Namespace(aa='rand-m9-mstd0.5-inc1', auto_resume=True, batch_size=512, cfg_path='iFormer_m.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://', distributed=True, drop_path=0, enable_wandb=False, epochs=300, 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_m', model_ema=False, model_ema_decay=0.9999, model_ema_eval=False, 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, 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=8)
Transform = 
RandomResizedCropAndInterpolation(size=(224, 224), scale=(0.08, 1.0), ratio=(0.75, 1.3333), interpolation=PIL.Image.BICUBIC)
RandomHorizontalFlip(p=0.5)
<timm.data.auto_augment.RandAugment object at 0x7f6e8054c220>
ToTensor()
Normalize(mean=tensor([0.4850, 0.4560, 0.4060]), std=tensor([0.2290, 0.2240, 0.2250]))
<timm.data.random_erasing.RandomErasing object at 0x7f6e80544e20>
---------------------------
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 = <torch.utils.data.distributed.DistributedSampler object at 0x7f6e80544ac0>
Mixup is activated!
Model = iFormer(
  (downsample_layers): ModuleList(
    (0): Sequential(
      (0): Conv2d_BN(
        (c): Conv2d(3, 24, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), bias=False)
        (bn): BatchNorm2d(24, 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(24, 96, kernel_size=(5, 5), stride=(2, 2), padding=(2, 2), bias=False)
          (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
        (act): GELU(approximate='none')
        (conv_pwl_bn2): Conv2d_BN(
          (c): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
    )
    (1): Sequential(
      (0): Conv2d_BN(
        (c): Conv2d(48, 96, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (2): Sequential(
      (0): Conv2d_BN(
        (c): Conv2d(96, 192, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (3): Sequential(
      (0): Conv2d_BN(
        (c): Conv2d(192, 384, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
        (bn): BatchNorm2d(384, 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(48, 48, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=48, bias=False)
                (bn): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(48, 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(48, 48, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=48, bias=False)
                (bn): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(48, 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(96, 96, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=96, bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(96, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, 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(96, 96, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=96, bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(96, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (9): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=192, bias=False)
              (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            )
          )
        )
      )
      (10): BasicBlock(
        (block): SHMABlock(
          (token_channel_mixer): Residual(
            (m): GAU2dv2(
              (q): Conv2d_BN(
                (c): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(192, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, 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(192, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (11): BasicBlock(
        (block): FFN2d(
          (channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(192, 576, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(576, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (12): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=192, bias=False)
              (bn): BatchNorm2d(192, 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(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(192, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, 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(192, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 576, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(576, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (15): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=192, bias=False)
              (bn): BatchNorm2d(192, 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(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(192, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, 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(192, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 576, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(576, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (18): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=192, bias=False)
              (bn): BatchNorm2d(192, 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(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(192, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, 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(192, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(192, 576, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(576, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (21): BasicBlock(
        (block): ConvBlock(
          (token_channel_mixer): Residual(
            (m): Sequential(
              (0): Conv2d_BN(
                (c): Conv2d(192, 192, kernel_size=(7, 7), stride=(1, 1), padding=(3, 3), groups=192, bias=False)
                (bn): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): Conv2d_BN(
                (c): Conv2d(192, 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)
              )
              (2): GELU(approximate='none')
              (3): Conv2d_BN(
                (c): Conv2d(768, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(192, 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(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=384, bias=False)
              (bn): BatchNorm2d(384, 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(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(384, 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)
              )
              (gate_act): Sigmoid()
              (attn_drop): Dropout(p=0.0, inplace=False)
              (proj): Conv2d_BN(
                (c): Conv2d(384, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, 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(384, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(1152, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
      (3): BasicBlock(
        (block): RepCPE(
          (cpe): Residual(
            (m): Conv2d_BN(
              (c): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=384, bias=False)
              (bn): BatchNorm2d(384, 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(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (k): Conv2d_BN(
                (c): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (v_gate): Conv2d_BN(
                (c): Conv2d(384, 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)
              )
              (gate_act): Sigmoid()
              (attn_drop): Dropout(p=0.0, inplace=False)
              (proj): Conv2d_BN(
                (c): Conv2d(384, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, 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(384, 1152, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(1152, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
              (1): GELU(approximate='none')
              (2): Conv2d_BN(
                (c): Conv2d(1152, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)
                (bn): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
              )
            )
          )
        )
      )
    )
  )
  (classifier): Classfier(
    (classifier): BN_Linear(
      (bn): BatchNorm1d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (l): Linear(in_features=384, out_features=1000, bias=True)
    )
  )
)
number of params: 8907424
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.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.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.cpe.m.c.weight",
      "stages.2.10.block.token_channel_mixer.m.q.c.weight",
      "stages.2.10.block.token_channel_mixer.m.k.c.weight",
      "stages.2.10.block.token_channel_mixer.m.v_gate.c.weight",
      "stages.2.10.block.token_channel_mixer.m.proj.c.weight",
      "stages.2.11.block.channel_mixer.m.0.c.weight",
      "stages.2.11.block.channel_mixer.m.2.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.v_gate.c.weight",
      "stages.2.13.block.token_channel_mixer.m.proj.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.v_gate.c.weight",
      "stages.2.16.block.token_channel_mixer.m.proj.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.v_gate.c.weight",
      "stages.2.19.block.token_channel_mixer.m.proj.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.token_channel_mixer.m.0.c.weight",
      "stages.2.21.block.token_channel_mixer.m.1.c.weight",
      "stages.2.21.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.v_gate.c.weight",
      "stages.3.1.block.token_channel_mixer.m.proj.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.v_gate.c.weight",
      "stages.3.4.block.token_channel_mixer.m.proj.c.weight",
      "stages.3.5.block.channel_mixer.m.0.c.weight",
      "stages.3.5.block.channel_mixer.m.2.c.weight",
      "classifier.classifier.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.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.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.cpe.m.bn.weight",
      "stages.2.9.block.cpe.m.bn.bias",
      "stages.2.10.block.token_channel_mixer.m.q.bn.weight",
      "stages.2.10.block.token_channel_mixer.m.q.bn.bias",
      "stages.2.10.block.token_channel_mixer.m.k.bn.weight",
      "stages.2.10.block.token_channel_mixer.m.k.bn.bias",
      "stages.2.10.block.token_channel_mixer.m.v_gate.bn.weight",
      "stages.2.10.block.token_channel_mixer.m.v_gate.bn.bias",
      "stages.2.10.block.token_channel_mixer.m.proj.bn.weight",
      "stages.2.10.block.token_channel_mixer.m.proj.bn.bias",
      "stages.2.11.block.channel_mixer.m.0.bn.weight",
      "stages.2.11.block.channel_mixer.m.0.bn.bias",
      "stages.2.11.block.channel_mixer.m.2.bn.weight",
      "stages.2.11.block.channel_mixer.m.2.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.v_gate.bn.weight",
      "stages.2.13.block.token_channel_mixer.m.v_gate.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.2.14.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.v_gate.bn.weight",
      "stages.2.16.block.token_channel_mixer.m.v_gate.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.2.17.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.v_gate.bn.weight",
      "stages.2.19.block.token_channel_mixer.m.v_gate.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.2.20.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.token_channel_mixer.m.0.bn.weight",
      "stages.2.21.block.token_channel_mixer.m.0.bn.bias",
      "stages.2.21.block.token_channel_mixer.m.1.bn.weight",
      "stages.2.21.block.token_channel_mixer.m.1.bn.bias",
      "stages.2.21.block.token_channel_mixer.m.3.bn.weight",
      "stages.2.21.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.v_gate.bn.weight",
      "stages.3.1.block.token_channel_mixer.m.v_gate.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.3.2.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.v_gate.bn.weight",
      "stages.3.4.block.token_channel_mixer.m.v_gate.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.3.5.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",
      "classifier.classifier.bn.weight",
      "classifier.classifier.bn.bias",
      "classifier.classifier.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
criterion = SoftTargetCrossEntropy()
Auto resume checkpoint: 
Start training for 300 epochs
Epoch: [0]  [  0/312]  eta: 4:34:44  lr: 0.000000  min_lr: 0.000000  loss: 6.9577 (6.9577)  weight_decay: 0.0500 (0.0500)  time: 52.8344  data: 17.5547  max mem: 41794
Epoch: [0]  [ 10/312]  eta: 0:32:19  lr: 0.000006  min_lr: 0.000006  loss: 6.9755 (6.9768)  weight_decay: 0.0500 (0.0500)  time: 6.4215  data: 1.5969  max mem: 41794
Epoch: [0]  [ 20/312]  eta: 0:18:41  lr: 0.000013  min_lr: 0.000013  loss: 6.9672 (6.9706)  weight_decay: 0.0500 (0.0500)  time: 1.3926  data: 0.0013  max mem: 41794
Epoch: [0]  [ 30/312]  eta: 0:13:45  lr: 0.000019  min_lr: 0.000019  loss: 6.9627 (6.9676)  weight_decay: 0.0500 (0.0500)  time: 1.0051  data: 0.0014  max mem: 41794
Epoch: [0]  [ 40/312]  eta: 0:11:08  lr: 0.000026  min_lr: 0.000026  loss: 6.9519 (6.9605)  weight_decay: 0.0500 (0.0500)  time: 1.0060  data: 0.0013  max mem: 41794
Epoch: [0]  [ 50/312]  eta: 0:09:29  lr: 0.000032  min_lr: 0.000032  loss: 6.9230 (6.9524)  weight_decay: 0.0500 (0.0500)  time: 1.0059  data: 0.0012  max mem: 41794
Epoch: [0]  [ 60/312]  eta: 0:08:18  lr: 0.000038  min_lr: 0.000038  loss: 6.9139 (6.9454)  weight_decay: 0.0500 (0.0500)  time: 0.9976  data: 0.0012  max mem: 41794
Epoch: [0]  [ 70/312]  eta: 0:07:07  lr: 0.000045  min_lr: 0.000045  loss: 6.8988 (6.9367)  weight_decay: 0.0500 (0.0500)  time: 0.7311  data: 0.0009  max mem: 41794
Epoch: [0]  [ 80/312]  eta: 0:06:14  lr: 0.000051  min_lr: 0.000051  loss: 6.8783 (6.9289)  weight_decay: 0.0500 (0.0500)  time: 0.4929  data: 0.0007  max mem: 41794
Epoch: [0]  [ 90/312]  eta: 0:05:31  lr: 0.000058  min_lr: 0.000058  loss: 6.8721 (6.9221)  weight_decay: 0.0500 (0.0500)  time: 0.5213  data: 0.0078  max mem: 41794
Epoch: [0]  [100/312]  eta: 0:04:56  lr: 0.000064  min_lr: 0.000064  loss: 6.8637 (6.9162)  weight_decay: 0.0500 (0.0500)  time: 0.5326  data: 0.0138  max mem: 41794
Epoch: [0]  [110/312]  eta: 0:04:26  lr: 0.000071  min_lr: 0.000071  loss: 6.8509 (6.9094)  weight_decay: 0.0500 (0.0500)  time: 0.5294  data: 0.0088  max mem: 41794
Epoch: [0]  [120/312]  eta: 0:04:00  lr: 0.000077  min_lr: 0.000077  loss: 6.8419 (6.9033)  weight_decay: 0.0500 (0.0500)  time: 0.5222  data: 0.0027  max mem: 41794
Epoch: [0]  [130/312]  eta: 0:03:38  lr: 0.000083  min_lr: 0.000083  loss: 6.8165 (6.8964)  weight_decay: 0.0500 (0.0500)  time: 0.5514  data: 0.0028  max mem: 41794
Epoch: [0]  [140/312]  eta: 0:03:23  lr: 0.000090  min_lr: 0.000090  loss: 6.8032 (6.8898)  weight_decay: 0.0500 (0.0500)  time: 0.7572  data: 0.0105  max mem: 41794
Epoch: [0]  [150/312]  eta: 0:03:05  lr: 0.000096  min_lr: 0.000096  loss: 6.7843 (6.8822)  weight_decay: 0.0500 (0.0500)  time: 0.7599  data: 0.0278  max mem: 41794
Epoch: [0]  [160/312]  eta: 0:02:51  lr: 0.000103  min_lr: 0.000103  loss: 6.7752 (6.8761)  weight_decay: 0.0500 (0.0500)  time: 0.7595  data: 0.0259  max mem: 41794
Epoch: [0]  [170/312]  eta: 0:02:37  lr: 0.000109  min_lr: 0.000109  loss: 6.7771 (6.8697)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.0191  max mem: 41794
Epoch: [0]  [180/312]  eta: 0:02:22  lr: 0.000115  min_lr: 0.000115  loss: 6.7451 (6.8617)  weight_decay: 0.0500 (0.0500)  time: 0.6984  data: 0.0370  max mem: 41794
Epoch: [0]  [190/312]  eta: 0:02:10  lr: 0.000122  min_lr: 0.000122  loss: 6.7139 (6.8551)  weight_decay: 0.0500 (0.0500)  time: 0.7406  data: 0.0363  max mem: 41794
Epoch: [0]  [200/312]  eta: 0:01:57  lr: 0.000128  min_lr: 0.000128  loss: 6.7029 (6.8472)  weight_decay: 0.0500 (0.0500)  time: 0.7631  data: 0.0163  max mem: 41794
Epoch: [0]  [210/312]  eta: 0:01:45  lr: 0.000135  min_lr: 0.000135  loss: 6.6868 (6.8397)  weight_decay: 0.0500 (0.0500)  time: 0.7174  data: 0.0241  max mem: 41794
Epoch: [0]  [220/312]  eta: 0:01:34  lr: 0.000141  min_lr: 0.000141  loss: 6.6815 (6.8327)  weight_decay: 0.0500 (0.0500)  time: 0.7948  data: 0.0766  max mem: 41794
Epoch: [0]  [230/312]  eta: 0:01:22  lr: 0.000147  min_lr: 0.000147  loss: 6.6893 (6.8259)  weight_decay: 0.0500 (0.0500)  time: 0.7402  data: 0.1043  max mem: 41794
Epoch: [0]  [240/312]  eta: 0:01:12  lr: 0.000154  min_lr: 0.000154  loss: 6.6821 (6.8189)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.1237  max mem: 41794
Epoch: [0]  [250/312]  eta: 0:01:01  lr: 0.000160  min_lr: 0.000160  loss: 6.6637 (6.8117)  weight_decay: 0.0500 (0.0500)  time: 0.7770  data: 0.1472  max mem: 41794
Epoch: [0]  [260/312]  eta: 0:00:50  lr: 0.000167  min_lr: 0.000167  loss: 6.6022 (6.8046)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.1296  max mem: 41794
Epoch: [0]  [270/312]  eta: 0:00:40  lr: 0.000173  min_lr: 0.000173  loss: 6.5700 (6.7969)  weight_decay: 0.0500 (0.0500)  time: 0.7506  data: 0.1718  max mem: 41794
Epoch: [0]  [280/312]  eta: 0:00:30  lr: 0.000180  min_lr: 0.000180  loss: 6.6878 (6.7921)  weight_decay: 0.0500 (0.0500)  time: 0.7559  data: 0.1360  max mem: 41794
Epoch: [0]  [290/312]  eta: 0:00:21  lr: 0.000186  min_lr: 0.000186  loss: 6.6878 (6.7849)  weight_decay: 0.0500 (0.0500)  time: 0.7324  data: 0.0816  max mem: 41794
Epoch: [0]  [300/312]  eta: 0:00:11  lr: 0.000192  min_lr: 0.000192  loss: 6.5341 (6.7773)  weight_decay: 0.0500 (0.0500)  time: 0.7584  data: 0.1023  max mem: 41794
Epoch: [0]  [310/312]  eta: 0:00:01  lr: 0.000199  min_lr: 0.000199  loss: 6.5735 (6.7718)  weight_decay: 0.0500 (0.0500)  time: 0.5801  data: 0.0431  max mem: 41794
Epoch: [0]  [311/312]  eta: 0:00:00  lr: 0.000199  min_lr: 0.000199  loss: 6.5764 (6.7715)  weight_decay: 0.0500 (0.0500)  time: 0.5791  data: 0.0431  max mem: 41794
Epoch: [0] Total time: 0:04:53 (0.9411 s / it)
Averaged stats: lr: 0.000199  min_lr: 0.000199  loss: 6.5764 (6.7763)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:02:28  loss: 5.8466 (5.8466)  acc1: 2.3438 (2.3438)  acc5: 8.9844 (8.9844)  time: 16.4640  data: 7.4667  max mem: 41794
Test:  [8/9]  eta: 0:00:02  loss: 6.0999 (5.9561)  acc1: 2.3438 (2.5920)  acc5: 8.4635 (9.2000)  time: 2.2482  data: 0.8297  max mem: 41794
Test: Total time: 0:00:20 (2.2607 s / it)
* Acc@1 2.540 Acc@5 9.088 loss 5.958
Accuracy of the model on the 50000 test images: 2.5%
Max accuracy: 2.54%
Epoch: [1]  [  0/312]  eta: 2:11:07  lr: 0.000200  min_lr: 0.000200  loss: 6.6977 (6.6977)  weight_decay: 0.0500 (0.0500)  time: 25.2153  data: 12.6378  max mem: 41794
Epoch: [1]  [ 10/312]  eta: 0:16:13  lr: 0.000206  min_lr: 0.000206  loss: 6.6722 (6.6523)  weight_decay: 0.0500 (0.0500)  time: 3.2241  data: 1.1494  max mem: 41794
Epoch: [1]  [ 20/312]  eta: 0:10:35  lr: 0.000213  min_lr: 0.000213  loss: 6.6372 (6.6084)  weight_decay: 0.0500 (0.0500)  time: 1.0232  data: 0.0006  max mem: 41794
Epoch: [1]  [ 30/312]  eta: 0:08:28  lr: 0.000219  min_lr: 0.000219  loss: 6.5643 (6.5971)  weight_decay: 0.0500 (0.0500)  time: 1.0196  data: 0.0006  max mem: 41794
Epoch: [1]  [ 40/312]  eta: 0:06:54  lr: 0.000226  min_lr: 0.000226  loss: 6.5632 (6.5867)  weight_decay: 0.0500 (0.0500)  time: 0.8418  data: 0.0006  max mem: 41794
Epoch: [1]  [ 50/312]  eta: 0:05:46  lr: 0.000232  min_lr: 0.000232  loss: 6.5329 (6.5768)  weight_decay: 0.0500 (0.0500)  time: 0.5810  data: 0.0006  max mem: 41794
Epoch: [1]  [ 60/312]  eta: 0:05:00  lr: 0.000238  min_lr: 0.000238  loss: 6.5224 (6.5680)  weight_decay: 0.0500 (0.0500)  time: 0.5081  data: 0.0017  max mem: 41794
Epoch: [1]  [ 70/312]  eta: 0:04:25  lr: 0.000245  min_lr: 0.000245  loss: 6.5224 (6.5585)  weight_decay: 0.0500 (0.0500)  time: 0.5209  data: 0.0050  max mem: 41794
Epoch: [1]  [ 80/312]  eta: 0:03:57  lr: 0.000251  min_lr: 0.000251  loss: 6.4945 (6.5477)  weight_decay: 0.0500 (0.0500)  time: 0.5149  data: 0.0054  max mem: 41794
Epoch: [1]  [ 90/312]  eta: 0:03:34  lr: 0.000258  min_lr: 0.000258  loss: 6.5184 (6.5499)  weight_decay: 0.0500 (0.0500)  time: 0.5109  data: 0.0021  max mem: 41794
Epoch: [1]  [100/312]  eta: 0:03:15  lr: 0.000264  min_lr: 0.000264  loss: 6.5376 (6.5494)  weight_decay: 0.0500 (0.0500)  time: 0.5177  data: 0.0036  max mem: 41794
Epoch: [1]  [110/312]  eta: 0:03:03  lr: 0.000271  min_lr: 0.000271  loss: 6.5245 (6.5378)  weight_decay: 0.0500 (0.0500)  time: 0.6483  data: 0.0399  max mem: 41794
Epoch: [1]  [120/312]  eta: 0:02:51  lr: 0.000277  min_lr: 0.000277  loss: 6.5345 (6.5412)  weight_decay: 0.0500 (0.0500)  time: 0.7370  data: 0.0425  max mem: 41794
Epoch: [1]  [130/312]  eta: 0:02:40  lr: 0.000283  min_lr: 0.000283  loss: 6.5695 (6.5379)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.0489  max mem: 41794
Epoch: [1]  [140/312]  eta: 0:02:29  lr: 0.000290  min_lr: 0.000290  loss: 6.4620 (6.5276)  weight_decay: 0.0500 (0.0500)  time: 0.7160  data: 0.0582  max mem: 41794
Epoch: [1]  [150/312]  eta: 0:02:19  lr: 0.000296  min_lr: 0.000296  loss: 6.3561 (6.5160)  weight_decay: 0.0500 (0.0500)  time: 0.7414  data: 0.0361  max mem: 41794
Epoch: [1]  [160/312]  eta: 0:02:09  lr: 0.000303  min_lr: 0.000303  loss: 6.3548 (6.5067)  weight_decay: 0.0500 (0.0500)  time: 0.7566  data: 0.0571  max mem: 41794
Epoch: [1]  [170/312]  eta: 0:01:59  lr: 0.000309  min_lr: 0.000309  loss: 6.4033 (6.5048)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.0509  max mem: 41794
Epoch: [1]  [180/312]  eta: 0:01:50  lr: 0.000315  min_lr: 0.000315  loss: 6.4859 (6.5022)  weight_decay: 0.0500 (0.0500)  time: 0.7321  data: 0.0151  max mem: 41794
Epoch: [1]  [190/312]  eta: 0:01:41  lr: 0.000322  min_lr: 0.000322  loss: 6.4059 (6.4951)  weight_decay: 0.0500 (0.0500)  time: 0.7374  data: 0.0385  max mem: 41794
Epoch: [1]  [200/312]  eta: 0:01:32  lr: 0.000328  min_lr: 0.000328  loss: 6.4325 (6.4967)  weight_decay: 0.0500 (0.0500)  time: 0.7058  data: 0.0559  max mem: 41794
Epoch: [1]  [210/312]  eta: 0:01:23  lr: 0.000335  min_lr: 0.000335  loss: 6.4663 (6.4902)  weight_decay: 0.0500 (0.0500)  time: 0.7131  data: 0.0475  max mem: 41794
Epoch: [1]  [220/312]  eta: 0:01:15  lr: 0.000341  min_lr: 0.000341  loss: 6.4481 (6.4846)  weight_decay: 0.0500 (0.0500)  time: 0.7037  data: 0.0649  max mem: 41794
Epoch: [1]  [230/312]  eta: 0:01:06  lr: 0.000347  min_lr: 0.000347  loss: 6.4411 (6.4771)  weight_decay: 0.0500 (0.0500)  time: 0.7510  data: 0.0983  max mem: 41794
Epoch: [1]  [240/312]  eta: 0:00:58  lr: 0.000354  min_lr: 0.000354  loss: 6.4102 (6.4723)  weight_decay: 0.0500 (0.0500)  time: 0.7667  data: 0.1258  max mem: 41794
Epoch: [1]  [250/312]  eta: 0:00:50  lr: 0.000360  min_lr: 0.000360  loss: 6.4492 (6.4708)  weight_decay: 0.0500 (0.0500)  time: 0.7307  data: 0.0848  max mem: 41794
Epoch: [1]  [260/312]  eta: 0:00:41  lr: 0.000367  min_lr: 0.000367  loss: 6.4460 (6.4668)  weight_decay: 0.0500 (0.0500)  time: 0.7000  data: 0.0607  max mem: 41794
Epoch: [1]  [270/312]  eta: 0:00:33  lr: 0.000373  min_lr: 0.000373  loss: 6.4326 (6.4652)  weight_decay: 0.0500 (0.0500)  time: 0.7005  data: 0.0421  max mem: 41794
Epoch: [1]  [280/312]  eta: 0:00:25  lr: 0.000380  min_lr: 0.000380  loss: 6.4201 (6.4608)  weight_decay: 0.0500 (0.0500)  time: 0.7375  data: 0.0439  max mem: 41794
Epoch: [1]  [290/312]  eta: 0:00:17  lr: 0.000386  min_lr: 0.000386  loss: 6.3172 (6.4540)  weight_decay: 0.0500 (0.0500)  time: 0.7194  data: 0.0755  max mem: 41794
Epoch: [1]  [300/312]  eta: 0:00:09  lr: 0.000392  min_lr: 0.000392  loss: 6.2877 (6.4497)  weight_decay: 0.0500 (0.0500)  time: 0.7004  data: 0.0597  max mem: 41794
Epoch: [1]  [310/312]  eta: 0:00:01  lr: 0.000399  min_lr: 0.000399  loss: 6.3610 (6.4480)  weight_decay: 0.0500 (0.0500)  time: 0.5893  data: 0.0248  max mem: 41794
Epoch: [1]  [311/312]  eta: 0:00:00  lr: 0.000399  min_lr: 0.000399  loss: 6.3742 (6.4478)  weight_decay: 0.0500 (0.0500)  time: 0.5397  data: 0.0248  max mem: 41794
Epoch: [1] Total time: 0:04:05 (0.7875 s / it)
Averaged stats: lr: 0.000399  min_lr: 0.000399  loss: 6.3742 (6.4372)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 5.2733 (5.2733)  acc1: 5.9896 (5.9896)  acc5: 21.2240 (21.2240)  time: 8.8629  data: 8.6558  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 5.2295 (5.2293)  acc1: 7.5521 (7.7440)  acc5: 21.2240 (20.8160)  time: 1.1345  data: 0.9618  max mem: 41794
Test: Total time: 0:00:10 (1.1720 s / it)
* Acc@1 7.574 Acc@5 20.618 loss 5.241
Accuracy of the model on the 50000 test images: 7.6%
Max accuracy: 7.57%
Epoch: [2]  [  0/312]  eta: 1:20:31  lr: 0.000400  min_lr: 0.000400  loss: 6.4295 (6.4295)  weight_decay: 0.0500 (0.0500)  time: 15.4846  data: 15.0094  max mem: 41794
Epoch: [2]  [ 10/312]  eta: 0:10:33  lr: 0.000406  min_lr: 0.000406  loss: 6.2025 (6.2939)  weight_decay: 0.0500 (0.0500)  time: 2.0993  data: 1.4297  max mem: 41794
Epoch: [2]  [ 20/312]  eta: 0:07:07  lr: 0.000413  min_lr: 0.000413  loss: 6.1819 (6.2315)  weight_decay: 0.0500 (0.0500)  time: 0.7644  data: 0.1459  max mem: 41794
Epoch: [2]  [ 30/312]  eta: 0:05:25  lr: 0.000419  min_lr: 0.000419  loss: 6.1629 (6.2346)  weight_decay: 0.0500 (0.0500)  time: 0.6349  data: 0.1124  max mem: 41794
Epoch: [2]  [ 40/312]  eta: 0:04:44  lr: 0.000426  min_lr: 0.000426  loss: 6.1081 (6.2021)  weight_decay: 0.0500 (0.0500)  time: 0.6092  data: 0.1063  max mem: 41794
Epoch: [2]  [ 50/312]  eta: 0:04:21  lr: 0.000432  min_lr: 0.000432  loss: 6.2782 (6.2336)  weight_decay: 0.0500 (0.0500)  time: 0.7521  data: 0.2456  max mem: 41794
Epoch: [2]  [ 60/312]  eta: 0:03:50  lr: 0.000439  min_lr: 0.000439  loss: 6.2640 (6.2237)  weight_decay: 0.0500 (0.0500)  time: 0.6382  data: 0.1420  max mem: 41794
Epoch: [2]  [ 70/312]  eta: 0:03:43  lr: 0.000445  min_lr: 0.000445  loss: 6.0978 (6.2119)  weight_decay: 0.0500 (0.0500)  time: 0.7429  data: 0.1621  max mem: 41794
Epoch: [2]  [ 80/312]  eta: 0:03:29  lr: 0.000451  min_lr: 0.000451  loss: 6.0621 (6.2029)  weight_decay: 0.0500 (0.0500)  time: 0.8658  data: 0.2547  max mem: 41794
Epoch: [2]  [ 90/312]  eta: 0:03:11  lr: 0.000458  min_lr: 0.000458  loss: 5.9930 (6.1909)  weight_decay: 0.0500 (0.0500)  time: 0.6412  data: 0.1051  max mem: 41794
Epoch: [2]  [100/312]  eta: 0:03:04  lr: 0.000464  min_lr: 0.000464  loss: 6.0816 (6.1834)  weight_decay: 0.0500 (0.0500)  time: 0.7372  data: 0.1625  max mem: 41794
Epoch: [2]  [110/312]  eta: 0:02:48  lr: 0.000471  min_lr: 0.000471  loss: 6.1636 (6.1871)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.1506  max mem: 41794
Epoch: [2]  [120/312]  eta: 0:02:39  lr: 0.000477  min_lr: 0.000477  loss: 6.3043 (6.1925)  weight_decay: 0.0500 (0.0500)  time: 0.6524  data: 0.1243  max mem: 41794
Epoch: [2]  [130/312]  eta: 0:02:33  lr: 0.000483  min_lr: 0.000483  loss: 6.3043 (6.1850)  weight_decay: 0.0500 (0.0500)  time: 0.8739  data: 0.2857  max mem: 41794
Epoch: [2]  [140/312]  eta: 0:02:20  lr: 0.000490  min_lr: 0.000490  loss: 6.2598 (6.1911)  weight_decay: 0.0500 (0.0500)  time: 0.7054  data: 0.1624  max mem: 41794
Epoch: [2]  [150/312]  eta: 0:02:12  lr: 0.000496  min_lr: 0.000496  loss: 6.3184 (6.1899)  weight_decay: 0.0500 (0.0500)  time: 0.6742  data: 0.1002  max mem: 41794
Epoch: [2]  [160/312]  eta: 0:02:04  lr: 0.000503  min_lr: 0.000503  loss: 6.1581 (6.1840)  weight_decay: 0.0500 (0.0500)  time: 0.8560  data: 0.1347  max mem: 41794
Epoch: [2]  [170/312]  eta: 0:01:54  lr: 0.000509  min_lr: 0.000509  loss: 6.1143 (6.1827)  weight_decay: 0.0500 (0.0500)  time: 0.7272  data: 0.1011  max mem: 41794
Epoch: [2]  [180/312]  eta: 0:01:46  lr: 0.000515  min_lr: 0.000515  loss: 6.0958 (6.1759)  weight_decay: 0.0500 (0.0500)  time: 0.6773  data: 0.1318  max mem: 41794
Epoch: [2]  [190/312]  eta: 0:01:36  lr: 0.000522  min_lr: 0.000522  loss: 5.9510 (6.1625)  weight_decay: 0.0500 (0.0500)  time: 0.6195  data: 0.0661  max mem: 41794
Epoch: [2]  [200/312]  eta: 0:01:29  lr: 0.000528  min_lr: 0.000528  loss: 5.8064 (6.1572)  weight_decay: 0.0500 (0.0500)  time: 0.7432  data: 0.1049  max mem: 41794
Epoch: [2]  [210/312]  eta: 0:01:21  lr: 0.000535  min_lr: 0.000535  loss: 5.9914 (6.1469)  weight_decay: 0.0500 (0.0500)  time: 0.9183  data: 0.1578  max mem: 41794
Epoch: [2]  [220/312]  eta: 0:01:12  lr: 0.000541  min_lr: 0.000541  loss: 5.9758 (6.1391)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.0536  max mem: 41794
Epoch: [2]  [230/312]  eta: 0:01:04  lr: 0.000548  min_lr: 0.000548  loss: 6.1354 (6.1396)  weight_decay: 0.0500 (0.0500)  time: 0.7021  data: 0.0663  max mem: 41794
Epoch: [2]  [240/312]  eta: 0:00:56  lr: 0.000554  min_lr: 0.000554  loss: 6.1212 (6.1340)  weight_decay: 0.0500 (0.0500)  time: 0.8403  data: 0.0727  max mem: 41794
Epoch: [2]  [250/312]  eta: 0:00:48  lr: 0.000560  min_lr: 0.000560  loss: 6.0412 (6.1318)  weight_decay: 0.0500 (0.0500)  time: 0.6990  data: 0.0648  max mem: 41794
Epoch: [2]  [260/312]  eta: 0:00:40  lr: 0.000567  min_lr: 0.000567  loss: 6.0277 (6.1239)  weight_decay: 0.0500 (0.0500)  time: 0.7561  data: 0.1133  max mem: 41794
Epoch: [2]  [270/312]  eta: 0:00:32  lr: 0.000573  min_lr: 0.000573  loss: 6.0209 (6.1194)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.0556  max mem: 41794
Epoch: [2]  [280/312]  eta: 0:00:24  lr: 0.000580  min_lr: 0.000580  loss: 6.1117 (6.1166)  weight_decay: 0.0500 (0.0500)  time: 0.6654  data: 0.0569  max mem: 41794
Epoch: [2]  [290/312]  eta: 0:00:17  lr: 0.000586  min_lr: 0.000586  loss: 6.1339 (6.1121)  weight_decay: 0.0500 (0.0500)  time: 0.7764  data: 0.0891  max mem: 41794
Epoch: [2]  [300/312]  eta: 0:00:09  lr: 0.000592  min_lr: 0.000592  loss: 6.2037 (6.1140)  weight_decay: 0.0500 (0.0500)  time: 0.5988  data: 0.0325  max mem: 41794
Epoch: [2]  [310/312]  eta: 0:00:01  lr: 0.000599  min_lr: 0.000599  loss: 6.1526 (6.1102)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0001  max mem: 41794
Epoch: [2]  [311/312]  eta: 0:00:00  lr: 0.000599  min_lr: 0.000599  loss: 6.1515 (6.1079)  weight_decay: 0.0500 (0.0500)  time: 0.4626  data: 0.0001  max mem: 41794
Epoch: [2] Total time: 0:03:58 (0.7638 s / it)
Averaged stats: lr: 0.000599  min_lr: 0.000599  loss: 6.1515 (6.1242)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 4.1671 (4.1671)  acc1: 18.2292 (18.2292)  acc5: 42.1875 (42.1875)  time: 8.9330  data: 8.7349  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 4.4695 (4.4148)  acc1: 16.1458 (15.1520)  acc5: 35.0260 (34.4320)  time: 1.1479  data: 0.9706  max mem: 41794
Test: Total time: 0:00:10 (1.1916 s / it)
* Acc@1 15.194 Acc@5 34.872 loss 4.397
Accuracy of the model on the 50000 test images: 15.2%
Max accuracy: 15.19%
Epoch: [3]  [  0/312]  eta: 1:18:54  lr: 0.000600  min_lr: 0.000600  loss: 6.2792 (6.2792)  weight_decay: 0.0500 (0.0500)  time: 15.1749  data: 14.7057  max mem: 41794
Epoch: [3]  [ 10/312]  eta: 0:10:01  lr: 0.000607  min_lr: 0.000607  loss: 6.0722 (5.8865)  weight_decay: 0.0500 (0.0500)  time: 1.9919  data: 1.3406  max mem: 41794
Epoch: [3]  [ 20/312]  eta: 0:06:55  lr: 0.000613  min_lr: 0.000613  loss: 5.8301 (5.8572)  weight_decay: 0.0500 (0.0500)  time: 0.7368  data: 0.0769  max mem: 41794
Epoch: [3]  [ 30/312]  eta: 0:05:18  lr: 0.000619  min_lr: 0.000619  loss: 5.9585 (5.9115)  weight_decay: 0.0500 (0.0500)  time: 0.6568  data: 0.0752  max mem: 41794
Epoch: [3]  [ 40/312]  eta: 0:04:56  lr: 0.000626  min_lr: 0.000626  loss: 6.1536 (5.9472)  weight_decay: 0.0500 (0.0500)  time: 0.7374  data: 0.1369  max mem: 41794
Epoch: [3]  [ 50/312]  eta: 0:04:32  lr: 0.000632  min_lr: 0.000632  loss: 6.1531 (5.9472)  weight_decay: 0.0500 (0.0500)  time: 0.8979  data: 0.2215  max mem: 41794
Epoch: [3]  [ 60/312]  eta: 0:04:00  lr: 0.000639  min_lr: 0.000639  loss: 6.0493 (5.9373)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.0852  max mem: 41794
Epoch: [3]  [ 70/312]  eta: 0:03:46  lr: 0.000645  min_lr: 0.000645  loss: 6.0249 (5.9375)  weight_decay: 0.0500 (0.0500)  time: 0.6740  data: 0.0581  max mem: 41794
Epoch: [3]  [ 80/312]  eta: 0:03:33  lr: 0.000651  min_lr: 0.000651  loss: 6.0530 (5.9498)  weight_decay: 0.0500 (0.0500)  time: 0.8135  data: 0.1152  max mem: 41794
Epoch: [3]  [ 90/312]  eta: 0:03:18  lr: 0.000658  min_lr: 0.000658  loss: 6.0059 (5.9278)  weight_decay: 0.0500 (0.0500)  time: 0.7397  data: 0.0956  max mem: 41794
Epoch: [3]  [100/312]  eta: 0:03:06  lr: 0.000664  min_lr: 0.000664  loss: 5.9248 (5.9248)  weight_decay: 0.0500 (0.0500)  time: 0.7026  data: 0.0607  max mem: 41794
Epoch: [3]  [110/312]  eta: 0:02:51  lr: 0.000671  min_lr: 0.000671  loss: 6.0801 (5.9262)  weight_decay: 0.0500 (0.0500)  time: 0.6569  data: 0.0516  max mem: 41794
Epoch: [3]  [120/312]  eta: 0:02:43  lr: 0.000677  min_lr: 0.000677  loss: 6.1228 (5.9348)  weight_decay: 0.0500 (0.0500)  time: 0.7234  data: 0.0973  max mem: 41794
Epoch: [3]  [130/312]  eta: 0:02:35  lr: 0.000683  min_lr: 0.000683  loss: 6.0452 (5.9313)  weight_decay: 0.0500 (0.0500)  time: 0.8647  data: 0.1440  max mem: 41794
Epoch: [3]  [140/312]  eta: 0:02:22  lr: 0.000690  min_lr: 0.000690  loss: 5.8518 (5.9189)  weight_decay: 0.0500 (0.0500)  time: 0.6804  data: 0.0760  max mem: 41794
Epoch: [3]  [150/312]  eta: 0:02:15  lr: 0.000696  min_lr: 0.000696  loss: 5.8102 (5.9140)  weight_decay: 0.0500 (0.0500)  time: 0.7255  data: 0.0587  max mem: 41794
Epoch: [3]  [160/312]  eta: 0:02:05  lr: 0.000703  min_lr: 0.000703  loss: 5.9986 (5.9098)  weight_decay: 0.0500 (0.0500)  time: 0.7969  data: 0.0609  max mem: 41794
Epoch: [3]  [170/312]  eta: 0:01:56  lr: 0.000709  min_lr: 0.000709  loss: 5.9835 (5.9072)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.0292  max mem: 41794
Epoch: [3]  [180/312]  eta: 0:01:47  lr: 0.000715  min_lr: 0.000715  loss: 6.0222 (5.9100)  weight_decay: 0.0500 (0.0500)  time: 0.7390  data: 0.0317  max mem: 41794
Epoch: [3]  [190/312]  eta: 0:01:37  lr: 0.000722  min_lr: 0.000722  loss: 6.0222 (5.9104)  weight_decay: 0.0500 (0.0500)  time: 0.6478  data: 0.0053  max mem: 41794
Epoch: [3]  [200/312]  eta: 0:01:29  lr: 0.000728  min_lr: 0.000728  loss: 5.7661 (5.9056)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.0044  max mem: 41794
Epoch: [3]  [210/312]  eta: 0:01:21  lr: 0.000735  min_lr: 0.000735  loss: 5.9866 (5.9048)  weight_decay: 0.0500 (0.0500)  time: 0.7828  data: 0.0064  max mem: 41794
Epoch: [3]  [220/312]  eta: 0:01:12  lr: 0.000741  min_lr: 0.000741  loss: 6.0106 (5.9013)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.0149  max mem: 41794
Epoch: [3]  [230/312]  eta: 0:01:04  lr: 0.000748  min_lr: 0.000748  loss: 5.8367 (5.8896)  weight_decay: 0.0500 (0.0500)  time: 0.7237  data: 0.0169  max mem: 41794
Epoch: [3]  [240/312]  eta: 0:00:56  lr: 0.000754  min_lr: 0.000754  loss: 5.9952 (5.8962)  weight_decay: 0.0500 (0.0500)  time: 0.7450  data: 0.0272  max mem: 41794
Epoch: [3]  [250/312]  eta: 0:00:48  lr: 0.000760  min_lr: 0.000760  loss: 6.0878 (5.9022)  weight_decay: 0.0500 (0.0500)  time: 0.6689  data: 0.0258  max mem: 41794
Epoch: [3]  [260/312]  eta: 0:00:40  lr: 0.000767  min_lr: 0.000767  loss: 5.8928 (5.8963)  weight_decay: 0.0500 (0.0500)  time: 0.7756  data: 0.0234  max mem: 41794
Epoch: [3]  [270/312]  eta: 0:00:32  lr: 0.000773  min_lr: 0.000773  loss: 5.8832 (5.8985)  weight_decay: 0.0500 (0.0500)  time: 0.6708  data: 0.0210  max mem: 41794
Epoch: [3]  [280/312]  eta: 0:00:24  lr: 0.000780  min_lr: 0.000780  loss: 5.9650 (5.8949)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.0289  max mem: 41794
Epoch: [3]  [290/312]  eta: 0:00:17  lr: 0.000786  min_lr: 0.000786  loss: 5.7445 (5.8866)  weight_decay: 0.0500 (0.0500)  time: 0.7805  data: 0.0382  max mem: 41794
Epoch: [3]  [300/312]  eta: 0:00:09  lr: 0.000792  min_lr: 0.000792  loss: 5.7108 (5.8806)  weight_decay: 0.0500 (0.0500)  time: 0.5881  data: 0.0101  max mem: 41794
Epoch: [3]  [310/312]  eta: 0:00:01  lr: 0.000799  min_lr: 0.000799  loss: 5.7856 (5.8789)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [3]  [311/312]  eta: 0:00:00  lr: 0.000799  min_lr: 0.000799  loss: 5.7856 (5.8775)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.0001  max mem: 41794
Epoch: [3] Total time: 0:03:57 (0.7603 s / it)
Averaged stats: lr: 0.000799  min_lr: 0.000799  loss: 5.7856 (5.8632)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 3.7247 (3.7247)  acc1: 26.3021 (26.3021)  acc5: 53.9062 (53.9062)  time: 8.5181  data: 8.3166  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 3.9189 (3.8752)  acc1: 22.9167 (23.2800)  acc5: 47.5260 (46.8640)  time: 1.1007  data: 0.9242  max mem: 41794
Test: Total time: 0:00:10 (1.1237 s / it)
* Acc@1 23.126 Acc@5 46.604 loss 3.871
Accuracy of the model on the 50000 test images: 23.1%
Max accuracy: 23.13%
Epoch: [4]  [  0/312]  eta: 1:19:13  lr: 0.000800  min_lr: 0.000800  loss: 4.8948 (4.8948)  weight_decay: 0.0500 (0.0500)  time: 15.2363  data: 11.8494  max mem: 41794
Epoch: [4]  [ 10/312]  eta: 0:11:56  lr: 0.000807  min_lr: 0.000807  loss: 5.8592 (5.6511)  weight_decay: 0.0500 (0.0500)  time: 2.3738  data: 1.2537  max mem: 41794
Epoch: [4]  [ 20/312]  eta: 0:07:47  lr: 0.000813  min_lr: 0.000813  loss: 5.9101 (5.7730)  weight_decay: 0.0500 (0.0500)  time: 0.9200  data: 0.2037  max mem: 41794
Epoch: [4]  [ 30/312]  eta: 0:05:53  lr: 0.000819  min_lr: 0.000819  loss: 5.9760 (5.7782)  weight_decay: 0.0500 (0.0500)  time: 0.6381  data: 0.1069  max mem: 41794
Epoch: [4]  [ 40/312]  eta: 0:04:51  lr: 0.000826  min_lr: 0.000826  loss: 5.9360 (5.8045)  weight_decay: 0.0500 (0.0500)  time: 0.5181  data: 0.0022  max mem: 41794
Epoch: [4]  [ 50/312]  eta: 0:04:13  lr: 0.000832  min_lr: 0.000832  loss: 5.8515 (5.7876)  weight_decay: 0.0500 (0.0500)  time: 0.5198  data: 0.0082  max mem: 41794
Epoch: [4]  [ 60/312]  eta: 0:03:45  lr: 0.000839  min_lr: 0.000839  loss: 5.7154 (5.7645)  weight_decay: 0.0500 (0.0500)  time: 0.5235  data: 0.0120  max mem: 41794
Epoch: [4]  [ 70/312]  eta: 0:03:26  lr: 0.000845  min_lr: 0.000845  loss: 5.5200 (5.7270)  weight_decay: 0.0500 (0.0500)  time: 0.5625  data: 0.0369  max mem: 41794
Epoch: [4]  [ 80/312]  eta: 0:03:16  lr: 0.000851  min_lr: 0.000851  loss: 5.5958 (5.7203)  weight_decay: 0.0500 (0.0500)  time: 0.7152  data: 0.1519  max mem: 41794
Epoch: [4]  [ 90/312]  eta: 0:03:03  lr: 0.000858  min_lr: 0.000858  loss: 5.9235 (5.7197)  weight_decay: 0.0500 (0.0500)  time: 0.7306  data: 0.1613  max mem: 41794
Epoch: [4]  [100/312]  eta: 0:02:55  lr: 0.000864  min_lr: 0.000864  loss: 5.8454 (5.7210)  weight_decay: 0.0500 (0.0500)  time: 0.7478  data: 0.1348  max mem: 41794
Epoch: [4]  [110/312]  eta: 0:02:42  lr: 0.000871  min_lr: 0.000871  loss: 5.7925 (5.7315)  weight_decay: 0.0500 (0.0500)  time: 0.7193  data: 0.1039  max mem: 41794
Epoch: [4]  [120/312]  eta: 0:02:35  lr: 0.000877  min_lr: 0.000877  loss: 5.9399 (5.7437)  weight_decay: 0.0500 (0.0500)  time: 0.7155  data: 0.0941  max mem: 41794
Epoch: [4]  [130/312]  eta: 0:02:27  lr: 0.000883  min_lr: 0.000883  loss: 5.8850 (5.7459)  weight_decay: 0.0500 (0.0500)  time: 0.8277  data: 0.1255  max mem: 41794
Epoch: [4]  [140/312]  eta: 0:02:16  lr: 0.000890  min_lr: 0.000890  loss: 5.8850 (5.7478)  weight_decay: 0.0500 (0.0500)  time: 0.7117  data: 0.0883  max mem: 41794
Epoch: [4]  [150/312]  eta: 0:02:09  lr: 0.000896  min_lr: 0.000896  loss: 5.7325 (5.7392)  weight_decay: 0.0500 (0.0500)  time: 0.7368  data: 0.1288  max mem: 41794
Epoch: [4]  [160/312]  eta: 0:02:00  lr: 0.000903  min_lr: 0.000903  loss: 5.7013 (5.7412)  weight_decay: 0.0500 (0.0500)  time: 0.7717  data: 0.0928  max mem: 41794
Epoch: [4]  [170/312]  eta: 0:01:51  lr: 0.000909  min_lr: 0.000909  loss: 5.7179 (5.7343)  weight_decay: 0.0500 (0.0500)  time: 0.6656  data: 0.0772  max mem: 41794
Epoch: [4]  [180/312]  eta: 0:01:44  lr: 0.000916  min_lr: 0.000916  loss: 5.6518 (5.7242)  weight_decay: 0.0500 (0.0500)  time: 0.7485  data: 0.1396  max mem: 41794
Epoch: [4]  [190/312]  eta: 0:01:34  lr: 0.000922  min_lr: 0.000922  loss: 5.6518 (5.7186)  weight_decay: 0.0500 (0.0500)  time: 0.7181  data: 0.0754  max mem: 41794
Epoch: [4]  [200/312]  eta: 0:01:27  lr: 0.000928  min_lr: 0.000928  loss: 5.8127 (5.7159)  weight_decay: 0.0500 (0.0500)  time: 0.6962  data: 0.0732  max mem: 41794
Epoch: [4]  [210/312]  eta: 0:01:19  lr: 0.000935  min_lr: 0.000935  loss: 5.8432 (5.7145)  weight_decay: 0.0500 (0.0500)  time: 0.7900  data: 0.1041  max mem: 41794
Epoch: [4]  [220/312]  eta: 0:01:11  lr: 0.000941  min_lr: 0.000941  loss: 5.7918 (5.7133)  weight_decay: 0.0500 (0.0500)  time: 0.7241  data: 0.0779  max mem: 41794
Epoch: [4]  [230/312]  eta: 0:01:03  lr: 0.000948  min_lr: 0.000948  loss: 5.7426 (5.6990)  weight_decay: 0.0500 (0.0500)  time: 0.7293  data: 0.0883  max mem: 41794
Epoch: [4]  [240/312]  eta: 0:00:55  lr: 0.000954  min_lr: 0.000954  loss: 5.2772 (5.6800)  weight_decay: 0.0500 (0.0500)  time: 0.7354  data: 0.0441  max mem: 41794
Epoch: [4]  [250/312]  eta: 0:00:47  lr: 0.000960  min_lr: 0.000960  loss: 5.2772 (5.6667)  weight_decay: 0.0500 (0.0500)  time: 0.6964  data: 0.0484  max mem: 41794
Epoch: [4]  [260/312]  eta: 0:00:40  lr: 0.000967  min_lr: 0.000967  loss: 5.5031 (5.6712)  weight_decay: 0.0500 (0.0500)  time: 0.7688  data: 0.0949  max mem: 41794
Epoch: [4]  [270/312]  eta: 0:00:32  lr: 0.000973  min_lr: 0.000973  loss: 5.7720 (5.6693)  weight_decay: 0.0500 (0.0500)  time: 0.7118  data: 0.0479  max mem: 41794
Epoch: [4]  [280/312]  eta: 0:00:24  lr: 0.000980  min_lr: 0.000980  loss: 5.6128 (5.6653)  weight_decay: 0.0500 (0.0500)  time: 0.6787  data: 0.0274  max mem: 41794
Epoch: [4]  [290/312]  eta: 0:00:16  lr: 0.000986  min_lr: 0.000986  loss: 5.5656 (5.6619)  weight_decay: 0.0500 (0.0500)  time: 0.8156  data: 0.0690  max mem: 41794
Epoch: [4]  [300/312]  eta: 0:00:09  lr: 0.000992  min_lr: 0.000992  loss: 5.4569 (5.6519)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.0419  max mem: 41794
Epoch: [4]  [310/312]  eta: 0:00:01  lr: 0.000999  min_lr: 0.000999  loss: 5.5822 (5.6512)  weight_decay: 0.0500 (0.0500)  time: 0.4994  data: 0.0001  max mem: 41794
Epoch: [4]  [311/312]  eta: 0:00:00  lr: 0.001000  min_lr: 0.001000  loss: 5.4569 (5.6492)  weight_decay: 0.0500 (0.0500)  time: 0.4722  data: 0.0001  max mem: 41794
Epoch: [4] Total time: 0:03:55 (0.7560 s / it)
Averaged stats: lr: 0.001000  min_lr: 0.001000  loss: 5.4569 (5.6264)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 3.1915 (3.1915)  acc1: 35.0260 (35.0260)  acc5: 59.7656 (59.7656)  time: 8.3656  data: 8.1591  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 3.4222 (3.3899)  acc1: 27.6042 (29.1680)  acc5: 58.2031 (54.4800)  time: 1.1159  data: 0.9450  max mem: 41794
Test: Total time: 0:00:10 (1.1253 s / it)
* Acc@1 29.696 Acc@5 54.950 loss 3.374
Accuracy of the model on the 50000 test images: 29.7%
Max accuracy: 29.70%
Epoch: [5]  [  0/312]  eta: 1:13:20  lr: 0.001000  min_lr: 0.001000  loss: 4.6099 (4.6099)  weight_decay: 0.0500 (0.0500)  time: 14.1044  data: 12.5844  max mem: 41794
Epoch: [5]  [ 10/312]  eta: 0:09:44  lr: 0.001007  min_lr: 0.001007  loss: 5.2264 (5.2036)  weight_decay: 0.0500 (0.0500)  time: 1.9348  data: 1.2743  max mem: 41794
Epoch: [5]  [ 20/312]  eta: 0:07:13  lr: 0.001013  min_lr: 0.001013  loss: 5.2734 (5.3119)  weight_decay: 0.0500 (0.0500)  time: 0.8536  data: 0.2168  max mem: 41794
Epoch: [5]  [ 30/312]  eta: 0:05:33  lr: 0.001019  min_lr: 0.001019  loss: 5.6352 (5.4109)  weight_decay: 0.0500 (0.0500)  time: 0.7708  data: 0.1455  max mem: 41794
Epoch: [5]  [ 40/312]  eta: 0:05:07  lr: 0.001026  min_lr: 0.001026  loss: 5.6937 (5.4889)  weight_decay: 0.0500 (0.0500)  time: 0.7573  data: 0.1516  max mem: 41794
Epoch: [5]  [ 50/312]  eta: 0:04:48  lr: 0.001032  min_lr: 0.001032  loss: 5.6112 (5.5053)  weight_decay: 0.0500 (0.0500)  time: 0.9745  data: 0.2589  max mem: 41794
Epoch: [5]  [ 60/312]  eta: 0:04:14  lr: 0.001039  min_lr: 0.001039  loss: 5.4863 (5.4935)  weight_decay: 0.0500 (0.0500)  time: 0.7633  data: 0.1080  max mem: 41794
Epoch: [5]  [ 70/312]  eta: 0:03:54  lr: 0.001045  min_lr: 0.001045  loss: 5.5571 (5.5071)  weight_decay: 0.0500 (0.0500)  time: 0.6334  data: 0.0930  max mem: 41794
Epoch: [5]  [ 80/312]  eta: 0:03:34  lr: 0.001051  min_lr: 0.001051  loss: 5.5798 (5.5064)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.1323  max mem: 41794
Epoch: [5]  [ 90/312]  eta: 0:03:17  lr: 0.001058  min_lr: 0.001058  loss: 5.2653 (5.4600)  weight_decay: 0.0500 (0.0500)  time: 0.5988  data: 0.0977  max mem: 41794
Epoch: [5]  [100/312]  eta: 0:03:06  lr: 0.001064  min_lr: 0.001064  loss: 5.0276 (5.4271)  weight_decay: 0.0500 (0.0500)  time: 0.7019  data: 0.2046  max mem: 41794
Epoch: [5]  [110/312]  eta: 0:02:50  lr: 0.001071  min_lr: 0.001071  loss: 5.2973 (5.4302)  weight_decay: 0.0500 (0.0500)  time: 0.6464  data: 0.1470  max mem: 41794
Epoch: [5]  [120/312]  eta: 0:02:42  lr: 0.001077  min_lr: 0.001077  loss: 5.7219 (5.4286)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1740  max mem: 41794
Epoch: [5]  [130/312]  eta: 0:02:34  lr: 0.001084  min_lr: 0.001084  loss: 5.6444 (5.4306)  weight_decay: 0.0500 (0.0500)  time: 0.8636  data: 0.3594  max mem: 41794
Epoch: [5]  [140/312]  eta: 0:02:21  lr: 0.001090  min_lr: 0.001090  loss: 5.7386 (5.4544)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.1872  max mem: 41794
Epoch: [5]  [150/312]  eta: 0:02:14  lr: 0.001096  min_lr: 0.001096  loss: 5.5568 (5.4468)  weight_decay: 0.0500 (0.0500)  time: 0.6942  data: 0.1938  max mem: 41794
Epoch: [5]  [160/312]  eta: 0:02:06  lr: 0.001103  min_lr: 0.001103  loss: 5.3105 (5.4357)  weight_decay: 0.0500 (0.0500)  time: 0.8941  data: 0.3968  max mem: 41794
Epoch: [5]  [170/312]  eta: 0:01:55  lr: 0.001109  min_lr: 0.001109  loss: 5.7086 (5.4467)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.2048  max mem: 41794
Epoch: [5]  [180/312]  eta: 0:01:47  lr: 0.001116  min_lr: 0.001116  loss: 5.7256 (5.4420)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1987  max mem: 41794
Epoch: [5]  [190/312]  eta: 0:01:37  lr: 0.001122  min_lr: 0.001122  loss: 5.2130 (5.4339)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.1987  max mem: 41794
Epoch: [5]  [200/312]  eta: 0:01:29  lr: 0.001128  min_lr: 0.001128  loss: 5.2219 (5.4313)  weight_decay: 0.0500 (0.0500)  time: 0.6687  data: 0.1773  max mem: 41794
Epoch: [5]  [210/312]  eta: 0:01:21  lr: 0.001135  min_lr: 0.001135  loss: 5.4927 (5.4349)  weight_decay: 0.0500 (0.0500)  time: 0.8491  data: 0.3541  max mem: 41794
Epoch: [5]  [220/312]  eta: 0:01:12  lr: 0.001141  min_lr: 0.001141  loss: 5.6325 (5.4365)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1775  max mem: 41794
Epoch: [5]  [230/312]  eta: 0:01:05  lr: 0.001148  min_lr: 0.001148  loss: 5.5305 (5.4285)  weight_decay: 0.0500 (0.0500)  time: 0.6991  data: 0.2055  max mem: 41794
Epoch: [5]  [240/312]  eta: 0:00:57  lr: 0.001154  min_lr: 0.001154  loss: 5.5430 (5.4354)  weight_decay: 0.0500 (0.0500)  time: 0.8650  data: 0.3720  max mem: 41794
Epoch: [5]  [250/312]  eta: 0:00:48  lr: 0.001160  min_lr: 0.001160  loss: 5.5091 (5.4263)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1903  max mem: 41794
Epoch: [5]  [260/312]  eta: 0:00:40  lr: 0.001167  min_lr: 0.001167  loss: 5.3385 (5.4271)  weight_decay: 0.0500 (0.0500)  time: 0.6900  data: 0.1950  max mem: 41794
Epoch: [5]  [270/312]  eta: 0:00:32  lr: 0.001173  min_lr: 0.001173  loss: 5.5710 (5.4306)  weight_decay: 0.0500 (0.0500)  time: 0.6657  data: 0.1718  max mem: 41794
Epoch: [5]  [280/312]  eta: 0:00:25  lr: 0.001180  min_lr: 0.001180  loss: 5.5776 (5.4262)  weight_decay: 0.0500 (0.0500)  time: 0.7049  data: 0.2075  max mem: 41794
Epoch: [5]  [290/312]  eta: 0:00:17  lr: 0.001186  min_lr: 0.001186  loss: 5.3577 (5.4156)  weight_decay: 0.0500 (0.0500)  time: 0.8534  data: 0.3480  max mem: 41794
Epoch: [5]  [300/312]  eta: 0:00:09  lr: 0.001192  min_lr: 0.001192  loss: 5.2268 (5.4108)  weight_decay: 0.0500 (0.0500)  time: 0.6286  data: 0.1408  max mem: 41794
Epoch: [5]  [310/312]  eta: 0:00:01  lr: 0.001199  min_lr: 0.001199  loss: 5.3750 (5.4007)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [5]  [311/312]  eta: 0:00:00  lr: 0.001200  min_lr: 0.001200  loss: 5.3750 (5.3989)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [5] Total time: 0:03:59 (0.7672 s / it)
Averaged stats: lr: 0.001200  min_lr: 0.001200  loss: 5.3750 (5.4174)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 2.7259 (2.7259)  acc1: 44.0104 (44.0104)  acc5: 68.2292 (68.2292)  time: 7.9735  data: 7.7664  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 3.2810 (3.1177)  acc1: 34.8958 (35.2160)  acc5: 59.3750 (60.3040)  time: 1.0535  data: 0.8825  max mem: 41794
Test: Total time: 0:00:09 (1.0643 s / it)
* Acc@1 35.126 Acc@5 60.694 loss 3.117
Accuracy of the model on the 50000 test images: 35.1%
Max accuracy: 35.13%
Epoch: [6]  [  0/312]  eta: 1:18:31  lr: 0.001200  min_lr: 0.001200  loss: 5.8862 (5.8862)  weight_decay: 0.0500 (0.0500)  time: 15.1004  data: 13.8504  max mem: 41794
Epoch: [6]  [ 10/312]  eta: 0:11:53  lr: 0.001207  min_lr: 0.001207  loss: 5.6446 (5.4773)  weight_decay: 0.0500 (0.0500)  time: 2.3637  data: 1.3882  max mem: 41794
Epoch: [6]  [ 20/312]  eta: 0:07:31  lr: 0.001213  min_lr: 0.001213  loss: 5.4817 (5.3630)  weight_decay: 0.0500 (0.0500)  time: 0.8690  data: 0.0768  max mem: 41794
Epoch: [6]  [ 30/312]  eta: 0:05:44  lr: 0.001219  min_lr: 0.001219  loss: 5.1834 (5.3125)  weight_decay: 0.0500 (0.0500)  time: 0.5918  data: 0.0132  max mem: 41794
Epoch: [6]  [ 40/312]  eta: 0:04:44  lr: 0.001226  min_lr: 0.001226  loss: 5.2075 (5.3107)  weight_decay: 0.0500 (0.0500)  time: 0.5209  data: 0.0100  max mem: 41794
Epoch: [6]  [ 50/312]  eta: 0:04:08  lr: 0.001232  min_lr: 0.001232  loss: 5.0223 (5.2328)  weight_decay: 0.0500 (0.0500)  time: 0.5308  data: 0.0242  max mem: 41794
Epoch: [6]  [ 60/312]  eta: 0:03:41  lr: 0.001239  min_lr: 0.001239  loss: 4.9207 (5.2131)  weight_decay: 0.0500 (0.0500)  time: 0.5317  data: 0.0236  max mem: 41794
Epoch: [6]  [ 70/312]  eta: 0:03:31  lr: 0.001245  min_lr: 0.001245  loss: 5.2063 (5.2159)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.1783  max mem: 41794
Epoch: [6]  [ 80/312]  eta: 0:03:20  lr: 0.001251  min_lr: 0.001251  loss: 5.2865 (5.1969)  weight_decay: 0.0500 (0.0500)  time: 0.8255  data: 0.3230  max mem: 41794
Epoch: [6]  [ 90/312]  eta: 0:03:04  lr: 0.001258  min_lr: 0.001258  loss: 5.2365 (5.2030)  weight_decay: 0.0500 (0.0500)  time: 0.6766  data: 0.1502  max mem: 41794
Epoch: [6]  [100/312]  eta: 0:02:57  lr: 0.001264  min_lr: 0.001264  loss: 5.2365 (5.1873)  weight_decay: 0.0500 (0.0500)  time: 0.7308  data: 0.1437  max mem: 41794
Epoch: [6]  [110/312]  eta: 0:02:43  lr: 0.001271  min_lr: 0.001271  loss: 5.0485 (5.1815)  weight_decay: 0.0500 (0.0500)  time: 0.7004  data: 0.1416  max mem: 41794
Epoch: [6]  [120/312]  eta: 0:02:36  lr: 0.001277  min_lr: 0.001277  loss: 5.1068 (5.1812)  weight_decay: 0.0500 (0.0500)  time: 0.6828  data: 0.1175  max mem: 41794
Epoch: [6]  [130/312]  eta: 0:02:28  lr: 0.001284  min_lr: 0.001284  loss: 5.3259 (5.1818)  weight_decay: 0.0500 (0.0500)  time: 0.8676  data: 0.2597  max mem: 41794
Epoch: [6]  [140/312]  eta: 0:02:16  lr: 0.001290  min_lr: 0.001290  loss: 5.4348 (5.2017)  weight_decay: 0.0500 (0.0500)  time: 0.6921  data: 0.1440  max mem: 41794
Epoch: [6]  [150/312]  eta: 0:02:09  lr: 0.001296  min_lr: 0.001296  loss: 5.5973 (5.2084)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.1331  max mem: 41794
Epoch: [6]  [160/312]  eta: 0:02:01  lr: 0.001303  min_lr: 0.001303  loss: 5.4937 (5.2206)  weight_decay: 0.0500 (0.0500)  time: 0.8039  data: 0.2448  max mem: 41794
Epoch: [6]  [170/312]  eta: 0:01:52  lr: 0.001309  min_lr: 0.001309  loss: 5.4756 (5.2229)  weight_decay: 0.0500 (0.0500)  time: 0.7176  data: 0.1198  max mem: 41794
Epoch: [6]  [180/312]  eta: 0:01:44  lr: 0.001316  min_lr: 0.001316  loss: 5.2128 (5.2213)  weight_decay: 0.0500 (0.0500)  time: 0.7766  data: 0.0992  max mem: 41794
Epoch: [6]  [190/312]  eta: 0:01:35  lr: 0.001322  min_lr: 0.001322  loss: 5.0207 (5.2072)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.0952  max mem: 41794
Epoch: [6]  [200/312]  eta: 0:01:27  lr: 0.001328  min_lr: 0.001328  loss: 5.2613 (5.2182)  weight_decay: 0.0500 (0.0500)  time: 0.6646  data: 0.0532  max mem: 41794
Epoch: [6]  [210/312]  eta: 0:01:20  lr: 0.001335  min_lr: 0.001335  loss: 5.2313 (5.2062)  weight_decay: 0.0500 (0.0500)  time: 0.8632  data: 0.1314  max mem: 41794
Epoch: [6]  [220/312]  eta: 0:01:11  lr: 0.001341  min_lr: 0.001341  loss: 5.4374 (5.2210)  weight_decay: 0.0500 (0.0500)  time: 0.6982  data: 0.0823  max mem: 41794
Epoch: [6]  [230/312]  eta: 0:01:03  lr: 0.001348  min_lr: 0.001348  loss: 5.4423 (5.2152)  weight_decay: 0.0500 (0.0500)  time: 0.7073  data: 0.0843  max mem: 41794
Epoch: [6]  [240/312]  eta: 0:00:55  lr: 0.001354  min_lr: 0.001354  loss: 5.1499 (5.2129)  weight_decay: 0.0500 (0.0500)  time: 0.7740  data: 0.1034  max mem: 41794
Epoch: [6]  [250/312]  eta: 0:00:47  lr: 0.001360  min_lr: 0.001360  loss: 5.3308 (5.2138)  weight_decay: 0.0500 (0.0500)  time: 0.6860  data: 0.0502  max mem: 41794
Epoch: [6]  [260/312]  eta: 0:00:40  lr: 0.001367  min_lr: 0.001367  loss: 5.4413 (5.2197)  weight_decay: 0.0500 (0.0500)  time: 0.8539  data: 0.0867  max mem: 41794
Epoch: [6]  [270/312]  eta: 0:00:32  lr: 0.001373  min_lr: 0.001373  loss: 5.5233 (5.2230)  weight_decay: 0.0500 (0.0500)  time: 0.7353  data: 0.0576  max mem: 41794
Epoch: [6]  [280/312]  eta: 0:00:24  lr: 0.001380  min_lr: 0.001380  loss: 5.3341 (5.2211)  weight_decay: 0.0500 (0.0500)  time: 0.6696  data: 0.0287  max mem: 41794
Epoch: [6]  [290/312]  eta: 0:00:17  lr: 0.001386  min_lr: 0.001386  loss: 5.3478 (5.2233)  weight_decay: 0.0500 (0.0500)  time: 0.8402  data: 0.0344  max mem: 41794
Epoch: [6]  [300/312]  eta: 0:00:09  lr: 0.001393  min_lr: 0.001393  loss: 5.3478 (5.2214)  weight_decay: 0.0500 (0.0500)  time: 0.6619  data: 0.0074  max mem: 41794
Epoch: [6]  [310/312]  eta: 0:00:01  lr: 0.001399  min_lr: 0.001399  loss: 5.3162 (5.2203)  weight_decay: 0.0500 (0.0500)  time: 0.4711  data: 0.0001  max mem: 41794
Epoch: [6]  [311/312]  eta: 0:00:00  lr: 0.001400  min_lr: 0.001400  loss: 5.2066 (5.2181)  weight_decay: 0.0500 (0.0500)  time: 0.4689  data: 0.0001  max mem: 41794
Epoch: [6] Total time: 0:03:57 (0.7614 s / it)
Averaged stats: lr: 0.001400  min_lr: 0.001400  loss: 5.2066 (5.2247)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 2.5558 (2.5558)  acc1: 48.4375 (48.4375)  acc5: 72.9167 (72.9167)  time: 8.9199  data: 8.7076  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.9569 (2.8299)  acc1: 40.3646 (40.8160)  acc5: 66.4062 (65.9040)  time: 1.1391  data: 0.9676  max mem: 41794
Test: Total time: 0:00:10 (1.1558 s / it)
* Acc@1 40.448 Acc@5 66.644 loss 2.828
Accuracy of the model on the 50000 test images: 40.4%
Max accuracy: 40.45%
Epoch: [7]  [  0/312]  eta: 1:22:13  lr: 0.001400  min_lr: 0.001400  loss: 5.5426 (5.5426)  weight_decay: 0.0500 (0.0500)  time: 15.8128  data: 11.9726  max mem: 41794
Epoch: [7]  [ 10/312]  eta: 0:10:56  lr: 0.001407  min_lr: 0.001407  loss: 5.5310 (5.3599)  weight_decay: 0.0500 (0.0500)  time: 2.1727  data: 1.3068  max mem: 41794
Epoch: [7]  [ 20/312]  eta: 0:07:28  lr: 0.001413  min_lr: 0.001413  loss: 5.2830 (5.1315)  weight_decay: 0.0500 (0.0500)  time: 0.8221  data: 0.1576  max mem: 41794
Epoch: [7]  [ 30/312]  eta: 0:05:42  lr: 0.001419  min_lr: 0.001419  loss: 5.1256 (5.1385)  weight_decay: 0.0500 (0.0500)  time: 0.6865  data: 0.0438  max mem: 41794
Epoch: [7]  [ 40/312]  eta: 0:04:47  lr: 0.001426  min_lr: 0.001426  loss: 4.8993 (5.0511)  weight_decay: 0.0500 (0.0500)  time: 0.5523  data: 0.0361  max mem: 41794
Epoch: [7]  [ 50/312]  eta: 0:04:24  lr: 0.001432  min_lr: 0.001432  loss: 4.8993 (5.0382)  weight_decay: 0.0500 (0.0500)  time: 0.6939  data: 0.1918  max mem: 41794
Epoch: [7]  [ 60/312]  eta: 0:03:53  lr: 0.001439  min_lr: 0.001439  loss: 4.9101 (5.0298)  weight_decay: 0.0500 (0.0500)  time: 0.6602  data: 0.1624  max mem: 41794
Epoch: [7]  [ 70/312]  eta: 0:03:41  lr: 0.001445  min_lr: 0.001445  loss: 4.7816 (5.0234)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.1715  max mem: 41794
Epoch: [7]  [ 80/312]  eta: 0:03:29  lr: 0.001452  min_lr: 0.001452  loss: 5.1540 (5.0472)  weight_decay: 0.0500 (0.0500)  time: 0.8317  data: 0.3346  max mem: 41794
Epoch: [7]  [ 90/312]  eta: 0:03:11  lr: 0.001458  min_lr: 0.001458  loss: 5.2349 (5.0675)  weight_decay: 0.0500 (0.0500)  time: 0.6654  data: 0.1665  max mem: 41794
Epoch: [7]  [100/312]  eta: 0:03:03  lr: 0.001464  min_lr: 0.001464  loss: 5.1108 (5.0514)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1969  max mem: 41794
Epoch: [7]  [110/312]  eta: 0:02:47  lr: 0.001471  min_lr: 0.001471  loss: 5.0777 (5.0503)  weight_decay: 0.0500 (0.0500)  time: 0.6876  data: 0.1941  max mem: 41794
Epoch: [7]  [120/312]  eta: 0:02:40  lr: 0.001477  min_lr: 0.001477  loss: 5.0777 (5.0472)  weight_decay: 0.0500 (0.0500)  time: 0.6845  data: 0.1878  max mem: 41794
Epoch: [7]  [130/312]  eta: 0:02:31  lr: 0.001484  min_lr: 0.001484  loss: 5.1621 (5.0500)  weight_decay: 0.0500 (0.0500)  time: 0.8664  data: 0.3639  max mem: 41794
Epoch: [7]  [140/312]  eta: 0:02:19  lr: 0.001490  min_lr: 0.001490  loss: 5.2446 (5.0680)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.1768  max mem: 41794
Epoch: [7]  [150/312]  eta: 0:02:12  lr: 0.001496  min_lr: 0.001496  loss: 5.3933 (5.0784)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.2057  max mem: 41794
Epoch: [7]  [160/312]  eta: 0:02:04  lr: 0.001503  min_lr: 0.001503  loss: 5.1302 (5.0720)  weight_decay: 0.0500 (0.0500)  time: 0.8713  data: 0.3661  max mem: 41794
Epoch: [7]  [170/312]  eta: 0:01:53  lr: 0.001509  min_lr: 0.001509  loss: 5.0164 (5.0726)  weight_decay: 0.0500 (0.0500)  time: 0.6668  data: 0.1651  max mem: 41794
Epoch: [7]  [180/312]  eta: 0:01:46  lr: 0.001516  min_lr: 0.001516  loss: 5.0148 (5.0668)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.1622  max mem: 41794
Epoch: [7]  [190/312]  eta: 0:01:36  lr: 0.001522  min_lr: 0.001522  loss: 4.9611 (5.0599)  weight_decay: 0.0500 (0.0500)  time: 0.6774  data: 0.1599  max mem: 41794
Epoch: [7]  [200/312]  eta: 0:01:29  lr: 0.001528  min_lr: 0.001528  loss: 5.1746 (5.0657)  weight_decay: 0.0500 (0.0500)  time: 0.7398  data: 0.1762  max mem: 41794
Epoch: [7]  [210/312]  eta: 0:01:21  lr: 0.001535  min_lr: 0.001535  loss: 5.1310 (5.0546)  weight_decay: 0.0500 (0.0500)  time: 0.9149  data: 0.3037  max mem: 41794
Epoch: [7]  [220/312]  eta: 0:01:12  lr: 0.001541  min_lr: 0.001541  loss: 5.1835 (5.0624)  weight_decay: 0.0500 (0.0500)  time: 0.6654  data: 0.1300  max mem: 41794
Epoch: [7]  [230/312]  eta: 0:01:04  lr: 0.001548  min_lr: 0.001548  loss: 5.4716 (5.0617)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1389  max mem: 41794
Epoch: [7]  [240/312]  eta: 0:00:56  lr: 0.001554  min_lr: 0.001554  loss: 4.8551 (5.0545)  weight_decay: 0.0500 (0.0500)  time: 0.8414  data: 0.2406  max mem: 41794
Epoch: [7]  [250/312]  eta: 0:00:48  lr: 0.001561  min_lr: 0.001561  loss: 5.0398 (5.0561)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1299  max mem: 41794
Epoch: [7]  [260/312]  eta: 0:00:40  lr: 0.001567  min_lr: 0.001567  loss: 5.2000 (5.0601)  weight_decay: 0.0500 (0.0500)  time: 0.7427  data: 0.1952  max mem: 41794
Epoch: [7]  [270/312]  eta: 0:00:32  lr: 0.001573  min_lr: 0.001573  loss: 5.3752 (5.0707)  weight_decay: 0.0500 (0.0500)  time: 0.7214  data: 0.1692  max mem: 41794
Epoch: [7]  [280/312]  eta: 0:00:24  lr: 0.001580  min_lr: 0.001580  loss: 4.9640 (5.0572)  weight_decay: 0.0500 (0.0500)  time: 0.6485  data: 0.1498  max mem: 41794
Epoch: [7]  [290/312]  eta: 0:00:17  lr: 0.001586  min_lr: 0.001586  loss: 4.7075 (5.0556)  weight_decay: 0.0500 (0.0500)  time: 0.7955  data: 0.2926  max mem: 41794
Epoch: [7]  [300/312]  eta: 0:00:09  lr: 0.001593  min_lr: 0.001593  loss: 5.1911 (5.0540)  weight_decay: 0.0500 (0.0500)  time: 0.6346  data: 0.1448  max mem: 41794
Epoch: [7]  [310/312]  eta: 0:00:01  lr: 0.001599  min_lr: 0.001599  loss: 5.0196 (5.0555)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [7]  [311/312]  eta: 0:00:00  lr: 0.001600  min_lr: 0.001600  loss: 5.0196 (5.0538)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [7] Total time: 0:03:57 (0.7624 s / it)
Averaged stats: lr: 0.001600  min_lr: 0.001600  loss: 5.0196 (5.0712)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 2.3538 (2.3538)  acc1: 54.4271 (54.4271)  acc5: 75.6510 (75.6510)  time: 8.7707  data: 8.5643  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.7869 (2.6299)  acc1: 43.0990 (44.0960)  acc5: 68.0990 (70.2400)  time: 1.1239  data: 0.9517  max mem: 41794
Test: Total time: 0:00:10 (1.1573 s / it)
* Acc@1 44.630 Acc@5 70.922 loss 2.632
Accuracy of the model on the 50000 test images: 44.6%
Max accuracy: 44.63%
Epoch: [8]  [  0/312]  eta: 1:19:50  lr: 0.001600  min_lr: 0.001600  loss: 4.5474 (4.5474)  weight_decay: 0.0500 (0.0500)  time: 15.3547  data: 11.8552  max mem: 41794
Epoch: [8]  [ 10/312]  eta: 0:11:08  lr: 0.001607  min_lr: 0.001607  loss: 4.5878 (4.8453)  weight_decay: 0.0500 (0.0500)  time: 2.2138  data: 1.4245  max mem: 41794
Epoch: [8]  [ 20/312]  eta: 0:07:38  lr: 0.001613  min_lr: 0.001613  loss: 4.9910 (4.9954)  weight_decay: 0.0500 (0.0500)  time: 0.8808  data: 0.2800  max mem: 41794
Epoch: [8]  [ 30/312]  eta: 0:05:51  lr: 0.001619  min_lr: 0.001619  loss: 5.2524 (5.0586)  weight_decay: 0.0500 (0.0500)  time: 0.7156  data: 0.1058  max mem: 41794
Epoch: [8]  [ 40/312]  eta: 0:04:51  lr: 0.001626  min_lr: 0.001626  loss: 5.2524 (5.0700)  weight_decay: 0.0500 (0.0500)  time: 0.5498  data: 0.0237  max mem: 41794
Epoch: [8]  [ 50/312]  eta: 0:04:25  lr: 0.001632  min_lr: 0.001632  loss: 5.0005 (5.0400)  weight_decay: 0.0500 (0.0500)  time: 0.6516  data: 0.0743  max mem: 41794
Epoch: [8]  [ 60/312]  eta: 0:03:54  lr: 0.001639  min_lr: 0.001639  loss: 5.3263 (5.0907)  weight_decay: 0.0500 (0.0500)  time: 0.6371  data: 0.0713  max mem: 41794
Epoch: [8]  [ 70/312]  eta: 0:03:42  lr: 0.001645  min_lr: 0.001645  loss: 5.3344 (5.1070)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.0824  max mem: 41794
Epoch: [8]  [ 80/312]  eta: 0:03:27  lr: 0.001652  min_lr: 0.001652  loss: 5.1860 (5.0941)  weight_decay: 0.0500 (0.0500)  time: 0.7860  data: 0.0787  max mem: 41794
Epoch: [8]  [ 90/312]  eta: 0:03:12  lr: 0.001658  min_lr: 0.001658  loss: 5.0435 (5.0863)  weight_decay: 0.0500 (0.0500)  time: 0.6850  data: 0.0844  max mem: 41794
Epoch: [8]  [100/312]  eta: 0:03:04  lr: 0.001664  min_lr: 0.001664  loss: 4.9810 (5.0603)  weight_decay: 0.0500 (0.0500)  time: 0.7628  data: 0.1632  max mem: 41794
Epoch: [8]  [110/312]  eta: 0:02:48  lr: 0.001671  min_lr: 0.001671  loss: 4.9634 (5.0614)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.0795  max mem: 41794
Epoch: [8]  [120/312]  eta: 0:02:41  lr: 0.001677  min_lr: 0.001677  loss: 5.1612 (5.0444)  weight_decay: 0.0500 (0.0500)  time: 0.7083  data: 0.0730  max mem: 41794
Epoch: [8]  [130/312]  eta: 0:02:32  lr: 0.001684  min_lr: 0.001684  loss: 5.0478 (5.0468)  weight_decay: 0.0500 (0.0500)  time: 0.8654  data: 0.1205  max mem: 41794
Epoch: [8]  [140/312]  eta: 0:02:20  lr: 0.001690  min_lr: 0.001690  loss: 5.0478 (5.0456)  weight_decay: 0.0500 (0.0500)  time: 0.6506  data: 0.0508  max mem: 41794
Epoch: [8]  [150/312]  eta: 0:02:12  lr: 0.001696  min_lr: 0.001696  loss: 5.2808 (5.0576)  weight_decay: 0.0500 (0.0500)  time: 0.6593  data: 0.0595  max mem: 41794
Epoch: [8]  [160/312]  eta: 0:02:03  lr: 0.001703  min_lr: 0.001703  loss: 5.3642 (5.0440)  weight_decay: 0.0500 (0.0500)  time: 0.8215  data: 0.0981  max mem: 41794
Epoch: [8]  [170/312]  eta: 0:01:54  lr: 0.001709  min_lr: 0.001709  loss: 5.0416 (5.0415)  weight_decay: 0.0500 (0.0500)  time: 0.7152  data: 0.0955  max mem: 41794
Epoch: [8]  [180/312]  eta: 0:01:46  lr: 0.001716  min_lr: 0.001716  loss: 5.1561 (5.0420)  weight_decay: 0.0500 (0.0500)  time: 0.7541  data: 0.1245  max mem: 41794
Epoch: [8]  [190/312]  eta: 0:01:36  lr: 0.001722  min_lr: 0.001722  loss: 5.2948 (5.0448)  weight_decay: 0.0500 (0.0500)  time: 0.7003  data: 0.0709  max mem: 41794
Epoch: [8]  [200/312]  eta: 0:01:28  lr: 0.001728  min_lr: 0.001728  loss: 5.2945 (5.0469)  weight_decay: 0.0500 (0.0500)  time: 0.6562  data: 0.0508  max mem: 41794
Epoch: [8]  [210/312]  eta: 0:01:21  lr: 0.001735  min_lr: 0.001735  loss: 5.3426 (5.0622)  weight_decay: 0.0500 (0.0500)  time: 0.8222  data: 0.0772  max mem: 41794
Epoch: [8]  [220/312]  eta: 0:01:12  lr: 0.001741  min_lr: 0.001741  loss: 5.1310 (5.0566)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.0609  max mem: 41794
Epoch: [8]  [230/312]  eta: 0:01:04  lr: 0.001748  min_lr: 0.001748  loss: 5.0747 (5.0598)  weight_decay: 0.0500 (0.0500)  time: 0.6762  data: 0.0842  max mem: 41794
Epoch: [8]  [240/312]  eta: 0:00:56  lr: 0.001754  min_lr: 0.001754  loss: 5.1178 (5.0519)  weight_decay: 0.0500 (0.0500)  time: 0.7746  data: 0.0504  max mem: 41794
Epoch: [8]  [250/312]  eta: 0:00:48  lr: 0.001761  min_lr: 0.001761  loss: 4.7177 (5.0422)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.0636  max mem: 41794
Epoch: [8]  [260/312]  eta: 0:00:40  lr: 0.001767  min_lr: 0.001767  loss: 4.8786 (5.0330)  weight_decay: 0.0500 (0.0500)  time: 0.7199  data: 0.0923  max mem: 41794
Epoch: [8]  [270/312]  eta: 0:00:32  lr: 0.001773  min_lr: 0.001773  loss: 5.0298 (5.0338)  weight_decay: 0.0500 (0.0500)  time: 0.6496  data: 0.0334  max mem: 41794
Epoch: [8]  [280/312]  eta: 0:00:24  lr: 0.001780  min_lr: 0.001780  loss: 5.1835 (5.0326)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.0731  max mem: 41794
Epoch: [8]  [290/312]  eta: 0:00:17  lr: 0.001786  min_lr: 0.001786  loss: 4.5538 (5.0168)  weight_decay: 0.0500 (0.0500)  time: 0.8469  data: 0.0811  max mem: 41794
Epoch: [8]  [300/312]  eta: 0:00:09  lr: 0.001793  min_lr: 0.001793  loss: 4.5393 (5.0069)  weight_decay: 0.0500 (0.0500)  time: 0.6542  data: 0.0316  max mem: 41794
Epoch: [8]  [310/312]  eta: 0:00:01  lr: 0.001799  min_lr: 0.001799  loss: 4.9900 (5.0034)  weight_decay: 0.0500 (0.0500)  time: 0.4856  data: 0.0193  max mem: 41794
Epoch: [8]  [311/312]  eta: 0:00:00  lr: 0.001800  min_lr: 0.001800  loss: 4.9900 (5.0036)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0001  max mem: 41794
Epoch: [8] Total time: 0:03:57 (0.7616 s / it)
Averaged stats: lr: 0.001800  min_lr: 0.001800  loss: 4.9900 (4.9370)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:23  loss: 2.2245 (2.2245)  acc1: 53.3854 (53.3854)  acc5: 77.2135 (77.2135)  time: 9.2849  data: 9.0727  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.5370 (2.4307)  acc1: 48.8281 (49.1680)  acc5: 74.2188 (73.4560)  time: 1.1799  data: 1.0081  max mem: 41794
Test: Total time: 0:00:10 (1.1920 s / it)
* Acc@1 48.450 Acc@5 73.964 loss 2.431
Accuracy of the model on the 50000 test images: 48.5%
Max accuracy: 48.45%
Epoch: [9]  [  0/312]  eta: 1:22:47  lr: 0.001800  min_lr: 0.001800  loss: 5.1028 (5.1028)  weight_decay: 0.0500 (0.0500)  time: 15.9208  data: 13.3386  max mem: 41794
Epoch: [9]  [ 10/312]  eta: 0:11:46  lr: 0.001807  min_lr: 0.001807  loss: 5.1028 (4.9310)  weight_decay: 0.0500 (0.0500)  time: 2.3392  data: 1.3538  max mem: 41794
Epoch: [9]  [ 20/312]  eta: 0:07:22  lr: 0.001813  min_lr: 0.001813  loss: 5.0832 (4.8984)  weight_decay: 0.0500 (0.0500)  time: 0.7951  data: 0.0919  max mem: 41794
Epoch: [9]  [ 30/312]  eta: 0:05:37  lr: 0.001820  min_lr: 0.001820  loss: 4.7194 (4.8581)  weight_decay: 0.0500 (0.0500)  time: 0.5688  data: 0.0146  max mem: 41794
Epoch: [9]  [ 40/312]  eta: 0:04:39  lr: 0.001826  min_lr: 0.001826  loss: 4.6510 (4.8129)  weight_decay: 0.0500 (0.0500)  time: 0.5173  data: 0.0007  max mem: 41794
Epoch: [9]  [ 50/312]  eta: 0:04:09  lr: 0.001832  min_lr: 0.001832  loss: 4.3259 (4.7358)  weight_decay: 0.0500 (0.0500)  time: 0.5750  data: 0.0169  max mem: 41794
Epoch: [9]  [ 60/312]  eta: 0:03:41  lr: 0.001839  min_lr: 0.001839  loss: 4.4563 (4.7216)  weight_decay: 0.0500 (0.0500)  time: 0.5766  data: 0.0169  max mem: 41794
Epoch: [9]  [ 70/312]  eta: 0:03:36  lr: 0.001845  min_lr: 0.001845  loss: 4.6457 (4.7195)  weight_decay: 0.0500 (0.0500)  time: 0.7391  data: 0.1311  max mem: 41794
Epoch: [9]  [ 80/312]  eta: 0:03:22  lr: 0.001852  min_lr: 0.001852  loss: 4.5929 (4.7061)  weight_decay: 0.0500 (0.0500)  time: 0.8547  data: 0.1377  max mem: 41794
Epoch: [9]  [ 90/312]  eta: 0:03:10  lr: 0.001858  min_lr: 0.001858  loss: 5.0371 (4.7539)  weight_decay: 0.0500 (0.0500)  time: 0.7294  data: 0.1114  max mem: 41794
Epoch: [9]  [100/312]  eta: 0:03:01  lr: 0.001864  min_lr: 0.001864  loss: 5.0371 (4.7412)  weight_decay: 0.0500 (0.0500)  time: 0.7874  data: 0.1740  max mem: 41794
Epoch: [9]  [110/312]  eta: 0:02:46  lr: 0.001871  min_lr: 0.001871  loss: 4.8295 (4.7587)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.0727  max mem: 41794
Epoch: [9]  [120/312]  eta: 0:02:39  lr: 0.001877  min_lr: 0.001877  loss: 5.1568 (4.7755)  weight_decay: 0.0500 (0.0500)  time: 0.6986  data: 0.0874  max mem: 41794
Epoch: [9]  [130/312]  eta: 0:02:30  lr: 0.001884  min_lr: 0.001884  loss: 5.1584 (4.8002)  weight_decay: 0.0500 (0.0500)  time: 0.8572  data: 0.1468  max mem: 41794
Epoch: [9]  [140/312]  eta: 0:02:18  lr: 0.001890  min_lr: 0.001890  loss: 4.9329 (4.7881)  weight_decay: 0.0500 (0.0500)  time: 0.6583  data: 0.0629  max mem: 41794
Epoch: [9]  [150/312]  eta: 0:02:11  lr: 0.001896  min_lr: 0.001896  loss: 4.7607 (4.7827)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.0874  max mem: 41794
Epoch: [9]  [160/312]  eta: 0:02:02  lr: 0.001903  min_lr: 0.001903  loss: 4.8194 (4.7870)  weight_decay: 0.0500 (0.0500)  time: 0.7998  data: 0.0875  max mem: 41794
Epoch: [9]  [170/312]  eta: 0:01:53  lr: 0.001909  min_lr: 0.001909  loss: 5.0163 (4.8000)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.0852  max mem: 41794
Epoch: [9]  [180/312]  eta: 0:01:45  lr: 0.001916  min_lr: 0.001916  loss: 4.8379 (4.7898)  weight_decay: 0.0500 (0.0500)  time: 0.7634  data: 0.1543  max mem: 41794
Epoch: [9]  [190/312]  eta: 0:01:35  lr: 0.001922  min_lr: 0.001922  loss: 4.8360 (4.7937)  weight_decay: 0.0500 (0.0500)  time: 0.6721  data: 0.0713  max mem: 41794
Epoch: [9]  [200/312]  eta: 0:01:28  lr: 0.001929  min_lr: 0.001929  loss: 4.7780 (4.7928)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.0890  max mem: 41794
Epoch: [9]  [210/312]  eta: 0:01:20  lr: 0.001935  min_lr: 0.001935  loss: 4.9587 (4.8042)  weight_decay: 0.0500 (0.0500)  time: 0.8157  data: 0.1740  max mem: 41794
Epoch: [9]  [220/312]  eta: 0:01:11  lr: 0.001941  min_lr: 0.001941  loss: 4.8810 (4.7928)  weight_decay: 0.0500 (0.0500)  time: 0.6447  data: 0.0869  max mem: 41794
Epoch: [9]  [230/312]  eta: 0:01:03  lr: 0.001948  min_lr: 0.001948  loss: 4.9271 (4.8027)  weight_decay: 0.0500 (0.0500)  time: 0.6993  data: 0.1019  max mem: 41794
Epoch: [9]  [240/312]  eta: 0:00:56  lr: 0.001954  min_lr: 0.001954  loss: 5.0009 (4.7977)  weight_decay: 0.0500 (0.0500)  time: 0.8697  data: 0.1729  max mem: 41794
Epoch: [9]  [250/312]  eta: 0:00:47  lr: 0.001961  min_lr: 0.001961  loss: 4.9518 (4.7996)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.0874  max mem: 41794
Epoch: [9]  [260/312]  eta: 0:00:40  lr: 0.001967  min_lr: 0.001967  loss: 4.8416 (4.7909)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.0795  max mem: 41794
Epoch: [9]  [270/312]  eta: 0:00:32  lr: 0.001973  min_lr: 0.001973  loss: 4.8186 (4.7898)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.0637  max mem: 41794
Epoch: [9]  [280/312]  eta: 0:00:24  lr: 0.001980  min_lr: 0.001980  loss: 4.6476 (4.7783)  weight_decay: 0.0500 (0.0500)  time: 0.7186  data: 0.0942  max mem: 41794
Epoch: [9]  [290/312]  eta: 0:00:17  lr: 0.001986  min_lr: 0.001986  loss: 4.9382 (4.7859)  weight_decay: 0.0500 (0.0500)  time: 0.8851  data: 0.1616  max mem: 41794
Epoch: [9]  [300/312]  eta: 0:00:09  lr: 0.001993  min_lr: 0.001993  loss: 4.9871 (4.7819)  weight_decay: 0.0500 (0.0500)  time: 0.6652  data: 0.0678  max mem: 41794
Epoch: [9]  [310/312]  eta: 0:00:01  lr: 0.001999  min_lr: 0.001999  loss: 4.8110 (4.7861)  weight_decay: 0.0500 (0.0500)  time: 0.4739  data: 0.0001  max mem: 41794
Epoch: [9]  [311/312]  eta: 0:00:00  lr: 0.002000  min_lr: 0.002000  loss: 4.8110 (4.7873)  weight_decay: 0.0500 (0.0500)  time: 0.4717  data: 0.0001  max mem: 41794
Epoch: [9] Total time: 0:03:56 (0.7591 s / it)
Averaged stats: lr: 0.002000  min_lr: 0.002000  loss: 4.8110 (4.8312)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 2.1964 (2.1964)  acc1: 57.2917 (57.2917)  acc5: 79.4271 (79.4271)  time: 9.1354  data: 8.9238  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.5455 (2.4069)  acc1: 48.1771 (48.8320)  acc5: 73.8281 (74.2240)  time: 1.1634  data: 0.9916  max mem: 41794
Test: Total time: 0:00:10 (1.1764 s / it)
* Acc@1 49.236 Acc@5 74.898 loss 2.401
Accuracy of the model on the 50000 test images: 49.2%
Max accuracy: 49.24%
Epoch: [10]  [  0/312]  eta: 1:20:37  lr: 0.002000  min_lr: 0.002000  loss: 4.9837 (4.9837)  weight_decay: 0.0500 (0.0500)  time: 15.5049  data: 13.3481  max mem: 41794
Epoch: [10]  [ 10/312]  eta: 0:10:40  lr: 0.002007  min_lr: 0.002007  loss: 4.7871 (4.7666)  weight_decay: 0.0500 (0.0500)  time: 2.1204  data: 1.3738  max mem: 41794
Epoch: [10]  [ 20/312]  eta: 0:07:06  lr: 0.002013  min_lr: 0.002013  loss: 4.6170 (4.6209)  weight_decay: 0.0500 (0.0500)  time: 0.7582  data: 0.1962  max mem: 41794
Epoch: [10]  [ 30/312]  eta: 0:05:25  lr: 0.002020  min_lr: 0.002020  loss: 4.7079 (4.6634)  weight_decay: 0.0500 (0.0500)  time: 0.6227  data: 0.1095  max mem: 41794
Epoch: [10]  [ 40/312]  eta: 0:04:43  lr: 0.002026  min_lr: 0.002026  loss: 4.9625 (4.6794)  weight_decay: 0.0500 (0.0500)  time: 0.6003  data: 0.1029  max mem: 41794
Epoch: [10]  [ 50/312]  eta: 0:04:23  lr: 0.002032  min_lr: 0.002032  loss: 4.8648 (4.7139)  weight_decay: 0.0500 (0.0500)  time: 0.7724  data: 0.2562  max mem: 41794
Epoch: [10]  [ 60/312]  eta: 0:03:51  lr: 0.002039  min_lr: 0.002039  loss: 4.8648 (4.7358)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.1552  max mem: 41794
Epoch: [10]  [ 70/312]  eta: 0:03:41  lr: 0.002045  min_lr: 0.002045  loss: 4.8395 (4.7102)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1662  max mem: 41794
Epoch: [10]  [ 80/312]  eta: 0:03:28  lr: 0.002052  min_lr: 0.002052  loss: 4.9069 (4.7340)  weight_decay: 0.0500 (0.0500)  time: 0.8350  data: 0.3081  max mem: 41794
Epoch: [10]  [ 90/312]  eta: 0:03:10  lr: 0.002058  min_lr: 0.002058  loss: 5.0912 (4.7431)  weight_decay: 0.0500 (0.0500)  time: 0.6498  data: 0.1547  max mem: 41794
Epoch: [10]  [100/312]  eta: 0:03:03  lr: 0.002064  min_lr: 0.002064  loss: 4.9529 (4.7582)  weight_decay: 0.0500 (0.0500)  time: 0.7185  data: 0.2235  max mem: 41794
Epoch: [10]  [110/312]  eta: 0:02:47  lr: 0.002071  min_lr: 0.002071  loss: 4.8976 (4.7661)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.2114  max mem: 41794
Epoch: [10]  [120/312]  eta: 0:02:40  lr: 0.002077  min_lr: 0.002077  loss: 5.1279 (4.8032)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.2061  max mem: 41794
Epoch: [10]  [130/312]  eta: 0:02:33  lr: 0.002084  min_lr: 0.002084  loss: 5.1522 (4.7989)  weight_decay: 0.0500 (0.0500)  time: 0.9249  data: 0.3595  max mem: 41794
Epoch: [10]  [140/312]  eta: 0:02:21  lr: 0.002090  min_lr: 0.002090  loss: 4.5904 (4.7961)  weight_decay: 0.0500 (0.0500)  time: 0.7249  data: 0.1626  max mem: 41794
Epoch: [10]  [150/312]  eta: 0:02:13  lr: 0.002096  min_lr: 0.002096  loss: 4.7598 (4.7827)  weight_decay: 0.0500 (0.0500)  time: 0.7173  data: 0.1439  max mem: 41794
Epoch: [10]  [160/312]  eta: 0:02:04  lr: 0.002103  min_lr: 0.002103  loss: 4.4982 (4.7757)  weight_decay: 0.0500 (0.0500)  time: 0.8013  data: 0.2035  max mem: 41794
Epoch: [10]  [170/312]  eta: 0:01:54  lr: 0.002109  min_lr: 0.002109  loss: 4.6212 (4.7764)  weight_decay: 0.0500 (0.0500)  time: 0.6423  data: 0.1247  max mem: 41794
Epoch: [10]  [180/312]  eta: 0:01:47  lr: 0.002116  min_lr: 0.002116  loss: 4.7651 (4.7776)  weight_decay: 0.0500 (0.0500)  time: 0.7640  data: 0.2253  max mem: 41794
Epoch: [10]  [190/312]  eta: 0:01:37  lr: 0.002122  min_lr: 0.002122  loss: 5.1070 (4.7905)  weight_decay: 0.0500 (0.0500)  time: 0.7108  data: 0.1695  max mem: 41794
Epoch: [10]  [200/312]  eta: 0:01:29  lr: 0.002129  min_lr: 0.002129  loss: 5.1070 (4.7934)  weight_decay: 0.0500 (0.0500)  time: 0.6811  data: 0.1248  max mem: 41794
Epoch: [10]  [210/312]  eta: 0:01:21  lr: 0.002135  min_lr: 0.002135  loss: 4.8314 (4.7746)  weight_decay: 0.0500 (0.0500)  time: 0.8781  data: 0.2225  max mem: 41794
Epoch: [10]  [220/312]  eta: 0:01:12  lr: 0.002141  min_lr: 0.002141  loss: 4.2901 (4.7588)  weight_decay: 0.0500 (0.0500)  time: 0.7094  data: 0.1123  max mem: 41794
Epoch: [10]  [230/312]  eta: 0:01:05  lr: 0.002148  min_lr: 0.002148  loss: 4.8010 (4.7613)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.0917  max mem: 41794
Epoch: [10]  [240/312]  eta: 0:00:56  lr: 0.002154  min_lr: 0.002154  loss: 4.9199 (4.7622)  weight_decay: 0.0500 (0.0500)  time: 0.7736  data: 0.1182  max mem: 41794
Epoch: [10]  [250/312]  eta: 0:00:48  lr: 0.002161  min_lr: 0.002161  loss: 4.8609 (4.7605)  weight_decay: 0.0500 (0.0500)  time: 0.6275  data: 0.0870  max mem: 41794
Epoch: [10]  [260/312]  eta: 0:00:41  lr: 0.002167  min_lr: 0.002167  loss: 4.8853 (4.7609)  weight_decay: 0.0500 (0.0500)  time: 0.7800  data: 0.1740  max mem: 41794
Epoch: [10]  [270/312]  eta: 0:00:32  lr: 0.002173  min_lr: 0.002173  loss: 4.9622 (4.7662)  weight_decay: 0.0500 (0.0500)  time: 0.7341  data: 0.1301  max mem: 41794
Epoch: [10]  [280/312]  eta: 0:00:24  lr: 0.002180  min_lr: 0.002180  loss: 4.5892 (4.7548)  weight_decay: 0.0500 (0.0500)  time: 0.6245  data: 0.0955  max mem: 41794
Epoch: [10]  [290/312]  eta: 0:00:17  lr: 0.002186  min_lr: 0.002186  loss: 4.7987 (4.7613)  weight_decay: 0.0500 (0.0500)  time: 0.7953  data: 0.1907  max mem: 41794
Epoch: [10]  [300/312]  eta: 0:00:09  lr: 0.002193  min_lr: 0.002193  loss: 5.0766 (4.7583)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.0975  max mem: 41794
Epoch: [10]  [310/312]  eta: 0:00:01  lr: 0.002199  min_lr: 0.002199  loss: 4.7799 (4.7527)  weight_decay: 0.0500 (0.0500)  time: 0.4670  data: 0.0001  max mem: 41794
Epoch: [10]  [311/312]  eta: 0:00:00  lr: 0.002200  min_lr: 0.002200  loss: 4.7799 (4.7518)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [10] Total time: 0:03:58 (0.7638 s / it)
Averaged stats: lr: 0.002200  min_lr: 0.002200  loss: 4.7799 (4.7414)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 1.8793 (1.8793)  acc1: 62.5000 (62.5000)  acc5: 81.3802 (81.3802)  time: 9.1496  data: 8.9377  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.3294 (2.2013)  acc1: 50.9115 (52.4160)  acc5: 78.7760 (77.9680)  time: 1.1646  data: 0.9932  max mem: 41794
Test: Total time: 0:00:10 (1.1999 s / it)
* Acc@1 52.786 Acc@5 77.628 loss 2.197
Accuracy of the model on the 50000 test images: 52.8%
Max accuracy: 52.79%
Epoch: [11]  [  0/312]  eta: 1:21:12  lr: 0.002200  min_lr: 0.002200  loss: 5.2405 (5.2405)  weight_decay: 0.0500 (0.0500)  time: 15.6168  data: 13.9497  max mem: 41794
Epoch: [11]  [ 10/312]  eta: 0:11:23  lr: 0.002207  min_lr: 0.002207  loss: 4.4078 (4.5885)  weight_decay: 0.0500 (0.0500)  time: 2.2629  data: 1.2954  max mem: 41794
Epoch: [11]  [ 20/312]  eta: 0:07:15  lr: 0.002213  min_lr: 0.002213  loss: 4.5071 (4.6046)  weight_decay: 0.0500 (0.0500)  time: 0.7840  data: 0.0372  max mem: 41794
Epoch: [11]  [ 30/312]  eta: 0:05:31  lr: 0.002220  min_lr: 0.002220  loss: 4.9420 (4.6676)  weight_decay: 0.0500 (0.0500)  time: 0.5762  data: 0.0247  max mem: 41794
Epoch: [11]  [ 40/312]  eta: 0:04:35  lr: 0.002226  min_lr: 0.002226  loss: 4.4783 (4.5772)  weight_decay: 0.0500 (0.0500)  time: 0.5130  data: 0.0081  max mem: 41794
Epoch: [11]  [ 50/312]  eta: 0:04:12  lr: 0.002232  min_lr: 0.002232  loss: 4.5610 (4.5914)  weight_decay: 0.0500 (0.0500)  time: 0.6320  data: 0.1189  max mem: 41794
Epoch: [11]  [ 60/312]  eta: 0:03:43  lr: 0.002239  min_lr: 0.002239  loss: 4.5914 (4.6169)  weight_decay: 0.0500 (0.0500)  time: 0.6280  data: 0.1143  max mem: 41794
Epoch: [11]  [ 70/312]  eta: 0:03:36  lr: 0.002245  min_lr: 0.002245  loss: 4.6648 (4.6102)  weight_decay: 0.0500 (0.0500)  time: 0.7300  data: 0.2191  max mem: 41794
Epoch: [11]  [ 80/312]  eta: 0:03:28  lr: 0.002252  min_lr: 0.002252  loss: 4.6930 (4.6257)  weight_decay: 0.0500 (0.0500)  time: 0.9301  data: 0.4093  max mem: 41794
Epoch: [11]  [ 90/312]  eta: 0:03:10  lr: 0.002258  min_lr: 0.002258  loss: 4.5477 (4.6073)  weight_decay: 0.0500 (0.0500)  time: 0.7131  data: 0.1916  max mem: 41794
Epoch: [11]  [100/312]  eta: 0:03:01  lr: 0.002264  min_lr: 0.002264  loss: 4.5886 (4.6208)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.1524  max mem: 41794
Epoch: [11]  [110/312]  eta: 0:02:46  lr: 0.002271  min_lr: 0.002271  loss: 4.5886 (4.6026)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.1541  max mem: 41794
Epoch: [11]  [120/312]  eta: 0:02:39  lr: 0.002277  min_lr: 0.002277  loss: 4.4743 (4.6069)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.1596  max mem: 41794
Epoch: [11]  [130/312]  eta: 0:02:32  lr: 0.002284  min_lr: 0.002284  loss: 4.9433 (4.6328)  weight_decay: 0.0500 (0.0500)  time: 0.8906  data: 0.3348  max mem: 41794
Epoch: [11]  [140/312]  eta: 0:02:19  lr: 0.002290  min_lr: 0.002290  loss: 4.9711 (4.6384)  weight_decay: 0.0500 (0.0500)  time: 0.7101  data: 0.1776  max mem: 41794
Epoch: [11]  [150/312]  eta: 0:02:12  lr: 0.002297  min_lr: 0.002297  loss: 4.8628 (4.6526)  weight_decay: 0.0500 (0.0500)  time: 0.7045  data: 0.1225  max mem: 41794
Epoch: [11]  [160/312]  eta: 0:02:04  lr: 0.002303  min_lr: 0.002303  loss: 4.9026 (4.6596)  weight_decay: 0.0500 (0.0500)  time: 0.8691  data: 0.2316  max mem: 41794
Epoch: [11]  [170/312]  eta: 0:01:53  lr: 0.002309  min_lr: 0.002309  loss: 4.8014 (4.6460)  weight_decay: 0.0500 (0.0500)  time: 0.6777  data: 0.1111  max mem: 41794
Epoch: [11]  [180/312]  eta: 0:01:46  lr: 0.002316  min_lr: 0.002316  loss: 4.8861 (4.6605)  weight_decay: 0.0500 (0.0500)  time: 0.7151  data: 0.1206  max mem: 41794
Epoch: [11]  [190/312]  eta: 0:01:36  lr: 0.002322  min_lr: 0.002322  loss: 4.8861 (4.6654)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.1238  max mem: 41794
Epoch: [11]  [200/312]  eta: 0:01:29  lr: 0.002329  min_lr: 0.002329  loss: 4.8669 (4.6789)  weight_decay: 0.0500 (0.0500)  time: 0.7016  data: 0.1094  max mem: 41794
Epoch: [11]  [210/312]  eta: 0:01:21  lr: 0.002335  min_lr: 0.002335  loss: 4.7994 (4.6785)  weight_decay: 0.0500 (0.0500)  time: 0.8856  data: 0.1941  max mem: 41794
Epoch: [11]  [220/312]  eta: 0:01:12  lr: 0.002341  min_lr: 0.002341  loss: 4.7685 (4.6858)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.0913  max mem: 41794
Epoch: [11]  [230/312]  eta: 0:01:04  lr: 0.002348  min_lr: 0.002348  loss: 4.9049 (4.6820)  weight_decay: 0.0500 (0.0500)  time: 0.6929  data: 0.0788  max mem: 41794
Epoch: [11]  [240/312]  eta: 0:00:57  lr: 0.002354  min_lr: 0.002354  loss: 4.9049 (4.6849)  weight_decay: 0.0500 (0.0500)  time: 0.8704  data: 0.1171  max mem: 41794
Epoch: [11]  [250/312]  eta: 0:00:48  lr: 0.002361  min_lr: 0.002361  loss: 4.8224 (4.6853)  weight_decay: 0.0500 (0.0500)  time: 0.7265  data: 0.0806  max mem: 41794
Epoch: [11]  [260/312]  eta: 0:00:41  lr: 0.002367  min_lr: 0.002367  loss: 4.7640 (4.6832)  weight_decay: 0.0500 (0.0500)  time: 0.7444  data: 0.0997  max mem: 41794
Epoch: [11]  [270/312]  eta: 0:00:32  lr: 0.002373  min_lr: 0.002373  loss: 4.7400 (4.6801)  weight_decay: 0.0500 (0.0500)  time: 0.7070  data: 0.0607  max mem: 41794
Epoch: [11]  [280/312]  eta: 0:00:25  lr: 0.002380  min_lr: 0.002380  loss: 4.7212 (4.6756)  weight_decay: 0.0500 (0.0500)  time: 0.6895  data: 0.0419  max mem: 41794
Epoch: [11]  [290/312]  eta: 0:00:17  lr: 0.002386  min_lr: 0.002386  loss: 4.6870 (4.6820)  weight_decay: 0.0500 (0.0500)  time: 0.8469  data: 0.0777  max mem: 41794
Epoch: [11]  [300/312]  eta: 0:00:09  lr: 0.002393  min_lr: 0.002393  loss: 4.9232 (4.6844)  weight_decay: 0.0500 (0.0500)  time: 0.6485  data: 0.0374  max mem: 41794
Epoch: [11]  [310/312]  eta: 0:00:01  lr: 0.002399  min_lr: 0.002399  loss: 4.8261 (4.6885)  weight_decay: 0.0500 (0.0500)  time: 0.4675  data: 0.0001  max mem: 41794
Epoch: [11]  [311/312]  eta: 0:00:00  lr: 0.002400  min_lr: 0.002400  loss: 4.8261 (4.6886)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0001  max mem: 41794
Epoch: [11] Total time: 0:04:00 (0.7701 s / it)
Averaged stats: lr: 0.002400  min_lr: 0.002400  loss: 4.8261 (4.6528)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.8507 (1.8507)  acc1: 63.4115 (63.4115)  acc5: 84.7656 (84.7656)  time: 8.9609  data: 8.7490  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.3547 (2.2039)  acc1: 51.3021 (53.6800)  acc5: 76.1719 (78.3200)  time: 1.1436  data: 0.9722  max mem: 41794
Test: Total time: 0:00:10 (1.1611 s / it)
* Acc@1 54.046 Acc@5 78.624 loss 2.200
Accuracy of the model on the 50000 test images: 54.0%
Max accuracy: 54.05%
Epoch: [12]  [  0/312]  eta: 1:20:34  lr: 0.002400  min_lr: 0.002400  loss: 4.8868 (4.8868)  weight_decay: 0.0500 (0.0500)  time: 15.4951  data: 13.7992  max mem: 41794
Epoch: [12]  [ 10/312]  eta: 0:11:22  lr: 0.002407  min_lr: 0.002407  loss: 4.5309 (4.4586)  weight_decay: 0.0500 (0.0500)  time: 2.2583  data: 1.2798  max mem: 41794
Epoch: [12]  [ 20/312]  eta: 0:07:17  lr: 0.002413  min_lr: 0.002413  loss: 4.7766 (4.5902)  weight_decay: 0.0500 (0.0500)  time: 0.7995  data: 0.0822  max mem: 41794
Epoch: [12]  [ 30/312]  eta: 0:05:35  lr: 0.002420  min_lr: 0.002420  loss: 4.8576 (4.5946)  weight_decay: 0.0500 (0.0500)  time: 0.6046  data: 0.0686  max mem: 41794
Epoch: [12]  [ 40/312]  eta: 0:04:39  lr: 0.002426  min_lr: 0.002426  loss: 4.7882 (4.6118)  weight_decay: 0.0500 (0.0500)  time: 0.5303  data: 0.0154  max mem: 41794
Epoch: [12]  [ 50/312]  eta: 0:04:19  lr: 0.002432  min_lr: 0.002432  loss: 4.7882 (4.6259)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.1794  max mem: 41794
Epoch: [12]  [ 60/312]  eta: 0:03:49  lr: 0.002439  min_lr: 0.002439  loss: 4.5921 (4.6188)  weight_decay: 0.0500 (0.0500)  time: 0.6704  data: 0.1678  max mem: 41794
Epoch: [12]  [ 70/312]  eta: 0:03:38  lr: 0.002445  min_lr: 0.002445  loss: 4.4467 (4.5775)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.1751  max mem: 41794
Epoch: [12]  [ 80/312]  eta: 0:03:28  lr: 0.002452  min_lr: 0.002452  loss: 4.7215 (4.6121)  weight_decay: 0.0500 (0.0500)  time: 0.8588  data: 0.3520  max mem: 41794
Epoch: [12]  [ 90/312]  eta: 0:03:10  lr: 0.002458  min_lr: 0.002458  loss: 4.8915 (4.6164)  weight_decay: 0.0500 (0.0500)  time: 0.6880  data: 0.1944  max mem: 41794
Epoch: [12]  [100/312]  eta: 0:03:02  lr: 0.002464  min_lr: 0.002464  loss: 4.8172 (4.5954)  weight_decay: 0.0500 (0.0500)  time: 0.7202  data: 0.2225  max mem: 41794
Epoch: [12]  [110/312]  eta: 0:02:47  lr: 0.002471  min_lr: 0.002471  loss: 4.8638 (4.6033)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.2089  max mem: 41794
Epoch: [12]  [120/312]  eta: 0:02:40  lr: 0.002477  min_lr: 0.002477  loss: 4.6986 (4.5851)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.2020  max mem: 41794
Epoch: [12]  [130/312]  eta: 0:02:31  lr: 0.002484  min_lr: 0.002484  loss: 4.6986 (4.5889)  weight_decay: 0.0500 (0.0500)  time: 0.8399  data: 0.3339  max mem: 41794
Epoch: [12]  [140/312]  eta: 0:02:18  lr: 0.002490  min_lr: 0.002490  loss: 4.7520 (4.5753)  weight_decay: 0.0500 (0.0500)  time: 0.6349  data: 0.1326  max mem: 41794
Epoch: [12]  [150/312]  eta: 0:02:12  lr: 0.002497  min_lr: 0.002497  loss: 4.4697 (4.5723)  weight_decay: 0.0500 (0.0500)  time: 0.7307  data: 0.2287  max mem: 41794
Epoch: [12]  [160/312]  eta: 0:02:04  lr: 0.002503  min_lr: 0.002503  loss: 4.5488 (4.5610)  weight_decay: 0.0500 (0.0500)  time: 0.9091  data: 0.4106  max mem: 41794
Epoch: [12]  [170/312]  eta: 0:01:53  lr: 0.002509  min_lr: 0.002509  loss: 4.5211 (4.5653)  weight_decay: 0.0500 (0.0500)  time: 0.6722  data: 0.1828  max mem: 41794
Epoch: [12]  [180/312]  eta: 0:01:45  lr: 0.002516  min_lr: 0.002516  loss: 4.5211 (4.5633)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.1672  max mem: 41794
Epoch: [12]  [190/312]  eta: 0:01:35  lr: 0.002522  min_lr: 0.002522  loss: 4.8135 (4.5595)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.1671  max mem: 41794
Epoch: [12]  [200/312]  eta: 0:01:28  lr: 0.002529  min_lr: 0.002529  loss: 4.6008 (4.5496)  weight_decay: 0.0500 (0.0500)  time: 0.6844  data: 0.1880  max mem: 41794
Epoch: [12]  [210/312]  eta: 0:01:20  lr: 0.002535  min_lr: 0.002535  loss: 4.5611 (4.5431)  weight_decay: 0.0500 (0.0500)  time: 0.8641  data: 0.3662  max mem: 41794
Epoch: [12]  [220/312]  eta: 0:01:11  lr: 0.002541  min_lr: 0.002541  loss: 4.6108 (4.5521)  weight_decay: 0.0500 (0.0500)  time: 0.6787  data: 0.1790  max mem: 41794
Epoch: [12]  [230/312]  eta: 0:01:04  lr: 0.002548  min_lr: 0.002548  loss: 4.7133 (4.5584)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.1696  max mem: 41794
Epoch: [12]  [240/312]  eta: 0:00:56  lr: 0.002554  min_lr: 0.002554  loss: 4.6978 (4.5612)  weight_decay: 0.0500 (0.0500)  time: 0.8822  data: 0.3804  max mem: 41794
Epoch: [12]  [250/312]  eta: 0:00:48  lr: 0.002561  min_lr: 0.002561  loss: 4.6768 (4.5654)  weight_decay: 0.0500 (0.0500)  time: 0.7142  data: 0.2143  max mem: 41794
Epoch: [12]  [260/312]  eta: 0:00:40  lr: 0.002567  min_lr: 0.002567  loss: 4.9390 (4.5646)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1893  max mem: 41794
Epoch: [12]  [270/312]  eta: 0:00:32  lr: 0.002573  min_lr: 0.002573  loss: 4.4578 (4.5566)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1864  max mem: 41794
Epoch: [12]  [280/312]  eta: 0:00:24  lr: 0.002580  min_lr: 0.002580  loss: 4.7046 (4.5667)  weight_decay: 0.0500 (0.0500)  time: 0.6928  data: 0.1922  max mem: 41794
Epoch: [12]  [290/312]  eta: 0:00:17  lr: 0.002586  min_lr: 0.002586  loss: 4.5891 (4.5599)  weight_decay: 0.0500 (0.0500)  time: 0.8744  data: 0.3689  max mem: 41794
Epoch: [12]  [300/312]  eta: 0:00:09  lr: 0.002593  min_lr: 0.002593  loss: 4.6766 (4.5718)  weight_decay: 0.0500 (0.0500)  time: 0.6662  data: 0.1770  max mem: 41794
Epoch: [12]  [310/312]  eta: 0:00:01  lr: 0.002599  min_lr: 0.002599  loss: 4.9161 (4.5831)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [12]  [311/312]  eta: 0:00:00  lr: 0.002600  min_lr: 0.002600  loss: 4.9161 (4.5857)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.0001  max mem: 41794
Epoch: [12] Total time: 0:03:58 (0.7641 s / it)
Averaged stats: lr: 0.002600  min_lr: 0.002600  loss: 4.9161 (4.5911)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.8724 (1.8724)  acc1: 64.1927 (64.1927)  acc5: 85.9375 (85.9375)  time: 8.6209  data: 8.4114  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.3659 (2.2204)  acc1: 52.6042 (54.4960)  acc5: 79.1667 (79.5200)  time: 1.1058  data: 0.9347  max mem: 41794
Test: Total time: 0:00:10 (1.1316 s / it)
* Acc@1 55.322 Acc@5 79.686 loss 2.224
Accuracy of the model on the 50000 test images: 55.3%
Max accuracy: 55.32%
Epoch: [13]  [  0/312]  eta: 1:15:15  lr: 0.002600  min_lr: 0.002600  loss: 4.8941 (4.8941)  weight_decay: 0.0500 (0.0500)  time: 14.4736  data: 12.9728  max mem: 41794
Epoch: [13]  [ 10/312]  eta: 0:09:50  lr: 0.002607  min_lr: 0.002607  loss: 4.0424 (4.2872)  weight_decay: 0.0500 (0.0500)  time: 1.9549  data: 1.2888  max mem: 41794
Epoch: [13]  [ 20/312]  eta: 0:07:02  lr: 0.002613  min_lr: 0.002613  loss: 4.0424 (4.3407)  weight_decay: 0.0500 (0.0500)  time: 0.7973  data: 0.2023  max mem: 41794
Epoch: [13]  [ 30/312]  eta: 0:05:24  lr: 0.002620  min_lr: 0.002620  loss: 4.2696 (4.2712)  weight_decay: 0.0500 (0.0500)  time: 0.7057  data: 0.1424  max mem: 41794
Epoch: [13]  [ 40/312]  eta: 0:05:03  lr: 0.002626  min_lr: 0.002626  loss: 4.3780 (4.3437)  weight_decay: 0.0500 (0.0500)  time: 0.7648  data: 0.1497  max mem: 41794
Epoch: [13]  [ 50/312]  eta: 0:04:44  lr: 0.002632  min_lr: 0.002632  loss: 4.4781 (4.3144)  weight_decay: 0.0500 (0.0500)  time: 0.9843  data: 0.2607  max mem: 41794
Epoch: [13]  [ 60/312]  eta: 0:04:10  lr: 0.002639  min_lr: 0.002639  loss: 4.4014 (4.3591)  weight_decay: 0.0500 (0.0500)  time: 0.7450  data: 0.1116  max mem: 41794
Epoch: [13]  [ 70/312]  eta: 0:03:50  lr: 0.002645  min_lr: 0.002645  loss: 4.5910 (4.3909)  weight_decay: 0.0500 (0.0500)  time: 0.6222  data: 0.0739  max mem: 41794
Epoch: [13]  [ 80/312]  eta: 0:03:32  lr: 0.002652  min_lr: 0.002652  loss: 4.5226 (4.3834)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.0754  max mem: 41794
Epoch: [13]  [ 90/312]  eta: 0:03:15  lr: 0.002658  min_lr: 0.002658  loss: 4.4849 (4.3904)  weight_decay: 0.0500 (0.0500)  time: 0.6231  data: 0.0591  max mem: 41794
Epoch: [13]  [100/312]  eta: 0:03:07  lr: 0.002665  min_lr: 0.002665  loss: 4.5804 (4.4243)  weight_decay: 0.0500 (0.0500)  time: 0.7689  data: 0.1425  max mem: 41794
Epoch: [13]  [110/312]  eta: 0:02:52  lr: 0.002671  min_lr: 0.002671  loss: 4.6359 (4.4402)  weight_decay: 0.0500 (0.0500)  time: 0.7189  data: 0.0869  max mem: 41794
Epoch: [13]  [120/312]  eta: 0:02:43  lr: 0.002677  min_lr: 0.002677  loss: 4.8311 (4.4686)  weight_decay: 0.0500 (0.0500)  time: 0.6715  data: 0.0737  max mem: 41794
Epoch: [13]  [130/312]  eta: 0:02:34  lr: 0.002684  min_lr: 0.002684  loss: 4.7336 (4.4703)  weight_decay: 0.0500 (0.0500)  time: 0.8419  data: 0.1339  max mem: 41794
Epoch: [13]  [140/312]  eta: 0:02:22  lr: 0.002690  min_lr: 0.002690  loss: 4.4956 (4.4699)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.0622  max mem: 41794
Epoch: [13]  [150/312]  eta: 0:02:13  lr: 0.002697  min_lr: 0.002697  loss: 4.5790 (4.4713)  weight_decay: 0.0500 (0.0500)  time: 0.6555  data: 0.0812  max mem: 41794
Epoch: [13]  [160/312]  eta: 0:02:04  lr: 0.002703  min_lr: 0.002703  loss: 4.6598 (4.4823)  weight_decay: 0.0500 (0.0500)  time: 0.7484  data: 0.0840  max mem: 41794
Epoch: [13]  [170/312]  eta: 0:01:54  lr: 0.002709  min_lr: 0.002709  loss: 4.6598 (4.4808)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.0746  max mem: 41794
Epoch: [13]  [180/312]  eta: 0:01:47  lr: 0.002716  min_lr: 0.002716  loss: 4.4379 (4.4657)  weight_decay: 0.0500 (0.0500)  time: 0.7812  data: 0.1236  max mem: 41794
Epoch: [13]  [190/312]  eta: 0:01:37  lr: 0.002722  min_lr: 0.002722  loss: 4.4379 (4.4746)  weight_decay: 0.0500 (0.0500)  time: 0.7341  data: 0.0777  max mem: 41794
Epoch: [13]  [200/312]  eta: 0:01:29  lr: 0.002729  min_lr: 0.002729  loss: 4.9599 (4.4892)  weight_decay: 0.0500 (0.0500)  time: 0.6626  data: 0.0481  max mem: 41794
Epoch: [13]  [210/312]  eta: 0:01:21  lr: 0.002735  min_lr: 0.002735  loss: 4.5566 (4.4788)  weight_decay: 0.0500 (0.0500)  time: 0.8345  data: 0.0775  max mem: 41794
Epoch: [13]  [220/312]  eta: 0:01:12  lr: 0.002741  min_lr: 0.002741  loss: 4.3223 (4.4767)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.0637  max mem: 41794
Epoch: [13]  [230/312]  eta: 0:01:04  lr: 0.002748  min_lr: 0.002748  loss: 4.5190 (4.4788)  weight_decay: 0.0500 (0.0500)  time: 0.6653  data: 0.0309  max mem: 41794
Epoch: [13]  [240/312]  eta: 0:00:56  lr: 0.002754  min_lr: 0.002754  loss: 4.6624 (4.4827)  weight_decay: 0.0500 (0.0500)  time: 0.7889  data: 0.0276  max mem: 41794
