7048) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [383] [220/312] eta: 0:01:08 lr: 0.000231 min_lr: 0.000231 loss: 1.6771 (1.6948) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [383] [230/312] eta: 0:01:00 lr: 0.000231 min_lr: 0.000231 loss: 1.5242 (1.6946) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [383] [240/312] eta: 0:00:53 lr: 0.000231 min_lr: 0.000231 loss: 1.5242 (1.6928) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [383] [250/312] eta: 0:00:45 lr: 0.000230 min_lr: 0.000230 loss: 1.8224 (1.6970) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [383] [260/312] eta: 0:00:38 lr: 0.000230 min_lr: 0.000230 loss: 1.8224 (1.7004) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [383] [270/312] eta: 0:00:30 lr: 0.000230 min_lr: 0.000230 loss: 1.7939 (1.7021) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [383] [280/312] eta: 0:00:23 lr: 0.000230 min_lr: 0.000230 loss: 1.8635 (1.7046) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0009 max mem: 64948 Epoch: [383] [290/312] eta: 0:00:16 lr: 0.000229 min_lr: 0.000229 loss: 1.8635 (1.7096) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0008 max mem: 64948 Epoch: [383] [300/312] eta: 0:00:08 lr: 0.000229 min_lr: 0.000229 loss: 1.7608 (1.7094) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [383] [310/312] eta: 0:00:01 lr: 0.000229 min_lr: 0.000229 loss: 1.6931 (1.7098) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [383] [311/312] eta: 0:00:00 lr: 0.000229 min_lr: 0.000229 loss: 1.6693 (1.7091) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [383] Total time: 0:03:48 (0.7314 s / it) Averaged stats: lr: 0.000229 min_lr: 0.000229 loss: 1.6693 (1.7124) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4589 (0.4589) acc1: 87.7604 (87.7604) acc5: 98.4375 (98.4375) time: 4.6292 data: 4.4095 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6427 (0.6497) acc1: 84.1146 (82.3680) acc5: 96.8750 (96.5440) time: 0.6661 data: 0.4900 max mem: 64948 Test: Total time: 0:00:06 (0.6876 s / it) * Acc@1 83.264 Acc@5 96.602 loss 0.632 Accuracy of the model on the 50000 test images: 83.3% Max accuracy: 83.31% Test: [0/9] eta: 0:00:42 loss: 0.4565 (0.4565) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.7396 data: 4.5229 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6253 (0.6444) acc1: 84.6354 (82.2720) acc5: 97.1354 (96.6400) time: 0.6787 data: 0.5026 max mem: 64948 Test: Total time: 0:00:06 (0.6900 s / it) * Acc@1 83.442 Acc@5 96.654 loss 0.622 Accuracy of the model EMA on 50000 test images: 83.4% Epoch: [384] [ 0/312] eta: 0:56:55 lr: 0.000229 min_lr: 0.000229 loss: 1.5454 (1.5454) weight_decay: 0.0500 (0.0500) time: 10.9466 data: 9.3169 max mem: 64948 Epoch: [384] [ 10/312] eta: 0:08:28 lr: 0.000229 min_lr: 0.000229 loss: 1.5454 (1.5552) weight_decay: 0.0500 (0.0500) time: 1.6846 data: 0.8474 max mem: 64948 Epoch: [384] [ 20/312] eta: 0:05:54 lr: 0.000229 min_lr: 0.000229 loss: 1.5733 (1.5947) weight_decay: 0.0500 (0.0500) time: 0.7285 data: 0.0004 max mem: 64948 Epoch: [384] [ 30/312] eta: 0:04:55 lr: 0.000228 min_lr: 0.000228 loss: 1.7541 (1.6244) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0003 max mem: 64948 Epoch: [384] [ 40/312] eta: 0:04:21 lr: 0.000228 min_lr: 0.000228 loss: 1.6992 (1.6397) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [384] [ 50/312] eta: 0:03:58 lr: 0.000228 min_lr: 0.000228 loss: 1.6992 (1.6517) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [384] [ 60/312] eta: 0:03:40 lr: 0.000228 min_lr: 0.000228 loss: 1.6255 (1.6462) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [384] [ 70/312] eta: 0:03:25 lr: 0.000227 min_lr: 0.000227 loss: 1.8062 (1.6633) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [384] [ 80/312] eta: 0:03:12 lr: 0.000227 min_lr: 0.000227 loss: 1.8091 (1.6675) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [384] [ 90/312] eta: 0:03:01 lr: 0.000227 min_lr: 0.000227 loss: 1.6406 (1.6607) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [384] [100/312] eta: 0:02:50 lr: 0.000227 min_lr: 0.000227 loss: 1.7508 (1.6722) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [384] [110/312] eta: 0:02:40 lr: 0.000227 min_lr: 0.000227 loss: 1.8423 (1.6689) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [384] [120/312] eta: 0:02:31 lr: 0.000226 min_lr: 0.000226 loss: 1.7982 (1.6728) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [384] [130/312] eta: 0:02:21 lr: 0.000226 min_lr: 0.000226 loss: 1.8690 (1.6883) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [384] [140/312] eta: 0:02:13 lr: 0.000226 min_lr: 0.000226 loss: 1.8690 (1.6890) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [384] [150/312] eta: 0:02:04 lr: 0.000226 min_lr: 0.000226 loss: 1.7297 (1.6839) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [384] [160/312] eta: 0:01:56 lr: 0.000226 min_lr: 0.000226 loss: 1.7297 (1.6883) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [384] [170/312] eta: 0:01:47 lr: 0.000225 min_lr: 0.000225 loss: 1.8485 (1.6923) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [384] [180/312] eta: 0:01:39 lr: 0.000225 min_lr: 0.000225 loss: 1.8993 (1.7005) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [384] [190/312] eta: 0:01:31 lr: 0.000225 min_lr: 0.000225 loss: 1.8945 (1.6987) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [384] [200/312] eta: 0:01:24 lr: 0.000225 min_lr: 0.000225 loss: 1.5917 (1.6897) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [384] [210/312] eta: 0:01:16 lr: 0.000224 min_lr: 0.000224 loss: 1.6281 (1.6957) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [384] [220/312] eta: 0:01:08 lr: 0.000224 min_lr: 0.000224 loss: 1.8190 (1.6912) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [384] [230/312] eta: 0:01:00 lr: 0.000224 min_lr: 0.000224 loss: 1.7272 (1.6887) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [384] [240/312] eta: 0:00:53 lr: 0.000224 min_lr: 0.000224 loss: 1.7272 (1.6944) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [384] [250/312] eta: 0:00:45 lr: 0.000224 min_lr: 0.000224 loss: 1.8031 (1.6993) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [384] [260/312] eta: 0:00:38 lr: 0.000223 min_lr: 0.000223 loss: 1.8640 (1.7011) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [384] [270/312] eta: 0:00:30 lr: 0.000223 min_lr: 0.000223 loss: 1.7646 (1.7026) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [384] [280/312] eta: 0:00:23 lr: 0.000223 min_lr: 0.000223 loss: 1.7213 (1.6994) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [384] [290/312] eta: 0:00:16 lr: 0.000223 min_lr: 0.000223 loss: 1.7254 (1.7014) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [384] [300/312] eta: 0:00:08 lr: 0.000223 min_lr: 0.000223 loss: 1.7753 (1.6999) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [384] [310/312] eta: 0:00:01 lr: 0.000222 min_lr: 0.000222 loss: 1.7617 (1.7015) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [384] [311/312] eta: 0:00:00 lr: 0.000222 min_lr: 0.000222 loss: 1.7753 (1.7026) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [384] Total time: 0:03:48 (0.7331 s / it) Averaged stats: lr: 0.000222 min_lr: 0.000222 loss: 1.7753 (1.7099) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4601 (0.4601) acc1: 87.7604 (87.7604) acc5: 97.6562 (97.6562) time: 4.4445 data: 4.2244 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6626 (0.6636) acc1: 83.5938 (82.4640) acc5: 96.8750 (96.3520) time: 0.6455 data: 0.4695 max mem: 64948 Test: Total time: 0:00:06 (0.6671 s / it) * Acc@1 83.242 Acc@5 96.462 loss 0.640 Accuracy of the model on the 50000 test images: 83.2% Max accuracy: 83.31% Test: [0/9] eta: 0:00:45 loss: 0.4564 (0.4564) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 5.0201 data: 4.8060 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6253 (0.6444) acc1: 84.6354 (82.3040) acc5: 97.1354 (96.6400) time: 0.7100 data: 0.5341 max mem: 64948 Test: Total time: 0:00:06 (0.7233 s / it) * Acc@1 83.462 Acc@5 96.658 loss 0.622 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.46% Epoch: [385] [ 0/312] eta: 0:48:31 lr: 0.000222 min_lr: 0.000222 loss: 1.9692 (1.9692) weight_decay: 0.0500 (0.0500) time: 9.3333 data: 8.1208 max mem: 64948 Epoch: [385] [ 10/312] eta: 0:07:45 lr: 0.000222 min_lr: 0.000222 loss: 1.7631 (1.7323) weight_decay: 0.0500 (0.0500) time: 1.5415 data: 0.7386 max mem: 64948 Epoch: [385] [ 20/312] eta: 0:05:32 lr: 0.000222 min_lr: 0.000222 loss: 1.6183 (1.6220) weight_decay: 0.0500 (0.0500) time: 0.7284 data: 0.0003 max mem: 64948 Epoch: [385] [ 30/312] eta: 0:04:40 lr: 0.000222 min_lr: 0.000222 loss: 1.4866 (1.6107) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [385] [ 40/312] eta: 0:04:10 lr: 0.000221 min_lr: 0.000221 loss: 1.6598 (1.6290) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [385] [ 50/312] eta: 0:03:50 lr: 0.000221 min_lr: 0.000221 loss: 1.6598 (1.6097) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [385] [ 60/312] eta: 0:03:33 lr: 0.000221 min_lr: 0.000221 loss: 1.7325 (1.6284) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [385] [ 70/312] eta: 0:03:20 lr: 0.000221 min_lr: 0.000221 loss: 1.7753 (1.6375) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [385] [ 80/312] eta: 0:03:07 lr: 0.000221 min_lr: 0.000221 loss: 1.7753 (1.6499) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [385] [ 90/312] eta: 0:02:57 lr: 0.000220 min_lr: 0.000220 loss: 1.7169 (1.6548) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [385] [100/312] eta: 0:02:47 lr: 0.000220 min_lr: 0.000220 loss: 1.8421 (1.6815) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [385] [110/312] eta: 0:02:37 lr: 0.000220 min_lr: 0.000220 loss: 1.8616 (1.6771) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [385] [120/312] eta: 0:02:28 lr: 0.000220 min_lr: 0.000220 loss: 1.7096 (1.6796) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [385] [130/312] eta: 0:02:19 lr: 0.000219 min_lr: 0.000219 loss: 1.8390 (1.6961) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [385] [140/312] eta: 0:02:11 lr: 0.000219 min_lr: 0.000219 loss: 1.8421 (1.6936) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [385] [150/312] eta: 0:02:02 lr: 0.000219 min_lr: 0.000219 loss: 1.8090 (1.7015) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [385] [160/312] eta: 0:01:54 lr: 0.000219 min_lr: 0.000219 loss: 1.8401 (1.7022) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [385] [170/312] eta: 0:01:46 lr: 0.000219 min_lr: 0.000219 loss: 1.7879 (1.6977) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [385] [180/312] eta: 0:01:38 lr: 0.000218 min_lr: 0.000218 loss: 1.7879 (1.7037) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [385] [190/312] eta: 0:01:30 lr: 0.000218 min_lr: 0.000218 loss: 1.7909 (1.6999) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [385] [200/312] eta: 0:01:23 lr: 0.000218 min_lr: 0.000218 loss: 1.5794 (1.6977) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [385] [210/312] eta: 0:01:15 lr: 0.000218 min_lr: 0.000218 loss: 1.6994 (1.7014) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [385] [220/312] eta: 0:01:07 lr: 0.000218 min_lr: 0.000218 loss: 1.5110 (1.6911) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [385] [230/312] eta: 0:01:00 lr: 0.000217 min_lr: 0.000217 loss: 1.5460 (1.6955) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [385] [240/312] eta: 0:00:52 lr: 0.000217 min_lr: 0.000217 loss: 1.8516 (1.6973) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [385] [250/312] eta: 0:00:45 lr: 0.000217 min_lr: 0.000217 loss: 1.8127 (1.6982) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [385] [260/312] eta: 0:00:38 lr: 0.000217 min_lr: 0.000217 loss: 1.7714 (1.6967) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [385] [270/312] eta: 0:00:30 lr: 0.000217 min_lr: 0.000217 loss: 1.7955 (1.6995) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [385] [280/312] eta: 0:00:23 lr: 0.000216 min_lr: 0.000216 loss: 1.8257 (1.7006) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [385] [290/312] eta: 0:00:16 lr: 0.000216 min_lr: 0.000216 loss: 1.8480 (1.7038) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [385] [300/312] eta: 0:00:08 lr: 0.000216 min_lr: 0.000216 loss: 1.7472 (1.7014) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [385] [310/312] eta: 0:00:01 lr: 0.000216 min_lr: 0.000216 loss: 1.7472 (1.7037) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [385] [311/312] eta: 0:00:00 lr: 0.000216 min_lr: 0.000216 loss: 1.7760 (1.7040) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [385] Total time: 0:03:47 (0.7280 s / it) Averaged stats: lr: 0.000216 min_lr: 0.000216 loss: 1.7760 (1.7085) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4537 (0.4537) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 4.6805 data: 4.4669 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6460 (0.6534) acc1: 83.5938 (82.4640) acc5: 96.8750 (96.5440) time: 0.6713 data: 0.4964 max mem: 64948 Test: Total time: 0:00:06 (0.6953 s / it) * Acc@1 83.404 Acc@5 96.560 loss 0.635 Accuracy of the model on the 50000 test images: 83.4% Max accuracy: 83.40% Test: [0/9] eta: 0:00:41 loss: 0.4564 (0.4564) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.5998 data: 4.3912 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6252 (0.6444) acc1: 84.6354 (82.3360) acc5: 97.1354 (96.6400) time: 0.6652 data: 0.4880 max mem: 64948 Test: Total time: 0:00:06 (0.6732 s / it) * Acc@1 83.482 Acc@5 96.662 loss 0.622 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.48% Epoch: [386] [ 0/312] eta: 0:55:02 lr: 0.000216 min_lr: 0.000216 loss: 1.8170 (1.8170) weight_decay: 0.0500 (0.0500) time: 10.5863 data: 9.8398 max mem: 64948 Epoch: [386] [ 10/312] eta: 0:08:09 lr: 0.000215 min_lr: 0.000215 loss: 1.6488 (1.6482) weight_decay: 0.0500 (0.0500) time: 1.6207 data: 0.8948 max mem: 64948 Epoch: [386] [ 20/312] eta: 0:05:44 lr: 0.000215 min_lr: 0.000215 loss: 1.7509 (1.7581) weight_decay: 0.0500 (0.0500) time: 0.7103 data: 0.0003 max mem: 64948 Epoch: [386] [ 30/312] eta: 0:04:48 lr: 0.000215 min_lr: 0.000215 loss: 1.9035 (1.7646) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [386] [ 40/312] eta: 0:04:16 lr: 0.000215 min_lr: 0.000215 loss: 1.9258 (1.7832) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [386] [ 50/312] eta: 0:03:54 lr: 0.000215 min_lr: 0.000215 loss: 1.8010 (1.7673) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [386] [ 60/312] eta: 0:03:37 lr: 0.000214 min_lr: 0.000214 loss: 1.7838 (1.7437) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [386] [ 70/312] eta: 0:03:23 lr: 0.000214 min_lr: 0.000214 loss: 1.6073 (1.7381) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [386] [ 80/312] eta: 0:03:10 lr: 0.000214 min_lr: 0.000214 loss: 1.7382 (1.7430) weight_decay: 0.0500 (0.0500) time: 0.7013 data: 0.0004 max mem: 64948 Epoch: [386] [ 90/312] eta: 0:02:59 lr: 0.000214 min_lr: 0.000214 loss: 1.6801 (1.7270) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [386] [100/312] eta: 0:02:49 lr: 0.000214 min_lr: 0.000214 loss: 1.7112 (1.7365) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [386] [110/312] eta: 0:02:39 lr: 0.000213 min_lr: 0.000213 loss: 1.7959 (1.7356) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [386] [120/312] eta: 0:02:29 lr: 0.000213 min_lr: 0.000213 loss: 1.7442 (1.7211) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [386] [130/312] eta: 0:02:20 lr: 0.000213 min_lr: 0.000213 loss: 1.6686 (1.7209) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [386] [140/312] eta: 0:02:12 lr: 0.000213 min_lr: 0.000213 loss: 1.6686 (1.7104) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [386] [150/312] eta: 0:02:03 lr: 0.000212 min_lr: 0.000212 loss: 1.4698 (1.7006) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [386] [160/312] eta: 0:01:55 lr: 0.000212 min_lr: 0.000212 loss: 1.7570 (1.7021) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [386] [170/312] eta: 0:01:47 lr: 0.000212 min_lr: 0.000212 loss: 1.8015 (1.7048) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [386] [180/312] eta: 0:01:39 lr: 0.000212 min_lr: 0.000212 loss: 1.7429 (1.7036) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [386] [190/312] eta: 0:01:31 lr: 0.000212 min_lr: 0.000212 loss: 1.7473 (1.7044) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [386] [200/312] eta: 0:01:23 lr: 0.000211 min_lr: 0.000211 loss: 1.7993 (1.7077) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [386] [210/312] eta: 0:01:15 lr: 0.000211 min_lr: 0.000211 loss: 1.8129 (1.7129) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [386] [220/312] eta: 0:01:08 lr: 0.000211 min_lr: 0.000211 loss: 1.9301 (1.7237) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [386] [230/312] eta: 0:01:00 lr: 0.000211 min_lr: 0.000211 loss: 1.9301 (1.7265) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [386] [240/312] eta: 0:00:53 lr: 0.000211 min_lr: 0.000211 loss: 1.7962 (1.7284) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [386] [250/312] eta: 0:00:45 lr: 0.000210 min_lr: 0.000210 loss: 1.8148 (1.7277) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [386] [260/312] eta: 0:00:38 lr: 0.000210 min_lr: 0.000210 loss: 1.7959 (1.7282) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [386] [270/312] eta: 0:00:30 lr: 0.000210 min_lr: 0.000210 loss: 1.8053 (1.7284) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [386] [280/312] eta: 0:00:23 lr: 0.000210 min_lr: 0.000210 loss: 1.8326 (1.7313) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [386] [290/312] eta: 0:00:16 lr: 0.000210 min_lr: 0.000210 loss: 1.8893 (1.7325) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [386] [300/312] eta: 0:00:08 lr: 0.000209 min_lr: 0.000209 loss: 1.7742 (1.7335) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [386] [310/312] eta: 0:00:01 lr: 0.000209 min_lr: 0.000209 loss: 1.7742 (1.7378) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [386] [311/312] eta: 0:00:00 lr: 0.000209 min_lr: 0.000209 loss: 1.7742 (1.7380) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [386] Total time: 0:03:47 (0.7307 s / it) Averaged stats: lr: 0.000209 min_lr: 0.000209 loss: 1.7742 (1.7105) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4692 (0.4692) acc1: 86.4583 (86.4583) acc5: 97.9167 (97.9167) time: 4.6420 data: 4.4356 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6555 (0.6570) acc1: 84.3750 (82.2080) acc5: 97.3958 (96.6720) time: 0.6670 data: 0.4929 max mem: 64948 Test: Total time: 0:00:06 (0.6897 s / it) * Acc@1 83.118 Acc@5 96.492 loss 0.640 Accuracy of the model on the 50000 test images: 83.1% Max accuracy: 83.40% Test: [0/9] eta: 0:00:42 loss: 0.4561 (0.4561) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.7714 data: 4.5485 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6251 (0.6441) acc1: 84.8958 (82.3360) acc5: 97.1354 (96.6720) time: 0.6820 data: 0.5055 max mem: 64948 Test: Total time: 0:00:06 (0.6964 s / it) * Acc@1 83.484 Acc@5 96.670 loss 0.622 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.48% Epoch: [387] [ 0/312] eta: 0:53:01 lr: 0.000209 min_lr: 0.000209 loss: 1.9106 (1.9106) weight_decay: 0.0500 (0.0500) time: 10.1986 data: 9.4095 max mem: 64948 Epoch: [387] [ 10/312] eta: 0:07:54 lr: 0.000209 min_lr: 0.000209 loss: 1.6433 (1.7141) weight_decay: 0.0500 (0.0500) time: 1.5710 data: 0.8558 max mem: 64948 Epoch: [387] [ 20/312] eta: 0:05:36 lr: 0.000209 min_lr: 0.000209 loss: 1.6077 (1.7003) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0003 max mem: 64948 Epoch: [387] [ 30/312] eta: 0:04:43 lr: 0.000208 min_lr: 0.000208 loss: 1.6077 (1.6822) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [387] [ 40/312] eta: 0:04:12 lr: 0.000208 min_lr: 0.000208 loss: 1.5015 (1.6472) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [387] [ 50/312] eta: 0:03:51 lr: 0.000208 min_lr: 0.000208 loss: 1.5015 (1.6445) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [387] [ 60/312] eta: 0:03:34 lr: 0.000208 min_lr: 0.000208 loss: 1.7699 (1.6815) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [387] [ 70/312] eta: 0:03:21 lr: 0.000208 min_lr: 0.000208 loss: 1.7699 (1.6740) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [387] [ 80/312] eta: 0:03:08 lr: 0.000207 min_lr: 0.000207 loss: 1.6703 (1.6787) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [387] [ 90/312] eta: 0:02:57 lr: 0.000207 min_lr: 0.000207 loss: 1.7549 (1.6758) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [387] [100/312] eta: 0:02:47 lr: 0.000207 min_lr: 0.000207 loss: 1.8126 (1.6906) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [387] [110/312] eta: 0:02:37 lr: 0.000207 min_lr: 0.000207 loss: 1.8478 (1.6965) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [387] [120/312] eta: 0:02:28 lr: 0.000207 min_lr: 0.000207 loss: 1.7866 (1.7058) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [387] [130/312] eta: 0:02:19 lr: 0.000206 min_lr: 0.000206 loss: 1.7660 (1.7098) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [387] [140/312] eta: 0:02:11 lr: 0.000206 min_lr: 0.000206 loss: 1.7740 (1.7178) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [387] [150/312] eta: 0:02:02 lr: 0.000206 min_lr: 0.000206 loss: 1.7740 (1.7204) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [387] [160/312] eta: 0:01:54 lr: 0.000206 min_lr: 0.000206 loss: 1.7508 (1.7200) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [387] [170/312] eta: 0:01:46 lr: 0.000206 min_lr: 0.000206 loss: 1.7348 (1.7192) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [387] [180/312] eta: 0:01:38 lr: 0.000205 min_lr: 0.000205 loss: 1.8315 (1.7224) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [387] [190/312] eta: 0:01:30 lr: 0.000205 min_lr: 0.000205 loss: 1.8507 (1.7271) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [387] [200/312] eta: 0:01:23 lr: 0.000205 min_lr: 0.000205 loss: 1.7595 (1.7249) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [387] [210/312] eta: 0:01:15 lr: 0.000205 min_lr: 0.000205 loss: 1.7717 (1.7247) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [387] [220/312] eta: 0:01:07 lr: 0.000205 min_lr: 0.000205 loss: 1.7203 (1.7225) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [387] [230/312] eta: 0:01:00 lr: 0.000204 min_lr: 0.000204 loss: 1.7228 (1.7210) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [387] [240/312] eta: 0:00:52 lr: 0.000204 min_lr: 0.000204 loss: 1.7825 (1.7221) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [387] [250/312] eta: 0:00:45 lr: 0.000204 min_lr: 0.000204 loss: 1.7839 (1.7252) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [387] [260/312] eta: 0:00:38 lr: 0.000204 min_lr: 0.000204 loss: 1.7490 (1.7207) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [387] [270/312] eta: 0:00:30 lr: 0.000204 min_lr: 0.000204 loss: 1.7398 (1.7186) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [387] [280/312] eta: 0:00:23 lr: 0.000203 min_lr: 0.000203 loss: 1.6259 (1.7159) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0009 max mem: 64948 Epoch: [387] [290/312] eta: 0:00:16 lr: 0.000203 min_lr: 0.000203 loss: 1.6801 (1.7161) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [387] [300/312] eta: 0:00:08 lr: 0.000203 min_lr: 0.000203 loss: 1.6290 (1.7146) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [387] [310/312] eta: 0:00:01 lr: 0.000203 min_lr: 0.000203 loss: 1.6629 (1.7164) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0002 max mem: 64948 Epoch: [387] [311/312] eta: 0:00:00 lr: 0.000203 min_lr: 0.000203 loss: 1.7077 (1.7171) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [387] Total time: 0:03:47 (0.7287 s / it) Averaged stats: lr: 0.000203 min_lr: 0.000203 loss: 1.7077 (1.7096) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4531 (0.4531) acc1: 88.2812 (88.2812) acc5: 98.1771 (98.1771) time: 4.4746 data: 4.2629 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6558 (0.6483) acc1: 84.8958 (82.6560) acc5: 96.8750 (96.5120) time: 0.6484 data: 0.4738 max mem: 64948 Test: Total time: 0:00:06 (0.6695 s / it) * Acc@1 83.264 Acc@5 96.498 loss 0.635 Accuracy of the model on the 50000 test images: 83.3% Max accuracy: 83.40% Test: [0/9] eta: 0:00:42 loss: 0.4562 (0.4562) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.7633 data: 4.5546 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6249 (0.6439) acc1: 84.6354 (82.2720) acc5: 97.1354 (96.6720) time: 0.6805 data: 0.5062 max mem: 64948 Test: Total time: 0:00:06 (0.6875 s / it) * Acc@1 83.484 Acc@5 96.672 loss 0.622 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.48% Epoch: [388] [ 0/312] eta: 0:53:47 lr: 0.000203 min_lr: 0.000203 loss: 1.1088 (1.1088) weight_decay: 0.0500 (0.0500) time: 10.3430 data: 9.5559 max mem: 64948 Epoch: [388] [ 10/312] eta: 0:07:59 lr: 0.000202 min_lr: 0.000202 loss: 1.6787 (1.6157) weight_decay: 0.0500 (0.0500) time: 1.5882 data: 0.8691 max mem: 64948 Epoch: [388] [ 20/312] eta: 0:05:40 lr: 0.000202 min_lr: 0.000202 loss: 1.6893 (1.7050) weight_decay: 0.0500 (0.0500) time: 0.7082 data: 0.0003 max mem: 64948 Epoch: [388] [ 30/312] eta: 0:04:46 lr: 0.000202 min_lr: 0.000202 loss: 1.8570 (1.7053) weight_decay: 0.0500 (0.0500) time: 0.7020 data: 0.0003 max mem: 64948 Epoch: [388] [ 40/312] eta: 0:04:15 lr: 0.000202 min_lr: 0.000202 loss: 1.6648 (1.6810) weight_decay: 0.0500 (0.0500) time: 0.6999 data: 0.0004 max mem: 64948 Epoch: [388] [ 50/312] eta: 0:03:53 lr: 0.000202 min_lr: 0.000202 loss: 1.6648 (1.6978) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [388] [ 60/312] eta: 0:03:36 lr: 0.000201 min_lr: 0.000201 loss: 1.8539 (1.7030) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [388] [ 70/312] eta: 0:03:22 lr: 0.000201 min_lr: 0.000201 loss: 1.8202 (1.7097) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [388] [ 80/312] eta: 0:03:09 lr: 0.000201 min_lr: 0.000201 loss: 1.6082 (1.6882) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [388] [ 90/312] eta: 0:02:58 lr: 0.000201 min_lr: 0.000201 loss: 1.5679 (1.6772) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [388] [100/312] eta: 0:02:48 lr: 0.000201 min_lr: 0.000201 loss: 1.6833 (1.6848) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [388] [110/312] eta: 0:02:38 lr: 0.000200 min_lr: 0.000200 loss: 1.9056 (1.7020) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [388] [120/312] eta: 0:02:29 lr: 0.000200 min_lr: 0.000200 loss: 1.8610 (1.7127) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [388] [130/312] eta: 0:02:20 lr: 0.000200 min_lr: 0.000200 loss: 1.7958 (1.7122) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [388] [140/312] eta: 0:02:11 lr: 0.000200 min_lr: 0.000200 loss: 1.6877 (1.7129) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [388] [150/312] eta: 0:02:03 lr: 0.000200 min_lr: 0.000200 loss: 1.8034 (1.7198) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [388] [160/312] eta: 0:01:55 lr: 0.000199 min_lr: 0.000199 loss: 1.7820 (1.7176) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [388] [170/312] eta: 0:01:46 lr: 0.000199 min_lr: 0.000199 loss: 1.6904 (1.7106) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [388] [180/312] eta: 0:01:38 lr: 0.000199 min_lr: 0.000199 loss: 1.7839 (1.7159) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [388] [190/312] eta: 0:01:31 lr: 0.000199 min_lr: 0.000199 loss: 1.7806 (1.7118) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [388] [200/312] eta: 0:01:23 lr: 0.000199 min_lr: 0.000199 loss: 1.7472 (1.7124) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [388] [210/312] eta: 0:01:15 lr: 0.000198 min_lr: 0.000198 loss: 1.7226 (1.7026) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [388] [220/312] eta: 0:01:08 lr: 0.000198 min_lr: 0.000198 loss: 1.6468 (1.7057) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [388] [230/312] eta: 0:01:00 lr: 0.000198 min_lr: 0.000198 loss: 1.8355 (1.7080) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [388] [240/312] eta: 0:00:52 lr: 0.000198 min_lr: 0.000198 loss: 1.8388 (1.7141) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [388] [250/312] eta: 0:00:45 lr: 0.000198 min_lr: 0.000198 loss: 1.8325 (1.7155) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [388] [260/312] eta: 0:00:38 lr: 0.000197 min_lr: 0.000197 loss: 1.5865 (1.7097) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [388] [270/312] eta: 0:00:30 lr: 0.000197 min_lr: 0.000197 loss: 1.5793 (1.7104) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [388] [280/312] eta: 0:00:23 lr: 0.000197 min_lr: 0.000197 loss: 1.6047 (1.7068) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0010 max mem: 64948 Epoch: [388] [290/312] eta: 0:00:16 lr: 0.000197 min_lr: 0.000197 loss: 1.6589 (1.7075) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0008 max mem: 64948 Epoch: [388] [300/312] eta: 0:00:08 lr: 0.000197 min_lr: 0.000197 loss: 1.6589 (1.7061) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0001 max mem: 64948 Epoch: [388] [310/312] eta: 0:00:01 lr: 0.000196 min_lr: 0.000196 loss: 1.7984 (1.7129) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [388] [311/312] eta: 0:00:00 lr: 0.000196 min_lr: 0.000196 loss: 1.8358 (1.7136) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [388] Total time: 0:03:47 (0.7294 s / it) Averaged stats: lr: 0.000196 min_lr: 0.000196 loss: 1.8358 (1.7051) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4785 (0.4785) acc1: 87.7604 (87.7604) acc5: 98.4375 (98.4375) time: 4.6198 data: 4.4066 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6378 (0.6635) acc1: 82.2917 (82.1440) acc5: 96.8750 (96.4800) time: 0.6645 data: 0.4897 max mem: 64948 Test: Total time: 0:00:06 (0.6872 s / it) * Acc@1 83.292 Acc@5 96.498 loss 0.635 Accuracy of the model on the 50000 test images: 83.3% Max accuracy: 83.40% Test: [0/9] eta: 0:00:43 loss: 0.4560 (0.4560) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.8366 data: 4.6185 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6247 (0.6438) acc1: 84.8958 (82.2720) acc5: 97.1354 (96.6720) time: 0.6894 data: 0.5133 max mem: 64948 Test: Total time: 0:00:06 (0.6975 s / it) * Acc@1 83.496 Acc@5 96.676 loss 0.622 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.50% Epoch: [389] [ 0/312] eta: 0:48:34 lr: 0.000196 min_lr: 0.000196 loss: 1.2963 (1.2963) weight_decay: 0.0500 (0.0500) time: 9.3421 data: 8.5403 max mem: 64948 Epoch: [389] [ 10/312] eta: 0:07:35 lr: 0.000196 min_lr: 0.000196 loss: 1.6240 (1.6655) weight_decay: 0.0500 (0.0500) time: 1.5099 data: 0.7767 max mem: 64948 Epoch: [389] [ 20/312] eta: 0:05:28 lr: 0.000196 min_lr: 0.000196 loss: 1.6447 (1.6739) weight_decay: 0.0500 (0.0500) time: 0.7126 data: 0.0004 max mem: 64948 Epoch: [389] [ 30/312] eta: 0:04:37 lr: 0.000196 min_lr: 0.000196 loss: 1.6447 (1.6704) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [389] [ 40/312] eta: 0:04:08 lr: 0.000195 min_lr: 0.000195 loss: 1.6592 (1.6617) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [389] [ 50/312] eta: 0:03:48 lr: 0.000195 min_lr: 0.000195 loss: 1.6687 (1.6495) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [389] [ 60/312] eta: 0:03:32 lr: 0.000195 min_lr: 0.000195 loss: 1.6060 (1.6565) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [389] [ 70/312] eta: 0:03:18 lr: 0.000195 min_lr: 0.000195 loss: 1.6486 (1.6591) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [389] [ 80/312] eta: 0:03:07 lr: 0.000195 min_lr: 0.000195 loss: 1.7074 (1.6457) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [389] [ 90/312] eta: 0:02:56 lr: 0.000194 min_lr: 0.000194 loss: 1.8478 (1.6763) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [389] [100/312] eta: 0:02:46 lr: 0.000194 min_lr: 0.000194 loss: 1.8630 (1.6905) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [389] [110/312] eta: 0:02:36 lr: 0.000194 min_lr: 0.000194 loss: 1.8292 (1.7039) weight_decay: 0.0500 (0.0500) time: 0.7008 data: 0.0004 max mem: 64948 Epoch: [389] [120/312] eta: 0:02:27 lr: 0.000194 min_lr: 0.000194 loss: 1.8050 (1.7074) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [389] [130/312] eta: 0:02:19 lr: 0.000194 min_lr: 0.000194 loss: 1.7296 (1.7105) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [389] [140/312] eta: 0:02:10 lr: 0.000193 min_lr: 0.000193 loss: 1.6819 (1.7071) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [389] [150/312] eta: 0:02:02 lr: 0.000193 min_lr: 0.000193 loss: 1.7121 (1.7086) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [389] [160/312] eta: 0:01:54 lr: 0.000193 min_lr: 0.000193 loss: 1.7398 (1.7138) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [389] [170/312] eta: 0:01:46 lr: 0.000193 min_lr: 0.000193 loss: 1.8385 (1.7163) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [389] [180/312] eta: 0:01:38 lr: 0.000193 min_lr: 0.000193 loss: 1.8166 (1.7187) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [389] [190/312] eta: 0:01:30 lr: 0.000192 min_lr: 0.000192 loss: 1.7793 (1.7196) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [389] [200/312] eta: 0:01:22 lr: 0.000192 min_lr: 0.000192 loss: 1.6080 (1.7056) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [389] [210/312] eta: 0:01:15 lr: 0.000192 min_lr: 0.000192 loss: 1.6636 (1.7117) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [389] [220/312] eta: 0:01:07 lr: 0.000192 min_lr: 0.000192 loss: 1.8620 (1.7171) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [389] [230/312] eta: 0:01:00 lr: 0.000192 min_lr: 0.000192 loss: 1.8417 (1.7208) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [389] [240/312] eta: 0:00:52 lr: 0.000191 min_lr: 0.000191 loss: 1.7130 (1.7144) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [389] [250/312] eta: 0:00:45 lr: 0.000191 min_lr: 0.000191 loss: 1.7409 (1.7172) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [389] [260/312] eta: 0:00:37 lr: 0.000191 min_lr: 0.000191 loss: 1.8579 (1.7206) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [389] [270/312] eta: 0:00:30 lr: 0.000191 min_lr: 0.000191 loss: 1.8782 (1.7196) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [389] [280/312] eta: 0:00:23 lr: 0.000191 min_lr: 0.000191 loss: 1.8069 (1.7229) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [389] [290/312] eta: 0:00:15 lr: 0.000190 min_lr: 0.000190 loss: 1.8534 (1.7271) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [389] [300/312] eta: 0:00:08 lr: 0.000190 min_lr: 0.000190 loss: 1.7809 (1.7265) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [389] [310/312] eta: 0:00:01 lr: 0.000190 min_lr: 0.000190 loss: 1.7711 (1.7273) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [389] [311/312] eta: 0:00:00 lr: 0.000190 min_lr: 0.000190 loss: 1.7711 (1.7287) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [389] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.000190 min_lr: 0.000190 loss: 1.7711 (1.7027) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4550 (0.4550) acc1: 88.8021 (88.8021) acc5: 98.6979 (98.6979) time: 4.8340 data: 4.6202 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6500 (0.6525) acc1: 84.1146 (82.6560) acc5: 96.8750 (96.5440) time: 0.6884 data: 0.5134 max mem: 64948 Test: Total time: 0:00:06 (0.7128 s / it) * Acc@1 83.346 Acc@5 96.546 loss 0.634 Accuracy of the model on the 50000 test images: 83.3% Max accuracy: 83.40% Test: [0/9] eta: 0:00:42 loss: 0.4563 (0.4563) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.6691 data: 4.4510 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6247 (0.6438) acc1: 84.3750 (82.2080) acc5: 97.1354 (96.6720) time: 0.6759 data: 0.4947 max mem: 64948 Test: Total time: 0:00:06 (0.6885 s / it) * Acc@1 83.494 Acc@5 96.672 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.5% Epoch: [390] [ 0/312] eta: 0:54:20 lr: 0.000190 min_lr: 0.000190 loss: 1.4173 (1.4173) weight_decay: 0.0500 (0.0500) time: 10.4515 data: 9.4733 max mem: 64948 Epoch: [390] [ 10/312] eta: 0:08:14 lr: 0.000190 min_lr: 0.000190 loss: 1.7777 (1.6804) weight_decay: 0.0500 (0.0500) time: 1.6378 data: 0.8616 max mem: 64948 Epoch: [390] [ 20/312] eta: 0:05:47 lr: 0.000190 min_lr: 0.000190 loss: 1.6748 (1.6238) weight_decay: 0.0500 (0.0500) time: 0.7273 data: 0.0004 max mem: 64948 Epoch: [390] [ 30/312] eta: 0:04:50 lr: 0.000189 min_lr: 0.000189 loss: 1.6509 (1.6675) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [390] [ 40/312] eta: 0:04:17 lr: 0.000189 min_lr: 0.000189 loss: 1.7839 (1.6718) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [390] [ 50/312] eta: 0:03:55 lr: 0.000189 min_lr: 0.000189 loss: 1.7494 (1.6925) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [390] [ 60/312] eta: 0:03:38 lr: 0.000189 min_lr: 0.000189 loss: 1.7743 (1.6993) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [390] [ 70/312] eta: 0:03:23 lr: 0.000189 min_lr: 0.000189 loss: 1.7743 (1.7095) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0005 max mem: 64948 Epoch: [390] [ 80/312] eta: 0:03:11 lr: 0.000188 min_lr: 0.000188 loss: 1.7074 (1.7061) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0005 max mem: 64948 Epoch: [390] [ 90/312] eta: 0:02:59 lr: 0.000188 min_lr: 0.000188 loss: 1.6694 (1.6956) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [390] [100/312] eta: 0:02:49 lr: 0.000188 min_lr: 0.000188 loss: 1.5946 (1.6799) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [390] [110/312] eta: 0:02:39 lr: 0.000188 min_lr: 0.000188 loss: 1.6981 (1.6866) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0003 max mem: 64948 Epoch: [390] [120/312] eta: 0:02:30 lr: 0.000188 min_lr: 0.000188 loss: 1.8046 (1.6901) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [390] [130/312] eta: 0:02:21 lr: 0.000187 min_lr: 0.000187 loss: 1.8339 (1.7014) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [390] [140/312] eta: 0:02:12 lr: 0.000187 min_lr: 0.000187 loss: 1.8782 (1.7036) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [390] [150/312] eta: 0:02:03 lr: 0.000187 min_lr: 0.000187 loss: 1.7274 (1.6964) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [390] [160/312] eta: 0:01:55 lr: 0.000187 min_lr: 0.000187 loss: 1.5790 (1.6994) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [390] [170/312] eta: 0:01:47 lr: 0.000187 min_lr: 0.000187 loss: 1.7322 (1.6999) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [390] [180/312] eta: 0:01:39 lr: 0.000186 min_lr: 0.000186 loss: 1.6539 (1.6944) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [390] [190/312] eta: 0:01:31 lr: 0.000186 min_lr: 0.000186 loss: 1.6033 (1.6929) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [390] [200/312] eta: 0:01:23 lr: 0.000186 min_lr: 0.000186 loss: 1.8261 (1.6944) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [390] [210/312] eta: 0:01:15 lr: 0.000186 min_lr: 0.000186 loss: 1.5834 (1.6858) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [390] [220/312] eta: 0:01:08 lr: 0.000186 min_lr: 0.000186 loss: 1.6973 (1.6915) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [390] [230/312] eta: 0:01:00 lr: 0.000186 min_lr: 0.000186 loss: 1.8025 (1.6938) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [390] [240/312] eta: 0:00:53 lr: 0.000185 min_lr: 0.000185 loss: 1.7420 (1.6922) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [390] [250/312] eta: 0:00:45 lr: 0.000185 min_lr: 0.000185 loss: 1.5571 (1.6914) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [390] [260/312] eta: 0:00:38 lr: 0.000185 min_lr: 0.000185 loss: 1.6959 (1.6916) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [390] [270/312] eta: 0:00:30 lr: 0.000185 min_lr: 0.000185 loss: 1.7570 (1.6998) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [390] [280/312] eta: 0:00:23 lr: 0.000185 min_lr: 0.000185 loss: 1.8427 (1.7029) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [390] [290/312] eta: 0:00:16 lr: 0.000184 min_lr: 0.000184 loss: 1.8138 (1.7044) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [390] [300/312] eta: 0:00:08 lr: 0.000184 min_lr: 0.000184 loss: 1.7448 (1.7045) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [390] [310/312] eta: 0:00:01 lr: 0.000184 min_lr: 0.000184 loss: 1.6874 (1.7056) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [390] [311/312] eta: 0:00:00 lr: 0.000184 min_lr: 0.000184 loss: 1.6870 (1.7034) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [390] Total time: 0:03:48 (0.7308 s / it) Averaged stats: lr: 0.000184 min_lr: 0.000184 loss: 1.6870 (1.7030) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4607 (0.4607) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 4.6869 data: 4.4834 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6433 (0.6584) acc1: 83.8542 (82.4320) acc5: 97.1354 (96.5120) time: 0.6720 data: 0.4982 max mem: 64948 Test: Total time: 0:00:06 (0.6944 s / it) * Acc@1 83.302 Acc@5 96.552 loss 0.638 Accuracy of the model on the 50000 test images: 83.3% Max accuracy: 83.40% Test: [0/9] eta: 0:00:44 loss: 0.4563 (0.4563) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.9124 data: 4.7100 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6246 (0.6436) acc1: 84.3750 (82.2400) acc5: 97.1354 (96.7040) time: 0.6971 data: 0.5234 max mem: 64948 Test: Total time: 0:00:06 (0.7050 s / it) * Acc@1 83.502 Acc@5 96.676 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.50% Epoch: [391] [ 0/312] eta: 0:52:41 lr: 0.000184 min_lr: 0.000184 loss: 0.9939 (0.9939) weight_decay: 0.0500 (0.0500) time: 10.1343 data: 9.3362 max mem: 64948 Epoch: [391] [ 10/312] eta: 0:07:53 lr: 0.000184 min_lr: 0.000184 loss: 1.6234 (1.6484) weight_decay: 0.0500 (0.0500) time: 1.5691 data: 0.8491 max mem: 64948 Epoch: [391] [ 20/312] eta: 0:05:36 lr: 0.000184 min_lr: 0.000184 loss: 1.6954 (1.6827) weight_decay: 0.0500 (0.0500) time: 0.7047 data: 0.0004 max mem: 64948 Epoch: [391] [ 30/312] eta: 0:04:43 lr: 0.000183 min_lr: 0.000183 loss: 1.7188 (1.7308) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [391] [ 40/312] eta: 0:04:13 lr: 0.000183 min_lr: 0.000183 loss: 1.7188 (1.7122) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [391] [ 50/312] eta: 0:03:51 lr: 0.000183 min_lr: 0.000183 loss: 1.6444 (1.6753) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [391] [ 60/312] eta: 0:03:35 lr: 0.000183 min_lr: 0.000183 loss: 1.4496 (1.6348) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [391] [ 70/312] eta: 0:03:21 lr: 0.000183 min_lr: 0.000183 loss: 1.5383 (1.6324) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [391] [ 80/312] eta: 0:03:08 lr: 0.000182 min_lr: 0.000182 loss: 1.6012 (1.6326) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [391] [ 90/312] eta: 0:02:57 lr: 0.000182 min_lr: 0.000182 loss: 1.5927 (1.6329) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [391] [100/312] eta: 0:02:47 lr: 0.000182 min_lr: 0.000182 loss: 1.7235 (1.6412) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [391] [110/312] eta: 0:02:38 lr: 0.000182 min_lr: 0.000182 loss: 1.7687 (1.6540) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [391] [120/312] eta: 0:02:28 lr: 0.000182 min_lr: 0.000182 loss: 1.7217 (1.6454) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [391] [130/312] eta: 0:02:19 lr: 0.000181 min_lr: 0.000181 loss: 1.7782 (1.6555) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [391] [140/312] eta: 0:02:11 lr: 0.000181 min_lr: 0.000181 loss: 1.8279 (1.6632) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [391] [150/312] eta: 0:02:02 lr: 0.000181 min_lr: 0.000181 loss: 1.8042 (1.6645) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [391] [160/312] eta: 0:01:54 lr: 0.000181 min_lr: 0.000181 loss: 1.5949 (1.6640) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [391] [170/312] eta: 0:01:46 lr: 0.000181 min_lr: 0.000181 loss: 1.7225 (1.6768) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [391] [180/312] eta: 0:01:38 lr: 0.000180 min_lr: 0.000180 loss: 1.7225 (1.6704) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [391] [190/312] eta: 0:01:30 lr: 0.000180 min_lr: 0.000180 loss: 1.7658 (1.6778) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [391] [200/312] eta: 0:01:23 lr: 0.000180 min_lr: 0.000180 loss: 1.8321 (1.6792) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [391] [210/312] eta: 0:01:15 lr: 0.000180 min_lr: 0.000180 loss: 1.8230 (1.6854) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [391] [220/312] eta: 0:01:07 lr: 0.000180 min_lr: 0.000180 loss: 1.8505 (1.6905) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [391] [230/312] eta: 0:01:00 lr: 0.000179 min_lr: 0.000179 loss: 1.8511 (1.6943) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [391] [240/312] eta: 0:00:52 lr: 0.000179 min_lr: 0.000179 loss: 1.8063 (1.6974) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [391] [250/312] eta: 0:00:45 lr: 0.000179 min_lr: 0.000179 loss: 1.7759 (1.6934) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [391] [260/312] eta: 0:00:38 lr: 0.000179 min_lr: 0.000179 loss: 1.8228 (1.6994) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [391] [270/312] eta: 0:00:30 lr: 0.000179 min_lr: 0.000179 loss: 1.8228 (1.6992) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [391] [280/312] eta: 0:00:23 lr: 0.000178 min_lr: 0.000178 loss: 1.6545 (1.6963) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [391] [290/312] eta: 0:00:16 lr: 0.000178 min_lr: 0.000178 loss: 1.7250 (1.6921) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0008 max mem: 64948 Epoch: [391] [300/312] eta: 0:00:08 lr: 0.000178 min_lr: 0.000178 loss: 1.3788 (1.6846) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [391] [310/312] eta: 0:00:01 lr: 0.000178 min_lr: 0.000178 loss: 1.6692 (1.6858) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [391] [311/312] eta: 0:00:00 lr: 0.000178 min_lr: 0.000178 loss: 1.7327 (1.6862) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [391] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.000178 min_lr: 0.000178 loss: 1.7327 (1.6965) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4671 (0.4671) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 4.7539 data: 4.5415 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6272 (0.6597) acc1: 83.3333 (82.3680) acc5: 97.1354 (96.7360) time: 0.6795 data: 0.5047 max mem: 64948 Test: Total time: 0:00:06 (0.7038 s / it) * Acc@1 