Epoch: [13]  [250/312]  eta: 0:00:48  lr: 0.002761  min_lr: 0.002761  loss: 4.3948 (4.4699)  weight_decay: 0.0500 (0.0500)  time: 0.7145  data: 0.0573  max mem: 41794
Epoch: [13]  [260/312]  eta: 0:00:40  lr: 0.002767  min_lr: 0.002767  loss: 4.6475 (4.4818)  weight_decay: 0.0500 (0.0500)  time: 0.7430  data: 0.0867  max mem: 41794
Epoch: [13]  [270/312]  eta: 0:00:32  lr: 0.002774  min_lr: 0.002774  loss: 4.9001 (4.4870)  weight_decay: 0.0500 (0.0500)  time: 0.6623  data: 0.0384  max mem: 41794
Epoch: [13]  [280/312]  eta: 0:00:24  lr: 0.002780  min_lr: 0.002780  loss: 4.6019 (4.4805)  weight_decay: 0.0500 (0.0500)  time: 0.7025  data: 0.0509  max mem: 41794
Epoch: [13]  [290/312]  eta: 0:00:17  lr: 0.002786  min_lr: 0.002786  loss: 4.3323 (4.4743)  weight_decay: 0.0500 (0.0500)  time: 0.8056  data: 0.0591  max mem: 41794
Epoch: [13]  [300/312]  eta: 0:00:09  lr: 0.002793  min_lr: 0.002793  loss: 4.6790 (4.4766)  weight_decay: 0.0500 (0.0500)  time: 0.5950  data: 0.0120  max mem: 41794
Epoch: [13]  [310/312]  eta: 0:00:01  lr: 0.002799  min_lr: 0.002799  loss: 4.7281 (4.4740)  weight_decay: 0.0500 (0.0500)  time: 0.4692  data: 0.0001  max mem: 41794
Epoch: [13]  [311/312]  eta: 0:00:00  lr: 0.002800  min_lr: 0.002800  loss: 4.7281 (4.4739)  weight_decay: 0.0500 (0.0500)  time: 0.4682  data: 0.0001  max mem: 41794
Epoch: [13] Total time: 0:03:58 (0.7637 s / it)
Averaged stats: lr: 0.002800  min_lr: 0.002800  loss: 4.7281 (4.5393)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.8083 (1.8083)  acc1: 63.8021 (63.8021)  acc5: 84.6354 (84.6354)  time: 8.5419  data: 8.3361  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.2178 (2.0694)  acc1: 53.7760 (55.5680)  acc5: 79.1667 (80.2080)  time: 1.0991  data: 0.9263  max mem: 41794
Test: Total time: 0:00:10 (1.1131 s / it)
* Acc@1 56.338 Acc@5 80.596 loss 2.060
Accuracy of the model on the 50000 test images: 56.3%
Max accuracy: 56.34%
Epoch: [14]  [  0/312]  eta: 1:23:51  lr: 0.002800  min_lr: 0.002800  loss: 4.7321 (4.7321)  weight_decay: 0.0500 (0.0500)  time: 16.1268  data: 15.6353  max mem: 41794
Epoch: [14]  [ 10/312]  eta: 0:11:21  lr: 0.002807  min_lr: 0.002807  loss: 4.5898 (4.4493)  weight_decay: 0.0500 (0.0500)  time: 2.2568  data: 1.4222  max mem: 41794
Epoch: [14]  [ 20/312]  eta: 0:07:20  lr: 0.002813  min_lr: 0.002813  loss: 4.5589 (4.4499)  weight_decay: 0.0500 (0.0500)  time: 0.7786  data: 0.0657  max mem: 41794
Epoch: [14]  [ 30/312]  eta: 0:05:35  lr: 0.002820  min_lr: 0.002820  loss: 4.7354 (4.5303)  weight_decay: 0.0500 (0.0500)  time: 0.6033  data: 0.0656  max mem: 41794
Epoch: [14]  [ 40/312]  eta: 0:04:44  lr: 0.002826  min_lr: 0.002826  loss: 4.7521 (4.5620)  weight_decay: 0.0500 (0.0500)  time: 0.5592  data: 0.0106  max mem: 41794
Epoch: [14]  [ 50/312]  eta: 0:04:23  lr: 0.002833  min_lr: 0.002833  loss: 4.5477 (4.5009)  weight_decay: 0.0500 (0.0500)  time: 0.7201  data: 0.1274  max mem: 41794
Epoch: [14]  [ 60/312]  eta: 0:03:51  lr: 0.002839  min_lr: 0.002839  loss: 4.4136 (4.4833)  weight_decay: 0.0500 (0.0500)  time: 0.6636  data: 0.1174  max mem: 41794
Epoch: [14]  [ 70/312]  eta: 0:03:40  lr: 0.002845  min_lr: 0.002845  loss: 4.7522 (4.5022)  weight_decay: 0.0500 (0.0500)  time: 0.6726  data: 0.1225  max mem: 41794
Epoch: [14]  [ 80/312]  eta: 0:03:27  lr: 0.002852  min_lr: 0.002852  loss: 4.7944 (4.5214)  weight_decay: 0.0500 (0.0500)  time: 0.8104  data: 0.2079  max mem: 41794
Epoch: [14]  [ 90/312]  eta: 0:03:11  lr: 0.002858  min_lr: 0.002858  loss: 4.7295 (4.4827)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.1385  max mem: 41794
Epoch: [14]  [100/312]  eta: 0:03:03  lr: 0.002865  min_lr: 0.002865  loss: 4.6806 (4.5074)  weight_decay: 0.0500 (0.0500)  time: 0.7421  data: 0.1680  max mem: 41794
Epoch: [14]  [110/312]  eta: 0:02:47  lr: 0.002871  min_lr: 0.002871  loss: 4.7721 (4.5125)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.1164  max mem: 41794
Epoch: [14]  [120/312]  eta: 0:02:40  lr: 0.002877  min_lr: 0.002877  loss: 4.6623 (4.5051)  weight_decay: 0.0500 (0.0500)  time: 0.7116  data: 0.1194  max mem: 41794
Epoch: [14]  [130/312]  eta: 0:02:33  lr: 0.002884  min_lr: 0.002884  loss: 4.6184 (4.5037)  weight_decay: 0.0500 (0.0500)  time: 0.9043  data: 0.2091  max mem: 41794
Epoch: [14]  [140/312]  eta: 0:02:20  lr: 0.002890  min_lr: 0.002890  loss: 4.6725 (4.5060)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.0913  max mem: 41794
Epoch: [14]  [150/312]  eta: 0:02:12  lr: 0.002897  min_lr: 0.002897  loss: 4.7927 (4.5005)  weight_decay: 0.0500 (0.0500)  time: 0.6584  data: 0.0615  max mem: 41794
Epoch: [14]  [160/312]  eta: 0:02:03  lr: 0.002903  min_lr: 0.002903  loss: 4.7636 (4.5036)  weight_decay: 0.0500 (0.0500)  time: 0.7681  data: 0.0778  max mem: 41794
Epoch: [14]  [170/312]  eta: 0:01:54  lr: 0.002909  min_lr: 0.002909  loss: 4.4502 (4.4898)  weight_decay: 0.0500 (0.0500)  time: 0.6988  data: 0.1123  max mem: 41794
Epoch: [14]  [180/312]  eta: 0:01:45  lr: 0.002916  min_lr: 0.002916  loss: 4.5485 (4.4978)  weight_decay: 0.0500 (0.0500)  time: 0.7404  data: 0.1451  max mem: 41794
Epoch: [14]  [190/312]  eta: 0:01:35  lr: 0.002922  min_lr: 0.002922  loss: 4.8109 (4.5065)  weight_decay: 0.0500 (0.0500)  time: 0.6406  data: 0.0498  max mem: 41794
Epoch: [14]  [200/312]  eta: 0:01:28  lr: 0.002929  min_lr: 0.002929  loss: 4.7764 (4.5073)  weight_decay: 0.0500 (0.0500)  time: 0.6581  data: 0.0629  max mem: 41794
Epoch: [14]  [210/312]  eta: 0:01:20  lr: 0.002935  min_lr: 0.002935  loss: 4.6562 (4.5080)  weight_decay: 0.0500 (0.0500)  time: 0.8490  data: 0.1315  max mem: 41794
Epoch: [14]  [220/312]  eta: 0:01:11  lr: 0.002941  min_lr: 0.002941  loss: 4.5986 (4.5126)  weight_decay: 0.0500 (0.0500)  time: 0.6965  data: 0.0835  max mem: 41794
Epoch: [14]  [230/312]  eta: 0:01:04  lr: 0.002948  min_lr: 0.002948  loss: 4.6596 (4.5153)  weight_decay: 0.0500 (0.0500)  time: 0.7386  data: 0.0807  max mem: 41794
Epoch: [14]  [240/312]  eta: 0:00:56  lr: 0.002954  min_lr: 0.002954  loss: 4.4336 (4.5011)  weight_decay: 0.0500 (0.0500)  time: 0.8648  data: 0.0683  max mem: 41794
Epoch: [14]  [250/312]  eta: 0:00:48  lr: 0.002961  min_lr: 0.002961  loss: 4.4914 (4.5089)  weight_decay: 0.0500 (0.0500)  time: 0.6576  data: 0.0230  max mem: 41794
Epoch: [14]  [260/312]  eta: 0:00:40  lr: 0.002967  min_lr: 0.002967  loss: 4.6728 (4.5076)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.0420  max mem: 41794
Epoch: [14]  [270/312]  eta: 0:00:32  lr: 0.002974  min_lr: 0.002974  loss: 4.7010 (4.5117)  weight_decay: 0.0500 (0.0500)  time: 0.6659  data: 0.0243  max mem: 41794
Epoch: [14]  [280/312]  eta: 0:00:24  lr: 0.002980  min_lr: 0.002980  loss: 4.7010 (4.5093)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.0250  max mem: 41794
Epoch: [14]  [290/312]  eta: 0:00:17  lr: 0.002986  min_lr: 0.002986  loss: 4.6814 (4.5165)  weight_decay: 0.0500 (0.0500)  time: 0.8589  data: 0.0354  max mem: 41794
Epoch: [14]  [300/312]  eta: 0:00:09  lr: 0.002993  min_lr: 0.002993  loss: 4.6814 (4.5193)  weight_decay: 0.0500 (0.0500)  time: 0.6839  data: 0.0137  max mem: 41794
Epoch: [14]  [310/312]  eta: 0:00:01  lr: 0.002999  min_lr: 0.002999  loss: 4.5956 (4.5189)  weight_decay: 0.0500 (0.0500)  time: 0.4692  data: 0.0001  max mem: 41794
Epoch: [14]  [311/312]  eta: 0:00:00  lr: 0.003000  min_lr: 0.003000  loss: 4.5956 (4.5210)  weight_decay: 0.0500 (0.0500)  time: 0.4689  data: 0.0001  max mem: 41794
Epoch: [14] Total time: 0:03:57 (0.7613 s / it)
Averaged stats: lr: 0.003000  min_lr: 0.003000  loss: 4.5956 (4.5026)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 1.9575 (1.9575)  acc1: 63.1510 (63.1510)  acc5: 85.1562 (85.1562)  time: 9.1905  data: 8.9787  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.1762 (2.1203)  acc1: 55.3385 (56.0960)  acc5: 79.4271 (80.3840)  time: 1.1693  data: 0.9977  max mem: 41794
Test: Total time: 0:00:10 (1.1851 s / it)
* Acc@1 56.502 Acc@5 80.900 loss 2.108
Accuracy of the model on the 50000 test images: 56.5%
Max accuracy: 56.50%
Epoch: [15]  [  0/312]  eta: 1:20:52  lr: 0.003000  min_lr: 0.003000  loss: 4.2923 (4.2923)  weight_decay: 0.0500 (0.0500)  time: 15.5518  data: 15.0703  max mem: 41794
Epoch: [15]  [ 10/312]  eta: 0:10:56  lr: 0.003007  min_lr: 0.003007  loss: 4.7647 (4.5768)  weight_decay: 0.0500 (0.0500)  time: 2.1750  data: 1.3707  max mem: 41794
Epoch: [15]  [ 20/312]  eta: 0:07:18  lr: 0.003013  min_lr: 0.003013  loss: 4.8159 (4.6766)  weight_decay: 0.0500 (0.0500)  time: 0.7985  data: 0.1201  max mem: 41794
Epoch: [15]  [ 30/312]  eta: 0:05:35  lr: 0.003020  min_lr: 0.003020  loss: 4.6304 (4.5581)  weight_decay: 0.0500 (0.0500)  time: 0.6458  data: 0.1201  max mem: 41794
Epoch: [15]  [ 40/312]  eta: 0:04:44  lr: 0.003026  min_lr: 0.003026  loss: 4.6482 (4.5627)  weight_decay: 0.0500 (0.0500)  time: 0.5709  data: 0.0547  max mem: 41794
Epoch: [15]  [ 50/312]  eta: 0:04:20  lr: 0.003033  min_lr: 0.003033  loss: 4.6482 (4.5033)  weight_decay: 0.0500 (0.0500)  time: 0.6952  data: 0.1915  max mem: 41794
Epoch: [15]  [ 60/312]  eta: 0:03:50  lr: 0.003039  min_lr: 0.003039  loss: 4.5617 (4.5334)  weight_decay: 0.0500 (0.0500)  time: 0.6401  data: 0.1375  max mem: 41794
Epoch: [15]  [ 70/312]  eta: 0:03:40  lr: 0.003045  min_lr: 0.003045  loss: 4.8076 (4.5187)  weight_decay: 0.0500 (0.0500)  time: 0.6953  data: 0.1573  max mem: 41794
Epoch: [15]  [ 80/312]  eta: 0:03:27  lr: 0.003052  min_lr: 0.003052  loss: 4.7677 (4.4867)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.2801  max mem: 41794
Epoch: [15]  [ 90/312]  eta: 0:03:13  lr: 0.003058  min_lr: 0.003058  loss: 4.4222 (4.4738)  weight_decay: 0.0500 (0.0500)  time: 0.7302  data: 0.1512  max mem: 41794
Epoch: [15]  [100/312]  eta: 0:03:03  lr: 0.003065  min_lr: 0.003065  loss: 4.4543 (4.4692)  weight_decay: 0.0500 (0.0500)  time: 0.7387  data: 0.1479  max mem: 41794
Epoch: [15]  [110/312]  eta: 0:02:47  lr: 0.003071  min_lr: 0.003071  loss: 4.5932 (4.4711)  weight_decay: 0.0500 (0.0500)  time: 0.6485  data: 0.1201  max mem: 41794
Epoch: [15]  [120/312]  eta: 0:02:42  lr: 0.003077  min_lr: 0.003077  loss: 4.6087 (4.4664)  weight_decay: 0.0500 (0.0500)  time: 0.7397  data: 0.1943  max mem: 41794
Epoch: [15]  [130/312]  eta: 0:02:34  lr: 0.003084  min_lr: 0.003084  loss: 4.5320 (4.4603)  weight_decay: 0.0500 (0.0500)  time: 0.9651  data: 0.3264  max mem: 41794
Epoch: [15]  [140/312]  eta: 0:02:22  lr: 0.003090  min_lr: 0.003090  loss: 4.5708 (4.4598)  weight_decay: 0.0500 (0.0500)  time: 0.7206  data: 0.1340  max mem: 41794
Epoch: [15]  [150/312]  eta: 0:02:14  lr: 0.003097  min_lr: 0.003097  loss: 4.4233 (4.4476)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.0862  max mem: 41794
Epoch: [15]  [160/312]  eta: 0:02:04  lr: 0.003103  min_lr: 0.003103  loss: 4.4233 (4.4535)  weight_decay: 0.0500 (0.0500)  time: 0.7443  data: 0.0849  max mem: 41794
Epoch: [15]  [170/312]  eta: 0:01:55  lr: 0.003109  min_lr: 0.003109  loss: 4.7630 (4.4516)  weight_decay: 0.0500 (0.0500)  time: 0.6790  data: 0.0768  max mem: 41794
Epoch: [15]  [180/312]  eta: 0:01:47  lr: 0.003116  min_lr: 0.003116  loss: 4.3247 (4.4339)  weight_decay: 0.0500 (0.0500)  time: 0.8078  data: 0.1905  max mem: 41794
Epoch: [15]  [190/312]  eta: 0:01:37  lr: 0.003122  min_lr: 0.003122  loss: 4.1018 (4.4147)  weight_decay: 0.0500 (0.0500)  time: 0.6953  data: 0.1172  max mem: 41794
Epoch: [15]  [200/312]  eta: 0:01:29  lr: 0.003129  min_lr: 0.003129  loss: 4.1248 (4.4143)  weight_decay: 0.0500 (0.0500)  time: 0.6345  data: 0.1016  max mem: 41794
Epoch: [15]  [210/312]  eta: 0:01:21  lr: 0.003135  min_lr: 0.003135  loss: 4.4110 (4.4111)  weight_decay: 0.0500 (0.0500)  time: 0.8087  data: 0.2449  max mem: 41794
Epoch: [15]  [220/312]  eta: 0:01:12  lr: 0.003142  min_lr: 0.003142  loss: 4.2187 (4.4012)  weight_decay: 0.0500 (0.0500)  time: 0.6718  data: 0.1468  max mem: 41794
Epoch: [15]  [230/312]  eta: 0:01:04  lr: 0.003148  min_lr: 0.003148  loss: 4.1669 (4.3990)  weight_decay: 0.0500 (0.0500)  time: 0.6685  data: 0.1439  max mem: 41794
Epoch: [15]  [240/312]  eta: 0:00:56  lr: 0.003154  min_lr: 0.003154  loss: 4.4855 (4.4060)  weight_decay: 0.0500 (0.0500)  time: 0.7685  data: 0.2354  max mem: 41794
Epoch: [15]  [250/312]  eta: 0:00:48  lr: 0.003161  min_lr: 0.003161  loss: 4.5896 (4.4077)  weight_decay: 0.0500 (0.0500)  time: 0.6726  data: 0.1704  max mem: 41794
Epoch: [15]  [260/312]  eta: 0:00:40  lr: 0.003167  min_lr: 0.003167  loss: 4.7338 (4.4216)  weight_decay: 0.0500 (0.0500)  time: 0.7521  data: 0.2588  max mem: 41794
Epoch: [15]  [270/312]  eta: 0:00:32  lr: 0.003174  min_lr: 0.003174  loss: 4.7338 (4.4224)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.1804  max mem: 41794
Epoch: [15]  [280/312]  eta: 0:00:24  lr: 0.003180  min_lr: 0.003180  loss: 4.5429 (4.4243)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1695  max mem: 41794
Epoch: [15]  [290/312]  eta: 0:00:17  lr: 0.003186  min_lr: 0.003186  loss: 4.3308 (4.4162)  weight_decay: 0.0500 (0.0500)  time: 0.8186  data: 0.3012  max mem: 41794
Epoch: [15]  [300/312]  eta: 0:00:09  lr: 0.003193  min_lr: 0.003193  loss: 4.7162 (4.4278)  weight_decay: 0.0500 (0.0500)  time: 0.6306  data: 0.1321  max mem: 41794
Epoch: [15]  [310/312]  eta: 0:00:01  lr: 0.003199  min_lr: 0.003199  loss: 4.7162 (4.4304)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [15]  [311/312]  eta: 0:00:00  lr: 0.003200  min_lr: 0.003200  loss: 4.7045 (4.4306)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [15] Total time: 0:03:57 (0.7602 s / it)
Averaged stats: lr: 0.003200  min_lr: 0.003200  loss: 4.7045 (4.4166)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.6655 (1.6655)  acc1: 67.8385 (67.8385)  acc5: 87.8906 (87.8906)  time: 8.4628  data: 8.2638  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.1088 (2.0048)  acc1: 56.9010 (57.7280)  acc5: 81.3802 (81.6800)  time: 1.0959  data: 0.9183  max mem: 41794
Test: Total time: 0:00:10 (1.1364 s / it)
* Acc@1 57.980 Acc@5 81.894 loss 2.002
Accuracy of the model on the 50000 test images: 58.0%
Max accuracy: 57.98%
Epoch: [16]  [  0/312]  eta: 1:22:58  lr: 0.003201  min_lr: 0.003201  loss: 4.3947 (4.3947)  weight_decay: 0.0500 (0.0500)  time: 15.9570  data: 15.4562  max mem: 41794
Epoch: [16]  [ 10/312]  eta: 0:09:51  lr: 0.003207  min_lr: 0.003207  loss: 4.6294 (4.5409)  weight_decay: 0.0500 (0.0500)  time: 1.9587  data: 1.4540  max mem: 41794
Epoch: [16]  [ 20/312]  eta: 0:07:06  lr: 0.003213  min_lr: 0.003213  loss: 4.6294 (4.5181)  weight_decay: 0.0500 (0.0500)  time: 0.7346  data: 0.2200  max mem: 41794
Epoch: [16]  [ 30/312]  eta: 0:05:25  lr: 0.003220  min_lr: 0.003220  loss: 4.5019 (4.4645)  weight_decay: 0.0500 (0.0500)  time: 0.7101  data: 0.1935  max mem: 41794
Epoch: [16]  [ 40/312]  eta: 0:04:59  lr: 0.003226  min_lr: 0.003226  loss: 4.2860 (4.4099)  weight_decay: 0.0500 (0.0500)  time: 0.7251  data: 0.2150  max mem: 41794
Epoch: [16]  [ 50/312]  eta: 0:04:40  lr: 0.003233  min_lr: 0.003233  loss: 4.4293 (4.3974)  weight_decay: 0.0500 (0.0500)  time: 0.9443  data: 0.4281  max mem: 41794
Epoch: [16]  [ 60/312]  eta: 0:04:07  lr: 0.003239  min_lr: 0.003239  loss: 4.4293 (4.3774)  weight_decay: 0.0500 (0.0500)  time: 0.7352  data: 0.2138  max mem: 41794
Epoch: [16]  [ 70/312]  eta: 0:03:54  lr: 0.003245  min_lr: 0.003245  loss: 4.5432 (4.4082)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.1809  max mem: 41794
Epoch: [16]  [ 80/312]  eta: 0:03:40  lr: 0.003252  min_lr: 0.003252  loss: 4.6818 (4.3795)  weight_decay: 0.0500 (0.0500)  time: 0.8558  data: 0.3341  max mem: 41794
Epoch: [16]  [ 90/312]  eta: 0:03:19  lr: 0.003258  min_lr: 0.003258  loss: 4.6152 (4.3921)  weight_decay: 0.0500 (0.0500)  time: 0.6616  data: 0.1538  max mem: 41794
Epoch: [16]  [100/312]  eta: 0:03:10  lr: 0.003265  min_lr: 0.003265  loss: 4.4523 (4.3649)  weight_decay: 0.0500 (0.0500)  time: 0.7011  data: 0.1959  max mem: 41794
Epoch: [16]  [110/312]  eta: 0:02:55  lr: 0.003271  min_lr: 0.003271  loss: 4.2217 (4.3644)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1959  max mem: 41794
Epoch: [16]  [120/312]  eta: 0:02:45  lr: 0.003277  min_lr: 0.003277  loss: 4.2759 (4.3687)  weight_decay: 0.0500 (0.0500)  time: 0.6588  data: 0.1529  max mem: 41794
Epoch: [16]  [130/312]  eta: 0:02:37  lr: 0.003284  min_lr: 0.003284  loss: 4.2759 (4.3617)  weight_decay: 0.0500 (0.0500)  time: 0.8420  data: 0.3425  max mem: 41794
Epoch: [16]  [140/312]  eta: 0:02:23  lr: 0.003290  min_lr: 0.003290  loss: 4.3695 (4.3779)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.1902  max mem: 41794
Epoch: [16]  [150/312]  eta: 0:02:15  lr: 0.003297  min_lr: 0.003297  loss: 4.6935 (4.3919)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.1788  max mem: 41794
Epoch: [16]  [160/312]  eta: 0:02:07  lr: 0.003303  min_lr: 0.003303  loss: 4.5644 (4.3872)  weight_decay: 0.0500 (0.0500)  time: 0.8474  data: 0.3436  max mem: 41794
Epoch: [16]  [170/312]  eta: 0:01:56  lr: 0.003310  min_lr: 0.003310  loss: 4.4844 (4.3997)  weight_decay: 0.0500 (0.0500)  time: 0.6671  data: 0.1655  max mem: 41794
Epoch: [16]  [180/312]  eta: 0:01:48  lr: 0.003316  min_lr: 0.003316  loss: 4.4844 (4.3929)  weight_decay: 0.0500 (0.0500)  time: 0.6625  data: 0.1652  max mem: 41794
Epoch: [16]  [190/312]  eta: 0:01:37  lr: 0.003322  min_lr: 0.003322  loss: 4.4150 (4.4043)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.1652  max mem: 41794
Epoch: [16]  [200/312]  eta: 0:01:29  lr: 0.003329  min_lr: 0.003329  loss: 4.3534 (4.3903)  weight_decay: 0.0500 (0.0500)  time: 0.6571  data: 0.1579  max mem: 41794
Epoch: [16]  [210/312]  eta: 0:01:22  lr: 0.003335  min_lr: 0.003335  loss: 4.4673 (4.4023)  weight_decay: 0.0500 (0.0500)  time: 0.8287  data: 0.3294  max mem: 41794
Epoch: [16]  [220/312]  eta: 0:01:12  lr: 0.003342  min_lr: 0.003342  loss: 4.6801 (4.4031)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1721  max mem: 41794
Epoch: [16]  [230/312]  eta: 0:01:04  lr: 0.003348  min_lr: 0.003348  loss: 4.5818 (4.4143)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1651  max mem: 41794
Epoch: [16]  [240/312]  eta: 0:00:57  lr: 0.003354  min_lr: 0.003354  loss: 4.5419 (4.4078)  weight_decay: 0.0500 (0.0500)  time: 0.8355  data: 0.3348  max mem: 41794
Epoch: [16]  [250/312]  eta: 0:00:48  lr: 0.003361  min_lr: 0.003361  loss: 4.6258 (4.4139)  weight_decay: 0.0500 (0.0500)  time: 0.6612  data: 0.1703  max mem: 41794
Epoch: [16]  [260/312]  eta: 0:00:40  lr: 0.003367  min_lr: 0.003367  loss: 4.6537 (4.4140)  weight_decay: 0.0500 (0.0500)  time: 0.6719  data: 0.1864  max mem: 41794
Epoch: [16]  [270/312]  eta: 0:00:32  lr: 0.003374  min_lr: 0.003374  loss: 4.5696 (4.4165)  weight_decay: 0.0500 (0.0500)  time: 0.6742  data: 0.1864  max mem: 41794
Epoch: [16]  [280/312]  eta: 0:00:24  lr: 0.003380  min_lr: 0.003380  loss: 4.5301 (4.4167)  weight_decay: 0.0500 (0.0500)  time: 0.6771  data: 0.1910  max mem: 41794
Epoch: [16]  [290/312]  eta: 0:00:17  lr: 0.003386  min_lr: 0.003386  loss: 4.5084 (4.4141)  weight_decay: 0.0500 (0.0500)  time: 0.8187  data: 0.3184  max mem: 41794
Epoch: [16]  [300/312]  eta: 0:00:09  lr: 0.003393  min_lr: 0.003393  loss: 4.6287 (4.4209)  weight_decay: 0.0500 (0.0500)  time: 0.6214  data: 0.1278  max mem: 41794
Epoch: [16]  [310/312]  eta: 0:00:01  lr: 0.003399  min_lr: 0.003399  loss: 4.6691 (4.4273)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0001  max mem: 41794
Epoch: [16]  [311/312]  eta: 0:00:00  lr: 0.003400  min_lr: 0.003400  loss: 4.6374 (4.4276)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [16] Total time: 0:03:57 (0.7622 s / it)
Averaged stats: lr: 0.003400  min_lr: 0.003400  loss: 4.6374 (4.4021)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.7695 (1.7695)  acc1: 64.7135 (64.7135)  acc5: 87.1094 (87.1094)  time: 8.4022  data: 8.1885  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.2287 (2.0515)  acc1: 55.8594 (57.8080)  acc5: 81.5104 (81.8560)  time: 1.0819  data: 0.9099  max mem: 41794
Test: Total time: 0:00:09 (1.0932 s / it)
* Acc@1 57.974 Acc@5 81.984 loss 2.047
Accuracy of the model on the 50000 test images: 58.0%
Max accuracy: 57.98%
Epoch: [17]  [  0/312]  eta: 1:21:04  lr: 0.003401  min_lr: 0.003401  loss: 4.1309 (4.1309)  weight_decay: 0.0500 (0.0500)  time: 15.5927  data: 13.1495  max mem: 41794
Epoch: [17]  [ 10/312]  eta: 0:10:26  lr: 0.003407  min_lr: 0.003407  loss: 4.1309 (4.1651)  weight_decay: 0.0500 (0.0500)  time: 2.0754  data: 1.4081  max mem: 41794
Epoch: [17]  [ 20/312]  eta: 0:07:27  lr: 0.003413  min_lr: 0.003413  loss: 4.3585 (4.2512)  weight_decay: 0.0500 (0.0500)  time: 0.8297  data: 0.2526  max mem: 41794
Epoch: [17]  [ 30/312]  eta: 0:05:41  lr: 0.003420  min_lr: 0.003420  loss: 4.3198 (4.2370)  weight_decay: 0.0500 (0.0500)  time: 0.7380  data: 0.1426  max mem: 41794
Epoch: [17]  [ 40/312]  eta: 0:05:07  lr: 0.003426  min_lr: 0.003426  loss: 4.5342 (4.3079)  weight_decay: 0.0500 (0.0500)  time: 0.7098  data: 0.0755  max mem: 41794
Epoch: [17]  [ 50/312]  eta: 0:04:36  lr: 0.003433  min_lr: 0.003433  loss: 4.6794 (4.3448)  weight_decay: 0.0500 (0.0500)  time: 0.8113  data: 0.1412  max mem: 41794
Epoch: [17]  [ 60/312]  eta: 0:04:02  lr: 0.003439  min_lr: 0.003439  loss: 4.4771 (4.3316)  weight_decay: 0.0500 (0.0500)  time: 0.6189  data: 0.0731  max mem: 41794
Epoch: [17]  [ 70/312]  eta: 0:03:52  lr: 0.003445  min_lr: 0.003445  loss: 4.1802 (4.3234)  weight_decay: 0.0500 (0.0500)  time: 0.7267  data: 0.1076  max mem: 41794
Epoch: [17]  [ 80/312]  eta: 0:03:37  lr: 0.003452  min_lr: 0.003452  loss: 4.2092 (4.3100)  weight_decay: 0.0500 (0.0500)  time: 0.8556  data: 0.1142  max mem: 41794
Epoch: [17]  [ 90/312]  eta: 0:03:20  lr: 0.003458  min_lr: 0.003458  loss: 4.3889 (4.3162)  weight_decay: 0.0500 (0.0500)  time: 0.6945  data: 0.0786  max mem: 41794
Epoch: [17]  [100/312]  eta: 0:03:12  lr: 0.003465  min_lr: 0.003465  loss: 4.5550 (4.3424)  weight_decay: 0.0500 (0.0500)  time: 0.7817  data: 0.1562  max mem: 41794
Epoch: [17]  [110/312]  eta: 0:02:56  lr: 0.003471  min_lr: 0.003471  loss: 4.6071 (4.3517)  weight_decay: 0.0500 (0.0500)  time: 0.7240  data: 0.0924  max mem: 41794
Epoch: [17]  [120/312]  eta: 0:02:45  lr: 0.003477  min_lr: 0.003477  loss: 4.3708 (4.3290)  weight_decay: 0.0500 (0.0500)  time: 0.6432  data: 0.0636  max mem: 41794
Epoch: [17]  [130/312]  eta: 0:02:37  lr: 0.003484  min_lr: 0.003484  loss: 4.4677 (4.3333)  weight_decay: 0.0500 (0.0500)  time: 0.8417  data: 0.1225  max mem: 41794
Epoch: [17]  [140/312]  eta: 0:02:24  lr: 0.003490  min_lr: 0.003490  loss: 4.4891 (4.3340)  weight_decay: 0.0500 (0.0500)  time: 0.7098  data: 0.0694  max mem: 41794
Epoch: [17]  [150/312]  eta: 0:02:16  lr: 0.003497  min_lr: 0.003497  loss: 4.4891 (4.3405)  weight_decay: 0.0500 (0.0500)  time: 0.6812  data: 0.0701  max mem: 41794
Epoch: [17]  [160/312]  eta: 0:02:07  lr: 0.003503  min_lr: 0.003503  loss: 4.6653 (4.3501)  weight_decay: 0.0500 (0.0500)  time: 0.8058  data: 0.0712  max mem: 41794
Epoch: [17]  [170/312]  eta: 0:01:56  lr: 0.003510  min_lr: 0.003510  loss: 4.6801 (4.3551)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.0423  max mem: 41794
Epoch: [17]  [180/312]  eta: 0:01:49  lr: 0.003516  min_lr: 0.003516  loss: 4.2990 (4.3371)  weight_decay: 0.0500 (0.0500)  time: 0.7600  data: 0.1294  max mem: 41794
Epoch: [17]  [190/312]  eta: 0:01:39  lr: 0.003522  min_lr: 0.003522  loss: 4.4228 (4.3487)  weight_decay: 0.0500 (0.0500)  time: 0.7313  data: 0.0976  max mem: 41794
Epoch: [17]  [200/312]  eta: 0:01:30  lr: 0.003529  min_lr: 0.003529  loss: 4.5944 (4.3473)  weight_decay: 0.0500 (0.0500)  time: 0.6579  data: 0.0536  max mem: 41794
Epoch: [17]  [210/312]  eta: 0:01:22  lr: 0.003535  min_lr: 0.003535  loss: 4.1934 (4.3401)  weight_decay: 0.0500 (0.0500)  time: 0.8226  data: 0.1049  max mem: 41794
Epoch: [17]  [220/312]  eta: 0:01:13  lr: 0.003542  min_lr: 0.003542  loss: 4.5244 (4.3505)  weight_decay: 0.0500 (0.0500)  time: 0.6713  data: 0.0584  max mem: 41794
Epoch: [17]  [230/312]  eta: 0:01:05  lr: 0.003548  min_lr: 0.003548  loss: 4.4591 (4.3377)  weight_decay: 0.0500 (0.0500)  time: 0.6962  data: 0.0695  max mem: 41794
Epoch: [17]  [240/312]  eta: 0:00:57  lr: 0.003554  min_lr: 0.003554  loss: 4.3640 (4.3339)  weight_decay: 0.0500 (0.0500)  time: 0.8148  data: 0.0813  max mem: 41794
Epoch: [17]  [250/312]  eta: 0:00:49  lr: 0.003561  min_lr: 0.003561  loss: 4.3640 (4.3282)  weight_decay: 0.0500 (0.0500)  time: 0.6692  data: 0.0623  max mem: 41794
Epoch: [17]  [260/312]  eta: 0:00:41  lr: 0.003567  min_lr: 0.003567  loss: 4.3370 (4.3306)  weight_decay: 0.0500 (0.0500)  time: 0.7247  data: 0.1124  max mem: 41794
Epoch: [17]  [270/312]  eta: 0:00:32  lr: 0.003574  min_lr: 0.003574  loss: 4.3023 (4.3227)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.0625  max mem: 41794
Epoch: [17]  [280/312]  eta: 0:00:25  lr: 0.003580  min_lr: 0.003580  loss: 4.3492 (4.3244)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.0539  max mem: 41794
Epoch: [17]  [290/312]  eta: 0:00:17  lr: 0.003586  min_lr: 0.003586  loss: 4.5251 (4.3316)  weight_decay: 0.0500 (0.0500)  time: 0.7905  data: 0.0878  max mem: 41794
Epoch: [17]  [300/312]  eta: 0:00:09  lr: 0.003593  min_lr: 0.003593  loss: 4.4952 (4.3251)  weight_decay: 0.0500 (0.0500)  time: 0.5884  data: 0.0342  max mem: 41794
Epoch: [17]  [310/312]  eta: 0:00:01  lr: 0.003599  min_lr: 0.003599  loss: 4.4045 (4.3277)  weight_decay: 0.0500 (0.0500)  time: 0.4616  data: 0.0001  max mem: 41794
Epoch: [17]  [311/312]  eta: 0:00:00  lr: 0.003600  min_lr: 0.003600  loss: 4.5733 (4.3291)  weight_decay: 0.0500 (0.0500)  time: 0.4614  data: 0.0001  max mem: 41794
Epoch: [17] Total time: 0:04:00 (0.7695 s / it)
Averaged stats: lr: 0.003600  min_lr: 0.003600  loss: 4.5733 (4.3568)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.6145 (1.6145)  acc1: 66.4062 (66.4062)  acc5: 86.4583 (86.4583)  time: 8.6387  data: 8.4292  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.1553 (1.9615)  acc1: 55.9896 (57.5360)  acc5: 80.7292 (82.0320)  time: 1.1096  data: 0.9367  max mem: 41794
Test: Total time: 0:00:10 (1.1518 s / it)
* Acc@1 58.062 Acc@5 82.240 loss 1.954
Accuracy of the model on the 50000 test images: 58.1%
Max accuracy: 58.06%
Epoch: [18]  [  0/312]  eta: 1:19:16  lr: 0.003601  min_lr: 0.003601  loss: 3.5842 (3.5842)  weight_decay: 0.0500 (0.0500)  time: 15.2436  data: 12.6404  max mem: 41794
Epoch: [18]  [ 10/312]  eta: 0:12:21  lr: 0.003607  min_lr: 0.003607  loss: 4.5697 (4.4439)  weight_decay: 0.0500 (0.0500)  time: 2.4560  data: 1.3927  max mem: 41794
Epoch: [18]  [ 20/312]  eta: 0:07:32  lr: 0.003613  min_lr: 0.003613  loss: 4.6235 (4.5192)  weight_decay: 0.0500 (0.0500)  time: 0.8651  data: 0.1446  max mem: 41794
Epoch: [18]  [ 30/312]  eta: 0:05:42  lr: 0.003620  min_lr: 0.003620  loss: 4.5641 (4.4188)  weight_decay: 0.0500 (0.0500)  time: 0.5316  data: 0.0167  max mem: 41794
Epoch: [18]  [ 40/312]  eta: 0:04:44  lr: 0.003626  min_lr: 0.003626  loss: 4.2964 (4.3731)  weight_decay: 0.0500 (0.0500)  time: 0.5136  data: 0.0107  max mem: 41794
Epoch: [18]  [ 50/312]  eta: 0:04:06  lr: 0.003633  min_lr: 0.003633  loss: 4.1966 (4.3745)  weight_decay: 0.0500 (0.0500)  time: 0.5161  data: 0.0114  max mem: 41794
Epoch: [18]  [ 60/312]  eta: 0:03:39  lr: 0.003639  min_lr: 0.003639  loss: 4.1966 (4.3620)  weight_decay: 0.0500 (0.0500)  time: 0.5141  data: 0.0084  max mem: 41794
Epoch: [18]  [ 70/312]  eta: 0:03:19  lr: 0.003645  min_lr: 0.003645  loss: 4.4211 (4.3662)  weight_decay: 0.0500 (0.0500)  time: 0.5288  data: 0.0022  max mem: 41794
Epoch: [18]  [ 80/312]  eta: 0:03:10  lr: 0.003652  min_lr: 0.003652  loss: 4.4399 (4.3725)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.0243  max mem: 41794
Epoch: [18]  [ 90/312]  eta: 0:03:00  lr: 0.003658  min_lr: 0.003658  loss: 4.5548 (4.3732)  weight_decay: 0.0500 (0.0500)  time: 0.7654  data: 0.0367  max mem: 41794
Epoch: [18]  [100/312]  eta: 0:02:49  lr: 0.003665  min_lr: 0.003665  loss: 4.3046 (4.3444)  weight_decay: 0.0500 (0.0500)  time: 0.7281  data: 0.0413  max mem: 41794
Epoch: [18]  [110/312]  eta: 0:02:41  lr: 0.003671  min_lr: 0.003671  loss: 4.1739 (4.3371)  weight_decay: 0.0500 (0.0500)  time: 0.7382  data: 0.0513  max mem: 41794
Epoch: [18]  [120/312]  eta: 0:02:31  lr: 0.003678  min_lr: 0.003678  loss: 4.1563 (4.3153)  weight_decay: 0.0500 (0.0500)  time: 0.7323  data: 0.0463  max mem: 41794
Epoch: [18]  [130/312]  eta: 0:02:25  lr: 0.003684  min_lr: 0.003684  loss: 4.3043 (4.3158)  weight_decay: 0.0500 (0.0500)  time: 0.7887  data: 0.0319  max mem: 41794
Epoch: [18]  [140/312]  eta: 0:02:15  lr: 0.003690  min_lr: 0.003690  loss: 4.4970 (4.3248)  weight_decay: 0.0500 (0.0500)  time: 0.7586  data: 0.0103  max mem: 41794
Epoch: [18]  [150/312]  eta: 0:02:07  lr: 0.003697  min_lr: 0.003697  loss: 4.5467 (4.3351)  weight_decay: 0.0500 (0.0500)  time: 0.7229  data: 0.0315  max mem: 41794
Epoch: [18]  [160/312]  eta: 0:01:59  lr: 0.003703  min_lr: 0.003703  loss: 4.4892 (4.3311)  weight_decay: 0.0500 (0.0500)  time: 0.7994  data: 0.0698  max mem: 41794
Epoch: [18]  [170/312]  eta: 0:01:49  lr: 0.003710  min_lr: 0.003710  loss: 4.4124 (4.3208)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.0404  max mem: 41794
Epoch: [18]  [180/312]  eta: 0:01:42  lr: 0.003716  min_lr: 0.003716  loss: 4.1482 (4.3120)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.0165  max mem: 41794
Epoch: [18]  [190/312]  eta: 0:01:34  lr: 0.003722  min_lr: 0.003722  loss: 4.2688 (4.3218)  weight_decay: 0.0500 (0.0500)  time: 0.7648  data: 0.0374  max mem: 41794
Epoch: [18]  [200/312]  eta: 0:01:26  lr: 0.003729  min_lr: 0.003729  loss: 4.2188 (4.3070)  weight_decay: 0.0500 (0.0500)  time: 0.6924  data: 0.0268  max mem: 41794
Epoch: [18]  [210/312]  eta: 0:01:18  lr: 0.003735  min_lr: 0.003735  loss: 4.5950 (4.3265)  weight_decay: 0.0500 (0.0500)  time: 0.7382  data: 0.0452  max mem: 41794
Epoch: [18]  [220/312]  eta: 0:01:10  lr: 0.003742  min_lr: 0.003742  loss: 4.6617 (4.3339)  weight_decay: 0.0500 (0.0500)  time: 0.7292  data: 0.0475  max mem: 41794
Epoch: [18]  [230/312]  eta: 0:01:03  lr: 0.003748  min_lr: 0.003748  loss: 4.4213 (4.3309)  weight_decay: 0.0500 (0.0500)  time: 0.7503  data: 0.0307  max mem: 41794
Epoch: [18]  [240/312]  eta: 0:00:55  lr: 0.003754  min_lr: 0.003754  loss: 4.2703 (4.3259)  weight_decay: 0.0500 (0.0500)  time: 0.7979  data: 0.0365  max mem: 41794
Epoch: [18]  [250/312]  eta: 0:00:47  lr: 0.003761  min_lr: 0.003761  loss: 4.5759 (4.3432)  weight_decay: 0.0500 (0.0500)  time: 0.6860  data: 0.0152  max mem: 41794
Epoch: [18]  [260/312]  eta: 0:00:39  lr: 0.003767  min_lr: 0.003767  loss: 4.7724 (4.3590)  weight_decay: 0.0500 (0.0500)  time: 0.7086  data: 0.0134  max mem: 41794
Epoch: [18]  [270/312]  eta: 0:00:31  lr: 0.003774  min_lr: 0.003774  loss: 4.6594 (4.3627)  weight_decay: 0.0500 (0.0500)  time: 0.7533  data: 0.0121  max mem: 41794
Epoch: [18]  [280/312]  eta: 0:00:24  lr: 0.003780  min_lr: 0.003780  loss: 4.2567 (4.3468)  weight_decay: 0.0500 (0.0500)  time: 0.7217  data: 0.0328  max mem: 41794
Epoch: [18]  [290/312]  eta: 0:00:16  lr: 0.003787  min_lr: 0.003787  loss: 3.9426 (4.3464)  weight_decay: 0.0500 (0.0500)  time: 0.7719  data: 0.0393  max mem: 41794
Epoch: [18]  [300/312]  eta: 0:00:09  lr: 0.003793  min_lr: 0.003793  loss: 4.4283 (4.3449)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.0069  max mem: 41794
Epoch: [18]  [310/312]  eta: 0:00:01  lr: 0.003799  min_lr: 0.003799  loss: 4.5420 (4.3459)  weight_decay: 0.0500 (0.0500)  time: 0.5012  data: 0.0001  max mem: 41794
Epoch: [18]  [311/312]  eta: 0:00:00  lr: 0.003800  min_lr: 0.003800  loss: 4.6415 (4.3472)  weight_decay: 0.0500 (0.0500)  time: 0.5002  data: 0.0001  max mem: 41794
Epoch: [18] Total time: 0:03:54 (0.7520 s / it)
Averaged stats: lr: 0.003800  min_lr: 0.003800  loss: 4.6415 (4.3262)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.5612 (1.5612)  acc1: 67.7083 (67.7083)  acc5: 88.8021 (88.8021)  time: 8.3581  data: 8.1454  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.1216 (1.9124)  acc1: 57.6823 (59.3760)  acc5: 79.5573 (82.6720)  time: 1.0767  data: 0.9051  max mem: 41794
Test: Total time: 0:00:09 (1.0859 s / it)
* Acc@1 59.470 Acc@5 82.830 loss 1.902
Accuracy of the model on the 50000 test images: 59.5%
Max accuracy: 59.47%
Epoch: [19]  [  0/312]  eta: 1:21:03  lr: 0.003801  min_lr: 0.003801  loss: 4.8327 (4.8327)  weight_decay: 0.0500 (0.0500)  time: 15.5878  data: 14.9165  max mem: 41794
Epoch: [19]  [ 10/312]  eta: 0:10:16  lr: 0.003807  min_lr: 0.003807  loss: 4.6489 (4.5488)  weight_decay: 0.0500 (0.0500)  time: 2.0423  data: 1.3566  max mem: 41794
Epoch: [19]  [ 20/312]  eta: 0:07:21  lr: 0.003813  min_lr: 0.003813  loss: 4.4872 (4.4406)  weight_decay: 0.0500 (0.0500)  time: 0.8098  data: 0.1123  max mem: 41794
Epoch: [19]  [ 30/312]  eta: 0:05:37  lr: 0.003820  min_lr: 0.003820  loss: 4.3634 (4.3011)  weight_decay: 0.0500 (0.0500)  time: 0.7320  data: 0.1123  max mem: 41794
Epoch: [19]  [ 40/312]  eta: 0:05:03  lr: 0.003826  min_lr: 0.003826  loss: 4.4940 (4.3540)  weight_decay: 0.0500 (0.0500)  time: 0.6967  data: 0.0951  max mem: 41794
Epoch: [19]  [ 50/312]  eta: 0:04:31  lr: 0.003833  min_lr: 0.003833  loss: 4.6401 (4.3561)  weight_decay: 0.0500 (0.0500)  time: 0.7840  data: 0.1256  max mem: 41794
Epoch: [19]  [ 60/312]  eta: 0:04:04  lr: 0.003839  min_lr: 0.003839  loss: 4.3930 (4.3705)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.0890  max mem: 41794
Epoch: [19]  [ 70/312]  eta: 0:03:49  lr: 0.003845  min_lr: 0.003845  loss: 4.3143 (4.3570)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.1471  max mem: 41794
Epoch: [19]  [ 80/312]  eta: 0:03:32  lr: 0.003852  min_lr: 0.003852  loss: 4.5394 (4.3654)  weight_decay: 0.0500 (0.0500)  time: 0.7491  data: 0.0931  max mem: 41794
Epoch: [19]  [ 90/312]  eta: 0:03:20  lr: 0.003858  min_lr: 0.003858  loss: 4.5394 (4.3690)  weight_decay: 0.0500 (0.0500)  time: 0.7387  data: 0.0902  max mem: 41794
Epoch: [19]  [100/312]  eta: 0:03:08  lr: 0.003865  min_lr: 0.003865  loss: 4.3219 (4.3434)  weight_decay: 0.0500 (0.0500)  time: 0.7914  data: 0.1439  max mem: 41794
Epoch: [19]  [110/312]  eta: 0:02:52  lr: 0.003871  min_lr: 0.003871  loss: 4.1386 (4.3195)  weight_decay: 0.0500 (0.0500)  time: 0.6464  data: 0.0618  max mem: 41794
Epoch: [19]  [120/312]  eta: 0:02:43  lr: 0.003878  min_lr: 0.003878  loss: 4.1163 (4.3108)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0948  max mem: 41794
Epoch: [19]  [130/312]  eta: 0:02:35  lr: 0.003884  min_lr: 0.003884  loss: 4.0982 (4.3073)  weight_decay: 0.0500 (0.0500)  time: 0.8424  data: 0.1725  max mem: 41794
Epoch: [19]  [140/312]  eta: 0:02:22  lr: 0.003890  min_lr: 0.003890  loss: 4.6200 (4.3352)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.0819  max mem: 41794
Epoch: [19]  [150/312]  eta: 0:02:14  lr: 0.003897  min_lr: 0.003897  loss: 4.6520 (4.3441)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.0801  max mem: 41794
Epoch: [19]  [160/312]  eta: 0:02:04  lr: 0.003903  min_lr: 0.003903  loss: 4.4350 (4.3435)  weight_decay: 0.0500 (0.0500)  time: 0.7417  data: 0.1210  max mem: 41794
Epoch: [19]  [170/312]  eta: 0:01:54  lr: 0.003910  min_lr: 0.003910  loss: 4.4058 (4.3374)  weight_decay: 0.0500 (0.0500)  time: 0.6347  data: 0.0618  max mem: 41794
Epoch: [19]  [180/312]  eta: 0:01:47  lr: 0.003916  min_lr: 0.003916  loss: 3.9098 (4.3226)  weight_decay: 0.0500 (0.0500)  time: 0.8058  data: 0.1328  max mem: 41794
Epoch: [19]  [190/312]  eta: 0:01:37  lr: 0.003922  min_lr: 0.003922  loss: 4.1943 (4.3209)  weight_decay: 0.0500 (0.0500)  time: 0.7218  data: 0.1140  max mem: 41794
Epoch: [19]  [200/312]  eta: 0:01:29  lr: 0.003929  min_lr: 0.003929  loss: 4.1943 (4.3148)  weight_decay: 0.0500 (0.0500)  time: 0.6766  data: 0.0676  max mem: 41794
Epoch: [19]  [210/312]  eta: 0:01:22  lr: 0.003935  min_lr: 0.003935  loss: 4.0269 (4.3020)  weight_decay: 0.0500 (0.0500)  time: 0.8749  data: 0.1163  max mem: 41794
Epoch: [19]  [220/312]  eta: 0:01:12  lr: 0.003942  min_lr: 0.003942  loss: 4.1150 (4.2994)  weight_decay: 0.0500 (0.0500)  time: 0.7003  data: 0.0523  max mem: 41794
Epoch: [19]  [230/312]  eta: 0:01:04  lr: 0.003948  min_lr: 0.003948  loss: 4.1575 (4.2951)  weight_decay: 0.0500 (0.0500)  time: 0.6425  data: 0.0308  max mem: 41794
Epoch: [19]  [240/312]  eta: 0:00:56  lr: 0.003954  min_lr: 0.003954  loss: 4.1827 (4.2985)  weight_decay: 0.0500 (0.0500)  time: 0.6802  data: 0.0712  max mem: 41794
Epoch: [19]  [250/312]  eta: 0:00:48  lr: 0.003961  min_lr: 0.003961  loss: 4.4628 (4.3070)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.0530  max mem: 41794
Epoch: [19]  [260/312]  eta: 0:00:41  lr: 0.003967  min_lr: 0.003967  loss: 4.5518 (4.3087)  weight_decay: 0.0500 (0.0500)  time: 0.8499  data: 0.0701  max mem: 41794
Epoch: [19]  [270/312]  eta: 0:00:32  lr: 0.003974  min_lr: 0.003974  loss: 4.5414 (4.3129)  weight_decay: 0.0500 (0.0500)  time: 0.7265  data: 0.0598  max mem: 41794
Epoch: [19]  [280/312]  eta: 0:00:25  lr: 0.003980  min_lr: 0.003980  loss: 4.2312 (4.3010)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.0523  max mem: 41794
Epoch: [19]  [290/312]  eta: 0:00:17  lr: 0.003987  min_lr: 0.003987  loss: 3.9554 (4.2914)  weight_decay: 0.0500 (0.0500)  time: 0.7923  data: 0.0522  max mem: 41794
Epoch: [19]  [300/312]  eta: 0:00:09  lr: 0.003993  min_lr: 0.003993  loss: 4.2969 (4.2869)  weight_decay: 0.0500 (0.0500)  time: 0.6011  data: 0.0151  max mem: 41794
Epoch: [19]  [310/312]  eta: 0:00:01  lr: 0.003999  min_lr: 0.003999  loss: 4.3754 (4.2923)  weight_decay: 0.0500 (0.0500)  time: 0.4797  data: 0.0150  max mem: 41794
Epoch: [19]  [311/312]  eta: 0:00:00  lr: 0.004000  min_lr: 0.004000  loss: 4.3754 (4.2928)  weight_decay: 0.0500 (0.0500)  time: 0.4647  data: 0.0001  max mem: 41794
Epoch: [19] Total time: 0:03:59 (0.7665 s / it)
Averaged stats: lr: 0.004000  min_lr: 0.004000  loss: 4.3754 (4.2873)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.7770 (1.7770)  acc1: 64.9740 (64.9740)  acc5: 85.5469 (85.5469)  time: 7.9393  data: 7.7250  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.1715 (2.0550)  acc1: 55.8594 (57.7760)  acc5: 82.1615 (81.6320)  time: 1.0308  data: 0.8584  max mem: 41794
Test: Total time: 0:00:09 (1.0437 s / it)
* Acc@1 58.242 Acc@5 82.118 loss 2.058
Accuracy of the model on the 50000 test images: 58.2%
Max accuracy: 59.47%
Epoch: [20]  [  0/312]  eta: 1:22:01  lr: 0.004000  min_lr: 0.004000  loss: 4.4679 (4.4679)  weight_decay: 0.0500 (0.0500)  time: 15.7735  data: 14.9851  max mem: 41794
Epoch: [20]  [ 10/312]  eta: 0:11:30  lr: 0.004000  min_lr: 0.004000  loss: 4.6644 (4.5857)  weight_decay: 0.0500 (0.0500)  time: 2.2869  data: 1.3628  max mem: 41794
Epoch: [20]  [ 20/312]  eta: 0:07:20  lr: 0.004000  min_lr: 0.004000  loss: 4.3482 (4.4301)  weight_decay: 0.0500 (0.0500)  time: 0.7964  data: 0.0323  max mem: 41794
Epoch: [20]  [ 30/312]  eta: 0:05:37  lr: 0.004000  min_lr: 0.004000  loss: 4.3482 (4.4847)  weight_decay: 0.0500 (0.0500)  time: 0.5975  data: 0.0373  max mem: 41794
Epoch: [20]  [ 40/312]  eta: 0:04:40  lr: 0.004000  min_lr: 0.004000  loss: 4.5281 (4.4215)  weight_decay: 0.0500 (0.0500)  time: 0.5280  data: 0.0151  max mem: 41794
Epoch: [20]  [ 50/312]  eta: 0:04:18  lr: 0.004000  min_lr: 0.004000  loss: 4.4314 (4.4347)  weight_decay: 0.0500 (0.0500)  time: 0.6625  data: 0.0777  max mem: 41794
Epoch: [20]  [ 60/312]  eta: 0:03:48  lr: 0.004000  min_lr: 0.004000  loss: 4.5235 (4.4548)  weight_decay: 0.0500 (0.0500)  time: 0.6578  data: 0.0697  max mem: 41794
Epoch: [20]  [ 70/312]  eta: 0:03:36  lr: 0.004000  min_lr: 0.004000  loss: 4.6611 (4.4969)  weight_decay: 0.0500 (0.0500)  time: 0.6559  data: 0.0939  max mem: 41794
Epoch: [20]  [ 80/312]  eta: 0:03:23  lr: 0.004000  min_lr: 0.004000  loss: 4.5617 (4.4542)  weight_decay: 0.0500 (0.0500)  time: 0.7870  data: 0.2245  max mem: 41794
Epoch: [20]  [ 90/312]  eta: 0:03:07  lr: 0.004000  min_lr: 0.004000  loss: 4.4686 (4.4427)  weight_decay: 0.0500 (0.0500)  time: 0.6638  data: 0.1346  max mem: 41794
Epoch: [20]  [100/312]  eta: 0:02:59  lr: 0.004000  min_lr: 0.004000  loss: 4.4248 (4.4258)  weight_decay: 0.0500 (0.0500)  time: 0.7113  data: 0.1668  max mem: 41794
Epoch: [20]  [110/312]  eta: 0:02:44  lr: 0.004000  min_lr: 0.004000  loss: 4.4248 (4.4137)  weight_decay: 0.0500 (0.0500)  time: 0.6769  data: 0.1680  max mem: 41794
Epoch: [20]  [120/312]  eta: 0:02:36  lr: 0.004000  min_lr: 0.004000  loss: 4.5566 (4.4018)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.1697  max mem: 41794
Epoch: [20]  [130/312]  eta: 0:02:29  lr: 0.004000  min_lr: 0.004000  loss: 4.6043 (4.3985)  weight_decay: 0.0500 (0.0500)  time: 0.8692  data: 0.3407  max mem: 41794
Epoch: [20]  [140/312]  eta: 0:02:17  lr: 0.004000  min_lr: 0.004000  loss: 4.6057 (4.4069)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.1746  max mem: 41794
Epoch: [20]  [150/312]  eta: 0:02:10  lr: 0.004000  min_lr: 0.004000  loss: 4.5147 (4.3945)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1459  max mem: 41794
Epoch: [20]  [160/312]  eta: 0:02:02  lr: 0.004000  min_lr: 0.004000  loss: 4.1509 (4.3845)  weight_decay: 0.0500 (0.0500)  time: 0.8314  data: 0.2928  max mem: 41794
Epoch: [20]  [170/312]  eta: 0:01:52  lr: 0.004000  min_lr: 0.004000  loss: 4.2995 (4.3799)  weight_decay: 0.0500 (0.0500)  time: 0.6833  data: 0.1514  max mem: 41794
Epoch: [20]  [180/312]  eta: 0:01:44  lr: 0.004000  min_lr: 0.004000  loss: 4.5769 (4.3916)  weight_decay: 0.0500 (0.0500)  time: 0.7277  data: 0.1341  max mem: 41794
Epoch: [20]  [190/312]  eta: 0:01:35  lr: 0.004000  min_lr: 0.004000  loss: 4.4513 (4.3753)  weight_decay: 0.0500 (0.0500)  time: 0.6868  data: 0.1303  max mem: 41794
Epoch: [20]  [200/312]  eta: 0:01:27  lr: 0.004000  min_lr: 0.004000  loss: 4.1786 (4.3695)  weight_decay: 0.0500 (0.0500)  time: 0.6539  data: 0.1335  max mem: 41794
Epoch: [20]  [210/312]  eta: 0:01:19  lr: 0.004000  min_lr: 0.004000  loss: 4.2111 (4.3531)  weight_decay: 0.0500 (0.0500)  time: 0.8228  data: 0.2942  max mem: 41794
Epoch: [20]  [220/312]  eta: 0:01:10  lr: 0.004000  min_lr: 0.004000  loss: 4.0861 (4.3396)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.1641  max mem: 41794
Epoch: [20]  [230/312]  eta: 0:01:03  lr: 0.004000  min_lr: 0.004000  loss: 4.0861 (4.3273)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.1567  max mem: 41794
Epoch: [20]  [240/312]  eta: 0:00:56  lr: 0.004000  min_lr: 0.004000  loss: 3.8730 (4.3114)  weight_decay: 0.0500 (0.0500)  time: 0.8762  data: 0.2595  max mem: 41794
Epoch: [20]  [250/312]  eta: 0:00:47  lr: 0.004000  min_lr: 0.004000  loss: 4.0032 (4.3101)  weight_decay: 0.0500 (0.0500)  time: 0.7131  data: 0.1318  max mem: 41794
Epoch: [20]  [260/312]  eta: 0:00:40  lr: 0.004000  min_lr: 0.004000  loss: 4.4042 (4.2992)  weight_decay: 0.0500 (0.0500)  time: 0.7276  data: 0.1217  max mem: 41794
Epoch: [20]  [270/312]  eta: 0:00:32  lr: 0.004000  min_lr: 0.004000  loss: 4.5247 (4.3074)  weight_decay: 0.0500 (0.0500)  time: 0.6860  data: 0.0961  max mem: 41794
Epoch: [20]  [280/312]  eta: 0:00:24  lr: 0.004000  min_lr: 0.004000  loss: 4.5324 (4.3034)  weight_decay: 0.0500 (0.0500)  time: 0.6879  data: 0.0812  max mem: 41794
Epoch: [20]  [290/312]  eta: 0:00:16  lr: 0.004000  min_lr: 0.004000  loss: 4.3672 (4.3026)  weight_decay: 0.0500 (0.0500)  time: 0.8299  data: 0.1778  max mem: 41794
Epoch: [20]  [300/312]  eta: 0:00:09  lr: 0.004000  min_lr: 0.004000  loss: 4.5323 (4.3086)  weight_decay: 0.0500 (0.0500)  time: 0.6325  data: 0.0970  max mem: 41794
Epoch: [20]  [310/312]  eta: 0:00:01  lr: 0.004000  min_lr: 0.004000  loss: 4.4819 (4.3114)  weight_decay: 0.0500 (0.0500)  time: 0.4700  data: 0.0001  max mem: 41794
Epoch: [20]  [311/312]  eta: 0:00:00  lr: 0.004000  min_lr: 0.004000  loss: 4.5469 (4.3128)  weight_decay: 0.0500 (0.0500)  time: 0.4681  data: 0.0001  max mem: 41794
Epoch: [20] Total time: 0:03:56 (0.7573 s / it)
Averaged stats: lr: 0.004000  min_lr: 0.004000  loss: 4.5469 (4.2715)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.9109 (1.9109)  acc1: 65.6250 (65.6250)  acc5: 86.4583 (86.4583)  time: 8.7839  data: 8.5861  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.2947 (2.1272)  acc1: 57.6823 (58.5920)  acc5: 79.6875 (81.2800)  time: 1.1356  data: 0.9541  max mem: 41794
Test: Total time: 0:00:10 (1.1874 s / it)
* Acc@1 57.510 Acc@5 81.316 loss 2.146
Accuracy of the model on the 50000 test images: 57.5%
Max accuracy: 59.47%
Epoch: [21]  [  0/312]  eta: 1:21:52  lr: 0.004000  min_lr: 0.004000  loss: 3.6241 (3.6241)  weight_decay: 0.0500 (0.0500)  time: 15.7463  data: 11.6755  max mem: 41794
Epoch: [21]  [ 10/312]  eta: 0:10:40  lr: 0.004000  min_lr: 0.004000  loss: 3.8993 (4.1507)  weight_decay: 0.0500 (0.0500)  time: 2.1213  data: 1.2975  max mem: 41794
Epoch: [21]  [ 20/312]  eta: 0:07:36  lr: 0.004000  min_lr: 0.004000  loss: 4.2517 (4.3416)  weight_decay: 0.0500 (0.0500)  time: 0.8535  data: 0.2133  max mem: 41794
Epoch: [21]  [ 30/312]  eta: 0:05:47  lr: 0.004000  min_lr: 0.004000  loss: 4.4229 (4.2301)  weight_decay: 0.0500 (0.0500)  time: 0.7441  data: 0.0838  max mem: 41794
Epoch: [21]  [ 40/312]  eta: 0:04:54  lr: 0.004000  min_lr: 0.004000  loss: 4.3853 (4.2780)  weight_decay: 0.0500 (0.0500)  time: 0.5781  data: 0.0007  max mem: 41794
Epoch: [21]  [ 50/312]  eta: 0:04:27  lr: 0.004000  min_lr: 0.004000  loss: 4.4383 (4.2665)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.0741  max mem: 41794
Epoch: [21]  [ 60/312]  eta: 0:03:55  lr: 0.004000  min_lr: 0.004000  loss: 4.1940 (4.2204)  weight_decay: 0.0500 (0.0500)  time: 0.6302  data: 0.0752  max mem: 41794
Epoch: [21]  [ 70/312]  eta: 0:03:44  lr: 0.004000  min_lr: 0.004000  loss: 4.3797 (4.2621)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1824  max mem: 41794
Epoch: [21]  [ 80/312]  eta: 0:03:36  lr: 0.004000  min_lr: 0.004000  loss: 4.4017 (4.2320)  weight_decay: 0.0500 (0.0500)  time: 0.9299  data: 0.4196  max mem: 41794
Epoch: [21]  [ 90/312]  eta: 0:03:16  lr: 0.004000  min_lr: 0.004000  loss: 4.4017 (4.2361)  weight_decay: 0.0500 (0.0500)  time: 0.7320  data: 0.2389  max mem: 41794
Epoch: [21]  [100/312]  eta: 0:03:06  lr: 0.004000  min_lr: 0.004000  loss: 4.1751 (4.2299)  weight_decay: 0.0500 (0.0500)  time: 0.6718  data: 0.1772  max mem: 41794
Epoch: [21]  [110/312]  eta: 0:02:50  lr: 0.004000  min_lr: 0.004000  loss: 4.2272 (4.2190)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.1772  max mem: 41794
Epoch: [21]  [120/312]  eta: 0:02:42  lr: 0.004000  min_lr: 0.004000  loss: 4.2272 (4.1989)  weight_decay: 0.0500 (0.0500)  time: 0.6824  data: 0.1860  max mem: 41794
Epoch: [21]  [130/312]  eta: 0:02:34  lr: 0.004000  min_lr: 0.004000  loss: 4.1579 (4.1859)  weight_decay: 0.0500 (0.0500)  time: 0.8773  data: 0.3821  max mem: 41794
Epoch: [21]  [140/312]  eta: 0:02:21  lr: 0.004000  min_lr: 0.004000  loss: 4.1184 (4.1713)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1966  max mem: 41794
Epoch: [21]  [150/312]  eta: 0:02:14  lr: 0.004000  min_lr: 0.004000  loss: 4.4601 (4.1904)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.1785  max mem: 41794
Epoch: [21]  [160/312]  eta: 0:02:06  lr: 0.004000  min_lr: 0.004000  loss: 4.4911 (4.1968)  weight_decay: 0.0500 (0.0500)  time: 0.8553  data: 0.3583  max mem: 41794
Epoch: [21]  [170/312]  eta: 0:01:54  lr: 0.004000  min_lr: 0.004000  loss: 4.4197 (4.2203)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.1843  max mem: 41794
Epoch: [21]  [180/312]  eta: 0:01:47  lr: 0.004000  min_lr: 0.004000  loss: 4.4629 (4.2218)  weight_decay: 0.0500 (0.0500)  time: 0.7130  data: 0.2226  max mem: 41794
Epoch: [21]  [190/312]  eta: 0:01:37  lr: 0.004000  min_lr: 0.004000  loss: 4.1920 (4.2112)  weight_decay: 0.0500 (0.0500)  time: 0.7108  data: 0.2191  max mem: 41794
Epoch: [21]  [200/312]  eta: 0:01:29  lr: 0.004000  min_lr: 0.004000  loss: 3.9460 (4.1931)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1712  max mem: 41794
Epoch: [21]  [210/312]  eta: 0:01:22  lr: 0.004000  min_lr: 0.004000  loss: 3.9460 (4.1913)  weight_decay: 0.0500 (0.0500)  time: 0.8540  data: 0.3537  max mem: 41794
Epoch: [21]  [220/312]  eta: 0:01:12  lr: 0.004000  min_lr: 0.004000  loss: 4.1761 (4.1818)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.1848  max mem: 41794
Epoch: [21]  [230/312]  eta: 0:01:05  lr: 0.004000  min_lr: 0.004000  loss: 4.1780 (4.1867)  weight_decay: 0.0500 (0.0500)  time: 0.7247  data: 0.2242  max mem: 41794
Epoch: [21]  [240/312]  eta: 0:00:57  lr: 0.004000  min_lr: 0.004000  loss: 4.5769 (4.2017)  weight_decay: 0.0500 (0.0500)  time: 0.9010  data: 0.4020  max mem: 41794
Epoch: [21]  [250/312]  eta: 0:00:48  lr: 0.004000  min_lr: 0.004000  loss: 4.5655 (4.2025)  weight_decay: 0.0500 (0.0500)  time: 0.6767  data: 0.1796  max mem: 41794
Epoch: [21]  [260/312]  eta: 0:00:41  lr: 0.004000  min_lr: 0.004000  loss: 4.3579 (4.1984)  weight_decay: 0.0500 (0.0500)  time: 0.6917  data: 0.1907  max mem: 41794
Epoch: [21]  [270/312]  eta: 0:00:32  lr: 0.004000  min_lr: 0.004000  loss: 4.3121 (4.2008)  weight_decay: 0.0500 (0.0500)  time: 0.6959  data: 0.1918  max mem: 41794
Epoch: [21]  [280/312]  eta: 0:00:25  lr: 0.004000  min_lr: 0.004000  loss: 4.0786 (4.1956)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.1730  max mem: 41794
Epoch: [21]  [290/312]  eta: 0:00:17  lr: 0.004000  min_lr: 0.004000  loss: 4.2047 (4.1990)  weight_decay: 0.0500 (0.0500)  time: 0.7884  data: 0.2868  max mem: 41794
Epoch: [21]  [300/312]  eta: 0:00:09  lr: 0.004000  min_lr: 0.004000  loss: 3.9851 (4.1883)  weight_decay: 0.0500 (0.0500)  time: 0.6001  data: 0.1153  max mem: 41794
Epoch: [21]  [310/312]  eta: 0:00:01  lr: 0.003999  min_lr: 0.003999  loss: 3.9087 (4.1870)  weight_decay: 0.0500 (0.0500)  time: 0.4637  data: 0.0001  max mem: 41794
Epoch: [21]  [311/312]  eta: 0:00:00  lr: 0.003999  min_lr: 0.003999  loss: 3.9087 (4.1864)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0001  max mem: 41794
Epoch: [21] Total time: 0:03:59 (0.7662 s / it)
Averaged stats: lr: 0.003999  min_lr: 0.003999  loss: 3.9087 (4.2001)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.8293 (1.8293)  acc1: 65.1042 (65.1042)  acc5: 85.2865 (85.2865)  time: 8.7083  data: 8.4960  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.0353 (1.8984)  acc1: 59.2448 (59.9200)  acc5: 82.9427 (83.1200)  time: 1.1160  data: 0.9441  max mem: 41794
Test: Total time: 0:00:10 (1.1387 s / it)
* Acc@1 59.740 Acc@5 83.102 loss 1.897
Accuracy of the model on the 50000 test images: 59.7%
Max accuracy: 59.74%
Epoch: [22]  [  0/312]  eta: 1:18:03  lr: 0.003999  min_lr: 0.003999  loss: 4.0095 (4.0095)  weight_decay: 0.0500 (0.0500)  time: 15.0110  data: 13.0273  max mem: 41794
Epoch: [22]  [ 10/312]  eta: 0:11:04  lr: 0.003999  min_lr: 0.003999  loss: 4.2747 (4.3714)  weight_decay: 0.0500 (0.0500)  time: 2.2005  data: 1.3786  max mem: 41794
Epoch: [22]  [ 20/312]  eta: 0:07:15  lr: 0.003999  min_lr: 0.003999  loss: 4.2744 (4.2077)  weight_decay: 0.0500 (0.0500)  time: 0.8139  data: 0.2057  max mem: 41794
Epoch: [22]  [ 30/312]  eta: 0:05:32  lr: 0.003999  min_lr: 0.003999  loss: 4.1969 (4.2056)  weight_decay: 0.0500 (0.0500)  time: 0.6197  data: 0.0992  max mem: 41794
Epoch: [22]  [ 40/312]  eta: 0:04:50  lr: 0.003999  min_lr: 0.003999  loss: 4.2460 (4.1521)  weight_decay: 0.0500 (0.0500)  time: 0.6249  data: 0.1096  max mem: 41794
Epoch: [22]  [ 50/312]  eta: 0:04:29  lr: 0.003999  min_lr: 0.003999  loss: 4.1258 (4.1540)  weight_decay: 0.0500 (0.0500)  time: 0.7895  data: 0.2803  max mem: 41794
Epoch: [22]  [ 60/312]  eta: 0:03:56  lr: 0.003999  min_lr: 0.003999  loss: 4.1258 (4.1424)  weight_decay: 0.0500 (0.0500)  time: 0.6704  data: 0.1713  max mem: 41794
Epoch: [22]  [ 70/312]  eta: 0:03:46  lr: 0.003999  min_lr: 0.003999  loss: 4.2134 (4.1697)  weight_decay: 0.0500 (0.0500)  time: 0.6986  data: 0.2055  max mem: 41794
Epoch: [22]  [ 80/312]  eta: 0:03:36  lr: 0.003999  min_lr: 0.003999  loss: 4.2029 (4.1444)  weight_decay: 0.0500 (0.0500)  time: 0.9211  data: 0.4155  max mem: 41794
Epoch: [22]  [ 90/312]  eta: 0:03:16  lr: 0.003999  min_lr: 0.003999  loss: 3.9972 (4.1198)  weight_decay: 0.0500 (0.0500)  time: 0.7056  data: 0.2107  max mem: 41794
Epoch: [22]  [100/312]  eta: 0:03:05  lr: 0.003999  min_lr: 0.003999  loss: 4.2538 (4.1402)  weight_decay: 0.0500 (0.0500)  time: 0.6420  data: 0.1424  max mem: 41794
Epoch: [22]  [110/312]  eta: 0:02:49  lr: 0.003999  min_lr: 0.003999  loss: 4.1873 (4.1275)  weight_decay: 0.0500 (0.0500)  time: 0.6450  data: 0.1424  max mem: 41794
Epoch: [22]  [120/312]  eta: 0:02:41  lr: 0.003999  min_lr: 0.003999  loss: 4.2350 (4.1397)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.1857  max mem: 41794
Epoch: [22]  [130/312]  eta: 0:02:33  lr: 0.003999  min_lr: 0.003999  loss: 4.2350 (4.1347)  weight_decay: 0.0500 (0.0500)  time: 0.8589  data: 0.3693  max mem: 41794
Epoch: [22]  [140/312]  eta: 0:02:20  lr: 0.003999  min_lr: 0.003999  loss: 3.8664 (4.1057)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.1843  max mem: 41794
Epoch: [22]  [150/312]  eta: 0:02:12  lr: 0.003999  min_lr: 0.003999  loss: 3.8585 (4.1106)  weight_decay: 0.0500 (0.0500)  time: 0.6604  data: 0.1656  max mem: 41794
Epoch: [22]  [160/312]  eta: 0:02:04  lr: 0.003999  min_lr: 0.003999  loss: 4.3612 (4.1100)  weight_decay: 0.0500 (0.0500)  time: 0.8446  data: 0.3508  max mem: 41794
Epoch: [22]  [170/312]  eta: 0:01:54  lr: 0.003999  min_lr: 0.003999  loss: 4.3150 (4.1172)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.2001  max mem: 41794
Epoch: [22]  [180/312]  eta: 0:01:46  lr: 0.003999  min_lr: 0.003999  loss: 4.2128 (4.1087)  weight_decay: 0.0500 (0.0500)  time: 0.7047  data: 0.2088  max mem: 41794
Epoch: [22]  [190/312]  eta: 0:01:36  lr: 0.003999  min_lr: 0.003999  loss: 4.2040 (4.1070)  weight_decay: 0.0500 (0.0500)  time: 0.6945  data: 0.1946  max mem: 41794
Epoch: [22]  [200/312]  eta: 0:01:28  lr: 0.003999  min_lr: 0.003999  loss: 4.5116 (4.1225)  weight_decay: 0.0500 (0.0500)  time: 0.6590  data: 0.1611  max mem: 41794
Epoch: [22]  [210/312]  eta: 0:01:21  lr: 0.003999  min_lr: 0.003999  loss: 4.3460 (4.1182)  weight_decay: 0.0500 (0.0500)  time: 0.8675  data: 0.3734  max mem: 41794
Epoch: [22]  [220/312]  eta: 0:01:12  lr: 0.003999  min_lr: 0.003999  loss: 4.0983 (4.1068)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.2129  max mem: 41794
Epoch: [22]  [230/312]  eta: 0:01:04  lr: 0.003999  min_lr: 0.003999  loss: 4.1793 (4.1128)  weight_decay: 0.0500 (0.0500)  time: 0.7022  data: 0.1937  max mem: 41794
Epoch: [22]  [240/312]  eta: 0:00:56  lr: 0.003999  min_lr: 0.003999  loss: 4.2874 (4.1172)  weight_decay: 0.0500 (0.0500)  time: 0.7985  data: 0.2764  max mem: 41794
Epoch: [22]  [250/312]  eta: 0:00:48  lr: 0.003999  min_lr: 0.003999  loss: 3.9273 (4.1102)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.1779  max mem: 41794
Epoch: [22]  [260/312]  eta: 0:00:40  lr: 0.003999  min_lr: 0.003999  loss: 3.8833 (4.1098)  weight_decay: 0.0500 (0.0500)  time: 0.7898  data: 0.2903  max mem: 41794
Epoch: [22]  [270/312]  eta: 0:00:32  lr: 0.003999  min_lr: 0.003999  loss: 4.0828 (4.1164)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.1958  max mem: 41794
Epoch: [22]  [280/312]  eta: 0:00:24  lr: 0.003999  min_lr: 0.003999  loss: 4.3042 (4.1156)  weight_decay: 0.0500 (0.0500)  time: 0.7003  data: 0.1585  max mem: 41794
Epoch: [22]  [290/312]  eta: 0:00:17  lr: 0.003999  min_lr: 0.003999  loss: 4.5221 (4.1363)  weight_decay: 0.0500 (0.0500)  time: 0.8266  data: 0.2695  max mem: 41794
Epoch: [22]  [300/312]  eta: 0:00:09  lr: 0.003999  min_lr: 0.003999  loss: 4.6265 (4.1491)  weight_decay: 0.0500 (0.0500)  time: 0.6138  data: 0.1114  max mem: 41794
Epoch: [22]  [310/312]  eta: 0:00:01  lr: 0.003999  min_lr: 0.003999  loss: 4.4631 (4.1537)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.0001  max mem: 41794
Epoch: [22]  [311/312]  eta: 0:00:00  lr: 0.003999  min_lr: 0.003999  loss: 4.4631 (4.1564)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [22] Total time: 0:03:58 (0.7653 s / it)
Averaged stats: lr: 0.003999  min_lr: 0.003999  loss: 4.4631 (4.1853)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.7974 (1.7974)  acc1: 64.7135 (64.7135)  acc5: 85.9375 (85.9375)  time: 8.0699  data: 7.8704  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.1965 (2.0080)  acc1: 57.0312 (58.7680)  acc5: 80.5990 (82.4480)  time: 1.0629  data: 0.8920  max mem: 41794
Test: Total time: 0:00:09 (1.0725 s / it)
* Acc@1 58.726 Acc@5 82.700 loss 2.016
Accuracy of the model on the 50000 test images: 58.7%
Max accuracy: 59.74%
Epoch: [23]  [  0/312]  eta: 1:22:50  lr: 0.003999  min_lr: 0.003999  loss: 4.3126 (4.3126)  weight_decay: 0.0500 (0.0500)  time: 15.9314  data: 15.3232  max mem: 41794
Epoch: [23]  [ 10/312]  eta: 0:11:07  lr: 0.003999  min_lr: 0.003999  loss: 3.9207 (3.7303)  weight_decay: 0.0500 (0.0500)  time: 2.2101  data: 1.3937  max mem: 41794
Epoch: [23]  [ 20/312]  eta: 0:07:20  lr: 0.003999  min_lr: 0.003999  loss: 4.1601 (4.0011)  weight_decay: 0.0500 (0.0500)  time: 0.7877  data: 0.1008  max mem: 41794
Epoch: [23]  [ 30/312]  eta: 0:05:34  lr: 0.003999  min_lr: 0.003999  loss: 4.1912 (3.9968)  weight_decay: 0.0500 (0.0500)  time: 0.6255  data: 0.1065  max mem: 41794
Epoch: [23]  [ 40/312]  eta: 0:04:49  lr: 0.003999  min_lr: 0.003999  loss: 4.1912 (4.0598)  weight_decay: 0.0500 (0.0500)  time: 0.6012  data: 0.0698  max mem: 41794
Epoch: [23]  [ 50/312]  eta: 0:04:28  lr: 0.003999  min_lr: 0.003999  loss: 4.2277 (4.0839)  weight_decay: 0.0500 (0.0500)  time: 0.7763  data: 0.1903  max mem: 41794
Epoch: [23]  [ 60/312]  eta: 0:03:56  lr: 0.003999  min_lr: 0.003999  loss: 4.2922 (4.1137)  weight_decay: 0.0500 (0.0500)  time: 0.6740  data: 0.1269  max mem: 41794
Epoch: [23]  [ 70/312]  eta: 0:03:45  lr: 0.003999  min_lr: 0.003999  loss: 4.2372 (4.1222)  weight_decay: 0.0500 (0.0500)  time: 0.6900  data: 0.0991  max mem: 41794
Epoch: [23]  [ 80/312]  eta: 0:03:34  lr: 0.003999  min_lr: 0.003999  loss: 4.2496 (4.1382)  weight_decay: 0.0500 (0.0500)  time: 0.8774  data: 0.2052  max mem: 41794
Epoch: [23]  [ 90/312]  eta: 0:03:14  lr: 0.003999  min_lr: 0.003999  loss: 4.5421 (4.1845)  weight_decay: 0.0500 (0.0500)  time: 0.6845  data: 0.1183  max mem: 41794
Epoch: [23]  [100/312]  eta: 0:03:06  lr: 0.003999  min_lr: 0.003999  loss: 4.5421 (4.1885)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1203  max mem: 41794
Epoch: [23]  [110/312]  eta: 0:02:50  lr: 0.003999  min_lr: 0.003999  loss: 3.8547 (4.1512)  weight_decay: 0.0500 (0.0500)  time: 0.7042  data: 0.1142  max mem: 41794
Epoch: [23]  [120/312]  eta: 0:02:43  lr: 0.003999  min_lr: 0.003999  loss: 4.1900 (4.1509)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1049  max mem: 41794
Epoch: [23]  [130/312]  eta: 0:02:35  lr: 0.003999  min_lr: 0.003999  loss: 4.2436 (4.1564)  weight_decay: 0.0500 (0.0500)  time: 0.8974  data: 0.1711  max mem: 41794
Epoch: [23]  [140/312]  eta: 0:02:22  lr: 0.003999  min_lr: 0.003999  loss: 4.0272 (4.1608)  weight_decay: 0.0500 (0.0500)  time: 0.7007  data: 0.0723  max mem: 41794
Epoch: [23]  [150/312]  eta: 0:02:14  lr: 0.003998  min_lr: 0.003998  loss: 4.0167 (4.1526)  weight_decay: 0.0500 (0.0500)  time: 0.6768  data: 0.0724  max mem: 41794
Epoch: [23]  [160/312]  eta: 0:02:05  lr: 0.003998  min_lr: 0.003998  loss: 4.0167 (4.1418)  weight_decay: 0.0500 (0.0500)  time: 0.8280  data: 0.0786  max mem: 41794
Epoch: [23]  [170/312]  eta: 0:01:55  lr: 0.003998  min_lr: 0.003998  loss: 4.2645 (4.1635)  weight_decay: 0.0500 (0.0500)  time: 0.6638  data: 0.0324  max mem: 41794
Epoch: [23]  [180/312]  eta: 0:01:47  lr: 0.003998  min_lr: 0.003998  loss: 4.4092 (4.1637)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.0726  max mem: 41794
Epoch: [23]  [190/312]  eta: 0:01:37  lr: 0.003998  min_lr: 0.003998  loss: 4.1795 (4.1585)  weight_decay: 0.0500 (0.0500)  time: 0.6708  data: 0.0486  max mem: 41794
Epoch: [23]  [200/312]  eta: 0:01:29  lr: 0.003998  min_lr: 0.003998  loss: 4.1795 (4.1614)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.0472  max mem: 41794
Epoch: [23]  [210/312]  eta: 0:01:21  lr: 0.003998  min_lr: 0.003998  loss: 4.3153 (4.1618)  weight_decay: 0.0500 (0.0500)  time: 0.8396  data: 0.0853  max mem: 41794
Epoch: [23]  [220/312]  eta: 0:01:12  lr: 0.003998  min_lr: 0.003998  loss: 4.2067 (4.1583)  weight_decay: 0.0500 (0.0500)  time: 0.6710  data: 0.0521  max mem: 41794
Epoch: [23]  [230/312]  eta: 0:01:04  lr: 0.003998  min_lr: 0.003998  loss: 4.0350 (4.1467)  weight_decay: 0.0500 (0.0500)  time: 0.7018  data: 0.0795  max mem: 41794
Epoch: [23]  [240/312]  eta: 0:00:57  lr: 0.003998  min_lr: 0.003998  loss: 4.2015 (4.1458)  weight_decay: 0.0500 (0.0500)  time: 0.8353  data: 0.0714  max mem: 41794
Epoch: [23]  [250/312]  eta: 0:00:48  lr: 0.003998  min_lr: 0.003998  loss: 4.3159 (4.1520)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.0576  max mem: 41794
Epoch: [23]  [260/312]  eta: 0:00:40  lr: 0.003998  min_lr: 0.003998  loss: 4.1777 (4.1436)  weight_decay: 0.0500 (0.0500)  time: 0.7388  data: 0.1224  max mem: 41794
Epoch: [23]  [270/312]  eta: 0:00:32  lr: 0.003998  min_lr: 0.003998  loss: 3.9756 (4.1379)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.0695  max mem: 41794
Epoch: [23]  [280/312]  eta: 0:00:24  lr: 0.003998  min_lr: 0.003998  loss: 4.0418 (4.1354)  weight_decay: 0.0500 (0.0500)  time: 0.6684  data: 0.0542  max mem: 41794
Epoch: [23]  [290/312]  eta: 0:00:17  lr: 0.003998  min_lr: 0.003998  loss: 4.3210 (4.1461)  weight_decay: 0.0500 (0.0500)  time: 0.8043  data: 0.0978  max mem: 41794
Epoch: [23]  [300/312]  eta: 0:00:09  lr: 0.003998  min_lr: 0.003998  loss: 4.3210 (4.1480)  weight_decay: 0.0500 (0.0500)  time: 0.6254  data: 0.0442  max mem: 41794
Epoch: [23]  [310/312]  eta: 0:00:01  lr: 0.003998  min_lr: 0.003998  loss: 4.1225 (4.1473)  weight_decay: 0.0500 (0.0500)  time: 0.4624  data: 0.0001  max mem: 41794
Epoch: [23]  [311/312]  eta: 0:00:00  lr: 0.003998  min_lr: 0.003998  loss: 4.0980 (4.1437)  weight_decay: 0.0500 (0.0500)  time: 0.4618  data: 0.0001  max mem: 41794
Epoch: [23] Total time: 0:03:58 (0.7653 s / it)
Averaged stats: lr: 0.003998  min_lr: 0.003998  loss: 4.0980 (4.1531)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.6106 (1.6106)  acc1: 67.8385 (67.8385)  acc5: 89.3229 (89.3229)  time: 8.5042  data: 8.2922  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8916 (1.7899)  acc1: 61.3281 (61.2800)  acc5: 83.9844 (85.0080)  time: 1.0930  data: 0.9214  max mem: 41794
Test: Total time: 0:00:10 (1.1223 s / it)
* Acc@1 61.526 Acc@5 84.588 loss 1.796
Accuracy of the model on the 50000 test images: 61.5%
Max accuracy: 61.53%
Epoch: [24]  [  0/312]  eta: 1:24:49  lr: 0.003998  min_lr: 0.003998  loss: 4.5381 (4.5381)  weight_decay: 0.0500 (0.0500)  time: 16.3115  data: 15.8280  max mem: 41794
Epoch: [24]  [ 10/312]  eta: 0:10:31  lr: 0.003998  min_lr: 0.003998  loss: 4.4020 (4.2015)  weight_decay: 0.0500 (0.0500)  time: 2.0913  data: 1.4396  max mem: 41794
Epoch: [24]  [ 20/312]  eta: 0:07:16  lr: 0.003998  min_lr: 0.003998  loss: 4.0850 (4.0580)  weight_decay: 0.0500 (0.0500)  time: 0.7557  data: 0.1564  max mem: 41794
Epoch: [24]  [ 30/312]  eta: 0:05:32  lr: 0.003998  min_lr: 0.003998  loss: 3.9716 (4.0704)  weight_decay: 0.0500 (0.0500)  time: 0.6758  data: 0.1600  max mem: 41794
Epoch: [24]  [ 40/312]  eta: 0:04:50  lr: 0.003998  min_lr: 0.003998  loss: 4.1380 (4.0607)  weight_decay: 0.0500 (0.0500)  time: 0.6213  data: 0.1270  max mem: 41794
Epoch: [24]  [ 50/312]  eta: 0:04:27  lr: 0.003998  min_lr: 0.003998  loss: 4.0401 (4.0365)  weight_decay: 0.0500 (0.0500)  time: 0.7815  data: 0.2712  max mem: 41794
Epoch: [24]  [ 60/312]  eta: 0:03:55  lr: 0.003998  min_lr: 0.003998  loss: 4.0779 (4.1011)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.1485  max mem: 41794
Epoch: [24]  [ 70/312]  eta: 0:03:43  lr: 0.003998  min_lr: 0.003998  loss: 4.1443 (4.0608)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1632  max mem: 41794
Epoch: [24]  [ 80/312]  eta: 0:03:32  lr: 0.003998  min_lr: 0.003998  loss: 3.8707 (4.0210)  weight_decay: 0.0500 (0.0500)  time: 0.8656  data: 0.3180  max mem: 41794
Epoch: [24]  [ 90/312]  eta: 0:03:13  lr: 0.003998  min_lr: 0.003998  loss: 4.0064 (4.0391)  weight_decay: 0.0500 (0.0500)  time: 0.6963  data: 0.1686  max mem: 41794
Epoch: [24]  [100/312]  eta: 0:03:07  lr: 0.003998  min_lr: 0.003998  loss: 4.3070 (4.0590)  weight_decay: 0.0500 (0.0500)  time: 0.7375  data: 0.1762  max mem: 41794
Epoch: [24]  [110/312]  eta: 0:02:51  lr: 0.003998  min_lr: 0.003998  loss: 4.2351 (4.0560)  weight_decay: 0.0500 (0.0500)  time: 0.7352  data: 0.1630  max mem: 41794
Epoch: [24]  [120/312]  eta: 0:02:42  lr: 0.003998  min_lr: 0.003998  loss: 4.2351 (4.0666)  weight_decay: 0.0500 (0.0500)  time: 0.6665  data: 0.1340  max mem: 41794
Epoch: [24]  [130/312]  eta: 0:02:33  lr: 0.003998  min_lr: 0.003998  loss: 4.3945 (4.0654)  weight_decay: 0.0500 (0.0500)  time: 0.8198  data: 0.2530  max mem: 41794
Epoch: [24]  [140/312]  eta: 0:02:21  lr: 0.003998  min_lr: 0.003998  loss: 4.3409 (4.0752)  weight_decay: 0.0500 (0.0500)  time: 0.6657  data: 0.1307  max mem: 41794
Epoch: [24]  [150/312]  eta: 0:02:13  lr: 0.003997  min_lr: 0.003997  loss: 4.3285 (4.0747)  weight_decay: 0.0500 (0.0500)  time: 0.6804  data: 0.1832  max mem: 41794
Epoch: [24]  [160/312]  eta: 0:02:05  lr: 0.003997  min_lr: 0.003997  loss: 4.0746 (4.0783)  weight_decay: 0.0500 (0.0500)  time: 0.8595  data: 0.3615  max mem: 41794
Epoch: [24]  [170/312]  eta: 0:01:54  lr: 0.003997  min_lr: 0.003997  loss: 4.2047 (4.0821)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.1900  max mem: 41794
Epoch: [24]  [180/312]  eta: 0:01:47  lr: 0.003997  min_lr: 0.003997  loss: 4.4048 (4.1020)  weight_decay: 0.0500 (0.0500)  time: 0.6892  data: 0.1900  max mem: 41794
Epoch: [24]  [190/312]  eta: 0:01:36  lr: 0.003997  min_lr: 0.003997  loss: 4.3424 (4.0891)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.1902  max mem: 41794
Epoch: [24]  [200/312]  eta: 0:01:29  lr: 0.003997  min_lr: 0.003997  loss: 4.0318 (4.0925)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1676  max mem: 41794
Epoch: [24]  [210/312]  eta: 0:01:21  lr: 0.003997  min_lr: 0.003997  loss: 4.1600 (4.1024)  weight_decay: 0.0500 (0.0500)  time: 0.8581  data: 0.3552  max mem: 41794
Epoch: [24]  [220/312]  eta: 0:01:12  lr: 0.003997  min_lr: 0.003997  loss: 4.4398 (4.1156)  weight_decay: 0.0500 (0.0500)  time: 0.6909  data: 0.1883  max mem: 41794
Epoch: [24]  [230/312]  eta: 0:01:04  lr: 0.003997  min_lr: 0.003997  loss: 4.0917 (4.0982)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1782  max mem: 41794
Epoch: [24]  [240/312]  eta: 0:00:57  lr: 0.003997  min_lr: 0.003997  loss: 3.7744 (4.0872)  weight_decay: 0.0500 (0.0500)  time: 0.8455  data: 0.3503  max mem: 41794
Epoch: [24]  [250/312]  eta: 0:00:48  lr: 0.003997  min_lr: 0.003997  loss: 4.0484 (4.0896)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.1863  max mem: 41794
Epoch: [24]  [260/312]  eta: 0:00:40  lr: 0.003997  min_lr: 0.003997  loss: 4.0474 (4.0872)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.2052  max mem: 41794
Epoch: [24]  [270/312]  eta: 0:00:32  lr: 0.003997  min_lr: 0.003997  loss: 3.9523 (4.0861)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.1917  max mem: 41794
Epoch: [24]  [280/312]  eta: 0:00:24  lr: 0.003997  min_lr: 0.003997  loss: 4.1637 (4.0854)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.1870  max mem: 41794
Epoch: [24]  [290/312]  eta: 0:00:17  lr: 0.003997  min_lr: 0.003997  loss: 4.1637 (4.0849)  weight_decay: 0.0500 (0.0500)  time: 0.8138  data: 0.3107  max mem: 41794
Epoch: [24]  [300/312]  eta: 0:00:09  lr: 0.003997  min_lr: 0.003997  loss: 4.0588 (4.0892)  weight_decay: 0.0500 (0.0500)  time: 0.6158  data: 0.1240  max mem: 41794
Epoch: [24]  [310/312]  eta: 0:00:01  lr: 0.003997  min_lr: 0.003997  loss: 4.0588 (4.0884)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [24]  [311/312]  eta: 0:00:00  lr: 0.003997  min_lr: 0.003997  loss: 4.0588 (4.0846)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [24] Total time: 0:03:57 (0.7623 s / it)
Averaged stats: lr: 0.003997  min_lr: 0.003997  loss: 4.0588 (4.1327)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.6655 (1.6655)  acc1: 66.5365 (66.5365)  acc5: 87.6302 (87.6302)  time: 8.7735  data: 8.5606  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.0638 (1.9322)  acc1: 59.6354 (59.5520)  acc5: 83.5938 (83.9200)  time: 1.1228  data: 0.9513  max mem: 41794
Test: Total time: 0:00:10 (1.1359 s / it)
* Acc@1 60.022 Acc@5 83.812 loss 1.941
Accuracy of the model on the 50000 test images: 60.0%
Max accuracy: 61.53%
Epoch: [25]  [  0/312]  eta: 1:19:30  lr: 0.003997  min_lr: 0.003997  loss: 4.9367 (4.9367)  weight_decay: 0.0500 (0.0500)  time: 15.2913  data: 13.8242  max mem: 41794
Epoch: [25]  [ 10/312]  eta: 0:10:07  lr: 0.003997  min_lr: 0.003997  loss: 4.1898 (4.2411)  weight_decay: 0.0500 (0.0500)  time: 2.0120  data: 1.3525  max mem: 41794
Epoch: [25]  [ 20/312]  eta: 0:06:57  lr: 0.003997  min_lr: 0.003997  loss: 4.1869 (4.2331)  weight_decay: 0.0500 (0.0500)  time: 0.7370  data: 0.1790  max mem: 41794
Epoch: [25]  [ 30/312]  eta: 0:05:15  lr: 0.003997  min_lr: 0.003997  loss: 4.1477 (4.1140)  weight_decay: 0.0500 (0.0500)  time: 0.6298  data: 0.1266  max mem: 41794
Epoch: [25]  [ 40/312]  eta: 0:04:56  lr: 0.003997  min_lr: 0.003997  loss: 4.0494 (4.0977)  weight_decay: 0.0500 (0.0500)  time: 0.7348  data: 0.2631  max mem: 41794
Epoch: [25]  [ 50/312]  eta: 0:04:38  lr: 0.003997  min_lr: 0.003997  loss: 4.4317 (4.1418)  weight_decay: 0.0500 (0.0500)  time: 0.9729  data: 0.4982  max mem: 41794
Epoch: [25]  [ 60/312]  eta: 0:04:03  lr: 0.003997  min_lr: 0.003997  loss: 4.2950 (4.1044)  weight_decay: 0.0500 (0.0500)  time: 0.7111  data: 0.2356  max mem: 41794
Epoch: [25]  [ 70/312]  eta: 0:03:51  lr: 0.003997  min_lr: 0.003997  loss: 4.2563 (4.1443)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.2011  max mem: 41794
Epoch: [25]  [ 80/312]  eta: 0:03:43  lr: 0.003997  min_lr: 0.003997  loss: 4.3023 (4.1462)  weight_decay: 0.0500 (0.0500)  time: 0.9519  data: 0.4693  max mem: 41794
Epoch: [25]  [ 90/312]  eta: 0:03:21  lr: 0.003996  min_lr: 0.003996  loss: 4.2299 (4.1378)  weight_decay: 0.0500 (0.0500)  time: 0.7461  data: 0.2687  max mem: 41794
Epoch: [25]  [100/312]  eta: 0:03:12  lr: 0.003996  min_lr: 0.003996  loss: 4.2487 (4.1459)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.2042  max mem: 41794
Epoch: [25]  [110/312]  eta: 0:02:55  lr: 0.003996  min_lr: 0.003996  loss: 4.2504 (4.1594)  weight_decay: 0.0500 (0.0500)  time: 0.6861  data: 0.2042  max mem: 41794
Epoch: [25]  [120/312]  eta: 0:02:47  lr: 0.003996  min_lr: 0.003996  loss: 4.0781 (4.1350)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.2044  max mem: 41794
Epoch: [25]  [130/312]  eta: 0:02:39  lr: 0.003996  min_lr: 0.003996  loss: 3.9806 (4.1293)  weight_decay: 0.0500 (0.0500)  time: 0.9037  data: 0.4197  max mem: 41794
Epoch: [25]  [140/312]  eta: 0:02:25  lr: 0.003996  min_lr: 0.003996  loss: 4.2170 (4.1321)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.2158  max mem: 41794
Epoch: [25]  [150/312]  eta: 0:02:17  lr: 0.003996  min_lr: 0.003996  loss: 4.2595 (4.1381)  weight_decay: 0.0500 (0.0500)  time: 0.6679  data: 0.1824  max mem: 41794
Epoch: [25]  [160/312]  eta: 0:02:09  lr: 0.003996  min_lr: 0.003996  loss: 4.2250 (4.1254)  weight_decay: 0.0500 (0.0500)  time: 0.8904  data: 0.4065  max mem: 41794
Epoch: [25]  [170/312]  eta: 0:01:57  lr: 0.003996  min_lr: 0.003996  loss: 3.8952 (4.1005)  weight_decay: 0.0500 (0.0500)  time: 0.7018  data: 0.2247  max mem: 41794
Epoch: [25]  [180/312]  eta: 0:01:50  lr: 0.003996  min_lr: 0.003996  loss: 3.8681 (4.1018)  weight_decay: 0.0500 (0.0500)  time: 0.7038  data: 0.2274  max mem: 41794
Epoch: [25]  [190/312]  eta: 0:01:39  lr: 0.003996  min_lr: 0.003996  loss: 4.0546 (4.0983)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.2275  max mem: 41794
Epoch: [25]  [200/312]  eta: 0:01:31  lr: 0.003996  min_lr: 0.003996  loss: 4.0546 (4.0960)  weight_decay: 0.0500 (0.0500)  time: 0.6881  data: 0.2046  max mem: 41794
Epoch: [25]  [210/312]  eta: 0:01:23  lr: 0.003996  min_lr: 0.003996  loss: 3.9549 (4.0915)  weight_decay: 0.0500 (0.0500)  time: 0.8829  data: 0.3816  max mem: 41794
Epoch: [25]  [220/312]  eta: 0:01:14  lr: 0.003996  min_lr: 0.003996  loss: 4.0414 (4.0958)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1776  max mem: 41794
Epoch: [25]  [230/312]  eta: 0:01:06  lr: 0.003996  min_lr: 0.003996  loss: 3.9206 (4.0876)  weight_decay: 0.0500 (0.0500)  time: 0.7069  data: 0.1992  max mem: 41794
Epoch: [25]  [240/312]  eta: 0:00:58  lr: 0.003996  min_lr: 0.003996  loss: 4.0690 (4.0893)  weight_decay: 0.0500 (0.0500)  time: 0.8665  data: 0.3521  max mem: 41794
Epoch: [25]  [250/312]  eta: 0:00:49  lr: 0.003996  min_lr: 0.003996  loss: 4.0690 (4.0771)  weight_decay: 0.0500 (0.0500)  time: 0.6478  data: 0.1535  max mem: 41794
Epoch: [25]  [260/312]  eta: 0:00:41  lr: 0.003996  min_lr: 0.003996  loss: 4.0558 (4.0719)  weight_decay: 0.0500 (0.0500)  time: 0.7311  data: 0.1988  max mem: 41794
Epoch: [25]  [270/312]  eta: 0:00:33  lr: 0.003996  min_lr: 0.003996  loss: 4.2971 (4.0812)  weight_decay: 0.0500 (0.0500)  time: 0.7301  data: 0.1988  max mem: 41794
Epoch: [25]  [280/312]  eta: 0:00:25  lr: 0.003996  min_lr: 0.003996  loss: 4.3604 (4.0838)  weight_decay: 0.0500 (0.0500)  time: 0.6731  data: 0.1592  max mem: 41794
Epoch: [25]  [290/312]  eta: 0:00:17  lr: 0.003996  min_lr: 0.003996  loss: 4.3585 (4.0912)  weight_decay: 0.0500 (0.0500)  time: 0.8055  data: 0.2866  max mem: 41794
Epoch: [25]  [300/312]  eta: 0:00:09  lr: 0.003996  min_lr: 0.003996  loss: 4.2607 (4.0914)  weight_decay: 0.0500 (0.0500)  time: 0.6182  data: 0.1277  max mem: 41794
Epoch: [25]  [310/312]  eta: 0:00:01  lr: 0.003995  min_lr: 0.003995  loss: 4.1346 (4.0868)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [25]  [311/312]  eta: 0:00:00  lr: 0.003995  min_lr: 0.003995  loss: 4.0079 (4.0865)  weight_decay: 0.0500 (0.0500)  time: 0.4656  data: 0.0001  max mem: 41794
Epoch: [25] Total time: 0:04:03 (0.7796 s / it)
Averaged stats: lr: 0.003995  min_lr: 0.003995  loss: 4.0079 (4.1149)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:25  loss: 1.4569 (1.4569)  acc1: 71.7448 (71.7448)  acc5: 89.8438 (89.8438)  time: 9.4759  data: 9.2701  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9649 (1.8165)  acc1: 62.8906 (62.1600)  acc5: 83.7240 (85.1040)  time: 1.2022  data: 1.0301  max mem: 41794
Test: Total time: 0:00:11 (1.2448 s / it)
* Acc@1 62.172 Acc@5 85.034 loss 1.819
Accuracy of the model on the 50000 test images: 62.2%
Max accuracy: 62.17%
Epoch: [26]  [  0/312]  eta: 1:16:36  lr: 0.003995  min_lr: 0.003995  loss: 4.5491 (4.5491)  weight_decay: 0.0500 (0.0500)  time: 14.7318  data: 13.6273  max mem: 41794
Epoch: [26]  [ 10/312]  eta: 0:09:49  lr: 0.003995  min_lr: 0.003995  loss: 4.2040 (4.1244)  weight_decay: 0.0500 (0.0500)  time: 1.9516  data: 1.2986  max mem: 41794
Epoch: [26]  [ 20/312]  eta: 0:06:52  lr: 0.003995  min_lr: 0.003995  loss: 4.0990 (4.0971)  weight_decay: 0.0500 (0.0500)  time: 0.7464  data: 0.1545  max mem: 41794
Epoch: [26]  [ 30/312]  eta: 0:05:13  lr: 0.003995  min_lr: 0.003995  loss: 3.9607 (4.0945)  weight_decay: 0.0500 (0.0500)  time: 0.6497  data: 0.1219  max mem: 41794
Epoch: [26]  [ 40/312]  eta: 0:04:54  lr: 0.003995  min_lr: 0.003995  loss: 4.0508 (4.0673)  weight_decay: 0.0500 (0.0500)  time: 0.7352  data: 0.1384  max mem: 41794
Epoch: [26]  [ 50/312]  eta: 0:04:38  lr: 0.003995  min_lr: 0.003995  loss: 3.9559 (4.0398)  weight_decay: 0.0500 (0.0500)  time: 0.9897  data: 0.2498  max mem: 41794
Epoch: [26]  [ 60/312]  eta: 0:04:05  lr: 0.003995  min_lr: 0.003995  loss: 4.1991 (4.0536)  weight_decay: 0.0500 (0.0500)  time: 0.7539  data: 0.1134  max mem: 41794
Epoch: [26]  [ 70/312]  eta: 0:03:48  lr: 0.003995  min_lr: 0.003995  loss: 4.1950 (4.0512)  weight_decay: 0.0500 (0.0500)  time: 0.6335  data: 0.1116  max mem: 41794
Epoch: [26]  [ 80/312]  eta: 0:03:35  lr: 0.003995  min_lr: 0.003995  loss: 4.2487 (4.0574)  weight_decay: 0.0500 (0.0500)  time: 0.7896  data: 0.1856  max mem: 41794
Epoch: [26]  [ 90/312]  eta: 0:03:20  lr: 0.003995  min_lr: 0.003995  loss: 4.2437 (4.0574)  weight_decay: 0.0500 (0.0500)  time: 0.7710  data: 0.1886  max mem: 41794
Epoch: [26]  [100/312]  eta: 0:03:12  lr: 0.003995  min_lr: 0.003995  loss: 4.0820 (4.0612)  weight_decay: 0.0500 (0.0500)  time: 0.8294  data: 0.2583  max mem: 41794
Epoch: [26]  [110/312]  eta: 0:02:56  lr: 0.003995  min_lr: 0.003995  loss: 4.1627 (4.0583)  weight_decay: 0.0500 (0.0500)  time: 0.7399  data: 0.1457  max mem: 41794
Epoch: [26]  [120/312]  eta: 0:02:47  lr: 0.003995  min_lr: 0.003995  loss: 3.7336 (4.0325)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.1289  max mem: 41794
Epoch: [26]  [130/312]  eta: 0:02:39  lr: 0.003995  min_lr: 0.003995  loss: 3.7336 (4.0290)  weight_decay: 0.0500 (0.0500)  time: 0.8673  data: 0.2358  max mem: 41794
Epoch: [26]  [140/312]  eta: 0:02:25  lr: 0.003995  min_lr: 0.003995  loss: 4.2435 (4.0389)  weight_decay: 0.0500 (0.0500)  time: 0.6915  data: 0.1075  max mem: 41794
Epoch: [26]  [150/312]  eta: 0:02:18  lr: 0.003995  min_lr: 0.003995  loss: 4.3067 (4.0545)  weight_decay: 0.0500 (0.0500)  time: 0.7203  data: 0.1384  max mem: 41794
Epoch: [26]  [160/312]  eta: 0:02:07  lr: 0.003995  min_lr: 0.003995  loss: 4.4555 (4.0776)  weight_decay: 0.0500 (0.0500)  time: 0.7885  data: 0.1402  max mem: 41794
Epoch: [26]  [170/312]  eta: 0:01:57  lr: 0.003995  min_lr: 0.003995  loss: 4.3197 (4.0787)  weight_decay: 0.0500 (0.0500)  time: 0.6433  data: 0.0795  max mem: 41794
Epoch: [26]  [180/312]  eta: 0:01:49  lr: 0.003995  min_lr: 0.003995  loss: 4.1407 (4.0886)  weight_decay: 0.0500 (0.0500)  time: 0.7734  data: 0.1942  max mem: 41794
Epoch: [26]  [190/312]  eta: 0:01:39  lr: 0.003995  min_lr: 0.003995  loss: 3.9827 (4.0777)  weight_decay: 0.0500 (0.0500)  time: 0.6965  data: 0.1172  max mem: 41794
Epoch: [26]  [200/312]  eta: 0:01:31  lr: 0.003994  min_lr: 0.003994  loss: 3.8825 (4.0710)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.0821  max mem: 41794
Epoch: [26]  [210/312]  eta: 0:01:23  lr: 0.003994  min_lr: 0.003994  loss: 4.2022 (4.0734)  weight_decay: 0.0500 (0.0500)  time: 0.8219  data: 0.1547  max mem: 41794
Epoch: [26]  [220/312]  eta: 0:01:13  lr: 0.003994  min_lr: 0.003994  loss: 4.2022 (4.0693)  weight_decay: 0.0500 (0.0500)  time: 0.6708  data: 0.0733  max mem: 41794
Epoch: [26]  [230/312]  eta: 0:01:06  lr: 0.003994  min_lr: 0.003994  loss: 4.1529 (4.0740)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.0873  max mem: 41794
Epoch: [26]  [240/312]  eta: 0:00:57  lr: 0.003994  min_lr: 0.003994  loss: 4.1115 (4.0658)  weight_decay: 0.0500 (0.0500)  time: 0.7976  data: 0.0915  max mem: 41794
Epoch: [26]  [250/312]  eta: 0:00:49  lr: 0.003994  min_lr: 0.003994  loss: 4.2514 (4.0780)  weight_decay: 0.0500 (0.0500)  time: 0.6553  data: 0.0557  max mem: 41794
Epoch: [26]  [260/312]  eta: 0:00:41  lr: 0.003994  min_lr: 0.003994  loss: 4.4082 (4.0731)  weight_decay: 0.0500 (0.0500)  time: 0.7072  data: 0.1214  max mem: 41794
Epoch: [26]  [270/312]  eta: 0:00:32  lr: 0.003994  min_lr: 0.003994  loss: 4.1304 (4.0748)  weight_decay: 0.0500 (0.0500)  time: 0.6566  data: 0.0729  max mem: 41794
Epoch: [26]  [280/312]  eta: 0:00:25  lr: 0.003994  min_lr: 0.003994  loss: 4.1304 (4.0810)  weight_decay: 0.0500 (0.0500)  time: 0.7465  data: 0.1135  max mem: 41794
Epoch: [26]  [290/312]  eta: 0:00:17  lr: 0.003994  min_lr: 0.003994  loss: 4.1009 (4.0825)  weight_decay: 0.0500 (0.0500)  time: 0.8300  data: 0.1332  max mem: 41794
Epoch: [26]  [300/312]  eta: 0:00:09  lr: 0.003994  min_lr: 0.003994  loss: 4.1576 (4.0854)  weight_decay: 0.0500 (0.0500)  time: 0.5675  data: 0.0224  max mem: 41794
Epoch: [26]  [310/312]  eta: 0:00:01  lr: 0.003994  min_lr: 0.003994  loss: 4.0457 (4.0774)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [26]  [311/312]  eta: 0:00:00  lr: 0.003994  min_lr: 0.003994  loss: 4.0457 (4.0785)  weight_decay: 0.0500 (0.0500)  time: 0.4627  data: 0.0001  max mem: 41794
Epoch: [26] Total time: 0:04:00 (0.7718 s / it)
Averaged stats: lr: 0.003994  min_lr: 0.003994  loss: 4.0457 (4.0781)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.9889 (1.9889)  acc1: 66.7969 (66.7969)  acc5: 86.8490 (86.8490)  time: 8.4850  data: 8.2725  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.2011 (2.1369)  acc1: 59.8958 (61.1680)  acc5: 83.3333 (84.0160)  time: 1.0912  data: 0.9192  max mem: 41794
Test: Total time: 0:00:10 (1.1249 s / it)
* Acc@1 61.444 Acc@5 84.548 loss 2.108
Accuracy of the model on the 50000 test images: 61.4%
Max accuracy: 62.17%
Epoch: [27]  [  0/312]  eta: 1:18:05  lr: 0.003994  min_lr: 0.003994  loss: 3.9564 (3.9564)  weight_decay: 0.0500 (0.0500)  time: 15.0162  data: 13.2623  max mem: 41794
Epoch: [27]  [ 10/312]  eta: 0:11:34  lr: 0.003994  min_lr: 0.003994  loss: 3.9564 (3.8717)  weight_decay: 0.0500 (0.0500)  time: 2.2995  data: 1.2782  max mem: 41794
Epoch: [27]  [ 20/312]  eta: 0:07:19  lr: 0.003994  min_lr: 0.003994  loss: 4.2485 (4.0971)  weight_decay: 0.0500 (0.0500)  time: 0.8313  data: 0.0758  max mem: 41794
Epoch: [27]  [ 30/312]  eta: 0:05:36  lr: 0.003994  min_lr: 0.003994  loss: 4.3548 (4.1429)  weight_decay: 0.0500 (0.0500)  time: 0.5854  data: 0.0401  max mem: 41794
Epoch: [27]  [ 40/312]  eta: 0:04:39  lr: 0.003994  min_lr: 0.003994  loss: 4.3428 (4.1497)  weight_decay: 0.0500 (0.0500)  time: 0.5267  data: 0.0079  max mem: 41794
Epoch: [27]  [ 50/312]  eta: 0:04:15  lr: 0.003994  min_lr: 0.003994  loss: 4.1998 (4.0928)  weight_decay: 0.0500 (0.0500)  time: 0.6340  data: 0.1022  max mem: 41794
Epoch: [27]  [ 60/312]  eta: 0:03:48  lr: 0.003993  min_lr: 0.003993  loss: 3.9020 (4.0495)  weight_decay: 0.0500 (0.0500)  time: 0.6580  data: 0.1025  max mem: 41794
Epoch: [27]  [ 70/312]  eta: 0:03:36  lr: 0.003993  min_lr: 0.003993  loss: 4.0452 (4.0918)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.1077  max mem: 41794
Epoch: [27]  [ 80/312]  eta: 0:03:27  lr: 0.003993  min_lr: 0.003993  loss: 4.3458 (4.1060)  weight_decay: 0.0500 (0.0500)  time: 0.8574  data: 0.2057  max mem: 41794
Epoch: [27]  [ 90/312]  eta: 0:03:10  lr: 0.003993  min_lr: 0.003993  loss: 4.1965 (4.1139)  weight_decay: 0.0500 (0.0500)  time: 0.7227  data: 0.1217  max mem: 41794
Epoch: [27]  [100/312]  eta: 0:03:03  lr: 0.003993  min_lr: 0.003993  loss: 4.1965 (4.1098)  weight_decay: 0.0500 (0.0500)  time: 0.7531  data: 0.0981  max mem: 41794
Epoch: [27]  [110/312]  eta: 0:02:48  lr: 0.003993  min_lr: 0.003993  loss: 4.2505 (4.1152)  weight_decay: 0.0500 (0.0500)  time: 0.7243  data: 0.0788  max mem: 41794
Epoch: [27]  [120/312]  eta: 0:02:40  lr: 0.003993  min_lr: 0.003993  loss: 4.3284 (4.1258)  weight_decay: 0.0500 (0.0500)  time: 0.6877  data: 0.0765  max mem: 41794
Epoch: [27]  [130/312]  eta: 0:02:32  lr: 0.003993  min_lr: 0.003993  loss: 4.3540 (4.1370)  weight_decay: 0.0500 (0.0500)  time: 0.8800  data: 0.1470  max mem: 41794
Epoch: [27]  [140/312]  eta: 0:02:21  lr: 0.003993  min_lr: 0.003993  loss: 4.2556 (4.1334)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.0758  max mem: 41794
Epoch: [27]  [150/312]  eta: 0:02:13  lr: 0.003993  min_lr: 0.003993  loss: 4.1896 (4.1295)  weight_decay: 0.0500 (0.0500)  time: 0.6942  data: 0.0585  max mem: 41794
Epoch: [27]  [160/312]  eta: 0:02:03  lr: 0.003993  min_lr: 0.003993  loss: 4.1448 (4.1068)  weight_decay: 0.0500 (0.0500)  time: 0.7507  data: 0.1231  max mem: 41794
Epoch: [27]  [170/312]  eta: 0:01:53  lr: 0.003993  min_lr: 0.003993  loss: 4.2191 (4.1263)  weight_decay: 0.0500 (0.0500)  time: 0.6386  data: 0.0738  max mem: 41794
Epoch: [27]  [180/312]  eta: 0:01:46  lr: 0.003993  min_lr: 0.003993  loss: 4.4464 (4.1319)  weight_decay: 0.0500 (0.0500)  time: 0.7324  data: 0.0985  max mem: 41794
Epoch: [27]  [190/312]  eta: 0:01:36  lr: 0.003993  min_lr: 0.003993  loss: 4.1534 (4.1275)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.0947  max mem: 41794
Epoch: [27]  [200/312]  eta: 0:01:28  lr: 0.003993  min_lr: 0.003993  loss: 4.1513 (4.1334)  weight_decay: 0.0500 (0.0500)  time: 0.6427  data: 0.1047  max mem: 41794
Epoch: [27]  [210/312]  eta: 0:01:20  lr: 0.003993  min_lr: 0.003993  loss: 4.1371 (4.1241)  weight_decay: 0.0500 (0.0500)  time: 0.8131  data: 0.2415  max mem: 41794
Epoch: [27]  [220/312]  eta: 0:01:11  lr: 0.003993  min_lr: 0.003993  loss: 4.2149 (4.1290)  weight_decay: 0.0500 (0.0500)  time: 0.6684  data: 0.1388  max mem: 41794
Epoch: [27]  [230/312]  eta: 0:01:04  lr: 0.003992  min_lr: 0.003992  loss: 4.2149 (4.1238)  weight_decay: 0.0500 (0.0500)  time: 0.6953  data: 0.1421  max mem: 41794
Epoch: [27]  [240/312]  eta: 0:00:56  lr: 0.003992  min_lr: 0.003992  loss: 3.9509 (4.1086)  weight_decay: 0.0500 (0.0500)  time: 0.8149  data: 0.2497  max mem: 41794
Epoch: [27]  [250/312]  eta: 0:00:47  lr: 0.003992  min_lr: 0.003992  loss: 4.0520 (4.1083)  weight_decay: 0.0500 (0.0500)  time: 0.6776  data: 0.1266  max mem: 41794
Epoch: [27]  [260/312]  eta: 0:00:40  lr: 0.003992  min_lr: 0.003992  loss: 3.9146 (4.0921)  weight_decay: 0.0500 (0.0500)  time: 0.7576  data: 0.1487  max mem: 41794
Epoch: [27]  [270/312]  eta: 0:00:32  lr: 0.003992  min_lr: 0.003992  loss: 3.7751 (4.0890)  weight_decay: 0.0500 (0.0500)  time: 0.6981  data: 0.1317  max mem: 41794
Epoch: [27]  [280/312]  eta: 0:00:24  lr: 0.003992  min_lr: 0.003992  loss: 4.3095 (4.0957)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.0987  max mem: 41794
Epoch: [27]  [290/312]  eta: 0:00:17  lr: 0.003992  min_lr: 0.003992  loss: 4.0601 (4.0903)  weight_decay: 0.0500 (0.0500)  time: 0.8714  data: 0.2220  max mem: 41794
Epoch: [27]  [300/312]  eta: 0:00:09  lr: 0.003992  min_lr: 0.003992  loss: 4.0034 (4.0940)  weight_decay: 0.0500 (0.0500)  time: 0.6671  data: 0.1237  max mem: 41794
Epoch: [27]  [310/312]  eta: 0:00:01  lr: 0.003992  min_lr: 0.003992  loss: 4.3102 (4.0932)  weight_decay: 0.0500 (0.0500)  time: 0.4693  data: 0.0001  max mem: 41794
Epoch: [27]  [311/312]  eta: 0:00:00  lr: 0.003992  min_lr: 0.003992  loss: 4.3470 (4.0941)  weight_decay: 0.0500 (0.0500)  time: 0.4690  data: 0.0001  max mem: 41794
Epoch: [27] Total time: 0:03:57 (0.7616 s / it)
Averaged stats: lr: 0.003992  min_lr: 0.003992  loss: 4.3470 (4.0678)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.7765 (1.7765)  acc1: 69.4010 (69.4010)  acc5: 88.8021 (88.8021)  time: 8.5454  data: 8.3456  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.0992 (1.9852)  acc1: 62.6302 (62.4000)  acc5: 82.5521 (84.7840)  time: 1.1039  data: 0.9274  max mem: 41794
Test: Total time: 0:00:10 (1.1490 s / it)
* Acc@1 62.090 Acc@5 84.672 loss 1.994
Accuracy of the model on the 50000 test images: 62.1%
Max accuracy: 62.17%
Epoch: [28]  [  0/312]  eta: 1:24:36  lr: 0.003992  min_lr: 0.003992  loss: 4.9033 (4.9033)  weight_decay: 0.0500 (0.0500)  time: 16.2707  data: 12.3154  max mem: 41794
Epoch: [28]  [ 10/312]  eta: 0:11:30  lr: 0.003992  min_lr: 0.003992  loss: 4.3208 (4.2546)  weight_decay: 0.0500 (0.0500)  time: 2.2878  data: 1.2781  max mem: 41794
Epoch: [28]  [ 20/312]  eta: 0:07:27  lr: 0.003992  min_lr: 0.003992  loss: 4.2736 (4.2303)  weight_decay: 0.0500 (0.0500)  time: 0.7942  data: 0.1158  max mem: 41794
Epoch: [28]  [ 30/312]  eta: 0:05:39  lr: 0.003992  min_lr: 0.003992  loss: 4.2348 (4.2077)  weight_decay: 0.0500 (0.0500)  time: 0.6095  data: 0.0315  max mem: 41794
Epoch: [28]  [ 40/312]  eta: 0:04:42  lr: 0.003992  min_lr: 0.003992  loss: 4.1602 (4.1883)  weight_decay: 0.0500 (0.0500)  time: 0.5199  data: 0.0080  max mem: 41794
Epoch: [28]  [ 50/312]  eta: 0:04:25  lr: 0.003992  min_lr: 0.003992  loss: 4.2634 (4.1708)  weight_decay: 0.0500 (0.0500)  time: 0.7119  data: 0.1026  max mem: 41794
Epoch: [28]  [ 60/312]  eta: 0:03:53  lr: 0.003992  min_lr: 0.003992  loss: 4.2803 (4.1640)  weight_decay: 0.0500 (0.0500)  time: 0.6960  data: 0.0978  max mem: 41794
Epoch: [28]  [ 70/312]  eta: 0:03:42  lr: 0.003991  min_lr: 0.003991  loss: 4.0188 (4.1234)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.0868  max mem: 41794
Epoch: [28]  [ 80/312]  eta: 0:03:28  lr: 0.003991  min_lr: 0.003991  loss: 3.8556 (4.1174)  weight_decay: 0.0500 (0.0500)  time: 0.8114  data: 0.1434  max mem: 41794
Epoch: [28]  [ 90/312]  eta: 0:03:11  lr: 0.003991  min_lr: 0.003991  loss: 4.0834 (4.1158)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.1002  max mem: 41794
Epoch: [28]  [100/312]  eta: 0:03:02  lr: 0.003991  min_lr: 0.003991  loss: 4.0963 (4.1116)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.1322  max mem: 41794
Epoch: [28]  [110/312]  eta: 0:02:47  lr: 0.003991  min_lr: 0.003991  loss: 4.2596 (4.1229)  weight_decay: 0.0500 (0.0500)  time: 0.6640  data: 0.0893  max mem: 41794
Epoch: [28]  [120/312]  eta: 0:02:40  lr: 0.003991  min_lr: 0.003991  loss: 4.1576 (4.1108)  weight_decay: 0.0500 (0.0500)  time: 0.7122  data: 0.1251  max mem: 41794
Epoch: [28]  [130/312]  eta: 0:02:32  lr: 0.003991  min_lr: 0.003991  loss: 3.9825 (4.1009)  weight_decay: 0.0500 (0.0500)  time: 0.8925  data: 0.2158  max mem: 41794
Epoch: [28]  [140/312]  eta: 0:02:19  lr: 0.003991  min_lr: 0.003991  loss: 4.1501 (4.0942)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.0914  max mem: 41794
Epoch: [28]  [150/312]  eta: 0:02:12  lr: 0.003991  min_lr: 0.003991  loss: 4.1712 (4.0904)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.0791  max mem: 41794
Epoch: [28]  [160/312]  eta: 0:02:04  lr: 0.003991  min_lr: 0.003991  loss: 4.1401 (4.0772)  weight_decay: 0.0500 (0.0500)  time: 0.8596  data: 0.0960  max mem: 41794
Epoch: [28]  [170/312]  eta: 0:01:54  lr: 0.003991  min_lr: 0.003991  loss: 3.8728 (4.0648)  weight_decay: 0.0500 (0.0500)  time: 0.7070  data: 0.0667  max mem: 41794
Epoch: [28]  [180/312]  eta: 0:01:46  lr: 0.003991  min_lr: 0.003991  loss: 4.1459 (4.0732)  weight_decay: 0.0500 (0.0500)  time: 0.7326  data: 0.1008  max mem: 41794
Epoch: [28]  [190/312]  eta: 0:01:36  lr: 0.003991  min_lr: 0.003991  loss: 4.1582 (4.0778)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.0517  max mem: 41794
Epoch: [28]  [200/312]  eta: 0:01:29  lr: 0.003991  min_lr: 0.003991  loss: 4.1582 (4.0870)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.0622  max mem: 41794
Epoch: [28]  [210/312]  eta: 0:01:21  lr: 0.003991  min_lr: 0.003991  loss: 3.9838 (4.0755)  weight_decay: 0.0500 (0.0500)  time: 0.8543  data: 0.1178  max mem: 41794
Epoch: [28]  [220/312]  eta: 0:01:12  lr: 0.003990  min_lr: 0.003990  loss: 3.7788 (4.0702)  weight_decay: 0.0500 (0.0500)  time: 0.6684  data: 0.0562  max mem: 41794
Epoch: [28]  [230/312]  eta: 0:01:04  lr: 0.003990  min_lr: 0.003990  loss: 4.4257 (4.0920)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.0326  max mem: 41794
Epoch: [28]  [240/312]  eta: 0:00:56  lr: 0.003990  min_lr: 0.003990  loss: 4.4053 (4.0816)  weight_decay: 0.0500 (0.0500)  time: 0.8324  data: 0.0327  max mem: 41794
Epoch: [28]  [250/312]  eta: 0:00:48  lr: 0.003990  min_lr: 0.003990  loss: 4.0693 (4.0828)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.0490  max mem: 41794
Epoch: [28]  [260/312]  eta: 0:00:40  lr: 0.003990  min_lr: 0.003990  loss: 4.0692 (4.0779)  weight_decay: 0.0500 (0.0500)  time: 0.7199  data: 0.1003  max mem: 41794
Epoch: [28]  [270/312]  eta: 0:00:32  lr: 0.003990  min_lr: 0.003990  loss: 3.8679 (4.0603)  weight_decay: 0.0500 (0.0500)  time: 0.6723  data: 0.0549  max mem: 41794
Epoch: [28]  [280/312]  eta: 0:00:24  lr: 0.003990  min_lr: 0.003990  loss: 3.8679 (4.0549)  weight_decay: 0.0500 (0.0500)  time: 0.6955  data: 0.0602  max mem: 41794
Epoch: [28]  [290/312]  eta: 0:00:17  lr: 0.003990  min_lr: 0.003990  loss: 4.1569 (4.0600)  weight_decay: 0.0500 (0.0500)  time: 0.8691  data: 0.0948  max mem: 41794
Epoch: [28]  [300/312]  eta: 0:00:09  lr: 0.003990  min_lr: 0.003990  loss: 4.2920 (4.0615)  weight_decay: 0.0500 (0.0500)  time: 0.6593  data: 0.0381  max mem: 41794
Epoch: [28]  [310/312]  eta: 0:00:01  lr: 0.003990  min_lr: 0.003990  loss: 3.9476 (4.0546)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0001  max mem: 41794
Epoch: [28]  [311/312]  eta: 0:00:00  lr: 0.003990  min_lr: 0.003990  loss: 3.7450 (4.0512)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [28] Total time: 0:03:58 (0.7652 s / it)
Averaged stats: lr: 0.003990  min_lr: 0.003990  loss: 3.7450 (4.0229)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:08  loss: 1.6039 (1.6039)  acc1: 70.4427 (70.4427)  acc5: 90.4948 (90.4948)  time: 7.5558  data: 7.3472  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9648 (1.8068)  acc1: 59.7656 (62.5280)  acc5: 85.2865 (85.8240)  time: 1.0310  data: 0.8599  max mem: 41794
Test: Total time: 0:00:09 (1.0421 s / it)
* Acc@1 62.872 Acc@5 85.320 loss 1.812
Accuracy of the model on the 50000 test images: 62.9%
Max accuracy: 62.87%
Epoch: [29]  [  0/312]  eta: 1:21:20  lr: 0.003990  min_lr: 0.003990  loss: 3.3366 (3.3366)  weight_decay: 0.0500 (0.0500)  time: 15.6438  data: 15.1478  max mem: 41794
Epoch: [29]  [ 10/312]  eta: 0:10:13  lr: 0.003990  min_lr: 0.003990  loss: 4.0175 (3.9440)  weight_decay: 0.0500 (0.0500)  time: 2.0299  data: 1.3777  max mem: 41794
Epoch: [29]  [ 20/312]  eta: 0:07:04  lr: 0.003990  min_lr: 0.003990  loss: 4.0780 (3.9497)  weight_decay: 0.0500 (0.0500)  time: 0.7441  data: 0.1560  max mem: 41794
Epoch: [29]  [ 30/312]  eta: 0:05:25  lr: 0.003990  min_lr: 0.003990  loss: 4.1579 (3.9994)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.1561  max mem: 41794
Epoch: [29]  [ 40/312]  eta: 0:04:51  lr: 0.003990  min_lr: 0.003990  loss: 4.2472 (4.0157)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.1666  max mem: 41794
Epoch: [29]  [ 50/312]  eta: 0:04:30  lr: 0.003989  min_lr: 0.003989  loss: 4.0674 (4.0149)  weight_decay: 0.0500 (0.0500)  time: 0.8495  data: 0.3556  max mem: 41794
Epoch: [29]  [ 60/312]  eta: 0:03:58  lr: 0.003989  min_lr: 0.003989  loss: 4.1059 (4.0542)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.1897  max mem: 41794
Epoch: [29]  [ 70/312]  eta: 0:03:46  lr: 0.003989  min_lr: 0.003989  loss: 4.0926 (4.0067)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1977  max mem: 41794
Epoch: [29]  [ 80/312]  eta: 0:03:35  lr: 0.003989  min_lr: 0.003989  loss: 3.8520 (4.0006)  weight_decay: 0.0500 (0.0500)  time: 0.8750  data: 0.3744  max mem: 41794
Epoch: [29]  [ 90/312]  eta: 0:03:15  lr: 0.003989  min_lr: 0.003989  loss: 4.0725 (3.9983)  weight_decay: 0.0500 (0.0500)  time: 0.6862  data: 0.1773  max mem: 41794
Epoch: [29]  [100/312]  eta: 0:03:07  lr: 0.003989  min_lr: 0.003989  loss: 4.0915 (3.9974)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.1778  max mem: 41794