83.356 Acc@5 96.606 loss 0.634 Accuracy of the model on the 50000 test images: 83.4% Max accuracy: 83.40% Test: [0/9] eta: 0:00:45 loss: 0.4562 (0.4562) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 5.0088 data: 4.8012 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6243 (0.6436) acc1: 84.3750 (82.2720) acc5: 97.1354 (96.7040) time: 0.7086 data: 0.5336 max mem: 64948 Test: Total time: 0:00:06 (0.7192 s / it) * Acc@1 83.512 Acc@5 96.686 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.51% Epoch: [392] [ 0/312] eta: 0:47:51 lr: 0.000178 min_lr: 0.000178 loss: 1.8406 (1.8406) weight_decay: 0.0500 (0.0500) time: 9.2045 data: 7.5210 max mem: 64948 Epoch: [392] [ 10/312] eta: 0:07:55 lr: 0.000178 min_lr: 0.000178 loss: 1.8406 (1.7125) weight_decay: 0.0500 (0.0500) time: 1.5736 data: 0.7321 max mem: 64948 Epoch: [392] [ 20/312] eta: 0:05:36 lr: 0.000177 min_lr: 0.000177 loss: 1.7471 (1.6870) weight_decay: 0.0500 (0.0500) time: 0.7515 data: 0.0268 max mem: 64948 Epoch: [392] [ 30/312] eta: 0:04:43 lr: 0.000177 min_lr: 0.000177 loss: 1.7062 (1.6695) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [392] [ 40/312] eta: 0:04:12 lr: 0.000177 min_lr: 0.000177 loss: 1.6274 (1.6635) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [392] [ 50/312] eta: 0:03:51 lr: 0.000177 min_lr: 0.000177 loss: 1.6419 (1.6766) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [392] [ 60/312] eta: 0:03:34 lr: 0.000177 min_lr: 0.000177 loss: 1.7732 (1.6790) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [392] [ 70/312] eta: 0:03:21 lr: 0.000177 min_lr: 0.000177 loss: 1.7732 (1.6805) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [392] [ 80/312] eta: 0:03:08 lr: 0.000176 min_lr: 0.000176 loss: 1.6047 (1.6662) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [392] [ 90/312] eta: 0:02:57 lr: 0.000176 min_lr: 0.000176 loss: 1.5270 (1.6705) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [392] [100/312] eta: 0:02:47 lr: 0.000176 min_lr: 0.000176 loss: 1.7435 (1.6695) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [392] [110/312] eta: 0:02:38 lr: 0.000176 min_lr: 0.000176 loss: 1.6762 (1.6630) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [392] [120/312] eta: 0:02:28 lr: 0.000176 min_lr: 0.000176 loss: 1.6824 (1.6644) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [392] [130/312] eta: 0:02:20 lr: 0.000175 min_lr: 0.000175 loss: 1.7248 (1.6655) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [392] [140/312] eta: 0:02:11 lr: 0.000175 min_lr: 0.000175 loss: 1.7248 (1.6686) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0003 max mem: 64948 Epoch: [392] [150/312] eta: 0:02:03 lr: 0.000175 min_lr: 0.000175 loss: 1.7332 (1.6632) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [392] [160/312] eta: 0:01:54 lr: 0.000175 min_lr: 0.000175 loss: 1.5867 (1.6586) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [392] [170/312] eta: 0:01:46 lr: 0.000175 min_lr: 0.000175 loss: 1.7453 (1.6649) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [392] [180/312] eta: 0:01:38 lr: 0.000174 min_lr: 0.000174 loss: 1.8065 (1.6711) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [392] [190/312] eta: 0:01:31 lr: 0.000174 min_lr: 0.000174 loss: 1.7635 (1.6713) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [392] [200/312] eta: 0:01:23 lr: 0.000174 min_lr: 0.000174 loss: 1.5470 (1.6617) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [392] [210/312] eta: 0:01:15 lr: 0.000174 min_lr: 0.000174 loss: 1.5470 (1.6653) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [392] [220/312] eta: 0:01:07 lr: 0.000174 min_lr: 0.000174 loss: 1.7707 (1.6725) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [392] [230/312] eta: 0:01:00 lr: 0.000173 min_lr: 0.000173 loss: 1.8278 (1.6787) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [392] [240/312] eta: 0:00:52 lr: 0.000173 min_lr: 0.000173 loss: 1.8278 (1.6814) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [392] [250/312] eta: 0:00:45 lr: 0.000173 min_lr: 0.000173 loss: 1.7527 (1.6812) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [392] [260/312] eta: 0:00:38 lr: 0.000173 min_lr: 0.000173 loss: 1.7211 (1.6828) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [392] [270/312] eta: 0:00:30 lr: 0.000173 min_lr: 0.000173 loss: 1.7232 (1.6843) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [392] [280/312] eta: 0:00:23 lr: 0.000172 min_lr: 0.000172 loss: 1.6647 (1.6790) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0009 max mem: 64948 Epoch: [392] [290/312] eta: 0:00:16 lr: 0.000172 min_lr: 0.000172 loss: 1.5626 (1.6761) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [392] [300/312] eta: 0:00:08 lr: 0.000172 min_lr: 0.000172 loss: 1.6677 (1.6752) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [392] [310/312] eta: 0:00:01 lr: 0.000172 min_lr: 0.000172 loss: 1.5282 (1.6714) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [392] [311/312] eta: 0:00:00 lr: 0.000172 min_lr: 0.000172 loss: 1.6916 (1.6720) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [392] Total time: 0:03:47 (0.7285 s / it) Averaged stats: lr: 0.000172 min_lr: 0.000172 loss: 1.6916 (1.7006) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:38 loss: 0.4491 (0.4491) acc1: 88.0208 (88.0208) acc5: 98.4375 (98.4375) time: 4.3109 data: 4.0911 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6320 (0.6501) acc1: 83.0729 (82.4320) acc5: 97.1354 (96.7360) time: 0.6308 data: 0.4547 max mem: 64948 Test: Total time: 0:00:05 (0.6521 s / it) * Acc@1 83.418 Acc@5 96.550 loss 0.634 Accuracy of the model on the 50000 test images: 83.4% Max accuracy: 83.42% Test: [0/9] eta: 0:00:42 loss: 0.4562 (0.4562) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.7662 data: 4.5484 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6241 (0.6436) acc1: 84.3750 (82.2720) acc5: 97.1354 (96.7040) time: 0.6814 data: 0.5055 max mem: 64948 Test: Total time: 0:00:06 (0.6898 s / it) * Acc@1 83.530 Acc@5 96.686 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.53% Epoch: [393] [ 0/312] eta: 0:48:56 lr: 0.000172 min_lr: 0.000172 loss: 1.9213 (1.9213) weight_decay: 0.0500 (0.0500) time: 9.4132 data: 8.6006 max mem: 64948 Epoch: [393] [ 10/312] eta: 0:07:42 lr: 0.000172 min_lr: 0.000172 loss: 1.8924 (1.7882) weight_decay: 0.0500 (0.0500) time: 1.5311 data: 0.7823 max mem: 64948 Epoch: [393] [ 20/312] eta: 0:05:31 lr: 0.000172 min_lr: 0.000172 loss: 1.8611 (1.8056) weight_decay: 0.0500 (0.0500) time: 0.7219 data: 0.0004 max mem: 64948 Epoch: [393] [ 30/312] eta: 0:04:40 lr: 0.000171 min_lr: 0.000171 loss: 1.7649 (1.7727) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0003 max mem: 64948 Epoch: [393] [ 40/312] eta: 0:04:10 lr: 0.000171 min_lr: 0.000171 loss: 1.6416 (1.7445) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [393] [ 50/312] eta: 0:03:49 lr: 0.000171 min_lr: 0.000171 loss: 1.6674 (1.7305) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [393] [ 60/312] eta: 0:03:33 lr: 0.000171 min_lr: 0.000171 loss: 1.6821 (1.7119) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [393] [ 70/312] eta: 0:03:20 lr: 0.000171 min_lr: 0.000171 loss: 1.7310 (1.7177) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0004 max mem: 64948 Epoch: [393] [ 80/312] eta: 0:03:08 lr: 0.000170 min_lr: 0.000170 loss: 1.6987 (1.7135) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [393] [ 90/312] eta: 0:02:57 lr: 0.000170 min_lr: 0.000170 loss: 1.6987 (1.7103) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [393] [100/312] eta: 0:02:47 lr: 0.000170 min_lr: 0.000170 loss: 1.6176 (1.6989) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [393] [110/312] eta: 0:02:37 lr: 0.000170 min_lr: 0.000170 loss: 1.6878 (1.7053) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [393] [120/312] eta: 0:02:28 lr: 0.000170 min_lr: 0.000170 loss: 1.7506 (1.7015) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [393] [130/312] eta: 0:02:19 lr: 0.000169 min_lr: 0.000169 loss: 1.6642 (1.6982) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [393] [140/312] eta: 0:02:11 lr: 0.000169 min_lr: 0.000169 loss: 1.6642 (1.6948) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [393] [150/312] eta: 0:02:02 lr: 0.000169 min_lr: 0.000169 loss: 1.6742 (1.6952) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [393] [160/312] eta: 0:01:54 lr: 0.000169 min_lr: 0.000169 loss: 1.7332 (1.6928) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [393] [170/312] eta: 0:01:46 lr: 0.000169 min_lr: 0.000169 loss: 1.8032 (1.6989) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [393] [180/312] eta: 0:01:38 lr: 0.000168 min_lr: 0.000168 loss: 1.8032 (1.6971) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [393] [190/312] eta: 0:01:30 lr: 0.000168 min_lr: 0.000168 loss: 1.7509 (1.6997) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [393] [200/312] eta: 0:01:23 lr: 0.000168 min_lr: 0.000168 loss: 1.8258 (1.7051) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [393] [210/312] eta: 0:01:15 lr: 0.000168 min_lr: 0.000168 loss: 1.8482 (1.7097) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [393] [220/312] eta: 0:01:07 lr: 0.000168 min_lr: 0.000168 loss: 1.7517 (1.7089) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [393] [230/312] eta: 0:01:00 lr: 0.000168 min_lr: 0.000168 loss: 1.7517 (1.7133) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [393] [240/312] eta: 0:00:52 lr: 0.000167 min_lr: 0.000167 loss: 1.6896 (1.7055) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [393] [250/312] eta: 0:00:45 lr: 0.000167 min_lr: 0.000167 loss: 1.5234 (1.7023) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [393] [260/312] eta: 0:00:38 lr: 0.000167 min_lr: 0.000167 loss: 1.6301 (1.7002) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [393] [270/312] eta: 0:00:30 lr: 0.000167 min_lr: 0.000167 loss: 1.7765 (1.7044) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [393] [280/312] eta: 0:00:23 lr: 0.000167 min_lr: 0.000167 loss: 1.8165 (1.7075) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [393] [290/312] eta: 0:00:16 lr: 0.000166 min_lr: 0.000166 loss: 1.6998 (1.7051) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [393] [300/312] eta: 0:00:08 lr: 0.000166 min_lr: 0.000166 loss: 1.6805 (1.7053) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [393] [310/312] eta: 0:00:01 lr: 0.000166 min_lr: 0.000166 loss: 1.6248 (1.7038) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [393] [311/312] eta: 0:00:00 lr: 0.000166 min_lr: 0.000166 loss: 1.6577 (1.7044) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [393] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.000166 min_lr: 0.000166 loss: 1.6577 (1.6961) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4610 (0.4610) acc1: 87.2396 (87.2396) acc5: 98.9583 (98.9583) time: 4.4896 data: 4.2708 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6277 (0.6436) acc1: 83.3333 (82.3040) acc5: 97.1354 (96.8960) time: 0.6502 data: 0.4746 max mem: 64948 Test: Total time: 0:00:06 (0.6723 s / it) * Acc@1 83.344 Acc@5 96.558 loss 0.632 Accuracy of the model on the 50000 test images: 83.3% Max accuracy: 83.42% Test: [0/9] eta: 0:00:44 loss: 0.4562 (0.4562) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.9787 data: 4.7738 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6238 (0.6434) acc1: 84.3750 (82.3040) acc5: 97.1354 (96.7040) time: 0.7141 data: 0.5402 max mem: 64948 Test: Total time: 0:00:06 (0.7294 s / it) * Acc@1 83.536 Acc@5 96.686 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.54% Epoch: [394] [ 0/312] eta: 0:44:46 lr: 0.000166 min_lr: 0.000166 loss: 1.7767 (1.7767) weight_decay: 0.0500 (0.0500) time: 8.6109 data: 7.6981 max mem: 64948 Epoch: [394] [ 10/312] eta: 0:07:47 lr: 0.000166 min_lr: 0.000166 loss: 1.8322 (1.6762) weight_decay: 0.0500 (0.0500) time: 1.5472 data: 0.8092 max mem: 64948 Epoch: [394] [ 20/312] eta: 0:05:33 lr: 0.000166 min_lr: 0.000166 loss: 1.8161 (1.7351) weight_decay: 0.0500 (0.0500) time: 0.7704 data: 0.0603 max mem: 64948 Epoch: [394] [ 30/312] eta: 0:04:41 lr: 0.000165 min_lr: 0.000165 loss: 1.7597 (1.7263) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [394] [ 40/312] eta: 0:04:11 lr: 0.000165 min_lr: 0.000165 loss: 1.7129 (1.6940) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [394] [ 50/312] eta: 0:03:50 lr: 0.000165 min_lr: 0.000165 loss: 1.7909 (1.7136) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [394] [ 60/312] eta: 0:03:34 lr: 0.000165 min_lr: 0.000165 loss: 1.7691 (1.7055) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [394] [ 70/312] eta: 0:03:20 lr: 0.000165 min_lr: 0.000165 loss: 1.7655 (1.7199) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0003 max mem: 64948 Epoch: [394] [ 80/312] eta: 0:03:08 lr: 0.000165 min_lr: 0.000165 loss: 1.9628 (1.7420) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [394] [ 90/312] eta: 0:02:57 lr: 0.000164 min_lr: 0.000164 loss: 1.9280 (1.7403) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [394] [100/312] eta: 0:02:47 lr: 0.000164 min_lr: 0.000164 loss: 1.7591 (1.7306) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [394] [110/312] eta: 0:02:37 lr: 0.000164 min_lr: 0.000164 loss: 1.7505 (1.7378) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [394] [120/312] eta: 0:02:28 lr: 0.000164 min_lr: 0.000164 loss: 1.8055 (1.7405) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [394] [130/312] eta: 0:02:19 lr: 0.000164 min_lr: 0.000164 loss: 1.7594 (1.7362) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [394] [140/312] eta: 0:02:11 lr: 0.000163 min_lr: 0.000163 loss: 1.6353 (1.7216) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [394] [150/312] eta: 0:02:02 lr: 0.000163 min_lr: 0.000163 loss: 1.6353 (1.7189) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [394] [160/312] eta: 0:01:54 lr: 0.000163 min_lr: 0.000163 loss: 1.5344 (1.7043) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [394] [170/312] eta: 0:01:46 lr: 0.000163 min_lr: 0.000163 loss: 1.6381 (1.7050) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [394] [180/312] eta: 0:01:38 lr: 0.000163 min_lr: 0.000163 loss: 1.7228 (1.7034) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [394] [190/312] eta: 0:01:30 lr: 0.000163 min_lr: 0.000163 loss: 1.6115 (1.7044) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [394] [200/312] eta: 0:01:23 lr: 0.000162 min_lr: 0.000162 loss: 1.6756 (1.7038) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [394] [210/312] eta: 0:01:15 lr: 0.000162 min_lr: 0.000162 loss: 1.6263 (1.7000) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [394] [220/312] eta: 0:01:07 lr: 0.000162 min_lr: 0.000162 loss: 1.5571 (1.6922) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [394] [230/312] eta: 0:01:00 lr: 0.000162 min_lr: 0.000162 loss: 1.6487 (1.6915) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [394] [240/312] eta: 0:00:52 lr: 0.000162 min_lr: 0.000162 loss: 1.7646 (1.6950) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [394] [250/312] eta: 0:00:45 lr: 0.000161 min_lr: 0.000161 loss: 1.7646 (1.6907) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [394] [260/312] eta: 0:00:38 lr: 0.000161 min_lr: 0.000161 loss: 1.7575 (1.6901) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [394] [270/312] eta: 0:00:30 lr: 0.000161 min_lr: 0.000161 loss: 1.8676 (1.6956) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [394] [280/312] eta: 0:00:23 lr: 0.000161 min_lr: 0.000161 loss: 1.6993 (1.6887) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0010 max mem: 64948 Epoch: [394] [290/312] eta: 0:00:16 lr: 0.000161 min_lr: 0.000161 loss: 1.5475 (1.6856) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [394] [300/312] eta: 0:00:08 lr: 0.000160 min_lr: 0.000160 loss: 1.6702 (1.6849) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [394] [310/312] eta: 0:00:01 lr: 0.000160 min_lr: 0.000160 loss: 1.7105 (1.6831) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [394] [311/312] eta: 0:00:00 lr: 0.000160 min_lr: 0.000160 loss: 1.7105 (1.6828) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [394] Total time: 0:03:47 (0.7284 s / it) Averaged stats: lr: 0.000160 min_lr: 0.000160 loss: 1.7105 (1.6961) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4642 (0.4642) acc1: 89.0625 (89.0625) acc5: 97.9167 (97.9167) time: 4.5098 data: 4.2987 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6233 (0.6533) acc1: 83.3333 (82.7520) acc5: 97.6562 (97.0240) time: 0.6524 data: 0.4777 max mem: 64948 Test: Total time: 0:00:06 (0.6776 s / it) * Acc@1 83.478 Acc@5 96.674 loss 0.630 Accuracy of the model on the 50000 test images: 83.5% Max accuracy: 83.48% Test: [0/9] eta: 0:00:40 loss: 0.4561 (0.4561) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.4762 data: 4.2607 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6235 (0.6432) acc1: 84.3750 (82.3360) acc5: 97.1354 (96.6720) time: 0.6487 data: 0.4735 max mem: 64948 Test: Total time: 0:00:05 (0.6558 s / it) * Acc@1 83.544 Acc@5 96.688 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.5% Max EMA accuracy: 83.54% Epoch: [395] [ 0/312] eta: 0:47:15 lr: 0.000160 min_lr: 0.000160 loss: 1.3438 (1.3438) weight_decay: 0.0500 (0.0500) time: 9.0882 data: 7.6216 max mem: 64948 Epoch: [395] [ 10/312] eta: 0:07:39 lr: 0.000160 min_lr: 0.000160 loss: 1.7478 (1.6774) weight_decay: 0.0500 (0.0500) time: 1.5220 data: 0.6933 max mem: 64948 Epoch: [395] [ 20/312] eta: 0:05:29 lr: 0.000160 min_lr: 0.000160 loss: 1.8708 (1.7464) weight_decay: 0.0500 (0.0500) time: 0.7301 data: 0.0004 max mem: 64948 Epoch: [395] [ 30/312] eta: 0:04:39 lr: 0.000160 min_lr: 0.000160 loss: 1.8590 (1.7076) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [395] [ 40/312] eta: 0:04:09 lr: 0.000160 min_lr: 0.000160 loss: 1.7375 (1.7160) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [395] [ 50/312] eta: 0:03:49 lr: 0.000159 min_lr: 0.000159 loss: 1.7738 (1.7243) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [395] [ 60/312] eta: 0:03:33 lr: 0.000159 min_lr: 0.000159 loss: 1.7841 (1.7108) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [395] [ 70/312] eta: 0:03:19 lr: 0.000159 min_lr: 0.000159 loss: 1.6004 (1.6999) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [395] [ 80/312] eta: 0:03:07 lr: 0.000159 min_lr: 0.000159 loss: 1.6724 (1.6979) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [395] [ 90/312] eta: 0:02:56 lr: 0.000159 min_lr: 0.000159 loss: 1.8438 (1.7038) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [395] [100/312] eta: 0:02:46 lr: 0.000158 min_lr: 0.000158 loss: 1.8438 (1.7020) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [395] [110/312] eta: 0:02:37 lr: 0.000158 min_lr: 0.000158 loss: 1.7063 (1.6969) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [395] [120/312] eta: 0:02:28 lr: 0.000158 min_lr: 0.000158 loss: 1.6739 (1.6867) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [395] [130/312] eta: 0:02:19 lr: 0.000158 min_lr: 0.000158 loss: 1.5893 (1.6848) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [395] [140/312] eta: 0:02:10 lr: 0.000158 min_lr: 0.000158 loss: 1.6977 (1.6904) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [395] [150/312] eta: 0:02:02 lr: 0.000158 min_lr: 0.000158 loss: 1.7826 (1.6967) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [395] [160/312] eta: 0:01:54 lr: 0.000157 min_lr: 0.000157 loss: 1.7396 (1.6910) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [395] [170/312] eta: 0:01:46 lr: 0.000157 min_lr: 0.000157 loss: 1.5392 (1.6843) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [395] [180/312] eta: 0:01:38 lr: 0.000157 min_lr: 0.000157 loss: 1.4721 (1.6754) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [395] [190/312] eta: 0:01:30 lr: 0.000157 min_lr: 0.000157 loss: 1.3815 (1.6708) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [395] [200/312] eta: 0:01:22 lr: 0.000157 min_lr: 0.000157 loss: 1.8103 (1.6669) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [395] [210/312] eta: 0:01:15 lr: 0.000156 min_lr: 0.000156 loss: 1.7501 (1.6673) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [395] [220/312] eta: 0:01:07 lr: 0.000156 min_lr: 0.000156 loss: 1.8067 (1.6748) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [395] [230/312] eta: 0:01:00 lr: 0.000156 min_lr: 0.000156 loss: 1.7319 (1.6732) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [395] [240/312] eta: 0:00:52 lr: 0.000156 min_lr: 0.000156 loss: 1.7319 (1.6793) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [395] [250/312] eta: 0:00:45 lr: 0.000156 min_lr: 0.000156 loss: 1.8403 (1.6816) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [395] [260/312] eta: 0:00:37 lr: 0.000156 min_lr: 0.000156 loss: 1.8008 (1.6806) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [395] [270/312] eta: 0:00:30 lr: 0.000155 min_lr: 0.000155 loss: 1.8921 (1.6859) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [395] [280/312] eta: 0:00:23 lr: 0.000155 min_lr: 0.000155 loss: 1.8921 (1.6866) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [395] [290/312] eta: 0:00:15 lr: 0.000155 min_lr: 0.000155 loss: 1.7938 (1.6907) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0008 max mem: 64948 Epoch: [395] [300/312] eta: 0:00:08 lr: 0.000155 min_lr: 0.000155 loss: 1.7595 (1.6896) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [395] [310/312] eta: 0:00:01 lr: 0.000155 min_lr: 0.000155 loss: 1.8343 (1.6965) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [395] [311/312] eta: 0:00:00 lr: 0.000155 min_lr: 0.000155 loss: 1.8581 (1.6972) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [395] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.000155 min_lr: 0.000155 loss: 1.8581 (1.6903) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4459 (0.4459) acc1: 88.8021 (88.8021) acc5: 98.4375 (98.4375) time: 4.4330 data: 4.2117 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6212 (0.6446) acc1: 84.3750 (82.9440) acc5: 97.3958 (96.7680) time: 0.6461 data: 0.4702 max mem: 64948 Test: Total time: 0:00:06 (0.6686 s / it) * Acc@1 83.490 Acc@5 96.622 loss 0.628 Accuracy of the model on the 50000 test images: 83.5% Max accuracy: 83.49% Test: [0/9] eta: 0:00:41 loss: 0.4562 (0.4562) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.6502 data: 4.4372 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6230 (0.6431) acc1: 84.3750 (82.4000) acc5: 97.1354 (96.6720) time: 0.6686 data: 0.4931 max mem: 64948 Test: Total time: 0:00:06 (0.6759 s / it) * Acc@1 83.570 Acc@5 96.686 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.57% Epoch: [396] [ 0/312] eta: 0:50:46 lr: 0.000155 min_lr: 0.000155 loss: 2.0687 (2.0687) weight_decay: 0.0500 (0.0500) time: 9.7630 data: 8.9448 max mem: 64948 Epoch: [396] [ 10/312] eta: 0:07:44 lr: 0.000154 min_lr: 0.000154 loss: 1.6723 (1.6182) weight_decay: 0.0500 (0.0500) time: 1.5380 data: 0.8136 max mem: 64948 Epoch: [396] [ 20/312] eta: 0:05:32 lr: 0.000154 min_lr: 0.000154 loss: 1.7012 (1.6792) weight_decay: 0.0500 (0.0500) time: 0.7062 data: 0.0004 max mem: 64948 Epoch: [396] [ 30/312] eta: 0:04:41 lr: 0.000154 min_lr: 0.000154 loss: 1.7186 (1.6515) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [396] [ 40/312] eta: 0:04:11 lr: 0.000154 min_lr: 0.000154 loss: 1.6762 (1.6735) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [396] [ 50/312] eta: 0:03:50 lr: 0.000154 min_lr: 0.000154 loss: 1.7664 (1.6844) weight_decay: 0.0500 (0.0500) time: 0.6990 data: 0.0004 max mem: 64948 Epoch: [396] [ 60/312] eta: 0:03:34 lr: 0.000154 min_lr: 0.000154 loss: 1.6725 (1.6780) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [396] [ 70/312] eta: 0:03:20 lr: 0.000153 min_lr: 0.000153 loss: 1.6464 (1.6767) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [396] [ 80/312] eta: 0:03:08 lr: 0.000153 min_lr: 0.000153 loss: 1.5489 (1.6668) weight_decay: 0.0500 (0.0500) time: 0.6996 data: 0.0004 max mem: 64948 Epoch: [396] [ 90/312] eta: 0:02:57 lr: 0.000153 min_lr: 0.000153 loss: 1.7869 (1.6880) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [396] [100/312] eta: 0:02:47 lr: 0.000153 min_lr: 0.000153 loss: 1.8456 (1.6962) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [396] [110/312] eta: 0:02:37 lr: 0.000153 min_lr: 0.000153 loss: 1.7652 (1.6886) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [396] [120/312] eta: 0:02:28 lr: 0.000152 min_lr: 0.000152 loss: 1.6743 (1.6912) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [396] [130/312] eta: 0:02:19 lr: 0.000152 min_lr: 0.000152 loss: 1.7429 (1.6956) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [396] [140/312] eta: 0:02:11 lr: 0.000152 min_lr: 0.000152 loss: 1.6943 (1.6889) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [396] [150/312] eta: 0:02:02 lr: 0.000152 min_lr: 0.000152 loss: 1.6065 (1.6890) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [396] [160/312] eta: 0:01:54 lr: 0.000152 min_lr: 0.000152 loss: 1.7396 (1.6917) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [396] [170/312] eta: 0:01:46 lr: 0.000152 min_lr: 0.000152 loss: 1.7768 (1.6926) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [396] [180/312] eta: 0:01:38 lr: 0.000151 min_lr: 0.000151 loss: 1.7159 (1.6887) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [396] [190/312] eta: 0:01:30 lr: 0.000151 min_lr: 0.000151 loss: 1.6447 (1.6830) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [396] [200/312] eta: 0:01:23 lr: 0.000151 min_lr: 0.000151 loss: 1.7351 (1.6848) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [396] [210/312] eta: 0:01:15 lr: 0.000151 min_lr: 0.000151 loss: 1.7243 (1.6799) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [396] [220/312] eta: 0:01:07 lr: 0.000151 min_lr: 0.000151 loss: 1.7759 (1.6898) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [396] [230/312] eta: 0:01:00 lr: 0.000150 min_lr: 0.000150 loss: 1.8348 (1.6891) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [396] [240/312] eta: 0:00:52 lr: 0.000150 min_lr: 0.000150 loss: 1.6618 (1.6890) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [396] [250/312] eta: 0:00:45 lr: 0.000150 min_lr: 0.000150 loss: 1.6618 (1.6858) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [396] [260/312] eta: 0:00:38 lr: 0.000150 min_lr: 0.000150 loss: 1.6148 (1.6795) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [396] [270/312] eta: 0:00:30 lr: 0.000150 min_lr: 0.000150 loss: 1.6433 (1.6817) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [396] [280/312] eta: 0:00:23 lr: 0.000150 min_lr: 0.000150 loss: 1.6707 (1.6783) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0009 max mem: 64948 Epoch: [396] [290/312] eta: 0:00:16 lr: 0.000149 min_lr: 0.000149 loss: 1.5793 (1.6793) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [396] [300/312] eta: 0:00:08 lr: 0.000149 min_lr: 0.000149 loss: 1.7293 (1.6813) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [396] [310/312] eta: 0:00:01 lr: 0.000149 min_lr: 0.000149 loss: 1.6563 (1.6803) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [396] [311/312] eta: 0:00:00 lr: 0.000149 min_lr: 0.000149 loss: 1.6563 (1.6808) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [396] Total time: 0:03:47 (0.7281 s / it) Averaged stats: lr: 0.000149 min_lr: 0.000149 loss: 1.6563 (1.6918) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4445 (0.4445) acc1: 89.3229 (89.3229) acc5: 97.9167 (97.9167) time: 4.5715 data: 4.3583 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6098 (0.6481) acc1: 84.8958 (83.2000) acc5: 97.1354 (96.5440) time: 0.6593 data: 0.4844 max mem: 64948 Test: Total time: 0:00:06 (0.6793 s / it) * Acc@1 83.444 Acc@5 96.590 loss 0.629 Accuracy of the model on the 50000 test images: 83.4% Max accuracy: 83.49% Test: [0/9] eta: 0:00:41 loss: 0.4562 (0.4562) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.5625 data: 4.3545 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6225 (0.6429) acc1: 84.3750 (82.3680) acc5: 96.8750 (96.6080) time: 0.6773 data: 0.5030 max mem: 64948 Test: Total time: 0:00:06 (0.6854 s / it) * Acc@1 83.572 Acc@5 96.672 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.57% Epoch: [397] [ 0/312] eta: 0:48:27 lr: 0.000149 min_lr: 0.000149 loss: 1.9123 (1.9123) weight_decay: 0.0500 (0.0500) time: 9.3198 data: 8.0308 max mem: 64948 Epoch: [397] [ 10/312] eta: 0:07:39 lr: 0.000149 min_lr: 0.000149 loss: 1.9039 (1.8313) weight_decay: 0.0500 (0.0500) time: 1.5201 data: 0.7400 max mem: 64948 Epoch: [397] [ 20/312] eta: 0:05:29 lr: 0.000149 min_lr: 0.000149 loss: 1.7965 (1.7695) weight_decay: 0.0500 (0.0500) time: 0.7198 data: 0.0056 max mem: 64948 Epoch: [397] [ 30/312] eta: 0:04:39 lr: 0.000149 min_lr: 0.000149 loss: 1.7570 (1.7475) weight_decay: 0.0500 (0.0500) time: 0.7008 data: 0.0004 max mem: 64948 Epoch: [397] [ 40/312] eta: 0:04:09 lr: 0.000148 min_lr: 0.000148 loss: 1.7834 (1.7589) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [397] [ 50/312] eta: 0:03:49 lr: 0.000148 min_lr: 0.000148 loss: 1.7834 (1.7510) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [397] [ 60/312] eta: 0:03:33 lr: 0.000148 min_lr: 0.000148 loss: 1.6897 (1.7357) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [397] [ 70/312] eta: 0:03:19 lr: 0.000148 min_lr: 0.000148 loss: 1.7621 (1.7453) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [397] [ 80/312] eta: 0:03:07 lr: 0.000148 min_lr: 0.000148 loss: 1.6914 (1.7133) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [397] [ 90/312] eta: 0:02:56 lr: 0.000147 min_lr: 0.000147 loss: 1.4191 (1.6974) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [397] [100/312] eta: 0:02:46 lr: 0.000147 min_lr: 0.000147 loss: 1.6357 (1.6890) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [397] [110/312] eta: 0:02:37 lr: 0.000147 min_lr: 0.000147 loss: 1.4341 (1.6759) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [397] [120/312] eta: 0:02:27 lr: 0.000147 min_lr: 0.000147 loss: 1.4381 (1.6601) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [397] [130/312] eta: 0:02:19 lr: 0.000147 min_lr: 0.000147 loss: 1.4401 (1.6503) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [397] [140/312] eta: 0:02:10 lr: 0.000147 min_lr: 0.000147 loss: 1.7263 (1.6638) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [397] [150/312] eta: 0:02:02 lr: 0.000146 min_lr: 0.000146 loss: 1.8031 (1.6726) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [397] [160/312] eta: 0:01:54 lr: 0.000146 min_lr: 0.000146 loss: 1.7057 (1.6653) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0003 max mem: 64948 Epoch: [397] [170/312] eta: 0:01:46 lr: 0.000146 min_lr: 0.000146 loss: 1.7371 (1.6752) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [397] [180/312] eta: 0:01:38 lr: 0.000146 min_lr: 0.000146 loss: 1.7800 (1.6747) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [397] [190/312] eta: 0:01:30 lr: 0.000146 min_lr: 0.000146 loss: 1.6933 (1.6760) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [397] [200/312] eta: 0:01:22 lr: 0.000146 min_lr: 0.000146 loss: 1.6904 (1.6731) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [397] [210/312] eta: 0:01:15 lr: 0.000145 min_lr: 0.000145 loss: 1.6001 (1.6667) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [397] [220/312] eta: 0:01:07 lr: 0.000145 min_lr: 0.000145 loss: 1.6385 (1.6708) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [397] [230/312] eta: 0:01:00 lr: 0.000145 min_lr: 0.000145 loss: 1.8038 (1.6732) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [397] [240/312] eta: 0:00:52 lr: 0.000145 min_lr: 0.000145 loss: 1.7654 (1.6752) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [397] [250/312] eta: 0:00:45 lr: 0.000145 min_lr: 0.000145 loss: 1.7654 (1.6750) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [397] [260/312] eta: 0:00:37 lr: 0.000144 min_lr: 0.000144 loss: 1.8149 (1.6727) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [397] [270/312] eta: 0:00:30 lr: 0.000144 min_lr: 0.000144 loss: 1.6772 (1.6709) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [397] [280/312] eta: 0:00:23 lr: 0.000144 min_lr: 0.000144 loss: 1.5794 (1.6685) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [397] [290/312] eta: 0:00:15 lr: 0.000144 min_lr: 0.000144 loss: 1.5215 (1.6670) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0008 max mem: 64948 Epoch: [397] [300/312] eta: 0:00:08 lr: 0.000144 min_lr: 0.000144 loss: 1.7324 (1.6698) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [397] [310/312] eta: 0:00:01 lr: 0.000144 min_lr: 0.000144 loss: 1.7324 (1.6707) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [397] [311/312] eta: 0:00:00 lr: 0.000144 min_lr: 0.000144 loss: 1.7324 (1.6706) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [397] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.000144 min_lr: 0.000144 loss: 1.7324 (1.6842) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4461 (0.4461) acc1: 88.0208 (88.0208) acc5: 98.6979 (98.6979) time: 4.6407 data: 4.4336 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6076 (0.6464) acc1: 84.6354 (82.6560) acc5: 97.1354 (96.7040) time: 0.6670 data: 0.4927 max mem: 64948 Test: Total time: 0:00:06 (0.6903 s / it) * Acc@1 83.420 Acc@5 96.634 loss 0.627 Accuracy of the model on the 50000 test images: 83.4% Max accuracy: 83.49% Test: [0/9] eta: 0:00:43 loss: 0.4561 (0.4561) acc1: 87.2396 (87.2396) acc5: 97.9167 (97.9167) time: 4.8425 data: 4.6243 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6218 (0.6427) acc1: 84.3750 (82.4000) acc5: 97.1354 (96.7040) time: 0.6894 data: 0.5139 max mem: 64948 Test: Total time: 0:00:06 (0.7011 s / it) * Acc@1 83.566 Acc@5 96.680 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [398] [ 0/312] eta: 0:55:28 lr: 0.000144 min_lr: 0.000144 loss: 2.2584 (2.2584) weight_decay: 0.0500 (0.0500) time: 10.6677 data: 7.9758 max mem: 64948 Epoch: [398] [ 10/312] eta: 0:08:11 lr: 0.000143 min_lr: 0.000143 loss: 1.8523 (1.8770) weight_decay: 0.0500 (0.0500) time: 1.6259 data: 0.7255 max mem: 64948 Epoch: [398] [ 20/312] eta: 0:05:46 lr: 0.000143 min_lr: 0.000143 loss: 1.8018 (1.8077) weight_decay: 0.0500 (0.0500) time: 0.7118 data: 0.0004 max mem: 64948 Epoch: [398] [ 30/312] eta: 0:04:50 lr: 0.000143 min_lr: 0.000143 loss: 1.7004 (1.7420) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [398] [ 40/312] eta: 0:04:17 lr: 0.000143 min_lr: 0.000143 loss: 1.7006 (1.7404) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [398] [ 50/312] eta: 0:03:55 lr: 0.000143 min_lr: 0.000143 loss: 1.8135 (1.7610) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [398] [ 60/312] eta: 0:03:38 lr: 0.000143 min_lr: 0.000143 loss: 1.8553 (1.7661) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [398] [ 70/312] eta: 0:03:23 lr: 0.000142 min_lr: 0.000142 loss: 1.8105 (1.7682) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [398] [ 80/312] eta: 0:03:11 lr: 0.000142 min_lr: 0.000142 loss: 1.8105 (1.7532) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [398] [ 90/312] eta: 0:02:59 lr: 0.000142 min_lr: 0.000142 loss: 1.6435 (1.7412) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [398] [100/312] eta: 0:02:49 lr: 0.000142 min_lr: 0.000142 loss: 1.6637 (1.7379) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [398] [110/312] eta: 0:02:39 lr: 0.000142 min_lr: 0.000142 loss: 1.7505 (1.7298) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [398] [120/312] eta: 0:02:29 lr: 0.000141 min_lr: 0.000141 loss: 1.8232 (1.7419) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [398] [130/312] eta: 0:02:20 lr: 0.000141 min_lr: 0.000141 loss: 1.8590 (1.7485) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [398] [140/312] eta: 0:02:12 lr: 0.000141 min_lr: 0.000141 loss: 1.7520 (1.7384) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [398] [150/312] eta: 0:02:03 lr: 0.000141 min_lr: 0.000141 loss: 1.5932 (1.7284) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [398] [160/312] eta: 0:01:55 lr: 0.000141 min_lr: 0.000141 loss: 1.6229 (1.7283) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [398] [170/312] eta: 0:01:47 lr: 0.000141 min_lr: 0.000141 loss: 1.7720 (1.7244) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [398] [180/312] eta: 0:01:39 lr: 0.000140 min_lr: 0.000140 loss: 1.6093 (1.7129) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [398] [190/312] eta: 0:01:31 lr: 0.000140 min_lr: 0.000140 loss: 1.6249 (1.7151) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [398] [200/312] eta: 0:01:23 lr: 0.000140 min_lr: 0.000140 loss: 1.7213 (1.6985) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [398] [210/312] eta: 0:01:15 lr: 0.000140 min_lr: 0.000140 loss: 1.3143 (1.6928) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [398] [220/312] eta: 0:01:08 lr: 0.000140 min_lr: 0.000140 loss: 1.8486 (1.6976) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [398] [230/312] eta: 0:01:00 lr: 0.000140 min_lr: 0.000140 loss: 1.8296 (1.6971) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [398] [240/312] eta: 0:00:53 lr: 0.000139 min_lr: 0.000139 loss: 1.7204 (1.6945) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [398] [250/312] eta: 0:00:45 lr: 0.000139 min_lr: 0.000139 loss: 1.8123 (1.6988) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [398] [260/312] eta: 0:00:38 lr: 0.000139 min_lr: 0.000139 loss: 1.8123 (1.6954) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [398] [270/312] eta: 0:00:30 lr: 0.000139 min_lr: 0.000139 loss: 1.7604 (1.6973) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [398] [280/312] eta: 0:00:23 lr: 0.000139 min_lr: 0.000139 loss: 1.7241 (1.6943) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [398] [290/312] eta: 0:00:16 lr: 0.000139 min_lr: 0.000139 loss: 1.6630 (1.6933) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [398] [300/312] eta: 0:00:08 lr: 0.000138 min_lr: 0.000138 loss: 1.6284 (1.6868) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [398] [310/312] eta: 0:00:01 lr: 0.000138 min_lr: 0.000138 loss: 1.6739 (1.6862) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [398] [311/312] eta: 0:00:00 lr: 0.000138 min_lr: 0.000138 loss: 1.6820 (1.6873) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [398] Total time: 0:03:48 (0.7311 s / it) Averaged stats: lr: 0.000138 min_lr: 0.000138 loss: 1.6820 (1.6896) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4335 (0.4335) acc1: 88.5417 (88.5417) acc5: 98.4375 (98.4375) time: 4.3987 data: 4.1834 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6197 (0.6463) acc1: 83.3333 (82.6560) acc5: 97.1354 (96.7360) time: 0.6400 data: 0.4649 max mem: 64948 Test: Total time: 0:00:05 (0.6652 s / it) * Acc@1 83.482 Acc@5 96.590 loss 0.628 Accuracy of the model on the 50000 test images: 83.5% Max accuracy: 83.49% Test: [0/9] eta: 0:00:44 loss: 0.4560 (0.4560) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.9645 data: 4.7594 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6213 (0.6424) acc1: 84.3750 (82.4320) acc5: 97.1354 (96.6720) time: 0.7029 data: 0.5289 max mem: 64948 Test: Total time: 0:00:06 (0.7178 s / it) * Acc@1 83.580 Acc@5 96.678 loss 0.621 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.58% Epoch: [399] [ 0/312] eta: 0:47:35 lr: 0.000138 min_lr: 0.000138 loss: 1.7908 (1.7908) weight_decay: 0.0500 (0.0500) time: 9.1523 data: 8.3726 max mem: 64948 Epoch: [399] [ 10/312] eta: 0:07:38 lr: 0.000138 min_lr: 0.000138 loss: 1.8422 (1.8511) weight_decay: 0.0500 (0.0500) time: 1.5182 data: 0.7615 max mem: 64948 Epoch: [399] [ 20/312] eta: 0:05:28 lr: 0.000138 min_lr: 0.000138 loss: 1.7957 (1.7706) weight_decay: 0.0500 (0.0500) time: 0.7251 data: 0.0004 max mem: 64948 Epoch: [399] [ 30/312] eta: 0:04:38 lr: 0.000138 min_lr: 0.000138 loss: 1.6346 (1.7325) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [399] [ 40/312] eta: 0:04:09 lr: 0.000138 min_lr: 0.000138 loss: 1.4675 (1.6867) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [399] [ 50/312] eta: 0:03:48 lr: 0.000137 min_lr: 0.000137 loss: 1.4784 (1.6725) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [399] [ 60/312] eta: 0:03:32 lr: 0.000137 min_lr: 0.000137 loss: 1.8135 (1.6884) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [399] [ 70/312] eta: 0:03:19 lr: 0.000137 min_lr: 0.000137 loss: 1.7223 (1.6669) weight_decay: 0.0500 (0.0500) time: 0.7008 data: 0.0004 max mem: 64948 Epoch: [399] [ 80/312] eta: 0:03:07 lr: 0.000137 min_lr: 0.000137 loss: 1.6806 (1.6738) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [399] [ 90/312] eta: 0:02:56 lr: 0.000137 min_lr: 0.000137 loss: 1.6667 (1.6585) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [399] [100/312] eta: 0:02:46 lr: 0.000137 min_lr: 0.000137 loss: 1.7283 (1.6763) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [399] [110/312] eta: 0:02:37 lr: 0.000136 min_lr: 0.000136 loss: 1.9241 (1.6936) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [399] [120/312] eta: 0:02:27 lr: 0.000136 min_lr: 0.000136 loss: 1.7784 (1.6901) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [399] [130/312] eta: 0:02:19 lr: 0.000136 min_lr: 0.000136 loss: 1.6789 (1.6784) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [399] [140/312] eta: 0:02:10 lr: 0.000136 min_lr: 0.000136 loss: 1.7438 (1.6819) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [399] [150/312] eta: 0:02:02 lr: 0.000136 min_lr: 0.000136 loss: 1.8456 (1.6912) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [399] [160/312] eta: 0:01:54 lr: 0.000135 min_lr: 0.000135 loss: 1.8456 (1.6893) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [399] [170/312] eta: 0:01:46 lr: 0.000135 min_lr: 0.000135 loss: 1.7559 (1.6988) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [399] [180/312] eta: 0:01:38 lr: 0.000135 min_lr: 0.000135 loss: 1.7858 (1.6978) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [399] [190/312] eta: 0:01:30 lr: 0.000135 min_lr: 0.000135 loss: 1.8034 (1.7063) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [399] [200/312] eta: 0:01:22 lr: 0.000135 min_lr: 0.000135 loss: 1.7857 (1.7023) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [399] [210/312] eta: 0:01:15 lr: 0.000135 min_lr: 0.000135 loss: 1.6911 (1.7016) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [399] [220/312] eta: 0:01:07 lr: 0.000134 min_lr: 0.000134 loss: 1.8101 (1.7059) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [399] [230/312] eta: 0:01:00 lr: 0.000134 min_lr: 0.000134 loss: 1.7516 (1.7054) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [399] [240/312] eta: 0:00:52 lr: 0.000134 min_lr: 0.000134 loss: 1.6991 (1.7038) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [399] [250/312] eta: 0:00:45 lr: 0.000134 min_lr: 0.000134 loss: 1.6818 (1.7023) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [399] [260/312] eta: 0:00:37 lr: 0.000134 min_lr: 0.000134 loss: 1.7463 (1.7049) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [399] [270/312] eta: 0:00:30 lr: 0.000134 min_lr: 0.000134 loss: 1.7463 (1.6992) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [399] [280/312] eta: 0:00:23 lr: 0.000133 min_lr: 0.000133 loss: 1.5777 (1.6995) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0010 max mem: 64948 Epoch: [399] [290/312] eta: 0:00:15 lr: 0.000133 min_lr: 0.000133 loss: 1.5832 (1.6944) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [399] [300/312] eta: 0:00:08 lr: 0.000133 min_lr: 0.000133 loss: 1.6966 (1.6988) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [399] [310/312] eta: 0:00:01 lr: 0.000133 min_lr: 0.000133 loss: 1.8820 (1.7040) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [399] [311/312] eta: 0:00:00 lr: 0.000133 min_lr: 0.000133 loss: 1.8803 (1.7046) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [399] Total time: 0:03:46 (0.7265 s / it) Averaged stats: lr: 0.000133 min_lr: 0.000133 loss: 1.8803 (1.6890) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4541 (0.4541) acc1: 88.0208 (88.0208) acc5: 98.4375 (98.4375) time: 4.4091 data: 4.1939 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6169 (0.6430) acc1: 83.8542 (82.7200) acc5: 97.3958 (96.7680) time: 0.6419 data: 0.4661 max mem: 64948 Test: Total time: 0:00:05 (0.6664 s / it) * Acc@1 83.570 Acc@5 96.612 loss 0.627 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.57% Test: [0/9] eta: 0:00:43 loss: 0.4558 (0.4558) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.8383 data: 4.6273 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6208 (0.6421) acc1: 84.3750 (82.4640) acc5: 97.3958 (96.7040) time: 0.6892 data: 0.5142 max mem: 64948 Test: Total time: 0:00:06 (0.6999 s / it) * Acc@1 83.568 Acc@5 96.678 loss 0.620 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [400] [ 0/312] eta: 0:56:34 lr: 0.000133 min_lr: 0.000133 loss: 1.6994 (1.6994) weight_decay: 0.0500 (0.0500) time: 10.8800 data: 9.2768 max mem: 64948 Epoch: [400] [ 10/312] eta: 0:08:24 lr: 0.000133 min_lr: 0.000133 loss: 1.6994 (1.6048) weight_decay: 0.0500 (0.0500) time: 1.6689 data: 0.8437 max mem: 64948 Epoch: [400] [ 20/312] eta: 0:05:51 lr: 0.000133 min_lr: 0.000133 loss: 1.7117 (1.6944) weight_decay: 0.0500 (0.0500) time: 0.7213 data: 0.0004 max mem: 64948 Epoch: [400] [ 30/312] eta: 0:04:53 lr: 0.000132 min_lr: 0.000132 loss: 1.7171 (1.6865) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [400] [ 40/312] eta: 0:04:20 lr: 0.000132 min_lr: 0.000132 loss: 1.7516 (1.6946) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [400] [ 50/312] eta: 0:03:57 lr: 0.000132 min_lr: 0.000132 loss: 1.8212 (1.7153) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [400] [ 60/312] eta: 0:03:39 lr: 0.000132 min_lr: 0.000132 loss: 1.8600 (1.7264) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [400] [ 70/312] eta: 0:03:25 lr: 0.000132 min_lr: 0.000132 loss: 1.8734 (1.7226) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [400] [ 80/312] eta: 0:03:12 lr: 0.000132 min_lr: 0.000132 loss: 1.8108 (1.7185) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [400] [ 90/312] eta: 0:03:00 lr: 0.000131 min_lr: 0.000131 loss: 1.7504 (1.7186) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [400] [100/312] eta: 0:02:50 lr: 0.000131 min_lr: 0.000131 loss: 1.6748 (1.6972) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [400] [110/312] eta: 0:02:40 lr: 0.000131 min_lr: 0.000131 loss: 1.6528 (1.6962) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [400] [120/312] eta: 0:02:30 lr: 0.000131 min_lr: 0.000131 loss: 1.6824 (1.6920) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [400] [130/312] eta: 0:02:21 lr: 0.000131 min_lr: 0.000131 loss: 1.7437 (1.6956) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [400] [140/312] eta: 0:02:12 lr: 0.000131 min_lr: 0.000131 loss: 1.8266 (1.7043) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [400] [150/312] eta: 0:02:04 lr: 0.000130 min_lr: 0.000130 loss: 1.8422 (1.6993) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [400] [160/312] eta: 0:01:55 lr: 0.000130 min_lr: 0.000130 loss: 1.7949 (1.6947) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [400] [170/312] eta: 0:01:47 lr: 0.000130 min_lr: 0.000130 loss: 1.7788 (1.6938) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [400] [180/312] eta: 0:01:39 lr: 0.000130 min_lr: 0.000130 loss: 1.8194 (1.7004) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [400] [190/312] eta: 0:01:31 lr: 0.000130 min_lr: 0.000130 loss: 1.8332 (1.7054) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [400] [200/312] eta: 0:01:23 lr: 0.000130 min_lr: 0.000130 loss: 1.7912 (1.7034) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [400] [210/312] eta: 0:01:16 lr: 0.000129 min_lr: 0.000129 loss: 1.6655 (1.7017) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [400] [220/312] eta: 0:01:08 lr: 0.000129 min_lr: 0.000129 loss: 1.6655 (1.7002) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [400] [230/312] eta: 0:01:00 lr: 0.000129 min_lr: 0.000129 loss: 1.7056 (1.7046) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [400] [240/312] eta: 0:00:53 lr: 0.000129 min_lr: 0.000129 loss: 1.7209 (1.7058) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [400] [250/312] eta: 0:00:45 lr: 0.000129 min_lr: 0.000129 loss: 1.8069 (1.7113) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [400] [260/312] eta: 0:00:38 lr: 0.000129 min_lr: 0.000129 loss: 1.8564 (1.7148) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [400] [270/312] eta: 0:00:30 lr: 0.000128 min_lr: 0.000128 loss: 1.7591 (1.7134) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0004 max mem: 64948 Epoch: [400] [280/312] eta: 0:00:23 lr: 0.000128 min_lr: 0.000128 loss: 1.8131 (1.7196) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0009 max mem: 64948 Epoch: [400] [290/312] eta: 0:00:16 lr: 0.000128 min_lr: 0.000128 loss: 1.7725 (1.7185) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [400] [300/312] eta: 0:00:08 lr: 0.000128 min_lr: 0.000128 loss: 1.6259 (1.7144) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [400] [310/312] eta: 0:00:01 lr: 0.000128 min_lr: 0.000128 loss: 1.6471 (1.7120) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [400] [311/312] eta: 0:00:00 lr: 0.000128 min_lr: 0.000128 loss: 1.6660 (1.7133) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [400] Total time: 0:03:48 (0.7326 s / it) Averaged stats: lr: 0.000128 min_lr: 0.000128 loss: 1.6660 (1.6859) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4350 (0.4350) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 4.7178 data: 4.5109 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6174 (0.6482) acc1: 84.3750 (82.8160) acc5: 97.3958 (96.8320) time: 0.6755 data: 0.5013 max mem: 64948 Test: Total time: 0:00:06 (0.6985 s / it) * Acc@1 83.366 Acc@5 96.644 loss 0.628 Accuracy of the model on the 50000 test images: 83.4% Max accuracy: 83.57% Test: [0/9] eta: 0:00:42 loss: 0.4556 (0.4556) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.6988 data: 4.4886 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6203 (0.6419) acc1: 84.3750 (82.4320) acc5: 97.3958 (96.7040) time: 0.6899 data: 0.5151 max mem: 64948 Test: Total time: 0:00:06 (0.6988 s / it) * Acc@1 83.558 Acc@5 96.680 loss 0.620 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [401] [ 0/312] eta: 0:53:25 lr: 0.000128 min_lr: 0.000128 loss: 2.0864 (2.0864) weight_decay: 0.0500 (0.0500) time: 10.2731 data: 6.6632 max mem: 64948 Epoch: [401] [ 10/312] eta: 0:08:12 lr: 0.000128 min_lr: 0.000128 loss: 1.6857 (1.6489) weight_decay: 0.0500 (0.0500) time: 1.6300 data: 0.6062 max mem: 64948 Epoch: [401] [ 20/312] eta: 0:05:45 lr: 0.000127 min_lr: 0.000127 loss: 1.6971 (1.7136) weight_decay: 0.0500 (0.0500) time: 0.7299 data: 0.0005 max mem: 64948 Epoch: [401] [ 30/312] eta: 0:04:49 lr: 0.000127 min_lr: 0.000127 loss: 1.7757 (1.6889) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [401] [ 40/312] eta: 0:04:17 lr: 0.000127 min_lr: 0.000127 loss: 1.7638 (1.7049) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [401] [ 50/312] eta: 0:03:55 lr: 0.000127 min_lr: 0.000127 loss: 1.7638 (1.7049) weight_decay: 0.0500 (0.0500) time: 0.7005 data: 0.0004 max mem: 64948 Epoch: [401] [ 60/312] eta: 0:03:37 lr: 0.000127 min_lr: 0.000127 loss: 1.7417 (1.6908) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [401] [ 70/312] eta: 0:03:23 lr: 0.000127 min_lr: 0.000127 loss: 1.6425 (1.6777) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [401] [ 80/312] eta: 0:03:10 lr: 0.000126 min_lr: 0.000126 loss: 1.7317 (1.7001) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [401] [ 90/312] eta: 0:02:59 lr: 0.000126 min_lr: 0.000126 loss: 1.7572 (1.6831) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [401] [100/312] eta: 0:02:49 lr: 0.000126 min_lr: 0.000126 loss: 1.3892 (1.6668) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [401] [110/312] eta: 0:02:39 lr: 0.000126 min_lr: 0.000126 loss: 1.5964 (1.6723) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [401] [120/312] eta: 0:02:29 lr: 0.000126 min_lr: 0.000126 loss: 1.5964 (1.6575) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [401] [130/312] eta: 0:02:20 lr: 0.000126 min_lr: 0.000126 loss: 1.4761 (1.6524) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [401] [140/312] eta: 0:02:12 lr: 0.000125 min_lr: 0.000125 loss: 1.6861 (1.6562) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [401] [150/312] eta: 0:02:03 lr: 0.000125 min_lr: 0.000125 loss: 1.7581 (1.6567) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [401] [160/312] eta: 0:01:55 lr: 0.000125 min_lr: 0.000125 loss: 1.5687 (1.6474) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [401] [170/312] eta: 0:01:47 lr: 0.000125 min_lr: 0.000125 loss: 1.6412 (1.6569) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [401] [180/312] eta: 0:01:39 lr: 0.000125 min_lr: 0.000125 loss: 1.8130 (1.6631) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [401] [190/312] eta: 0:01:31 lr: 0.000125 min_lr: 0.000125 loss: 1.6662 (1.6565) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [401] [200/312] eta: 0:01:23 lr: 0.000125 min_lr: 0.000125 loss: 1.7163 (1.6598) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [401] [210/312] eta: 0:01:15 lr: 0.000124 min_lr: 0.000124 loss: 1.8250 (1.6634) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [401] [220/312] eta: 0:01:08 lr: 0.000124 min_lr: 0.000124 loss: 1.7798 (1.6669) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [401] [230/312] eta: 0:01:00 lr: 0.000124 min_lr: 0.000124 loss: 1.7788 (1.6725) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [401] [240/312] eta: 0:00:53 lr: 0.000124 min_lr: 0.000124 loss: 1.6959 (1.6654) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [401] [250/312] eta: 0:00:45 lr: 0.000124 min_lr: 0.000124 loss: 1.5541 (1.6647) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [401] [260/312] eta: 0:00:38 lr: 0.000124 min_lr: 0.000124 loss: 1.6168 (1.6615) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [401] [270/312] eta: 0:00:30 lr: 0.000123 min_lr: 0.000123 loss: 1.6168 (1.6617) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [401] [280/312] eta: 0:00:23 lr: 0.000123 min_lr: 0.000123 loss: 1.8252 (1.6673) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0010 max mem: 64948 Epoch: [401] [290/312] eta: 0:00:16 lr: 0.000123 min_lr: 0.000123 loss: 1.8252 (1.6638) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0009 max mem: 64948 Epoch: [401] [300/312] eta: 0:00:08 lr: 0.000123 min_lr: 0.000123 loss: 1.7339 (1.6684) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [401] [310/312] eta: 0:00:01 lr: 0.000123 min_lr: 0.000123 loss: 1.7339 (1.6697) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [401] [311/312] eta: 0:00:00 lr: 0.000123 min_lr: 0.000123 loss: 1.7339 (1.6704) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [401] Total time: 0:03:48 (0.7310 s / it) Averaged stats: lr: 0.000123 min_lr: 0.000123 loss: 1.7339 (1.6732) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4527 (0.4527) acc1: 89.3229 (89.3229) acc5: 98.1771 (98.1771) time: 4.5610 data: 4.3416 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6059 (0.6375) acc1: 84.3750 (83.2640) acc5: 97.6562 (96.8640) time: 0.6584 data: 0.4825 max mem: 64948 Test: Total time: 0:00:06 (0.6817 s / it) * Acc@1 83.546 Acc@5 96.642 loss 0.625 Accuracy of the model on the 50000 test images: 83.5% Max accuracy: 83.57% Test: [0/9] eta: 0:00:46 loss: 0.4554 (0.4554) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 5.2004 data: 4.9820 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6199 (0.6416) acc1: 84.3750 (82.4960) acc5: 97.3958 (96.6720) time: 0.7291 data: 0.5536 max mem: 64948 Test: Total time: 0:00:06 (0.7401 s / it) * Acc@1 83.574 Acc@5 96.688 loss 0.620 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [402] [ 0/312] eta: 0:55:34 lr: 0.000123 min_lr: 0.000123 loss: 1.6510 (1.6510) weight_decay: 0.0500 (0.0500) time: 10.6889 data: 8.5719 max mem: 64948 Epoch: [402] [ 10/312] eta: 0:08:11 lr: 0.000123 min_lr: 0.000123 loss: 1.7848 (1.7229) weight_decay: 0.0500 (0.0500) time: 1.6279 data: 0.7797 max mem: 64948 Epoch: [402] [ 20/312] eta: 0:05:45 lr: 0.000122 min_lr: 0.000122 loss: 1.7958 (1.7134) weight_decay: 0.0500 (0.0500) time: 0.7089 data: 0.0004 max mem: 64948 Epoch: [402] [ 30/312] eta: 0:04:49 lr: 0.000122 min_lr: 0.000122 loss: 1.7821 (1.7345) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [402] [ 40/312] eta: 0:04:17 lr: 0.000122 min_lr: 0.000122 loss: 1.7485 (1.7283) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [402] [ 50/312] eta: 0:03:54 lr: 0.000122 min_lr: 0.000122 loss: 1.6418 (1.6865) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [402] [ 60/312] eta: 0:03:37 lr: 0.000122 min_lr: 0.000122 loss: 1.6418 (1.6761) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [402] [ 70/312] eta: 0:03:23 lr: 0.000122 min_lr: 0.000122 loss: 1.6456 (1.6791) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [402] [ 80/312] eta: 0:03:10 lr: 0.000121 min_lr: 0.000121 loss: 1.6912 (1.6740) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [402] [ 90/312] eta: 0:02:59 lr: 0.000121 min_lr: 0.000121 loss: 1.6912 (1.6808) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [402] [100/312] eta: 0:02:49 lr: 0.000121 min_lr: 0.000121 loss: 1.6875 (1.6778) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [402] [110/312] eta: 0:02:39 lr: 0.000121 min_lr: 0.000121 loss: 1.6281 (1.6735) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [402] [120/312] eta: 0:02:29 lr: 0.000121 min_lr: 0.000121 loss: 1.6281 (1.6719) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [402] [130/312] eta: 0:02:20 lr: 0.000121 min_lr: 0.000121 loss: 1.5505 (1.6711) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [402] [140/312] eta: 0:02:12 lr: 0.000120 min_lr: 0.000120 loss: 1.7881 (1.6825) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [402] [150/312] eta: 0:02:03 lr: 0.000120 min_lr: 0.000120 loss: 1.7881 (1.6867) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [402] [160/312] eta: 0:01:55 lr: 0.000120 min_lr: 0.000120 loss: 1.6983 (1.6874) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [402] [170/312] eta: 0:01:47 lr: 0.000120 min_lr: 0.000120 loss: 1.7801 (1.6940) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [402] [180/312] eta: 0:01:39 lr: 0.000120 min_lr: 0.000120 loss: 1.8539 (1.6969) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [402] [190/312] eta: 0:01:31 lr: 0.000120 min_lr: 0.000120 loss: 1.7522 (1.6915) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [402] [200/312] eta: 0:01:23 lr: 0.000120 min_lr: 0.000120 loss: 1.7622 (1.6985) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [402] [210/312] eta: 0:01:15 lr: 0.000119 min_lr: 0.000119 loss: 1.8092 (1.7008) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [402] [220/312] eta: 0:01:08 lr: 0.000119 min_lr: 0.000119 loss: 1.7718 (1.7040) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [402] [230/312] eta: 0:01:00 lr: 0.000119 min_lr: 0.000119 loss: 1.7764 (1.7051) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [402] [240/312] eta: 0:00:53 lr: 0.000119 min_lr: 0.000119 loss: 1.7764 (1.7055) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [402] [250/312] eta: 0:00:45 lr: 0.000119 min_lr: 0.000119 loss: 1.8254 (1.7083) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [402] [260/312] eta: 0:00:38 lr: 0.000119 min_lr: 0.000119 loss: 1.7691 (1.7083) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [402] [270/312] eta: 0:00:30 lr: 0.000118 min_lr: 0.000118 loss: 1.7344 (1.7086) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [402] [280/312] eta: 0:00:23 lr: 0.000118 min_lr: 0.000118 loss: 1.6635 (1.7037) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0010 max mem: 64948 Epoch: [402] [290/312] eta: 0:00:16 lr: 0.000118 min_lr: 0.000118 loss: 1.5227 (1.6957) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0009 max mem: 64948 Epoch: [402] [300/312] eta: 0:00:08 lr: 0.000118 min_lr: 0.000118 loss: 1.6959 (1.6991) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [402] [310/312] eta: 0:00:01 lr: 0.000118 min_lr: 0.000118 loss: 1.6667 (1.6946) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [402] [311/312] eta: 0:00:00 lr: 0.000118 min_lr: 0.000118 loss: 1.6667 (1.6947) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [402] Total time: 0:03:48 (0.7308 s / it) Averaged stats: lr: 0.000118 min_lr: 0.000118 loss: 1.6667 (1.6777) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4474 (0.4474) acc1: 88.0208 (88.0208) acc5: 97.6562 (97.6562) time: 4.7065 data: 4.4938 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6220 (0.6455) acc1: 83.8542 (82.5600) acc5: 97.1354 (96.6720) time: 0.6742 data: 0.4994 max mem: 64948 Test: Total time: 0:00:06 (0.6982 s / it) * Acc@1 83.548 Acc@5 96.646 loss 0.630 Accuracy of the model on the 50000 test images: 83.5% Max accuracy: 83.57% Test: [0/9] eta: 0:00:44 loss: 0.4555 (0.4555) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.8950 data: 4.6756 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6194 (0.6414) acc1: 84.3750 (82.4640) acc5: 97.1354 (96.6400) time: 0.6952 data: 0.5196 max mem: 64948 Test: Total time: 0:00:06 (0.7071 s / it) * Acc@1 83.586 Acc@5 96.688 loss 0.620 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.59% Epoch: [403] [ 0/312] eta: 0:54:12 lr: 0.000118 min_lr: 0.000118 loss: 2.1625 (2.1625) weight_decay: 0.0500 (0.0500) time: 10.4257 data: 9.6451 max mem: 64948 Epoch: [403] [ 10/312] eta: 0:08:01 lr: 0.000118 min_lr: 0.000118 loss: 1.6620 (1.7615) weight_decay: 0.0500 (0.0500) time: 1.5945 data: 0.8771 max mem: 64948 Epoch: [403] [ 20/312] eta: 0:05:40 lr: 0.000117 min_lr: 0.000117 loss: 1.6620 (1.7108) weight_decay: 0.0500 (0.0500) time: 0.7041 data: 0.0004 max mem: 64948 Epoch: [403] [ 30/312] eta: 0:04:46 lr: 0.000117 min_lr: 0.000117 loss: 1.8330 (1.7713) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [403] [ 40/312] eta: 0:04:14 lr: 0.000117 min_lr: 0.000117 loss: 1.8265 (1.7575) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [403] [ 50/312] eta: 0:03:52 lr: 0.000117 min_lr: 0.000117 loss: 1.7704 (1.7527) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [403] [ 60/312] eta: 0:03:36 lr: 0.000117 min_lr: 0.000117 loss: 1.7983 (1.7461) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [403] [ 70/312] eta: 0:03:22 lr: 0.000117 min_lr: 0.000117 loss: 1.7525 (1.7309) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [403] [ 80/312] eta: 0:03:09 lr: 0.000116 min_lr: 0.000116 loss: 1.6827 (1.7252) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [403] [ 90/312] eta: 0:02:58 lr: 0.000116 min_lr: 0.000116 loss: 1.7655 (1.7301) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [403] [100/312] eta: 0:02:48 lr: 0.000116 min_lr: 0.000116 loss: 1.7708 (1.7195) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [403] [110/312] eta: 0:02:38 lr: 0.000116 min_lr: 0.000116 loss: 1.5961 (1.7086) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [403] [120/312] eta: 0:02:29 lr: 0.000116 min_lr: 0.000116 loss: 1.5701 (1.6938) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [403] [130/312] eta: 0:02:20 lr: 0.000116 min_lr: 0.000116 loss: 1.6411 (1.6926) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [403] [140/312] eta: 0:02:11 lr: 0.000116 min_lr: 0.000116 loss: 1.5081 (1.6798) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [403] [150/312] eta: 0:02:03 lr: 0.000115 min_lr: 0.000115 loss: 1.6319 (1.6820) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [403] [160/312] eta: 0:01:55 lr: 0.000115 min_lr: 0.000115 loss: 1.7110 (1.6773) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [403] [170/312] eta: 0:01:46 lr: 0.000115 min_lr: 0.000115 loss: 1.6007 (1.6688) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [403] [180/312] eta: 0:01:38 lr: 0.000115 min_lr: 0.000115 loss: 1.5796 (1.6665) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [403] [190/312] eta: 0:01:31 lr: 0.000115 min_lr: 0.000115 loss: 1.5118 (1.6563) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [403] [200/312] eta: 0:01:23 lr: 0.000115 min_lr: 0.000115 loss: 1.5436 (1.6582) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [403] [210/312] eta: 0:01:15 lr: 0.000114 min_lr: 0.000114 loss: 1.7706 (1.6611) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [403] [220/312] eta: 0:01:08 lr: 0.000114 min_lr: 0.000114 loss: 1.5745 (1.6599) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [403] [230/312] eta: 0:01:00 lr: 0.000114 min_lr: 0.000114 loss: 1.6536 (1.6619) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [403] [240/312] eta: 0:00:52 lr: 0.000114 min_lr: 0.000114 loss: 1.7197 (1.6611) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [403] [250/312] eta: 0:00:45 lr: 0.000114 min_lr: 0.000114 loss: 1.6603 (1.6613) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [403] [260/312] eta: 0:00:38 lr: 0.000114 min_lr: 0.000114 loss: 1.7112 (1.6609) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [403] [270/312] eta: 0:00:30 lr: 0.000114 min_lr: 0.000114 loss: 1.7469 (1.6632) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [403] [280/312] eta: 0:00:23 lr: 0.000113 min_lr: 0.000113 loss: 1.7778 (1.6631) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0009 max mem: 64948 Epoch: [403] [290/312] eta: 0:00:16 lr: 0.000113 min_lr: 0.000113 loss: 1.7778 (1.6655) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [403] [300/312] eta: 0:00:08 lr: 0.000113 min_lr: 0.000113 loss: 1.7884 (1.6683) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [403] [310/312] eta: 0:00:01 lr: 0.000113 min_lr: 0.000113 loss: 1.8228 (1.6720) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [403] [311/312] eta: 0:00:00 lr: 0.000113 min_lr: 0.000113 loss: 1.8221 (1.6722) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [403] Total time: 0:03:47 (0.7294 s / it) Averaged stats: lr: 0.000113 min_lr: 0.000113 loss: 1.8221 (1.6728) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4452 (0.4452) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.5898 data: 4.3758 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6074 (0.6437) acc1: 84.1146 (82.9760) acc5: 96.8750 (96.6400) time: 0.6612 data: 0.4863 max mem: 64948 Test: Total time: 0:00:06 (0.6843 s / it) * Acc@1 83.498 Acc@5 96.638 loss 0.627 Accuracy of the model on the 50000 test images: 83.5% Max accuracy: 83.57% Test: [0/9] eta: 0:00:44 loss: 0.4554 (0.4554) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.9761 data: 4.7747 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6190 (0.6411) acc1: 84.3750 (82.4960) acc5: 97.1354 (96.6400) time: 0.7041 data: 0.5306 max mem: 64948 Test: Total time: 0:00:06 (0.7120 s / it) * Acc@1 83.584 Acc@5 96.684 loss 0.620 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [404] [ 0/312] eta: 0:54:42 lr: 0.000113 min_lr: 0.000113 loss: 1.2349 (1.2349) weight_decay: 0.0500 (0.0500) time: 10.5214 data: 8.6309 max mem: 64948 Epoch: [404] [ 10/312] eta: 0:08:06 lr: 0.000113 min_lr: 0.000113 loss: 1.6811 (1.6637) weight_decay: 0.0500 (0.0500) time: 1.6122 data: 0.7851 max mem: 64948 Epoch: [404] [ 20/312] eta: 0:05:43 lr: 0.000113 min_lr: 0.000113 loss: 1.6291 (1.6472) weight_decay: 0.0500 (0.0500) time: 0.7079 data: 0.0004 max mem: 64948 Epoch: [404] [ 30/312] eta: 0:04:47 lr: 0.000112 min_lr: 0.000112 loss: 1.5758 (1.6542) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [404] [ 40/312] eta: 0:04:16 lr: 0.000112 min_lr: 0.000112 loss: 1.7060 (1.6490) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [404] [ 50/312] eta: 0:03:53 lr: 0.000112 min_lr: 0.000112 loss: 1.7723 (1.6468) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [404] [ 60/312] eta: 0:03:36 lr: 0.000112 min_lr: 0.000112 loss: 1.5325 (1.6355) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [404] [ 70/312] eta: 0:03:22 lr: 0.000112 min_lr: 0.000112 loss: 1.6199 (1.6304) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [404] [ 80/312] eta: 0:03:10 lr: 0.000112 min_lr: 0.000112 loss: 1.6199 (1.6188) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [404] [ 90/312] eta: 0:02:58 lr: 0.000111 min_lr: 0.000111 loss: 1.6499 (1.6357) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [404] [100/312] eta: 0:02:48 lr: 0.000111 min_lr: 0.000111 loss: 1.7101 (1.6362) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [404] [110/312] eta: 0:02:38 lr: 0.000111 min_lr: 0.000111 loss: 1.7194 (1.6490) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [404] [120/312] eta: 0:02:29 lr: 0.000111 min_lr: 0.000111 loss: 1.7513 (1.6606) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [404] [130/312] eta: 0:02:20 lr: 0.000111 min_lr: 0.000111 loss: 1.8058 (1.6697) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [404] [140/312] eta: 0:02:11 lr: 0.000111 min_lr: 0.000111 loss: 1.8812 (1.6873) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [404] [150/312] eta: 0:02:03 lr: 0.000111 min_lr: 0.000111 loss: 1.7086 (1.6791) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [404] [160/312] eta: 0:01:55 lr: 0.000110 min_lr: 0.000110 loss: 1.4907 (1.6793) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [404] [170/312] eta: 0:01:47 lr: 0.000110 min_lr: 0.000110 loss: 1.8206 (1.6829) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [404] [180/312] eta: 0:01:39 lr: 0.000110 min_lr: 0.000110 loss: 1.8051 (1.6794) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [404] [190/312] eta: 0:01:31 lr: 0.000110 min_lr: 0.000110 loss: 1.8051 (1.6862) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [404] [200/312] eta: 0:01:23 lr: 0.000110 min_lr: 0.000110 loss: 1.7638 (1.6805) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [404] [210/312] eta: 0:01:15 lr: 0.000110 min_lr: 0.000110 loss: 1.5264 (1.6782) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [404] [220/312] eta: 0:01:08 lr: 0.000109 min_lr: 0.000109 loss: 1.4813 (1.6753) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [404] [230/312] eta: 0:01:00 lr: 0.000109 min_lr: 0.000109 loss: 1.4813 (1.6710) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0003 max mem: 64948 Epoch: [404] [240/312] eta: 0:00:53 lr: 0.000109 min_lr: 0.000109 loss: 1.7048 (1.6716) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0003 max mem: 64948 Epoch: [404] [250/312] eta: 0:00:45 lr: 0.000109 min_lr: 0.000109 loss: 1.7687 (1.6763) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [404] [260/312] eta: 0:00:38 lr: 0.000109 min_lr: 0.000109 loss: 1.7687 (1.6770) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [404] [270/312] eta: 0:00:30 lr: 0.000109 min_lr: 0.000109 loss: 1.6285 (1.6774) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [404] [280/312] eta: 0:00:23 lr: 0.000109 min_lr: 0.000109 loss: 1.6085 (1.6710) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0009 max mem: 64948 Epoch: [404] [290/312] eta: 0:00:16 lr: 0.000108 min_lr: 0.000108 loss: 1.6085 (1.6718) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [404] [300/312] eta: 0:00:08 lr: 0.000108 min_lr: 0.000108 loss: 1.7476 (1.6728) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [404] [310/312] eta: 0:00:01 lr: 0.000108 min_lr: 0.000108 loss: 1.6968 (1.6695) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [404] [311/312] eta: 0:00:00 lr: 0.000108 min_lr: 0.000108 loss: 1.6545 (1.6694) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [404] Total time: 0:03:47 (0.7302 s / it) Averaged stats: lr: 0.000108 min_lr: 0.000108 loss: 1.6545 (1.6743) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4335 (0.4335) acc1: 88.2812 (88.2812) acc5: 98.1771 (98.1771) time: 4.7843 data: 4.5650 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6130 (0.6385) acc1: 84.1146 (82.9440) acc5: 97.6562 (96.9280) time: 0.6834 data: 0.5073 max mem: 64948 Test: Total time: 0:00:06 (0.7036 s / it) * Acc@1 83.560 Acc@5 96.702 loss 0.626 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.57% Test: [0/9] eta: 0:00:47 loss: 0.4552 (0.4552) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 5.2855 data: 5.0832 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6186 (0.6410) acc1: 84.6354 (82.5920) acc5: 97.1354 (96.6400) time: 0.7386 data: 0.5649 max mem: 64948 Test: Total time: 0:00:06 (0.7472 s / it) * Acc@1 83.602 Acc@5 96.694 loss 0.620 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.60% Epoch: [405] [ 0/312] eta: 0:49:12 lr: 0.000108 min_lr: 0.000108 loss: 1.8529 (1.8529) weight_decay: 0.0500 (0.0500) time: 9.4627 data: 8.1416 max mem: 64948 Epoch: [405] [ 10/312] eta: 0:07:52 lr: 0.000108 min_lr: 0.000108 loss: 1.5853 (1.6042) weight_decay: 0.0500 (0.0500) time: 1.5633 data: 0.7405 max mem: 64948 Epoch: [405] [ 20/312] eta: 0:05:36 lr: 0.000108 min_lr: 0.000108 loss: 1.5853 (1.5721) weight_decay: 0.0500 (0.0500) time: 0.7360 data: 0.0004 max mem: 64948 Epoch: [405] [ 30/312] eta: 0:04:43 lr: 0.000108 min_lr: 0.000108 loss: 1.6851 (1.6567) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [405] [ 40/312] eta: 0:04:12 lr: 0.000107 min_lr: 0.000107 loss: 1.7475 (1.6531) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [405] [ 50/312] eta: 0:03:51 lr: 0.000107 min_lr: 0.000107 loss: 1.7249 (1.6777) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [405] [ 60/312] eta: 0:03:34 lr: 0.000107 min_lr: 0.000107 loss: 1.8643 (1.7043) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [405] [ 70/312] eta: 0:03:20 lr: 0.000107 min_lr: 0.000107 loss: 1.7695 (1.7037) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [405] [ 80/312] eta: 0:03:08 lr: 0.000107 min_lr: 0.000107 loss: 1.6902 (1.7023) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [405] [ 90/312] eta: 0:02:57 lr: 0.000107 min_lr: 0.000107 loss: 1.6550 (1.6987) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [405] [100/312] eta: 0:02:47 lr: 0.000107 min_lr: 0.000107 loss: 1.6718 (1.6960) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [405] [110/312] eta: 0:02:37 lr: 0.000106 min_lr: 0.000106 loss: 1.8340 (1.7062) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [405] [120/312] eta: 0:02:28 lr: 0.000106 min_lr: 0.000106 loss: 1.7647 (1.7031) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [405] [130/312] eta: 0:02:19 lr: 0.000106 min_lr: 0.000106 loss: 1.7035 (1.7015) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [405] [140/312] eta: 0:02:11 lr: 0.000106 min_lr: 0.000106 loss: 1.6878 (1.6976) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [405] [150/312] eta: 0:02:02 lr: 0.000106 min_lr: 0.000106 loss: 1.6940 (1.7017) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [405] [160/312] eta: 0:01:54 lr: 0.000106 min_lr: 0.000106 loss: 1.7551 (1.6992) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [405] [170/312] eta: 0:01:46 lr: 0.000106 min_lr: 0.000106 loss: 1.7255 (1.6985) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [405] [180/312] eta: 0:01:38 lr: 0.000105 min_lr: 0.000105 loss: 1.6852 (1.6931) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [405] [190/312] eta: 0:01:30 lr: 0.000105 min_lr: 0.000105 loss: 1.6959 (1.6933) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [405] [200/312] eta: 0:01:23 lr: 0.000105 min_lr: 0.000105 loss: 1.6711 (1.6903) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [405] [210/312] eta: 0:01:15 lr: 0.000105 min_lr: 0.000105 loss: 1.6711 (1.6899) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [405] [220/312] eta: 0:01:07 lr: 0.000105 min_lr: 0.000105 loss: 1.7979 (1.6958) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [405] [230/312] eta: 0:01:00 lr: 0.000105 min_lr: 0.000105 loss: 1.7050 (1.6852) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [405] [240/312] eta: 0:00:52 lr: 0.000104 min_lr: 0.000104 loss: 1.4317 (1.6812) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [405] [250/312] eta: 0:00:45 lr: 0.000104 min_lr: 0.000104 loss: 1.6943 (1.6775) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [405] [260/312] eta: 0:00:38 lr: 0.000104 min_lr: 0.000104 loss: 1.6943 (1.6765) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [405] [270/312] eta: 0:00:30 lr: 0.000104 min_lr: 0.000104 loss: 1.7181 (1.6788) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [405] [280/312] eta: 0:00:23 lr: 0.000104 min_lr: 0.000104 loss: 1.5889 (1.6771) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [405] [290/312] eta: 0:00:16 lr: 0.000104 min_lr: 0.000104 loss: 1.5631 (1.6741) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0008 max mem: 64948 Epoch: [405] [300/312] eta: 0:00:08 lr: 0.000104 min_lr: 0.000104 loss: 1.6436 (1.6732) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [405] [310/312] eta: 0:00:01 lr: 0.000103 min_lr: 0.000103 loss: 1.7738 (1.6741) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [405] [311/312] eta: 0:00:00 lr: 0.000103 min_lr: 0.000103 loss: 1.7738 (1.6733) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [405] Total time: 0:03:47 (0.7288 s / it) Averaged stats: lr: 0.000103 min_lr: 0.000103 loss: 1.7738 (1.6688) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4390 (0.4390) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.6227 data: 4.4156 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6054 (0.6420) acc1: 84.3750 (82.9760) acc5: 97.3958 (96.6720) time: 0.6649 data: 0.4907 max mem: 64948 Test: Total time: 0:00:06 (0.6911 s / it) * Acc@1 83.598 Acc@5 96.646 loss 0.626 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.60% Test: [0/9] eta: 0:00:41 loss: 0.4548 (0.4548) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.6409 data: 4.4239 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6181 (0.6406) acc1: 84.3750 (82.5280) acc5: 97.1354 (96.6400) time: 0.6689 data: 0.4916 max mem: 64948 Test: Total time: 0:00:06 (0.6765 s / it) * Acc@1 83.592 Acc@5 96.698 loss 0.620 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [406] [ 0/312] eta: 0:53:26 lr: 0.000103 min_lr: 0.000103 loss: 1.5630 (1.5630) weight_decay: 0.0500 (0.0500) time: 10.2771 data: 6.9652 max mem: 64948 Epoch: [406] [ 10/312] eta: 0:08:03 lr: 0.000103 min_lr: 0.000103 loss: 1.8039 (1.7843) weight_decay: 0.0500 (0.0500) time: 1.6000 data: 0.6337 max mem: 64948 Epoch: [406] [ 20/312] eta: 0:05:41 lr: 0.000103 min_lr: 0.000103 loss: 1.8039 (1.7821) weight_decay: 0.0500 (0.0500) time: 0.7133 data: 0.0004 max mem: 64948 Epoch: [406] [ 30/312] eta: 0:04:46 lr: 0.000103 min_lr: 0.000103 loss: 1.6481 (1.6983) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [406] [ 40/312] eta: 0:04:15 lr: 0.000103 min_lr: 0.000103 loss: 1.5648 (1.6625) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [406] [ 50/312] eta: 0:03:53 lr: 0.000103 min_lr: 0.000103 loss: 1.6046 (1.6728) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [406] [ 60/312] eta: 0:03:36 lr: 0.000103 min_lr: 0.000103 loss: 1.7117 (1.6619) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [406] [ 70/312] eta: 0:03:22 lr: 0.000102 min_lr: 0.000102 loss: 1.7117 (1.6574) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [406] [ 80/312] eta: 0:03:09 lr: 0.000102 min_lr: 0.000102 loss: 1.7442 (1.6621) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [406] [ 90/312] eta: 0:02:58 lr: 0.000102 min_lr: 0.000102 loss: 1.7506 (1.6692) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [406] [100/312] eta: 0:02:48 lr: 0.000102 min_lr: 0.000102 loss: 1.7055 (1.6582) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [406] [110/312] eta: 0:02:38 lr: 0.000102 min_lr: 0.000102 loss: 1.7612 (1.6632) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [406] [120/312] eta: 0:02:29 lr: 0.000102 min_lr: 0.000102 loss: 1.7978 (1.6693) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [406] [130/312] eta: 0:02:20 lr: 0.000102 min_lr: 0.000102 loss: 1.6629 (1.6669) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [406] [140/312] eta: 0:02:11 lr: 0.000101 min_lr: 0.000101 loss: 1.7061 (1.6747) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [406] [150/312] eta: 0:02:03 lr: 0.000101 min_lr: 0.000101 loss: 1.7061 (1.6643) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [406] [160/312] eta: 0:01:55 lr: 0.000101 min_lr: 0.000101 loss: 1.7139 (1.6661) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [406] [170/312] eta: 0:01:47 lr: 0.000101 min_lr: 0.000101 loss: 1.6692 (1.6635) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [406] [180/312] eta: 0:01:39 lr: 0.000101 min_lr: 0.000101 loss: 1.6575 (1.6625) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [406] [190/312] eta: 0:01:31 lr: 0.000101 min_lr: 0.000101 loss: 1.7516 (1.6642) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [406] [200/312] eta: 0:01:23 lr: 0.000100 min_lr: 0.000100 loss: 1.7516 (1.6612) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [406] [210/312] eta: 0:01:15 lr: 0.000100 min_lr: 0.000100 loss: 1.7654 (1.6633) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [406] [220/312] eta: 0:01:08 lr: 0.000100 min_lr: 0.000100 loss: 1.8493 (1.6665) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [406] [230/312] eta: 0:01:00 lr: 0.000100 min_lr: 0.000100 loss: 1.7102 (1.6667) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [406] [240/312] eta: 0:00:53 lr: 0.000100 min_lr: 0.000100 loss: 1.6865 (1.6670) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [406] [250/312] eta: 0:00:45 lr: 0.000100 min_lr: 0.000100 loss: 1.8240 (1.6756) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [406] [260/312] eta: 0:00:38 lr: 0.000100 min_lr: 0.000100 loss: 1.7745 (1.6714) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [406] [270/312] eta: 0:00:30 lr: 0.000099 min_lr: 0.000099 loss: 1.6323 (1.6730) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [406] [280/312] eta: 0:00:23 lr: 0.000099 min_lr: 0.000099 loss: 1.6520 (1.6704) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0009 max mem: 64948 Epoch: [406] [290/312] eta: 0:00:16 lr: 0.000099 min_lr: 0.000099 loss: 1.6244 (1.6689) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [406] [300/312] eta: 0:00:08 lr: 0.000099 min_lr: 0.000099 loss: 1.7034 (1.6719) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [406] [310/312] eta: 0:00:01 lr: 0.000099 min_lr: 0.000099 loss: 1.7423 (1.6723) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [406] [311/312] eta: 0:00:00 lr: 0.000099 min_lr: 0.000099 loss: 1.7141 (1.6718) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [406] Total time: 0:03:47 (0.7298 s / it) Averaged stats: lr: 0.000099 min_lr: 0.000099 loss: 1.7141 (1.6731) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4495 (0.4495) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 4.5978 data: 4.3867 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6047 (0.6408) acc1: 85.1562 (83.2320) acc5: 97.6562 (97.0240) time: 0.6622 data: 0.4875 max mem: 64948 Test: Total time: 0:00:06 (0.6833 s / it) * Acc@1 83.578 Acc@5 96.752 loss 0.626 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.60% Test: [0/9] eta: 0:00:44 loss: 0.4545 (0.4545) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 4.9523 data: 4.7488 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6175 (0.6404) acc1: 84.3750 (82.5280) acc5: 97.1354 (96.6720) time: 0.7017 data: 0.5277 max mem: 64948 Test: Total time: 0:00:06 (0.7131 s / it) * Acc@1 83.590 Acc@5 96.700 loss 0.620 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [407] [ 0/312] eta: 1:00:25 lr: 0.000099 min_lr: 0.000099 loss: 1.4877 (1.4877) weight_decay: 0.0500 (0.0500) time: 11.6201 data: 7.8349 max mem: 64948 Epoch: [407] [ 10/312] eta: 0:08:33 lr: 0.000099 min_lr: 0.000099 loss: 1.7840 (1.6790) weight_decay: 0.0500 (0.0500) time: 1.7009 data: 0.7152 max mem: 64948 Epoch: [407] [ 20/312] eta: 0:05:56 lr: 0.000099 min_lr: 0.000099 loss: 1.6602 (1.6032) weight_decay: 0.0500 (0.0500) time: 0.7023 data: 0.0018 max mem: 64948 Epoch: [407] [ 30/312] eta: 0:04:56 lr: 0.000098 min_lr: 0.000098 loss: 1.5691 (1.5857) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [407] [ 40/312] eta: 0:04:22 lr: 0.000098 min_lr: 0.000098 loss: 1.6206 (1.5819) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [407] [ 50/312] eta: 0:03:59 lr: 0.000098 min_lr: 0.000098 loss: 1.6206 (1.5943) weight_decay: 0.0500 (0.0500) time: 0.7017 data: 0.0004 max mem: 64948 Epoch: [407] [ 60/312] eta: 0:03:41 lr: 0.000098 min_lr: 0.000098 loss: 1.8157 (1.6373) weight_decay: 0.0500 (0.0500) time: 0.6992 data: 0.0004 max mem: 64948 Epoch: [407] [ 70/312] eta: 0:03:26 lr: 0.000098 min_lr: 0.000098 loss: 1.8342 (1.6528) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [407] [ 80/312] eta: 0:03:13 lr: 0.000098 min_lr: 0.000098 loss: 1.8989 (1.6724) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [407] [ 90/312] eta: 0:03:01 lr: 0.000098 min_lr: 0.000098 loss: 1.8547 (1.6715) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [407] [100/312] eta: 0:02:50 lr: 0.000097 min_lr: 0.000097 loss: 1.6801 (1.6774) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [407] [110/312] eta: 0:02:40 lr: 0.000097 min_lr: 0.000097 loss: 1.6675 (1.6693) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [407] [120/312] eta: 0:02:31 lr: 0.000097 min_lr: 0.000097 loss: 1.7009 (1.6767) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0005 max mem: 64948 Epoch: [407] [130/312] eta: 0:02:22 lr: 0.000097 min_lr: 0.000097 loss: 1.8417 (1.6879) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0005 max mem: 64948 Epoch: [407] [140/312] eta: 0:02:13 lr: 0.000097 min_lr: 0.000097 loss: 1.7121 (1.6871) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [407] [150/312] eta: 0:02:04 lr: 0.000097 min_lr: 0.000097 loss: 1.7864 (1.7005) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [407] [160/312] eta: 0:01:56 lr: 0.000097 min_lr: 0.000097 loss: 1.7884 (1.6947) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [407] [170/312] eta: 0:01:48 lr: 0.000096 min_lr: 0.000096 loss: 1.5827 (1.6894) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [407] [180/312] eta: 0:01:39 lr: 0.000096 min_lr: 0.000096 loss: 1.6316 (1.6862) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [407] [190/312] eta: 0:01:31 lr: 0.000096 min_lr: 0.000096 loss: 1.6406 (1.6806) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [407] [200/312] eta: 0:01:24 lr: 0.000096 min_lr: 0.000096 loss: 1.6607 (1.6780) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [407] [210/312] eta: 0:01:16 lr: 0.000096 min_lr: 0.000096 loss: 1.8482 (1.6799) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [407] [220/312] eta: 0:01:08 lr: 0.000096 min_lr: 0.000096 loss: 1.8292 (1.6847) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [407] [230/312] eta: 0:01:00 lr: 0.000096 min_lr: 0.000096 loss: 1.8031 (1.6841) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [407] [240/312] eta: 0:00:53 lr: 0.000095 min_lr: 0.000095 loss: 1.6841 (1.6806) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [407] [250/312] eta: 0:00:45 lr: 0.000095 min_lr: 0.000095 loss: 1.7941 (1.6815) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [407] [260/312] eta: 0:00:38 lr: 0.000095 min_lr: 0.000095 loss: 1.6581 (1.6791) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [407] [270/312] eta: 0:00:30 lr: 0.000095 min_lr: 0.000095 loss: 1.6161 (1.6758) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [407] [280/312] eta: 0:00:23 lr: 0.000095 min_lr: 0.000095 loss: 1.5669 (1.6724) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [407] [290/312] eta: 0:00:16 lr: 0.000095 min_lr: 0.000095 loss: 1.6901 (1.6745) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0008 max mem: 64948 Epoch: [407] [300/312] eta: 0:00:08 lr: 0.000095 min_lr: 0.000095 loss: 1.7067 (1.6703) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [407] [310/312] eta: 0:00:01 lr: 0.000094 min_lr: 0.000094 loss: 1.7475 (1.6740) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [407] [311/312] eta: 0:00:00 lr: 0.000094 min_lr: 0.000094 loss: 1.7475 (1.6743) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [407] Total time: 0:03:48 (0.7339 s / it) Averaged stats: lr: 0.000094 min_lr: 0.000094 loss: 1.7475 (1.6755) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4491 (0.4491) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.7673 data: 4.5491 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6090 (0.6424) acc1: 84.1146 (82.7520) acc5: 97.1354 (96.8000) time: 0.6810 data: 0.5055 max mem: 64948 Test: Total time: 0:00:06 (0.7065 s / it) * Acc@1 83.516 Acc@5 96.662 loss 0.626 Accuracy of the model on the 50000 test images: 83.5% Max accuracy: 83.60% Test: [0/9] eta: 0:00:45 loss: 0.4542 (0.4542) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 5.0146 data: 4.8063 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6170 (0.6402) acc1: 84.3750 (82.5600) acc5: 97.1354 (96.5760) time: 0.7084 data: 0.5341 max mem: 64948 Test: Total time: 0:00:06 (0.7197 s / it) * Acc@1 83.598 Acc@5 96.688 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [408] [ 0/312] eta: 0:56:17 lr: 0.000094 min_lr: 0.000094 loss: 1.6488 (1.6488) weight_decay: 0.0500 (0.0500) time: 10.8253 data: 7.7576 max mem: 64948 Epoch: [408] [ 10/312] eta: 0:08:15 lr: 0.000094 min_lr: 0.000094 loss: 1.7539 (1.6768) weight_decay: 0.0500 (0.0500) time: 1.6411 data: 0.7056 max mem: 64948 Epoch: [408] [ 20/312] eta: 0:05:47 lr: 0.000094 min_lr: 0.000094 loss: 1.7883 (1.7410) weight_decay: 0.0500 (0.0500) time: 0.7090 data: 0.0004 max mem: 64948 Epoch: [408] [ 30/312] eta: 0:04:51 lr: 0.000094 min_lr: 0.000094 loss: 1.7677 (1.7033) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [408] [ 40/312] eta: 0:04:18 lr: 0.000094 min_lr: 0.000094 loss: 1.7235 (1.6688) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [408] [ 50/312] eta: 0:03:56 lr: 0.000094 min_lr: 0.000094 loss: 1.7256 (1.6741) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [408] [ 60/312] eta: 0:03:38 lr: 0.000094 min_lr: 0.000094 loss: 1.7840 (1.6891) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [408] [ 70/312] eta: 0:03:24 lr: 0.000093 min_lr: 0.000093 loss: 1.7603 (1.6814) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [408] [ 80/312] eta: 0:03:11 lr: 0.000093 min_lr: 0.000093 loss: 1.7212 (1.6818) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [408] [ 90/312] eta: 0:02:59 lr: 0.000093 min_lr: 0.000093 loss: 1.7567 (1.6842) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [408] [100/312] eta: 0:02:49 lr: 0.000093 min_lr: 0.000093 loss: 1.7408 (1.6734) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [408] [110/312] eta: 0:02:39 lr: 0.000093 min_lr: 0.000093 loss: 1.7408 (1.6816) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [408] [120/312] eta: 0:02:30 lr: 0.000093 min_lr: 0.000093 loss: 1.7280 (1.6665) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [408] [130/312] eta: 0:02:21 lr: 0.000093 min_lr: 0.000093 loss: 1.7245 (1.6720) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [408] [140/312] eta: 0:02:12 lr: 0.000092 min_lr: 0.000092 loss: 1.7608 (1.6728) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [408] [150/312] eta: 0:02:03 lr: 0.000092 min_lr: 0.000092 loss: 1.6364 (1.6722) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [408] [160/312] eta: 0:01:55 lr: 0.000092 min_lr: 0.000092 loss: 1.5875 (1.6669) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [408] [170/312] eta: 0:01:47 lr: 0.000092 min_lr: 0.000092 loss: 1.6296 (1.6656) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [408] [180/312] eta: 0:01:39 lr: 0.000092 min_lr: 0.000092 loss: 1.7326 (1.6750) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [408] [190/312] eta: 0:01:31 lr: 0.000092 min_lr: 0.000092 loss: 1.8320 (1.6723) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [408] [200/312] eta: 0:01:23 lr: 0.000092 min_lr: 0.000092 loss: 1.7499 (1.6693) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [408] [210/312] eta: 0:01:15 lr: 0.000091 min_lr: 0.000091 loss: 1.6852 (1.6673) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [408] [220/312] eta: 0:01:08 lr: 0.000091 min_lr: 0.000091 loss: 1.7746 (1.6697) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [408] [230/312] eta: 0:01:00 lr: 0.000091 min_lr: 0.000091 loss: 1.8302 (1.6765) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [408] [240/312] eta: 0:00:53 lr: 0.000091 min_lr: 0.000091 loss: 1.7767 (1.6712) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [408] [250/312] eta: 0:00:45 lr: 0.000091 min_lr: 0.000091 loss: 1.7351 (1.6741) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [408] [260/312] eta: 