Epoch: [29]  [110/312]  eta: 0:02:51  lr: 0.003989  min_lr: 0.003989  loss: 3.8798 (3.9812)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.1778  max mem: 41794
Epoch: [29]  [120/312]  eta: 0:02:42  lr: 0.003989  min_lr: 0.003989  loss: 3.8798 (3.9950)  weight_decay: 0.0500 (0.0500)  time: 0.6653  data: 0.1726  max mem: 41794
Epoch: [29]  [130/312]  eta: 0:02:34  lr: 0.003989  min_lr: 0.003989  loss: 4.0560 (3.9774)  weight_decay: 0.0500 (0.0500)  time: 0.8505  data: 0.3528  max mem: 41794
Epoch: [29]  [140/312]  eta: 0:02:21  lr: 0.003989  min_lr: 0.003989  loss: 3.9692 (3.9778)  weight_decay: 0.0500 (0.0500)  time: 0.6806  data: 0.1809  max mem: 41794
Epoch: [29]  [150/312]  eta: 0:02:14  lr: 0.003989  min_lr: 0.003989  loss: 3.9692 (3.9833)  weight_decay: 0.0500 (0.0500)  time: 0.6930  data: 0.1968  max mem: 41794
Epoch: [29]  [160/312]  eta: 0:02:05  lr: 0.003989  min_lr: 0.003989  loss: 4.0727 (3.9837)  weight_decay: 0.0500 (0.0500)  time: 0.8493  data: 0.3269  max mem: 41794
Epoch: [29]  [170/312]  eta: 0:01:54  lr: 0.003989  min_lr: 0.003989  loss: 4.3630 (4.0039)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1308  max mem: 41794
Epoch: [29]  [180/312]  eta: 0:01:47  lr: 0.003988  min_lr: 0.003988  loss: 4.2211 (4.0009)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.1604  max mem: 41794
Epoch: [29]  [190/312]  eta: 0:01:36  lr: 0.003988  min_lr: 0.003988  loss: 3.7949 (3.9968)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.1604  max mem: 41794
Epoch: [29]  [200/312]  eta: 0:01:29  lr: 0.003988  min_lr: 0.003988  loss: 3.7907 (3.9854)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.1858  max mem: 41794
Epoch: [29]  [210/312]  eta: 0:01:22  lr: 0.003988  min_lr: 0.003988  loss: 3.9763 (3.9886)  weight_decay: 0.0500 (0.0500)  time: 0.9049  data: 0.3411  max mem: 41794
Epoch: [29]  [220/312]  eta: 0:01:12  lr: 0.003988  min_lr: 0.003988  loss: 4.2069 (3.9972)  weight_decay: 0.0500 (0.0500)  time: 0.6977  data: 0.1560  max mem: 41794
Epoch: [29]  [230/312]  eta: 0:01:05  lr: 0.003988  min_lr: 0.003988  loss: 4.3449 (4.0065)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.1253  max mem: 41794
Epoch: [29]  [240/312]  eta: 0:00:57  lr: 0.003988  min_lr: 0.003988  loss: 4.1963 (4.0065)  weight_decay: 0.0500 (0.0500)  time: 0.8290  data: 0.2002  max mem: 41794
Epoch: [29]  [250/312]  eta: 0:00:48  lr: 0.003988  min_lr: 0.003988  loss: 4.0882 (3.9985)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.0995  max mem: 41794
Epoch: [29]  [260/312]  eta: 0:00:41  lr: 0.003988  min_lr: 0.003988  loss: 4.2006 (4.0080)  weight_decay: 0.0500 (0.0500)  time: 0.7460  data: 0.1607  max mem: 41794
Epoch: [29]  [270/312]  eta: 0:00:32  lr: 0.003988  min_lr: 0.003988  loss: 4.1954 (4.0008)  weight_decay: 0.0500 (0.0500)  time: 0.7124  data: 0.1372  max mem: 41794
Epoch: [29]  [280/312]  eta: 0:00:25  lr: 0.003988  min_lr: 0.003988  loss: 3.7220 (3.9918)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.1223  max mem: 41794
Epoch: [29]  [290/312]  eta: 0:00:17  lr: 0.003988  min_lr: 0.003988  loss: 3.9094 (3.9886)  weight_decay: 0.0500 (0.0500)  time: 0.7517  data: 0.1826  max mem: 41794
Epoch: [29]  [300/312]  eta: 0:00:09  lr: 0.003988  min_lr: 0.003988  loss: 3.8156 (3.9834)  weight_decay: 0.0500 (0.0500)  time: 0.5682  data: 0.0610  max mem: 41794
Epoch: [29]  [310/312]  eta: 0:00:01  lr: 0.003987  min_lr: 0.003987  loss: 4.0685 (3.9878)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.0001  max mem: 41794
Epoch: [29]  [311/312]  eta: 0:00:00  lr: 0.003987  min_lr: 0.003987  loss: 4.0685 (3.9885)  weight_decay: 0.0500 (0.0500)  time: 0.4656  data: 0.0001  max mem: 41794
Epoch: [29] Total time: 0:03:58 (0.7629 s / it)
Averaged stats: lr: 0.003987  min_lr: 0.003987  loss: 4.0685 (4.0091)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.6307 (1.6307)  acc1: 70.5729 (70.5729)  acc5: 89.5833 (89.5833)  time: 8.2319  data: 8.0189  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.0151 (1.9110)  acc1: 60.0260 (62.0480)  acc5: 84.1146 (84.8480)  time: 1.0627  data: 0.8911  max mem: 41794
Test: Total time: 0:00:09 (1.0790 s / it)
* Acc@1 62.782 Acc@5 85.240 loss 1.896
Accuracy of the model on the 50000 test images: 62.8%
Max accuracy: 62.87%
Epoch: [30]  [  0/312]  eta: 1:21:21  lr: 0.003987  min_lr: 0.003987  loss: 4.7959 (4.7959)  weight_decay: 0.0500 (0.0500)  time: 15.6453  data: 15.1642  max mem: 41794
Epoch: [30]  [ 10/312]  eta: 0:11:48  lr: 0.003987  min_lr: 0.003987  loss: 4.3253 (4.2547)  weight_decay: 0.0500 (0.0500)  time: 2.3466  data: 1.3854  max mem: 41794
Epoch: [30]  [ 20/312]  eta: 0:07:30  lr: 0.003987  min_lr: 0.003987  loss: 4.2587 (4.0888)  weight_decay: 0.0500 (0.0500)  time: 0.8373  data: 0.0269  max mem: 41794
Epoch: [30]  [ 30/312]  eta: 0:05:43  lr: 0.003987  min_lr: 0.003987  loss: 3.9658 (4.0835)  weight_decay: 0.0500 (0.0500)  time: 0.5969  data: 0.0235  max mem: 41794
Epoch: [30]  [ 40/312]  eta: 0:04:45  lr: 0.003987  min_lr: 0.003987  loss: 4.1388 (4.0619)  weight_decay: 0.0500 (0.0500)  time: 0.5298  data: 0.0040  max mem: 41794
Epoch: [30]  [ 50/312]  eta: 0:04:12  lr: 0.003987  min_lr: 0.003987  loss: 4.2790 (4.0894)  weight_decay: 0.0500 (0.0500)  time: 0.5735  data: 0.0252  max mem: 41794
Epoch: [30]  [ 60/312]  eta: 0:03:45  lr: 0.003987  min_lr: 0.003987  loss: 4.3804 (4.1219)  weight_decay: 0.0500 (0.0500)  time: 0.5751  data: 0.0219  max mem: 41794
Epoch: [30]  [ 70/312]  eta: 0:03:32  lr: 0.003987  min_lr: 0.003987  loss: 4.2529 (4.1309)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.1278  max mem: 41794
Epoch: [30]  [ 80/312]  eta: 0:03:20  lr: 0.003987  min_lr: 0.003987  loss: 4.2140 (4.1182)  weight_decay: 0.0500 (0.0500)  time: 0.7781  data: 0.2549  max mem: 41794
Epoch: [30]  [ 90/312]  eta: 0:03:07  lr: 0.003987  min_lr: 0.003987  loss: 4.1605 (4.1184)  weight_decay: 0.0500 (0.0500)  time: 0.7197  data: 0.1433  max mem: 41794
Epoch: [30]  [100/312]  eta: 0:02:59  lr: 0.003987  min_lr: 0.003987  loss: 3.8181 (4.0854)  weight_decay: 0.0500 (0.0500)  time: 0.7643  data: 0.1070  max mem: 41794
Epoch: [30]  [110/312]  eta: 0:02:44  lr: 0.003987  min_lr: 0.003987  loss: 4.0407 (4.0986)  weight_decay: 0.0500 (0.0500)  time: 0.6910  data: 0.1085  max mem: 41794
Epoch: [30]  [120/312]  eta: 0:02:37  lr: 0.003986  min_lr: 0.003986  loss: 4.2515 (4.1049)  weight_decay: 0.0500 (0.0500)  time: 0.6961  data: 0.1002  max mem: 41794
Epoch: [30]  [130/312]  eta: 0:02:28  lr: 0.003986  min_lr: 0.003986  loss: 4.1958 (4.1039)  weight_decay: 0.0500 (0.0500)  time: 0.8258  data: 0.1458  max mem: 41794
Epoch: [30]  [140/312]  eta: 0:02:17  lr: 0.003986  min_lr: 0.003986  loss: 4.0356 (4.0994)  weight_decay: 0.0500 (0.0500)  time: 0.6790  data: 0.1001  max mem: 41794
Epoch: [30]  [150/312]  eta: 0:02:10  lr: 0.003986  min_lr: 0.003986  loss: 3.6146 (4.0514)  weight_decay: 0.0500 (0.0500)  time: 0.7027  data: 0.1283  max mem: 41794
Epoch: [30]  [160/312]  eta: 0:02:00  lr: 0.003986  min_lr: 0.003986  loss: 3.4280 (4.0402)  weight_decay: 0.0500 (0.0500)  time: 0.7512  data: 0.1575  max mem: 41794
Epoch: [30]  [170/312]  eta: 0:01:51  lr: 0.003986  min_lr: 0.003986  loss: 4.0165 (4.0428)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.1115  max mem: 41794
Epoch: [30]  [180/312]  eta: 0:01:44  lr: 0.003986  min_lr: 0.003986  loss: 4.1899 (4.0558)  weight_decay: 0.0500 (0.0500)  time: 0.7941  data: 0.1735  max mem: 41794
Epoch: [30]  [190/312]  eta: 0:01:35  lr: 0.003986  min_lr: 0.003986  loss: 4.0444 (4.0474)  weight_decay: 0.0500 (0.0500)  time: 0.7053  data: 0.1304  max mem: 41794
Epoch: [30]  [200/312]  eta: 0:01:27  lr: 0.003986  min_lr: 0.003986  loss: 3.9144 (4.0495)  weight_decay: 0.0500 (0.0500)  time: 0.6833  data: 0.0972  max mem: 41794
Epoch: [30]  [210/312]  eta: 0:01:20  lr: 0.003986  min_lr: 0.003986  loss: 4.2613 (4.0507)  weight_decay: 0.0500 (0.0500)  time: 0.8763  data: 0.1797  max mem: 41794
Epoch: [30]  [220/312]  eta: 0:01:11  lr: 0.003986  min_lr: 0.003986  loss: 3.9416 (4.0441)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.0889  max mem: 41794
Epoch: [30]  [230/312]  eta: 0:01:03  lr: 0.003986  min_lr: 0.003986  loss: 4.0213 (4.0419)  weight_decay: 0.0500 (0.0500)  time: 0.6794  data: 0.1015  max mem: 41794
Epoch: [30]  [240/312]  eta: 0:00:56  lr: 0.003985  min_lr: 0.003985  loss: 4.1009 (4.0364)  weight_decay: 0.0500 (0.0500)  time: 0.8077  data: 0.1113  max mem: 41794
Epoch: [30]  [250/312]  eta: 0:00:47  lr: 0.003985  min_lr: 0.003985  loss: 4.1009 (4.0357)  weight_decay: 0.0500 (0.0500)  time: 0.7149  data: 0.0900  max mem: 41794
Epoch: [30]  [260/312]  eta: 0:00:40  lr: 0.003985  min_lr: 0.003985  loss: 3.8771 (4.0244)  weight_decay: 0.0500 (0.0500)  time: 0.7646  data: 0.1407  max mem: 41794
Epoch: [30]  [270/312]  eta: 0:00:32  lr: 0.003985  min_lr: 0.003985  loss: 4.0600 (4.0321)  weight_decay: 0.0500 (0.0500)  time: 0.6810  data: 0.0661  max mem: 41794
Epoch: [30]  [280/312]  eta: 0:00:24  lr: 0.003985  min_lr: 0.003985  loss: 3.8155 (4.0160)  weight_decay: 0.0500 (0.0500)  time: 0.6874  data: 0.0571  max mem: 41794
Epoch: [30]  [290/312]  eta: 0:00:17  lr: 0.003985  min_lr: 0.003985  loss: 3.8155 (4.0186)  weight_decay: 0.0500 (0.0500)  time: 0.8586  data: 0.0943  max mem: 41794
Epoch: [30]  [300/312]  eta: 0:00:09  lr: 0.003985  min_lr: 0.003985  loss: 4.1489 (4.0222)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.0387  max mem: 41794
Epoch: [30]  [310/312]  eta: 0:00:01  lr: 0.003985  min_lr: 0.003985  loss: 4.2987 (4.0308)  weight_decay: 0.0500 (0.0500)  time: 0.4702  data: 0.0001  max mem: 41794
Epoch: [30]  [311/312]  eta: 0:00:00  lr: 0.003985  min_lr: 0.003985  loss: 4.2987 (4.0317)  weight_decay: 0.0500 (0.0500)  time: 0.4672  data: 0.0001  max mem: 41794
Epoch: [30] Total time: 0:03:57 (0.7599 s / it)
Averaged stats: lr: 0.003985  min_lr: 0.003985  loss: 4.2987 (3.9964)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.7781 (1.7781)  acc1: 68.6198 (68.6198)  acc5: 89.7135 (89.7135)  time: 8.7265  data: 8.5209  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.0056 (1.8617)  acc1: 63.1510 (62.9920)  acc5: 84.2448 (85.1840)  time: 1.1207  data: 0.9468  max mem: 41794
Test: Total time: 0:00:10 (1.1543 s / it)
* Acc@1 62.424 Acc@5 85.212 loss 1.866
Accuracy of the model on the 50000 test images: 62.4%
Max accuracy: 62.87%
Epoch: [31]  [  0/312]  eta: 1:17:51  lr: 0.003985  min_lr: 0.003985  loss: 3.5810 (3.5810)  weight_decay: 0.0500 (0.0500)  time: 14.9726  data: 13.0388  max mem: 41794
Epoch: [31]  [ 10/312]  eta: 0:10:36  lr: 0.003985  min_lr: 0.003985  loss: 4.3198 (4.1907)  weight_decay: 0.0500 (0.0500)  time: 2.1064  data: 1.2980  max mem: 41794
Epoch: [31]  [ 20/312]  eta: 0:07:21  lr: 0.003985  min_lr: 0.003985  loss: 4.1874 (4.1118)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.1316  max mem: 41794
Epoch: [31]  [ 30/312]  eta: 0:05:38  lr: 0.003985  min_lr: 0.003985  loss: 4.1967 (4.1514)  weight_decay: 0.0500 (0.0500)  time: 0.7011  data: 0.0700  max mem: 41794
Epoch: [31]  [ 40/312]  eta: 0:04:56  lr: 0.003984  min_lr: 0.003984  loss: 4.2070 (4.1587)  weight_decay: 0.0500 (0.0500)  time: 0.6437  data: 0.0986  max mem: 41794
Epoch: [31]  [ 50/312]  eta: 0:04:34  lr: 0.003984  min_lr: 0.003984  loss: 4.0312 (4.0811)  weight_decay: 0.0500 (0.0500)  time: 0.8077  data: 0.1947  max mem: 41794
Epoch: [31]  [ 60/312]  eta: 0:04:01  lr: 0.003984  min_lr: 0.003984  loss: 4.1550 (4.0947)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.0968  max mem: 41794
Epoch: [31]  [ 70/312]  eta: 0:03:49  lr: 0.003984  min_lr: 0.003984  loss: 4.1913 (4.0770)  weight_decay: 0.0500 (0.0500)  time: 0.7042  data: 0.1051  max mem: 41794
Epoch: [31]  [ 80/312]  eta: 0:03:36  lr: 0.003984  min_lr: 0.003984  loss: 4.0195 (4.0552)  weight_decay: 0.0500 (0.0500)  time: 0.8556  data: 0.1617  max mem: 41794
Epoch: [31]  [ 90/312]  eta: 0:03:20  lr: 0.003984  min_lr: 0.003984  loss: 4.0497 (4.0558)  weight_decay: 0.0500 (0.0500)  time: 0.7398  data: 0.1050  max mem: 41794
Epoch: [31]  [100/312]  eta: 0:03:12  lr: 0.003984  min_lr: 0.003984  loss: 3.9057 (4.0270)  weight_decay: 0.0500 (0.0500)  time: 0.8048  data: 0.1411  max mem: 41794
Epoch: [31]  [110/312]  eta: 0:02:56  lr: 0.003984  min_lr: 0.003984  loss: 3.8488 (4.0201)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.0992  max mem: 41794
Epoch: [31]  [120/312]  eta: 0:02:48  lr: 0.003984  min_lr: 0.003984  loss: 4.0443 (4.0267)  weight_decay: 0.0500 (0.0500)  time: 0.7149  data: 0.0911  max mem: 41794
Epoch: [31]  [130/312]  eta: 0:02:37  lr: 0.003984  min_lr: 0.003984  loss: 4.1034 (4.0231)  weight_decay: 0.0500 (0.0500)  time: 0.8383  data: 0.1539  max mem: 41794
Epoch: [31]  [140/312]  eta: 0:02:24  lr: 0.003984  min_lr: 0.003984  loss: 4.1034 (4.0129)  weight_decay: 0.0500 (0.0500)  time: 0.6405  data: 0.0694  max mem: 41794
Epoch: [31]  [150/312]  eta: 0:02:15  lr: 0.003983  min_lr: 0.003983  loss: 4.0748 (4.0122)  weight_decay: 0.0500 (0.0500)  time: 0.6430  data: 0.0766  max mem: 41794
Epoch: [31]  [160/312]  eta: 0:02:06  lr: 0.003983  min_lr: 0.003983  loss: 4.0748 (4.0071)  weight_decay: 0.0500 (0.0500)  time: 0.7424  data: 0.0792  max mem: 41794
Epoch: [31]  [170/312]  eta: 0:01:56  lr: 0.003983  min_lr: 0.003983  loss: 4.0366 (4.0057)  weight_decay: 0.0500 (0.0500)  time: 0.6633  data: 0.0705  max mem: 41794
Epoch: [31]  [180/312]  eta: 0:01:48  lr: 0.003983  min_lr: 0.003983  loss: 3.7705 (3.9909)  weight_decay: 0.0500 (0.0500)  time: 0.7910  data: 0.1749  max mem: 41794
Epoch: [31]  [190/312]  eta: 0:01:38  lr: 0.003983  min_lr: 0.003983  loss: 4.0252 (3.9891)  weight_decay: 0.0500 (0.0500)  time: 0.7374  data: 0.1102  max mem: 41794
Epoch: [31]  [200/312]  eta: 0:01:30  lr: 0.003983  min_lr: 0.003983  loss: 3.9001 (3.9840)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.0673  max mem: 41794
Epoch: [31]  [210/312]  eta: 0:01:22  lr: 0.003983  min_lr: 0.003983  loss: 4.2078 (3.9918)  weight_decay: 0.0500 (0.0500)  time: 0.8184  data: 0.1295  max mem: 41794
Epoch: [31]  [220/312]  eta: 0:01:13  lr: 0.003983  min_lr: 0.003983  loss: 4.2309 (3.9949)  weight_decay: 0.0500 (0.0500)  time: 0.6405  data: 0.0654  max mem: 41794
Epoch: [31]  [230/312]  eta: 0:01:05  lr: 0.003983  min_lr: 0.003983  loss: 4.0218 (3.9834)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.0841  max mem: 41794
Epoch: [31]  [240/312]  eta: 0:00:56  lr: 0.003983  min_lr: 0.003983  loss: 3.9114 (3.9881)  weight_decay: 0.0500 (0.0500)  time: 0.7418  data: 0.0862  max mem: 41794
Epoch: [31]  [250/312]  eta: 0:00:48  lr: 0.003982  min_lr: 0.003982  loss: 4.3541 (3.9920)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.0823  max mem: 41794
Epoch: [31]  [260/312]  eta: 0:00:41  lr: 0.003982  min_lr: 0.003982  loss: 4.2871 (3.9945)  weight_decay: 0.0500 (0.0500)  time: 0.8377  data: 0.1697  max mem: 41794
Epoch: [31]  [270/312]  eta: 0:00:32  lr: 0.003982  min_lr: 0.003982  loss: 4.2185 (3.9929)  weight_decay: 0.0500 (0.0500)  time: 0.7427  data: 0.0902  max mem: 41794
Epoch: [31]  [280/312]  eta: 0:00:25  lr: 0.003982  min_lr: 0.003982  loss: 4.0323 (3.9918)  weight_decay: 0.0500 (0.0500)  time: 0.7078  data: 0.0948  max mem: 41794
Epoch: [31]  [290/312]  eta: 0:00:17  lr: 0.003982  min_lr: 0.003982  loss: 3.7945 (3.9818)  weight_decay: 0.0500 (0.0500)  time: 0.7595  data: 0.1253  max mem: 41794
Epoch: [31]  [300/312]  eta: 0:00:09  lr: 0.003982  min_lr: 0.003982  loss: 3.8251 (3.9821)  weight_decay: 0.0500 (0.0500)  time: 0.5427  data: 0.0309  max mem: 41794
Epoch: [31]  [310/312]  eta: 0:00:01  lr: 0.003982  min_lr: 0.003982  loss: 4.2125 (3.9887)  weight_decay: 0.0500 (0.0500)  time: 0.4614  data: 0.0001  max mem: 41794
Epoch: [31]  [311/312]  eta: 0:00:00  lr: 0.003982  min_lr: 0.003982  loss: 4.2125 (3.9875)  weight_decay: 0.0500 (0.0500)  time: 0.4615  data: 0.0001  max mem: 41794
Epoch: [31] Total time: 0:03:59 (0.7684 s / it)
Averaged stats: lr: 0.003982  min_lr: 0.003982  loss: 4.2125 (3.9794)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.5715 (1.5715)  acc1: 67.1875 (67.1875)  acc5: 88.9323 (88.9323)  time: 7.9820  data: 7.7732  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9989 (1.7990)  acc1: 61.0677 (62.0160)  acc5: 83.4635 (85.3120)  time: 1.0350  data: 0.8638  max mem: 41794
Test: Total time: 0:00:09 (1.0594 s / it)
* Acc@1 62.148 Acc@5 85.268 loss 1.793
Accuracy of the model on the 50000 test images: 62.1%
Max accuracy: 62.87%
Epoch: [32]  [  0/312]  eta: 1:19:11  lr: 0.003982  min_lr: 0.003982  loss: 3.4599 (3.4599)  weight_decay: 0.0500 (0.0500)  time: 15.2293  data: 13.7910  max mem: 41794
Epoch: [32]  [ 10/312]  eta: 0:10:48  lr: 0.003982  min_lr: 0.003982  loss: 3.9038 (3.7887)  weight_decay: 0.0500 (0.0500)  time: 2.1473  data: 1.3551  max mem: 41794
Epoch: [32]  [ 20/312]  eta: 0:07:27  lr: 0.003982  min_lr: 0.003982  loss: 3.9038 (3.8535)  weight_decay: 0.0500 (0.0500)  time: 0.8475  data: 0.1064  max mem: 41794
Epoch: [32]  [ 30/312]  eta: 0:05:41  lr: 0.003982  min_lr: 0.003982  loss: 4.0650 (3.9571)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.0512  max mem: 41794
Epoch: [32]  [ 40/312]  eta: 0:04:49  lr: 0.003982  min_lr: 0.003982  loss: 4.2354 (3.9866)  weight_decay: 0.0500 (0.0500)  time: 0.5755  data: 0.0011  max mem: 41794
Epoch: [32]  [ 50/312]  eta: 0:04:28  lr: 0.003981  min_lr: 0.003981  loss: 4.2897 (4.0402)  weight_decay: 0.0500 (0.0500)  time: 0.7354  data: 0.0598  max mem: 41794
Epoch: [32]  [ 60/312]  eta: 0:03:55  lr: 0.003981  min_lr: 0.003981  loss: 4.3184 (4.0376)  weight_decay: 0.0500 (0.0500)  time: 0.6709  data: 0.0596  max mem: 41794
Epoch: [32]  [ 70/312]  eta: 0:03:45  lr: 0.003981  min_lr: 0.003981  loss: 4.2000 (4.0210)  weight_decay: 0.0500 (0.0500)  time: 0.6939  data: 0.0438  max mem: 41794
Epoch: [32]  [ 80/312]  eta: 0:03:30  lr: 0.003981  min_lr: 0.003981  loss: 4.1704 (4.0500)  weight_decay: 0.0500 (0.0500)  time: 0.8197  data: 0.0439  max mem: 41794
Epoch: [32]  [ 90/312]  eta: 0:03:15  lr: 0.003981  min_lr: 0.003981  loss: 4.1704 (4.0528)  weight_decay: 0.0500 (0.0500)  time: 0.6978  data: 0.0491  max mem: 41794
Epoch: [32]  [100/312]  eta: 0:03:06  lr: 0.003981  min_lr: 0.003981  loss: 4.1158 (4.0623)  weight_decay: 0.0500 (0.0500)  time: 0.7722  data: 0.1395  max mem: 41794
Epoch: [32]  [110/312]  eta: 0:02:50  lr: 0.003981  min_lr: 0.003981  loss: 4.1646 (4.0588)  weight_decay: 0.0500 (0.0500)  time: 0.6955  data: 0.0912  max mem: 41794
Epoch: [32]  [120/312]  eta: 0:02:43  lr: 0.003981  min_lr: 0.003981  loss: 4.1239 (4.0361)  weight_decay: 0.0500 (0.0500)  time: 0.6965  data: 0.1134  max mem: 41794
Epoch: [32]  [130/312]  eta: 0:02:34  lr: 0.003981  min_lr: 0.003981  loss: 4.0762 (4.0474)  weight_decay: 0.0500 (0.0500)  time: 0.8642  data: 0.2037  max mem: 41794
Epoch: [32]  [140/312]  eta: 0:02:21  lr: 0.003981  min_lr: 0.003981  loss: 4.0762 (4.0495)  weight_decay: 0.0500 (0.0500)  time: 0.6665  data: 0.0961  max mem: 41794
Epoch: [32]  [150/312]  eta: 0:02:14  lr: 0.003980  min_lr: 0.003980  loss: 4.0742 (4.0424)  weight_decay: 0.0500 (0.0500)  time: 0.7036  data: 0.0989  max mem: 41794
Epoch: [32]  [160/312]  eta: 0:02:04  lr: 0.003980  min_lr: 0.003980  loss: 4.0944 (4.0396)  weight_decay: 0.0500 (0.0500)  time: 0.7945  data: 0.1117  max mem: 41794
Epoch: [32]  [170/312]  eta: 0:01:56  lr: 0.003980  min_lr: 0.003980  loss: 3.9693 (4.0304)  weight_decay: 0.0500 (0.0500)  time: 0.7483  data: 0.0875  max mem: 41794
Epoch: [32]  [180/312]  eta: 0:01:48  lr: 0.003980  min_lr: 0.003980  loss: 3.7847 (4.0194)  weight_decay: 0.0500 (0.0500)  time: 0.8182  data: 0.1256  max mem: 41794
Epoch: [32]  [190/312]  eta: 0:01:38  lr: 0.003980  min_lr: 0.003980  loss: 4.0610 (4.0331)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.0568  max mem: 41794
Epoch: [32]  [200/312]  eta: 0:01:30  lr: 0.003980  min_lr: 0.003980  loss: 4.2147 (4.0299)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.0581  max mem: 41794
Epoch: [32]  [210/312]  eta: 0:01:22  lr: 0.003980  min_lr: 0.003980  loss: 4.2147 (4.0330)  weight_decay: 0.0500 (0.0500)  time: 0.8167  data: 0.0942  max mem: 41794
Epoch: [32]  [220/312]  eta: 0:01:13  lr: 0.003980  min_lr: 0.003980  loss: 4.2085 (4.0381)  weight_decay: 0.0500 (0.0500)  time: 0.6712  data: 0.0627  max mem: 41794
Epoch: [32]  [230/312]  eta: 0:01:04  lr: 0.003980  min_lr: 0.003980  loss: 4.1236 (4.0319)  weight_decay: 0.0500 (0.0500)  time: 0.6584  data: 0.1032  max mem: 41794
Epoch: [32]  [240/312]  eta: 0:00:56  lr: 0.003980  min_lr: 0.003980  loss: 3.9957 (4.0244)  weight_decay: 0.0500 (0.0500)  time: 0.7686  data: 0.1388  max mem: 41794
Epoch: [32]  [250/312]  eta: 0:00:48  lr: 0.003979  min_lr: 0.003979  loss: 3.8870 (4.0143)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.1526  max mem: 41794
Epoch: [32]  [260/312]  eta: 0:00:41  lr: 0.003979  min_lr: 0.003979  loss: 4.0103 (4.0135)  weight_decay: 0.0500 (0.0500)  time: 0.7899  data: 0.2121  max mem: 41794
Epoch: [32]  [270/312]  eta: 0:00:32  lr: 0.003979  min_lr: 0.003979  loss: 4.3259 (4.0315)  weight_decay: 0.0500 (0.0500)  time: 0.7062  data: 0.1216  max mem: 41794
Epoch: [32]  [280/312]  eta: 0:00:25  lr: 0.003979  min_lr: 0.003979  loss: 4.3760 (4.0403)  weight_decay: 0.0500 (0.0500)  time: 0.7264  data: 0.1258  max mem: 41794
Epoch: [32]  [290/312]  eta: 0:00:17  lr: 0.003979  min_lr: 0.003979  loss: 3.9928 (4.0257)  weight_decay: 0.0500 (0.0500)  time: 0.8240  data: 0.1902  max mem: 41794
Epoch: [32]  [300/312]  eta: 0:00:09  lr: 0.003979  min_lr: 0.003979  loss: 3.9937 (4.0301)  weight_decay: 0.0500 (0.0500)  time: 0.5861  data: 0.0648  max mem: 41794
Epoch: [32]  [310/312]  eta: 0:00:01  lr: 0.003979  min_lr: 0.003979  loss: 4.1473 (4.0251)  weight_decay: 0.0500 (0.0500)  time: 0.4621  data: 0.0001  max mem: 41794
Epoch: [32]  [311/312]  eta: 0:00:00  lr: 0.003979  min_lr: 0.003979  loss: 4.1579 (4.0276)  weight_decay: 0.0500 (0.0500)  time: 0.4626  data: 0.0001  max mem: 41794
Epoch: [32] Total time: 0:04:00 (0.7697 s / it)
Averaged stats: lr: 0.003979  min_lr: 0.003979  loss: 4.1579 (3.9643)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.5484 (1.5484)  acc1: 71.8750 (71.8750)  acc5: 90.3646 (90.3646)  time: 8.5935  data: 8.3814  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9204 (1.8010)  acc1: 63.5417 (63.2640)  acc5: 84.7656 (85.8560)  time: 1.1056  data: 0.9314  max mem: 41794
Test: Total time: 0:00:10 (1.1390 s / it)
* Acc@1 63.752 Acc@5 86.092 loss 1.804
Accuracy of the model on the 50000 test images: 63.8%
Max accuracy: 63.75%
Epoch: [33]  [  0/312]  eta: 1:19:41  lr: 0.003979  min_lr: 0.003979  loss: 3.8323 (3.8323)  weight_decay: 0.0500 (0.0500)  time: 15.3245  data: 12.0758  max mem: 41794
Epoch: [33]  [ 10/312]  eta: 0:11:14  lr: 0.003979  min_lr: 0.003979  loss: 4.3696 (4.3203)  weight_decay: 0.0500 (0.0500)  time: 2.2323  data: 1.3134  max mem: 41794
Epoch: [33]  [ 20/312]  eta: 0:07:22  lr: 0.003979  min_lr: 0.003979  loss: 4.0060 (3.9665)  weight_decay: 0.0500 (0.0500)  time: 0.8249  data: 0.1993  max mem: 41794
Epoch: [33]  [ 30/312]  eta: 0:05:37  lr: 0.003978  min_lr: 0.003978  loss: 3.7464 (3.9493)  weight_decay: 0.0500 (0.0500)  time: 0.6258  data: 0.0881  max mem: 41794
Epoch: [33]  [ 40/312]  eta: 0:04:42  lr: 0.003978  min_lr: 0.003978  loss: 3.9568 (3.9638)  weight_decay: 0.0500 (0.0500)  time: 0.5346  data: 0.0296  max mem: 41794
Epoch: [33]  [ 50/312]  eta: 0:04:19  lr: 0.003978  min_lr: 0.003978  loss: 4.0585 (3.9903)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.1381  max mem: 41794
Epoch: [33]  [ 60/312]  eta: 0:03:48  lr: 0.003978  min_lr: 0.003978  loss: 4.1010 (4.0035)  weight_decay: 0.0500 (0.0500)  time: 0.6444  data: 0.1163  max mem: 41794
Epoch: [33]  [ 70/312]  eta: 0:03:38  lr: 0.003978  min_lr: 0.003978  loss: 4.0735 (4.0000)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.1841  max mem: 41794
Epoch: [33]  [ 80/312]  eta: 0:03:30  lr: 0.003978  min_lr: 0.003978  loss: 3.9293 (3.9598)  weight_decay: 0.0500 (0.0500)  time: 0.8996  data: 0.3998  max mem: 41794
Epoch: [33]  [ 90/312]  eta: 0:03:11  lr: 0.003978  min_lr: 0.003978  loss: 3.5089 (3.9305)  weight_decay: 0.0500 (0.0500)  time: 0.7271  data: 0.2181  max mem: 41794
Epoch: [33]  [100/312]  eta: 0:03:05  lr: 0.003978  min_lr: 0.003978  loss: 3.9210 (3.9438)  weight_decay: 0.0500 (0.0500)  time: 0.7521  data: 0.2340  max mem: 41794
Epoch: [33]  [110/312]  eta: 0:02:49  lr: 0.003978  min_lr: 0.003978  loss: 3.8547 (3.9134)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.2322  max mem: 41794
Epoch: [33]  [120/312]  eta: 0:02:40  lr: 0.003977  min_lr: 0.003977  loss: 3.8008 (3.9279)  weight_decay: 0.0500 (0.0500)  time: 0.6470  data: 0.1492  max mem: 41794
Epoch: [33]  [130/312]  eta: 0:02:33  lr: 0.003977  min_lr: 0.003977  loss: 4.0163 (3.9247)  weight_decay: 0.0500 (0.0500)  time: 0.8567  data: 0.3533  max mem: 41794
Epoch: [33]  [140/312]  eta: 0:02:21  lr: 0.003977  min_lr: 0.003977  loss: 3.8906 (3.9166)  weight_decay: 0.0500 (0.0500)  time: 0.7106  data: 0.2046  max mem: 41794
Epoch: [33]  [150/312]  eta: 0:02:13  lr: 0.003977  min_lr: 0.003977  loss: 3.8906 (3.9138)  weight_decay: 0.0500 (0.0500)  time: 0.7079  data: 0.2050  max mem: 41794
Epoch: [33]  [160/312]  eta: 0:02:05  lr: 0.003977  min_lr: 0.003977  loss: 3.9149 (3.9151)  weight_decay: 0.0500 (0.0500)  time: 0.8848  data: 0.3860  max mem: 41794
Epoch: [33]  [170/312]  eta: 0:01:54  lr: 0.003977  min_lr: 0.003977  loss: 3.8770 (3.9116)  weight_decay: 0.0500 (0.0500)  time: 0.6850  data: 0.1816  max mem: 41794
Epoch: [33]  [180/312]  eta: 0:01:47  lr: 0.003977  min_lr: 0.003977  loss: 4.2460 (3.9309)  weight_decay: 0.0500 (0.0500)  time: 0.6955  data: 0.1911  max mem: 41794
Epoch: [33]  [190/312]  eta: 0:01:37  lr: 0.003977  min_lr: 0.003977  loss: 4.2632 (3.9347)  weight_decay: 0.0500 (0.0500)  time: 0.6955  data: 0.1910  max mem: 41794
Epoch: [33]  [200/312]  eta: 0:01:29  lr: 0.003977  min_lr: 0.003977  loss: 3.8619 (3.9260)  weight_decay: 0.0500 (0.0500)  time: 0.6735  data: 0.1689  max mem: 41794
Epoch: [33]  [210/312]  eta: 0:01:21  lr: 0.003977  min_lr: 0.003977  loss: 4.1070 (3.9377)  weight_decay: 0.0500 (0.0500)  time: 0.8568  data: 0.3561  max mem: 41794
Epoch: [33]  [220/312]  eta: 0:01:12  lr: 0.003976  min_lr: 0.003976  loss: 4.1279 (3.9400)  weight_decay: 0.0500 (0.0500)  time: 0.6869  data: 0.1878  max mem: 41794
Epoch: [33]  [230/312]  eta: 0:01:05  lr: 0.003976  min_lr: 0.003976  loss: 4.0379 (3.9368)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.1878  max mem: 41794
Epoch: [33]  [240/312]  eta: 0:00:57  lr: 0.003976  min_lr: 0.003976  loss: 3.6117 (3.9396)  weight_decay: 0.0500 (0.0500)  time: 0.8662  data: 0.3686  max mem: 41794
Epoch: [33]  [250/312]  eta: 0:00:48  lr: 0.003976  min_lr: 0.003976  loss: 3.5733 (3.9348)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.1815  max mem: 41794
Epoch: [33]  [260/312]  eta: 0:00:40  lr: 0.003976  min_lr: 0.003976  loss: 4.1486 (3.9409)  weight_decay: 0.0500 (0.0500)  time: 0.7001  data: 0.1957  max mem: 41794
Epoch: [33]  [270/312]  eta: 0:00:32  lr: 0.003976  min_lr: 0.003976  loss: 3.5753 (3.9206)  weight_decay: 0.0500 (0.0500)  time: 0.7001  data: 0.1958  max mem: 41794
Epoch: [33]  [280/312]  eta: 0:00:24  lr: 0.003976  min_lr: 0.003976  loss: 3.5184 (3.9239)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.1782  max mem: 41794
Epoch: [33]  [290/312]  eta: 0:00:17  lr: 0.003976  min_lr: 0.003976  loss: 4.0772 (3.9263)  weight_decay: 0.0500 (0.0500)  time: 0.8565  data: 0.3521  max mem: 41794
Epoch: [33]  [300/312]  eta: 0:00:09  lr: 0.003976  min_lr: 0.003976  loss: 3.8267 (3.9151)  weight_decay: 0.0500 (0.0500)  time: 0.6636  data: 0.1745  max mem: 41794
Epoch: [33]  [310/312]  eta: 0:00:01  lr: 0.003975  min_lr: 0.003975  loss: 3.7031 (3.9141)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [33]  [311/312]  eta: 0:00:00  lr: 0.003975  min_lr: 0.003975  loss: 3.7031 (3.9158)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [33] Total time: 0:03:59 (0.7682 s / it)
Averaged stats: lr: 0.003975  min_lr: 0.003975  loss: 3.7031 (3.9620)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.4199 (1.4199)  acc1: 72.9167 (72.9167)  acc5: 90.6250 (90.6250)  time: 8.3985  data: 8.1865  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7941 (1.7002)  acc1: 62.5000 (63.8240)  acc5: 85.6771 (86.0960)  time: 1.0937  data: 0.9097  max mem: 41794
Test: Total time: 0:00:09 (1.1067 s / it)
* Acc@1 64.354 Acc@5 86.204 loss 1.696
Accuracy of the model on the 50000 test images: 64.4%
Max accuracy: 64.35%
Epoch: [34]  [  0/312]  eta: 1:19:46  lr: 0.003975  min_lr: 0.003975  loss: 4.3085 (4.3085)  weight_decay: 0.0500 (0.0500)  time: 15.3400  data: 12.3446  max mem: 41794
Epoch: [34]  [ 10/312]  eta: 0:11:02  lr: 0.003975  min_lr: 0.003975  loss: 3.8684 (3.8386)  weight_decay: 0.0500 (0.0500)  time: 2.1950  data: 1.4096  max mem: 41794
Epoch: [34]  [ 20/312]  eta: 0:07:21  lr: 0.003975  min_lr: 0.003975  loss: 3.8473 (3.8579)  weight_decay: 0.0500 (0.0500)  time: 0.8190  data: 0.2255  max mem: 41794
Epoch: [34]  [ 30/312]  eta: 0:05:35  lr: 0.003975  min_lr: 0.003975  loss: 4.2509 (4.0103)  weight_decay: 0.0500 (0.0500)  time: 0.6394  data: 0.0678  max mem: 41794
Epoch: [34]  [ 40/312]  eta: 0:04:44  lr: 0.003975  min_lr: 0.003975  loss: 4.1443 (3.9521)  weight_decay: 0.0500 (0.0500)  time: 0.5599  data: 0.0202  max mem: 41794
Epoch: [34]  [ 50/312]  eta: 0:04:23  lr: 0.003975  min_lr: 0.003975  loss: 3.7223 (3.9125)  weight_decay: 0.0500 (0.0500)  time: 0.7195  data: 0.1051  max mem: 41794
Epoch: [34]  [ 60/312]  eta: 0:03:52  lr: 0.003975  min_lr: 0.003975  loss: 3.8018 (3.9188)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0856  max mem: 41794
Epoch: [34]  [ 70/312]  eta: 0:03:44  lr: 0.003975  min_lr: 0.003975  loss: 4.0633 (3.9219)  weight_decay: 0.0500 (0.0500)  time: 0.7290  data: 0.1506  max mem: 41794
Epoch: [34]  [ 80/312]  eta: 0:03:27  lr: 0.003974  min_lr: 0.003974  loss: 4.1240 (3.9686)  weight_decay: 0.0500 (0.0500)  time: 0.8095  data: 0.1581  max mem: 41794
Epoch: [34]  [ 90/312]  eta: 0:03:12  lr: 0.003974  min_lr: 0.003974  loss: 4.1494 (3.9723)  weight_decay: 0.0500 (0.0500)  time: 0.6556  data: 0.0860  max mem: 41794
Epoch: [34]  [100/312]  eta: 0:03:04  lr: 0.003974  min_lr: 0.003974  loss: 4.0854 (3.9816)  weight_decay: 0.0500 (0.0500)  time: 0.7673  data: 0.2081  max mem: 41794
Epoch: [34]  [110/312]  eta: 0:02:48  lr: 0.003974  min_lr: 0.003974  loss: 4.1305 (3.9705)  weight_decay: 0.0500 (0.0500)  time: 0.6916  data: 0.1303  max mem: 41794
Epoch: [34]  [120/312]  eta: 0:02:41  lr: 0.003974  min_lr: 0.003974  loss: 4.1499 (3.9634)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.1573  max mem: 41794
Epoch: [34]  [130/312]  eta: 0:02:34  lr: 0.003974  min_lr: 0.003974  loss: 4.1720 (3.9778)  weight_decay: 0.0500 (0.0500)  time: 0.9210  data: 0.3531  max mem: 41794
Epoch: [34]  [140/312]  eta: 0:02:21  lr: 0.003974  min_lr: 0.003974  loss: 4.1863 (3.9798)  weight_decay: 0.0500 (0.0500)  time: 0.7168  data: 0.1977  max mem: 41794
Epoch: [34]  [150/312]  eta: 0:02:14  lr: 0.003974  min_lr: 0.003974  loss: 4.0549 (3.9820)  weight_decay: 0.0500 (0.0500)  time: 0.7267  data: 0.1575  max mem: 41794
Epoch: [34]  [160/312]  eta: 0:02:05  lr: 0.003974  min_lr: 0.003974  loss: 4.0126 (3.9775)  weight_decay: 0.0500 (0.0500)  time: 0.8246  data: 0.1563  max mem: 41794
Epoch: [34]  [170/312]  eta: 0:01:56  lr: 0.003973  min_lr: 0.003973  loss: 3.9578 (3.9675)  weight_decay: 0.0500 (0.0500)  time: 0.6952  data: 0.1092  max mem: 41794
Epoch: [34]  [180/312]  eta: 0:01:47  lr: 0.003973  min_lr: 0.003973  loss: 3.9983 (3.9728)  weight_decay: 0.0500 (0.0500)  time: 0.7658  data: 0.2190  max mem: 41794
Epoch: [34]  [190/312]  eta: 0:01:37  lr: 0.003973  min_lr: 0.003973  loss: 3.9755 (3.9544)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.1106  max mem: 41794
Epoch: [34]  [200/312]  eta: 0:01:29  lr: 0.003973  min_lr: 0.003973  loss: 4.0754 (3.9538)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1261  max mem: 41794
Epoch: [34]  [210/312]  eta: 0:01:21  lr: 0.003973  min_lr: 0.003973  loss: 4.1782 (3.9565)  weight_decay: 0.0500 (0.0500)  time: 0.8237  data: 0.2309  max mem: 41794
Epoch: [34]  [220/312]  eta: 0:01:12  lr: 0.003973  min_lr: 0.003973  loss: 3.8813 (3.9427)  weight_decay: 0.0500 (0.0500)  time: 0.6638  data: 0.1168  max mem: 41794
Epoch: [34]  [230/312]  eta: 0:01:05  lr: 0.003973  min_lr: 0.003973  loss: 4.0025 (3.9483)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.1496  max mem: 41794
Epoch: [34]  [240/312]  eta: 0:00:56  lr: 0.003973  min_lr: 0.003973  loss: 4.2194 (3.9498)  weight_decay: 0.0500 (0.0500)  time: 0.7308  data: 0.1382  max mem: 41794
Epoch: [34]  [250/312]  eta: 0:00:48  lr: 0.003972  min_lr: 0.003972  loss: 4.1512 (3.9520)  weight_decay: 0.0500 (0.0500)  time: 0.6641  data: 0.1298  max mem: 41794
Epoch: [34]  [260/312]  eta: 0:00:40  lr: 0.003972  min_lr: 0.003972  loss: 4.1512 (3.9574)  weight_decay: 0.0500 (0.0500)  time: 0.7950  data: 0.2770  max mem: 41794
Epoch: [34]  [270/312]  eta: 0:00:32  lr: 0.003972  min_lr: 0.003972  loss: 4.1733 (3.9589)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.1479  max mem: 41794
Epoch: [34]  [280/312]  eta: 0:00:24  lr: 0.003972  min_lr: 0.003972  loss: 4.1418 (3.9580)  weight_decay: 0.0500 (0.0500)  time: 0.7103  data: 0.1684  max mem: 41794
Epoch: [34]  [290/312]  eta: 0:00:17  lr: 0.003972  min_lr: 0.003972  loss: 4.1081 (3.9513)  weight_decay: 0.0500 (0.0500)  time: 0.8288  data: 0.2404  max mem: 41794
Epoch: [34]  [300/312]  eta: 0:00:09  lr: 0.003972  min_lr: 0.003972  loss: 3.9364 (3.9527)  weight_decay: 0.0500 (0.0500)  time: 0.6015  data: 0.0723  max mem: 41794
Epoch: [34]  [310/312]  eta: 0:00:01  lr: 0.003972  min_lr: 0.003972  loss: 3.9788 (3.9453)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [34]  [311/312]  eta: 0:00:00  lr: 0.003972  min_lr: 0.003972  loss: 3.9788 (3.9431)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [34] Total time: 0:03:58 (0.7641 s / it)
Averaged stats: lr: 0.003972  min_lr: 0.003972  loss: 3.9788 (3.9607)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.6183 (1.6183)  acc1: 71.0938 (71.0938)  acc5: 90.4948 (90.4948)  time: 8.3490  data: 8.1426  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9111 (1.7977)  acc1: 62.3698 (64.4160)  acc5: 86.1979 (85.9680)  time: 1.0768  data: 0.9048  max mem: 41794
Test: Total time: 0:00:09 (1.1088 s / it)
* Acc@1 64.022 Acc@5 86.218 loss 1.810
Accuracy of the model on the 50000 test images: 64.0%
Max accuracy: 64.35%
Epoch: [35]  [  0/312]  eta: 1:24:00  lr: 0.003972  min_lr: 0.003972  loss: 3.4878 (3.4878)  weight_decay: 0.0500 (0.0500)  time: 16.1563  data: 12.8807  max mem: 41794
Epoch: [35]  [ 10/312]  eta: 0:11:24  lr: 0.003972  min_lr: 0.003972  loss: 3.7006 (3.7519)  weight_decay: 0.0500 (0.0500)  time: 2.2651  data: 1.3784  max mem: 41794
Epoch: [35]  [ 20/312]  eta: 0:07:45  lr: 0.003972  min_lr: 0.003972  loss: 3.8059 (3.7948)  weight_decay: 0.0500 (0.0500)  time: 0.8656  data: 0.1969  max mem: 41794
Epoch: [35]  [ 30/312]  eta: 0:05:53  lr: 0.003971  min_lr: 0.003971  loss: 4.0459 (3.9108)  weight_decay: 0.0500 (0.0500)  time: 0.6945  data: 0.0881  max mem: 41794
Epoch: [35]  [ 40/312]  eta: 0:04:51  lr: 0.003971  min_lr: 0.003971  loss: 4.0793 (3.8926)  weight_decay: 0.0500 (0.0500)  time: 0.5261  data: 0.0078  max mem: 41794
Epoch: [35]  [ 50/312]  eta: 0:04:28  lr: 0.003971  min_lr: 0.003971  loss: 4.0793 (3.9209)  weight_decay: 0.0500 (0.0500)  time: 0.6711  data: 0.0534  max mem: 41794
Epoch: [35]  [ 60/312]  eta: 0:03:56  lr: 0.003971  min_lr: 0.003971  loss: 4.1886 (3.9478)  weight_decay: 0.0500 (0.0500)  time: 0.6608  data: 0.0512  max mem: 41794
Epoch: [35]  [ 70/312]  eta: 0:03:45  lr: 0.003971  min_lr: 0.003971  loss: 4.0163 (3.9134)  weight_decay: 0.0500 (0.0500)  time: 0.6912  data: 0.0586  max mem: 41794
Epoch: [35]  [ 80/312]  eta: 0:03:31  lr: 0.003971  min_lr: 0.003971  loss: 4.0163 (3.9250)  weight_decay: 0.0500 (0.0500)  time: 0.8378  data: 0.0616  max mem: 41794
Epoch: [35]  [ 90/312]  eta: 0:03:14  lr: 0.003971  min_lr: 0.003971  loss: 4.1258 (3.9543)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.0578  max mem: 41794
Epoch: [35]  [100/312]  eta: 0:03:06  lr: 0.003971  min_lr: 0.003971  loss: 4.1875 (3.9744)  weight_decay: 0.0500 (0.0500)  time: 0.7377  data: 0.1037  max mem: 41794
Epoch: [35]  [110/312]  eta: 0:02:50  lr: 0.003970  min_lr: 0.003970  loss: 4.0193 (3.9762)  weight_decay: 0.0500 (0.0500)  time: 0.6872  data: 0.0511  max mem: 41794
Epoch: [35]  [120/312]  eta: 0:02:41  lr: 0.003970  min_lr: 0.003970  loss: 3.9477 (3.9610)  weight_decay: 0.0500 (0.0500)  time: 0.6483  data: 0.0222  max mem: 41794
Epoch: [35]  [130/312]  eta: 0:02:34  lr: 0.003970  min_lr: 0.003970  loss: 3.9438 (3.9665)  weight_decay: 0.0500 (0.0500)  time: 0.8632  data: 0.0499  max mem: 41794
Epoch: [35]  [140/312]  eta: 0:02:21  lr: 0.003970  min_lr: 0.003970  loss: 4.1389 (3.9560)  weight_decay: 0.0500 (0.0500)  time: 0.7114  data: 0.0299  max mem: 41794
Epoch: [35]  [150/312]  eta: 0:02:14  lr: 0.003970  min_lr: 0.003970  loss: 4.1811 (3.9657)  weight_decay: 0.0500 (0.0500)  time: 0.7262  data: 0.0621  max mem: 41794
Epoch: [35]  [160/312]  eta: 0:02:05  lr: 0.003970  min_lr: 0.003970  loss: 4.1485 (3.9537)  weight_decay: 0.0500 (0.0500)  time: 0.8539  data: 0.0620  max mem: 41794
Epoch: [35]  [170/312]  eta: 0:01:55  lr: 0.003970  min_lr: 0.003970  loss: 3.6552 (3.9477)  weight_decay: 0.0500 (0.0500)  time: 0.6556  data: 0.0324  max mem: 41794
Epoch: [35]  [180/312]  eta: 0:01:47  lr: 0.003970  min_lr: 0.003970  loss: 4.1996 (3.9552)  weight_decay: 0.0500 (0.0500)  time: 0.7150  data: 0.0926  max mem: 41794
Epoch: [35]  [190/312]  eta: 0:01:37  lr: 0.003969  min_lr: 0.003969  loss: 3.9720 (3.9438)  weight_decay: 0.0500 (0.0500)  time: 0.6881  data: 0.0622  max mem: 41794
Epoch: [35]  [200/312]  eta: 0:01:30  lr: 0.003969  min_lr: 0.003969  loss: 3.9347 (3.9399)  weight_decay: 0.0500 (0.0500)  time: 0.7140  data: 0.0740  max mem: 41794
Epoch: [35]  [210/312]  eta: 0:01:22  lr: 0.003969  min_lr: 0.003969  loss: 4.0560 (3.9498)  weight_decay: 0.0500 (0.0500)  time: 0.8705  data: 0.1190  max mem: 41794
Epoch: [35]  [220/312]  eta: 0:01:12  lr: 0.003969  min_lr: 0.003969  loss: 4.0070 (3.9457)  weight_decay: 0.0500 (0.0500)  time: 0.6556  data: 0.0475  max mem: 41794
Epoch: [35]  [230/312]  eta: 0:01:05  lr: 0.003969  min_lr: 0.003969  loss: 4.0889 (3.9587)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.0789  max mem: 41794
Epoch: [35]  [240/312]  eta: 0:00:57  lr: 0.003969  min_lr: 0.003969  loss: 3.9460 (3.9549)  weight_decay: 0.0500 (0.0500)  time: 0.8223  data: 0.0801  max mem: 41794
Epoch: [35]  [250/312]  eta: 0:00:48  lr: 0.003969  min_lr: 0.003969  loss: 3.9159 (3.9532)  weight_decay: 0.0500 (0.0500)  time: 0.6595  data: 0.0544  max mem: 41794
Epoch: [35]  [260/312]  eta: 0:00:40  lr: 0.003969  min_lr: 0.003969  loss: 3.9159 (3.9471)  weight_decay: 0.0500 (0.0500)  time: 0.7084  data: 0.1159  max mem: 41794
Epoch: [35]  [270/312]  eta: 0:00:32  lr: 0.003968  min_lr: 0.003968  loss: 3.9106 (3.9376)  weight_decay: 0.0500 (0.0500)  time: 0.6623  data: 0.0639  max mem: 41794
Epoch: [35]  [280/312]  eta: 0:00:24  lr: 0.003968  min_lr: 0.003968  loss: 3.9753 (3.9406)  weight_decay: 0.0500 (0.0500)  time: 0.7043  data: 0.0735  max mem: 41794
Epoch: [35]  [290/312]  eta: 0:00:17  lr: 0.003968  min_lr: 0.003968  loss: 4.1872 (3.9351)  weight_decay: 0.0500 (0.0500)  time: 0.8602  data: 0.1345  max mem: 41794
Epoch: [35]  [300/312]  eta: 0:00:09  lr: 0.003968  min_lr: 0.003968  loss: 4.0632 (3.9361)  weight_decay: 0.0500 (0.0500)  time: 0.6398  data: 0.0613  max mem: 41794
Epoch: [35]  [310/312]  eta: 0:00:01  lr: 0.003968  min_lr: 0.003968  loss: 4.1845 (3.9428)  weight_decay: 0.0500 (0.0500)  time: 0.4642  data: 0.0001  max mem: 41794
Epoch: [35]  [311/312]  eta: 0:00:00  lr: 0.003968  min_lr: 0.003968  loss: 4.1845 (3.9442)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0001  max mem: 41794
Epoch: [35] Total time: 0:03:59 (0.7672 s / it)
Averaged stats: lr: 0.003968  min_lr: 0.003968  loss: 4.1845 (3.9130)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 1.5344 (1.5344)  acc1: 72.9167 (72.9167)  acc5: 89.3229 (89.3229)  time: 7.8145  data: 7.6024  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.0077 (1.8780)  acc1: 62.7604 (64.0320)  acc5: 86.5885 (85.9200)  time: 1.0200  data: 0.8448  max mem: 41794
Test: Total time: 0:00:09 (1.0301 s / it)
* Acc@1 63.652 Acc@5 86.206 loss 1.869
Accuracy of the model on the 50000 test images: 63.7%
Max accuracy: 64.35%
Epoch: [36]  [  0/312]  eta: 1:22:54  lr: 0.003968  min_lr: 0.003968  loss: 4.2320 (4.2320)  weight_decay: 0.0500 (0.0500)  time: 15.9440  data: 13.8615  max mem: 41794
Epoch: [36]  [ 10/312]  eta: 0:11:33  lr: 0.003968  min_lr: 0.003968  loss: 3.9321 (3.7797)  weight_decay: 0.0500 (0.0500)  time: 2.2962  data: 1.3429  max mem: 41794
Epoch: [36]  [ 20/312]  eta: 0:07:31  lr: 0.003968  min_lr: 0.003968  loss: 3.7600 (3.7944)  weight_decay: 0.0500 (0.0500)  time: 0.8269  data: 0.1013  max mem: 41794
Epoch: [36]  [ 30/312]  eta: 0:05:45  lr: 0.003967  min_lr: 0.003967  loss: 3.9422 (3.7990)  weight_decay: 0.0500 (0.0500)  time: 0.6338  data: 0.0657  max mem: 41794
Epoch: [36]  [ 40/312]  eta: 0:04:44  lr: 0.003967  min_lr: 0.003967  loss: 3.9422 (3.8377)  weight_decay: 0.0500 (0.0500)  time: 0.5191  data: 0.0103  max mem: 41794
Epoch: [36]  [ 50/312]  eta: 0:04:20  lr: 0.003967  min_lr: 0.003967  loss: 3.8978 (3.8018)  weight_decay: 0.0500 (0.0500)  time: 0.6360  data: 0.0512  max mem: 41794
Epoch: [36]  [ 60/312]  eta: 0:03:50  lr: 0.003967  min_lr: 0.003967  loss: 4.0279 (3.8498)  weight_decay: 0.0500 (0.0500)  time: 0.6415  data: 0.0511  max mem: 41794
Epoch: [36]  [ 70/312]  eta: 0:03:42  lr: 0.003967  min_lr: 0.003967  loss: 4.1546 (3.8748)  weight_decay: 0.0500 (0.0500)  time: 0.7320  data: 0.0699  max mem: 41794
Epoch: [36]  [ 80/312]  eta: 0:03:27  lr: 0.003967  min_lr: 0.003967  loss: 4.0998 (3.8864)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.1077  max mem: 41794
Epoch: [36]  [ 90/312]  eta: 0:03:11  lr: 0.003967  min_lr: 0.003967  loss: 3.9498 (3.8895)  weight_decay: 0.0500 (0.0500)  time: 0.6581  data: 0.0426  max mem: 41794
Epoch: [36]  [100/312]  eta: 0:03:03  lr: 0.003967  min_lr: 0.003967  loss: 3.9098 (3.8863)  weight_decay: 0.0500 (0.0500)  time: 0.7499  data: 0.0813  max mem: 41794
Epoch: [36]  [110/312]  eta: 0:02:48  lr: 0.003966  min_lr: 0.003966  loss: 4.0669 (3.9118)  weight_decay: 0.0500 (0.0500)  time: 0.7013  data: 0.0772  max mem: 41794
Epoch: [36]  [120/312]  eta: 0:02:40  lr: 0.003966  min_lr: 0.003966  loss: 4.2271 (3.9297)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.0745  max mem: 41794
Epoch: [36]  [130/312]  eta: 0:02:33  lr: 0.003966  min_lr: 0.003966  loss: 4.1203 (3.9278)  weight_decay: 0.0500 (0.0500)  time: 0.8943  data: 0.1266  max mem: 41794
Epoch: [36]  [140/312]  eta: 0:02:20  lr: 0.003966  min_lr: 0.003966  loss: 4.1512 (3.9364)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.0585  max mem: 41794
Epoch: [36]  [150/312]  eta: 0:02:13  lr: 0.003966  min_lr: 0.003966  loss: 4.1858 (3.9376)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.0597  max mem: 41794
Epoch: [36]  [160/312]  eta: 0:02:03  lr: 0.003966  min_lr: 0.003966  loss: 3.9479 (3.9312)  weight_decay: 0.0500 (0.0500)  time: 0.7825  data: 0.0867  max mem: 41794
Epoch: [36]  [170/312]  eta: 0:01:53  lr: 0.003966  min_lr: 0.003966  loss: 3.9060 (3.9260)  weight_decay: 0.0500 (0.0500)  time: 0.6425  data: 0.0534  max mem: 41794
Epoch: [36]  [180/312]  eta: 0:01:46  lr: 0.003966  min_lr: 0.003966  loss: 3.8038 (3.9218)  weight_decay: 0.0500 (0.0500)  time: 0.7134  data: 0.0983  max mem: 41794
Epoch: [36]  [190/312]  eta: 0:01:36  lr: 0.003965  min_lr: 0.003965  loss: 4.2519 (3.9396)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0783  max mem: 41794
Epoch: [36]  [200/312]  eta: 0:01:28  lr: 0.003965  min_lr: 0.003965  loss: 4.1743 (3.9410)  weight_decay: 0.0500 (0.0500)  time: 0.6703  data: 0.0988  max mem: 41794
Epoch: [36]  [210/312]  eta: 0:01:20  lr: 0.003965  min_lr: 0.003965  loss: 3.7392 (3.9359)  weight_decay: 0.0500 (0.0500)  time: 0.8399  data: 0.2051  max mem: 41794
Epoch: [36]  [220/312]  eta: 0:01:11  lr: 0.003965  min_lr: 0.003965  loss: 4.0137 (3.9399)  weight_decay: 0.0500 (0.0500)  time: 0.6692  data: 0.1068  max mem: 41794
Epoch: [36]  [230/312]  eta: 0:01:04  lr: 0.003965  min_lr: 0.003965  loss: 4.1871 (3.9488)  weight_decay: 0.0500 (0.0500)  time: 0.6960  data: 0.1333  max mem: 41794
Epoch: [36]  [240/312]  eta: 0:00:56  lr: 0.003965  min_lr: 0.003965  loss: 4.0773 (3.9467)  weight_decay: 0.0500 (0.0500)  time: 0.8098  data: 0.2246  max mem: 41794
Epoch: [36]  [250/312]  eta: 0:00:48  lr: 0.003965  min_lr: 0.003965  loss: 4.0773 (3.9489)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.1490  max mem: 41794
Epoch: [36]  [260/312]  eta: 0:00:40  lr: 0.003964  min_lr: 0.003964  loss: 4.2249 (3.9577)  weight_decay: 0.0500 (0.0500)  time: 0.7358  data: 0.2251  max mem: 41794
Epoch: [36]  [270/312]  eta: 0:00:32  lr: 0.003964  min_lr: 0.003964  loss: 4.2964 (3.9735)  weight_decay: 0.0500 (0.0500)  time: 0.6808  data: 0.1680  max mem: 41794
Epoch: [36]  [280/312]  eta: 0:00:24  lr: 0.003964  min_lr: 0.003964  loss: 4.2556 (3.9801)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.1634  max mem: 41794
Epoch: [36]  [290/312]  eta: 0:00:17  lr: 0.003964  min_lr: 0.003964  loss: 3.9999 (3.9711)  weight_decay: 0.0500 (0.0500)  time: 0.8826  data: 0.3501  max mem: 41794
Epoch: [36]  [300/312]  eta: 0:00:09  lr: 0.003964  min_lr: 0.003964  loss: 3.8732 (3.9638)  weight_decay: 0.0500 (0.0500)  time: 0.6956  data: 0.1871  max mem: 41794
Epoch: [36]  [310/312]  eta: 0:00:01  lr: 0.003964  min_lr: 0.003964  loss: 4.0599 (3.9663)  weight_decay: 0.0500 (0.0500)  time: 0.4676  data: 0.0001  max mem: 41794
Epoch: [36]  [311/312]  eta: 0:00:00  lr: 0.003964  min_lr: 0.003964  loss: 3.8918 (3.9646)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.0001  max mem: 41794
Epoch: [36] Total time: 0:03:57 (0.7612 s / it)
Averaged stats: lr: 0.003964  min_lr: 0.003964  loss: 3.8918 (3.9531)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.7445 (1.7445)  acc1: 73.5677 (73.5677)  acc5: 90.2344 (90.2344)  time: 8.6609  data: 8.4487  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.1681 (1.9881)  acc1: 62.8906 (64.5280)  acc5: 85.0260 (86.3040)  time: 1.1103  data: 0.9388  max mem: 41794
Test: Total time: 0:00:10 (1.1488 s / it)
* Acc@1 64.014 Acc@5 86.180 loss 1.988
Accuracy of the model on the 50000 test images: 64.0%
Max accuracy: 64.35%
Epoch: [37]  [  0/312]  eta: 1:22:40  lr: 0.003964  min_lr: 0.003964  loss: 4.8277 (4.8277)  weight_decay: 0.0500 (0.0500)  time: 15.8987  data: 12.9625  max mem: 41794
Epoch: [37]  [ 10/312]  eta: 0:11:27  lr: 0.003964  min_lr: 0.003964  loss: 4.2315 (4.1421)  weight_decay: 0.0500 (0.0500)  time: 2.2774  data: 1.4341  max mem: 41794
Epoch: [37]  [ 20/312]  eta: 0:07:23  lr: 0.003963  min_lr: 0.003963  loss: 3.9473 (3.9355)  weight_decay: 0.0500 (0.0500)  time: 0.8004  data: 0.1990  max mem: 41794
Epoch: [37]  [ 30/312]  eta: 0:05:39  lr: 0.003963  min_lr: 0.003963  loss: 3.6777 (3.9079)  weight_decay: 0.0500 (0.0500)  time: 0.6133  data: 0.0626  max mem: 41794
Epoch: [37]  [ 40/312]  eta: 0:04:41  lr: 0.003963  min_lr: 0.003963  loss: 3.7104 (3.8202)  weight_decay: 0.0500 (0.0500)  time: 0.5248  data: 0.0047  max mem: 41794
Epoch: [37]  [ 50/312]  eta: 0:04:22  lr: 0.003963  min_lr: 0.003963  loss: 3.7869 (3.8098)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.1002  max mem: 41794
Epoch: [37]  [ 60/312]  eta: 0:03:51  lr: 0.003963  min_lr: 0.003963  loss: 3.9356 (3.8479)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.1001  max mem: 41794
Epoch: [37]  [ 70/312]  eta: 0:03:41  lr: 0.003963  min_lr: 0.003963  loss: 3.9586 (3.8530)  weight_decay: 0.0500 (0.0500)  time: 0.6945  data: 0.0874  max mem: 41794
Epoch: [37]  [ 80/312]  eta: 0:03:25  lr: 0.003963  min_lr: 0.003963  loss: 3.7663 (3.8321)  weight_decay: 0.0500 (0.0500)  time: 0.7943  data: 0.0886  max mem: 41794
Epoch: [37]  [ 90/312]  eta: 0:03:12  lr: 0.003963  min_lr: 0.003963  loss: 3.7920 (3.8416)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.0936  max mem: 41794
Epoch: [37]  [100/312]  eta: 0:03:03  lr: 0.003962  min_lr: 0.003962  loss: 3.9357 (3.8521)  weight_decay: 0.0500 (0.0500)  time: 0.7842  data: 0.1547  max mem: 41794
Epoch: [37]  [110/312]  eta: 0:02:48  lr: 0.003962  min_lr: 0.003962  loss: 4.1168 (3.8459)  weight_decay: 0.0500 (0.0500)  time: 0.6891  data: 0.0665  max mem: 41794
Epoch: [37]  [120/312]  eta: 0:02:40  lr: 0.003962  min_lr: 0.003962  loss: 4.1633 (3.8780)  weight_decay: 0.0500 (0.0500)  time: 0.6660  data: 0.0649  max mem: 41794
Epoch: [37]  [130/312]  eta: 0:02:32  lr: 0.003962  min_lr: 0.003962  loss: 3.8563 (3.8659)  weight_decay: 0.0500 (0.0500)  time: 0.8444  data: 0.1395  max mem: 41794
Epoch: [37]  [140/312]  eta: 0:02:19  lr: 0.003962  min_lr: 0.003962  loss: 3.6555 (3.8617)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.0789  max mem: 41794
Epoch: [37]  [150/312]  eta: 0:02:12  lr: 0.003962  min_lr: 0.003962  loss: 3.9436 (3.8700)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.0953  max mem: 41794
Epoch: [37]  [160/312]  eta: 0:02:03  lr: 0.003962  min_lr: 0.003962  loss: 3.9436 (3.8750)  weight_decay: 0.0500 (0.0500)  time: 0.7887  data: 0.1109  max mem: 41794
Epoch: [37]  [170/312]  eta: 0:01:54  lr: 0.003961  min_lr: 0.003961  loss: 4.0811 (3.8795)  weight_decay: 0.0500 (0.0500)  time: 0.7063  data: 0.1150  max mem: 41794
Epoch: [37]  [180/312]  eta: 0:01:46  lr: 0.003961  min_lr: 0.003961  loss: 3.7094 (3.8612)  weight_decay: 0.0500 (0.0500)  time: 0.7702  data: 0.1974  max mem: 41794
Epoch: [37]  [190/312]  eta: 0:01:36  lr: 0.003961  min_lr: 0.003961  loss: 3.6946 (3.8589)  weight_decay: 0.0500 (0.0500)  time: 0.6642  data: 0.0986  max mem: 41794
Epoch: [37]  [200/312]  eta: 0:01:28  lr: 0.003961  min_lr: 0.003961  loss: 3.9103 (3.8632)  weight_decay: 0.0500 (0.0500)  time: 0.6747  data: 0.1139  max mem: 41794
Epoch: [37]  [210/312]  eta: 0:01:20  lr: 0.003961  min_lr: 0.003961  loss: 3.8938 (3.8514)  weight_decay: 0.0500 (0.0500)  time: 0.8209  data: 0.2168  max mem: 41794
Epoch: [37]  [220/312]  eta: 0:01:11  lr: 0.003961  min_lr: 0.003961  loss: 3.9065 (3.8604)  weight_decay: 0.0500 (0.0500)  time: 0.6288  data: 0.1035  max mem: 41794
Epoch: [37]  [230/312]  eta: 0:01:04  lr: 0.003961  min_lr: 0.003961  loss: 3.9998 (3.8578)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1302  max mem: 41794
Epoch: [37]  [240/312]  eta: 0:00:56  lr: 0.003960  min_lr: 0.003960  loss: 4.0057 (3.8671)  weight_decay: 0.0500 (0.0500)  time: 0.8096  data: 0.1526  max mem: 41794
Epoch: [37]  [250/312]  eta: 0:00:48  lr: 0.003960  min_lr: 0.003960  loss: 3.9245 (3.8624)  weight_decay: 0.0500 (0.0500)  time: 0.7015  data: 0.1105  max mem: 41794
Epoch: [37]  [260/312]  eta: 0:00:40  lr: 0.003960  min_lr: 0.003960  loss: 3.8583 (3.8670)  weight_decay: 0.0500 (0.0500)  time: 0.7789  data: 0.1822  max mem: 41794
Epoch: [37]  [270/312]  eta: 0:00:32  lr: 0.003960  min_lr: 0.003960  loss: 4.1112 (3.8782)  weight_decay: 0.0500 (0.0500)  time: 0.6952  data: 0.0960  max mem: 41794
Epoch: [37]  [280/312]  eta: 0:00:24  lr: 0.003960  min_lr: 0.003960  loss: 4.2725 (3.8799)  weight_decay: 0.0500 (0.0500)  time: 0.6997  data: 0.0701  max mem: 41794
Epoch: [37]  [290/312]  eta: 0:00:17  lr: 0.003960  min_lr: 0.003960  loss: 3.8665 (3.8766)  weight_decay: 0.0500 (0.0500)  time: 0.8676  data: 0.1008  max mem: 41794
Epoch: [37]  [300/312]  eta: 0:00:09  lr: 0.003960  min_lr: 0.003960  loss: 3.9775 (3.8804)  weight_decay: 0.0500 (0.0500)  time: 0.6542  data: 0.0324  max mem: 41794
Epoch: [37]  [310/312]  eta: 0:00:01  lr: 0.003959  min_lr: 0.003959  loss: 4.1222 (3.8898)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [37]  [311/312]  eta: 0:00:00  lr: 0.003959  min_lr: 0.003959  loss: 4.0856 (3.8865)  weight_decay: 0.0500 (0.0500)  time: 0.4670  data: 0.0001  max mem: 41794
Epoch: [37] Total time: 0:03:57 (0.7620 s / it)
Averaged stats: lr: 0.003959  min_lr: 0.003959  loss: 4.0856 (3.9230)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.5826 (1.5826)  acc1: 71.4844 (71.4844)  acc5: 90.7552 (90.7552)  time: 8.6931  data: 8.4837  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9676 (1.8606)  acc1: 61.4583 (63.7120)  acc5: 85.4167 (85.6640)  time: 1.1138  data: 0.9427  max mem: 41794
Test: Total time: 0:00:10 (1.1256 s / it)
* Acc@1 63.792 Acc@5 85.916 loss 1.862
Accuracy of the model on the 50000 test images: 63.8%
Max accuracy: 64.35%
Epoch: [38]  [  0/312]  eta: 1:19:16  lr: 0.003959  min_lr: 0.003959  loss: 3.0348 (3.0348)  weight_decay: 0.0500 (0.0500)  time: 15.2446  data: 14.5861  max mem: 41794
Epoch: [38]  [ 10/312]  eta: 0:10:13  lr: 0.003959  min_lr: 0.003959  loss: 3.7380 (3.6488)  weight_decay: 0.0500 (0.0500)  time: 2.0308  data: 1.4000  max mem: 41794
Epoch: [38]  [ 20/312]  eta: 0:07:02  lr: 0.003959  min_lr: 0.003959  loss: 3.8443 (3.7992)  weight_decay: 0.0500 (0.0500)  time: 0.7577  data: 0.1672  max mem: 41794
Epoch: [38]  [ 30/312]  eta: 0:05:23  lr: 0.003959  min_lr: 0.003959  loss: 3.9828 (3.7758)  weight_decay: 0.0500 (0.0500)  time: 0.6594  data: 0.1268  max mem: 41794
Epoch: [38]  [ 40/312]  eta: 0:04:52  lr: 0.003959  min_lr: 0.003959  loss: 4.2062 (3.8202)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1486  max mem: 41794
Epoch: [38]  [ 50/312]  eta: 0:04:32  lr: 0.003959  min_lr: 0.003959  loss: 4.1758 (3.8711)  weight_decay: 0.0500 (0.0500)  time: 0.8717  data: 0.3066  max mem: 41794
Epoch: [38]  [ 60/312]  eta: 0:03:58  lr: 0.003958  min_lr: 0.003958  loss: 4.0363 (3.8633)  weight_decay: 0.0500 (0.0500)  time: 0.6812  data: 0.1587  max mem: 41794
Epoch: [38]  [ 70/312]  eta: 0:03:46  lr: 0.003958  min_lr: 0.003958  loss: 3.9696 (3.8882)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.1737  max mem: 41794
Epoch: [38]  [ 80/312]  eta: 0:03:33  lr: 0.003958  min_lr: 0.003958  loss: 3.8188 (3.8615)  weight_decay: 0.0500 (0.0500)  time: 0.8409  data: 0.3334  max mem: 41794
Epoch: [38]  [ 90/312]  eta: 0:03:14  lr: 0.003958  min_lr: 0.003958  loss: 3.8797 (3.8767)  weight_decay: 0.0500 (0.0500)  time: 0.6618  data: 0.1750  max mem: 41794
Epoch: [38]  [100/312]  eta: 0:03:04  lr: 0.003958  min_lr: 0.003958  loss: 4.0166 (3.8717)  weight_decay: 0.0500 (0.0500)  time: 0.6677  data: 0.1820  max mem: 41794
Epoch: [38]  [110/312]  eta: 0:02:48  lr: 0.003958  min_lr: 0.003958  loss: 4.0684 (3.8847)  weight_decay: 0.0500 (0.0500)  time: 0.6553  data: 0.1674  max mem: 41794
Epoch: [38]  [120/312]  eta: 0:02:40  lr: 0.003958  min_lr: 0.003958  loss: 4.0684 (3.8904)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.1836  max mem: 41794
Epoch: [38]  [130/312]  eta: 0:02:32  lr: 0.003957  min_lr: 0.003957  loss: 3.7359 (3.8725)  weight_decay: 0.0500 (0.0500)  time: 0.8679  data: 0.3765  max mem: 41794
Epoch: [38]  [140/312]  eta: 0:02:20  lr: 0.003957  min_lr: 0.003957  loss: 3.7359 (3.8714)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.1936  max mem: 41794
Epoch: [38]  [150/312]  eta: 0:02:12  lr: 0.003957  min_lr: 0.003957  loss: 4.2617 (3.8844)  weight_decay: 0.0500 (0.0500)  time: 0.6807  data: 0.1971  max mem: 41794
Epoch: [38]  [160/312]  eta: 0:02:04  lr: 0.003957  min_lr: 0.003957  loss: 4.1479 (3.8863)  weight_decay: 0.0500 (0.0500)  time: 0.8321  data: 0.3428  max mem: 41794
Epoch: [38]  [170/312]  eta: 0:01:54  lr: 0.003957  min_lr: 0.003957  loss: 3.7878 (3.8840)  weight_decay: 0.0500 (0.0500)  time: 0.7041  data: 0.1844  max mem: 41794
Epoch: [38]  [180/312]  eta: 0:01:46  lr: 0.003957  min_lr: 0.003957  loss: 3.9650 (3.8852)  weight_decay: 0.0500 (0.0500)  time: 0.7327  data: 0.1848  max mem: 41794
Epoch: [38]  [190/312]  eta: 0:01:36  lr: 0.003957  min_lr: 0.003957  loss: 3.9954 (3.8894)  weight_decay: 0.0500 (0.0500)  time: 0.6658  data: 0.1468  max mem: 41794
Epoch: [38]  [200/312]  eta: 0:01:28  lr: 0.003956  min_lr: 0.003956  loss: 3.9954 (3.8963)  weight_decay: 0.0500 (0.0500)  time: 0.6663  data: 0.1727  max mem: 41794
Epoch: [38]  [210/312]  eta: 0:01:21  lr: 0.003956  min_lr: 0.003956  loss: 4.0117 (3.8993)  weight_decay: 0.0500 (0.0500)  time: 0.8645  data: 0.3636  max mem: 41794
Epoch: [38]  [220/312]  eta: 0:01:12  lr: 0.003956  min_lr: 0.003956  loss: 4.1315 (3.9116)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.1915  max mem: 41794
Epoch: [38]  [230/312]  eta: 0:01:04  lr: 0.003956  min_lr: 0.003956  loss: 4.0207 (3.9051)  weight_decay: 0.0500 (0.0500)  time: 0.6802  data: 0.1779  max mem: 41794
Epoch: [38]  [240/312]  eta: 0:00:56  lr: 0.003956  min_lr: 0.003956  loss: 3.7693 (3.8914)  weight_decay: 0.0500 (0.0500)  time: 0.8067  data: 0.2999  max mem: 41794
Epoch: [38]  [250/312]  eta: 0:00:48  lr: 0.003956  min_lr: 0.003956  loss: 3.5572 (3.8840)  weight_decay: 0.0500 (0.0500)  time: 0.6823  data: 0.1739  max mem: 41794
Epoch: [38]  [260/312]  eta: 0:00:40  lr: 0.003956  min_lr: 0.003956  loss: 3.7508 (3.8856)  weight_decay: 0.0500 (0.0500)  time: 0.7480  data: 0.2054  max mem: 41794
Epoch: [38]  [270/312]  eta: 0:00:32  lr: 0.003955  min_lr: 0.003955  loss: 3.8082 (3.8801)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.1540  max mem: 41794
Epoch: [38]  [280/312]  eta: 0:00:24  lr: 0.003955  min_lr: 0.003955  loss: 3.7519 (3.8702)  weight_decay: 0.0500 (0.0500)  time: 0.6906  data: 0.1369  max mem: 41794
Epoch: [38]  [290/312]  eta: 0:00:17  lr: 0.003955  min_lr: 0.003955  loss: 3.6933 (3.8703)  weight_decay: 0.0500 (0.0500)  time: 0.7948  data: 0.2111  max mem: 41794
Epoch: [38]  [300/312]  eta: 0:00:09  lr: 0.003955  min_lr: 0.003955  loss: 3.7875 (3.8677)  weight_decay: 0.0500 (0.0500)  time: 0.5876  data: 0.0746  max mem: 41794
Epoch: [38]  [310/312]  eta: 0:00:01  lr: 0.003955  min_lr: 0.003955  loss: 3.9347 (3.8699)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0001  max mem: 41794
Epoch: [38]  [311/312]  eta: 0:00:00  lr: 0.003955  min_lr: 0.003955  loss: 3.9347 (3.8700)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0001  max mem: 41794
Epoch: [38] Total time: 0:03:56 (0.7596 s / it)
Averaged stats: lr: 0.003955  min_lr: 0.003955  loss: 3.9347 (3.9003)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.5367 (1.5367)  acc1: 71.0938 (71.0938)  acc5: 92.4479 (92.4479)  time: 8.0564  data: 7.8458  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9048 (1.7825)  acc1: 61.4583 (64.7360)  acc5: 86.9792 (86.7360)  time: 1.0431  data: 0.8719  max mem: 41794
Test: Total time: 0:00:09 (1.0555 s / it)
* Acc@1 64.766 Acc@5 86.652 loss 1.792
Accuracy of the model on the 50000 test images: 64.8%
Max accuracy: 64.77%
Epoch: [39]  [  0/312]  eta: 1:19:55  lr: 0.003955  min_lr: 0.003955  loss: 4.0678 (4.0678)  weight_decay: 0.0500 (0.0500)  time: 15.3714  data: 14.8409  max mem: 41794
Epoch: [39]  [ 10/312]  eta: 0:12:00  lr: 0.003955  min_lr: 0.003955  loss: 3.6848 (3.7476)  weight_decay: 0.0500 (0.0500)  time: 2.3854  data: 1.3539  max mem: 41794
Epoch: [39]  [ 20/312]  eta: 0:07:20  lr: 0.003954  min_lr: 0.003954  loss: 3.6094 (3.6996)  weight_decay: 0.0500 (0.0500)  time: 0.8165  data: 0.0031  max mem: 41794
Epoch: [39]  [ 30/312]  eta: 0:05:34  lr: 0.003954  min_lr: 0.003954  loss: 3.6094 (3.6932)  weight_decay: 0.0500 (0.0500)  time: 0.5275  data: 0.0010  max mem: 41794
Epoch: [39]  [ 40/312]  eta: 0:04:38  lr: 0.003954  min_lr: 0.003954  loss: 3.6211 (3.6905)  weight_decay: 0.0500 (0.0500)  time: 0.5131  data: 0.0010  max mem: 41794
Epoch: [39]  [ 50/312]  eta: 0:04:01  lr: 0.003954  min_lr: 0.003954  loss: 3.8124 (3.7189)  weight_decay: 0.0500 (0.0500)  time: 0.5133  data: 0.0008  max mem: 41794
Epoch: [39]  [ 60/312]  eta: 0:03:36  lr: 0.003954  min_lr: 0.003954  loss: 3.8401 (3.7344)  weight_decay: 0.0500 (0.0500)  time: 0.5196  data: 0.0011  max mem: 41794
Epoch: [39]  [ 70/312]  eta: 0:03:24  lr: 0.003954  min_lr: 0.003954  loss: 3.8564 (3.7612)  weight_decay: 0.0500 (0.0500)  time: 0.6485  data: 0.1374  max mem: 41794
Epoch: [39]  [ 80/312]  eta: 0:03:18  lr: 0.003954  min_lr: 0.003954  loss: 4.1079 (3.7759)  weight_decay: 0.0500 (0.0500)  time: 0.8557  data: 0.3249  max mem: 41794
Epoch: [39]  [ 90/312]  eta: 0:03:03  lr: 0.003953  min_lr: 0.003953  loss: 4.0534 (3.8056)  weight_decay: 0.0500 (0.0500)  time: 0.7626  data: 0.2049  max mem: 41794
Epoch: [39]  [100/312]  eta: 0:02:54  lr: 0.003953  min_lr: 0.003953  loss: 4.1251 (3.8265)  weight_decay: 0.0500 (0.0500)  time: 0.6788  data: 0.1578  max mem: 41794
Epoch: [39]  [110/312]  eta: 0:02:42  lr: 0.003953  min_lr: 0.003953  loss: 3.9343 (3.8048)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.1940  max mem: 41794
Epoch: [39]  [120/312]  eta: 0:02:33  lr: 0.003953  min_lr: 0.003953  loss: 3.9343 (3.8288)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.1640  max mem: 41794
Epoch: [39]  [130/312]  eta: 0:02:25  lr: 0.003953  min_lr: 0.003953  loss: 4.0743 (3.8405)  weight_decay: 0.0500 (0.0500)  time: 0.7896  data: 0.2562  max mem: 41794
Epoch: [39]  [140/312]  eta: 0:02:16  lr: 0.003953  min_lr: 0.003953  loss: 4.0391 (3.8452)  weight_decay: 0.0500 (0.0500)  time: 0.7555  data: 0.2084  max mem: 41794
Epoch: [39]  [150/312]  eta: 0:02:07  lr: 0.003952  min_lr: 0.003952  loss: 3.9485 (3.8508)  weight_decay: 0.0500 (0.0500)  time: 0.7127  data: 0.1406  max mem: 41794
Epoch: [39]  [160/312]  eta: 0:02:00  lr: 0.003952  min_lr: 0.003952  loss: 4.1573 (3.8645)  weight_decay: 0.0500 (0.0500)  time: 0.7910  data: 0.1815  max mem: 41794
Epoch: [39]  [170/312]  eta: 0:01:51  lr: 0.003952  min_lr: 0.003952  loss: 4.1244 (3.8576)  weight_decay: 0.0500 (0.0500)  time: 0.7622  data: 0.1753  max mem: 41794
Epoch: [39]  [180/312]  eta: 0:01:43  lr: 0.003952  min_lr: 0.003952  loss: 4.1110 (3.8644)  weight_decay: 0.0500 (0.0500)  time: 0.7026  data: 0.1123  max mem: 41794
Epoch: [39]  [190/312]  eta: 0:01:35  lr: 0.003952  min_lr: 0.003952  loss: 4.0080 (3.8623)  weight_decay: 0.0500 (0.0500)  time: 0.7698  data: 0.0916  max mem: 41794
Epoch: [39]  [200/312]  eta: 0:01:26  lr: 0.003952  min_lr: 0.003952  loss: 3.7319 (3.8437)  weight_decay: 0.0500 (0.0500)  time: 0.6905  data: 0.0860  max mem: 41794
Epoch: [39]  [210/312]  eta: 0:01:19  lr: 0.003951  min_lr: 0.003951  loss: 3.6546 (3.8379)  weight_decay: 0.0500 (0.0500)  time: 0.6879  data: 0.1241  max mem: 41794
Epoch: [39]  [220/312]  eta: 0:01:11  lr: 0.003951  min_lr: 0.003951  loss: 4.0232 (3.8396)  weight_decay: 0.0500 (0.0500)  time: 0.8210  data: 0.1809  max mem: 41794
Epoch: [39]  [230/312]  eta: 0:01:03  lr: 0.003951  min_lr: 0.003951  loss: 3.9689 (3.8313)  weight_decay: 0.0500 (0.0500)  time: 0.7299  data: 0.1338  max mem: 41794
Epoch: [39]  [240/312]  eta: 0:00:55  lr: 0.003951  min_lr: 0.003951  loss: 3.8897 (3.8383)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.1150  max mem: 41794
Epoch: [39]  [250/312]  eta: 0:00:47  lr: 0.003951  min_lr: 0.003951  loss: 3.8406 (3.8299)  weight_decay: 0.0500 (0.0500)  time: 0.7481  data: 0.1736  max mem: 41794
Epoch: [39]  [260/312]  eta: 0:00:39  lr: 0.003951  min_lr: 0.003951  loss: 3.8638 (3.8335)  weight_decay: 0.0500 (0.0500)  time: 0.7124  data: 0.1382  max mem: 41794
Epoch: [39]  [270/312]  eta: 0:00:32  lr: 0.003951  min_lr: 0.003951  loss: 3.7789 (3.8230)  weight_decay: 0.0500 (0.0500)  time: 0.7298  data: 0.1120  max mem: 41794
Epoch: [39]  [280/312]  eta: 0:00:24  lr: 0.003950  min_lr: 0.003950  loss: 3.7923 (3.8255)  weight_decay: 0.0500 (0.0500)  time: 0.7201  data: 0.1028  max mem: 41794
Epoch: [39]  [290/312]  eta: 0:00:16  lr: 0.003950  min_lr: 0.003950  loss: 3.9078 (3.8207)  weight_decay: 0.0500 (0.0500)  time: 0.7784  data: 0.0727  max mem: 41794
Epoch: [39]  [300/312]  eta: 0:00:09  lr: 0.003950  min_lr: 0.003950  loss: 3.9576 (3.8274)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.0556  max mem: 41794
Epoch: [39]  [310/312]  eta: 0:00:01  lr: 0.003950  min_lr: 0.003950  loss: 3.8981 (3.8267)  weight_decay: 0.0500 (0.0500)  time: 0.4871  data: 0.0120  max mem: 41794
Epoch: [39]  [311/312]  eta: 0:00:00  lr: 0.003950  min_lr: 0.003950  loss: 3.9235 (3.8270)  weight_decay: 0.0500 (0.0500)  time: 0.4862  data: 0.0120  max mem: 41794
Epoch: [39] Total time: 0:03:55 (0.7558 s / it)
Averaged stats: lr: 0.003950  min_lr: 0.003950  loss: 3.9235 (3.8830)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.8143 (1.8143)  acc1: 71.8750 (71.8750)  acc5: 89.9740 (89.9740)  time: 8.2085  data: 8.0031  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9706 (1.9651)  acc1: 64.0625 (63.8400)  acc5: 87.1094 (86.1920)  time: 1.0601  data: 0.8893  max mem: 41794
Test: Total time: 0:00:09 (1.0709 s / it)
* Acc@1 63.666 Acc@5 85.988 loss 1.979
Accuracy of the model on the 50000 test images: 63.7%
Max accuracy: 64.77%
Epoch: [40]  [  0/312]  eta: 1:18:23  lr: 0.003950  min_lr: 0.003950  loss: 3.5016 (3.5016)  weight_decay: 0.0500 (0.0500)  time: 15.0753  data: 14.5767  max mem: 41794
Epoch: [40]  [ 10/312]  eta: 0:10:18  lr: 0.003950  min_lr: 0.003950  loss: 3.8036 (3.6802)  weight_decay: 0.0500 (0.0500)  time: 2.0492  data: 1.3837  max mem: 41794
Epoch: [40]  [ 20/312]  eta: 0:07:18  lr: 0.003950  min_lr: 0.003950  loss: 3.9678 (3.8049)  weight_decay: 0.0500 (0.0500)  time: 0.8233  data: 0.1464  max mem: 41794
Epoch: [40]  [ 30/312]  eta: 0:05:35  lr: 0.003949  min_lr: 0.003949  loss: 4.0538 (3.8609)  weight_decay: 0.0500 (0.0500)  time: 0.7163  data: 0.1160  max mem: 41794
Epoch: [40]  [ 40/312]  eta: 0:04:57  lr: 0.003949  min_lr: 0.003949  loss: 4.0971 (3.9166)  weight_decay: 0.0500 (0.0500)  time: 0.6653  data: 0.0971  max mem: 41794
Epoch: [40]  [ 50/312]  eta: 0:04:34  lr: 0.003949  min_lr: 0.003949  loss: 4.0971 (3.9335)  weight_decay: 0.0500 (0.0500)  time: 0.8293  data: 0.2198  max mem: 41794
Epoch: [40]  [ 60/312]  eta: 0:04:01  lr: 0.003949  min_lr: 0.003949  loss: 3.9258 (3.9041)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.1248  max mem: 41794
Epoch: [40]  [ 70/312]  eta: 0:03:46  lr: 0.003949  min_lr: 0.003949  loss: 3.8746 (3.9074)  weight_decay: 0.0500 (0.0500)  time: 0.6544  data: 0.1419  max mem: 41794
Epoch: [40]  [ 80/312]  eta: 0:03:37  lr: 0.003949  min_lr: 0.003949  loss: 4.1221 (3.8907)  weight_decay: 0.0500 (0.0500)  time: 0.8706  data: 0.3516  max mem: 41794
Epoch: [40]  [ 90/312]  eta: 0:03:16  lr: 0.003948  min_lr: 0.003948  loss: 3.8497 (3.8730)  weight_decay: 0.0500 (0.0500)  time: 0.7007  data: 0.2103  max mem: 41794
Epoch: [40]  [100/312]  eta: 0:03:08  lr: 0.003948  min_lr: 0.003948  loss: 3.9705 (3.8944)  weight_decay: 0.0500 (0.0500)  time: 0.6954  data: 0.2098  max mem: 41794
Epoch: [40]  [110/312]  eta: 0:02:52  lr: 0.003948  min_lr: 0.003948  loss: 4.1612 (3.9131)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.2098  max mem: 41794
Epoch: [40]  [120/312]  eta: 0:02:44  lr: 0.003948  min_lr: 0.003948  loss: 3.8358 (3.9031)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.2165  max mem: 41794
Epoch: [40]  [130/312]  eta: 0:02:36  lr: 0.003948  min_lr: 0.003948  loss: 3.7011 (3.8870)  weight_decay: 0.0500 (0.0500)  time: 0.8940  data: 0.3965  max mem: 41794
Epoch: [40]  [140/312]  eta: 0:02:23  lr: 0.003948  min_lr: 0.003948  loss: 4.0729 (3.8922)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.1807  max mem: 41794
Epoch: [40]  [150/312]  eta: 0:02:14  lr: 0.003947  min_lr: 0.003947  loss: 4.1886 (3.8974)  weight_decay: 0.0500 (0.0500)  time: 0.6652  data: 0.1748  max mem: 41794
Epoch: [40]  [160/312]  eta: 0:02:06  lr: 0.003947  min_lr: 0.003947  loss: 4.0492 (3.8920)  weight_decay: 0.0500 (0.0500)  time: 0.8246  data: 0.3331  max mem: 41794
Epoch: [40]  [170/312]  eta: 0:01:55  lr: 0.003947  min_lr: 0.003947  loss: 3.9810 (3.8923)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.2054  max mem: 41794
Epoch: [40]  [180/312]  eta: 0:01:48  lr: 0.003947  min_lr: 0.003947  loss: 4.0031 (3.8961)  weight_decay: 0.0500 (0.0500)  time: 0.7648  data: 0.2448  max mem: 41794
Epoch: [40]  [190/312]  eta: 0:01:38  lr: 0.003947  min_lr: 0.003947  loss: 4.0445 (3.9024)  weight_decay: 0.0500 (0.0500)  time: 0.7299  data: 0.2029  max mem: 41794
Epoch: [40]  [200/312]  eta: 0:01:30  lr: 0.003947  min_lr: 0.003947  loss: 4.1128 (3.9125)  weight_decay: 0.0500 (0.0500)  time: 0.6629  data: 0.1426  max mem: 41794
Epoch: [40]  [210/312]  eta: 0:01:22  lr: 0.003946  min_lr: 0.003946  loss: 4.0826 (3.9122)  weight_decay: 0.0500 (0.0500)  time: 0.8456  data: 0.2878  max mem: 41794
Epoch: [40]  [220/312]  eta: 0:01:13  lr: 0.003946  min_lr: 0.003946  loss: 4.0714 (3.9141)  weight_decay: 0.0500 (0.0500)  time: 0.6844  data: 0.1504  max mem: 41794
Epoch: [40]  [230/312]  eta: 0:01:05  lr: 0.003946  min_lr: 0.003946  loss: 4.0166 (3.9078)  weight_decay: 0.0500 (0.0500)  time: 0.6919  data: 0.1529  max mem: 41794
Epoch: [40]  [240/312]  eta: 0:00:57  lr: 0.003946  min_lr: 0.003946  loss: 4.0166 (3.9171)  weight_decay: 0.0500 (0.0500)  time: 0.8474  data: 0.2544  max mem: 41794
Epoch: [40]  [250/312]  eta: 0:00:49  lr: 0.003946  min_lr: 0.003946  loss: 4.1467 (3.9175)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1280  max mem: 41794
Epoch: [40]  [260/312]  eta: 0:00:41  lr: 0.003946  min_lr: 0.003946  loss: 4.0763 (3.9178)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.1695  max mem: 41794
Epoch: [40]  [270/312]  eta: 0:00:32  lr: 0.003945  min_lr: 0.003945  loss: 3.5217 (3.9039)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.1436  max mem: 41794
Epoch: [40]  [280/312]  eta: 0:00:25  lr: 0.003945  min_lr: 0.003945  loss: 3.7142 (3.9059)  weight_decay: 0.0500 (0.0500)  time: 0.6347  data: 0.1381  max mem: 41794
Epoch: [40]  [290/312]  eta: 0:00:17  lr: 0.003945  min_lr: 0.003945  loss: 4.1185 (3.9171)  weight_decay: 0.0500 (0.0500)  time: 0.7891  data: 0.2925  max mem: 41794
Epoch: [40]  [300/312]  eta: 0:00:09  lr: 0.003945  min_lr: 0.003945  loss: 4.1185 (3.9098)  weight_decay: 0.0500 (0.0500)  time: 0.6405  data: 0.1548  max mem: 41794
Epoch: [40]  [310/312]  eta: 0:00:01  lr: 0.003945  min_lr: 0.003945  loss: 3.6506 (3.9037)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [40]  [311/312]  eta: 0:00:00  lr: 0.003945  min_lr: 0.003945  loss: 3.6506 (3.9033)  weight_decay: 0.0500 (0.0500)  time: 0.4641  data: 0.0001  max mem: 41794
Epoch: [40] Total time: 0:03:59 (0.7673 s / it)
Averaged stats: lr: 0.003945  min_lr: 0.003945  loss: 3.6506 (3.8895)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.5869 (1.5869)  acc1: 73.8281 (73.8281)  acc5: 90.4948 (90.4948)  time: 8.9606  data: 8.7487  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8788 (1.7546)  acc1: 63.4115 (64.4480)  acc5: 85.0260 (86.7200)  time: 1.1436  data: 0.9722  max mem: 41794
Test: Total time: 0:00:10 (1.1593 s / it)
* Acc@1 63.942 Acc@5 86.514 loss 1.763
Accuracy of the model on the 50000 test images: 63.9%
Max accuracy: 64.77%
Epoch: [41]  [  0/312]  eta: 1:19:47  lr: 0.003945  min_lr: 0.003945  loss: 3.0697 (3.0697)  weight_decay: 0.0500 (0.0500)  time: 15.3440  data: 14.4468  max mem: 41794
Epoch: [41]  [ 10/312]  eta: 0:10:58  lr: 0.003945  min_lr: 0.003945  loss: 3.2544 (3.6129)  weight_decay: 0.0500 (0.0500)  time: 2.1817  data: 1.3446  max mem: 41794
Epoch: [41]  [ 20/312]  eta: 0:07:14  lr: 0.003944  min_lr: 0.003944  loss: 4.0075 (3.8149)  weight_decay: 0.0500 (0.0500)  time: 0.7960  data: 0.0740  max mem: 41794
Epoch: [41]  [ 30/312]  eta: 0:05:30  lr: 0.003944  min_lr: 0.003944  loss: 4.0941 (3.8295)  weight_decay: 0.0500 (0.0500)  time: 0.6178  data: 0.0571  max mem: 41794
Epoch: [41]  [ 40/312]  eta: 0:04:49  lr: 0.003944  min_lr: 0.003944  loss: 4.0866 (3.8929)  weight_decay: 0.0500 (0.0500)  time: 0.6209  data: 0.0858  max mem: 41794
Epoch: [41]  [ 50/312]  eta: 0:04:25  lr: 0.003944  min_lr: 0.003944  loss: 3.8941 (3.8488)  weight_decay: 0.0500 (0.0500)  time: 0.7662  data: 0.2193  max mem: 41794
Epoch: [41]  [ 60/312]  eta: 0:03:54  lr: 0.003944  min_lr: 0.003944  loss: 3.9094 (3.8908)  weight_decay: 0.0500 (0.0500)  time: 0.6485  data: 0.1342  max mem: 41794
Epoch: [41]  [ 70/312]  eta: 0:03:45  lr: 0.003944  min_lr: 0.003944  loss: 3.6846 (3.8394)  weight_decay: 0.0500 (0.0500)  time: 0.7237  data: 0.1633  max mem: 41794
Epoch: [41]  [ 80/312]  eta: 0:03:33  lr: 0.003943  min_lr: 0.003943  loss: 3.8653 (3.8708)  weight_decay: 0.0500 (0.0500)  time: 0.8873  data: 0.3138  max mem: 41794
Epoch: [41]  [ 90/312]  eta: 0:03:13  lr: 0.003943  min_lr: 0.003943  loss: 4.0708 (3.8728)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.1511  max mem: 41794
Epoch: [41]  [100/312]  eta: 0:03:03  lr: 0.003943  min_lr: 0.003943  loss: 4.0974 (3.8728)  weight_decay: 0.0500 (0.0500)  time: 0.6579  data: 0.1530  max mem: 41794
Epoch: [41]  [110/312]  eta: 0:02:48  lr: 0.003943  min_lr: 0.003943  loss: 4.1337 (3.8890)  weight_decay: 0.0500 (0.0500)  time: 0.6499  data: 0.1530  max mem: 41794
Epoch: [41]  [120/312]  eta: 0:02:42  lr: 0.003943  min_lr: 0.003943  loss: 4.2178 (3.8996)  weight_decay: 0.0500 (0.0500)  time: 0.7297  data: 0.1819  max mem: 41794
Epoch: [41]  [130/312]  eta: 0:02:34  lr: 0.003943  min_lr: 0.003943  loss: 4.0832 (3.9155)  weight_decay: 0.0500 (0.0500)  time: 0.9248  data: 0.3112  max mem: 41794
Epoch: [41]  [140/312]  eta: 0:02:21  lr: 0.003942  min_lr: 0.003942  loss: 3.8363 (3.8943)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.1299  max mem: 41794
Epoch: [41]  [150/312]  eta: 0:02:13  lr: 0.003942  min_lr: 0.003942  loss: 3.7677 (3.8994)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1656  max mem: 41794
Epoch: [41]  [160/312]  eta: 0:02:05  lr: 0.003942  min_lr: 0.003942  loss: 4.0829 (3.8945)  weight_decay: 0.0500 (0.0500)  time: 0.8378  data: 0.3138  max mem: 41794
Epoch: [41]  [170/312]  eta: 0:01:54  lr: 0.003942  min_lr: 0.003942  loss: 3.9190 (3.9030)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1488  max mem: 41794
Epoch: [41]  [180/312]  eta: 0:01:47  lr: 0.003942  min_lr: 0.003942  loss: 3.9883 (3.9030)  weight_decay: 0.0500 (0.0500)  time: 0.7152  data: 0.1348  max mem: 41794
Epoch: [41]  [190/312]  eta: 0:01:36  lr: 0.003942  min_lr: 0.003942  loss: 3.9604 (3.8881)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1349  max mem: 41794
Epoch: [41]  [200/312]  eta: 0:01:29  lr: 0.003941  min_lr: 0.003941  loss: 3.7449 (3.8797)  weight_decay: 0.0500 (0.0500)  time: 0.6711  data: 0.1427  max mem: 41794
Epoch: [41]  [210/312]  eta: 0:01:21  lr: 0.003941  min_lr: 0.003941  loss: 3.8052 (3.8670)  weight_decay: 0.0500 (0.0500)  time: 0.8364  data: 0.2970  max mem: 41794
Epoch: [41]  [220/312]  eta: 0:01:12  lr: 0.003941  min_lr: 0.003941  loss: 4.1594 (3.8778)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.1549  max mem: 41794
Epoch: [41]  [230/312]  eta: 0:01:04  lr: 0.003941  min_lr: 0.003941  loss: 4.2592 (3.8968)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.1457  max mem: 41794
Epoch: [41]  [240/312]  eta: 0:00:56  lr: 0.003941  min_lr: 0.003941  loss: 4.1308 (3.8969)  weight_decay: 0.0500 (0.0500)  time: 0.8361  data: 0.3135  max mem: 41794
Epoch: [41]  [250/312]  eta: 0:00:48  lr: 0.003940  min_lr: 0.003940  loss: 4.0295 (3.8989)  weight_decay: 0.0500 (0.0500)  time: 0.6712  data: 0.1684  max mem: 41794
Epoch: [41]  [260/312]  eta: 0:00:40  lr: 0.003940  min_lr: 0.003940  loss: 4.1403 (3.9043)  weight_decay: 0.0500 (0.0500)  time: 0.6784  data: 0.1750  max mem: 41794
Epoch: [41]  [270/312]  eta: 0:00:32  lr: 0.003940  min_lr: 0.003940  loss: 4.0955 (3.9007)  weight_decay: 0.0500 (0.0500)  time: 0.6740  data: 0.1750  max mem: 41794
Epoch: [41]  [280/312]  eta: 0:00:24  lr: 0.003940  min_lr: 0.003940  loss: 3.5535 (3.8906)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.1814  max mem: 41794
Epoch: [41]  [290/312]  eta: 0:00:17  lr: 0.003940  min_lr: 0.003940  loss: 3.8237 (3.8851)  weight_decay: 0.0500 (0.0500)  time: 0.8256  data: 0.3260  max mem: 41794
Epoch: [41]  [300/312]  eta: 0:00:09  lr: 0.003940  min_lr: 0.003940  loss: 4.1262 (3.8907)  weight_decay: 0.0500 (0.0500)  time: 0.6285  data: 0.1450  max mem: 41794
Epoch: [41]  [310/312]  eta: 0:00:01  lr: 0.003939  min_lr: 0.003939  loss: 4.1464 (3.8959)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [41]  [311/312]  eta: 0:00:00  lr: 0.003939  min_lr: 0.003939  loss: 4.1464 (3.8978)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [41] Total time: 0:03:57 (0.7609 s / it)
Averaged stats: lr: 0.003939  min_lr: 0.003939  loss: 4.1464 (3.8596)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.6987 (1.6987)  acc1: 71.8750 (71.8750)  acc5: 90.2344 (90.2344)  time: 7.9721  data: 7.7599  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 2.0857 (1.9320)  acc1: 63.0208 (64.1120)  acc5: 85.2865 (86.5120)  time: 1.0620  data: 0.8800  max mem: 41794
Test: Total time: 0:00:09 (1.0720 s / it)
* Acc@1 63.936 Acc@5 86.218 loss 1.947
Accuracy of the model on the 50000 test images: 63.9%
Max accuracy: 64.77%
Epoch: [42]  [  0/312]  eta: 1:20:31  lr: 0.003939  min_lr: 0.003939  loss: 4.6714 (4.6714)  weight_decay: 0.0500 (0.0500)  time: 15.4867  data: 11.6037  max mem: 41794
Epoch: [42]  [ 10/312]  eta: 0:10:15  lr: 0.003939  min_lr: 0.003939  loss: 3.8623 (3.7479)  weight_decay: 0.0500 (0.0500)  time: 2.0380  data: 1.2435  max mem: 41794
Epoch: [42]  [ 20/312]  eta: 0:07:07  lr: 0.003939  min_lr: 0.003939  loss: 3.8623 (3.8202)  weight_decay: 0.0500 (0.0500)  time: 0.7614  data: 0.1498  max mem: 41794
Epoch: [42]  [ 30/312]  eta: 0:05:25  lr: 0.003939  min_lr: 0.003939  loss: 3.9520 (3.7991)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.0465  max mem: 41794
Epoch: [42]  [ 40/312]  eta: 0:04:45  lr: 0.003939  min_lr: 0.003939  loss: 4.0103 (3.8286)  weight_decay: 0.0500 (0.0500)  time: 0.6165  data: 0.0361  max mem: 41794
Epoch: [42]  [ 50/312]  eta: 0:04:28  lr: 0.003939  min_lr: 0.003939  loss: 4.2362 (3.9135)  weight_decay: 0.0500 (0.0500)  time: 0.8202  data: 0.0837  max mem: 41794
Epoch: [42]  [ 60/312]  eta: 0:03:56  lr: 0.003938  min_lr: 0.003938  loss: 3.9911 (3.8656)  weight_decay: 0.0500 (0.0500)  time: 0.7139  data: 0.0610  max mem: 41794
Epoch: [42]  [ 70/312]  eta: 0:03:49  lr: 0.003938  min_lr: 0.003938  loss: 3.6374 (3.8504)  weight_decay: 0.0500 (0.0500)  time: 0.7544  data: 0.0813  max mem: 41794
Epoch: [42]  [ 80/312]  eta: 0:03:35  lr: 0.003938  min_lr: 0.003938  loss: 4.0217 (3.8835)  weight_decay: 0.0500 (0.0500)  time: 0.8925  data: 0.0730  max mem: 41794
Epoch: [42]  [ 90/312]  eta: 0:03:16  lr: 0.003938  min_lr: 0.003938  loss: 3.8144 (3.8508)  weight_decay: 0.0500 (0.0500)  time: 0.6613  data: 0.0330  max mem: 41794
Epoch: [42]  [100/312]  eta: 0:03:07  lr: 0.003938  min_lr: 0.003938  loss: 3.7611 (3.8470)  weight_decay: 0.0500 (0.0500)  time: 0.7036  data: 0.0661  max mem: 41794
Epoch: [42]  [110/312]  eta: 0:02:51  lr: 0.003937  min_lr: 0.003937  loss: 4.0736 (3.8834)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.0477  max mem: 41794
Epoch: [42]  [120/312]  eta: 0:02:43  lr: 0.003937  min_lr: 0.003937  loss: 4.0736 (3.8827)  weight_decay: 0.0500 (0.0500)  time: 0.6961  data: 0.0614  max mem: 41794
Epoch: [42]  [130/312]  eta: 0:02:34  lr: 0.003937  min_lr: 0.003937  loss: 3.8087 (3.8850)  weight_decay: 0.0500 (0.0500)  time: 0.8498  data: 0.0758  max mem: 41794
Epoch: [42]  [140/312]  eta: 0:02:22  lr: 0.003937  min_lr: 0.003937  loss: 3.7312 (3.8567)  weight_decay: 0.0500 (0.0500)  time: 0.6711  data: 0.0473  max mem: 41794
Epoch: [42]  [150/312]  eta: 0:02:14  lr: 0.003937  min_lr: 0.003937  loss: 3.7158 (3.8593)  weight_decay: 0.0500 (0.0500)  time: 0.7075  data: 0.0737  max mem: 41794
Epoch: [42]  [160/312]  eta: 0:02:05  lr: 0.003937  min_lr: 0.003937  loss: 3.9321 (3.8633)  weight_decay: 0.0500 (0.0500)  time: 0.8215  data: 0.0509  max mem: 41794
Epoch: [42]  [170/312]  eta: 0:01:55  lr: 0.003936  min_lr: 0.003936  loss: 3.7664 (3.8463)  weight_decay: 0.0500 (0.0500)  time: 0.6703  data: 0.0518  max mem: 41794
Epoch: [42]  [180/312]  eta: 0:01:47  lr: 0.003936  min_lr: 0.003936  loss: 3.6471 (3.8444)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.1121  max mem: 41794
Epoch: [42]  [190/312]  eta: 0:01:37  lr: 0.003936  min_lr: 0.003936  loss: 4.1687 (3.8563)  weight_decay: 0.0500 (0.0500)  time: 0.6505  data: 0.0622  max mem: 41794
Epoch: [42]  [200/312]  eta: 0:01:29  lr: 0.003936  min_lr: 0.003936  loss: 4.0260 (3.8416)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.0702  max mem: 41794
Epoch: [42]  [210/312]  eta: 0:01:21  lr: 0.003936  min_lr: 0.003936  loss: 3.5099 (3.8282)  weight_decay: 0.0500 (0.0500)  time: 0.8444  data: 0.0918  max mem: 41794
Epoch: [42]  [220/312]  eta: 0:01:12  lr: 0.003935  min_lr: 0.003935  loss: 3.8518 (3.8320)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.0563  max mem: 41794
Epoch: [42]  [230/312]  eta: 0:01:05  lr: 0.003935  min_lr: 0.003935  loss: 3.7946 (3.8295)  weight_decay: 0.0500 (0.0500)  time: 0.7293  data: 0.1052  max mem: 41794
Epoch: [42]  [240/312]  eta: 0:00:57  lr: 0.003935  min_lr: 0.003935  loss: 3.7946 (3.8297)  weight_decay: 0.0500 (0.0500)  time: 0.8445  data: 0.0737  max mem: 41794
Epoch: [42]  [250/312]  eta: 0:00:48  lr: 0.003935  min_lr: 0.003935  loss: 3.9197 (3.8332)  weight_decay: 0.0500 (0.0500)  time: 0.7004  data: 0.0587  max mem: 41794
Epoch: [42]  [260/312]  eta: 0:00:41  lr: 0.003935  min_lr: 0.003935  loss: 3.9157 (3.8343)  weight_decay: 0.0500 (0.0500)  time: 0.7482  data: 0.1260  max mem: 41794
Epoch: [42]  [270/312]  eta: 0:00:32  lr: 0.003935  min_lr: 0.003935  loss: 4.2168 (3.8484)  weight_decay: 0.0500 (0.0500)  time: 0.6988  data: 0.0693  max mem: 41794
Epoch: [42]  [280/312]  eta: 0:00:25  lr: 0.003934  min_lr: 0.003934  loss: 4.2328 (3.8552)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.0733  max mem: 41794
Epoch: [42]  [290/312]  eta: 0:00:17  lr: 0.003934  min_lr: 0.003934  loss: 3.8928 (3.8477)  weight_decay: 0.0500 (0.0500)  time: 0.7828  data: 0.1033  max mem: 41794
Epoch: [42]  [300/312]  eta: 0:00:09  lr: 0.003934  min_lr: 0.003934  loss: 3.3059 (3.8308)  weight_decay: 0.0500 (0.0500)  time: 0.6023  data: 0.0371  max mem: 41794
Epoch: [42]  [310/312]  eta: 0:00:01  lr: 0.003934  min_lr: 0.003934  loss: 3.4450 (3.8300)  weight_decay: 0.0500 (0.0500)  time: 0.4671  data: 0.0068  max mem: 41794
Epoch: [42]  [311/312]  eta: 0:00:00  lr: 0.003934  min_lr: 0.003934  loss: 3.4985 (3.8296)  weight_decay: 0.0500 (0.0500)  time: 0.4610  data: 0.0001  max mem: 41794
Epoch: [42] Total time: 0:03:59 (0.7662 s / it)
Averaged stats: lr: 0.003934  min_lr: 0.003934  loss: 3.4985 (3.8767)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2668 (1.2668)  acc1: 76.5625 (76.5625)  acc5: 92.9688 (92.9688)  time: 8.5080  data: 8.3093  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7431 (1.6003)  acc1: 66.2760 (66.8640)  acc5: 86.7188 (87.5360)  time: 1.1236  data: 0.9536  max mem: 41794
Test: Total time: 0:00:10 (1.1387 s / it)
* Acc@1 66.504 Acc@5 87.764 loss 1.612
Accuracy of the model on the 50000 test images: 66.5%
Max accuracy: 66.50%
Epoch: [43]  [  0/312]  eta: 1:20:24  lr: 0.003934  min_lr: 0.003934  loss: 4.0525 (4.0525)  weight_decay: 0.0500 (0.0500)  time: 15.4646  data: 13.6813  max mem: 41794
Epoch: [43]  [ 10/312]  eta: 0:10:41  lr: 0.003934  min_lr: 0.003934  loss: 3.9434 (3.7304)  weight_decay: 0.0500 (0.0500)  time: 2.1228  data: 1.3372  max mem: 41794
Epoch: [43]  [ 20/312]  eta: 0:07:11  lr: 0.003933  min_lr: 0.003933  loss: 4.0767 (3.9816)  weight_decay: 0.0500 (0.0500)  time: 0.7796  data: 0.1036  max mem: 41794
Epoch: [43]  [ 30/312]  eta: 0:05:29  lr: 0.003933  min_lr: 0.003933  loss: 4.1220 (4.0021)  weight_decay: 0.0500 (0.0500)  time: 0.6427  data: 0.0530  max mem: 41794
Epoch: [43]  [ 40/312]  eta: 0:04:42  lr: 0.003933  min_lr: 0.003933  loss: 3.8547 (3.9636)  weight_decay: 0.0500 (0.0500)  time: 0.5793  data: 0.0282  max mem: 41794
Epoch: [43]  [ 50/312]  eta: 0:04:20  lr: 0.003933  min_lr: 0.003933  loss: 4.0720 (3.9709)  weight_decay: 0.0500 (0.0500)  time: 0.7218  data: 0.0879  max mem: 41794
Epoch: [43]  [ 60/312]  eta: 0:03:51  lr: 0.003933  min_lr: 0.003933  loss: 4.0720 (3.9356)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0862  max mem: 41794
Epoch: [43]  [ 70/312]  eta: 0:03:41  lr: 0.003932  min_lr: 0.003932  loss: 3.5997 (3.8919)  weight_decay: 0.0500 (0.0500)  time: 0.7204  data: 0.1393  max mem: 41794
Epoch: [43]  [ 80/312]  eta: 0:03:30  lr: 0.003932  min_lr: 0.003932  loss: 3.9417 (3.9101)  weight_decay: 0.0500 (0.0500)  time: 0.8698  data: 0.1863  max mem: 41794
Epoch: [43]  [ 90/312]  eta: 0:03:13  lr: 0.003932  min_lr: 0.003932  loss: 4.0787 (3.9081)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.1159  max mem: 41794
Epoch: [43]  [100/312]  eta: 0:03:04  lr: 0.003932  min_lr: 0.003932  loss: 3.8831 (3.8941)  weight_decay: 0.0500 (0.0500)  time: 0.7288  data: 0.1542  max mem: 41794
Epoch: [43]  [110/312]  eta: 0:02:49  lr: 0.003932  min_lr: 0.003932  loss: 3.4707 (3.8380)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.1113  max mem: 41794
Epoch: [43]  [120/312]  eta: 0:02:41  lr: 0.003932  min_lr: 0.003932  loss: 3.5896 (3.8494)  weight_decay: 0.0500 (0.0500)  time: 0.7028  data: 0.1186  max mem: 41794
Epoch: [43]  [130/312]  eta: 0:02:33  lr: 0.003931  min_lr: 0.003931  loss: 3.8816 (3.8347)  weight_decay: 0.0500 (0.0500)  time: 0.8924  data: 0.2374  max mem: 41794
Epoch: [43]  [140/312]  eta: 0:02:21  lr: 0.003931  min_lr: 0.003931  loss: 3.8816 (3.8409)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1221  max mem: 41794
Epoch: [43]  [150/312]  eta: 0:02:13  lr: 0.003931  min_lr: 0.003931  loss: 3.8084 (3.8286)  weight_decay: 0.0500 (0.0500)  time: 0.6939  data: 0.0956  max mem: 41794
Epoch: [43]  [160/312]  eta: 0:02:04  lr: 0.003931  min_lr: 0.003931  loss: 3.7384 (3.8298)  weight_decay: 0.0500 (0.0500)  time: 0.8123  data: 0.2099  max mem: 41794
Epoch: [43]  [170/312]  eta: 0:01:54  lr: 0.003931  min_lr: 0.003931  loss: 3.9495 (3.8278)  weight_decay: 0.0500 (0.0500)  time: 0.6824  data: 0.1175  max mem: 41794
Epoch: [43]  [180/312]  eta: 0:01:46  lr: 0.003930  min_lr: 0.003930  loss: 3.5299 (3.8156)  weight_decay: 0.0500 (0.0500)  time: 0.7168  data: 0.1059  max mem: 41794
Epoch: [43]  [190/312]  eta: 0:01:36  lr: 0.003930  min_lr: 0.003930  loss: 3.4924 (3.8128)  weight_decay: 0.0500 (0.0500)  time: 0.6489  data: 0.1059  max mem: 41794
Epoch: [43]  [200/312]  eta: 0:01:28  lr: 0.003930  min_lr: 0.003930  loss: 4.0419 (3.8171)  weight_decay: 0.0500 (0.0500)  time: 0.6662  data: 0.1206  max mem: 41794
Epoch: [43]  [210/312]  eta: 0:01:21  lr: 0.003930  min_lr: 0.003930  loss: 4.0387 (3.8234)  weight_decay: 0.0500 (0.0500)  time: 0.8545  data: 0.2312  max mem: 41794
Epoch: [43]  [220/312]  eta: 0:01:12  lr: 0.003930  min_lr: 0.003930  loss: 3.9964 (3.8176)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.1127  max mem: 41794
Epoch: [43]  [230/312]  eta: 0:01:04  lr: 0.003930  min_lr: 0.003930  loss: 3.3763 (3.8014)  weight_decay: 0.0500 (0.0500)  time: 0.6443  data: 0.1282  max mem: 41794
Epoch: [43]  [240/312]  eta: 0:00:56  lr: 0.003929  min_lr: 0.003929  loss: 3.2320 (3.7829)  weight_decay: 0.0500 (0.0500)  time: 0.8368  data: 0.2674  max mem: 41794
Epoch: [43]  [250/312]  eta: 0:00:48  lr: 0.003929  min_lr: 0.003929  loss: 3.7401 (3.7878)  weight_decay: 0.0500 (0.0500)  time: 0.7267  data: 0.1414  max mem: 41794
Epoch: [43]  [260/312]  eta: 0:00:40  lr: 0.003929  min_lr: 0.003929  loss: 4.0847 (3.7940)  weight_decay: 0.0500 (0.0500)  time: 0.7278  data: 0.1063  max mem: 41794
Epoch: [43]  [270/312]  eta: 0:00:32  lr: 0.003929  min_lr: 0.003929  loss: 4.1132 (3.7967)  weight_decay: 0.0500 (0.0500)  time: 0.6921  data: 0.1076  max mem: 41794
Epoch: [43]  [280/312]  eta: 0:00:24  lr: 0.003929  min_lr: 0.003929  loss: 4.0586 (3.7962)  weight_decay: 0.0500 (0.0500)  time: 0.6979  data: 0.0913  max mem: 41794
Epoch: [43]  [290/312]  eta: 0:00:17  lr: 0.003928  min_lr: 0.003928  loss: 3.6203 (3.7909)  weight_decay: 0.0500 (0.0500)  time: 0.8133  data: 0.1610  max mem: 41794
Epoch: [43]  [300/312]  eta: 0:00:09  lr: 0.003928  min_lr: 0.003928  loss: 3.8088 (3.8006)  weight_decay: 0.0500 (0.0500)  time: 0.6024  data: 0.0715  max mem: 41794
Epoch: [43]  [310/312]  eta: 0:00:01  lr: 0.003928  min_lr: 0.003928  loss: 4.0205 (3.8046)  weight_decay: 0.0500 (0.0500)  time: 0.4632  data: 0.0001  max mem: 41794
Epoch: [43]  [311/312]  eta: 0:00:00  lr: 0.003928  min_lr: 0.003928  loss: 3.9848 (3.8007)  weight_decay: 0.0500 (0.0500)  time: 0.4629  data: 0.0001  max mem: 41794
Epoch: [43] Total time: 0:03:57 (0.7618 s / it)
Averaged stats: lr: 0.003928  min_lr: 0.003928  loss: 3.9848 (3.8397)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:08  loss: 1.3044 (1.3044)  acc1: 74.8698 (74.8698)  acc5: 91.2760 (91.2760)  time: 7.5926  data: 7.3803  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6257 (1.5706)  acc1: 65.7552 (65.8400)  acc5: 86.7188 (87.2640)  time: 1.0357  data: 0.8597  max mem: 41794
Test: Total time: 0:00:09 (1.0455 s / it)
* Acc@1 65.712 Acc@5 87.410 loss 1.578
Accuracy of the model on the 50000 test images: 65.7%
Max accuracy: 66.50%
Epoch: [44]  [  0/312]  eta: 1:24:55  lr: 0.003928  min_lr: 0.003928  loss: 3.9691 (3.9691)  weight_decay: 0.0500 (0.0500)  time: 16.3331  data: 13.5718  max mem: 41794
Epoch: [44]  [ 10/312]  eta: 0:10:40  lr: 0.003928  min_lr: 0.003928  loss: 3.9029 (3.8040)  weight_decay: 0.0500 (0.0500)  time: 2.1212  data: 1.3455  max mem: 41794
Epoch: [44]  [ 20/312]  eta: 0:07:29  lr: 0.003928  min_lr: 0.003928  loss: 3.8187 (3.8577)  weight_decay: 0.0500 (0.0500)  time: 0.7993  data: 0.1358  max mem: 41794
Epoch: [44]  [ 30/312]  eta: 0:05:42  lr: 0.003927  min_lr: 0.003927  loss: 3.8187 (3.8723)  weight_decay: 0.0500 (0.0500)  time: 0.7143  data: 0.0760  max mem: 41794
Epoch: [44]  [ 40/312]  eta: 0:05:01  lr: 0.003927  min_lr: 0.003927  loss: 3.8282 (3.8844)  weight_decay: 0.0500 (0.0500)  time: 0.6526  data: 0.0322  max mem: 41794
Epoch: [44]  [ 50/312]  eta: 0:04:33  lr: 0.003927  min_lr: 0.003927  loss: 4.3573 (3.9488)  weight_decay: 0.0500 (0.0500)  time: 0.7826  data: 0.0508  max mem: 41794
Epoch: [44]  [ 60/312]  eta: 0:04:01  lr: 0.003927  min_lr: 0.003927  loss: 4.2708 (3.9266)  weight_decay: 0.0500 (0.0500)  time: 0.6562  data: 0.0315  max mem: 41794
Epoch: [44]  [ 70/312]  eta: 0:03:48  lr: 0.003927  min_lr: 0.003927  loss: 3.7881 (3.9171)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.0924  max mem: 41794
Epoch: [44]  [ 80/312]  eta: 0:03:32  lr: 0.003926  min_lr: 0.003926  loss: 3.5765 (3.8759)  weight_decay: 0.0500 (0.0500)  time: 0.7832  data: 0.1754  max mem: 41794
Epoch: [44]  [ 90/312]  eta: 0:03:16  lr: 0.003926  min_lr: 0.003926  loss: 3.6729 (3.8571)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1153  max mem: 41794
Epoch: [44]  [100/312]  eta: 0:03:07  lr: 0.003926  min_lr: 0.003926  loss: 3.8796 (3.8488)  weight_decay: 0.0500 (0.0500)  time: 0.7559  data: 0.1062  max mem: 41794
Epoch: [44]  [110/312]  eta: 0:02:52  lr: 0.003926  min_lr: 0.003926  loss: 3.9162 (3.8467)  weight_decay: 0.0500 (0.0500)  time: 0.7061  data: 0.1174  max mem: 41794
Epoch: [44]  [120/312]  eta: 0:02:44  lr: 0.003926  min_lr: 0.003926  loss: 4.0070 (3.8483)  weight_decay: 0.0500 (0.0500)  time: 0.7330  data: 0.1388  max mem: 41794
Epoch: [44]  [130/312]  eta: 0:02:36  lr: 0.003925  min_lr: 0.003925  loss: 4.0070 (3.8680)  weight_decay: 0.0500 (0.0500)  time: 0.9036  data: 0.1899  max mem: 41794
Epoch: [44]  [140/312]  eta: 0:02:24  lr: 0.003925  min_lr: 0.003925  loss: 3.8956 (3.8621)  weight_decay: 0.0500 (0.0500)  time: 0.7066  data: 0.0837  max mem: 41794
Epoch: [44]  [150/312]  eta: 0:02:15  lr: 0.003925  min_lr: 0.003925  loss: 3.7961 (3.8559)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.0773  max mem: 41794
Epoch: [44]  [160/312]  eta: 0:02:06  lr: 0.003925  min_lr: 0.003925  loss: 3.7700 (3.8467)  weight_decay: 0.0500 (0.0500)  time: 0.7891  data: 0.2075  max mem: 41794
Epoch: [44]  [170/312]  eta: 0:01:55  lr: 0.003925  min_lr: 0.003925  loss: 4.0431 (3.8567)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1310  max mem: 41794
Epoch: [44]  [180/312]  eta: 0:01:48  lr: 0.003924  min_lr: 0.003924  loss: 4.0069 (3.8493)  weight_decay: 0.0500 (0.0500)  time: 0.7138  data: 0.1251  max mem: 41794
Epoch: [44]  [190/312]  eta: 0:01:37  lr: 0.003924  min_lr: 0.003924  loss: 3.8932 (3.8460)  weight_decay: 0.0500 (0.0500)  time: 0.6757  data: 0.1251  max mem: 41794
Epoch: [44]  [200/312]  eta: 0:01:30  lr: 0.003924  min_lr: 0.003924  loss: 3.8297 (3.8415)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.1403  max mem: 41794
Epoch: [44]  [210/312]  eta: 0:01:22  lr: 0.003924  min_lr: 0.003924  loss: 3.8297 (3.8369)  weight_decay: 0.0500 (0.0500)  time: 0.9042  data: 0.3320  max mem: 41794
Epoch: [44]  [220/312]  eta: 0:01:13  lr: 0.003924  min_lr: 0.003924  loss: 3.8480 (3.8288)  weight_decay: 0.0500 (0.0500)  time: 0.7202  data: 0.1923  max mem: 41794
Epoch: [44]  [230/312]  eta: 0:01:05  lr: 0.003923  min_lr: 0.003923  loss: 3.8434 (3.8221)  weight_decay: 0.0500 (0.0500)  time: 0.6953  data: 0.1497  max mem: 41794
Epoch: [44]  [240/312]  eta: 0:00:57  lr: 0.003923  min_lr: 0.003923  loss: 3.9037 (3.8219)  weight_decay: 0.0500 (0.0500)  time: 0.7804  data: 0.2257  max mem: 41794
Epoch: [44]  [250/312]  eta: 0:00:49  lr: 0.003923  min_lr: 0.003923  loss: 3.9037 (3.8138)  weight_decay: 0.0500 (0.0500)  time: 0.7041  data: 0.1472  max mem: 41794
Epoch: [44]  [260/312]  eta: 0:00:41  lr: 0.003923  min_lr: 0.003923  loss: 4.0316 (3.8237)  weight_decay: 0.0500 (0.0500)  time: 0.8158  data: 0.2094  max mem: 41794
Epoch: [44]  [270/312]  eta: 0:00:33  lr: 0.003923  min_lr: 0.003923  loss: 4.1634 (3.8395)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.1464  max mem: 41794
Epoch: [44]  [280/312]  eta: 0:00:25  lr: 0.003922  min_lr: 0.003922  loss: 4.0947 (3.8403)  weight_decay: 0.0500 (0.0500)  time: 0.7008  data: 0.1300  max mem: 41794
Epoch: [44]  [290/312]  eta: 0:00:17  lr: 0.003922  min_lr: 0.003922  loss: 3.9996 (3.8441)  weight_decay: 0.0500 (0.0500)  time: 0.7983  data: 0.1757  max mem: 41794
Epoch: [44]  [300/312]  eta: 0:00:09  lr: 0.003922  min_lr: 0.003922  loss: 3.8861 (3.8455)  weight_decay: 0.0500 (0.0500)  time: 0.5839  data: 0.0537  max mem: 41794
Epoch: [44]  [310/312]  eta: 0:00:01  lr: 0.003922  min_lr: 0.003922  loss: 3.8422 (3.8448)  weight_decay: 0.0500 (0.0500)  time: 0.4621  data: 0.0001  max mem: 41794
Epoch: [44]  [311/312]  eta: 0:00:00  lr: 0.003922  min_lr: 0.003922  loss: 3.8822 (3.8458)  weight_decay: 0.0500 (0.0500)  time: 0.4613  data: 0.0001  max mem: 41794
Epoch: [44] Total time: 0:04:00 (0.7712 s / it)