0:00:38 lr: 0.000091 min_lr: 0.000091 loss: 1.8411 (1.6821) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [408] [270/312] eta: 0:00:30 lr: 0.000091 min_lr: 0.000091 loss: 1.7705 (1.6773) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [408] [280/312] eta: 0:00:23 lr: 0.000090 min_lr: 0.000090 loss: 1.6791 (1.6773) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0009 max mem: 64948 Epoch: [408] [290/312] eta: 0:00:16 lr: 0.000090 min_lr: 0.000090 loss: 1.7531 (1.6809) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0008 max mem: 64948 Epoch: [408] [300/312] eta: 0:00:08 lr: 0.000090 min_lr: 0.000090 loss: 1.8017 (1.6802) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [408] [310/312] eta: 0:00:01 lr: 0.000090 min_lr: 0.000090 loss: 1.6435 (1.6759) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [408] [311/312] eta: 0:00:00 lr: 0.000090 min_lr: 0.000090 loss: 1.7321 (1.6768) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [408] Total time: 0:03:48 (0.7312 s / it) Averaged stats: lr: 0.000090 min_lr: 0.000090 loss: 1.7321 (1.6659) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:47 loss: 0.4457 (0.4457) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 5.2747 data: 5.0688 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6231 (0.6460) acc1: 84.6354 (82.8480) acc5: 97.3958 (96.9280) time: 0.7373 data: 0.5633 max mem: 64948 Test: Total time: 0:00:06 (0.7594 s / it) * Acc@1 83.580 Acc@5 96.688 loss 0.627 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.60% Test: [0/9] eta: 0:00:45 loss: 0.4540 (0.4540) acc1: 87.2396 (87.2396) acc5: 97.6562 (97.6562) time: 5.1103 data: 4.9041 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6167 (0.6400) acc1: 84.3750 (82.5920) acc5: 97.1354 (96.6080) time: 0.7192 data: 0.5450 max mem: 64948 Test: Total time: 0:00:06 (0.7381 s / it) * Acc@1 83.602 Acc@5 96.696 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.60% Epoch: [409] [ 0/312] eta: 0:45:23 lr: 0.000090 min_lr: 0.000090 loss: 1.8913 (1.8913) weight_decay: 0.0500 (0.0500) time: 8.7281 data: 7.9894 max mem: 64948 Epoch: [409] [ 10/312] eta: 0:07:17 lr: 0.000090 min_lr: 0.000090 loss: 1.7178 (1.6354) weight_decay: 0.0500 (0.0500) time: 1.4495 data: 0.7267 max mem: 64948 Epoch: [409] [ 20/312] eta: 0:05:18 lr: 0.000090 min_lr: 0.000090 loss: 1.6949 (1.6191) weight_decay: 0.0500 (0.0500) time: 0.7082 data: 0.0004 max mem: 64948 Epoch: [409] [ 30/312] eta: 0:04:31 lr: 0.000090 min_lr: 0.000090 loss: 1.7722 (1.6789) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [409] [ 40/312] eta: 0:04:03 lr: 0.000089 min_lr: 0.000089 loss: 1.8113 (1.6949) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [409] [ 50/312] eta: 0:03:44 lr: 0.000089 min_lr: 0.000089 loss: 1.7560 (1.6917) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [409] [ 60/312] eta: 0:03:29 lr: 0.000089 min_lr: 0.000089 loss: 1.4487 (1.6441) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [409] [ 70/312] eta: 0:03:16 lr: 0.000089 min_lr: 0.000089 loss: 1.4548 (1.6428) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [409] [ 80/312] eta: 0:03:05 lr: 0.000089 min_lr: 0.000089 loss: 1.5756 (1.6405) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [409] [ 90/312] eta: 0:02:54 lr: 0.000089 min_lr: 0.000089 loss: 1.5911 (1.6411) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [409] [100/312] eta: 0:02:44 lr: 0.000089 min_lr: 0.000089 loss: 1.7264 (1.6534) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [409] [110/312] eta: 0:02:35 lr: 0.000089 min_lr: 0.000089 loss: 1.6623 (1.6390) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [409] [120/312] eta: 0:02:26 lr: 0.000088 min_lr: 0.000088 loss: 1.4711 (1.6378) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [409] [130/312] eta: 0:02:18 lr: 0.000088 min_lr: 0.000088 loss: 1.6414 (1.6339) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [409] [140/312] eta: 0:02:09 lr: 0.000088 min_lr: 0.000088 loss: 1.7133 (1.6396) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [409] [150/312] eta: 0:02:01 lr: 0.000088 min_lr: 0.000088 loss: 1.7675 (1.6331) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [409] [160/312] eta: 0:01:53 lr: 0.000088 min_lr: 0.000088 loss: 1.7048 (1.6420) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [409] [170/312] eta: 0:01:45 lr: 0.000088 min_lr: 0.000088 loss: 1.7048 (1.6359) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [409] [180/312] eta: 0:01:37 lr: 0.000088 min_lr: 0.000088 loss: 1.5240 (1.6369) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [409] [190/312] eta: 0:01:30 lr: 0.000087 min_lr: 0.000087 loss: 1.6987 (1.6386) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [409] [200/312] eta: 0:01:22 lr: 0.000087 min_lr: 0.000087 loss: 1.6845 (1.6390) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [409] [210/312] eta: 0:01:14 lr: 0.000087 min_lr: 0.000087 loss: 1.6316 (1.6363) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [409] [220/312] eta: 0:01:07 lr: 0.000087 min_lr: 0.000087 loss: 1.6924 (1.6426) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [409] [230/312] eta: 0:00:59 lr: 0.000087 min_lr: 0.000087 loss: 1.7665 (1.6509) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [409] [240/312] eta: 0:00:52 lr: 0.000087 min_lr: 0.000087 loss: 1.7665 (1.6539) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [409] [250/312] eta: 0:00:45 lr: 0.000087 min_lr: 0.000087 loss: 1.6404 (1.6509) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [409] [260/312] eta: 0:00:37 lr: 0.000086 min_lr: 0.000086 loss: 1.6659 (1.6561) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [409] [270/312] eta: 0:00:30 lr: 0.000086 min_lr: 0.000086 loss: 1.8657 (1.6622) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [409] [280/312] eta: 0:00:23 lr: 0.000086 min_lr: 0.000086 loss: 1.8657 (1.6663) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0011 max mem: 64948 Epoch: [409] [290/312] eta: 0:00:15 lr: 0.000086 min_lr: 0.000086 loss: 1.8612 (1.6715) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0010 max mem: 64948 Epoch: [409] [300/312] eta: 0:00:08 lr: 0.000086 min_lr: 0.000086 loss: 1.8064 (1.6753) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [409] [310/312] eta: 0:00:01 lr: 0.000086 min_lr: 0.000086 loss: 1.7459 (1.6760) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [409] [311/312] eta: 0:00:00 lr: 0.000086 min_lr: 0.000086 loss: 1.7540 (1.6763) weight_decay: 0.0500 (0.0500) time: 0.6919 data: 0.0001 max mem: 64948 Epoch: [409] Total time: 0:03:46 (0.7247 s / it) Averaged stats: lr: 0.000086 min_lr: 0.000086 loss: 1.7540 (1.6691) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4412 (0.4412) acc1: 88.8021 (88.8021) acc5: 98.4375 (98.4375) time: 4.6302 data: 4.4250 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6139 (0.6425) acc1: 84.3750 (82.6560) acc5: 97.1354 (96.8960) time: 0.6657 data: 0.4918 max mem: 64948 Test: Total time: 0:00:06 (0.6866 s / it) * Acc@1 83.584 Acc@5 96.672 loss 0.624 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.60% Test: [0/9] eta: 0:00:44 loss: 0.4536 (0.4536) acc1: 87.5000 (87.5000) acc5: 97.6562 (97.6562) time: 4.8984 data: 4.6923 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6164 (0.6398) acc1: 84.3750 (82.6880) acc5: 97.1354 (96.6080) time: 0.6956 data: 0.5215 max mem: 64948 Test: Total time: 0:00:06 (0.7057 s / it) * Acc@1 83.608 Acc@5 96.692 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.61% Epoch: [410] [ 0/312] eta: 0:45:38 lr: 0.000086 min_lr: 0.000086 loss: 0.9899 (0.9899) weight_decay: 0.0500 (0.0500) time: 8.7759 data: 7.7805 max mem: 64948 Epoch: [410] [ 10/312] eta: 0:07:22 lr: 0.000086 min_lr: 0.000086 loss: 1.8139 (1.7172) weight_decay: 0.0500 (0.0500) time: 1.4660 data: 0.7077 max mem: 64948 Epoch: [410] [ 20/312] eta: 0:05:20 lr: 0.000086 min_lr: 0.000086 loss: 1.7711 (1.7258) weight_decay: 0.0500 (0.0500) time: 0.7144 data: 0.0004 max mem: 64948 Epoch: [410] [ 30/312] eta: 0:04:33 lr: 0.000085 min_lr: 0.000085 loss: 1.7667 (1.7417) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [410] [ 40/312] eta: 0:04:05 lr: 0.000085 min_lr: 0.000085 loss: 1.6782 (1.6940) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [410] [ 50/312] eta: 0:03:46 lr: 0.000085 min_lr: 0.000085 loss: 1.6559 (1.6861) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [410] [ 60/312] eta: 0:03:30 lr: 0.000085 min_lr: 0.000085 loss: 1.6462 (1.6692) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [410] [ 70/312] eta: 0:03:17 lr: 0.000085 min_lr: 0.000085 loss: 1.7378 (1.6689) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [410] [ 80/312] eta: 0:03:05 lr: 0.000085 min_lr: 0.000085 loss: 1.8230 (1.6940) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [410] [ 90/312] eta: 0:02:55 lr: 0.000085 min_lr: 0.000085 loss: 1.8082 (1.6862) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [410] [100/312] eta: 0:02:45 lr: 0.000084 min_lr: 0.000084 loss: 1.7499 (1.6865) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [410] [110/312] eta: 0:02:36 lr: 0.000084 min_lr: 0.000084 loss: 1.7265 (1.6909) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [410] [120/312] eta: 0:02:27 lr: 0.000084 min_lr: 0.000084 loss: 1.6684 (1.6885) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [410] [130/312] eta: 0:02:18 lr: 0.000084 min_lr: 0.000084 loss: 1.6640 (1.6891) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [410] [140/312] eta: 0:02:10 lr: 0.000084 min_lr: 0.000084 loss: 1.6641 (1.6855) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [410] [150/312] eta: 0:02:01 lr: 0.000084 min_lr: 0.000084 loss: 1.5018 (1.6809) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [410] [160/312] eta: 0:01:53 lr: 0.000084 min_lr: 0.000084 loss: 1.5834 (1.6809) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [410] [170/312] eta: 0:01:45 lr: 0.000083 min_lr: 0.000083 loss: 1.5695 (1.6687) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [410] [180/312] eta: 0:01:37 lr: 0.000083 min_lr: 0.000083 loss: 1.5511 (1.6735) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [410] [190/312] eta: 0:01:30 lr: 0.000083 min_lr: 0.000083 loss: 1.5996 (1.6659) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [410] [200/312] eta: 0:01:22 lr: 0.000083 min_lr: 0.000083 loss: 1.5996 (1.6649) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [410] [210/312] eta: 0:01:15 lr: 0.000083 min_lr: 0.000083 loss: 1.6824 (1.6624) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [410] [220/312] eta: 0:01:07 lr: 0.000083 min_lr: 0.000083 loss: 1.7344 (1.6654) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0005 max mem: 64948 Epoch: [410] [230/312] eta: 0:01:00 lr: 0.000083 min_lr: 0.000083 loss: 1.6989 (1.6620) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [410] [240/312] eta: 0:00:52 lr: 0.000083 min_lr: 0.000083 loss: 1.6989 (1.6614) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [410] [250/312] eta: 0:00:45 lr: 0.000082 min_lr: 0.000082 loss: 1.6832 (1.6596) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [410] [260/312] eta: 0:00:37 lr: 0.000082 min_lr: 0.000082 loss: 1.7131 (1.6643) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [410] [270/312] eta: 0:00:30 lr: 0.000082 min_lr: 0.000082 loss: 1.7131 (1.6644) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [410] [280/312] eta: 0:00:23 lr: 0.000082 min_lr: 0.000082 loss: 1.7027 (1.6668) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0006 max mem: 64948 Epoch: [410] [290/312] eta: 0:00:15 lr: 0.000082 min_lr: 0.000082 loss: 1.7027 (1.6673) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0005 max mem: 64948 Epoch: [410] [300/312] eta: 0:00:08 lr: 0.000082 min_lr: 0.000082 loss: 1.8525 (1.6746) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [410] [310/312] eta: 0:00:01 lr: 0.000082 min_lr: 0.000082 loss: 1.8525 (1.6748) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [410] [311/312] eta: 0:00:00 lr: 0.000082 min_lr: 0.000082 loss: 1.8237 (1.6740) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [410] Total time: 0:03:46 (0.7254 s / it) Averaged stats: lr: 0.000082 min_lr: 0.000082 loss: 1.8237 (1.6643) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4416 (0.4416) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.6547 data: 4.4506 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6053 (0.6422) acc1: 83.5938 (82.7200) acc5: 97.6562 (96.8320) time: 0.6685 data: 0.4946 max mem: 64948 Test: Total time: 0:00:06 (0.6907 s / it) * Acc@1 83.594 Acc@5 96.658 loss 0.623 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.60% Test: [0/9] eta: 0:00:46 loss: 0.4534 (0.4534) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 5.1983 data: 4.9803 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6162 (0.6398) acc1: 84.3750 (82.7200) acc5: 97.1354 (96.6080) time: 0.7294 data: 0.5535 max mem: 64948 Test: Total time: 0:00:06 (0.7391 s / it) * Acc@1 83.612 Acc@5 96.684 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.61% Epoch: [411] [ 0/312] eta: 0:47:16 lr: 0.000082 min_lr: 0.000082 loss: 1.8907 (1.8907) weight_decay: 0.0500 (0.0500) time: 9.0922 data: 7.9276 max mem: 64948 Epoch: [411] [ 10/312] eta: 0:07:28 lr: 0.000081 min_lr: 0.000081 loss: 1.5935 (1.5663) weight_decay: 0.0500 (0.0500) time: 1.4841 data: 0.7212 max mem: 64948 Epoch: [411] [ 20/312] eta: 0:05:23 lr: 0.000081 min_lr: 0.000081 loss: 1.5935 (1.5667) weight_decay: 0.0500 (0.0500) time: 0.7104 data: 0.0004 max mem: 64948 Epoch: [411] [ 30/312] eta: 0:04:35 lr: 0.000081 min_lr: 0.000081 loss: 1.7756 (1.6350) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [411] [ 40/312] eta: 0:04:06 lr: 0.000081 min_lr: 0.000081 loss: 1.8244 (1.6445) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [411] [ 50/312] eta: 0:03:46 lr: 0.000081 min_lr: 0.000081 loss: 1.6848 (1.6469) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [411] [ 60/312] eta: 0:03:31 lr: 0.000081 min_lr: 0.000081 loss: 1.6453 (1.6379) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [411] [ 70/312] eta: 0:03:17 lr: 0.000081 min_lr: 0.000081 loss: 1.6332 (1.6407) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [411] [ 80/312] eta: 0:03:06 lr: 0.000081 min_lr: 0.000081 loss: 1.5942 (1.6304) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [411] [ 90/312] eta: 0:02:55 lr: 0.000080 min_lr: 0.000080 loss: 1.5343 (1.6168) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [411] [100/312] eta: 0:02:45 lr: 0.000080 min_lr: 0.000080 loss: 1.7450 (1.6313) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [411] [110/312] eta: 0:02:36 lr: 0.000080 min_lr: 0.000080 loss: 1.7810 (1.6325) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [411] [120/312] eta: 0:02:27 lr: 0.000080 min_lr: 0.000080 loss: 1.7114 (1.6389) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [411] [130/312] eta: 0:02:18 lr: 0.000080 min_lr: 0.000080 loss: 1.7376 (1.6444) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [411] [140/312] eta: 0:02:10 lr: 0.000080 min_lr: 0.000080 loss: 1.7937 (1.6537) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [411] [150/312] eta: 0:02:02 lr: 0.000080 min_lr: 0.000080 loss: 1.7980 (1.6602) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [411] [160/312] eta: 0:01:53 lr: 0.000080 min_lr: 0.000080 loss: 1.7797 (1.6671) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [411] [170/312] eta: 0:01:45 lr: 0.000079 min_lr: 0.000079 loss: 1.7582 (1.6688) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [411] [180/312] eta: 0:01:38 lr: 0.000079 min_lr: 0.000079 loss: 1.7357 (1.6644) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [411] [190/312] eta: 0:01:30 lr: 0.000079 min_lr: 0.000079 loss: 1.7357 (1.6649) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [411] [200/312] eta: 0:01:22 lr: 0.000079 min_lr: 0.000079 loss: 1.8496 (1.6652) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [411] [210/312] eta: 0:01:15 lr: 0.000079 min_lr: 0.000079 loss: 1.8695 (1.6704) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [411] [220/312] eta: 0:01:07 lr: 0.000079 min_lr: 0.000079 loss: 1.8565 (1.6717) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [411] [230/312] eta: 0:01:00 lr: 0.000079 min_lr: 0.000079 loss: 1.6735 (1.6615) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [411] [240/312] eta: 0:00:52 lr: 0.000078 min_lr: 0.000078 loss: 1.5265 (1.6592) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [411] [250/312] eta: 0:00:45 lr: 0.000078 min_lr: 0.000078 loss: 1.6194 (1.6608) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [411] [260/312] eta: 0:00:37 lr: 0.000078 min_lr: 0.000078 loss: 1.7193 (1.6645) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [411] [270/312] eta: 0:00:30 lr: 0.000078 min_lr: 0.000078 loss: 1.8602 (1.6736) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [411] [280/312] eta: 0:00:23 lr: 0.000078 min_lr: 0.000078 loss: 1.8303 (1.6746) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [411] [290/312] eta: 0:00:15 lr: 0.000078 min_lr: 0.000078 loss: 1.8303 (1.6789) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0008 max mem: 64948 Epoch: [411] [300/312] eta: 0:00:08 lr: 0.000078 min_lr: 0.000078 loss: 1.7180 (1.6755) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [411] [310/312] eta: 0:00:01 lr: 0.000078 min_lr: 0.000078 loss: 1.7180 (1.6810) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [411] [311/312] eta: 0:00:00 lr: 0.000078 min_lr: 0.000078 loss: 1.7180 (1.6823) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [411] Total time: 0:03:46 (0.7257 s / it) Averaged stats: lr: 0.000078 min_lr: 0.000078 loss: 1.7180 (1.6670) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4367 (0.4367) acc1: 89.5833 (89.5833) acc5: 98.1771 (98.1771) time: 4.6836 data: 4.4780 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6209 (0.6450) acc1: 84.3750 (83.0720) acc5: 97.1354 (96.8320) time: 0.6716 data: 0.4977 max mem: 64948 Test: Total time: 0:00:06 (0.6949 s / it) * Acc@1 83.576 Acc@5 96.654 loss 0.623 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.60% Test: [0/9] eta: 0:00:42 loss: 0.4530 (0.4530) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.7129 data: 4.5063 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6158 (0.6395) acc1: 84.3750 (82.7520) acc5: 96.8750 (96.5440) time: 0.6750 data: 0.5008 max mem: 64948 Test: Total time: 0:00:06 (0.6833 s / it) * Acc@1 83.618 Acc@5 96.696 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.62% Epoch: [412] [ 0/312] eta: 0:44:57 lr: 0.000078 min_lr: 0.000078 loss: 1.5866 (1.5866) weight_decay: 0.0500 (0.0500) time: 8.6455 data: 7.4428 max mem: 64948 Epoch: [412] [ 10/312] eta: 0:07:17 lr: 0.000077 min_lr: 0.000077 loss: 1.7503 (1.7047) weight_decay: 0.0500 (0.0500) time: 1.4483 data: 0.6770 max mem: 64948 Epoch: [412] [ 20/312] eta: 0:05:17 lr: 0.000077 min_lr: 0.000077 loss: 1.7503 (1.6659) weight_decay: 0.0500 (0.0500) time: 0.7111 data: 0.0004 max mem: 64948 Epoch: [412] [ 30/312] eta: 0:04:31 lr: 0.000077 min_lr: 0.000077 loss: 1.7563 (1.6807) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [412] [ 40/312] eta: 0:04:03 lr: 0.000077 min_lr: 0.000077 loss: 1.7897 (1.6692) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [412] [ 50/312] eta: 0:03:44 lr: 0.000077 min_lr: 0.000077 loss: 1.7170 (1.6578) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [412] [ 60/312] eta: 0:03:29 lr: 0.000077 min_lr: 0.000077 loss: 1.5577 (1.6435) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [412] [ 70/312] eta: 0:03:16 lr: 0.000077 min_lr: 0.000077 loss: 1.6721 (1.6456) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [412] [ 80/312] eta: 0:03:05 lr: 0.000077 min_lr: 0.000077 loss: 1.7271 (1.6534) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [412] [ 90/312] eta: 0:02:54 lr: 0.000076 min_lr: 0.000076 loss: 1.6986 (1.6594) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [412] [100/312] eta: 0:02:44 lr: 0.000076 min_lr: 0.000076 loss: 1.7220 (1.6593) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [412] [110/312] eta: 0:02:35 lr: 0.000076 min_lr: 0.000076 loss: 1.7472 (1.6590) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [412] [120/312] eta: 0:02:26 lr: 0.000076 min_lr: 0.000076 loss: 1.7465 (1.6682) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [412] [130/312] eta: 0:02:18 lr: 0.000076 min_lr: 0.000076 loss: 1.7816 (1.6655) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [412] [140/312] eta: 0:02:09 lr: 0.000076 min_lr: 0.000076 loss: 1.7816 (1.6613) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [412] [150/312] eta: 0:02:01 lr: 0.000076 min_lr: 0.000076 loss: 1.7060 (1.6594) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [412] [160/312] eta: 0:01:53 lr: 0.000076 min_lr: 0.000076 loss: 1.7200 (1.6601) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [412] [170/312] eta: 0:01:45 lr: 0.000075 min_lr: 0.000075 loss: 1.7200 (1.6597) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [412] [180/312] eta: 0:01:37 lr: 0.000075 min_lr: 0.000075 loss: 1.6956 (1.6589) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [412] [190/312] eta: 0:01:30 lr: 0.000075 min_lr: 0.000075 loss: 1.7139 (1.6609) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [412] [200/312] eta: 0:01:22 lr: 0.000075 min_lr: 0.000075 loss: 1.6020 (1.6527) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [412] [210/312] eta: 0:01:14 lr: 0.000075 min_lr: 0.000075 loss: 1.5949 (1.6469) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [412] [220/312] eta: 0:01:07 lr: 0.000075 min_lr: 0.000075 loss: 1.6153 (1.6449) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [412] [230/312] eta: 0:00:59 lr: 0.000075 min_lr: 0.000075 loss: 1.6742 (1.6519) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [412] [240/312] eta: 0:00:52 lr: 0.000075 min_lr: 0.000075 loss: 1.8503 (1.6565) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [412] [250/312] eta: 0:00:45 lr: 0.000074 min_lr: 0.000074 loss: 1.5172 (1.6493) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [412] [260/312] eta: 0:00:37 lr: 0.000074 min_lr: 0.000074 loss: 1.5136 (1.6477) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [412] [270/312] eta: 0:00:30 lr: 0.000074 min_lr: 0.000074 loss: 1.5259 (1.6424) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [412] [280/312] eta: 0:00:23 lr: 0.000074 min_lr: 0.000074 loss: 1.6369 (1.6434) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [412] [290/312] eta: 0:00:15 lr: 0.000074 min_lr: 0.000074 loss: 1.6538 (1.6430) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [412] [300/312] eta: 0:00:08 lr: 0.000074 min_lr: 0.000074 loss: 1.6551 (1.6426) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [412] [310/312] eta: 0:00:01 lr: 0.000074 min_lr: 0.000074 loss: 1.7674 (1.6485) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [412] [311/312] eta: 0:00:00 lr: 0.000074 min_lr: 0.000074 loss: 1.7674 (1.6486) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [412] Total time: 0:03:45 (0.7243 s / it) Averaged stats: lr: 0.000074 min_lr: 0.000074 loss: 1.7674 (1.6598) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4435 (0.4435) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 4.7659 data: 4.5500 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6186 (0.6413) acc1: 83.8542 (82.8800) acc5: 97.1354 (96.6720) time: 0.6808 data: 0.5056 max mem: 64948 Test: Total time: 0:00:06 (0.7030 s / it) * Acc@1 83.608 Acc@5 96.668 loss 0.625 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.61% Test: [0/9] eta: 0:00:40 loss: 0.4526 (0.4526) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.5239 data: 4.3060 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6155 (0.6394) acc1: 84.3750 (82.7200) acc5: 96.8750 (96.5440) time: 0.6539 data: 0.4785 max mem: 64948 Test: Total time: 0:00:05 (0.6629 s / it) * Acc@1 83.608 Acc@5 96.704 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [413] [ 0/312] eta: 0:57:00 lr: 0.000074 min_lr: 0.000074 loss: 2.0076 (2.0076) weight_decay: 0.0500 (0.0500) time: 10.9637 data: 6.9906 max mem: 64948 Epoch: [413] [ 10/312] eta: 0:08:29 lr: 0.000073 min_lr: 0.000073 loss: 1.7464 (1.6076) weight_decay: 0.0500 (0.0500) time: 1.6874 data: 0.6359 max mem: 64948 Epoch: [413] [ 20/312] eta: 0:05:55 lr: 0.000073 min_lr: 0.000073 loss: 1.7270 (1.6479) weight_decay: 0.0500 (0.0500) time: 0.7299 data: 0.0004 max mem: 64948 Epoch: [413] [ 30/312] eta: 0:04:56 lr: 0.000073 min_lr: 0.000073 loss: 1.6824 (1.6478) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [413] [ 40/312] eta: 0:04:22 lr: 0.000073 min_lr: 0.000073 loss: 1.5523 (1.6296) weight_decay: 0.0500 (0.0500) time: 0.6993 data: 0.0004 max mem: 64948 Epoch: [413] [ 50/312] eta: 0:03:58 lr: 0.000073 min_lr: 0.000073 loss: 1.5523 (1.6144) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [413] [ 60/312] eta: 0:03:40 lr: 0.000073 min_lr: 0.000073 loss: 1.7749 (1.6501) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [413] [ 70/312] eta: 0:03:25 lr: 0.000073 min_lr: 0.000073 loss: 1.8535 (1.6776) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [413] [ 80/312] eta: 0:03:13 lr: 0.000073 min_lr: 0.000073 loss: 1.8321 (1.6701) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [413] [ 90/312] eta: 0:03:01 lr: 0.000072 min_lr: 0.000072 loss: 1.8198 (1.6791) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [413] [100/312] eta: 0:02:50 lr: 0.000072 min_lr: 0.000072 loss: 1.6329 (1.6587) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [413] [110/312] eta: 0:02:40 lr: 0.000072 min_lr: 0.000072 loss: 1.5366 (1.6424) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [413] [120/312] eta: 0:02:30 lr: 0.000072 min_lr: 0.000072 loss: 1.7298 (1.6584) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [413] [130/312] eta: 0:02:21 lr: 0.000072 min_lr: 0.000072 loss: 1.7934 (1.6493) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [413] [140/312] eta: 0:02:13 lr: 0.000072 min_lr: 0.000072 loss: 1.7130 (1.6469) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [413] [150/312] eta: 0:02:04 lr: 0.000072 min_lr: 0.000072 loss: 1.7730 (1.6521) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [413] [160/312] eta: 0:01:56 lr: 0.000072 min_lr: 0.000072 loss: 1.8054 (1.6594) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [413] [170/312] eta: 0:01:47 lr: 0.000072 min_lr: 0.000072 loss: 1.7090 (1.6533) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0003 max mem: 64948 Epoch: [413] [180/312] eta: 0:01:39 lr: 0.000071 min_lr: 0.000071 loss: 1.5898 (1.6506) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [413] [190/312] eta: 0:01:31 lr: 0.000071 min_lr: 0.000071 loss: 1.6795 (1.6509) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [413] [200/312] eta: 0:01:24 lr: 0.000071 min_lr: 0.000071 loss: 1.6315 (1.6481) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [413] [210/312] eta: 0:01:16 lr: 0.000071 min_lr: 0.000071 loss: 1.7468 (1.6559) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [413] [220/312] eta: 0:01:08 lr: 0.000071 min_lr: 0.000071 loss: 1.7944 (1.6601) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [413] [230/312] eta: 0:01:00 lr: 0.000071 min_lr: 0.000071 loss: 1.6734 (1.6575) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [413] [240/312] eta: 0:00:53 lr: 0.000071 min_lr: 0.000071 loss: 1.7145 (1.6589) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [413] [250/312] eta: 0:00:45 lr: 0.000071 min_lr: 0.000071 loss: 1.7502 (1.6597) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [413] [260/312] eta: 0:00:38 lr: 0.000070 min_lr: 0.000070 loss: 1.7405 (1.6629) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [413] [270/312] eta: 0:00:30 lr: 0.000070 min_lr: 0.000070 loss: 1.6968 (1.6568) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [413] [280/312] eta: 0:00:23 lr: 0.000070 min_lr: 0.000070 loss: 1.5468 (1.6540) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0010 max mem: 64948 Epoch: [413] [290/312] eta: 0:00:16 lr: 0.000070 min_lr: 0.000070 loss: 1.7335 (1.6566) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [413] [300/312] eta: 0:00:08 lr: 0.000070 min_lr: 0.000070 loss: 1.8076 (1.6569) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [413] [310/312] eta: 0:00:01 lr: 0.000070 min_lr: 0.000070 loss: 1.4970 (1.6515) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [413] [311/312] eta: 0:00:00 lr: 0.000070 min_lr: 0.000070 loss: 1.5278 (1.6517) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [413] Total time: 0:03:48 (0.7325 s / it) Averaged stats: lr: 0.000070 min_lr: 0.000070 loss: 1.5278 (1.6634) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4451 (0.4451) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 4.6089 data: 4.3859 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6101 (0.6409) acc1: 84.3750 (83.1360) acc5: 97.3958 (96.8640) time: 0.6638 data: 0.4874 max mem: 64948 Test: Total time: 0:00:06 (0.6852 s / it) * Acc@1 83.704 Acc@5 96.694 loss 0.624 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.70% Test: [0/9] eta: 0:00:39 loss: 0.4524 (0.4524) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.4046 data: 4.1910 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6151 (0.6391) acc1: 84.3750 (82.7200) acc5: 96.8750 (96.5440) time: 0.6407 data: 0.4658 max mem: 64948 Test: Total time: 0:00:05 (0.6482 s / it) * Acc@1 83.620 Acc@5 96.710 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.62% Epoch: [414] [ 0/312] eta: 0:51:29 lr: 0.000070 min_lr: 0.000070 loss: 1.2758 (1.2758) weight_decay: 0.0500 (0.0500) time: 9.9037 data: 9.1013 max mem: 64948 Epoch: [414] [ 10/312] eta: 0:07:52 lr: 0.000070 min_lr: 0.000070 loss: 1.7848 (1.5989) weight_decay: 0.0500 (0.0500) time: 1.5645 data: 0.8418 max mem: 64948 Epoch: [414] [ 20/312] eta: 0:05:36 lr: 0.000070 min_lr: 0.000070 loss: 1.6107 (1.5796) weight_decay: 0.0500 (0.0500) time: 0.7138 data: 0.0081 max mem: 64948 Epoch: [414] [ 30/312] eta: 0:04:42 lr: 0.000069 min_lr: 0.000069 loss: 1.6210 (1.6429) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [414] [ 40/312] eta: 0:04:12 lr: 0.000069 min_lr: 0.000069 loss: 1.7964 (1.6202) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0003 max mem: 64948 Epoch: [414] [ 50/312] eta: 0:03:51 lr: 0.000069 min_lr: 0.000069 loss: 1.4799 (1.6139) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [414] [ 60/312] eta: 0:03:34 lr: 0.000069 min_lr: 0.000069 loss: 1.7460 (1.6372) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [414] [ 70/312] eta: 0:03:20 lr: 0.000069 min_lr: 0.000069 loss: 1.7696 (1.6516) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [414] [ 80/312] eta: 0:03:08 lr: 0.000069 min_lr: 0.000069 loss: 1.7408 (1.6481) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [414] [ 90/312] eta: 0:02:57 lr: 0.000069 min_lr: 0.000069 loss: 1.7378 (1.6688) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [414] [100/312] eta: 0:02:47 lr: 0.000069 min_lr: 0.000069 loss: 1.8762 (1.6792) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [414] [110/312] eta: 0:02:37 lr: 0.000068 min_lr: 0.000068 loss: 1.7356 (1.6787) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [414] [120/312] eta: 0:02:28 lr: 0.000068 min_lr: 0.000068 loss: 1.7339 (1.6790) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [414] [130/312] eta: 0:02:19 lr: 0.000068 min_lr: 0.000068 loss: 1.5974 (1.6669) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [414] [140/312] eta: 0:02:11 lr: 0.000068 min_lr: 0.000068 loss: 1.5890 (1.6571) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [414] [150/312] eta: 0:02:02 lr: 0.000068 min_lr: 0.000068 loss: 1.6502 (1.6537) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [414] [160/312] eta: 0:01:54 lr: 0.000068 min_lr: 0.000068 loss: 1.6200 (1.6489) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [414] [170/312] eta: 0:01:46 lr: 0.000068 min_lr: 0.000068 loss: 1.7033 (1.6509) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [414] [180/312] eta: 0:01:38 lr: 0.000068 min_lr: 0.000068 loss: 1.7168 (1.6512) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [414] [190/312] eta: 0:01:30 lr: 0.000067 min_lr: 0.000067 loss: 1.7214 (1.6519) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [414] [200/312] eta: 0:01:23 lr: 0.000067 min_lr: 0.000067 loss: 1.7622 (1.6539) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [414] [210/312] eta: 0:01:15 lr: 0.000067 min_lr: 0.000067 loss: 1.6836 (1.6518) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [414] [220/312] eta: 0:01:07 lr: 0.000067 min_lr: 0.000067 loss: 1.6226 (1.6465) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [414] [230/312] eta: 0:01:00 lr: 0.000067 min_lr: 0.000067 loss: 1.5988 (1.6479) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [414] [240/312] eta: 0:00:52 lr: 0.000067 min_lr: 0.000067 loss: 1.6944 (1.6469) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [414] [250/312] eta: 0:00:45 lr: 0.000067 min_lr: 0.000067 loss: 1.8105 (1.6489) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [414] [260/312] eta: 0:00:38 lr: 0.000067 min_lr: 0.000067 loss: 1.8160 (1.6543) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [414] [270/312] eta: 0:00:30 lr: 0.000067 min_lr: 0.000067 loss: 1.7878 (1.6518) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [414] [280/312] eta: 0:00:23 lr: 0.000066 min_lr: 0.000066 loss: 1.5894 (1.6541) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0009 max mem: 64948 Epoch: [414] [290/312] eta: 0:00:16 lr: 0.000066 min_lr: 0.000066 loss: 1.5999 (1.6507) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [414] [300/312] eta: 0:00:08 lr: 0.000066 min_lr: 0.000066 loss: 1.5999 (1.6475) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [414] [310/312] eta: 0:00:01 lr: 0.000066 min_lr: 0.000066 loss: 1.6924 (1.6486) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0001 max mem: 64948 Epoch: [414] [311/312] eta: 0:00:00 lr: 0.000066 min_lr: 0.000066 loss: 1.6924 (1.6492) weight_decay: 0.0500 (0.0500) time: 0.6923 data: 0.0001 max mem: 64948 Epoch: [414] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.000066 min_lr: 0.000066 loss: 1.6924 (1.6686) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4460 (0.4460) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 4.6553 data: 4.4471 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6113 (0.6423) acc1: 84.6354 (83.2320) acc5: 97.3958 (96.8320) time: 0.6685 data: 0.4942 max mem: 64948 Test: Total time: 0:00:06 (0.6911 s / it) * Acc@1 83.648 Acc@5 96.670 loss 0.625 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.70% Test: [0/9] eta: 0:00:45 loss: 0.4521 (0.4521) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 5.0456 data: 4.8277 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6148 (0.6389) acc1: 84.3750 (82.8480) acc5: 96.8750 (96.5440) time: 0.7122 data: 0.5365 max mem: 64948 Test: Total time: 0:00:06 (0.7212 s / it) * Acc@1 83.630 Acc@5 96.710 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.63% Epoch: [415] [ 0/312] eta: 0:45:43 lr: 0.000066 min_lr: 0.000066 loss: 2.0279 (2.0279) weight_decay: 0.0500 (0.0500) time: 8.7927 data: 7.9321 max mem: 64948 Epoch: [415] [ 10/312] eta: 0:07:43 lr: 0.000066 min_lr: 0.000066 loss: 1.7529 (1.6924) weight_decay: 0.0500 (0.0500) time: 1.5357 data: 0.7587 max mem: 64948 Epoch: [415] [ 20/312] eta: 0:05:31 lr: 0.000066 min_lr: 0.000066 loss: 1.7749 (1.7680) weight_decay: 0.0500 (0.0500) time: 0.7518 data: 0.0208 max mem: 64948 Epoch: [415] [ 30/312] eta: 0:04:40 lr: 0.000066 min_lr: 0.000066 loss: 1.8375 (1.7438) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [415] [ 40/312] eta: 0:04:10 lr: 0.000066 min_lr: 0.000066 loss: 1.6896 (1.7359) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [415] [ 50/312] eta: 0:03:49 lr: 0.000065 min_lr: 0.000065 loss: 1.7936 (1.7296) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [415] [ 60/312] eta: 0:03:33 lr: 0.000065 min_lr: 0.000065 loss: 1.7723 (1.7186) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [415] [ 70/312] eta: 0:03:19 lr: 0.000065 min_lr: 0.000065 loss: 1.7380 (1.7058) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [415] [ 80/312] eta: 0:03:07 lr: 0.000065 min_lr: 0.000065 loss: 1.6846 (1.6917) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [415] [ 90/312] eta: 0:02:57 lr: 0.000065 min_lr: 0.000065 loss: 1.6684 (1.6843) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [415] [100/312] eta: 0:02:46 lr: 0.000065 min_lr: 0.000065 loss: 1.7681 (1.6959) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [415] [110/312] eta: 0:02:37 lr: 0.000065 min_lr: 0.000065 loss: 1.7508 (1.6928) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [415] [120/312] eta: 0:02:28 lr: 0.000065 min_lr: 0.000065 loss: 1.6560 (1.6898) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [415] [130/312] eta: 0:02:19 lr: 0.000064 min_lr: 0.000064 loss: 1.5478 (1.6742) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [415] [140/312] eta: 0:02:10 lr: 0.000064 min_lr: 0.000064 loss: 1.6022 (1.6719) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [415] [150/312] eta: 0:02:02 lr: 0.000064 min_lr: 0.000064 loss: 1.6866 (1.6729) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [415] [160/312] eta: 0:01:54 lr: 0.000064 min_lr: 0.000064 loss: 1.6449 (1.6635) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [415] [170/312] eta: 0:01:46 lr: 0.000064 min_lr: 0.000064 loss: 1.4214 (1.6540) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [415] [180/312] eta: 0:01:38 lr: 0.000064 min_lr: 0.000064 loss: 1.4569 (1.6471) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [415] [190/312] eta: 0:01:30 lr: 0.000064 min_lr: 0.000064 loss: 1.6330 (1.6487) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [415] [200/312] eta: 0:01:22 lr: 0.000064 min_lr: 0.000064 loss: 1.7229 (1.6525) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [415] [210/312] eta: 0:01:15 lr: 0.000064 min_lr: 0.000064 loss: 1.8207 (1.6613) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [415] [220/312] eta: 0:01:07 lr: 0.000063 min_lr: 0.000063 loss: 1.8125 (1.6601) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [415] [230/312] eta: 0:01:00 lr: 0.000063 min_lr: 0.000063 loss: 1.4738 (1.6565) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [415] [240/312] eta: 0:00:52 lr: 0.000063 min_lr: 0.000063 loss: 1.7234 (1.6619) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [415] [250/312] eta: 0:00:45 lr: 0.000063 min_lr: 0.000063 loss: 1.7142 (1.6543) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [415] [260/312] eta: 0:00:37 lr: 0.000063 min_lr: 0.000063 loss: 1.5458 (1.6561) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [415] [270/312] eta: 0:00:30 lr: 0.000063 min_lr: 0.000063 loss: 1.5458 (1.6523) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [415] [280/312] eta: 0:00:23 lr: 0.000063 min_lr: 0.000063 loss: 1.6997 (1.6556) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [415] [290/312] eta: 0:00:15 lr: 0.000063 min_lr: 0.000063 loss: 1.7673 (1.6539) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [415] [300/312] eta: 0:00:08 lr: 0.000063 min_lr: 0.000063 loss: 1.6772 (1.6547) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [415] [310/312] eta: 0:00:01 lr: 0.000062 min_lr: 0.000062 loss: 1.7994 (1.6566) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [415] [311/312] eta: 0:00:00 lr: 0.000062 min_lr: 0.000062 loss: 1.8297 (1.6573) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [415] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.000062 min_lr: 0.000062 loss: 1.8297 (1.6640) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4482 (0.4482) acc1: 88.8021 (88.8021) acc5: 98.4375 (98.4375) time: 4.5370 data: 4.3256 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6214 (0.6423) acc1: 84.3750 (83.1680) acc5: 97.1354 (96.8640) time: 0.6554 data: 0.4807 max mem: 64948 Test: Total time: 0:00:06 (0.6809 s / it) * Acc@1 83.658 Acc@5 96.658 loss 0.623 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.70% Test: [0/9] eta: 0:00:45 loss: 0.4518 (0.4518) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 5.0715 data: 4.8547 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6144 (0.6388) acc1: 84.3750 (82.8480) acc5: 96.8750 (96.5440) time: 0.7148 data: 0.5395 max mem: 64948 Test: Total time: 0:00:06 (0.7302 s / it) * Acc@1 83.630 Acc@5 96.716 loss 0.619 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.63% Epoch: [416] [ 0/312] eta: 0:48:51 lr: 0.000062 min_lr: 0.000062 loss: 1.9249 (1.9249) weight_decay: 0.0500 (0.0500) time: 9.3948 data: 7.5272 max mem: 64948 Epoch: [416] [ 10/312] eta: 0:07:43 lr: 0.000062 min_lr: 0.000062 loss: 1.8407 (1.8693) weight_decay: 0.0500 (0.0500) time: 1.5357 data: 0.7120 max mem: 64948 Epoch: [416] [ 20/312] eta: 0:05:31 lr: 0.000062 min_lr: 0.000062 loss: 1.8374 (1.8296) weight_decay: 0.0500 (0.0500) time: 0.7216 data: 0.0154 max mem: 64948 Epoch: [416] [ 30/312] eta: 0:04:39 lr: 0.000062 min_lr: 0.000062 loss: 1.7568 (1.7873) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [416] [ 40/312] eta: 0:04:10 lr: 0.000062 min_lr: 0.000062 loss: 1.7962 (1.7732) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [416] [ 50/312] eta: 0:03:49 lr: 0.000062 min_lr: 0.000062 loss: 1.8094 (1.7746) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [416] [ 60/312] eta: 0:03:33 lr: 0.000062 min_lr: 0.000062 loss: 1.7988 (1.7595) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [416] [ 70/312] eta: 0:03:19 lr: 0.000062 min_lr: 0.000062 loss: 1.7553 (1.7529) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [416] [ 80/312] eta: 0:03:07 lr: 0.000061 min_lr: 0.000061 loss: 1.7543 (1.7441) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [416] [ 90/312] eta: 0:02:56 lr: 0.000061 min_lr: 0.000061 loss: 1.6691 (1.7305) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [416] [100/312] eta: 0:02:46 lr: 0.000061 min_lr: 0.000061 loss: 1.7099 (1.7337) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [416] [110/312] eta: 0:02:37 lr: 0.000061 min_lr: 0.000061 loss: 1.6751 (1.7177) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [416] [120/312] eta: 0:02:28 lr: 0.000061 min_lr: 0.000061 loss: 1.5911 (1.7116) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [416] [130/312] eta: 0:02:19 lr: 0.000061 min_lr: 0.000061 loss: 1.5911 (1.6988) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [416] [140/312] eta: 0:02:10 lr: 0.000061 min_lr: 0.000061 loss: 1.5724 (1.6913) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [416] [150/312] eta: 0:02:02 lr: 0.000061 min_lr: 0.000061 loss: 1.5724 (1.6865) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [416] [160/312] eta: 0:01:54 lr: 0.000061 min_lr: 0.000061 loss: 1.6779 (1.6891) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [416] [170/312] eta: 0:01:46 lr: 0.000060 min_lr: 0.000060 loss: 1.7608 (1.6912) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [416] [180/312] eta: 0:01:38 lr: 0.000060 min_lr: 0.000060 loss: 1.7104 (1.6919) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [416] [190/312] eta: 0:01:30 lr: 0.000060 min_lr: 0.000060 loss: 1.7077 (1.6951) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [416] [200/312] eta: 0:01:23 lr: 0.000060 min_lr: 0.000060 loss: 1.6930 (1.6904) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [416] [210/312] eta: 0:01:15 lr: 0.000060 min_lr: 0.000060 loss: 1.5856 (1.6941) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [416] [220/312] eta: 0:01:07 lr: 0.000060 min_lr: 0.000060 loss: 1.8136 (1.6995) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [416] [230/312] eta: 0:01:00 lr: 0.000060 min_lr: 0.000060 loss: 1.7126 (1.6987) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [416] [240/312] eta: 0:00:52 lr: 0.000060 min_lr: 0.000060 loss: 1.6950 (1.6947) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [416] [250/312] eta: 0:00:45 lr: 0.000060 min_lr: 0.000060 loss: 1.7163 (1.6919) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [416] [260/312] eta: 0:00:37 lr: 0.000059 min_lr: 0.000059 loss: 1.7163 (1.6934) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [416] [270/312] eta: 0:00:30 lr: 0.000059 min_lr: 0.000059 loss: 1.7571 (1.6942) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [416] [280/312] eta: 0:00:23 lr: 0.000059 min_lr: 0.000059 loss: 1.6846 (1.6893) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [416] [290/312] eta: 0:00:15 lr: 0.000059 min_lr: 0.000059 loss: 1.6827 (1.6907) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [416] [300/312] eta: 0:00:08 lr: 0.000059 min_lr: 0.000059 loss: 1.7146 (1.6894) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [416] [310/312] eta: 0:00:01 lr: 0.000059 min_lr: 0.000059 loss: 1.7146 (1.6879) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [416] [311/312] eta: 0:00:00 lr: 0.000059 min_lr: 0.000059 loss: 1.7125 (1.6880) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [416] Total time: 0:03:46 (0.7270 s / it) Averaged stats: lr: 0.000059 min_lr: 0.000059 loss: 1.7125 (1.6705) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4411 (0.4411) acc1: 88.0208 (88.0208) acc5: 98.1771 (98.1771) time: 4.7123 data: 4.5066 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6087 (0.6337) acc1: 85.1562 (83.1680) acc5: 97.1354 (96.8960) time: 0.6748 data: 0.5008 max mem: 64948 Test: Total time: 0:00:06 (0.6998 s / it) * Acc@1 83.708 Acc@5 96.716 loss 0.621 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.71% Test: [0/9] eta: 0:00:42 loss: 0.4515 (0.4515) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.6774 data: 4.4652 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6138 (0.6386) acc1: 84.6354 (82.9120) acc5: 96.8750 (96.5440) time: 0.6710 data: 0.4962 max mem: 64948 Test: Total time: 0:00:06 (0.6808 s / it) * Acc@1 83.642 Acc@5 96.710 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.6% Max EMA accuracy: 83.64% Epoch: [417] [ 0/312] eta: 0:52:05 lr: 0.000059 min_lr: 0.000059 loss: 1.5737 (1.5737) weight_decay: 0.0500 (0.0500) time: 10.0169 data: 9.2656 max mem: 64948 Epoch: [417] [ 10/312] eta: 0:07:48 lr: 0.000059 min_lr: 0.000059 loss: 1.5826 (1.5988) weight_decay: 0.0500 (0.0500) time: 1.5526 data: 0.8427 max mem: 64948 Epoch: [417] [ 20/312] eta: 0:05:34 lr: 0.000059 min_lr: 0.000059 loss: 1.5082 (1.5481) weight_decay: 0.0500 (0.0500) time: 0.7003 data: 0.0004 max mem: 64948 Epoch: [417] [ 30/312] eta: 0:04:41 lr: 0.000058 min_lr: 0.000058 loss: 1.5082 (1.5773) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [417] [ 40/312] eta: 0:04:12 lr: 0.000058 min_lr: 0.000058 loss: 1.6822 (1.5890) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [417] [ 50/312] eta: 0:03:51 lr: 0.000058 min_lr: 0.000058 loss: 1.5938 (1.5613) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [417] [ 60/312] eta: 0:03:34 lr: 0.000058 min_lr: 0.000058 loss: 1.7229 (1.5931) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [417] [ 70/312] eta: 0:03:20 lr: 0.000058 min_lr: 0.000058 loss: 1.6985 (1.5805) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [417] [ 80/312] eta: 0:03:08 lr: 0.000058 min_lr: 0.000058 loss: 1.6399 (1.5965) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [417] [ 90/312] eta: 0:02:57 lr: 0.000058 min_lr: 0.000058 loss: 1.8594 (1.6059) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [417] [100/312] eta: 0:02:47 lr: 0.000058 min_lr: 0.000058 loss: 1.8275 (1.6167) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [417] [110/312] eta: 0:02:37 lr: 0.000058 min_lr: 0.000058 loss: 1.7001 (1.6214) weight_decay: 0.0500 (0.0500) time: 0.6998 data: 0.0004 max mem: 64948 Epoch: [417] [120/312] eta: 0:02:28 lr: 0.000057 min_lr: 0.000057 loss: 1.6069 (1.6203) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [417] [130/312] eta: 0:02:19 lr: 0.000057 min_lr: 0.000057 loss: 1.7251 (1.6334) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [417] [140/312] eta: 0:02:11 lr: 0.000057 min_lr: 0.000057 loss: 1.7364 (1.6327) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [417] [150/312] eta: 0:02:02 lr: 0.000057 min_lr: 0.000057 loss: 1.7363 (1.6338) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [417] [160/312] eta: 0:01:54 lr: 0.000057 min_lr: 0.000057 loss: 1.8409 (1.6446) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [417] [170/312] eta: 0:01:46 lr: 0.000057 min_lr: 0.000057 loss: 1.8009 (1.6488) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [417] [180/312] eta: 0:01:38 lr: 0.000057 min_lr: 0.000057 loss: 1.6101 (1.6465) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [417] [190/312] eta: 0:01:30 lr: 0.000057 min_lr: 0.000057 loss: 1.5912 (1.6470) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [417] [200/312] eta: 0:01:23 lr: 0.000057 min_lr: 0.000057 loss: 1.7001 (1.6495) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0003 max mem: 64948 Epoch: [417] [210/312] eta: 0:01:15 lr: 0.000057 min_lr: 0.000057 loss: 1.7052 (1.6482) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [417] [220/312] eta: 0:01:07 lr: 0.000056 min_lr: 0.000056 loss: 1.5329 (1.6466) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [417] [230/312] eta: 0:01:00 lr: 0.000056 min_lr: 0.000056 loss: 1.7279 (1.6525) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [417] [240/312] eta: 0:00:52 lr: 0.000056 min_lr: 0.000056 loss: 1.7471 (1.6470) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [417] [250/312] eta: 0:00:45 lr: 0.000056 min_lr: 0.000056 loss: 1.6992 (1.6489) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [417] [260/312] eta: 0:00:38 lr: 0.000056 min_lr: 0.000056 loss: 1.7701 (1.6467) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [417] [270/312] eta: 0:00:30 lr: 0.000056 min_lr: 0.000056 loss: 1.7701 (1.6508) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [417] [280/312] eta: 0:00:23 lr: 0.000056 min_lr: 0.000056 loss: 1.7631 (1.6517) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0009 max mem: 64948 Epoch: [417] [290/312] eta: 0:00:16 lr: 0.000056 min_lr: 0.000056 loss: 1.7519 (1.6542) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [417] [300/312] eta: 0:00:08 lr: 0.000056 min_lr: 0.000056 loss: 1.7519 (1.6553) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [417] [310/312] eta: 0:00:01 lr: 0.000055 min_lr: 0.000055 loss: 1.7280 (1.6553) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [417] [311/312] eta: 0:00:00 lr: 0.000055 min_lr: 0.000055 loss: 1.7467 (1.6558) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [417] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.000055 min_lr: 0.000055 loss: 1.7467 (1.6606) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4369 (0.4369) acc1: 89.3229 (89.3229) acc5: 98.1771 (98.1771) time: 4.5777 data: 4.3672 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6117 (0.6331) acc1: 84.1146 (83.1360) acc5: 97.3958 (96.8960) time: 0.6599 data: 0.4853 max mem: 64948 Test: Total time: 0:00:06 (0.6859 s / it) * Acc@1 83.690 Acc@5 96.684 loss 0.619 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.71% Test: [0/9] eta: 0:00:43 loss: 0.4512 (0.4512) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.8185 data: 4.6004 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6133 (0.6383) acc1: 84.6354 (82.9120) acc5: 96.8750 (96.5440) time: 0.6870 data: 0.5112 max mem: 64948 Test: Total time: 0:00:06 (0.6941 s / it) * Acc@1 83.638 Acc@5 96.706 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [418] [ 0/312] eta: 0:56:08 lr: 0.000055 min_lr: 0.000055 loss: 1.4124 (1.4124) weight_decay: 0.0500 (0.0500) time: 10.7962 data: 8.6020 max mem: 64948 Epoch: [418] [ 10/312] eta: 0:08:20 lr: 0.000055 min_lr: 0.000055 loss: 1.4911 (1.4830) weight_decay: 0.0500 (0.0500) time: 1.6568 data: 0.7824 max mem: 64948 Epoch: [418] [ 20/312] eta: 0:05:49 lr: 0.000055 min_lr: 0.000055 loss: 1.5172 (1.5715) weight_decay: 0.0500 (0.0500) time: 0.7171 data: 0.0004 max mem: 64948 Epoch: [418] [ 30/312] eta: 0:04:52 lr: 0.000055 min_lr: 0.000055 loss: 1.7227 (1.6081) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [418] [ 40/312] eta: 0:04:19 lr: 0.000055 min_lr: 0.000055 loss: 1.6365 (1.5646) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [418] [ 50/312] eta: 0:03:56 lr: 0.000055 min_lr: 0.000055 loss: 1.4347 (1.5784) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [418] [ 60/312] eta: 0:03:38 lr: 0.000055 min_lr: 0.000055 loss: 1.8375 (1.6290) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [418] [ 70/312] eta: 0:03:24 lr: 0.000055 min_lr: 0.000055 loss: 1.8390 (1.6398) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [418] [ 80/312] eta: 0:03:11 lr: 0.000055 min_lr: 0.000055 loss: 1.7497 (1.6530) weight_decay: 0.0500 (0.0500) time: 0.7008 data: 0.0004 max mem: 64948 Epoch: [418] [ 90/312] eta: 0:03:00 lr: 0.000054 min_lr: 0.000054 loss: 1.6351 (1.6429) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [418] [100/312] eta: 0:02:49 lr: 0.000054 min_lr: 0.000054 loss: 1.5030 (1.6356) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [418] [110/312] eta: 0:02:39 lr: 0.000054 min_lr: 0.000054 loss: 1.6737 (1.6333) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [418] [120/312] eta: 0:02:30 lr: 0.000054 min_lr: 0.000054 loss: 1.6737 (1.6328) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [418] [130/312] eta: 0:02:21 lr: 0.000054 min_lr: 0.000054 loss: 1.4945 (1.6155) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [418] [140/312] eta: 0:02:12 lr: 0.000054 min_lr: 0.000054 loss: 1.6094 (1.6194) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [418] [150/312] eta: 0:02:04 lr: 0.000054 min_lr: 0.000054 loss: 1.6941 (1.6217) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [418] [160/312] eta: 0:01:55 lr: 0.000054 min_lr: 0.000054 loss: 1.6781 (1.6175) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [418] [170/312] eta: 0:01:47 lr: 0.000054 min_lr: 0.000054 loss: 1.7154 (1.6284) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [418] [180/312] eta: 0:01:39 lr: 0.000053 min_lr: 0.000053 loss: 1.8061 (1.6385) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [418] [190/312] eta: 0:01:31 lr: 0.000053 min_lr: 0.000053 loss: 1.8049 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [418] [200/312] eta: 0:01:23 lr: 0.000053 min_lr: 0.000053 loss: 1.7588 (1.6489) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [418] [210/312] eta: 0:01:16 lr: 0.000053 min_lr: 0.000053 loss: 1.6999 (1.6515) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [418] [220/312] eta: 0:01:08 lr: 0.000053 min_lr: 0.000053 loss: 1.6774 (1.6521) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [418] [230/312] eta: 0:01:00 lr: 0.000053 min_lr: 0.000053 loss: 1.6882 (1.6531) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [418] [240/312] eta: 0:00:53 lr: 0.000053 min_lr: 0.000053 loss: 1.8110 (1.6557) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [418] [250/312] eta: 0:00:45 lr: 0.000053 min_lr: 0.000053 loss: 1.8535 (1.6555) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [418] [260/312] eta: 0:00:38 lr: 0.000053 min_lr: 0.000053 loss: 1.8285 (1.6551) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [418] [270/312] eta: 0:00:30 lr: 0.000053 min_lr: 0.000053 loss: 1.8053 (1.6547) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [418] [280/312] eta: 0:00:23 lr: 0.000052 min_lr: 0.000052 loss: 1.6433 (1.6519) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [418] [290/312] eta: 0:00:16 lr: 0.000052 min_lr: 0.000052 loss: 1.6211 (1.6481) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [418] [300/312] eta: 0:00:08 lr: 0.000052 min_lr: 0.000052 loss: 1.7868 (1.6561) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [418] [310/312] eta: 0:00:01 lr: 0.000052 min_lr: 0.000052 loss: 1.8327 (1.6601) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [418] [311/312] eta: 0:00:00 lr: 0.000052 min_lr: 0.000052 loss: 1.8327 (1.6609) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [418] Total time: 0:03:48 (0.7319 s / it) Averaged stats: lr: 0.000052 min_lr: 0.000052 loss: 1.8327 (1.6592) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4448 (0.4448) acc1: 89.3229 (89.3229) acc5: 97.9167 (97.9167) time: 4.7183 data: 4.5035 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6171 (0.6391) acc1: 84.3750 (83.1040) acc5: 97.6562 (96.9600) time: 0.6756 data: 0.5005 max mem: 64948 Test: Total time: 0:00:06 (0.7109 s / it) * Acc@1 83.604 Acc@5 96.712 loss 0.622 Accuracy of the model on the 50000 test images: 83.6% Max accuracy: 83.71% Test: [0/9] eta: 0:00:43 loss: 0.4509 (0.4509) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.7980 data: 4.5866 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6127 (0.6382) acc1: 84.8958 (82.9440) acc5: 96.8750 (96.5760) time: 0.6851 data: 0.5097 max mem: 64948 Test: Total time: 0:00:06 (0.6927 s / it) * Acc@1 83.636 Acc@5 96.694 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.6% Epoch: [419] [ 0/312] eta: 0:54:35 lr: 0.000052 min_lr: 0.000052 loss: 1.5049 (1.5049) weight_decay: 0.0500 (0.0500) time: 10.4979 data: 8.2823 max mem: 64948 Epoch: [419] [ 10/312] eta: 0:08:08 lr: 0.000052 min_lr: 0.000052 loss: 1.5049 (1.4660) weight_decay: 0.0500 (0.0500) time: 1.6164 data: 0.7534 max mem: 64948 Epoch: [419] [ 20/312] eta: 0:05:43 lr: 0.000052 min_lr: 0.000052 loss: 1.5341 (1.4896) weight_decay: 0.0500 (0.0500) time: 0.7106 data: 0.0004 max mem: 64948 Epoch: [419] [ 30/312] eta: 0:04:48 lr: 0.000052 min_lr: 0.000052 loss: 1.5168 (1.4800) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [419] [ 40/312] eta: 0:04:16 lr: 0.000052 min_lr: 0.000052 loss: 1.5917 (1.5177) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [419] [ 50/312] eta: 0:03:53 lr: 0.000052 min_lr: 0.000052 loss: 1.6480 (1.5311) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [419] [ 60/312] eta: 0:03:37 lr: 0.000051 min_lr: 0.000051 loss: 1.6956 (1.5654) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [419] [ 70/312] eta: 0:03:23 lr: 0.000051 min_lr: 0.000051 loss: 1.6747 (1.5760) weight_decay: 0.0500 (0.0500) time: 0.7012 data: 0.0004 max mem: 64948 Epoch: [419] [ 80/312] eta: 0:03:10 lr: 0.000051 min_lr: 0.000051 loss: 1.6747 (1.5845) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0003 max mem: 64948 Epoch: [419] [ 90/312] eta: 0:02:59 lr: 0.000051 min_lr: 0.000051 loss: 1.6379 (1.5774) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0003 max mem: 64948 Epoch: [419] [100/312] eta: 0:02:48 lr: 0.000051 min_lr: 0.000051 loss: 1.5320 (1.5879) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [419] [110/312] eta: 0:02:39 lr: 0.000051 min_lr: 0.000051 loss: 1.6213 (1.5952) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [419] [120/312] eta: 0:02:29 lr: 0.000051 min_lr: 0.000051 loss: 1.6481 (1.6045) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [419] [130/312] eta: 0:02:20 lr: 0.000051 min_lr: 0.000051 loss: 1.7494 (1.6171) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [419] [140/312] eta: 0:02:12 lr: 0.000051 min_lr: 0.000051 loss: 1.7436 (1.6090) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [419] [150/312] eta: 0:02:03 lr: 0.000050 min_lr: 0.000050 loss: 1.6536 (1.6182) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [419] [160/312] eta: 0:01:55 lr: 0.000050 min_lr: 0.000050 loss: 1.6408 (1.6168) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [419] [170/312] eta: 0:01:47 lr: 0.000050 min_lr: 0.000050 loss: 1.6989 (1.6266) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [419] [180/312] eta: 0:01:39 lr: 0.000050 min_lr: 0.000050 loss: 1.7913 (1.6324) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [419] [190/312] eta: 0:01:31 lr: 0.000050 min_lr: 0.000050 loss: 1.7632 (1.6331) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [419] [200/312] eta: 0:01:23 lr: 0.000050 min_lr: 0.000050 loss: 1.7303 (1.6323) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [419] [210/312] eta: 0:01:15 lr: 0.000050 min_lr: 0.000050 loss: 1.7715 (1.6379) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [419] [220/312] eta: 0:01:08 lr: 0.000050 min_lr: 0.000050 loss: 1.7715 (1.6409) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [419] [230/312] eta: 0:01:00 lr: 0.000050 min_lr: 0.000050 loss: 1.7600 (1.6416) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [419] [240/312] eta: 0:00:53 lr: 0.000050 min_lr: 0.000050 loss: 1.5459 (1.6330) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [419] [250/312] eta: 0:00:45 lr: 0.000049 min_lr: 0.000049 loss: 1.5856 (1.6356) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [419] [260/312] eta: 0:00:38 lr: 0.000049 min_lr: 0.000049 loss: 1.7526 (1.6331) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [419] [270/312] eta: 0:00:30 lr: 0.000049 min_lr: 0.000049 loss: 1.8082 (1.6386) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [419] [280/312] eta: 0:00:23 lr: 0.000049 min_lr: 0.000049 loss: 1.8216 (1.6370) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0009 max mem: 64948 Epoch: [419] [290/312] eta: 0:00:16 lr: 0.000049 min_lr: 0.000049 loss: 1.6393 (1.6355) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [419] [300/312] eta: 0:00:08 lr: 0.000049 min_lr: 0.000049 loss: 1.6335 (1.6323) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [419] [310/312] eta: 0:00:01 lr: 0.000049 min_lr: 0.000049 loss: 1.6134 (1.6285) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [419] [311/312] eta: 0:00:00 lr: 0.000049 min_lr: 0.000049 loss: 1.6134 (1.6293) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [419] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.000049 min_lr: 0.000049 loss: 1.6134 (1.6557) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:51 loss: 0.4383 (0.4383) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 5.7476 data: 5.5273 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6166 (0.6409) acc1: 83.5938 (83.1040) acc5: 97.3958 (96.8960) time: 0.7899 data: 0.6142 max mem: 64948 Test: Total time: 0:00:07 (0.8093 s / it) * Acc@1 83.706 Acc@5 96.700 loss 0.622 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.71% Test: [0/9] eta: 0:00:46 loss: 0.4506 (0.4506) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 5.1178 data: 4.8997 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6125 (0.6381) acc1: 84.8958 (82.9440) acc5: 96.8750 (96.5760) time: 0.7287 data: 0.5533 max mem: 64948 Test: Total time: 0:00:06 (0.7407 s / it) * Acc@1 83.652 Acc@5 96.692 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.7% Max EMA accuracy: 83.65% Epoch: [420] [ 0/312] eta: 0:46:25 lr: 0.000049 min_lr: 0.000049 loss: 1.6887 (1.6887) weight_decay: 0.0500 (0.0500) time: 8.9280 data: 8.1705 max mem: 64948 Epoch: [420] [ 10/312] eta: 0:07:21 lr: 0.000049 min_lr: 0.000049 loss: 1.7649 (1.6728) weight_decay: 0.0500 (0.0500) time: 1.4605 data: 0.7431 max mem: 64948 Epoch: [420] [ 20/312] eta: 0:05:20 lr: 0.000049 min_lr: 0.000049 loss: 1.7419 (1.5976) weight_decay: 0.0500 (0.0500) time: 0.7057 data: 0.0004 max mem: 64948 Epoch: [420] [ 30/312] eta: 0:04:32 lr: 0.000049 min_lr: 0.000049 loss: 1.5312 (1.6203) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0003 max mem: 64948 Epoch: [420] [ 40/312] eta: 0:04:05 lr: 0.000048 min_lr: 0.000048 loss: 1.6594 (1.6074) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [420] [ 50/312] eta: 0:03:45 lr: 0.000048 min_lr: 0.000048 loss: 1.6427 (1.6269) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [420] [ 60/312] eta: 0:03:30 lr: 0.000048 min_lr: 0.000048 loss: 1.6824 (1.6299) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [420] [ 70/312] eta: 0:03:17 lr: 0.000048 min_lr: 0.000048 loss: 1.6692 (1.6338) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [420] [ 80/312] eta: 0:03:05 lr: 0.000048 min_lr: 0.000048 loss: 1.6605 (1.6429) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [420] [ 90/312] eta: 0:02:55 lr: 0.000048 min_lr: 0.000048 loss: 1.6605 (1.6303) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [420] [100/312] eta: 0:02:45 lr: 0.000048 min_lr: 0.000048 loss: 1.8245 (1.6549) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [420] [110/312] eta: 0:02:35 lr: 0.000048 min_lr: 0.000048 loss: 1.8105 (1.6526) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [420] [120/312] eta: 0:02:26 lr: 0.000048 min_lr: 0.000048 loss: 1.7215 (1.6595) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [420] [130/312] eta: 0:02:18 lr: 0.000048 min_lr: 0.000048 loss: 1.6210 (1.6554) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [420] [140/312] eta: 0:02:09 lr: 0.000047 min_lr: 0.000047 loss: 1.6084 (1.6535) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [420] [150/312] eta: 0:02:01 lr: 0.000047 min_lr: 0.000047 loss: 1.5956 (1.6497) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [420] [160/312] eta: 0:01:53 lr: 0.000047 min_lr: 0.000047 loss: 1.5851 (1.6369) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [420] [170/312] eta: 0:01:45 lr: 0.000047 min_lr: 0.000047 loss: 1.4790 (1.6320) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [420] [180/312] eta: 0:01:37 lr: 0.000047 min_lr: 0.000047 loss: 1.7589 (1.6367) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [420] [190/312] eta: 0:01:30 lr: 0.000047 min_lr: 0.000047 loss: 1.7119 (1.6392) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [420] [200/312] eta: 0:01:22 lr: 0.000047 min_lr: 0.000047 loss: 1.6749 (1.6382) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [420] [210/312] eta: 0:01:14 lr: 0.000047 min_lr: 0.000047 loss: 1.8131 (1.6453) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [420] [220/312] eta: 0:01:07 lr: 0.000047 min_lr: 0.000047 loss: 1.7443 (1.6453) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [420] [230/312] eta: 0:00:59 lr: 0.000047 min_lr: 0.000047 loss: 1.6810 (1.6442) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [420] [240/312] eta: 0:00:52 lr: 0.000046 min_lr: 0.000046 loss: 1.6906 (1.6439) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [420] [250/312] eta: 0:00:45 lr: 0.000046 min_lr: 0.000046 loss: 1.6906 (1.6430) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [420] [260/312] eta: 0:00:37 lr: 0.000046 min_lr: 0.000046 loss: 1.7141 (1.6448) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [420] [270/312] eta: 0:00:30 lr: 0.000046 min_lr: 0.000046 loss: 1.8079 (1.6472) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [420] [280/312] eta: 0:00:23 lr: 0.000046 min_lr: 0.000046 loss: 1.8193 (1.6554) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0009 max mem: 64948 Epoch: [420] [290/312] eta: 0:00:15 lr: 0.000046 min_lr: 0.000046 loss: 1.7915 (1.6521) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0008 max mem: 64948 Epoch: [420] [300/312] eta: 0:00:08 lr: 0.000046 min_lr: 0.000046 loss: 1.7142 (1.6533) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [420] [310/312] eta: 0:00:01 lr: 0.000046 min_lr: 0.000046 loss: 1.5981 (1.6484) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [420] [311/312] eta: 0:00:00 lr: 0.000046 min_lr: 0.000046 loss: 1.6980 (1.6499) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [420] Total time: 0:03:45 (0.7243 s / it) Averaged stats: lr: 0.000046 min_lr: 0.000046 loss: 1.6980 (1.6603) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:48 loss: 0.4346 (0.4346) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 5.4135 data: 5.2015 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6132 (0.6376) acc1: 84.6354 (83.0400) acc5: 97.1354 (96.8640) time: 0.7528 data: 0.5780 max mem: 64948 Test: Total time: 0:00:07 (0.7848 s / it) * Acc@1 83.764 Acc@5 96.736 loss 0.621 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.76% Test: [0/9] eta: 0:00:42 loss: 0.4503 (0.4503) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.7148 data: 4.4993 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6121 (0.6379) acc1: 84.8958 (83.0400) acc5: 96.8750 (96.6080) time: 0.6752 data: 0.5000 max mem: 64948 Test: Total time: 0:00:06 (0.6830 s / it) * Acc@1 83.662 Acc@5 96.690 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.7% Max EMA accuracy: 83.66% Epoch: [421] [ 0/312] eta: 0:47:06 lr: 0.000046 min_lr: 0.000046 loss: 1.3367 (1.3367) weight_decay: 0.0500 (0.0500) time: 9.0598 data: 8.3009 max mem: 64948 Epoch: [421] [ 10/312] eta: 0:07:24 lr: 0.000046 min_lr: 0.000046 loss: 1.8128 (1.7363) weight_decay: 0.0500 (0.0500) time: 1.4725 data: 0.7550 max mem: 64948 Epoch: [421] [ 20/312] eta: 0:05:21 lr: 0.000046 min_lr: 0.000046 loss: 1.7916 (1.7475) weight_decay: 0.0500 (0.0500) time: 0.7044 data: 0.0004 max mem: 64948 Epoch: [421] [ 30/312] eta: 0:04:33 lr: 0.000045 min_lr: 0.000045 loss: 1.6556 (1.6927) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0003 max mem: 64948 Epoch: [421] [ 40/312] eta: 0:04:05 lr: 0.000045 min_lr: 0.000045 loss: 1.6506 (1.6814) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [421] [ 50/312] eta: 0:03:46 lr: 0.000045 min_lr: 0.000045 loss: 1.7119 (1.6578) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [421] [ 60/312] eta: 0:03:30 lr: 0.000045 min_lr: 0.000045 loss: 1.3668 (1.6163) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [421] [ 70/312] eta: 0:03:17 lr: 0.000045 min_lr: 0.000045 loss: 1.6184 (1.6299) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0003 max mem: 64948 Epoch: [421] [ 80/312] eta: 0:03:05 lr: 0.000045 min_lr: 0.000045 loss: 1.6184 (1.6047) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [421] [ 90/312] eta: 0:02:55 lr: 0.000045 min_lr: 0.000045 loss: 1.3818 (1.5956) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [421] [100/312] eta: 0:02:45 lr: 0.000045 min_lr: 0.000045 loss: 1.6132 (1.5895) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [421] [110/312] eta: 0:02:35 lr: 0.000045 min_lr: 0.000045 loss: 1.7246 (1.5998) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [421] [120/312] eta: 0:02:26 lr: 0.000045 min_lr: 0.000045 loss: 1.7355 (1.6021) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [421] [130/312] eta: 0:02:18 lr: 0.000044 min_lr: 0.000044 loss: 1.7355 (1.6091) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [421] [140/312] eta: 0:02:10 lr: 0.000044 min_lr: 0.000044 loss: 1.6275 (1.6079) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [421] [150/312] eta: 0:02:01 lr: 0.000044 min_lr: 0.000044 loss: 1.6275 (1.6109) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [421] [160/312] eta: 0:01:53 lr: 0.000044 min_lr: 0.000044 loss: 1.6595 (1.6101) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [421] [170/312] eta: 0:01:45 lr: 0.000044 min_lr: 0.000044 loss: 1.6827 (1.6153) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [421] [180/312] eta: 0:01:38 lr: 0.000044 min_lr: 0.000044 loss: 1.7478 (1.6222) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [421] [190/312] eta: 0:01:30 lr: 0.000044 min_lr: 0.000044 loss: 1.7378 (1.6209) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [421] [200/312] eta: 0:01:22 lr: 0.000044 min_lr: 0.000044 loss: 1.5427 (1.6197) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [421] [210/312] eta: 0:01:15 lr: 0.000044 min_lr: 0.000044 loss: 1.6272 (1.6186) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [421] [220/312] eta: 0:01:07 lr: 0.000044 min_lr: 0.000044 loss: 1.6852 (1.6244) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [421] [230/312] eta: 0:01:00 lr: 0.000043 min_lr: 0.000043 loss: 1.6821 (1.6182) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [421] [240/312] eta: 0:00:52 lr: 0.000043 min_lr: 0.000043 loss: 1.6821 (1.6273) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [421] [250/312] eta: 0:00:45 lr: 0.000043 min_lr: 0.000043 loss: 1.8088 (1.6271) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [421] [260/312] eta: 0:00:37 lr: 0.000043 min_lr: 0.000043 loss: 1.6258 (1.6248) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [421] [270/312] eta: 0:00:30 lr: 0.000043 min_lr: 0.000043 loss: 1.5627 (1.6239) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [421] [280/312] eta: 0:00:23 lr: 0.000043 min_lr: 0.000043 loss: 1.6721 (1.6257) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0009 max mem: 64948 Epoch: [421] [290/312] eta: 0:00:15 lr: 0.000043 min_lr: 0.000043 loss: 1.7643 (1.6330) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0008 max mem: 64948 Epoch: [421] [300/312] eta: 0:00:08 lr: 0.000043 min_lr: 0.000043 loss: 1.7370 (1.6322) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [421] [310/312] eta: 0:00:01 lr: 0.000043 min_lr: 0.000043 loss: 1.6780 (1.6359) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [421] [311/312] eta: 0:00:00 lr: 0.000043 min_lr: 0.000043 loss: 1.6892 (1.6366) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [421] Total time: 0:03:46 (0.7252 s / it) Averaged stats: lr: 0.000043 min_lr: 0.000043 loss: 1.6892 (1.6560) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4324 (0.4324) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 4.6776 data: 4.4577 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6195 (0.6367) acc1: 84.1146 (82.7520) acc5: 97.3958 (96.8640) time: 0.6710 data: 0.4954 max mem: 64948 Test: Total time: 0:00:06 (0.6958 s / it) * Acc@1 83.710 Acc@5 96.696 loss 0.619 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.76% Test: [0/9] eta: 0:00:44 loss: 0.4499 (0.4499) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.9031 data: 4.6849 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6120 (0.6378) acc1: 84.8958 (83.0400) acc5: 96.8750 (96.6080) time: 0.6990 data: 0.5206 max mem: 64948 Test: Total time: 0:00:06 (0.7088 s / it) * Acc@1 83.688 Acc@5 96.694 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.7% Max EMA accuracy: 83.69% Epoch: [422] [ 0/312] eta: 0:51:27 lr: 0.000043 min_lr: 0.000043 loss: 1.5282 (1.5282) weight_decay: 0.0500 (0.0500) time: 9.8966 data: 9.1198 max mem: 64948 Epoch: [422] [ 10/312] eta: 0:07:46 lr: 0.000043 min_lr: 0.000043 loss: 1.5289 (1.4746) weight_decay: 0.0500 (0.0500) time: 1.5462 data: 0.8294 max mem: 64948 Epoch: [422] [ 20/312] eta: 0:05:33 lr: 0.000043 min_lr: 0.000043 loss: 1.5360 (1.5341) weight_decay: 0.0500 (0.0500) time: 0.7032 data: 0.0003 max mem: 64948 Epoch: [422] [ 30/312] eta: 0:04:41 lr: 0.000042 min_lr: 0.000042 loss: 1.7397 (1.5920) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [422] [ 40/312] eta: 0:04:11 lr: 0.000042 min_lr: 0.000042 loss: 1.6490 (1.5513) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [422] [ 50/312] eta: 0:03:50 lr: 0.000042 min_lr: 0.000042 loss: 1.3346 (1.5367) weight_decay: 0.0500 (0.0500) time: 0.6987 data: 0.0004 max mem: 64948 Epoch: [422] [ 60/312] eta: 0:03:34 lr: 0.000042 min_lr: 0.000042 loss: 1.6141 (1.5543) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [422] [ 70/312] eta: 0:03:20 lr: 0.000042 min_lr: 0.000042 loss: 1.6512 (1.5661) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [422] [ 80/312] eta: 0:03:08 lr: 0.000042 min_lr: 0.000042 loss: 1.7484 (1.5925) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [422] [ 90/312] eta: 0:02:57 lr: 0.000042 min_lr: 0.000042 loss: 1.7306 (1.5805) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [422] [100/312] eta: 0:02:47 lr: 0.000042 min_lr: 0.000042 loss: 1.6355 (1.5966) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [422] [110/312] eta: 0:02:37 lr: 0.000042 min_lr: 0.000042 loss: 1.7111 (1.6000) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [422] [120/312] eta: 0:02:28 lr: 0.000042 min_lr: 0.000042 loss: 1.7111 (1.6050) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [422] [130/312] eta: 0:02:19 lr: 0.000041 min_lr: 0.000041 loss: 1.8257 (1.6087) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [422] [140/312] eta: 0:02:11 lr: 0.000041 min_lr: 0.000041 loss: 1.7790 (1.6060) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [422] [150/312] eta: 0:02:02 lr: 0.000041 min_lr: 0.000041 loss: 1.7380 (1.6081) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [422] [160/312] eta: 0:01:54 lr: 0.000041 min_lr: 0.000041 loss: 1.7404 (1.6098) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [422] [170/312] eta: 0:01:46 lr: 0.000041 min_lr: 0.000041 loss: 1.7586 (1.6212) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [422] [180/312] eta: 0:01:38 lr: 0.000041 min_lr: 0.000041 loss: 1.7565 (1.6216) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [422] [190/312] eta: 0:01:30 lr: 0.000041 min_lr: 0.000041 loss: 1.6907 (1.6234) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [422] [200/312] eta: 0:01:23 lr: 0.000041 min_lr: 0.000041 loss: 1.5968 (1.6181) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [422] [210/312] eta: 0:01:15 lr: 0.000041 min_lr: 0.000041 loss: 1.5968 (1.6215) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [422] [220/312] eta: 0:01:07 lr: 0.000041 min_lr: 0.000041 loss: 1.6958 (1.6247) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [422] [230/312] eta: 0:01:00 lr: 0.000041 min_lr: 0.000041 loss: 1.6280 (1.6238) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [422] [240/312] eta: 0:00:52 lr: 0.000040 min_lr: 0.000040 loss: 1.4995 (1.6205) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [422] [250/312] eta: 0:00:45 lr: 0.000040 min_lr: 0.000040 loss: 1.6592 (1.6194) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [422] [260/312] eta: 0:00:38 lr: 0.000040 min_lr: 0.000040 loss: 1.6987 (1.6234) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [422] [270/312] eta: 0:00:30 lr: 0.000040 min_lr: 0.000040 loss: 1.6733 (1.6225) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [422] [280/312] eta: 0:00:23 lr: 0.000040 min_lr: 0.000040 loss: 1.6004 (1.6193) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0006 max mem: 64948 Epoch: [422] [290/312] eta: 0:00:15 lr: 0.000040 min_lr: 0.000040 loss: 1.6481 (1.6193) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0005 max mem: 64948 Epoch: [422] [300/312] eta: 0:00:08 lr: 0.000040 min_lr: 0.000040 loss: 1.6481 (1.6220) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [422] [310/312] eta: 0:00:01 lr: 0.000040 min_lr: 0.000040 loss: 1.6239 (1.6216) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [422] [311/312] eta: 0:00:00 lr: 0.000040 min_lr: 0.000040 loss: 1.7835 (1.6223) weight_decay: 0.0500 (0.0500) time: 0.6903 data: 0.0001 max mem: 64948 Epoch: [422] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.000040 min_lr: 0.000040 loss: 1.7835 (1.6448) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4394 (0.4394) acc1: 87.7604 (87.7604) acc5: 98.1771 (98.1771) time: 4.5395 data: 4.3201 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6089 (0.6377) acc1: 83.8542 (82.6240) acc5: 97.1354 (96.8000) time: 0.6558 data: 0.4801 max mem: 64948 Test: Total time: 0:00:06 (0.6809 s / it) * Acc@1 83.724 Acc@5 96.704 loss 0.620 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.76% Test: [0/9] eta: 0:00:47 loss: 0.4496 (0.4496) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 5.2235 data: 5.0153 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6117 (0.6377) acc1: 84.8958 (83.1040) acc5: 96.8750 (96.6080) time: 0.7316 data: 0.5573 max mem: 64948 Test: Total time: 0:00:06 (0.7480 s / it) * Acc@1 83.682 Acc@5 96.704 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.7% Epoch: [423] [ 0/312] eta: 0:46:47 lr: 0.000040 min_lr: 0.000040 loss: 1.3973 (1.3973) weight_decay: 0.0500 (0.0500) time: 8.9990 data: 6.9234 max mem: 64948 Epoch: [423] [ 10/312] eta: 0:07:29 lr: 0.000040 min_lr: 0.000040 loss: 1.7284 (1.6724) weight_decay: 0.0500 (0.0500) time: 1.4890 data: 0.6299 max mem: 64948 Epoch: [423] [ 20/312] eta: 0:05:24 lr: 0.000040 min_lr: 0.000040 loss: 1.6056 (1.6236) weight_decay: 0.0500 (0.0500) time: 0.7175 data: 0.0004 max mem: 64948 Epoch: [423] [ 30/312] eta: 0:04:35 lr: 0.000040 min_lr: 0.000040 loss: 1.7500 (1.6525) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0003 max mem: 64948 Epoch: [423] [ 40/312] eta: 0:04:06 lr: 0.000039 min_lr: 0.000039 loss: 1.7609 (1.6542) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [423] [ 50/312] eta: 0:03:47 lr: 0.000039 min_lr: 0.000039 loss: 1.7317 (1.6528) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0003 max mem: 64948 Epoch: [423] [ 60/312] eta: 0:03:31 lr: 0.000039 min_lr: 0.000039 loss: 1.6933 (1.6468) weight_decay: 0.0500 (0.0500) time: 0.6989 data: 0.0004 max mem: 64948 Epoch: [423] [ 70/312] eta: 0:03:18 lr: 0.000039 min_lr: 0.000039 loss: 1.6933 (1.6457) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [423] [ 80/312] eta: 0:03:06 lr: 0.000039 min_lr: 0.000039 loss: 1.7111 (1.6522) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [423] [ 90/312] eta: 0:02:55 lr: 0.000039 min_lr: 0.000039 loss: 1.6943 (1.6514) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [423] [100/312] eta: 0:02:45 lr: 0.000039 min_lr: 0.000039 loss: 1.6943 (1.6651) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [423] [110/312] eta: 0:02:36 lr: 0.000039 min_lr: 0.000039 loss: 1.7721 (1.6730) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [423] [120/312] eta: 0:02:27 lr: 0.000039 min_lr: 0.000039 loss: 1.8162 (1.6795) weight_decay: 0.0500 (0.0500) time: 0.6994 data: 0.0004 max mem: 64948 Epoch: [423] [130/312] eta: 0:02:18 lr: 0.000039 min_lr: 0.000039 loss: 1.8023 (1.6886) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [423] [140/312] eta: 0:02:10 lr: 0.000039 min_lr: 0.000039 loss: 1.7656 (1.6794) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [423] [150/312] eta: 0:02:02 lr: 0.000038 min_lr: 0.000038 loss: 1.6215 (1.6720) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [423] [160/312] eta: 0:01:54 lr: 0.000038 min_lr: 0.000038 loss: 1.6215 (1.6675) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [423] [170/312] eta: 0:01:46 lr: 0.000038 min_lr: 0.000038 loss: 1.5998 (1.6573) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [423] [180/312] eta: 0:01:38 lr: 0.000038 min_lr: 0.000038 loss: 1.6804 (1.6656) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [423] [190/312] eta: 0:01:30 lr: 0.000038 min_lr: 0.000038 loss: 1.7786 (1.6656) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [423] [200/312] eta: 0:01:22 lr: 0.000038 min_lr: 0.000038 loss: 1.7343 (1.6690) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [423] [210/312] eta: 0:01:15 lr: 0.000038 min_lr: 0.000038 loss: 1.7734 (1.6681) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [423] [220/312] eta: 0:01:07 lr: 0.000038 min_lr: 0.000038 loss: 1.7739 (1.6694) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [423] [230/312] eta: 0:01:00 lr: 0.000038 min_lr: 0.000038 loss: 1.7064 (1.6652) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [423] [240/312] eta: 0:00:52 lr: 0.000038 min_lr: 0.000038 loss: 1.6452 (1.6682) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [423] [250/312] eta: 0:00:45 lr: 0.000038 min_lr: 0.000038 loss: 1.7803 (1.6687) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [423] [260/312] eta: 0:00:37 lr: 0.000037 min_lr: 0.000037 loss: 1.7037 (1.6702) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [423] [270/312] eta: 0:00:30 lr: 0.000037 min_lr: 0.000037 loss: 1.7663 (1.6742) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [423] [280/312] eta: 0:00:23 lr: 0.000037 min_lr: 0.000037 loss: 1.5340 (1.6680) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0010 max mem: 64948 Epoch: [423] [290/312] eta: 0:00:15 lr: 0.000037 min_lr: 0.000037 loss: 1.5534 (1.6683) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0008 max mem: 64948 Epoch: [423] [300/312] eta: 0:00:08 lr: 0.000037 min_lr: 0.000037 loss: 1.7524 (1.6732) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [423] [310/312] eta: 0:00:01 lr: 0.000037 min_lr: 0.000037 loss: 1.7207 (1.6706) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [423] [311/312] eta: 0:00:00 lr: 0.000037 min_lr: 0.000037 loss: 1.7197 (1.6707) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [423] Total time: 0:03:46 (0.7262 s / it) Averaged stats: lr: 0.000037 min_lr: 0.000037 loss: 1.7197 (1.6572) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4423 (0.4423) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.4149 data: 4.1905 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6122 (0.6385) acc1: 84.3750 (83.0720) acc5: 96.8750 (96.7040) time: 0.6418 data: 0.4657 max mem: 64948 Test: Total time: 0:00:05 (0.6524 s / it) * Acc@1 83.778 Acc@5 96.720 loss 0.620 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.78% Test: [0/9] eta: 0:00:38 loss: 0.4492 (0.4492) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.2727 data: 4.0649 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6115 (0.6376) acc1: 84.6354 (83.0720) acc5: 96.8750 (96.6400) time: 0.6378 data: 0.4635 max mem: 64948 Test: Total time: 0:00:05 (0.6461 s / it) * Acc@1 83.704 Acc@5 96.704 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.7% Max EMA accuracy: 83.70% Epoch: [424] [ 0/312] eta: 0:49:16 lr: 0.000037 min_lr: 0.000037 loss: 1.9279 (1.9279) weight_decay: 0.0500 (0.0500) time: 9.4763 data: 8.6739 max mem: 64948 Epoch: [424] [ 10/312] eta: 0:07:41 lr: 0.000037 min_lr: 0.000037 loss: 1.6068 (1.5566) weight_decay: 0.0500 (0.0500) time: 1.5285 data: 0.7889 max mem: 64948 Epoch: [424] [ 20/312] eta: 0:05:30 lr: 0.000037 min_lr: 0.000037 loss: 1.7570 (1.7041) weight_decay: 0.0500 (0.0500) time: 0.7138 data: 0.0004 max mem: 64948 Epoch: [424] [ 30/312] eta: 0:04:39 lr: 0.000037 min_lr: 0.000037 loss: 1.7570 (1.6588) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [424] [ 40/312] eta: 0:04:09 lr: 0.000037 min_lr: 0.000037 loss: 1.5071 (1.6065) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [424] [ 50/312] eta: 0:03:49 lr: 0.000037 min_lr: 0.000037 loss: 1.5959 (1.6167) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [424] [ 60/312] eta: 0:03:33 lr: 0.000036 min_lr: 0.000036 loss: 1.7512 (1.6420) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [424] [ 70/312] eta: 0:03:19 lr: 0.000036 min_lr: 0.000036 loss: 1.6907 (1.6329) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [424] [ 80/312] eta: 0:03:07 lr: 0.000036 min_lr: 0.000036 loss: 1.5801 (1.6273) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [424] [ 90/312] eta: 0:02:57 lr: 0.000036 min_lr: 0.000036 loss: 1.5928 (1.6363) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [424] [100/312] eta: 0:02:46 lr: 0.000036 min_lr: 0.000036 loss: 1.7467 (1.6333) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [424] [110/312] eta: 0:02:37 lr: 0.000036 min_lr: 0.000036 loss: 1.7494 (1.6425) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [424] [120/312] eta: 0:02:28 lr: 0.000036 min_lr: 0.000036 loss: 1.6760 (1.6410) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [424] [130/312] eta: 0:02:19 lr: 0.000036 min_lr: 0.000036 loss: 1.7006 (1.6483) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [424] [140/312] eta: 0:02:10 lr: 0.000036 min_lr: 0.000036 loss: 1.6928 (1.6434) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [424] [150/312] eta: 0:02:02 lr: 0.000036 min_lr: 0.000036 loss: 1.6864 (1.6531) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [424] [160/312] eta: 0:01:54 lr: 0.000036 min_lr: 0.000036 loss: 1.7891 (1.6553) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [424] [170/312] eta: 0:01:46 lr: 0.000035 min_lr: 0.000035 loss: 1.8401 (1.6583) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [424] [180/312] eta: 0:01:38 lr: 0.000035 min_lr: 0.000035 loss: 1.8251 (1.6579) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [424] [190/312] eta: 0:01:30 lr: 0.000035 min_lr: 0.000035 loss: 1.6584 (1.6564) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [424] [200/312] eta: 0:01:22 lr: 0.000035 min_lr: 0.000035 loss: 1.7782 (1.6609) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [424] [210/312] eta: 0:01:15 lr: 0.000035 min_lr: 0.000035 loss: 1.8092 (1.6638) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [424] [220/312] eta: 0:01:07 lr: 0.000035 min_lr: 0.000035 loss: 1.8092 (1.6671) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [424] [230/312] eta: 0:01:00 lr: 0.000035 min_lr: 0.000035 loss: 1.7704 (1.6693) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [424] [240/312] eta: 0:00:52 lr: 0.000035 min_lr: 0.000035 loss: 1.7784 (1.6743) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [424] [250/312] eta: 0:00:45 lr: 0.000035 min_lr: 0.000035 loss: 1.7209 (1.6688) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [424] [260/312] eta: 0:00:37 lr: 0.000035 min_lr: 0.000035 loss: 1.5323 (1.6679) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [424] [270/312] eta: 0:00:30 lr: 0.000035 min_lr: 0.000035 loss: 1.7637 (1.6701) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [424] [280/312] eta: 0:00:23 lr: 0.000035 min_lr: 0.000035 loss: 1.8277 (1.6724) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0009 max mem: 64948 Epoch: [424] [290/312] eta: 0:00:15 lr: 0.000034 min_lr: 0.000034 loss: 1.6786 (1.6685) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0008 max mem: 64948 Epoch: [424] [300/312] eta: 0:00:08 lr: 0.000034 min_lr: 0.000034 loss: 1.7124 (1.6716) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [424] [310/312] eta: 0:00:01 lr: 0.000034 min_lr: 0.000034 loss: 1.7124 (1.6686) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [424] [311/312] eta: 0:00:00 lr: 0.000034 min_lr: 0.000034 loss: 1.6973 (1.6678) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [424] Total time: 0:03:46 (0.7269 s / it) Averaged stats: lr: 0.000034 min_lr: 0.000034 loss: 1.6973 (1.6490) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4383 (0.4383) acc1: 89.3229 (89.3229) acc5: 97.9167 (97.9167) time: 4.5470 data: 4.3315 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6270 (0.6399) acc1: 83.3333 (83.0080) acc5: 97.1354 (96.7680) time: 0.6565 data: 0.4814 max mem: 64948 Test: Total time: 0:00:06 (0.6798 s / it) * Acc@1 83.792 Acc@5 96.696 loss 0.620 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.79% Test: [0/9] eta: 0:00:42 loss: 0.4489 (0.4489) acc1: 87.5000 (87.5000) acc5: 97.9167 (97.9167) time: 4.6842 data: 4.4664 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6113 (0.6376) acc1: 84.6354 (83.0400) acc5: 96.8750 (96.6400) time: 0.6720 data: 0.4964 max mem: 64948 Test: Total time: 0:00:06 (0.6792 s / it) * Acc@1 83.716 Acc@5 96.708 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.7% Max EMA accuracy: 83.72% Epoch: [425] [ 0/312] eta: 0:48:18 lr: 0.000034 min_lr: 0.000034 loss: 1.3128 (1.3128) weight_decay: 0.0500 (0.0500) time: 9.2890 data: 8.3252 max mem: 64948 Epoch: [425] [ 10/312] eta: 0:07:42 lr: 0.000034 min_lr: 0.000034 loss: 1.7233 (1.6038) weight_decay: 0.0500 (0.0500) time: 1.5324 data: 0.7573 max mem: 64948 Epoch: [425] [ 20/312] eta: 0:05:31 lr: 0.000034 min_lr: 0.000034 loss: 1.5245 (1.5590) weight_decay: 0.0500 (0.0500) time: 0.7278 data: 0.0004 max mem: 64948 Epoch: [425] [ 30/312] eta: 0:04:39 lr: 0.000034 min_lr: 0.000034 loss: 1.5576 (1.5987) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [425] [ 40/312] eta: 0:04:10 lr: 0.000034 min_lr: 0.000034 loss: 1.7536 (1.6180) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0003 max mem: 64948 Epoch: [425] [ 50/312] eta: 0:03:50 lr: 0.000034 min_lr: 0.000034 loss: 1.7461 (1.6373) weight_decay: 0.0500 (0.0500) time: 0.7017 data: 0.0004 max mem: 64948 Epoch: [425] [ 60/312] eta: 0:03:33 lr: 0.000034 min_lr: 0.000034 loss: 1.8206 (1.6567) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [425] [ 70/312] eta: 0:03:20 lr: 0.000034 min_lr: 0.000034 loss: 1.7558 (1.6587) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [425] [ 80/312] eta: 0:03:08 lr: 0.000034 min_lr: 0.000034 loss: 1.5688 (1.6371) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0003 max mem: 64948 Epoch: [425] [ 90/312] eta: 0:02:57 lr: 0.000033 min_lr: 0.000033 loss: 1.3895 (1.6115) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [425] [100/312] eta: 0:02:47 lr: 0.000033 min_lr: 0.000033 loss: 1.4629 (1.6133) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [425] [110/312] eta: 0:02:37 lr: 0.000033 min_lr: 0.000033 loss: 1.5645 (1.6087) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0003 max mem: 64948 Epoch: [425] [120/312] eta: 0:02:28 lr: 0.000033 min_lr: 0.000033 loss: 1.6802 (1.6179) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [425] [130/312] eta: 0:02:19 lr: 0.000033 min_lr: 0.000033 loss: 1.7329 (1.6170) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [425] [140/312] eta: 0:02:11 lr: 0.000033 min_lr: 0.000033 loss: 1.5304 (1.6100) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [425] [150/312] eta: 0:02:02 lr: 0.000033 min_lr: 0.000033 loss: 1.5304 (1.6060) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [425] [160/312] eta: 0:01:54 lr: 0.000033 min_lr: 0.000033 loss: 1.3700 (1.5969) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [425] [170/312] eta: 0:01:46 lr: 0.000033 min_lr: 0.000033 loss: 1.7632 (1.6107) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [425] [180/312] eta: 0:01:38 lr: 0.000033 min_lr: 0.000033 loss: 1.7632 (1.6093) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [425] [190/312] eta: 0:01:30 lr: 0.000033 min_lr: 0.000033 loss: 1.6784 (1.6088) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [425] [200/312] eta: 0:01:23 lr: 0.000033 min_lr: 0.000033 loss: 1.7371 (1.6140) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [425] [210/312] eta: 0:01:15 lr: 0.000032 min_lr: 0.000032 loss: 1.7048 (1.6140) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [425] [220/312] eta: 0:01:07 lr: 0.000032 min_lr: 0.000032 loss: 1.7653 (1.6214) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [425] [230/312] eta: 0:01:00 lr: 0.000032 min_lr: 0.000032 loss: 1.7844 (1.6209) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [425] [240/312] eta: 0:00:52 lr: 0.000032 min_lr: 0.000032 loss: 1.7643 (1.6294) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0005 max mem: 64948 Epoch: [425] [250/312] eta: 0:00:45 lr: 0.000032 min_lr: 0.000032 loss: 1.8379 (1.6364) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [425] [260/312] eta: 0:00:37 lr: 0.000032 min_lr: 0.000032 loss: 1.7682 (1.6374) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [425] [270/312] eta: 0:00:30 lr: 0.000032 min_lr: 0.000032 loss: 1.7290 (1.6381) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [425] [280/312] eta: 0:00:23 lr: 0.000032 min_lr: 0.000032 loss: 1.7293 (1.6423) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0009 max mem: 64948 Epoch: [425] [290/312] eta: 0:00:15 lr: 0.000032 min_lr: 0.000032 loss: 1.7660 (1.6449) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0008 max mem: 64948 Epoch: [425] [300/312] eta: 0:00:08 lr: 0.000032 min_lr: 0.000032 loss: 1.7451 (1.6441) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [425] [310/312] eta: 0:00:01 lr: 0.000032 min_lr: 0.000032 loss: 1.6106 (1.6437) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [425] [311/312] eta: 0:00:00 lr: 0.000032 min_lr: 0.000032 loss: 1.6106 (1.6419) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [425] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.000032 min_lr: 0.000032 loss: 1.6106 (1.6462) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4470 (0.4470) acc1: 88.2812 (88.2812) acc5: 97.9167 (97.9167) time: 4.6744 data: 4.4692 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6205 (0.6399) acc1: 84.1146 (83.0720) acc5: 97.1354 (96.8320) time: 0.6707 data: 0.4967 max mem: 64948 Test: Total time: 0:00:06 (0.6936 s / it) * Acc@1 83.824 Acc@5 96.742 loss 0.621 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.82% Test: [0/9] eta: 0:00:42 loss: 0.4487 (0.4487) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.7070 data: 4.5012 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6111 (0.6375) acc1: 84.6354 (83.0400) acc5: 97.1354 (96.7040) time: 0.6750 data: 0.5002 max mem: 64948 Test: Total time: 0:00:06 (0.6822 s / it) * Acc@1 83.714 Acc@5 96.704 loss 0.618 Accuracy of the model EMA on 50000 test images: 83.7% Epoch: [426] [ 0/312] eta: 0:54:28 lr: 0.000032 min_lr: 0.000032 loss: 2.0517 (2.0517) weight_decay: 0.0500 (0.0500) time: 10.4772 data: 7.1012 max mem: 64948 Epoch: [426] [ 10/312] eta: 0:08:07 lr: 0.000032 min_lr: 0.000032 loss: 1.7958 (1.7513) weight_decay: 0.0500 (0.0500) time: 1.6128 data: 0.6460 max mem: 64948 Epoch: [426] [ 20/312] eta: 0:05:43 lr: 0.000031 min_lr: 0.000031 loss: 1.6832 (1.7042) weight_decay: 0.0500 (0.0500) time: 0.7098 data: 0.0004 max mem: 64948 Epoch: [426] [ 30/312] eta: 0:04:47 lr: 0.000031 min_lr: 0.000031 loss: 1.6832 (1.7264) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [426] [ 40/312] eta: 0:04:16 lr: 0.000031 min_lr: 0.000031 loss: 1.7751 (1.7371) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [426] [ 50/312] eta: 0:03:54 lr: 0.000031 min_lr: 0.000031 loss: 1.7434 (1.7086) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [426] [ 60/312] eta: 0:03:36 lr: 0.000031 min_lr: 0.000031 loss: 1.6466 (1.6927) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [426] [ 70/312] eta: 0:03:22 lr: 0.000031 min_lr: 0.000031 loss: 1.6860 (1.6904) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [426] [ 80/312] eta: 0:03:10 lr: 0.000031 min_lr: 0.000031 loss: 1.7660 (1.6938) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [426] [ 90/312] eta: 0:02:59 lr: 0.000031 min_lr: 0.000031 loss: 1.7439 (1.6904) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [426] [100/312] eta: 0:02:48 lr: 0.000031 min_lr: 0.000031 loss: 1.6015 (1.6739) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [426] [110/312] eta: 0:02:38 lr: 0.000031 min_lr: 0.000031 loss: 1.7246 (1.6787) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [426] [120/312] eta: 0:02:29 lr: 0.000031 min_lr: 0.000031 loss: 1.7795 (1.6766) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [426] [130/312] eta: 0:02:20 lr: 0.000031 min_lr: 0.000031 loss: 1.7481 (1.6773) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [426] [140/312] eta: 0:02:11 lr: 0.000031 min_lr: 0.000031 loss: 1.7509 (1.6823) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [426] [150/312] eta: 0:02:03 lr: 0.000030 min_lr: 0.000030 loss: 1.7714 (1.6869) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [426] [160/312] eta: 0:01:55 lr: 0.000030 min_lr: 0.000030 loss: 1.7260 (1.6806) weight_decay: 0.0500 (0.0500) time: 0.7004 data: 0.0004 max mem: 64948 Epoch: [426] [170/312] eta: 0:01:47 lr: 0.000030 min_lr: 0.000030 loss: 1.5746 (1.6724) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [426] [180/312] eta: 0:01:39 lr: 0.000030 min_lr: 0.000030 loss: 1.5746 (1.6696) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [426] [190/312] eta: 0:01:31 lr: 0.000030 min_lr: 0.000030 loss: 1.8118 (1.6831) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [426] [200/312] eta: 0:01:23 lr: 0.000030 min_lr: 0.000030 loss: 1.8913 (1.6909) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [426] [210/312] eta: 0:01:15 lr: 0.000030 min_lr: 0.000030 loss: 1.7691 (1.6903) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [426] [220/312] eta: 0:01:08 lr: 0.000030 min_lr: 0.000030 loss: 1.4865 (1.6858) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [426] [230/312] eta: 0:01:00 lr: 0.000030 min_lr: 0.000030 loss: 1.7053 (1.6927) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [426] [240/312] eta: 0:00:53 lr: 0.000030 min_lr: 0.000030 loss: 1.7691 (1.6919) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [426] [250/312] eta: 0:00:45 lr: 0.000030 min_lr: 0.000030 loss: 1.6626 (1.6831) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [426] [260/312] eta: 0:00:38 lr: 0.000030 min_lr: 0.000030 loss: 1.5665 (1.6778) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [426] [270/312] eta: 0:00:30 lr: 0.000029 min_lr: 0.000029 loss: 1.5397 (1.6702) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [426] [280/312] eta: 0:00:23 lr: 0.000029 min_lr: 0.000029 loss: 1.7110 (1.6753) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [426] [290/312] eta: 0:00:16 lr: 0.000029 min_lr: 0.000029 loss: 1.7723 (1.6734) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [426] [300/312] eta: 0:00:08 lr: 0.000029 min_lr: 0.000029 loss: 1.6568 (1.6726) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [426] [310/312] eta: 0:00:01 lr: 0.000029 min_lr: 0.000029 loss: 1.7778 (1.6765) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [426] [311/312] eta: 0:00:00 lr: 0.000029 min_lr: 0.000029 loss: 1.7323 (1.6764) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [426] Total time: 0:03:47 (0.7305 s / it) Averaged stats: lr: 0.000029 min_lr: 0.000029 loss: 1.7323 (1.6558) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4362 (0.4362) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 4.7573 data: 4.5404 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6161 (0.6383) acc1: 84.3750 (82.9760) acc5: 97.1354 (96.8320) time: 0.6806 data: 0.5046 max mem: 64948 Test: Total time: 0:00:06 (0.7043 s / it) * Acc@1 83.740 Acc@5 96.714 loss 0.619 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.82% Test: [0/9] eta: 0:00:46 loss: 0.4483 (0.4483) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.1713 data: 4.9533 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6109 (0.6374) acc1: 84.6354 (83.0400) acc5: 97.1354 (96.7040) time: 0.7262 data: 0.5505 max mem: 64948 Test: Total time: 0:00:06 (0.7348 s / it) * Acc@1 83.706 Acc@5 96.716 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.7% Epoch: [427] [ 0/312] eta: 0:51:07 lr: 0.000029 min_lr: 0.000029 loss: 1.8966 (1.8966) weight_decay: 0.0500 (0.0500) time: 9.8302 data: 8.1577 max mem: 64948 Epoch: [427] [ 10/312] eta: 0:07:55 lr: 0.000029 min_lr: 0.000029 loss: 1.7462 (1.6931) weight_decay: 0.0500 (0.0500) time: 1.5742 data: 0.7421 max mem: 64948 Epoch: [427] [ 20/312] eta: 0:05:37 lr: 0.000029 min_lr: 0.000029 loss: 1.7462 (1.7511) weight_decay: 0.0500 (0.0500) time: 0.7214 data: 0.0004 max mem: 64948 Epoch: [427] [ 30/312] eta: 0:04:43 lr: 0.000029 min_lr: 0.000029 loss: 1.7365 (1.7245) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [427] [ 40/312] eta: 0:04:13 lr: 0.000029 min_lr: 0.000029 loss: 1.7356 (1.6969) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [427] [ 50/312] eta: 0:03:51 lr: 0.000029 min_lr: 0.000029 loss: 1.6884 (1.6891) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [427] [ 60/312] eta: 0:03:35 lr: 0.000029 min_lr: 0.000029 loss: 1.7782 (1.6987) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [427] [ 70/312] eta: 0:03:21 lr: 0.000029 min_lr: 0.000029 loss: 1.7926 (1.7223) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [427] [ 80/312] eta: 0:03:09 lr: 0.000029 min_lr: 0.000029 loss: 1.7405 (1.7056) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [427] [ 90/312] eta: 0:02:57 lr: 0.000028 min_lr: 0.000028 loss: 1.5855 (1.6936) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [427] [100/312] eta: 0:02:47 lr: 0.000028 min_lr: 0.000028 loss: 1.5665 (1.6726) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [427] [110/312] eta: 0:02:38 lr: 0.000028 min_lr: 0.000028 loss: 1.6507 (1.6717) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [427] [120/312] eta: 0:02:28 lr: 0.000028 min_lr: 0.000028 loss: 1.7876 (1.6794) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [427] [130/312] eta: 0:02:19 lr: 0.000028 min_lr: 0.000028 loss: 1.7845 (1.6884) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [427] [140/312] eta: 0:02:11 lr: 0.000028 min_lr: 0.000028 loss: 1.7824 (1.6901) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [427] [150/312] eta: 0:02:02 lr: 0.000028 min_lr: 0.000028 loss: 1.6828 (1.6802) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [427] [160/312] eta: 0:01:54 lr: 0.000028 min_lr: 0.000028 loss: 1.6572 (1.6866) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [427] [170/312] eta: 0:01:46 lr: 0.000028 min_lr: 0.000028 loss: 1.6572 (1.6813) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0003 max mem: 64948 Epoch: [427] [180/312] eta: 0:01:38 lr: 0.000028 min_lr: 0.000028 loss: 1.5687 (1.6775) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [427] [190/312] eta: 0:01:30 lr: 0.000028 min_lr: 0.000028 loss: 1.7460 (1.6802) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [427] [200/312] eta: 0:01:23 lr: 0.000028 min_lr: 0.000028 loss: 1.7786 (1.6796) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [427] [210/312] eta: 0:01:15 lr: 0.000028 min_lr: 0.000028 loss: 1.7450 (1.6826) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [427] [220/312] eta: 0:01:07 lr: 0.000027 min_lr: 0.000027 loss: 1.7071 (1.6800) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [427] [230/312] eta: 0:01:00 lr: 0.000027 min_lr: 0.000027 loss: 1.6271 (1.6775) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [427] [240/312] eta: 0:00:52 lr: 0.000027 min_lr: 0.000027 loss: 1.6271 (1.6720) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [427] [250/312] eta: 0:00:45 lr: 0.000027 min_lr: 0.000027 loss: 1.6704 (1.6751) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [427] [260/312] eta: 0:00:38 lr: 0.000027 min_lr: 0.000027 loss: 1.8064 (1.6759) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [427] [270/312] eta: 0:00:30 lr: 0.000027 min_lr: 0.000027 loss: 1.8106 (1.6814) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [427] [280/312] eta: 0:00:23 lr: 0.000027 min_lr: 0.000027 loss: 1.8093 (1.6789) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [427] [290/312] eta: 0:00:16 lr: 0.000027 min_lr: 0.000027 loss: 1.7148 (1.6826) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0008 max mem: 64948 Epoch: [427] [300/312] eta: 0:00:08 lr: 0.000027 min_lr: 0.000027 loss: 1.7062 (1.6773) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [427] [310/312] eta: 0:00:01 lr: 0.000027 min_lr: 0.000027 loss: 1.6953 (1.6788) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [427] [311/312] eta: 0:00:00 lr: 0.000027 min_lr: 0.000027 loss: 1.5982 (1.6786) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [427] Total time: 0:03:47 (0.7283 s / it) Averaged stats: lr: 0.000027 min_lr: 0.000027 loss: 1.5982 (1.6611) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.4383 (0.4383) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 4.9832 data: 4.7767 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6122 (0.6379) acc1: 84.3750 (83.0080) acc5: 97.1354 (96.6720) time: 0.7050 data: 0.5308 max mem: 64948 Test: Total time: 0:00:06 (0.7276 s / it) * Acc@1 83.762 Acc@5 96.682 loss 0.620 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.82% Test: [0/9] eta: 0:00:46 loss: 0.4481 (0.4481) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.1375 data: 4.9194 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6106 (0.6374) acc1: 84.6354 (82.9760) acc5: 97.1354 (96.7680) time: 0.7267 data: 0.5467 max mem: 64948 Test: Total time: 0:00:06 (0.7367 s / it) * Acc@1 83.718 Acc@5 96.716 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.7% Max EMA accuracy: 83.72% Epoch: [428] [ 0/312] eta: 0:43:55 lr: 0.000027 min_lr: 0.000027 loss: 1.0089 (1.0089) weight_decay: 0.0500 (0.0500) time: 8.4477 data: 7.6579 max mem: 64948 Epoch: [428] [ 10/312] eta: 0:07:14 lr: 0.000027 min_lr: 0.000027 loss: 1.7412 (1.5879) weight_decay: 0.0500 (0.0500) time: 1.4400 data: 0.6965 max mem: 64948 Epoch: [428] [ 20/312] eta: 0:05:16 lr: 0.000027 min_lr: 0.000027 loss: 1.5788 (1.5585) weight_decay: 0.0500 (0.0500) time: 0.7161 data: 0.0004 max mem: 64948 Epoch: [428] [ 30/312] eta: 0:04:30 lr: 0.000027 min_lr: 0.000027 loss: 1.4683 (1.5855) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [428] [ 40/312] eta: 0:04:03 lr: 0.000026 min_lr: 0.000026 loss: 1.7160 (1.5957) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [428] [ 50/312] eta: 0:03:44 lr: 0.000026 min_lr: 0.000026 loss: 1.7137 (1.6259) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [428] [ 60/312] eta: 0:03:28 lr: 0.000026 min_lr: 0.000026 loss: 1.7574 (1.6374) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [428] [ 70/312] eta: 0:03:16 lr: 0.000026 min_lr: 0.000026 loss: 1.8264 (1.6531) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [428] [ 80/312] eta: 0:03:04 lr: 0.000026 min_lr: 0.000026 loss: 1.7011 (1.6452) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [428] [ 90/312] eta: 0:02:54 lr: 0.000026 min_lr: 0.000026 loss: 1.5689 (1.6368) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [428] [100/312] eta: 0:02:44 lr: 0.000026 min_lr: 0.000026 loss: 1.6370 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [428] [110/312] eta: 0:02:35 lr: 0.000026 min_lr: 0.000026 loss: 1.7561 (1.6519) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [428] [120/312] eta: 0:02:26 lr: 0.000026 min_lr: 0.000026 loss: 1.7561 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0004 max mem: 64948 Epoch: [428] [130/312] eta: 0:02:17 lr: 0.000026 min_lr: 0.000026 loss: 1.6860 (1.6488) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [428] [140/312] eta: 0:02:09 lr: 0.000026 min_lr: 0.000026 loss: 1.7043 (1.6508) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [428] [150/312] eta: 0:02:01 lr: 0.000026 min_lr: 0.000026 loss: 1.7379 (1.6561) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [428] [160/312] eta: 0:01:53 lr: 0.000026 min_lr: 0.000026 loss: 1.6804 (1.6533) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [428] [170/312] eta: 0:01:45 lr: 0.000026 min_lr: 0.000026 loss: 1.6572 (1.6549) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [428] [180/312] eta: 0:01:37 lr: 0.000025 min_lr: 0.000025 loss: 1.6833 (1.6522) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [428] [190/312] eta: 0:01:30 lr: 0.000025 min_lr: 0.000025 loss: 1.4740 (1.6439) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [428] [200/312] eta: 0:01:22 lr: 0.000025 min_lr: 0.000025 loss: 1.7115 (1.6507) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [428] [210/312] eta: 0:01:14 lr: 0.000025 min_lr: 0.000025 loss: 1.7293 (1.6449) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [428] [220/312] eta: 0:01:07 lr: 0.000025 min_lr: 0.000025 loss: 1.7306 (1.6458) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [428] [230/312] eta: 0:00:59 lr: 0.000025 min_lr: 0.000025 loss: 1.6024 (1.6457) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [428] [240/312] eta: 0:00:52 lr: 0.000025 min_lr: 0.000025 loss: 1.5598 (1.6431) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [428] [250/312] eta: 0:00:45 lr: 0.000025 min_lr: 0.000025 loss: 1.6496 (1.6414) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [428] [260/312] eta: 0:00:37 lr: 0.000025 min_lr: 0.000025 loss: 1.6917 (1.6397) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [428] [270/312] eta: 0:00:30 lr: 0.000025 min_lr: 0.000025 loss: 1.6917 (1.6357) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [428] [280/312] eta: 0:00:23 lr: 0.000025 min_lr: 0.000025 loss: 1.6514 (1.6386) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0009 max mem: 64948 Epoch: [428] [290/312] eta: 0:00:15 lr: 0.000025 min_lr: 0.000025 loss: 1.7594 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [428] [300/312] eta: 0:00:08 lr: 0.000025 min_lr: 0.000025 loss: 1.7223 (1.6449) weight_decay: 0.0500 (0.0500) time: 0.6918 data: 0.0001 max mem: 64948 Epoch: [428] [310/312] eta: 0:00:01 lr: 0.000025 min_lr: 0.000025 loss: 1.6541 (1.6424) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [428] [311/312] eta: 0:00:00 lr: 0.000024 min_lr: 0.000024 loss: 1.6541 (1.6426) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [428] Total time: 0:03:45 (0.7235 s / it) Averaged stats: lr: 0.000024 min_lr: 0.000024 loss: 1.6541 (1.6521) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4419 (0.4419) acc1: 89.3229 (89.3229) acc5: 97.9167 (97.9167) time: 4.4485 data: 4.2306 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6160 (0.6363) acc1: 84.3750 (83.1040) acc5: 97.1354 (96.7680) time: 0.6538 data: 0.4784 max mem: 64948 Test: Total time: 0:00:06 (0.6832 s / it) * Acc@1 83.798 Acc@5 96.728 loss 0.621 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.82% Test: [0/9] eta: 0:00:46 loss: 0.4479 (0.4479) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.1249 data: 4.9226 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6103 (0.6372) acc1: 84.6354 (82.9120) acc5: 97.1354 (96.7680) time: 0.7207 data: 0.5470 max mem: 64948 Test: Total time: 0:00:06 (0.7335 s / it) * Acc@1 83.724 Acc@5 96.720 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.7% Max EMA accuracy: 83.72% Epoch: [429] [ 0/312] eta: 0:43:31 lr: 0.000024 min_lr: 0.000024 loss: 1.1086 (1.1086) weight_decay: 0.0500 (0.0500) time: 8.3698 data: 7.1642 max mem: 64948 Epoch: [429] [ 10/312] eta: 0:07:20 lr: 0.000024 min_lr: 0.000024 loss: 1.7700 (1.6201) weight_decay: 0.0500 (0.0500) time: 1.4595 data: 0.6518 max mem: 64948 Epoch: [429] [ 20/312] eta: 0:05:19 lr: 0.000024 min_lr: 0.000024 loss: 1.7260 (1.6056) weight_decay: 0.0500 (0.0500) time: 0.7305 data: 0.0004 max mem: 64948 Epoch: [429] [ 30/312] eta: 0:04:32 lr: 0.000024 min_lr: 0.000024 loss: 1.6821 (1.6039) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [429] [ 40/312] eta: 0:04:04 lr: 0.000024 min_lr: 0.000024 loss: 1.6408 (1.5959) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [429] [ 50/312] eta: 0:03:45 lr: 0.000024 min_lr: 0.000024 loss: 1.6408 (1.6081) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [429] [ 60/312] eta: 0:03:29 lr: 0.000024 min_lr: 0.000024 loss: 1.6628 (1.6030) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [429] [ 70/312] eta: 0:03:16 lr: 0.000024 min_lr: 0.000024 loss: 1.6728 (1.6164) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [429] [ 80/312] eta: 0:03:05 lr: 0.000024 min_lr: 0.000024 loss: 1.7096 (1.6258) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [429] [ 90/312] eta: 0:02:54 lr: 0.000024 min_lr: 0.000024 loss: 1.7457 (1.6362) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [429] [100/312] eta: 0:02:45 lr: 0.000024 min_lr: 0.000024 loss: 1.6639 (1.6264) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [429] [110/312] eta: 0:02:35 lr: 0.000024 min_lr: 0.000024 loss: 1.6531 (1.6328) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [429] [120/312] eta: 0:02:26 lr: 0.000024 min_lr: 0.000024 loss: 1.6531 (1.6219) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [429] [130/312] eta: 0:02:18 lr: 0.000024 min_lr: 0.000024 loss: 1.4411 (1.6255) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [429] [140/312] eta: 0:02:09 lr: 0.000023 min_lr: 0.000023 loss: 1.4873 (1.6247) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [429] [150/312] eta: 0:02:01 lr: 0.000023 min_lr: 0.000023 loss: 1.5193 (1.6261) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [429] [160/312] eta: 0:01:53 lr: 0.000023 min_lr: 0.000023 loss: 1.5461 (1.6262) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [429] [170/312] eta: 0:01:45 lr: 0.000023 min_lr: 0.000023 loss: 1.5461 (1.6200) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [429] [180/312] eta: 0:01:37 lr: 0.000023 min_lr: 0.000023 loss: 1.6628 (1.6235) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [429] [190/312] eta: 0:01:30 lr: 0.000023 min_lr: 0.000023 loss: 1.8723 (1.6344) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [429] [200/312] eta: 0:01:22 lr: 0.000023 min_lr: 0.000023 loss: 1.8723 (1.6279) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [429] [210/312] eta: 0:01:14 lr: 0.000023 min_lr: 0.000023 loss: 1.3913 (1.6210) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [429] [220/312] eta: 0:01:07 lr: 0.000023 min_lr: 0.000023 loss: 1.6255 (1.6287) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [429] [230/312] eta: 0:00:59 lr: 0.000023 min_lr: 0.000023 loss: 1.8039 (1.6273) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [429] [240/312] eta: 0:00:52 lr: 0.000023 min_lr: 0.000023 loss: 1.5422 (1.6258) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [429] [250/312] eta: 0:00:45 lr: 0.000023 min_lr: 0.000023 loss: 1.7184 (1.6343) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [429] [260/312] eta: 0:00:37 lr: 0.000023 min_lr: 0.000023 loss: 1.7605 (1.6371) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [429] [270/312] eta: 0:00:30 lr: 0.000023 min_lr: 0.000023 loss: 1.6769 (1.6371) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [429] [280/312] eta: 0:00:23 lr: 0.000023 min_lr: 0.000023 loss: 1.7721 (1.6381) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0010 max mem: 64948 Epoch: [429] [290/312] eta: 0:00:15 lr: 0.000022 min_lr: 0.000022 loss: 1.7699 (1.6382) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [429] [300/312] eta: 0:00:08 lr: 0.000022 min_lr: 0.000022 loss: 1.7814 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [429] [310/312] eta: 0:00:01 lr: 0.000022 min_lr: 0.000022 loss: 1.8308 (1.6460) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [429] [311/312] eta: 0:00:00 lr: 0.000022 min_lr: 0.000022 loss: 1.8308 (1.6470) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [429] Total time: 0:03:46 (0.7244 s / it) Averaged stats: lr: 0.000022 min_lr: 0.000022 loss: 1.8308 (1.6547) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4389 (0.4389) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 4.8884 data: 4.6791 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6075 (0.6361) acc1: 84.8958 (83.0080) acc5: 97.1354 (96.7360) time: 0.6945 data: 0.5200 max mem: 64948 Test: Total time: 0:00:06 (0.7179 s / it) * Acc@1 83.800 Acc@5 96.686 loss 0.619 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.82% Test: [0/9] eta: 0:00:47 loss: 0.4476 (0.4476) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 5.2277 data: 5.0097 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6098 (0.6370) acc1: 84.6354 (82.9440) acc5: 97.1354 (96.7040) time: 0.7325 data: 0.5567 max mem: 64948 Test: Total time: 0:00:06 (0.7401 s / it) * Acc@1 83.724 Acc@5 96.716 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.7% Max EMA accuracy: 83.72% Epoch: [430] [ 0/312] eta: 0:43:06 lr: 0.000022 min_lr: 0.000022 loss: 1.0166 (1.0166) weight_decay: 0.0500 (0.0500) time: 8.2909 data: 7.5027 max mem: 64948 Epoch: [430] [ 10/312] eta: 0:07:26 lr: 0.000022 min_lr: 0.000022 loss: 1.6998 (1.6316) weight_decay: 0.0500 (0.0500) time: 1.4783 data: 0.7543 max mem: 64948 Epoch: [430] [ 20/312] eta: 0:05:22 lr: 0.000022 min_lr: 0.000022 loss: 1.6998 (1.6735) weight_decay: 0.0500 (0.0500) time: 0.7444 data: 0.0399 max mem: 64948 Epoch: [430] [ 30/312] eta: 0:04:34 lr: 0.000022 min_lr: 0.000022 loss: 1.7349 (1.6920) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0003 max mem: 64948 Epoch: [430] [ 40/312] eta: 0:04:06 lr: 0.000022 min_lr: 0.000022 loss: 1.7358 (1.6918) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [430] [ 50/312] eta: 0:03:46 lr: 0.000022 min_lr: 0.000022 loss: 1.6849 (1.6681) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [430] [ 60/312] eta: 0:03:31 lr: 0.000022 min_lr: 0.000022 loss: 1.6542 (1.6646) weight_decay: 0.0500 (0.0500) time: 0.7012 data: 0.0004 max mem: 64948 Epoch: [430] [ 70/312] eta: 0:03:18 lr: 0.000022 min_lr: 0.000022 loss: 1.6542 (1.6623) weight_decay: 0.0500 (0.0500) time: 0.7028 data: 0.0004 max mem: 64948 Epoch: [430] [ 80/312] eta: 0:03:06 lr: 0.000022 min_lr: 0.000022 loss: 1.7449 (1.6669) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [430] [ 90/312] eta: 0:02:55 lr: 0.000022 min_lr: 0.000022 loss: 1.7449 (1.6613) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [430] [100/312] eta: 0:02:45 lr: 0.000022 min_lr: 0.000022 loss: 1.6905 (1.6558) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [430] [110/312] eta: 0:02:36 lr: 0.000022 min_lr: 0.000022 loss: 1.4981 (1.6487) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [430] [120/312] eta: 0:02:27 lr: 0.000021 min_lr: 0.000021 loss: 1.7556 (1.6550) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [430] [130/312] eta: 0:02:18 lr: 0.000021 min_lr: 0.000021 loss: 1.7556 (1.6526) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [430] [140/312] eta: 0:02:10 lr: 0.000021 min_lr: 0.000021 loss: 1.6808 (1.6482) weight_decay: 0.0500 (0.0500) time: 0.7013 data: 0.0004 max mem: 64948 Epoch: [430] [150/312] eta: 0:02:02 lr: 0.000021 min_lr: 0.000021 loss: 1.6808 (1.6505) weight_decay: 0.0500 (0.0500) time: 0.7013 data: 0.0004 max mem: 64948 Epoch: [430] [160/312] eta: 0:01:53 lr: 0.000021 min_lr: 0.000021 loss: 1.5604 (1.6413) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [430] [170/312] eta: 0:01:46 lr: 0.000021 min_lr: 0.000021 loss: 1.5604 (1.6459) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [430] [180/312] eta: 0:01:38 lr: 0.000021 min_lr: 0.000021 loss: 1.6156 (1.6440) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [430] [190/312] eta: 0:01:30 lr: 0.000021 min_lr: 0.000021 loss: 1.6124 (1.6428) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [430] [200/312] eta: 0:01:22 lr: 0.000021 min_lr: 0.000021 loss: 1.7780 (1.6477) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [430] [210/312] eta: 0:01:15 lr: 0.000021 min_lr: 0.000021 loss: 1.5663 (1.6442) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [430] [220/312] eta: 0:01:07 lr: 0.000021 min_lr: 0.000021 loss: 1.5128 (1.6402) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [430] [230/312] eta: 0:01:00 lr: 0.000021 min_lr: 0.000021 loss: 1.6479 (1.6425) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [430] [240/312] eta: 0:00:52 lr: 0.000021 min_lr: 0.000021 loss: 1.8014 (1.6454) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [430] [250/312] eta: 0:00:45 lr: 0.000021 min_lr: 0.000021 loss: 1.6631 (1.6431) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [430] [260/312] eta: 0:00:37 lr: 0.000021 min_lr: 0.000021 loss: 1.5579 (1.6423) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [430] [270/312] eta: 0:00:30 lr: 0.000021 min_lr: 0.000021 loss: 1.7393 (1.6437) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [430] [280/312] eta: 0:00:23 lr: 0.000020 min_lr: 0.000020 loss: 1.7393 (1.6438) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0009 max mem: 64948 Epoch: [430] [290/312] eta: 0:00:15 lr: 0.000020 min_lr: 0.000020 loss: 1.7011 (1.6411) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0008 max mem: 64948 Epoch: [430] [300/312] eta: 0:00:08 lr: 0.000020 min_lr: 0.000020 loss: 1.6942 (1.6402) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [430] [310/312] eta: 0:00:01 lr: 0.000020 min_lr: 0.000020 loss: 1.6379 (1.6346) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [430] [311/312] eta: 0:00:00 lr: 0.000020 min_lr: 0.000020 loss: 1.6379 (1.6348) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [430] Total time: 0:03:46 (0.7259 s / it) Averaged stats: lr: 0.000020 min_lr: 0.000020 loss: 1.6379 (1.6506) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4381 (0.4381) acc1: 89.0625 (89.0625) acc5: 97.9167 (97.9167) time: 4.5196 data: 4.2991 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6074 (0.6343) acc1: 84.1146 (82.9120) acc5: 97.3958 (96.8960) time: 0.6535 data: 0.4778 max mem: 64948 Test: Total time: 0:00:06 (0.6789 s / it) * Acc@1 83.774 Acc@5 96.762 loss 0.621 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.82% Test: [0/9] eta: 0:00:42 loss: 0.4472 (0.4472) acc1: 87.7604 (87.7604) acc5: 97.9167 (97.9167) time: 4.7523 data: 4.5436 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6092 (0.6368) acc1: 84.6354 (82.9440) acc5: 97.1354 (96.7040) time: 0.6793 data: 0.5049 max mem: 64948 Test: Total time: 0:00:06 (0.6885 s / it) * Acc@1 83.750 Acc@5 96.716 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.75% Epoch: [431] [ 0/312] eta: 0:47:20 lr: 0.000020 min_lr: 0.000020 loss: 1.7647 (1.7647) weight_decay: 0.0500 (0.0500) time: 9.1042 data: 7.5078 max mem: 64948 Epoch: [431] [ 10/312] eta: 0:07:29 lr: 0.000020 min_lr: 0.000020 loss: 1.7505 (1.7128) weight_decay: 0.0500 (0.0500) time: 1.4896 data: 0.6830 max mem: 64948 Epoch: [431] [ 20/312] eta: 0:05:24 lr: 0.000020 min_lr: 0.000020 loss: 1.7862 (1.7081) weight_decay: 0.0500 (0.0500) time: 0.7118 data: 0.0004 max mem: 64948 Epoch: [431] [ 30/312] eta: 0:04:35 lr: 0.000020 min_lr: 0.000020 loss: 1.8039 (1.7453) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [431] [ 40/312] eta: 0:04:07 lr: 0.000020 min_lr: 0.000020 loss: 1.7602 (1.7166) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [431] [ 50/312] eta: 0:03:47 lr: 0.000020 min_lr: 0.000020 loss: 1.7020 (1.6911) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [431] [ 60/312] eta: 0:03:31 lr: 0.000020 min_lr: 0.000020 loss: 1.7178 (1.6854) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [431] [ 70/312] eta: 0:03:18 lr: 0.000020 min_lr: 0.000020 loss: 1.6608 (1.6641) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [431] [ 80/312] eta: 0:03:06 lr: 0.000020 min_lr: 0.000020 loss: 1.5679 (1.6622) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [431] [ 90/312] eta: 0:02:55 lr: 0.000020 min_lr: 0.000020 loss: 1.5736 (1.6601) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [431] [100/312] eta: 0:02:45 lr: 0.000020 min_lr: 0.000020 loss: 1.5607 (1.6504) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [431] [110/312] eta: 0:02:36 lr: 0.000020 min_lr: 0.000020 loss: 1.5700 (1.6441) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [431] [120/312] eta: 0:02:27 lr: 0.000019 min_lr: 0.000019 loss: 1.5964 (1.6410) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [431] [130/312] eta: 0:02:18 lr: 0.000019 min_lr: 0.000019 loss: 1.5941 (1.6360) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [431] [140/312] eta: 0:02:10 lr: 0.000019 min_lr: 0.000019 loss: 1.6010 (1.6323) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [431] [150/312] eta: 0:02:02 lr: 0.000019 min_lr: 0.000019 loss: 1.4935 (1.6280) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [431] [160/312] eta: 0:01:53 lr: 0.000019 min_lr: 0.000019 loss: 1.6990 (1.6334) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [431] [170/312] eta: 0:01:46 lr: 0.000019 min_lr: 0.000019 loss: 1.7305 (1.6338) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [431] [180/312] eta: 0:01:38 lr: 0.000019 min_lr: 0.000019 loss: 1.7744 (1.6386) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [431] [190/312] eta: 0:01:30 lr: 0.000019 min_lr: 0.000019 loss: 1.7313 (1.6328) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [431] [200/312] eta: 0:01:22 lr: 0.000019 min_lr: 0.000019 loss: 1.7300 (1.6409) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [431] [210/312] eta: 0:01:15 lr: 0.000019 min_lr: 0.000019 loss: 1.8058 (1.6444) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [431] [220/312] eta: 0:01:07 lr: 0.000019 min_lr: 0.000019 loss: 1.6900 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [431] [230/312] eta: 0:01:00 lr: 0.000019 min_lr: 0.000019 loss: 1.6493 (1.6431) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [431] [240/312] eta: 0:00:52 lr: 0.000019 min_lr: 0.000019 loss: 1.7079 (1.6416) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [431] [250/312] eta: 0:00:45 lr: 0.000019 min_lr: 0.000019 loss: 1.7079 (1.6444) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [431] [260/312] eta: 0:00:37 lr: 0.000019 min_lr: 0.000019 loss: 1.6310 (1.6403) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [431] [270/312] eta: 0:00:30 lr: 0.000019 min_lr: 0.000019 loss: 1.6752 (1.6435) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [431] [280/312] eta: 0:00:23 lr: 0.000018 min_lr: 0.000018 loss: 1.7707 (1.6432) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0010 max mem: 64948 Epoch: [431] [290/312] eta: 0:00:15 lr: 0.000018 min_lr: 0.000018 loss: 1.6293 (1.6429) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0009 max mem: 64948 Epoch: [431] [300/312] eta: 0:00:08 lr: 0.000018 min_lr: 0.000018 loss: 1.7479 (1.6493) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [431] [310/312] eta: 0:00:01 lr: 0.000018 min_lr: 0.000018 loss: 1.7919 (1.6539) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [431] [311/312] eta: 0:00:00 lr: 0.000018 min_lr: 0.000018 loss: 1.7919 (1.6537) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [431] Total time: 0:03:46 (0.7257 s / it) Averaged stats: lr: 0.000018 min_lr: 0.000018 loss: 1.7919 (1.6523) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:50 loss: 0.4388 (0.4388) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 5.6593 data: 5.4386 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6118 (0.6309) acc1: 84.1146 (82.8480) acc5: 97.1354 (96.7360) time: 0.7801 data: 0.6044 max mem: 64948 Test: Total time: 0:00:07 (0.8008 s / it) * Acc@1 83.774 Acc@5 96.698 loss 0.618 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.82% Test: [0/9] eta: 0:00:49 loss: 0.4470 (0.4470) acc1: 88.0208 (88.0208) acc5: 97.9167 (97.9167) time: 5.4526 data: 5.2346 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6090 (0.6366) acc1: 84.6354 (82.9440) acc5: 97.1354 (96.7360) time: 0.7593 data: 0.5817 max mem: 64948 Test: Total time: 0:00:06 (0.7708 s / it) * Acc@1 83.746 Acc@5 96.718 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.7% Epoch: [432] [ 0/312] eta: 0:48:56 lr: 0.000018 min_lr: 0.000018 loss: 1.9613 (1.9613) weight_decay: 0.0500 (0.0500) time: 9.4113 data: 7.9328 max mem: 64948 Epoch: [432] [ 10/312] eta: 0:07:45 lr: 0.000018 min_lr: 0.000018 loss: 1.8316 (1.7266) weight_decay: 0.0500 (0.0500) time: 1.5401 data: 0.7216 max mem: 64948 Epoch: [432] [ 20/312] eta: 0:05:31 lr: 0.000018 min_lr: 0.000018 loss: 1.7492 (1.6533) weight_decay: 0.0500 (0.0500) time: 0.7224 data: 0.0004 max mem: 64948 Epoch: [432] [ 30/312] eta: 0:04:40 lr: 0.000018 min_lr: 0.000018 loss: 1.7477 (1.7092) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0003 max mem: 64948 Epoch: [432] [ 40/312] eta: 0:04:10 lr: 0.000018 min_lr: 0.000018 loss: 1.7260 (1.6789) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [432] [ 50/312] eta: 0:03:49 lr: 0.000018 min_lr: 0.000018 loss: 1.6219 (1.6777) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [432] [ 60/312] eta: 0:03:33 lr: 0.000018 min_lr: 0.000018 loss: 1.5920 (1.6538) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [432] [ 70/312] eta: 0:03:19 lr: 0.000018 min_lr: 0.000018 loss: 1.6067 (1.6441) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [432] [ 80/312] eta: 0:03:07 lr: 0.000018 min_lr: 0.000018 loss: 1.7398 (1.6548) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [432] [ 90/312] eta: 0:02:56 lr: 0.000018 min_lr: 0.000018 loss: 1.7955 (1.6628) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [432] [100/312] eta: 0:02:46 lr: 0.000018 min_lr: 0.000018 loss: 1.7589 (1.6678) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [432] [110/312] eta: 0:02:37 lr: 0.000018 min_lr: 0.000018 loss: 1.7589 (1.6653) weight_decay: 0.0500 (0.0500) time: 0.6988 data: 0.0004 max mem: 64948 Epoch: [432] [120/312] eta: 0:02:28 lr: 0.000018 min_lr: 0.000018 loss: 1.6642 (1.6529) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [432] [130/312] eta: 0:02:19 lr: 0.000017 min_lr: 0.000017 loss: 1.5498 (1.6515) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [432] [140/312] eta: 0:02:10 lr: 0.000017 min_lr: 0.000017 loss: 1.6761 (1.6587) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [432] [150/312] eta: 0:02:02 lr: 0.000017 min_lr: 0.000017 loss: 1.8112 (1.6646) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [432] [160/312] eta: 0:01:54 lr: 0.000017 min_lr: 0.000017 loss: 1.8491 (1.6720) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [432] [170/312] eta: 0:01:46 lr: 0.000017 min_lr: 0.000017 loss: 1.6702 (1.6639) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [432] [180/312] eta: 0:01:38 lr: 0.000017 min_lr: 0.000017 loss: 1.6132 (1.6630) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [432] [190/312] eta: 0:01:30 lr: 0.000017 min_lr: 0.000017 loss: 1.5333 (1.6505) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [432] [200/312] eta: 0:01:23 lr: 0.000017 min_lr: 0.000017 loss: 1.5733 (1.6489) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [432] [210/312] eta: 0:01:15 lr: 0.000017 min_lr: 0.000017 loss: 1.7225 (1.6511) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [432] [220/312] eta: 0:01:07 lr: 0.000017 min_lr: 0.000017 loss: 1.7158 (1.6512) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [432] [230/312] eta: 0:01:00 lr: 0.000017 min_lr: 0.000017 loss: 1.5982 (1.6502) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [432] [240/312] eta: 0:00:52 lr: 0.000017 min_lr: 0.000017 loss: 1.6189 (1.6510) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [432] [250/312] eta: 0:00:45 lr: 0.000017 min_lr: 0.000017 loss: 1.7329 (1.6539) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [432] [260/312] eta: 0:00:38 lr: 0.000017 min_lr: 0.000017 loss: 1.7779 (1.6574) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [432] [270/312] eta: 0:00:30 lr: 0.000017 min_lr: 0.000017 loss: 1.7290 (1.6540) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [432] [280/312] eta: 0:00:23 lr: 0.000017 min_lr: 0.000017 loss: 1.6730 (1.6520) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0009 max mem: 64948 Epoch: [432] [290/312] eta: 0:00:15 lr: 0.000017 min_lr: 0.000017 loss: 1.4882 (1.6448) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0008 max mem: 64948 Epoch: [432] [300/312] eta: 0:00:08 lr: 0.000016 min_lr: 0.000016 loss: 1.5373 (1.6431) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [432] [310/312] eta: 0:00:01 lr: 0.000016 min_lr: 0.000016 loss: 1.6469 (1.6466) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [432] [311/312] eta: 0:00:00 lr: 0.000016 min_lr: 0.000016 loss: 1.6251 (1.6447) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [432] Total time: 0:03:47 (0.7276 s / it) Averaged stats: lr: 0.000016 min_lr: 0.000016 loss: 1.6251 (1.6501) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4362 (0.4362) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.8185 data: 4.5992 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6196 (0.6354) acc1: 84.1146 (82.8800) acc5: 97.1354 (96.8640) time: 0.6872 data: 0.5111 max mem: 64948 Test: Total time: 0:00:06 (0.7113 s / it) * Acc@1 83.848 Acc@5 96.730 loss 0.620 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:40 loss: 0.4466 (0.4466) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 4.5395 data: 4.3354 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6092 (0.6364) acc1: 84.6354 (83.0080) acc5: 97.3958 (96.7680) time: 0.6557 data: 0.4818 max mem: 64948 Test: Total time: 0:00:05 (0.6641 s / it) * Acc@1 83.752 Acc@5 96.726 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.75% Epoch: [433] [ 0/312] eta: 0:52:32 lr: 0.000016 min_lr: 0.000016 loss: 1.8151 (1.8151) weight_decay: 0.0500 (0.0500) time: 10.1035 data: 9.3251 max mem: 64948 Epoch: [433] [ 10/312] eta: 0:08:04 lr: 0.000016 min_lr: 0.000016 loss: 1.8069 (1.6949) weight_decay: 0.0500 (0.0500) time: 1.6036 data: 0.8480 max mem: 64948 Epoch: [433] [ 20/312] eta: 0:05:42 lr: 0.000016 min_lr: 0.000016 loss: 1.7014 (1.6706) weight_decay: 0.0500 (0.0500) time: 0.7279 data: 0.0004 max mem: 64948 Epoch: [433] [ 30/312] eta: 0:04:47 lr: 0.000016 min_lr: 0.000016 loss: 1.7014 (1.6845) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [433] [ 40/312] eta: 0:04:15 lr: 0.000016 min_lr: 0.000016 loss: 1.7720 (1.6870) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [433] [ 50/312] eta: 0:03:54 lr: 0.000016 min_lr: 0.000016 loss: 1.6908 (1.6869) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [433] [ 60/312] eta: 0:03:36 lr: 0.000016 min_lr: 0.000016 loss: 1.7430 (1.7015) weight_decay: 0.0500 (0.0500) time: 0.6983 data: 0.0004 max mem: 64948 Epoch: [433] [ 70/312] eta: 0:03:22 lr: 0.000016 min_lr: 0.000016 loss: 1.7430 (1.6900) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [433] [ 80/312] eta: 0:03:10 lr: 0.000016 min_lr: 0.000016 loss: 1.5942 (1.6696) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [433] [ 90/312] eta: 0:02:59 lr: 0.000016 min_lr: 0.000016 loss: 1.6591 (1.6705) weight_decay: 0.0500 (0.0500) time: 0.6971 data: 0.0004 max mem: 64948 Epoch: [433] [100/312] eta: 0:02:48 lr: 0.000016 min_lr: 0.000016 loss: 1.7361 (1.6709) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [433] [110/312] eta: 0:02:38 lr: 0.000016 min_lr: 0.000016 loss: 1.6204 (1.6519) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [433] [120/312] eta: 0:02:29 lr: 0.000016 min_lr: 0.000016 loss: 1.4845 (1.6546) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [433] [130/312] eta: 0:02:20 lr: 0.000016 min_lr: 0.000016 loss: 1.7210 (1.6565) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [433] [140/312] eta: 0:02:11 lr: 0.000016 min_lr: 0.000016 loss: 1.7210 (1.6591) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [433] [150/312] eta: 0:02:03 lr: 0.000016 min_lr: 0.000016 loss: 1.6572 (1.6551) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [433] [160/312] eta: 0:01:55 lr: 0.000015 min_lr: 0.000015 loss: 1.6075 (1.6539) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [433] [170/312] eta: 0:01:47 lr: 0.000015 min_lr: 0.000015 loss: 1.7038 (1.6563) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [433] [180/312] eta: 0:01:39 lr: 0.000015 min_lr: 0.000015 loss: 1.7286 (1.6531) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [433] [190/312] eta: 0:01:31 lr: 0.000015 min_lr: 0.000015 loss: 1.7286 (1.6557) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [433] [200/312] eta: 0:01:23 lr: 0.000015 min_lr: 0.000015 loss: 1.7828 (1.6579) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [433] [210/312] eta: 0:01:15 lr: 0.000015 min_lr: 0.000015 loss: 1.7274 (1.6556) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [433] [220/312] eta: 0:01:08 lr: 0.000015 min_lr: 0.000015 loss: 1.7274 (1.6565) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [433] [230/312] eta: 0:01:00 lr: 0.000015 min_lr: 0.000015 loss: 1.7402 (1.6591) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [433] [240/312] eta: 0:00:53 lr: 0.000015 min_lr: 0.000015 loss: 1.7441 (1.6581) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [433] [250/312] eta: 0:00:45 lr: 0.000015 min_lr: 0.000015 loss: 1.7396 (1.6568) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [433] [260/312] eta: 0:00:38 lr: 0.000015 min_lr: 0.000015 loss: 1.7758 (1.6637) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [433] [270/312] eta: 0:00:30 lr: 0.000015 min_lr: 0.000015 loss: 1.7758 (1.6619) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [433] [280/312] eta: 