Averaged stats: lr: 0.003922  min_lr: 0.003922  loss: 3.8822 (3.8208)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.6348 (1.6348)  acc1: 73.1771 (73.1771)  acc5: 92.4479 (92.4479)  time: 8.1671  data: 7.9544  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9985 (1.8507)  acc1: 62.5000 (64.7520)  acc5: 85.0260 (87.0240)  time: 1.0554  data: 0.8839  max mem: 41794
Test: Total time: 0:00:09 (1.0724 s / it)
* Acc@1 64.564 Acc@5 86.804 loss 1.867
Accuracy of the model on the 50000 test images: 64.6%
Max accuracy: 66.50%
Epoch: [45]  [  0/312]  eta: 1:26:18  lr: 0.003922  min_lr: 0.003922  loss: 2.8296 (2.8296)  weight_decay: 0.0500 (0.0500)  time: 16.5963  data: 14.1537  max mem: 41794
Epoch: [45]  [ 10/312]  eta: 0:11:15  lr: 0.003922  min_lr: 0.003922  loss: 3.8418 (3.7919)  weight_decay: 0.0500 (0.0500)  time: 2.2368  data: 1.4087  max mem: 41794
Epoch: [45]  [ 20/312]  eta: 0:07:35  lr: 0.003921  min_lr: 0.003921  loss: 3.7293 (3.7279)  weight_decay: 0.0500 (0.0500)  time: 0.8079  data: 0.1390  max mem: 41794
Epoch: [45]  [ 30/312]  eta: 0:05:45  lr: 0.003921  min_lr: 0.003921  loss: 3.9155 (3.8603)  weight_decay: 0.0500 (0.0500)  time: 0.6690  data: 0.0784  max mem: 41794
Epoch: [45]  [ 40/312]  eta: 0:04:49  lr: 0.003921  min_lr: 0.003921  loss: 4.1240 (3.8926)  weight_decay: 0.0500 (0.0500)  time: 0.5450  data: 0.0444  max mem: 41794
Epoch: [45]  [ 50/312]  eta: 0:04:25  lr: 0.003921  min_lr: 0.003921  loss: 3.9810 (3.9363)  weight_decay: 0.0500 (0.0500)  time: 0.6891  data: 0.1945  max mem: 41794
Epoch: [45]  [ 60/312]  eta: 0:03:54  lr: 0.003921  min_lr: 0.003921  loss: 3.9233 (3.9128)  weight_decay: 0.0500 (0.0500)  time: 0.6506  data: 0.1570  max mem: 41794
Epoch: [45]  [ 70/312]  eta: 0:03:43  lr: 0.003920  min_lr: 0.003920  loss: 3.5808 (3.8780)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.2055  max mem: 41794
Epoch: [45]  [ 80/312]  eta: 0:03:30  lr: 0.003920  min_lr: 0.003920  loss: 3.5142 (3.8534)  weight_decay: 0.0500 (0.0500)  time: 0.8444  data: 0.3555  max mem: 41794
Epoch: [45]  [ 90/312]  eta: 0:03:14  lr: 0.003920  min_lr: 0.003920  loss: 3.9617 (3.8596)  weight_decay: 0.0500 (0.0500)  time: 0.6972  data: 0.2057  max mem: 41794
Epoch: [45]  [100/312]  eta: 0:03:05  lr: 0.003920  min_lr: 0.003920  loss: 3.9775 (3.8623)  weight_decay: 0.0500 (0.0500)  time: 0.7436  data: 0.2391  max mem: 41794
Epoch: [45]  [110/312]  eta: 0:02:49  lr: 0.003920  min_lr: 0.003920  loss: 4.0560 (3.8799)  weight_decay: 0.0500 (0.0500)  time: 0.6861  data: 0.1862  max mem: 41794
Epoch: [45]  [120/312]  eta: 0:02:43  lr: 0.003919  min_lr: 0.003919  loss: 4.0539 (3.8761)  weight_decay: 0.0500 (0.0500)  time: 0.7232  data: 0.1766  max mem: 41794
Epoch: [45]  [130/312]  eta: 0:02:34  lr: 0.003919  min_lr: 0.003919  loss: 3.9379 (3.8647)  weight_decay: 0.0500 (0.0500)  time: 0.8993  data: 0.2914  max mem: 41794
Epoch: [45]  [140/312]  eta: 0:02:21  lr: 0.003919  min_lr: 0.003919  loss: 3.7187 (3.8512)  weight_decay: 0.0500 (0.0500)  time: 0.6662  data: 0.1204  max mem: 41794
Epoch: [45]  [150/312]  eta: 0:02:14  lr: 0.003919  min_lr: 0.003919  loss: 4.0358 (3.8710)  weight_decay: 0.0500 (0.0500)  time: 0.7191  data: 0.1895  max mem: 41794
Epoch: [45]  [160/312]  eta: 0:02:05  lr: 0.003919  min_lr: 0.003919  loss: 4.1124 (3.8831)  weight_decay: 0.0500 (0.0500)  time: 0.8140  data: 0.2558  max mem: 41794
Epoch: [45]  [170/312]  eta: 0:01:55  lr: 0.003918  min_lr: 0.003918  loss: 3.7090 (3.8511)  weight_decay: 0.0500 (0.0500)  time: 0.6677  data: 0.1560  max mem: 41794
Epoch: [45]  [180/312]  eta: 0:01:47  lr: 0.003918  min_lr: 0.003918  loss: 3.3443 (3.8401)  weight_decay: 0.0500 (0.0500)  time: 0.7565  data: 0.2697  max mem: 41794
Epoch: [45]  [190/312]  eta: 0:01:37  lr: 0.003918  min_lr: 0.003918  loss: 3.8822 (3.8318)  weight_decay: 0.0500 (0.0500)  time: 0.6760  data: 0.1833  max mem: 41794
Epoch: [45]  [200/312]  eta: 0:01:29  lr: 0.003918  min_lr: 0.003918  loss: 3.9844 (3.8373)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1865  max mem: 41794
Epoch: [45]  [210/312]  eta: 0:01:22  lr: 0.003918  min_lr: 0.003918  loss: 3.9998 (3.8416)  weight_decay: 0.0500 (0.0500)  time: 0.8595  data: 0.3259  max mem: 41794
Epoch: [45]  [220/312]  eta: 0:01:13  lr: 0.003917  min_lr: 0.003917  loss: 3.8324 (3.8392)  weight_decay: 0.0500 (0.0500)  time: 0.7252  data: 0.1826  max mem: 41794
Epoch: [45]  [230/312]  eta: 0:01:05  lr: 0.003917  min_lr: 0.003917  loss: 3.7198 (3.8364)  weight_decay: 0.0500 (0.0500)  time: 0.7317  data: 0.2024  max mem: 41794
Epoch: [45]  [240/312]  eta: 0:00:56  lr: 0.003917  min_lr: 0.003917  loss: 3.5357 (3.8306)  weight_decay: 0.0500 (0.0500)  time: 0.7411  data: 0.1703  max mem: 41794
Epoch: [45]  [250/312]  eta: 0:00:49  lr: 0.003917  min_lr: 0.003917  loss: 3.7472 (3.8387)  weight_decay: 0.0500 (0.0500)  time: 0.7006  data: 0.1654  max mem: 41794
Epoch: [45]  [260/312]  eta: 0:00:41  lr: 0.003917  min_lr: 0.003917  loss: 3.7472 (3.8403)  weight_decay: 0.0500 (0.0500)  time: 0.8065  data: 0.2926  max mem: 41794
Epoch: [45]  [270/312]  eta: 0:00:32  lr: 0.003916  min_lr: 0.003916  loss: 3.7127 (3.8283)  weight_decay: 0.0500 (0.0500)  time: 0.6595  data: 0.1396  max mem: 41794
Epoch: [45]  [280/312]  eta: 0:00:25  lr: 0.003916  min_lr: 0.003916  loss: 3.8251 (3.8343)  weight_decay: 0.0500 (0.0500)  time: 0.7114  data: 0.1716  max mem: 41794
Epoch: [45]  [290/312]  eta: 0:00:17  lr: 0.003916  min_lr: 0.003916  loss: 3.9843 (3.8355)  weight_decay: 0.0500 (0.0500)  time: 0.8038  data: 0.2088  max mem: 41794
Epoch: [45]  [300/312]  eta: 0:00:09  lr: 0.003916  min_lr: 0.003916  loss: 3.8846 (3.8338)  weight_decay: 0.0500 (0.0500)  time: 0.6134  data: 0.0768  max mem: 41794
Epoch: [45]  [310/312]  eta: 0:00:01  lr: 0.003916  min_lr: 0.003916  loss: 3.8437 (3.8362)  weight_decay: 0.0500 (0.0500)  time: 0.5009  data: 0.0380  max mem: 41794
Epoch: [45]  [311/312]  eta: 0:00:00  lr: 0.003916  min_lr: 0.003916  loss: 3.8437 (3.8378)  weight_decay: 0.0500 (0.0500)  time: 0.4620  data: 0.0001  max mem: 41794
Epoch: [45] Total time: 0:04:00 (0.7699 s / it)
Averaged stats: lr: 0.003916  min_lr: 0.003916  loss: 3.8437 (3.8366)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.5208 (1.5208)  acc1: 69.6615 (69.6615)  acc5: 90.2344 (90.2344)  time: 8.7182  data: 8.5062  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9239 (1.7696)  acc1: 64.5833 (64.7840)  acc5: 85.6771 (87.1040)  time: 1.1167  data: 0.9452  max mem: 41794
Test: Total time: 0:00:10 (1.1337 s / it)
* Acc@1 64.248 Acc@5 86.738 loss 1.786
Accuracy of the model on the 50000 test images: 64.2%
Max accuracy: 66.50%
Epoch: [46]  [  0/312]  eta: 1:23:14  lr: 0.003916  min_lr: 0.003916  loss: 3.8053 (3.8053)  weight_decay: 0.0500 (0.0500)  time: 16.0079  data: 12.8274  max mem: 41794
Epoch: [46]  [ 10/312]  eta: 0:11:34  lr: 0.003915  min_lr: 0.003915  loss: 3.9964 (3.7354)  weight_decay: 0.0500 (0.0500)  time: 2.3012  data: 1.3420  max mem: 41794
Epoch: [46]  [ 20/312]  eta: 0:07:18  lr: 0.003915  min_lr: 0.003915  loss: 3.9964 (3.8210)  weight_decay: 0.0500 (0.0500)  time: 0.7777  data: 0.1003  max mem: 41794
Epoch: [46]  [ 30/312]  eta: 0:05:34  lr: 0.003915  min_lr: 0.003915  loss: 4.1465 (3.8637)  weight_decay: 0.0500 (0.0500)  time: 0.5730  data: 0.0039  max mem: 41794
Epoch: [46]  [ 40/312]  eta: 0:04:37  lr: 0.003915  min_lr: 0.003915  loss: 4.0503 (3.8515)  weight_decay: 0.0500 (0.0500)  time: 0.5158  data: 0.0103  max mem: 41794
Epoch: [46]  [ 50/312]  eta: 0:04:17  lr: 0.003914  min_lr: 0.003914  loss: 3.9340 (3.8984)  weight_decay: 0.0500 (0.0500)  time: 0.6676  data: 0.0909  max mem: 41794
Epoch: [46]  [ 60/312]  eta: 0:03:47  lr: 0.003914  min_lr: 0.003914  loss: 4.0105 (3.9038)  weight_decay: 0.0500 (0.0500)  time: 0.6591  data: 0.0813  max mem: 41794
Epoch: [46]  [ 70/312]  eta: 0:03:41  lr: 0.003914  min_lr: 0.003914  loss: 4.0821 (3.8898)  weight_decay: 0.0500 (0.0500)  time: 0.7399  data: 0.0901  max mem: 41794
Epoch: [46]  [ 80/312]  eta: 0:03:21  lr: 0.003914  min_lr: 0.003914  loss: 3.9867 (3.9010)  weight_decay: 0.0500 (0.0500)  time: 0.7732  data: 0.0901  max mem: 41794
Epoch: [46]  [ 90/312]  eta: 0:03:10  lr: 0.003914  min_lr: 0.003914  loss: 3.9867 (3.8834)  weight_decay: 0.0500 (0.0500)  time: 0.6482  data: 0.0596  max mem: 41794
Epoch: [46]  [100/312]  eta: 0:03:01  lr: 0.003913  min_lr: 0.003913  loss: 4.0194 (3.8979)  weight_decay: 0.0500 (0.0500)  time: 0.7941  data: 0.1578  max mem: 41794
Epoch: [46]  [110/312]  eta: 0:02:46  lr: 0.003913  min_lr: 0.003913  loss: 4.0529 (3.8994)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.1011  max mem: 41794
Epoch: [46]  [120/312]  eta: 0:02:39  lr: 0.003913  min_lr: 0.003913  loss: 4.1731 (3.9295)  weight_decay: 0.0500 (0.0500)  time: 0.7053  data: 0.0795  max mem: 41794
Epoch: [46]  [130/312]  eta: 0:02:31  lr: 0.003913  min_lr: 0.003913  loss: 4.1349 (3.9202)  weight_decay: 0.0500 (0.0500)  time: 0.8932  data: 0.1557  max mem: 41794
Epoch: [46]  [140/312]  eta: 0:02:19  lr: 0.003913  min_lr: 0.003913  loss: 4.0506 (3.9321)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.0788  max mem: 41794
Epoch: [46]  [150/312]  eta: 0:02:12  lr: 0.003912  min_lr: 0.003912  loss: 3.8944 (3.9227)  weight_decay: 0.0500 (0.0500)  time: 0.7196  data: 0.0987  max mem: 41794
Epoch: [46]  [160/312]  eta: 0:02:01  lr: 0.003912  min_lr: 0.003912  loss: 3.8463 (3.9172)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.1001  max mem: 41794
Epoch: [46]  [170/312]  eta: 0:01:53  lr: 0.003912  min_lr: 0.003912  loss: 3.9099 (3.9228)  weight_decay: 0.0500 (0.0500)  time: 0.6762  data: 0.0772  max mem: 41794
Epoch: [46]  [180/312]  eta: 0:01:46  lr: 0.003912  min_lr: 0.003912  loss: 4.1068 (3.9235)  weight_decay: 0.0500 (0.0500)  time: 0.8620  data: 0.1420  max mem: 41794
Epoch: [46]  [190/312]  eta: 0:01:36  lr: 0.003912  min_lr: 0.003912  loss: 4.1068 (3.9199)  weight_decay: 0.0500 (0.0500)  time: 0.6786  data: 0.0667  max mem: 41794
Epoch: [46]  [200/312]  eta: 0:01:28  lr: 0.003911  min_lr: 0.003911  loss: 4.0106 (3.9216)  weight_decay: 0.0500 (0.0500)  time: 0.6532  data: 0.0765  max mem: 41794
Epoch: [46]  [210/312]  eta: 0:01:21  lr: 0.003911  min_lr: 0.003911  loss: 3.9026 (3.9138)  weight_decay: 0.0500 (0.0500)  time: 0.8535  data: 0.1639  max mem: 41794
Epoch: [46]  [220/312]  eta: 0:01:11  lr: 0.003911  min_lr: 0.003911  loss: 3.9026 (3.9074)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.0881  max mem: 41794
Epoch: [46]  [230/312]  eta: 0:01:04  lr: 0.003911  min_lr: 0.003911  loss: 3.9365 (3.9067)  weight_decay: 0.0500 (0.0500)  time: 0.6694  data: 0.0853  max mem: 41794
Epoch: [46]  [240/312]  eta: 0:00:55  lr: 0.003910  min_lr: 0.003910  loss: 3.9615 (3.9072)  weight_decay: 0.0500 (0.0500)  time: 0.7018  data: 0.0996  max mem: 41794
Epoch: [46]  [250/312]  eta: 0:00:48  lr: 0.003910  min_lr: 0.003910  loss: 4.1195 (3.9123)  weight_decay: 0.0500 (0.0500)  time: 0.6743  data: 0.0873  max mem: 41794
Epoch: [46]  [260/312]  eta: 0:00:40  lr: 0.003910  min_lr: 0.003910  loss: 4.1195 (3.9243)  weight_decay: 0.0500 (0.0500)  time: 0.8497  data: 0.1574  max mem: 41794
Epoch: [46]  [270/312]  eta: 0:00:32  lr: 0.003910  min_lr: 0.003910  loss: 4.0536 (3.9222)  weight_decay: 0.0500 (0.0500)  time: 0.7056  data: 0.0869  max mem: 41794
Epoch: [46]  [280/312]  eta: 0:00:24  lr: 0.003910  min_lr: 0.003910  loss: 3.9187 (3.9160)  weight_decay: 0.0500 (0.0500)  time: 0.6879  data: 0.1053  max mem: 41794
Epoch: [46]  [290/312]  eta: 0:00:17  lr: 0.003909  min_lr: 0.003909  loss: 3.8845 (3.9103)  weight_decay: 0.0500 (0.0500)  time: 0.8758  data: 0.1789  max mem: 41794
Epoch: [46]  [300/312]  eta: 0:00:09  lr: 0.003909  min_lr: 0.003909  loss: 3.8730 (3.9026)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.0756  max mem: 41794
Epoch: [46]  [310/312]  eta: 0:00:01  lr: 0.003909  min_lr: 0.003909  loss: 3.8822 (3.8999)  weight_decay: 0.0500 (0.0500)  time: 0.4689  data: 0.0001  max mem: 41794
Epoch: [46]  [311/312]  eta: 0:00:00  lr: 0.003909  min_lr: 0.003909  loss: 3.9230 (3.9008)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0001  max mem: 41794
Epoch: [46] Total time: 0:03:58 (0.7634 s / it)
Averaged stats: lr: 0.003909  min_lr: 0.003909  loss: 3.9230 (3.8212)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.5492 (1.5492)  acc1: 71.4844 (71.4844)  acc5: 91.5365 (91.5365)  time: 8.5454  data: 8.3460  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7994 (1.7567)  acc1: 64.3229 (65.2640)  acc5: 87.6302 (87.4560)  time: 1.1038  data: 0.9274  max mem: 41794
Test: Total time: 0:00:10 (1.1455 s / it)
* Acc@1 64.760 Acc@5 87.148 loss 1.755
Accuracy of the model on the 50000 test images: 64.8%
Max accuracy: 66.50%
Epoch: [47]  [  0/312]  eta: 1:17:54  lr: 0.003909  min_lr: 0.003909  loss: 4.0513 (4.0513)  weight_decay: 0.0500 (0.0500)  time: 14.9829  data: 12.1929  max mem: 41794
Epoch: [47]  [ 10/312]  eta: 0:11:22  lr: 0.003909  min_lr: 0.003909  loss: 3.8748 (3.7068)  weight_decay: 0.0500 (0.0500)  time: 2.2599  data: 1.3472  max mem: 41794
Epoch: [47]  [ 20/312]  eta: 0:07:37  lr: 0.003909  min_lr: 0.003909  loss: 3.6980 (3.6186)  weight_decay: 0.0500 (0.0500)  time: 0.8969  data: 0.2093  max mem: 41794
Epoch: [47]  [ 30/312]  eta: 0:05:48  lr: 0.003908  min_lr: 0.003908  loss: 3.3702 (3.5267)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.0876  max mem: 41794
Epoch: [47]  [ 40/312]  eta: 0:04:47  lr: 0.003908  min_lr: 0.003908  loss: 3.5464 (3.5534)  weight_decay: 0.0500 (0.0500)  time: 0.5242  data: 0.0099  max mem: 41794
Epoch: [47]  [ 50/312]  eta: 0:04:16  lr: 0.003908  min_lr: 0.003908  loss: 3.8476 (3.5986)  weight_decay: 0.0500 (0.0500)  time: 0.5792  data: 0.0091  max mem: 41794
Epoch: [47]  [ 60/312]  eta: 0:03:49  lr: 0.003908  min_lr: 0.003908  loss: 3.8409 (3.6071)  weight_decay: 0.0500 (0.0500)  time: 0.6083  data: 0.0455  max mem: 41794
Epoch: [47]  [ 70/312]  eta: 0:03:38  lr: 0.003907  min_lr: 0.003907  loss: 3.8409 (3.6276)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.1324  max mem: 41794
Epoch: [47]  [ 80/312]  eta: 0:03:24  lr: 0.003907  min_lr: 0.003907  loss: 3.9965 (3.6530)  weight_decay: 0.0500 (0.0500)  time: 0.7908  data: 0.1254  max mem: 41794
Epoch: [47]  [ 90/312]  eta: 0:03:10  lr: 0.003907  min_lr: 0.003907  loss: 3.8948 (3.6744)  weight_decay: 0.0500 (0.0500)  time: 0.7037  data: 0.1144  max mem: 41794
Epoch: [47]  [100/312]  eta: 0:03:01  lr: 0.003907  min_lr: 0.003907  loss: 3.9517 (3.6752)  weight_decay: 0.0500 (0.0500)  time: 0.7495  data: 0.1738  max mem: 41794
Epoch: [47]  [110/312]  eta: 0:02:46  lr: 0.003907  min_lr: 0.003907  loss: 3.4481 (3.6656)  weight_decay: 0.0500 (0.0500)  time: 0.6632  data: 0.0894  max mem: 41794
Epoch: [47]  [120/312]  eta: 0:02:39  lr: 0.003906  min_lr: 0.003906  loss: 3.6072 (3.6642)  weight_decay: 0.0500 (0.0500)  time: 0.7172  data: 0.0980  max mem: 41794
Epoch: [47]  [130/312]  eta: 0:02:31  lr: 0.003906  min_lr: 0.003906  loss: 3.8454 (3.6987)  weight_decay: 0.0500 (0.0500)  time: 0.8705  data: 0.1491  max mem: 41794
Epoch: [47]  [140/312]  eta: 0:02:19  lr: 0.003906  min_lr: 0.003906  loss: 3.9945 (3.7148)  weight_decay: 0.0500 (0.0500)  time: 0.6616  data: 0.0612  max mem: 41794
Epoch: [47]  [150/312]  eta: 0:02:10  lr: 0.003906  min_lr: 0.003906  loss: 3.9222 (3.7174)  weight_decay: 0.0500 (0.0500)  time: 0.6541  data: 0.0898  max mem: 41794
Epoch: [47]  [160/312]  eta: 0:02:02  lr: 0.003905  min_lr: 0.003905  loss: 3.3574 (3.7001)  weight_decay: 0.0500 (0.0500)  time: 0.8091  data: 0.1123  max mem: 41794
Epoch: [47]  [170/312]  eta: 0:01:53  lr: 0.003905  min_lr: 0.003905  loss: 3.8472 (3.7206)  weight_decay: 0.0500 (0.0500)  time: 0.7577  data: 0.1278  max mem: 41794
Epoch: [47]  [180/312]  eta: 0:01:45  lr: 0.003905  min_lr: 0.003905  loss: 4.0433 (3.7352)  weight_decay: 0.0500 (0.0500)  time: 0.7362  data: 0.1550  max mem: 41794
Epoch: [47]  [190/312]  eta: 0:01:35  lr: 0.003905  min_lr: 0.003905  loss: 3.9757 (3.7420)  weight_decay: 0.0500 (0.0500)  time: 0.6395  data: 0.0598  max mem: 41794
Epoch: [47]  [200/312]  eta: 0:01:28  lr: 0.003905  min_lr: 0.003905  loss: 3.9314 (3.7503)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.0868  max mem: 41794
Epoch: [47]  [210/312]  eta: 0:01:20  lr: 0.003904  min_lr: 0.003904  loss: 3.8773 (3.7551)  weight_decay: 0.0500 (0.0500)  time: 0.8238  data: 0.1160  max mem: 41794
Epoch: [47]  [220/312]  eta: 0:01:11  lr: 0.003904  min_lr: 0.003904  loss: 3.9162 (3.7584)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.0954  max mem: 41794
Epoch: [47]  [230/312]  eta: 0:01:04  lr: 0.003904  min_lr: 0.003904  loss: 3.9920 (3.7566)  weight_decay: 0.0500 (0.0500)  time: 0.7259  data: 0.1790  max mem: 41794
Epoch: [47]  [240/312]  eta: 0:00:56  lr: 0.003904  min_lr: 0.003904  loss: 3.9920 (3.7573)  weight_decay: 0.0500 (0.0500)  time: 0.7982  data: 0.1301  max mem: 41794
Epoch: [47]  [250/312]  eta: 0:00:48  lr: 0.003904  min_lr: 0.003904  loss: 4.1239 (3.7697)  weight_decay: 0.0500 (0.0500)  time: 0.7266  data: 0.1175  max mem: 41794
Epoch: [47]  [260/312]  eta: 0:00:40  lr: 0.003903  min_lr: 0.003903  loss: 4.1096 (3.7706)  weight_decay: 0.0500 (0.0500)  time: 0.8053  data: 0.1933  max mem: 41794
Epoch: [47]  [270/312]  eta: 0:00:32  lr: 0.003903  min_lr: 0.003903  loss: 3.7722 (3.7623)  weight_decay: 0.0500 (0.0500)  time: 0.7113  data: 0.0932  max mem: 41794
Epoch: [47]  [280/312]  eta: 0:00:24  lr: 0.003903  min_lr: 0.003903  loss: 3.7722 (3.7654)  weight_decay: 0.0500 (0.0500)  time: 0.6740  data: 0.0920  max mem: 41794
Epoch: [47]  [290/312]  eta: 0:00:17  lr: 0.003903  min_lr: 0.003903  loss: 4.0075 (3.7715)  weight_decay: 0.0500 (0.0500)  time: 0.7945  data: 0.1054  max mem: 41794
Epoch: [47]  [300/312]  eta: 0:00:09  lr: 0.003902  min_lr: 0.003902  loss: 4.0075 (3.7759)  weight_decay: 0.0500 (0.0500)  time: 0.6451  data: 0.0539  max mem: 41794
Epoch: [47]  [310/312]  eta: 0:00:01  lr: 0.003902  min_lr: 0.003902  loss: 3.6838 (3.7679)  weight_decay: 0.0500 (0.0500)  time: 0.5055  data: 0.0403  max mem: 41794
Epoch: [47]  [311/312]  eta: 0:00:00  lr: 0.003902  min_lr: 0.003902  loss: 3.6838 (3.7704)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [47] Total time: 0:03:58 (0.7630 s / it)
Averaged stats: lr: 0.003902  min_lr: 0.003902  loss: 3.6838 (3.8037)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.5603 (1.5603)  acc1: 72.7865 (72.7865)  acc5: 90.6250 (90.6250)  time: 8.3328  data: 8.1228  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8114 (1.7838)  acc1: 65.3646 (65.9040)  acc5: 87.8906 (87.2640)  time: 1.0739  data: 0.9026  max mem: 41794
Test: Total time: 0:00:09 (1.0860 s / it)
* Acc@1 65.732 Acc@5 87.420 loss 1.771
Accuracy of the model on the 50000 test images: 65.7%
Max accuracy: 66.50%
Epoch: [48]  [  0/312]  eta: 1:19:16  lr: 0.003902  min_lr: 0.003902  loss: 3.5984 (3.5984)  weight_decay: 0.0500 (0.0500)  time: 15.2439  data: 14.7357  max mem: 41794
Epoch: [48]  [ 10/312]  eta: 0:10:56  lr: 0.003902  min_lr: 0.003902  loss: 3.9469 (3.7990)  weight_decay: 0.0500 (0.0500)  time: 2.1746  data: 1.3403  max mem: 41794
Epoch: [48]  [ 20/312]  eta: 0:07:29  lr: 0.003902  min_lr: 0.003902  loss: 3.9540 (3.8195)  weight_decay: 0.0500 (0.0500)  time: 0.8535  data: 0.0646  max mem: 41794
Epoch: [48]  [ 30/312]  eta: 0:05:43  lr: 0.003901  min_lr: 0.003901  loss: 4.1062 (3.8825)  weight_decay: 0.0500 (0.0500)  time: 0.6927  data: 0.0749  max mem: 41794
Epoch: [48]  [ 40/312]  eta: 0:04:54  lr: 0.003901  min_lr: 0.003901  loss: 4.1029 (3.8618)  weight_decay: 0.0500 (0.0500)  time: 0.6019  data: 0.0471  max mem: 41794
Epoch: [48]  [ 50/312]  eta: 0:04:26  lr: 0.003901  min_lr: 0.003901  loss: 3.7893 (3.8231)  weight_decay: 0.0500 (0.0500)  time: 0.7056  data: 0.1504  max mem: 41794
Epoch: [48]  [ 60/312]  eta: 0:03:54  lr: 0.003901  min_lr: 0.003901  loss: 3.8340 (3.8219)  weight_decay: 0.0500 (0.0500)  time: 0.6149  data: 0.1144  max mem: 41794
Epoch: [48]  [ 70/312]  eta: 0:03:41  lr: 0.003901  min_lr: 0.003901  loss: 3.9185 (3.8031)  weight_decay: 0.0500 (0.0500)  time: 0.6580  data: 0.1635  max mem: 41794
Epoch: [48]  [ 80/312]  eta: 0:03:29  lr: 0.003900  min_lr: 0.003900  loss: 3.6896 (3.7827)  weight_decay: 0.0500 (0.0500)  time: 0.8318  data: 0.3303  max mem: 41794
Epoch: [48]  [ 90/312]  eta: 0:03:12  lr: 0.003900  min_lr: 0.003900  loss: 3.5381 (3.7485)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.1673  max mem: 41794
Epoch: [48]  [100/312]  eta: 0:03:03  lr: 0.003900  min_lr: 0.003900  loss: 3.9567 (3.7619)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.1456  max mem: 41794
Epoch: [48]  [110/312]  eta: 0:02:47  lr: 0.003900  min_lr: 0.003900  loss: 3.9298 (3.7566)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.1457  max mem: 41794
Epoch: [48]  [120/312]  eta: 0:02:40  lr: 0.003899  min_lr: 0.003899  loss: 3.9778 (3.7819)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.1642  max mem: 41794
Epoch: [48]  [130/312]  eta: 0:02:32  lr: 0.003899  min_lr: 0.003899  loss: 3.9779 (3.7633)  weight_decay: 0.0500 (0.0500)  time: 0.8699  data: 0.3368  max mem: 41794
Epoch: [48]  [140/312]  eta: 0:02:19  lr: 0.003899  min_lr: 0.003899  loss: 4.0671 (3.7868)  weight_decay: 0.0500 (0.0500)  time: 0.6660  data: 0.1732  max mem: 41794
Epoch: [48]  [150/312]  eta: 0:02:12  lr: 0.003899  min_lr: 0.003899  loss: 4.0671 (3.7858)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.2074  max mem: 41794
Epoch: [48]  [160/312]  eta: 0:02:04  lr: 0.003899  min_lr: 0.003899  loss: 3.6881 (3.7750)  weight_decay: 0.0500 (0.0500)  time: 0.8918  data: 0.3961  max mem: 41794
Epoch: [48]  [170/312]  eta: 0:01:53  lr: 0.003898  min_lr: 0.003898  loss: 3.9042 (3.7880)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1893  max mem: 41794
Epoch: [48]  [180/312]  eta: 0:01:46  lr: 0.003898  min_lr: 0.003898  loss: 4.0507 (3.7992)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.1697  max mem: 41794
Epoch: [48]  [190/312]  eta: 0:01:36  lr: 0.003898  min_lr: 0.003898  loss: 3.9502 (3.8024)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.1697  max mem: 41794
Epoch: [48]  [200/312]  eta: 0:01:28  lr: 0.003898  min_lr: 0.003898  loss: 3.9502 (3.8073)  weight_decay: 0.0500 (0.0500)  time: 0.6886  data: 0.1986  max mem: 41794
Epoch: [48]  [210/312]  eta: 0:01:21  lr: 0.003897  min_lr: 0.003897  loss: 4.0117 (3.8138)  weight_decay: 0.0500 (0.0500)  time: 0.8706  data: 0.3767  max mem: 41794
Epoch: [48]  [220/312]  eta: 0:01:11  lr: 0.003897  min_lr: 0.003897  loss: 3.9788 (3.8078)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1800  max mem: 41794
Epoch: [48]  [230/312]  eta: 0:01:04  lr: 0.003897  min_lr: 0.003897  loss: 3.9210 (3.8066)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.1937  max mem: 41794
Epoch: [48]  [240/312]  eta: 0:00:56  lr: 0.003897  min_lr: 0.003897  loss: 3.8913 (3.8038)  weight_decay: 0.0500 (0.0500)  time: 0.8554  data: 0.3635  max mem: 41794
Epoch: [48]  [250/312]  eta: 0:00:48  lr: 0.003897  min_lr: 0.003897  loss: 3.9079 (3.8089)  weight_decay: 0.0500 (0.0500)  time: 0.6721  data: 0.1839  max mem: 41794
Epoch: [48]  [260/312]  eta: 0:00:40  lr: 0.003896  min_lr: 0.003896  loss: 3.9446 (3.8108)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.1563  max mem: 41794
Epoch: [48]  [270/312]  eta: 0:00:32  lr: 0.003896  min_lr: 0.003896  loss: 4.0528 (3.8141)  weight_decay: 0.0500 (0.0500)  time: 0.6632  data: 0.1440  max mem: 41794
Epoch: [48]  [280/312]  eta: 0:00:24  lr: 0.003896  min_lr: 0.003896  loss: 4.0611 (3.8143)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.2112  max mem: 41794
Epoch: [48]  [290/312]  eta: 0:00:17  lr: 0.003896  min_lr: 0.003896  loss: 3.9936 (3.8144)  weight_decay: 0.0500 (0.0500)  time: 0.8955  data: 0.3426  max mem: 41794
Epoch: [48]  [300/312]  eta: 0:00:09  lr: 0.003895  min_lr: 0.003895  loss: 4.0264 (3.8214)  weight_decay: 0.0500 (0.0500)  time: 0.6623  data: 0.1318  max mem: 41794
Epoch: [48]  [310/312]  eta: 0:00:01  lr: 0.003895  min_lr: 0.003895  loss: 4.1132 (3.8205)  weight_decay: 0.0500 (0.0500)  time: 0.4656  data: 0.0001  max mem: 41794
Epoch: [48]  [311/312]  eta: 0:00:00  lr: 0.003895  min_lr: 0.003895  loss: 4.0284 (3.8198)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [48] Total time: 0:03:58 (0.7630 s / it)
Averaged stats: lr: 0.003895  min_lr: 0.003895  loss: 4.0284 (3.8157)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.2334 (1.2334)  acc1: 75.2604 (75.2604)  acc5: 91.6667 (91.6667)  time: 8.1030  data: 7.8910  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7556 (1.6004)  acc1: 65.7552 (66.8800)  acc5: 88.0208 (88.2400)  time: 1.0607  data: 0.8826  max mem: 41794
Test: Total time: 0:00:09 (1.0764 s / it)
* Acc@1 66.584 Acc@5 88.070 loss 1.606
Accuracy of the model on the 50000 test images: 66.6%
Max accuracy: 66.58%
Epoch: [49]  [  0/312]  eta: 1:15:53  lr: 0.003895  min_lr: 0.003895  loss: 4.1931 (4.1931)  weight_decay: 0.0500 (0.0500)  time: 14.5956  data: 12.7526  max mem: 41794
Epoch: [49]  [ 10/312]  eta: 0:10:33  lr: 0.003895  min_lr: 0.003895  loss: 4.0220 (3.9796)  weight_decay: 0.0500 (0.0500)  time: 2.0982  data: 1.4469  max mem: 41794
Epoch: [49]  [ 20/312]  eta: 0:07:27  lr: 0.003895  min_lr: 0.003895  loss: 3.8033 (3.7529)  weight_decay: 0.0500 (0.0500)  time: 0.8791  data: 0.2287  max mem: 41794
Epoch: [49]  [ 30/312]  eta: 0:05:41  lr: 0.003894  min_lr: 0.003894  loss: 3.9034 (3.7858)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.0742  max mem: 41794
Epoch: [49]  [ 40/312]  eta: 0:04:55  lr: 0.003894  min_lr: 0.003894  loss: 3.8644 (3.7248)  weight_decay: 0.0500 (0.0500)  time: 0.6196  data: 0.0641  max mem: 41794
Epoch: [49]  [ 50/312]  eta: 0:04:27  lr: 0.003894  min_lr: 0.003894  loss: 3.7633 (3.7186)  weight_decay: 0.0500 (0.0500)  time: 0.7249  data: 0.1157  max mem: 41794
Epoch: [49]  [ 60/312]  eta: 0:03:55  lr: 0.003894  min_lr: 0.003894  loss: 3.5817 (3.7044)  weight_decay: 0.0500 (0.0500)  time: 0.6198  data: 0.0557  max mem: 41794
Epoch: [49]  [ 70/312]  eta: 0:03:44  lr: 0.003893  min_lr: 0.003893  loss: 4.0140 (3.7669)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1510  max mem: 41794
Epoch: [49]  [ 80/312]  eta: 0:03:29  lr: 0.003893  min_lr: 0.003893  loss: 4.0214 (3.7737)  weight_decay: 0.0500 (0.0500)  time: 0.8166  data: 0.2419  max mem: 41794
Epoch: [49]  [ 90/312]  eta: 0:03:14  lr: 0.003893  min_lr: 0.003893  loss: 3.9370 (3.7965)  weight_decay: 0.0500 (0.0500)  time: 0.7018  data: 0.1771  max mem: 41794
Epoch: [49]  [100/312]  eta: 0:03:06  lr: 0.003893  min_lr: 0.003893  loss: 3.8744 (3.7947)  weight_decay: 0.0500 (0.0500)  time: 0.7780  data: 0.2487  max mem: 41794
Epoch: [49]  [110/312]  eta: 0:02:50  lr: 0.003893  min_lr: 0.003893  loss: 3.6593 (3.7638)  weight_decay: 0.0500 (0.0500)  time: 0.6903  data: 0.1632  max mem: 41794
Epoch: [49]  [120/312]  eta: 0:02:41  lr: 0.003892  min_lr: 0.003892  loss: 3.4186 (3.7618)  weight_decay: 0.0500 (0.0500)  time: 0.6398  data: 0.1541  max mem: 41794
Epoch: [49]  [130/312]  eta: 0:02:33  lr: 0.003892  min_lr: 0.003892  loss: 3.9183 (3.7763)  weight_decay: 0.0500 (0.0500)  time: 0.8489  data: 0.3582  max mem: 41794
Epoch: [49]  [140/312]  eta: 0:02:20  lr: 0.003892  min_lr: 0.003892  loss: 3.9072 (3.7617)  weight_decay: 0.0500 (0.0500)  time: 0.6952  data: 0.2047  max mem: 41794
Epoch: [49]  [150/312]  eta: 0:02:13  lr: 0.003892  min_lr: 0.003892  loss: 3.4563 (3.7645)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.2168  max mem: 41794
Epoch: [49]  [160/312]  eta: 0:02:03  lr: 0.003891  min_lr: 0.003891  loss: 3.8167 (3.7733)  weight_decay: 0.0500 (0.0500)  time: 0.7852  data: 0.2942  max mem: 41794
Epoch: [49]  [170/312]  eta: 0:01:54  lr: 0.003891  min_lr: 0.003891  loss: 3.9501 (3.7613)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1919  max mem: 41794
Epoch: [49]  [180/312]  eta: 0:01:47  lr: 0.003891  min_lr: 0.003891  loss: 3.6000 (3.7525)  weight_decay: 0.0500 (0.0500)  time: 0.7942  data: 0.3035  max mem: 41794
Epoch: [49]  [190/312]  eta: 0:01:37  lr: 0.003891  min_lr: 0.003891  loss: 3.6000 (3.7456)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.1896  max mem: 41794
Epoch: [49]  [200/312]  eta: 0:01:29  lr: 0.003890  min_lr: 0.003890  loss: 3.8976 (3.7594)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.1988  max mem: 41794
Epoch: [49]  [210/312]  eta: 0:01:21  lr: 0.003890  min_lr: 0.003890  loss: 4.0537 (3.7729)  weight_decay: 0.0500 (0.0500)  time: 0.8340  data: 0.3362  max mem: 41794
Epoch: [49]  [220/312]  eta: 0:01:12  lr: 0.003890  min_lr: 0.003890  loss: 4.0350 (3.7686)  weight_decay: 0.0500 (0.0500)  time: 0.6696  data: 0.1702  max mem: 41794
Epoch: [49]  [230/312]  eta: 0:01:04  lr: 0.003890  min_lr: 0.003890  loss: 4.0350 (3.7757)  weight_decay: 0.0500 (0.0500)  time: 0.6995  data: 0.1690  max mem: 41794
Epoch: [49]  [240/312]  eta: 0:00:56  lr: 0.003889  min_lr: 0.003889  loss: 4.0530 (3.7838)  weight_decay: 0.0500 (0.0500)  time: 0.7998  data: 0.2669  max mem: 41794
Epoch: [49]  [250/312]  eta: 0:00:48  lr: 0.003889  min_lr: 0.003889  loss: 3.9451 (3.7820)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1433  max mem: 41794
Epoch: [49]  [260/312]  eta: 0:00:40  lr: 0.003889  min_lr: 0.003889  loss: 3.9372 (3.7880)  weight_decay: 0.0500 (0.0500)  time: 0.7399  data: 0.1763  max mem: 41794
Epoch: [49]  [270/312]  eta: 0:00:32  lr: 0.003889  min_lr: 0.003889  loss: 3.9372 (3.7856)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1637  max mem: 41794
Epoch: [49]  [280/312]  eta: 0:00:24  lr: 0.003889  min_lr: 0.003889  loss: 3.9924 (3.7949)  weight_decay: 0.0500 (0.0500)  time: 0.6603  data: 0.1616  max mem: 41794
Epoch: [49]  [290/312]  eta: 0:00:17  lr: 0.003888  min_lr: 0.003888  loss: 4.0622 (3.8012)  weight_decay: 0.0500 (0.0500)  time: 0.7937  data: 0.2915  max mem: 41794
Epoch: [49]  [300/312]  eta: 0:00:09  lr: 0.003888  min_lr: 0.003888  loss: 3.9017 (3.7914)  weight_decay: 0.0500 (0.0500)  time: 0.6231  data: 0.1303  max mem: 41794
Epoch: [49]  [310/312]  eta: 0:00:01  lr: 0.003888  min_lr: 0.003888  loss: 3.3330 (3.7862)  weight_decay: 0.0500 (0.0500)  time: 0.4687  data: 0.0001  max mem: 41794
Epoch: [49]  [311/312]  eta: 0:00:00  lr: 0.003888  min_lr: 0.003888  loss: 3.7056 (3.7868)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0001  max mem: 41794
Epoch: [49] Total time: 0:03:57 (0.7620 s / it)
Averaged stats: lr: 0.003888  min_lr: 0.003888  loss: 3.7056 (3.7968)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.3703 (1.3703)  acc1: 74.4792 (74.4792)  acc5: 92.1875 (92.1875)  time: 8.8219  data: 8.6132  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8407 (1.6648)  acc1: 61.9792 (65.4400)  acc5: 85.5469 (87.5200)  time: 1.1288  data: 0.9571  max mem: 41794
Test: Total time: 0:00:10 (1.1545 s / it)
* Acc@1 65.448 Acc@5 87.132 loss 1.669
Accuracy of the model on the 50000 test images: 65.4%
Max accuracy: 66.58%
Epoch: [50]  [  0/312]  eta: 1:23:05  lr: 0.003888  min_lr: 0.003888  loss: 3.7743 (3.7743)  weight_decay: 0.0500 (0.0500)  time: 15.9807  data: 12.2959  max mem: 41794
Epoch: [50]  [ 10/312]  eta: 0:11:08  lr: 0.003888  min_lr: 0.003888  loss: 3.9008 (3.7814)  weight_decay: 0.0500 (0.0500)  time: 2.2136  data: 1.3936  max mem: 41794
Epoch: [50]  [ 20/312]  eta: 0:07:22  lr: 0.003887  min_lr: 0.003887  loss: 3.9188 (3.8880)  weight_decay: 0.0500 (0.0500)  time: 0.7928  data: 0.2497  max mem: 41794
Epoch: [50]  [ 30/312]  eta: 0:05:36  lr: 0.003887  min_lr: 0.003887  loss: 4.1057 (3.9097)  weight_decay: 0.0500 (0.0500)  time: 0.6345  data: 0.1031  max mem: 41794
Epoch: [50]  [ 40/312]  eta: 0:04:51  lr: 0.003887  min_lr: 0.003887  loss: 3.8513 (3.8732)  weight_decay: 0.0500 (0.0500)  time: 0.6048  data: 0.0808  max mem: 41794
Epoch: [50]  [ 50/312]  eta: 0:04:29  lr: 0.003887  min_lr: 0.003887  loss: 3.4228 (3.7867)  weight_decay: 0.0500 (0.0500)  time: 0.7750  data: 0.2094  max mem: 41794
Epoch: [50]  [ 60/312]  eta: 0:03:57  lr: 0.003886  min_lr: 0.003886  loss: 3.7573 (3.8132)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.1340  max mem: 41794
Epoch: [50]  [ 70/312]  eta: 0:03:44  lr: 0.003886  min_lr: 0.003886  loss: 3.9167 (3.8119)  weight_decay: 0.0500 (0.0500)  time: 0.6598  data: 0.1066  max mem: 41794
Epoch: [50]  [ 80/312]  eta: 0:03:33  lr: 0.003886  min_lr: 0.003886  loss: 4.0345 (3.8139)  weight_decay: 0.0500 (0.0500)  time: 0.8627  data: 0.2250  max mem: 41794
Epoch: [50]  [ 90/312]  eta: 0:03:14  lr: 0.003886  min_lr: 0.003886  loss: 3.6801 (3.7869)  weight_decay: 0.0500 (0.0500)  time: 0.6922  data: 0.1207  max mem: 41794
Epoch: [50]  [100/312]  eta: 0:03:08  lr: 0.003885  min_lr: 0.003885  loss: 3.9147 (3.7876)  weight_decay: 0.0500 (0.0500)  time: 0.7544  data: 0.0822  max mem: 41794
Epoch: [50]  [110/312]  eta: 0:02:52  lr: 0.003885  min_lr: 0.003885  loss: 3.9446 (3.7965)  weight_decay: 0.0500 (0.0500)  time: 0.7665  data: 0.0872  max mem: 41794
Epoch: [50]  [120/312]  eta: 0:02:45  lr: 0.003885  min_lr: 0.003885  loss: 3.9446 (3.8015)  weight_decay: 0.0500 (0.0500)  time: 0.7191  data: 0.0245  max mem: 41794
Epoch: [50]  [130/312]  eta: 0:02:34  lr: 0.003885  min_lr: 0.003885  loss: 3.7640 (3.7847)  weight_decay: 0.0500 (0.0500)  time: 0.8231  data: 0.0192  max mem: 41794
Epoch: [50]  [140/312]  eta: 0:02:22  lr: 0.003884  min_lr: 0.003884  loss: 3.7643 (3.7880)  weight_decay: 0.0500 (0.0500)  time: 0.6240  data: 0.0160  max mem: 41794
Epoch: [50]  [150/312]  eta: 0:02:13  lr: 0.003884  min_lr: 0.003884  loss: 3.9896 (3.8000)  weight_decay: 0.0500 (0.0500)  time: 0.6562  data: 0.0181  max mem: 41794
Epoch: [50]  [160/312]  eta: 0:02:05  lr: 0.003884  min_lr: 0.003884  loss: 3.9896 (3.8033)  weight_decay: 0.0500 (0.0500)  time: 0.8080  data: 0.0042  max mem: 41794
Epoch: [50]  [170/312]  eta: 0:01:55  lr: 0.003884  min_lr: 0.003884  loss: 3.9459 (3.7988)  weight_decay: 0.0500 (0.0500)  time: 0.7404  data: 0.0194  max mem: 41794
Epoch: [50]  [180/312]  eta: 0:01:47  lr: 0.003883  min_lr: 0.003883  loss: 3.8039 (3.7909)  weight_decay: 0.0500 (0.0500)  time: 0.7377  data: 0.0322  max mem: 41794
Epoch: [50]  [190/312]  eta: 0:01:37  lr: 0.003883  min_lr: 0.003883  loss: 3.6560 (3.7864)  weight_decay: 0.0500 (0.0500)  time: 0.6642  data: 0.0143  max mem: 41794
Epoch: [50]  [200/312]  eta: 0:01:29  lr: 0.003883  min_lr: 0.003883  loss: 3.9918 (3.7989)  weight_decay: 0.0500 (0.0500)  time: 0.6864  data: 0.0043  max mem: 41794
Epoch: [50]  [210/312]  eta: 0:01:21  lr: 0.003883  min_lr: 0.003883  loss: 3.9595 (3.7953)  weight_decay: 0.0500 (0.0500)  time: 0.8458  data: 0.0055  max mem: 41794
Epoch: [50]  [220/312]  eta: 0:01:12  lr: 0.003883  min_lr: 0.003883  loss: 3.8364 (3.8020)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.0028  max mem: 41794
Epoch: [50]  [230/312]  eta: 0:01:05  lr: 0.003882  min_lr: 0.003882  loss: 3.9131 (3.8065)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.0028  max mem: 41794
Epoch: [50]  [240/312]  eta: 0:00:56  lr: 0.003882  min_lr: 0.003882  loss: 3.9311 (3.8126)  weight_decay: 0.0500 (0.0500)  time: 0.7801  data: 0.0067  max mem: 41794
Epoch: [50]  [250/312]  eta: 0:00:48  lr: 0.003882  min_lr: 0.003882  loss: 3.9961 (3.8083)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.0045  max mem: 41794
Epoch: [50]  [260/312]  eta: 0:00:40  lr: 0.003882  min_lr: 0.003882  loss: 4.0054 (3.8165)  weight_decay: 0.0500 (0.0500)  time: 0.7377  data: 0.0155  max mem: 41794
Epoch: [50]  [270/312]  eta: 0:00:32  lr: 0.003881  min_lr: 0.003881  loss: 3.9731 (3.8181)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.0156  max mem: 41794
Epoch: [50]  [280/312]  eta: 0:00:24  lr: 0.003881  min_lr: 0.003881  loss: 3.9716 (3.8270)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.0017  max mem: 41794
Epoch: [50]  [290/312]  eta: 0:00:17  lr: 0.003881  min_lr: 0.003881  loss: 4.0279 (3.8300)  weight_decay: 0.0500 (0.0500)  time: 0.8328  data: 0.0071  max mem: 41794
Epoch: [50]  [300/312]  eta: 0:00:09  lr: 0.003881  min_lr: 0.003881  loss: 4.0036 (3.8310)  weight_decay: 0.0500 (0.0500)  time: 0.6359  data: 0.0146  max mem: 41794
Epoch: [50]  [310/312]  eta: 0:00:01  lr: 0.003880  min_lr: 0.003880  loss: 4.0036 (3.8270)  weight_decay: 0.0500 (0.0500)  time: 0.4737  data: 0.0089  max mem: 41794
Epoch: [50]  [311/312]  eta: 0:00:00  lr: 0.003880  min_lr: 0.003880  loss: 3.6233 (3.8231)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [50] Total time: 0:03:58 (0.7660 s / it)
Averaged stats: lr: 0.003880  min_lr: 0.003880  loss: 3.6233 (3.8030)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 1.4484 (1.4484)  acc1: 73.0469 (73.0469)  acc5: 91.6667 (91.6667)  time: 9.1446  data: 8.9426  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7893 (1.6704)  acc1: 64.3229 (65.9680)  acc5: 88.5417 (88.1440)  time: 1.1687  data: 0.9937  max mem: 41794
Test: Total time: 0:00:10 (1.2142 s / it)
* Acc@1 66.378 Acc@5 87.786 loss 1.673
Accuracy of the model on the 50000 test images: 66.4%
Max accuracy: 66.58%
Epoch: [51]  [  0/312]  eta: 1:21:46  lr: 0.003880  min_lr: 0.003880  loss: 4.0790 (4.0790)  weight_decay: 0.0500 (0.0500)  time: 15.7259  data: 13.2561  max mem: 41794
Epoch: [51]  [ 10/312]  eta: 0:11:11  lr: 0.003880  min_lr: 0.003880  loss: 3.7889 (3.6846)  weight_decay: 0.0500 (0.0500)  time: 2.2219  data: 1.3199  max mem: 41794
Epoch: [51]  [ 20/312]  eta: 0:07:27  lr: 0.003880  min_lr: 0.003880  loss: 3.3851 (3.4952)  weight_decay: 0.0500 (0.0500)  time: 0.8241  data: 0.1581  max mem: 41794
Epoch: [51]  [ 30/312]  eta: 0:05:42  lr: 0.003880  min_lr: 0.003880  loss: 3.6786 (3.6029)  weight_decay: 0.0500 (0.0500)  time: 0.6582  data: 0.1031  max mem: 41794
Epoch: [51]  [ 40/312]  eta: 0:04:53  lr: 0.003879  min_lr: 0.003879  loss: 3.7680 (3.5947)  weight_decay: 0.0500 (0.0500)  time: 0.6024  data: 0.0142  max mem: 41794
Epoch: [51]  [ 50/312]  eta: 0:04:30  lr: 0.003879  min_lr: 0.003879  loss: 3.7249 (3.6389)  weight_decay: 0.0500 (0.0500)  time: 0.7482  data: 0.0471  max mem: 41794
Epoch: [51]  [ 60/312]  eta: 0:03:57  lr: 0.003879  min_lr: 0.003879  loss: 4.0641 (3.6511)  weight_decay: 0.0500 (0.0500)  time: 0.6564  data: 0.0414  max mem: 41794
Epoch: [51]  [ 70/312]  eta: 0:03:45  lr: 0.003879  min_lr: 0.003879  loss: 3.9905 (3.6467)  weight_decay: 0.0500 (0.0500)  time: 0.6844  data: 0.0686  max mem: 41794
Epoch: [51]  [ 80/312]  eta: 0:03:33  lr: 0.003878  min_lr: 0.003878  loss: 3.4877 (3.6329)  weight_decay: 0.0500 (0.0500)  time: 0.8509  data: 0.1626  max mem: 41794
Epoch: [51]  [ 90/312]  eta: 0:03:13  lr: 0.003878  min_lr: 0.003878  loss: 3.7397 (3.6787)  weight_decay: 0.0500 (0.0500)  time: 0.6576  data: 0.0947  max mem: 41794
Epoch: [51]  [100/312]  eta: 0:03:06  lr: 0.003878  min_lr: 0.003878  loss: 3.9945 (3.6740)  weight_decay: 0.0500 (0.0500)  time: 0.7170  data: 0.1227  max mem: 41794
Epoch: [51]  [110/312]  eta: 0:02:50  lr: 0.003878  min_lr: 0.003878  loss: 3.8932 (3.6934)  weight_decay: 0.0500 (0.0500)  time: 0.7156  data: 0.1246  max mem: 41794
Epoch: [51]  [120/312]  eta: 0:02:42  lr: 0.003877  min_lr: 0.003877  loss: 3.8932 (3.6910)  weight_decay: 0.0500 (0.0500)  time: 0.6783  data: 0.1080  max mem: 41794
Epoch: [51]  [130/312]  eta: 0:02:33  lr: 0.003877  min_lr: 0.003877  loss: 3.8753 (3.6979)  weight_decay: 0.0500 (0.0500)  time: 0.8303  data: 0.2344  max mem: 41794
Epoch: [51]  [140/312]  eta: 0:02:20  lr: 0.003877  min_lr: 0.003877  loss: 3.8753 (3.6995)  weight_decay: 0.0500 (0.0500)  time: 0.6449  data: 0.1289  max mem: 41794
Epoch: [51]  [150/312]  eta: 0:02:14  lr: 0.003877  min_lr: 0.003877  loss: 3.6796 (3.7004)  weight_decay: 0.0500 (0.0500)  time: 0.7362  data: 0.1878  max mem: 41794
Epoch: [51]  [160/312]  eta: 0:02:06  lr: 0.003876  min_lr: 0.003876  loss: 3.9663 (3.7056)  weight_decay: 0.0500 (0.0500)  time: 0.9202  data: 0.3540  max mem: 41794
Epoch: [51]  [170/312]  eta: 0:01:55  lr: 0.003876  min_lr: 0.003876  loss: 4.0640 (3.7249)  weight_decay: 0.0500 (0.0500)  time: 0.6724  data: 0.1669  max mem: 41794
Epoch: [51]  [180/312]  eta: 0:01:47  lr: 0.003876  min_lr: 0.003876  loss: 4.0696 (3.7230)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.1772  max mem: 41794
Epoch: [51]  [190/312]  eta: 0:01:37  lr: 0.003876  min_lr: 0.003876  loss: 3.8401 (3.7147)  weight_decay: 0.0500 (0.0500)  time: 0.6808  data: 0.1786  max mem: 41794
Epoch: [51]  [200/312]  eta: 0:01:29  lr: 0.003875  min_lr: 0.003875  loss: 3.6999 (3.7150)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1818  max mem: 41794
Epoch: [51]  [210/312]  eta: 0:01:22  lr: 0.003875  min_lr: 0.003875  loss: 3.7536 (3.7198)  weight_decay: 0.0500 (0.0500)  time: 0.8893  data: 0.3895  max mem: 41794
Epoch: [51]  [220/312]  eta: 0:01:12  lr: 0.003875  min_lr: 0.003875  loss: 3.8236 (3.7233)  weight_decay: 0.0500 (0.0500)  time: 0.7123  data: 0.2098  max mem: 41794
Epoch: [51]  [230/312]  eta: 0:01:05  lr: 0.003875  min_lr: 0.003875  loss: 4.0077 (3.7349)  weight_decay: 0.0500 (0.0500)  time: 0.6723  data: 0.1774  max mem: 41794
Epoch: [51]  [240/312]  eta: 0:00:57  lr: 0.003874  min_lr: 0.003874  loss: 4.0385 (3.7464)  weight_decay: 0.0500 (0.0500)  time: 0.8755  data: 0.3845  max mem: 41794
Epoch: [51]  [250/312]  eta: 0:00:48  lr: 0.003874  min_lr: 0.003874  loss: 3.9168 (3.7490)  weight_decay: 0.0500 (0.0500)  time: 0.6990  data: 0.2089  max mem: 41794
Epoch: [51]  [260/312]  eta: 0:00:41  lr: 0.003874  min_lr: 0.003874  loss: 3.8735 (3.7455)  weight_decay: 0.0500 (0.0500)  time: 0.6690  data: 0.1734  max mem: 41794
Epoch: [51]  [270/312]  eta: 0:00:32  lr: 0.003874  min_lr: 0.003874  loss: 3.5822 (3.7331)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.1724  max mem: 41794
Epoch: [51]  [280/312]  eta: 0:00:25  lr: 0.003873  min_lr: 0.003873  loss: 3.5822 (3.7387)  weight_decay: 0.0500 (0.0500)  time: 0.7061  data: 0.2046  max mem: 41794
Epoch: [51]  [290/312]  eta: 0:00:17  lr: 0.003873  min_lr: 0.003873  loss: 4.0870 (3.7481)  weight_decay: 0.0500 (0.0500)  time: 0.8271  data: 0.3181  max mem: 41794
Epoch: [51]  [300/312]  eta: 0:00:09  lr: 0.003873  min_lr: 0.003873  loss: 4.0429 (3.7520)  weight_decay: 0.0500 (0.0500)  time: 0.6061  data: 0.1139  max mem: 41794
Epoch: [51]  [310/312]  eta: 0:00:01  lr: 0.003873  min_lr: 0.003873  loss: 4.0125 (3.7500)  weight_decay: 0.0500 (0.0500)  time: 0.4644  data: 0.0001  max mem: 41794
Epoch: [51]  [311/312]  eta: 0:00:00  lr: 0.003873  min_lr: 0.003873  loss: 4.0125 (3.7502)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [51] Total time: 0:03:59 (0.7660 s / it)
Averaged stats: lr: 0.003873  min_lr: 0.003873  loss: 4.0125 (3.7952)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.5720 (1.5720)  acc1: 73.5677 (73.5677)  acc5: 92.7083 (92.7083)  time: 8.2412  data: 8.0275  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9310 (1.7721)  acc1: 62.7604 (64.9120)  acc5: 85.1562 (87.1360)  time: 1.0676  data: 0.8921  max mem: 41794
Test: Total time: 0:00:09 (1.0895 s / it)
* Acc@1 65.352 Acc@5 87.010 loss 1.775
Accuracy of the model on the 50000 test images: 65.4%
Max accuracy: 66.58%
Epoch: [52]  [  0/312]  eta: 1:14:44  lr: 0.003873  min_lr: 0.003873  loss: 3.7644 (3.7644)  weight_decay: 0.0500 (0.0500)  time: 14.3720  data: 13.1835  max mem: 41794
Epoch: [52]  [ 10/312]  eta: 0:10:38  lr: 0.003872  min_lr: 0.003872  loss: 3.8357 (3.7112)  weight_decay: 0.0500 (0.0500)  time: 2.1133  data: 1.3048  max mem: 41794
Epoch: [52]  [ 20/312]  eta: 0:07:31  lr: 0.003872  min_lr: 0.003872  loss: 3.8357 (3.6907)  weight_decay: 0.0500 (0.0500)  time: 0.9040  data: 0.1587  max mem: 41794
Epoch: [52]  [ 30/312]  eta: 0:05:43  lr: 0.003872  min_lr: 0.003872  loss: 3.5859 (3.6416)  weight_decay: 0.0500 (0.0500)  time: 0.7248  data: 0.1052  max mem: 41794
Epoch: [52]  [ 40/312]  eta: 0:04:53  lr: 0.003871  min_lr: 0.003871  loss: 3.7164 (3.7147)  weight_decay: 0.0500 (0.0500)  time: 0.5886  data: 0.0895  max mem: 41794
Epoch: [52]  [ 50/312]  eta: 0:04:25  lr: 0.003871  min_lr: 0.003871  loss: 3.8791 (3.6790)  weight_decay: 0.0500 (0.0500)  time: 0.6958  data: 0.1526  max mem: 41794
Epoch: [52]  [ 60/312]  eta: 0:03:54  lr: 0.003871  min_lr: 0.003871  loss: 3.8977 (3.7082)  weight_decay: 0.0500 (0.0500)  time: 0.6260  data: 0.0801  max mem: 41794
Epoch: [52]  [ 70/312]  eta: 0:03:42  lr: 0.003871  min_lr: 0.003871  loss: 3.7109 (3.7069)  weight_decay: 0.0500 (0.0500)  time: 0.6844  data: 0.0944  max mem: 41794
Epoch: [52]  [ 80/312]  eta: 0:03:31  lr: 0.003870  min_lr: 0.003870  loss: 3.6559 (3.6905)  weight_decay: 0.0500 (0.0500)  time: 0.8608  data: 0.1795  max mem: 41794
Epoch: [52]  [ 90/312]  eta: 0:03:12  lr: 0.003870  min_lr: 0.003870  loss: 3.7886 (3.7165)  weight_decay: 0.0500 (0.0500)  time: 0.6835  data: 0.1002  max mem: 41794
Epoch: [52]  [100/312]  eta: 0:03:04  lr: 0.003870  min_lr: 0.003870  loss: 3.8243 (3.7236)  weight_decay: 0.0500 (0.0500)  time: 0.6936  data: 0.1466  max mem: 41794
Epoch: [52]  [110/312]  eta: 0:02:48  lr: 0.003870  min_lr: 0.003870  loss: 3.8138 (3.7258)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1439  max mem: 41794
Epoch: [52]  [120/312]  eta: 0:02:42  lr: 0.003869  min_lr: 0.003869  loss: 3.8534 (3.7191)  weight_decay: 0.0500 (0.0500)  time: 0.7142  data: 0.1567  max mem: 41794
Epoch: [52]  [130/312]  eta: 0:02:35  lr: 0.003869  min_lr: 0.003869  loss: 3.3913 (3.7015)  weight_decay: 0.0500 (0.0500)  time: 0.9699  data: 0.3301  max mem: 41794
Epoch: [52]  [140/312]  eta: 0:02:22  lr: 0.003869  min_lr: 0.003869  loss: 3.8365 (3.7089)  weight_decay: 0.0500 (0.0500)  time: 0.7427  data: 0.1752  max mem: 41794
Epoch: [52]  [150/312]  eta: 0:02:15  lr: 0.003869  min_lr: 0.003869  loss: 3.9805 (3.7113)  weight_decay: 0.0500 (0.0500)  time: 0.6981  data: 0.1271  max mem: 41794
Epoch: [52]  [160/312]  eta: 0:02:05  lr: 0.003868  min_lr: 0.003868  loss: 3.7932 (3.7066)  weight_decay: 0.0500 (0.0500)  time: 0.7720  data: 0.1289  max mem: 41794
Epoch: [52]  [170/312]  eta: 0:01:55  lr: 0.003868  min_lr: 0.003868  loss: 3.8392 (3.7176)  weight_decay: 0.0500 (0.0500)  time: 0.6386  data: 0.0814  max mem: 41794
Epoch: [52]  [180/312]  eta: 0:01:46  lr: 0.003868  min_lr: 0.003868  loss: 3.8392 (3.7135)  weight_decay: 0.0500 (0.0500)  time: 0.7118  data: 0.1661  max mem: 41794
Epoch: [52]  [190/312]  eta: 0:01:37  lr: 0.003868  min_lr: 0.003868  loss: 3.7887 (3.7245)  weight_decay: 0.0500 (0.0500)  time: 0.6554  data: 0.1045  max mem: 41794
Epoch: [52]  [200/312]  eta: 0:01:29  lr: 0.003867  min_lr: 0.003867  loss: 3.6899 (3.7166)  weight_decay: 0.0500 (0.0500)  time: 0.7045  data: 0.1638  max mem: 41794
Epoch: [52]  [210/312]  eta: 0:01:21  lr: 0.003867  min_lr: 0.003867  loss: 3.5601 (3.7159)  weight_decay: 0.0500 (0.0500)  time: 0.8712  data: 0.2693  max mem: 41794
Epoch: [52]  [220/312]  eta: 0:01:12  lr: 0.003867  min_lr: 0.003867  loss: 3.9137 (3.7387)  weight_decay: 0.0500 (0.0500)  time: 0.6908  data: 0.1347  max mem: 41794
Epoch: [52]  [230/312]  eta: 0:01:05  lr: 0.003867  min_lr: 0.003867  loss: 3.8316 (3.7277)  weight_decay: 0.0500 (0.0500)  time: 0.7273  data: 0.1465  max mem: 41794
Epoch: [52]  [240/312]  eta: 0:00:56  lr: 0.003866  min_lr: 0.003866  loss: 3.6902 (3.7243)  weight_decay: 0.0500 (0.0500)  time: 0.8003  data: 0.1408  max mem: 41794
Epoch: [52]  [250/312]  eta: 0:00:48  lr: 0.003866  min_lr: 0.003866  loss: 3.9149 (3.7301)  weight_decay: 0.0500 (0.0500)  time: 0.6641  data: 0.0959  max mem: 41794
Epoch: [52]  [260/312]  eta: 0:00:40  lr: 0.003866  min_lr: 0.003866  loss: 3.7579 (3.7317)  weight_decay: 0.0500 (0.0500)  time: 0.7613  data: 0.1879  max mem: 41794
Epoch: [52]  [270/312]  eta: 0:00:32  lr: 0.003866  min_lr: 0.003866  loss: 3.7579 (3.7321)  weight_decay: 0.0500 (0.0500)  time: 0.7083  data: 0.1261  max mem: 41794
Epoch: [52]  [280/312]  eta: 0:00:25  lr: 0.003865  min_lr: 0.003865  loss: 3.9886 (3.7410)  weight_decay: 0.0500 (0.0500)  time: 0.7377  data: 0.1434  max mem: 41794
Epoch: [52]  [290/312]  eta: 0:00:17  lr: 0.003865  min_lr: 0.003865  loss: 4.1967 (3.7514)  weight_decay: 0.0500 (0.0500)  time: 0.7757  data: 0.1234  max mem: 41794
Epoch: [52]  [300/312]  eta: 0:00:09  lr: 0.003865  min_lr: 0.003865  loss: 3.9846 (3.7482)  weight_decay: 0.0500 (0.0500)  time: 0.5674  data: 0.0251  max mem: 41794
Epoch: [52]  [310/312]  eta: 0:00:01  lr: 0.003865  min_lr: 0.003865  loss: 3.5419 (3.7446)  weight_decay: 0.0500 (0.0500)  time: 0.4803  data: 0.0179  max mem: 41794
Epoch: [52]  [311/312]  eta: 0:00:00  lr: 0.003865  min_lr: 0.003865  loss: 3.4668 (3.7411)  weight_decay: 0.0500 (0.0500)  time: 0.4809  data: 0.0179  max mem: 41794
Epoch: [52] Total time: 0:03:59 (0.7661 s / it)
Averaged stats: lr: 0.003865  min_lr: 0.003865  loss: 3.4668 (3.7700)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.4787 (1.4787)  acc1: 72.7865 (72.7865)  acc5: 91.9271 (91.9271)  time: 8.4147  data: 8.2011  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8061 (1.7326)  acc1: 65.7552 (66.2080)  acc5: 86.8490 (87.2800)  time: 1.0834  data: 0.9113  max mem: 41794
Test: Total time: 0:00:09 (1.0972 s / it)
* Acc@1 65.978 Acc@5 87.800 loss 1.732
Accuracy of the model on the 50000 test images: 66.0%
Max accuracy: 66.58%
Epoch: [53]  [  0/312]  eta: 1:26:17  lr: 0.003865  min_lr: 0.003865  loss: 3.9246 (3.9246)  weight_decay: 0.0500 (0.0500)  time: 16.5960  data: 13.5454  max mem: 41794
Epoch: [53]  [ 10/312]  eta: 0:11:39  lr: 0.003864  min_lr: 0.003864  loss: 3.9246 (3.7131)  weight_decay: 0.0500 (0.0500)  time: 2.3173  data: 1.3516  max mem: 41794
Epoch: [53]  [ 20/312]  eta: 0:07:37  lr: 0.003864  min_lr: 0.003864  loss: 3.6095 (3.5858)  weight_decay: 0.0500 (0.0500)  time: 0.8144  data: 0.0760  max mem: 41794
Epoch: [53]  [ 30/312]  eta: 0:05:49  lr: 0.003864  min_lr: 0.003864  loss: 3.7420 (3.6627)  weight_decay: 0.0500 (0.0500)  time: 0.6451  data: 0.0194  max mem: 41794
Epoch: [53]  [ 40/312]  eta: 0:04:48  lr: 0.003863  min_lr: 0.003863  loss: 3.7479 (3.6881)  weight_decay: 0.0500 (0.0500)  time: 0.5336  data: 0.0104  max mem: 41794
Epoch: [53]  [ 50/312]  eta: 0:04:19  lr: 0.003863  min_lr: 0.003863  loss: 4.0379 (3.7640)  weight_decay: 0.0500 (0.0500)  time: 0.6093  data: 0.0824  max mem: 41794
Epoch: [53]  [ 60/312]  eta: 0:03:49  lr: 0.003863  min_lr: 0.003863  loss: 4.0651 (3.7265)  weight_decay: 0.0500 (0.0500)  time: 0.6045  data: 0.0818  max mem: 41794
Epoch: [53]  [ 70/312]  eta: 0:03:37  lr: 0.003863  min_lr: 0.003863  loss: 3.7713 (3.7399)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.1397  max mem: 41794
Epoch: [53]  [ 80/312]  eta: 0:03:28  lr: 0.003862  min_lr: 0.003862  loss: 3.8486 (3.7388)  weight_decay: 0.0500 (0.0500)  time: 0.8586  data: 0.3426  max mem: 41794
Epoch: [53]  [ 90/312]  eta: 0:03:10  lr: 0.003862  min_lr: 0.003862  loss: 3.9191 (3.7372)  weight_decay: 0.0500 (0.0500)  time: 0.7024  data: 0.2084  max mem: 41794
Epoch: [53]  [100/312]  eta: 0:03:02  lr: 0.003862  min_lr: 0.003862  loss: 3.9191 (3.7351)  weight_decay: 0.0500 (0.0500)  time: 0.6957  data: 0.1906  max mem: 41794
Epoch: [53]  [110/312]  eta: 0:02:46  lr: 0.003862  min_lr: 0.003862  loss: 3.9555 (3.7438)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1877  max mem: 41794
Epoch: [53]  [120/312]  eta: 0:02:39  lr: 0.003861  min_lr: 0.003861  loss: 4.0032 (3.7331)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1952  max mem: 41794
Epoch: [53]  [130/312]  eta: 0:02:32  lr: 0.003861  min_lr: 0.003861  loss: 3.9192 (3.7274)  weight_decay: 0.0500 (0.0500)  time: 0.9002  data: 0.3707  max mem: 41794
Epoch: [53]  [140/312]  eta: 0:02:19  lr: 0.003861  min_lr: 0.003861  loss: 3.8024 (3.7247)  weight_decay: 0.0500 (0.0500)  time: 0.7022  data: 0.1780  max mem: 41794
Epoch: [53]  [150/312]  eta: 0:02:12  lr: 0.003861  min_lr: 0.003861  loss: 3.6939 (3.7297)  weight_decay: 0.0500 (0.0500)  time: 0.6784  data: 0.1863  max mem: 41794
Epoch: [53]  [160/312]  eta: 0:02:04  lr: 0.003860  min_lr: 0.003860  loss: 3.9098 (3.7420)  weight_decay: 0.0500 (0.0500)  time: 0.8449  data: 0.3455  max mem: 41794
Epoch: [53]  [170/312]  eta: 0:01:53  lr: 0.003860  min_lr: 0.003860  loss: 4.0625 (3.7613)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.1650  max mem: 41794
Epoch: [53]  [180/312]  eta: 0:01:45  lr: 0.003860  min_lr: 0.003860  loss: 4.0087 (3.7626)  weight_decay: 0.0500 (0.0500)  time: 0.6771  data: 0.1779  max mem: 41794
Epoch: [53]  [190/312]  eta: 0:01:35  lr: 0.003860  min_lr: 0.003860  loss: 3.7517 (3.7484)  weight_decay: 0.0500 (0.0500)  time: 0.6704  data: 0.1727  max mem: 41794
Epoch: [53]  [200/312]  eta: 0:01:28  lr: 0.003859  min_lr: 0.003859  loss: 3.8038 (3.7458)  weight_decay: 0.0500 (0.0500)  time: 0.6774  data: 0.1780  max mem: 41794
Epoch: [53]  [210/312]  eta: 0:01:20  lr: 0.003859  min_lr: 0.003859  loss: 3.9144 (3.7546)  weight_decay: 0.0500 (0.0500)  time: 0.8690  data: 0.3704  max mem: 41794
Epoch: [53]  [220/312]  eta: 0:01:11  lr: 0.003859  min_lr: 0.003859  loss: 3.9833 (3.7656)  weight_decay: 0.0500 (0.0500)  time: 0.6886  data: 0.1930  max mem: 41794
Epoch: [53]  [230/312]  eta: 0:01:04  lr: 0.003858  min_lr: 0.003858  loss: 4.0356 (3.7729)  weight_decay: 0.0500 (0.0500)  time: 0.7300  data: 0.2372  max mem: 41794
Epoch: [53]  [240/312]  eta: 0:00:56  lr: 0.003858  min_lr: 0.003858  loss: 4.0927 (3.7875)  weight_decay: 0.0500 (0.0500)  time: 0.9008  data: 0.4101  max mem: 41794
Epoch: [53]  [250/312]  eta: 0:00:48  lr: 0.003858  min_lr: 0.003858  loss: 3.9041 (3.7787)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.1735  max mem: 41794
Epoch: [53]  [260/312]  eta: 0:00:40  lr: 0.003858  min_lr: 0.003858  loss: 3.3686 (3.7645)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.1966  max mem: 41794
Epoch: [53]  [270/312]  eta: 0:00:32  lr: 0.003857  min_lr: 0.003857  loss: 3.5659 (3.7657)  weight_decay: 0.0500 (0.0500)  time: 0.6959  data: 0.1979  max mem: 41794
Epoch: [53]  [280/312]  eta: 0:00:24  lr: 0.003857  min_lr: 0.003857  loss: 3.8481 (3.7518)  weight_decay: 0.0500 (0.0500)  time: 0.6862  data: 0.1880  max mem: 41794
Epoch: [53]  [290/312]  eta: 0:00:17  lr: 0.003857  min_lr: 0.003857  loss: 3.8178 (3.7577)  weight_decay: 0.0500 (0.0500)  time: 0.8519  data: 0.3562  max mem: 41794
Epoch: [53]  [300/312]  eta: 0:00:09  lr: 0.003857  min_lr: 0.003857  loss: 4.0630 (3.7633)  weight_decay: 0.0500 (0.0500)  time: 0.6537  data: 0.1699  max mem: 41794
Epoch: [53]  [310/312]  eta: 0:00:01  lr: 0.003856  min_lr: 0.003856  loss: 4.0431 (3.7674)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.0001  max mem: 41794
Epoch: [53]  [311/312]  eta: 0:00:00  lr: 0.003856  min_lr: 0.003856  loss: 4.0431 (3.7701)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.0001  max mem: 41794
Epoch: [53] Total time: 0:03:58 (0.7630 s / it)
Averaged stats: lr: 0.003856  min_lr: 0.003856  loss: 4.0431 (3.7821)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.4394 (1.4394)  acc1: 76.1719 (76.1719)  acc5: 93.0990 (93.0990)  time: 8.6595  data: 8.4591  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8783 (1.7991)  acc1: 64.7135 (65.9200)  acc5: 87.6302 (87.7280)  time: 1.1232  data: 0.9400  max mem: 41794
Test: Total time: 0:00:10 (1.1723 s / it)
* Acc@1 65.958 Acc@5 87.672 loss 1.801
Accuracy of the model on the 50000 test images: 66.0%
Max accuracy: 66.58%
Epoch: [54]  [  0/312]  eta: 1:20:11  lr: 0.003856  min_lr: 0.003856  loss: 4.4657 (4.4657)  weight_decay: 0.0500 (0.0500)  time: 15.4209  data: 11.9204  max mem: 41794
Epoch: [54]  [ 10/312]  eta: 0:11:11  lr: 0.003856  min_lr: 0.003856  loss: 3.6817 (3.5892)  weight_decay: 0.0500 (0.0500)  time: 2.2226  data: 1.3675  max mem: 41794
Epoch: [54]  [ 20/312]  eta: 0:07:25  lr: 0.003856  min_lr: 0.003856  loss: 3.8771 (3.7849)  weight_decay: 0.0500 (0.0500)  time: 0.8296  data: 0.2021  max mem: 41794
Epoch: [54]  [ 30/312]  eta: 0:05:39  lr: 0.003855  min_lr: 0.003855  loss: 4.0256 (3.7744)  weight_decay: 0.0500 (0.0500)  time: 0.6425  data: 0.0484  max mem: 41794
Epoch: [54]  [ 40/312]  eta: 0:04:47  lr: 0.003855  min_lr: 0.003855  loss: 3.8686 (3.7607)  weight_decay: 0.0500 (0.0500)  time: 0.5676  data: 0.0058  max mem: 41794
Epoch: [54]  [ 50/312]  eta: 0:04:19  lr: 0.003855  min_lr: 0.003855  loss: 3.7800 (3.7497)  weight_decay: 0.0500 (0.0500)  time: 0.6627  data: 0.0346  max mem: 41794
Epoch: [54]  [ 60/312]  eta: 0:03:54  lr: 0.003855  min_lr: 0.003855  loss: 3.5208 (3.6972)  weight_decay: 0.0500 (0.0500)  time: 0.6716  data: 0.0345  max mem: 41794
Epoch: [54]  [ 70/312]  eta: 0:03:43  lr: 0.003854  min_lr: 0.003854  loss: 3.9753 (3.7327)  weight_decay: 0.0500 (0.0500)  time: 0.7448  data: 0.0607  max mem: 41794
Epoch: [54]  [ 80/312]  eta: 0:03:29  lr: 0.003854  min_lr: 0.003854  loss: 3.9482 (3.7086)  weight_decay: 0.0500 (0.0500)  time: 0.8087  data: 0.0578  max mem: 41794
Epoch: [54]  [ 90/312]  eta: 0:03:14  lr: 0.003854  min_lr: 0.003854  loss: 3.5322 (3.6899)  weight_decay: 0.0500 (0.0500)  time: 0.7065  data: 0.0465  max mem: 41794
Epoch: [54]  [100/312]  eta: 0:03:05  lr: 0.003854  min_lr: 0.003854  loss: 3.7961 (3.7207)  weight_decay: 0.0500 (0.0500)  time: 0.7685  data: 0.1119  max mem: 41794
Epoch: [54]  [110/312]  eta: 0:02:49  lr: 0.003853  min_lr: 0.003853  loss: 4.0687 (3.7346)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.0660  max mem: 41794
Epoch: [54]  [120/312]  eta: 0:02:42  lr: 0.003853  min_lr: 0.003853  loss: 3.8589 (3.7222)  weight_decay: 0.0500 (0.0500)  time: 0.7079  data: 0.1235  max mem: 41794
Epoch: [54]  [130/312]  eta: 0:02:34  lr: 0.003853  min_lr: 0.003853  loss: 3.8619 (3.7331)  weight_decay: 0.0500 (0.0500)  time: 0.8956  data: 0.2108  max mem: 41794
Epoch: [54]  [140/312]  eta: 0:02:21  lr: 0.003852  min_lr: 0.003852  loss: 3.7936 (3.7173)  weight_decay: 0.0500 (0.0500)  time: 0.6917  data: 0.0880  max mem: 41794
Epoch: [54]  [150/312]  eta: 0:02:13  lr: 0.003852  min_lr: 0.003852  loss: 3.7443 (3.7273)  weight_decay: 0.0500 (0.0500)  time: 0.6688  data: 0.0627  max mem: 41794
Epoch: [54]  [160/312]  eta: 0:02:04  lr: 0.003852  min_lr: 0.003852  loss: 4.0005 (3.7279)  weight_decay: 0.0500 (0.0500)  time: 0.7948  data: 0.1354  max mem: 41794
Epoch: [54]  [170/312]  eta: 0:01:54  lr: 0.003852  min_lr: 0.003852  loss: 3.8540 (3.7212)  weight_decay: 0.0500 (0.0500)  time: 0.6498  data: 0.0986  max mem: 41794
Epoch: [54]  [180/312]  eta: 0:01:46  lr: 0.003851  min_lr: 0.003851  loss: 3.4782 (3.7075)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.1235  max mem: 41794
Epoch: [54]  [190/312]  eta: 0:01:36  lr: 0.003851  min_lr: 0.003851  loss: 3.4759 (3.6974)  weight_decay: 0.0500 (0.0500)  time: 0.6509  data: 0.0989  max mem: 41794
Epoch: [54]  [200/312]  eta: 0:01:28  lr: 0.003851  min_lr: 0.003851  loss: 3.7656 (3.7019)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.1258  max mem: 41794
Epoch: [54]  [210/312]  eta: 0:01:21  lr: 0.003851  min_lr: 0.003851  loss: 3.7656 (3.6990)  weight_decay: 0.0500 (0.0500)  time: 0.8688  data: 0.2249  max mem: 41794
Epoch: [54]  [220/312]  eta: 0:01:12  lr: 0.003850  min_lr: 0.003850  loss: 3.6042 (3.6993)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.1045  max mem: 41794
Epoch: [54]  [230/312]  eta: 0:01:04  lr: 0.003850  min_lr: 0.003850  loss: 3.8577 (3.7025)  weight_decay: 0.0500 (0.0500)  time: 0.6735  data: 0.0891  max mem: 41794
Epoch: [54]  [240/312]  eta: 0:00:56  lr: 0.003850  min_lr: 0.003850  loss: 3.9668 (3.7159)  weight_decay: 0.0500 (0.0500)  time: 0.8393  data: 0.1683  max mem: 41794
Epoch: [54]  [250/312]  eta: 0:00:48  lr: 0.003849  min_lr: 0.003849  loss: 3.9668 (3.7159)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.0987  max mem: 41794
Epoch: [54]  [260/312]  eta: 0:00:40  lr: 0.003849  min_lr: 0.003849  loss: 3.9017 (3.7196)  weight_decay: 0.0500 (0.0500)  time: 0.7191  data: 0.1368  max mem: 41794
Epoch: [54]  [270/312]  eta: 0:00:32  lr: 0.003849  min_lr: 0.003849  loss: 4.0273 (3.7307)  weight_decay: 0.0500 (0.0500)  time: 0.7085  data: 0.1222  max mem: 41794
Epoch: [54]  [280/312]  eta: 0:00:24  lr: 0.003849  min_lr: 0.003849  loss: 3.6764 (3.7231)  weight_decay: 0.0500 (0.0500)  time: 0.7051  data: 0.1099  max mem: 41794
Epoch: [54]  [290/312]  eta: 0:00:17  lr: 0.003848  min_lr: 0.003848  loss: 3.6764 (3.7261)  weight_decay: 0.0500 (0.0500)  time: 0.8516  data: 0.1856  max mem: 41794
Epoch: [54]  [300/312]  eta: 0:00:09  lr: 0.003848  min_lr: 0.003848  loss: 4.0605 (3.7293)  weight_decay: 0.0500 (0.0500)  time: 0.6321  data: 0.0761  max mem: 41794
Epoch: [54]  [310/312]  eta: 0:00:01  lr: 0.003848  min_lr: 0.003848  loss: 3.9857 (3.7363)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [54]  [311/312]  eta: 0:00:00  lr: 0.003848  min_lr: 0.003848  loss: 4.0605 (3.7389)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [54] Total time: 0:03:57 (0.7627 s / it)
Averaged stats: lr: 0.003848  min_lr: 0.003848  loss: 4.0605 (3.7566)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.5143 (1.5143)  acc1: 72.5260 (72.5260)  acc5: 90.3646 (90.3646)  time: 8.7548  data: 8.5430  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.9272 (1.7846)  acc1: 61.7188 (64.4000)  acc5: 86.5885 (86.6240)  time: 1.1214  data: 0.9493  max mem: 41794
Test: Total time: 0:00:10 (1.1495 s / it)
* Acc@1 64.688 Acc@5 86.600 loss 1.774
Accuracy of the model on the 50000 test images: 64.7%
Max accuracy: 66.58%
Epoch: [55]  [  0/312]  eta: 1:20:42  lr: 0.003848  min_lr: 0.003848  loss: 4.5238 (4.5238)  weight_decay: 0.0500 (0.0500)  time: 15.5194  data: 13.3509  max mem: 41794
Epoch: [55]  [ 10/312]  eta: 0:10:14  lr: 0.003848  min_lr: 0.003848  loss: 3.8254 (3.7513)  weight_decay: 0.0500 (0.0500)  time: 2.0344  data: 1.3468  max mem: 41794
Epoch: [55]  [ 20/312]  eta: 0:07:34  lr: 0.003847  min_lr: 0.003847  loss: 3.6323 (3.6842)  weight_decay: 0.0500 (0.0500)  time: 0.8577  data: 0.1887  max mem: 41794
Epoch: [55]  [ 30/312]  eta: 0:05:45  lr: 0.003847  min_lr: 0.003847  loss: 3.4699 (3.5778)  weight_decay: 0.0500 (0.0500)  time: 0.7796  data: 0.1178  max mem: 41794
Epoch: [55]  [ 40/312]  eta: 0:05:05  lr: 0.003847  min_lr: 0.003847  loss: 3.7654 (3.6430)  weight_decay: 0.0500 (0.0500)  time: 0.6680  data: 0.0674  max mem: 41794
Epoch: [55]  [ 50/312]  eta: 0:04:32  lr: 0.003846  min_lr: 0.003846  loss: 3.7991 (3.6736)  weight_decay: 0.0500 (0.0500)  time: 0.7576  data: 0.1380  max mem: 41794
Epoch: [55]  [ 60/312]  eta: 0:04:00  lr: 0.003846  min_lr: 0.003846  loss: 3.7991 (3.6800)  weight_decay: 0.0500 (0.0500)  time: 0.6046  data: 0.0731  max mem: 41794
Epoch: [55]  [ 70/312]  eta: 0:03:45  lr: 0.003846  min_lr: 0.003846  loss: 3.6755 (3.6629)  weight_decay: 0.0500 (0.0500)  time: 0.6542  data: 0.1017  max mem: 41794
Epoch: [55]  [ 80/312]  eta: 0:03:30  lr: 0.003846  min_lr: 0.003846  loss: 3.5846 (3.6755)  weight_decay: 0.0500 (0.0500)  time: 0.7639  data: 0.1863  max mem: 41794
Epoch: [55]  [ 90/312]  eta: 0:03:13  lr: 0.003845  min_lr: 0.003845  loss: 3.8463 (3.7014)  weight_decay: 0.0500 (0.0500)  time: 0.6474  data: 0.1300  max mem: 41794
Epoch: [55]  [100/312]  eta: 0:03:06  lr: 0.003845  min_lr: 0.003845  loss: 3.8269 (3.7187)  weight_decay: 0.0500 (0.0500)  time: 0.7675  data: 0.1542  max mem: 41794
Epoch: [55]  [110/312]  eta: 0:02:50  lr: 0.003845  min_lr: 0.003845  loss: 3.8078 (3.7113)  weight_decay: 0.0500 (0.0500)  time: 0.7326  data: 0.1094  max mem: 41794
Epoch: [55]  [120/312]  eta: 0:02:43  lr: 0.003844  min_lr: 0.003844  loss: 3.5755 (3.6997)  weight_decay: 0.0500 (0.0500)  time: 0.7101  data: 0.0793  max mem: 41794
Epoch: [55]  [130/312]  eta: 0:02:34  lr: 0.003844  min_lr: 0.003844  loss: 3.8582 (3.7205)  weight_decay: 0.0500 (0.0500)  time: 0.8816  data: 0.1470  max mem: 41794
Epoch: [55]  [140/312]  eta: 0:02:22  lr: 0.003844  min_lr: 0.003844  loss: 3.8635 (3.7221)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.0758  max mem: 41794
Epoch: [55]  [150/312]  eta: 0:02:13  lr: 0.003844  min_lr: 0.003844  loss: 3.8635 (3.7302)  weight_decay: 0.0500 (0.0500)  time: 0.6587  data: 0.1022  max mem: 41794
Epoch: [55]  [160/312]  eta: 0:02:04  lr: 0.003843  min_lr: 0.003843  loss: 3.9564 (3.7331)  weight_decay: 0.0500 (0.0500)  time: 0.7549  data: 0.1798  max mem: 41794
Epoch: [55]  [170/312]  eta: 0:01:54  lr: 0.003843  min_lr: 0.003843  loss: 3.8758 (3.7418)  weight_decay: 0.0500 (0.0500)  time: 0.6629  data: 0.1204  max mem: 41794
Epoch: [55]  [180/312]  eta: 0:01:46  lr: 0.003843  min_lr: 0.003843  loss: 3.8269 (3.7358)  weight_decay: 0.0500 (0.0500)  time: 0.7353  data: 0.1710  max mem: 41794
Epoch: [55]  [190/312]  eta: 0:01:36  lr: 0.003843  min_lr: 0.003843  loss: 3.6260 (3.7317)  weight_decay: 0.0500 (0.0500)  time: 0.6666  data: 0.1364  max mem: 41794
Epoch: [55]  [200/312]  eta: 0:01:29  lr: 0.003842  min_lr: 0.003842  loss: 3.7930 (3.7287)  weight_decay: 0.0500 (0.0500)  time: 0.6991  data: 0.1717  max mem: 41794
Epoch: [55]  [210/312]  eta: 0:01:22  lr: 0.003842  min_lr: 0.003842  loss: 3.9088 (3.7292)  weight_decay: 0.0500 (0.0500)  time: 0.9357  data: 0.3487  max mem: 41794
Epoch: [55]  [220/312]  eta: 0:01:12  lr: 0.003842  min_lr: 0.003842  loss: 3.8100 (3.7131)  weight_decay: 0.0500 (0.0500)  time: 0.7311  data: 0.1787  max mem: 41794
Epoch: [55]  [230/312]  eta: 0:01:05  lr: 0.003841  min_lr: 0.003841  loss: 3.4609 (3.7152)  weight_decay: 0.0500 (0.0500)  time: 0.7095  data: 0.1344  max mem: 41794
Epoch: [55]  [240/312]  eta: 0:00:57  lr: 0.003841  min_lr: 0.003841  loss: 3.8884 (3.7207)  weight_decay: 0.0500 (0.0500)  time: 0.8100  data: 0.1632  max mem: 41794
Epoch: [55]  [250/312]  eta: 0:00:48  lr: 0.003841  min_lr: 0.003841  loss: 3.9793 (3.7261)  weight_decay: 0.0500 (0.0500)  time: 0.6891  data: 0.1251  max mem: 41794
Epoch: [55]  [260/312]  eta: 0:00:41  lr: 0.003841  min_lr: 0.003841  loss: 3.8632 (3.7334)  weight_decay: 0.0500 (0.0500)  time: 0.7757  data: 0.2020  max mem: 41794
Epoch: [55]  [270/312]  eta: 0:00:32  lr: 0.003840  min_lr: 0.003840  loss: 3.8696 (3.7336)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.1084  max mem: 41794
Epoch: [55]  [280/312]  eta: 0:00:24  lr: 0.003840  min_lr: 0.003840  loss: 3.7414 (3.7357)  weight_decay: 0.0500 (0.0500)  time: 0.6414  data: 0.1151  max mem: 41794
Epoch: [55]  [290/312]  eta: 0:00:17  lr: 0.003840  min_lr: 0.003840  loss: 3.6044 (3.7272)  weight_decay: 0.0500 (0.0500)  time: 0.7942  data: 0.1984  max mem: 41794
Epoch: [55]  [300/312]  eta: 0:00:09  lr: 0.003839  min_lr: 0.003839  loss: 3.4629 (3.7173)  weight_decay: 0.0500 (0.0500)  time: 0.6340  data: 0.0849  max mem: 41794
Epoch: [55]  [310/312]  eta: 0:00:01  lr: 0.003839  min_lr: 0.003839  loss: 3.8304 (3.7249)  weight_decay: 0.0500 (0.0500)  time: 0.4640  data: 0.0001  max mem: 41794
Epoch: [55]  [311/312]  eta: 0:00:00  lr: 0.003839  min_lr: 0.003839  loss: 3.8304 (3.7210)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [55] Total time: 0:03:59 (0.7662 s / it)
Averaged stats: lr: 0.003839  min_lr: 0.003839  loss: 3.8304 (3.7253)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.4190 (1.4190)  acc1: 73.6979 (73.6979)  acc5: 91.9271 (91.9271)  time: 8.0945  data: 7.8817  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7701 (1.6308)  acc1: 67.1875 (67.4080)  acc5: 88.0208 (88.4480)  time: 1.0480  data: 0.8758  max mem: 41794
Test: Total time: 0:00:09 (1.0622 s / it)
* Acc@1 67.616 Acc@5 88.562 loss 1.642
Accuracy of the model on the 50000 test images: 67.6%
Max accuracy: 67.62%
Epoch: [56]  [  0/312]  eta: 1:14:33  lr: 0.003839  min_lr: 0.003839  loss: 3.5393 (3.5393)  weight_decay: 0.0500 (0.0500)  time: 14.3375  data: 12.4110  max mem: 41794
Epoch: [56]  [ 10/312]  eta: 0:10:54  lr: 0.003839  min_lr: 0.003839  loss: 3.8731 (3.7084)  weight_decay: 0.0500 (0.0500)  time: 2.1681  data: 1.2805  max mem: 41794
Epoch: [56]  [ 20/312]  eta: 0:07:14  lr: 0.003839  min_lr: 0.003839  loss: 3.9908 (3.7110)  weight_decay: 0.0500 (0.0500)  time: 0.8469  data: 0.0985  max mem: 41794
Epoch: [56]  [ 30/312]  eta: 0:05:39  lr: 0.003838  min_lr: 0.003838  loss: 3.6222 (3.6064)  weight_decay: 0.0500 (0.0500)  time: 0.6729  data: 0.0654  max mem: 41794
Epoch: [56]  [ 40/312]  eta: 0:04:41  lr: 0.003838  min_lr: 0.003838  loss: 3.6424 (3.6702)  weight_decay: 0.0500 (0.0500)  time: 0.5592  data: 0.0582  max mem: 41794
Epoch: [56]  [ 50/312]  eta: 0:04:18  lr: 0.003838  min_lr: 0.003838  loss: 4.0078 (3.7456)  weight_decay: 0.0500 (0.0500)  time: 0.6502  data: 0.0726  max mem: 41794
Epoch: [56]  [ 60/312]  eta: 0:03:48  lr: 0.003837  min_lr: 0.003837  loss: 4.0100 (3.7047)  weight_decay: 0.0500 (0.0500)  time: 0.6448  data: 0.0668  max mem: 41794
Epoch: [56]  [ 70/312]  eta: 0:03:36  lr: 0.003837  min_lr: 0.003837  loss: 4.0263 (3.7352)  weight_decay: 0.0500 (0.0500)  time: 0.6652  data: 0.0941  max mem: 41794
Epoch: [56]  [ 80/312]  eta: 0:03:22  lr: 0.003837  min_lr: 0.003837  loss: 3.7913 (3.7058)  weight_decay: 0.0500 (0.0500)  time: 0.7681  data: 0.1567  max mem: 41794
Epoch: [56]  [ 90/312]  eta: 0:03:09  lr: 0.003837  min_lr: 0.003837  loss: 3.4393 (3.6869)  weight_decay: 0.0500 (0.0500)  time: 0.6930  data: 0.1049  max mem: 41794
Epoch: [56]  [100/312]  eta: 0:03:00  lr: 0.003836  min_lr: 0.003836  loss: 4.0277 (3.7188)  weight_decay: 0.0500 (0.0500)  time: 0.7575  data: 0.1113  max mem: 41794
Epoch: [56]  [110/312]  eta: 0:02:45  lr: 0.003836  min_lr: 0.003836  loss: 4.0234 (3.7186)  weight_decay: 0.0500 (0.0500)  time: 0.6684  data: 0.0770  max mem: 41794
Epoch: [56]  [120/312]  eta: 0:02:39  lr: 0.003836  min_lr: 0.003836  loss: 3.8834 (3.7312)  weight_decay: 0.0500 (0.0500)  time: 0.7178  data: 0.1304  max mem: 41794
Epoch: [56]  [130/312]  eta: 0:02:31  lr: 0.003835  min_lr: 0.003835  loss: 3.9026 (3.7352)  weight_decay: 0.0500 (0.0500)  time: 0.8994  data: 0.2173  max mem: 41794
Epoch: [56]  [140/312]  eta: 0:02:19  lr: 0.003835  min_lr: 0.003835  loss: 3.8772 (3.7371)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.1136  max mem: 41794
Epoch: [56]  [150/312]  eta: 0:02:11  lr: 0.003835  min_lr: 0.003835  loss: 3.8813 (3.7357)  weight_decay: 0.0500 (0.0500)  time: 0.6788  data: 0.1136  max mem: 41794
Epoch: [56]  [160/312]  eta: 0:02:02  lr: 0.003835  min_lr: 0.003835  loss: 3.5969 (3.7248)  weight_decay: 0.0500 (0.0500)  time: 0.7918  data: 0.1664  max mem: 41794
Epoch: [56]  [170/312]  eta: 0:01:53  lr: 0.003834  min_lr: 0.003834  loss: 3.8647 (3.7412)  weight_decay: 0.0500 (0.0500)  time: 0.7007  data: 0.1094  max mem: 41794
Epoch: [56]  [180/312]  eta: 0:01:46  lr: 0.003834  min_lr: 0.003834  loss: 3.9085 (3.7312)  weight_decay: 0.0500 (0.0500)  time: 0.7749  data: 0.1146  max mem: 41794
Epoch: [56]  [190/312]  eta: 0:01:36  lr: 0.003834  min_lr: 0.003834  loss: 3.8760 (3.7382)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.0802  max mem: 41794
Epoch: [56]  [200/312]  eta: 0:01:28  lr: 0.003833  min_lr: 0.003833  loss: 3.9429 (3.7432)  weight_decay: 0.0500 (0.0500)  time: 0.6598  data: 0.1044  max mem: 41794
Epoch: [56]  [210/312]  eta: 0:01:20  lr: 0.003833  min_lr: 0.003833  loss: 3.8292 (3.7349)  weight_decay: 0.0500 (0.0500)  time: 0.8300  data: 0.2345  max mem: 41794
Epoch: [56]  [220/312]  eta: 0:01:11  lr: 0.003833  min_lr: 0.003833  loss: 3.0379 (3.7145)  weight_decay: 0.0500 (0.0500)  time: 0.6593  data: 0.1323  max mem: 41794
Epoch: [56]  [230/312]  eta: 0:01:04  lr: 0.003833  min_lr: 0.003833  loss: 3.4320 (3.7190)  weight_decay: 0.0500 (0.0500)  time: 0.7351  data: 0.1079  max mem: 41794
Epoch: [56]  [240/312]  eta: 0:00:56  lr: 0.003832  min_lr: 0.003832  loss: 3.9638 (3.7194)  weight_decay: 0.0500 (0.0500)  time: 0.8048  data: 0.1653  max mem: 41794
Epoch: [56]  [250/312]  eta: 0:00:48  lr: 0.003832  min_lr: 0.003832  loss: 3.7107 (3.7178)  weight_decay: 0.0500 (0.0500)  time: 0.6576  data: 0.0776  max mem: 41794
Epoch: [56]  [260/312]  eta: 0:00:40  lr: 0.003832  min_lr: 0.003832  loss: 3.6868 (3.7150)  weight_decay: 0.0500 (0.0500)  time: 0.7780  data: 0.0978  max mem: 41794
Epoch: [56]  [270/312]  eta: 0:00:32  lr: 0.003831  min_lr: 0.003831  loss: 3.8793 (3.7206)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.0782  max mem: 41794
Epoch: [56]  [280/312]  eta: 0:00:24  lr: 0.003831  min_lr: 0.003831  loss: 3.9186 (3.7269)  weight_decay: 0.0500 (0.0500)  time: 0.7346  data: 0.0984  max mem: 41794
Epoch: [56]  [290/312]  eta: 0:00:17  lr: 0.003831  min_lr: 0.003831  loss: 3.9584 (3.7326)  weight_decay: 0.0500 (0.0500)  time: 0.8640  data: 0.1492  max mem: 41794
Epoch: [56]  [300/312]  eta: 0:00:09  lr: 0.003831  min_lr: 0.003831  loss: 4.0420 (3.7346)  weight_decay: 0.0500 (0.0500)  time: 0.6150  data: 0.0512  max mem: 41794
Epoch: [56]  [310/312]  eta: 0:00:01  lr: 0.003830  min_lr: 0.003830  loss: 3.7242 (3.7294)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [56]  [311/312]  eta: 0:00:00  lr: 0.003830  min_lr: 0.003830  loss: 3.8770 (3.7298)  weight_decay: 0.0500 (0.0500)  time: 0.4644  data: 0.0001  max mem: 41794
Epoch: [56] Total time: 0:03:57 (0.7623 s / it)
Averaged stats: lr: 0.003830  min_lr: 0.003830  loss: 3.8770 (3.7357)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3044 (1.3044)  acc1: 73.8281 (73.8281)  acc5: 92.9688 (92.9688)  time: 8.5458  data: 8.3324  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7025 (1.6067)  acc1: 65.8854 (66.5760)  acc5: 87.7604 (88.4000)  time: 1.1119  data: 0.9259  max mem: 41794
Test: Total time: 0:00:10 (1.1445 s / it)
* Acc@1 67.004 Acc@5 88.480 loss 1.599
Accuracy of the model on the 50000 test images: 67.0%
Max accuracy: 67.62%
Epoch: [57]  [  0/312]  eta: 1:23:09  lr: 0.003830  min_lr: 0.003830  loss: 2.8196 (2.8196)  weight_decay: 0.0500 (0.0500)  time: 15.9917  data: 13.4171  max mem: 41794
Epoch: [57]  [ 10/312]  eta: 0:10:52  lr: 0.003830  min_lr: 0.003830  loss: 3.8772 (3.6233)  weight_decay: 0.0500 (0.0500)  time: 2.1615  data: 1.2587  max mem: 41794
Epoch: [57]  [ 20/312]  eta: 0:07:24  lr: 0.003830  min_lr: 0.003830  loss: 3.7680 (3.6632)  weight_decay: 0.0500 (0.0500)  time: 0.8005  data: 0.0820  max mem: 41794
Epoch: [57]  [ 30/312]  eta: 0:05:40  lr: 0.003829  min_lr: 0.003829  loss: 3.7368 (3.7038)  weight_decay: 0.0500 (0.0500)  time: 0.6837  data: 0.0652  max mem: 41794
Epoch: [57]  [ 40/312]  eta: 0:04:46  lr: 0.003829  min_lr: 0.003829  loss: 3.6723 (3.6322)  weight_decay: 0.0500 (0.0500)  time: 0.5617  data: 0.0078  max mem: 41794
Epoch: [57]  [ 50/312]  eta: 0:04:22  lr: 0.003829  min_lr: 0.003829  loss: 3.4686 (3.6428)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.0505  max mem: 41794
Epoch: [57]  [ 60/312]  eta: 0:03:56  lr: 0.003828  min_lr: 0.003828  loss: 3.8890 (3.6727)  weight_decay: 0.0500 (0.0500)  time: 0.7018  data: 0.1136  max mem: 41794
Epoch: [57]  [ 70/312]  eta: 0:03:46  lr: 0.003828  min_lr: 0.003828  loss: 3.7085 (3.6475)  weight_decay: 0.0500 (0.0500)  time: 0.7674  data: 0.1755  max mem: 41794
Epoch: [57]  [ 80/312]  eta: 0:03:31  lr: 0.003828  min_lr: 0.003828  loss: 3.4978 (3.6540)  weight_decay: 0.0500 (0.0500)  time: 0.8278  data: 0.1096  max mem: 41794
Epoch: [57]  [ 90/312]  eta: 0:03:16  lr: 0.003828  min_lr: 0.003828  loss: 3.8760 (3.6809)  weight_decay: 0.0500 (0.0500)  time: 0.7115  data: 0.0979  max mem: 41794
Epoch: [57]  [100/312]  eta: 0:03:07  lr: 0.003827  min_lr: 0.003827  loss: 3.7659 (3.6706)  weight_decay: 0.0500 (0.0500)  time: 0.7680  data: 0.1784  max mem: 41794
Epoch: [57]  [110/312]  eta: 0:02:51  lr: 0.003827  min_lr: 0.003827  loss: 3.7041 (3.6773)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.0812  max mem: 41794
Epoch: [57]  [120/312]  eta: 0:02:43  lr: 0.003827  min_lr: 0.003827  loss: 3.9687 (3.6835)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.0913  max mem: 41794
Epoch: [57]  [130/312]  eta: 0:02:32  lr: 0.003826  min_lr: 0.003826  loss: 3.9377 (3.7021)  weight_decay: 0.0500 (0.0500)  time: 0.7892  data: 0.1000  max mem: 41794
Epoch: [57]  [140/312]  eta: 0:02:22  lr: 0.003826  min_lr: 0.003826  loss: 3.9377 (3.7110)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1060  max mem: 41794
Epoch: [57]  [150/312]  eta: 0:02:14  lr: 0.003826  min_lr: 0.003826  loss: 3.7033 (3.6973)  weight_decay: 0.0500 (0.0500)  time: 0.7670  data: 0.1765  max mem: 41794
Epoch: [57]  [160/312]  eta: 0:02:04  lr: 0.003825  min_lr: 0.003825  loss: 3.7033 (3.7085)  weight_decay: 0.0500 (0.0500)  time: 0.7548  data: 0.0846  max mem: 41794
Epoch: [57]  [170/312]  eta: 0:01:55  lr: 0.003825  min_lr: 0.003825  loss: 3.6777 (3.6982)  weight_decay: 0.0500 (0.0500)  time: 0.7154  data: 0.0965  max mem: 41794
Epoch: [57]  [180/312]  eta: 0:01:48  lr: 0.003825  min_lr: 0.003825  loss: 3.5674 (3.6994)  weight_decay: 0.0500 (0.0500)  time: 0.8185  data: 0.1664  max mem: 41794
Epoch: [57]  [190/312]  eta: 0:01:37  lr: 0.003825  min_lr: 0.003825  loss: 3.9422 (3.7097)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.0772  max mem: 41794
Epoch: [57]  [200/312]  eta: 0:01:30  lr: 0.003824  min_lr: 0.003824  loss: 3.7945 (3.6935)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.0614  max mem: 41794
Epoch: [57]  [210/312]  eta: 0:01:21  lr: 0.003824  min_lr: 0.003824  loss: 3.5217 (3.6865)  weight_decay: 0.0500 (0.0500)  time: 0.7947  data: 0.0643  max mem: 41794
Epoch: [57]  [220/312]  eta: 0:01:13  lr: 0.003824  min_lr: 0.003824  loss: 3.9632 (3.6956)  weight_decay: 0.0500 (0.0500)  time: 0.7036  data: 0.0819  max mem: 41794
Epoch: [57]  [230/312]  eta: 0:01:05  lr: 0.003823  min_lr: 0.003823  loss: 3.9900 (3.6992)  weight_decay: 0.0500 (0.0500)  time: 0.7548  data: 0.1598  max mem: 41794
Epoch: [57]  [240/312]  eta: 0:00:57  lr: 0.003823  min_lr: 0.003823  loss: 3.8875 (3.7066)  weight_decay: 0.0500 (0.0500)  time: 0.7507  data: 0.0833  max mem: 41794
Epoch: [57]  [250/312]  eta: 0:00:48  lr: 0.003823  min_lr: 0.003823  loss: 3.7917 (3.7033)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0716  max mem: 41794
Epoch: [57]  [260/312]  eta: 0:00:41  lr: 0.003823  min_lr: 0.003823  loss: 3.7917 (3.7046)  weight_decay: 0.0500 (0.0500)  time: 0.7924  data: 0.1501  max mem: 41794
Epoch: [57]  [270/312]  eta: 0:00:32  lr: 0.003822  min_lr: 0.003822  loss: 3.8969 (3.6991)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.0805  max mem: 41794
Epoch: [57]  [280/312]  eta: 0:00:24  lr: 0.003822  min_lr: 0.003822  loss: 3.9483 (3.7018)  weight_decay: 0.0500 (0.0500)  time: 0.6143  data: 0.0663  max mem: 41794
Epoch: [57]  [290/312]  eta: 0:00:17  lr: 0.003822  min_lr: 0.003822  loss: 3.9596 (3.7103)  weight_decay: 0.0500 (0.0500)  time: 0.7419  data: 0.0946  max mem: 41794
Epoch: [57]  [300/312]  eta: 0:00:09  lr: 0.003821  min_lr: 0.003821  loss: 4.0213 (3.7123)  weight_decay: 0.0500 (0.0500)  time: 0.6199  data: 0.0405  max mem: 41794
Epoch: [57]  [310/312]  eta: 0:00:01  lr: 0.003821  min_lr: 0.003821  loss: 4.0213 (3.7233)  weight_decay: 0.0500 (0.0500)  time: 0.4719  data: 0.0104  max mem: 41794
Epoch: [57]  [311/312]  eta: 0:00:00  lr: 0.003821  min_lr: 0.003821  loss: 4.0213 (3.7246)  weight_decay: 0.0500 (0.0500)  time: 0.4620  data: 0.0001  max mem: 41794
Epoch: [57] Total time: 0:03:58 (0.7650 s / it)
Averaged stats: lr: 0.003821  min_lr: 0.003821  loss: 4.0213 (3.7524)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.4811 (1.4811)  acc1: 73.8281 (73.8281)  acc5: 91.5365 (91.5365)  time: 8.3907  data: 8.1779  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7718 (1.7070)  acc1: 66.2760 (66.6560)  acc5: 89.1927 (88.3040)  time: 1.1021  data: 0.9251  max mem: 41794
Test: Total time: 0:00:10 (1.1120 s / it)
* Acc@1 66.718 Acc@5 88.164 loss 1.695
Accuracy of the model on the 50000 test images: 66.7%
Max accuracy: 67.62%
Epoch: [58]  [  0/312]  eta: 1:24:53  lr: 0.003821  min_lr: 0.003821  loss: 4.0542 (4.0542)  weight_decay: 0.0500 (0.0500)  time: 16.3252  data: 13.5161  max mem: 41794
Epoch: [58]  [ 10/312]  eta: 0:11:34  lr: 0.003821  min_lr: 0.003821  loss: 4.0440 (3.8571)  weight_decay: 0.0500 (0.0500)  time: 2.2996  data: 1.3451  max mem: 41794
Epoch: [58]  [ 20/312]  eta: 0:07:18  lr: 0.003820  min_lr: 0.003820  loss: 3.9159 (3.8157)  weight_decay: 0.0500 (0.0500)  time: 0.7615  data: 0.0644  max mem: 41794
Epoch: [58]  [ 30/312]  eta: 0:05:34  lr: 0.003820  min_lr: 0.003820  loss: 3.8139 (3.7887)  weight_decay: 0.0500 (0.0500)  time: 0.5729  data: 0.0025  max mem: 41794
Epoch: [58]  [ 40/312]  eta: 0:04:40  lr: 0.003820  min_lr: 0.003820  loss: 3.7466 (3.7552)  weight_decay: 0.0500 (0.0500)  time: 0.5382  data: 0.0278  max mem: 41794
Epoch: [58]  [ 50/312]  eta: 0:04:22  lr: 0.003820  min_lr: 0.003820  loss: 3.7466 (3.7437)  weight_decay: 0.0500 (0.0500)  time: 0.7202  data: 0.2195  max mem: 41794
Epoch: [58]  [ 60/312]  eta: 0:03:52  lr: 0.003819  min_lr: 0.003819  loss: 3.7654 (3.7418)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.1941  max mem: 41794
Epoch: [58]  [ 70/312]  eta: 0:03:42  lr: 0.003819  min_lr: 0.003819  loss: 3.8405 (3.7255)  weight_decay: 0.0500 (0.0500)  time: 0.7062  data: 0.2016  max mem: 41794
Epoch: [58]  [ 80/312]  eta: 0:03:30  lr: 0.003819  min_lr: 0.003819  loss: 3.9053 (3.7502)  weight_decay: 0.0500 (0.0500)  time: 0.8705  data: 0.3638  max mem: 41794
Epoch: [58]  [ 90/312]  eta: 0:03:11  lr: 0.003818  min_lr: 0.003818  loss: 3.9053 (3.7503)  weight_decay: 0.0500 (0.0500)  time: 0.6687  data: 0.1681  max mem: 41794
Epoch: [58]  [100/312]  eta: 0:03:05  lr: 0.003818  min_lr: 0.003818  loss: 3.9035 (3.7621)  weight_decay: 0.0500 (0.0500)  time: 0.7354  data: 0.2291  max mem: 41794
Epoch: [58]  [110/312]  eta: 0:02:49  lr: 0.003818  min_lr: 0.003818  loss: 4.0048 (3.7656)  weight_decay: 0.0500 (0.0500)  time: 0.7362  data: 0.2239  max mem: 41794
Epoch: [58]  [120/312]  eta: 0:02:41  lr: 0.003817  min_lr: 0.003817  loss: 3.8898 (3.7842)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.1792  max mem: 41794
Epoch: [58]  [130/312]  eta: 0:02:34  lr: 0.003817  min_lr: 0.003817  loss: 4.0232 (3.7901)  weight_decay: 0.0500 (0.0500)  time: 0.8799  data: 0.3742  max mem: 41794
Epoch: [58]  [140/312]  eta: 0:02:21  lr: 0.003817  min_lr: 0.003817  loss: 4.0684 (3.7959)  weight_decay: 0.0500 (0.0500)  time: 0.6955  data: 0.1956  max mem: 41794
Epoch: [58]  [150/312]  eta: 0:02:13  lr: 0.003817  min_lr: 0.003817  loss: 4.0511 (3.8020)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.1791  max mem: 41794
Epoch: [58]  [160/312]  eta: 0:02:05  lr: 0.003816  min_lr: 0.003816  loss: 4.1138 (3.8131)  weight_decay: 0.0500 (0.0500)  time: 0.8602  data: 0.3657  max mem: 41794
Epoch: [58]  [170/312]  eta: 0:01:55  lr: 0.003816  min_lr: 0.003816  loss: 3.9812 (3.8229)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.2114  max mem: 41794
Epoch: [58]  [180/312]  eta: 0:01:46  lr: 0.003816  min_lr: 0.003816  loss: 3.9812 (3.8316)  weight_decay: 0.0500 (0.0500)  time: 0.6740  data: 0.1778  max mem: 41794
Epoch: [58]  [190/312]  eta: 0:01:36  lr: 0.003815  min_lr: 0.003815  loss: 4.0550 (3.8344)  weight_decay: 0.0500 (0.0500)  time: 0.6500  data: 0.1536  max mem: 41794
Epoch: [58]  [200/312]  eta: 0:01:29  lr: 0.003815  min_lr: 0.003815  loss: 3.6610 (3.8184)  weight_decay: 0.0500 (0.0500)  time: 0.7296  data: 0.1969  max mem: 41794
Epoch: [58]  [210/312]  eta: 0:01:21  lr: 0.003815  min_lr: 0.003815  loss: 3.5816 (3.8230)  weight_decay: 0.0500 (0.0500)  time: 0.8948  data: 0.3589  max mem: 41794
Epoch: [58]  [220/312]  eta: 0:01:12  lr: 0.003814  min_lr: 0.003814  loss: 3.7952 (3.8176)  weight_decay: 0.0500 (0.0500)  time: 0.6632  data: 0.1628  max mem: 41794
Epoch: [58]  [230/312]  eta: 0:01:05  lr: 0.003814  min_lr: 0.003814  loss: 3.9367 (3.8187)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.1578  max mem: 41794
Epoch: [58]  [240/312]  eta: 0:00:57  lr: 0.003814  min_lr: 0.003814  loss: 3.9687 (3.8172)  weight_decay: 0.0500 (0.0500)  time: 0.8705  data: 0.2566  max mem: 41794
Epoch: [58]  [250/312]  eta: 0:00:48  lr: 0.003813  min_lr: 0.003813  loss: 3.6459 (3.8036)  weight_decay: 0.0500 (0.0500)  time: 0.7006  data: 0.1286  max mem: 41794
Epoch: [58]  [260/312]  eta: 0:00:41  lr: 0.003813  min_lr: 0.003813  loss: 3.4032 (3.7968)  weight_decay: 0.0500 (0.0500)  time: 0.7241  data: 0.1393  max mem: 41794
Epoch: [58]  [270/312]  eta: 0:00:32  lr: 0.003813  min_lr: 0.003813  loss: 3.6388 (3.7930)  weight_decay: 0.0500 (0.0500)  time: 0.7005  data: 0.1103  max mem: 41794
Epoch: [58]  [280/312]  eta: 0:00:25  lr: 0.003813  min_lr: 0.003813  loss: 3.7837 (3.7935)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.0981  max mem: 41794
Epoch: [58]  [290/312]  eta: 0:00:17  lr: 0.003812  min_lr: 0.003812  loss: 4.0368 (3.7975)  weight_decay: 0.0500 (0.0500)  time: 0.8115  data: 0.1843  max mem: 41794
Epoch: [58]  [300/312]  eta: 0:00:09  lr: 0.003812  min_lr: 0.003812  loss: 4.0401 (3.7950)  weight_decay: 0.0500 (0.0500)  time: 0.6111  data: 0.0884  max mem: 41794
Epoch: [58]  [310/312]  eta: 0:00:01  lr: 0.003812  min_lr: 0.003812  loss: 3.8951 (3.7956)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0020  max mem: 41794
Epoch: [58]  [311/312]  eta: 0:00:00  lr: 0.003812  min_lr: 0.003812  loss: 3.8951 (3.7959)  weight_decay: 0.0500 (0.0500)  time: 0.4633  data: 0.0001  max mem: 41794
Epoch: [58] Total time: 0:03:59 (0.7677 s / it)
Averaged stats: lr: 0.003812  min_lr: 0.003812  loss: 3.8951 (3.7452)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3897 (1.3897)  acc1: 75.5208 (75.5208)  acc5: 91.9271 (91.9271)  time: 8.4668  data: 8.2535  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7997 (1.7441)  acc1: 65.1042 (66.6880)  acc5: 88.9323 (87.9520)  time: 1.0947  data: 0.9172  max mem: 41794
Test: Total time: 0:00:10 (1.1112 s / it)
* Acc@1 67.004 Acc@5 88.160 loss 1.738
Accuracy of the model on the 50000 test images: 67.0%
Max accuracy: 67.62%
Epoch: [59]  [  0/312]  eta: 1:21:15  lr: 0.003812  min_lr: 0.003812  loss: 3.5184 (3.5184)  weight_decay: 0.0500 (0.0500)  time: 15.6264  data: 14.6816  max mem: 41794
Epoch: [59]  [ 10/312]  eta: 0:11:16  lr: 0.003811  min_lr: 0.003811  loss: 4.0614 (3.8557)  weight_decay: 0.0500 (0.0500)  time: 2.2386  data: 1.3818  max mem: 41794
Epoch: [59]  [ 20/312]  eta: 0:07:20  lr: 0.003811  min_lr: 0.003811  loss: 3.7022 (3.7266)  weight_decay: 0.0500 (0.0500)  time: 0.8037  data: 0.1284  max mem: 41794
Epoch: [59]  [ 30/312]  eta: 0:05:34  lr: 0.003811  min_lr: 0.003811  loss: 3.7022 (3.7190)  weight_decay: 0.0500 (0.0500)  time: 0.6097  data: 0.1069  max mem: 41794
Epoch: [59]  [ 40/312]  eta: 0:04:46  lr: 0.003810  min_lr: 0.003810  loss: 3.7278 (3.6858)  weight_decay: 0.0500 (0.0500)  time: 0.5729  data: 0.0703  max mem: 41794
Epoch: [59]  [ 50/312]  eta: 0:04:22  lr: 0.003810  min_lr: 0.003810  loss: 3.5559 (3.6659)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.1839  max mem: 41794
Epoch: [59]  [ 60/312]  eta: 0:03:51  lr: 0.003810  min_lr: 0.003810  loss: 3.5559 (3.6346)  weight_decay: 0.0500 (0.0500)  time: 0.6380  data: 0.1183  max mem: 41794
Epoch: [59]  [ 70/312]  eta: 0:03:40  lr: 0.003809  min_lr: 0.003809  loss: 3.8973 (3.6978)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.1633  max mem: 41794
Epoch: [59]  [ 80/312]  eta: 0:03:30  lr: 0.003809  min_lr: 0.003809  loss: 4.0260 (3.6909)  weight_decay: 0.0500 (0.0500)  time: 0.8734  data: 0.3532  max mem: 41794
Epoch: [59]  [ 90/312]  eta: 0:03:13  lr: 0.003809  min_lr: 0.003809  loss: 3.6427 (3.7085)  weight_decay: 0.0500 (0.0500)  time: 0.7398  data: 0.1927  max mem: 41794
Epoch: [59]  [100/312]  eta: 0:03:04  lr: 0.003809  min_lr: 0.003809  loss: 3.9837 (3.7170)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.1700  max mem: 41794
Epoch: [59]  [110/312]  eta: 0:02:49  lr: 0.003808  min_lr: 0.003808  loss: 3.9521 (3.7355)  weight_decay: 0.0500 (0.0500)  time: 0.6802  data: 0.1678  max mem: 41794
Epoch: [59]  [120/312]  eta: 0:02:41  lr: 0.003808  min_lr: 0.003808  loss: 3.9183 (3.7394)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1810  max mem: 41794
Epoch: [59]  [130/312]  eta: 0:02:33  lr: 0.003808  min_lr: 0.003808  loss: 3.5435 (3.7185)  weight_decay: 0.0500 (0.0500)  time: 0.8546  data: 0.3487  max mem: 41794
Epoch: [59]  [140/312]  eta: 0:02:21  lr: 0.003807  min_lr: 0.003807  loss: 3.2967 (3.7023)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.1683  max mem: 41794
Epoch: [59]  [150/312]  eta: 0:02:13  lr: 0.003807  min_lr: 0.003807  loss: 3.8808 (3.7189)  weight_decay: 0.0500 (0.0500)  time: 0.7149  data: 0.1928  max mem: 41794
Epoch: [59]  [160/312]  eta: 0:02:05  lr: 0.003807  min_lr: 0.003807  loss: 3.8808 (3.7007)  weight_decay: 0.0500 (0.0500)  time: 0.8599  data: 0.3691  max mem: 41794
Epoch: [59]  [170/312]  eta: 0:01:54  lr: 0.003806  min_lr: 0.003806  loss: 3.5475 (3.7041)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1769  max mem: 41794
Epoch: [59]  [180/312]  eta: 0:01:47  lr: 0.003806  min_lr: 0.003806  loss: 3.6192 (3.7026)  weight_decay: 0.0500 (0.0500)  time: 0.7120  data: 0.2074  max mem: 41794
Epoch: [59]  [190/312]  eta: 0:01:37  lr: 0.003806  min_lr: 0.003806  loss: 3.6192 (3.7017)  weight_decay: 0.0500 (0.0500)  time: 0.7049  data: 0.2074  max mem: 41794
Epoch: [59]  [200/312]  eta: 0:01:29  lr: 0.003805  min_lr: 0.003805  loss: 3.9575 (3.7087)  weight_decay: 0.0500 (0.0500)  time: 0.6589  data: 0.1669  max mem: 41794
Epoch: [59]  [210/312]  eta: 0:01:21  lr: 0.003805  min_lr: 0.003805  loss: 3.9575 (3.7075)  weight_decay: 0.0500 (0.0500)  time: 0.8464  data: 0.3501  max mem: 41794
Epoch: [59]  [220/312]  eta: 0:01:12  lr: 0.003805  min_lr: 0.003805  loss: 3.7098 (3.6884)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.1860  max mem: 41794
Epoch: [59]  [230/312]  eta: 0:01:04  lr: 0.003805  min_lr: 0.003805  loss: 3.3664 (3.6890)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.1907  max mem: 41794
Epoch: [59]  [240/312]  eta: 0:00:57  lr: 0.003804  min_lr: 0.003804  loss: 3.8660 (3.6987)  weight_decay: 0.0500 (0.0500)  time: 0.8740  data: 0.3722  max mem: 41794
Epoch: [59]  [250/312]  eta: 0:00:48  lr: 0.003804  min_lr: 0.003804  loss: 3.9561 (3.7029)  weight_decay: 0.0500 (0.0500)  time: 0.6810  data: 0.1843  max mem: 41794
Epoch: [59]  [260/312]  eta: 0:00:40  lr: 0.003804  min_lr: 0.003804  loss: 3.8952 (3.7068)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.1887  max mem: 41794
Epoch: [59]  [270/312]  eta: 0:00:32  lr: 0.003803  min_lr: 0.003803  loss: 3.6665 (3.6972)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.1887  max mem: 41794
Epoch: [59]  [280/312]  eta: 0:00:24  lr: 0.003803  min_lr: 0.003803  loss: 3.7996 (3.7028)  weight_decay: 0.0500 (0.0500)  time: 0.6768  data: 0.1528  max mem: 41794
Epoch: [59]  [290/312]  eta: 0:00:17  lr: 0.003803  min_lr: 0.003803  loss: 3.9081 (3.7101)  weight_decay: 0.0500 (0.0500)  time: 0.8250  data: 0.2936  max mem: 41794
Epoch: [59]  [300/312]  eta: 0:00:09  lr: 0.003802  min_lr: 0.003802  loss: 3.8896 (3.7037)  weight_decay: 0.0500 (0.0500)  time: 0.6578  data: 0.1411  max mem: 41794
Epoch: [59]  [310/312]  eta: 0:00:01  lr: 0.003802  min_lr: 0.003802  loss: 3.3428 (3.6974)  weight_decay: 0.0500 (0.0500)  time: 0.4895  data: 0.0001  max mem: 41794
Epoch: [59]  [311/312]  eta: 0:00:00  lr: 0.003802  min_lr: 0.003802  loss: 3.4783 (3.6975)  weight_decay: 0.0500 (0.0500)  time: 0.4647  data: 0.0001  max mem: 41794
Epoch: [59] Total time: 0:03:58 (0.7659 s / it)
Averaged stats: lr: 0.003802  min_lr: 0.003802  loss: 3.4783 (3.7285)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.1853 (1.1853)  acc1: 75.6510 (75.6510)  acc5: 93.4896 (93.4896)  time: 8.7542  data: 8.5420  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5582 (1.5014)  acc1: 68.4896 (68.1920)  acc5: 89.8438 (89.1040)  time: 1.1209  data: 0.9492  max mem: 41794
Test: Total time: 0:00:10 (1.1369 s / it)
* Acc@1 68.568 Acc@5 89.264 loss 1.497
Accuracy of the model on the 50000 test images: 68.6%
Max accuracy: 68.57%
Epoch: [60]  [  0/312]  eta: 1:20:08  lr: 0.003802  min_lr: 0.003802  loss: 3.6807 (3.6807)  weight_decay: 0.0500 (0.0500)  time: 15.4115  data: 14.8031  max mem: 41794
Epoch: [60]  [ 10/312]  eta: 0:10:24  lr: 0.003802  min_lr: 0.003802  loss: 3.6807 (3.7180)  weight_decay: 0.0500 (0.0500)  time: 2.0682  data: 1.3462  max mem: 41794
Epoch: [60]  [ 20/312]  eta: 0:07:04  lr: 0.003801  min_lr: 0.003801  loss: 3.5641 (3.6346)  weight_decay: 0.0500 (0.0500)  time: 0.7568  data: 0.1393  max mem: 41794
Epoch: [60]  [ 30/312]  eta: 0:05:23  lr: 0.003801  min_lr: 0.003801  loss: 3.7023 (3.6470)  weight_decay: 0.0500 (0.0500)  time: 0.6399  data: 0.1396  max mem: 41794
Epoch: [60]  [ 40/312]  eta: 0:04:44  lr: 0.003801  min_lr: 0.003801  loss: 3.6141 (3.6254)  weight_decay: 0.0500 (0.0500)  time: 0.6131  data: 0.1161  max mem: 41794
Epoch: [60]  [ 50/312]  eta: 0:04:24  lr: 0.003800  min_lr: 0.003800  loss: 3.7638 (3.6624)  weight_decay: 0.0500 (0.0500)  time: 0.7949  data: 0.2557  max mem: 41794
Epoch: [60]  [ 60/312]  eta: 0:03:52  lr: 0.003800  min_lr: 0.003800  loss: 3.8876 (3.6813)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.1404  max mem: 41794
Epoch: [60]  [ 70/312]  eta: 0:03:42  lr: 0.003800  min_lr: 0.003800  loss: 3.7645 (3.6921)  weight_decay: 0.0500 (0.0500)  time: 0.6880  data: 0.1582  max mem: 41794
Epoch: [60]  [ 80/312]  eta: 0:03:29  lr: 0.003799  min_lr: 0.003799  loss: 3.8580 (3.7154)  weight_decay: 0.0500 (0.0500)  time: 0.8358  data: 0.2958  max mem: 41794
Epoch: [60]  [ 90/312]  eta: 0:03:12  lr: 0.003799  min_lr: 0.003799  loss: 3.8580 (3.7193)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1732  max mem: 41794
Epoch: [60]  [100/312]  eta: 0:03:04  lr: 0.003799  min_lr: 0.003799  loss: 3.6191 (3.7036)  weight_decay: 0.0500 (0.0500)  time: 0.7537  data: 0.2262  max mem: 41794
Epoch: [60]  [110/312]  eta: 0:02:48  lr: 0.003799  min_lr: 0.003799  loss: 3.6191 (3.6967)  weight_decay: 0.0500 (0.0500)  time: 0.6917  data: 0.1913  max mem: 41794
Epoch: [60]  [120/312]  eta: 0:02:41  lr: 0.003798  min_lr: 0.003798  loss: 3.8902 (3.6951)  weight_decay: 0.0500 (0.0500)  time: 0.6981  data: 0.1769  max mem: 41794
Epoch: [60]  [130/312]  eta: 0:02:33  lr: 0.003798  min_lr: 0.003798  loss: 3.7359 (3.6703)  weight_decay: 0.0500 (0.0500)  time: 0.8982  data: 0.3398  max mem: 41794
Epoch: [60]  [140/312]  eta: 0:02:21  lr: 0.003798  min_lr: 0.003798  loss: 3.7359 (3.6743)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.1635  max mem: 41794
Epoch: [60]  [150/312]  eta: 0:02:13  lr: 0.003797  min_lr: 0.003797  loss: 3.9325 (3.6884)  weight_decay: 0.0500 (0.0500)  time: 0.6662  data: 0.1604  max mem: 41794
Epoch: [60]  [160/312]  eta: 0:02:04  lr: 0.003797  min_lr: 0.003797  loss: 3.9695 (3.6902)  weight_decay: 0.0500 (0.0500)  time: 0.7942  data: 0.2759  max mem: 41794
Epoch: [60]  [170/312]  eta: 0:01:54  lr: 0.003797  min_lr: 0.003797  loss: 3.9397 (3.7103)  weight_decay: 0.0500 (0.0500)  time: 0.7047  data: 0.1783  max mem: 41794
Epoch: [60]  [180/312]  eta: 0:01:47  lr: 0.003796  min_lr: 0.003796  loss: 3.8862 (3.7172)  weight_decay: 0.0500 (0.0500)  time: 0.7839  data: 0.2091  max mem: 41794
Epoch: [60]  [190/312]  eta: 0:01:37  lr: 0.003796  min_lr: 0.003796  loss: 3.7565 (3.7106)  weight_decay: 0.0500 (0.0500)  time: 0.6922  data: 0.1478  max mem: 41794
Epoch: [60]  [200/312]  eta: 0:01:29  lr: 0.003796  min_lr: 0.003796  loss: 3.7524 (3.7075)  weight_decay: 0.0500 (0.0500)  time: 0.6640  data: 0.1615  max mem: 41794
Epoch: [60]  [210/312]  eta: 0:01:21  lr: 0.003795  min_lr: 0.003795  loss: 3.7884 (3.7166)  weight_decay: 0.0500 (0.0500)  time: 0.8198  data: 0.2934  max mem: 41794
Epoch: [60]  [220/312]  eta: 0:01:12  lr: 0.003795  min_lr: 0.003795  loss: 3.6304 (3.7044)  weight_decay: 0.0500 (0.0500)  time: 0.6449  data: 0.1338  max mem: 41794
Epoch: [60]  [230/312]  eta: 0:01:04  lr: 0.003795  min_lr: 0.003795  loss: 3.5920 (3.7068)  weight_decay: 0.0500 (0.0500)  time: 0.6635  data: 0.1478  max mem: 41794
Epoch: [60]  [240/312]  eta: 0:00:56  lr: 0.003794  min_lr: 0.003794  loss: 3.5690 (3.6939)  weight_decay: 0.0500 (0.0500)  time: 0.7874  data: 0.2646  max mem: 41794
Epoch: [60]  [250/312]  eta: 0:00:48  lr: 0.003794  min_lr: 0.003794  loss: 3.7143 (3.6995)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1622  max mem: 41794
Epoch: [60]  [260/312]  eta: 0:00:40  lr: 0.003794  min_lr: 0.003794  loss: 3.8495 (3.7049)  weight_decay: 0.0500 (0.0500)  time: 0.7773  data: 0.1551  max mem: 41794
Epoch: [60]  [270/312]  eta: 0:00:32  lr: 0.003793  min_lr: 0.003793  loss: 4.0143 (3.7013)  weight_decay: 0.0500 (0.0500)  time: 0.7060  data: 0.1108  max mem: 41794
Epoch: [60]  [280/312]  eta: 0:00:24  lr: 0.003793  min_lr: 0.003793  loss: 3.6848 (3.6969)  weight_decay: 0.0500 (0.0500)  time: 0.6498  data: 0.0847  max mem: 41794
Epoch: [60]  [290/312]  eta: 0:00:17  lr: 0.003793  min_lr: 0.003793  loss: 3.7897 (3.7016)  weight_decay: 0.0500 (0.0500)  time: 0.8171  data: 0.1962  max mem: 41794
Epoch: [60]  [300/312]  eta: 0:00:09  lr: 0.003793  min_lr: 0.003793  loss: 3.8338 (3.7044)  weight_decay: 0.0500 (0.0500)  time: 0.6492  data: 0.1119  max mem: 41794
Epoch: [60]  [310/312]  eta: 0:00:01  lr: 0.003792  min_lr: 0.003792  loss: 3.8338 (3.7014)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [60]  [311/312]  eta: 0:00:00  lr: 0.003792  min_lr: 0.003792  loss: 3.8545 (3.7024)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [60] Total time: 0:03:57 (0.7615 s / it)
Averaged stats: lr: 0.003792  min_lr: 0.003792  loss: 3.8545 (3.7313)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.5426 (1.5426)  acc1: 71.8750 (71.8750)  acc5: 91.2760 (91.2760)  time: 8.7971  data: 8.5906  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8300 (1.7437)  acc1: 65.7552 (66.6240)  acc5: 86.4583 (87.7600)  time: 1.1261  data: 0.9546  max mem: 41794
Test: Total time: 0:00:10 (1.1424 s / it)
* Acc@1 66.468 Acc@5 87.930 loss 1.744
Accuracy of the model on the 50000 test images: 66.5%
Max accuracy: 68.57%
Epoch: [61]  [  0/312]  eta: 1:24:19  lr: 0.003792  min_lr: 0.003792  loss: 4.1114 (4.1114)  weight_decay: 0.0500 (0.0500)  time: 16.2179  data: 12.4466  max mem: 41794
Epoch: [61]  [ 10/312]  eta: 0:11:38  lr: 0.003792  min_lr: 0.003792  loss: 3.3917 (3.4437)  weight_decay: 0.0500 (0.0500)  time: 2.3118  data: 1.2800  max mem: 41794
Epoch: [61]  [ 20/312]  eta: 0:07:22  lr: 0.003792  min_lr: 0.003792  loss: 3.4392 (3.5723)  weight_decay: 0.0500 (0.0500)  time: 0.7790  data: 0.0889  max mem: 41794
Epoch: [61]  [ 30/312]  eta: 0:05:37  lr: 0.003791  min_lr: 0.003791  loss: 3.8189 (3.6643)  weight_decay: 0.0500 (0.0500)  time: 0.5857  data: 0.0138  max mem: 41794
Epoch: [61]  [ 40/312]  eta: 0:04:40  lr: 0.003791  min_lr: 0.003791  loss: 3.9442 (3.7261)  weight_decay: 0.0500 (0.0500)  time: 0.5258  data: 0.0069  max mem: 41794
Epoch: [61]  [ 50/312]  eta: 0:04:13  lr: 0.003791  min_lr: 0.003791  loss: 3.9870 (3.7581)  weight_decay: 0.0500 (0.0500)  time: 0.6090  data: 0.0006  max mem: 41794
Epoch: [61]  [ 60/312]  eta: 0:03:47  lr: 0.003790  min_lr: 0.003790  loss: 3.6957 (3.7126)  weight_decay: 0.0500 (0.0500)  time: 0.6329  data: 0.0335  max mem: 41794
Epoch: [61]  [ 70/312]  eta: 0:03:37  lr: 0.003790  min_lr: 0.003790  loss: 3.7685 (3.7610)  weight_decay: 0.0500 (0.0500)  time: 0.7281  data: 0.1115  max mem: 41794
Epoch: [61]  [ 80/312]  eta: 0:03:23  lr: 0.003790  min_lr: 0.003790  loss: 4.0349 (3.7931)  weight_decay: 0.0500 (0.0500)  time: 0.8058  data: 0.0786  max mem: 41794
Epoch: [61]  [ 90/312]  eta: 0:03:09  lr: 0.003789  min_lr: 0.003789  loss: 3.9331 (3.7724)  weight_decay: 0.0500 (0.0500)  time: 0.6943  data: 0.0892  max mem: 41794
Epoch: [61]  [100/312]  eta: 0:03:01  lr: 0.003789  min_lr: 0.003789  loss: 3.5891 (3.7342)  weight_decay: 0.0500 (0.0500)  time: 0.7748  data: 0.1568  max mem: 41794
Epoch: [61]  [110/312]  eta: 0:02:46  lr: 0.003789  min_lr: 0.003789  loss: 3.4975 (3.7212)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.0737  max mem: 41794
Epoch: [61]  [120/312]  eta: 0:02:39  lr: 0.003788  min_lr: 0.003788  loss: 3.7365 (3.7207)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.0786  max mem: 41794
Epoch: [61]  [130/312]  eta: 0:02:30  lr: 0.003788  min_lr: 0.003788  loss: 3.8554 (3.7122)  weight_decay: 0.0500 (0.0500)  time: 0.8211  data: 0.0919  max mem: 41794
Epoch: [61]  [140/312]  eta: 0:02:18  lr: 0.003788  min_lr: 0.003788  loss: 4.0367 (3.7272)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.0396  max mem: 41794
Epoch: [61]  [150/312]  eta: 0:02:11  lr: 0.003787  min_lr: 0.003787  loss: 4.0367 (3.7345)  weight_decay: 0.0500 (0.0500)  time: 0.7172  data: 0.0763  max mem: 41794
Epoch: [61]  [160/312]  eta: 0:02:01  lr: 0.003787  min_lr: 0.003787  loss: 3.9684 (3.7375)  weight_decay: 0.0500 (0.0500)  time: 0.7693  data: 0.0561  max mem: 41794
Epoch: [61]  [170/312]  eta: 0:01:52  lr: 0.003787  min_lr: 0.003787  loss: 3.9574 (3.7454)  weight_decay: 0.0500 (0.0500)  time: 0.6602  data: 0.0628  max mem: 41794
Epoch: [61]  [180/312]  eta: 0:01:44  lr: 0.003786  min_lr: 0.003786  loss: 4.0093 (3.7474)  weight_decay: 0.0500 (0.0500)  time: 0.7494  data: 0.1381  max mem: 41794
Epoch: [61]  [190/312]  eta: 0:01:35  lr: 0.003786  min_lr: 0.003786  loss: 3.9681 (3.7613)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.0784  max mem: 41794
Epoch: [61]  [200/312]  eta: 0:01:27  lr: 0.003786  min_lr: 0.003786  loss: 3.9400 (3.7737)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.1111  max mem: 41794
Epoch: [61]  [210/312]  eta: 0:01:20  lr: 0.003785  min_lr: 0.003785  loss: 3.9025 (3.7713)  weight_decay: 0.0500 (0.0500)  time: 0.8651  data: 0.1937  max mem: 41794
Epoch: [61]  [220/312]  eta: 0:01:11  lr: 0.003785  min_lr: 0.003785  loss: 3.8719 (3.7583)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.0904  max mem: 41794
Epoch: [61]  [230/312]  eta: 0:01:04  lr: 0.003785  min_lr: 0.003785  loss: 3.7389 (3.7550)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.1184  max mem: 41794
Epoch: [61]  [240/312]  eta: 0:00:56  lr: 0.003784  min_lr: 0.003784  loss: 3.5899 (3.7476)  weight_decay: 0.0500 (0.0500)  time: 0.8701  data: 0.1356  max mem: 41794
Epoch: [61]  [250/312]  eta: 0:00:48  lr: 0.003784  min_lr: 0.003784  loss: 3.3758 (3.7376)  weight_decay: 0.0500 (0.0500)  time: 0.7353  data: 0.1013  max mem: 41794
Epoch: [61]  [260/312]  eta: 0:00:40  lr: 0.003784  min_lr: 0.003784  loss: 3.8292 (3.7408)  weight_decay: 0.0500 (0.0500)  time: 0.7355  data: 0.1418  max mem: 41794
Epoch: [61]  [270/312]  eta: 0:00:32  lr: 0.003783  min_lr: 0.003783  loss: 3.9122 (3.7423)  weight_decay: 0.0500 (0.0500)  time: 0.6605  data: 0.0653  max mem: 41794
Epoch: [61]  [280/312]  eta: 0:00:24  lr: 0.003783  min_lr: 0.003783  loss: 3.8545 (3.7435)  weight_decay: 0.0500 (0.0500)  time: 0.6905  data: 0.0714  max mem: 41794
Epoch: [61]  [290/312]  eta: 0:00:17  lr: 0.003783  min_lr: 0.003783  loss: 3.7100 (3.7389)  weight_decay: 0.0500 (0.0500)  time: 0.9249  data: 0.1791  max mem: 41794
Epoch: [61]  [300/312]  eta: 0:00:09  lr: 0.003782  min_lr: 0.003782  loss: 3.5736 (3.7312)  weight_decay: 0.0500 (0.0500)  time: 0.7192  data: 0.1103  max mem: 41794
Epoch: [61]  [310/312]  eta: 0:00:01  lr: 0.003782  min_lr: 0.003782  loss: 3.7244 (3.7357)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [61]  [311/312]  eta: 0:00:00  lr: 0.003782  min_lr: 0.003782  loss: 3.8459 (3.7364)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [61] Total time: 0:03:58 (0.7658 s / it)
Averaged stats: lr: 0.003782  min_lr: 0.003782  loss: 3.8459 (3.7389)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.2816 (1.2816)  acc1: 74.8698 (74.8698)  acc5: 92.8385 (92.8385)  time: 8.9618  data: 8.7498  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6102 (1.5134)  acc1: 67.1875 (68.0960)  acc5: 88.8021 (89.0400)  time: 1.1439  data: 0.9723  max mem: 41794
Test: Total time: 0:00:10 (1.1618 s / it)
* Acc@1 68.244 Acc@5 89.002 loss 1.516
Accuracy of the model on the 50000 test images: 68.2%
Max accuracy: 68.57%
Epoch: [62]  [  0/312]  eta: 1:22:03  lr: 0.003782  min_lr: 0.003782  loss: 4.1135 (4.1135)  weight_decay: 0.0500 (0.0500)  time: 15.7799  data: 14.0089  max mem: 41794
Epoch: [62]  [ 10/312]  eta: 0:11:21  lr: 0.003782  min_lr: 0.003782  loss: 3.9442 (3.7445)  weight_decay: 0.0500 (0.0500)  time: 2.2561  data: 1.2813  max mem: 41794
Epoch: [62]  [ 20/312]  eta: 0:07:20  lr: 0.003781  min_lr: 0.003781  loss: 3.8790 (3.7510)  weight_decay: 0.0500 (0.0500)  time: 0.7945  data: 0.0067  max mem: 41794
Epoch: [62]  [ 30/312]  eta: 0:05:35  lr: 0.003781  min_lr: 0.003781  loss: 3.5768 (3.6529)  weight_decay: 0.0500 (0.0500)  time: 0.6020  data: 0.0049  max mem: 41794
Epoch: [62]  [ 40/312]  eta: 0:04:39  lr: 0.003781  min_lr: 0.003781  loss: 3.5050 (3.6308)  weight_decay: 0.0500 (0.0500)  time: 0.5226  data: 0.0099  max mem: 41794
Epoch: [62]  [ 50/312]  eta: 0:04:16  lr: 0.003780  min_lr: 0.003780  loss: 3.5166 (3.5746)  weight_decay: 0.0500 (0.0500)  time: 0.6583  data: 0.1567  max mem: 41794
Epoch: [62]  [ 60/312]  eta: 0:03:47  lr: 0.003780  min_lr: 0.003780  loss: 3.6246 (3.6203)  weight_decay: 0.0500 (0.0500)  time: 0.6442  data: 0.1512  max mem: 41794
Epoch: [62]  [ 70/312]  eta: 0:03:37  lr: 0.003780  min_lr: 0.003780  loss: 3.9444 (3.6514)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1870  max mem: 41794
Epoch: [62]  [ 80/312]  eta: 0:03:27  lr: 0.003779  min_lr: 0.003779  loss: 4.0513 (3.7160)  weight_decay: 0.0500 (0.0500)  time: 0.8775  data: 0.3789  max mem: 41794
Epoch: [62]  [ 90/312]  eta: 0:03:09  lr: 0.003779  min_lr: 0.003779  loss: 4.0066 (3.7098)  weight_decay: 0.0500 (0.0500)  time: 0.6956  data: 0.1986  max mem: 41794
Epoch: [62]  [100/312]  eta: 0:03:03  lr: 0.003779  min_lr: 0.003779  loss: 3.8959 (3.7257)  weight_decay: 0.0500 (0.0500)  time: 0.7331  data: 0.2250  max mem: 41794
Epoch: [62]  [110/312]  eta: 0:02:47  lr: 0.003778  min_lr: 0.003778  loss: 3.9232 (3.7387)  weight_decay: 0.0500 (0.0500)  time: 0.7204  data: 0.2206  max mem: 41794
Epoch: [62]  [120/312]  eta: 0:02:40  lr: 0.003778  min_lr: 0.003778  loss: 3.9582 (3.7375)  weight_decay: 0.0500 (0.0500)  time: 0.7099  data: 0.1658  max mem: 41794
Epoch: [62]  [130/312]  eta: 0:02:32  lr: 0.003778  min_lr: 0.003778  loss: 3.9549 (3.7421)  weight_decay: 0.0500 (0.0500)  time: 0.8843  data: 0.2940  max mem: 41794
Epoch: [62]  [140/312]  eta: 0:02:20  lr: 0.003777  min_lr: 0.003777  loss: 4.0409 (3.7566)  weight_decay: 0.0500 (0.0500)  time: 0.6969  data: 0.1525  max mem: 41794
Epoch: [62]  [150/312]  eta: 0:02:12  lr: 0.003777  min_lr: 0.003777  loss: 3.9875 (3.7561)  weight_decay: 0.0500 (0.0500)  time: 0.7071  data: 0.1252  max mem: 41794
Epoch: [62]  [160/312]  eta: 0:02:04  lr: 0.003777  min_lr: 0.003777  loss: 3.7376 (3.7446)  weight_decay: 0.0500 (0.0500)  time: 0.8192  data: 0.2016  max mem: 41794
Epoch: [62]  [170/312]  eta: 0:01:54  lr: 0.003776  min_lr: 0.003776  loss: 3.7335 (3.7347)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.1318  max mem: 41794
Epoch: [62]  [180/312]  eta: 0:01:46  lr: 0.003776  min_lr: 0.003776  loss: 3.8202 (3.7371)  weight_decay: 0.0500 (0.0500)  time: 0.7451  data: 0.1299  max mem: 41794
Epoch: [62]  [190/312]  eta: 0:01:37  lr: 0.003776  min_lr: 0.003776  loss: 3.7917 (3.7416)  weight_decay: 0.0500 (0.0500)  time: 0.7133  data: 0.1213  max mem: 41794
Epoch: [62]  [200/312]  eta: 0:01:29  lr: 0.003775  min_lr: 0.003775  loss: 3.9928 (3.7479)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.1282  max mem: 41794
Epoch: [62]  [210/312]  eta: 0:01:21  lr: 0.003775  min_lr: 0.003775  loss: 3.9528 (3.7461)  weight_decay: 0.0500 (0.0500)  time: 0.8241  data: 0.2384  max mem: 41794
Epoch: [62]  [220/312]  eta: 0:01:12  lr: 0.003775  min_lr: 0.003775  loss: 3.9035 (3.7488)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.1334  max mem: 41794
Epoch: [62]  [230/312]  eta: 0:01:04  lr: 0.003774  min_lr: 0.003774  loss: 3.9035 (3.7562)  weight_decay: 0.0500 (0.0500)  time: 0.6851  data: 0.1377  max mem: 41794
Epoch: [62]  [240/312]  eta: 0:00:56  lr: 0.003774  min_lr: 0.003774  loss: 3.9156 (3.7497)  weight_decay: 0.0500 (0.0500)  time: 0.8061  data: 0.1955  max mem: 41794
Epoch: [62]  [250/312]  eta: 0:00:48  lr: 0.003774  min_lr: 0.003774  loss: 3.7341 (3.7508)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.1046  max mem: 41794
Epoch: [62]  [260/312]  eta: 0:00:40  lr: 0.003774  min_lr: 0.003774  loss: 3.7749 (3.7526)  weight_decay: 0.0500 (0.0500)  time: 0.7546  data: 0.1564  max mem: 41794
Epoch: [62]  [270/312]  eta: 0:00:32  lr: 0.003773  min_lr: 0.003773  loss: 3.7973 (3.7552)  weight_decay: 0.0500 (0.0500)  time: 0.7090  data: 0.1116  max mem: 41794
Epoch: [62]  [280/312]  eta: 0:00:24  lr: 0.003773  min_lr: 0.003773  loss: 3.7836 (3.7566)  weight_decay: 0.0500 (0.0500)  time: 0.6926  data: 0.0709  max mem: 41794
Epoch: [62]  [290/312]  eta: 0:00:17  lr: 0.003773  min_lr: 0.003773  loss: 3.8222 (3.7570)  weight_decay: 0.0500 (0.0500)  time: 0.8676  data: 0.1290  max mem: 41794
Epoch: [62]  [300/312]  eta: 0:00:09  lr: 0.003772  min_lr: 0.003772  loss: 3.8222 (3.7499)  weight_decay: 0.0500 (0.0500)  time: 0.6588  data: 0.0600  max mem: 41794
Epoch: [62]  [310/312]  eta: 0:00:01  lr: 0.003772  min_lr: 0.003772  loss: 3.5468 (3.7435)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [62]  [311/312]  eta: 0:00:00  lr: 0.003772  min_lr: 0.003772  loss: 3.5468 (3.7407)  weight_decay: 0.0500 (0.0500)  time: 0.4643  data: 0.0001  max mem: 41794
Epoch: [62] Total time: 0:03:58 (0.7654 s / it)
Averaged stats: lr: 0.003772  min_lr: 0.003772  loss: 3.5468 (3.7042)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.1520 (1.1520)  acc1: 76.4323 (76.4323)  acc5: 93.2292 (93.2292)  time: 8.0385  data: 7.8386  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6432 (1.4832)  acc1: 66.2760 (67.9520)  acc5: 88.5417 (89.1040)  time: 1.0532  data: 0.8711  max mem: 41794
Test: Total time: 0:00:09 (1.0652 s / it)
* Acc@1 68.422 Acc@5 89.196 loss 1.491
Accuracy of the model on the 50000 test images: 68.4%
Max accuracy: 68.57%
Epoch: [63]  [  0/312]  eta: 1:23:59  lr: 0.003772  min_lr: 0.003772  loss: 2.9693 (2.9693)  weight_decay: 0.0500 (0.0500)  time: 16.1509  data: 15.0895  max mem: 41794
Epoch: [63]  [ 10/312]  eta: 0:11:08  lr: 0.003771  min_lr: 0.003771  loss: 3.7648 (3.5667)  weight_decay: 0.0500 (0.0500)  time: 2.2124  data: 1.3726  max mem: 41794
Epoch: [63]  [ 20/312]  eta: 0:07:18  lr: 0.003771  min_lr: 0.003771  loss: 3.7648 (3.6107)  weight_decay: 0.0500 (0.0500)  time: 0.7685  data: 0.0859  max mem: 41794
Epoch: [63]  [ 30/312]  eta: 0:05:34  lr: 0.003771  min_lr: 0.003771  loss: 3.7442 (3.6485)  weight_decay: 0.0500 (0.0500)  time: 0.6245  data: 0.0899  max mem: 41794
Epoch: [63]  [ 40/312]  eta: 0:04:44  lr: 0.003770  min_lr: 0.003770  loss: 3.8397 (3.6799)  weight_decay: 0.0500 (0.0500)  time: 0.5673  data: 0.0145  max mem: 41794
Epoch: [63]  [ 50/312]  eta: 0:04:22  lr: 0.003770  min_lr: 0.003770  loss: 4.0357 (3.7179)  weight_decay: 0.0500 (0.0500)  time: 0.7169  data: 0.1055  max mem: 41794
Epoch: [63]  [ 60/312]  eta: 0:03:51  lr: 0.003770  min_lr: 0.003770  loss: 3.6565 (3.6524)  weight_decay: 0.0500 (0.0500)  time: 0.6616  data: 0.0959  max mem: 41794
Epoch: [63]  [ 70/312]  eta: 0:03:44  lr: 0.003769  min_lr: 0.003769  loss: 3.6565 (3.7002)  weight_decay: 0.0500 (0.0500)  time: 0.7313  data: 0.1524  max mem: 41794
Epoch: [63]  [ 80/312]  eta: 0:03:30  lr: 0.003769  min_lr: 0.003769  loss: 3.9475 (3.7042)  weight_decay: 0.0500 (0.0500)  time: 0.8642  data: 0.1918  max mem: 41794
Epoch: [63]  [ 90/312]  eta: 0:03:14  lr: 0.003769  min_lr: 0.003769  loss: 3.8835 (3.7106)  weight_decay: 0.0500 (0.0500)  time: 0.6985  data: 0.1097  max mem: 41794
Epoch: [63]  [100/312]  eta: 0:03:05  lr: 0.003768  min_lr: 0.003768  loss: 3.7642 (3.6981)  weight_decay: 0.0500 (0.0500)  time: 0.7588  data: 0.1788  max mem: 41794
Epoch: [63]  [110/312]  eta: 0:02:50  lr: 0.003768  min_lr: 0.003768  loss: 3.9518 (3.7220)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1090  max mem: 41794
Epoch: [63]  [120/312]  eta: 0:02:42  lr: 0.003768  min_lr: 0.003768  loss: 3.8881 (3.6980)  weight_decay: 0.0500 (0.0500)  time: 0.6823  data: 0.0798  max mem: 41794
Epoch: [63]  [130/312]  eta: 0:02:33  lr: 0.003767  min_lr: 0.003767  loss: 3.5145 (3.6944)  weight_decay: 0.0500 (0.0500)  time: 0.8615  data: 0.1651  max mem: 41794
Epoch: [63]  [140/312]  eta: 0:02:21  lr: 0.003767  min_lr: 0.003767  loss: 3.6597 (3.6793)  weight_decay: 0.0500 (0.0500)  time: 0.6758  data: 0.0860  max mem: 41794
Epoch: [63]  [150/312]  eta: 0:02:13  lr: 0.003767  min_lr: 0.003767  loss: 3.6879 (3.6757)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.0862  max mem: 41794
Epoch: [63]  [160/312]  eta: 0:02:04  lr: 0.003766  min_lr: 0.003766  loss: 3.6741 (3.6664)  weight_decay: 0.0500 (0.0500)  time: 0.7947  data: 0.1167  max mem: 41794
Epoch: [63]  [170/312]  eta: 0:01:54  lr: 0.003766  min_lr: 0.003766  loss: 3.6020 (3.6564)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.0819  max mem: 41794
Epoch: [63]  [180/312]  eta: 0:01:46  lr: 0.003766  min_lr: 0.003766  loss: 3.8613 (3.6627)  weight_decay: 0.0500 (0.0500)  time: 0.7390  data: 0.1504  max mem: 41794
Epoch: [63]  [190/312]  eta: 0:01:36  lr: 0.003765  min_lr: 0.003765  loss: 3.8143 (3.6621)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1000  max mem: 41794
Epoch: [63]  [200/312]  eta: 0:01:28  lr: 0.003765  min_lr: 0.003765  loss: 3.7545 (3.6570)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.0892  max mem: 41794
Epoch: [63]  [210/312]  eta: 0:01:21  lr: 0.003765  min_lr: 0.003765  loss: 3.8084 (3.6686)  weight_decay: 0.0500 (0.0500)  time: 0.8369  data: 0.1815  max mem: 41794
Epoch: [63]  [220/312]  eta: 0:01:12  lr: 0.003764  min_lr: 0.003764  loss: 3.8084 (3.6668)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.0933  max mem: 41794
Epoch: [63]  [230/312]  eta: 0:01:04  lr: 0.003764  min_lr: 0.003764  loss: 4.0205 (3.6800)  weight_decay: 0.0500 (0.0500)  time: 0.6628  data: 0.0999  max mem: 41794
Epoch: [63]  [240/312]  eta: 0:00:56  lr: 0.003764  min_lr: 0.003764  loss: 4.0221 (3.6893)  weight_decay: 0.0500 (0.0500)  time: 0.8492  data: 0.1968  max mem: 41794
Epoch: [63]  [250/312]  eta: 0:00:48  lr: 0.003763  min_lr: 0.003763  loss: 3.6184 (3.6876)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.1097  max mem: 41794
Epoch: [63]  [260/312]  eta: 0:00:40  lr: 0.003763  min_lr: 0.003763  loss: 3.7871 (3.6920)  weight_decay: 0.0500 (0.0500)  time: 0.7110  data: 0.1454  max mem: 41794
Epoch: [63]  [270/312]  eta: 0:00:32  lr: 0.003763  min_lr: 0.003763  loss: 3.7871 (3.6947)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.1333  max mem: 41794
Epoch: [63]  [280/312]  eta: 0:00:24  lr: 0.003762  min_lr: 0.003762  loss: 3.7425 (3.7017)  weight_decay: 0.0500 (0.0500)  time: 0.6886  data: 0.1100  max mem: 41794
Epoch: [63]  [290/312]  eta: 0:00:17  lr: 0.003762  min_lr: 0.003762  loss: 3.7021 (3.6951)  weight_decay: 0.0500 (0.0500)  time: 0.8369  data: 0.2124  max mem: 41794
Epoch: [63]  [300/312]  eta: 0:00:09  lr: 0.003762  min_lr: 0.003762  loss: 3.7109 (3.6997)  weight_decay: 0.0500 (0.0500)  time: 0.6295  data: 0.1028  max mem: 41794
Epoch: [63]  [310/312]  eta: 0:00:01  lr: 0.003761  min_lr: 0.003761  loss: 3.9114 (3.6993)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [63]  [311/312]  eta: 0:00:00  lr: 0.003761  min_lr: 0.003761  loss: 3.8730 (3.6986)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0001  max mem: 41794
Epoch: [63] Total time: 0:03:57 (0.7620 s / it)
Averaged stats: lr: 0.003761  min_lr: 0.003761  loss: 3.8730 (3.7134)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.2718 (1.2718)  acc1: 74.2188 (74.2188)  acc5: 93.6198 (93.6198)  time: 8.9435  data: 8.7305  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7667 (1.6145)  acc1: 67.7083 (67.7440)  acc5: 88.6719 (89.0240)  time: 1.1417  data: 0.9701  max mem: 41794
Test: Total time: 0:00:10 (1.1813 s / it)
* Acc@1 68.440 Acc@5 89.064 loss 1.622
Accuracy of the model on the 50000 test images: 68.4%
Max accuracy: 68.57%
Epoch: [64]  [  0/312]  eta: 1:20:42  lr: 0.003761  min_lr: 0.003761  loss: 3.5333 (3.5333)  weight_decay: 0.0500 (0.0500)  time: 15.5193  data: 12.3855  max mem: 41794
Epoch: [64]  [ 10/312]  eta: 0:11:13  lr: 0.003761  min_lr: 0.003761  loss: 3.6079 (3.6302)  weight_decay: 0.0500 (0.0500)  time: 2.2289  data: 1.2990  max mem: 41794
Epoch: [64]  [ 20/312]  eta: 0:07:19  lr: 0.003761  min_lr: 0.003761  loss: 3.6201 (3.5727)  weight_decay: 0.0500 (0.0500)  time: 0.8048  data: 0.1804  max mem: 41794
Epoch: [64]  [ 30/312]  eta: 0:05:33  lr: 0.003760  min_lr: 0.003760  loss: 3.6499 (3.5986)  weight_decay: 0.0500 (0.0500)  time: 0.6059  data: 0.0872  max mem: 41794
Epoch: [64]  [ 40/312]  eta: 0:04:46  lr: 0.003760  min_lr: 0.003760  loss: 3.6393 (3.6092)  weight_decay: 0.0500 (0.0500)  time: 0.5760  data: 0.0713  max mem: 41794
Epoch: [64]  [ 50/312]  eta: 0:04:24  lr: 0.003760  min_lr: 0.003760  loss: 3.8060 (3.6514)  weight_decay: 0.0500 (0.0500)  time: 0.7424  data: 0.2117  max mem: 41794
Epoch: [64]  [ 60/312]  eta: 0:03:53  lr: 0.003759  min_lr: 0.003759  loss: 3.9104 (3.6520)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.1427  max mem: 41794
Epoch: [64]  [ 70/312]  eta: 0:03:41  lr: 0.003759  min_lr: 0.003759  loss: 3.9531 (3.7025)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.1782  max mem: 41794
Epoch: [64]  [ 80/312]  eta: 0:03:33  lr: 0.003759  min_lr: 0.003759  loss: 4.0845 (3.7258)  weight_decay: 0.0500 (0.0500)  time: 0.8974  data: 0.4016  max mem: 41794
Epoch: [64]  [ 90/312]  eta: 0:03:13  lr: 0.003758  min_lr: 0.003758  loss: 3.8148 (3.7106)  weight_decay: 0.0500 (0.0500)  time: 0.7164  data: 0.2240  max mem: 41794
Epoch: [64]  [100/312]  eta: 0:03:05  lr: 0.003758  min_lr: 0.003758  loss: 3.8154 (3.7213)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1914  max mem: 41794
Epoch: [64]  [110/312]  eta: 0:02:49  lr: 0.003757  min_lr: 0.003757  loss: 3.8732 (3.7461)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1914  max mem: 41794
Epoch: [64]  [120/312]  eta: 0:02:41  lr: 0.003757  min_lr: 0.003757  loss: 3.7516 (3.7353)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.1858  max mem: 41794
Epoch: [64]  [130/312]  eta: 0:02:33  lr: 0.003757  min_lr: 0.003757  loss: 3.8187 (3.7601)  weight_decay: 0.0500 (0.0500)  time: 0.8762  data: 0.3730  max mem: 41794
Epoch: [64]  [140/312]  eta: 0:02:21  lr: 0.003756  min_lr: 0.003756  loss: 3.9643 (3.7508)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.1902  max mem: 41794
Epoch: [64]  [150/312]  eta: 0:02:13  lr: 0.003756  min_lr: 0.003756  loss: 3.3970 (3.7246)  weight_decay: 0.0500 (0.0500)  time: 0.6813  data: 0.1902  max mem: 41794
Epoch: [64]  [160/312]  eta: 0:02:05  lr: 0.003756  min_lr: 0.003756  loss: 3.5138 (3.7184)  weight_decay: 0.0500 (0.0500)  time: 0.8464  data: 0.3518  max mem: 41794
Epoch: [64]  [170/312]  eta: 0:01:54  lr: 0.003755  min_lr: 0.003755  loss: 3.7110 (3.7165)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.1646  max mem: 41794
Epoch: [64]  [180/312]  eta: 0:01:46  lr: 0.003755  min_lr: 0.003755  loss: 3.6293 (3.7146)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.1558  max mem: 41794
Epoch: [64]  [190/312]  eta: 0:01:36  lr: 0.003755  min_lr: 0.003755  loss: 3.7889 (3.7195)  weight_decay: 0.0500 (0.0500)  time: 0.6542  data: 0.1558  max mem: 41794
Epoch: [64]  [200/312]  eta: 0:01:29  lr: 0.003754  min_lr: 0.003754  loss: 3.7889 (3.7114)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.1940  max mem: 41794
Epoch: [64]  [210/312]  eta: 0:01:21  lr: 0.003754  min_lr: 0.003754  loss: 3.6596 (3.7064)  weight_decay: 0.0500 (0.0500)  time: 0.8767  data: 0.3722  max mem: 41794
Epoch: [64]  [220/312]  eta: 0:01:12  lr: 0.003754  min_lr: 0.003754  loss: 3.5034 (3.6978)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1788  max mem: 41794
Epoch: [64]  [230/312]  eta: 0:01:04  lr: 0.003753  min_lr: 0.003753  loss: 3.5202 (3.6983)  weight_decay: 0.0500 (0.0500)  time: 0.6778  data: 0.1809  max mem: 41794
Epoch: [64]  [240/312]  eta: 0:00:56  lr: 0.003753  min_lr: 0.003753  loss: 3.7616 (3.6928)  weight_decay: 0.0500 (0.0500)  time: 0.8614  data: 0.3584  max mem: 41794
Epoch: [64]  [250/312]  eta: 0:00:48  lr: 0.003753  min_lr: 0.003753  loss: 3.7840 (3.6933)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.1781  max mem: 41794
Epoch: [64]  [260/312]  eta: 0:00:40  lr: 0.003752  min_lr: 0.003752  loss: 3.8135 (3.6925)  weight_decay: 0.0500 (0.0500)  time: 0.7143  data: 0.2123  max mem: 41794
Epoch: [64]  [270/312]  eta: 0:00:32  lr: 0.003752  min_lr: 0.003752  loss: 3.6747 (3.6912)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.2123  max mem: 41794
Epoch: [64]  [280/312]  eta: 0:00:24  lr: 0.003752  min_lr: 0.003752  loss: 3.9172 (3.6986)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.1693  max mem: 41794
Epoch: [64]  [290/312]  eta: 0:00:17  lr: 0.003751  min_lr: 0.003751  loss: 3.8872 (3.6908)  weight_decay: 0.0500 (0.0500)  time: 0.8133  data: 0.3077  max mem: 41794
Epoch: [64]  [300/312]  eta: 0:00:09  lr: 0.003751  min_lr: 0.003751  loss: 3.5048 (3.6894)  weight_decay: 0.0500 (0.0500)  time: 0.6349  data: 0.1388  max mem: 41794
Epoch: [64]  [310/312]  eta: 0:00:01  lr: 0.003751  min_lr: 0.003751  loss: 3.6610 (3.6887)  weight_decay: 0.0500 (0.0500)  time: 0.4669  data: 0.0001  max mem: 41794
Epoch: [64]  [311/312]  eta: 0:00:00  lr: 0.003751  min_lr: 0.003751  loss: 3.6959 (3.6904)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [64] Total time: 0:03:57 (0.7627 s / it)
Averaged stats: lr: 0.003751  min_lr: 0.003751  loss: 3.6959 (3.6951)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:09  loss: 1.5513 (1.5513)  acc1: 73.5677 (73.5677)  acc5: 92.0573 (92.0573)  time: 7.7067  data: 7.4984  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7916 (1.7455)  acc1: 65.6250 (68.0320)  acc5: 87.8906 (88.6400)  time: 1.1117  data: 0.9407  max mem: 41794
Test: Total time: 0:00:10 (1.1223 s / it)
* Acc@1 68.134 Acc@5 88.762 loss 1.742
Accuracy of the model on the 50000 test images: 68.1%
Max accuracy: 68.57%
Epoch: [65]  [  0/312]  eta: 1:12:48  lr: 0.003751  min_lr: 0.003751  loss: 4.1884 (4.1884)  weight_decay: 0.0500 (0.0500)  time: 14.0013  data: 13.5361  max mem: 41794
Epoch: [65]  [ 10/312]  eta: 0:10:22  lr: 0.003750  min_lr: 0.003750  loss: 4.0917 (3.8245)  weight_decay: 0.0500 (0.0500)  time: 2.0603  data: 1.3076  max mem: 41794
Epoch: [65]  [ 20/312]  eta: 0:07:03  lr: 0.003750  min_lr: 0.003750  loss: 3.9338 (3.7880)  weight_decay: 0.0500 (0.0500)  time: 0.8226  data: 0.1373  max mem: 41794
Epoch: [65]  [ 30/312]  eta: 0:05:23  lr: 0.003749  min_lr: 0.003749  loss: 3.7860 (3.7037)  weight_decay: 0.0500 (0.0500)  time: 0.6423  data: 0.0953  max mem: 41794
Epoch: [65]  [ 40/312]  eta: 0:04:50  lr: 0.003749  min_lr: 0.003749  loss: 3.6223 (3.6858)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1361  max mem: 41794
Epoch: [65]  [ 50/312]  eta: 0:04:24  lr: 0.003749  min_lr: 0.003749  loss: 3.6865 (3.6884)  weight_decay: 0.0500 (0.0500)  time: 0.7975  data: 0.2518  max mem: 41794
Epoch: [65]  [ 60/312]  eta: 0:03:55  lr: 0.003748  min_lr: 0.003748  loss: 3.7830 (3.7394)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.1460  max mem: 41794
Epoch: [65]  [ 70/312]  eta: 0:03:48  lr: 0.003748  min_lr: 0.003748  loss: 3.6460 (3.6813)  weight_decay: 0.0500 (0.0500)  time: 0.7841  data: 0.2901  max mem: 41794
Epoch: [65]  [ 80/312]  eta: 0:03:34  lr: 0.003748  min_lr: 0.003748  loss: 3.1932 (3.6450)  weight_decay: 0.0500 (0.0500)  time: 0.9006  data: 0.3949  max mem: 41794
Epoch: [65]  [ 90/312]  eta: 0:03:16  lr: 0.003747  min_lr: 0.003747  loss: 3.5859 (3.6738)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1841  max mem: 41794
Epoch: [65]  [100/312]  eta: 0:03:09  lr: 0.003747  min_lr: 0.003747  loss: 3.9103 (3.6804)  weight_decay: 0.0500 (0.0500)  time: 0.7590  data: 0.2675  max mem: 41794
Epoch: [65]  [110/312]  eta: 0:02:53  lr: 0.003747  min_lr: 0.003747  loss: 3.8616 (3.6946)  weight_decay: 0.0500 (0.0500)  time: 0.7204  data: 0.2197  max mem: 41794
Epoch: [65]  [120/312]  eta: 0:02:44  lr: 0.003746  min_lr: 0.003746  loss: 3.7448 (3.6943)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1796  max mem: 41794
Epoch: [65]  [130/312]  eta: 0:02:36  lr: 0.003746  min_lr: 0.003746  loss: 3.5928 (3.6816)  weight_decay: 0.0500 (0.0500)  time: 0.8603  data: 0.3451  max mem: 41794
Epoch: [65]  [140/312]  eta: 0:02:23  lr: 0.003746  min_lr: 0.003746  loss: 3.7884 (3.6909)  weight_decay: 0.0500 (0.0500)  time: 0.7004  data: 0.1846  max mem: 41794
Epoch: [65]  [150/312]  eta: 0:02:14  lr: 0.003745  min_lr: 0.003745  loss: 3.9227 (3.7051)  weight_decay: 0.0500 (0.0500)  time: 0.6738  data: 0.1783  max mem: 41794
Epoch: [65]  [160/312]  eta: 0:02:06  lr: 0.003745  min_lr: 0.003745  loss: 3.7932 (3.6935)  weight_decay: 0.0500 (0.0500)  time: 0.8103  data: 0.3124  max mem: 41794
Epoch: [65]  [170/312]  eta: 0:01:56  lr: 0.003745  min_lr: 0.003745  loss: 3.5559 (3.6871)  weight_decay: 0.0500 (0.0500)  time: 0.7251  data: 0.2000  max mem: 41794
Epoch: [65]  [180/312]  eta: 0:01:49  lr: 0.003744  min_lr: 0.003744  loss: 3.7526 (3.6964)  weight_decay: 0.0500 (0.0500)  time: 0.7888  data: 0.2080  max mem: 41794
Epoch: [65]  [190/312]  eta: 0:01:38  lr: 0.003744  min_lr: 0.003744  loss: 3.9190 (3.7001)  weight_decay: 0.0500 (0.0500)  time: 0.7145  data: 0.1599  max mem: 41794
Epoch: [65]  [200/312]  eta: 0:01:30  lr: 0.003744  min_lr: 0.003744  loss: 3.9190 (3.7071)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1474  max mem: 41794
Epoch: [65]  [210/312]  eta: 0:01:22  lr: 0.003743  min_lr: 0.003743  loss: 3.7733 (3.7105)  weight_decay: 0.0500 (0.0500)  time: 0.8070  data: 0.2993  max mem: 41794
Epoch: [65]  [220/312]  eta: 0:01:13  lr: 0.003743  min_lr: 0.003743  loss: 3.5460 (3.7000)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.1687  max mem: 41794
Epoch: [65]  [230/312]  eta: 0:01:05  lr: 0.003742  min_lr: 0.003742  loss: 3.7115 (3.7051)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.1831  max mem: 41794
Epoch: [65]  [240/312]  eta: 0:00:57  lr: 0.003742  min_lr: 0.003742  loss: 3.8987 (3.7078)  weight_decay: 0.0500 (0.0500)  time: 0.7971  data: 0.2983  max mem: 41794
Epoch: [65]  [250/312]  eta: 0:00:48  lr: 0.003742  min_lr: 0.003742  loss: 3.8951 (3.7053)  weight_decay: 0.0500 (0.0500)  time: 0.6794  data: 0.1823  max mem: 41794
Epoch: [65]  [260/312]  eta: 0:00:41  lr: 0.003741  min_lr: 0.003741  loss: 4.0045 (3.7178)  weight_decay: 0.0500 (0.0500)  time: 0.7425  data: 0.2421  max mem: 41794
Epoch: [65]  [270/312]  eta: 0:00:32  lr: 0.003741  min_lr: 0.003741  loss: 3.8062 (3.7165)  weight_decay: 0.0500 (0.0500)  time: 0.6957  data: 0.1917  max mem: 41794
Epoch: [65]  [280/312]  eta: 0:00:25  lr: 0.003741  min_lr: 0.003741  loss: 3.4974 (3.7090)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.1747  max mem: 41794
Epoch: [65]  [290/312]  eta: 0:00:17  lr: 0.003740  min_lr: 0.003740  loss: 3.9183 (3.7189)  weight_decay: 0.0500 (0.0500)  time: 0.7948  data: 0.2826  max mem: 41794
Epoch: [65]  [300/312]  eta: 0:00:09  lr: 0.003740  min_lr: 0.003740  loss: 3.9365 (3.7177)  weight_decay: 0.0500 (0.0500)  time: 0.5999  data: 0.1084  max mem: 41794
Epoch: [65]  [310/312]  eta: 0:00:01  lr: 0.003740  min_lr: 0.003740  loss: 3.8408 (3.7252)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [65]  [311/312]  eta: 0:00:00  lr: 0.003740  min_lr: 0.003740  loss: 3.8408 (3.7223)  weight_decay: 0.0500 (0.0500)  time: 0.4650  data: 0.0001  max mem: 41794
Epoch: [65] Total time: 0:03:59 (0.7673 s / it)
Averaged stats: lr: 0.003740  min_lr: 0.003740  loss: 3.8408 (3.7098)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3968 (1.3968)  acc1: 76.9531 (76.9531)  acc5: 92.8385 (92.8385)  time: 8.5300  data: 8.3196  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7905 (1.7102)  acc1: 65.8854 (67.1680)  acc5: 87.7604 (88.1600)  time: 1.0962  data: 0.9245  max mem: 41794
Test: Total time: 0:00:10 (1.1298 s / it)
* Acc@1 67.330 Acc@5 88.468 loss 1.704
Accuracy of the model on the 50000 test images: 67.3%
Max accuracy: 68.57%
Epoch: [66]  [  0/312]  eta: 1:21:06  lr: 0.003740  min_lr: 0.003740  loss: 4.0192 (4.0192)  weight_decay: 0.0500 (0.0500)  time: 15.5966  data: 12.2818  max mem: 41794
Epoch: [66]  [ 10/312]  eta: 0:12:29  lr: 0.003739  min_lr: 0.003739  loss: 4.0192 (3.8855)  weight_decay: 0.0500 (0.0500)  time: 2.4814  data: 1.3489  max mem: 41794
Epoch: [66]  [ 20/312]  eta: 0:07:36  lr: 0.003739  min_lr: 0.003739  loss: 3.9081 (3.8869)  weight_decay: 0.0500 (0.0500)  time: 0.8600  data: 0.1401  max mem: 41794
Epoch: [66]  [ 30/312]  eta: 0:05:44  lr: 0.003738  min_lr: 0.003738  loss: 3.9081 (3.8692)  weight_decay: 0.0500 (0.0500)  time: 0.5274  data: 0.0144  max mem: 41794
Epoch: [66]  [ 40/312]  eta: 0:04:44  lr: 0.003738  min_lr: 0.003738  loss: 3.8342 (3.8035)  weight_decay: 0.0500 (0.0500)  time: 0.5080  data: 0.0048  max mem: 41794
Epoch: [66]  [ 50/312]  eta: 0:04:07  lr: 0.003738  min_lr: 0.003738  loss: 3.7219 (3.7527)  weight_decay: 0.0500 (0.0500)  time: 0.5148  data: 0.0069  max mem: 41794
Epoch: [66]  [ 60/312]  eta: 0:03:39  lr: 0.003737  min_lr: 0.003737  loss: 3.8375 (3.7778)  weight_decay: 0.0500 (0.0500)  time: 0.5126  data: 0.0060  max mem: 41794
Epoch: [66]  [ 70/312]  eta: 0:03:21  lr: 0.003737  min_lr: 0.003737  loss: 3.9855 (3.7758)  weight_decay: 0.0500 (0.0500)  time: 0.5488  data: 0.0304  max mem: 41794
Epoch: [66]  [ 80/312]  eta: 0:03:13  lr: 0.003737  min_lr: 0.003737  loss: 3.6152 (3.7507)  weight_decay: 0.0500 (0.0500)  time: 0.7160  data: 0.0636  max mem: 41794
Epoch: [66]  [ 90/312]  eta: 0:03:04  lr: 0.003736  min_lr: 0.003736  loss: 3.7684 (3.7647)  weight_decay: 0.0500 (0.0500)  time: 0.8245  data: 0.0742  max mem: 41794
Epoch: [66]  [100/312]  eta: 0:02:51  lr: 0.003736  min_lr: 0.003736  loss: 3.7793 (3.7566)  weight_decay: 0.0500 (0.0500)  time: 0.7120  data: 0.0628  max mem: 41794
Epoch: [66]  [110/312]  eta: 0:02:43  lr: 0.003736  min_lr: 0.003736  loss: 3.7888 (3.7557)  weight_decay: 0.0500 (0.0500)  time: 0.7204  data: 0.0707  max mem: 41794
Epoch: [66]  [120/312]  eta: 0:02:33  lr: 0.003735  min_lr: 0.003735  loss: 3.7888 (3.7484)  weight_decay: 0.0500 (0.0500)  time: 0.7596  data: 0.0680  max mem: 41794
Epoch: [66]  [130/312]  eta: 0:02:24  lr: 0.003735  min_lr: 0.003735  loss: 3.7206 (3.7389)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.0492  max mem: 41794
Epoch: [66]  [140/312]  eta: 0:02:16  lr: 0.003735  min_lr: 0.003735  loss: 3.7033 (3.7375)  weight_decay: 0.0500 (0.0500)  time: 0.7671  data: 0.0511  max mem: 41794
Epoch: [66]  [150/312]  eta: 0:02:07  lr: 0.003734  min_lr: 0.003734  loss: 3.6673 (3.7269)  weight_decay: 0.0500 (0.0500)  time: 0.7279  data: 0.0411  max mem: 41794
Epoch: [66]  [160/312]  eta: 0:01:58  lr: 0.003734  min_lr: 0.003734  loss: 3.7639 (3.7331)  weight_decay: 0.0500 (0.0500)  time: 0.6802  data: 0.0246  max mem: 41794
Epoch: [66]  [170/312]  eta: 0:01:50  lr: 0.003733  min_lr: 0.003733  loss: 3.8304 (3.7337)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.0139  max mem: 41794
Epoch: [66]  [180/312]  eta: 0:01:42  lr: 0.003733  min_lr: 0.003733  loss: 3.8222 (3.7337)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.0492  max mem: 41794
Epoch: [66]  [190/312]  eta: 0:01:34  lr: 0.003733  min_lr: 0.003733  loss: 3.9881 (3.7371)  weight_decay: 0.0500 (0.0500)  time: 0.7347  data: 0.0891  max mem: 41794
Epoch: [66]  [200/312]  eta: 0:01:26  lr: 0.003732  min_lr: 0.003732  loss: 4.0610 (3.7381)  weight_decay: 0.0500 (0.0500)  time: 0.7504  data: 0.0764  max mem: 41794
Epoch: [66]  [210/312]  eta: 0:01:18  lr: 0.003732  min_lr: 0.003732  loss: 3.3300 (3.7156)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.0689  max mem: 41794
Epoch: [66]  [220/312]  eta: 0:01:11  lr: 0.003732  min_lr: 0.003732  loss: 3.8763 (3.7248)  weight_decay: 0.0500 (0.0500)  time: 0.7747  data: 0.0970  max mem: 41794
Epoch: [66]  [230/312]  eta: 0:01:02  lr: 0.003731  min_lr: 0.003731  loss: 3.9449 (3.7306)  weight_decay: 0.0500 (0.0500)  time: 0.7375  data: 0.1094  max mem: 41794
Epoch: [66]  [240/312]  eta: 0:00:54  lr: 0.003731  min_lr: 0.003731  loss: 3.9018 (3.7287)  weight_decay: 0.0500 (0.0500)  time: 0.6582  data: 0.0862  max mem: 41794
Epoch: [66]  [250/312]  eta: 0:00:47  lr: 0.003731  min_lr: 0.003731  loss: 3.9803 (3.7378)  weight_decay: 0.0500 (0.0500)  time: 0.7481  data: 0.0954  max mem: 41794
Epoch: [66]  [260/312]  eta: 0:00:39  lr: 0.003730  min_lr: 0.003730  loss: 3.8174 (3.7273)  weight_decay: 0.0500 (0.0500)  time: 0.7258  data: 0.0982  max mem: 41794
Epoch: [66]  [270/312]  eta: 0:00:31  lr: 0.003730  min_lr: 0.003730  loss: 3.6374 (3.7208)  weight_decay: 0.0500 (0.0500)  time: 0.7428  data: 0.1002  max mem: 41794
Epoch: [66]  [280/312]  eta: 0:00:24  lr: 0.003730  min_lr: 0.003730  loss: 3.6374 (3.7144)  weight_decay: 0.0500 (0.0500)  time: 0.7187  data: 0.0814  max mem: 41794
Epoch: [66]  [290/312]  eta: 0:00:16  lr: 0.003729  min_lr: 0.003729  loss: 3.7431 (3.7133)  weight_decay: 0.0500 (0.0500)  time: 0.6886  data: 0.0825  max mem: 41794
Epoch: [66]  [300/312]  eta: 0:00:09  lr: 0.003729  min_lr: 0.003729  loss: 3.7789 (3.7174)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.1012  max mem: 41794
Epoch: [66]  [310/312]  eta: 0:00:01  lr: 0.003728  min_lr: 0.003728  loss: 3.7460 (3.7166)  weight_decay: 0.0500 (0.0500)  time: 0.5668  data: 0.0367  max mem: 41794
Epoch: [66]  [311/312]  eta: 0:00:00  lr: 0.003728  min_lr: 0.003728  loss: 3.7789 (3.7178)  weight_decay: 0.0500 (0.0500)  time: 0.5649  data: 0.0367  max mem: 41794
Epoch: [66] Total time: 0:03:54 (0.7510 s / it)
Averaged stats: lr: 0.003728  min_lr: 0.003728  loss: 3.7789 (3.6972)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.4189 (1.4189)  acc1: 73.1771 (73.1771)  acc5: 91.0156 (91.0156)  time: 8.7867  data: 8.5741  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6343 (1.5778)  acc1: 66.7969 (67.5200)  acc5: 88.1510 (87.6320)  time: 1.1250  data: 0.9528  max mem: 41794
Test: Total time: 0:00:10 (1.1533 s / it)
* Acc@1 66.674 Acc@5 88.042 loss 1.573
Accuracy of the model on the 50000 test images: 66.7%
Max accuracy: 68.57%
Epoch: [67]  [  0/312]  eta: 1:26:32  lr: 0.003728  min_lr: 0.003728  loss: 3.3890 (3.3890)  weight_decay: 0.0500 (0.0500)  time: 16.6418  data: 12.8390  max mem: 41794
Epoch: [67]  [ 10/312]  eta: 0:11:21  lr: 0.003728  min_lr: 0.003728  loss: 3.4808 (3.5992)  weight_decay: 0.0500 (0.0500)  time: 2.2561  data: 1.4361  max mem: 41794
Epoch: [67]  [ 20/312]  eta: 0:07:45  lr: 0.003728  min_lr: 0.003728  loss: 3.7645 (3.6822)  weight_decay: 0.0500 (0.0500)  time: 0.8406  data: 0.2229  max mem: 41794
Epoch: [67]  [ 30/312]  eta: 0:05:53  lr: 0.003727  min_lr: 0.003727  loss: 3.8566 (3.6212)  weight_decay: 0.0500 (0.0500)  time: 0.7040  data: 0.0899  max mem: 41794
Epoch: [67]  [ 40/312]  eta: 0:04:52  lr: 0.003727  min_lr: 0.003727  loss: 3.3185 (3.5636)  weight_decay: 0.0500 (0.0500)  time: 0.5354  data: 0.0225  max mem: 41794
Epoch: [67]  [ 50/312]  eta: 0:04:37  lr: 0.003727  min_lr: 0.003727  loss: 3.7671 (3.6352)  weight_decay: 0.0500 (0.0500)  time: 0.7581  data: 0.0987  max mem: 41794
Epoch: [67]  [ 60/312]  eta: 0:04:04  lr: 0.003726  min_lr: 0.003726  loss: 3.8793 (3.6256)  weight_decay: 0.0500 (0.0500)  time: 0.7453  data: 0.0947  max mem: 41794
Epoch: [67]  [ 70/312]  eta: 0:03:51  lr: 0.003726  min_lr: 0.003726  loss: 3.8441 (3.6238)  weight_decay: 0.0500 (0.0500)  time: 0.6878  data: 0.0786  max mem: 41794
Epoch: [67]  [ 80/312]  eta: 0:03:34  lr: 0.003725  min_lr: 0.003725  loss: 3.6311 (3.6245)  weight_decay: 0.0500 (0.0500)  time: 0.7984  data: 0.0754  max mem: 41794
Epoch: [67]  [ 90/312]  eta: 0:03:18  lr: 0.003725  min_lr: 0.003725  loss: 3.6454 (3.6315)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.0736  max mem: 41794
Epoch: [67]  [100/312]  eta: 0:03:08  lr: 0.003725  min_lr: 0.003725  loss: 3.8443 (3.6452)  weight_decay: 0.0500 (0.0500)  time: 0.7270  data: 0.1493  max mem: 41794
Epoch: [67]  [110/312]  eta: 0:02:51  lr: 0.003724  min_lr: 0.003724  loss: 3.7952 (3.6371)  weight_decay: 0.0500 (0.0500)  time: 0.6526  data: 0.0763  max mem: 41794
Epoch: [67]  [120/312]  eta: 0:02:45  lr: 0.003724  min_lr: 0.003724  loss: 3.6794 (3.6163)  weight_decay: 0.0500 (0.0500)  time: 0.7232  data: 0.1076  max mem: 41794
Epoch: [67]  [130/312]  eta: 0:02:36  lr: 0.003724  min_lr: 0.003724  loss: 3.7015 (3.6152)  weight_decay: 0.0500 (0.0500)  time: 0.8951  data: 0.1877  max mem: 41794
Epoch: [67]  [140/312]  eta: 0:02:23  lr: 0.003723  min_lr: 0.003723  loss: 3.8101 (3.6279)  weight_decay: 0.0500 (0.0500)  time: 0.6658  data: 0.0807  max mem: 41794
Epoch: [67]  [150/312]  eta: 0:02:15  lr: 0.003723  min_lr: 0.003723  loss: 3.7795 (3.6270)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.1028  max mem: 41794
Epoch: [67]  [160/312]  eta: 0:02:05  lr: 0.003723  min_lr: 0.003723  loss: 3.7667 (3.6359)  weight_decay: 0.0500 (0.0500)  time: 0.7591  data: 0.1027  max mem: 41794
Epoch: [67]  [170/312]  eta: 0:01:55  lr: 0.003722  min_lr: 0.003722  loss: 3.7667 (3.6280)  weight_decay: 0.0500 (0.0500)  time: 0.6515  data: 0.0893  max mem: 41794
Epoch: [67]  [180/312]  eta: 0:01:47  lr: 0.003722  min_lr: 0.003722  loss: 3.4399 (3.6221)  weight_decay: 0.0500 (0.0500)  time: 0.7751  data: 0.1964  max mem: 41794
Epoch: [67]  [190/312]  eta: 0:01:37  lr: 0.003721  min_lr: 0.003721  loss: 3.8193 (3.6272)  weight_decay: 0.0500 (0.0500)  time: 0.6935  data: 0.1078  max mem: 41794
Epoch: [67]  [200/312]  eta: 0:01:30  lr: 0.003721  min_lr: 0.003721  loss: 3.9062 (3.6285)  weight_decay: 0.0500 (0.0500)  time: 0.6776  data: 0.0942  max mem: 41794
Epoch: [67]  [210/312]  eta: 0:01:22  lr: 0.003721  min_lr: 0.003721  loss: 3.6952 (3.6233)  weight_decay: 0.0500 (0.0500)  time: 0.8304  data: 0.1544  max mem: 41794
Epoch: [67]  [220/312]  eta: 0:01:12  lr: 0.003720  min_lr: 0.003720  loss: 3.4342 (3.6133)  weight_decay: 0.0500 (0.0500)  time: 0.6515  data: 0.0609  max mem: 41794
Epoch: [67]  [230/312]  eta: 0:01:05  lr: 0.003720  min_lr: 0.003720  loss: 3.5189 (3.6152)  weight_decay: 0.0500 (0.0500)  time: 0.7106  data: 0.0759  max mem: 41794
Epoch: [67]  [240/312]  eta: 0:00:56  lr: 0.003720  min_lr: 0.003720  loss: 3.6669 (3.6175)  weight_decay: 0.0500 (0.0500)  time: 0.7977  data: 0.0760  max mem: 41794
Epoch: [67]  [250/312]  eta: 0:00:48  lr: 0.003719  min_lr: 0.003719  loss: 3.9366 (3.6279)  weight_decay: 0.0500 (0.0500)  time: 0.6502  data: 0.0678  max mem: 41794
Epoch: [67]  [260/312]  eta: 0:00:40  lr: 0.003719  min_lr: 0.003719  loss: 3.9153 (3.6299)  weight_decay: 0.0500 (0.0500)  time: 0.7488  data: 0.1398  max mem: 41794
Epoch: [67]  [270/312]  eta: 0:00:32  lr: 0.003719  min_lr: 0.003719  loss: 3.7672 (3.6294)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0726  max mem: 41794
Epoch: [67]  [280/312]  eta: 0:00:24  lr: 0.003718  min_lr: 0.003718  loss: 3.6554 (3.6285)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.0767  max mem: 41794
Epoch: [67]  [290/312]  eta: 0:00:17  lr: 0.003718  min_lr: 0.003718  loss: 3.9561 (3.6413)  weight_decay: 0.0500 (0.0500)  time: 0.8594  data: 0.1339  max mem: 41794
Epoch: [67]  [300/312]  eta: 0:00:09  lr: 0.003717  min_lr: 0.003717  loss: 4.0418 (3.6474)  weight_decay: 0.0500 (0.0500)  time: 0.6702  data: 0.0576  max mem: 41794
Epoch: [67]  [310/312]  eta: 0:00:01  lr: 0.003717  min_lr: 0.003717  loss: 3.9673 (3.6516)  weight_decay: 0.0500 (0.0500)  time: 0.4711  data: 0.0001  max mem: 41794
Epoch: [67]  [311/312]  eta: 0:00:00  lr: 0.003717  min_lr: 0.003717  loss: 3.9673 (3.6542)  weight_decay: 0.0500 (0.0500)  time: 0.4686  data: 0.0001  max mem: 41794
Epoch: [67] Total time: 0:03:59 (0.7678 s / it)
Averaged stats: lr: 0.003717  min_lr: 0.003717  loss: 3.9673 (3.6932)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.4162 (1.4162)  acc1: 75.1302 (75.1302)  acc5: 93.2292 (93.2292)  time: 8.4746  data: 8.2743  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7764 (1.6885)  acc1: 67.8385 (67.2640)  acc5: 88.0208 (88.3200)  time: 1.0895  data: 0.9194  max mem: 41794
Test: Total time: 0:00:09 (1.1023 s / it)
* Acc@1 67.472 Acc@5 88.672 loss 1.677
Accuracy of the model on the 50000 test images: 67.5%
Max accuracy: 68.57%
Epoch: [68]  [  0/312]  eta: 1:23:13  lr: 0.003717  min_lr: 0.003717  loss: 3.8039 (3.8039)  weight_decay: 0.0500 (0.0500)  time: 16.0050  data: 14.0933  max mem: 41794
Epoch: [68]  [ 10/312]  eta: 0:12:19  lr: 0.003717  min_lr: 0.003717  loss: 3.8437 (3.6527)  weight_decay: 0.0500 (0.0500)  time: 2.4479  data: 1.2822  max mem: 41794
Epoch: [68]  [ 20/312]  eta: 0:07:34  lr: 0.003716  min_lr: 0.003716  loss: 3.8437 (3.6493)  weight_decay: 0.0500 (0.0500)  time: 0.8348  data: 0.0015  max mem: 41794
Epoch: [68]  [ 30/312]  eta: 0:05:44  lr: 0.003716  min_lr: 0.003716  loss: 3.8123 (3.6910)  weight_decay: 0.0500 (0.0500)  time: 0.5485  data: 0.0030  max mem: 41794
Epoch: [68]  [ 40/312]  eta: 0:04:45  lr: 0.003715  min_lr: 0.003715  loss: 3.8123 (3.6790)  weight_decay: 0.0500 (0.0500)  time: 0.5180  data: 0.0056  max mem: 41794
Epoch: [68]  [ 50/312]  eta: 0:04:08  lr: 0.003715  min_lr: 0.003715  loss: 3.8407 (3.6777)  weight_decay: 0.0500 (0.0500)  time: 0.5198  data: 0.0040  max mem: 41794
Epoch: [68]  [ 60/312]  eta: 0:03:40  lr: 0.003715  min_lr: 0.003715  loss: 3.8407 (3.6915)  weight_decay: 0.0500 (0.0500)  time: 0.5164  data: 0.0007  max mem: 41794
Epoch: [68]  [ 70/312]  eta: 0:03:25  lr: 0.003714  min_lr: 0.003714  loss: 3.8553 (3.7106)  weight_decay: 0.0500 (0.0500)  time: 0.5960  data: 0.0886  max mem: 41794
Epoch: [68]  [ 80/312]  eta: 0:03:16  lr: 0.003714  min_lr: 0.003714  loss: 3.5903 (3.6696)  weight_decay: 0.0500 (0.0500)  time: 0.7641  data: 0.2563  max mem: 41794
Epoch: [68]  [ 90/312]  eta: 0:03:01  lr: 0.003714  min_lr: 0.003714  loss: 3.4887 (3.6412)  weight_decay: 0.0500 (0.0500)  time: 0.7122  data: 0.1938  max mem: 41794
Epoch: [68]  [100/312]  eta: 0:02:53  lr: 0.003713  min_lr: 0.003713  loss: 3.0391 (3.5904)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.2016  max mem: 41794
Epoch: [68]  [110/312]  eta: 0:02:42  lr: 0.003713  min_lr: 0.003713  loss: 3.2922 (3.6095)  weight_decay: 0.0500 (0.0500)  time: 0.7441  data: 0.2224  max mem: 41794
Epoch: [68]  [120/312]  eta: 0:02:33  lr: 0.003713  min_lr: 0.003713  loss: 3.8746 (3.6214)  weight_decay: 0.0500 (0.0500)  time: 0.7123  data: 0.1512  max mem: 41794
Epoch: [68]  [130/312]  eta: 0:02:25  lr: 0.003712  min_lr: 0.003712  loss: 3.7794 (3.6295)  weight_decay: 0.0500 (0.0500)  time: 0.7840  data: 0.2198  max mem: 41794
Epoch: [68]  [140/312]  eta: 0:02:16  lr: 0.003712  min_lr: 0.003712  loss: 3.8154 (3.6455)  weight_decay: 0.0500 (0.0500)  time: 0.7649  data: 0.1449  max mem: 41794
Epoch: [68]  [150/312]  eta: 0:02:08  lr: 0.003711  min_lr: 0.003711  loss: 3.9800 (3.6634)  weight_decay: 0.0500 (0.0500)  time: 0.7540  data: 0.0960  max mem: 41794
Epoch: [68]  [160/312]  eta: 0:02:00  lr: 0.003711  min_lr: 0.003711  loss: 3.9243 (3.6693)  weight_decay: 0.0500 (0.0500)  time: 0.7773  data: 0.1203  max mem: 41794
Epoch: [68]  [170/312]  eta: 0:01:52  lr: 0.003711  min_lr: 0.003711  loss: 3.6174 (3.6572)  weight_decay: 0.0500 (0.0500)  time: 0.7497  data: 0.0672  max mem: 41794
Epoch: [68]  [180/312]  eta: 0:01:43  lr: 0.003710  min_lr: 0.003710  loss: 3.5315 (3.6639)  weight_decay: 0.0500 (0.0500)  time: 0.7358  data: 0.0800  max mem: 41794
Epoch: [68]  [190/312]  eta: 0:01:35  lr: 0.003710  min_lr: 0.003710  loss: 3.4958 (3.6622)  weight_decay: 0.0500 (0.0500)  time: 0.6960  data: 0.0786  max mem: 41794
Epoch: [68]  [200/312]  eta: 0:01:26  lr: 0.003710  min_lr: 0.003710  loss: 3.4958 (3.6573)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.0506  max mem: 41794
Epoch: [68]  [210/312]  eta: 0:01:19  lr: 0.003709  min_lr: 0.003709  loss: 3.7356 (3.6569)  weight_decay: 0.0500 (0.0500)  time: 0.7872  data: 0.0941  max mem: 41794
Epoch: [68]  [220/312]  eta: 0:01:11  lr: 0.003709  min_lr: 0.003709  loss: 3.7356 (3.6538)  weight_decay: 0.0500 (0.0500)  time: 0.7546  data: 0.0584  max mem: 41794
Epoch: [68]  [230/312]  eta: 0:01:03  lr: 0.003708  min_lr: 0.003708  loss: 3.4867 (3.6501)  weight_decay: 0.0500 (0.0500)  time: 0.6578  data: 0.0792  max mem: 41794
Epoch: [68]  [240/312]  eta: 0:00:55  lr: 0.003708  min_lr: 0.003708  loss: 3.4867 (3.6506)  weight_decay: 0.0500 (0.0500)  time: 0.7850  data: 0.1105  max mem: 41794
Epoch: [68]  [250/312]  eta: 0:00:47  lr: 0.003708  min_lr: 0.003708  loss: 3.6267 (3.6403)  weight_decay: 0.0500 (0.0500)  time: 0.7895  data: 0.0550  max mem: 41794
Epoch: [68]  [260/312]  eta: 0:00:39  lr: 0.003707  min_lr: 0.003707  loss: 3.7399 (3.6464)  weight_decay: 0.0500 (0.0500)  time: 0.6646  data: 0.0690  max mem: 41794
Epoch: [68]  [270/312]  eta: 0:00:32  lr: 0.003707  min_lr: 0.003707  loss: 3.9130 (3.6499)  weight_decay: 0.0500 (0.0500)  time: 0.7153  data: 0.0488  max mem: 41794
Epoch: [68]  [280/312]  eta: 0:00:24  lr: 0.003707  min_lr: 0.003707  loss: 3.6737 (3.6427)  weight_decay: 0.0500 (0.0500)  time: 0.7157  data: 0.0547  max mem: 41794
Epoch: [68]  [290/312]  eta: 0:00:16  lr: 0.003706  min_lr: 0.003706  loss: 3.3045 (3.6383)  weight_decay: 0.0500 (0.0500)  time: 0.7680  data: 0.1301  max mem: 41794
Epoch: [68]  [300/312]  eta: 0:00:09  lr: 0.003706  min_lr: 0.003706  loss: 3.6710 (3.6342)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.0758  max mem: 41794
Epoch: [68]  [310/312]  eta: 0:00:01  lr: 0.003705  min_lr: 0.003705  loss: 3.8185 (3.6410)  weight_decay: 0.0500 (0.0500)  time: 0.4931  data: 0.0001  max mem: 41794
Epoch: [68]  [311/312]  eta: 0:00:00  lr: 0.003705  min_lr: 0.003705  loss: 3.8185 (3.6423)  weight_decay: 0.0500 (0.0500)  time: 0.4916  data: 0.0001  max mem: 41794
Epoch: [68] Total time: 0:03:55 (0.7561 s / it)
Averaged stats: lr: 0.003705  min_lr: 0.003705  loss: 3.8185 (3.7010)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.4827 (1.4827)  acc1: 72.1354 (72.1354)  acc5: 91.5365 (91.5365)  time: 8.5830  data: 8.3736  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8029 (1.7244)  acc1: 68.8802 (66.1920)  acc5: 86.8490 (87.6960)  time: 1.1017  data: 0.9305  max mem: 41794
Test: Total time: 0:00:10 (1.1425 s / it)
* Acc@1 66.222 Acc@5 87.646 loss 1.723
Accuracy of the model on the 50000 test images: 66.2%
Max accuracy: 68.57%
Epoch: [69]  [  0/312]  eta: 1:20:16  lr: 0.003705  min_lr: 0.003705  loss: 2.9475 (2.9475)  weight_decay: 0.0500 (0.0500)  time: 15.4359  data: 11.5262  max mem: 41794
Epoch: [69]  [ 10/312]  eta: 0:10:42  lr: 0.003705  min_lr: 0.003705  loss: 2.9494 (3.3273)  weight_decay: 0.0500 (0.0500)  time: 2.1285  data: 1.3129  max mem: 41794
Epoch: [69]  [ 20/312]  eta: 0:07:34  lr: 0.003705  min_lr: 0.003705  loss: 3.2433 (3.4557)  weight_decay: 0.0500 (0.0500)  time: 0.8635  data: 0.2269  max mem: 41794
Epoch: [69]  [ 30/312]  eta: 0:05:47  lr: 0.003704  min_lr: 0.003704  loss: 3.2823 (3.4290)  weight_decay: 0.0500 (0.0500)  time: 0.7404  data: 0.0863  max mem: 41794
Epoch: [69]  [ 40/312]  eta: 0:04:49  lr: 0.003704  min_lr: 0.003704  loss: 3.7257 (3.5396)  weight_decay: 0.0500 (0.0500)  time: 0.5449  data: 0.0172  max mem: 41794
Epoch: [69]  [ 50/312]  eta: 0:04:30  lr: 0.003703  min_lr: 0.003703  loss: 3.8660 (3.5767)  weight_decay: 0.0500 (0.0500)  time: 0.7196  data: 0.1453  max mem: 41794
Epoch: [69]  [ 60/312]  eta: 0:03:57  lr: 0.003703  min_lr: 0.003703  loss: 3.8380 (3.5899)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.1336  max mem: 41794
Epoch: [69]  [ 70/312]  eta: 0:03:47  lr: 0.003703  min_lr: 0.003703  loss: 3.5348 (3.5341)  weight_decay: 0.0500 (0.0500)  time: 0.7109  data: 0.1420  max mem: 41794
Epoch: [69]  [ 80/312]  eta: 0:03:34  lr: 0.003702  min_lr: 0.003702  loss: 3.1134 (3.5298)  weight_decay: 0.0500 (0.0500)  time: 0.8608  data: 0.2090  max mem: 41794
Epoch: [69]  [ 90/312]  eta: 0:03:17  lr: 0.003702  min_lr: 0.003702  loss: 3.3449 (3.5160)  weight_decay: 0.0500 (0.0500)  time: 0.7092  data: 0.1407  max mem: 41794
Epoch: [69]  [100/312]  eta: 0:03:07  lr: 0.003702  min_lr: 0.003702  loss: 3.7286 (3.5463)  weight_decay: 0.0500 (0.0500)  time: 0.7306  data: 0.1416  max mem: 41794
Epoch: [69]  [110/312]  eta: 0:02:51  lr: 0.003701  min_lr: 0.003701  loss: 3.9462 (3.5685)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0686  max mem: 41794
Epoch: [69]  [120/312]  eta: 0:02:42  lr: 0.003701  min_lr: 0.003701  loss: 3.9589 (3.5864)  weight_decay: 0.0500 (0.0500)  time: 0.6510  data: 0.0931  max mem: 41794
Epoch: [69]  [130/312]  eta: 0:02:32  lr: 0.003700  min_lr: 0.003700  loss: 3.7125 (3.5840)  weight_decay: 0.0500 (0.0500)  time: 0.7838  data: 0.1766  max mem: 41794
Epoch: [69]  [140/312]  eta: 0:02:21  lr: 0.003700  min_lr: 0.003700  loss: 3.4062 (3.5870)  weight_decay: 0.0500 (0.0500)  time: 0.6721  data: 0.0844  max mem: 41794
Epoch: [69]  [150/312]  eta: 0:02:13  lr: 0.003700  min_lr: 0.003700  loss: 3.6935 (3.6029)  weight_decay: 0.0500 (0.0500)  time: 0.7021  data: 0.1657  max mem: 41794
Epoch: [69]  [160/312]  eta: 0:02:04  lr: 0.003699  min_lr: 0.003699  loss: 3.7075 (3.6069)  weight_decay: 0.0500 (0.0500)  time: 0.8226  data: 0.3289  max mem: 41794
Epoch: [69]  [170/312]  eta: 0:01:54  lr: 0.003699  min_lr: 0.003699  loss: 3.8724 (3.6185)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.1640  max mem: 41794
Epoch: [69]  [180/312]  eta: 0:01:46  lr: 0.003699  min_lr: 0.003699  loss: 3.8291 (3.6211)  weight_decay: 0.0500 (0.0500)  time: 0.7169  data: 0.2050  max mem: 41794
Epoch: [69]  [190/312]  eta: 0:01:36  lr: 0.003698  min_lr: 0.003698  loss: 3.7586 (3.6346)  weight_decay: 0.0500 (0.0500)  time: 0.7009  data: 0.2053  max mem: 41794
Epoch: [69]  [200/312]  eta: 0:01:28  lr: 0.003698  min_lr: 0.003698  loss: 3.7596 (3.6376)  weight_decay: 0.0500 (0.0500)  time: 0.6482  data: 0.1615  max mem: 41794
Epoch: [69]  [210/312]  eta: 0:01:21  lr: 0.003697  min_lr: 0.003697  loss: 3.7596 (3.6427)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.3520  max mem: 41794
Epoch: [69]  [220/312]  eta: 0:01:12  lr: 0.003697  min_lr: 0.003697  loss: 3.7560 (3.6389)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.1913  max mem: 41794
Epoch: [69]  [230/312]  eta: 0:01:04  lr: 0.003697  min_lr: 0.003697  loss: 3.8113 (3.6467)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.2106  max mem: 41794
Epoch: [69]  [240/312]  eta: 0:00:56  lr: 0.003696  min_lr: 0.003696  loss: 3.9601 (3.6537)  weight_decay: 0.0500 (0.0500)  time: 0.8764  data: 0.3873  max mem: 41794
Epoch: [69]  [250/312]  eta: 0:00:48  lr: 0.003696  min_lr: 0.003696  loss: 4.0197 (3.6523)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1772  max mem: 41794
Epoch: [69]  [260/312]  eta: 0:00:40  lr: 0.003696  min_lr: 0.003696  loss: 3.9480 (3.6572)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.1872  max mem: 41794
Epoch: [69]  [270/312]  eta: 0:00:32  lr: 0.003695  min_lr: 0.003695  loss: 3.6660 (3.6464)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1872  max mem: 41794
Epoch: [69]  [280/312]  eta: 0:00:24  lr: 0.003695  min_lr: 0.003695  loss: 3.8201 (3.6541)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1908  max mem: 41794
Epoch: [69]  [290/312]  eta: 0:00:17  lr: 0.003694  min_lr: 0.003694  loss: 3.9968 (3.6655)  weight_decay: 0.0500 (0.0500)  time: 0.8237  data: 0.3213  max mem: 41794
Epoch: [69]  [300/312]  eta: 0:00:09  lr: 0.003694  min_lr: 0.003694  loss: 4.0114 (3.6743)  weight_decay: 0.0500 (0.0500)  time: 0.6204  data: 0.1309  max mem: 41794
Epoch: [69]  [310/312]  eta: 0:00:01  lr: 0.003694  min_lr: 0.003694  loss: 3.8716 (3.6681)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [69]  [311/312]  eta: 0:00:00  lr: 0.003694  min_lr: 0.003694  loss: 3.7056 (3.6669)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [69] Total time: 0:03:57 (0.7612 s / it)
Averaged stats: lr: 0.003694  min_lr: 0.003694  loss: 3.7056 (3.6754)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.2388 (1.2388)  acc1: 74.3490 (74.3490)  acc5: 91.6667 (91.6667)  time: 8.8494  data: 8.6513  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6227 (1.5226)  acc1: 69.0104 (68.0800)  acc5: 89.9740 (89.1840)  time: 1.1382  data: 0.9613  max mem: 41794
Test: Total time: 0:00:10 (1.1834 s / it)
* Acc@1 68.182 Acc@5 89.194 loss 1.520
Accuracy of the model on the 50000 test images: 68.2%
Max accuracy: 68.57%
Epoch: [70]  [  0/312]  eta: 1:25:03  lr: 0.003694  min_lr: 0.003694  loss: 4.1817 (4.1817)  weight_decay: 0.0500 (0.0500)  time: 16.3585  data: 13.4781  max mem: 41794
Epoch: [70]  [ 10/312]  eta: 0:11:54  lr: 0.003693  min_lr: 0.003693  loss: 3.9811 (3.5649)  weight_decay: 0.0500 (0.0500)  time: 2.3668  data: 1.4142  max mem: 41794
Epoch: [70]  [ 20/312]  eta: 0:07:46  lr: 0.003693  min_lr: 0.003693  loss: 3.3836 (3.5554)  weight_decay: 0.0500 (0.0500)  time: 0.8589  data: 0.1545  max mem: 41794
Epoch: [70]  [ 30/312]  eta: 0:05:53  lr: 0.003692  min_lr: 0.003692  loss: 3.6715 (3.6648)  weight_decay: 0.0500 (0.0500)  time: 0.6419  data: 0.0509  max mem: 41794
Epoch: [70]  [ 40/312]  eta: 0:04:51  lr: 0.003692  min_lr: 0.003692  loss: 3.9065 (3.6909)  weight_decay: 0.0500 (0.0500)  time: 0.5194  data: 0.0006  max mem: 41794
Epoch: [70]  [ 50/312]  eta: 0:04:17  lr: 0.003692  min_lr: 0.003692  loss: 3.8091 (3.6683)  weight_decay: 0.0500 (0.0500)  time: 0.5626  data: 0.0063  max mem: 41794
Epoch: [70]  [ 60/312]  eta: 0:03:47  lr: 0.003691  min_lr: 0.003691  loss: 3.8347 (3.7157)  weight_decay: 0.0500 (0.0500)  time: 0.5597  data: 0.0114  max mem: 41794
Epoch: [70]  [ 70/312]  eta: 0:03:33  lr: 0.003691  min_lr: 0.003691  loss: 3.7780 (3.6693)  weight_decay: 0.0500 (0.0500)  time: 0.6258  data: 0.0853  max mem: 41794
Epoch: [70]  [ 80/312]  eta: 0:03:23  lr: 0.003690  min_lr: 0.003690  loss: 3.6061 (3.6937)  weight_decay: 0.0500 (0.0500)  time: 0.7988  data: 0.2256  max mem: 41794
Epoch: [70]  [ 90/312]  eta: 0:03:07  lr: 0.003690  min_lr: 0.003690  loss: 3.9361 (3.7070)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.1517  max mem: 41794
Epoch: [70]  [100/312]  eta: 0:02:58  lr: 0.003690  min_lr: 0.003690  loss: 3.7843 (3.7043)  weight_decay: 0.0500 (0.0500)  time: 0.6956  data: 0.1295  max mem: 41794
Epoch: [70]  [110/312]  eta: 0:02:44  lr: 0.003689  min_lr: 0.003689  loss: 3.7679 (3.6908)  weight_decay: 0.0500 (0.0500)  time: 0.6679  data: 0.1237  max mem: 41794
Epoch: [70]  [120/312]  eta: 0:02:36  lr: 0.003689  min_lr: 0.003689  loss: 3.6421 (3.6891)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.1443  max mem: 41794
Epoch: [70]  [130/312]  eta: 0:02:28  lr: 0.003689  min_lr: 0.003689  loss: 3.5322 (3.6762)  weight_decay: 0.0500 (0.0500)  time: 0.8494  data: 0.3138  max mem: 41794
Epoch: [70]  [140/312]  eta: 0:02:17  lr: 0.003688  min_lr: 0.003688  loss: 3.6157 (3.6888)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.1792  max mem: 41794
Epoch: [70]  [150/312]  eta: 0:02:10  lr: 0.003688  min_lr: 0.003688  loss: 3.8221 (3.6922)  weight_decay: 0.0500 (0.0500)  time: 0.7218  data: 0.1445  max mem: 41794
Epoch: [70]  [160/312]  eta: 0:02:01  lr: 0.003687  min_lr: 0.003687  loss: 3.7745 (3.6898)  weight_decay: 0.0500 (0.0500)  time: 0.8364  data: 0.2377  max mem: 41794
Epoch: [70]  [170/312]  eta: 0:01:52  lr: 0.003687  min_lr: 0.003687  loss: 4.0466 (3.7059)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1083  max mem: 41794
Epoch: [70]  [180/312]  eta: 0:01:44  lr: 0.003687  min_lr: 0.003687  loss: 4.0062 (3.7016)  weight_decay: 0.0500 (0.0500)  time: 0.7327  data: 0.1363  max mem: 41794
Epoch: [70]  [190/312]  eta: 0:01:34  lr: 0.003686  min_lr: 0.003686  loss: 3.4884 (3.6884)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.1323  max mem: 41794
Epoch: [70]  [200/312]  eta: 0:01:27  lr: 0.003686  min_lr: 0.003686  loss: 3.4851 (3.6776)  weight_decay: 0.0500 (0.0500)  time: 0.6471  data: 0.1318  max mem: 41794
Epoch: [70]  [210/312]  eta: 0:01:19  lr: 0.003685  min_lr: 0.003685  loss: 3.7535 (3.6790)  weight_decay: 0.0500 (0.0500)  time: 0.8202  data: 0.2880  max mem: 41794
Epoch: [70]  [220/312]  eta: 0:01:10  lr: 0.003685  min_lr: 0.003685  loss: 3.8291 (3.6792)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.1623  max mem: 41794
Epoch: [70]  [230/312]  eta: 0:01:03  lr: 0.003685  min_lr: 0.003685  loss: 3.8749 (3.6899)  weight_decay: 0.0500 (0.0500)  time: 0.6742  data: 0.1731  max mem: 41794
Epoch: [70]  [240/312]  eta: 0:00:55  lr: 0.003684  min_lr: 0.003684  loss: 3.8984 (3.6802)  weight_decay: 0.0500 (0.0500)  time: 0.8570  data: 0.3449  max mem: 41794
Epoch: [70]  [250/312]  eta: 0:00:47  lr: 0.003684  min_lr: 0.003684  loss: 3.3736 (3.6755)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1783  max mem: 41794
Epoch: [70]  [260/312]  eta: 0:00:40  lr: 0.003684  min_lr: 0.003684  loss: 3.3736 (3.6717)  weight_decay: 0.0500 (0.0500)  time: 0.6972  data: 0.1507  max mem: 41794
Epoch: [70]  [270/312]  eta: 0:00:31  lr: 0.003683  min_lr: 0.003683  loss: 3.7709 (3.6767)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.1489  max mem: 41794
Epoch: [70]  [280/312]  eta: 0:00:24  lr: 0.003683  min_lr: 0.003683  loss: 3.7709 (3.6686)  weight_decay: 0.0500 (0.0500)  time: 0.6635  data: 0.1273  max mem: 41794
Epoch: [70]  [290/312]  eta: 0:00:16  lr: 0.003682  min_lr: 0.003682  loss: 3.3945 (3.6656)  weight_decay: 0.0500 (0.0500)  time: 0.8698  data: 0.2496  max mem: 41794
Epoch: [70]  [300/312]  eta: 0:00:09  lr: 0.003682  min_lr: 0.003682  loss: 3.3945 (3.6617)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.1235  max mem: 41794
Epoch: [70]  [310/312]  eta: 0:00:01  lr: 0.003682  min_lr: 0.003682  loss: 3.8685 (3.6669)  weight_decay: 0.0500 (0.0500)  time: 0.4724  data: 0.0010  max mem: 41794
Epoch: [70]  [311/312]  eta: 0:00:00  lr: 0.003682  min_lr: 0.003682  loss: 3.8685 (3.6688)  weight_decay: 0.0500 (0.0500)  time: 0.4717  data: 0.0010  max mem: 41794
Epoch: [70] Total time: 0:03:55 (0.7562 s / it)
Averaged stats: lr: 0.003682  min_lr: 0.003682  loss: 3.8685 (3.6852)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1812 (1.1812)  acc1: 75.9115 (75.9115)  acc5: 94.2708 (94.2708)  time: 8.5200  data: 8.3072  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7080 (1.5641)  acc1: 67.8385 (68.4640)  acc5: 88.2812 (88.7200)  time: 1.1187  data: 0.9419  max mem: 41794
Test: Total time: 0:00:10 (1.1407 s / it)
* Acc@1 68.628 Acc@5 89.084 loss 1.553
Accuracy of the model on the 50000 test images: 68.6%
Max accuracy: 68.63%
Epoch: [71]  [  0/312]  eta: 1:20:06  lr: 0.003681  min_lr: 0.003681  loss: 2.8775 (2.8775)  weight_decay: 0.0500 (0.0500)  time: 15.4069  data: 13.0366  max mem: 41794
Epoch: [71]  [ 10/312]  eta: 0:11:01  lr: 0.003681  min_lr: 0.003681  loss: 3.4700 (3.4209)  weight_decay: 0.0500 (0.0500)  time: 2.1897  data: 1.3744  max mem: 41794
Epoch: [71]  [ 20/312]  eta: 0:07:27  lr: 0.003681  min_lr: 0.003681  loss: 3.8681 (3.5923)  weight_decay: 0.0500 (0.0500)  time: 0.8376  data: 0.1591  max mem: 41794
Epoch: [71]  [ 30/312]  eta: 0:05:41  lr: 0.003680  min_lr: 0.003680  loss: 3.9631 (3.6897)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.0555  max mem: 41794
Epoch: [71]  [ 40/312]  eta: 0:04:48  lr: 0.003680  min_lr: 0.003680  loss: 3.8551 (3.6488)  weight_decay: 0.0500 (0.0500)  time: 0.5690  data: 0.0343  max mem: 41794
Epoch: [71]  [ 50/312]  eta: 0:04:26  lr: 0.003680  min_lr: 0.003680  loss: 3.7566 (3.6448)  weight_decay: 0.0500 (0.0500)  time: 0.7210  data: 0.1944  max mem: 41794
Epoch: [71]  [ 60/312]  eta: 0:03:55  lr: 0.003679  min_lr: 0.003679  loss: 3.9356 (3.6398)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1608  max mem: 41794
Epoch: [71]  [ 70/312]  eta: 0:03:44  lr: 0.003679  min_lr: 0.003679  loss: 3.9356 (3.6653)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.1948  max mem: 41794
Epoch: [71]  [ 80/312]  eta: 0:03:31  lr: 0.003678  min_lr: 0.003678  loss: 3.7144 (3.6590)  weight_decay: 0.0500 (0.0500)  time: 0.8456  data: 0.3525  max mem: 41794
Epoch: [71]  [ 90/312]  eta: 0:03:13  lr: 0.003678  min_lr: 0.003678  loss: 3.6917 (3.6576)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.1790  max mem: 41794
Epoch: [71]  [100/312]  eta: 0:03:04  lr: 0.003678  min_lr: 0.003678  loss: 3.7957 (3.6840)  weight_decay: 0.0500 (0.0500)  time: 0.7017  data: 0.2016  max mem: 41794
Epoch: [71]  [110/312]  eta: 0:02:48  lr: 0.003677  min_lr: 0.003677  loss: 4.0039 (3.7161)  weight_decay: 0.0500 (0.0500)  time: 0.6762  data: 0.1825  max mem: 41794
Epoch: [71]  [120/312]  eta: 0:02:42  lr: 0.003677  min_lr: 0.003677  loss: 3.9809 (3.7375)  weight_decay: 0.0500 (0.0500)  time: 0.7153  data: 0.2037  max mem: 41794
Epoch: [71]  [130/312]  eta: 0:02:32  lr: 0.003676  min_lr: 0.003676  loss: 3.9826 (3.7562)  weight_decay: 0.0500 (0.0500)  time: 0.8499  data: 0.3121  max mem: 41794
Epoch: [71]  [140/312]  eta: 0:02:20  lr: 0.003676  min_lr: 0.003676  loss: 3.7294 (3.7347)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.1490  max mem: 41794
Epoch: [71]  [150/312]  eta: 0:02:13  lr: 0.003676  min_lr: 0.003676  loss: 3.4141 (3.7185)  weight_decay: 0.0500 (0.0500)  time: 0.7213  data: 0.2340  max mem: 41794
Epoch: [71]  [160/312]  eta: 0:02:03  lr: 0.003675  min_lr: 0.003675  loss: 3.5702 (3.7205)  weight_decay: 0.0500 (0.0500)  time: 0.7968  data: 0.2891  max mem: 41794
Epoch: [71]  [170/312]  eta: 0:01:54  lr: 0.003675  min_lr: 0.003675  loss: 3.8809 (3.7233)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.1851  max mem: 41794
Epoch: [71]  [180/312]  eta: 0:01:46  lr: 0.003674  min_lr: 0.003674  loss: 4.0617 (3.7409)  weight_decay: 0.0500 (0.0500)  time: 0.7648  data: 0.2747  max mem: 41794
Epoch: [71]  [190/312]  eta: 0:01:36  lr: 0.003674  min_lr: 0.003674  loss: 3.9938 (3.7450)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.1838  max mem: 41794
Epoch: [71]  [200/312]  eta: 0:01:28  lr: 0.003674  min_lr: 0.003674  loss: 3.6300 (3.7395)  weight_decay: 0.0500 (0.0500)  time: 0.6451  data: 0.1555  max mem: 41794
Epoch: [71]  [210/312]  eta: 0:01:21  lr: 0.003673  min_lr: 0.003673  loss: 3.8285 (3.7463)  weight_decay: 0.0500 (0.0500)  time: 0.8152  data: 0.3212  max mem: 41794
Epoch: [71]  [220/312]  eta: 0:01:12  lr: 0.003673  min_lr: 0.003673  loss: 3.8285 (3.7386)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.1789  max mem: 41794
Epoch: [71]  [230/312]  eta: 0:01:04  lr: 0.003672  min_lr: 0.003672  loss: 3.7444 (3.7408)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.2138  max mem: 41794
Epoch: [71]  [240/312]  eta: 0:00:56  lr: 0.003672  min_lr: 0.003672  loss: 3.8034 (3.7415)  weight_decay: 0.0500 (0.0500)  time: 0.8548  data: 0.3525  max mem: 41794
Epoch: [71]  [250/312]  eta: 0:00:48  lr: 0.003672  min_lr: 0.003672  loss: 3.8034 (3.7433)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.2038  max mem: 41794
Epoch: [71]  [260/312]  eta: 0:00:40  lr: 0.003671  min_lr: 0.003671  loss: 3.8897 (3.7367)  weight_decay: 0.0500 (0.0500)  time: 0.7156  data: 0.2250  max mem: 41794
Epoch: [71]  [270/312]  eta: 0:00:32  lr: 0.003671  min_lr: 0.003671  loss: 3.7846 (3.7398)  weight_decay: 0.0500 (0.0500)  time: 0.6661  data: 0.1730  max mem: 41794
Epoch: [71]  [280/312]  eta: 0:00:24  lr: 0.003670  min_lr: 0.003670  loss: 3.7857 (3.7399)  weight_decay: 0.0500 (0.0500)  time: 0.6585  data: 0.1732  max mem: 41794
Epoch: [71]  [290/312]  eta: 0:00:17  lr: 0.003670  min_lr: 0.003670  loss: 3.9233 (3.7426)  weight_decay: 0.0500 (0.0500)  time: 0.8271  data: 0.3345  max mem: 41794
Epoch: [71]  [300/312]  eta: 0:00:09  lr: 0.003670  min_lr: 0.003670  loss: 3.9612 (3.7473)  weight_decay: 0.0500 (0.0500)  time: 0.6475  data: 0.1617  max mem: 41794
Epoch: [71]  [310/312]  eta: 0:00:01  lr: 0.003669  min_lr: 0.003669  loss: 3.9408 (3.7514)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [71]  [311/312]  eta: 0:00:00  lr: 0.003669  min_lr: 0.003669  loss: 3.9408 (3.7540)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [71] Total time: 0:03:57 (0.7611 s / it)
Averaged stats: lr: 0.003669  min_lr: 0.003669  loss: 3.9408 (3.6955)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:09  loss: 1.4715 (1.4715)  acc1: 75.7812 (75.7812)  acc5: 91.6667 (91.6667)  time: 7.7384  data: 7.5329  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.8788 (1.7404)  acc1: 67.0573 (67.4720)  acc5: 87.3698 (88.3200)  time: 1.0337  data: 0.8608  max mem: 41794
Test: Total time: 0:00:09 (1.0442 s / it)
* Acc@1 67.712 Acc@5 88.622 loss 1.726
Accuracy of the model on the 50000 test images: 67.7%
Max accuracy: 68.63%
Epoch: [72]  [  0/312]  eta: 1:24:53  lr: 0.003669  min_lr: 0.003669  loss: 2.9991 (2.9991)  weight_decay: 0.0500 (0.0500)  time: 16.3239  data: 15.8487  max mem: 41794
Epoch: [72]  [ 10/312]  eta: 0:11:27  lr: 0.003669  min_lr: 0.003669  loss: 3.6890 (3.5421)  weight_decay: 0.0500 (0.0500)  time: 2.2757  data: 1.4446  max mem: 41794
Epoch: [72]  [ 20/312]  eta: 0:07:12  lr: 0.003668  min_lr: 0.003668  loss: 3.8990 (3.6455)  weight_decay: 0.0500 (0.0500)  time: 0.7398  data: 0.0374  max mem: 41794
Epoch: [72]  [ 30/312]  eta: 0:05:30  lr: 0.003668  min_lr: 0.003668  loss: 3.8751 (3.6053)  weight_decay: 0.0500 (0.0500)  time: 0.5649  data: 0.0376  max mem: 41794
Epoch: [72]  [ 40/312]  eta: 0:04:35  lr: 0.003668  min_lr: 0.003668  loss: 3.5697 (3.6314)  weight_decay: 0.0500 (0.0500)  time: 0.5181  data: 0.0132  max mem: 41794
Epoch: [72]  [ 50/312]  eta: 0:04:12  lr: 0.003667  min_lr: 0.003667  loss: 3.9398 (3.7169)  weight_decay: 0.0500 (0.0500)  time: 0.6408  data: 0.0610  max mem: 41794
Epoch: [72]  [ 60/312]  eta: 0:03:47  lr: 0.003667  min_lr: 0.003667  loss: 3.8874 (3.7148)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.1020  max mem: 41794
Epoch: [72]  [ 70/312]  eta: 0:03:37  lr: 0.003666  min_lr: 0.003666  loss: 3.6829 (3.6988)  weight_decay: 0.0500 (0.0500)  time: 0.7308  data: 0.1259  max mem: 41794
Epoch: [72]  [ 80/312]  eta: 0:03:23  lr: 0.003666  min_lr: 0.003666  loss: 3.4863 (3.6759)  weight_decay: 0.0500 (0.0500)  time: 0.7937  data: 0.1195  max mem: 41794
Epoch: [72]  [ 90/312]  eta: 0:03:08  lr: 0.003666  min_lr: 0.003666  loss: 3.5874 (3.6752)  weight_decay: 0.0500 (0.0500)  time: 0.6747  data: 0.0790  max mem: 41794
Epoch: [72]  [100/312]  eta: 0:03:01  lr: 0.003665  min_lr: 0.003665  loss: 3.5874 (3.6527)  weight_decay: 0.0500 (0.0500)  time: 0.7772  data: 0.1380  max mem: 41794
Epoch: [72]  [110/312]  eta: 0:02:46  lr: 0.003665  min_lr: 0.003665  loss: 3.6396 (3.6672)  weight_decay: 0.0500 (0.0500)  time: 0.7155  data: 0.1050  max mem: 41794
Epoch: [72]  [120/312]  eta: 0:02:38  lr: 0.003664  min_lr: 0.003664  loss: 3.9255 (3.6657)  weight_decay: 0.0500 (0.0500)  time: 0.6794  data: 0.0976  max mem: 41794
Epoch: [72]  [130/312]  eta: 0:02:31  lr: 0.003664  min_lr: 0.003664  loss: 3.1542 (3.6328)  weight_decay: 0.0500 (0.0500)  time: 0.8695  data: 0.2004  max mem: 41794
Epoch: [72]  [140/312]  eta: 0:02:19  lr: 0.003664  min_lr: 0.003664  loss: 3.5661 (3.6475)  weight_decay: 0.0500 (0.0500)  time: 0.6980  data: 0.1131  max mem: 41794
Epoch: [72]  [150/312]  eta: 0:02:12  lr: 0.003663  min_lr: 0.003663  loss: 3.9152 (3.6526)  weight_decay: 0.0500 (0.0500)  time: 0.7095  data: 0.0930  max mem: 41794
Epoch: [72]  [160/312]  eta: 0:02:03  lr: 0.003663  min_lr: 0.003663  loss: 3.9163 (3.6559)  weight_decay: 0.0500 (0.0500)  time: 0.8278  data: 0.0996  max mem: 41794
Epoch: [72]  [170/312]  eta: 0:01:54  lr: 0.003662  min_lr: 0.003662  loss: 3.8909 (3.6690)  weight_decay: 0.0500 (0.0500)  time: 0.7089  data: 0.0868  max mem: 41794
Epoch: [72]  [180/312]  eta: 0:01:46  lr: 0.003662  min_lr: 0.003662  loss: 3.8826 (3.6755)  weight_decay: 0.0500 (0.0500)  time: 0.7895  data: 0.1534  max mem: 41794
Epoch: [72]  [190/312]  eta: 0:01:36  lr: 0.003662  min_lr: 0.003662  loss: 3.8826 (3.6870)  weight_decay: 0.0500 (0.0500)  time: 0.7069  data: 0.0836  max mem: 41794
Epoch: [72]  [200/312]  eta: 0:01:28  lr: 0.003661  min_lr: 0.003661  loss: 3.8855 (3.6856)  weight_decay: 0.0500 (0.0500)  time: 0.6520  data: 0.0807  max mem: 41794
Epoch: [72]  [210/312]  eta: 0:01:21  lr: 0.003661  min_lr: 0.003661  loss: 3.5100 (3.6770)  weight_decay: 0.0500 (0.0500)  time: 0.8235  data: 0.1348  max mem: 41794
Epoch: [72]  [220/312]  eta: 0:01:12  lr: 0.003660  min_lr: 0.003660  loss: 3.5100 (3.6667)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.0894  max mem: 41794
Epoch: [72]  [230/312]  eta: 0:01:04  lr: 0.003660  min_lr: 0.003660  loss: 3.5022 (3.6523)  weight_decay: 0.0500 (0.0500)  time: 0.7418  data: 0.1051  max mem: 41794
Epoch: [72]  [240/312]  eta: 0:00:56  lr: 0.003660  min_lr: 0.003660  loss: 3.5022 (3.6515)  weight_decay: 0.0500 (0.0500)  time: 0.8373  data: 0.0900  max mem: 41794
Epoch: [72]  [250/312]  eta: 0:00:48  lr: 0.003659  min_lr: 0.003659  loss: 3.6576 (3.6480)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.0908  max mem: 41794
Epoch: [72]  [260/312]  eta: 0:00:40  lr: 0.003659  min_lr: 0.003659  loss: 3.8529 (3.6503)  weight_decay: 0.0500 (0.0500)  time: 0.7350  data: 0.1171  max mem: 41794
Epoch: [72]  [270/312]  eta: 0:00:32  lr: 0.003658  min_lr: 0.003658  loss: 3.7739 (3.6530)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.0465  max mem: 41794
Epoch: [72]  [280/312]  eta: 0:00:24  lr: 0.003658  min_lr: 0.003658  loss: 3.7739 (3.6580)  weight_decay: 0.0500 (0.0500)  time: 0.7215  data: 0.0658  max mem: 41794
Epoch: [72]  [290/312]  eta: 0:00:17  lr: 0.003658  min_lr: 0.003658  loss: 3.8956 (3.6616)  weight_decay: 0.0500 (0.0500)  time: 0.8360  data: 0.1016  max mem: 41794
Epoch: [72]  [300/312]  eta: 0:00:09  lr: 0.003657  min_lr: 0.003657  loss: 3.9253 (3.6642)  weight_decay: 0.0500 (0.0500)  time: 0.6009  data: 0.0362  max mem: 41794
Epoch: [72]  [310/312]  eta: 0:00:01  lr: 0.003657  min_lr: 0.003657  loss: 3.6965 (3.6556)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0001  max mem: 41794
Epoch: [72]  [311/312]  eta: 0:00:00  lr: 0.003657  min_lr: 0.003657  loss: 3.6965 (3.6554)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [72] Total time: 0:03:58 (0.7644 s / it)
Averaged stats: lr: 0.003657  min_lr: 0.003657  loss: 3.6965 (3.6652)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.2331 (1.2331)  acc1: 76.4323 (76.4323)  acc5: 92.7083 (92.7083)  time: 8.6766  data: 8.4679  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5391 (1.4502)  acc1: 67.4479 (68.8000)  acc5: 89.3229 (89.6000)  time: 1.1121  data: 0.9410  max mem: 41794
Test: Total time: 0:00:10 (1.1236 s / it)
* Acc@1 69.300 Acc@5 89.774 loss 1.438
Accuracy of the model on the 50000 test images: 69.3%
Max accuracy: 69.30%
Epoch: [73]  [  0/312]  eta: 1:23:51  lr: 0.003657  min_lr: 0.003657  loss: 4.2911 (4.2911)  weight_decay: 0.0500 (0.0500)  time: 16.1279  data: 15.5911  max mem: 41794
Epoch: [73]  [ 10/312]  eta: 0:12:45  lr: 0.003656  min_lr: 0.003656  loss: 3.9458 (3.9603)  weight_decay: 0.0500 (0.0500)  time: 2.5336  data: 1.4179  max mem: 41794
Epoch: [73]  [ 20/312]  eta: 0:07:43  lr: 0.003656  min_lr: 0.003656  loss: 3.8212 (3.7455)  weight_decay: 0.0500 (0.0500)  time: 0.8614  data: 0.0023  max mem: 41794
Epoch: [73]  [ 30/312]  eta: 0:05:51  lr: 0.003656  min_lr: 0.003656  loss: 3.9468 (3.8300)  weight_decay: 0.0500 (0.0500)  time: 0.5401  data: 0.0059  max mem: 41794
Epoch: [73]  [ 40/312]  eta: 0:04:49  lr: 0.003655  min_lr: 0.003655  loss: 4.0345 (3.8130)  weight_decay: 0.0500 (0.0500)  time: 0.5162  data: 0.0043  max mem: 41794
Epoch: [73]  [ 50/312]  eta: 0:04:10  lr: 0.003655  min_lr: 0.003655  loss: 3.5691 (3.7144)  weight_decay: 0.0500 (0.0500)  time: 0.5075  data: 0.0024  max mem: 41794
Epoch: [73]  [ 60/312]  eta: 0:03:42  lr: 0.003654  min_lr: 0.003654  loss: 3.5592 (3.7089)  weight_decay: 0.0500 (0.0500)  time: 0.5128  data: 0.0025  max mem: 41794
Epoch: [73]  [ 70/312]  eta: 0:03:24  lr: 0.003654  min_lr: 0.003654  loss: 3.6412 (3.7044)  weight_decay: 0.0500 (0.0500)  time: 0.5559  data: 0.0019  max mem: 41794
Epoch: [73]  [ 80/312]  eta: 0:03:12  lr: 0.003654  min_lr: 0.003654  loss: 3.8975 (3.7094)  weight_decay: 0.0500 (0.0500)  time: 0.6633  data: 0.0304  max mem: 41794
Epoch: [73]  [ 90/312]  eta: 0:03:00  lr: 0.003653  min_lr: 0.003653  loss: 3.7777 (3.6771)  weight_decay: 0.0500 (0.0500)  time: 0.6972  data: 0.0291  max mem: 41794
Epoch: [73]  [100/312]  eta: 0:02:52  lr: 0.003653  min_lr: 0.003653  loss: 3.2753 (3.6422)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.0336  max mem: 41794
Epoch: [73]  [110/312]  eta: 0:02:42  lr: 0.003652  min_lr: 0.003652  loss: 3.5230 (3.6318)  weight_decay: 0.0500 (0.0500)  time: 0.7739  data: 0.0606  max mem: 41794
Epoch: [73]  [120/312]  eta: 0:02:32  lr: 0.003652  min_lr: 0.003652  loss: 3.7338 (3.6431)  weight_decay: 0.0500 (0.0500)  time: 0.6984  data: 0.0303  max mem: 41794
Epoch: [73]  [130/312]  eta: 0:02:24  lr: 0.003652  min_lr: 0.003652  loss: 3.9327 (3.6393)  weight_decay: 0.0500 (0.0500)  time: 0.7259  data: 0.0957  max mem: 41794
Epoch: [73]  [140/312]  eta: 0:02:14  lr: 0.003651  min_lr: 0.003651  loss: 3.8733 (3.6588)  weight_decay: 0.0500 (0.0500)  time: 0.7162  data: 0.1170  max mem: 41794
Epoch: [73]  [150/312]  eta: 0:02:06  lr: 0.003651  min_lr: 0.003651  loss: 3.6770 (3.6494)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.0781  max mem: 41794
Epoch: [73]  [160/312]  eta: 0:01:58  lr: 0.003650  min_lr: 0.003650  loss: 3.7633 (3.6525)  weight_decay: 0.0500 (0.0500)  time: 0.7754  data: 0.1072  max mem: 41794
Epoch: [73]  [170/312]  eta: 0:01:49  lr: 0.003650  min_lr: 0.003650  loss: 3.8348 (3.6574)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.0837  max mem: 41794
Epoch: [73]  [180/312]  eta: 0:01:41  lr: 0.003649  min_lr: 0.003649  loss: 3.8791 (3.6548)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.0938  max mem: 41794
Epoch: [73]  [190/312]  eta: 0:01:34  lr: 0.003649  min_lr: 0.003649  loss: 3.8791 (3.6596)  weight_decay: 0.0500 (0.0500)  time: 0.7551  data: 0.1190  max mem: 41794
Epoch: [73]  [200/312]  eta: 0:01:25  lr: 0.003649  min_lr: 0.003649  loss: 3.7888 (3.6544)  weight_decay: 0.0500 (0.0500)  time: 0.7194  data: 0.0922  max mem: 41794
Epoch: [73]  [210/312]  eta: 0:01:18  lr: 0.003648  min_lr: 0.003648  loss: 3.6240 (3.6558)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.0706  max mem: 41794
Epoch: [73]  [220/312]  eta: 0:01:10  lr: 0.003648  min_lr: 0.003648  loss: 3.7973 (3.6575)  weight_decay: 0.0500 (0.0500)  time: 0.7705  data: 0.0553  max mem: 41794
Epoch: [73]  [230/312]  eta: 0:01:02  lr: 0.003647  min_lr: 0.003647  loss: 3.8307 (3.6603)  weight_decay: 0.0500 (0.0500)  time: 0.7654  data: 0.0434  max mem: 41794
Epoch: [73]  [240/312]  eta: 0:00:55  lr: 0.003647  min_lr: 0.003647  loss: 3.9230 (3.6708)  weight_decay: 0.0500 (0.0500)  time: 0.7592  data: 0.0764  max mem: 41794
Epoch: [73]  [250/312]  eta: 0:00:47  lr: 0.003647  min_lr: 0.003647  loss: 3.7884 (3.6632)  weight_decay: 0.0500 (0.0500)  time: 0.7880  data: 0.0702  max mem: 41794
Epoch: [73]  [260/312]  eta: 0:00:39  lr: 0.003646  min_lr: 0.003646  loss: 3.5723 (3.6599)  weight_decay: 0.0500 (0.0500)  time: 0.7485  data: 0.0511  max mem: 41794
Epoch: [73]  [270/312]  eta: 0:00:32  lr: 0.003646  min_lr: 0.003646  loss: 3.6614 (3.6578)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.0516  max mem: 41794
Epoch: [73]  [280/312]  eta: 0:00:24  lr: 0.003645  min_lr: 0.003645  loss: 3.4833 (3.6464)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.0373  max mem: 41794
Epoch: [73]  [290/312]  eta: 0:00:16  lr: 0.003645  min_lr: 0.003645  loss: 3.4145 (3.6439)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.0457  max mem: 41794
Epoch: [73]  [300/312]  eta: 0:00:09  lr: 0.003645  min_lr: 0.003645  loss: 3.6925 (3.6413)  weight_decay: 0.0500 (0.0500)  time: 0.7060  data: 0.0290  max mem: 41794
Epoch: [73]  [310/312]  eta: 0:00:01  lr: 0.003644  min_lr: 0.003644  loss: 3.9173 (3.6519)  weight_decay: 0.0500 (0.0500)  time: 0.5648  data: 0.0042  max mem: 41794
Epoch: [73]  [311/312]  eta: 0:00:00  lr: 0.003644  min_lr: 0.003644  loss: 3.8724 (3.6486)  weight_decay: 0.0500 (0.0500)  time: 0.5085  data: 0.0042  max mem: 41794
Epoch: [73] Total time: 0:03:54 (0.7528 s / it)
Averaged stats: lr: 0.003644  min_lr: 0.003644  loss: 3.8724 (3.6729)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3175 (1.3175)  acc1: 74.4792 (74.4792)  acc5: 91.7969 (91.7969)  time: 8.4701  data: 8.2631  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6637 (1.5789)  acc1: 66.2760 (67.2000)  acc5: 88.0208 (88.3680)  time: 1.0898  data: 0.9182  max mem: 41794
Test: Total time: 0:00:09 (1.1080 s / it)
* Acc@1 67.296 Acc@5 88.670 loss 1.587
Accuracy of the model on the 50000 test images: 67.3%
Max accuracy: 69.30%
Epoch: [74]  [  0/312]  eta: 1:18:38  lr: 0.003644  min_lr: 0.003644  loss: 3.4760 (3.4760)  weight_decay: 0.0500 (0.0500)  time: 15.1228  data: 12.9841  max mem: 41794
Epoch: [74]  [ 10/312]  eta: 0:11:21  lr: 0.003644  min_lr: 0.003644  loss: 3.9705 (3.6457)  weight_decay: 0.0500 (0.0500)  time: 2.2576  data: 1.2447  max mem: 41794
Epoch: [74]  [ 20/312]  eta: 0:07:44  lr: 0.003643  min_lr: 0.003643  loss: 3.9772 (3.7401)  weight_decay: 0.0500 (0.0500)  time: 0.9125  data: 0.0610  max mem: 41794
Epoch: [74]  [ 30/312]  eta: 0:05:54  lr: 0.003643  min_lr: 0.003643  loss: 3.7903 (3.6187)  weight_decay: 0.0500 (0.0500)  time: 0.7077  data: 0.0343  max mem: 41794
Epoch: [74]  [ 40/312]  eta: 0:04:52  lr: 0.003642  min_lr: 0.003642  loss: 3.6456 (3.6484)  weight_decay: 0.0500 (0.0500)  time: 0.5387  data: 0.0177  max mem: 41794
Epoch: [74]  [ 50/312]  eta: 0:04:24  lr: 0.003642  min_lr: 0.003642  loss: 3.6456 (3.6334)  weight_decay: 0.0500 (0.0500)  time: 0.6281  data: 0.0345  max mem: 41794
Epoch: [74]  [ 60/312]  eta: 0:03:54  lr: 0.003642  min_lr: 0.003642  loss: 3.7347 (3.6449)  weight_decay: 0.0500 (0.0500)  time: 0.6279  data: 0.0350  max mem: 41794
Epoch: [74]  [ 70/312]  eta: 0:03:41  lr: 0.003641  min_lr: 0.003641  loss: 3.7294 (3.6505)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.1311  max mem: 41794
Epoch: [74]  [ 80/312]  eta: 0:03:28  lr: 0.003641  min_lr: 0.003641  loss: 3.6432 (3.6276)  weight_decay: 0.0500 (0.0500)  time: 0.8064  data: 0.2676  max mem: 41794
Epoch: [74]  [ 90/312]  eta: 0:03:12  lr: 0.003640  min_lr: 0.003640  loss: 3.6432 (3.6344)  weight_decay: 0.0500 (0.0500)  time: 0.6931  data: 0.1499  max mem: 41794
Epoch: [74]  [100/312]  eta: 0:03:03  lr: 0.003640  min_lr: 0.003640  loss: 3.6791 (3.6187)  weight_decay: 0.0500 (0.0500)  time: 0.7289  data: 0.1697  max mem: 41794
Epoch: [74]  [110/312]  eta: 0:02:47  lr: 0.003640  min_lr: 0.003640  loss: 3.8295 (3.6419)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.1662  max mem: 41794
Epoch: [74]  [120/312]  eta: 0:02:41  lr: 0.003639  min_lr: 0.003639  loss: 3.9362 (3.6369)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.1458  max mem: 41794
Epoch: [74]  [130/312]  eta: 0:02:32  lr: 0.003639  min_lr: 0.003639  loss: 3.7457 (3.6408)  weight_decay: 0.0500 (0.0500)  time: 0.8860  data: 0.2627  max mem: 41794
Epoch: [74]  [140/312]  eta: 0:02:20  lr: 0.003638  min_lr: 0.003638  loss: 3.8032 (3.6538)  weight_decay: 0.0500 (0.0500)  time: 0.6794  data: 0.1202  max mem: 41794
Epoch: [74]  [150/312]  eta: 0:02:11  lr: 0.003638  min_lr: 0.003638  loss: 3.7686 (3.6639)  weight_decay: 0.0500 (0.0500)  time: 0.6426  data: 0.1316  max mem: 41794
Epoch: [74]  [160/312]  eta: 0:02:03  lr: 0.003638  min_lr: 0.003638  loss: 3.6308 (3.6585)  weight_decay: 0.0500 (0.0500)  time: 0.8090  data: 0.3033  max mem: 41794
Epoch: [74]  [170/312]  eta: 0:01:53  lr: 0.003637  min_lr: 0.003637  loss: 3.6308 (3.6596)  weight_decay: 0.0500 (0.0500)  time: 0.6861  data: 0.1750  max mem: 41794
Epoch: [74]  [180/312]  eta: 0:01:46  lr: 0.003637  min_lr: 0.003637  loss: 3.8058 (3.6734)  weight_decay: 0.0500 (0.0500)  time: 0.7208  data: 0.1875  max mem: 41794
Epoch: [74]  [190/312]  eta: 0:01:36  lr: 0.003636  min_lr: 0.003636  loss: 3.9046 (3.6774)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.1875  max mem: 41794
Epoch: [74]  [200/312]  eta: 0:01:28  lr: 0.003636  min_lr: 0.003636  loss: 3.6826 (3.6546)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.1578  max mem: 41794
Epoch: [74]  [210/312]  eta: 0:01:20  lr: 0.003635  min_lr: 0.003635  loss: 3.7134 (3.6515)  weight_decay: 0.0500 (0.0500)  time: 0.8400  data: 0.2926  max mem: 41794
Epoch: [74]  [220/312]  eta: 0:01:11  lr: 0.003635  min_lr: 0.003635  loss: 3.8326 (3.6607)  weight_decay: 0.0500 (0.0500)  time: 0.6552  data: 0.1354  max mem: 41794
Epoch: [74]  [230/312]  eta: 0:01:04  lr: 0.003635  min_lr: 0.003635  loss: 3.9595 (3.6713)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.1525  max mem: 41794
Epoch: [74]  [240/312]  eta: 0:00:56  lr: 0.003634  min_lr: 0.003634  loss: 3.8739 (3.6719)  weight_decay: 0.0500 (0.0500)  time: 0.8549  data: 0.2730  max mem: 41794
Epoch: [74]  [250/312]  eta: 0:00:48  lr: 0.003634  min_lr: 0.003634  loss: 3.7721 (3.6659)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1223  max mem: 41794
Epoch: [74]  [260/312]  eta: 0:00:40  lr: 0.003633  min_lr: 0.003633  loss: 3.8847 (3.6726)  weight_decay: 0.0500 (0.0500)  time: 0.7239  data: 0.1254  max mem: 41794
Epoch: [74]  [270/312]  eta: 0:00:32  lr: 0.003633  min_lr: 0.003633  loss: 3.8076 (3.6678)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.1240  max mem: 41794
Epoch: [74]  [280/312]  eta: 0:00:24  lr: 0.003633  min_lr: 0.003633  loss: 3.7810 (3.6700)  weight_decay: 0.0500 (0.0500)  time: 0.6691  data: 0.1194  max mem: 41794
Epoch: [74]  [290/312]  eta: 0:00:17  lr: 0.003632  min_lr: 0.003632  loss: 3.8356 (3.6644)  weight_decay: 0.0500 (0.0500)  time: 0.8497  data: 0.2488  max mem: 41794
Epoch: [74]  [300/312]  eta: 0:00:09  lr: 0.003632  min_lr: 0.003632  loss: 3.8090 (3.6599)  weight_decay: 0.0500 (0.0500)  time: 0.6738  data: 0.1377  max mem: 41794
Epoch: [74]  [310/312]  eta: 0:00:01  lr: 0.003631  min_lr: 0.003631  loss: 3.7031 (3.6567)  weight_decay: 0.0500 (0.0500)  time: 0.4738  data: 0.0080  max mem: 41794
Epoch: [74]  [311/312]  eta: 0:00:00  lr: 0.003631  min_lr: 0.003631  loss: 3.2278 (3.6531)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [74] Total time: 0:03:57 (0.7624 s / it)
Averaged stats: lr: 0.003631  min_lr: 0.003631  loss: 3.2278 (3.6691)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.2420 (1.2420)  acc1: 76.3021 (76.3021)  acc5: 92.1875 (92.1875)  time: 8.6024  data: 8.3934  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6323 (1.4771)  acc1: 68.0990 (69.6160)  acc5: 89.7135 (89.6800)  time: 1.1038  data: 0.9327  max mem: 41794
Test: Total time: 0:00:10 (1.1363 s / it)
* Acc@1 69.398 Acc@5 89.778 loss 1.474
Accuracy of the model on the 50000 test images: 69.4%
Max accuracy: 69.40%
Epoch: [75]  [  0/312]  eta: 1:11:53  lr: 0.003631  min_lr: 0.003631  loss: 3.8506 (3.8506)  weight_decay: 0.0500 (0.0500)  time: 13.8243  data: 13.3466  max mem: 41794
Epoch: [75]  [ 10/312]  eta: 0:09:37  lr: 0.003631  min_lr: 0.003631  loss: 3.5338 (3.4642)  weight_decay: 0.0500 (0.0500)  time: 1.9114  data: 1.3755  max mem: 41794
Epoch: [75]  [ 20/312]  eta: 0:06:55  lr: 0.003630  min_lr: 0.003630  loss: 3.4382 (3.4540)  weight_decay: 0.0500 (0.0500)  time: 0.8033  data: 0.2742  max mem: 41794
Epoch: [75]  [ 30/312]  eta: 0:05:19  lr: 0.003630  min_lr: 0.003630  loss: 3.7187 (3.4619)  weight_decay: 0.0500 (0.0500)  time: 0.7037  data: 0.1854  max mem: 41794
Epoch: [75]  [ 40/312]  eta: 0:04:48  lr: 0.003630  min_lr: 0.003630  loss: 3.6853 (3.4880)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1699  max mem: 41794
Epoch: [75]  [ 50/312]  eta: 0:04:37  lr: 0.003629  min_lr: 0.003629  loss: 3.7244 (3.5405)  weight_decay: 0.0500 (0.0500)  time: 0.9472  data: 0.3779  max mem: 41794
Epoch: [75]  [ 60/312]  eta: 0:04:04  lr: 0.003629  min_lr: 0.003629  loss: 3.8938 (3.5997)  weight_decay: 0.0500 (0.0500)  time: 0.7799  data: 0.2087  max mem: 41794
Epoch: [75]  [ 70/312]  eta: 0:03:53  lr: 0.003628  min_lr: 0.003628  loss: 3.9297 (3.6082)  weight_decay: 0.0500 (0.0500)  time: 0.7303  data: 0.1417  max mem: 41794
Epoch: [75]  [ 80/312]  eta: 0:03:35  lr: 0.003628  min_lr: 0.003628  loss: 3.7716 (3.6236)  weight_decay: 0.0500 (0.0500)  time: 0.8158  data: 0.1565  max mem: 41794
Epoch: [75]  [ 90/312]  eta: 0:03:19  lr: 0.003627  min_lr: 0.003627  loss: 3.7716 (3.6198)  weight_decay: 0.0500 (0.0500)  time: 0.6567  data: 0.0859  max mem: 41794
Epoch: [75]  [100/312]  eta: 0:03:10  lr: 0.003627  min_lr: 0.003627  loss: 3.5277 (3.5897)  weight_decay: 0.0500 (0.0500)  time: 0.7776  data: 0.1658  max mem: 41794
Epoch: [75]  [110/312]  eta: 0:02:55  lr: 0.003627  min_lr: 0.003627  loss: 3.2583 (3.5941)  weight_decay: 0.0500 (0.0500)  time: 0.7235  data: 0.0986  max mem: 41794
Epoch: [75]  [120/312]  eta: 0:02:44  lr: 0.003626  min_lr: 0.003626  loss: 3.7154 (3.6015)  weight_decay: 0.0500 (0.0500)  time: 0.6486  data: 0.0528  max mem: 41794
Epoch: [75]  [130/312]  eta: 0:02:35  lr: 0.003626  min_lr: 0.003626  loss: 3.7176 (3.6095)  weight_decay: 0.0500 (0.0500)  time: 0.7698  data: 0.0909  max mem: 41794
Epoch: [75]  [140/312]  eta: 0:02:22  lr: 0.003625  min_lr: 0.003625  loss: 3.8128 (3.6099)  weight_decay: 0.0500 (0.0500)  time: 0.6299  data: 0.0420  max mem: 41794
Epoch: [75]  [150/312]  eta: 0:02:14  lr: 0.003625  min_lr: 0.003625  loss: 3.6308 (3.6197)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.0691  max mem: 41794
Epoch: [75]  [160/312]  eta: 0:02:04  lr: 0.003625  min_lr: 0.003625  loss: 3.7234 (3.6308)  weight_decay: 0.0500 (0.0500)  time: 0.7928  data: 0.0768  max mem: 41794
Epoch: [75]  [170/312]  eta: 0:01:55  lr: 0.003624  min_lr: 0.003624  loss: 3.8301 (3.6346)  weight_decay: 0.0500 (0.0500)  time: 0.7135  data: 0.1095  max mem: 41794
Epoch: [75]  [180/312]  eta: 0:01:47  lr: 0.003624  min_lr: 0.003624  loss: 3.8190 (3.6478)  weight_decay: 0.0500 (0.0500)  time: 0.7302  data: 0.1797  max mem: 41794
Epoch: [75]  [190/312]  eta: 0:01:37  lr: 0.003623  min_lr: 0.003623  loss: 3.7357 (3.6497)  weight_decay: 0.0500 (0.0500)  time: 0.6293  data: 0.0785  max mem: 41794
Epoch: [75]  [200/312]  eta: 0:01:30  lr: 0.003623  min_lr: 0.003623  loss: 3.7357 (3.6581)  weight_decay: 0.0500 (0.0500)  time: 0.7699  data: 0.1224  max mem: 41794
Epoch: [75]  [210/312]  eta: 0:01:21  lr: 0.003622  min_lr: 0.003622  loss: 3.8729 (3.6694)  weight_decay: 0.0500 (0.0500)  time: 0.8816  data: 0.1824  max mem: 41794
Epoch: [75]  [220/312]  eta: 0:01:12  lr: 0.003622  min_lr: 0.003622  loss: 3.9356 (3.6757)  weight_decay: 0.0500 (0.0500)  time: 0.6054  data: 0.0607  max mem: 41794
Epoch: [75]  [230/312]  eta: 0:01:05  lr: 0.003622  min_lr: 0.003622  loss: 3.7799 (3.6736)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.1021  max mem: 41794
Epoch: [75]  [240/312]  eta: 0:00:56  lr: 0.003621  min_lr: 0.003621  loss: 3.5940 (3.6761)  weight_decay: 0.0500 (0.0500)  time: 0.7976  data: 0.1087  max mem: 41794
Epoch: [75]  [250/312]  eta: 0:00:48  lr: 0.003621  min_lr: 0.003621  loss: 3.5940 (3.6669)  weight_decay: 0.0500 (0.0500)  time: 0.6436  data: 0.0563  max mem: 41794
Epoch: [75]  [260/312]  eta: 0:00:40  lr: 0.003620  min_lr: 0.003620  loss: 3.7034 (3.6630)  weight_decay: 0.0500 (0.0500)  time: 0.7065  data: 0.0953  max mem: 41794
Epoch: [75]  [270/312]  eta: 0:00:32  lr: 0.003620  min_lr: 0.003620  loss: 3.7480 (3.6683)  weight_decay: 0.0500 (0.0500)  time: 0.6564  data: 0.0479  max mem: 41794
Epoch: [75]  [280/312]  eta: 0:00:24  lr: 0.003619  min_lr: 0.003619  loss: 4.0331 (3.6799)  weight_decay: 0.0500 (0.0500)  time: 0.7165  data: 0.0979  max mem: 41794
Epoch: [75]  [290/312]  eta: 0:00:17  lr: 0.003619  min_lr: 0.003619  loss: 4.0331 (3.6787)  weight_decay: 0.0500 (0.0500)  time: 0.8239  data: 0.1257  max mem: 41794
Epoch: [75]  [300/312]  eta: 0:00:09  lr: 0.003619  min_lr: 0.003619  loss: 3.6873 (3.6796)  weight_decay: 0.0500 (0.0500)  time: 0.5861  data: 0.0298  max mem: 41794
Epoch: [75]  [310/312]  eta: 0:00:01  lr: 0.003618  min_lr: 0.003618  loss: 3.6561 (3.6748)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [75]  [311/312]  eta: 0:00:00  lr: 0.003618  min_lr: 0.003618  loss: 3.6561 (3.6751)  weight_decay: 0.0500 (0.0500)  time: 0.4641  data: 0.0001  max mem: 41794
Epoch: [75] Total time: 0:03:57 (0.7624 s / it)
Averaged stats: lr: 0.003618  min_lr: 0.003618  loss: 3.6561 (3.6672)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.2219 (1.2219)  acc1: 74.2188 (74.2188)  acc5: 92.0573 (92.0573)  time: 8.2681  data: 8.0667  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6175 (1.5011)  acc1: 67.0573 (68.0640)  acc5: 88.0208 (89.1040)  time: 1.0771  data: 0.8964  max mem: 41794
Test: Total time: 0:00:10 (1.1166 s / it)
* Acc@1 68.188 Acc@5 89.062 loss 1.517
Accuracy of the model on the 50000 test images: 68.2%
Max accuracy: 69.40%
Epoch: [76]  [  0/312]  eta: 1:24:27  lr: 0.003618  min_lr: 0.003618  loss: 3.7286 (3.7286)  weight_decay: 0.0500 (0.0500)  time: 16.2414  data: 12.3445  max mem: 41794
Epoch: [76]  [ 10/312]  eta: 0:11:44  lr: 0.003618  min_lr: 0.003618  loss: 4.0232 (3.7765)  weight_decay: 0.0500 (0.0500)  time: 2.3321  data: 1.3296  max mem: 41794
Epoch: [76]  [ 20/312]  eta: 0:07:33  lr: 0.003617  min_lr: 0.003617  loss: 3.7253 (3.6345)  weight_decay: 0.0500 (0.0500)  time: 0.8199  data: 0.1302  max mem: 41794
Epoch: [76]  [ 30/312]  eta: 0:05:45  lr: 0.003617  min_lr: 0.003617  loss: 3.5104 (3.6281)  weight_decay: 0.0500 (0.0500)  time: 0.6185  data: 0.0259  max mem: 41794
Epoch: [76]  [ 40/312]  eta: 0:04:46  lr: 0.003616  min_lr: 0.003616  loss: 3.5813 (3.6260)  weight_decay: 0.0500 (0.0500)  time: 0.5301  data: 0.0184  max mem: 41794
Epoch: [76]  [ 50/312]  eta: 0:04:18  lr: 0.003616  min_lr: 0.003616  loss: 3.7503 (3.6173)  weight_decay: 0.0500 (0.0500)  time: 0.6173  data: 0.0156  max mem: 41794
Epoch: [76]  [ 60/312]  eta: 0:03:49  lr: 0.003616  min_lr: 0.003616  loss: 3.8060 (3.6298)  weight_decay: 0.0500 (0.0500)  time: 0.6104  data: 0.0101  max mem: 41794