0:00:23 lr: 0.000015 min_lr: 0.000015 loss: 1.7021 (1.6659) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [433] [290/312] eta: 0:00:16 lr: 0.000015 min_lr: 0.000015 loss: 1.6340 (1.6622) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [433] [300/312] eta: 0:00:08 lr: 0.000015 min_lr: 0.000015 loss: 1.6305 (1.6662) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [433] [310/312] eta: 0:00:01 lr: 0.000015 min_lr: 0.000015 loss: 1.7446 (1.6699) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [433] [311/312] eta: 0:00:00 lr: 0.000015 min_lr: 0.000015 loss: 1.7237 (1.6679) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [433] Total time: 0:03:47 (0.7299 s / it) Averaged stats: lr: 0.000015 min_lr: 0.000015 loss: 1.7237 (1.6476) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:47 loss: 0.4370 (0.4370) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 5.2732 data: 5.0534 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6135 (0.6343) acc1: 84.3750 (83.0400) acc5: 97.1354 (96.7680) time: 0.7376 data: 0.5616 max mem: 64948 Test: Total time: 0:00:06 (0.7650 s / it) * Acc@1 83.808 Acc@5 96.712 loss 0.619 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:48 loss: 0.4462 (0.4462) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 5.4133 data: 5.1991 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6092 (0.6363) acc1: 84.8958 (83.0080) acc5: 97.3958 (96.7680) time: 0.7527 data: 0.5778 max mem: 64948 Test: Total time: 0:00:06 (0.7602 s / it) * Acc@1 83.768 Acc@5 96.732 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.77% Epoch: [434] [ 0/312] eta: 0:44:34 lr: 0.000015 min_lr: 0.000015 loss: 1.8859 (1.8859) weight_decay: 0.0500 (0.0500) time: 8.5718 data: 7.8337 max mem: 64948 Epoch: [434] [ 10/312] eta: 0:07:17 lr: 0.000015 min_lr: 0.000015 loss: 1.7635 (1.6382) weight_decay: 0.0500 (0.0500) time: 1.4484 data: 0.7125 max mem: 64948 Epoch: [434] [ 20/312] eta: 0:05:17 lr: 0.000015 min_lr: 0.000015 loss: 1.7489 (1.6694) weight_decay: 0.0500 (0.0500) time: 0.7139 data: 0.0004 max mem: 64948 Epoch: [434] [ 30/312] eta: 0:04:30 lr: 0.000014 min_lr: 0.000014 loss: 1.8199 (1.7061) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0003 max mem: 64948 Epoch: [434] [ 40/312] eta: 0:04:03 lr: 0.000014 min_lr: 0.000014 loss: 1.7784 (1.6870) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [434] [ 50/312] eta: 0:03:44 lr: 0.000014 min_lr: 0.000014 loss: 1.7330 (1.6905) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [434] [ 60/312] eta: 0:03:29 lr: 0.000014 min_lr: 0.000014 loss: 1.7330 (1.6857) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [434] [ 70/312] eta: 0:03:16 lr: 0.000014 min_lr: 0.000014 loss: 1.4740 (1.6381) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [434] [ 80/312] eta: 0:03:04 lr: 0.000014 min_lr: 0.000014 loss: 1.3783 (1.6336) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [434] [ 90/312] eta: 0:02:54 lr: 0.000014 min_lr: 0.000014 loss: 1.4538 (1.6183) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [434] [100/312] eta: 0:02:44 lr: 0.000014 min_lr: 0.000014 loss: 1.4772 (1.6174) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [434] [110/312] eta: 0:02:35 lr: 0.000014 min_lr: 0.000014 loss: 1.7352 (1.6370) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [434] [120/312] eta: 0:02:26 lr: 0.000014 min_lr: 0.000014 loss: 1.7038 (1.6307) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0003 max mem: 64948 Epoch: [434] [130/312] eta: 0:02:18 lr: 0.000014 min_lr: 0.000014 loss: 1.6585 (1.6345) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0003 max mem: 64948 Epoch: [434] [140/312] eta: 0:02:09 lr: 0.000014 min_lr: 0.000014 loss: 1.6952 (1.6483) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [434] [150/312] eta: 0:02:01 lr: 0.000014 min_lr: 0.000014 loss: 1.8558 (1.6570) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [434] [160/312] eta: 0:01:53 lr: 0.000014 min_lr: 0.000014 loss: 1.7998 (1.6552) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [434] [170/312] eta: 0:01:45 lr: 0.000014 min_lr: 0.000014 loss: 1.5086 (1.6456) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [434] [180/312] eta: 0:01:37 lr: 0.000014 min_lr: 0.000014 loss: 1.5391 (1.6469) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [434] [190/312] eta: 0:01:30 lr: 0.000014 min_lr: 0.000014 loss: 1.7748 (1.6529) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [434] [200/312] eta: 0:01:22 lr: 0.000014 min_lr: 0.000014 loss: 1.7224 (1.6524) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [434] [210/312] eta: 0:01:14 lr: 0.000014 min_lr: 0.000014 loss: 1.6594 (1.6505) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [434] [220/312] eta: 0:01:07 lr: 0.000013 min_lr: 0.000013 loss: 1.7229 (1.6531) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [434] [230/312] eta: 0:00:59 lr: 0.000013 min_lr: 0.000013 loss: 1.7982 (1.6580) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [434] [240/312] eta: 0:00:52 lr: 0.000013 min_lr: 0.000013 loss: 1.8127 (1.6607) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [434] [250/312] eta: 0:00:45 lr: 0.000013 min_lr: 0.000013 loss: 1.7817 (1.6603) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [434] [260/312] eta: 0:00:37 lr: 0.000013 min_lr: 0.000013 loss: 1.6933 (1.6602) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [434] [270/312] eta: 0:00:30 lr: 0.000013 min_lr: 0.000013 loss: 1.8778 (1.6682) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [434] [280/312] eta: 0:00:23 lr: 0.000013 min_lr: 0.000013 loss: 1.8234 (1.6685) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0009 max mem: 64948 Epoch: [434] [290/312] eta: 0:00:15 lr: 0.000013 min_lr: 0.000013 loss: 1.5870 (1.6634) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0008 max mem: 64948 Epoch: [434] [300/312] eta: 0:00:08 lr: 0.000013 min_lr: 0.000013 loss: 1.6467 (1.6646) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [434] [310/312] eta: 0:00:01 lr: 0.000013 min_lr: 0.000013 loss: 1.7626 (1.6687) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [434] [311/312] eta: 0:00:00 lr: 0.000013 min_lr: 0.000013 loss: 1.7547 (1.6686) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [434] Total time: 0:03:45 (0.7240 s / it) Averaged stats: lr: 0.000013 min_lr: 0.000013 loss: 1.7547 (1.6482) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:47 loss: 0.4354 (0.4354) acc1: 89.0625 (89.0625) acc5: 97.9167 (97.9167) time: 5.2394 data: 5.0199 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6156 (0.6341) acc1: 84.3750 (83.1360) acc5: 97.1354 (96.7040) time: 0.7341 data: 0.5578 max mem: 64948 Test: Total time: 0:00:06 (0.7590 s / it) * Acc@1 83.756 Acc@5 96.684 loss 0.619 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:45 loss: 0.4458 (0.4458) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 5.0770 data: 4.8708 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6092 (0.6361) acc1: 84.8958 (82.9760) acc5: 97.3958 (96.7680) time: 0.7154 data: 0.5413 max mem: 64948 Test: Total time: 0:00:06 (0.7257 s / it) * Acc@1 83.762 Acc@5 96.726 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [435] [ 0/312] eta: 0:49:25 lr: 0.000013 min_lr: 0.000013 loss: 1.9166 (1.9166) weight_decay: 0.0500 (0.0500) time: 9.5038 data: 6.0884 max mem: 64948 Epoch: [435] [ 10/312] eta: 0:07:49 lr: 0.000013 min_lr: 0.000013 loss: 1.7795 (1.6929) weight_decay: 0.0500 (0.0500) time: 1.5540 data: 0.5540 max mem: 64948 Epoch: [435] [ 20/312] eta: 0:05:34 lr: 0.000013 min_lr: 0.000013 loss: 1.7126 (1.6378) weight_decay: 0.0500 (0.0500) time: 0.7275 data: 0.0004 max mem: 64948 Epoch: [435] [ 30/312] eta: 0:04:42 lr: 0.000013 min_lr: 0.000013 loss: 1.5443 (1.6272) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [435] [ 40/312] eta: 0:04:11 lr: 0.000013 min_lr: 0.000013 loss: 1.7881 (1.6728) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0003 max mem: 64948 Epoch: [435] [ 50/312] eta: 0:03:51 lr: 0.000013 min_lr: 0.000013 loss: 1.7881 (1.6847) weight_decay: 0.0500 (0.0500) time: 0.6995 data: 0.0003 max mem: 64948 Epoch: [435] [ 60/312] eta: 0:03:34 lr: 0.000013 min_lr: 0.000013 loss: 1.7777 (1.6854) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [435] [ 70/312] eta: 0:03:20 lr: 0.000013 min_lr: 0.000013 loss: 1.6304 (1.6573) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [435] [ 80/312] eta: 0:03:08 lr: 0.000013 min_lr: 0.000013 loss: 1.4641 (1.6465) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [435] [ 90/312] eta: 0:02:57 lr: 0.000013 min_lr: 0.000013 loss: 1.6493 (1.6444) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [435] [100/312] eta: 0:02:47 lr: 0.000012 min_lr: 0.000012 loss: 1.7065 (1.6440) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [435] [110/312] eta: 0:02:37 lr: 0.000012 min_lr: 0.000012 loss: 1.6646 (1.6421) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [435] [120/312] eta: 0:02:28 lr: 0.000012 min_lr: 0.000012 loss: 1.6759 (1.6510) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [435] [130/312] eta: 0:02:19 lr: 0.000012 min_lr: 0.000012 loss: 1.7183 (1.6481) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [435] [140/312] eta: 0:02:11 lr: 0.000012 min_lr: 0.000012 loss: 1.6165 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [435] [150/312] eta: 0:02:02 lr: 0.000012 min_lr: 0.000012 loss: 1.6165 (1.6420) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [435] [160/312] eta: 0:01:54 lr: 0.000012 min_lr: 0.000012 loss: 1.5777 (1.6378) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [435] [170/312] eta: 0:01:46 lr: 0.000012 min_lr: 0.000012 loss: 1.5660 (1.6318) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [435] [180/312] eta: 0:01:38 lr: 0.000012 min_lr: 0.000012 loss: 1.6269 (1.6323) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [435] [190/312] eta: 0:01:30 lr: 0.000012 min_lr: 0.000012 loss: 1.7662 (1.6382) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [435] [200/312] eta: 0:01:23 lr: 0.000012 min_lr: 0.000012 loss: 1.7662 (1.6402) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [435] [210/312] eta: 0:01:15 lr: 0.000012 min_lr: 0.000012 loss: 1.7898 (1.6460) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [435] [220/312] eta: 0:01:07 lr: 0.000012 min_lr: 0.000012 loss: 1.7707 (1.6459) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [435] [230/312] eta: 0:01:00 lr: 0.000012 min_lr: 0.000012 loss: 1.7707 (1.6525) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [435] [240/312] eta: 0:00:52 lr: 0.000012 min_lr: 0.000012 loss: 1.7876 (1.6540) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [435] [250/312] eta: 0:00:45 lr: 0.000012 min_lr: 0.000012 loss: 1.7409 (1.6580) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [435] [260/312] eta: 0:00:37 lr: 0.000012 min_lr: 0.000012 loss: 1.7195 (1.6591) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [435] [270/312] eta: 0:00:30 lr: 0.000012 min_lr: 0.000012 loss: 1.7033 (1.6589) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0003 max mem: 64948 Epoch: [435] [280/312] eta: 0:00:23 lr: 0.000012 min_lr: 0.000012 loss: 1.6891 (1.6560) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0009 max mem: 64948 Epoch: [435] [290/312] eta: 0:00:15 lr: 0.000012 min_lr: 0.000012 loss: 1.5560 (1.6513) weight_decay: 0.0500 (0.0500) time: 0.6924 data: 0.0008 max mem: 64948 Epoch: [435] [300/312] eta: 0:00:08 lr: 0.000012 min_lr: 0.000012 loss: 1.6570 (1.6550) weight_decay: 0.0500 (0.0500) time: 0.6901 data: 0.0001 max mem: 64948 Epoch: [435] [310/312] eta: 0:00:01 lr: 0.000011 min_lr: 0.000011 loss: 1.6570 (1.6518) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [435] [311/312] eta: 0:00:00 lr: 0.000011 min_lr: 0.000011 loss: 1.6583 (1.6524) weight_decay: 0.0500 (0.0500) time: 0.6913 data: 0.0001 max mem: 64948 Epoch: [435] Total time: 0:03:46 (0.7273 s / it) Averaged stats: lr: 0.000011 min_lr: 0.000011 loss: 1.6583 (1.6502) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4336 (0.4336) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 4.8172 data: 4.6061 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6130 (0.6345) acc1: 84.3750 (83.0720) acc5: 97.1354 (96.7680) time: 0.6865 data: 0.5119 max mem: 64948 Test: Total time: 0:00:06 (0.7144 s / it) * Acc@1 83.806 Acc@5 96.718 loss 0.619 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:47 loss: 0.4454 (0.4454) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 5.2910 data: 5.0730 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6094 (0.6359) acc1: 84.8958 (82.9760) acc5: 97.3958 (96.7680) time: 0.7393 data: 0.5638 max mem: 64948 Test: Total time: 0:00:06 (0.7529 s / it) * Acc@1 83.780 Acc@5 96.726 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.78% Epoch: [436] [ 0/312] eta: 0:45:54 lr: 0.000011 min_lr: 0.000011 loss: 1.5870 (1.5870) weight_decay: 0.0500 (0.0500) time: 8.8296 data: 7.4308 max mem: 64948 Epoch: [436] [ 10/312] eta: 0:07:27 lr: 0.000011 min_lr: 0.000011 loss: 1.5870 (1.5658) weight_decay: 0.0500 (0.0500) time: 1.4819 data: 0.6759 max mem: 64948 Epoch: [436] [ 20/312] eta: 0:05:23 lr: 0.000011 min_lr: 0.000011 loss: 1.6860 (1.6381) weight_decay: 0.0500 (0.0500) time: 0.7210 data: 0.0004 max mem: 64948 Epoch: [436] [ 30/312] eta: 0:04:34 lr: 0.000011 min_lr: 0.000011 loss: 1.7359 (1.6651) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [436] [ 40/312] eta: 0:04:06 lr: 0.000011 min_lr: 0.000011 loss: 1.7208 (1.6530) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0003 max mem: 64948 Epoch: [436] [ 50/312] eta: 0:03:46 lr: 0.000011 min_lr: 0.000011 loss: 1.6962 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [436] [ 60/312] eta: 0:03:31 lr: 0.000011 min_lr: 0.000011 loss: 1.7114 (1.6574) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [436] [ 70/312] eta: 0:03:17 lr: 0.000011 min_lr: 0.000011 loss: 1.7730 (1.6867) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [436] [ 80/312] eta: 0:03:06 lr: 0.000011 min_lr: 0.000011 loss: 1.8299 (1.6902) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [436] [ 90/312] eta: 0:02:55 lr: 0.000011 min_lr: 0.000011 loss: 1.7781 (1.6873) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [436] [100/312] eta: 0:02:45 lr: 0.000011 min_lr: 0.000011 loss: 1.7087 (1.6845) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [436] [110/312] eta: 0:02:36 lr: 0.000011 min_lr: 0.000011 loss: 1.5842 (1.6635) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [436] [120/312] eta: 0:02:27 lr: 0.000011 min_lr: 0.000011 loss: 1.5551 (1.6593) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [436] [130/312] eta: 0:02:18 lr: 0.000011 min_lr: 0.000011 loss: 1.7556 (1.6624) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [436] [140/312] eta: 0:02:10 lr: 0.000011 min_lr: 0.000011 loss: 1.5586 (1.6502) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [436] [150/312] eta: 0:02:01 lr: 0.000011 min_lr: 0.000011 loss: 1.6439 (1.6542) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [436] [160/312] eta: 0:01:53 lr: 0.000011 min_lr: 0.000011 loss: 1.7701 (1.6535) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [436] [170/312] eta: 0:01:45 lr: 0.000011 min_lr: 0.000011 loss: 1.8149 (1.6645) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [436] [180/312] eta: 0:01:38 lr: 0.000011 min_lr: 0.000011 loss: 1.7697 (1.6590) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [436] [190/312] eta: 0:01:30 lr: 0.000011 min_lr: 0.000011 loss: 1.5802 (1.6583) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0003 max mem: 64948 Epoch: [436] [200/312] eta: 0:01:22 lr: 0.000011 min_lr: 0.000011 loss: 1.7351 (1.6606) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [436] [210/312] eta: 0:01:15 lr: 0.000010 min_lr: 0.000010 loss: 1.6247 (1.6474) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [436] [220/312] eta: 0:01:07 lr: 0.000010 min_lr: 0.000010 loss: 1.6971 (1.6517) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0003 max mem: 64948 Epoch: [436] [230/312] eta: 0:01:00 lr: 0.000010 min_lr: 0.000010 loss: 1.7224 (1.6550) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0003 max mem: 64948 Epoch: [436] [240/312] eta: 0:00:52 lr: 0.000010 min_lr: 0.000010 loss: 1.7906 (1.6653) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [436] [250/312] eta: 0:00:45 lr: 0.000010 min_lr: 0.000010 loss: 1.8356 (1.6650) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [436] [260/312] eta: 0:00:37 lr: 0.000010 min_lr: 0.000010 loss: 1.8166 (1.6713) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [436] [270/312] eta: 0:00:30 lr: 0.000010 min_lr: 0.000010 loss: 1.8439 (1.6708) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [436] [280/312] eta: 0:00:23 lr: 0.000010 min_lr: 0.000010 loss: 1.6049 (1.6660) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0009 max mem: 64948 Epoch: [436] [290/312] eta: 0:00:15 lr: 0.000010 min_lr: 0.000010 loss: 1.5786 (1.6665) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0008 max mem: 64948 Epoch: [436] [300/312] eta: 0:00:08 lr: 0.000010 min_lr: 0.000010 loss: 1.7456 (1.6702) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [436] [310/312] eta: 0:00:01 lr: 0.000010 min_lr: 0.000010 loss: 1.7795 (1.6719) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [436] [311/312] eta: 0:00:00 lr: 0.000010 min_lr: 0.000010 loss: 1.7795 (1.6713) weight_decay: 0.0500 (0.0500) time: 0.6911 data: 0.0001 max mem: 64948 Epoch: [436] Total time: 0:03:46 (0.7249 s / it) Averaged stats: lr: 0.000010 min_lr: 0.000010 loss: 1.7795 (1.6487) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:47 loss: 0.4386 (0.4386) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 5.2965 data: 5.0796 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6169 (0.6361) acc1: 83.8542 (82.9760) acc5: 97.3958 (96.8640) time: 0.7397 data: 0.5645 max mem: 64948 Test: Total time: 0:00:06 (0.7630 s / it) * Acc@1 83.784 Acc@5 96.716 loss 0.619 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:47 loss: 0.4452 (0.4452) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 5.3225 data: 5.1046 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6092 (0.6358) acc1: 84.8958 (83.0080) acc5: 97.3958 (96.7680) time: 0.7429 data: 0.5673 max mem: 64948 Test: Total time: 0:00:06 (0.7549 s / it) * Acc@1 83.790 Acc@5 96.724 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.79% Epoch: [437] [ 0/312] eta: 0:45:59 lr: 0.000010 min_lr: 0.000010 loss: 1.0658 (1.0658) weight_decay: 0.0500 (0.0500) time: 8.8434 data: 8.0980 max mem: 64948 Epoch: [437] [ 10/312] eta: 0:07:21 lr: 0.000010 min_lr: 0.000010 loss: 1.7587 (1.6218) weight_decay: 0.0500 (0.0500) time: 1.4633 data: 0.7366 max mem: 64948 Epoch: [437] [ 20/312] eta: 0:05:20 lr: 0.000010 min_lr: 0.000010 loss: 1.6443 (1.5837) weight_decay: 0.0500 (0.0500) time: 0.7087 data: 0.0004 max mem: 64948 Epoch: [437] [ 30/312] eta: 0:04:32 lr: 0.000010 min_lr: 0.000010 loss: 1.6015 (1.6251) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [437] [ 40/312] eta: 0:04:04 lr: 0.000010 min_lr: 0.000010 loss: 1.7640 (1.6268) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0003 max mem: 64948 Epoch: [437] [ 50/312] eta: 0:03:45 lr: 0.000010 min_lr: 0.000010 loss: 1.7350 (1.6215) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [437] [ 60/312] eta: 0:03:29 lr: 0.000010 min_lr: 0.000010 loss: 1.7362 (1.6275) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [437] [ 70/312] eta: 0:03:16 lr: 0.000010 min_lr: 0.000010 loss: 1.7829 (1.6424) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [437] [ 80/312] eta: 0:03:05 lr: 0.000010 min_lr: 0.000010 loss: 1.7893 (1.6504) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [437] [ 90/312] eta: 0:02:54 lr: 0.000010 min_lr: 0.000010 loss: 1.6255 (1.6421) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [437] [100/312] eta: 0:02:44 lr: 0.000010 min_lr: 0.000010 loss: 1.6473 (1.6443) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [437] [110/312] eta: 0:02:35 lr: 0.000010 min_lr: 0.000010 loss: 1.5559 (1.6269) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [437] [120/312] eta: 0:02:26 lr: 0.000009 min_lr: 0.000009 loss: 1.5903 (1.6376) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [437] [130/312] eta: 0:02:18 lr: 0.000009 min_lr: 0.000009 loss: 1.7283 (1.6340) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [437] [140/312] eta: 0:02:09 lr: 0.000009 min_lr: 0.000009 loss: 1.6978 (1.6345) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [437] [150/312] eta: 0:02:01 lr: 0.000009 min_lr: 0.000009 loss: 1.7720 (1.6313) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [437] [160/312] eta: 0:01:53 lr: 0.000009 min_lr: 0.000009 loss: 1.5417 (1.6139) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [437] [170/312] eta: 0:01:45 lr: 0.000009 min_lr: 0.000009 loss: 1.4313 (1.6090) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [437] [180/312] eta: 0:01:37 lr: 0.000009 min_lr: 0.000009 loss: 1.6627 (1.6094) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [437] [190/312] eta: 0:01:30 lr: 0.000009 min_lr: 0.000009 loss: 1.6667 (1.6100) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [437] [200/312] eta: 0:01:22 lr: 0.000009 min_lr: 0.000009 loss: 1.5737 (1.6076) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [437] [210/312] eta: 0:01:14 lr: 0.000009 min_lr: 0.000009 loss: 1.5737 (1.6074) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0003 max mem: 64948 Epoch: [437] [220/312] eta: 0:01:07 lr: 0.000009 min_lr: 0.000009 loss: 1.7601 (1.6179) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [437] [230/312] eta: 0:00:59 lr: 0.000009 min_lr: 0.000009 loss: 1.8207 (1.6222) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [437] [240/312] eta: 0:00:52 lr: 0.000009 min_lr: 0.000009 loss: 1.8189 (1.6313) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [437] [250/312] eta: 0:00:45 lr: 0.000009 min_lr: 0.000009 loss: 1.7878 (1.6302) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [437] [260/312] eta: 0:00:37 lr: 0.000009 min_lr: 0.000009 loss: 1.6598 (1.6351) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0003 max mem: 64948 Epoch: [437] [270/312] eta: 0:00:30 lr: 0.000009 min_lr: 0.000009 loss: 1.6179 (1.6333) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [437] [280/312] eta: 0:00:23 lr: 0.000009 min_lr: 0.000009 loss: 1.5379 (1.6318) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [437] [290/312] eta: 0:00:15 lr: 0.000009 min_lr: 0.000009 loss: 1.5442 (1.6300) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [437] [300/312] eta: 0:00:08 lr: 0.000009 min_lr: 0.000009 loss: 1.6121 (1.6337) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [437] [310/312] eta: 0:00:01 lr: 0.000009 min_lr: 0.000009 loss: 1.6784 (1.6346) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [437] [311/312] eta: 0:00:00 lr: 0.000009 min_lr: 0.000009 loss: 1.6926 (1.6348) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [437] Total time: 0:03:45 (0.7243 s / it) Averaged stats: lr: 0.000009 min_lr: 0.000009 loss: 1.6926 (1.6467) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4385 (0.4385) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 4.7125 data: 4.5081 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6087 (0.6322) acc1: 84.1146 (83.0720) acc5: 97.1354 (96.8000) time: 0.6749 data: 0.5010 max mem: 64948 Test: Total time: 0:00:06 (0.6996 s / it) * Acc@1 83.824 Acc@5 96.738 loss 0.618 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:45 loss: 0.4449 (0.4449) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 5.0215 data: 4.8034 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6093 (0.6357) acc1: 84.8958 (83.0400) acc5: 97.3958 (96.7680) time: 0.7093 data: 0.5338 max mem: 64948 Test: Total time: 0:00:06 (0.7167 s / it) * Acc@1 83.788 Acc@5 96.722 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [438] [ 0/312] eta: 0:52:36 lr: 0.000009 min_lr: 0.000009 loss: 1.1600 (1.1600) weight_decay: 0.0500 (0.0500) time: 10.1160 data: 6.9119 max mem: 64948 Epoch: [438] [ 10/312] eta: 0:08:15 lr: 0.000009 min_lr: 0.000009 loss: 1.4109 (1.4399) weight_decay: 0.0500 (0.0500) time: 1.6396 data: 0.6288 max mem: 64948 Epoch: [438] [ 20/312] eta: 0:05:48 lr: 0.000009 min_lr: 0.000009 loss: 1.6286 (1.5002) weight_decay: 0.0500 (0.0500) time: 0.7480 data: 0.0004 max mem: 64948 Epoch: [438] [ 30/312] eta: 0:04:51 lr: 0.000009 min_lr: 0.000009 loss: 1.6698 (1.5265) weight_decay: 0.0500 (0.0500) time: 0.7000 data: 0.0004 max mem: 64948 Epoch: [438] [ 40/312] eta: 0:04:18 lr: 0.000009 min_lr: 0.000009 loss: 1.6748 (1.5689) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [438] [ 50/312] eta: 0:03:56 lr: 0.000008 min_lr: 0.000008 loss: 1.6689 (1.5723) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [438] [ 60/312] eta: 0:03:38 lr: 0.000008 min_lr: 0.000008 loss: 1.6291 (1.5863) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [438] [ 70/312] eta: 0:03:24 lr: 0.000008 min_lr: 0.000008 loss: 1.7739 (1.6318) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [438] [ 80/312] eta: 0:03:11 lr: 0.000008 min_lr: 0.000008 loss: 1.7410 (1.6202) weight_decay: 0.0500 (0.0500) time: 0.6981 data: 0.0004 max mem: 64948 Epoch: [438] [ 90/312] eta: 0:03:00 lr: 0.000008 min_lr: 0.000008 loss: 1.5189 (1.6163) weight_decay: 0.0500 (0.0500) time: 0.6979 data: 0.0004 max mem: 64948 Epoch: [438] [100/312] eta: 0:02:49 lr: 0.000008 min_lr: 0.000008 loss: 1.5724 (1.6243) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [438] [110/312] eta: 0:02:39 lr: 0.000008 min_lr: 0.000008 loss: 1.7606 (1.6286) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [438] [120/312] eta: 0:02:30 lr: 0.000008 min_lr: 0.000008 loss: 1.7652 (1.6351) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [438] [130/312] eta: 0:02:21 lr: 0.000008 min_lr: 0.000008 loss: 1.7552 (1.6321) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [438] [140/312] eta: 0:02:12 lr: 0.000008 min_lr: 0.000008 loss: 1.7038 (1.6303) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [438] [150/312] eta: 0:02:03 lr: 0.000008 min_lr: 0.000008 loss: 1.7146 (1.6369) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [438] [160/312] eta: 0:01:55 lr: 0.000008 min_lr: 0.000008 loss: 1.6942 (1.6230) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [438] [170/312] eta: 0:01:47 lr: 0.000008 min_lr: 0.000008 loss: 1.6806 (1.6332) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [438] [180/312] eta: 0:01:39 lr: 0.000008 min_lr: 0.000008 loss: 1.6806 (1.6330) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [438] [190/312] eta: 0:01:31 lr: 0.000008 min_lr: 0.000008 loss: 1.5664 (1.6346) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [438] [200/312] eta: 0:01:23 lr: 0.000008 min_lr: 0.000008 loss: 1.6484 (1.6343) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [438] [210/312] eta: 0:01:16 lr: 0.000008 min_lr: 0.000008 loss: 1.6671 (1.6400) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [438] [220/312] eta: 0:01:08 lr: 0.000008 min_lr: 0.000008 loss: 1.8594 (1.6478) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [438] [230/312] eta: 0:01:00 lr: 0.000008 min_lr: 0.000008 loss: 1.6966 (1.6405) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [438] [240/312] eta: 0:00:53 lr: 0.000008 min_lr: 0.000008 loss: 1.5756 (1.6402) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [438] [250/312] eta: 0:00:45 lr: 0.000008 min_lr: 0.000008 loss: 1.5756 (1.6379) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [438] [260/312] eta: 0:00:38 lr: 0.000008 min_lr: 0.000008 loss: 1.5536 (1.6381) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [438] [270/312] eta: 0:00:30 lr: 0.000008 min_lr: 0.000008 loss: 1.4519 (1.6368) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [438] [280/312] eta: 0:00:23 lr: 0.000008 min_lr: 0.000008 loss: 1.6728 (1.6404) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0006 max mem: 64948 Epoch: [438] [290/312] eta: 0:00:16 lr: 0.000008 min_lr: 0.000008 loss: 1.7601 (1.6452) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0005 max mem: 64948 Epoch: [438] [300/312] eta: 0:00:08 lr: 0.000007 min_lr: 0.000007 loss: 1.6234 (1.6387) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [438] [310/312] eta: 0:00:01 lr: 0.000007 min_lr: 0.000007 loss: 1.6122 (1.6425) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [438] [311/312] eta: 0:00:00 lr: 0.000007 min_lr: 0.000007 loss: 1.6185 (1.6430) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [438] Total time: 0:03:48 (0.7316 s / it) Averaged stats: lr: 0.000007 min_lr: 0.000007 loss: 1.6185 (1.6502) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4393 (0.4393) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 4.6166 data: 4.4029 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6093 (0.6349) acc1: 84.3750 (83.1360) acc5: 97.3958 (96.8000) time: 0.6643 data: 0.4893 max mem: 64948 Test: Total time: 0:00:06 (0.6879 s / it) * Acc@1 83.790 Acc@5 96.730 loss 0.618 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:47 loss: 0.4447 (0.4447) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 5.3162 data: 5.1018 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6093 (0.6356) acc1: 84.8958 (83.0080) acc5: 97.3958 (96.7680) time: 0.7421 data: 0.5670 max mem: 64948 Test: Total time: 0:00:06 (0.7494 s / it) * Acc@1 83.776 Acc@5 96.730 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [439] [ 0/312] eta: 0:58:24 lr: 0.000007 min_lr: 0.000007 loss: 1.7834 (1.7834) weight_decay: 0.0500 (0.0500) time: 11.2327 data: 6.5354 max mem: 64948 Epoch: [439] [ 10/312] eta: 0:08:23 lr: 0.000007 min_lr: 0.000007 loss: 1.6801 (1.6774) weight_decay: 0.0500 (0.0500) time: 1.6683 data: 0.5945 max mem: 64948 Epoch: [439] [ 20/312] eta: 0:05:51 lr: 0.000007 min_lr: 0.000007 loss: 1.6560 (1.6678) weight_decay: 0.0500 (0.0500) time: 0.7022 data: 0.0004 max mem: 64948 Epoch: [439] [ 30/312] eta: 0:04:52 lr: 0.000007 min_lr: 0.000007 loss: 1.7289 (1.6800) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0003 max mem: 64948 Epoch: [439] [ 40/312] eta: 0:04:19 lr: 0.000007 min_lr: 0.000007 loss: 1.7560 (1.6750) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [439] [ 50/312] eta: 0:03:56 lr: 0.000007 min_lr: 0.000007 loss: 1.5131 (1.6459) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [439] [ 60/312] eta: 0:03:39 lr: 0.000007 min_lr: 0.000007 loss: 1.7761 (1.6844) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [439] [ 70/312] eta: 0:03:24 lr: 0.000007 min_lr: 0.000007 loss: 1.7760 (1.6781) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [439] [ 80/312] eta: 0:03:11 lr: 0.000007 min_lr: 0.000007 loss: 1.6819 (1.6846) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [439] [ 90/312] eta: 0:03:00 lr: 0.000007 min_lr: 0.000007 loss: 1.7214 (1.6861) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [439] [100/312] eta: 0:02:49 lr: 0.000007 min_lr: 0.000007 loss: 1.6550 (1.6812) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [439] [110/312] eta: 0:02:39 lr: 0.000007 min_lr: 0.000007 loss: 1.6550 (1.6719) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [439] [120/312] eta: 0:02:30 lr: 0.000007 min_lr: 0.000007 loss: 1.7001 (1.6732) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [439] [130/312] eta: 0:02:21 lr: 0.000007 min_lr: 0.000007 loss: 1.7001 (1.6680) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [439] [140/312] eta: 0:02:12 lr: 0.000007 min_lr: 0.000007 loss: 1.7777 (1.6749) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [439] [150/312] eta: 0:02:04 lr: 0.000007 min_lr: 0.000007 loss: 1.7941 (1.6703) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [439] [160/312] eta: 0:01:55 lr: 0.000007 min_lr: 0.000007 loss: 1.6777 (1.6721) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [439] [170/312] eta: 0:01:47 lr: 0.000007 min_lr: 0.000007 loss: 1.6225 (1.6647) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [439] [180/312] eta: 0:01:39 lr: 0.000007 min_lr: 0.000007 loss: 1.5652 (1.6643) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [439] [190/312] eta: 0:01:31 lr: 0.000007 min_lr: 0.000007 loss: 1.7510 (1.6623) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [439] [200/312] eta: 0:01:23 lr: 0.000007 min_lr: 0.000007 loss: 1.8108 (1.6714) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [439] [210/312] eta: 0:01:16 lr: 0.000007 min_lr: 0.000007 loss: 1.8131 (1.6676) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [439] [220/312] eta: 0:01:08 lr: 0.000007 min_lr: 0.000007 loss: 1.5750 (1.6635) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [439] [230/312] eta: 0:01:00 lr: 0.000007 min_lr: 0.000007 loss: 1.6043 (1.6585) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [439] [240/312] eta: 0:00:53 lr: 0.000007 min_lr: 0.000007 loss: 1.5212 (1.6547) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [439] [250/312] eta: 0:00:45 lr: 0.000007 min_lr: 0.000007 loss: 1.7098 (1.6559) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [439] [260/312] eta: 0:00:38 lr: 0.000007 min_lr: 0.000007 loss: 1.7559 (1.6605) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [439] [270/312] eta: 0:00:30 lr: 0.000006 min_lr: 0.000006 loss: 1.6367 (1.6528) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [439] [280/312] eta: 0:00:23 lr: 0.000006 min_lr: 0.000006 loss: 1.6249 (1.6580) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0016 max mem: 64948 Epoch: [439] [290/312] eta: 0:00:16 lr: 0.000006 min_lr: 0.000006 loss: 1.6814 (1.6542) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0015 max mem: 64948 Epoch: [439] [300/312] eta: 0:00:08 lr: 0.000006 min_lr: 0.000006 loss: 1.6949 (1.6550) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [439] [310/312] eta: 0:00:01 lr: 0.000006 min_lr: 0.000006 loss: 1.7659 (1.6587) weight_decay: 0.0500 (0.0500) time: 0.6910 data: 0.0001 max mem: 64948 Epoch: [439] [311/312] eta: 0:00:00 lr: 0.000006 min_lr: 0.000006 loss: 1.7659 (1.6585) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [439] Total time: 0:03:48 (0.7315 s / it) Averaged stats: lr: 0.000006 min_lr: 0.000006 loss: 1.7659 (1.6465) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4423 (0.4423) acc1: 89.3229 (89.3229) acc5: 97.9167 (97.9167) time: 4.4613 data: 4.2448 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6158 (0.6363) acc1: 84.6354 (83.0080) acc5: 97.1354 (96.8000) time: 0.6475 data: 0.4717 max mem: 64948 Test: Total time: 0:00:06 (0.6695 s / it) * Acc@1 83.820 Acc@5 96.706 loss 0.619 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:43 loss: 0.4444 (0.4444) acc1: 88.5417 (88.5417) acc5: 97.9167 (97.9167) time: 4.8550 data: 4.6395 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6093 (0.6354) acc1: 84.8958 (83.0400) acc5: 97.3958 (96.8000) time: 0.6908 data: 0.5156 max mem: 64948 Test: Total time: 0:00:06 (0.6981 s / it) * Acc@1 83.784 Acc@5 96.734 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [440] [ 0/312] eta: 0:54:04 lr: 0.000006 min_lr: 0.000006 loss: 1.8061 (1.8061) weight_decay: 0.0500 (0.0500) time: 10.3989 data: 9.0664 max mem: 64948 Epoch: [440] [ 10/312] eta: 0:08:05 lr: 0.000006 min_lr: 0.000006 loss: 1.8061 (1.7388) weight_decay: 0.0500 (0.0500) time: 1.6087 data: 0.8246 max mem: 64948 Epoch: [440] [ 20/312] eta: 0:05:42 lr: 0.000006 min_lr: 0.000006 loss: 1.7994 (1.7488) weight_decay: 0.0500 (0.0500) time: 0.7125 data: 0.0004 max mem: 64948 Epoch: [440] [ 30/312] eta: 0:04:47 lr: 0.000006 min_lr: 0.000006 loss: 1.6362 (1.6345) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [440] [ 40/312] eta: 0:04:15 lr: 0.000006 min_lr: 0.000006 loss: 1.5254 (1.6423) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [440] [ 50/312] eta: 0:03:53 lr: 0.000006 min_lr: 0.000006 loss: 1.7885 (1.6401) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [440] [ 60/312] eta: 0:03:36 lr: 0.000006 min_lr: 0.000006 loss: 1.7362 (1.6483) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [440] [ 70/312] eta: 0:03:22 lr: 0.000006 min_lr: 0.000006 loss: 1.7362 (1.6571) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [440] [ 80/312] eta: 0:03:10 lr: 0.000006 min_lr: 0.000006 loss: 1.6803 (1.6514) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [440] [ 90/312] eta: 0:02:58 lr: 0.000006 min_lr: 0.000006 loss: 1.6126 (1.6402) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [440] [100/312] eta: 0:02:48 lr: 0.000006 min_lr: 0.000006 loss: 1.6126 (1.6258) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0003 max mem: 64948 Epoch: [440] [110/312] eta: 0:02:38 lr: 0.000006 min_lr: 0.000006 loss: 1.7136 (1.6184) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [440] [120/312] eta: 0:02:29 lr: 0.000006 min_lr: 0.000006 loss: 1.7301 (1.6245) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [440] [130/312] eta: 0:02:20 lr: 0.000006 min_lr: 0.000006 loss: 1.6316 (1.6200) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [440] [140/312] eta: 0:02:11 lr: 0.000006 min_lr: 0.000006 loss: 1.5731 (1.6209) weight_decay: 0.0500 (0.0500) time: 0.7002 data: 0.0004 max mem: 64948 Epoch: [440] [150/312] eta: 0:02:03 lr: 0.000006 min_lr: 0.000006 loss: 1.6982 (1.6288) weight_decay: 0.0500 (0.0500) time: 0.6974 data: 0.0004 max mem: 64948 Epoch: [440] [160/312] eta: 0:01:55 lr: 0.000006 min_lr: 0.000006 loss: 1.6982 (1.6313) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [440] [170/312] eta: 0:01:47 lr: 0.000006 min_lr: 0.000006 loss: 1.6640 (1.6355) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [440] [180/312] eta: 0:01:39 lr: 0.000006 min_lr: 0.000006 loss: 1.7638 (1.6387) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [440] [190/312] eta: 0:01:31 lr: 0.000006 min_lr: 0.000006 loss: 1.7958 (1.6400) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [440] [200/312] eta: 0:01:23 lr: 0.000006 min_lr: 0.000006 loss: 1.7413 (1.6390) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [440] [210/312] eta: 0:01:15 lr: 0.000006 min_lr: 0.000006 loss: 1.7475 (1.6414) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0003 max mem: 64948 Epoch: [440] [220/312] eta: 0:01:08 lr: 0.000006 min_lr: 0.000006 loss: 1.6778 (1.6401) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [440] [230/312] eta: 0:01:00 lr: 0.000006 min_lr: 0.000006 loss: 1.6550 (1.6401) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [440] [240/312] eta: 0:00:53 lr: 0.000006 min_lr: 0.000006 loss: 1.7529 (1.6426) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [440] [250/312] eta: 0:00:45 lr: 0.000006 min_lr: 0.000006 loss: 1.6911 (1.6378) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [440] [260/312] eta: 0:00:38 lr: 0.000005 min_lr: 0.000005 loss: 1.6322 (1.6377) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0004 max mem: 64948 Epoch: [440] [270/312] eta: 0:00:30 lr: 0.000005 min_lr: 0.000005 loss: 1.6322 (1.6333) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [440] [280/312] eta: 0:00:23 lr: 0.000005 min_lr: 0.000005 loss: 1.6411 (1.6313) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0010 max mem: 64948 Epoch: [440] [290/312] eta: 0:00:16 lr: 0.000005 min_lr: 0.000005 loss: 1.6666 (1.6323) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0009 max mem: 64948 Epoch: [440] [300/312] eta: 0:00:08 lr: 0.000005 min_lr: 0.000005 loss: 1.7224 (1.6341) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [440] [310/312] eta: 0:00:01 lr: 0.000005 min_lr: 0.000005 loss: 1.5370 (1.6307) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [440] [311/312] eta: 0:00:00 lr: 0.000005 min_lr: 0.000005 loss: 1.6048 (1.6313) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [440] Total time: 0:03:47 (0.7296 s / it) Averaged stats: lr: 0.000005 min_lr: 0.000005 loss: 1.6048 (1.6498) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4394 (0.4394) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 4.5101 data: 4.2905 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6131 (0.6356) acc1: 84.1146 (82.9760) acc5: 97.1354 (96.8640) time: 0.6524 data: 0.4768 max mem: 64948 Test: Total time: 0:00:06 (0.6814 s / it) * Acc@1 83.806 Acc@5 96.760 loss 0.620 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:43 loss: 0.4443 (0.4443) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.8271 data: 4.6213 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6094 (0.6353) acc1: 84.6354 (83.0400) acc5: 97.3958 (96.8000) time: 0.6949 data: 0.5209 max mem: 64948 Test: Total time: 0:00:06 (0.7035 s / it) * Acc@1 83.790 Acc@5 96.734 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [441] [ 0/312] eta: 0:54:48 lr: 0.000005 min_lr: 0.000005 loss: 1.4773 (1.4773) weight_decay: 0.0500 (0.0500) time: 10.5392 data: 7.2413 max mem: 64948 Epoch: [441] [ 10/312] eta: 0:08:10 lr: 0.000005 min_lr: 0.000005 loss: 1.8258 (1.7034) weight_decay: 0.0500 (0.0500) time: 1.6225 data: 0.6588 max mem: 64948 Epoch: [441] [ 20/312] eta: 0:05:44 lr: 0.000005 min_lr: 0.000005 loss: 1.8182 (1.6903) weight_decay: 0.0500 (0.0500) time: 0.7123 data: 0.0004 max mem: 64948 Epoch: [441] [ 30/312] eta: 0:04:48 lr: 0.000005 min_lr: 0.000005 loss: 1.7838 (1.6837) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0003 max mem: 64948 Epoch: [441] [ 40/312] eta: 0:04:17 lr: 0.000005 min_lr: 0.000005 loss: 1.7575 (1.6748) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [441] [ 50/312] eta: 0:03:54 lr: 0.000005 min_lr: 0.000005 loss: 1.7145 (1.6526) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [441] [ 60/312] eta: 0:03:37 lr: 0.000005 min_lr: 0.000005 loss: 1.6236 (1.6411) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [441] [ 70/312] eta: 0:03:23 lr: 0.000005 min_lr: 0.000005 loss: 1.6236 (1.6503) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [441] [ 80/312] eta: 0:03:10 lr: 0.000005 min_lr: 0.000005 loss: 1.6502 (1.6474) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [441] [ 90/312] eta: 0:02:59 lr: 0.000005 min_lr: 0.000005 loss: 1.6963 (1.6504) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0003 max mem: 64948 Epoch: [441] [100/312] eta: 0:02:49 lr: 0.000005 min_lr: 0.000005 loss: 1.7562 (1.6529) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [441] [110/312] eta: 0:02:39 lr: 0.000005 min_lr: 0.000005 loss: 1.6606 (1.6439) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [441] [120/312] eta: 0:02:29 lr: 0.000005 min_lr: 0.000005 loss: 1.6606 (1.6523) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [441] [130/312] eta: 0:02:20 lr: 0.000005 min_lr: 0.000005 loss: 1.7310 (1.6497) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [441] [140/312] eta: 0:02:12 lr: 0.000005 min_lr: 0.000005 loss: 1.7310 (1.6530) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [441] [150/312] eta: 0:02:03 lr: 0.000005 min_lr: 0.000005 loss: 1.8137 (1.6617) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [441] [160/312] eta: 0:01:55 lr: 0.000005 min_lr: 0.000005 loss: 1.7530 (1.6594) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [441] [170/312] eta: 0:01:47 lr: 0.000005 min_lr: 0.000005 loss: 1.6084 (1.6586) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [441] [180/312] eta: 0:01:39 lr: 0.000005 min_lr: 0.000005 loss: 1.5483 (1.6539) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0004 max mem: 64948 Epoch: [441] [190/312] eta: 0:01:31 lr: 0.000005 min_lr: 0.000005 loss: 1.5483 (1.6515) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [441] [200/312] eta: 0:01:23 lr: 0.000005 min_lr: 0.000005 loss: 1.6450 (1.6537) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [441] [210/312] eta: 0:01:15 lr: 0.000005 min_lr: 0.000005 loss: 1.7060 (1.6546) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [441] [220/312] eta: 0:01:08 lr: 0.000005 min_lr: 0.000005 loss: 1.6959 (1.6529) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [441] [230/312] eta: 0:01:00 lr: 0.000005 min_lr: 0.000005 loss: 1.7074 (1.6531) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [441] [240/312] eta: 0:00:53 lr: 0.000005 min_lr: 0.000005 loss: 1.7235 (1.6553) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [441] [250/312] eta: 0:00:45 lr: 0.000005 min_lr: 0.000005 loss: 1.7231 (1.6535) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [441] [260/312] eta: 0:00:38 lr: 0.000005 min_lr: 0.000005 loss: 1.7337 (1.6549) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [441] [270/312] eta: 0:00:30 lr: 0.000005 min_lr: 0.000005 loss: 1.6063 (1.6479) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0004 max mem: 64948 Epoch: [441] [280/312] eta: 0:00:23 lr: 0.000005 min_lr: 0.000005 loss: 1.6012 (1.6472) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [441] [290/312] eta: 0:00:16 lr: 0.000004 min_lr: 0.000004 loss: 1.7834 (1.6471) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0008 max mem: 64948 Epoch: [441] [300/312] eta: 0:00:08 lr: 0.000004 min_lr: 0.000004 loss: 1.6732 (1.6472) weight_decay: 0.0500 (0.0500) time: 0.6920 data: 0.0001 max mem: 64948 Epoch: [441] [310/312] eta: 0:00:01 lr: 0.000004 min_lr: 0.000004 loss: 1.6804 (1.6486) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [441] [311/312] eta: 0:00:00 lr: 0.000004 min_lr: 0.000004 loss: 1.6804 (1.6487) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [441] Total time: 0:03:47 (0.7299 s / it) Averaged stats: lr: 0.000004 min_lr: 0.000004 loss: 1.6804 (1.6460) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:40 loss: 0.4381 (0.4381) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.4806 data: 4.2704 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6104 (0.6338) acc1: 83.5938 (82.9440) acc5: 97.1354 (96.7360) time: 0.6491 data: 0.4746 max mem: 64948 Test: Total time: 0:00:06 (0.6771 s / it) * Acc@1 83.800 Acc@5 96.748 loss 0.617 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:41 loss: 0.4441 (0.4441) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.6654 data: 4.4474 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6093 (0.6353) acc1: 84.6354 (83.0400) acc5: 97.3958 (96.8000) time: 0.6700 data: 0.4943 max mem: 64948 Test: Total time: 0:00:06 (0.6802 s / it) * Acc@1 83.772 Acc@5 96.734 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [442] [ 0/312] eta: 0:54:30 lr: 0.000004 min_lr: 0.000004 loss: 1.7207 (1.7207) weight_decay: 