Epoch: [76]  [ 70/312]  eta: 0:03:38  lr: 0.003615  min_lr: 0.003615  loss: 3.5945 (3.6176)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.0369  max mem: 41794
Epoch: [76]  [ 80/312]  eta: 0:03:26  lr: 0.003615  min_lr: 0.003615  loss: 3.4579 (3.5924)  weight_decay: 0.0500 (0.0500)  time: 0.8349  data: 0.0506  max mem: 41794
Epoch: [76]  [ 90/312]  eta: 0:03:10  lr: 0.003614  min_lr: 0.003614  loss: 3.6213 (3.5856)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.0524  max mem: 41794
Epoch: [76]  [100/312]  eta: 0:03:00  lr: 0.003614  min_lr: 0.003614  loss: 3.6672 (3.5908)  weight_decay: 0.0500 (0.0500)  time: 0.6997  data: 0.0686  max mem: 41794
Epoch: [76]  [110/312]  eta: 0:02:46  lr: 0.003613  min_lr: 0.003613  loss: 3.6570 (3.5895)  weight_decay: 0.0500 (0.0500)  time: 0.6887  data: 0.0681  max mem: 41794
Epoch: [76]  [120/312]  eta: 0:02:39  lr: 0.003613  min_lr: 0.003613  loss: 3.8223 (3.6063)  weight_decay: 0.0500 (0.0500)  time: 0.7178  data: 0.1052  max mem: 41794
Epoch: [76]  [130/312]  eta: 0:02:29  lr: 0.003613  min_lr: 0.003613  loss: 3.8601 (3.6063)  weight_decay: 0.0500 (0.0500)  time: 0.8078  data: 0.0921  max mem: 41794
Epoch: [76]  [140/312]  eta: 0:02:18  lr: 0.003612  min_lr: 0.003612  loss: 3.4473 (3.5889)  weight_decay: 0.0500 (0.0500)  time: 0.6629  data: 0.0687  max mem: 41794
Epoch: [76]  [150/312]  eta: 0:02:11  lr: 0.003612  min_lr: 0.003612  loss: 3.6570 (3.6099)  weight_decay: 0.0500 (0.0500)  time: 0.7391  data: 0.1669  max mem: 41794
Epoch: [76]  [160/312]  eta: 0:02:02  lr: 0.003611  min_lr: 0.003611  loss: 3.9733 (3.6275)  weight_decay: 0.0500 (0.0500)  time: 0.8030  data: 0.1901  max mem: 41794
Epoch: [76]  [170/312]  eta: 0:01:53  lr: 0.003611  min_lr: 0.003611  loss: 3.9133 (3.6250)  weight_decay: 0.0500 (0.0500)  time: 0.7177  data: 0.1405  max mem: 41794
Epoch: [76]  [180/312]  eta: 0:01:45  lr: 0.003610  min_lr: 0.003610  loss: 3.8766 (3.6327)  weight_decay: 0.0500 (0.0500)  time: 0.7139  data: 0.1123  max mem: 41794
Epoch: [76]  [190/312]  eta: 0:01:36  lr: 0.003610  min_lr: 0.003610  loss: 3.8484 (3.6336)  weight_decay: 0.0500 (0.0500)  time: 0.6714  data: 0.1086  max mem: 41794
Epoch: [76]  [200/312]  eta: 0:01:28  lr: 0.003610  min_lr: 0.003610  loss: 3.6837 (3.6281)  weight_decay: 0.0500 (0.0500)  time: 0.7558  data: 0.1803  max mem: 41794
Epoch: [76]  [210/312]  eta: 0:01:20  lr: 0.003609  min_lr: 0.003609  loss: 3.6369 (3.6352)  weight_decay: 0.0500 (0.0500)  time: 0.8383  data: 0.1606  max mem: 41794
Epoch: [76]  [220/312]  eta: 0:01:12  lr: 0.003609  min_lr: 0.003609  loss: 3.8282 (3.6391)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.0877  max mem: 41794
Epoch: [76]  [230/312]  eta: 0:01:04  lr: 0.003608  min_lr: 0.003608  loss: 3.7106 (3.6399)  weight_decay: 0.0500 (0.0500)  time: 0.7356  data: 0.1311  max mem: 41794
Epoch: [76]  [240/312]  eta: 0:00:56  lr: 0.003608  min_lr: 0.003608  loss: 3.7106 (3.6423)  weight_decay: 0.0500 (0.0500)  time: 0.7696  data: 0.1100  max mem: 41794
Epoch: [76]  [250/312]  eta: 0:00:48  lr: 0.003607  min_lr: 0.003607  loss: 3.9118 (3.6533)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.0809  max mem: 41794
Epoch: [76]  [260/312]  eta: 0:00:40  lr: 0.003607  min_lr: 0.003607  loss: 3.9118 (3.6509)  weight_decay: 0.0500 (0.0500)  time: 0.7589  data: 0.1084  max mem: 41794
Epoch: [76]  [270/312]  eta: 0:00:32  lr: 0.003607  min_lr: 0.003607  loss: 3.9788 (3.6569)  weight_decay: 0.0500 (0.0500)  time: 0.6806  data: 0.0743  max mem: 41794
Epoch: [76]  [280/312]  eta: 0:00:24  lr: 0.003606  min_lr: 0.003606  loss: 3.7998 (3.6544)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.1116  max mem: 41794
Epoch: [76]  [290/312]  eta: 0:00:17  lr: 0.003606  min_lr: 0.003606  loss: 3.8294 (3.6562)  weight_decay: 0.0500 (0.0500)  time: 0.8429  data: 0.1975  max mem: 41794
Epoch: [76]  [300/312]  eta: 0:00:09  lr: 0.003605  min_lr: 0.003605  loss: 3.9077 (3.6647)  weight_decay: 0.0500 (0.0500)  time: 0.6729  data: 0.1281  max mem: 41794
Epoch: [76]  [310/312]  eta: 0:00:01  lr: 0.003605  min_lr: 0.003605  loss: 3.8273 (3.6581)  weight_decay: 0.0500 (0.0500)  time: 0.4791  data: 0.0149  max mem: 41794
Epoch: [76]  [311/312]  eta: 0:00:00  lr: 0.003605  min_lr: 0.003605  loss: 3.8666 (3.6606)  weight_decay: 0.0500 (0.0500)  time: 0.4764  data: 0.0124  max mem: 41794
Epoch: [76] Total time: 0:03:58 (0.7633 s / it)
Averaged stats: lr: 0.003605  min_lr: 0.003605  loss: 3.8666 (3.6706)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.2784 (1.2784)  acc1: 77.3438 (77.3438)  acc5: 92.9688 (92.9688)  time: 8.1436  data: 7.9339  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7390 (1.6001)  acc1: 65.8854 (68.7040)  acc5: 87.1094 (88.8640)  time: 1.0529  data: 0.8816  max mem: 41794
Test: Total time: 0:00:09 (1.0631 s / it)
* Acc@1 68.678 Acc@5 89.172 loss 1.606
Accuracy of the model on the 50000 test images: 68.7%
Max accuracy: 69.40%
Epoch: [77]  [  0/312]  eta: 1:21:47  lr: 0.003605  min_lr: 0.003605  loss: 4.2926 (4.2926)  weight_decay: 0.0500 (0.0500)  time: 15.7287  data: 14.1439  max mem: 41794
Epoch: [77]  [ 10/312]  eta: 0:11:56  lr: 0.003604  min_lr: 0.003604  loss: 3.1695 (3.3275)  weight_decay: 0.0500 (0.0500)  time: 2.3730  data: 1.3723  max mem: 41794
Epoch: [77]  [ 20/312]  eta: 0:07:47  lr: 0.003604  min_lr: 0.003604  loss: 3.3142 (3.5093)  weight_decay: 0.0500 (0.0500)  time: 0.8963  data: 0.1295  max mem: 41794
Epoch: [77]  [ 30/312]  eta: 0:05:54  lr: 0.003604  min_lr: 0.003604  loss: 3.8196 (3.6008)  weight_decay: 0.0500 (0.0500)  time: 0.6431  data: 0.0956  max mem: 41794
Epoch: [77]  [ 40/312]  eta: 0:04:51  lr: 0.003603  min_lr: 0.003603  loss: 3.7402 (3.5701)  weight_decay: 0.0500 (0.0500)  time: 0.5177  data: 0.0152  max mem: 41794
Epoch: [77]  [ 50/312]  eta: 0:04:16  lr: 0.003603  min_lr: 0.003603  loss: 3.6991 (3.5764)  weight_decay: 0.0500 (0.0500)  time: 0.5520  data: 0.0020  max mem: 41794
Epoch: [77]  [ 60/312]  eta: 0:03:47  lr: 0.003602  min_lr: 0.003602  loss: 3.6991 (3.5713)  weight_decay: 0.0500 (0.0500)  time: 0.5498  data: 0.0035  max mem: 41794
Epoch: [77]  [ 70/312]  eta: 0:03:38  lr: 0.003602  min_lr: 0.003602  loss: 3.7831 (3.6312)  weight_decay: 0.0500 (0.0500)  time: 0.6986  data: 0.0890  max mem: 41794
Epoch: [77]  [ 80/312]  eta: 0:03:22  lr: 0.003601  min_lr: 0.003601  loss: 3.8167 (3.6391)  weight_decay: 0.0500 (0.0500)  time: 0.7891  data: 0.0862  max mem: 41794
Epoch: [77]  [ 90/312]  eta: 0:03:08  lr: 0.003601  min_lr: 0.003601  loss: 3.7566 (3.6409)  weight_decay: 0.0500 (0.0500)  time: 0.6546  data: 0.0586  max mem: 41794
Epoch: [77]  [100/312]  eta: 0:02:59  lr: 0.003601  min_lr: 0.003601  loss: 3.7455 (3.6207)  weight_decay: 0.0500 (0.0500)  time: 0.7384  data: 0.1487  max mem: 41794
Epoch: [77]  [110/312]  eta: 0:02:45  lr: 0.003600  min_lr: 0.003600  loss: 3.7026 (3.6259)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.0951  max mem: 41794
Epoch: [77]  [120/312]  eta: 0:02:37  lr: 0.003600  min_lr: 0.003600  loss: 3.7762 (3.6445)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1275  max mem: 41794
Epoch: [77]  [130/312]  eta: 0:02:28  lr: 0.003599  min_lr: 0.003599  loss: 3.8807 (3.6466)  weight_decay: 0.0500 (0.0500)  time: 0.8139  data: 0.2223  max mem: 41794
Epoch: [77]  [140/312]  eta: 0:02:17  lr: 0.003599  min_lr: 0.003599  loss: 3.9516 (3.6732)  weight_decay: 0.0500 (0.0500)  time: 0.6572  data: 0.1195  max mem: 41794
Epoch: [77]  [150/312]  eta: 0:02:10  lr: 0.003598  min_lr: 0.003598  loss: 3.8563 (3.6548)  weight_decay: 0.0500 (0.0500)  time: 0.7223  data: 0.1395  max mem: 41794
Epoch: [77]  [160/312]  eta: 0:02:02  lr: 0.003598  min_lr: 0.003598  loss: 3.7097 (3.6668)  weight_decay: 0.0500 (0.0500)  time: 0.8428  data: 0.1862  max mem: 41794
Epoch: [77]  [170/312]  eta: 0:01:52  lr: 0.003598  min_lr: 0.003598  loss: 3.7561 (3.6655)  weight_decay: 0.0500 (0.0500)  time: 0.6864  data: 0.1077  max mem: 41794
Epoch: [77]  [180/312]  eta: 0:01:44  lr: 0.003597  min_lr: 0.003597  loss: 3.8662 (3.6677)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.1363  max mem: 41794
Epoch: [77]  [190/312]  eta: 0:01:34  lr: 0.003597  min_lr: 0.003597  loss: 3.8103 (3.6677)  weight_decay: 0.0500 (0.0500)  time: 0.6623  data: 0.1032  max mem: 41794
Epoch: [77]  [200/312]  eta: 0:01:27  lr: 0.003596  min_lr: 0.003596  loss: 3.6722 (3.6656)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.1314  max mem: 41794
Epoch: [77]  [210/312]  eta: 0:01:19  lr: 0.003596  min_lr: 0.003596  loss: 3.6722 (3.6507)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.2557  max mem: 41794
Epoch: [77]  [220/312]  eta: 0:01:10  lr: 0.003595  min_lr: 0.003595  loss: 3.7695 (3.6513)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.1382  max mem: 41794
Epoch: [77]  [230/312]  eta: 0:01:03  lr: 0.003595  min_lr: 0.003595  loss: 3.7710 (3.6514)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.1456  max mem: 41794
Epoch: [77]  [240/312]  eta: 0:00:55  lr: 0.003594  min_lr: 0.003594  loss: 3.8565 (3.6590)  weight_decay: 0.0500 (0.0500)  time: 0.7979  data: 0.2402  max mem: 41794
Epoch: [77]  [250/312]  eta: 0:00:47  lr: 0.003594  min_lr: 0.003594  loss: 3.7731 (3.6625)  weight_decay: 0.0500 (0.0500)  time: 0.7184  data: 0.1553  max mem: 41794
Epoch: [77]  [260/312]  eta: 0:00:40  lr: 0.003594  min_lr: 0.003594  loss: 3.7245 (3.6593)  weight_decay: 0.0500 (0.0500)  time: 0.7750  data: 0.1930  max mem: 41794
Epoch: [77]  [270/312]  eta: 0:00:32  lr: 0.003593  min_lr: 0.003593  loss: 3.6399 (3.6559)  weight_decay: 0.0500 (0.0500)  time: 0.6906  data: 0.1406  max mem: 41794
Epoch: [77]  [280/312]  eta: 0:00:24  lr: 0.003593  min_lr: 0.003593  loss: 3.8295 (3.6652)  weight_decay: 0.0500 (0.0500)  time: 0.6678  data: 0.1132  max mem: 41794
Epoch: [77]  [290/312]  eta: 0:00:16  lr: 0.003592  min_lr: 0.003592  loss: 3.8641 (3.6685)  weight_decay: 0.0500 (0.0500)  time: 0.8696  data: 0.2376  max mem: 41794
Epoch: [77]  [300/312]  eta: 0:00:09  lr: 0.003592  min_lr: 0.003592  loss: 3.8612 (3.6729)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.1268  max mem: 41794
Epoch: [77]  [310/312]  eta: 0:00:01  lr: 0.003591  min_lr: 0.003591  loss: 3.6416 (3.6681)  weight_decay: 0.0500 (0.0500)  time: 0.4713  data: 0.0001  max mem: 41794
Epoch: [77]  [311/312]  eta: 0:00:00  lr: 0.003591  min_lr: 0.003591  loss: 3.6939 (3.6690)  weight_decay: 0.0500 (0.0500)  time: 0.4716  data: 0.0001  max mem: 41794
Epoch: [77] Total time: 0:03:56 (0.7586 s / it)
Averaged stats: lr: 0.003591  min_lr: 0.003591  loss: 3.6939 (3.6637)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3430 (1.3430)  acc1: 75.7812 (75.7812)  acc5: 93.4896 (93.4896)  time: 8.4802  data: 8.2743  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6673 (1.5466)  acc1: 68.3594 (68.5920)  acc5: 88.1510 (89.3600)  time: 1.0910  data: 0.9195  max mem: 41794
Test: Total time: 0:00:10 (1.1373 s / it)
* Acc@1 68.962 Acc@5 89.380 loss 1.558
Accuracy of the model on the 50000 test images: 69.0%
Max accuracy: 69.40%
Epoch: [78]  [  0/312]  eta: 1:24:38  lr: 0.003591  min_lr: 0.003591  loss: 2.5413 (2.5413)  weight_decay: 0.0500 (0.0500)  time: 16.2770  data: 13.1550  max mem: 41794
Epoch: [78]  [ 10/312]  eta: 0:11:15  lr: 0.003591  min_lr: 0.003591  loss: 3.8671 (3.6717)  weight_decay: 0.0500 (0.0500)  time: 2.2368  data: 1.2664  max mem: 41794
Epoch: [78]  [ 20/312]  eta: 0:07:38  lr: 0.003590  min_lr: 0.003590  loss: 3.8075 (3.5773)  weight_decay: 0.0500 (0.0500)  time: 0.8365  data: 0.0476  max mem: 41794
Epoch: [78]  [ 30/312]  eta: 0:05:50  lr: 0.003590  min_lr: 0.003590  loss: 3.6157 (3.6015)  weight_decay: 0.0500 (0.0500)  time: 0.6952  data: 0.0221  max mem: 41794
Epoch: [78]  [ 40/312]  eta: 0:04:52  lr: 0.003590  min_lr: 0.003590  loss: 3.6057 (3.6102)  weight_decay: 0.0500 (0.0500)  time: 0.5565  data: 0.0225  max mem: 41794
Epoch: [78]  [ 50/312]  eta: 0:04:26  lr: 0.003589  min_lr: 0.003589  loss: 3.2356 (3.5433)  weight_decay: 0.0500 (0.0500)  time: 0.6669  data: 0.0871  max mem: 41794
Epoch: [78]  [ 60/312]  eta: 0:03:54  lr: 0.003589  min_lr: 0.003589  loss: 3.4830 (3.5867)  weight_decay: 0.0500 (0.0500)  time: 0.6299  data: 0.0782  max mem: 41794
Epoch: [78]  [ 70/312]  eta: 0:03:42  lr: 0.003588  min_lr: 0.003588  loss: 3.8035 (3.5995)  weight_decay: 0.0500 (0.0500)  time: 0.6677  data: 0.1215  max mem: 41794
Epoch: [78]  [ 80/312]  eta: 0:03:30  lr: 0.003588  min_lr: 0.003588  loss: 3.8035 (3.6139)  weight_decay: 0.0500 (0.0500)  time: 0.8414  data: 0.2073  max mem: 41794
Epoch: [78]  [ 90/312]  eta: 0:03:13  lr: 0.003587  min_lr: 0.003587  loss: 3.6102 (3.6023)  weight_decay: 0.0500 (0.0500)  time: 0.7119  data: 0.1299  max mem: 41794
Epoch: [78]  [100/312]  eta: 0:03:04  lr: 0.003587  min_lr: 0.003587  loss: 3.5829 (3.5874)  weight_decay: 0.0500 (0.0500)  time: 0.7231  data: 0.1347  max mem: 41794
Epoch: [78]  [110/312]  eta: 0:02:49  lr: 0.003587  min_lr: 0.003587  loss: 3.4826 (3.5741)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.0928  max mem: 41794
Epoch: [78]  [120/312]  eta: 0:02:42  lr: 0.003586  min_lr: 0.003586  loss: 3.1067 (3.5600)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.1113  max mem: 41794
Epoch: [78]  [130/312]  eta: 0:02:34  lr: 0.003586  min_lr: 0.003586  loss: 3.4080 (3.5694)  weight_decay: 0.0500 (0.0500)  time: 0.9114  data: 0.1885  max mem: 41794
Epoch: [78]  [140/312]  eta: 0:02:21  lr: 0.003585  min_lr: 0.003585  loss: 3.8146 (3.5933)  weight_decay: 0.0500 (0.0500)  time: 0.7015  data: 0.0795  max mem: 41794
Epoch: [78]  [150/312]  eta: 0:02:12  lr: 0.003585  min_lr: 0.003585  loss: 3.8146 (3.5936)  weight_decay: 0.0500 (0.0500)  time: 0.6335  data: 0.0809  max mem: 41794
Epoch: [78]  [160/312]  eta: 0:02:04  lr: 0.003584  min_lr: 0.003584  loss: 3.7857 (3.5994)  weight_decay: 0.0500 (0.0500)  time: 0.7648  data: 0.1659  max mem: 41794
Epoch: [78]  [170/312]  eta: 0:01:54  lr: 0.003584  min_lr: 0.003584  loss: 3.7857 (3.6028)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.1036  max mem: 41794
Epoch: [78]  [180/312]  eta: 0:01:46  lr: 0.003583  min_lr: 0.003583  loss: 3.8889 (3.6087)  weight_decay: 0.0500 (0.0500)  time: 0.7258  data: 0.1307  max mem: 41794
Epoch: [78]  [190/312]  eta: 0:01:36  lr: 0.003583  min_lr: 0.003583  loss: 3.7463 (3.6116)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.1126  max mem: 41794
Epoch: [78]  [200/312]  eta: 0:01:28  lr: 0.003583  min_lr: 0.003583  loss: 3.8413 (3.6247)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.1185  max mem: 41794
Epoch: [78]  [210/312]  eta: 0:01:20  lr: 0.003582  min_lr: 0.003582  loss: 3.9216 (3.6344)  weight_decay: 0.0500 (0.0500)  time: 0.8344  data: 0.2118  max mem: 41794
Epoch: [78]  [220/312]  eta: 0:01:11  lr: 0.003582  min_lr: 0.003582  loss: 3.8366 (3.6360)  weight_decay: 0.0500 (0.0500)  time: 0.6773  data: 0.0988  max mem: 41794
Epoch: [78]  [230/312]  eta: 0:01:04  lr: 0.003581  min_lr: 0.003581  loss: 3.6246 (3.6331)  weight_decay: 0.0500 (0.0500)  time: 0.7290  data: 0.1129  max mem: 41794
Epoch: [78]  [240/312]  eta: 0:00:56  lr: 0.003581  min_lr: 0.003581  loss: 3.9128 (3.6438)  weight_decay: 0.0500 (0.0500)  time: 0.8405  data: 0.2209  max mem: 41794
Epoch: [78]  [250/312]  eta: 0:00:48  lr: 0.003580  min_lr: 0.003580  loss: 3.9128 (3.6383)  weight_decay: 0.0500 (0.0500)  time: 0.6594  data: 0.1136  max mem: 41794
Epoch: [78]  [260/312]  eta: 0:00:40  lr: 0.003580  min_lr: 0.003580  loss: 3.7884 (3.6431)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.1203  max mem: 41794
Epoch: [78]  [270/312]  eta: 0:00:32  lr: 0.003580  min_lr: 0.003580  loss: 3.6409 (3.6378)  weight_decay: 0.0500 (0.0500)  time: 0.6760  data: 0.1203  max mem: 41794
Epoch: [78]  [280/312]  eta: 0:00:24  lr: 0.003579  min_lr: 0.003579  loss: 3.5369 (3.6337)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.0975  max mem: 41794
Epoch: [78]  [290/312]  eta: 0:00:17  lr: 0.003579  min_lr: 0.003579  loss: 3.8306 (3.6443)  weight_decay: 0.0500 (0.0500)  time: 0.8453  data: 0.1615  max mem: 41794
Epoch: [78]  [300/312]  eta: 0:00:09  lr: 0.003578  min_lr: 0.003578  loss: 3.9027 (3.6474)  weight_decay: 0.0500 (0.0500)  time: 0.6426  data: 0.0644  max mem: 41794
Epoch: [78]  [310/312]  eta: 0:00:01  lr: 0.003578  min_lr: 0.003578  loss: 3.7153 (3.6499)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [78]  [311/312]  eta: 0:00:00  lr: 0.003578  min_lr: 0.003578  loss: 3.7153 (3.6512)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0001  max mem: 41794
Epoch: [78] Total time: 0:03:57 (0.7611 s / it)
Averaged stats: lr: 0.003578  min_lr: 0.003578  loss: 3.7153 (3.6515)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.3244 (1.3244)  acc1: 74.2188 (74.2188)  acc5: 93.4896 (93.4896)  time: 8.5993  data: 8.3997  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7419 (1.5681)  acc1: 67.0573 (68.7200)  acc5: 88.9323 (89.4080)  time: 1.1093  data: 0.9334  max mem: 41794
Test: Total time: 0:00:10 (1.1561 s / it)
* Acc@1 69.252 Acc@5 89.354 loss 1.561
Accuracy of the model on the 50000 test images: 69.3%
Max accuracy: 69.40%
Epoch: [79]  [  0/312]  eta: 1:18:31  lr: 0.003578  min_lr: 0.003578  loss: 3.3320 (3.3320)  weight_decay: 0.0500 (0.0500)  time: 15.1011  data: 14.4762  max mem: 41794
Epoch: [79]  [ 10/312]  eta: 0:10:19  lr: 0.003577  min_lr: 0.003577  loss: 4.2781 (3.8328)  weight_decay: 0.0500 (0.0500)  time: 2.0505  data: 1.3872  max mem: 41794
Epoch: [79]  [ 20/312]  eta: 0:07:16  lr: 0.003577  min_lr: 0.003577  loss: 4.0311 (3.7631)  weight_decay: 0.0500 (0.0500)  time: 0.8148  data: 0.1585  max mem: 41794
Epoch: [79]  [ 30/312]  eta: 0:05:33  lr: 0.003576  min_lr: 0.003576  loss: 3.7455 (3.6939)  weight_decay: 0.0500 (0.0500)  time: 0.7026  data: 0.1221  max mem: 41794
Epoch: [79]  [ 40/312]  eta: 0:04:50  lr: 0.003576  min_lr: 0.003576  loss: 3.6294 (3.6824)  weight_decay: 0.0500 (0.0500)  time: 0.6230  data: 0.0840  max mem: 41794
Epoch: [79]  [ 50/312]  eta: 0:04:34  lr: 0.003575  min_lr: 0.003575  loss: 3.8063 (3.6954)  weight_decay: 0.0500 (0.0500)  time: 0.8419  data: 0.2069  max mem: 41794
Epoch: [79]  [ 60/312]  eta: 0:04:01  lr: 0.003575  min_lr: 0.003575  loss: 3.8206 (3.6783)  weight_decay: 0.0500 (0.0500)  time: 0.7282  data: 0.1260  max mem: 41794
Epoch: [79]  [ 70/312]  eta: 0:03:52  lr: 0.003575  min_lr: 0.003575  loss: 3.8686 (3.6855)  weight_decay: 0.0500 (0.0500)  time: 0.7335  data: 0.1106  max mem: 41794
Epoch: [79]  [ 80/312]  eta: 0:03:36  lr: 0.003574  min_lr: 0.003574  loss: 3.8686 (3.6900)  weight_decay: 0.0500 (0.0500)  time: 0.8505  data: 0.2117  max mem: 41794
Epoch: [79]  [ 90/312]  eta: 0:03:17  lr: 0.003574  min_lr: 0.003574  loss: 3.8789 (3.7298)  weight_decay: 0.0500 (0.0500)  time: 0.6349  data: 0.1149  max mem: 41794
Epoch: [79]  [100/312]  eta: 0:03:07  lr: 0.003573  min_lr: 0.003573  loss: 4.0339 (3.7543)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1361  max mem: 41794
Epoch: [79]  [110/312]  eta: 0:02:51  lr: 0.003573  min_lr: 0.003573  loss: 3.9566 (3.7538)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.1230  max mem: 41794
Epoch: [79]  [120/312]  eta: 0:02:42  lr: 0.003572  min_lr: 0.003572  loss: 3.7508 (3.7472)  weight_decay: 0.0500 (0.0500)  time: 0.6632  data: 0.1709  max mem: 41794
Epoch: [79]  [130/312]  eta: 0:02:35  lr: 0.003572  min_lr: 0.003572  loss: 3.8356 (3.7588)  weight_decay: 0.0500 (0.0500)  time: 0.8745  data: 0.3847  max mem: 41794
Epoch: [79]  [140/312]  eta: 0:02:22  lr: 0.003571  min_lr: 0.003571  loss: 3.3160 (3.7057)  weight_decay: 0.0500 (0.0500)  time: 0.7016  data: 0.2144  max mem: 41794
Epoch: [79]  [150/312]  eta: 0:02:14  lr: 0.003571  min_lr: 0.003571  loss: 3.2414 (3.6956)  weight_decay: 0.0500 (0.0500)  time: 0.6880  data: 0.2014  max mem: 41794
Epoch: [79]  [160/312]  eta: 0:02:06  lr: 0.003571  min_lr: 0.003571  loss: 3.5300 (3.6831)  weight_decay: 0.0500 (0.0500)  time: 0.8907  data: 0.3919  max mem: 41794
Epoch: [79]  [170/312]  eta: 0:01:55  lr: 0.003570  min_lr: 0.003570  loss: 3.7906 (3.6857)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.1911  max mem: 41794
Epoch: [79]  [180/312]  eta: 0:01:47  lr: 0.003570  min_lr: 0.003570  loss: 3.7906 (3.6803)  weight_decay: 0.0500 (0.0500)  time: 0.6605  data: 0.1719  max mem: 41794
Epoch: [79]  [190/312]  eta: 0:01:37  lr: 0.003569  min_lr: 0.003569  loss: 3.6400 (3.6769)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.1719  max mem: 41794
Epoch: [79]  [200/312]  eta: 0:01:29  lr: 0.003569  min_lr: 0.003569  loss: 3.7526 (3.6757)  weight_decay: 0.0500 (0.0500)  time: 0.6865  data: 0.1929  max mem: 41794
Epoch: [79]  [210/312]  eta: 0:01:22  lr: 0.003568  min_lr: 0.003568  loss: 3.8350 (3.6825)  weight_decay: 0.0500 (0.0500)  time: 0.8590  data: 0.3653  max mem: 41794
Epoch: [79]  [220/312]  eta: 0:01:12  lr: 0.003568  min_lr: 0.003568  loss: 3.8950 (3.6887)  weight_decay: 0.0500 (0.0500)  time: 0.6652  data: 0.1730  max mem: 41794
Epoch: [79]  [230/312]  eta: 0:01:05  lr: 0.003567  min_lr: 0.003567  loss: 3.8950 (3.6848)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1887  max mem: 41794
Epoch: [79]  [240/312]  eta: 0:00:57  lr: 0.003567  min_lr: 0.003567  loss: 3.9219 (3.6914)  weight_decay: 0.0500 (0.0500)  time: 0.8907  data: 0.4009  max mem: 41794
Epoch: [79]  [250/312]  eta: 0:00:48  lr: 0.003567  min_lr: 0.003567  loss: 3.8688 (3.6789)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.2137  max mem: 41794
Epoch: [79]  [260/312]  eta: 0:00:41  lr: 0.003566  min_lr: 0.003566  loss: 3.5972 (3.6759)  weight_decay: 0.0500 (0.0500)  time: 0.6787  data: 0.1836  max mem: 41794
Epoch: [79]  [270/312]  eta: 0:00:32  lr: 0.003566  min_lr: 0.003566  loss: 3.4903 (3.6684)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1827  max mem: 41794
Epoch: [79]  [280/312]  eta: 0:00:25  lr: 0.003565  min_lr: 0.003565  loss: 3.5872 (3.6701)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.1893  max mem: 41794
Epoch: [79]  [290/312]  eta: 0:00:17  lr: 0.003565  min_lr: 0.003565  loss: 3.8903 (3.6655)  weight_decay: 0.0500 (0.0500)  time: 0.8049  data: 0.3082  max mem: 41794
Epoch: [79]  [300/312]  eta: 0:00:09  lr: 0.003564  min_lr: 0.003564  loss: 3.8718 (3.6686)  weight_decay: 0.0500 (0.0500)  time: 0.6077  data: 0.1193  max mem: 41794
Epoch: [79]  [310/312]  eta: 0:00:01  lr: 0.003564  min_lr: 0.003564  loss: 3.8718 (3.6783)  weight_decay: 0.0500 (0.0500)  time: 0.4638  data: 0.0001  max mem: 41794
Epoch: [79]  [311/312]  eta: 0:00:00  lr: 0.003564  min_lr: 0.003564  loss: 3.8718 (3.6794)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [79] Total time: 0:03:58 (0.7651 s / it)
Averaged stats: lr: 0.003564  min_lr: 0.003564  loss: 3.8718 (3.6445)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3131 (1.3131)  acc1: 76.3021 (76.3021)  acc5: 92.3177 (92.3177)  time: 8.4662  data: 8.2644  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5262 (1.5344)  acc1: 67.9688 (68.1440)  acc5: 90.1042 (89.0240)  time: 1.0929  data: 0.9183  max mem: 41794
Test: Total time: 0:00:10 (1.1416 s / it)
* Acc@1 68.238 Acc@5 89.006 loss 1.534
Accuracy of the model on the 50000 test images: 68.2%
Max accuracy: 69.40%
Epoch: [80]  [  0/312]  eta: 1:22:35  lr: 0.003564  min_lr: 0.003564  loss: 3.9270 (3.9270)  weight_decay: 0.0500 (0.0500)  time: 15.8841  data: 11.5310  max mem: 41794
Epoch: [80]  [ 10/312]  eta: 0:11:05  lr: 0.003563  min_lr: 0.003563  loss: 3.8764 (3.6848)  weight_decay: 0.0500 (0.0500)  time: 2.2028  data: 1.3443  max mem: 41794
Epoch: [80]  [ 20/312]  eta: 0:07:27  lr: 0.003563  min_lr: 0.003563  loss: 3.8764 (3.8216)  weight_decay: 0.0500 (0.0500)  time: 0.8155  data: 0.1910  max mem: 41794
Epoch: [80]  [ 30/312]  eta: 0:05:41  lr: 0.003562  min_lr: 0.003562  loss: 3.7632 (3.6722)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0333  max mem: 41794
Epoch: [80]  [ 40/312]  eta: 0:04:46  lr: 0.003562  min_lr: 0.003562  loss: 3.7673 (3.7061)  weight_decay: 0.0500 (0.0500)  time: 0.5522  data: 0.0065  max mem: 41794
Epoch: [80]  [ 50/312]  eta: 0:04:23  lr: 0.003562  min_lr: 0.003562  loss: 3.8657 (3.7063)  weight_decay: 0.0500 (0.0500)  time: 0.6839  data: 0.0029  max mem: 41794
Epoch: [80]  [ 60/312]  eta: 0:03:52  lr: 0.003561  min_lr: 0.003561  loss: 3.8151 (3.6857)  weight_decay: 0.0500 (0.0500)  time: 0.6473  data: 0.0040  max mem: 41794
Epoch: [80]  [ 70/312]  eta: 0:03:44  lr: 0.003561  min_lr: 0.003561  loss: 3.7639 (3.6800)  weight_decay: 0.0500 (0.0500)  time: 0.7386  data: 0.0197  max mem: 41794
Epoch: [80]  [ 80/312]  eta: 0:03:29  lr: 0.003560  min_lr: 0.003560  loss: 3.8184 (3.6889)  weight_decay: 0.0500 (0.0500)  time: 0.8447  data: 0.0208  max mem: 41794
Epoch: [80]  [ 90/312]  eta: 0:03:14  lr: 0.003560  min_lr: 0.003560  loss: 3.8421 (3.6791)  weight_decay: 0.0500 (0.0500)  time: 0.6828  data: 0.0356  max mem: 41794
Epoch: [80]  [100/312]  eta: 0:03:04  lr: 0.003559  min_lr: 0.003559  loss: 3.4286 (3.6420)  weight_decay: 0.0500 (0.0500)  time: 0.7528  data: 0.0665  max mem: 41794
Epoch: [80]  [110/312]  eta: 0:02:49  lr: 0.003559  min_lr: 0.003559  loss: 3.3342 (3.6313)  weight_decay: 0.0500 (0.0500)  time: 0.6807  data: 0.0348  max mem: 41794
Epoch: [80]  [120/312]  eta: 0:02:42  lr: 0.003558  min_lr: 0.003558  loss: 3.6505 (3.6288)  weight_decay: 0.0500 (0.0500)  time: 0.7248  data: 0.0420  max mem: 41794
Epoch: [80]  [130/312]  eta: 0:02:34  lr: 0.003558  min_lr: 0.003558  loss: 3.7886 (3.6364)  weight_decay: 0.0500 (0.0500)  time: 0.8822  data: 0.0454  max mem: 41794
Epoch: [80]  [140/312]  eta: 0:02:21  lr: 0.003557  min_lr: 0.003557  loss: 3.7886 (3.6378)  weight_decay: 0.0500 (0.0500)  time: 0.6727  data: 0.0218  max mem: 41794
Epoch: [80]  [150/312]  eta: 0:02:13  lr: 0.003557  min_lr: 0.003557  loss: 3.6729 (3.6235)  weight_decay: 0.0500 (0.0500)  time: 0.6661  data: 0.0212  max mem: 41794
Epoch: [80]  [160/312]  eta: 0:02:03  lr: 0.003557  min_lr: 0.003557  loss: 3.0565 (3.5982)  weight_decay: 0.0500 (0.0500)  time: 0.7474  data: 0.0183  max mem: 41794
Epoch: [80]  [170/312]  eta: 0:01:54  lr: 0.003556  min_lr: 0.003556  loss: 3.3810 (3.6002)  weight_decay: 0.0500 (0.0500)  time: 0.6703  data: 0.0311  max mem: 41794
Epoch: [80]  [180/312]  eta: 0:01:47  lr: 0.003556  min_lr: 0.003556  loss: 3.5955 (3.6002)  weight_decay: 0.0500 (0.0500)  time: 0.7955  data: 0.0182  max mem: 41794
Epoch: [80]  [190/312]  eta: 0:01:37  lr: 0.003555  min_lr: 0.003555  loss: 3.5182 (3.5965)  weight_decay: 0.0500 (0.0500)  time: 0.7716  data: 0.0567  max mem: 41794
Epoch: [80]  [200/312]  eta: 0:01:28  lr: 0.003555  min_lr: 0.003555  loss: 3.5487 (3.5936)  weight_decay: 0.0500 (0.0500)  time: 0.6339  data: 0.0589  max mem: 41794
Epoch: [80]  [210/312]  eta: 0:01:21  lr: 0.003554  min_lr: 0.003554  loss: 3.7732 (3.6027)  weight_decay: 0.0500 (0.0500)  time: 0.7330  data: 0.0103  max mem: 41794
Epoch: [80]  [220/312]  eta: 0:01:12  lr: 0.003554  min_lr: 0.003554  loss: 3.9331 (3.6074)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.0431  max mem: 41794
Epoch: [80]  [230/312]  eta: 0:01:04  lr: 0.003553  min_lr: 0.003553  loss: 3.7253 (3.5939)  weight_decay: 0.0500 (0.0500)  time: 0.7404  data: 0.0955  max mem: 41794
Epoch: [80]  [240/312]  eta: 0:00:56  lr: 0.003553  min_lr: 0.003553  loss: 3.7253 (3.6047)  weight_decay: 0.0500 (0.0500)  time: 0.8205  data: 0.0603  max mem: 41794
Epoch: [80]  [250/312]  eta: 0:00:48  lr: 0.003553  min_lr: 0.003553  loss: 3.8786 (3.6175)  weight_decay: 0.0500 (0.0500)  time: 0.6683  data: 0.0214  max mem: 41794
Epoch: [80]  [260/312]  eta: 0:00:40  lr: 0.003552  min_lr: 0.003552  loss: 3.9107 (3.6318)  weight_decay: 0.0500 (0.0500)  time: 0.7180  data: 0.0330  max mem: 41794
Epoch: [80]  [270/312]  eta: 0:00:32  lr: 0.003552  min_lr: 0.003552  loss: 3.9831 (3.6399)  weight_decay: 0.0500 (0.0500)  time: 0.6551  data: 0.0165  max mem: 41794
Epoch: [80]  [280/312]  eta: 0:00:24  lr: 0.003551  min_lr: 0.003551  loss: 3.8110 (3.6422)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.0295  max mem: 41794
Epoch: [80]  [290/312]  eta: 0:00:17  lr: 0.003551  min_lr: 0.003551  loss: 3.7473 (3.6452)  weight_decay: 0.0500 (0.0500)  time: 0.8287  data: 0.0405  max mem: 41794
Epoch: [80]  [300/312]  eta: 0:00:09  lr: 0.003550  min_lr: 0.003550  loss: 3.7264 (3.6410)  weight_decay: 0.0500 (0.0500)  time: 0.6200  data: 0.0139  max mem: 41794
Epoch: [80]  [310/312]  eta: 0:00:01  lr: 0.003550  min_lr: 0.003550  loss: 3.6741 (3.6384)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [80]  [311/312]  eta: 0:00:00  lr: 0.003550  min_lr: 0.003550  loss: 3.6741 (3.6363)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [80] Total time: 0:03:57 (0.7611 s / it)
Averaged stats: lr: 0.003550  min_lr: 0.003550  loss: 3.6741 (3.6661)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2665 (1.2665)  acc1: 75.2604 (75.2604)  acc5: 92.7083 (92.7083)  time: 8.4789  data: 8.2662  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5930 (1.5033)  acc1: 67.4479 (68.3200)  acc5: 89.1927 (88.8640)  time: 1.0905  data: 0.9185  max mem: 41794
Test: Total time: 0:00:10 (1.1211 s / it)
* Acc@1 68.568 Acc@5 89.518 loss 1.484
Accuracy of the model on the 50000 test images: 68.6%
Max accuracy: 69.40%
Epoch: [81]  [  0/312]  eta: 1:18:05  lr: 0.003550  min_lr: 0.003550  loss: 4.3454 (4.3454)  weight_decay: 0.0500 (0.0500)  time: 15.0167  data: 12.6896  max mem: 41794
Epoch: [81]  [ 10/312]  eta: 0:10:30  lr: 0.003549  min_lr: 0.003549  loss: 3.9608 (3.8087)  weight_decay: 0.0500 (0.0500)  time: 2.0885  data: 1.2990  max mem: 41794
Epoch: [81]  [ 20/312]  eta: 0:07:02  lr: 0.003549  min_lr: 0.003549  loss: 3.8618 (3.7156)  weight_decay: 0.0500 (0.0500)  time: 0.7678  data: 0.0990  max mem: 41794
Epoch: [81]  [ 30/312]  eta: 0:05:22  lr: 0.003548  min_lr: 0.003548  loss: 3.7738 (3.7121)  weight_decay: 0.0500 (0.0500)  time: 0.6237  data: 0.0240  max mem: 41794
Epoch: [81]  [ 40/312]  eta: 0:04:45  lr: 0.003548  min_lr: 0.003548  loss: 3.6473 (3.6860)  weight_decay: 0.0500 (0.0500)  time: 0.6310  data: 0.0329  max mem: 41794
Epoch: [81]  [ 50/312]  eta: 0:04:22  lr: 0.003547  min_lr: 0.003547  loss: 3.5783 (3.6636)  weight_decay: 0.0500 (0.0500)  time: 0.7863  data: 0.1091  max mem: 41794
Epoch: [81]  [ 60/312]  eta: 0:03:52  lr: 0.003547  min_lr: 0.003547  loss: 3.6375 (3.6588)  weight_decay: 0.0500 (0.0500)  time: 0.6621  data: 0.0910  max mem: 41794
Epoch: [81]  [ 70/312]  eta: 0:03:46  lr: 0.003546  min_lr: 0.003546  loss: 3.7622 (3.6481)  weight_decay: 0.0500 (0.0500)  time: 0.7669  data: 0.1739  max mem: 41794
Epoch: [81]  [ 80/312]  eta: 0:03:35  lr: 0.003546  min_lr: 0.003546  loss: 3.8598 (3.6728)  weight_decay: 0.0500 (0.0500)  time: 0.9439  data: 0.2617  max mem: 41794
Epoch: [81]  [ 90/312]  eta: 0:03:15  lr: 0.003546  min_lr: 0.003546  loss: 3.8728 (3.6834)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.0979  max mem: 41794
Epoch: [81]  [100/312]  eta: 0:03:07  lr: 0.003545  min_lr: 0.003545  loss: 3.7912 (3.6618)  weight_decay: 0.0500 (0.0500)  time: 0.7002  data: 0.1281  max mem: 41794
Epoch: [81]  [110/312]  eta: 0:02:51  lr: 0.003545  min_lr: 0.003545  loss: 3.5921 (3.6383)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1281  max mem: 41794
Epoch: [81]  [120/312]  eta: 0:02:41  lr: 0.003544  min_lr: 0.003544  loss: 3.5337 (3.6338)  weight_decay: 0.0500 (0.0500)  time: 0.6319  data: 0.1366  max mem: 41794
Epoch: [81]  [130/312]  eta: 0:02:33  lr: 0.003544  min_lr: 0.003544  loss: 3.7099 (3.6418)  weight_decay: 0.0500 (0.0500)  time: 0.8240  data: 0.2799  max mem: 41794
Epoch: [81]  [140/312]  eta: 0:02:20  lr: 0.003543  min_lr: 0.003543  loss: 3.5960 (3.6338)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.1439  max mem: 41794
Epoch: [81]  [150/312]  eta: 0:02:13  lr: 0.003543  min_lr: 0.003543  loss: 3.5084 (3.6181)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1602  max mem: 41794
Epoch: [81]  [160/312]  eta: 0:02:05  lr: 0.003542  min_lr: 0.003542  loss: 3.7202 (3.6314)  weight_decay: 0.0500 (0.0500)  time: 0.8864  data: 0.3051  max mem: 41794
Epoch: [81]  [170/312]  eta: 0:01:54  lr: 0.003542  min_lr: 0.003542  loss: 3.9000 (3.6527)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.1455  max mem: 41794
Epoch: [81]  [180/312]  eta: 0:01:47  lr: 0.003541  min_lr: 0.003541  loss: 3.8229 (3.6537)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.1512  max mem: 41794
Epoch: [81]  [190/312]  eta: 0:01:36  lr: 0.003541  min_lr: 0.003541  loss: 3.6616 (3.6424)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.1512  max mem: 41794
Epoch: [81]  [200/312]  eta: 0:01:29  lr: 0.003541  min_lr: 0.003541  loss: 3.3642 (3.6299)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.1574  max mem: 41794
Epoch: [81]  [210/312]  eta: 0:01:21  lr: 0.003540  min_lr: 0.003540  loss: 3.7939 (3.6381)  weight_decay: 0.0500 (0.0500)  time: 0.8595  data: 0.3306  max mem: 41794
Epoch: [81]  [220/312]  eta: 0:01:12  lr: 0.003540  min_lr: 0.003540  loss: 3.9338 (3.6375)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.1739  max mem: 41794
Epoch: [81]  [230/312]  eta: 0:01:04  lr: 0.003539  min_lr: 0.003539  loss: 3.3134 (3.6231)  weight_decay: 0.0500 (0.0500)  time: 0.6929  data: 0.1559  max mem: 41794
Epoch: [81]  [240/312]  eta: 0:00:57  lr: 0.003539  min_lr: 0.003539  loss: 3.2924 (3.6177)  weight_decay: 0.0500 (0.0500)  time: 0.8667  data: 0.3031  max mem: 41794
Epoch: [81]  [250/312]  eta: 0:00:48  lr: 0.003538  min_lr: 0.003538  loss: 3.6355 (3.6113)  weight_decay: 0.0500 (0.0500)  time: 0.6665  data: 0.1478  max mem: 41794
Epoch: [81]  [260/312]  eta: 0:00:40  lr: 0.003538  min_lr: 0.003538  loss: 3.3085 (3.5940)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.1805  max mem: 41794
Epoch: [81]  [270/312]  eta: 0:00:32  lr: 0.003537  min_lr: 0.003537  loss: 3.1998 (3.5882)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1819  max mem: 41794
Epoch: [81]  [280/312]  eta: 0:00:24  lr: 0.003537  min_lr: 0.003537  loss: 3.6721 (3.5928)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.1424  max mem: 41794
Epoch: [81]  [290/312]  eta: 0:00:17  lr: 0.003536  min_lr: 0.003536  loss: 3.8228 (3.5894)  weight_decay: 0.0500 (0.0500)  time: 0.7893  data: 0.2643  max mem: 41794
Epoch: [81]  [300/312]  eta: 0:00:09  lr: 0.003536  min_lr: 0.003536  loss: 3.8228 (3.5922)  weight_decay: 0.0500 (0.0500)  time: 0.6081  data: 0.1237  max mem: 41794
Epoch: [81]  [310/312]  eta: 0:00:01  lr: 0.003535  min_lr: 0.003535  loss: 3.8313 (3.5958)  weight_decay: 0.0500 (0.0500)  time: 0.4669  data: 0.0001  max mem: 41794
Epoch: [81]  [311/312]  eta: 0:00:00  lr: 0.003535  min_lr: 0.003535  loss: 3.8279 (3.5962)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [81] Total time: 0:03:57 (0.7611 s / it)
Averaged stats: lr: 0.003535  min_lr: 0.003535  loss: 3.8279 (3.6274)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:09  loss: 1.3956 (1.3956)  acc1: 75.3906 (75.3906)  acc5: 91.6667 (91.6667)  time: 7.6827  data: 7.4696  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6712 (1.5874)  acc1: 67.8385 (68.1120)  acc5: 88.8021 (88.8800)  time: 1.0176  data: 0.8459  max mem: 41794
Test: Total time: 0:00:09 (1.0287 s / it)
* Acc@1 68.188 Acc@5 89.008 loss 1.594
Accuracy of the model on the 50000 test images: 68.2%
Max accuracy: 69.40%
Epoch: [82]  [  0/312]  eta: 1:24:40  lr: 0.003535  min_lr: 0.003535  loss: 3.1391 (3.1391)  weight_decay: 0.0500 (0.0500)  time: 16.2841  data: 12.1133  max mem: 41794
Epoch: [82]  [ 10/312]  eta: 0:11:04  lr: 0.003535  min_lr: 0.003535  loss: 3.8336 (3.7486)  weight_decay: 0.0500 (0.0500)  time: 2.2011  data: 1.2682  max mem: 41794
Epoch: [82]  [ 20/312]  eta: 0:07:20  lr: 0.003534  min_lr: 0.003534  loss: 3.6204 (3.5806)  weight_decay: 0.0500 (0.0500)  time: 0.7693  data: 0.1028  max mem: 41794
Epoch: [82]  [ 30/312]  eta: 0:05:45  lr: 0.003534  min_lr: 0.003534  loss: 3.8499 (3.7192)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.0596  max mem: 41794
Epoch: [82]  [ 40/312]  eta: 0:04:47  lr: 0.003534  min_lr: 0.003534  loss: 3.8658 (3.6929)  weight_decay: 0.0500 (0.0500)  time: 0.5821  data: 0.0510  max mem: 41794
Epoch: [82]  [ 50/312]  eta: 0:04:24  lr: 0.003533  min_lr: 0.003533  loss: 3.7597 (3.6826)  weight_decay: 0.0500 (0.0500)  time: 0.6719  data: 0.0239  max mem: 41794
Epoch: [82]  [ 60/312]  eta: 0:03:53  lr: 0.003533  min_lr: 0.003533  loss: 3.6628 (3.6373)  weight_decay: 0.0500 (0.0500)  time: 0.6584  data: 0.0252  max mem: 41794
Epoch: [82]  [ 70/312]  eta: 0:03:45  lr: 0.003532  min_lr: 0.003532  loss: 3.6412 (3.6323)  weight_decay: 0.0500 (0.0500)  time: 0.7292  data: 0.0991  max mem: 41794
Epoch: [82]  [ 80/312]  eta: 0:03:29  lr: 0.003532  min_lr: 0.003532  loss: 3.7454 (3.6234)  weight_decay: 0.0500 (0.0500)  time: 0.8321  data: 0.1334  max mem: 41794
Epoch: [82]  [ 90/312]  eta: 0:03:12  lr: 0.003531  min_lr: 0.003531  loss: 3.7601 (3.6073)  weight_decay: 0.0500 (0.0500)  time: 0.6360  data: 0.0697  max mem: 41794
Epoch: [82]  [100/312]  eta: 0:03:01  lr: 0.003531  min_lr: 0.003531  loss: 3.2714 (3.5808)  weight_decay: 0.0500 (0.0500)  time: 0.6660  data: 0.1082  max mem: 41794
Epoch: [82]  [110/312]  eta: 0:02:48  lr: 0.003530  min_lr: 0.003530  loss: 3.3248 (3.5733)  weight_decay: 0.0500 (0.0500)  time: 0.6803  data: 0.1316  max mem: 41794
Epoch: [82]  [120/312]  eta: 0:02:40  lr: 0.003530  min_lr: 0.003530  loss: 3.7022 (3.5849)  weight_decay: 0.0500 (0.0500)  time: 0.7331  data: 0.1972  max mem: 41794
Epoch: [82]  [130/312]  eta: 0:02:30  lr: 0.003529  min_lr: 0.003529  loss: 3.8601 (3.5970)  weight_decay: 0.0500 (0.0500)  time: 0.7991  data: 0.2041  max mem: 41794
Epoch: [82]  [140/312]  eta: 0:02:19  lr: 0.003529  min_lr: 0.003529  loss: 3.8601 (3.6119)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1363  max mem: 41794
Epoch: [82]  [150/312]  eta: 0:02:12  lr: 0.003528  min_lr: 0.003528  loss: 4.1379 (3.6260)  weight_decay: 0.0500 (0.0500)  time: 0.7639  data: 0.2188  max mem: 41794
Epoch: [82]  [160/312]  eta: 0:02:03  lr: 0.003528  min_lr: 0.003528  loss: 3.7597 (3.6274)  weight_decay: 0.0500 (0.0500)  time: 0.7828  data: 0.1626  max mem: 41794
Epoch: [82]  [170/312]  eta: 0:01:54  lr: 0.003528  min_lr: 0.003528  loss: 3.7597 (3.6177)  weight_decay: 0.0500 (0.0500)  time: 0.6892  data: 0.1270  max mem: 41794
Epoch: [82]  [180/312]  eta: 0:01:45  lr: 0.003527  min_lr: 0.003527  loss: 3.8140 (3.6114)  weight_decay: 0.0500 (0.0500)  time: 0.7177  data: 0.2132  max mem: 41794
Epoch: [82]  [190/312]  eta: 0:01:36  lr: 0.003527  min_lr: 0.003527  loss: 3.5044 (3.6072)  weight_decay: 0.0500 (0.0500)  time: 0.6453  data: 0.1363  max mem: 41794
Epoch: [82]  [200/312]  eta: 0:01:28  lr: 0.003526  min_lr: 0.003526  loss: 3.5856 (3.6104)  weight_decay: 0.0500 (0.0500)  time: 0.6943  data: 0.1861  max mem: 41794
Epoch: [82]  [210/312]  eta: 0:01:20  lr: 0.003526  min_lr: 0.003526  loss: 3.4610 (3.6060)  weight_decay: 0.0500 (0.0500)  time: 0.8492  data: 0.3456  max mem: 41794
Epoch: [82]  [220/312]  eta: 0:01:11  lr: 0.003525  min_lr: 0.003525  loss: 3.6791 (3.6166)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.1926  max mem: 41794
Epoch: [82]  [230/312]  eta: 0:01:04  lr: 0.003525  min_lr: 0.003525  loss: 3.6791 (3.6135)  weight_decay: 0.0500 (0.0500)  time: 0.6692  data: 0.1762  max mem: 41794
Epoch: [82]  [240/312]  eta: 0:00:56  lr: 0.003524  min_lr: 0.003524  loss: 3.6614 (3.6144)  weight_decay: 0.0500 (0.0500)  time: 0.8026  data: 0.3080  max mem: 41794
Epoch: [82]  [250/312]  eta: 0:00:47  lr: 0.003524  min_lr: 0.003524  loss: 3.7607 (3.6155)  weight_decay: 0.0500 (0.0500)  time: 0.6809  data: 0.1930  max mem: 41794
Epoch: [82]  [260/312]  eta: 0:00:40  lr: 0.003523  min_lr: 0.003523  loss: 3.5949 (3.6187)  weight_decay: 0.0500 (0.0500)  time: 0.7485  data: 0.2596  max mem: 41794
Epoch: [82]  [270/312]  eta: 0:00:32  lr: 0.003523  min_lr: 0.003523  loss: 3.9259 (3.6318)  weight_decay: 0.0500 (0.0500)  time: 0.6888  data: 0.1996  max mem: 41794
Epoch: [82]  [280/312]  eta: 0:00:24  lr: 0.003522  min_lr: 0.003522  loss: 4.0104 (3.6323)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.1845  max mem: 41794
Epoch: [82]  [290/312]  eta: 0:00:17  lr: 0.003522  min_lr: 0.003522  loss: 3.9810 (3.6393)  weight_decay: 0.0500 (0.0500)  time: 0.8664  data: 0.3758  max mem: 41794
Epoch: [82]  [300/312]  eta: 0:00:09  lr: 0.003521  min_lr: 0.003521  loss: 3.3775 (3.6214)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.1917  max mem: 41794
Epoch: [82]  [310/312]  eta: 0:00:01  lr: 0.003521  min_lr: 0.003521  loss: 3.3063 (3.6175)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [82]  [311/312]  eta: 0:00:00  lr: 0.003521  min_lr: 0.003521  loss: 3.3063 (3.6177)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [82] Total time: 0:03:57 (0.7601 s / it)
Averaged stats: lr: 0.003521  min_lr: 0.003521  loss: 3.3063 (3.6328)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.2944 (1.2944)  acc1: 74.7396 (74.7396)  acc5: 91.2760 (91.2760)  time: 7.9496  data: 7.7368  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6054 (1.5613)  acc1: 69.2708 (69.0560)  acc5: 89.5833 (89.2000)  time: 1.0349  data: 0.8597  max mem: 41794
Test: Total time: 0:00:09 (1.0493 s / it)
* Acc@1 68.842 Acc@5 89.380 loss 1.566
Accuracy of the model on the 50000 test images: 68.8%
Max accuracy: 69.40%
Epoch: [83]  [  0/312]  eta: 1:23:29  lr: 0.003521  min_lr: 0.003521  loss: 3.9259 (3.9259)  weight_decay: 0.0500 (0.0500)  time: 16.0563  data: 15.2708  max mem: 41794
Epoch: [83]  [ 10/312]  eta: 0:10:58  lr: 0.003520  min_lr: 0.003520  loss: 3.4563 (3.3153)  weight_decay: 0.0500 (0.0500)  time: 2.1808  data: 1.3889  max mem: 41794
Epoch: [83]  [ 20/312]  eta: 0:07:11  lr: 0.003520  min_lr: 0.003520  loss: 3.5890 (3.4886)  weight_decay: 0.0500 (0.0500)  time: 0.7496  data: 0.0951  max mem: 41794
Epoch: [83]  [ 30/312]  eta: 0:05:29  lr: 0.003520  min_lr: 0.003520  loss: 3.6801 (3.5334)  weight_decay: 0.0500 (0.0500)  time: 0.6129  data: 0.1007  max mem: 41794
Epoch: [83]  [ 40/312]  eta: 0:04:46  lr: 0.003519  min_lr: 0.003519  loss: 3.6801 (3.5217)  weight_decay: 0.0500 (0.0500)  time: 0.6072  data: 0.0802  max mem: 41794
Epoch: [83]  [ 50/312]  eta: 0:04:22  lr: 0.003519  min_lr: 0.003519  loss: 3.7322 (3.5361)  weight_decay: 0.0500 (0.0500)  time: 0.7424  data: 0.2162  max mem: 41794
Epoch: [83]  [ 60/312]  eta: 0:03:51  lr: 0.003518  min_lr: 0.003518  loss: 3.4356 (3.5023)  weight_decay: 0.0500 (0.0500)  time: 0.6399  data: 0.1426  max mem: 41794
Epoch: [83]  [ 70/312]  eta: 0:03:41  lr: 0.003518  min_lr: 0.003518  loss: 3.4893 (3.5291)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.1586  max mem: 41794
Epoch: [83]  [ 80/312]  eta: 0:03:29  lr: 0.003517  min_lr: 0.003517  loss: 3.6883 (3.5221)  weight_decay: 0.0500 (0.0500)  time: 0.8649  data: 0.2544  max mem: 41794
Epoch: [83]  [ 90/312]  eta: 0:03:15  lr: 0.003517  min_lr: 0.003517  loss: 3.4788 (3.5061)  weight_decay: 0.0500 (0.0500)  time: 0.7727  data: 0.1425  max mem: 41794
Epoch: [83]  [100/312]  eta: 0:03:07  lr: 0.003516  min_lr: 0.003516  loss: 3.5309 (3.5021)  weight_decay: 0.0500 (0.0500)  time: 0.7998  data: 0.1142  max mem: 41794
Epoch: [83]  [110/312]  eta: 0:02:51  lr: 0.003516  min_lr: 0.003516  loss: 3.6685 (3.5111)  weight_decay: 0.0500 (0.0500)  time: 0.6988  data: 0.0739  max mem: 41794
Epoch: [83]  [120/312]  eta: 0:02:43  lr: 0.003515  min_lr: 0.003515  loss: 3.7083 (3.5079)  weight_decay: 0.0500 (0.0500)  time: 0.6918  data: 0.0625  max mem: 41794
Epoch: [83]  [130/312]  eta: 0:02:35  lr: 0.003515  min_lr: 0.003515  loss: 3.7342 (3.5293)  weight_decay: 0.0500 (0.0500)  time: 0.8705  data: 0.0886  max mem: 41794
Epoch: [83]  [140/312]  eta: 0:02:22  lr: 0.003514  min_lr: 0.003514  loss: 3.8382 (3.5356)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.0336  max mem: 41794
Epoch: [83]  [150/312]  eta: 0:02:15  lr: 0.003514  min_lr: 0.003514  loss: 3.5808 (3.5303)  weight_decay: 0.0500 (0.0500)  time: 0.7192  data: 0.0491  max mem: 41794
Epoch: [83]  [160/312]  eta: 0:02:05  lr: 0.003513  min_lr: 0.003513  loss: 3.4591 (3.5272)  weight_decay: 0.0500 (0.0500)  time: 0.8032  data: 0.0524  max mem: 41794
Epoch: [83]  [170/312]  eta: 0:01:55  lr: 0.003513  min_lr: 0.003513  loss: 3.4314 (3.5240)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.0350  max mem: 41794
Epoch: [83]  [180/312]  eta: 0:01:47  lr: 0.003512  min_lr: 0.003512  loss: 3.6935 (3.5322)  weight_decay: 0.0500 (0.0500)  time: 0.7278  data: 0.0578  max mem: 41794
Epoch: [83]  [190/312]  eta: 0:01:37  lr: 0.003512  min_lr: 0.003512  loss: 3.8226 (3.5460)  weight_decay: 0.0500 (0.0500)  time: 0.6455  data: 0.0296  max mem: 41794
Epoch: [83]  [200/312]  eta: 0:01:29  lr: 0.003512  min_lr: 0.003512  loss: 3.7310 (3.5497)  weight_decay: 0.0500 (0.0500)  time: 0.6534  data: 0.0432  max mem: 41794
Epoch: [83]  [210/312]  eta: 0:01:21  lr: 0.003511  min_lr: 0.003511  loss: 3.6423 (3.5507)  weight_decay: 0.0500 (0.0500)  time: 0.8132  data: 0.1032  max mem: 41794
Epoch: [83]  [220/312]  eta: 0:01:12  lr: 0.003511  min_lr: 0.003511  loss: 3.5414 (3.5477)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.0652  max mem: 41794
Epoch: [83]  [230/312]  eta: 0:01:04  lr: 0.003510  min_lr: 0.003510  loss: 3.7084 (3.5503)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.0691  max mem: 41794
Epoch: [83]  [240/312]  eta: 0:00:56  lr: 0.003510  min_lr: 0.003510  loss: 3.6376 (3.5500)  weight_decay: 0.0500 (0.0500)  time: 0.8096  data: 0.0661  max mem: 41794
Epoch: [83]  [250/312]  eta: 0:00:48  lr: 0.003509  min_lr: 0.003509  loss: 3.3903 (3.5455)  weight_decay: 0.0500 (0.0500)  time: 0.6604  data: 0.0467  max mem: 41794
Epoch: [83]  [260/312]  eta: 0:00:40  lr: 0.003509  min_lr: 0.003509  loss: 3.4199 (3.5470)  weight_decay: 0.0500 (0.0500)  time: 0.7698  data: 0.1233  max mem: 41794
Epoch: [83]  [270/312]  eta: 0:00:32  lr: 0.003508  min_lr: 0.003508  loss: 3.7920 (3.5477)  weight_decay: 0.0500 (0.0500)  time: 0.7296  data: 0.0783  max mem: 41794
Epoch: [83]  [280/312]  eta: 0:00:24  lr: 0.003508  min_lr: 0.003508  loss: 3.7851 (3.5594)  weight_decay: 0.0500 (0.0500)  time: 0.6434  data: 0.0448  max mem: 41794
Epoch: [83]  [290/312]  eta: 0:00:17  lr: 0.003507  min_lr: 0.003507  loss: 3.8612 (3.5624)  weight_decay: 0.0500 (0.0500)  time: 0.8004  data: 0.0867  max mem: 41794
Epoch: [83]  [300/312]  eta: 0:00:09  lr: 0.003507  min_lr: 0.003507  loss: 3.8334 (3.5640)  weight_decay: 0.0500 (0.0500)  time: 0.6475  data: 0.0433  max mem: 41794
Epoch: [83]  [310/312]  eta: 0:00:01  lr: 0.003506  min_lr: 0.003506  loss: 3.4995 (3.5601)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0001  max mem: 41794
Epoch: [83]  [311/312]  eta: 0:00:00  lr: 0.003506  min_lr: 0.003506  loss: 3.4995 (3.5610)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [83] Total time: 0:03:58 (0.7643 s / it)
Averaged stats: lr: 0.003506  min_lr: 0.003506  loss: 3.4995 (3.6352)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.2435 (1.2435)  acc1: 73.9583 (73.9583)  acc5: 92.8385 (92.8385)  time: 8.4122  data: 8.2028  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6629 (1.5464)  acc1: 65.7552 (67.3920)  acc5: 88.0208 (88.7200)  time: 1.0826  data: 0.9115  max mem: 41794
Test: Total time: 0:00:09 (1.0993 s / it)
* Acc@1 67.826 Acc@5 88.840 loss 1.543
Accuracy of the model on the 50000 test images: 67.8%
Max accuracy: 69.40%
Epoch: [84]  [  0/312]  eta: 1:23:31  lr: 0.003506  min_lr: 0.003506  loss: 3.4262 (3.4262)  weight_decay: 0.0500 (0.0500)  time: 16.0622  data: 14.4664  max mem: 41794
Epoch: [84]  [ 10/312]  eta: 0:11:46  lr: 0.003506  min_lr: 0.003506  loss: 3.8749 (3.6708)  weight_decay: 0.0500 (0.0500)  time: 2.3380  data: 1.3351  max mem: 41794
Epoch: [84]  [ 20/312]  eta: 0:07:14  lr: 0.003505  min_lr: 0.003505  loss: 3.8007 (3.6316)  weight_decay: 0.0500 (0.0500)  time: 0.7594  data: 0.0294  max mem: 41794
Epoch: [84]  [ 30/312]  eta: 0:05:32  lr: 0.003505  min_lr: 0.003505  loss: 3.7139 (3.5992)  weight_decay: 0.0500 (0.0500)  time: 0.5412  data: 0.0207  max mem: 41794
Epoch: [84]  [ 40/312]  eta: 0:04:36  lr: 0.003504  min_lr: 0.003504  loss: 3.7660 (3.6277)  weight_decay: 0.0500 (0.0500)  time: 0.5181  data: 0.0067  max mem: 41794
Epoch: [84]  [ 50/312]  eta: 0:04:10  lr: 0.003504  min_lr: 0.003504  loss: 3.8818 (3.6829)  weight_decay: 0.0500 (0.0500)  time: 0.6091  data: 0.0560  max mem: 41794
Epoch: [84]  [ 60/312]  eta: 0:03:41  lr: 0.003503  min_lr: 0.003503  loss: 3.8952 (3.6843)  weight_decay: 0.0500 (0.0500)  time: 0.6058  data: 0.0548  max mem: 41794
Epoch: [84]  [ 70/312]  eta: 0:03:32  lr: 0.003503  min_lr: 0.003503  loss: 3.8348 (3.6979)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1417  max mem: 41794
Epoch: [84]  [ 80/312]  eta: 0:03:23  lr: 0.003502  min_lr: 0.003502  loss: 3.8491 (3.7325)  weight_decay: 0.0500 (0.0500)  time: 0.8576  data: 0.2930  max mem: 41794
Epoch: [84]  [ 90/312]  eta: 0:03:06  lr: 0.003502  min_lr: 0.003502  loss: 3.7554 (3.7040)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.1571  max mem: 41794
Epoch: [84]  [100/312]  eta: 0:02:58  lr: 0.003502  min_lr: 0.003502  loss: 3.6262 (3.7083)  weight_decay: 0.0500 (0.0500)  time: 0.7116  data: 0.1527  max mem: 41794
Epoch: [84]  [110/312]  eta: 0:02:44  lr: 0.003501  min_lr: 0.003501  loss: 3.5347 (3.6845)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.1561  max mem: 41794
Epoch: [84]  [120/312]  eta: 0:02:37  lr: 0.003501  min_lr: 0.003501  loss: 3.7584 (3.6902)  weight_decay: 0.0500 (0.0500)  time: 0.6931  data: 0.1585  max mem: 41794
Epoch: [84]  [130/312]  eta: 0:02:29  lr: 0.003500  min_lr: 0.003500  loss: 3.7965 (3.6869)  weight_decay: 0.0500 (0.0500)  time: 0.8666  data: 0.3039  max mem: 41794
Epoch: [84]  [140/312]  eta: 0:02:17  lr: 0.003500  min_lr: 0.003500  loss: 3.8131 (3.6938)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.1568  max mem: 41794
Epoch: [84]  [150/312]  eta: 0:02:10  lr: 0.003499  min_lr: 0.003499  loss: 3.6398 (3.6799)  weight_decay: 0.0500 (0.0500)  time: 0.7084  data: 0.1541  max mem: 41794
Epoch: [84]  [160/312]  eta: 0:02:02  lr: 0.003499  min_lr: 0.003499  loss: 3.7128 (3.6838)  weight_decay: 0.0500 (0.0500)  time: 0.8756  data: 0.2816  max mem: 41794
Epoch: [84]  [170/312]  eta: 0:01:52  lr: 0.003498  min_lr: 0.003498  loss: 3.9008 (3.6891)  weight_decay: 0.0500 (0.0500)  time: 0.6844  data: 0.1356  max mem: 41794
Epoch: [84]  [180/312]  eta: 0:01:44  lr: 0.003498  min_lr: 0.003498  loss: 3.8540 (3.6902)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.1331  max mem: 41794
Epoch: [84]  [190/312]  eta: 0:01:35  lr: 0.003497  min_lr: 0.003497  loss: 3.6575 (3.6860)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1308  max mem: 41794
Epoch: [84]  [200/312]  eta: 0:01:27  lr: 0.003497  min_lr: 0.003497  loss: 3.4283 (3.6775)  weight_decay: 0.0500 (0.0500)  time: 0.6655  data: 0.1275  max mem: 41794
Epoch: [84]  [210/312]  eta: 0:01:20  lr: 0.003496  min_lr: 0.003496  loss: 3.9624 (3.6865)  weight_decay: 0.0500 (0.0500)  time: 0.8548  data: 0.2732  max mem: 41794
Epoch: [84]  [220/312]  eta: 0:01:11  lr: 0.003496  min_lr: 0.003496  loss: 3.8489 (3.6803)  weight_decay: 0.0500 (0.0500)  time: 0.6900  data: 0.1463  max mem: 41794
Epoch: [84]  [230/312]  eta: 0:01:03  lr: 0.003495  min_lr: 0.003495  loss: 3.7553 (3.6854)  weight_decay: 0.0500 (0.0500)  time: 0.7007  data: 0.1287  max mem: 41794
Epoch: [84]  [240/312]  eta: 0:00:56  lr: 0.003495  min_lr: 0.003495  loss: 3.7553 (3.6824)  weight_decay: 0.0500 (0.0500)  time: 0.8691  data: 0.2273  max mem: 41794
Epoch: [84]  [250/312]  eta: 0:00:47  lr: 0.003494  min_lr: 0.003494  loss: 3.7626 (3.6838)  weight_decay: 0.0500 (0.0500)  time: 0.6694  data: 0.1006  max mem: 41794
Epoch: [84]  [260/312]  eta: 0:00:40  lr: 0.003494  min_lr: 0.003494  loss: 3.8262 (3.6853)  weight_decay: 0.0500 (0.0500)  time: 0.6804  data: 0.1296  max mem: 41794
Epoch: [84]  [270/312]  eta: 0:00:32  lr: 0.003493  min_lr: 0.003493  loss: 3.6114 (3.6693)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.1283  max mem: 41794
Epoch: [84]  [280/312]  eta: 0:00:24  lr: 0.003493  min_lr: 0.003493  loss: 3.4318 (3.6674)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.1181  max mem: 41794
Epoch: [84]  [290/312]  eta: 0:00:16  lr: 0.003492  min_lr: 0.003492  loss: 3.6133 (3.6645)  weight_decay: 0.0500 (0.0500)  time: 0.8704  data: 0.2724  max mem: 41794
Epoch: [84]  [300/312]  eta: 0:00:09  lr: 0.003492  min_lr: 0.003492  loss: 3.7862 (3.6667)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1548  max mem: 41794
Epoch: [84]  [310/312]  eta: 0:00:01  lr: 0.003492  min_lr: 0.003492  loss: 3.8783 (3.6616)  weight_decay: 0.0500 (0.0500)  time: 0.4693  data: 0.0001  max mem: 41794
Epoch: [84]  [311/312]  eta: 0:00:00  lr: 0.003491  min_lr: 0.003491  loss: 3.8783 (3.6631)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [84] Total time: 0:03:56 (0.7571 s / it)
Averaged stats: lr: 0.003491  min_lr: 0.003491  loss: 3.8783 (3.6192)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.3076 (1.3076)  acc1: 78.1250 (78.1250)  acc5: 92.4479 (92.4479)  time: 8.1572  data: 7.9521  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6266 (1.5619)  acc1: 67.9688 (68.4800)  acc5: 90.1042 (89.3280)  time: 1.0688  data: 0.8836  max mem: 41794
Test: Total time: 0:00:09 (1.0997 s / it)
* Acc@1 69.184 Acc@5 89.520 loss 1.550
Accuracy of the model on the 50000 test images: 69.2%
Max accuracy: 69.40%
Epoch: [85]  [  0/312]  eta: 1:23:30  lr: 0.003491  min_lr: 0.003491  loss: 3.3534 (3.3534)  weight_decay: 0.0500 (0.0500)  time: 16.0595  data: 12.6217  max mem: 41794
Epoch: [85]  [ 10/312]  eta: 0:11:16  lr: 0.003491  min_lr: 0.003491  loss: 3.6106 (3.6427)  weight_decay: 0.0500 (0.0500)  time: 2.2415  data: 1.4192  max mem: 41794
Epoch: [85]  [ 20/312]  eta: 0:07:48  lr: 0.003490  min_lr: 0.003490  loss: 3.7449 (3.7612)  weight_decay: 0.0500 (0.0500)  time: 0.8802  data: 0.1954  max mem: 41794
Epoch: [85]  [ 30/312]  eta: 0:05:59  lr: 0.003490  min_lr: 0.003490  loss: 3.8771 (3.7454)  weight_decay: 0.0500 (0.0500)  time: 0.7453  data: 0.0549  max mem: 41794
Epoch: [85]  [ 40/312]  eta: 0:04:55  lr: 0.003489  min_lr: 0.003489  loss: 3.8481 (3.7160)  weight_decay: 0.0500 (0.0500)  time: 0.5475  data: 0.0125  max mem: 41794
Epoch: [85]  [ 50/312]  eta: 0:04:21  lr: 0.003489  min_lr: 0.003489  loss: 3.9001 (3.7284)  weight_decay: 0.0500 (0.0500)  time: 0.5686  data: 0.0089  max mem: 41794
Epoch: [85]  [ 60/312]  eta: 0:03:50  lr: 0.003489  min_lr: 0.003489  loss: 3.7614 (3.7085)  weight_decay: 0.0500 (0.0500)  time: 0.5586  data: 0.0079  max mem: 41794
Epoch: [85]  [ 70/312]  eta: 0:03:37  lr: 0.003488  min_lr: 0.003488  loss: 3.6413 (3.7026)  weight_decay: 0.0500 (0.0500)  time: 0.6428  data: 0.0366  max mem: 41794
Epoch: [85]  [ 80/312]  eta: 0:03:24  lr: 0.003488  min_lr: 0.003488  loss: 3.7647 (3.6799)  weight_decay: 0.0500 (0.0500)  time: 0.7877  data: 0.0732  max mem: 41794
Epoch: [85]  [ 90/312]  eta: 0:03:09  lr: 0.003487  min_lr: 0.003487  loss: 3.6867 (3.6569)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.0692  max mem: 41794
Epoch: [85]  [100/312]  eta: 0:03:00  lr: 0.003487  min_lr: 0.003487  loss: 3.4475 (3.6423)  weight_decay: 0.0500 (0.0500)  time: 0.7192  data: 0.0840  max mem: 41794
Epoch: [85]  [110/312]  eta: 0:02:45  lr: 0.003486  min_lr: 0.003486  loss: 3.4471 (3.6292)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.0544  max mem: 41794
Epoch: [85]  [120/312]  eta: 0:02:38  lr: 0.003486  min_lr: 0.003486  loss: 3.3694 (3.5995)  weight_decay: 0.0500 (0.0500)  time: 0.7060  data: 0.0822  max mem: 41794
Epoch: [85]  [130/312]  eta: 0:02:31  lr: 0.003485  min_lr: 0.003485  loss: 3.5788 (3.6077)  weight_decay: 0.0500 (0.0500)  time: 0.9000  data: 0.1715  max mem: 41794
Epoch: [85]  [140/312]  eta: 0:02:18  lr: 0.003485  min_lr: 0.003485  loss: 3.8359 (3.6330)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.0899  max mem: 41794
Epoch: [85]  [150/312]  eta: 0:02:11  lr: 0.003484  min_lr: 0.003484  loss: 3.9906 (3.6472)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.0985  max mem: 41794
Epoch: [85]  [160/312]  eta: 0:02:02  lr: 0.003484  min_lr: 0.003484  loss: 3.9906 (3.6606)  weight_decay: 0.0500 (0.0500)  time: 0.7840  data: 0.1254  max mem: 41794
Epoch: [85]  [170/312]  eta: 0:01:53  lr: 0.003483  min_lr: 0.003483  loss: 3.6296 (3.6577)  weight_decay: 0.0500 (0.0500)  time: 0.7094  data: 0.0921  max mem: 41794
Epoch: [85]  [180/312]  eta: 0:01:46  lr: 0.003483  min_lr: 0.003483  loss: 3.6296 (3.6605)  weight_decay: 0.0500 (0.0500)  time: 0.7954  data: 0.1288  max mem: 41794
Epoch: [85]  [190/312]  eta: 0:01:36  lr: 0.003482  min_lr: 0.003482  loss: 3.8349 (3.6633)  weight_decay: 0.0500 (0.0500)  time: 0.6809  data: 0.0645  max mem: 41794
Epoch: [85]  [200/312]  eta: 0:01:28  lr: 0.003482  min_lr: 0.003482  loss: 3.6717 (3.6644)  weight_decay: 0.0500 (0.0500)  time: 0.6602  data: 0.0495  max mem: 41794
Epoch: [85]  [210/312]  eta: 0:01:21  lr: 0.003481  min_lr: 0.003481  loss: 3.7563 (3.6695)  weight_decay: 0.0500 (0.0500)  time: 0.8644  data: 0.1386  max mem: 41794
Epoch: [85]  [220/312]  eta: 0:01:11  lr: 0.003481  min_lr: 0.003481  loss: 3.5762 (3.6651)  weight_decay: 0.0500 (0.0500)  time: 0.7040  data: 0.0934  max mem: 41794
Epoch: [85]  [230/312]  eta: 0:01:04  lr: 0.003480  min_lr: 0.003480  loss: 3.3612 (3.6633)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.0628  max mem: 41794
Epoch: [85]  [240/312]  eta: 0:00:56  lr: 0.003480  min_lr: 0.003480  loss: 3.4665 (3.6490)  weight_decay: 0.0500 (0.0500)  time: 0.8204  data: 0.1069  max mem: 41794
Epoch: [85]  [250/312]  eta: 0:00:48  lr: 0.003479  min_lr: 0.003479  loss: 3.5383 (3.6505)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.0632  max mem: 41794
Epoch: [85]  [260/312]  eta: 0:00:40  lr: 0.003479  min_lr: 0.003479  loss: 3.6834 (3.6519)  weight_decay: 0.0500 (0.0500)  time: 0.7436  data: 0.0930  max mem: 41794
Epoch: [85]  [270/312]  eta: 0:00:32  lr: 0.003478  min_lr: 0.003478  loss: 3.8478 (3.6591)  weight_decay: 0.0500 (0.0500)  time: 0.6900  data: 0.0823  max mem: 41794
Epoch: [85]  [280/312]  eta: 0:00:24  lr: 0.003478  min_lr: 0.003478  loss: 3.8389 (3.6571)  weight_decay: 0.0500 (0.0500)  time: 0.6543  data: 0.0596  max mem: 41794
Epoch: [85]  [290/312]  eta: 0:00:17  lr: 0.003477  min_lr: 0.003477  loss: 3.6737 (3.6619)  weight_decay: 0.0500 (0.0500)  time: 0.8750  data: 0.1629  max mem: 41794
Epoch: [85]  [300/312]  eta: 0:00:09  lr: 0.003477  min_lr: 0.003477  loss: 3.8777 (3.6626)  weight_decay: 0.0500 (0.0500)  time: 0.7076  data: 0.1080  max mem: 41794
Epoch: [85]  [310/312]  eta: 0:00:01  lr: 0.003476  min_lr: 0.003476  loss: 3.8412 (3.6660)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0001  max mem: 41794
Epoch: [85]  [311/312]  eta: 0:00:00  lr: 0.003476  min_lr: 0.003476  loss: 3.8160 (3.6619)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0001  max mem: 41794
Epoch: [85] Total time: 0:03:58 (0.7639 s / it)
Averaged stats: lr: 0.003476  min_lr: 0.003476  loss: 3.8160 (3.6367)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1975 (1.1975)  acc1: 75.9115 (75.9115)  acc5: 92.7083 (92.7083)  time: 8.5137  data: 8.3005  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4524 (1.3593)  acc1: 69.6615 (70.1120)  acc5: 90.1042 (90.0160)  time: 1.0948  data: 0.9224  max mem: 41794
Test: Total time: 0:00:10 (1.1350 s / it)
* Acc@1 70.278 Acc@5 90.022 loss 1.361
Accuracy of the model on the 50000 test images: 70.3%
Max accuracy: 70.28%
Epoch: [86]  [  0/312]  eta: 1:22:40  lr: 0.003476  min_lr: 0.003476  loss: 2.3769 (2.3769)  weight_decay: 0.0500 (0.0500)  time: 15.8989  data: 11.9347  max mem: 41794
Epoch: [86]  [ 10/312]  eta: 0:11:46  lr: 0.003476  min_lr: 0.003476  loss: 3.4130 (3.3750)  weight_decay: 0.0500 (0.0500)  time: 2.3384  data: 1.3296  max mem: 41794
Epoch: [86]  [ 20/312]  eta: 0:07:27  lr: 0.003475  min_lr: 0.003475  loss: 3.6337 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.8129  data: 0.1417  max mem: 41794
Epoch: [86]  [ 30/312]  eta: 0:05:45  lr: 0.003475  min_lr: 0.003475  loss: 3.8172 (3.6595)  weight_decay: 0.0500 (0.0500)  time: 0.6144  data: 0.0512  max mem: 41794
Epoch: [86]  [ 40/312]  eta: 0:04:45  lr: 0.003474  min_lr: 0.003474  loss: 3.7836 (3.6800)  weight_decay: 0.0500 (0.0500)  time: 0.5449  data: 0.0457  max mem: 41794
Epoch: [86]  [ 50/312]  eta: 0:04:09  lr: 0.003474  min_lr: 0.003474  loss: 3.7836 (3.6897)  weight_decay: 0.0500 (0.0500)  time: 0.5247  data: 0.0118  max mem: 41794
Epoch: [86]  [ 60/312]  eta: 0:03:41  lr: 0.003473  min_lr: 0.003473  loss: 3.6831 (3.6439)  weight_decay: 0.0500 (0.0500)  time: 0.5298  data: 0.0162  max mem: 41794
Epoch: [86]  [ 70/312]  eta: 0:03:32  lr: 0.003473  min_lr: 0.003473  loss: 3.6688 (3.6750)  weight_decay: 0.0500 (0.0500)  time: 0.6930  data: 0.1604  max mem: 41794
Epoch: [86]  [ 80/312]  eta: 0:03:22  lr: 0.003472  min_lr: 0.003472  loss: 3.6750 (3.6608)  weight_decay: 0.0500 (0.0500)  time: 0.8546  data: 0.2715  max mem: 41794
Epoch: [86]  [ 90/312]  eta: 0:03:07  lr: 0.003472  min_lr: 0.003472  loss: 3.7600 (3.6636)  weight_decay: 0.0500 (0.0500)  time: 0.7336  data: 0.1795  max mem: 41794
Epoch: [86]  [100/312]  eta: 0:03:00  lr: 0.003472  min_lr: 0.003472  loss: 3.8378 (3.6450)  weight_decay: 0.0500 (0.0500)  time: 0.7607  data: 0.1901  max mem: 41794
Epoch: [86]  [110/312]  eta: 0:02:45  lr: 0.003471  min_lr: 0.003471  loss: 3.6622 (3.6435)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.1281  max mem: 41794
Epoch: [86]  [120/312]  eta: 0:02:37  lr: 0.003471  min_lr: 0.003471  loss: 3.7999 (3.6585)  weight_decay: 0.0500 (0.0500)  time: 0.6683  data: 0.1080  max mem: 41794
Epoch: [86]  [130/312]  eta: 0:02:28  lr: 0.003470  min_lr: 0.003470  loss: 3.9797 (3.6627)  weight_decay: 0.0500 (0.0500)  time: 0.8129  data: 0.2017  max mem: 41794
Epoch: [86]  [140/312]  eta: 0:02:17  lr: 0.003470  min_lr: 0.003470  loss: 3.7187 (3.6596)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.1247  max mem: 41794
Epoch: [86]  [150/312]  eta: 0:02:10  lr: 0.003469  min_lr: 0.003469  loss: 3.6252 (3.6485)  weight_decay: 0.0500 (0.0500)  time: 0.7460  data: 0.1469  max mem: 41794
Epoch: [86]  [160/312]  eta: 0:02:02  lr: 0.003469  min_lr: 0.003469  loss: 3.6252 (3.6508)  weight_decay: 0.0500 (0.0500)  time: 0.8535  data: 0.1326  max mem: 41794
Epoch: [86]  [170/312]  eta: 0:01:52  lr: 0.003468  min_lr: 0.003468  loss: 3.8095 (3.6539)  weight_decay: 0.0500 (0.0500)  time: 0.6870  data: 0.0599  max mem: 41794
Epoch: [86]  [180/312]  eta: 0:01:45  lr: 0.003468  min_lr: 0.003468  loss: 3.6794 (3.6440)  weight_decay: 0.0500 (0.0500)  time: 0.7236  data: 0.1040  max mem: 41794
Epoch: [86]  [190/312]  eta: 0:01:35  lr: 0.003467  min_lr: 0.003467  loss: 3.4233 (3.6379)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.0608  max mem: 41794
Epoch: [86]  [200/312]  eta: 0:01:27  lr: 0.003467  min_lr: 0.003467  loss: 3.7285 (3.6443)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.0625  max mem: 41794
Epoch: [86]  [210/312]  eta: 0:01:20  lr: 0.003466  min_lr: 0.003466  loss: 3.8010 (3.6395)  weight_decay: 0.0500 (0.0500)  time: 0.8592  data: 0.1291  max mem: 41794
Epoch: [86]  [220/312]  eta: 0:01:11  lr: 0.003466  min_lr: 0.003466  loss: 3.8010 (3.6482)  weight_decay: 0.0500 (0.0500)  time: 0.6874  data: 0.0673  max mem: 41794
Epoch: [86]  [230/312]  eta: 0:01:03  lr: 0.003465  min_lr: 0.003465  loss: 3.7658 (3.6423)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.0731  max mem: 41794
Epoch: [86]  [240/312]  eta: 0:00:56  lr: 0.003465  min_lr: 0.003465  loss: 3.4726 (3.6341)  weight_decay: 0.0500 (0.0500)  time: 0.8246  data: 0.0782  max mem: 41794
Epoch: [86]  [250/312]  eta: 0:00:47  lr: 0.003464  min_lr: 0.003464  loss: 3.8060 (3.6390)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.0587  max mem: 41794
Epoch: [86]  [260/312]  eta: 0:00:40  lr: 0.003464  min_lr: 0.003464  loss: 3.9602 (3.6537)  weight_decay: 0.0500 (0.0500)  time: 0.7443  data: 0.1120  max mem: 41794
Epoch: [86]  [270/312]  eta: 0:00:32  lr: 0.003463  min_lr: 0.003463  loss: 3.8701 (3.6541)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.0591  max mem: 41794
Epoch: [86]  [280/312]  eta: 0:00:24  lr: 0.003463  min_lr: 0.003463  loss: 3.8404 (3.6631)  weight_decay: 0.0500 (0.0500)  time: 0.6835  data: 0.0570  max mem: 41794
Epoch: [86]  [290/312]  eta: 0:00:17  lr: 0.003462  min_lr: 0.003462  loss: 3.8559 (3.6670)  weight_decay: 0.0500 (0.0500)  time: 0.8644  data: 0.1259  max mem: 41794
Epoch: [86]  [300/312]  eta: 0:00:09  lr: 0.003462  min_lr: 0.003462  loss: 3.7470 (3.6643)  weight_decay: 0.0500 (0.0500)  time: 0.6609  data: 0.0693  max mem: 41794
Epoch: [86]  [310/312]  eta: 0:00:01  lr: 0.003461  min_lr: 0.003461  loss: 3.7470 (3.6647)  weight_decay: 0.0500 (0.0500)  time: 0.4669  data: 0.0001  max mem: 41794
Epoch: [86]  [311/312]  eta: 0:00:00  lr: 0.003461  min_lr: 0.003461  loss: 3.7470 (3.6608)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [86] Total time: 0:03:57 (0.7605 s / it)
Averaged stats: lr: 0.003461  min_lr: 0.003461  loss: 3.7470 (3.6130)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.1337 (1.1337)  acc1: 77.8646 (77.8646)  acc5: 92.8385 (92.8385)  time: 8.6189  data: 8.4056  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5297 (1.4211)  acc1: 67.8385 (68.9280)  acc5: 88.0208 (88.8480)  time: 1.1058  data: 0.9341  max mem: 41794
Test: Total time: 0:00:10 (1.1493 s / it)
* Acc@1 68.650 Acc@5 89.276 loss 1.417
Accuracy of the model on the 50000 test images: 68.7%
Max accuracy: 70.28%
Epoch: [87]  [  0/312]  eta: 1:20:25  lr: 0.003461  min_lr: 0.003461  loss: 3.4431 (3.4431)  weight_decay: 0.0500 (0.0500)  time: 15.4666  data: 13.7616  max mem: 41794
Epoch: [87]  [ 10/312]  eta: 0:11:13  lr: 0.003461  min_lr: 0.003461  loss: 3.7694 (3.8954)  weight_decay: 0.0500 (0.0500)  time: 2.2317  data: 1.3052  max mem: 41794
Epoch: [87]  [ 20/312]  eta: 0:07:06  lr: 0.003460  min_lr: 0.003460  loss: 3.7694 (3.8105)  weight_decay: 0.0500 (0.0500)  time: 0.7611  data: 0.0756  max mem: 41794
Epoch: [87]  [ 30/312]  eta: 0:05:26  lr: 0.003460  min_lr: 0.003460  loss: 3.5636 (3.7145)  weight_decay: 0.0500 (0.0500)  time: 0.5676  data: 0.0489  max mem: 41794
Epoch: [87]  [ 40/312]  eta: 0:04:40  lr: 0.003459  min_lr: 0.003459  loss: 3.5518 (3.6682)  weight_decay: 0.0500 (0.0500)  time: 0.5801  data: 0.0789  max mem: 41794
Epoch: [87]  [ 50/312]  eta: 0:04:19  lr: 0.003459  min_lr: 0.003459  loss: 3.7403 (3.6828)  weight_decay: 0.0500 (0.0500)  time: 0.7309  data: 0.2016  max mem: 41794
Epoch: [87]  [ 60/312]  eta: 0:03:48  lr: 0.003458  min_lr: 0.003458  loss: 3.7652 (3.6787)  weight_decay: 0.0500 (0.0500)  time: 0.6546  data: 0.1290  max mem: 41794
Epoch: [87]  [ 70/312]  eta: 0:03:38  lr: 0.003458  min_lr: 0.003458  loss: 3.5418 (3.6359)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.1668  max mem: 41794
Epoch: [87]  [ 80/312]  eta: 0:03:30  lr: 0.003457  min_lr: 0.003457  loss: 3.2830 (3.6046)  weight_decay: 0.0500 (0.0500)  time: 0.9070  data: 0.3856  max mem: 41794
Epoch: [87]  [ 90/312]  eta: 0:03:11  lr: 0.003457  min_lr: 0.003457  loss: 3.2830 (3.5758)  weight_decay: 0.0500 (0.0500)  time: 0.7151  data: 0.2224  max mem: 41794
Epoch: [87]  [100/312]  eta: 0:03:02  lr: 0.003456  min_lr: 0.003456  loss: 3.2263 (3.5481)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.1662  max mem: 41794
Epoch: [87]  [110/312]  eta: 0:02:47  lr: 0.003456  min_lr: 0.003456  loss: 3.7060 (3.5785)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.1682  max mem: 41794
Epoch: [87]  [120/312]  eta: 0:02:39  lr: 0.003455  min_lr: 0.003455  loss: 3.8208 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.1959  max mem: 41794
Epoch: [87]  [130/312]  eta: 0:02:32  lr: 0.003455  min_lr: 0.003455  loss: 3.7300 (3.6010)  weight_decay: 0.0500 (0.0500)  time: 0.8905  data: 0.3897  max mem: 41794
Epoch: [87]  [140/312]  eta: 0:02:19  lr: 0.003454  min_lr: 0.003454  loss: 3.7072 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1964  max mem: 41794
Epoch: [87]  [150/312]  eta: 0:02:12  lr: 0.003454  min_lr: 0.003454  loss: 3.7688 (3.5948)  weight_decay: 0.0500 (0.0500)  time: 0.6930  data: 0.1969  max mem: 41794
Epoch: [87]  [160/312]  eta: 0:02:04  lr: 0.003453  min_lr: 0.003453  loss: 3.5852 (3.5925)  weight_decay: 0.0500 (0.0500)  time: 0.8450  data: 0.3591  max mem: 41794
Epoch: [87]  [170/312]  eta: 0:01:53  lr: 0.003453  min_lr: 0.003453  loss: 3.5416 (3.5885)  weight_decay: 0.0500 (0.0500)  time: 0.6513  data: 0.1628  max mem: 41794
Epoch: [87]  [180/312]  eta: 0:01:46  lr: 0.003452  min_lr: 0.003452  loss: 3.8383 (3.6044)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.2101  max mem: 41794
Epoch: [87]  [190/312]  eta: 0:01:36  lr: 0.003452  min_lr: 0.003452  loss: 3.8383 (3.6016)  weight_decay: 0.0500 (0.0500)  time: 0.7013  data: 0.2115  max mem: 41794
Epoch: [87]  [200/312]  eta: 0:01:28  lr: 0.003451  min_lr: 0.003451  loss: 3.5512 (3.5913)  weight_decay: 0.0500 (0.0500)  time: 0.6528  data: 0.1633  max mem: 41794
Epoch: [87]  [210/312]  eta: 0:01:20  lr: 0.003451  min_lr: 0.003451  loss: 3.3953 (3.5848)  weight_decay: 0.0500 (0.0500)  time: 0.8349  data: 0.3346  max mem: 41794
Epoch: [87]  [220/312]  eta: 0:01:11  lr: 0.003450  min_lr: 0.003450  loss: 3.6596 (3.5933)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1734  max mem: 41794
Epoch: [87]  [230/312]  eta: 0:01:04  lr: 0.003450  min_lr: 0.003450  loss: 3.6596 (3.5935)  weight_decay: 0.0500 (0.0500)  time: 0.6960  data: 0.1955  max mem: 41794
Epoch: [87]  [240/312]  eta: 0:00:56  lr: 0.003449  min_lr: 0.003449  loss: 3.5184 (3.5938)  weight_decay: 0.0500 (0.0500)  time: 0.8352  data: 0.3428  max mem: 41794
Epoch: [87]  [250/312]  eta: 0:00:48  lr: 0.003449  min_lr: 0.003449  loss: 3.8319 (3.6026)  weight_decay: 0.0500 (0.0500)  time: 0.6694  data: 0.1493  max mem: 41794
Epoch: [87]  [260/312]  eta: 0:00:40  lr: 0.003448  min_lr: 0.003448  loss: 3.7774 (3.6041)  weight_decay: 0.0500 (0.0500)  time: 0.7298  data: 0.1582  max mem: 41794
Epoch: [87]  [270/312]  eta: 0:00:32  lr: 0.003448  min_lr: 0.003448  loss: 3.7774 (3.6013)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.1569  max mem: 41794
Epoch: [87]  [280/312]  eta: 0:00:24  lr: 0.003447  min_lr: 0.003447  loss: 3.7501 (3.6009)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1339  max mem: 41794
Epoch: [87]  [290/312]  eta: 0:00:17  lr: 0.003447  min_lr: 0.003447  loss: 3.6602 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.8277  data: 0.2374  max mem: 41794
Epoch: [87]  [300/312]  eta: 0:00:09  lr: 0.003446  min_lr: 0.003446  loss: 3.6602 (3.5981)  weight_decay: 0.0500 (0.0500)  time: 0.6226  data: 0.1039  max mem: 41794
Epoch: [87]  [310/312]  eta: 0:00:01  lr: 0.003446  min_lr: 0.003446  loss: 3.8513 (3.5985)  weight_decay: 0.0500 (0.0500)  time: 0.4849  data: 0.0001  max mem: 41794
Epoch: [87]  [311/312]  eta: 0:00:00  lr: 0.003446  min_lr: 0.003446  loss: 3.8410 (3.5993)  weight_decay: 0.0500 (0.0500)  time: 0.4675  data: 0.0001  max mem: 41794
Epoch: [87] Total time: 0:03:57 (0.7604 s / it)
Averaged stats: lr: 0.003446  min_lr: 0.003446  loss: 3.8410 (3.6332)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.1542 (1.1542)  acc1: 76.3021 (76.3021)  acc5: 93.7500 (93.7500)  time: 8.8199  data: 8.6079  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6529 (1.4916)  acc1: 65.8854 (68.4000)  acc5: 88.2812 (88.8640)  time: 1.1285  data: 0.9565  max mem: 41794
Test: Total time: 0:00:10 (1.1624 s / it)
* Acc@1 68.624 Acc@5 89.270 loss 1.493
Accuracy of the model on the 50000 test images: 68.6%
Max accuracy: 70.28%
Epoch: [88]  [  0/312]  eta: 1:24:33  lr: 0.003446  min_lr: 0.003446  loss: 2.5843 (2.5843)  weight_decay: 0.0500 (0.0500)  time: 16.2600  data: 15.7477  max mem: 41794
Epoch: [88]  [ 10/312]  eta: 0:11:31  lr: 0.003445  min_lr: 0.003445  loss: 3.2928 (3.3037)  weight_decay: 0.0500 (0.0500)  time: 2.2904  data: 1.4323  max mem: 41794
Epoch: [88]  [ 20/312]  eta: 0:07:32  lr: 0.003445  min_lr: 0.003445  loss: 3.4884 (3.4223)  weight_decay: 0.0500 (0.0500)  time: 0.8138  data: 0.0415  max mem: 41794
Epoch: [88]  [ 30/312]  eta: 0:05:45  lr: 0.003444  min_lr: 0.003444  loss: 3.7069 (3.5214)  weight_decay: 0.0500 (0.0500)  time: 0.6377  data: 0.0511  max mem: 41794
Epoch: [88]  [ 40/312]  eta: 0:04:45  lr: 0.003444  min_lr: 0.003444  loss: 3.8317 (3.5600)  weight_decay: 0.0500 (0.0500)  time: 0.5228  data: 0.0119  max mem: 41794
Epoch: [88]  [ 50/312]  eta: 0:04:22  lr: 0.003443  min_lr: 0.003443  loss: 3.7162 (3.5609)  weight_decay: 0.0500 (0.0500)  time: 0.6570  data: 0.0864  max mem: 41794
Epoch: [88]  [ 60/312]  eta: 0:03:51  lr: 0.003443  min_lr: 0.003443  loss: 3.5042 (3.5425)  weight_decay: 0.0500 (0.0500)  time: 0.6501  data: 0.0848  max mem: 41794
Epoch: [88]  [ 70/312]  eta: 0:03:38  lr: 0.003442  min_lr: 0.003442  loss: 3.7988 (3.5605)  weight_decay: 0.0500 (0.0500)  time: 0.6449  data: 0.0662  max mem: 41794
Epoch: [88]  [ 80/312]  eta: 0:03:24  lr: 0.003442  min_lr: 0.003442  loss: 3.9038 (3.5832)  weight_decay: 0.0500 (0.0500)  time: 0.7746  data: 0.0724  max mem: 41794
Epoch: [88]  [ 90/312]  eta: 0:03:09  lr: 0.003441  min_lr: 0.003441  loss: 3.9523 (3.6007)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.0669  max mem: 41794
Epoch: [88]  [100/312]  eta: 0:03:02  lr: 0.003441  min_lr: 0.003441  loss: 3.9240 (3.6232)  weight_decay: 0.0500 (0.0500)  time: 0.7728  data: 0.1308  max mem: 41794
Epoch: [88]  [110/312]  eta: 0:02:47  lr: 0.003440  min_lr: 0.003440  loss: 3.9091 (3.6095)  weight_decay: 0.0500 (0.0500)  time: 0.7026  data: 0.0740  max mem: 41794
Epoch: [88]  [120/312]  eta: 0:02:38  lr: 0.003440  min_lr: 0.003440  loss: 3.5800 (3.5971)  weight_decay: 0.0500 (0.0500)  time: 0.6569  data: 0.0538  max mem: 41794
Epoch: [88]  [130/312]  eta: 0:02:31  lr: 0.003439  min_lr: 0.003439  loss: 3.5353 (3.6040)  weight_decay: 0.0500 (0.0500)  time: 0.8643  data: 0.1083  max mem: 41794
Epoch: [88]  [140/312]  eta: 0:02:19  lr: 0.003439  min_lr: 0.003439  loss: 3.5353 (3.5976)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.0615  max mem: 41794
Epoch: [88]  [150/312]  eta: 0:02:11  lr: 0.003438  min_lr: 0.003438  loss: 3.5518 (3.6024)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.0664  max mem: 41794
Epoch: [88]  [160/312]  eta: 0:02:02  lr: 0.003438  min_lr: 0.003438  loss: 3.7138 (3.5956)  weight_decay: 0.0500 (0.0500)  time: 0.7762  data: 0.0632  max mem: 41794
Epoch: [88]  [170/312]  eta: 0:01:53  lr: 0.003437  min_lr: 0.003437  loss: 3.5200 (3.5928)  weight_decay: 0.0500 (0.0500)  time: 0.7232  data: 0.1001  max mem: 41794
Epoch: [88]  [180/312]  eta: 0:01:45  lr: 0.003437  min_lr: 0.003437  loss: 3.7407 (3.6031)  weight_decay: 0.0500 (0.0500)  time: 0.7645  data: 0.1722  max mem: 41794
Epoch: [88]  [190/312]  eta: 0:01:35  lr: 0.003436  min_lr: 0.003436  loss: 3.6247 (3.5920)  weight_decay: 0.0500 (0.0500)  time: 0.6399  data: 0.0727  max mem: 41794
Epoch: [88]  [200/312]  eta: 0:01:28  lr: 0.003436  min_lr: 0.003436  loss: 3.4639 (3.5996)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.0919  max mem: 41794
Epoch: [88]  [210/312]  eta: 0:01:21  lr: 0.003435  min_lr: 0.003435  loss: 3.8904 (3.6088)  weight_decay: 0.0500 (0.0500)  time: 0.8906  data: 0.1816  max mem: 41794
Epoch: [88]  [220/312]  eta: 0:01:11  lr: 0.003435  min_lr: 0.003435  loss: 3.6644 (3.5907)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0903  max mem: 41794
Epoch: [88]  [230/312]  eta: 0:01:04  lr: 0.003434  min_lr: 0.003434  loss: 3.3275 (3.5936)  weight_decay: 0.0500 (0.0500)  time: 0.7041  data: 0.0931  max mem: 41794
Epoch: [88]  [240/312]  eta: 0:00:56  lr: 0.003434  min_lr: 0.003434  loss: 3.7695 (3.5959)  weight_decay: 0.0500 (0.0500)  time: 0.8019  data: 0.0931  max mem: 41794
Epoch: [88]  [250/312]  eta: 0:00:48  lr: 0.003433  min_lr: 0.003433  loss: 3.7266 (3.6010)  weight_decay: 0.0500 (0.0500)  time: 0.6460  data: 0.0558  max mem: 41794
Epoch: [88]  [260/312]  eta: 0:00:40  lr: 0.003433  min_lr: 0.003433  loss: 3.7029 (3.5948)  weight_decay: 0.0500 (0.0500)  time: 0.7547  data: 0.1320  max mem: 41794
Epoch: [88]  [270/312]  eta: 0:00:32  lr: 0.003432  min_lr: 0.003432  loss: 3.7749 (3.5975)  weight_decay: 0.0500 (0.0500)  time: 0.7002  data: 0.0768  max mem: 41794
Epoch: [88]  [280/312]  eta: 0:00:24  lr: 0.003432  min_lr: 0.003432  loss: 3.5834 (3.5961)  weight_decay: 0.0500 (0.0500)  time: 0.6783  data: 0.0792  max mem: 41794
Epoch: [88]  [290/312]  eta: 0:00:17  lr: 0.003431  min_lr: 0.003431  loss: 3.5322 (3.5973)  weight_decay: 0.0500 (0.0500)  time: 0.8578  data: 0.1474  max mem: 41794
Epoch: [88]  [300/312]  eta: 0:00:09  lr: 0.003431  min_lr: 0.003431  loss: 3.6475 (3.5923)  weight_decay: 0.0500 (0.0500)  time: 0.6683  data: 0.0686  max mem: 41794
Epoch: [88]  [310/312]  eta: 0:00:01  lr: 0.003430  min_lr: 0.003430  loss: 3.4906 (3.5890)  weight_decay: 0.0500 (0.0500)  time: 0.4692  data: 0.0001  max mem: 41794
Epoch: [88]  [311/312]  eta: 0:00:00  lr: 0.003430  min_lr: 0.003430  loss: 3.4474 (3.5862)  weight_decay: 0.0500 (0.0500)  time: 0.4679  data: 0.0001  max mem: 41794
Epoch: [88] Total time: 0:03:57 (0.7626 s / it)
Averaged stats: lr: 0.003430  min_lr: 0.003430  loss: 3.4474 (3.6271)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:09  loss: 1.2001 (1.2001)  acc1: 77.2135 (77.2135)  acc5: 92.1875 (92.1875)  time: 7.7028  data: 7.5044  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5640 (1.4456)  acc1: 68.8802 (69.6640)  acc5: 88.9323 (89.7600)  time: 1.0635  data: 0.8935  max mem: 41794
Test: Total time: 0:00:09 (1.0767 s / it)
* Acc@1 69.526 Acc@5 89.850 loss 1.445
Accuracy of the model on the 50000 test images: 69.5%
Max accuracy: 70.28%
Epoch: [89]  [  0/312]  eta: 1:27:02  lr: 0.003430  min_lr: 0.003430  loss: 4.2882 (4.2882)  weight_decay: 0.0500 (0.0500)  time: 16.7388  data: 15.7346  max mem: 41794
Epoch: [89]  [ 10/312]  eta: 0:10:51  lr: 0.003430  min_lr: 0.003430  loss: 3.8591 (3.7577)  weight_decay: 0.0500 (0.0500)  time: 2.1569  data: 1.4311  max mem: 41794
Epoch: [89]  [ 20/312]  eta: 0:07:44  lr: 0.003429  min_lr: 0.003429  loss: 3.7445 (3.6575)  weight_decay: 0.0500 (0.0500)  time: 0.8321  data: 0.0746  max mem: 41794
Epoch: [89]  [ 30/312]  eta: 0:05:53  lr: 0.003429  min_lr: 0.003429  loss: 3.5526 (3.6251)  weight_decay: 0.0500 (0.0500)  time: 0.7586  data: 0.0854  max mem: 41794
Epoch: [89]  [ 40/312]  eta: 0:04:58  lr: 0.003428  min_lr: 0.003428  loss: 3.7640 (3.6719)  weight_decay: 0.0500 (0.0500)  time: 0.5784  data: 0.0130  max mem: 41794
Epoch: [89]  [ 50/312]  eta: 0:04:36  lr: 0.003428  min_lr: 0.003428  loss: 3.8211 (3.6618)  weight_decay: 0.0500 (0.0500)  time: 0.7426  data: 0.0134  max mem: 41794
Epoch: [89]  [ 60/312]  eta: 0:04:02  lr: 0.003427  min_lr: 0.003427  loss: 3.8418 (3.6798)  weight_decay: 0.0500 (0.0500)  time: 0.6929  data: 0.0142  max mem: 41794
Epoch: [89]  [ 70/312]  eta: 0:03:50  lr: 0.003427  min_lr: 0.003427  loss: 3.8418 (3.6699)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.0072  max mem: 41794
Epoch: [89]  [ 80/312]  eta: 0:03:39  lr: 0.003426  min_lr: 0.003426  loss: 3.8752 (3.6850)  weight_decay: 0.0500 (0.0500)  time: 0.8879  data: 0.0112  max mem: 41794
Epoch: [89]  [ 90/312]  eta: 0:03:22  lr: 0.003426  min_lr: 0.003426  loss: 3.8760 (3.7040)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.0105  max mem: 41794
Epoch: [89]  [100/312]  eta: 0:03:09  lr: 0.003425  min_lr: 0.003425  loss: 3.8990 (3.7132)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.0107  max mem: 41794
Epoch: [89]  [110/312]  eta: 0:02:54  lr: 0.003425  min_lr: 0.003425  loss: 3.9749 (3.7197)  weight_decay: 0.0500 (0.0500)  time: 0.6339  data: 0.0072  max mem: 41794
Epoch: [89]  [120/312]  eta: 0:02:46  lr: 0.003424  min_lr: 0.003424  loss: 3.7199 (3.7026)  weight_decay: 0.0500 (0.0500)  time: 0.7160  data: 0.0048  max mem: 41794
Epoch: [89]  [130/312]  eta: 0:02:36  lr: 0.003424  min_lr: 0.003424  loss: 3.5674 (3.6821)  weight_decay: 0.0500 (0.0500)  time: 0.8643  data: 0.0124  max mem: 41794
Epoch: [89]  [140/312]  eta: 0:02:24  lr: 0.003423  min_lr: 0.003423  loss: 3.4070 (3.6610)  weight_decay: 0.0500 (0.0500)  time: 0.6779  data: 0.0348  max mem: 41794
Epoch: [89]  [150/312]  eta: 0:02:15  lr: 0.003423  min_lr: 0.003423  loss: 3.5243 (3.6604)  weight_decay: 0.0500 (0.0500)  time: 0.6743  data: 0.0272  max mem: 41794
Epoch: [89]  [160/312]  eta: 0:02:07  lr: 0.003422  min_lr: 0.003422  loss: 3.6811 (3.6593)  weight_decay: 0.0500 (0.0500)  time: 0.8198  data: 0.0077  max mem: 41794
Epoch: [89]  [170/312]  eta: 0:01:57  lr: 0.003422  min_lr: 0.003422  loss: 3.5837 (3.6506)  weight_decay: 0.0500 (0.0500)  time: 0.7377  data: 0.0187  max mem: 41794
Epoch: [89]  [180/312]  eta: 0:01:48  lr: 0.003421  min_lr: 0.003421  loss: 3.6820 (3.6572)  weight_decay: 0.0500 (0.0500)  time: 0.7067  data: 0.0116  max mem: 41794
Epoch: [89]  [190/312]  eta: 0:01:38  lr: 0.003421  min_lr: 0.003421  loss: 3.6337 (3.6455)  weight_decay: 0.0500 (0.0500)  time: 0.6399  data: 0.0024  max mem: 41794
Epoch: [89]  [200/312]  eta: 0:01:31  lr: 0.003420  min_lr: 0.003420  loss: 3.8025 (3.6540)  weight_decay: 0.0500 (0.0500)  time: 0.7610  data: 0.0035  max mem: 41794
Epoch: [89]  [210/312]  eta: 0:01:23  lr: 0.003420  min_lr: 0.003420  loss: 3.8337 (3.6588)  weight_decay: 0.0500 (0.0500)  time: 0.9179  data: 0.0066  max mem: 41794
Epoch: [89]  [220/312]  eta: 0:01:13  lr: 0.003419  min_lr: 0.003419  loss: 3.5262 (3.6491)  weight_decay: 0.0500 (0.0500)  time: 0.6676  data: 0.0143  max mem: 41794
Epoch: [89]  [230/312]  eta: 0:01:06  lr: 0.003419  min_lr: 0.003419  loss: 3.4803 (3.6500)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.0095  max mem: 41794
Epoch: [89]  [240/312]  eta: 0:00:57  lr: 0.003418  min_lr: 0.003418  loss: 3.7390 (3.6518)  weight_decay: 0.0500 (0.0500)  time: 0.8225  data: 0.0025  max mem: 41794
Epoch: [89]  [250/312]  eta: 0:00:49  lr: 0.003418  min_lr: 0.003418  loss: 3.6679 (3.6458)  weight_decay: 0.0500 (0.0500)  time: 0.6671  data: 0.0101  max mem: 41794
Epoch: [89]  [260/312]  eta: 0:00:41  lr: 0.003417  min_lr: 0.003417  loss: 3.3448 (3.6316)  weight_decay: 0.0500 (0.0500)  time: 0.7050  data: 0.0086  max mem: 41794
Epoch: [89]  [270/312]  eta: 0:00:33  lr: 0.003416  min_lr: 0.003416  loss: 3.5025 (3.6349)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.0058  max mem: 41794
Epoch: [89]  [280/312]  eta: 0:00:25  lr: 0.003416  min_lr: 0.003416  loss: 3.8346 (3.6384)  weight_decay: 0.0500 (0.0500)  time: 0.6959  data: 0.0122  max mem: 41794
Epoch: [89]  [290/312]  eta: 0:00:17  lr: 0.003415  min_lr: 0.003415  loss: 3.6579 (3.6397)  weight_decay: 0.0500 (0.0500)  time: 0.7748  data: 0.0071  max mem: 41794
Epoch: [89]  [300/312]  eta: 0:00:09  lr: 0.003415  min_lr: 0.003415  loss: 3.6540 (3.6323)  weight_decay: 0.0500 (0.0500)  time: 0.5641  data: 0.0002  max mem: 41794
Epoch: [89]  [310/312]  eta: 0:00:01  lr: 0.003414  min_lr: 0.003414  loss: 3.8042 (3.6373)  weight_decay: 0.0500 (0.0500)  time: 0.4621  data: 0.0001  max mem: 41794
Epoch: [89]  [311/312]  eta: 0:00:00  lr: 0.003414  min_lr: 0.003414  loss: 3.6695 (3.6349)  weight_decay: 0.0500 (0.0500)  time: 0.4617  data: 0.0001  max mem: 41794
Epoch: [89] Total time: 0:04:01 (0.7725 s / it)
Averaged stats: lr: 0.003414  min_lr: 0.003414  loss: 3.6695 (3.6051)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.4731 (1.4731)  acc1: 73.1771 (73.1771)  acc5: 92.0573 (92.0573)  time: 8.7396  data: 8.5327  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7734 (1.7187)  acc1: 67.3177 (68.2720)  acc5: 90.3646 (88.7840)  time: 1.1194  data: 0.9482  max mem: 41794
Test: Total time: 0:00:10 (1.1304 s / it)
* Acc@1 68.494 Acc@5 88.958 loss 1.724
Accuracy of the model on the 50000 test images: 68.5%
Max accuracy: 70.28%
Epoch: [90]  [  0/312]  eta: 1:19:56  lr: 0.003414  min_lr: 0.003414  loss: 2.9332 (2.9332)  weight_decay: 0.0500 (0.0500)  time: 15.3726  data: 13.2747  max mem: 41794
Epoch: [90]  [ 10/312]  eta: 0:11:00  lr: 0.003414  min_lr: 0.003414  loss: 3.8732 (3.8834)  weight_decay: 0.0500 (0.0500)  time: 2.1863  data: 1.2613  max mem: 41794
Epoch: [90]  [ 20/312]  eta: 0:07:34  lr: 0.003413  min_lr: 0.003413  loss: 3.7052 (3.5958)  weight_decay: 0.0500 (0.0500)  time: 0.8659  data: 0.1296  max mem: 41794
Epoch: [90]  [ 30/312]  eta: 0:05:46  lr: 0.003413  min_lr: 0.003413  loss: 3.5395 (3.6084)  weight_decay: 0.0500 (0.0500)  time: 0.7020  data: 0.1018  max mem: 41794
Epoch: [90]  [ 40/312]  eta: 0:04:58  lr: 0.003412  min_lr: 0.003412  loss: 3.6852 (3.5833)  weight_decay: 0.0500 (0.0500)  time: 0.6115  data: 0.1001  max mem: 41794
Epoch: [90]  [ 50/312]  eta: 0:04:33  lr: 0.003412  min_lr: 0.003412  loss: 3.5312 (3.5697)  weight_decay: 0.0500 (0.0500)  time: 0.7556  data: 0.2348  max mem: 41794
Epoch: [90]  [ 60/312]  eta: 0:03:59  lr: 0.003411  min_lr: 0.003411  loss: 3.3425 (3.5139)  weight_decay: 0.0500 (0.0500)  time: 0.6552  data: 0.1372  max mem: 41794
Epoch: [90]  [ 70/312]  eta: 0:03:48  lr: 0.003411  min_lr: 0.003411  loss: 3.6932 (3.5727)  weight_decay: 0.0500 (0.0500)  time: 0.6936  data: 0.1334  max mem: 41794
Epoch: [90]  [ 80/312]  eta: 0:03:32  lr: 0.003410  min_lr: 0.003410  loss: 3.8789 (3.5616)  weight_decay: 0.0500 (0.0500)  time: 0.7999  data: 0.1335  max mem: 41794
Epoch: [90]  [ 90/312]  eta: 0:03:17  lr: 0.003410  min_lr: 0.003410  loss: 3.7144 (3.5711)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1007  max mem: 41794
Epoch: [90]  [100/312]  eta: 0:03:09  lr: 0.003409  min_lr: 0.003409  loss: 3.7144 (3.5854)  weight_decay: 0.0500 (0.0500)  time: 0.8148  data: 0.2189  max mem: 41794
Epoch: [90]  [110/312]  eta: 0:02:53  lr: 0.003409  min_lr: 0.003409  loss: 3.6077 (3.5777)  weight_decay: 0.0500 (0.0500)  time: 0.7196  data: 0.1189  max mem: 41794
Epoch: [90]  [120/312]  eta: 0:02:47  lr: 0.003408  min_lr: 0.003408  loss: 3.6077 (3.5713)  weight_decay: 0.0500 (0.0500)  time: 0.7705  data: 0.1040  max mem: 41794
Epoch: [90]  [130/312]  eta: 0:02:36  lr: 0.003408  min_lr: 0.003408  loss: 3.7114 (3.5766)  weight_decay: 0.0500 (0.0500)  time: 0.8779  data: 0.1065  max mem: 41794
Epoch: [90]  [140/312]  eta: 0:02:24  lr: 0.003407  min_lr: 0.003407  loss: 3.4579 (3.5659)  weight_decay: 0.0500 (0.0500)  time: 0.6445  data: 0.0404  max mem: 41794
Epoch: [90]  [150/312]  eta: 0:02:15  lr: 0.003407  min_lr: 0.003407  loss: 3.6851 (3.5840)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.0707  max mem: 41794
Epoch: [90]  [160/312]  eta: 0:02:07  lr: 0.003406  min_lr: 0.003406  loss: 3.8083 (3.5754)  weight_decay: 0.0500 (0.0500)  time: 0.8069  data: 0.0335  max mem: 41794
Epoch: [90]  [170/312]  eta: 0:01:56  lr: 0.003406  min_lr: 0.003406  loss: 3.6354 (3.5745)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.0403  max mem: 41794
Epoch: [90]  [180/312]  eta: 0:01:48  lr: 0.003405  min_lr: 0.003405  loss: 3.7488 (3.5911)  weight_decay: 0.0500 (0.0500)  time: 0.6770  data: 0.0459  max mem: 41794
Epoch: [90]  [190/312]  eta: 0:01:38  lr: 0.003405  min_lr: 0.003405  loss: 3.7749 (3.6004)  weight_decay: 0.0500 (0.0500)  time: 0.6687  data: 0.0407  max mem: 41794
Epoch: [90]  [200/312]  eta: 0:01:30  lr: 0.003404  min_lr: 0.003404  loss: 3.7461 (3.6056)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.0735  max mem: 41794
Epoch: [90]  [210/312]  eta: 0:01:22  lr: 0.003404  min_lr: 0.003404  loss: 3.5609 (3.5947)  weight_decay: 0.0500 (0.0500)  time: 0.8017  data: 0.0391  max mem: 41794
Epoch: [90]  [220/312]  eta: 0:01:13  lr: 0.003403  min_lr: 0.003403  loss: 3.4856 (3.5915)  weight_decay: 0.0500 (0.0500)  time: 0.7151  data: 0.0766  max mem: 41794
Epoch: [90]  [230/312]  eta: 0:01:05  lr: 0.003403  min_lr: 0.003403  loss: 3.7572 (3.5964)  weight_decay: 0.0500 (0.0500)  time: 0.6872  data: 0.0792  max mem: 41794
Epoch: [90]  [240/312]  eta: 0:00:57  lr: 0.003402  min_lr: 0.003402  loss: 3.7125 (3.5917)  weight_decay: 0.0500 (0.0500)  time: 0.7862  data: 0.0460  max mem: 41794
Epoch: [90]  [250/312]  eta: 0:00:48  lr: 0.003402  min_lr: 0.003402  loss: 3.5700 (3.5956)  weight_decay: 0.0500 (0.0500)  time: 0.7207  data: 0.0943  max mem: 41794
Epoch: [90]  [260/312]  eta: 0:00:41  lr: 0.003401  min_lr: 0.003401  loss: 3.6477 (3.5923)  weight_decay: 0.0500 (0.0500)  time: 0.6916  data: 0.0568  max mem: 41794
Epoch: [90]  [270/312]  eta: 0:00:32  lr: 0.003401  min_lr: 0.003401  loss: 3.4497 (3.5807)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.0624  max mem: 41794
Epoch: [90]  [280/312]  eta: 0:00:25  lr: 0.003400  min_lr: 0.003400  loss: 3.7303 (3.5898)  weight_decay: 0.0500 (0.0500)  time: 0.7015  data: 0.0951  max mem: 41794
Epoch: [90]  [290/312]  eta: 0:00:17  lr: 0.003400  min_lr: 0.003400  loss: 3.8506 (3.5958)  weight_decay: 0.0500 (0.0500)  time: 0.7320  data: 0.0411  max mem: 41794
Epoch: [90]  [300/312]  eta: 0:00:09  lr: 0.003399  min_lr: 0.003399  loss: 3.6325 (3.5964)  weight_decay: 0.0500 (0.0500)  time: 0.5807  data: 0.0105  max mem: 41794
Epoch: [90]  [310/312]  eta: 0:00:01  lr: 0.003399  min_lr: 0.003399  loss: 3.5642 (3.5926)  weight_decay: 0.0500 (0.0500)  time: 0.4714  data: 0.0077  max mem: 41794
Epoch: [90]  [311/312]  eta: 0:00:00  lr: 0.003398  min_lr: 0.003398  loss: 3.5642 (3.5900)  weight_decay: 0.0500 (0.0500)  time: 0.4706  data: 0.0077  max mem: 41794
Epoch: [90] Total time: 0:03:59 (0.7667 s / it)
Averaged stats: lr: 0.003398  min_lr: 0.003398  loss: 3.5642 (3.6104)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.3481 (1.3481)  acc1: 75.3906 (75.3906)  acc5: 92.1875 (92.1875)  time: 8.6920  data: 8.4791  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6974 (1.6013)  acc1: 67.0573 (67.6640)  acc5: 88.1510 (88.6560)  time: 1.1138  data: 0.9422  max mem: 41794
Test: Total time: 0:00:10 (1.1371 s / it)
* Acc@1 67.980 Acc@5 88.922 loss 1.596
Accuracy of the model on the 50000 test images: 68.0%
Max accuracy: 70.28%
Epoch: [91]  [  0/312]  eta: 1:22:46  lr: 0.003398  min_lr: 0.003398  loss: 3.4141 (3.4141)  weight_decay: 0.0500 (0.0500)  time: 15.9196  data: 15.4432  max mem: 41794
Epoch: [91]  [ 10/312]  eta: 0:11:40  lr: 0.003398  min_lr: 0.003398  loss: 3.7601 (3.5549)  weight_decay: 0.0500 (0.0500)  time: 2.3208  data: 1.4047  max mem: 41794
Epoch: [91]  [ 20/312]  eta: 0:07:30  lr: 0.003397  min_lr: 0.003397  loss: 3.6844 (3.5494)  weight_decay: 0.0500 (0.0500)  time: 0.8244  data: 0.0959  max mem: 41794
Epoch: [91]  [ 30/312]  eta: 0:05:41  lr: 0.003397  min_lr: 0.003397  loss: 3.5812 (3.5553)  weight_decay: 0.0500 (0.0500)  time: 0.5994  data: 0.0958  max mem: 41794
Epoch: [91]  [ 40/312]  eta: 0:04:43  lr: 0.003396  min_lr: 0.003396  loss: 3.6429 (3.5436)  weight_decay: 0.0500 (0.0500)  time: 0.5181  data: 0.0119  max mem: 41794
Epoch: [91]  [ 50/312]  eta: 0:04:17  lr: 0.003396  min_lr: 0.003396  loss: 3.4540 (3.5238)  weight_decay: 0.0500 (0.0500)  time: 0.6344  data: 0.1281  max mem: 41794
Epoch: [91]  [ 60/312]  eta: 0:03:48  lr: 0.003395  min_lr: 0.003395  loss: 3.4152 (3.5287)  weight_decay: 0.0500 (0.0500)  time: 0.6249  data: 0.1169  max mem: 41794
Epoch: [91]  [ 70/312]  eta: 0:03:38  lr: 0.003395  min_lr: 0.003395  loss: 3.7467 (3.5477)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.1825  max mem: 41794
Epoch: [91]  [ 80/312]  eta: 0:03:27  lr: 0.003394  min_lr: 0.003394  loss: 3.7850 (3.5522)  weight_decay: 0.0500 (0.0500)  time: 0.8513  data: 0.3549  max mem: 41794
Epoch: [91]  [ 90/312]  eta: 0:03:08  lr: 0.003394  min_lr: 0.003394  loss: 3.6464 (3.5447)  weight_decay: 0.0500 (0.0500)  time: 0.6678  data: 0.1730  max mem: 41794
Epoch: [91]  [100/312]  eta: 0:03:01  lr: 0.003393  min_lr: 0.003393  loss: 3.6464 (3.5561)  weight_decay: 0.0500 (0.0500)  time: 0.7088  data: 0.2051  max mem: 41794
Epoch: [91]  [110/312]  eta: 0:02:46  lr: 0.003393  min_lr: 0.003393  loss: 3.7530 (3.5640)  weight_decay: 0.0500 (0.0500)  time: 0.7016  data: 0.2051  max mem: 41794
Epoch: [91]  [120/312]  eta: 0:02:39  lr: 0.003392  min_lr: 0.003392  loss: 3.8211 (3.5824)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1968  max mem: 41794
Epoch: [91]  [130/312]  eta: 0:02:30  lr: 0.003392  min_lr: 0.003392  loss: 3.6158 (3.5763)  weight_decay: 0.0500 (0.0500)  time: 0.8608  data: 0.3668  max mem: 41794
Epoch: [91]  [140/312]  eta: 0:02:18  lr: 0.003391  min_lr: 0.003391  loss: 3.5509 (3.5740)  weight_decay: 0.0500 (0.0500)  time: 0.6663  data: 0.1708  max mem: 41794
Epoch: [91]  [150/312]  eta: 0:02:11  lr: 0.003391  min_lr: 0.003391  loss: 3.6224 (3.5823)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1781  max mem: 41794
Epoch: [91]  [160/312]  eta: 0:02:03  lr: 0.003390  min_lr: 0.003390  loss: 3.6482 (3.5822)  weight_decay: 0.0500 (0.0500)  time: 0.8640  data: 0.3605  max mem: 41794
Epoch: [91]  [170/312]  eta: 0:01:52  lr: 0.003390  min_lr: 0.003390  loss: 3.8840 (3.5980)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1870  max mem: 41794
Epoch: [91]  [180/312]  eta: 0:01:45  lr: 0.003389  min_lr: 0.003389  loss: 3.8842 (3.5881)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.2041  max mem: 41794
Epoch: [91]  [190/312]  eta: 0:01:35  lr: 0.003389  min_lr: 0.003389  loss: 3.7868 (3.5959)  weight_decay: 0.0500 (0.0500)  time: 0.6919  data: 0.2003  max mem: 41794
Epoch: [91]  [200/312]  eta: 0:01:28  lr: 0.003388  min_lr: 0.003388  loss: 3.7460 (3.5962)  weight_decay: 0.0500 (0.0500)  time: 0.6819  data: 0.1879  max mem: 41794
Epoch: [91]  [210/312]  eta: 0:01:20  lr: 0.003388  min_lr: 0.003388  loss: 3.7544 (3.5999)  weight_decay: 0.0500 (0.0500)  time: 0.8582  data: 0.3597  max mem: 41794
Epoch: [91]  [220/312]  eta: 0:01:11  lr: 0.003387  min_lr: 0.003387  loss: 3.7708 (3.6098)  weight_decay: 0.0500 (0.0500)  time: 0.6708  data: 0.1724  max mem: 41794
Epoch: [91]  [230/312]  eta: 0:01:04  lr: 0.003387  min_lr: 0.003387  loss: 3.8633 (3.6246)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1790  max mem: 41794
Epoch: [91]  [240/312]  eta: 0:00:56  lr: 0.003386  min_lr: 0.003386  loss: 3.8032 (3.6210)  weight_decay: 0.0500 (0.0500)  time: 0.8617  data: 0.3614  max mem: 41794
Epoch: [91]  [250/312]  eta: 0:00:47  lr: 0.003385  min_lr: 0.003385  loss: 3.7230 (3.6211)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.1830  max mem: 41794
Epoch: [91]  [260/312]  eta: 0:00:40  lr: 0.003385  min_lr: 0.003385  loss: 3.7371 (3.6219)  weight_decay: 0.0500 (0.0500)  time: 0.6921  data: 0.1914  max mem: 41794
Epoch: [91]  [270/312]  eta: 0:00:32  lr: 0.003384  min_lr: 0.003384  loss: 3.7158 (3.6296)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.1915  max mem: 41794
Epoch: [91]  [280/312]  eta: 0:00:24  lr: 0.003384  min_lr: 0.003384  loss: 3.6117 (3.6228)  weight_decay: 0.0500 (0.0500)  time: 0.6887  data: 0.1920  max mem: 41794
Epoch: [91]  [290/312]  eta: 0:00:17  lr: 0.003383  min_lr: 0.003383  loss: 3.5490 (3.6256)  weight_decay: 0.0500 (0.0500)  time: 0.8587  data: 0.3545  max mem: 41794