0.0500 (0.0500) time: 10.4817 data: 7.9988 max mem: 64948 Epoch: [442] [ 10/312] eta: 0:08:15 lr: 0.000004 min_lr: 0.000004 loss: 1.5791 (1.5596) weight_decay: 0.0500 (0.0500) time: 1.6413 data: 0.7276 max mem: 64948 Epoch: [442] [ 20/312] eta: 0:05:48 lr: 0.000004 min_lr: 0.000004 loss: 1.5791 (1.5791) weight_decay: 0.0500 (0.0500) time: 0.7289 data: 0.0004 max mem: 64948 Epoch: [442] [ 30/312] eta: 0:04:51 lr: 0.000004 min_lr: 0.000004 loss: 1.5479 (1.5523) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [442] [ 40/312] eta: 0:04:18 lr: 0.000004 min_lr: 0.000004 loss: 1.7338 (1.6045) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [442] [ 50/312] eta: 0:03:55 lr: 0.000004 min_lr: 0.000004 loss: 1.6911 (1.5667) weight_decay: 0.0500 (0.0500) time: 0.6962 data: 0.0003 max mem: 64948 Epoch: [442] [ 60/312] eta: 0:03:38 lr: 0.000004 min_lr: 0.000004 loss: 1.6744 (1.5817) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [442] [ 70/312] eta: 0:03:24 lr: 0.000004 min_lr: 0.000004 loss: 1.7134 (1.5856) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [442] [ 80/312] eta: 0:03:11 lr: 0.000004 min_lr: 0.000004 loss: 1.5322 (1.5823) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [442] [ 90/312] eta: 0:02:59 lr: 0.000004 min_lr: 0.000004 loss: 1.6610 (1.5970) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [442] [100/312] eta: 0:02:49 lr: 0.000004 min_lr: 0.000004 loss: 1.7809 (1.6171) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [442] [110/312] eta: 0:02:39 lr: 0.000004 min_lr: 0.000004 loss: 1.8120 (1.6279) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [442] [120/312] eta: 0:02:30 lr: 0.000004 min_lr: 0.000004 loss: 1.7963 (1.6324) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [442] [130/312] eta: 0:02:21 lr: 0.000004 min_lr: 0.000004 loss: 1.7460 (1.6394) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [442] [140/312] eta: 0:02:12 lr: 0.000004 min_lr: 0.000004 loss: 1.7015 (1.6435) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0003 max mem: 64948 Epoch: [442] [150/312] eta: 0:02:03 lr: 0.000004 min_lr: 0.000004 loss: 1.6186 (1.6358) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [442] [160/312] eta: 0:01:55 lr: 0.000004 min_lr: 0.000004 loss: 1.6235 (1.6373) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [442] [170/312] eta: 0:01:47 lr: 0.000004 min_lr: 0.000004 loss: 1.6821 (1.6377) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [442] [180/312] eta: 0:01:39 lr: 0.000004 min_lr: 0.000004 loss: 1.5589 (1.6374) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0003 max mem: 64948 Epoch: [442] [190/312] eta: 0:01:31 lr: 0.000004 min_lr: 0.000004 loss: 1.7438 (1.6428) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [442] [200/312] eta: 0:01:23 lr: 0.000004 min_lr: 0.000004 loss: 1.7438 (1.6446) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [442] [210/312] eta: 0:01:15 lr: 0.000004 min_lr: 0.000004 loss: 1.7160 (1.6467) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [442] [220/312] eta: 0:01:08 lr: 0.000004 min_lr: 0.000004 loss: 1.6624 (1.6496) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [442] [230/312] eta: 0:01:00 lr: 0.000004 min_lr: 0.000004 loss: 1.6624 (1.6491) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [442] [240/312] eta: 0:00:53 lr: 0.000004 min_lr: 0.000004 loss: 1.7415 (1.6530) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [442] [250/312] eta: 0:00:45 lr: 0.000004 min_lr: 0.000004 loss: 1.6818 (1.6523) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [442] [260/312] eta: 0:00:38 lr: 0.000004 min_lr: 0.000004 loss: 1.5180 (1.6472) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [442] [270/312] eta: 0:00:30 lr: 0.000004 min_lr: 0.000004 loss: 1.5573 (1.6488) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [442] [280/312] eta: 0:00:23 lr: 0.000004 min_lr: 0.000004 loss: 1.5573 (1.6476) weight_decay: 0.0500 (0.0500) time: 0.6975 data: 0.0009 max mem: 64948 Epoch: [442] [290/312] eta: 0:00:16 lr: 0.000004 min_lr: 0.000004 loss: 1.5568 (1.6455) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0008 max mem: 64948 Epoch: [442] [300/312] eta: 0:00:08 lr: 0.000004 min_lr: 0.000004 loss: 1.5698 (1.6435) weight_decay: 0.0500 (0.0500) time: 0.6912 data: 0.0001 max mem: 64948 Epoch: [442] [310/312] eta: 0:00:01 lr: 0.000004 min_lr: 0.000004 loss: 1.7178 (1.6491) weight_decay: 0.0500 (0.0500) time: 0.6914 data: 0.0001 max mem: 64948 Epoch: [442] [311/312] eta: 0:00:00 lr: 0.000004 min_lr: 0.000004 loss: 1.7178 (1.6486) weight_decay: 0.0500 (0.0500) time: 0.6915 data: 0.0001 max mem: 64948 Epoch: [442] Total time: 0:03:48 (0.7311 s / it) Averaged stats: lr: 0.000004 min_lr: 0.000004 loss: 1.7178 (1.6433) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4409 (0.4409) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 4.6137 data: 4.4044 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6123 (0.6370) acc1: 84.3750 (82.9760) acc5: 97.1354 (96.7360) time: 0.6639 data: 0.4895 max mem: 64948 Test: Total time: 0:00:06 (0.6894 s / it) * Acc@1 83.744 Acc@5 96.714 loss 0.619 Accuracy of the model on the 50000 test images: 83.7% Max accuracy: 83.85% Test: [0/9] eta: 0:00:46 loss: 0.4439 (0.4439) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 5.1931 data: 4.9753 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6094 (0.6352) acc1: 84.6354 (83.0720) acc5: 97.3958 (96.8000) time: 0.7287 data: 0.5529 max mem: 64948 Test: Total time: 0:00:06 (0.7364 s / it) * Acc@1 83.784 Acc@5 96.734 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [443] [ 0/312] eta: 0:54:32 lr: 0.000004 min_lr: 0.000004 loss: 1.0142 (1.0142) weight_decay: 0.0500 (0.0500) time: 10.4882 data: 7.1126 max mem: 64948 Epoch: [443] [ 10/312] eta: 0:08:06 lr: 0.000004 min_lr: 0.000004 loss: 1.6247 (1.6007) weight_decay: 0.0500 (0.0500) time: 1.6113 data: 0.6470 max mem: 64948 Epoch: [443] [ 20/312] eta: 0:05:43 lr: 0.000004 min_lr: 0.000004 loss: 1.7541 (1.6574) weight_decay: 0.0500 (0.0500) time: 0.7119 data: 0.0004 max mem: 64948 Epoch: [443] [ 30/312] eta: 0:04:48 lr: 0.000004 min_lr: 0.000004 loss: 1.7998 (1.6856) weight_decay: 0.0500 (0.0500) time: 0.6991 data: 0.0004 max mem: 64948 Epoch: [443] [ 40/312] eta: 0:04:16 lr: 0.000004 min_lr: 0.000004 loss: 1.6428 (1.6677) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [443] [ 50/312] eta: 0:03:54 lr: 0.000003 min_lr: 0.000003 loss: 1.6428 (1.6544) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [443] [ 60/312] eta: 0:03:37 lr: 0.000003 min_lr: 0.000003 loss: 1.6540 (1.6374) weight_decay: 0.0500 (0.0500) time: 0.6956 data: 0.0004 max mem: 64948 Epoch: [443] [ 70/312] eta: 0:03:22 lr: 0.000003 min_lr: 0.000003 loss: 1.6218 (1.6362) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [443] [ 80/312] eta: 0:03:10 lr: 0.000003 min_lr: 0.000003 loss: 1.5994 (1.6297) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [443] [ 90/312] eta: 0:02:58 lr: 0.000003 min_lr: 0.000003 loss: 1.5994 (1.6257) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [443] [100/312] eta: 0:02:48 lr: 0.000003 min_lr: 0.000003 loss: 1.7046 (1.6333) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [443] [110/312] eta: 0:02:38 lr: 0.000003 min_lr: 0.000003 loss: 1.6745 (1.6367) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [443] [120/312] eta: 0:02:29 lr: 0.000003 min_lr: 0.000003 loss: 1.6745 (1.6338) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [443] [130/312] eta: 0:02:20 lr: 0.000003 min_lr: 0.000003 loss: 1.5627 (1.6264) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [443] [140/312] eta: 0:02:11 lr: 0.000003 min_lr: 0.000003 loss: 1.4986 (1.6158) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [443] [150/312] eta: 0:02:03 lr: 0.000003 min_lr: 0.000003 loss: 1.5181 (1.6141) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [443] [160/312] eta: 0:01:55 lr: 0.000003 min_lr: 0.000003 loss: 1.4656 (1.6025) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [443] [170/312] eta: 0:01:47 lr: 0.000003 min_lr: 0.000003 loss: 1.5749 (1.6065) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [443] [180/312] eta: 0:01:39 lr: 0.000003 min_lr: 0.000003 loss: 1.7701 (1.6092) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [443] [190/312] eta: 0:01:31 lr: 0.000003 min_lr: 0.000003 loss: 1.7910 (1.6137) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [443] [200/312] eta: 0:01:23 lr: 0.000003 min_lr: 0.000003 loss: 1.8060 (1.6111) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [443] [210/312] eta: 0:01:15 lr: 0.000003 min_lr: 0.000003 loss: 1.3089 (1.6026) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [443] [220/312] eta: 0:01:08 lr: 0.000003 min_lr: 0.000003 loss: 1.5578 (1.6074) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [443] [230/312] eta: 0:01:00 lr: 0.000003 min_lr: 0.000003 loss: 1.7722 (1.6163) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [443] [240/312] eta: 0:00:53 lr: 0.000003 min_lr: 0.000003 loss: 1.7024 (1.6176) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [443] [250/312] eta: 0:00:45 lr: 0.000003 min_lr: 0.000003 loss: 1.6879 (1.6240) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [443] [260/312] eta: 0:00:38 lr: 0.000003 min_lr: 0.000003 loss: 1.6686 (1.6188) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [443] [270/312] eta: 0:00:30 lr: 0.000003 min_lr: 0.000003 loss: 1.6450 (1.6212) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [443] [280/312] eta: 0:00:23 lr: 0.000003 min_lr: 0.000003 loss: 1.6450 (1.6128) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0010 max mem: 64948 Epoch: [443] [290/312] eta: 0:00:16 lr: 0.000003 min_lr: 0.000003 loss: 1.3397 (1.6111) weight_decay: 0.0500 (0.0500) time: 0.6921 data: 0.0008 max mem: 64948 Epoch: [443] [300/312] eta: 0:00:08 lr: 0.000003 min_lr: 0.000003 loss: 1.7988 (1.6140) weight_decay: 0.0500 (0.0500) time: 0.6905 data: 0.0001 max mem: 64948 Epoch: [443] [310/312] eta: 0:00:01 lr: 0.000003 min_lr: 0.000003 loss: 1.7988 (1.6149) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [443] [311/312] eta: 0:00:00 lr: 0.000003 min_lr: 0.000003 loss: 1.7807 (1.6135) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [443] Total time: 0:03:47 (0.7302 s / it) Averaged stats: lr: 0.000003 min_lr: 0.000003 loss: 1.7807 (1.6418) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4386 (0.4386) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 4.6117 data: 4.3910 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6125 (0.6345) acc1: 84.1146 (83.0400) acc5: 97.1354 (96.8000) time: 0.6644 data: 0.4880 max mem: 64948 Test: Total time: 0:00:06 (0.6919 s / it) * Acc@1 83.820 Acc@5 96.724 loss 0.619 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:45 loss: 0.4437 (0.4437) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 5.0548 data: 4.8432 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6094 (0.6352) acc1: 84.6354 (83.1040) acc5: 97.3958 (96.7680) time: 0.7129 data: 0.5382 max mem: 64948 Test: Total time: 0:00:06 (0.7248 s / it) * Acc@1 83.788 Acc@5 96.734 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [444] [ 0/312] eta: 0:51:47 lr: 0.000003 min_lr: 0.000003 loss: 1.7020 (1.7020) weight_decay: 0.0500 (0.0500) time: 9.9606 data: 7.4850 max mem: 64948 Epoch: [444] [ 10/312] eta: 0:08:06 lr: 0.000003 min_lr: 0.000003 loss: 1.7995 (1.7175) weight_decay: 0.0500 (0.0500) time: 1.6107 data: 0.6808 max mem: 64948 Epoch: [444] [ 20/312] eta: 0:05:42 lr: 0.000003 min_lr: 0.000003 loss: 1.7092 (1.6062) weight_decay: 0.0500 (0.0500) time: 0.7349 data: 0.0004 max mem: 64948 Epoch: [444] [ 30/312] eta: 0:04:47 lr: 0.000003 min_lr: 0.000003 loss: 1.6716 (1.6489) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0003 max mem: 64948 Epoch: [444] [ 40/312] eta: 0:04:15 lr: 0.000003 min_lr: 0.000003 loss: 1.6622 (1.6085) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [444] [ 50/312] eta: 0:03:54 lr: 0.000003 min_lr: 0.000003 loss: 1.5103 (1.6350) weight_decay: 0.0500 (0.0500) time: 0.7012 data: 0.0004 max mem: 64948 Epoch: [444] [ 60/312] eta: 0:03:37 lr: 0.000003 min_lr: 0.000003 loss: 1.8353 (1.6439) weight_decay: 0.0500 (0.0500) time: 0.7018 data: 0.0004 max mem: 64948 Epoch: [444] [ 70/312] eta: 0:03:23 lr: 0.000003 min_lr: 0.000003 loss: 1.6817 (1.6288) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [444] [ 80/312] eta: 0:03:10 lr: 0.000003 min_lr: 0.000003 loss: 1.4343 (1.6008) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [444] [ 90/312] eta: 0:02:59 lr: 0.000003 min_lr: 0.000003 loss: 1.4343 (1.5930) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [444] [100/312] eta: 0:02:48 lr: 0.000003 min_lr: 0.000003 loss: 1.5506 (1.5902) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [444] [110/312] eta: 0:02:38 lr: 0.000003 min_lr: 0.000003 loss: 1.5495 (1.5906) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0004 max mem: 64948 Epoch: [444] [120/312] eta: 0:02:29 lr: 0.000003 min_lr: 0.000003 loss: 1.6202 (1.5945) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [444] [130/312] eta: 0:02:20 lr: 0.000003 min_lr: 0.000003 loss: 1.6202 (1.5946) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [444] [140/312] eta: 0:02:11 lr: 0.000003 min_lr: 0.000003 loss: 1.5611 (1.5928) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [444] [150/312] eta: 0:02:03 lr: 0.000003 min_lr: 0.000003 loss: 1.7142 (1.5951) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [444] [160/312] eta: 0:01:55 lr: 0.000003 min_lr: 0.000003 loss: 1.6743 (1.5966) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [444] [170/312] eta: 0:01:47 lr: 0.000003 min_lr: 0.000003 loss: 1.6743 (1.6067) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [444] [180/312] eta: 0:01:39 lr: 0.000003 min_lr: 0.000003 loss: 1.5741 (1.5942) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [444] [190/312] eta: 0:01:31 lr: 0.000003 min_lr: 0.000003 loss: 1.4590 (1.5951) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [444] [200/312] eta: 0:01:23 lr: 0.000003 min_lr: 0.000003 loss: 1.6848 (1.5960) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [444] [210/312] eta: 0:01:15 lr: 0.000003 min_lr: 0.000003 loss: 1.7214 (1.5996) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [444] [220/312] eta: 0:01:08 lr: 0.000002 min_lr: 0.000002 loss: 1.7819 (1.6054) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0004 max mem: 64948 Epoch: [444] [230/312] eta: 0:01:00 lr: 0.000002 min_lr: 0.000002 loss: 1.7173 (1.6064) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [444] [240/312] eta: 0:00:53 lr: 0.000002 min_lr: 0.000002 loss: 1.5356 (1.6026) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [444] [250/312] eta: 0:00:45 lr: 0.000002 min_lr: 0.000002 loss: 1.4623 (1.6008) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [444] [260/312] eta: 0:00:38 lr: 0.000002 min_lr: 0.000002 loss: 1.6106 (1.6013) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [444] [270/312] eta: 0:00:30 lr: 0.000002 min_lr: 0.000002 loss: 1.6374 (1.6022) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0004 max mem: 64948 Epoch: [444] [280/312] eta: 0:00:23 lr: 0.000002 min_lr: 0.000002 loss: 1.6166 (1.6000) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0009 max mem: 64948 Epoch: [444] [290/312] eta: 0:00:16 lr: 0.000002 min_lr: 0.000002 loss: 1.6579 (1.6042) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0008 max mem: 64948 Epoch: [444] [300/312] eta: 0:00:08 lr: 0.000002 min_lr: 0.000002 loss: 1.6963 (1.6048) weight_decay: 0.0500 (0.0500) time: 0.6906 data: 0.0001 max mem: 64948 Epoch: [444] [310/312] eta: 0:00:01 lr: 0.000002 min_lr: 0.000002 loss: 1.7693 (1.6078) weight_decay: 0.0500 (0.0500) time: 0.6907 data: 0.0001 max mem: 64948 Epoch: [444] [311/312] eta: 0:00:00 lr: 0.000002 min_lr: 0.000002 loss: 1.7369 (1.6082) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [444] Total time: 0:03:47 (0.7297 s / it) Averaged stats: lr: 0.000002 min_lr: 0.000002 loss: 1.7369 (1.6453) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:39 loss: 0.4369 (0.4369) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 4.3910 data: 4.1815 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6093 (0.6331) acc1: 84.1146 (83.0400) acc5: 97.3958 (96.8640) time: 0.6392 data: 0.4647 max mem: 64948 Test: Total time: 0:00:05 (0.6626 s / it) * Acc@1 83.814 Acc@5 96.736 loss 0.617 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:44 loss: 0.4435 (0.4435) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.9804 data: 4.7750 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6095 (0.6351) acc1: 84.6354 (83.1040) acc5: 97.3958 (96.7680) time: 0.7051 data: 0.5306 max mem: 64948 Test: Total time: 0:00:06 (0.7126 s / it) * Acc@1 83.800 Acc@5 96.734 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.80% Epoch: [445] [ 0/312] eta: 0:54:48 lr: 0.000002 min_lr: 0.000002 loss: 1.4797 (1.4797) weight_decay: 0.0500 (0.0500) time: 10.5412 data: 9.7730 max mem: 64948 Epoch: [445] [ 10/312] eta: 0:08:03 lr: 0.000002 min_lr: 0.000002 loss: 1.6793 (1.5773) weight_decay: 0.0500 (0.0500) time: 1.6004 data: 0.8888 max mem: 64948 Epoch: [445] [ 20/312] eta: 0:05:41 lr: 0.000002 min_lr: 0.000002 loss: 1.6793 (1.5952) weight_decay: 0.0500 (0.0500) time: 0.7001 data: 0.0003 max mem: 64948 Epoch: [445] [ 30/312] eta: 0:04:46 lr: 0.000002 min_lr: 0.000002 loss: 1.6338 (1.5865) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0003 max mem: 64948 Epoch: [445] [ 40/312] eta: 0:04:15 lr: 0.000002 min_lr: 0.000002 loss: 1.6200 (1.5912) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0003 max mem: 64948 Epoch: [445] [ 50/312] eta: 0:03:53 lr: 0.000002 min_lr: 0.000002 loss: 1.7738 (1.6089) weight_decay: 0.0500 (0.0500) time: 0.6980 data: 0.0004 max mem: 64948 Epoch: [445] [ 60/312] eta: 0:03:36 lr: 0.000002 min_lr: 0.000002 loss: 1.5683 (1.5866) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [445] [ 70/312] eta: 0:03:22 lr: 0.000002 min_lr: 0.000002 loss: 1.4110 (1.5920) weight_decay: 0.0500 (0.0500) time: 0.6977 data: 0.0004 max mem: 64948 Epoch: [445] [ 80/312] eta: 0:03:10 lr: 0.000002 min_lr: 0.000002 loss: 1.8497 (1.6197) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0004 max mem: 64948 Epoch: [445] [ 90/312] eta: 0:02:58 lr: 0.000002 min_lr: 0.000002 loss: 1.7623 (1.6236) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0003 max mem: 64948 Epoch: [445] [100/312] eta: 0:02:48 lr: 0.000002 min_lr: 0.000002 loss: 1.7082 (1.6291) weight_decay: 0.0500 (0.0500) time: 0.6926 data: 0.0004 max mem: 64948 Epoch: [445] [110/312] eta: 0:02:38 lr: 0.000002 min_lr: 0.000002 loss: 1.7082 (1.6254) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [445] [120/312] eta: 0:02:29 lr: 0.000002 min_lr: 0.000002 loss: 1.6038 (1.6301) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [445] [130/312] eta: 0:02:20 lr: 0.000002 min_lr: 0.000002 loss: 1.6454 (1.6293) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [445] [140/312] eta: 0:02:11 lr: 0.000002 min_lr: 0.000002 loss: 1.6454 (1.6323) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [445] [150/312] eta: 0:02:03 lr: 0.000002 min_lr: 0.000002 loss: 1.6392 (1.6225) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [445] [160/312] eta: 0:01:55 lr: 0.000002 min_lr: 0.000002 loss: 1.5244 (1.6171) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [445] [170/312] eta: 0:01:46 lr: 0.000002 min_lr: 0.000002 loss: 1.6576 (1.6187) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [445] [180/312] eta: 0:01:39 lr: 0.000002 min_lr: 0.000002 loss: 1.6369 (1.6205) weight_decay: 0.0500 (0.0500) time: 0.6942 data: 0.0004 max mem: 64948 Epoch: [445] [190/312] eta: 0:01:31 lr: 0.000002 min_lr: 0.000002 loss: 1.7870 (1.6256) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [445] [200/312] eta: 0:01:23 lr: 0.000002 min_lr: 0.000002 loss: 1.7870 (1.6319) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [445] [210/312] eta: 0:01:15 lr: 0.000002 min_lr: 0.000002 loss: 1.7394 (1.6355) weight_decay: 0.0500 (0.0500) time: 0.6969 data: 0.0004 max mem: 64948 Epoch: [445] [220/312] eta: 0:01:08 lr: 0.000002 min_lr: 0.000002 loss: 1.6625 (1.6355) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [445] [230/312] eta: 0:01:00 lr: 0.000002 min_lr: 0.000002 loss: 1.6471 (1.6379) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [445] [240/312] eta: 0:00:53 lr: 0.000002 min_lr: 0.000002 loss: 1.7395 (1.6410) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [445] [250/312] eta: 0:00:45 lr: 0.000002 min_lr: 0.000002 loss: 1.6958 (1.6369) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [445] [260/312] eta: 0:00:38 lr: 0.000002 min_lr: 0.000002 loss: 1.5818 (1.6344) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [445] [270/312] eta: 0:00:30 lr: 0.000002 min_lr: 0.000002 loss: 1.7396 (1.6404) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [445] [280/312] eta: 0:00:23 lr: 0.000002 min_lr: 0.000002 loss: 1.8749 (1.6489) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0009 max mem: 64948 Epoch: [445] [290/312] eta: 0:00:16 lr: 0.000002 min_lr: 0.000002 loss: 1.7539 (1.6448) weight_decay: 0.0500 (0.0500) time: 0.6928 data: 0.0008 max mem: 64948 Epoch: [445] [300/312] eta: 0:00:08 lr: 0.000002 min_lr: 0.000002 loss: 1.3952 (1.6401) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [445] [310/312] eta: 0:00:01 lr: 0.000002 min_lr: 0.000002 loss: 1.5909 (1.6431) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [445] [311/312] eta: 0:00:00 lr: 0.000002 min_lr: 0.000002 loss: 1.5909 (1.6429) weight_decay: 0.0500 (0.0500) time: 0.6922 data: 0.0001 max mem: 64948 Epoch: [445] Total time: 0:03:47 (0.7295 s / it) Averaged stats: lr: 0.000002 min_lr: 0.000002 loss: 1.5909 (1.6456) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4388 (0.4388) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.5646 data: 4.3576 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6114 (0.6330) acc1: 84.3750 (83.0400) acc5: 97.1354 (96.8000) time: 0.6585 data: 0.4843 max mem: 64948 Test: Total time: 0:00:06 (0.6840 s / it) * Acc@1 83.808 Acc@5 96.756 loss 0.618 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:44 loss: 0.4433 (0.4433) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 4.9029 data: 4.6973 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6096 (0.6350) acc1: 84.6354 (83.1040) acc5: 97.3958 (96.7680) time: 0.6960 data: 0.5220 max mem: 64948 Test: Total time: 0:00:06 (0.7110 s / it) * Acc@1 83.804 Acc@5 96.734 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.80% Epoch: [446] [ 0/312] eta: 0:50:59 lr: 0.000002 min_lr: 0.000002 loss: 1.7907 (1.7907) weight_decay: 0.0500 (0.0500) time: 9.8059 data: 8.3088 max mem: 64948 Epoch: [446] [ 10/312] eta: 0:07:46 lr: 0.000002 min_lr: 0.000002 loss: 1.6687 (1.4966) weight_decay: 0.0500 (0.0500) time: 1.5437 data: 0.7557 max mem: 64948 Epoch: [446] [ 20/312] eta: 0:05:32 lr: 0.000002 min_lr: 0.000002 loss: 1.6811 (1.5886) weight_decay: 0.0500 (0.0500) time: 0.7047 data: 0.0004 max mem: 64948 Epoch: [446] [ 30/312] eta: 0:04:40 lr: 0.000002 min_lr: 0.000002 loss: 1.7121 (1.5946) weight_decay: 0.0500 (0.0500) time: 0.6934 data: 0.0003 max mem: 64948 Epoch: [446] [ 40/312] eta: 0:04:10 lr: 0.000002 min_lr: 0.000002 loss: 1.7121 (1.6197) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0003 max mem: 64948 Epoch: [446] [ 50/312] eta: 0:03:49 lr: 0.000002 min_lr: 0.000002 loss: 1.7088 (1.5965) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [446] [ 60/312] eta: 0:03:33 lr: 0.000002 min_lr: 0.000002 loss: 1.5587 (1.5843) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [446] [ 70/312] eta: 0:03:19 lr: 0.000002 min_lr: 0.000002 loss: 1.5587 (1.5734) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [446] [ 80/312] eta: 0:03:08 lr: 0.000002 min_lr: 0.000002 loss: 1.6156 (1.5857) weight_decay: 0.0500 (0.0500) time: 0.7012 data: 0.0004 max mem: 64948 Epoch: [446] [ 90/312] eta: 0:02:57 lr: 0.000002 min_lr: 0.000002 loss: 1.6216 (1.5890) weight_decay: 0.0500 (0.0500) time: 0.6997 data: 0.0004 max mem: 64948 Epoch: [446] [100/312] eta: 0:02:47 lr: 0.000002 min_lr: 0.000002 loss: 1.8242 (1.6144) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [446] [110/312] eta: 0:02:37 lr: 0.000002 min_lr: 0.000002 loss: 1.7798 (1.6183) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [446] [120/312] eta: 0:02:28 lr: 0.000002 min_lr: 0.000002 loss: 1.8120 (1.6435) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [446] [130/312] eta: 0:02:19 lr: 0.000002 min_lr: 0.000002 loss: 1.8536 (1.6463) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0004 max mem: 64948 Epoch: [446] [140/312] eta: 0:02:10 lr: 0.000002 min_lr: 0.000002 loss: 1.6870 (1.6338) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [446] [150/312] eta: 0:02:02 lr: 0.000002 min_lr: 0.000002 loss: 1.6870 (1.6384) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [446] [160/312] eta: 0:01:54 lr: 0.000002 min_lr: 0.000002 loss: 1.7301 (1.6363) weight_decay: 0.0500 (0.0500) time: 0.6967 data: 0.0004 max mem: 64948 Epoch: [446] [170/312] eta: 0:01:46 lr: 0.000002 min_lr: 0.000002 loss: 1.7708 (1.6405) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [446] [180/312] eta: 0:01:38 lr: 0.000002 min_lr: 0.000002 loss: 1.6868 (1.6391) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [446] [190/312] eta: 0:01:30 lr: 0.000002 min_lr: 0.000002 loss: 1.6685 (1.6434) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [446] [200/312] eta: 0:01:23 lr: 0.000002 min_lr: 0.000002 loss: 1.6911 (1.6458) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [446] [210/312] eta: 0:01:15 lr: 0.000002 min_lr: 0.000002 loss: 1.6786 (1.6470) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [446] [220/312] eta: 0:01:07 lr: 0.000002 min_lr: 0.000002 loss: 1.6882 (1.6460) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [446] [230/312] eta: 0:01:00 lr: 0.000002 min_lr: 0.000002 loss: 1.7384 (1.6489) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [446] [240/312] eta: 0:00:52 lr: 0.000002 min_lr: 0.000002 loss: 1.7324 (1.6474) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [446] [250/312] eta: 0:00:45 lr: 0.000002 min_lr: 0.000002 loss: 1.6254 (1.6461) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [446] [260/312] eta: 0:00:38 lr: 0.000002 min_lr: 0.000002 loss: 1.6277 (1.6484) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [446] [270/312] eta: 0:00:30 lr: 0.000002 min_lr: 0.000002 loss: 1.6688 (1.6467) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [446] [280/312] eta: 0:00:23 lr: 0.000002 min_lr: 0.000002 loss: 1.6688 (1.6474) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0010 max mem: 64948 Epoch: [446] [290/312] eta: 0:00:15 lr: 0.000002 min_lr: 0.000002 loss: 1.6541 (1.6480) weight_decay: 0.0500 (0.0500) time: 0.6929 data: 0.0009 max mem: 64948 Epoch: [446] [300/312] eta: 0:00:08 lr: 0.000001 min_lr: 0.000001 loss: 1.7984 (1.6509) weight_decay: 0.0500 (0.0500) time: 0.6916 data: 0.0001 max mem: 64948 Epoch: [446] [310/312] eta: 0:00:01 lr: 0.000001 min_lr: 0.000001 loss: 1.7616 (1.6509) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [446] [311/312] eta: 0:00:00 lr: 0.000001 min_lr: 0.000001 loss: 1.7616 (1.6518) weight_decay: 0.0500 (0.0500) time: 0.6917 data: 0.0001 max mem: 64948 Epoch: [446] Total time: 0:03:47 (0.7277 s / it) Averaged stats: lr: 0.000001 min_lr: 0.000001 loss: 1.7616 (1.6420) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:42 loss: 0.4348 (0.4348) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 4.7482 data: 4.5326 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6094 (0.6335) acc1: 84.3750 (83.0080) acc5: 97.1354 (96.8000) time: 0.6788 data: 0.5037 max mem: 64948 Test: Total time: 0:00:06 (0.7071 s / it) * Acc@1 83.786 Acc@5 96.740 loss 0.617 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:46 loss: 0.4431 (0.4431) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 5.1389 data: 4.9255 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6096 (0.6350) acc1: 84.6354 (83.1040) acc5: 97.3958 (96.7680) time: 0.7222 data: 0.5474 max mem: 64948 Test: Total time: 0:00:06 (0.7351 s / it) * Acc@1 83.796 Acc@5 96.736 loss 0.617 Accuracy of the model EMA on 50000 test images: 83.8% Epoch: [447] [ 0/312] eta: 0:53:18 lr: 0.000001 min_lr: 0.000001 loss: 1.7000 (1.7000) weight_decay: 0.0500 (0.0500) time: 10.2511 data: 7.3490 max mem: 64948 Epoch: [447] [ 10/312] eta: 0:07:58 lr: 0.000001 min_lr: 0.000001 loss: 1.7000 (1.6498) weight_decay: 0.0500 (0.0500) time: 1.5848 data: 0.6685 max mem: 64948 Epoch: [447] [ 20/312] eta: 0:05:39 lr: 0.000001 min_lr: 0.000001 loss: 1.6377 (1.6295) weight_decay: 0.0500 (0.0500) time: 0.7065 data: 0.0004 max mem: 64948 Epoch: [447] [ 30/312] eta: 0:04:45 lr: 0.000001 min_lr: 0.000001 loss: 1.6377 (1.6209) weight_decay: 0.0500 (0.0500) time: 0.6953 data: 0.0004 max mem: 64948 Epoch: [447] [ 40/312] eta: 0:04:13 lr: 0.000001 min_lr: 0.000001 loss: 1.7536 (1.6465) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [447] [ 50/312] eta: 0:03:52 lr: 0.000001 min_lr: 0.000001 loss: 1.6951 (1.6138) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0004 max mem: 64948 Epoch: [447] [ 60/312] eta: 0:03:35 lr: 0.000001 min_lr: 0.000001 loss: 1.5795 (1.6215) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [447] [ 70/312] eta: 0:03:21 lr: 0.000001 min_lr: 0.000001 loss: 1.7515 (1.6449) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [447] [ 80/312] eta: 0:03:09 lr: 0.000001 min_lr: 0.000001 loss: 1.7726 (1.6434) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [447] [ 90/312] eta: 0:02:58 lr: 0.000001 min_lr: 0.000001 loss: 1.6574 (1.6451) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [447] [100/312] eta: 0:02:47 lr: 0.000001 min_lr: 0.000001 loss: 1.6538 (1.6437) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [447] [110/312] eta: 0:02:38 lr: 0.000001 min_lr: 0.000001 loss: 1.6687 (1.6436) weight_decay: 0.0500 (0.0500) time: 0.6973 data: 0.0004 max mem: 64948 Epoch: [447] [120/312] eta: 0:02:29 lr: 0.000001 min_lr: 0.000001 loss: 1.6687 (1.6360) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [447] [130/312] eta: 0:02:20 lr: 0.000001 min_lr: 0.000001 loss: 1.4594 (1.6228) weight_decay: 0.0500 (0.0500) time: 0.6943 data: 0.0004 max mem: 64948 Epoch: [447] [140/312] eta: 0:02:11 lr: 0.000001 min_lr: 0.000001 loss: 1.7163 (1.6257) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [447] [150/312] eta: 0:02:03 lr: 0.000001 min_lr: 0.000001 loss: 1.8002 (1.6303) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [447] [160/312] eta: 0:01:54 lr: 0.000001 min_lr: 0.000001 loss: 1.7217 (1.6346) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [447] [170/312] eta: 0:01:46 lr: 0.000001 min_lr: 0.000001 loss: 1.7417 (1.6378) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [447] [180/312] eta: 0:01:38 lr: 0.000001 min_lr: 0.000001 loss: 1.5765 (1.6343) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [447] [190/312] eta: 0:01:31 lr: 0.000001 min_lr: 0.000001 loss: 1.5321 (1.6315) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [447] [200/312] eta: 0:01:23 lr: 0.000001 min_lr: 0.000001 loss: 1.5321 (1.6307) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [447] [210/312] eta: 0:01:15 lr: 0.000001 min_lr: 0.000001 loss: 1.6269 (1.6304) weight_decay: 0.0500 (0.0500) time: 0.6982 data: 0.0004 max mem: 64948 Epoch: [447] [220/312] eta: 0:01:08 lr: 0.000001 min_lr: 0.000001 loss: 1.6046 (1.6270) weight_decay: 0.0500 (0.0500) time: 0.6978 data: 0.0003 max mem: 64948 Epoch: [447] [230/312] eta: 0:01:00 lr: 0.000001 min_lr: 0.000001 loss: 1.7331 (1.6310) weight_decay: 0.0500 (0.0500) time: 0.6935 data: 0.0003 max mem: 64948 Epoch: [447] [240/312] eta: 0:00:52 lr: 0.000001 min_lr: 0.000001 loss: 1.7331 (1.6304) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0003 max mem: 64948 Epoch: [447] [250/312] eta: 0:00:45 lr: 0.000001 min_lr: 0.000001 loss: 1.6819 (1.6367) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [447] [260/312] eta: 0:00:38 lr: 0.000001 min_lr: 0.000001 loss: 1.7134 (1.6324) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0004 max mem: 64948 Epoch: [447] [270/312] eta: 0:00:30 lr: 0.000001 min_lr: 0.000001 loss: 1.6578 (1.6312) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [447] [280/312] eta: 0:00:23 lr: 0.000001 min_lr: 0.000001 loss: 1.6578 (1.6324) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0009 max mem: 64948 Epoch: [447] [290/312] eta: 0:00:16 lr: 0.000001 min_lr: 0.000001 loss: 1.6129 (1.6297) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0008 max mem: 64948 Epoch: [447] [300/312] eta: 0:00:08 lr: 0.000001 min_lr: 0.000001 loss: 1.6065 (1.6302) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [447] [310/312] eta: 0:00:01 lr: 0.000001 min_lr: 0.000001 loss: 1.7846 (1.6321) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0001 max mem: 64948 Epoch: [447] [311/312] eta: 0:00:00 lr: 0.000001 min_lr: 0.000001 loss: 1.7866 (1.6329) weight_decay: 0.0500 (0.0500) time: 0.6925 data: 0.0001 max mem: 64948 Epoch: [447] Total time: 0:03:47 (0.7291 s / it) Averaged stats: lr: 0.000001 min_lr: 0.000001 loss: 1.7866 (1.6441) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:44 loss: 0.4416 (0.4416) acc1: 88.8021 (88.8021) acc5: 97.9167 (97.9167) time: 4.9042 data: 4.6882 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6134 (0.6343) acc1: 84.3750 (83.0080) acc5: 97.1354 (96.7680) time: 0.6969 data: 0.5210 max mem: 64948 Test: Total time: 0:00:06 (0.7238 s / it) * Acc@1 83.774 Acc@5 96.732 loss 0.617 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:46 loss: 0.4430 (0.4430) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 5.1443 data: 4.9264 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6096 (0.6349) acc1: 84.6354 (83.1360) acc5: 97.3958 (96.7680) time: 0.7234 data: 0.5475 max mem: 64948 Test: Total time: 0:00:06 (0.7309 s / it) * Acc@1 83.808 Acc@5 96.748 loss 0.616 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.81% Epoch: [448] [ 0/312] eta: 0:44:25 lr: 0.000001 min_lr: 0.000001 loss: 1.7737 (1.7737) weight_decay: 0.0500 (0.0500) time: 8.5435 data: 7.5893 max mem: 64948 Epoch: [448] [ 10/312] eta: 0:07:29 lr: 0.000001 min_lr: 0.000001 loss: 1.7123 (1.6511) weight_decay: 0.0500 (0.0500) time: 1.4879 data: 0.6957 max mem: 64948 Epoch: [448] [ 20/312] eta: 0:05:23 lr: 0.000001 min_lr: 0.000001 loss: 1.6229 (1.6632) weight_decay: 0.0500 (0.0500) time: 0.7376 data: 0.0034 max mem: 64948 Epoch: [448] [ 30/312] eta: 0:04:34 lr: 0.000001 min_lr: 0.000001 loss: 1.6799 (1.6721) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0003 max mem: 64948 Epoch: [448] [ 40/312] eta: 0:04:06 lr: 0.000001 min_lr: 0.000001 loss: 1.6416 (1.6380) weight_decay: 0.0500 (0.0500) time: 0.6938 data: 0.0003 max mem: 64948 Epoch: [448] [ 50/312] eta: 0:03:46 lr: 0.000001 min_lr: 0.000001 loss: 1.5509 (1.6421) weight_decay: 0.0500 (0.0500) time: 0.6952 data: 0.0004 max mem: 64948 Epoch: [448] [ 60/312] eta: 0:03:31 lr: 0.000001 min_lr: 0.000001 loss: 1.7911 (1.6784) weight_decay: 0.0500 (0.0500) time: 0.6986 data: 0.0004 max mem: 64948 Epoch: [448] [ 70/312] eta: 0:03:18 lr: 0.000001 min_lr: 0.000001 loss: 1.8380 (1.6901) weight_decay: 0.0500 (0.0500) time: 0.6984 data: 0.0004 max mem: 64948 Epoch: [448] [ 80/312] eta: 0:03:06 lr: 0.000001 min_lr: 0.000001 loss: 1.7657 (1.6874) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [448] [ 90/312] eta: 0:02:55 lr: 0.000001 min_lr: 0.000001 loss: 1.7287 (1.6808) weight_decay: 0.0500 (0.0500) time: 0.6960 data: 0.0004 max mem: 64948 Epoch: [448] [100/312] eta: 0:02:45 lr: 0.000001 min_lr: 0.000001 loss: 1.7449 (1.6850) weight_decay: 0.0500 (0.0500) time: 0.6961 data: 0.0004 max mem: 64948 Epoch: [448] [110/312] eta: 0:02:36 lr: 0.000001 min_lr: 0.000001 loss: 1.7449 (1.6856) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [448] [120/312] eta: 0:02:27 lr: 0.000001 min_lr: 0.000001 loss: 1.7669 (1.6914) weight_decay: 0.0500 (0.0500) time: 0.6957 data: 0.0004 max mem: 64948 Epoch: [448] [130/312] eta: 0:02:18 lr: 0.000001 min_lr: 0.000001 loss: 1.8613 (1.7017) weight_decay: 0.0500 (0.0500) time: 0.6972 data: 0.0004 max mem: 64948 Epoch: [448] [140/312] eta: 0:02:10 lr: 0.000001 min_lr: 0.000001 loss: 1.9001 (1.7030) weight_decay: 0.0500 (0.0500) time: 0.6976 data: 0.0004 max mem: 64948 Epoch: [448] [150/312] eta: 0:02:02 lr: 0.000001 min_lr: 0.000001 loss: 1.6172 (1.6911) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [448] [160/312] eta: 0:01:54 lr: 0.000001 min_lr: 0.000001 loss: 1.7306 (1.6904) weight_decay: 0.0500 (0.0500) time: 0.6966 data: 0.0004 max mem: 64948 Epoch: [448] [170/312] eta: 0:01:46 lr: 0.000001 min_lr: 0.000001 loss: 1.7306 (1.6830) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0004 max mem: 64948 Epoch: [448] [180/312] eta: 0:01:38 lr: 0.000001 min_lr: 0.000001 loss: 1.5812 (1.6766) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [448] [190/312] eta: 0:01:30 lr: 0.000001 min_lr: 0.000001 loss: 1.6059 (1.6703) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [448] [200/312] eta: 0:01:22 lr: 0.000001 min_lr: 0.000001 loss: 1.4668 (1.6579) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [448] [210/312] eta: 0:01:15 lr: 0.000001 min_lr: 0.000001 loss: 1.5885 (1.6603) weight_decay: 0.0500 (0.0500) time: 0.6949 data: 0.0004 max mem: 64948 Epoch: [448] [220/312] eta: 0:01:07 lr: 0.000001 min_lr: 0.000001 loss: 1.7054 (1.6621) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [448] [230/312] eta: 0:01:00 lr: 0.000001 min_lr: 0.000001 loss: 1.6794 (1.6553) weight_decay: 0.0500 (0.0500) time: 0.6927 data: 0.0004 max mem: 64948 Epoch: [448] [240/312] eta: 0:00:52 lr: 0.000001 min_lr: 0.000001 loss: 1.7727 (1.6601) weight_decay: 0.0500 (0.0500) time: 0.6931 data: 0.0004 max mem: 64948 Epoch: [448] [250/312] eta: 0:00:45 lr: 0.000001 min_lr: 0.000001 loss: 1.7983 (1.6566) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [448] [260/312] eta: 0:00:37 lr: 0.000001 min_lr: 0.000001 loss: 1.7082 (1.6600) weight_decay: 0.0500 (0.0500) time: 0.6945 data: 0.0004 max mem: 64948 Epoch: [448] [270/312] eta: 0:00:30 lr: 0.000001 min_lr: 0.000001 loss: 1.7231 (1.6623) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0004 max mem: 64948 Epoch: [448] [280/312] eta: 0:00:23 lr: 0.000001 min_lr: 0.000001 loss: 1.7601 (1.6656) weight_decay: 0.0500 (0.0500) time: 0.6940 data: 0.0006 max mem: 64948 Epoch: [448] [290/312] eta: 0:00:15 lr: 0.000001 min_lr: 0.000001 loss: 1.8077 (1.6712) weight_decay: 0.0500 (0.0500) time: 0.6930 data: 0.0005 max mem: 64948 Epoch: [448] [300/312] eta: 0:00:08 lr: 0.000001 min_lr: 0.000001 loss: 1.7319 (1.6695) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [448] [310/312] eta: 0:00:01 lr: 0.000001 min_lr: 0.000001 loss: 1.6763 (1.6692) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [448] [311/312] eta: 0:00:00 lr: 0.000001 min_lr: 0.000001 loss: 1.6954 (1.6696) weight_decay: 0.0500 (0.0500) time: 0.6904 data: 0.0001 max mem: 64948 Epoch: [448] Total time: 0:03:46 (0.7258 s / it) Averaged stats: lr: 0.000001 min_lr: 0.000001 loss: 1.6954 (1.6447) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:43 loss: 0.4351 (0.4351) acc1: 88.5417 (88.5417) acc5: 98.1771 (98.1771) time: 4.7831 data: 4.5638 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6108 (0.6333) acc1: 83.8542 (82.9760) acc5: 97.1354 (96.8640) time: 0.6831 data: 0.5072 max mem: 64948 Test: Total time: 0:00:06 (0.7049 s / it) * Acc@1 83.802 Acc@5 96.736 loss 0.617 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:45 loss: 0.4427 (0.4427) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 5.0280 data: 4.8162 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6096 (0.6349) acc1: 84.6354 (83.1680) acc5: 97.3958 (96.7680) time: 0.7099 data: 0.5352 max mem: 64948 Test: Total time: 0:00:06 (0.7174 s / it) * Acc@1 83.818 Acc@5 96.742 loss 0.616 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.82% Epoch: [449] [ 0/312] eta: 0:53:41 lr: 0.000001 min_lr: 0.000001 loss: 1.6690 (1.6690) weight_decay: 0.0500 (0.0500) time: 10.3238 data: 9.5375 max mem: 64948 Epoch: [449] [ 10/312] eta: 0:07:57 lr: 0.000001 min_lr: 0.000001 loss: 1.5961 (1.5937) weight_decay: 0.0500 (0.0500) time: 1.5811 data: 0.8673 max mem: 64948 Epoch: [449] [ 20/312] eta: 0:05:38 lr: 0.000001 min_lr: 0.000001 loss: 1.4869 (1.5457) weight_decay: 0.0500 (0.0500) time: 0.7006 data: 0.0003 max mem: 64948 Epoch: [449] [ 30/312] eta: 0:04:44 lr: 0.000001 min_lr: 0.000001 loss: 1.5718 (1.5912) weight_decay: 0.0500 (0.0500) time: 0.6951 data: 0.0003 max mem: 64948 Epoch: [449] [ 40/312] eta: 0:04:13 lr: 0.000001 min_lr: 0.000001 loss: 1.6469 (1.5818) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [449] [ 50/312] eta: 0:03:52 lr: 0.000001 min_lr: 0.000001 loss: 1.5618 (1.5839) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [449] [ 60/312] eta: 0:03:35 lr: 0.000001 min_lr: 0.000001 loss: 1.5400 (1.5753) weight_decay: 0.0500 (0.0500) time: 0.6968 data: 0.0004 max mem: 64948 Epoch: [449] [ 70/312] eta: 0:03:21 lr: 0.000001 min_lr: 0.000001 loss: 1.6419 (1.5817) weight_decay: 0.0500 (0.0500) time: 0.6959 data: 0.0004 max mem: 64948 Epoch: [449] [ 80/312] eta: 0:03:09 lr: 0.000001 min_lr: 0.000001 loss: 1.6103 (1.5845) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [449] [ 90/312] eta: 0:02:58 lr: 0.000001 min_lr: 0.000001 loss: 1.6926 (1.5899) weight_decay: 0.0500 (0.0500) time: 0.6958 data: 0.0004 max mem: 64948 Epoch: [449] [100/312] eta: 0:02:47 lr: 0.000001 min_lr: 0.000001 loss: 1.6946 (1.6035) weight_decay: 0.0500 (0.0500) time: 0.6950 data: 0.0004 max mem: 64948 Epoch: [449] [110/312] eta: 0:02:38 lr: 0.000001 min_lr: 0.000001 loss: 1.6955 (1.6088) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [449] [120/312] eta: 0:02:29 lr: 0.000001 min_lr: 0.000001 loss: 1.7671 (1.6146) weight_decay: 0.0500 (0.0500) time: 0.6964 data: 0.0004 max mem: 64948 Epoch: [449] [130/312] eta: 0:02:20 lr: 0.000001 min_lr: 0.000001 loss: 1.7082 (1.6136) weight_decay: 0.0500 (0.0500) time: 0.6970 data: 0.0004 max mem: 64948 Epoch: [449] [140/312] eta: 0:02:11 lr: 0.000001 min_lr: 0.000001 loss: 1.6098 (1.6091) weight_decay: 0.0500 (0.0500) time: 0.6947 data: 0.0004 max mem: 64948 Epoch: [449] [150/312] eta: 0:02:03 lr: 0.000001 min_lr: 0.000001 loss: 1.5901 (1.6044) weight_decay: 0.0500 (0.0500) time: 0.6936 data: 0.0004 max mem: 64948 Epoch: [449] [160/312] eta: 0:01:54 lr: 0.000001 min_lr: 0.000001 loss: 1.6582 (1.6051) weight_decay: 0.0500 (0.0500) time: 0.6932 data: 0.0004 max mem: 64948 Epoch: [449] [170/312] eta: 0:01:46 lr: 0.000001 min_lr: 0.000001 loss: 1.7521 (1.6129) weight_decay: 0.0500 (0.0500) time: 0.6939 data: 0.0004 max mem: 64948 Epoch: [449] [180/312] eta: 0:01:38 lr: 0.000001 min_lr: 0.000001 loss: 1.7564 (1.6192) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [449] [190/312] eta: 0:01:31 lr: 0.000001 min_lr: 0.000001 loss: 1.8388 (1.6226) weight_decay: 0.0500 (0.0500) time: 0.6941 data: 0.0004 max mem: 64948 Epoch: [449] [200/312] eta: 0:01:23 lr: 0.000001 min_lr: 0.000001 loss: 1.6829 (1.6202) weight_decay: 0.0500 (0.0500) time: 0.6944 data: 0.0004 max mem: 64948 Epoch: [449] [210/312] eta: 0:01:15 lr: 0.000001 min_lr: 0.000001 loss: 1.5255 (1.6156) weight_decay: 0.0500 (0.0500) time: 0.6937 data: 0.0004 max mem: 64948 Epoch: [449] [220/312] eta: 0:01:07 lr: 0.000001 min_lr: 0.000001 loss: 1.6187 (1.6167) weight_decay: 0.0500 (0.0500) time: 0.6946 data: 0.0004 max mem: 64948 Epoch: [449] [230/312] eta: 0:01:00 lr: 0.000001 min_lr: 0.000001 loss: 1.6582 (1.6167) weight_decay: 0.0500 (0.0500) time: 0.6955 data: 0.0004 max mem: 64948 Epoch: [449] [240/312] eta: 0:00:52 lr: 0.000001 min_lr: 0.000001 loss: 1.7570 (1.6200) weight_decay: 0.0500 (0.0500) time: 0.6963 data: 0.0004 max mem: 64948 Epoch: [449] [250/312] eta: 0:00:45 lr: 0.000001 min_lr: 0.000001 loss: 1.6806 (1.6184) weight_decay: 0.0500 (0.0500) time: 0.6965 data: 0.0004 max mem: 64948 Epoch: [449] [260/312] eta: 0:00:38 lr: 0.000001 min_lr: 0.000001 loss: 1.5456 (1.6163) weight_decay: 0.0500 (0.0500) time: 0.6954 data: 0.0004 max mem: 64948 Epoch: [449] [270/312] eta: 0:00:30 lr: 0.000001 min_lr: 0.000001 loss: 1.6052 (1.6200) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0004 max mem: 64948 Epoch: [449] [280/312] eta: 0:00:23 lr: 0.000001 min_lr: 0.000001 loss: 1.6230 (1.6167) weight_decay: 0.0500 (0.0500) time: 0.6948 data: 0.0009 max mem: 64948 Epoch: [449] [290/312] eta: 0:00:16 lr: 0.000001 min_lr: 0.000001 loss: 1.4936 (1.6152) weight_decay: 0.0500 (0.0500) time: 0.6933 data: 0.0008 max mem: 64948 Epoch: [449] [300/312] eta: 0:00:08 lr: 0.000001 min_lr: 0.000001 loss: 1.6650 (1.6178) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [449] [310/312] eta: 0:00:01 lr: 0.000001 min_lr: 0.000001 loss: 1.7712 (1.6214) weight_decay: 0.0500 (0.0500) time: 0.6909 data: 0.0001 max mem: 64948 Epoch: [449] [311/312] eta: 0:00:00 lr: 0.000001 min_lr: 0.000001 loss: 1.7298 (1.6214) weight_decay: 0.0500 (0.0500) time: 0.6908 data: 0.0001 max mem: 64948 Epoch: [449] Total time: 0:03:47 (0.7290 s / it) Averaged stats: lr: 0.000001 min_lr: 0.000001 loss: 1.7298 (1.6467) weight_decay: 0.0500 (0.0500) Test: [0/9] eta: 0:00:41 loss: 0.4394 (0.4394) acc1: 89.0625 (89.0625) acc5: 98.1771 (98.1771) time: 4.6506 data: 4.4396 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6147 (0.6364) acc1: 84.6354 (83.0400) acc5: 97.3958 (96.8320) time: 0.6680 data: 0.4934 max mem: 64948 Test: Total time: 0:00:06 (0.6923 s / it) * Acc@1 83.788 Acc@5 96.746 loss 0.619 Accuracy of the model on the 50000 test images: 83.8% Max accuracy: 83.85% Test: [0/9] eta: 0:00:45 loss: 0.4425 (0.4425) acc1: 88.8021 (88.8021) acc5: 98.1771 (98.1771) time: 5.0166 data: 4.7986 max mem: 64948 Test: [8/9] eta: 0:00:00 loss: 0.6095 (0.6348) acc1: 84.6354 (83.1680) acc5: 97.3958 (96.7680) time: 0.7094 data: 0.5333 max mem: 64948 Test: Total time: 0:00:06 (0.7205 s / it) * Acc@1 83.828 Acc@5 96.742 loss 0.616 Accuracy of the model EMA on 50000 test images: 83.8% Max EMA accuracy: 83.83% Training time 1 day, 6:21:12