Epoch: [91]  [300/312]  eta: 0:00:09  lr: 0.003383  min_lr: 0.003383  loss: 3.6957 (3.6295)  weight_decay: 0.0500 (0.0500)  time: 0.6542  data: 0.1629  max mem: 41794
Epoch: [91]  [310/312]  eta: 0:00:01  lr: 0.003382  min_lr: 0.003382  loss: 3.8541 (3.6341)  weight_decay: 0.0500 (0.0500)  time: 0.4643  data: 0.0001  max mem: 41794
Epoch: [91]  [311/312]  eta: 0:00:00  lr: 0.003382  min_lr: 0.003382  loss: 3.8878 (3.6355)  weight_decay: 0.0500 (0.0500)  time: 0.4641  data: 0.0001  max mem: 41794
Epoch: [91] Total time: 0:03:57 (0.7597 s / it)
Averaged stats: lr: 0.003382  min_lr: 0.003382  loss: 3.8878 (3.6097)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.4276 (1.4276)  acc1: 74.7396 (74.7396)  acc5: 92.7083 (92.7083)  time: 8.6369  data: 8.4270  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7693 (1.6439)  acc1: 64.9740 (67.9840)  acc5: 88.4115 (88.9280)  time: 1.1077  data: 0.9364  max mem: 41794
Test: Total time: 0:00:10 (1.1458 s / it)
* Acc@1 68.392 Acc@5 89.064 loss 1.631
Accuracy of the model on the 50000 test images: 68.4%
Max accuracy: 70.28%
Epoch: [92]  [  0/312]  eta: 1:22:21  lr: 0.003382  min_lr: 0.003382  loss: 4.0927 (4.0927)  weight_decay: 0.0500 (0.0500)  time: 15.8369  data: 14.6235  max mem: 41794
Epoch: [92]  [ 10/312]  eta: 0:11:26  lr: 0.003382  min_lr: 0.003382  loss: 3.6630 (3.4642)  weight_decay: 0.0500 (0.0500)  time: 2.2738  data: 1.3302  max mem: 41794
Epoch: [92]  [ 20/312]  eta: 0:07:27  lr: 0.003381  min_lr: 0.003381  loss: 3.6630 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.8187  data: 0.0226  max mem: 41794
Epoch: [92]  [ 30/312]  eta: 0:05:41  lr: 0.003381  min_lr: 0.003381  loss: 4.0299 (3.7070)  weight_decay: 0.0500 (0.0500)  time: 0.6248  data: 0.0309  max mem: 41794
Epoch: [92]  [ 40/312]  eta: 0:04:45  lr: 0.003380  min_lr: 0.003380  loss: 3.7633 (3.6406)  weight_decay: 0.0500 (0.0500)  time: 0.5393  data: 0.0114  max mem: 41794
Epoch: [92]  [ 50/312]  eta: 0:04:20  lr: 0.003380  min_lr: 0.003380  loss: 3.7322 (3.6509)  weight_decay: 0.0500 (0.0500)  time: 0.6583  data: 0.0640  max mem: 41794
Epoch: [92]  [ 60/312]  eta: 0:03:49  lr: 0.003379  min_lr: 0.003379  loss: 3.6923 (3.6008)  weight_decay: 0.0500 (0.0500)  time: 0.6329  data: 0.0715  max mem: 41794
Epoch: [92]  [ 70/312]  eta: 0:03:40  lr: 0.003379  min_lr: 0.003379  loss: 3.4248 (3.5645)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.1254  max mem: 41794
Epoch: [92]  [ 80/312]  eta: 0:03:26  lr: 0.003378  min_lr: 0.003378  loss: 3.4248 (3.5580)  weight_decay: 0.0500 (0.0500)  time: 0.8302  data: 0.2040  max mem: 41794
Epoch: [92]  [ 90/312]  eta: 0:03:09  lr: 0.003378  min_lr: 0.003378  loss: 3.6797 (3.5667)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.1213  max mem: 41794
Epoch: [92]  [100/312]  eta: 0:03:00  lr: 0.003377  min_lr: 0.003377  loss: 3.7747 (3.5743)  weight_decay: 0.0500 (0.0500)  time: 0.6908  data: 0.1435  max mem: 41794
Epoch: [92]  [110/312]  eta: 0:02:45  lr: 0.003377  min_lr: 0.003377  loss: 3.7449 (3.5682)  weight_decay: 0.0500 (0.0500)  time: 0.6611  data: 0.1125  max mem: 41794
Epoch: [92]  [120/312]  eta: 0:02:39  lr: 0.003376  min_lr: 0.003376  loss: 3.6494 (3.5695)  weight_decay: 0.0500 (0.0500)  time: 0.7086  data: 0.1410  max mem: 41794
Epoch: [92]  [130/312]  eta: 0:02:30  lr: 0.003376  min_lr: 0.003376  loss: 3.8393 (3.5753)  weight_decay: 0.0500 (0.0500)  time: 0.8517  data: 0.2523  max mem: 41794
Epoch: [92]  [140/312]  eta: 0:02:19  lr: 0.003375  min_lr: 0.003375  loss: 3.6583 (3.5676)  weight_decay: 0.0500 (0.0500)  time: 0.6931  data: 0.1168  max mem: 41794
Epoch: [92]  [150/312]  eta: 0:02:10  lr: 0.003374  min_lr: 0.003374  loss: 3.6583 (3.5811)  weight_decay: 0.0500 (0.0500)  time: 0.6953  data: 0.0566  max mem: 41794
Epoch: [92]  [160/312]  eta: 0:02:02  lr: 0.003374  min_lr: 0.003374  loss: 3.9886 (3.5987)  weight_decay: 0.0500 (0.0500)  time: 0.7980  data: 0.1453  max mem: 41794
Epoch: [92]  [170/312]  eta: 0:01:52  lr: 0.003373  min_lr: 0.003373  loss: 3.9886 (3.6119)  weight_decay: 0.0500 (0.0500)  time: 0.6743  data: 0.1133  max mem: 41794
Epoch: [92]  [180/312]  eta: 0:01:45  lr: 0.003373  min_lr: 0.003373  loss: 3.7090 (3.6080)  weight_decay: 0.0500 (0.0500)  time: 0.7366  data: 0.1465  max mem: 41794
Epoch: [92]  [190/312]  eta: 0:01:35  lr: 0.003372  min_lr: 0.003372  loss: 3.5842 (3.5942)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1260  max mem: 41794
Epoch: [92]  [200/312]  eta: 0:01:28  lr: 0.003372  min_lr: 0.003372  loss: 3.5019 (3.5938)  weight_decay: 0.0500 (0.0500)  time: 0.7009  data: 0.1079  max mem: 41794
Epoch: [92]  [210/312]  eta: 0:01:20  lr: 0.003371  min_lr: 0.003371  loss: 3.5197 (3.5908)  weight_decay: 0.0500 (0.0500)  time: 0.8051  data: 0.1094  max mem: 41794
Epoch: [92]  [220/312]  eta: 0:01:11  lr: 0.003371  min_lr: 0.003371  loss: 3.5225 (3.5801)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.0685  max mem: 41794
Epoch: [92]  [230/312]  eta: 0:01:04  lr: 0.003370  min_lr: 0.003370  loss: 3.5225 (3.5834)  weight_decay: 0.0500 (0.0500)  time: 0.7642  data: 0.1579  max mem: 41794
Epoch: [92]  [240/312]  eta: 0:00:56  lr: 0.003370  min_lr: 0.003370  loss: 3.4405 (3.5752)  weight_decay: 0.0500 (0.0500)  time: 0.8267  data: 0.1083  max mem: 41794
Epoch: [92]  [250/312]  eta: 0:00:48  lr: 0.003369  min_lr: 0.003369  loss: 3.4115 (3.5686)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.0762  max mem: 41794
Epoch: [92]  [260/312]  eta: 0:00:40  lr: 0.003369  min_lr: 0.003369  loss: 3.6709 (3.5777)  weight_decay: 0.0500 (0.0500)  time: 0.7329  data: 0.1104  max mem: 41794
Epoch: [92]  [270/312]  eta: 0:00:32  lr: 0.003368  min_lr: 0.003368  loss: 3.6709 (3.5744)  weight_decay: 0.0500 (0.0500)  time: 0.6766  data: 0.0516  max mem: 41794
Epoch: [92]  [280/312]  eta: 0:00:24  lr: 0.003368  min_lr: 0.003368  loss: 3.5927 (3.5824)  weight_decay: 0.0500 (0.0500)  time: 0.6704  data: 0.0487  max mem: 41794
Epoch: [92]  [290/312]  eta: 0:00:17  lr: 0.003367  min_lr: 0.003367  loss: 3.8306 (3.5811)  weight_decay: 0.0500 (0.0500)  time: 0.8292  data: 0.0503  max mem: 41794
Epoch: [92]  [300/312]  eta: 0:00:09  lr: 0.003367  min_lr: 0.003367  loss: 3.5882 (3.5808)  weight_decay: 0.0500 (0.0500)  time: 0.6534  data: 0.0087  max mem: 41794
Epoch: [92]  [310/312]  eta: 0:00:01  lr: 0.003366  min_lr: 0.003366  loss: 3.5882 (3.5772)  weight_decay: 0.0500 (0.0500)  time: 0.4731  data: 0.0069  max mem: 41794
Epoch: [92]  [311/312]  eta: 0:00:00  lr: 0.003366  min_lr: 0.003366  loss: 3.5882 (3.5760)  weight_decay: 0.0500 (0.0500)  time: 0.4720  data: 0.0069  max mem: 41794
Epoch: [92] Total time: 0:03:57 (0.7602 s / it)
Averaged stats: lr: 0.003366  min_lr: 0.003366  loss: 3.5882 (3.5919)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.2793 (1.2793)  acc1: 75.7812 (75.7812)  acc5: 93.2292 (93.2292)  time: 8.2716  data: 8.0588  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6625 (1.5286)  acc1: 67.9688 (68.6880)  acc5: 89.1927 (89.5040)  time: 1.0672  data: 0.8955  max mem: 41794
Test: Total time: 0:00:09 (1.0804 s / it)
* Acc@1 69.290 Acc@5 89.632 loss 1.508
Accuracy of the model on the 50000 test images: 69.3%
Max accuracy: 70.28%
Epoch: [93]  [  0/312]  eta: 1:21:10  lr: 0.003366  min_lr: 0.003366  loss: 3.9672 (3.9672)  weight_decay: 0.0500 (0.0500)  time: 15.6093  data: 15.1366  max mem: 41794
Epoch: [93]  [ 10/312]  eta: 0:10:19  lr: 0.003365  min_lr: 0.003365  loss: 3.8723 (3.6821)  weight_decay: 0.0500 (0.0500)  time: 2.0511  data: 1.3767  max mem: 41794
Epoch: [93]  [ 20/312]  eta: 0:07:06  lr: 0.003365  min_lr: 0.003365  loss: 3.6330 (3.6496)  weight_decay: 0.0500 (0.0500)  time: 0.7517  data: 0.1423  max mem: 41794
Epoch: [93]  [ 30/312]  eta: 0:05:24  lr: 0.003364  min_lr: 0.003364  loss: 3.5849 (3.6185)  weight_decay: 0.0500 (0.0500)  time: 0.6541  data: 0.1473  max mem: 41794
Epoch: [93]  [ 40/312]  eta: 0:04:53  lr: 0.003364  min_lr: 0.003364  loss: 3.6597 (3.6216)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1665  max mem: 41794
Epoch: [93]  [ 50/312]  eta: 0:04:35  lr: 0.003363  min_lr: 0.003363  loss: 3.7455 (3.6435)  weight_decay: 0.0500 (0.0500)  time: 0.9013  data: 0.2977  max mem: 41794
Epoch: [93]  [ 60/312]  eta: 0:04:01  lr: 0.003363  min_lr: 0.003363  loss: 3.6249 (3.6055)  weight_decay: 0.0500 (0.0500)  time: 0.7158  data: 0.1368  max mem: 41794
Epoch: [93]  [ 70/312]  eta: 0:03:47  lr: 0.003362  min_lr: 0.003362  loss: 3.8125 (3.6424)  weight_decay: 0.0500 (0.0500)  time: 0.6486  data: 0.1605  max mem: 41794
Epoch: [93]  [ 80/312]  eta: 0:03:35  lr: 0.003362  min_lr: 0.003362  loss: 3.8425 (3.6309)  weight_decay: 0.0500 (0.0500)  time: 0.8414  data: 0.3251  max mem: 41794
Epoch: [93]  [ 90/312]  eta: 0:03:15  lr: 0.003361  min_lr: 0.003361  loss: 3.4710 (3.6105)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1652  max mem: 41794
Epoch: [93]  [100/312]  eta: 0:03:06  lr: 0.003361  min_lr: 0.003361  loss: 3.6958 (3.6141)  weight_decay: 0.0500 (0.0500)  time: 0.6728  data: 0.1528  max mem: 41794
Epoch: [93]  [110/312]  eta: 0:02:50  lr: 0.003360  min_lr: 0.003360  loss: 3.5703 (3.6001)  weight_decay: 0.0500 (0.0500)  time: 0.6813  data: 0.1528  max mem: 41794
Epoch: [93]  [120/312]  eta: 0:02:43  lr: 0.003360  min_lr: 0.003360  loss: 3.5703 (3.5920)  weight_decay: 0.0500 (0.0500)  time: 0.7030  data: 0.1384  max mem: 41794
Epoch: [93]  [130/312]  eta: 0:02:34  lr: 0.003359  min_lr: 0.003359  loss: 3.7813 (3.6067)  weight_decay: 0.0500 (0.0500)  time: 0.8715  data: 0.2720  max mem: 41794
Epoch: [93]  [140/312]  eta: 0:02:21  lr: 0.003359  min_lr: 0.003359  loss: 3.8370 (3.6097)  weight_decay: 0.0500 (0.0500)  time: 0.6657  data: 0.1342  max mem: 41794
Epoch: [93]  [150/312]  eta: 0:02:13  lr: 0.003358  min_lr: 0.003358  loss: 3.7410 (3.6057)  weight_decay: 0.0500 (0.0500)  time: 0.6587  data: 0.1198  max mem: 41794
Epoch: [93]  [160/312]  eta: 0:02:05  lr: 0.003358  min_lr: 0.003358  loss: 3.5604 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.8236  data: 0.2814  max mem: 41794
Epoch: [93]  [170/312]  eta: 0:01:54  lr: 0.003357  min_lr: 0.003357  loss: 3.4724 (3.5966)  weight_decay: 0.0500 (0.0500)  time: 0.6609  data: 0.1625  max mem: 41794
Epoch: [93]  [180/312]  eta: 0:01:46  lr: 0.003356  min_lr: 0.003356  loss: 3.3905 (3.5845)  weight_decay: 0.0500 (0.0500)  time: 0.6919  data: 0.1945  max mem: 41794
Epoch: [93]  [190/312]  eta: 0:01:36  lr: 0.003356  min_lr: 0.003356  loss: 3.8362 (3.5930)  weight_decay: 0.0500 (0.0500)  time: 0.6886  data: 0.1941  max mem: 41794
Epoch: [93]  [200/312]  eta: 0:01:29  lr: 0.003355  min_lr: 0.003355  loss: 3.7600 (3.5855)  weight_decay: 0.0500 (0.0500)  time: 0.6720  data: 0.1830  max mem: 41794
Epoch: [93]  [210/312]  eta: 0:01:21  lr: 0.003355  min_lr: 0.003355  loss: 3.6478 (3.5825)  weight_decay: 0.0500 (0.0500)  time: 0.8598  data: 0.3683  max mem: 41794
Epoch: [93]  [220/312]  eta: 0:01:12  lr: 0.003354  min_lr: 0.003354  loss: 3.6478 (3.5878)  weight_decay: 0.0500 (0.0500)  time: 0.6794  data: 0.1859  max mem: 41794
Epoch: [93]  [230/312]  eta: 0:01:04  lr: 0.003354  min_lr: 0.003354  loss: 3.7400 (3.5789)  weight_decay: 0.0500 (0.0500)  time: 0.6794  data: 0.1904  max mem: 41794
Epoch: [93]  [240/312]  eta: 0:00:56  lr: 0.003353  min_lr: 0.003353  loss: 3.7431 (3.5893)  weight_decay: 0.0500 (0.0500)  time: 0.8538  data: 0.3588  max mem: 41794
Epoch: [93]  [250/312]  eta: 0:00:48  lr: 0.003353  min_lr: 0.003353  loss: 3.6075 (3.5788)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.1707  max mem: 41794
Epoch: [93]  [260/312]  eta: 0:00:40  lr: 0.003352  min_lr: 0.003352  loss: 3.7324 (3.5954)  weight_decay: 0.0500 (0.0500)  time: 0.6786  data: 0.1833  max mem: 41794
Epoch: [93]  [270/312]  eta: 0:00:32  lr: 0.003352  min_lr: 0.003352  loss: 3.9294 (3.6078)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1817  max mem: 41794
Epoch: [93]  [280/312]  eta: 0:00:24  lr: 0.003351  min_lr: 0.003351  loss: 3.8506 (3.6144)  weight_decay: 0.0500 (0.0500)  time: 0.6913  data: 0.1822  max mem: 41794
Epoch: [93]  [290/312]  eta: 0:00:17  lr: 0.003351  min_lr: 0.003351  loss: 3.7997 (3.6120)  weight_decay: 0.0500 (0.0500)  time: 0.8458  data: 0.3308  max mem: 41794
Epoch: [93]  [300/312]  eta: 0:00:09  lr: 0.003350  min_lr: 0.003350  loss: 3.5695 (3.6142)  weight_decay: 0.0500 (0.0500)  time: 0.6404  data: 0.1490  max mem: 41794
Epoch: [93]  [310/312]  eta: 0:00:01  lr: 0.003350  min_lr: 0.003350  loss: 3.5579 (3.6132)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0001  max mem: 41794
Epoch: [93]  [311/312]  eta: 0:00:00  lr: 0.003350  min_lr: 0.003350  loss: 3.5577 (3.6128)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [93] Total time: 0:03:58 (0.7632 s / it)
Averaged stats: lr: 0.003350  min_lr: 0.003350  loss: 3.5577 (3.6032)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.3029 (1.3029)  acc1: 78.6458 (78.6458)  acc5: 93.3594 (93.3594)  time: 8.3409  data: 8.1287  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7723 (1.6861)  acc1: 66.9271 (69.4240)  acc5: 88.6719 (89.3760)  time: 1.0746  data: 0.9033  max mem: 41794
Test: Total time: 0:00:09 (1.1044 s / it)
* Acc@1 69.428 Acc@5 89.598 loss 1.684
Accuracy of the model on the 50000 test images: 69.4%
Max accuracy: 70.28%
Epoch: [94]  [  0/312]  eta: 1:24:25  lr: 0.003350  min_lr: 0.003350  loss: 3.9744 (3.9744)  weight_decay: 0.0500 (0.0500)  time: 16.2367  data: 12.3180  max mem: 41794
Epoch: [94]  [ 10/312]  eta: 0:10:54  lr: 0.003349  min_lr: 0.003349  loss: 3.8576 (3.8221)  weight_decay: 0.0500 (0.0500)  time: 2.1683  data: 1.3309  max mem: 41794
Epoch: [94]  [ 20/312]  eta: 0:07:57  lr: 0.003348  min_lr: 0.003348  loss: 3.7996 (3.7890)  weight_decay: 0.0500 (0.0500)  time: 0.9039  data: 0.1916  max mem: 41794
Epoch: [94]  [ 30/312]  eta: 0:06:04  lr: 0.003348  min_lr: 0.003348  loss: 3.7841 (3.6765)  weight_decay: 0.0500 (0.0500)  time: 0.8123  data: 0.0886  max mem: 41794
Epoch: [94]  [ 40/312]  eta: 0:05:00  lr: 0.003347  min_lr: 0.003347  loss: 3.4828 (3.6439)  weight_decay: 0.0500 (0.0500)  time: 0.5508  data: 0.0158  max mem: 41794
Epoch: [94]  [ 50/312]  eta: 0:04:22  lr: 0.003347  min_lr: 0.003347  loss: 3.6704 (3.6446)  weight_decay: 0.0500 (0.0500)  time: 0.5520  data: 0.0078  max mem: 41794
Epoch: [94]  [ 60/312]  eta: 0:03:51  lr: 0.003346  min_lr: 0.003346  loss: 3.6704 (3.6019)  weight_decay: 0.0500 (0.0500)  time: 0.5327  data: 0.0055  max mem: 41794
Epoch: [94]  [ 70/312]  eta: 0:03:38  lr: 0.003346  min_lr: 0.003346  loss: 3.7534 (3.6311)  weight_decay: 0.0500 (0.0500)  time: 0.6522  data: 0.0336  max mem: 41794
Epoch: [94]  [ 80/312]  eta: 0:03:27  lr: 0.003345  min_lr: 0.003345  loss: 3.8423 (3.6099)  weight_decay: 0.0500 (0.0500)  time: 0.8191  data: 0.0845  max mem: 41794
Epoch: [94]  [ 90/312]  eta: 0:03:09  lr: 0.003345  min_lr: 0.003345  loss: 3.5551 (3.6080)  weight_decay: 0.0500 (0.0500)  time: 0.6661  data: 0.0542  max mem: 41794
Epoch: [94]  [100/312]  eta: 0:03:01  lr: 0.003344  min_lr: 0.003344  loss: 3.8143 (3.6062)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.0878  max mem: 41794
Epoch: [94]  [110/312]  eta: 0:02:45  lr: 0.003344  min_lr: 0.003344  loss: 3.8248 (3.6038)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.0851  max mem: 41794
Epoch: [94]  [120/312]  eta: 0:02:38  lr: 0.003343  min_lr: 0.003343  loss: 3.4414 (3.5744)  weight_decay: 0.0500 (0.0500)  time: 0.6761  data: 0.0818  max mem: 41794
Epoch: [94]  [130/312]  eta: 0:02:29  lr: 0.003343  min_lr: 0.003343  loss: 3.4414 (3.5757)  weight_decay: 0.0500 (0.0500)  time: 0.8356  data: 0.1419  max mem: 41794
Epoch: [94]  [140/312]  eta: 0:02:18  lr: 0.003342  min_lr: 0.003342  loss: 3.7451 (3.5793)  weight_decay: 0.0500 (0.0500)  time: 0.6685  data: 0.0726  max mem: 41794
Epoch: [94]  [150/312]  eta: 0:02:11  lr: 0.003342  min_lr: 0.003342  loss: 3.7740 (3.5910)  weight_decay: 0.0500 (0.0500)  time: 0.7116  data: 0.1163  max mem: 41794
Epoch: [94]  [160/312]  eta: 0:02:03  lr: 0.003341  min_lr: 0.003341  loss: 3.7449 (3.5843)  weight_decay: 0.0500 (0.0500)  time: 0.8621  data: 0.1956  max mem: 41794
Epoch: [94]  [170/312]  eta: 0:01:52  lr: 0.003340  min_lr: 0.003340  loss: 3.8066 (3.6024)  weight_decay: 0.0500 (0.0500)  time: 0.6726  data: 0.1004  max mem: 41794
Epoch: [94]  [180/312]  eta: 0:01:45  lr: 0.003340  min_lr: 0.003340  loss: 3.8440 (3.6120)  weight_decay: 0.0500 (0.0500)  time: 0.7033  data: 0.1257  max mem: 41794
Epoch: [94]  [190/312]  eta: 0:01:35  lr: 0.003339  min_lr: 0.003339  loss: 3.8008 (3.6063)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.1172  max mem: 41794
Epoch: [94]  [200/312]  eta: 0:01:28  lr: 0.003339  min_lr: 0.003339  loss: 3.4321 (3.5993)  weight_decay: 0.0500 (0.0500)  time: 0.7003  data: 0.1089  max mem: 41794
Epoch: [94]  [210/312]  eta: 0:01:20  lr: 0.003338  min_lr: 0.003338  loss: 3.5319 (3.5972)  weight_decay: 0.0500 (0.0500)  time: 0.8937  data: 0.1980  max mem: 41794
Epoch: [94]  [220/312]  eta: 0:01:11  lr: 0.003338  min_lr: 0.003338  loss: 3.6474 (3.5906)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.0916  max mem: 41794
Epoch: [94]  [230/312]  eta: 0:01:04  lr: 0.003337  min_lr: 0.003337  loss: 3.6743 (3.5957)  weight_decay: 0.0500 (0.0500)  time: 0.7183  data: 0.1004  max mem: 41794
Epoch: [94]  [240/312]  eta: 0:00:56  lr: 0.003337  min_lr: 0.003337  loss: 3.6743 (3.5960)  weight_decay: 0.0500 (0.0500)  time: 0.8874  data: 0.1681  max mem: 41794
Epoch: [94]  [250/312]  eta: 0:00:48  lr: 0.003336  min_lr: 0.003336  loss: 3.7209 (3.6057)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.0800  max mem: 41794
Epoch: [94]  [260/312]  eta: 0:00:40  lr: 0.003336  min_lr: 0.003336  loss: 3.7739 (3.6030)  weight_decay: 0.0500 (0.0500)  time: 0.6735  data: 0.0837  max mem: 41794
Epoch: [94]  [270/312]  eta: 0:00:32  lr: 0.003335  min_lr: 0.003335  loss: 3.6809 (3.5978)  weight_decay: 0.0500 (0.0500)  time: 0.6604  data: 0.0760  max mem: 41794
Epoch: [94]  [280/312]  eta: 0:00:24  lr: 0.003335  min_lr: 0.003335  loss: 3.4001 (3.5918)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.1059  max mem: 41794
Epoch: [94]  [290/312]  eta: 0:00:17  lr: 0.003334  min_lr: 0.003334  loss: 3.4001 (3.5845)  weight_decay: 0.0500 (0.0500)  time: 0.8873  data: 0.2116  max mem: 41794
Epoch: [94]  [300/312]  eta: 0:00:09  lr: 0.003334  min_lr: 0.003334  loss: 3.2707 (3.5754)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1082  max mem: 41794
Epoch: [94]  [310/312]  eta: 0:00:01  lr: 0.003333  min_lr: 0.003333  loss: 3.6260 (3.5882)  weight_decay: 0.0500 (0.0500)  time: 0.4721  data: 0.0001  max mem: 41794
Epoch: [94]  [311/312]  eta: 0:00:00  lr: 0.003333  min_lr: 0.003333  loss: 3.9206 (3.5893)  weight_decay: 0.0500 (0.0500)  time: 0.4693  data: 0.0001  max mem: 41794
Epoch: [94] Total time: 0:03:58 (0.7634 s / it)
Averaged stats: lr: 0.003333  min_lr: 0.003333  loss: 3.9206 (3.5939)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.3121 (1.3121)  acc1: 76.6927 (76.6927)  acc5: 92.0573 (92.0573)  time: 8.5620  data: 8.3567  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5074 (1.4629)  acc1: 68.3594 (69.1520)  acc5: 90.3646 (89.6640)  time: 1.1002  data: 0.9286  max mem: 41794
Test: Total time: 0:00:10 (1.1152 s / it)
* Acc@1 69.866 Acc@5 90.124 loss 1.456
Accuracy of the model on the 50000 test images: 69.9%
Max accuracy: 70.28%
Epoch: [95]  [  0/312]  eta: 1:23:04  lr: 0.003333  min_lr: 0.003333  loss: 3.5719 (3.5719)  weight_decay: 0.0500 (0.0500)  time: 15.9768  data: 14.7453  max mem: 41794
Epoch: [95]  [ 10/312]  eta: 0:10:52  lr: 0.003332  min_lr: 0.003332  loss: 3.6149 (3.6400)  weight_decay: 0.0500 (0.0500)  time: 2.1605  data: 1.3535  max mem: 41794
Epoch: [95]  [ 20/312]  eta: 0:07:17  lr: 0.003332  min_lr: 0.003332  loss: 3.5065 (3.5331)  weight_decay: 0.0500 (0.0500)  time: 0.7732  data: 0.0722  max mem: 41794
Epoch: [95]  [ 30/312]  eta: 0:05:32  lr: 0.003331  min_lr: 0.003331  loss: 3.5886 (3.5955)  weight_decay: 0.0500 (0.0500)  time: 0.6395  data: 0.0681  max mem: 41794
Epoch: [95]  [ 40/312]  eta: 0:04:48  lr: 0.003331  min_lr: 0.003331  loss: 3.6882 (3.6200)  weight_decay: 0.0500 (0.0500)  time: 0.6029  data: 0.0246  max mem: 41794
Epoch: [95]  [ 50/312]  eta: 0:04:25  lr: 0.003330  min_lr: 0.003330  loss: 3.7383 (3.6371)  weight_decay: 0.0500 (0.0500)  time: 0.7587  data: 0.1169  max mem: 41794
Epoch: [95]  [ 60/312]  eta: 0:03:55  lr: 0.003330  min_lr: 0.003330  loss: 3.8828 (3.6525)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1128  max mem: 41794
Epoch: [95]  [ 70/312]  eta: 0:03:43  lr: 0.003329  min_lr: 0.003329  loss: 3.7546 (3.6352)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.0922  max mem: 41794
Epoch: [95]  [ 80/312]  eta: 0:03:31  lr: 0.003329  min_lr: 0.003329  loss: 3.7636 (3.6433)  weight_decay: 0.0500 (0.0500)  time: 0.8507  data: 0.1229  max mem: 41794
Epoch: [95]  [ 90/312]  eta: 0:03:13  lr: 0.003328  min_lr: 0.003328  loss: 3.5978 (3.6266)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.0780  max mem: 41794
Epoch: [95]  [100/312]  eta: 0:03:03  lr: 0.003327  min_lr: 0.003327  loss: 3.5404 (3.6096)  weight_decay: 0.0500 (0.0500)  time: 0.6802  data: 0.0914  max mem: 41794
Epoch: [95]  [110/312]  eta: 0:02:48  lr: 0.003327  min_lr: 0.003327  loss: 3.5920 (3.6099)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.0636  max mem: 41794
Epoch: [95]  [120/312]  eta: 0:02:40  lr: 0.003326  min_lr: 0.003326  loss: 3.5920 (3.6013)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.0832  max mem: 41794
Epoch: [95]  [130/312]  eta: 0:02:32  lr: 0.003326  min_lr: 0.003326  loss: 3.6862 (3.6044)  weight_decay: 0.0500 (0.0500)  time: 0.8647  data: 0.1934  max mem: 41794
Epoch: [95]  [140/312]  eta: 0:02:20  lr: 0.003325  min_lr: 0.003325  loss: 3.5511 (3.5752)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1144  max mem: 41794
Epoch: [95]  [150/312]  eta: 0:02:12  lr: 0.003325  min_lr: 0.003325  loss: 3.5214 (3.5855)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1104  max mem: 41794
Epoch: [95]  [160/312]  eta: 0:02:04  lr: 0.003324  min_lr: 0.003324  loss: 3.7352 (3.5920)  weight_decay: 0.0500 (0.0500)  time: 0.8587  data: 0.1918  max mem: 41794
Epoch: [95]  [170/312]  eta: 0:01:54  lr: 0.003324  min_lr: 0.003324  loss: 3.7360 (3.6045)  weight_decay: 0.0500 (0.0500)  time: 0.7090  data: 0.1235  max mem: 41794
Epoch: [95]  [180/312]  eta: 0:01:46  lr: 0.003323  min_lr: 0.003323  loss: 3.7769 (3.6007)  weight_decay: 0.0500 (0.0500)  time: 0.7230  data: 0.1378  max mem: 41794
Epoch: [95]  [190/312]  eta: 0:01:36  lr: 0.003323  min_lr: 0.003323  loss: 3.7967 (3.6042)  weight_decay: 0.0500 (0.0500)  time: 0.6917  data: 0.0999  max mem: 41794
Epoch: [95]  [200/312]  eta: 0:01:29  lr: 0.003322  min_lr: 0.003322  loss: 3.8881 (3.6084)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.0755  max mem: 41794
Epoch: [95]  [210/312]  eta: 0:01:21  lr: 0.003322  min_lr: 0.003322  loss: 3.7736 (3.6100)  weight_decay: 0.0500 (0.0500)  time: 0.8748  data: 0.1482  max mem: 41794
Epoch: [95]  [220/312]  eta: 0:01:12  lr: 0.003321  min_lr: 0.003321  loss: 3.6319 (3.6080)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.0751  max mem: 41794
Epoch: [95]  [230/312]  eta: 0:01:04  lr: 0.003320  min_lr: 0.003320  loss: 3.6319 (3.6039)  weight_decay: 0.0500 (0.0500)  time: 0.7001  data: 0.0802  max mem: 41794
Epoch: [95]  [240/312]  eta: 0:00:57  lr: 0.003320  min_lr: 0.003320  loss: 3.6667 (3.6063)  weight_decay: 0.0500 (0.0500)  time: 0.8451  data: 0.1524  max mem: 41794
Epoch: [95]  [250/312]  eta: 0:00:48  lr: 0.003319  min_lr: 0.003319  loss: 3.5346 (3.5971)  weight_decay: 0.0500 (0.0500)  time: 0.6447  data: 0.0753  max mem: 41794
Epoch: [95]  [260/312]  eta: 0:00:40  lr: 0.003319  min_lr: 0.003319  loss: 3.3224 (3.5933)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.0946  max mem: 41794
Epoch: [95]  [270/312]  eta: 0:00:32  lr: 0.003318  min_lr: 0.003318  loss: 3.6923 (3.5955)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.0922  max mem: 41794
Epoch: [95]  [280/312]  eta: 0:00:24  lr: 0.003318  min_lr: 0.003318  loss: 3.7439 (3.5881)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.0949  max mem: 41794
Epoch: [95]  [290/312]  eta: 0:00:17  lr: 0.003317  min_lr: 0.003317  loss: 3.4982 (3.5856)  weight_decay: 0.0500 (0.0500)  time: 0.8266  data: 0.1988  max mem: 41794
Epoch: [95]  [300/312]  eta: 0:00:09  lr: 0.003317  min_lr: 0.003317  loss: 3.5998 (3.5888)  weight_decay: 0.0500 (0.0500)  time: 0.6271  data: 0.1043  max mem: 41794
Epoch: [95]  [310/312]  eta: 0:00:01  lr: 0.003316  min_lr: 0.003316  loss: 3.8451 (3.5952)  weight_decay: 0.0500 (0.0500)  time: 0.4638  data: 0.0001  max mem: 41794
Epoch: [95]  [311/312]  eta: 0:00:00  lr: 0.003316  min_lr: 0.003316  loss: 3.8451 (3.5959)  weight_decay: 0.0500 (0.0500)  time: 0.4638  data: 0.0001  max mem: 41794
Epoch: [95] Total time: 0:03:58 (0.7628 s / it)
Averaged stats: lr: 0.003316  min_lr: 0.003316  loss: 3.8451 (3.5977)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.4612 (1.4612)  acc1: 77.8646 (77.8646)  acc5: 93.4896 (93.4896)  time: 8.1666  data: 7.9633  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7929 (1.6419)  acc1: 66.4062 (69.4240)  acc5: 88.8021 (89.9200)  time: 1.0653  data: 0.8849  max mem: 41794
Test: Total time: 0:00:09 (1.0841 s / it)
* Acc@1 69.724 Acc@5 89.954 loss 1.641
Accuracy of the model on the 50000 test images: 69.7%
Max accuracy: 70.28%
Epoch: [96]  [  0/312]  eta: 1:22:33  lr: 0.003316  min_lr: 0.003316  loss: 4.2877 (4.2877)  weight_decay: 0.0500 (0.0500)  time: 15.8768  data: 12.5865  max mem: 41794
Epoch: [96]  [ 10/312]  eta: 0:10:36  lr: 0.003316  min_lr: 0.003316  loss: 3.9351 (3.6523)  weight_decay: 0.0500 (0.0500)  time: 2.1073  data: 1.3329  max mem: 41794
Epoch: [96]  [ 20/312]  eta: 0:07:35  lr: 0.003315  min_lr: 0.003315  loss: 3.9457 (3.7496)  weight_decay: 0.0500 (0.0500)  time: 0.8426  data: 0.1668  max mem: 41794
Epoch: [96]  [ 30/312]  eta: 0:05:48  lr: 0.003314  min_lr: 0.003314  loss: 3.8373 (3.6368)  weight_decay: 0.0500 (0.0500)  time: 0.7549  data: 0.0634  max mem: 41794
Epoch: [96]  [ 40/312]  eta: 0:04:56  lr: 0.003314  min_lr: 0.003314  loss: 3.7468 (3.6448)  weight_decay: 0.0500 (0.0500)  time: 0.6008  data: 0.0124  max mem: 41794
Epoch: [96]  [ 50/312]  eta: 0:04:27  lr: 0.003313  min_lr: 0.003313  loss: 3.7633 (3.6611)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.0192  max mem: 41794
Epoch: [96]  [ 60/312]  eta: 0:03:55  lr: 0.003313  min_lr: 0.003313  loss: 3.6379 (3.6520)  weight_decay: 0.0500 (0.0500)  time: 0.6097  data: 0.0075  max mem: 41794
Epoch: [96]  [ 70/312]  eta: 0:03:43  lr: 0.003312  min_lr: 0.003312  loss: 3.6635 (3.6442)  weight_decay: 0.0500 (0.0500)  time: 0.6724  data: 0.0342  max mem: 41794
Epoch: [96]  [ 80/312]  eta: 0:03:31  lr: 0.003312  min_lr: 0.003312  loss: 3.7311 (3.6257)  weight_decay: 0.0500 (0.0500)  time: 0.8403  data: 0.0438  max mem: 41794
Epoch: [96]  [ 90/312]  eta: 0:03:16  lr: 0.003311  min_lr: 0.003311  loss: 3.3012 (3.6024)  weight_decay: 0.0500 (0.0500)  time: 0.7390  data: 0.0508  max mem: 41794
Epoch: [96]  [100/312]  eta: 0:03:05  lr: 0.003311  min_lr: 0.003311  loss: 3.7530 (3.6096)  weight_decay: 0.0500 (0.0500)  time: 0.7222  data: 0.0816  max mem: 41794
Epoch: [96]  [110/312]  eta: 0:02:49  lr: 0.003310  min_lr: 0.003310  loss: 3.8074 (3.6266)  weight_decay: 0.0500 (0.0500)  time: 0.6389  data: 0.0410  max mem: 41794
Epoch: [96]  [120/312]  eta: 0:02:40  lr: 0.003310  min_lr: 0.003310  loss: 3.7139 (3.6165)  weight_decay: 0.0500 (0.0500)  time: 0.6540  data: 0.0444  max mem: 41794
Epoch: [96]  [130/312]  eta: 0:02:31  lr: 0.003309  min_lr: 0.003309  loss: 3.4915 (3.6017)  weight_decay: 0.0500 (0.0500)  time: 0.8015  data: 0.0795  max mem: 41794
Epoch: [96]  [140/312]  eta: 0:02:19  lr: 0.003308  min_lr: 0.003308  loss: 3.6163 (3.5985)  weight_decay: 0.0500 (0.0500)  time: 0.6545  data: 0.0480  max mem: 41794
Epoch: [96]  [150/312]  eta: 0:02:12  lr: 0.003308  min_lr: 0.003308  loss: 3.8085 (3.6218)  weight_decay: 0.0500 (0.0500)  time: 0.7316  data: 0.0781  max mem: 41794
Epoch: [96]  [160/312]  eta: 0:02:04  lr: 0.003307  min_lr: 0.003307  loss: 3.9228 (3.6316)  weight_decay: 0.0500 (0.0500)  time: 0.8497  data: 0.0690  max mem: 41794
Epoch: [96]  [170/312]  eta: 0:01:54  lr: 0.003307  min_lr: 0.003307  loss: 3.8913 (3.6337)  weight_decay: 0.0500 (0.0500)  time: 0.6968  data: 0.0699  max mem: 41794
Epoch: [96]  [180/312]  eta: 0:01:46  lr: 0.003306  min_lr: 0.003306  loss: 3.8913 (3.6431)  weight_decay: 0.0500 (0.0500)  time: 0.7499  data: 0.0885  max mem: 41794
Epoch: [96]  [190/312]  eta: 0:01:36  lr: 0.003306  min_lr: 0.003306  loss: 3.7964 (3.6415)  weight_decay: 0.0500 (0.0500)  time: 0.7000  data: 0.0443  max mem: 41794
Epoch: [96]  [200/312]  eta: 0:01:29  lr: 0.003305  min_lr: 0.003305  loss: 3.7284 (3.6423)  weight_decay: 0.0500 (0.0500)  time: 0.6880  data: 0.0655  max mem: 41794
Epoch: [96]  [210/312]  eta: 0:01:21  lr: 0.003305  min_lr: 0.003305  loss: 3.7284 (3.6404)  weight_decay: 0.0500 (0.0500)  time: 0.7975  data: 0.0627  max mem: 41794
Epoch: [96]  [220/312]  eta: 0:01:12  lr: 0.003304  min_lr: 0.003304  loss: 3.7654 (3.6395)  weight_decay: 0.0500 (0.0500)  time: 0.6503  data: 0.0447  max mem: 41794
Epoch: [96]  [230/312]  eta: 0:01:04  lr: 0.003304  min_lr: 0.003304  loss: 3.7104 (3.6342)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.0572  max mem: 41794
Epoch: [96]  [240/312]  eta: 0:00:56  lr: 0.003303  min_lr: 0.003303  loss: 3.6506 (3.6409)  weight_decay: 0.0500 (0.0500)  time: 0.7935  data: 0.0365  max mem: 41794
Epoch: [96]  [250/312]  eta: 0:00:48  lr: 0.003302  min_lr: 0.003302  loss: 3.5603 (3.6387)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.0364  max mem: 41794
Epoch: [96]  [260/312]  eta: 0:00:40  lr: 0.003302  min_lr: 0.003302  loss: 3.5532 (3.6447)  weight_decay: 0.0500 (0.0500)  time: 0.7728  data: 0.0678  max mem: 41794
Epoch: [96]  [270/312]  eta: 0:00:32  lr: 0.003301  min_lr: 0.003301  loss: 3.4367 (3.6319)  weight_decay: 0.0500 (0.0500)  time: 0.6946  data: 0.0378  max mem: 41794
Epoch: [96]  [280/312]  eta: 0:00:24  lr: 0.003301  min_lr: 0.003301  loss: 3.2993 (3.6233)  weight_decay: 0.0500 (0.0500)  time: 0.6322  data: 0.0608  max mem: 41794
Epoch: [96]  [290/312]  eta: 0:00:17  lr: 0.003300  min_lr: 0.003300  loss: 3.5112 (3.6205)  weight_decay: 0.0500 (0.0500)  time: 0.8042  data: 0.1307  max mem: 41794
Epoch: [96]  [300/312]  eta: 0:00:09  lr: 0.003300  min_lr: 0.003300  loss: 3.6938 (3.6288)  weight_decay: 0.0500 (0.0500)  time: 0.6543  data: 0.0718  max mem: 41794
Epoch: [96]  [310/312]  eta: 0:00:01  lr: 0.003299  min_lr: 0.003299  loss: 3.8883 (3.6332)  weight_decay: 0.0500 (0.0500)  time: 0.4624  data: 0.0001  max mem: 41794
Epoch: [96]  [311/312]  eta: 0:00:00  lr: 0.003299  min_lr: 0.003299  loss: 3.8883 (3.6333)  weight_decay: 0.0500 (0.0500)  time: 0.4621  data: 0.0001  max mem: 41794
Epoch: [96] Total time: 0:03:57 (0.7606 s / it)
Averaged stats: lr: 0.003299  min_lr: 0.003299  loss: 3.8883 (3.6192)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.2135 (1.2135)  acc1: 78.1250 (78.1250)  acc5: 93.0990 (93.0990)  time: 8.4145  data: 8.1902  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6378 (1.5518)  acc1: 69.2708 (69.0720)  acc5: 89.4531 (89.2960)  time: 1.0966  data: 0.9101  max mem: 41794
Test: Total time: 0:00:10 (1.1350 s / it)
* Acc@1 69.014 Acc@5 89.404 loss 1.558
Accuracy of the model on the 50000 test images: 69.0%
Max accuracy: 70.28%
Epoch: [97]  [  0/312]  eta: 1:20:21  lr: 0.003299  min_lr: 0.003299  loss: 3.5619 (3.5619)  weight_decay: 0.0500 (0.0500)  time: 15.4535  data: 14.9542  max mem: 41794
Epoch: [97]  [ 10/312]  eta: 0:11:05  lr: 0.003299  min_lr: 0.003299  loss: 3.6199 (3.4914)  weight_decay: 0.0500 (0.0500)  time: 2.2021  data: 1.3696  max mem: 41794
Epoch: [97]  [ 20/312]  eta: 0:07:10  lr: 0.003298  min_lr: 0.003298  loss: 3.6199 (3.4767)  weight_decay: 0.0500 (0.0500)  time: 0.7753  data: 0.0803  max mem: 41794
Epoch: [97]  [ 30/312]  eta: 0:05:28  lr: 0.003297  min_lr: 0.003297  loss: 3.4759 (3.4710)  weight_decay: 0.0500 (0.0500)  time: 0.5927  data: 0.0751  max mem: 41794
Epoch: [97]  [ 40/312]  eta: 0:04:45  lr: 0.003297  min_lr: 0.003297  loss: 3.6133 (3.5268)  weight_decay: 0.0500 (0.0500)  time: 0.6049  data: 0.0628  max mem: 41794
Epoch: [97]  [ 50/312]  eta: 0:04:27  lr: 0.003296  min_lr: 0.003296  loss: 3.8342 (3.5356)  weight_decay: 0.0500 (0.0500)  time: 0.7951  data: 0.1763  max mem: 41794
Epoch: [97]  [ 60/312]  eta: 0:03:56  lr: 0.003296  min_lr: 0.003296  loss: 3.7195 (3.5631)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1258  max mem: 41794
Epoch: [97]  [ 70/312]  eta: 0:03:48  lr: 0.003295  min_lr: 0.003295  loss: 3.7195 (3.5683)  weight_decay: 0.0500 (0.0500)  time: 0.7594  data: 0.1296  max mem: 41794
Epoch: [97]  [ 80/312]  eta: 0:03:34  lr: 0.003295  min_lr: 0.003295  loss: 3.6933 (3.5813)  weight_decay: 0.0500 (0.0500)  time: 0.8865  data: 0.1411  max mem: 41794
Epoch: [97]  [ 90/312]  eta: 0:03:17  lr: 0.003294  min_lr: 0.003294  loss: 3.5648 (3.5649)  weight_decay: 0.0500 (0.0500)  time: 0.6922  data: 0.0891  max mem: 41794
Epoch: [97]  [100/312]  eta: 0:03:08  lr: 0.003294  min_lr: 0.003294  loss: 3.5648 (3.5916)  weight_decay: 0.0500 (0.0500)  time: 0.7513  data: 0.1648  max mem: 41794
Epoch: [97]  [110/312]  eta: 0:02:52  lr: 0.003293  min_lr: 0.003293  loss: 3.6821 (3.5839)  weight_decay: 0.0500 (0.0500)  time: 0.6962  data: 0.0994  max mem: 41794
Epoch: [97]  [120/312]  eta: 0:02:43  lr: 0.003292  min_lr: 0.003292  loss: 3.6119 (3.5801)  weight_decay: 0.0500 (0.0500)  time: 0.6624  data: 0.0861  max mem: 41794
Epoch: [97]  [130/312]  eta: 0:02:34  lr: 0.003292  min_lr: 0.003292  loss: 3.6119 (3.5835)  weight_decay: 0.0500 (0.0500)  time: 0.8141  data: 0.1600  max mem: 41794
Epoch: [97]  [140/312]  eta: 0:02:21  lr: 0.003291  min_lr: 0.003291  loss: 3.5867 (3.5707)  weight_decay: 0.0500 (0.0500)  time: 0.6504  data: 0.0780  max mem: 41794
Epoch: [97]  [150/312]  eta: 0:02:13  lr: 0.003291  min_lr: 0.003291  loss: 3.2371 (3.5566)  weight_decay: 0.0500 (0.0500)  time: 0.6723  data: 0.1163  max mem: 41794
Epoch: [97]  [160/312]  eta: 0:02:04  lr: 0.003290  min_lr: 0.003290  loss: 3.5977 (3.5710)  weight_decay: 0.0500 (0.0500)  time: 0.7921  data: 0.1487  max mem: 41794
Epoch: [97]  [170/312]  eta: 0:01:54  lr: 0.003290  min_lr: 0.003290  loss: 3.7780 (3.5778)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.1042  max mem: 41794
Epoch: [97]  [180/312]  eta: 0:01:47  lr: 0.003289  min_lr: 0.003289  loss: 3.7508 (3.5853)  weight_decay: 0.0500 (0.0500)  time: 0.7491  data: 0.1778  max mem: 41794
Epoch: [97]  [190/312]  eta: 0:01:37  lr: 0.003289  min_lr: 0.003289  loss: 3.6905 (3.5721)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.1100  max mem: 41794
Epoch: [97]  [200/312]  eta: 0:01:29  lr: 0.003288  min_lr: 0.003288  loss: 3.6768 (3.5742)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1084  max mem: 41794
Epoch: [97]  [210/312]  eta: 0:01:21  lr: 0.003288  min_lr: 0.003288  loss: 3.8180 (3.5762)  weight_decay: 0.0500 (0.0500)  time: 0.8388  data: 0.1709  max mem: 41794
Epoch: [97]  [220/312]  eta: 0:01:12  lr: 0.003287  min_lr: 0.003287  loss: 3.7876 (3.5855)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.1077  max mem: 41794
Epoch: [97]  [230/312]  eta: 0:01:05  lr: 0.003286  min_lr: 0.003286  loss: 3.7876 (3.5866)  weight_decay: 0.0500 (0.0500)  time: 0.7394  data: 0.1393  max mem: 41794
Epoch: [97]  [240/312]  eta: 0:00:56  lr: 0.003286  min_lr: 0.003286  loss: 3.5861 (3.5858)  weight_decay: 0.0500 (0.0500)  time: 0.7848  data: 0.0994  max mem: 41794
Epoch: [97]  [250/312]  eta: 0:00:48  lr: 0.003285  min_lr: 0.003285  loss: 3.3135 (3.5751)  weight_decay: 0.0500 (0.0500)  time: 0.6583  data: 0.0732  max mem: 41794
Epoch: [97]  [260/312]  eta: 0:00:40  lr: 0.003285  min_lr: 0.003285  loss: 3.1333 (3.5667)  weight_decay: 0.0500 (0.0500)  time: 0.7483  data: 0.1461  max mem: 41794
Epoch: [97]  [270/312]  eta: 0:00:32  lr: 0.003284  min_lr: 0.003284  loss: 3.1333 (3.5574)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.0781  max mem: 41794
Epoch: [97]  [280/312]  eta: 0:00:24  lr: 0.003284  min_lr: 0.003284  loss: 3.2807 (3.5604)  weight_decay: 0.0500 (0.0500)  time: 0.6807  data: 0.0850  max mem: 41794
Epoch: [97]  [290/312]  eta: 0:00:17  lr: 0.003283  min_lr: 0.003283  loss: 3.7170 (3.5638)  weight_decay: 0.0500 (0.0500)  time: 0.7890  data: 0.1072  max mem: 41794
Epoch: [97]  [300/312]  eta: 0:00:09  lr: 0.003283  min_lr: 0.003283  loss: 3.5821 (3.5562)  weight_decay: 0.0500 (0.0500)  time: 0.6109  data: 0.0443  max mem: 41794
Epoch: [97]  [310/312]  eta: 0:00:01  lr: 0.003282  min_lr: 0.003282  loss: 3.3552 (3.5496)  weight_decay: 0.0500 (0.0500)  time: 0.4870  data: 0.0218  max mem: 41794
Epoch: [97]  [311/312]  eta: 0:00:00  lr: 0.003282  min_lr: 0.003282  loss: 3.3552 (3.5509)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.0001  max mem: 41794
Epoch: [97] Total time: 0:03:58 (0.7641 s / it)
Averaged stats: lr: 0.003282  min_lr: 0.003282  loss: 3.3552 (3.5887)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:06  loss: 1.1836 (1.1836)  acc1: 75.3906 (75.3906)  acc5: 93.3594 (93.3594)  time: 7.4024  data: 7.2015  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5262 (1.4261)  acc1: 70.3125 (69.8080)  acc5: 89.3229 (90.1760)  time: 1.0076  data: 0.8372  max mem: 41794
Test: Total time: 0:00:09 (1.0177 s / it)
* Acc@1 70.148 Acc@5 90.120 loss 1.426
Accuracy of the model on the 50000 test images: 70.1%
Max accuracy: 70.28%
Epoch: [98]  [  0/312]  eta: 1:23:50  lr: 0.003282  min_lr: 0.003282  loss: 3.7143 (3.7143)  weight_decay: 0.0500 (0.0500)  time: 16.1221  data: 14.6619  max mem: 41794
Epoch: [98]  [ 10/312]  eta: 0:10:01  lr: 0.003281  min_lr: 0.003281  loss: 3.7143 (3.5896)  weight_decay: 0.0500 (0.0500)  time: 1.9907  data: 1.3607  max mem: 41794
Epoch: [98]  [ 20/312]  eta: 0:07:01  lr: 0.003281  min_lr: 0.003281  loss: 3.7123 (3.5973)  weight_decay: 0.0500 (0.0500)  time: 0.7107  data: 0.1487  max mem: 41794
Epoch: [98]  [ 30/312]  eta: 0:05:22  lr: 0.003280  min_lr: 0.003280  loss: 3.7505 (3.5972)  weight_decay: 0.0500 (0.0500)  time: 0.6772  data: 0.1337  max mem: 41794
Epoch: [98]  [ 40/312]  eta: 0:04:58  lr: 0.003280  min_lr: 0.003280  loss: 3.7704 (3.5911)  weight_decay: 0.0500 (0.0500)  time: 0.7342  data: 0.0971  max mem: 41794
Epoch: [98]  [ 50/312]  eta: 0:04:40  lr: 0.003279  min_lr: 0.003279  loss: 3.7065 (3.6173)  weight_decay: 0.0500 (0.0500)  time: 0.9539  data: 0.1625  max mem: 41794
Epoch: [98]  [ 60/312]  eta: 0:04:08  lr: 0.003279  min_lr: 0.003279  loss: 3.5818 (3.5855)  weight_decay: 0.0500 (0.0500)  time: 0.7548  data: 0.0775  max mem: 41794
Epoch: [98]  [ 70/312]  eta: 0:03:51  lr: 0.003278  min_lr: 0.003278  loss: 3.5398 (3.5747)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.0676  max mem: 41794
Epoch: [98]  [ 80/312]  eta: 0:03:34  lr: 0.003278  min_lr: 0.003278  loss: 3.7831 (3.6210)  weight_decay: 0.0500 (0.0500)  time: 0.7369  data: 0.0604  max mem: 41794
Epoch: [98]  [ 90/312]  eta: 0:03:17  lr: 0.003277  min_lr: 0.003277  loss: 3.8490 (3.6280)  weight_decay: 0.0500 (0.0500)  time: 0.6567  data: 0.0665  max mem: 41794
Epoch: [98]  [100/312]  eta: 0:03:07  lr: 0.003276  min_lr: 0.003276  loss: 3.8146 (3.6084)  weight_decay: 0.0500 (0.0500)  time: 0.7256  data: 0.1437  max mem: 41794
Epoch: [98]  [110/312]  eta: 0:02:51  lr: 0.003276  min_lr: 0.003276  loss: 3.1446 (3.5689)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.0838  max mem: 41794
Epoch: [98]  [120/312]  eta: 0:02:43  lr: 0.003275  min_lr: 0.003275  loss: 3.2477 (3.5737)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.0623  max mem: 41794
Epoch: [98]  [130/312]  eta: 0:02:34  lr: 0.003275  min_lr: 0.003275  loss: 3.7024 (3.5685)  weight_decay: 0.0500 (0.0500)  time: 0.8303  data: 0.0894  max mem: 41794
Epoch: [98]  [140/312]  eta: 0:02:22  lr: 0.003274  min_lr: 0.003274  loss: 3.7193 (3.5768)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.0571  max mem: 41794
Epoch: [98]  [150/312]  eta: 0:02:14  lr: 0.003274  min_lr: 0.003274  loss: 3.6881 (3.5840)  weight_decay: 0.0500 (0.0500)  time: 0.7143  data: 0.1213  max mem: 41794
Epoch: [98]  [160/312]  eta: 0:02:05  lr: 0.003273  min_lr: 0.003273  loss: 3.6198 (3.5818)  weight_decay: 0.0500 (0.0500)  time: 0.8063  data: 0.1661  max mem: 41794
Epoch: [98]  [170/312]  eta: 0:01:54  lr: 0.003273  min_lr: 0.003273  loss: 3.6645 (3.5825)  weight_decay: 0.0500 (0.0500)  time: 0.6147  data: 0.0764  max mem: 41794
Epoch: [98]  [180/312]  eta: 0:01:47  lr: 0.003272  min_lr: 0.003272  loss: 3.7437 (3.5872)  weight_decay: 0.0500 (0.0500)  time: 0.7374  data: 0.1559  max mem: 41794
Epoch: [98]  [190/312]  eta: 0:01:37  lr: 0.003271  min_lr: 0.003271  loss: 3.7031 (3.5914)  weight_decay: 0.0500 (0.0500)  time: 0.7367  data: 0.1539  max mem: 41794
Epoch: [98]  [200/312]  eta: 0:01:29  lr: 0.003271  min_lr: 0.003271  loss: 3.7857 (3.5994)  weight_decay: 0.0500 (0.0500)  time: 0.6644  data: 0.1005  max mem: 41794
Epoch: [98]  [210/312]  eta: 0:01:21  lr: 0.003270  min_lr: 0.003270  loss: 3.6644 (3.5977)  weight_decay: 0.0500 (0.0500)  time: 0.8253  data: 0.2131  max mem: 41794
Epoch: [98]  [220/312]  eta: 0:01:12  lr: 0.003270  min_lr: 0.003270  loss: 3.6627 (3.6070)  weight_decay: 0.0500 (0.0500)  time: 0.6559  data: 0.1145  max mem: 41794
Epoch: [98]  [230/312]  eta: 0:01:04  lr: 0.003269  min_lr: 0.003269  loss: 3.7761 (3.6128)  weight_decay: 0.0500 (0.0500)  time: 0.6625  data: 0.1651  max mem: 41794
Epoch: [98]  [240/312]  eta: 0:00:56  lr: 0.003269  min_lr: 0.003269  loss: 3.8169 (3.6298)  weight_decay: 0.0500 (0.0500)  time: 0.8153  data: 0.3161  max mem: 41794
Epoch: [98]  [250/312]  eta: 0:00:48  lr: 0.003268  min_lr: 0.003268  loss: 3.8712 (3.6379)  weight_decay: 0.0500 (0.0500)  time: 0.6466  data: 0.1516  max mem: 41794
Epoch: [98]  [260/312]  eta: 0:00:40  lr: 0.003268  min_lr: 0.003268  loss: 3.8414 (3.6333)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.2019  max mem: 41794
Epoch: [98]  [270/312]  eta: 0:00:32  lr: 0.003267  min_lr: 0.003267  loss: 3.6992 (3.6330)  weight_decay: 0.0500 (0.0500)  time: 0.6980  data: 0.2019  max mem: 41794
Epoch: [98]  [280/312]  eta: 0:00:24  lr: 0.003266  min_lr: 0.003266  loss: 3.6992 (3.6326)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.2034  max mem: 41794
Epoch: [98]  [290/312]  eta: 0:00:17  lr: 0.003266  min_lr: 0.003266  loss: 3.8694 (3.6438)  weight_decay: 0.0500 (0.0500)  time: 0.8567  data: 0.3621  max mem: 41794
Epoch: [98]  [300/312]  eta: 0:00:09  lr: 0.003265  min_lr: 0.003265  loss: 3.8762 (3.6475)  weight_decay: 0.0500 (0.0500)  time: 0.6457  data: 0.1591  max mem: 41794
Epoch: [98]  [310/312]  eta: 0:00:01  lr: 0.003265  min_lr: 0.003265  loss: 3.8144 (3.6432)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [98]  [311/312]  eta: 0:00:00  lr: 0.003265  min_lr: 0.003265  loss: 3.8279 (3.6443)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [98] Total time: 0:03:57 (0.7623 s / it)
Averaged stats: lr: 0.003265  min_lr: 0.003265  loss: 3.8279 (3.5970)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.3469 (1.3469)  acc1: 76.6927 (76.6927)  acc5: 94.6615 (94.6615)  time: 8.0934  data: 7.8823  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6677 (1.6264)  acc1: 70.4427 (70.2560)  acc5: 89.3229 (90.1440)  time: 1.0480  data: 0.8759  max mem: 41794
Test: Total time: 0:00:09 (1.0599 s / it)
* Acc@1 70.158 Acc@5 90.122 loss 1.629
Accuracy of the model on the 50000 test images: 70.2%
Max accuracy: 70.28%
Epoch: [99]  [  0/312]  eta: 1:23:26  lr: 0.003265  min_lr: 0.003265  loss: 3.6686 (3.6686)  weight_decay: 0.0500 (0.0500)  time: 16.0451  data: 15.1276  max mem: 41794
Epoch: [99]  [ 10/312]  eta: 0:10:59  lr: 0.003264  min_lr: 0.003264  loss: 3.4766 (3.4018)  weight_decay: 0.0500 (0.0500)  time: 2.1835  data: 1.3759  max mem: 41794
Epoch: [99]  [ 20/312]  eta: 0:07:16  lr: 0.003264  min_lr: 0.003264  loss: 3.6067 (3.5068)  weight_decay: 0.0500 (0.0500)  time: 0.7676  data: 0.1210  max mem: 41794
Epoch: [99]  [ 30/312]  eta: 0:05:31  lr: 0.003263  min_lr: 0.003263  loss: 3.7995 (3.5716)  weight_decay: 0.0500 (0.0500)  time: 0.6196  data: 0.1209  max mem: 41794
Epoch: [99]  [ 40/312]  eta: 0:04:45  lr: 0.003262  min_lr: 0.003262  loss: 3.5852 (3.5649)  weight_decay: 0.0500 (0.0500)  time: 0.5854  data: 0.0886  max mem: 41794
Epoch: [99]  [ 50/312]  eta: 0:04:23  lr: 0.003262  min_lr: 0.003262  loss: 3.7312 (3.5732)  weight_decay: 0.0500 (0.0500)  time: 0.7478  data: 0.2296  max mem: 41794
Epoch: [99]  [ 60/312]  eta: 0:03:52  lr: 0.003261  min_lr: 0.003261  loss: 3.7312 (3.5633)  weight_decay: 0.0500 (0.0500)  time: 0.6617  data: 0.1417  max mem: 41794
Epoch: [99]  [ 70/312]  eta: 0:03:41  lr: 0.003261  min_lr: 0.003261  loss: 3.5960 (3.5355)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.1559  max mem: 41794
Epoch: [99]  [ 80/312]  eta: 0:03:32  lr: 0.003260  min_lr: 0.003260  loss: 3.5942 (3.5286)  weight_decay: 0.0500 (0.0500)  time: 0.9001  data: 0.2708  max mem: 41794
Epoch: [99]  [ 90/312]  eta: 0:03:14  lr: 0.003260  min_lr: 0.003260  loss: 3.6052 (3.5423)  weight_decay: 0.0500 (0.0500)  time: 0.7382  data: 0.1418  max mem: 41794
Epoch: [99]  [100/312]  eta: 0:03:07  lr: 0.003259  min_lr: 0.003259  loss: 3.6016 (3.5347)  weight_decay: 0.0500 (0.0500)  time: 0.7602  data: 0.1287  max mem: 41794
Epoch: [99]  [110/312]  eta: 0:02:51  lr: 0.003258  min_lr: 0.003258  loss: 3.5460 (3.5311)  weight_decay: 0.0500 (0.0500)  time: 0.7330  data: 0.1025  max mem: 41794
Epoch: [99]  [120/312]  eta: 0:02:43  lr: 0.003258  min_lr: 0.003258  loss: 3.4633 (3.5213)  weight_decay: 0.0500 (0.0500)  time: 0.6673  data: 0.0521  max mem: 41794
Epoch: [99]  [130/312]  eta: 0:02:34  lr: 0.003257  min_lr: 0.003257  loss: 3.6500 (3.5348)  weight_decay: 0.0500 (0.0500)  time: 0.8327  data: 0.0963  max mem: 41794
Epoch: [99]  [140/312]  eta: 0:02:22  lr: 0.003257  min_lr: 0.003257  loss: 3.6688 (3.5368)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.0767  max mem: 41794
Epoch: [99]  [150/312]  eta: 0:02:13  lr: 0.003256  min_lr: 0.003256  loss: 3.6321 (3.5435)  weight_decay: 0.0500 (0.0500)  time: 0.6833  data: 0.0950  max mem: 41794
Epoch: [99]  [160/312]  eta: 0:02:04  lr: 0.003256  min_lr: 0.003256  loss: 3.8644 (3.5744)  weight_decay: 0.0500 (0.0500)  time: 0.7601  data: 0.0709  max mem: 41794
Epoch: [99]  [170/312]  eta: 0:01:54  lr: 0.003255  min_lr: 0.003255  loss: 3.9007 (3.5827)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.0831  max mem: 41794
Epoch: [99]  [180/312]  eta: 0:01:47  lr: 0.003255  min_lr: 0.003255  loss: 3.8755 (3.5942)  weight_decay: 0.0500 (0.0500)  time: 0.7820  data: 0.1649  max mem: 41794
Epoch: [99]  [190/312]  eta: 0:01:37  lr: 0.003254  min_lr: 0.003254  loss: 3.7156 (3.5915)  weight_decay: 0.0500 (0.0500)  time: 0.7064  data: 0.0905  max mem: 41794
Epoch: [99]  [200/312]  eta: 0:01:29  lr: 0.003253  min_lr: 0.003253  loss: 3.7156 (3.6015)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.0821  max mem: 41794
Epoch: [99]  [210/312]  eta: 0:01:21  lr: 0.003253  min_lr: 0.003253  loss: 3.7593 (3.6077)  weight_decay: 0.0500 (0.0500)  time: 0.8448  data: 0.1784  max mem: 41794
Epoch: [99]  [220/312]  eta: 0:01:12  lr: 0.003252  min_lr: 0.003252  loss: 3.8128 (3.6187)  weight_decay: 0.0500 (0.0500)  time: 0.6651  data: 0.0972  max mem: 41794
Epoch: [99]  [230/312]  eta: 0:01:04  lr: 0.003252  min_lr: 0.003252  loss: 3.8036 (3.6183)  weight_decay: 0.0500 (0.0500)  time: 0.6483  data: 0.0844  max mem: 41794
Epoch: [99]  [240/312]  eta: 0:00:56  lr: 0.003251  min_lr: 0.003251  loss: 3.8625 (3.6270)  weight_decay: 0.0500 (0.0500)  time: 0.7007  data: 0.0844  max mem: 41794
Epoch: [99]  [250/312]  eta: 0:00:48  lr: 0.003251  min_lr: 0.003251  loss: 3.9347 (3.6357)  weight_decay: 0.0500 (0.0500)  time: 0.6585  data: 0.1076  max mem: 41794
Epoch: [99]  [260/312]  eta: 0:00:40  lr: 0.003250  min_lr: 0.003250  loss: 3.8061 (3.6361)  weight_decay: 0.0500 (0.0500)  time: 0.7857  data: 0.2193  max mem: 41794
Epoch: [99]  [270/312]  eta: 0:00:32  lr: 0.003250  min_lr: 0.003250  loss: 3.8041 (3.6378)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.1125  max mem: 41794
Epoch: [99]  [280/312]  eta: 0:00:24  lr: 0.003249  min_lr: 0.003249  loss: 3.6365 (3.6299)  weight_decay: 0.0500 (0.0500)  time: 0.7135  data: 0.1134  max mem: 41794
Epoch: [99]  [290/312]  eta: 0:00:17  lr: 0.003248  min_lr: 0.003248  loss: 3.5288 (3.6256)  weight_decay: 0.0500 (0.0500)  time: 0.8373  data: 0.1798  max mem: 41794
Epoch: [99]  [300/312]  eta: 0:00:09  lr: 0.003248  min_lr: 0.003248  loss: 3.5288 (3.6209)  weight_decay: 0.0500 (0.0500)  time: 0.6102  data: 0.0669  max mem: 41794
Epoch: [99]  [310/312]  eta: 0:00:01  lr: 0.003247  min_lr: 0.003247  loss: 3.4028 (3.6166)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [99]  [311/312]  eta: 0:00:00  lr: 0.003247  min_lr: 0.003247  loss: 3.4028 (3.6174)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [99] Total time: 0:03:57 (0.7616 s / it)
Averaged stats: lr: 0.003247  min_lr: 0.003247  loss: 3.4028 (3.5855)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2799 (1.2799)  acc1: 77.4740 (77.4740)  acc5: 92.9688 (92.9688)  time: 8.4657  data: 8.2545  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5337 (1.5184)  acc1: 69.5312 (69.7280)  acc5: 91.4062 (90.0960)  time: 1.0916  data: 0.9173  max mem: 41794
Test: Total time: 0:00:10 (1.1365 s / it)
* Acc@1 69.510 Acc@5 89.704 loss 1.526
Accuracy of the model on the 50000 test images: 69.5%
Max accuracy: 70.28%
Epoch: [100]  [  0/312]  eta: 1:24:43  lr: 0.003247  min_lr: 0.003247  loss: 2.8634 (2.8634)  weight_decay: 0.0500 (0.0500)  time: 16.2942  data: 11.6750  max mem: 41794
Epoch: [100]  [ 10/312]  eta: 0:12:04  lr: 0.003247  min_lr: 0.003247  loss: 3.6119 (3.3395)  weight_decay: 0.0500 (0.0500)  time: 2.3974  data: 1.3248  max mem: 41794
Epoch: [100]  [ 20/312]  eta: 0:08:06  lr: 0.003246  min_lr: 0.003246  loss: 3.6119 (3.4409)  weight_decay: 0.0500 (0.0500)  time: 0.9359  data: 0.2376  max mem: 41794
Epoch: [100]  [ 30/312]  eta: 0:06:11  lr: 0.003245  min_lr: 0.003245  loss: 3.6790 (3.5054)  weight_decay: 0.0500 (0.0500)  time: 0.7223  data: 0.0930  max mem: 41794
Epoch: [100]  [ 40/312]  eta: 0:05:04  lr: 0.003245  min_lr: 0.003245  loss: 3.6855 (3.5444)  weight_decay: 0.0500 (0.0500)  time: 0.5409  data: 0.0006  max mem: 41794
Epoch: [100]  [ 50/312]  eta: 0:04:21  lr: 0.003244  min_lr: 0.003244  loss: 3.8002 (3.5716)  weight_decay: 0.0500 (0.0500)  time: 0.5062  data: 0.0009  max mem: 41794
Epoch: [100]  [ 60/312]  eta: 0:03:51  lr: 0.003244  min_lr: 0.003244  loss: 3.8053 (3.5719)  weight_decay: 0.0500 (0.0500)  time: 0.5075  data: 0.0064  max mem: 41794
Epoch: [100]  [ 70/312]  eta: 0:03:31  lr: 0.003243  min_lr: 0.003243  loss: 3.8053 (3.5697)  weight_decay: 0.0500 (0.0500)  time: 0.5529  data: 0.0111  max mem: 41794
Epoch: [100]  [ 80/312]  eta: 0:03:23  lr: 0.003243  min_lr: 0.003243  loss: 3.7570 (3.5893)  weight_decay: 0.0500 (0.0500)  time: 0.7535  data: 0.0070  max mem: 41794
Epoch: [100]  [ 90/312]  eta: 0:03:08  lr: 0.003242  min_lr: 0.003242  loss: 3.5632 (3.5781)  weight_decay: 0.0500 (0.0500)  time: 0.7573  data: 0.0407  max mem: 41794
Epoch: [100]  [100/312]  eta: 0:02:59  lr: 0.003242  min_lr: 0.003242  loss: 3.2118 (3.5419)  weight_decay: 0.0500 (0.0500)  time: 0.7337  data: 0.0998  max mem: 41794
Epoch: [100]  [110/312]  eta: 0:02:45  lr: 0.003241  min_lr: 0.003241  loss: 3.5692 (3.5486)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.0666  max mem: 41794
Epoch: [100]  [120/312]  eta: 0:02:38  lr: 0.003240  min_lr: 0.003240  loss: 3.5671 (3.5231)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.0698  max mem: 41794
Epoch: [100]  [130/312]  eta: 0:02:30  lr: 0.003240  min_lr: 0.003240  loss: 3.6482 (3.5408)  weight_decay: 0.0500 (0.0500)  time: 0.8714  data: 0.1104  max mem: 41794
Epoch: [100]  [140/312]  eta: 0:02:18  lr: 0.003239  min_lr: 0.003239  loss: 3.9443 (3.5704)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.0504  max mem: 41794
Epoch: [100]  [150/312]  eta: 0:02:10  lr: 0.003239  min_lr: 0.003239  loss: 3.6589 (3.5578)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.0494  max mem: 41794
Epoch: [100]  [160/312]  eta: 0:02:02  lr: 0.003238  min_lr: 0.003238  loss: 3.6141 (3.5641)  weight_decay: 0.0500 (0.0500)  time: 0.8268  data: 0.0486  max mem: 41794
Epoch: [100]  [170/312]  eta: 0:01:52  lr: 0.003238  min_lr: 0.003238  loss: 3.7377 (3.5662)  weight_decay: 0.0500 (0.0500)  time: 0.6766  data: 0.0395  max mem: 41794
Epoch: [100]  [180/312]  eta: 0:01:44  lr: 0.003237  min_lr: 0.003237  loss: 3.6258 (3.5620)  weight_decay: 0.0500 (0.0500)  time: 0.7016  data: 0.0630  max mem: 41794
Epoch: [100]  [190/312]  eta: 0:01:35  lr: 0.003236  min_lr: 0.003236  loss: 3.8032 (3.5811)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0329  max mem: 41794
Epoch: [100]  [200/312]  eta: 0:01:28  lr: 0.003236  min_lr: 0.003236  loss: 3.7606 (3.5715)  weight_decay: 0.0500 (0.0500)  time: 0.7120  data: 0.0562  max mem: 41794
Epoch: [100]  [210/312]  eta: 0:01:20  lr: 0.003235  min_lr: 0.003235  loss: 3.4424 (3.5706)  weight_decay: 0.0500 (0.0500)  time: 0.8771  data: 0.0761  max mem: 41794
Epoch: [100]  [220/312]  eta: 0:01:11  lr: 0.003235  min_lr: 0.003235  loss: 3.6797 (3.5759)  weight_decay: 0.0500 (0.0500)  time: 0.6762  data: 0.0287  max mem: 41794
Epoch: [100]  [230/312]  eta: 0:01:03  lr: 0.003234  min_lr: 0.003234  loss: 3.9170 (3.5851)  weight_decay: 0.0500 (0.0500)  time: 0.6738  data: 0.0242  max mem: 41794
Epoch: [100]  [240/312]  eta: 0:00:56  lr: 0.003234  min_lr: 0.003234  loss: 4.0479 (3.5944)  weight_decay: 0.0500 (0.0500)  time: 0.8178  data: 0.0230  max mem: 41794
Epoch: [100]  [250/312]  eta: 0:00:47  lr: 0.003233  min_lr: 0.003233  loss: 3.6894 (3.5905)  weight_decay: 0.0500 (0.0500)  time: 0.6677  data: 0.0253  max mem: 41794
Epoch: [100]  [260/312]  eta: 0:00:40  lr: 0.003232  min_lr: 0.003232  loss: 3.6246 (3.5931)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.0565  max mem: 41794
Epoch: [100]  [270/312]  eta: 0:00:32  lr: 0.003232  min_lr: 0.003232  loss: 3.7553 (3.5897)  weight_decay: 0.0500 (0.0500)  time: 0.7090  data: 0.0331  max mem: 41794
Epoch: [100]  [280/312]  eta: 0:00:24  lr: 0.003231  min_lr: 0.003231  loss: 3.6438 (3.5957)  weight_decay: 0.0500 (0.0500)  time: 0.6769  data: 0.0201  max mem: 41794
Epoch: [100]  [290/312]  eta: 0:00:17  lr: 0.003231  min_lr: 0.003231  loss: 3.6438 (3.5925)  weight_decay: 0.0500 (0.0500)  time: 0.9039  data: 0.0627  max mem: 41794
Epoch: [100]  [300/312]  eta: 0:00:09  lr: 0.003230  min_lr: 0.003230  loss: 3.7206 (3.5946)  weight_decay: 0.0500 (0.0500)  time: 0.7350  data: 0.0430  max mem: 41794
Epoch: [100]  [310/312]  eta: 0:00:01  lr: 0.003230  min_lr: 0.003230  loss: 3.6441 (3.5920)  weight_decay: 0.0500 (0.0500)  time: 0.4732  data: 0.0001  max mem: 41794
Epoch: [100]  [311/312]  eta: 0:00:00  lr: 0.003230  min_lr: 0.003230  loss: 3.6303 (3.5895)  weight_decay: 0.0500 (0.0500)  time: 0.4716  data: 0.0001  max mem: 41794
Epoch: [100] Total time: 0:03:57 (0.7623 s / it)
Averaged stats: lr: 0.003230  min_lr: 0.003230  loss: 3.6303 (3.5865)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.2951 (1.2951)  acc1: 76.1719 (76.1719)  acc5: 93.6198 (93.6198)  time: 7.9992  data: 7.7989  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5406 (1.4511)  acc1: 68.6198 (68.8640)  acc5: 90.1042 (89.7280)  time: 1.0387  data: 0.8667  max mem: 41794
Test: Total time: 0:00:09 (1.0468 s / it)
* Acc@1 69.118 Acc@5 89.810 loss 1.455
Accuracy of the model on the 50000 test images: 69.1%
Max accuracy: 70.28%
Epoch: [101]  [  0/312]  eta: 1:21:23  lr: 0.003230  min_lr: 0.003230  loss: 3.4413 (3.4413)  weight_decay: 0.0500 (0.0500)  time: 15.6530  data: 12.2728  max mem: 41794
Epoch: [101]  [ 10/312]  eta: 0:12:01  lr: 0.003229  min_lr: 0.003229  loss: 3.6691 (3.7666)  weight_decay: 0.0500 (0.0500)  time: 2.3883  data: 1.2577  max mem: 41794
Epoch: [101]  [ 20/312]  eta: 0:07:31  lr: 0.003228  min_lr: 0.003228  loss: 3.6455 (3.6756)  weight_decay: 0.0500 (0.0500)  time: 0.8399  data: 0.0816  max mem: 41794
Epoch: [101]  [ 30/312]  eta: 0:05:44  lr: 0.003228  min_lr: 0.003228  loss: 3.6260 (3.5872)  weight_decay: 0.0500 (0.0500)  time: 0.5784  data: 0.0169  max mem: 41794
Epoch: [101]  [ 40/312]  eta: 0:04:45  lr: 0.003227  min_lr: 0.003227  loss: 3.0066 (3.5137)  weight_decay: 0.0500 (0.0500)  time: 0.5297  data: 0.0230  max mem: 41794
Epoch: [101]  [ 50/312]  eta: 0:04:13  lr: 0.003227  min_lr: 0.003227  loss: 3.5299 (3.5097)  weight_decay: 0.0500 (0.0500)  time: 0.5756  data: 0.0128  max mem: 41794
Epoch: [101]  [ 60/312]  eta: 0:03:45  lr: 0.003226  min_lr: 0.003226  loss: 3.7129 (3.5375)  weight_decay: 0.0500 (0.0500)  time: 0.5740  data: 0.0064  max mem: 41794
Epoch: [101]  [ 70/312]  eta: 0:03:33  lr: 0.003226  min_lr: 0.003226  loss: 3.8670 (3.5835)  weight_decay: 0.0500 (0.0500)  time: 0.6583  data: 0.0771  max mem: 41794
Epoch: [101]  [ 80/312]  eta: 0:03:23  lr: 0.003225  min_lr: 0.003225  loss: 3.8907 (3.6087)  weight_decay: 0.0500 (0.0500)  time: 0.8274  data: 0.1536  max mem: 41794
Epoch: [101]  [ 90/312]  eta: 0:03:05  lr: 0.003224  min_lr: 0.003224  loss: 3.7459 (3.5966)  weight_decay: 0.0500 (0.0500)  time: 0.6826  data: 0.0835  max mem: 41794
Epoch: [101]  [100/312]  eta: 0:02:58  lr: 0.003224  min_lr: 0.003224  loss: 3.3732 (3.5703)  weight_decay: 0.0500 (0.0500)  time: 0.7037  data: 0.0583  max mem: 41794
Epoch: [101]  [110/312]  eta: 0:02:44  lr: 0.003223  min_lr: 0.003223  loss: 3.2736 (3.5623)  weight_decay: 0.0500 (0.0500)  time: 0.7019  data: 0.0549  max mem: 41794
Epoch: [101]  [120/312]  eta: 0:02:36  lr: 0.003223  min_lr: 0.003223  loss: 3.4180 (3.5575)  weight_decay: 0.0500 (0.0500)  time: 0.6888  data: 0.0387  max mem: 41794
Epoch: [101]  [130/312]  eta: 0:02:28  lr: 0.003222  min_lr: 0.003222  loss: 3.5914 (3.5692)  weight_decay: 0.0500 (0.0500)  time: 0.8396  data: 0.0632  max mem: 41794
Epoch: [101]  [140/312]  eta: 0:02:16  lr: 0.003222  min_lr: 0.003222  loss: 3.8142 (3.5821)  weight_decay: 0.0500 (0.0500)  time: 0.6602  data: 0.0319  max mem: 41794
Epoch: [101]  [150/312]  eta: 0:02:09  lr: 0.003221  min_lr: 0.003221  loss: 3.7796 (3.5738)  weight_decay: 0.0500 (0.0500)  time: 0.6921  data: 0.0434  max mem: 41794
Epoch: [101]  [160/312]  eta: 0:02:02  lr: 0.003220  min_lr: 0.003220  loss: 3.6626 (3.5812)  weight_decay: 0.0500 (0.0500)  time: 0.8599  data: 0.0367  max mem: 41794
Epoch: [101]  [170/312]  eta: 0:01:52  lr: 0.003220  min_lr: 0.003220  loss: 3.8684 (3.5979)  weight_decay: 0.0500 (0.0500)  time: 0.7143  data: 0.0253  max mem: 41794
Epoch: [101]  [180/312]  eta: 0:01:44  lr: 0.003219  min_lr: 0.003219  loss: 3.6708 (3.5877)  weight_decay: 0.0500 (0.0500)  time: 0.7292  data: 0.0371  max mem: 41794
Epoch: [101]  [190/312]  eta: 0:01:35  lr: 0.003219  min_lr: 0.003219  loss: 3.6445 (3.6014)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.0125  max mem: 41794
Epoch: [101]  [200/312]  eta: 0:01:27  lr: 0.003218  min_lr: 0.003218  loss: 3.8177 (3.6033)  weight_decay: 0.0500 (0.0500)  time: 0.6982  data: 0.0092  max mem: 41794
Epoch: [101]  [210/312]  eta: 0:01:20  lr: 0.003218  min_lr: 0.003218  loss: 3.7410 (3.6023)  weight_decay: 0.0500 (0.0500)  time: 0.8727  data: 0.0146  max mem: 41794
Epoch: [101]  [220/312]  eta: 0:01:11  lr: 0.003217  min_lr: 0.003217  loss: 3.6423 (3.5972)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.0061  max mem: 41794
Epoch: [101]  [230/312]  eta: 0:01:03  lr: 0.003216  min_lr: 0.003216  loss: 3.5621 (3.5895)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.0156  max mem: 41794
Epoch: [101]  [240/312]  eta: 0:00:56  lr: 0.003216  min_lr: 0.003216  loss: 3.5621 (3.5957)  weight_decay: 0.0500 (0.0500)  time: 0.8272  data: 0.0156  max mem: 41794
Epoch: [101]  [250/312]  eta: 0:00:47  lr: 0.003215  min_lr: 0.003215  loss: 3.8191 (3.6000)  weight_decay: 0.0500 (0.0500)  time: 0.7057  data: 0.0097  max mem: 41794
Epoch: [101]  [260/312]  eta: 0:00:40  lr: 0.003215  min_lr: 0.003215  loss: 3.8680 (3.6086)  weight_decay: 0.0500 (0.0500)  time: 0.7502  data: 0.0282  max mem: 41794
Epoch: [101]  [270/312]  eta: 0:00:32  lr: 0.003214  min_lr: 0.003214  loss: 3.8735 (3.6097)  weight_decay: 0.0500 (0.0500)  time: 0.7046  data: 0.0251  max mem: 41794
Epoch: [101]  [280/312]  eta: 0:00:24  lr: 0.003214  min_lr: 0.003214  loss: 3.8102 (3.6151)  weight_decay: 0.0500 (0.0500)  time: 0.6478  data: 0.0185  max mem: 41794
Epoch: [101]  [290/312]  eta: 0:00:17  lr: 0.003213  min_lr: 0.003213  loss: 3.7061 (3.6167)  weight_decay: 0.0500 (0.0500)  time: 0.8817  data: 0.0127  max mem: 41794
Epoch: [101]  [300/312]  eta: 0:00:09  lr: 0.003212  min_lr: 0.003212  loss: 3.6081 (3.6137)  weight_decay: 0.0500 (0.0500)  time: 0.7310  data: 0.0044  max mem: 41794
Epoch: [101]  [310/312]  eta: 0:00:01  lr: 0.003212  min_lr: 0.003212  loss: 3.6635 (3.6138)  weight_decay: 0.0500 (0.0500)  time: 0.4739  data: 0.0039  max mem: 41794
Epoch: [101]  [311/312]  eta: 0:00:00  lr: 0.003212  min_lr: 0.003212  loss: 3.6606 (3.6124)  weight_decay: 0.0500 (0.0500)  time: 0.4677  data: 0.0039  max mem: 41794
Epoch: [101] Total time: 0:03:58 (0.7632 s / it)
Averaged stats: lr: 0.003212  min_lr: 0.003212  loss: 3.6606 (3.5832)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:08  loss: 1.3688 (1.3688)  acc1: 76.3021 (76.3021)  acc5: 92.8385 (92.8385)  time: 7.6643  data: 7.4518  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7277 (1.6178)  acc1: 67.3177 (69.3280)  acc5: 89.8438 (89.8080)  time: 1.0548  data: 0.8830  max mem: 41794
Test: Total time: 0:00:09 (1.0871 s / it)
* Acc@1 69.396 Acc@5 89.762 loss 1.623
Accuracy of the model on the 50000 test images: 69.4%
Max accuracy: 70.28%
Epoch: [102]  [  0/312]  eta: 1:21:46  lr: 0.003212  min_lr: 0.003212  loss: 4.1126 (4.1126)  weight_decay: 0.0500 (0.0500)  time: 15.7252  data: 14.1198  max mem: 41794
Epoch: [102]  [ 10/312]  eta: 0:10:36  lr: 0.003211  min_lr: 0.003211  loss: 3.8853 (3.7169)  weight_decay: 0.0500 (0.0500)  time: 2.1073  data: 1.3560  max mem: 41794
Epoch: [102]  [ 20/312]  eta: 0:07:16  lr: 0.003211  min_lr: 0.003211  loss: 3.4864 (3.5879)  weight_decay: 0.0500 (0.0500)  time: 0.7832  data: 0.1570  max mem: 41794
Epoch: [102]  [ 30/312]  eta: 0:05:33  lr: 0.003210  min_lr: 0.003210  loss: 3.4223 (3.4967)  weight_decay: 0.0500 (0.0500)  time: 0.6741  data: 0.1224  max mem: 41794
Epoch: [102]  [ 40/312]  eta: 0:04:46  lr: 0.003209  min_lr: 0.003209  loss: 3.4310 (3.5111)  weight_decay: 0.0500 (0.0500)  time: 0.5880  data: 0.0781  max mem: 41794
Epoch: [102]  [ 50/312]  eta: 0:04:27  lr: 0.003209  min_lr: 0.003209  loss: 3.6718 (3.5268)  weight_decay: 0.0500 (0.0500)  time: 0.7670  data: 0.2594  max mem: 41794
Epoch: [102]  [ 60/312]  eta: 0:03:54  lr: 0.003208  min_lr: 0.003208  loss: 3.5875 (3.5148)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.1868  max mem: 41794
Epoch: [102]  [ 70/312]  eta: 0:03:41  lr: 0.003208  min_lr: 0.003208  loss: 3.3702 (3.5025)  weight_decay: 0.0500 (0.0500)  time: 0.6489  data: 0.1560  max mem: 41794
Epoch: [102]  [ 80/312]  eta: 0:03:31  lr: 0.003207  min_lr: 0.003207  loss: 3.3702 (3.5103)  weight_decay: 0.0500 (0.0500)  time: 0.8427  data: 0.3143  max mem: 41794
Epoch: [102]  [ 90/312]  eta: 0:03:12  lr: 0.003207  min_lr: 0.003207  loss: 3.6193 (3.5276)  weight_decay: 0.0500 (0.0500)  time: 0.7028  data: 0.1831  max mem: 41794
Epoch: [102]  [100/312]  eta: 0:03:05  lr: 0.003206  min_lr: 0.003206  loss: 3.8039 (3.5533)  weight_decay: 0.0500 (0.0500)  time: 0.7320  data: 0.1978  max mem: 41794
Epoch: [102]  [110/312]  eta: 0:02:49  lr: 0.003205  min_lr: 0.003205  loss: 3.8213 (3.5667)  weight_decay: 0.0500 (0.0500)  time: 0.7135  data: 0.1735  max mem: 41794
Epoch: [102]  [120/312]  eta: 0:02:41  lr: 0.003205  min_lr: 0.003205  loss: 3.6558 (3.5722)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.1461  max mem: 41794
Epoch: [102]  [130/312]  eta: 0:02:34  lr: 0.003204  min_lr: 0.003204  loss: 3.6558 (3.5819)  weight_decay: 0.0500 (0.0500)  time: 0.8927  data: 0.2859  max mem: 41794
Epoch: [102]  [140/312]  eta: 0:02:21  lr: 0.003204  min_lr: 0.003204  loss: 3.7621 (3.5700)  weight_decay: 0.0500 (0.0500)  time: 0.7078  data: 0.1419  max mem: 41794
Epoch: [102]  [150/312]  eta: 0:02:13  lr: 0.003203  min_lr: 0.003203  loss: 3.5331 (3.5624)  weight_decay: 0.0500 (0.0500)  time: 0.6530  data: 0.1338  max mem: 41794
Epoch: [102]  [160/312]  eta: 0:02:04  lr: 0.003203  min_lr: 0.003203  loss: 3.1127 (3.5397)  weight_decay: 0.0500 (0.0500)  time: 0.7948  data: 0.2507  max mem: 41794
Epoch: [102]  [170/312]  eta: 0:01:54  lr: 0.003202  min_lr: 0.003202  loss: 3.3016 (3.5437)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.1447  max mem: 41794
Epoch: [102]  [180/312]  eta: 0:01:47  lr: 0.003201  min_lr: 0.003201  loss: 3.6354 (3.5460)  weight_decay: 0.0500 (0.0500)  time: 0.7441  data: 0.1654  max mem: 41794
Epoch: [102]  [190/312]  eta: 0:01:37  lr: 0.003201  min_lr: 0.003201  loss: 3.6017 (3.5342)  weight_decay: 0.0500 (0.0500)  time: 0.7189  data: 0.1397  max mem: 41794
Epoch: [102]  [200/312]  eta: 0:01:28  lr: 0.003200  min_lr: 0.003200  loss: 3.6577 (3.5404)  weight_decay: 0.0500 (0.0500)  time: 0.6345  data: 0.1249  max mem: 41794
Epoch: [102]  [210/312]  eta: 0:01:21  lr: 0.003200  min_lr: 0.003200  loss: 3.7629 (3.5422)  weight_decay: 0.0500 (0.0500)  time: 0.8136  data: 0.2810  max mem: 41794
Epoch: [102]  [220/312]  eta: 0:01:12  lr: 0.003199  min_lr: 0.003199  loss: 3.4536 (3.5330)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1567  max mem: 41794
Epoch: [102]  [230/312]  eta: 0:01:04  lr: 0.003199  min_lr: 0.003199  loss: 3.4733 (3.5339)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.1492  max mem: 41794
Epoch: [102]  [240/312]  eta: 0:00:56  lr: 0.003198  min_lr: 0.003198  loss: 3.6053 (3.5383)  weight_decay: 0.0500 (0.0500)  time: 0.8550  data: 0.3245  max mem: 41794
Epoch: [102]  [250/312]  eta: 0:00:48  lr: 0.003197  min_lr: 0.003197  loss: 3.6778 (3.5444)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.1759  max mem: 41794
Epoch: [102]  [260/312]  eta: 0:00:40  lr: 0.003197  min_lr: 0.003197  loss: 3.7056 (3.5485)  weight_decay: 0.0500 (0.0500)  time: 0.7128  data: 0.1955  max mem: 41794
Epoch: [102]  [270/312]  eta: 0:00:32  lr: 0.003196  min_lr: 0.003196  loss: 3.6900 (3.5489)  weight_decay: 0.0500 (0.0500)  time: 0.7065  data: 0.1969  max mem: 41794
Epoch: [102]  [280/312]  eta: 0:00:24  lr: 0.003196  min_lr: 0.003196  loss: 3.7017 (3.5553)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.1792  max mem: 41794
Epoch: [102]  [290/312]  eta: 0:00:17  lr: 0.003195  min_lr: 0.003195  loss: 3.7017 (3.5595)  weight_decay: 0.0500 (0.0500)  time: 0.8124  data: 0.2865  max mem: 41794
Epoch: [102]  [300/312]  eta: 0:00:09  lr: 0.003195  min_lr: 0.003195  loss: 3.8119 (3.5728)  weight_decay: 0.0500 (0.0500)  time: 0.6100  data: 0.1091  max mem: 41794
Epoch: [102]  [310/312]  eta: 0:00:01  lr: 0.003194  min_lr: 0.003194  loss: 3.8040 (3.5717)  weight_decay: 0.0500 (0.0500)  time: 0.4656  data: 0.0001  max mem: 41794
Epoch: [102]  [311/312]  eta: 0:00:00  lr: 0.003194  min_lr: 0.003194  loss: 3.7568 (3.5708)  weight_decay: 0.0500 (0.0500)  time: 0.4626  data: 0.0001  max mem: 41794
Epoch: [102] Total time: 0:03:57 (0.7617 s / it)
Averaged stats: lr: 0.003194  min_lr: 0.003194  loss: 3.7568 (3.5735)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.2840 (1.2840)  acc1: 75.7812 (75.7812)  acc5: 91.4062 (91.4062)  time: 8.6493  data: 8.4510  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6481 (1.5626)  acc1: 67.5781 (68.7200)  acc5: 87.7604 (88.9120)  time: 1.1160  data: 0.9391  max mem: 41794
Test: Total time: 0:00:10 (1.1590 s / it)
* Acc@1 69.076 Acc@5 89.386 loss 1.544
Accuracy of the model on the 50000 test images: 69.1%
Max accuracy: 70.28%
Epoch: [103]  [  0/312]  eta: 1:19:40  lr: 0.003194  min_lr: 0.003194  loss: 4.2132 (4.2132)  weight_decay: 0.0500 (0.0500)  time: 15.3229  data: 12.9241  max mem: 41794
Epoch: [103]  [ 10/312]  eta: 0:10:43  lr: 0.003193  min_lr: 0.003193  loss: 3.8401 (3.7506)  weight_decay: 0.0500 (0.0500)  time: 2.1304  data: 1.3408  max mem: 41794
Epoch: [103]  [ 20/312]  eta: 0:07:10  lr: 0.003193  min_lr: 0.003193  loss: 3.8044 (3.7067)  weight_decay: 0.0500 (0.0500)  time: 0.7837  data: 0.1851  max mem: 41794
Epoch: [103]  [ 30/312]  eta: 0:05:28  lr: 0.003192  min_lr: 0.003192  loss: 3.7753 (3.7158)  weight_decay: 0.0500 (0.0500)  time: 0.6348  data: 0.0985  max mem: 41794
Epoch: [103]  [ 40/312]  eta: 0:04:46  lr: 0.003192  min_lr: 0.003192  loss: 3.7873 (3.7089)  weight_decay: 0.0500 (0.0500)  time: 0.6063  data: 0.0663  max mem: 41794
Epoch: [103]  [ 50/312]  eta: 0:04:27  lr: 0.003191  min_lr: 0.003191  loss: 3.8582 (3.7084)  weight_decay: 0.0500 (0.0500)  time: 0.7955  data: 0.1978  max mem: 41794
Epoch: [103]  [ 60/312]  eta: 0:03:55  lr: 0.003190  min_lr: 0.003190  loss: 3.8618 (3.7022)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.1365  max mem: 41794
Epoch: [103]  [ 70/312]  eta: 0:03:43  lr: 0.003190  min_lr: 0.003190  loss: 3.7357 (3.6811)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.0900  max mem: 41794
Epoch: [103]  [ 80/312]  eta: 0:03:28  lr: 0.003189  min_lr: 0.003189  loss: 3.5833 (3.6836)  weight_decay: 0.0500 (0.0500)  time: 0.7945  data: 0.1315  max mem: 41794
Epoch: [103]  [ 90/312]  eta: 0:03:14  lr: 0.003189  min_lr: 0.003189  loss: 3.7641 (3.6795)  weight_decay: 0.0500 (0.0500)  time: 0.7088  data: 0.1106  max mem: 41794
Epoch: [103]  [100/312]  eta: 0:03:05  lr: 0.003188  min_lr: 0.003188  loss: 3.6191 (3.6463)  weight_decay: 0.0500 (0.0500)  time: 0.7848  data: 0.1813  max mem: 41794
Epoch: [103]  [110/312]  eta: 0:02:50  lr: 0.003187  min_lr: 0.003187  loss: 3.7015 (3.6539)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.1129  max mem: 41794
Epoch: [103]  [120/312]  eta: 0:02:43  lr: 0.003187  min_lr: 0.003187  loss: 3.7253 (3.6460)  weight_decay: 0.0500 (0.0500)  time: 0.7105  data: 0.1280  max mem: 41794
Epoch: [103]  [130/312]  eta: 0:02:33  lr: 0.003186  min_lr: 0.003186  loss: 3.7253 (3.6438)  weight_decay: 0.0500 (0.0500)  time: 0.8586  data: 0.2105  max mem: 41794
Epoch: [103]  [140/312]  eta: 0:02:21  lr: 0.003186  min_lr: 0.003186  loss: 3.7639 (3.6467)  weight_decay: 0.0500 (0.0500)  time: 0.6440  data: 0.0831  max mem: 41794
Epoch: [103]  [150/312]  eta: 0:02:13  lr: 0.003185  min_lr: 0.003185  loss: 3.5959 (3.6358)  weight_decay: 0.0500 (0.0500)  time: 0.6975  data: 0.0979  max mem: 41794
Epoch: [103]  [160/312]  eta: 0:02:03  lr: 0.003185  min_lr: 0.003185  loss: 3.5959 (3.6301)  weight_decay: 0.0500 (0.0500)  time: 0.7804  data: 0.0980  max mem: 41794
Epoch: [103]  [170/312]  eta: 0:01:54  lr: 0.003184  min_lr: 0.003184  loss: 3.6915 (3.6370)  weight_decay: 0.0500 (0.0500)  time: 0.6865  data: 0.0853  max mem: 41794
Epoch: [103]  [180/312]  eta: 0:01:46  lr: 0.003183  min_lr: 0.003183  loss: 3.8220 (3.6438)  weight_decay: 0.0500 (0.0500)  time: 0.7689  data: 0.1833  max mem: 41794
Epoch: [103]  [190/312]  eta: 0:01:36  lr: 0.003183  min_lr: 0.003183  loss: 3.8444 (3.6409)  weight_decay: 0.0500 (0.0500)  time: 0.6607  data: 0.0987  max mem: 41794
Epoch: [103]  [200/312]  eta: 0:01:29  lr: 0.003182  min_lr: 0.003182  loss: 3.8497 (3.6486)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.0914  max mem: 41794
Epoch: [103]  [210/312]  eta: 0:01:21  lr: 0.003182  min_lr: 0.003182  loss: 3.8767 (3.6591)  weight_decay: 0.0500 (0.0500)  time: 0.8836  data: 0.1986  max mem: 41794
Epoch: [103]  [220/312]  eta: 0:01:12  lr: 0.003181  min_lr: 0.003181  loss: 3.7280 (3.6484)  weight_decay: 0.0500 (0.0500)  time: 0.7057  data: 0.1079  max mem: 41794
Epoch: [103]  [230/312]  eta: 0:01:04  lr: 0.003181  min_lr: 0.003181  loss: 3.5052 (3.6376)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.0929  max mem: 41794
Epoch: [103]  [240/312]  eta: 0:00:56  lr: 0.003180  min_lr: 0.003180  loss: 3.4600 (3.6354)  weight_decay: 0.0500 (0.0500)  time: 0.7333  data: 0.0929  max mem: 41794
Epoch: [103]  [250/312]  eta: 0:00:48  lr: 0.003179  min_lr: 0.003179  loss: 3.6861 (3.6398)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.0772  max mem: 41794
Epoch: [103]  [260/312]  eta: 0:00:40  lr: 0.003179  min_lr: 0.003179  loss: 3.7108 (3.6339)  weight_decay: 0.0500 (0.0500)  time: 0.8042  data: 0.1769  max mem: 41794
Epoch: [103]  [270/312]  eta: 0:00:32  lr: 0.003178  min_lr: 0.003178  loss: 3.7866 (3.6360)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1004  max mem: 41794
Epoch: [103]  [280/312]  eta: 0:00:24  lr: 0.003178  min_lr: 0.003178  loss: 3.8562 (3.6305)  weight_decay: 0.0500 (0.0500)  time: 0.6624  data: 0.0874  max mem: 41794
Epoch: [103]  [290/312]  eta: 0:00:17  lr: 0.003177  min_lr: 0.003177  loss: 3.5569 (3.6308)  weight_decay: 0.0500 (0.0500)  time: 0.7758  data: 0.1533  max mem: 41794
Epoch: [103]  [300/312]  eta: 0:00:09  lr: 0.003176  min_lr: 0.003176  loss: 3.6377 (3.6245)  weight_decay: 0.0500 (0.0500)  time: 0.6003  data: 0.0663  max mem: 41794
Epoch: [103]  [310/312]  eta: 0:00:01  lr: 0.003176  min_lr: 0.003176  loss: 3.8173 (3.6319)  weight_decay: 0.0500 (0.0500)  time: 0.4669  data: 0.0001  max mem: 41794
Epoch: [103]  [311/312]  eta: 0:00:00  lr: 0.003176  min_lr: 0.003176  loss: 3.8173 (3.6291)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.0001  max mem: 41794
Epoch: [103] Total time: 0:03:57 (0.7608 s / it)
Averaged stats: lr: 0.003176  min_lr: 0.003176  loss: 3.8173 (3.5792)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.2098 (1.2098)  acc1: 75.1302 (75.1302)  acc5: 93.7500 (93.7500)  time: 8.2837  data: 8.0718  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5967 (1.4962)  acc1: 67.4479 (69.0880)  acc5: 89.1927 (89.7440)  time: 1.0684  data: 0.8970  max mem: 41794
Test: Total time: 0:00:09 (1.1087 s / it)
* Acc@1 68.990 Acc@5 89.558 loss 1.510
Accuracy of the model on the 50000 test images: 69.0%
Max accuracy: 70.28%
Epoch: [104]  [  0/312]  eta: 1:21:28  lr: 0.003176  min_lr: 0.003176  loss: 2.9044 (2.9044)  weight_decay: 0.0500 (0.0500)  time: 15.6670  data: 14.7618  max mem: 41794
Epoch: [104]  [ 10/312]  eta: 0:11:04  lr: 0.003175  min_lr: 0.003175  loss: 3.2805 (3.4182)  weight_decay: 0.0500 (0.0500)  time: 2.1999  data: 1.3557  max mem: 41794
Epoch: [104]  [ 20/312]  eta: 0:07:22  lr: 0.003175  min_lr: 0.003175  loss: 3.4477 (3.5076)  weight_decay: 0.0500 (0.0500)  time: 0.8065  data: 0.1237  max mem: 41794
Epoch: [104]  [ 30/312]  eta: 0:05:37  lr: 0.003174  min_lr: 0.003174  loss: 3.8179 (3.5454)  weight_decay: 0.0500 (0.0500)  time: 0.6453  data: 0.1165  max mem: 41794
Epoch: [104]  [ 40/312]  eta: 0:04:46  lr: 0.003173  min_lr: 0.003173  loss: 3.7132 (3.5010)  weight_decay: 0.0500 (0.0500)  time: 0.5657  data: 0.0562  max mem: 41794
Epoch: [104]  [ 50/312]  eta: 0:04:23  lr: 0.003173  min_lr: 0.003173  loss: 3.2549 (3.4782)  weight_decay: 0.0500 (0.0500)  time: 0.7101  data: 0.2177  max mem: 41794
Epoch: [104]  [ 60/312]  eta: 0:03:52  lr: 0.003172  min_lr: 0.003172  loss: 3.5662 (3.4999)  weight_decay: 0.0500 (0.0500)  time: 0.6570  data: 0.1621  max mem: 41794
Epoch: [104]  [ 70/312]  eta: 0:03:42  lr: 0.003172  min_lr: 0.003172  loss: 3.4804 (3.4897)  weight_decay: 0.0500 (0.0500)  time: 0.7018  data: 0.2068  max mem: 41794
Epoch: [104]  [ 80/312]  eta: 0:03:30  lr: 0.003171  min_lr: 0.003171  loss: 3.4804 (3.4895)  weight_decay: 0.0500 (0.0500)  time: 0.8704  data: 0.3723  max mem: 41794
Epoch: [104]  [ 90/312]  eta: 0:03:12  lr: 0.003171  min_lr: 0.003171  loss: 3.5699 (3.4870)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.1818  max mem: 41794
Epoch: [104]  [100/312]  eta: 0:03:05  lr: 0.003170  min_lr: 0.003170  loss: 3.6000 (3.4776)  weight_decay: 0.0500 (0.0500)  time: 0.7240  data: 0.2312  max mem: 41794
Epoch: [104]  [110/312]  eta: 0:02:49  lr: 0.003169  min_lr: 0.003169  loss: 3.4292 (3.4622)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.2155  max mem: 41794
Epoch: [104]  [120/312]  eta: 0:02:40  lr: 0.003169  min_lr: 0.003169  loss: 3.7015 (3.4915)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1661  max mem: 41794
Epoch: [104]  [130/312]  eta: 0:02:33  lr: 0.003168  min_lr: 0.003168  loss: 3.7842 (3.5122)  weight_decay: 0.0500 (0.0500)  time: 0.8755  data: 0.3764  max mem: 41794
Epoch: [104]  [140/312]  eta: 0:02:20  lr: 0.003168  min_lr: 0.003168  loss: 3.7377 (3.5184)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.2110  max mem: 41794
Epoch: [104]  [150/312]  eta: 0:02:13  lr: 0.003167  min_lr: 0.003167  loss: 3.4836 (3.5145)  weight_decay: 0.0500 (0.0500)  time: 0.6986  data: 0.1896  max mem: 41794
Epoch: [104]  [160/312]  eta: 0:02:05  lr: 0.003166  min_lr: 0.003166  loss: 3.4310 (3.5121)  weight_decay: 0.0500 (0.0500)  time: 0.8799  data: 0.3047  max mem: 41794
Epoch: [104]  [170/312]  eta: 0:01:55  lr: 0.003166  min_lr: 0.003166  loss: 3.3510 (3.5065)  weight_decay: 0.0500 (0.0500)  time: 0.7176  data: 0.1632  max mem: 41794
Epoch: [104]  [180/312]  eta: 0:01:48  lr: 0.003165  min_lr: 0.003165  loss: 3.3279 (3.5081)  weight_decay: 0.0500 (0.0500)  time: 0.7725  data: 0.1665  max mem: 41794
Epoch: [104]  [190/312]  eta: 0:01:38  lr: 0.003165  min_lr: 0.003165  loss: 3.8513 (3.5309)  weight_decay: 0.0500 (0.0500)  time: 0.7296  data: 0.1191  max mem: 41794
Epoch: [104]  [200/312]  eta: 0:01:30  lr: 0.003164  min_lr: 0.003164  loss: 3.8907 (3.5375)  weight_decay: 0.0500 (0.0500)  time: 0.6618  data: 0.0833  max mem: 41794
Epoch: [104]  [210/312]  eta: 0:01:22  lr: 0.003164  min_lr: 0.003164  loss: 3.6286 (3.5314)  weight_decay: 0.0500 (0.0500)  time: 0.8672  data: 0.1530  max mem: 41794
Epoch: [104]  [220/312]  eta: 0:01:13  lr: 0.003163  min_lr: 0.003163  loss: 3.6625 (3.5360)  weight_decay: 0.0500 (0.0500)  time: 0.7077  data: 0.0718  max mem: 41794
Epoch: [104]  [230/312]  eta: 0:01:05  lr: 0.003162  min_lr: 0.003162  loss: 3.7599 (3.5388)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.0646  max mem: 41794
Epoch: [104]  [240/312]  eta: 0:00:57  lr: 0.003162  min_lr: 0.003162  loss: 3.6735 (3.5408)  weight_decay: 0.0500 (0.0500)  time: 0.8050  data: 0.0665  max mem: 41794
Epoch: [104]  [250/312]  eta: 0:00:48  lr: 0.003161  min_lr: 0.003161  loss: 3.4701 (3.5374)  weight_decay: 0.0500 (0.0500)  time: 0.6918  data: 0.0747  max mem: 41794
Epoch: [104]  [260/312]  eta: 0:00:41  lr: 0.003161  min_lr: 0.003161  loss: 3.4701 (3.5355)  weight_decay: 0.0500 (0.0500)  time: 0.7022  data: 0.1290  max mem: 41794
Epoch: [104]  [270/312]  eta: 0:00:32  lr: 0.003160  min_lr: 0.003160  loss: 3.6967 (3.5423)  weight_decay: 0.0500 (0.0500)  time: 0.6341  data: 0.0582  max mem: 41794
Epoch: [104]  [280/312]  eta: 0:00:25  lr: 0.003159  min_lr: 0.003159  loss: 3.7406 (3.5505)  weight_decay: 0.0500 (0.0500)  time: 0.7261  data: 0.1117  max mem: 41794
Epoch: [104]  [290/312]  eta: 0:00:17  lr: 0.003159  min_lr: 0.003159  loss: 3.7859 (3.5557)  weight_decay: 0.0500 (0.0500)  time: 0.8215  data: 0.1430  max mem: 41794
Epoch: [104]  [300/312]  eta: 0:00:09  lr: 0.003158  min_lr: 0.003158  loss: 3.6011 (3.5528)  weight_decay: 0.0500 (0.0500)  time: 0.5795  data: 0.0317  max mem: 41794
Epoch: [104]  [310/312]  eta: 0:00:01  lr: 0.003158  min_lr: 0.003158  loss: 3.5335 (3.5519)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0001  max mem: 41794
Epoch: [104]  [311/312]  eta: 0:00:00  lr: 0.003158  min_lr: 0.003158  loss: 3.5335 (3.5538)  weight_decay: 0.0500 (0.0500)  time: 0.4630  data: 0.0001  max mem: 41794
Epoch: [104] Total time: 0:03:59 (0.7669 s / it)
Averaged stats: lr: 0.003158  min_lr: 0.003158  loss: 3.5335 (3.5827)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.3369 (1.3369)  acc1: 78.1250 (78.1250)  acc5: 92.4479 (92.4479)  time: 8.0190  data: 7.8060  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7470 (1.6467)  acc1: 68.8802 (69.3280)  acc5: 89.7135 (89.6160)  time: 1.0684  data: 0.8968  max mem: 41794
Test: Total time: 0:00:09 (1.0773 s / it)
* Acc@1 69.806 Acc@5 89.912 loss 1.632
Accuracy of the model on the 50000 test images: 69.8%
Max accuracy: 70.28%
Epoch: [105]  [  0/312]  eta: 1:21:16  lr: 0.003158  min_lr: 0.003158  loss: 2.4697 (2.4697)  weight_decay: 0.0500 (0.0500)  time: 15.6308  data: 13.2316  max mem: 41794
Epoch: [105]  [ 10/312]  eta: 0:11:08  lr: 0.003157  min_lr: 0.003157  loss: 3.7414 (3.4880)  weight_decay: 0.0500 (0.0500)  time: 2.2122  data: 1.2040  max mem: 41794
Epoch: [105]  [ 20/312]  eta: 0:07:29  lr: 0.003156  min_lr: 0.003156  loss: 3.7414 (3.5801)  weight_decay: 0.0500 (0.0500)  time: 0.8361  data: 0.0370  max mem: 41794
Epoch: [105]  [ 30/312]  eta: 0:05:42  lr: 0.003156  min_lr: 0.003156  loss: 3.6271 (3.5815)  weight_decay: 0.0500 (0.0500)  time: 0.6663  data: 0.0414  max mem: 41794
Epoch: [105]  [ 40/312]  eta: 0:04:43  lr: 0.003155  min_lr: 0.003155  loss: 3.7509 (3.5917)  weight_decay: 0.0500 (0.0500)  time: 0.5196  data: 0.0074  max mem: 41794
Epoch: [105]  [ 50/312]  eta: 0:04:20  lr: 0.003155  min_lr: 0.003155  loss: 3.6876 (3.5696)  weight_decay: 0.0500 (0.0500)  time: 0.6484  data: 0.0824  max mem: 41794
Epoch: [105]  [ 60/312]  eta: 0:03:49  lr: 0.003154  min_lr: 0.003154  loss: 3.6876 (3.5969)  weight_decay: 0.0500 (0.0500)  time: 0.6422  data: 0.0835  max mem: 41794
Epoch: [105]  [ 70/312]  eta: 0:03:39  lr: 0.003153  min_lr: 0.003153  loss: 3.8421 (3.6471)  weight_decay: 0.0500 (0.0500)  time: 0.6877  data: 0.1309  max mem: 41794
Epoch: [105]  [ 80/312]  eta: 0:03:28  lr: 0.003153  min_lr: 0.003153  loss: 3.8250 (3.6218)  weight_decay: 0.0500 (0.0500)  time: 0.8521  data: 0.2817  max mem: 41794
Epoch: [105]  [ 90/312]  eta: 0:03:09  lr: 0.003152  min_lr: 0.003152  loss: 3.9456 (3.6587)  weight_decay: 0.0500 (0.0500)  time: 0.6646  data: 0.1546  max mem: 41794
Epoch: [105]  [100/312]  eta: 0:03:01  lr: 0.003152  min_lr: 0.003152  loss: 3.8597 (3.6654)  weight_decay: 0.0500 (0.0500)  time: 0.6978  data: 0.1708  max mem: 41794
Epoch: [105]  [110/312]  eta: 0:02:46  lr: 0.003151  min_lr: 0.003151  loss: 3.7867 (3.6533)  weight_decay: 0.0500 (0.0500)  time: 0.6905  data: 0.1707  max mem: 41794
Epoch: [105]  [120/312]  eta: 0:02:39  lr: 0.003151  min_lr: 0.003151  loss: 3.4741 (3.6277)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1831  max mem: 41794
Epoch: [105]  [130/312]  eta: 0:02:31  lr: 0.003150  min_lr: 0.003150  loss: 3.4663 (3.6159)  weight_decay: 0.0500 (0.0500)  time: 0.8850  data: 0.3175  max mem: 41794
Epoch: [105]  [140/312]  eta: 0:02:19  lr: 0.003149  min_lr: 0.003149  loss: 3.5679 (3.6054)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.1351  max mem: 41794
Epoch: [105]  [150/312]  eta: 0:02:12  lr: 0.003149  min_lr: 0.003149  loss: 3.5679 (3.6062)  weight_decay: 0.0500 (0.0500)  time: 0.7174  data: 0.1345  max mem: 41794
Epoch: [105]  [160/312]  eta: 0:02:03  lr: 0.003148  min_lr: 0.003148  loss: 3.5331 (3.6006)  weight_decay: 0.0500 (0.0500)  time: 0.8503  data: 0.2288  max mem: 41794
Epoch: [105]  [170/312]  eta: 0:01:53  lr: 0.003148  min_lr: 0.003148  loss: 3.6320 (3.5882)  weight_decay: 0.0500 (0.0500)  time: 0.6407  data: 0.0950  max mem: 41794
Epoch: [105]  [180/312]  eta: 0:01:46  lr: 0.003147  min_lr: 0.003147  loss: 3.6979 (3.6066)  weight_decay: 0.0500 (0.0500)  time: 0.7294  data: 0.1339  max mem: 41794
Epoch: [105]  [190/312]  eta: 0:01:36  lr: 0.003146  min_lr: 0.003146  loss: 3.8986 (3.6017)  weight_decay: 0.0500 (0.0500)  time: 0.7232  data: 0.1355  max mem: 41794
Epoch: [105]  [200/312]  eta: 0:01:29  lr: 0.003146  min_lr: 0.003146  loss: 3.3237 (3.5901)  weight_decay: 0.0500 (0.0500)  time: 0.7280  data: 0.0793  max mem: 41794
Epoch: [105]  [210/312]  eta: 0:01:21  lr: 0.003145  min_lr: 0.003145  loss: 3.5582 (3.5884)  weight_decay: 0.0500 (0.0500)  time: 0.9128  data: 0.1190  max mem: 41794
Epoch: [105]  [220/312]  eta: 0:01:12  lr: 0.003145  min_lr: 0.003145  loss: 3.5170 (3.5792)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.0467  max mem: 41794
Epoch: [105]  [230/312]  eta: 0:01:04  lr: 0.003144  min_lr: 0.003144  loss: 3.4574 (3.5745)  weight_decay: 0.0500 (0.0500)  time: 0.6737  data: 0.0503  max mem: 41794
Epoch: [105]  [240/312]  eta: 0:00:56  lr: 0.003143  min_lr: 0.003143  loss: 3.5435 (3.5682)  weight_decay: 0.0500 (0.0500)  time: 0.8095  data: 0.0457  max mem: 41794
Epoch: [105]  [250/312]  eta: 0:00:48  lr: 0.003143  min_lr: 0.003143  loss: 3.7271 (3.5759)  weight_decay: 0.0500 (0.0500)  time: 0.7083  data: 0.0392  max mem: 41794
Epoch: [105]  [260/312]  eta: 0:00:40  lr: 0.003142  min_lr: 0.003142  loss: 3.6495 (3.5719)  weight_decay: 0.0500 (0.0500)  time: 0.7389  data: 0.0639  max mem: 41794
Epoch: [105]  [270/312]  eta: 0:00:32  lr: 0.003142  min_lr: 0.003142  loss: 3.7941 (3.5796)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.0299  max mem: 41794
Epoch: [105]  [280/312]  eta: 0:00:24  lr: 0.003141  min_lr: 0.003141  loss: 3.7963 (3.5799)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.0402  max mem: 41794
Epoch: [105]  [290/312]  eta: 0:00:17  lr: 0.003140  min_lr: 0.003140  loss: 3.7409 (3.5926)  weight_decay: 0.0500 (0.0500)  time: 0.8286  data: 0.0354  max mem: 41794
Epoch: [105]  [300/312]  eta: 0:00:09  lr: 0.003140  min_lr: 0.003140  loss: 3.7467 (3.6010)  weight_decay: 0.0500 (0.0500)  time: 0.5924  data: 0.0002  max mem: 41794
Epoch: [105]  [310/312]  eta: 0:00:01  lr: 0.003139  min_lr: 0.003139  loss: 3.6381 (3.5891)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [105]  [311/312]  eta: 0:00:00  lr: 0.003139  min_lr: 0.003139  loss: 3.6867 (3.5900)  weight_decay: 0.0500 (0.0500)  time: 0.4643  data: 0.0001  max mem: 41794
Epoch: [105] Total time: 0:03:58 (0.7641 s / it)
Averaged stats: lr: 0.003139  min_lr: 0.003139  loss: 3.6867 (3.5664)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.2011 (1.2011)  acc1: 74.7396 (74.7396)  acc5: 93.3594 (93.3594)  time: 8.4389  data: 8.2264  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5043 (1.3888)  acc1: 71.6146 (71.1520)  acc5: 90.6250 (90.8160)  time: 1.0859  data: 0.9141  max mem: 41794
Test: Total time: 0:00:09 (1.1033 s / it)
* Acc@1 70.740 Acc@5 90.418 loss 1.412
Accuracy of the model on the 50000 test images: 70.7%
Max accuracy: 70.74%
Epoch: [106]  [  0/312]  eta: 1:21:47  lr: 0.003139  min_lr: 0.003139  loss: 3.1129 (3.1129)  weight_decay: 0.0500 (0.0500)  time: 15.7292  data: 12.1732  max mem: 41794
Epoch: [106]  [ 10/312]  eta: 0:10:48  lr: 0.003139  min_lr: 0.003139  loss: 3.7264 (3.6449)  weight_decay: 0.0500 (0.0500)  time: 2.1487  data: 1.3527  max mem: 41794
Epoch: [106]  [ 20/312]  eta: 0:07:36  lr: 0.003138  min_lr: 0.003138  loss: 3.7634 (3.6667)  weight_decay: 0.0500 (0.0500)  time: 0.8551  data: 0.2460  max mem: 41794
Epoch: [106]  [ 30/312]  eta: 0:05:48  lr: 0.003137  min_lr: 0.003137  loss: 3.6866 (3.6193)  weight_decay: 0.0500 (0.0500)  time: 0.7344  data: 0.1158  max mem: 41794
Epoch: [106]  [ 40/312]  eta: 0:04:50  lr: 0.003137  min_lr: 0.003137  loss: 3.8069 (3.6800)  weight_decay: 0.0500 (0.0500)  time: 0.5516  data: 0.0208  max mem: 41794
Epoch: [106]  [ 50/312]  eta: 0:04:21  lr: 0.003136  min_lr: 0.003136  loss: 3.8319 (3.6675)  weight_decay: 0.0500 (0.0500)  time: 0.6295  data: 0.1067  max mem: 41794
Epoch: [106]  [ 60/312]  eta: 0:03:51  lr: 0.003136  min_lr: 0.003136  loss: 3.6920 (3.6493)  weight_decay: 0.0500 (0.0500)  time: 0.6061  data: 0.0914  max mem: 41794
Epoch: [106]  [ 70/312]  eta: 0:03:37  lr: 0.003135  min_lr: 0.003135  loss: 3.3559 (3.5743)  weight_decay: 0.0500 (0.0500)  time: 0.6422  data: 0.1379  max mem: 41794
Epoch: [106]  [ 80/312]  eta: 0:03:26  lr: 0.003134  min_lr: 0.003134  loss: 3.3213 (3.5445)  weight_decay: 0.0500 (0.0500)  time: 0.8105  data: 0.3143  max mem: 41794
Epoch: [106]  [ 90/312]  eta: 0:03:08  lr: 0.003134  min_lr: 0.003134  loss: 3.6533 (3.5628)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1771  max mem: 41794
Epoch: [106]  [100/312]  eta: 0:03:01  lr: 0.003133  min_lr: 0.003133  loss: 3.6533 (3.5630)  weight_decay: 0.0500 (0.0500)  time: 0.7033  data: 0.1712  max mem: 41794
Epoch: [106]  [110/312]  eta: 0:02:45  lr: 0.003133  min_lr: 0.003133  loss: 3.6370 (3.5602)  weight_decay: 0.0500 (0.0500)  time: 0.6903  data: 0.1711  max mem: 41794
Epoch: [106]  [120/312]  eta: 0:02:38  lr: 0.003132  min_lr: 0.003132  loss: 3.7144 (3.5647)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.1836  max mem: 41794
Epoch: [106]  [130/312]  eta: 0:02:31  lr: 0.003131  min_lr: 0.003131  loss: 3.6995 (3.5644)  weight_decay: 0.0500 (0.0500)  time: 0.8909  data: 0.3937  max mem: 41794
Epoch: [106]  [140/312]  eta: 0:02:18  lr: 0.003131  min_lr: 0.003131  loss: 3.5712 (3.5536)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.2108  max mem: 41794
Epoch: [106]  [150/312]  eta: 0:02:11  lr: 0.003130  min_lr: 0.003130  loss: 3.2947 (3.5394)  weight_decay: 0.0500 (0.0500)  time: 0.6931  data: 0.1953  max mem: 41794
Epoch: [106]  [160/312]  eta: 0:02:03  lr: 0.003130  min_lr: 0.003130  loss: 3.4988 (3.5417)  weight_decay: 0.0500 (0.0500)  time: 0.8571  data: 0.3622  max mem: 41794
Epoch: [106]  [170/312]  eta: 0:01:53  lr: 0.003129  min_lr: 0.003129  loss: 3.7231 (3.5600)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1708  max mem: 41794
Epoch: [106]  [180/312]  eta: 0:01:45  lr: 0.003129  min_lr: 0.003129  loss: 3.7661 (3.5565)  weight_decay: 0.0500 (0.0500)  time: 0.6977  data: 0.1971  max mem: 41794
Epoch: [106]  [190/312]  eta: 0:01:35  lr: 0.003128  min_lr: 0.003128  loss: 3.5521 (3.5597)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.1937  max mem: 41794
Epoch: [106]  [200/312]  eta: 0:01:27  lr: 0.003127  min_lr: 0.003127  loss: 3.5521 (3.5633)  weight_decay: 0.0500 (0.0500)  time: 0.6433  data: 0.1353  max mem: 41794
Epoch: [106]  [210/312]  eta: 0:01:20  lr: 0.003127  min_lr: 0.003127  loss: 3.4980 (3.5651)  weight_decay: 0.0500 (0.0500)  time: 0.8456  data: 0.3277  max mem: 41794
Epoch: [106]  [220/312]  eta: 0:01:11  lr: 0.003126  min_lr: 0.003126  loss: 3.6515 (3.5693)  weight_decay: 0.0500 (0.0500)  time: 0.6957  data: 0.1943  max mem: 41794
Epoch: [106]  [230/312]  eta: 0:01:04  lr: 0.003126  min_lr: 0.003126  loss: 3.6819 (3.5754)  weight_decay: 0.0500 (0.0500)  time: 0.6886  data: 0.1910  max mem: 41794
Epoch: [106]  [240/312]  eta: 0:00:56  lr: 0.003125  min_lr: 0.003125  loss: 3.6122 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.8603  data: 0.3630  max mem: 41794
Epoch: [106]  [250/312]  eta: 0:00:47  lr: 0.003124  min_lr: 0.003124  loss: 3.7586 (3.5858)  weight_decay: 0.0500 (0.0500)  time: 0.6829  data: 0.1887  max mem: 41794
Epoch: [106]  [260/312]  eta: 0:00:40  lr: 0.003124  min_lr: 0.003124  loss: 3.9327 (3.5929)  weight_decay: 0.0500 (0.0500)  time: 0.7121  data: 0.2147  max mem: 41794
Epoch: [106]  [270/312]  eta: 0:00:32  lr: 0.003123  min_lr: 0.003123  loss: 3.7794 (3.5900)  weight_decay: 0.0500 (0.0500)  time: 0.7050  data: 0.1999  max mem: 41794
Epoch: [106]  [280/312]  eta: 0:00:24  lr: 0.003123  min_lr: 0.003123  loss: 3.6916 (3.5964)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1852  max mem: 41794
Epoch: [106]  [290/312]  eta: 0:00:17  lr: 0.003122  min_lr: 0.003122  loss: 3.6101 (3.5907)  weight_decay: 0.0500 (0.0500)  time: 0.8529  data: 0.3517  max mem: 41794
Epoch: [106]  [300/312]  eta: 0:00:09  lr: 0.003121  min_lr: 0.003121  loss: 3.2002 (3.5777)  weight_decay: 0.0500 (0.0500)  time: 0.6535  data: 0.1668  max mem: 41794
Epoch: [106]  [310/312]  eta: 0:00:01  lr: 0.003121  min_lr: 0.003121  loss: 3.5182 (3.5837)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [106]  [311/312]  eta: 0:00:00  lr: 0.003121  min_lr: 0.003121  loss: 3.5856 (3.5837)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [106] Total time: 0:03:57 (0.7600 s / it)
Averaged stats: lr: 0.003121  min_lr: 0.003121  loss: 3.5856 (3.5643)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.3970 (1.3970)  acc1: 76.9531 (76.9531)  acc5: 92.4479 (92.4479)  time: 8.7988  data: 8.5870  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7172 (1.6269)  acc1: 70.8333 (70.0160)  acc5: 90.6250 (90.1920)  time: 1.1256  data: 0.9542  max mem: 41794
Test: Total time: 0:00:10 (1.1395 s / it)
* Acc@1 69.968 Acc@5 90.120 loss 1.631
Accuracy of the model on the 50000 test images: 70.0%
Max accuracy: 70.74%
Epoch: [107]  [  0/312]  eta: 1:22:36  lr: 0.003121  min_lr: 0.003121  loss: 3.9122 (3.9122)  weight_decay: 0.0500 (0.0500)  time: 15.8850  data: 13.2171  max mem: 41794
Epoch: [107]  [ 10/312]  eta: 0:10:49  lr: 0.003120  min_lr: 0.003120  loss: 3.7174 (3.5621)  weight_decay: 0.0500 (0.0500)  time: 2.1514  data: 1.3522  max mem: 41794
Epoch: [107]  [ 20/312]  eta: 0:07:59  lr: 0.003119  min_lr: 0.003119  loss: 3.6100 (3.5049)  weight_decay: 0.0500 (0.0500)  time: 0.9294  data: 0.1132  max mem: 41794
Epoch: [107]  [ 30/312]  eta: 0:06:11  lr: 0.003119  min_lr: 0.003119  loss: 3.3207 (3.3830)  weight_decay: 0.0500 (0.0500)  time: 0.8600  data: 0.0918  max mem: 41794
Epoch: [107]  [ 40/312]  eta: 0:05:05  lr: 0.003118  min_lr: 0.003118  loss: 3.4317 (3.4107)  weight_decay: 0.0500 (0.0500)  time: 0.5809  data: 0.0682  max mem: 41794
Epoch: [107]  [ 50/312]  eta: 0:04:22  lr: 0.003118  min_lr: 0.003118  loss: 3.6308 (3.4136)  weight_decay: 0.0500 (0.0500)  time: 0.5138  data: 0.0072  max mem: 41794
Epoch: [107]  [ 60/312]  eta: 0:03:52  lr: 0.003117  min_lr: 0.003117  loss: 3.5024 (3.4228)  weight_decay: 0.0500 (0.0500)  time: 0.5035  data: 0.0007  max mem: 41794
Epoch: [107]  [ 70/312]  eta: 0:03:29  lr: 0.003116  min_lr: 0.003116  loss: 3.5411 (3.4337)  weight_decay: 0.0500 (0.0500)  time: 0.5150  data: 0.0039  max mem: 41794
Epoch: [107]  [ 80/312]  eta: 0:03:18  lr: 0.003116  min_lr: 0.003116  loss: 3.5725 (3.4577)  weight_decay: 0.0500 (0.0500)  time: 0.6610  data: 0.0798  max mem: 41794
Epoch: [107]  [ 90/312]  eta: 0:03:04  lr: 0.003115  min_lr: 0.003115  loss: 3.8044 (3.5096)  weight_decay: 0.0500 (0.0500)  time: 0.7058  data: 0.0962  max mem: 41794
Epoch: [107]  [100/312]  eta: 0:02:55  lr: 0.003115  min_lr: 0.003115  loss: 3.9467 (3.5483)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.0990  max mem: 41794
Epoch: [107]  [110/312]  eta: 0:02:43  lr: 0.003114  min_lr: 0.003114  loss: 3.7852 (3.5342)  weight_decay: 0.0500 (0.0500)  time: 0.7280  data: 0.1049  max mem: 41794
Epoch: [107]  [120/312]  eta: 0:02:34  lr: 0.003114  min_lr: 0.003114  loss: 3.7633 (3.5567)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.1073  max mem: 41794
Epoch: [107]  [130/312]  eta: 0:02:26  lr: 0.003113  min_lr: 0.003113  loss: 3.8209 (3.5650)  weight_decay: 0.0500 (0.0500)  time: 0.7752  data: 0.1823  max mem: 41794
Epoch: [107]  [140/312]  eta: 0:02:16  lr: 0.003112  min_lr: 0.003112  loss: 3.7873 (3.5707)  weight_decay: 0.0500 (0.0500)  time: 0.7339  data: 0.1339  max mem: 41794
Epoch: [107]  [150/312]  eta: 0:02:08  lr: 0.003112  min_lr: 0.003112  loss: 3.7381 (3.5784)  weight_decay: 0.0500 (0.0500)  time: 0.7047  data: 0.1578  max mem: 41794
Epoch: [107]  [160/312]  eta: 0:02:01  lr: 0.003111  min_lr: 0.003111  loss: 3.7732 (3.5809)  weight_decay: 0.0500 (0.0500)  time: 0.8196  data: 0.2510  max mem: 41794
Epoch: [107]  [170/312]  eta: 0:01:51  lr: 0.003111  min_lr: 0.003111  loss: 3.6452 (3.5740)  weight_decay: 0.0500 (0.0500)  time: 0.7242  data: 0.1542  max mem: 41794
Epoch: [107]  [180/312]  eta: 0:01:44  lr: 0.003110  min_lr: 0.003110  loss: 3.6452 (3.5831)  weight_decay: 0.0500 (0.0500)  time: 0.7208  data: 0.1192  max mem: 41794
Epoch: [107]  [190/312]  eta: 0:01:35  lr: 0.003109  min_lr: 0.003109  loss: 3.7747 (3.5872)  weight_decay: 0.0500 (0.0500)  time: 0.7446  data: 0.1225  max mem: 41794
Epoch: [107]  [200/312]  eta: 0:01:27  lr: 0.003109  min_lr: 0.003109  loss: 3.7493 (3.5865)  weight_decay: 0.0500 (0.0500)  time: 0.7062  data: 0.1035  max mem: 41794
Epoch: [107]  [210/312]  eta: 0:01:19  lr: 0.003108  min_lr: 0.003108  loss: 3.7711 (3.5934)  weight_decay: 0.0500 (0.0500)  time: 0.7830  data: 0.1439  max mem: 41794
Epoch: [107]  [220/312]  eta: 0:01:11  lr: 0.003108  min_lr: 0.003108  loss: 3.7964 (3.5943)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.1124  max mem: 41794
Epoch: [107]  [230/312]  eta: 0:01:03  lr: 0.003107  min_lr: 0.003107  loss: 3.9217 (3.6078)  weight_decay: 0.0500 (0.0500)  time: 0.6960  data: 0.1272  max mem: 41794
Epoch: [107]  [240/312]  eta: 0:00:55  lr: 0.003106  min_lr: 0.003106  loss: 3.9186 (3.6081)  weight_decay: 0.0500 (0.0500)  time: 0.7937  data: 0.1642  max mem: 41794
Epoch: [107]  [250/312]  eta: 0:00:47  lr: 0.003106  min_lr: 0.003106  loss: 3.8041 (3.6081)  weight_decay: 0.0500 (0.0500)  time: 0.7212  data: 0.1265  max mem: 41794
Epoch: [107]  [260/312]  eta: 0:00:40  lr: 0.003105  min_lr: 0.003105  loss: 3.6193 (3.5913)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.1201  max mem: 41794
Epoch: [107]  [270/312]  eta: 0:00:32  lr: 0.003105  min_lr: 0.003105  loss: 3.1724 (3.5860)  weight_decay: 0.0500 (0.0500)  time: 0.7368  data: 0.1003  max mem: 41794
Epoch: [107]  [280/312]  eta: 0:00:24  lr: 0.003104  min_lr: 0.003104  loss: 3.5689 (3.5875)  weight_decay: 0.0500 (0.0500)  time: 0.7145  data: 0.0962  max mem: 41794
Epoch: [107]  [290/312]  eta: 0:00:16  lr: 0.003103  min_lr: 0.003103  loss: 3.6177 (3.5858)  weight_decay: 0.0500 (0.0500)  time: 0.7879  data: 0.1417  max mem: 41794
Epoch: [107]  [300/312]  eta: 0:00:09  lr: 0.003103  min_lr: 0.003103  loss: 3.5506 (3.5793)  weight_decay: 0.0500 (0.0500)  time: 0.6844  data: 0.0811  max mem: 41794
Epoch: [107]  [310/312]  eta: 0:00:01  lr: 0.003102  min_lr: 0.003102  loss: 3.5818 (3.5767)  weight_decay: 0.0500 (0.0500)  time: 0.4956  data: 0.0065  max mem: 41794
Epoch: [107]  [311/312]  eta: 0:00:00  lr: 0.003102  min_lr: 0.003102  loss: 3.5818 (3.5765)  weight_decay: 0.0500 (0.0500)  time: 0.4945  data: 0.0065  max mem: 41794
Epoch: [107] Total time: 0:03:56 (0.7571 s / it)
Averaged stats: lr: 0.003102  min_lr: 0.003102  loss: 3.5818 (3.5611)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2373 (1.2373)  acc1: 76.6927 (76.6927)  acc5: 93.6198 (93.6198)  time: 8.5168  data: 8.3042  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5963 (1.4959)  acc1: 69.6615 (69.9200)  acc5: 90.1042 (89.9520)  time: 1.0943  data: 0.9228  max mem: 41794
Test: Total time: 0:00:10 (1.1284 s / it)
* Acc@1 70.086 Acc@5 90.098 loss 1.495
Accuracy of the model on the 50000 test images: 70.1%
Max accuracy: 70.74%
Epoch: [108]  [  0/312]  eta: 1:14:43  lr: 0.003102  min_lr: 0.003102  loss: 3.9251 (3.9251)  weight_decay: 0.0500 (0.0500)  time: 14.3688  data: 12.7707  max mem: 41794
Epoch: [108]  [ 10/312]  eta: 0:11:10  lr: 0.003101  min_lr: 0.003101  loss: 3.8083 (3.5154)  weight_decay: 0.0500 (0.0500)  time: 2.2187  data: 1.3058  max mem: 41794
Epoch: [108]  [ 20/312]  eta: 0:07:31  lr: 0.003101  min_lr: 0.003101  loss: 3.6635 (3.5276)  weight_decay: 0.0500 (0.0500)  time: 0.9037  data: 0.1232  max mem: 41794
Epoch: [108]  [ 30/312]  eta: 0:05:45  lr: 0.003100  min_lr: 0.003100  loss: 3.3937 (3.4097)  weight_decay: 0.0500 (0.0500)  time: 0.6778  data: 0.0439  max mem: 41794
Epoch: [108]  [ 40/312]  eta: 0:04:49  lr: 0.003100  min_lr: 0.003100  loss: 3.3937 (3.4325)  weight_decay: 0.0500 (0.0500)  time: 0.5602  data: 0.0201  max mem: 41794
Epoch: [108]  [ 50/312]  eta: 0:04:22  lr: 0.003099  min_lr: 0.003099  loss: 3.8080 (3.5189)  weight_decay: 0.0500 (0.0500)  time: 0.6610  data: 0.1282  max mem: 41794
Epoch: [108]  [ 60/312]  eta: 0:03:52  lr: 0.003098  min_lr: 0.003098  loss: 3.6379 (3.4785)  weight_decay: 0.0500 (0.0500)  time: 0.6295  data: 0.1138  max mem: 41794
Epoch: [108]  [ 70/312]  eta: 0:03:44  lr: 0.003098  min_lr: 0.003098  loss: 3.4173 (3.4656)  weight_decay: 0.0500 (0.0500)  time: 0.7326  data: 0.1749  max mem: 41794
Epoch: [108]  [ 80/312]  eta: 0:03:30  lr: 0.003097  min_lr: 0.003097  loss: 3.5559 (3.4779)  weight_decay: 0.0500 (0.0500)  time: 0.8703  data: 0.2729  max mem: 41794
Epoch: [108]  [ 90/312]  eta: 0:03:13  lr: 0.003097  min_lr: 0.003097  loss: 3.4771 (3.4593)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1366  max mem: 41794
Epoch: [108]  [100/312]  eta: 0:03:07  lr: 0.003096  min_lr: 0.003096  loss: 3.4771 (3.4549)  weight_decay: 0.0500 (0.0500)  time: 0.7743  data: 0.1630  max mem: 41794
Epoch: [108]  [110/312]  eta: 0:02:51  lr: 0.003095  min_lr: 0.003095  loss: 3.8587 (3.5008)  weight_decay: 0.0500 (0.0500)  time: 0.7364  data: 0.1313  max mem: 41794
Epoch: [108]  [120/312]  eta: 0:02:43  lr: 0.003095  min_lr: 0.003095  loss: 3.8849 (3.5076)  weight_decay: 0.0500 (0.0500)  time: 0.6814  data: 0.0668  max mem: 41794
Epoch: [108]  [130/312]  eta: 0:02:32  lr: 0.003094  min_lr: 0.003094  loss: 3.7636 (3.5136)  weight_decay: 0.0500 (0.0500)  time: 0.7915  data: 0.0781  max mem: 41794
Epoch: [108]  [140/312]  eta: 0:02:21  lr: 0.003094  min_lr: 0.003094  loss: 3.8469 (3.5210)  weight_decay: 0.0500 (0.0500)  time: 0.6598  data: 0.0738  max mem: 41794
Epoch: [108]  [150/312]  eta: 0:02:13  lr: 0.003093  min_lr: 0.003093  loss: 3.7418 (3.5268)  weight_decay: 0.0500 (0.0500)  time: 0.7383  data: 0.1519  max mem: 41794
Epoch: [108]  [160/312]  eta: 0:02:04  lr: 0.003092  min_lr: 0.003092  loss: 3.7418 (3.5377)  weight_decay: 0.0500 (0.0500)  time: 0.7910  data: 0.1122  max mem: 41794
Epoch: [108]  [170/312]  eta: 0:01:54  lr: 0.003092  min_lr: 0.003092  loss: 3.8271 (3.5557)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.0875  max mem: 41794
Epoch: [108]  [180/312]  eta: 0:01:46  lr: 0.003091  min_lr: 0.003091  loss: 3.8427 (3.5659)  weight_decay: 0.0500 (0.0500)  time: 0.7154  data: 0.1683  max mem: 41794
Epoch: [108]  [190/312]  eta: 0:01:36  lr: 0.003091  min_lr: 0.003091  loss: 3.6776 (3.5604)  weight_decay: 0.0500 (0.0500)  time: 0.6500  data: 0.1023  max mem: 41794
Epoch: [108]  [200/312]  eta: 0:01:28  lr: 0.003090  min_lr: 0.003090  loss: 3.6572 (3.5672)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.1214  max mem: 41794
Epoch: [108]  [210/312]  eta: 0:01:21  lr: 0.003089  min_lr: 0.003089  loss: 3.6572 (3.5554)  weight_decay: 0.0500 (0.0500)  time: 0.8347  data: 0.1852  max mem: 41794
Epoch: [108]  [220/312]  eta: 0:01:12  lr: 0.003089  min_lr: 0.003089  loss: 3.3848 (3.5456)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.1117  max mem: 41794
Epoch: [108]  [230/312]  eta: 0:01:04  lr: 0.003088  min_lr: 0.003088  loss: 3.3872 (3.5437)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.1614  max mem: 41794
Epoch: [108]  [240/312]  eta: 0:00:56  lr: 0.003088  min_lr: 0.003088  loss: 3.6768 (3.5492)  weight_decay: 0.0500 (0.0500)  time: 0.7531  data: 0.2277  max mem: 41794
Epoch: [108]  [250/312]  eta: 0:00:48  lr: 0.003087  min_lr: 0.003087  loss: 3.6768 (3.5405)  weight_decay: 0.0500 (0.0500)  time: 0.6676  data: 0.1515  max mem: 41794
Epoch: [108]  [260/312]  eta: 0:00:40  lr: 0.003086  min_lr: 0.003086  loss: 2.9717 (3.5227)  weight_decay: 0.0500 (0.0500)  time: 0.7364  data: 0.1938  max mem: 41794
Epoch: [108]  [270/312]  eta: 0:00:32  lr: 0.003086  min_lr: 0.003086  loss: 3.2996 (3.5286)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.1565  max mem: 41794
Epoch: [108]  [280/312]  eta: 0:00:24  lr: 0.003085  min_lr: 0.003085  loss: 3.7274 (3.5324)  weight_decay: 0.0500 (0.0500)  time: 0.6598  data: 0.1537  max mem: 41794
Epoch: [108]  [290/312]  eta: 0:00:16  lr: 0.003085  min_lr: 0.003085  loss: 3.5402 (3.5280)  weight_decay: 0.0500 (0.0500)  time: 0.8060  data: 0.3023  max mem: 41794
Epoch: [108]  [300/312]  eta: 0:00:09  lr: 0.003084  min_lr: 0.003084  loss: 3.5049 (3.5247)  weight_decay: 0.0500 (0.0500)  time: 0.6672  data: 0.1855  max mem: 41794
Epoch: [108]  [310/312]  eta: 0:00:01  lr: 0.003083  min_lr: 0.003083  loss: 3.7808 (3.5311)  weight_decay: 0.0500 (0.0500)  time: 0.5012  data: 0.0366  max mem: 41794
Epoch: [108]  [311/312]  eta: 0:00:00  lr: 0.003083  min_lr: 0.003083  loss: 3.6344 (3.5312)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [108] Total time: 0:03:56 (0.7593 s / it)
Averaged stats: lr: 0.003083  min_lr: 0.003083  loss: 3.6344 (3.5503)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3946 (1.3946)  acc1: 79.0365 (79.0365)  acc5: 94.4010 (94.4010)  time: 8.5026  data: 8.3001  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6679 (1.6213)  acc1: 67.8385 (69.7920)  acc5: 91.0156 (90.0640)  time: 1.0934  data: 0.9223  max mem: 41794
Test: Total time: 0:00:09 (1.1046 s / it)
* Acc@1 70.234 Acc@5 90.072 loss 1.624
Accuracy of the model on the 50000 test images: 70.2%
Max accuracy: 70.74%
Epoch: [109]  [  0/312]  eta: 1:19:17  lr: 0.003083  min_lr: 0.003083  loss: 2.7978 (2.7978)  weight_decay: 0.0500 (0.0500)  time: 15.2496  data: 13.4384  max mem: 41794
Epoch: [109]  [ 10/312]  eta: 0:10:53  lr: 0.003083  min_lr: 0.003083  loss: 3.6309 (3.4294)  weight_decay: 0.0500 (0.0500)  time: 2.1637  data: 1.3077  max mem: 41794
Epoch: [109]  [ 20/312]  eta: 0:07:09  lr: 0.003082  min_lr: 0.003082  loss: 3.5343 (3.4530)  weight_decay: 0.0500 (0.0500)  time: 0.7807  data: 0.1424  max mem: 41794
Epoch: [109]  [ 30/312]  eta: 0:05:27  lr: 0.003081  min_lr: 0.003081  loss: 3.6194 (3.5047)  weight_decay: 0.0500 (0.0500)  time: 0.6096  data: 0.0994  max mem: 41794
Epoch: [109]  [ 40/312]  eta: 0:04:43  lr: 0.003081  min_lr: 0.003081  loss: 3.6791 (3.4980)  weight_decay: 0.0500 (0.0500)  time: 0.5964  data: 0.0791  max mem: 41794
Epoch: [109]  [ 50/312]  eta: 0:04:25  lr: 0.003080  min_lr: 0.003080  loss: 3.6791 (3.4922)  weight_decay: 0.0500 (0.0500)  time: 0.7798  data: 0.2279  max mem: 41794
Epoch: [109]  [ 60/312]  eta: 0:03:53  lr: 0.003080  min_lr: 0.003080  loss: 3.6255 (3.5187)  weight_decay: 0.0500 (0.0500)  time: 0.6850  data: 0.1535  max mem: 41794
Epoch: [109]  [ 70/312]  eta: 0:03:41  lr: 0.003079  min_lr: 0.003079  loss: 3.6158 (3.5328)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1398  max mem: 41794
Epoch: [109]  [ 80/312]  eta: 0:03:31  lr: 0.003078  min_lr: 0.003078  loss: 3.3671 (3.5100)  weight_decay: 0.0500 (0.0500)  time: 0.8705  data: 0.2855  max mem: 41794
Epoch: [109]  [ 90/312]  eta: 0:03:12  lr: 0.003078  min_lr: 0.003078  loss: 3.5007 (3.5380)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1463  max mem: 41794
Epoch: [109]  [100/312]  eta: 0:03:05  lr: 0.003077  min_lr: 0.003077  loss: 3.5747 (3.5116)  weight_decay: 0.0500 (0.0500)  time: 0.7145  data: 0.1505  max mem: 41794
Epoch: [109]  [110/312]  eta: 0:02:49  lr: 0.003077  min_lr: 0.003077  loss: 3.5153 (3.5150)  weight_decay: 0.0500 (0.0500)  time: 0.7088  data: 0.1505  max mem: 41794
Epoch: [109]  [120/312]  eta: 0:02:41  lr: 0.003076  min_lr: 0.003076  loss: 3.6038 (3.5127)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.1240  max mem: 41794
Epoch: [109]  [130/312]  eta: 0:02:33  lr: 0.003075  min_lr: 0.003075  loss: 3.7600 (3.5340)  weight_decay: 0.0500 (0.0500)  time: 0.8652  data: 0.2360  max mem: 41794
Epoch: [109]  [140/312]  eta: 0:02:20  lr: 0.003075  min_lr: 0.003075  loss: 3.6820 (3.5203)  weight_decay: 0.0500 (0.0500)  time: 0.6778  data: 0.1126  max mem: 41794
Epoch: [109]  [150/312]  eta: 0:02:13  lr: 0.003074  min_lr: 0.003074  loss: 3.6054 (3.5319)  weight_decay: 0.0500 (0.0500)  time: 0.7089  data: 0.1398  max mem: 41794
Epoch: [109]  [160/312]  eta: 0:02:04  lr: 0.003074  min_lr: 0.003074  loss: 3.8308 (3.5303)  weight_decay: 0.0500 (0.0500)  time: 0.8406  data: 0.2214  max mem: 41794
Epoch: [109]  [170/312]  eta: 0:01:54  lr: 0.003073  min_lr: 0.003073  loss: 3.4007 (3.5183)  weight_decay: 0.0500 (0.0500)  time: 0.6731  data: 0.0939  max mem: 41794
Epoch: [109]  [180/312]  eta: 0:01:47  lr: 0.003072  min_lr: 0.003072  loss: 3.7030 (3.5354)  weight_decay: 0.0500 (0.0500)  time: 0.7667  data: 0.1265  max mem: 41794
Epoch: [109]  [190/312]  eta: 0:01:37  lr: 0.003072  min_lr: 0.003072  loss: 3.6605 (3.5114)  weight_decay: 0.0500 (0.0500)  time: 0.7285  data: 0.1166  max mem: 41794
Epoch: [109]  [200/312]  eta: 0:01:30  lr: 0.003071  min_lr: 0.003071  loss: 3.2368 (3.5109)  weight_decay: 0.0500 (0.0500)  time: 0.7182  data: 0.0652  max mem: 41794
Epoch: [109]  [210/312]  eta: 0:01:22  lr: 0.003071  min_lr: 0.003071  loss: 3.4260 (3.5081)  weight_decay: 0.0500 (0.0500)  time: 0.8963  data: 0.1113  max mem: 41794
Epoch: [109]  [220/312]  eta: 0:01:13  lr: 0.003070  min_lr: 0.003070  loss: 3.2087 (3.4894)  weight_decay: 0.0500 (0.0500)  time: 0.6753  data: 0.0485  max mem: 41794
Epoch: [109]  [230/312]  eta: 0:01:05  lr: 0.003069  min_lr: 0.003069  loss: 3.5345 (3.5017)  weight_decay: 0.0500 (0.0500)  time: 0.6829  data: 0.0423  max mem: 41794
Epoch: [109]  [240/312]  eta: 0:00:56  lr: 0.003069  min_lr: 0.003069  loss: 3.8676 (3.5011)  weight_decay: 0.0500 (0.0500)  time: 0.7412  data: 0.0440  max mem: 41794
Epoch: [109]  [250/312]  eta: 0:00:48  lr: 0.003068  min_lr: 0.003068  loss: 3.7048 (3.5001)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.0369  max mem: 41794
Epoch: [109]  [260/312]  eta: 0:00:41  lr: 0.003067  min_lr: 0.003067  loss: 3.5511 (3.5019)  weight_decay: 0.0500 (0.0500)  time: 0.8010  data: 0.0824  max mem: 41794
Epoch: [109]  [270/312]  eta: 0:00:32  lr: 0.003067  min_lr: 0.003067  loss: 3.7373 (3.5084)  weight_decay: 0.0500 (0.0500)  time: 0.6990  data: 0.0490  max mem: 41794
Epoch: [109]  [280/312]  eta: 0:00:25  lr: 0.003066  min_lr: 0.003066  loss: 3.6867 (3.5150)  weight_decay: 0.0500 (0.0500)  time: 0.6626  data: 0.0293  max mem: 41794
Epoch: [109]  [290/312]  eta: 0:00:17  lr: 0.003066  min_lr: 0.003066  loss: 3.6171 (3.5126)  weight_decay: 0.0500 (0.0500)  time: 0.7755  data: 0.0434  max mem: 41794
Epoch: [109]  [300/312]  eta: 0:00:09  lr: 0.003065  min_lr: 0.003065  loss: 3.6171 (3.5216)  weight_decay: 0.0500 (0.0500)  time: 0.6022  data: 0.0158  max mem: 41794
Epoch: [109]  [310/312]  eta: 0:00:01  lr: 0.003064  min_lr: 0.003064  loss: 3.7900 (3.5254)  weight_decay: 0.0500 (0.0500)  time: 0.4643  data: 0.0001  max mem: 41794
Epoch: [109]  [311/312]  eta: 0:00:00  lr: 0.003064  min_lr: 0.003064  loss: 3.7274 (3.5261)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0001  max mem: 41794
Epoch: [109] Total time: 0:03:58 (0.7657 s / it)
Averaged stats: lr: 0.003064  min_lr: 0.003064  loss: 3.7274 (3.5457)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.4008 (1.4008)  acc1: 77.0833 (77.0833)  acc5: 93.4896 (93.4896)  time: 8.3564  data: 8.1434  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6395 (1.5744)  acc1: 69.4010 (70.0480)  acc5: 90.8854 (90.9120)  time: 1.0767  data: 0.9049  max mem: 41794
Test: Total time: 0:00:09 (1.0878 s / it)
* Acc@1 70.484 Acc@5 90.480 loss 1.574
Accuracy of the model on the 50000 test images: 70.5%
Max accuracy: 70.74%
Epoch: [110]  [  0/312]  eta: 1:22:29  lr: 0.003064  min_lr: 0.003064  loss: 3.8213 (3.8213)  weight_decay: 0.0500 (0.0500)  time: 15.8653  data: 12.6369  max mem: 41794
Epoch: [110]  [ 10/312]  eta: 0:12:30  lr: 0.003064  min_lr: 0.003064  loss: 3.8136 (3.6975)  weight_decay: 0.0500 (0.0500)  time: 2.4849  data: 1.3981  max mem: 41794
Epoch: [110]  [ 20/312]  eta: 0:07:38  lr: 0.003063  min_lr: 0.003063  loss: 3.7378 (3.4582)  weight_decay: 0.0500 (0.0500)  time: 0.8559  data: 0.1580  max mem: 41794
Epoch: [110]  [ 30/312]  eta: 0:05:47  lr: 0.003062  min_lr: 0.003062  loss: 3.3059 (3.4415)  weight_decay: 0.0500 (0.0500)  time: 0.5448  data: 0.0280  max mem: 41794
Epoch: [110]  [ 40/312]  eta: 0:04:48  lr: 0.003062  min_lr: 0.003062  loss: 3.7846 (3.5119)  weight_decay: 0.0500 (0.0500)  time: 0.5225  data: 0.0111  max mem: 41794
Epoch: [110]  [ 50/312]  eta: 0:04:09  lr: 0.003061  min_lr: 0.003061  loss: 3.9196 (3.5698)  weight_decay: 0.0500 (0.0500)  time: 0.5142  data: 0.0123  max mem: 41794
Epoch: [110]  [ 60/312]  eta: 0:03:41  lr: 0.003061  min_lr: 0.003061  loss: 3.8644 (3.5881)  weight_decay: 0.0500 (0.0500)  time: 0.5034  data: 0.0087  max mem: 41794
Epoch: [110]  [ 70/312]  eta: 0:03:24  lr: 0.003060  min_lr: 0.003060  loss: 3.7247 (3.5848)  weight_decay: 0.0500 (0.0500)  time: 0.5783  data: 0.0007  max mem: 41794
Epoch: [110]  [ 80/312]  eta: 0:03:15  lr: 0.003059  min_lr: 0.003059  loss: 3.5548 (3.5967)  weight_decay: 0.0500 (0.0500)  time: 0.7347  data: 0.0057  max mem: 41794
Epoch: [110]  [ 90/312]  eta: 0:03:01  lr: 0.003059  min_lr: 0.003059  loss: 3.6873 (3.6106)  weight_decay: 0.0500 (0.0500)  time: 0.7124  data: 0.0276  max mem: 41794
Epoch: [110]  [100/312]  eta: 0:02:52  lr: 0.003058  min_lr: 0.003058  loss: 3.5901 (3.5827)  weight_decay: 0.0500 (0.0500)  time: 0.6927  data: 0.0399  max mem: 41794
Epoch: [110]  [110/312]  eta: 0:02:43  lr: 0.003058  min_lr: 0.003058  loss: 3.3676 (3.5644)  weight_decay: 0.0500 (0.0500)  time: 0.7641  data: 0.0234  max mem: 41794
Epoch: [110]  [120/312]  eta: 0:02:33  lr: 0.003057  min_lr: 0.003057  loss: 3.4618 (3.5618)  weight_decay: 0.0500 (0.0500)  time: 0.7220  data: 0.0280  max mem: 41794
Epoch: [110]  [130/312]  eta: 0:02:25  lr: 0.003056  min_lr: 0.003056  loss: 3.8369 (3.5706)  weight_decay: 0.0500 (0.0500)  time: 0.7606  data: 0.0390  max mem: 41794
Epoch: [110]  [140/312]  eta: 0:02:17  lr: 0.003056  min_lr: 0.003056  loss: 3.6307 (3.5587)  weight_decay: 0.0500 (0.0500)  time: 0.8098  data: 0.0206  max mem: 41794
Epoch: [110]  [150/312]  eta: 0:02:07  lr: 0.003055  min_lr: 0.003055  loss: 3.3883 (3.5476)  weight_decay: 0.0500 (0.0500)  time: 0.7163  data: 0.0472  max mem: 41794
Epoch: [110]  [160/312]  eta: 0:02:00  lr: 0.003055  min_lr: 0.003055  loss: 3.5690 (3.5591)  weight_decay: 0.0500 (0.0500)  time: 0.7273  data: 0.0803  max mem: 41794
Epoch: [110]  [170/312]  eta: 0:01:51  lr: 0.003054  min_lr: 0.003054  loss: 3.8642 (3.5714)  weight_decay: 0.0500 (0.0500)  time: 0.7432  data: 0.0895  max mem: 41794
Epoch: [110]  [180/312]  eta: 0:01:43  lr: 0.003053  min_lr: 0.003053  loss: 3.7513 (3.5699)  weight_decay: 0.0500 (0.0500)  time: 0.7150  data: 0.1308  max mem: 41794
Epoch: [110]  [190/312]  eta: 0:01:35  lr: 0.003053  min_lr: 0.003053  loss: 3.4621 (3.5569)  weight_decay: 0.0500 (0.0500)  time: 0.7709  data: 0.1233  max mem: 41794
Epoch: [110]  [200/312]  eta: 0:01:26  lr: 0.003052  min_lr: 0.003052  loss: 3.5267 (3.5634)  weight_decay: 0.0500 (0.0500)  time: 0.7139  data: 0.0860  max mem: 41794
Epoch: [110]  [210/312]  eta: 0:01:19  lr: 0.003051  min_lr: 0.003051  loss: 3.5326 (3.5596)  weight_decay: 0.0500 (0.0500)  time: 0.7549  data: 0.0933  max mem: 41794
Epoch: [110]  [220/312]  eta: 0:01:11  lr: 0.003051  min_lr: 0.003051  loss: 3.5647 (3.5616)  weight_decay: 0.0500 (0.0500)  time: 0.8136  data: 0.0811  max mem: 41794
Epoch: [110]  [230/312]  eta: 0:01:03  lr: 0.003050  min_lr: 0.003050  loss: 3.6912 (3.5636)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.0556  max mem: 41794
Epoch: [110]  [240/312]  eta: 0:00:55  lr: 0.003050  min_lr: 0.003050  loss: 3.6649 (3.5578)  weight_decay: 0.0500 (0.0500)  time: 0.7528  data: 0.0664  max mem: 41794
Epoch: [110]  [250/312]  eta: 0:00:47  lr: 0.003049  min_lr: 0.003049  loss: 3.6275 (3.5526)  weight_decay: 0.0500 (0.0500)  time: 0.7180  data: 0.0397  max mem: 41794
Epoch: [110]  [260/312]  eta: 0:00:39  lr: 0.003048  min_lr: 0.003048  loss: 3.6340 (3.5520)  weight_decay: 0.0500 (0.0500)  time: 0.6404  data: 0.0405  max mem: 41794
Epoch: [110]  [270/312]  eta: 0:00:32  lr: 0.003048  min_lr: 0.003048  loss: 3.6534 (3.5561)  weight_decay: 0.0500 (0.0500)  time: 0.7632  data: 0.0551  max mem: 41794
Epoch: [110]  [280/312]  eta: 0:00:24  lr: 0.003047  min_lr: 0.003047  loss: 3.6854 (3.5509)  weight_decay: 0.0500 (0.0500)  time: 0.7515  data: 0.0573  max mem: 41794
Epoch: [110]  [290/312]  eta: 0:00:16  lr: 0.003047  min_lr: 0.003047  loss: 3.5091 (3.5493)  weight_decay: 0.0500 (0.0500)  time: 0.7458  data: 0.0986  max mem: 41794
Epoch: [110]  [300/312]  eta: 0:00:09  lr: 0.003046  min_lr: 0.003046  loss: 3.4655 (3.5387)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.0563  max mem: 41794
Epoch: [110]  [310/312]  eta: 0:00:01  lr: 0.003045  min_lr: 0.003045  loss: 3.5487 (3.5361)  weight_decay: 0.0500 (0.0500)  time: 0.5229  data: 0.0002  max mem: 41794
Epoch: [110]  [311/312]  eta: 0:00:00  lr: 0.003045  min_lr: 0.003045  loss: 3.5845 (3.5370)  weight_decay: 0.0500 (0.0500)  time: 0.5214  data: 0.0002  max mem: 41794
Epoch: [110] Total time: 0:03:56 (0.7565 s / it)
Averaged stats: lr: 0.003045  min_lr: 0.003045  loss: 3.5845 (3.5535)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.1174 (1.1174)  acc1: 78.3854 (78.3854)  acc5: 94.2708 (94.2708)  time: 8.6122  data: 8.4026  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5313 (1.4559)  acc1: 69.9219 (70.6720)  acc5: 89.9740 (90.2720)  time: 1.1050  data: 0.9337  max mem: 41794
Test: Total time: 0:00:10 (1.1142 s / it)
* Acc@1 70.622 Acc@5 90.320 loss 1.458
Accuracy of the model on the 50000 test images: 70.6%
Max accuracy: 70.74%
Epoch: [111]  [  0/312]  eta: 1:24:17  lr: 0.003045  min_lr: 0.003045  loss: 3.9526 (3.9526)  weight_decay: 0.0500 (0.0500)  time: 16.2115  data: 15.7481  max mem: 41794
Epoch: [111]  [ 10/312]  eta: 0:11:04  lr: 0.003045  min_lr: 0.003045  loss: 3.1875 (3.3195)  weight_decay: 0.0500 (0.0500)  time: 2.1987  data: 1.4323  max mem: 41794
Epoch: [111]  [ 20/312]  eta: 0:07:27  lr: 0.003044  min_lr: 0.003044  loss: 3.1953 (3.3852)  weight_decay: 0.0500 (0.0500)  time: 0.7970  data: 0.0716  max mem: 41794
Epoch: [111]  [ 30/312]  eta: 0:05:40  lr: 0.003043  min_lr: 0.003043  loss: 3.4462 (3.3891)  weight_decay: 0.0500 (0.0500)  time: 0.6624  data: 0.0716  max mem: 41794
Epoch: [111]  [ 40/312]  eta: 0:04:46  lr: 0.003043  min_lr: 0.003043  loss: 3.6157 (3.4782)  weight_decay: 0.0500 (0.0500)  time: 0.5532  data: 0.0100  max mem: 41794
Epoch: [111]  [ 50/312]  eta: 0:04:23  lr: 0.003042  min_lr: 0.003042  loss: 3.6707 (3.4432)  weight_decay: 0.0500 (0.0500)  time: 0.6939  data: 0.0736  max mem: 41794
Epoch: [111]  [ 60/312]  eta: 0:03:51  lr: 0.003042  min_lr: 0.003042  loss: 3.7199 (3.5263)  weight_decay: 0.0500 (0.0500)  time: 0.6454  data: 0.0656  max mem: 41794
Epoch: [111]  [ 70/312]  eta: 0:03:44  lr: 0.003041  min_lr: 0.003041  loss: 3.8579 (3.5691)  weight_decay: 0.0500 (0.0500)  time: 0.7299  data: 0.1106  max mem: 41794
Epoch: [111]  [ 80/312]  eta: 0:03:34  lr: 0.003040  min_lr: 0.003040  loss: 3.9108 (3.6167)  weight_decay: 0.0500 (0.0500)  time: 0.9346  data: 0.1196  max mem: 41794
Epoch: [111]  [ 90/312]  eta: 0:03:17  lr: 0.003040  min_lr: 0.003040  loss: 3.8332 (3.6074)  weight_decay: 0.0500 (0.0500)  time: 0.7600  data: 0.0767  max mem: 41794
Epoch: [111]  [100/312]  eta: 0:03:07  lr: 0.003039  min_lr: 0.003039  loss: 3.7107 (3.6106)  weight_decay: 0.0500 (0.0500)  time: 0.7326  data: 0.0978  max mem: 41794
Epoch: [111]  [110/312]  eta: 0:02:52  lr: 0.003038  min_lr: 0.003038  loss: 3.6000 (3.5871)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0356  max mem: 41794
Epoch: [111]  [120/312]  eta: 0:02:42  lr: 0.003038  min_lr: 0.003038  loss: 3.5061 (3.5942)  weight_decay: 0.0500 (0.0500)  time: 0.6526  data: 0.0356  max mem: 41794
Epoch: [111]  [130/312]  eta: 0:02:33  lr: 0.003037  min_lr: 0.003037  loss: 3.7779 (3.6028)  weight_decay: 0.0500 (0.0500)  time: 0.8121  data: 0.1036  max mem: 41794
Epoch: [111]  [140/312]  eta: 0:02:21  lr: 0.003037  min_lr: 0.003037  loss: 3.8729 (3.6093)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0721  max mem: 41794
Epoch: [111]  [150/312]  eta: 0:02:13  lr: 0.003036  min_lr: 0.003036  loss: 3.8035 (3.6146)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1013  max mem: 41794
Epoch: [111]  [160/312]  eta: 0:02:05  lr: 0.003035  min_lr: 0.003035  loss: 3.5825 (3.5980)  weight_decay: 0.0500 (0.0500)  time: 0.8509  data: 0.1944  max mem: 41794
Epoch: [111]  [170/312]  eta: 0:01:54  lr: 0.003035  min_lr: 0.003035  loss: 3.3169 (3.5841)  weight_decay: 0.0500 (0.0500)  time: 0.6742  data: 0.0942  max mem: 41794
Epoch: [111]  [180/312]  eta: 0:01:46  lr: 0.003034  min_lr: 0.003034  loss: 3.7243 (3.5896)  weight_decay: 0.0500 (0.0500)  time: 0.6888  data: 0.0937  max mem: 41794
Epoch: [111]  [190/312]  eta: 0:01:36  lr: 0.003034  min_lr: 0.003034  loss: 3.7222 (3.5821)  weight_decay: 0.0500 (0.0500)  time: 0.6862  data: 0.0931  max mem: 41794
Epoch: [111]  [200/312]  eta: 0:01:29  lr: 0.003033  min_lr: 0.003033  loss: 3.5163 (3.5841)  weight_decay: 0.0500 (0.0500)  time: 0.6779  data: 0.0761  max mem: 41794
Epoch: [111]  [210/312]  eta: 0:01:21  lr: 0.003032  min_lr: 0.003032  loss: 3.6154 (3.5851)  weight_decay: 0.0500 (0.0500)  time: 0.8598  data: 0.1472  max mem: 41794
Epoch: [111]  [220/312]  eta: 0:01:12  lr: 0.003032  min_lr: 0.003032  loss: 3.6318 (3.5835)  weight_decay: 0.0500 (0.0500)  time: 0.6714  data: 0.0717  max mem: 41794
Epoch: [111]  [230/312]  eta: 0:01:04  lr: 0.003031  min_lr: 0.003031  loss: 3.7075 (3.5894)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.0780  max mem: 41794
Epoch: [111]  [240/312]  eta: 0:00:56  lr: 0.003030  min_lr: 0.003030  loss: 3.7826 (3.5949)  weight_decay: 0.0500 (0.0500)  time: 0.8481  data: 0.1558  max mem: 41794
Epoch: [111]  [250/312]  eta: 0:00:48  lr: 0.003030  min_lr: 0.003030  loss: 3.6867 (3.5889)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0787  max mem: 41794
Epoch: [111]  [260/312]  eta: 0:00:40  lr: 0.003029  min_lr: 0.003029  loss: 3.3520 (3.5792)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.0851  max mem: 41794
Epoch: [111]  [270/312]  eta: 0:00:32  lr: 0.003029  min_lr: 0.003029  loss: 3.5373 (3.5841)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.0849  max mem: 41794
Epoch: [111]  [280/312]  eta: 0:00:24  lr: 0.003028  min_lr: 0.003028  loss: 3.4241 (3.5774)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.1097  max mem: 41794
Epoch: [111]  [290/312]  eta: 0:00:17  lr: 0.003027  min_lr: 0.003027  loss: 3.5317 (3.5807)  weight_decay: 0.0500 (0.0500)  time: 0.8636  data: 0.2284  max mem: 41794
Epoch: [111]  [300/312]  eta: 0:00:09  lr: 0.003027  min_lr: 0.003027  loss: 3.8046 (3.5851)  weight_decay: 0.0500 (0.0500)  time: 0.6608  data: 0.1191  max mem: 41794
Epoch: [111]  [310/312]  eta: 0:00:01  lr: 0.003026  min_lr: 0.003026  loss: 3.4797 (3.5792)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [111]  [311/312]  eta: 0:00:00  lr: 0.003026  min_lr: 0.003026  loss: 3.4797 (3.5803)  weight_decay: 0.0500 (0.0500)  time: 0.4656  data: 0.0001  max mem: 41794
Epoch: [111] Total time: 0:03:57 (0.7628 s / it)
Averaged stats: lr: 0.003026  min_lr: 0.003026  loss: 3.4797 (3.5459)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.1745 (1.1745)  acc1: 78.1250 (78.1250)  acc5: 94.2708 (94.2708)  time: 8.1968  data: 7.9877  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5394 (1.4527)  acc1: 70.0521 (70.8960)  acc5: 90.2344 (90.7040)  time: 1.0661  data: 0.8938  max mem: 41794
Test: Total time: 0:00:09 (1.1046 s / it)
* Acc@1 71.180 Acc@5 90.768 loss 1.445
Accuracy of the model on the 50000 test images: 71.2%
Max accuracy: 71.18%
Epoch: [112]  [  0/312]  eta: 1:21:07  lr: 0.003026  min_lr: 0.003026  loss: 3.7435 (3.7435)  weight_decay: 0.0500 (0.0500)  time: 15.5999  data: 15.1383  max mem: 41794
Epoch: [112]  [ 10/312]  eta: 0:11:04  lr: 0.003025  min_lr: 0.003025  loss: 3.8722 (3.8667)  weight_decay: 0.0500 (0.0500)  time: 2.1987  data: 1.3807  max mem: 41794
Epoch: [112]  [ 20/312]  eta: 0:07:12  lr: 0.003025  min_lr: 0.003025  loss: 3.8006 (3.8005)  weight_decay: 0.0500 (0.0500)  time: 0.7747  data: 0.0343  max mem: 41794
Epoch: [112]  [ 30/312]  eta: 0:05:29  lr: 0.003024  min_lr: 0.003024  loss: 3.7633 (3.7370)  weight_decay: 0.0500 (0.0500)  time: 0.5997  data: 0.0321  max mem: 41794
Epoch: [112]  [ 40/312]  eta: 0:04:39  lr: 0.003024  min_lr: 0.003024  loss: 3.2850 (3.5966)  weight_decay: 0.0500 (0.0500)  time: 0.5530  data: 0.0506  max mem: 41794
Epoch: [112]  [ 50/312]  eta: 0:04:15  lr: 0.003023  min_lr: 0.003023  loss: 3.1875 (3.5666)  weight_decay: 0.0500 (0.0500)  time: 0.6735  data: 0.1816  max mem: 41794
Epoch: [112]  [ 60/312]  eta: 0:03:45  lr: 0.003022  min_lr: 0.003022  loss: 3.6299 (3.5750)  weight_decay: 0.0500 (0.0500)  time: 0.6179  data: 0.1316  max mem: 41794
Epoch: [112]  [ 70/312]  eta: 0:03:36  lr: 0.003022  min_lr: 0.003022  loss: 3.5789 (3.5368)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.2045  max mem: 41794
Epoch: [112]  [ 80/312]  eta: 0:03:29  lr: 0.003021  min_lr: 0.003021  loss: 3.3643 (3.5255)  weight_decay: 0.0500 (0.0500)  time: 0.9266  data: 0.4277  max mem: 41794
Epoch: [112]  [ 90/312]  eta: 0:03:10  lr: 0.003020  min_lr: 0.003020  loss: 3.5417 (3.5180)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.2238  max mem: 41794
Epoch: [112]  [100/312]  eta: 0:03:02  lr: 0.003020  min_lr: 0.003020  loss: 3.6943 (3.5100)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1868  max mem: 41794
Epoch: [112]  [110/312]  eta: 0:02:46  lr: 0.003019  min_lr: 0.003019  loss: 3.6931 (3.5137)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.1868  max mem: 41794
Epoch: [112]  [120/312]  eta: 0:02:39  lr: 0.003019  min_lr: 0.003019  loss: 3.4509 (3.5020)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.1865  max mem: 41794
Epoch: [112]  [130/312]  eta: 0:02:31  lr: 0.003018  min_lr: 0.003018  loss: 3.7112 (3.5164)  weight_decay: 0.0500 (0.0500)  time: 0.8556  data: 0.3571  max mem: 41794
Epoch: [112]  [140/312]  eta: 0:02:18  lr: 0.003017  min_lr: 0.003017  loss: 3.7087 (3.4981)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1714  max mem: 41794
Epoch: [112]  [150/312]  eta: 0:02:11  lr: 0.003017  min_lr: 0.003017  loss: 3.5114 (3.5164)  weight_decay: 0.0500 (0.0500)  time: 0.6787  data: 0.1815  max mem: 41794
Epoch: [112]  [160/312]  eta: 0:02:03  lr: 0.003016  min_lr: 0.003016  loss: 3.5114 (3.5064)  weight_decay: 0.0500 (0.0500)  time: 0.8758  data: 0.3819  max mem: 41794
Epoch: [112]  [170/312]  eta: 0:01:53  lr: 0.003016  min_lr: 0.003016  loss: 3.6987 (3.5168)  weight_decay: 0.0500 (0.0500)  time: 0.7283  data: 0.2369  max mem: 41794
Epoch: [112]  [180/312]  eta: 0:01:45  lr: 0.003015  min_lr: 0.003015  loss: 3.7427 (3.5349)  weight_decay: 0.0500 (0.0500)  time: 0.6939  data: 0.1943  max mem: 41794
Epoch: [112]  [190/312]  eta: 0:01:35  lr: 0.003014  min_lr: 0.003014  loss: 3.8865 (3.5491)  weight_decay: 0.0500 (0.0500)  time: 0.6562  data: 0.1586  max mem: 41794
Epoch: [112]  [200/312]  eta: 0:01:28  lr: 0.003014  min_lr: 0.003014  loss: 3.7821 (3.5516)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.1929  max mem: 41794
Epoch: [112]  [210/312]  eta: 0:01:20  lr: 0.003013  min_lr: 0.003013  loss: 3.7569 (3.5554)  weight_decay: 0.0500 (0.0500)  time: 0.8656  data: 0.3723  max mem: 41794
Epoch: [112]  [220/312]  eta: 0:01:11  lr: 0.003012  min_lr: 0.003012  loss: 3.7569 (3.5609)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.1800  max mem: 41794
Epoch: [112]  [230/312]  eta: 0:01:04  lr: 0.003012  min_lr: 0.003012  loss: 3.8003 (3.5664)  weight_decay: 0.0500 (0.0500)  time: 0.7080  data: 0.2077  max mem: 41794
Epoch: [112]  [240/312]  eta: 0:00:56  lr: 0.003011  min_lr: 0.003011  loss: 3.7920 (3.5742)  weight_decay: 0.0500 (0.0500)  time: 0.8866  data: 0.3903  max mem: 41794
Epoch: [112]  [250/312]  eta: 0:00:48  lr: 0.003011  min_lr: 0.003011  loss: 3.7076 (3.5797)  weight_decay: 0.0500 (0.0500)  time: 0.6812  data: 0.1887  max mem: 41794
Epoch: [112]  [260/312]  eta: 0:00:40  lr: 0.003010  min_lr: 0.003010  loss: 3.7411 (3.5877)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.1981  max mem: 41794
Epoch: [112]  [270/312]  eta: 0:00:32  lr: 0.003009  min_lr: 0.003009  loss: 3.8676 (3.5900)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.1927  max mem: 41794
Epoch: [112]  [280/312]  eta: 0:00:24  lr: 0.003009  min_lr: 0.003009  loss: 3.6254 (3.5892)  weight_decay: 0.0500 (0.0500)  time: 0.7023  data: 0.2028  max mem: 41794
Epoch: [112]  [290/312]  eta: 0:00:17  lr: 0.003008  min_lr: 0.003008  loss: 3.5547 (3.5900)  weight_decay: 0.0500 (0.0500)  time: 0.8604  data: 0.3564  max mem: 41794
Epoch: [112]  [300/312]  eta: 0:00:09  lr: 0.003007  min_lr: 0.003007  loss: 3.7562 (3.5939)  weight_decay: 0.0500 (0.0500)  time: 0.6441  data: 0.1540  max mem: 41794
Epoch: [112]  [310/312]  eta: 0:00:01  lr: 0.003007  min_lr: 0.003007  loss: 3.8385 (3.6050)  weight_decay: 0.0500 (0.0500)  time: 0.4647  data: 0.0001  max mem: 41794
Epoch: [112]  [311/312]  eta: 0:00:00  lr: 0.003007  min_lr: 0.003007  loss: 3.8321 (3.6021)  weight_decay: 0.0500 (0.0500)  time: 0.4637  data: 0.0001  max mem: 41794
Epoch: [112] Total time: 0:03:57 (0.7622 s / it)
Averaged stats: lr: 0.003007  min_lr: 0.003007  loss: 3.8321 (3.5614)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2400 (1.2400)  acc1: 77.2135 (77.2135)  acc5: 93.7500 (93.7500)  time: 8.5192  data: 8.3069  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7093 (1.5491)  acc1: 69.0104 (70.7680)  acc5: 90.7552 (90.8800)  time: 1.0950  data: 0.9231  max mem: 41794
Test: Total time: 0:00:10 (1.1332 s / it)
* Acc@1 70.644 Acc@5 90.454 loss 1.550
Accuracy of the model on the 50000 test images: 70.6%
Max accuracy: 71.18%
Epoch: [113]  [  0/312]  eta: 1:22:55  lr: 0.003007  min_lr: 0.003007  loss: 4.1550 (4.1550)  weight_decay: 0.0500 (0.0500)  time: 15.9477  data: 11.8549  max mem: 41794
Epoch: [113]  [ 10/312]  eta: 0:11:22  lr: 0.003006  min_lr: 0.003006  loss: 3.7360 (3.5138)  weight_decay: 0.0500 (0.0500)  time: 2.2595  data: 1.4109  max mem: 41794
Epoch: [113]  [ 20/312]  eta: 0:07:35  lr: 0.003005  min_lr: 0.003005  loss: 3.5528 (3.4585)  weight_decay: 0.0500 (0.0500)  time: 0.8419  data: 0.2959  max mem: 41794
Epoch: [113]  [ 30/312]  eta: 0:05:46  lr: 0.003005  min_lr: 0.003005  loss: 3.4142 (3.4320)  weight_decay: 0.0500 (0.0500)  time: 0.6611  data: 0.1247  max mem: 41794
Epoch: [113]  [ 40/312]  eta: 0:04:47  lr: 0.003004  min_lr: 0.003004  loss: 3.2926 (3.4232)  weight_decay: 0.0500 (0.0500)  time: 0.5284  data: 0.0224  max mem: 41794
Epoch: [113]  [ 50/312]  eta: 0:04:25  lr: 0.003004  min_lr: 0.003004  loss: 3.4586 (3.4431)  weight_decay: 0.0500 (0.0500)  time: 0.6784  data: 0.1076  max mem: 41794
Epoch: [113]  [ 60/312]  eta: 0:03:54  lr: 0.003003  min_lr: 0.003003  loss: 3.4586 (3.4718)  weight_decay: 0.0500 (0.0500)  time: 0.6678  data: 0.0993  max mem: 41794
Epoch: [113]  [ 70/312]  eta: 0:03:39  lr: 0.003002  min_lr: 0.003002  loss: 3.4758 (3.4613)  weight_decay: 0.0500 (0.0500)  time: 0.6397  data: 0.1130  max mem: 41794
Epoch: [113]  [ 80/312]  eta: 0:03:28  lr: 0.003002  min_lr: 0.003002  loss: 3.6507 (3.4588)  weight_decay: 0.0500 (0.0500)  time: 0.8010  data: 0.2120  max mem: 41794
Epoch: [113]  [ 90/312]  eta: 0:03:13  lr: 0.003001  min_lr: 0.003001  loss: 3.4382 (3.4512)  weight_decay: 0.0500 (0.0500)  time: 0.7417  data: 0.1327  max mem: 41794
Epoch: [113]  [100/312]  eta: 0:03:03  lr: 0.003000  min_lr: 0.003000  loss: 3.2637 (3.4323)  weight_decay: 0.0500 (0.0500)  time: 0.7327  data: 0.1642  max mem: 41794
Epoch: [113]  [110/312]  eta: 0:02:47  lr: 0.003000  min_lr: 0.003000  loss: 3.3298 (3.4320)  weight_decay: 0.0500 (0.0500)  time: 0.6484  data: 0.1341  max mem: 41794
Epoch: [113]  [120/312]  eta: 0:02:40  lr: 0.002999  min_lr: 0.002999  loss: 3.3298 (3.4303)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.1604  max mem: 41794
Epoch: [113]  [130/312]  eta: 0:02:31  lr: 0.002999  min_lr: 0.002999  loss: 3.4574 (3.4384)  weight_decay: 0.0500 (0.0500)  time: 0.8418  data: 0.3024  max mem: 41794
Epoch: [113]  [140/312]  eta: 0:02:19  lr: 0.002998  min_lr: 0.002998  loss: 3.2237 (3.4090)  weight_decay: 0.0500 (0.0500)  time: 0.6409  data: 0.1438  max mem: 41794
Epoch: [113]  [150/312]  eta: 0:02:12  lr: 0.002997  min_lr: 0.002997  loss: 3.2237 (3.4123)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.2077  max mem: 41794
Epoch: [113]  [160/312]  eta: 0:02:03  lr: 0.002997  min_lr: 0.002997  loss: 3.6227 (3.4230)  weight_decay: 0.0500 (0.0500)  time: 0.8495  data: 0.3559  max mem: 41794
Epoch: [113]  [170/312]  eta: 0:01:53  lr: 0.002996  min_lr: 0.002996  loss: 3.5874 (3.4313)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.1765  max mem: 41794
Epoch: [113]  [180/312]  eta: 0:01:46  lr: 0.002996  min_lr: 0.002996  loss: 3.7540 (3.4443)  weight_decay: 0.0500 (0.0500)  time: 0.7552  data: 0.1970  max mem: 41794
Epoch: [113]  [190/312]  eta: 0:01:36  lr: 0.002995  min_lr: 0.002995  loss: 3.7299 (3.4501)  weight_decay: 0.0500 (0.0500)  time: 0.7106  data: 0.1694  max mem: 41794
Epoch: [113]  [200/312]  eta: 0:01:28  lr: 0.002994  min_lr: 0.002994  loss: 3.5046 (3.4434)  weight_decay: 0.0500 (0.0500)  time: 0.6577  data: 0.1513  max mem: 41794
Epoch: [113]  [210/312]  eta: 0:01:21  lr: 0.002994  min_lr: 0.002994  loss: 3.5573 (3.4543)  weight_decay: 0.0500 (0.0500)  time: 0.8448  data: 0.3159  max mem: 41794
Epoch: [113]  [220/312]  eta: 0:01:12  lr: 0.002993  min_lr: 0.002993  loss: 3.7029 (3.4647)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.1688  max mem: 41794
Epoch: [113]  [230/312]  eta: 0:01:04  lr: 0.002992  min_lr: 0.002992  loss: 3.7316 (3.4715)  weight_decay: 0.0500 (0.0500)  time: 0.7179  data: 0.1688  max mem: 41794
Epoch: [113]  [240/312]  eta: 0:00:56  lr: 0.002992  min_lr: 0.002992  loss: 3.7316 (3.4743)  weight_decay: 0.0500 (0.0500)  time: 0.8644  data: 0.3114  max mem: 41794
Epoch: [113]  [250/312]  eta: 0:00:48  lr: 0.002991  min_lr: 0.002991  loss: 3.5956 (3.4723)  weight_decay: 0.0500 (0.0500)  time: 0.6447  data: 0.1469  max mem: 41794
Epoch: [113]  [260/312]  eta: 0:00:40  lr: 0.002991  min_lr: 0.002991  loss: 3.6524 (3.4781)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.1813  max mem: 41794
Epoch: [113]  [270/312]  eta: 0:00:32  lr: 0.002990  min_lr: 0.002990  loss: 3.7396 (3.4816)  weight_decay: 0.0500 (0.0500)  time: 0.6790  data: 0.1813  max mem: 41794
Epoch: [113]  [280/312]  eta: 0:00:24  lr: 0.002989  min_lr: 0.002989  loss: 3.1845 (3.4698)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.1780  max mem: 41794
Epoch: [113]  [290/312]  eta: 0:00:17  lr: 0.002989  min_lr: 0.002989  loss: 3.4218 (3.4779)  weight_decay: 0.0500 (0.0500)  time: 0.8556  data: 0.3382  max mem: 41794
Epoch: [113]  [300/312]  eta: 0:00:09  lr: 0.002988  min_lr: 0.002988  loss: 3.7488 (3.4794)  weight_decay: 0.0500 (0.0500)  time: 0.6762  data: 0.1687  max mem: 41794
Epoch: [113]  [310/312]  eta: 0:00:01  lr: 0.002987  min_lr: 0.002987  loss: 3.6355 (3.4841)  weight_decay: 0.0500 (0.0500)  time: 0.4760  data: 0.0083  max mem: 41794
Epoch: [113]  [311/312]  eta: 0:00:00  lr: 0.002987  min_lr: 0.002987  loss: 3.7401 (3.4862)  weight_decay: 0.0500 (0.0500)  time: 0.4745  data: 0.0083  max mem: 41794
Epoch: [113] Total time: 0:03:58 (0.7629 s / it)
Averaged stats: lr: 0.002987  min_lr: 0.002987  loss: 3.7401 (3.5389)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.4145 (1.4145)  acc1: 77.9948 (77.9948)  acc5: 93.6198 (93.6198)  time: 8.7911  data: 8.5828  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6556 (1.6116)  acc1: 70.4427 (70.7520)  acc5: 91.1458 (90.8480)  time: 1.1256  data: 0.9537  max mem: 41794
Test: Total time: 0:00:10 (1.1699 s / it)
* Acc@1 70.704 Acc@5 90.606 loss 1.618
Accuracy of the model on the 50000 test images: 70.7%
Max accuracy: 71.18%
Epoch: [114]  [  0/312]  eta: 1:22:45  lr: 0.002987  min_lr: 0.002987  loss: 3.1476 (3.1476)  weight_decay: 0.0500 (0.0500)  time: 15.9140  data: 13.9713  max mem: 41794
Epoch: [114]  [ 10/312]  eta: 0:10:45  lr: 0.002987  min_lr: 0.002987  loss: 3.7055 (3.6364)  weight_decay: 0.0500 (0.0500)  time: 2.1385  data: 1.2728  max mem: 41794
Epoch: [114]  [ 20/312]  eta: 0:07:11  lr: 0.002986  min_lr: 0.002986  loss: 3.7350 (3.6062)  weight_decay: 0.0500 (0.0500)  time: 0.7547  data: 0.1196  max mem: 41794
Epoch: [114]  [ 30/312]  eta: 0:05:28  lr: 0.002985  min_lr: 0.002985  loss: 3.8615 (3.6486)  weight_decay: 0.0500 (0.0500)  time: 0.6298  data: 0.1185  max mem: 41794
Epoch: [114]  [ 40/312]  eta: 0:04:50  lr: 0.002985  min_lr: 0.002985  loss: 3.8615 (3.7112)  weight_decay: 0.0500 (0.0500)  time: 0.6392  data: 0.1381  max mem: 41794
Epoch: [114]  [ 50/312]  eta: 0:04:27  lr: 0.002984  min_lr: 0.002984  loss: 3.7478 (3.6524)  weight_decay: 0.0500 (0.0500)  time: 0.7950  data: 0.2690  max mem: 41794
Epoch: [114]  [ 60/312]  eta: 0:03:55  lr: 0.002984  min_lr: 0.002984  loss: 3.7400 (3.6929)  weight_decay: 0.0500 (0.0500)  time: 0.6551  data: 0.1358  max mem: 41794
Epoch: [114]  [ 70/312]  eta: 0:03:47  lr: 0.002983  min_lr: 0.002983  loss: 3.8020 (3.6662)  weight_decay: 0.0500 (0.0500)  time: 0.7336  data: 0.1895  max mem: 41794
Epoch: [114]  [ 80/312]  eta: 0:03:31  lr: 0.002982  min_lr: 0.002982  loss: 3.3384 (3.6005)  weight_decay: 0.0500 (0.0500)  time: 0.8508  data: 0.1937  max mem: 41794
Epoch: [114]  [ 90/312]  eta: 0:03:17  lr: 0.002982  min_lr: 0.002982  loss: 3.6224 (3.6132)  weight_decay: 0.0500 (0.0500)  time: 0.7233  data: 0.1305  max mem: 41794
Epoch: [114]  [100/312]  eta: 0:03:07  lr: 0.002981  min_lr: 0.002981  loss: 3.5910 (3.5709)  weight_decay: 0.0500 (0.0500)  time: 0.7622  data: 0.1909  max mem: 41794
Epoch: [114]  [110/312]  eta: 0:02:52  lr: 0.002980  min_lr: 0.002980  loss: 3.1646 (3.5574)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0904  max mem: 41794
Epoch: [114]  [120/312]  eta: 0:02:44  lr: 0.002980  min_lr: 0.002980  loss: 3.6949 (3.5481)  weight_decay: 0.0500 (0.0500)  time: 0.7293  data: 0.1316  max mem: 41794
Epoch: [114]  [130/312]  eta: 0:02:35  lr: 0.002979  min_lr: 0.002979  loss: 3.6949 (3.5550)  weight_decay: 0.0500 (0.0500)  time: 0.8586  data: 0.1663  max mem: 41794
Epoch: [114]  [140/312]  eta: 0:02:23  lr: 0.002979  min_lr: 0.002979  loss: 3.7661 (3.5510)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.0839  max mem: 41794
Epoch: [114]  [150/312]  eta: 0:02:14  lr: 0.002978  min_lr: 0.002978  loss: 3.2202 (3.5206)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.1236  max mem: 41794
Epoch: [114]  [160/312]  eta: 0:02:04  lr: 0.002977  min_lr: 0.002977  loss: 3.4182 (3.5321)  weight_decay: 0.0500 (0.0500)  time: 0.7555  data: 0.1010  max mem: 41794
Epoch: [114]  [170/312]  eta: 0:01:55  lr: 0.002977  min_lr: 0.002977  loss: 3.6501 (3.5211)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.0997  max mem: 41794
Epoch: [114]  [180/312]  eta: 0:01:47  lr: 0.002976  min_lr: 0.002976  loss: 3.7504 (3.5296)  weight_decay: 0.0500 (0.0500)  time: 0.7694  data: 0.1730  max mem: 41794
Epoch: [114]  [190/312]  eta: 0:01:37  lr: 0.002975  min_lr: 0.002975  loss: 3.7799 (3.5429)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.0942  max mem: 41794
Epoch: [114]  [200/312]  eta: 0:01:30  lr: 0.002975  min_lr: 0.002975  loss: 3.5389 (3.5138)  weight_decay: 0.0500 (0.0500)  time: 0.6987  data: 0.1094  max mem: 41794
Epoch: [114]  [210/312]  eta: 0:01:22  lr: 0.002974  min_lr: 0.002974  loss: 3.3304 (3.5130)  weight_decay: 0.0500 (0.0500)  time: 0.8885  data: 0.1804  max mem: 41794
Epoch: [114]  [220/312]  eta: 0:01:13  lr: 0.002973  min_lr: 0.002973  loss: 3.5657 (3.5190)  weight_decay: 0.0500 (0.0500)  time: 0.7099  data: 0.0872  max mem: 41794
Epoch: [114]  [230/312]  eta: 0:01:05  lr: 0.002973  min_lr: 0.002973  loss: 3.6967 (3.5290)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.0775  max mem: 41794
Epoch: [114]  [240/312]  eta: 0:00:57  lr: 0.002972  min_lr: 0.002972  loss: 3.8065 (3.5283)  weight_decay: 0.0500 (0.0500)  time: 0.7933  data: 0.0821  max mem: 41794
Epoch: [114]  [250/312]  eta: 0:00:48  lr: 0.002972  min_lr: 0.002972  loss: 3.3814 (3.5240)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.0786  max mem: 41794
Epoch: [114]  [260/312]  eta: 0:00:41  lr: 0.002971  min_lr: 0.002971  loss: 3.5224 (3.5283)  weight_decay: 0.0500 (0.0500)  time: 0.7569  data: 0.1373  max mem: 41794
Epoch: [114]  [270/312]  eta: 0:00:32  lr: 0.002970  min_lr: 0.002970  loss: 3.6501 (3.5289)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.0666  max mem: 41794
Epoch: [114]  [280/312]  eta: 0:00:25  lr: 0.002970  min_lr: 0.002970  loss: 3.6501 (3.5299)  weight_decay: 0.0500 (0.0500)  time: 0.7200  data: 0.0832  max mem: 41794
Epoch: [114]  [290/312]  eta: 0:00:17  lr: 0.002969  min_lr: 0.002969  loss: 3.4944 (3.5204)  weight_decay: 0.0500 (0.0500)  time: 0.7603  data: 0.0874  max mem: 41794
Epoch: [114]  [300/312]  eta: 0:00:09  lr: 0.002968  min_lr: 0.002968  loss: 3.3759 (3.5152)  weight_decay: 0.0500 (0.0500)  time: 0.5289  data: 0.0072  max mem: 41794
Epoch: [114]  [310/312]  eta: 0:00:01  lr: 0.002968  min_lr: 0.002968  loss: 3.5199 (3.5178)  weight_decay: 0.0500 (0.0500)  time: 0.4630  data: 0.0001  max mem: 41794
Epoch: [114]  [311/312]  eta: 0:00:00  lr: 0.002968  min_lr: 0.002968  loss: 3.5199 (3.5168)  weight_decay: 0.0500 (0.0500)  time: 0.4632  data: 0.0001  max mem: 41794
Epoch: [114] Total time: 0:03:59 (0.7666 s / it)
Averaged stats: lr: 0.002968  min_lr: 0.002968  loss: 3.5199 (3.5334)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.2842 (1.2842)  acc1: 77.0833 (77.0833)  acc5: 94.5312 (94.5312)  time: 8.7167  data: 8.5048  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6018 (1.5181)  acc1: 71.2240 (70.4640)  acc5: 89.4531 (90.2080)  time: 1.1167  data: 0.9450  max mem: 41794
Test: Total time: 0:00:10 (1.1365 s / it)
* Acc@1 70.434 Acc@5 90.180 loss 1.518
Accuracy of the model on the 50000 test images: 70.4%
Max accuracy: 71.18%
Epoch: [115]  [  0/312]  eta: 1:17:58  lr: 0.002968  min_lr: 0.002968  loss: 2.9702 (2.9702)  weight_decay: 0.0500 (0.0500)  time: 14.9958  data: 12.0793  max mem: 41794
Epoch: [115]  [ 10/312]  eta: 0:11:06  lr: 0.002967  min_lr: 0.002967  loss: 2.9702 (3.2015)  weight_decay: 0.0500 (0.0500)  time: 2.2067  data: 1.2526  max mem: 41794
Epoch: [115]  [ 20/312]  eta: 0:07:10  lr: 0.002966  min_lr: 0.002966  loss: 3.3618 (3.2698)  weight_decay: 0.0500 (0.0500)  time: 0.7986  data: 0.1006  max mem: 41794
Epoch: [115]  [ 30/312]  eta: 0:05:28  lr: 0.002966  min_lr: 0.002966  loss: 3.6756 (3.3663)  weight_decay: 0.0500 (0.0500)  time: 0.5946  data: 0.0167  max mem: 41794
Epoch: [115]  [ 40/312]  eta: 0:04:48  lr: 0.002965  min_lr: 0.002965  loss: 3.7157 (3.4053)  weight_decay: 0.0500 (0.0500)  time: 0.6231  data: 0.0810  max mem: 41794
Epoch: [115]  [ 50/312]  eta: 0:04:21  lr: 0.002965  min_lr: 0.002965  loss: 3.6374 (3.3894)  weight_decay: 0.0500 (0.0500)  time: 0.7399  data: 0.1608  max mem: 41794
Epoch: [115]  [ 60/312]  eta: 0:03:54  lr: 0.002964  min_lr: 0.002964  loss: 3.6374 (3.4434)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.1259  max mem: 41794
Epoch: [115]  [ 70/312]  eta: 0:03:43  lr: 0.002963  min_lr: 0.002963  loss: 3.6348 (3.4505)  weight_decay: 0.0500 (0.0500)  time: 0.7333  data: 0.1896  max mem: 41794
Epoch: [115]  [ 80/312]  eta: 0:03:28  lr: 0.002963  min_lr: 0.002963  loss: 3.7559 (3.4990)  weight_decay: 0.0500 (0.0500)  time: 0.7903  data: 0.2377  max mem: 41794
Epoch: [115]  [ 90/312]  eta: 0:03:13  lr: 0.002962  min_lr: 0.002962  loss: 3.7704 (3.4736)  weight_decay: 0.0500 (0.0500)  time: 0.6946  data: 0.1646  max mem: 41794
Epoch: [115]  [100/312]  eta: 0:03:05  lr: 0.002961  min_lr: 0.002961  loss: 3.0230 (3.4348)  weight_decay: 0.0500 (0.0500)  time: 0.7823  data: 0.2268  max mem: 41794
Epoch: [115]  [110/312]  eta: 0:02:49  lr: 0.002961  min_lr: 0.002961  loss: 3.3727 (3.4514)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.1557  max mem: 41794
Epoch: [115]  [120/312]  eta: 0:02:41  lr: 0.002960  min_lr: 0.002960  loss: 3.5528 (3.4347)  weight_decay: 0.0500 (0.0500)  time: 0.6807  data: 0.1865  max mem: 41794
Epoch: [115]  [130/312]  eta: 0:02:33  lr: 0.002960  min_lr: 0.002960  loss: 3.4223 (3.4408)  weight_decay: 0.0500 (0.0500)  time: 0.8587  data: 0.3458  max mem: 41794
Epoch: [115]  [140/312]  eta: 0:02:20  lr: 0.002959  min_lr: 0.002959  loss: 3.5041 (3.4332)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.1710  max mem: 41794
Epoch: [115]  [150/312]  eta: 0:02:14  lr: 0.002958  min_lr: 0.002958  loss: 3.7016 (3.4581)  weight_decay: 0.0500 (0.0500)  time: 0.7447  data: 0.2407  max mem: 41794
Epoch: [115]  [160/312]  eta: 0:02:03  lr: 0.002958  min_lr: 0.002958  loss: 3.7604 (3.4640)  weight_decay: 0.0500 (0.0500)  time: 0.7893  data: 0.2769  max mem: 41794
Epoch: [115]  [170/312]  eta: 0:01:55  lr: 0.002957  min_lr: 0.002957  loss: 3.7420 (3.4747)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.1685  max mem: 41794
Epoch: [115]  [180/312]  eta: 0:01:47  lr: 0.002956  min_lr: 0.002956  loss: 3.5089 (3.4794)  weight_decay: 0.0500 (0.0500)  time: 0.8118  data: 0.2636  max mem: 41794
Epoch: [115]  [190/312]  eta: 0:01:37  lr: 0.002956  min_lr: 0.002956  loss: 3.4502 (3.4793)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.1430  max mem: 41794
Epoch: [115]  [200/312]  eta: 0:01:29  lr: 0.002955  min_lr: 0.002955  loss: 3.3126 (3.4654)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.1431  max mem: 41794
Epoch: [115]  [210/312]  eta: 0:01:21  lr: 0.002954  min_lr: 0.002954  loss: 3.5126 (3.4717)  weight_decay: 0.0500 (0.0500)  time: 0.8092  data: 0.2453  max mem: 41794
Epoch: [115]  [220/312]  eta: 0:01:12  lr: 0.002954  min_lr: 0.002954  loss: 3.5126 (3.4669)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.1441  max mem: 41794
Epoch: [115]  [230/312]  eta: 0:01:04  lr: 0.002953  min_lr: 0.002953  loss: 3.3646 (3.4686)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.1969  max mem: 41794
Epoch: [115]  [240/312]  eta: 0:00:56  lr: 0.002953  min_lr: 0.002953  loss: 3.4803 (3.4665)  weight_decay: 0.0500 (0.0500)  time: 0.7548  data: 0.2270  max mem: 41794
Epoch: [115]  [250/312]  eta: 0:00:48  lr: 0.002952  min_lr: 0.002952  loss: 3.4803 (3.4619)  weight_decay: 0.0500 (0.0500)  time: 0.7494  data: 0.1652  max mem: 41794
Epoch: [115]  [260/312]  eta: 0:00:41  lr: 0.002951  min_lr: 0.002951  loss: 3.6279 (3.4649)  weight_decay: 0.0500 (0.0500)  time: 0.8448  data: 0.1868  max mem: 41794
Epoch: [115]  [270/312]  eta: 0:00:32  lr: 0.002951  min_lr: 0.002951  loss: 3.5952 (3.4663)  weight_decay: 0.0500 (0.0500)  time: 0.6788  data: 0.0954  max mem: 41794
Epoch: [115]  [280/312]  eta: 0:00:24  lr: 0.002950  min_lr: 0.002950  loss: 3.6985 (3.4776)  weight_decay: 0.0500 (0.0500)  time: 0.6268  data: 0.0820  max mem: 41794
Epoch: [115]  [290/312]  eta: 0:00:17  lr: 0.002949  min_lr: 0.002949  loss: 3.7750 (3.4774)  weight_decay: 0.0500 (0.0500)  time: 0.7500  data: 0.1931  max mem: 41794
Epoch: [115]  [300/312]  eta: 0:00:09  lr: 0.002949  min_lr: 0.002949  loss: 3.3518 (3.4752)  weight_decay: 0.0500 (0.0500)  time: 0.6196  data: 0.1133  max mem: 41794
Epoch: [115]  [310/312]  eta: 0:00:01  lr: 0.002948  min_lr: 0.002948  loss: 3.4807 (3.4808)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [115]  [311/312]  eta: 0:00:00  lr: 0.002948  min_lr: 0.002948  loss: 3.5127 (3.4814)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [115] Total time: 0:03:58 (0.7638 s / it)
Averaged stats: lr: 0.002948  min_lr: 0.002948  loss: 3.5127 (3.5368)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.1959 (1.1959)  acc1: 77.7344 (77.7344)  acc5: 93.6198 (93.6198)  time: 8.5814  data: 8.3685  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5577 (1.4399)  acc1: 68.3594 (70.1440)  acc5: 89.9740 (90.7040)  time: 1.1021  data: 0.9299  max mem: 41794
Test: Total time: 0:00:10 (1.1417 s / it)
* Acc@1 70.360 Acc@5 90.390 loss 1.436
Accuracy of the model on the 50000 test images: 70.4%
Max accuracy: 71.18%
Epoch: [116]  [  0/312]  eta: 1:21:42  lr: 0.002948  min_lr: 0.002948  loss: 3.6815 (3.6815)  weight_decay: 0.0500 (0.0500)  time: 15.7127  data: 13.0239  max mem: 41794
Epoch: [116]  [ 10/312]  eta: 0:11:18  lr: 0.002947  min_lr: 0.002947  loss: 3.5220 (3.3750)  weight_decay: 0.0500 (0.0500)  time: 2.2477  data: 1.3349  max mem: 41794
Epoch: [116]  [ 20/312]  eta: 0:07:23  lr: 0.002947  min_lr: 0.002947  loss: 3.5220 (3.4806)  weight_decay: 0.0500 (0.0500)  time: 0.8087  data: 0.1055  max mem: 41794
Epoch: [116]  [ 30/312]  eta: 0:05:41  lr: 0.002946  min_lr: 0.002946  loss: 3.5892 (3.5036)  weight_decay: 0.0500 (0.0500)  time: 0.6424  data: 0.0294  max mem: 41794
Epoch: [116]  [ 40/312]  eta: 0:04:45  lr: 0.002945  min_lr: 0.002945  loss: 3.5001 (3.4615)  weight_decay: 0.0500 (0.0500)  time: 0.5611  data: 0.0088  max mem: 41794
Epoch: [116]  [ 50/312]  eta: 0:04:22  lr: 0.002945  min_lr: 0.002945  loss: 3.4793 (3.4796)  weight_decay: 0.0500 (0.0500)  time: 0.6723  data: 0.0604  max mem: 41794
Epoch: [116]  [ 60/312]  eta: 0:03:51  lr: 0.002944  min_lr: 0.002944  loss: 3.6055 (3.4891)  weight_decay: 0.0500 (0.0500)  time: 0.6417  data: 0.0588  max mem: 41794
Epoch: [116]  [ 70/312]  eta: 0:03:40  lr: 0.002944  min_lr: 0.002944  loss: 3.6455 (3.5200)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.0693  max mem: 41794
Epoch: [116]  [ 80/312]  eta: 0:03:27  lr: 0.002943  min_lr: 0.002943  loss: 3.7787 (3.5479)  weight_decay: 0.0500 (0.0500)  time: 0.8336  data: 0.1154  max mem: 41794
Epoch: [116]  [ 90/312]  eta: 0:03:10  lr: 0.002942  min_lr: 0.002942  loss: 3.7623 (3.5632)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.0534  max mem: 41794
Epoch: [116]  [100/312]  eta: 0:03:04  lr: 0.002942  min_lr: 0.002942  loss: 3.6292 (3.5559)  weight_decay: 0.0500 (0.0500)  time: 0.7687  data: 0.0510  max mem: 41794
Epoch: [116]  [110/312]  eta: 0:02:49  lr: 0.002941  min_lr: 0.002941  loss: 3.7662 (3.5583)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.0444  max mem: 41794
Epoch: [116]  [120/312]  eta: 0:02:39  lr: 0.002940  min_lr: 0.002940  loss: 3.2107 (3.5374)  weight_decay: 0.0500 (0.0500)  time: 0.6416  data: 0.0092  max mem: 41794
Epoch: [116]  [130/312]  eta: 0:02:31  lr: 0.002940  min_lr: 0.002940  loss: 3.1255 (3.5212)  weight_decay: 0.0500 (0.0500)  time: 0.8130  data: 0.0327  max mem: 41794
Epoch: [116]  [140/312]  eta: 0:02:19  lr: 0.002939  min_lr: 0.002939  loss: 3.5220 (3.5191)  weight_decay: 0.0500 (0.0500)  time: 0.6892  data: 0.0423  max mem: 41794
Epoch: [116]  [150/312]  eta: 0:02:12  lr: 0.002938  min_lr: 0.002938  loss: 3.3156 (3.4999)  weight_decay: 0.0500 (0.0500)  time: 0.7156  data: 0.0629  max mem: 41794
Epoch: [116]  [160/312]  eta: 0:02:04  lr: 0.002938  min_lr: 0.002938  loss: 3.3156 (3.5051)  weight_decay: 0.0500 (0.0500)  time: 0.8524  data: 0.0578  max mem: 41794
Epoch: [116]  [170/312]  eta: 0:01:54  lr: 0.002937  min_lr: 0.002937  loss: 3.5902 (3.4998)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.0197  max mem: 41794
Epoch: [116]  [180/312]  eta: 0:01:45  lr: 0.002937  min_lr: 0.002937  loss: 3.6320 (3.5093)  weight_decay: 0.0500 (0.0500)  time: 0.6805  data: 0.0343  max mem: 41794
Epoch: [116]  [190/312]  eta: 0:01:36  lr: 0.002936  min_lr: 0.002936  loss: 3.7514 (3.5143)  weight_decay: 0.0500 (0.0500)  time: 0.6662  data: 0.0294  max mem: 41794
Epoch: [116]  [200/312]  eta: 0:01:28  lr: 0.002935  min_lr: 0.002935  loss: 3.6339 (3.5027)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.0539  max mem: 41794
Epoch: [116]  [210/312]  eta: 0:01:20  lr: 0.002935  min_lr: 0.002935  loss: 3.4963 (3.5063)  weight_decay: 0.0500 (0.0500)  time: 0.8416  data: 0.0989  max mem: 41794
Epoch: [116]  [220/312]  eta: 0:01:11  lr: 0.002934  min_lr: 0.002934  loss: 3.8202 (3.5071)  weight_decay: 0.0500 (0.0500)  time: 0.6854  data: 0.0468  max mem: 41794
Epoch: [116]  [230/312]  eta: 0:01:04  lr: 0.002933  min_lr: 0.002933  loss: 3.8012 (3.5106)  weight_decay: 0.0500 (0.0500)  time: 0.6706  data: 0.0385  max mem: 41794
Epoch: [116]  [240/312]  eta: 0:00:56  lr: 0.002933  min_lr: 0.002933  loss: 3.6887 (3.5124)  weight_decay: 0.0500 (0.0500)  time: 0.8239  data: 0.0706  max mem: 41794
Epoch: [116]  [250/312]  eta: 0:00:48  lr: 0.002932  min_lr: 0.002932  loss: 3.6675 (3.5212)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.0375  max mem: 41794
Epoch: [116]  [260/312]  eta: 0:00:40  lr: 0.002931  min_lr: 0.002931  loss: 3.6762 (3.5232)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.0213  max mem: 41794
Epoch: [116]  [270/312]  eta: 0:00:32  lr: 0.002931  min_lr: 0.002931  loss: 3.5023 (3.5227)  weight_decay: 0.0500 (0.0500)  time: 0.6942  data: 0.0178  max mem: 41794
Epoch: [116]  [280/312]  eta: 0:00:24  lr: 0.002930  min_lr: 0.002930  loss: 3.4302 (3.5128)  weight_decay: 0.0500 (0.0500)  time: 0.6742  data: 0.0142  max mem: 41794
Epoch: [116]  [290/312]  eta: 0:00:17  lr: 0.002930  min_lr: 0.002930  loss: 3.4583 (3.5054)  weight_decay: 0.0500 (0.0500)  time: 0.8445  data: 0.0308  max mem: 41794
Epoch: [116]  [300/312]  eta: 0:00:09  lr: 0.002929  min_lr: 0.002929  loss: 3.6314 (3.5088)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.0183  max mem: 41794
Epoch: [116]  [310/312]  eta: 0:00:01  lr: 0.002928  min_lr: 0.002928  loss: 3.7049 (3.5112)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0001  max mem: 41794
Epoch: [116]  [311/312]  eta: 0:00:00  lr: 0.002928  min_lr: 0.002928  loss: 3.7049 (3.5120)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [116] Total time: 0:03:57 (0.7611 s / it)
Averaged stats: lr: 0.002928  min_lr: 0.002928  loss: 3.7049 (3.5200)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1335 (1.1335)  acc1: 78.5156 (78.5156)  acc5: 93.4896 (93.4896)  time: 8.5408  data: 8.3308  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5622 (1.4421)  acc1: 72.3958 (71.4400)  acc5: 90.1042 (90.4480)  time: 1.0973  data: 0.9258  max mem: 41794
Test: Total time: 0:00:10 (1.1428 s / it)
* Acc@1 70.840 Acc@5 90.440 loss 1.459
Accuracy of the model on the 50000 test images: 70.8%
Max accuracy: 71.18%
Epoch: [117]  [  0/312]  eta: 1:25:52  lr: 0.002928  min_lr: 0.002928  loss: 4.2974 (4.2974)  weight_decay: 0.0500 (0.0500)  time: 16.5146  data: 16.0337  max mem: 41794
Epoch: [117]  [ 10/312]  eta: 0:11:26  lr: 0.002928  min_lr: 0.002928  loss: 3.8007 (3.6090)  weight_decay: 0.0500 (0.0500)  time: 2.2738  data: 1.4586  max mem: 41794
Epoch: [117]  [ 20/312]  eta: 0:07:28  lr: 0.002927  min_lr: 0.002927  loss: 3.8007 (3.6965)  weight_decay: 0.0500 (0.0500)  time: 0.7876  data: 0.1205  max mem: 41794
Epoch: [117]  [ 30/312]  eta: 0:05:42  lr: 0.002926  min_lr: 0.002926  loss: 3.4762 (3.5670)  weight_decay: 0.0500 (0.0500)  time: 0.6320  data: 0.1243  max mem: 41794
Epoch: [117]  [ 40/312]  eta: 0:04:52  lr: 0.002926  min_lr: 0.002926  loss: 3.4738 (3.5798)  weight_decay: 0.0500 (0.0500)  time: 0.5927  data: 0.0808  max mem: 41794
Epoch: [117]  [ 50/312]  eta: 0:04:29  lr: 0.002925  min_lr: 0.002925  loss: 3.5145 (3.5525)  weight_decay: 0.0500 (0.0500)  time: 0.7443  data: 0.2402  max mem: 41794
Epoch: [117]  [ 60/312]  eta: 0:03:56  lr: 0.002924  min_lr: 0.002924  loss: 3.3929 (3.5122)  weight_decay: 0.0500 (0.0500)  time: 0.6602  data: 0.1641  max mem: 41794
Epoch: [117]  [ 70/312]  eta: 0:03:44  lr: 0.002924  min_lr: 0.002924  loss: 3.7480 (3.5514)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.1745  max mem: 41794
Epoch: [117]  [ 80/312]  eta: 0:03:32  lr: 0.002923  min_lr: 0.002923  loss: 3.7621 (3.5342)  weight_decay: 0.0500 (0.0500)  time: 0.8413  data: 0.3464  max mem: 41794
Epoch: [117]  [ 90/312]  eta: 0:03:12  lr: 0.002922  min_lr: 0.002922  loss: 3.7484 (3.5483)  weight_decay: 0.0500 (0.0500)  time: 0.6632  data: 0.1725  max mem: 41794
Epoch: [117]  [100/312]  eta: 0:03:04  lr: 0.002922  min_lr: 0.002922  loss: 3.8528 (3.5717)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.1779  max mem: 41794
Epoch: [117]  [110/312]  eta: 0:02:48  lr: 0.002921  min_lr: 0.002921  loss: 3.7146 (3.5654)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.1780  max mem: 41794
Epoch: [117]  [120/312]  eta: 0:02:39  lr: 0.002921  min_lr: 0.002921  loss: 3.6401 (3.5558)  weight_decay: 0.0500 (0.0500)  time: 0.6378  data: 0.1414  max mem: 41794
Epoch: [117]  [130/312]  eta: 0:02:32  lr: 0.002920  min_lr: 0.002920  loss: 3.3841 (3.5329)  weight_decay: 0.0500 (0.0500)  time: 0.8463  data: 0.3478  max mem: 41794
Epoch: [117]  [140/312]  eta: 0:02:19  lr: 0.002919  min_lr: 0.002919  loss: 3.4125 (3.5296)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.2071  max mem: 41794
Epoch: [117]  [150/312]  eta: 0:02:11  lr: 0.002919  min_lr: 0.002919  loss: 3.6828 (3.5378)  weight_decay: 0.0500 (0.0500)  time: 0.6689  data: 0.1680  max mem: 41794
Epoch: [117]  [160/312]  eta: 0:02:04  lr: 0.002918  min_lr: 0.002918  loss: 3.7673 (3.5493)  weight_decay: 0.0500 (0.0500)  time: 0.8550  data: 0.3572  max mem: 41794
Epoch: [117]  [170/312]  eta: 0:01:53  lr: 0.002917  min_lr: 0.002917  loss: 3.7357 (3.5524)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.1898  max mem: 41794
Epoch: [117]  [180/312]  eta: 0:01:45  lr: 0.002917  min_lr: 0.002917  loss: 3.5089 (3.5353)  weight_decay: 0.0500 (0.0500)  time: 0.6670  data: 0.1677  max mem: 41794
Epoch: [117]  [190/312]  eta: 0:01:35  lr: 0.002916  min_lr: 0.002916  loss: 3.4374 (3.5354)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.1679  max mem: 41794
Epoch: [117]  [200/312]  eta: 0:01:28  lr: 0.002915  min_lr: 0.002915  loss: 3.5265 (3.5289)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.1872  max mem: 41794
Epoch: [117]  [210/312]  eta: 0:01:20  lr: 0.002915  min_lr: 0.002915  loss: 3.3391 (3.5161)  weight_decay: 0.0500 (0.0500)  time: 0.8745  data: 0.3702  max mem: 41794
Epoch: [117]  [220/312]  eta: 0:01:11  lr: 0.002914  min_lr: 0.002914  loss: 3.5989 (3.5265)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1838  max mem: 41794
Epoch: [117]  [230/312]  eta: 0:01:04  lr: 0.002914  min_lr: 0.002914  loss: 3.5989 (3.5109)  weight_decay: 0.0500 (0.0500)  time: 0.6803  data: 0.1845  max mem: 41794
Epoch: [117]  [240/312]  eta: 0:00:56  lr: 0.002913  min_lr: 0.002913  loss: 3.4807 (3.5125)  weight_decay: 0.0500 (0.0500)  time: 0.8681  data: 0.3827  max mem: 41794
Epoch: [117]  [250/312]  eta: 0:00:48  lr: 0.002912  min_lr: 0.002912  loss: 3.6545 (3.5228)  weight_decay: 0.0500 (0.0500)  time: 0.6878  data: 0.2006  max mem: 41794
Epoch: [117]  [260/312]  eta: 0:00:40  lr: 0.002912  min_lr: 0.002912  loss: 3.8592 (3.5329)  weight_decay: 0.0500 (0.0500)  time: 0.6703  data: 0.1730  max mem: 41794
Epoch: [117]  [270/312]  eta: 0:00:32  lr: 0.002911  min_lr: 0.002911  loss: 3.8017 (3.5361)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.1713  max mem: 41794
Epoch: [117]  [280/312]  eta: 0:00:24  lr: 0.002910  min_lr: 0.002910  loss: 3.4876 (3.5274)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.1877  max mem: 41794
Epoch: [117]  [290/312]  eta: 0:00:17  lr: 0.002910  min_lr: 0.002910  loss: 3.3905 (3.5290)  weight_decay: 0.0500 (0.0500)  time: 0.8795  data: 0.3839  max mem: 41794
Epoch: [117]  [300/312]  eta: 0:00:09  lr: 0.002909  min_lr: 0.002909  loss: 3.4352 (3.5263)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1967  max mem: 41794
Epoch: [117]  [310/312]  eta: 0:00:01  lr: 0.002908  min_lr: 0.002908  loss: 3.4352 (3.5179)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [117]  [311/312]  eta: 0:00:00  lr: 0.002908  min_lr: 0.002908  loss: 3.4352 (3.5161)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [117] Total time: 0:03:57 (0.7606 s / it)
Averaged stats: lr: 0.002908  min_lr: 0.002908  loss: 3.4352 (3.5379)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:09  loss: 1.1312 (1.1312)  acc1: 78.2552 (78.2552)  acc5: 94.0104 (94.0104)  time: 7.7251  data: 7.5175  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4684 (1.3705)  acc1: 69.5312 (70.9120)  acc5: 90.7552 (90.6240)  time: 1.0415  data: 0.8702  max mem: 41794
Test: Total time: 0:00:09 (1.0554 s / it)
* Acc@1 70.848 Acc@5 90.510 loss 1.370
Accuracy of the model on the 50000 test images: 70.8%
Max accuracy: 71.18%
Epoch: [118]  [  0/312]  eta: 1:22:25  lr: 0.002908  min_lr: 0.002908  loss: 4.1105 (4.1105)  weight_decay: 0.0500 (0.0500)  time: 15.8495  data: 14.9231  max mem: 41794
Epoch: [118]  [ 10/312]  eta: 0:11:29  lr: 0.002908  min_lr: 0.002908  loss: 3.7538 (3.5909)  weight_decay: 0.0500 (0.0500)  time: 2.2829  data: 1.3573  max mem: 41794
Epoch: [118]  [ 20/312]  eta: 0:07:24  lr: 0.002907  min_lr: 0.002907  loss: 3.5865 (3.4533)  weight_decay: 0.0500 (0.0500)  time: 0.8057  data: 0.0901  max mem: 41794
Epoch: [118]  [ 30/312]  eta: 0:05:39  lr: 0.002906  min_lr: 0.002906  loss: 3.6526 (3.5263)  weight_decay: 0.0500 (0.0500)  time: 0.6128  data: 0.0953  max mem: 41794
Epoch: [118]  [ 40/312]  eta: 0:04:42  lr: 0.002906  min_lr: 0.002906  loss: 3.6783 (3.4743)  weight_decay: 0.0500 (0.0500)  time: 0.5274  data: 0.0121  max mem: 41794
Epoch: [118]  [ 50/312]  eta: 0:04:19  lr: 0.002905  min_lr: 0.002905  loss: 3.5331 (3.5163)  weight_decay: 0.0500 (0.0500)  time: 0.6530  data: 0.0986  max mem: 41794
Epoch: [118]  [ 60/312]  eta: 0:03:48  lr: 0.002904  min_lr: 0.002904  loss: 3.5777 (3.5134)  weight_decay: 0.0500 (0.0500)  time: 0.6399  data: 0.0963  max mem: 41794
Epoch: [118]  [ 70/312]  eta: 0:03:38  lr: 0.002904  min_lr: 0.002904  loss: 3.6126 (3.5213)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1475  max mem: 41794
Epoch: [118]  [ 80/312]  eta: 0:03:25  lr: 0.002903  min_lr: 0.002903  loss: 3.6549 (3.5337)  weight_decay: 0.0500 (0.0500)  time: 0.8206  data: 0.2786  max mem: 41794
Epoch: [118]  [ 90/312]  eta: 0:03:09  lr: 0.002902  min_lr: 0.002902  loss: 3.7238 (3.5148)  weight_decay: 0.0500 (0.0500)  time: 0.6778  data: 0.1356  max mem: 41794
Epoch: [118]  [100/312]  eta: 0:03:03  lr: 0.002902  min_lr: 0.002902  loss: 3.7214 (3.5316)  weight_decay: 0.0500 (0.0500)  time: 0.7733  data: 0.1011  max mem: 41794
Epoch: [118]  [110/312]  eta: 0:02:47  lr: 0.002901  min_lr: 0.002901  loss: 3.6380 (3.5119)  weight_decay: 0.0500 (0.0500)  time: 0.7217  data: 0.1011  max mem: 41794
Epoch: [118]  [120/312]  eta: 0:02:40  lr: 0.002901  min_lr: 0.002901  loss: 3.8192 (3.5555)  weight_decay: 0.0500 (0.0500)  time: 0.6924  data: 0.0675  max mem: 41794
Epoch: [118]  [130/312]  eta: 0:02:32  lr: 0.002900  min_lr: 0.002900  loss: 3.8192 (3.5330)  weight_decay: 0.0500 (0.0500)  time: 0.8805  data: 0.0978  max mem: 41794
Epoch: [118]  [140/312]  eta: 0:02:20  lr: 0.002899  min_lr: 0.002899  loss: 3.3419 (3.5327)  weight_decay: 0.0500 (0.0500)  time: 0.7033  data: 0.0572  max mem: 41794
Epoch: [118]  [150/312]  eta: 0:02:11  lr: 0.002899  min_lr: 0.002899  loss: 3.5670 (3.5519)  weight_decay: 0.0500 (0.0500)  time: 0.6543  data: 0.0667  max mem: 41794
Epoch: [118]  [160/312]  eta: 0:02:02  lr: 0.002898  min_lr: 0.002898  loss: 3.6125 (3.5440)  weight_decay: 0.0500 (0.0500)  time: 0.7505  data: 0.0934  max mem: 41794
Epoch: [118]  [170/312]  eta: 0:01:53  lr: 0.002897  min_lr: 0.002897  loss: 3.6125 (3.5542)  weight_decay: 0.0500 (0.0500)  time: 0.6691  data: 0.0536  max mem: 41794
Epoch: [118]  [180/312]  eta: 0:01:46  lr: 0.002897  min_lr: 0.002897  loss: 3.8437 (3.5648)  weight_decay: 0.0500 (0.0500)  time: 0.7708  data: 0.0589  max mem: 41794
Epoch: [118]  [190/312]  eta: 0:01:36  lr: 0.002896  min_lr: 0.002896  loss: 3.7230 (3.5623)  weight_decay: 0.0500 (0.0500)  time: 0.7109  data: 0.0588  max mem: 41794
Epoch: [118]  [200/312]  eta: 0:01:28  lr: 0.002895  min_lr: 0.002895  loss: 3.6166 (3.5557)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.0574  max mem: 41794
Epoch: [118]  [210/312]  eta: 0:01:20  lr: 0.002895  min_lr: 0.002895  loss: 3.3823 (3.5433)  weight_decay: 0.0500 (0.0500)  time: 0.8425  data: 0.0576  max mem: 41794
Epoch: [118]  [220/312]  eta: 0:01:11  lr: 0.002894  min_lr: 0.002894  loss: 3.5529 (3.5471)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.0310  max mem: 41794
Epoch: [118]  [230/312]  eta: 0:01:04  lr: 0.002893  min_lr: 0.002893  loss: 3.4728 (3.5357)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0715  max mem: 41794
Epoch: [118]  [240/312]  eta: 0:00:56  lr: 0.002893  min_lr: 0.002893  loss: 3.4431 (3.5379)  weight_decay: 0.0500 (0.0500)  time: 0.7807  data: 0.0593  max mem: 41794
Epoch: [118]  [250/312]  eta: 0:00:48  lr: 0.002892  min_lr: 0.002892  loss: 3.7449 (3.5351)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.0553  max mem: 41794
Epoch: [118]  [260/312]  eta: 0:00:40  lr: 0.002892  min_lr: 0.002892  loss: 3.7262 (3.5381)  weight_decay: 0.0500 (0.0500)  time: 0.7270  data: 0.0826  max mem: 41794
Epoch: [118]  [270/312]  eta: 0:00:32  lr: 0.002891  min_lr: 0.002891  loss: 3.6307 (3.5369)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.0464  max mem: 41794
Epoch: [118]  [280/312]  eta: 0:00:24  lr: 0.002890  min_lr: 0.002890  loss: 3.4879 (3.5303)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.0738  max mem: 41794
Epoch: [118]  [290/312]  eta: 0:00:16  lr: 0.002890  min_lr: 0.002890  loss: 3.4122 (3.5281)  weight_decay: 0.0500 (0.0500)  time: 0.8423  data: 0.1141  max mem: 41794
Epoch: [118]  [300/312]  eta: 0:00:09  lr: 0.002889  min_lr: 0.002889  loss: 3.5740 (3.5346)  weight_decay: 0.0500 (0.0500)  time: 0.6220  data: 0.0412  max mem: 41794
Epoch: [118]  [310/312]  eta: 0:00:01  lr: 0.002888  min_lr: 0.002888  loss: 3.4918 (3.5245)  weight_decay: 0.0500 (0.0500)  time: 0.4694  data: 0.0001  max mem: 41794
Epoch: [118]  [311/312]  eta: 0:00:00  lr: 0.002888  min_lr: 0.002888  loss: 3.3785 (3.5239)  weight_decay: 0.0500 (0.0500)  time: 0.4681  data: 0.0001  max mem: 41794
Epoch: [118] Total time: 0:03:56 (0.7586 s / it)
Averaged stats: lr: 0.002888  min_lr: 0.002888  loss: 3.3785 (3.5253)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.1647 (1.1647)  acc1: 75.0000 (75.0000)  acc5: 93.8802 (93.8802)  time: 8.3798  data: 8.1667  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4430 (1.4073)  acc1: 70.4427 (70.5120)  acc5: 90.1042 (90.5120)  time: 1.0909  data: 0.9194  max mem: 41794
Test: Total time: 0:00:09 (1.0989 s / it)
* Acc@1 70.840 Acc@5 90.406 loss 1.393
Accuracy of the model on the 50000 test images: 70.8%
Max accuracy: 71.18%
Epoch: [119]  [  0/312]  eta: 1:16:55  lr: 0.002888  min_lr: 0.002888  loss: 3.9037 (3.9037)  weight_decay: 0.0500 (0.0500)  time: 14.7922  data: 12.5182  max mem: 41794
Epoch: [119]  [ 10/312]  eta: 0:10:33  lr: 0.002888  min_lr: 0.002888  loss: 3.9308 (3.7623)  weight_decay: 0.0500 (0.0500)  time: 2.0977  data: 1.3835  max mem: 41794
Epoch: [119]  [ 20/312]  eta: 0:07:15  lr: 0.002887  min_lr: 0.002887  loss: 3.6659 (3.5773)  weight_decay: 0.0500 (0.0500)  time: 0.8280  data: 0.2476  max mem: 41794
Epoch: [119]  [ 30/312]  eta: 0:05:33  lr: 0.002886  min_lr: 0.002886  loss: 3.5489 (3.5864)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.1176  max mem: 41794
Epoch: [119]  [ 40/312]  eta: 0:04:49  lr: 0.002886  min_lr: 0.002886  loss: 3.6146 (3.5356)  weight_decay: 0.0500 (0.0500)  time: 0.6175  data: 0.0716  max mem: 41794
Epoch: [119]  [ 50/312]  eta: 0:04:28  lr: 0.002885  min_lr: 0.002885  loss: 3.6424 (3.5897)  weight_decay: 0.0500 (0.0500)  time: 0.7816  data: 0.1876  max mem: 41794
Epoch: [119]  [ 60/312]  eta: 0:03:56  lr: 0.002884  min_lr: 0.002884  loss: 3.7872 (3.5714)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1213  max mem: 41794
Epoch: [119]  [ 70/312]  eta: 0:03:51  lr: 0.002884  min_lr: 0.002884  loss: 3.3213 (3.5215)  weight_decay: 0.0500 (0.0500)  time: 0.7748  data: 0.1145  max mem: 41794
Epoch: [119]  [ 80/312]  eta: 0:03:33  lr: 0.002883  min_lr: 0.002883  loss: 3.2878 (3.4887)  weight_decay: 0.0500 (0.0500)  time: 0.8683  data: 0.1230  max mem: 41794
Epoch: [119]  [ 90/312]  eta: 0:03:19  lr: 0.002882  min_lr: 0.002882  loss: 3.5321 (3.5085)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.0399  max mem: 41794
Epoch: [119]  [100/312]  eta: 0:03:09  lr: 0.002882  min_lr: 0.002882  loss: 3.7170 (3.5070)  weight_decay: 0.0500 (0.0500)  time: 0.7849  data: 0.1257  max mem: 41794
Epoch: [119]  [110/312]  eta: 0:02:54  lr: 0.002881  min_lr: 0.002881  loss: 3.5131 (3.5140)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.0948  max mem: 41794
Epoch: [119]  [120/312]  eta: 0:02:44  lr: 0.002880  min_lr: 0.002880  loss: 3.5095 (3.5012)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.0679  max mem: 41794
Epoch: [119]  [130/312]  eta: 0:02:33  lr: 0.002880  min_lr: 0.002880  loss: 3.3256 (3.5003)  weight_decay: 0.0500 (0.0500)  time: 0.7546  data: 0.1834  max mem: 41794
Epoch: [119]  [140/312]  eta: 0:02:22  lr: 0.002879  min_lr: 0.002879  loss: 3.5057 (3.5054)  weight_decay: 0.0500 (0.0500)  time: 0.6431  data: 0.1177  max mem: 41794
Epoch: [119]  [150/312]  eta: 0:02:13  lr: 0.002879  min_lr: 0.002879  loss: 3.6109 (3.5134)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.1463  max mem: 41794
Epoch: [119]  [160/312]  eta: 0:02:04  lr: 0.002878  min_lr: 0.002878  loss: 3.6673 (3.5252)  weight_decay: 0.0500 (0.0500)  time: 0.7544  data: 0.2137  max mem: 41794
Epoch: [119]  [170/312]  eta: 0:01:55  lr: 0.002877  min_lr: 0.002877  loss: 3.6673 (3.5257)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.1345  max mem: 41794
Epoch: [119]  [180/312]  eta: 0:01:48  lr: 0.002877  min_lr: 0.002877  loss: 3.5686 (3.5292)  weight_decay: 0.0500 (0.0500)  time: 0.8306  data: 0.1981  max mem: 41794
Epoch: [119]  [190/312]  eta: 0:01:37  lr: 0.002876  min_lr: 0.002876  loss: 3.6893 (3.5380)  weight_decay: 0.0500 (0.0500)  time: 0.7274  data: 0.1331  max mem: 41794
Epoch: [119]  [200/312]  eta: 0:01:29  lr: 0.002875  min_lr: 0.002875  loss: 3.6893 (3.5428)  weight_decay: 0.0500 (0.0500)  time: 0.6428  data: 0.1094  max mem: 41794
Epoch: [119]  [210/312]  eta: 0:01:21  lr: 0.002875  min_lr: 0.002875  loss: 3.4979 (3.5334)  weight_decay: 0.0500 (0.0500)  time: 0.8109  data: 0.2366  max mem: 41794
Epoch: [119]  [220/312]  eta: 0:01:12  lr: 0.002874  min_lr: 0.002874  loss: 3.5722 (3.5427)  weight_decay: 0.0500 (0.0500)  time: 0.6565  data: 0.1278  max mem: 41794
Epoch: [119]  [230/312]  eta: 0:01:04  lr: 0.002873  min_lr: 0.002873  loss: 3.6763 (3.5454)  weight_decay: 0.0500 (0.0500)  time: 0.6541  data: 0.1592  max mem: 41794
Epoch: [119]  [240/312]  eta: 0:00:56  lr: 0.002873  min_lr: 0.002873  loss: 3.5473 (3.5406)  weight_decay: 0.0500 (0.0500)  time: 0.7793  data: 0.2755  max mem: 41794
Epoch: [119]  [250/312]  eta: 0:00:48  lr: 0.002872  min_lr: 0.002872  loss: 3.3540 (3.5309)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1942  max mem: 41794
Epoch: [119]  [260/312]  eta: 0:00:40  lr: 0.002871  min_lr: 0.002871  loss: 3.5069 (3.5366)  weight_decay: 0.0500 (0.0500)  time: 0.7580  data: 0.2699  max mem: 41794
Epoch: [119]  [270/312]  eta: 0:00:32  lr: 0.002871  min_lr: 0.002871  loss: 3.8504 (3.5430)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1927  max mem: 41794
Epoch: [119]  [280/312]  eta: 0:00:24  lr: 0.002870  min_lr: 0.002870  loss: 3.4398 (3.5353)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.2050  max mem: 41794
Epoch: [119]  [290/312]  eta: 0:00:17  lr: 0.002869  min_lr: 0.002869  loss: 3.5321 (3.5388)  weight_decay: 0.0500 (0.0500)  time: 0.8180  data: 0.3241  max mem: 41794
Epoch: [119]  [300/312]  eta: 0:00:09  lr: 0.002869  min_lr: 0.002869  loss: 3.6815 (3.5404)  weight_decay: 0.0500 (0.0500)  time: 0.6088  data: 0.1195  max mem: 41794
Epoch: [119]  [310/312]  eta: 0:00:01  lr: 0.002868  min_lr: 0.002868  loss: 3.5894 (3.5340)  weight_decay: 0.0500 (0.0500)  time: 0.4669  data: 0.0001  max mem: 41794
Epoch: [119]  [311/312]  eta: 0:00:00  lr: 0.002868  min_lr: 0.002868  loss: 3.5894 (3.5363)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [119] Total time: 0:03:58 (0.7632 s / it)
Averaged stats: lr: 0.002868  min_lr: 0.002868  loss: 3.5894 (3.5257)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1078 (1.1078)  acc1: 76.8229 (76.8229)  acc5: 94.1406 (94.1406)  time: 8.2779  data: 8.0718  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5336 (1.4040)  acc1: 69.2708 (70.4480)  acc5: 89.5833 (90.5440)  time: 1.0702  data: 0.8970  max mem: 41794
Test: Total time: 0:00:09 (1.1021 s / it)
* Acc@1 70.770 Acc@5 90.250 loss 1.421
Accuracy of the model on the 50000 test images: 70.8%
Max accuracy: 71.18%
Epoch: [120]  [  0/312]  eta: 1:19:09  lr: 0.002868  min_lr: 0.002868  loss: 3.9589 (3.9589)  weight_decay: 0.0500 (0.0500)  time: 15.2234  data: 12.8274  max mem: 41794
Epoch: [120]  [ 10/312]  eta: 0:11:51  lr: 0.002867  min_lr: 0.002867  loss: 3.4811 (3.4724)  weight_decay: 0.0500 (0.0500)  time: 2.3569  data: 1.3505  max mem: 41794
Epoch: [120]  [ 20/312]  eta: 0:07:23  lr: 0.002867  min_lr: 0.002867  loss: 3.4811 (3.4616)  weight_decay: 0.0500 (0.0500)  time: 0.8351  data: 0.1356  max mem: 41794
Epoch: [120]  [ 30/312]  eta: 0:05:38  lr: 0.002866  min_lr: 0.002866  loss: 3.6162 (3.4311)  weight_decay: 0.0500 (0.0500)  time: 0.5643  data: 0.0419  max mem: 41794
Epoch: [120]  [ 40/312]  eta: 0:04:40  lr: 0.002865  min_lr: 0.002865  loss: 3.5114 (3.4383)  weight_decay: 0.0500 (0.0500)  time: 0.5191  data: 0.0122  max mem: 41794
Epoch: [120]  [ 50/312]  eta: 0:04:11  lr: 0.002865  min_lr: 0.002865  loss: 3.4566 (3.4393)  weight_decay: 0.0500 (0.0500)  time: 0.5884  data: 0.0295  max mem: 41794
Epoch: [120]  [ 60/312]  eta: 0:03:43  lr: 0.002864  min_lr: 0.002864  loss: 3.3894 (3.4167)  weight_decay: 0.0500 (0.0500)  time: 0.5885  data: 0.0254  max mem: 41794
Epoch: [120]  [ 70/312]  eta: 0:03:34  lr: 0.002864  min_lr: 0.002864  loss: 3.5047 (3.4359)  weight_decay: 0.0500 (0.0500)  time: 0.7002  data: 0.1006  max mem: 41794
Epoch: [120]  [ 80/312]  eta: 0:03:20  lr: 0.002863  min_lr: 0.002863  loss: 3.5047 (3.4298)  weight_decay: 0.0500 (0.0500)  time: 0.7979  data: 0.1249  max mem: 41794
Epoch: [120]  [ 90/312]  eta: 0:03:07  lr: 0.002862  min_lr: 0.002862  loss: 3.5176 (3.4428)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.0691  max mem: 41794
Epoch: [120]  [100/312]  eta: 0:02:59  lr: 0.002862  min_lr: 0.002862  loss: 3.6006 (3.4532)  weight_decay: 0.0500 (0.0500)  time: 0.7711  data: 0.1172  max mem: 41794
Epoch: [120]  [110/312]  eta: 0:02:44  lr: 0.002861  min_lr: 0.002861  loss: 3.7622 (3.4601)  weight_decay: 0.0500 (0.0500)  time: 0.6891  data: 0.0768  max mem: 41794
Epoch: [120]  [120/312]  eta: 0:02:36  lr: 0.002860  min_lr: 0.002860  loss: 3.6160 (3.4681)  weight_decay: 0.0500 (0.0500)  time: 0.6654  data: 0.0885  max mem: 41794
Epoch: [120]  [130/312]  eta: 0:02:27  lr: 0.002860  min_lr: 0.002860  loss: 3.6139 (3.4683)  weight_decay: 0.0500 (0.0500)  time: 0.7881  data: 0.1836  max mem: 41794
Epoch: [120]  [140/312]  eta: 0:02:15  lr: 0.002859  min_lr: 0.002859  loss: 3.6751 (3.4839)  weight_decay: 0.0500 (0.0500)  time: 0.6307  data: 0.0995  max mem: 41794
Epoch: [120]  [150/312]  eta: 0:02:09  lr: 0.002858  min_lr: 0.002858  loss: 3.7557 (3.4979)  weight_decay: 0.0500 (0.0500)  time: 0.7332  data: 0.0981  max mem: 41794
Epoch: [120]  [160/312]  eta: 0:02:00  lr: 0.002858  min_lr: 0.002858  loss: 3.6313 (3.4744)  weight_decay: 0.0500 (0.0500)  time: 0.8225  data: 0.1466  max mem: 41794
Epoch: [120]  [170/312]  eta: 0:01:51  lr: 0.002857  min_lr: 0.002857  loss: 3.6313 (3.4842)  weight_decay: 0.0500 (0.0500)  time: 0.6383  data: 0.0779  max mem: 41794
Epoch: [120]  [180/312]  eta: 0:01:44  lr: 0.002856  min_lr: 0.002856  loss: 3.6934 (3.4808)  weight_decay: 0.0500 (0.0500)  time: 0.7649  data: 0.1291  max mem: 41794
Epoch: [120]  [190/312]  eta: 0:01:34  lr: 0.002856  min_lr: 0.002856  loss: 3.5822 (3.4796)  weight_decay: 0.0500 (0.0500)  time: 0.7222  data: 0.1004  max mem: 41794
Epoch: [120]  [200/312]  eta: 0:01:26  lr: 0.002855  min_lr: 0.002855  loss: 3.7888 (3.4969)  weight_decay: 0.0500 (0.0500)  time: 0.6428  data: 0.0633  max mem: 41794
Epoch: [120]  [210/312]  eta: 0:01:19  lr: 0.002854  min_lr: 0.002854  loss: 3.8069 (3.5085)  weight_decay: 0.0500 (0.0500)  time: 0.8102  data: 0.1129  max mem: 41794
Epoch: [120]  [220/312]  eta: 0:01:10  lr: 0.002854  min_lr: 0.002854  loss: 3.6344 (3.5104)  weight_decay: 0.0500 (0.0500)  time: 0.7087  data: 0.0905  max mem: 41794
Epoch: [120]  [230/312]  eta: 0:01:03  lr: 0.002853  min_lr: 0.002853  loss: 3.5965 (3.5112)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.0995  max mem: 41794
Epoch: [120]  [240/312]  eta: 0:00:55  lr: 0.002852  min_lr: 0.002852  loss: 3.3394 (3.5028)  weight_decay: 0.0500 (0.0500)  time: 0.7540  data: 0.1002  max mem: 41794
Epoch: [120]  [250/312]  eta: 0:00:47  lr: 0.002852  min_lr: 0.002852  loss: 3.5085 (3.5120)  weight_decay: 0.0500 (0.0500)  time: 0.6759  data: 0.0841  max mem: 41794
Epoch: [120]  [260/312]  eta: 0:00:39  lr: 0.002851  min_lr: 0.002851  loss: 3.7263 (3.5162)  weight_decay: 0.0500 (0.0500)  time: 0.7656  data: 0.1149  max mem: 41794
Epoch: [120]  [270/312]  eta: 0:00:31  lr: 0.002851  min_lr: 0.002851  loss: 3.7137 (3.5149)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.0724  max mem: 41794
Epoch: [120]  [280/312]  eta: 0:00:24  lr: 0.002850  min_lr: 0.002850  loss: 3.6727 (3.5222)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.0604  max mem: 41794
Epoch: [120]  [290/312]  eta: 0:00:16  lr: 0.002849  min_lr: 0.002849  loss: 3.7137 (3.5235)  weight_decay: 0.0500 (0.0500)  time: 0.8481  data: 0.1035  max mem: 41794
Epoch: [120]  [300/312]  eta: 0:00:09  lr: 0.002849  min_lr: 0.002849  loss: 3.6173 (3.5203)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.0599  max mem: 41794
Epoch: [120]  [310/312]  eta: 0:00:01  lr: 0.002848  min_lr: 0.002848  loss: 3.3637 (3.5053)  weight_decay: 0.0500 (0.0500)  time: 0.4849  data: 0.0165  max mem: 41794
Epoch: [120]  [311/312]  eta: 0:00:00  lr: 0.002848  min_lr: 0.002848  loss: 3.0218 (3.5026)  weight_decay: 0.0500 (0.0500)  time: 0.4688  data: 0.0001  max mem: 41794
Epoch: [120] Total time: 0:03:54 (0.7528 s / it)
Averaged stats: lr: 0.002848  min_lr: 0.002848  loss: 3.0218 (3.5194)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.1041 (1.1041)  acc1: 75.3906 (75.3906)  acc5: 93.0990 (93.0990)  time: 8.4324  data: 8.2345  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3138 (1.3061)  acc1: 71.7448 (70.5280)  acc5: 90.8854 (90.6400)  time: 1.0926  data: 0.9150  max mem: 41794
Test: Total time: 0:00:10 (1.1376 s / it)
* Acc@1 70.722 Acc@5 90.432 loss 1.295
Accuracy of the model on the 50000 test images: 70.7%
Max accuracy: 71.18%
Epoch: [121]  [  0/312]  eta: 1:19:34  lr: 0.002848  min_lr: 0.002848  loss: 3.2380 (3.2380)  weight_decay: 0.0500 (0.0500)  time: 15.3022  data: 13.0808  max mem: 41794
Epoch: [121]  [ 10/312]  eta: 0:10:52  lr: 0.002847  min_lr: 0.002847  loss: 3.7464 (3.5948)  weight_decay: 0.0500 (0.0500)  time: 2.1614  data: 1.3382  max mem: 41794
Epoch: [121]  [ 20/312]  eta: 0:07:17  lr: 0.002846  min_lr: 0.002846  loss: 3.7464 (3.5554)  weight_decay: 0.0500 (0.0500)  time: 0.8076  data: 0.1714  max mem: 41794
Epoch: [121]  [ 30/312]  eta: 0:05:31  lr: 0.002846  min_lr: 0.002846  loss: 3.3351 (3.5233)  weight_decay: 0.0500 (0.0500)  time: 0.6352  data: 0.0905  max mem: 41794
Epoch: [121]  [ 40/312]  eta: 0:04:50  lr: 0.002845  min_lr: 0.002845  loss: 3.4854 (3.5074)  weight_decay: 0.0500 (0.0500)  time: 0.6187  data: 0.0257  max mem: 41794
Epoch: [121]  [ 50/312]  eta: 0:04:29  lr: 0.002845  min_lr: 0.002845  loss: 3.5698 (3.5402)  weight_decay: 0.0500 (0.0500)  time: 0.8024  data: 0.1096  max mem: 41794
Epoch: [121]  [ 60/312]  eta: 0:03:57  lr: 0.002844  min_lr: 0.002844  loss: 3.5465 (3.4912)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.0853  max mem: 41794
Epoch: [121]  [ 70/312]  eta: 0:03:45  lr: 0.002843  min_lr: 0.002843  loss: 3.4119 (3.4935)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.0979  max mem: 41794
Epoch: [121]  [ 80/312]  eta: 0:03:33  lr: 0.002843  min_lr: 0.002843  loss: 3.4119 (3.4755)  weight_decay: 0.0500 (0.0500)  time: 0.8549  data: 0.1797  max mem: 41794
Epoch: [121]  [ 90/312]  eta: 0:03:16  lr: 0.002842  min_lr: 0.002842  loss: 3.3619 (3.4632)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.1366  max mem: 41794
Epoch: [121]  [100/312]  eta: 0:03:07  lr: 0.002841  min_lr: 0.002841  loss: 3.5978 (3.4855)  weight_decay: 0.0500 (0.0500)  time: 0.7395  data: 0.1259  max mem: 41794
Epoch: [121]  [110/312]  eta: 0:02:52  lr: 0.002841  min_lr: 0.002841  loss: 3.7277 (3.4882)  weight_decay: 0.0500 (0.0500)  time: 0.6974  data: 0.0717  max mem: 41794
Epoch: [121]  [120/312]  eta: 0:02:44  lr: 0.002840  min_lr: 0.002840  loss: 3.6399 (3.4925)  weight_decay: 0.0500 (0.0500)  time: 0.6988  data: 0.0838  max mem: 41794
Epoch: [121]  [130/312]  eta: 0:02:36  lr: 0.002839  min_lr: 0.002839  loss: 3.8176 (3.5200)  weight_decay: 0.0500 (0.0500)  time: 0.8931  data: 0.1449  max mem: 41794
Epoch: [121]  [140/312]  eta: 0:02:23  lr: 0.002839  min_lr: 0.002839  loss: 3.7574 (3.5187)  weight_decay: 0.0500 (0.0500)  time: 0.7072  data: 0.0631  max mem: 41794
Epoch: [121]  [150/312]  eta: 0:02:14  lr: 0.002838  min_lr: 0.002838  loss: 3.6982 (3.5075)  weight_decay: 0.0500 (0.0500)  time: 0.6547  data: 0.0567  max mem: 41794
Epoch: [121]  [160/312]  eta: 0:02:04  lr: 0.002837  min_lr: 0.002837  loss: 3.6796 (3.5102)  weight_decay: 0.0500 (0.0500)  time: 0.7080  data: 0.0570  max mem: 41794
Epoch: [121]  [170/312]  eta: 0:01:54  lr: 0.002837  min_lr: 0.002837  loss: 3.6285 (3.5155)  weight_decay: 0.0500 (0.0500)  time: 0.6424  data: 0.0727  max mem: 41794
Epoch: [121]  [180/312]  eta: 0:01:46  lr: 0.002836  min_lr: 0.002836  loss: 3.6285 (3.5159)  weight_decay: 0.0500 (0.0500)  time: 0.7436  data: 0.1782  max mem: 41794
Epoch: [121]  [190/312]  eta: 0:01:36  lr: 0.002835  min_lr: 0.002835  loss: 3.5901 (3.5107)  weight_decay: 0.0500 (0.0500)  time: 0.6509  data: 0.1077  max mem: 41794
Epoch: [121]  [200/312]  eta: 0:01:29  lr: 0.002835  min_lr: 0.002835  loss: 3.1757 (3.4916)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.1068  max mem: 41794
Epoch: [121]  [210/312]  eta: 0:01:22  lr: 0.002834  min_lr: 0.002834  loss: 3.4936 (3.5039)  weight_decay: 0.0500 (0.0500)  time: 0.9125  data: 0.2124  max mem: 41794
Epoch: [121]  [220/312]  eta: 0:01:12  lr: 0.002833  min_lr: 0.002833  loss: 3.6027 (3.4947)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.1082  max mem: 41794
Epoch: [121]  [230/312]  eta: 0:01:05  lr: 0.002833  min_lr: 0.002833  loss: 3.3439 (3.4938)  weight_decay: 0.0500 (0.0500)  time: 0.6743  data: 0.0683  max mem: 41794
Epoch: [121]  [240/312]  eta: 0:00:56  lr: 0.002832  min_lr: 0.002832  loss: 3.5515 (3.4955)  weight_decay: 0.0500 (0.0500)  time: 0.7037  data: 0.0664  max mem: 41794
Epoch: [121]  [250/312]  eta: 0:00:48  lr: 0.002831  min_lr: 0.002831  loss: 3.6824 (3.4989)  weight_decay: 0.0500 (0.0500)  time: 0.6512  data: 0.0548  max mem: 41794
Epoch: [121]  [260/312]  eta: 0:00:40  lr: 0.002831  min_lr: 0.002831  loss: 3.4430 (3.4882)  weight_decay: 0.0500 (0.0500)  time: 0.7720  data: 0.1347  max mem: 41794
Epoch: [121]  [270/312]  eta: 0:00:32  lr: 0.002830  min_lr: 0.002830  loss: 3.5044 (3.4938)  weight_decay: 0.0500 (0.0500)  time: 0.6471  data: 0.0805  max mem: 41794
Epoch: [121]  [280/312]  eta: 0:00:24  lr: 0.002830  min_lr: 0.002830  loss: 3.6239 (3.4873)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0848  max mem: 41794
Epoch: [121]  [290/312]  eta: 0:00:17  lr: 0.002829  min_lr: 0.002829  loss: 3.4105 (3.4854)  weight_decay: 0.0500 (0.0500)  time: 0.8341  data: 0.1312  max mem: 41794
Epoch: [121]  [300/312]  eta: 0:00:09  lr: 0.002828  min_lr: 0.002828  loss: 3.4337 (3.4813)  weight_decay: 0.0500 (0.0500)  time: 0.6561  data: 0.0467  max mem: 41794
Epoch: [121]  [310/312]  eta: 0:00:01  lr: 0.002828  min_lr: 0.002828  loss: 3.4337 (3.4826)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [121]  [311/312]  eta: 0:00:00  lr: 0.002827  min_lr: 0.002827  loss: 3.4753 (3.4827)  weight_decay: 0.0500 (0.0500)  time: 0.4650  data: 0.0001  max mem: 41794
Epoch: [121] Total time: 0:03:57 (0.7616 s / it)
Averaged stats: lr: 0.002827  min_lr: 0.002827  loss: 3.4753 (3.4937)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.2422 (1.2422)  acc1: 78.7760 (78.7760)  acc5: 94.0104 (94.0104)  time: 8.3251  data: 8.1195  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5869 (1.4674)  acc1: 69.1406 (70.9440)  acc5: 89.1927 (90.4480)  time: 1.0763  data: 0.9022  max mem: 41794
Test: Total time: 0:00:10 (1.1127 s / it)
* Acc@1 70.892 Acc@5 90.430 loss 1.472
Accuracy of the model on the 50000 test images: 70.9%
Max accuracy: 71.18%
Epoch: [122]  [  0/312]  eta: 1:23:34  lr: 0.002827  min_lr: 0.002827  loss: 3.8627 (3.8627)  weight_decay: 0.0500 (0.0500)  time: 16.0732  data: 12.9828  max mem: 41794
Epoch: [122]  [ 10/312]  eta: 0:12:57  lr: 0.002827  min_lr: 0.002827  loss: 3.7544 (3.3770)  weight_decay: 0.0500 (0.0500)  time: 2.5735  data: 1.2112  max mem: 41794
Epoch: [122]  [ 20/312]  eta: 0:07:49  lr: 0.002826  min_lr: 0.002826  loss: 3.2944 (3.3476)  weight_decay: 0.0500 (0.0500)  time: 0.8835  data: 0.0399  max mem: 41794
Epoch: [122]  [ 30/312]  eta: 0:05:54  lr: 0.002825  min_lr: 0.002825  loss: 3.4573 (3.4254)  weight_decay: 0.0500 (0.0500)  time: 0.5308  data: 0.0266  max mem: 41794
Epoch: [122]  [ 40/312]  eta: 0:04:51  lr: 0.002825  min_lr: 0.002825  loss: 3.7636 (3.5271)  weight_decay: 0.0500 (0.0500)  time: 0.5070  data: 0.0059  max mem: 41794
Epoch: [122]  [ 50/312]  eta: 0:04:11  lr: 0.002824  min_lr: 0.002824  loss: 3.7682 (3.5283)  weight_decay: 0.0500 (0.0500)  time: 0.4984  data: 0.0026  max mem: 41794
Epoch: [122]  [ 60/312]  eta: 0:03:42  lr: 0.002823  min_lr: 0.002823  loss: 3.7682 (3.5530)  weight_decay: 0.0500 (0.0500)  time: 0.5005  data: 0.0017  max mem: 41794
Epoch: [122]  [ 70/312]  eta: 0:03:21  lr: 0.002823  min_lr: 0.002823  loss: 3.7454 (3.5503)  weight_decay: 0.0500 (0.0500)  time: 0.5140  data: 0.0016  max mem: 41794
Epoch: [122]  [ 80/312]  eta: 0:03:11  lr: 0.002822  min_lr: 0.002822  loss: 3.6968 (3.5538)  weight_decay: 0.0500 (0.0500)  time: 0.6439  data: 0.0350  max mem: 41794
Epoch: [122]  [ 90/312]  eta: 0:02:59  lr: 0.002822  min_lr: 0.002822  loss: 3.8158 (3.5805)  weight_decay: 0.0500 (0.0500)  time: 0.7253  data: 0.0484  max mem: 41794
Epoch: [122]  [100/312]  eta: 0:02:51  lr: 0.002821  min_lr: 0.002821  loss: 3.7231 (3.5778)  weight_decay: 0.0500 (0.0500)  time: 0.7492  data: 0.0230  max mem: 41794
Epoch: [122]  [110/312]  eta: 0:02:42  lr: 0.002820  min_lr: 0.002820  loss: 3.5083 (3.5734)  weight_decay: 0.0500 (0.0500)  time: 0.7677  data: 0.0191  max mem: 41794
Epoch: [122]  [120/312]  eta: 0:02:33  lr: 0.002820  min_lr: 0.002820  loss: 3.5213 (3.5527)  weight_decay: 0.0500 (0.0500)  time: 0.7647  data: 0.0593  max mem: 41794
Epoch: [122]  [130/312]  eta: 0:02:23  lr: 0.002819  min_lr: 0.002819  loss: 3.5860 (3.5527)  weight_decay: 0.0500 (0.0500)  time: 0.7188  data: 0.0565  max mem: 41794
Epoch: [122]  [140/312]  eta: 0:02:15  lr: 0.002818  min_lr: 0.002818  loss: 3.6484 (3.5533)  weight_decay: 0.0500 (0.0500)  time: 0.6969  data: 0.0523  max mem: 41794
Epoch: [122]  [150/312]  eta: 0:02:07  lr: 0.002818  min_lr: 0.002818  loss: 3.7444 (3.5634)  weight_decay: 0.0500 (0.0500)  time: 0.7487  data: 0.0631  max mem: 41794
Epoch: [122]  [160/312]  eta: 0:01:58  lr: 0.002817  min_lr: 0.002817  loss: 3.7131 (3.5511)  weight_decay: 0.0500 (0.0500)  time: 0.7218  data: 0.0451  max mem: 41794
Epoch: [122]  [170/312]  eta: 0:01:50  lr: 0.002816  min_lr: 0.002816  loss: 3.7079 (3.5603)  weight_decay: 0.0500 (0.0500)  time: 0.7257  data: 0.0755  max mem: 41794
Epoch: [122]  [180/312]  eta: 0:01:42  lr: 0.002816  min_lr: 0.002816  loss: 3.6820 (3.5597)  weight_decay: 0.0500 (0.0500)  time: 0.7807  data: 0.0975  max mem: 41794
Epoch: [122]  [190/312]  eta: 0:01:34  lr: 0.002815  min_lr: 0.002815  loss: 3.6736 (3.5598)  weight_decay: 0.0500 (0.0500)  time: 0.7668  data: 0.0725  max mem: 41794
Epoch: [122]  [200/312]  eta: 0:01:26  lr: 0.002814  min_lr: 0.002814  loss: 3.6199 (3.5615)  weight_decay: 0.0500 (0.0500)  time: 0.6994  data: 0.0369  max mem: 41794
Epoch: [122]  [210/312]  eta: 0:01:18  lr: 0.002814  min_lr: 0.002814  loss: 3.7477 (3.5678)  weight_decay: 0.0500 (0.0500)  time: 0.7235  data: 0.0293  max mem: 41794
Epoch: [122]  [220/312]  eta: 0:01:10  lr: 0.002813  min_lr: 0.002813  loss: 3.7484 (3.5693)  weight_decay: 0.0500 (0.0500)  time: 0.7188  data: 0.0487  max mem: 41794
Epoch: [122]  [230/312]  eta: 0:01:03  lr: 0.002812  min_lr: 0.002812  loss: 3.4868 (3.5606)  weight_decay: 0.0500 (0.0500)  time: 0.7506  data: 0.0513  max mem: 41794
Epoch: [122]  [240/312]  eta: 0:00:54  lr: 0.002812  min_lr: 0.002812  loss: 3.3405 (3.5514)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.0598  max mem: 41794
Epoch: [122]  [250/312]  eta: 0:00:47  lr: 0.002811  min_lr: 0.002811  loss: 3.3413 (3.5450)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.0582  max mem: 41794
Epoch: [122]  [260/312]  eta: 0:00:39  lr: 0.002810  min_lr: 0.002810  loss: 3.4144 (3.5357)  weight_decay: 0.0500 (0.0500)  time: 0.7852  data: 0.0851  max mem: 41794
Epoch: [122]  [270/312]  eta: 0:00:31  lr: 0.002810  min_lr: 0.002810  loss: 3.4144 (3.5290)  weight_decay: 0.0500 (0.0500)  time: 0.7283  data: 0.1203  max mem: 41794
Epoch: [122]  [280/312]  eta: 0:00:24  lr: 0.002809  min_lr: 0.002809  loss: 3.4710 (3.5260)  weight_decay: 0.0500 (0.0500)  time: 0.6931  data: 0.0638  max mem: 41794
Epoch: [122]  [290/312]  eta: 0:00:16  lr: 0.002808  min_lr: 0.002808  loss: 3.5814 (3.5271)  weight_decay: 0.0500 (0.0500)  time: 0.7960  data: 0.0596  max mem: 41794
Epoch: [122]  [300/312]  eta: 0:00:09  lr: 0.002808  min_lr: 0.002808  loss: 3.5250 (3.5281)  weight_decay: 0.0500 (0.0500)  time: 0.7223  data: 0.0976  max mem: 41794
Epoch: [122]  [310/312]  eta: 0:00:01  lr: 0.002807  min_lr: 0.002807  loss: 3.5250 (3.5294)  weight_decay: 0.0500 (0.0500)  time: 0.5357  data: 0.0499  max mem: 41794
Epoch: [122]  [311/312]  eta: 0:00:00  lr: 0.002807  min_lr: 0.002807  loss: 3.4587 (3.5269)  weight_decay: 0.0500 (0.0500)  time: 0.5152  data: 0.0302  max mem: 41794
Epoch: [122] Total time: 0:03:54 (0.7531 s / it)
Averaged stats: lr: 0.002807  min_lr: 0.002807  loss: 3.4587 (3.5347)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 1.3090 (1.3090)  acc1: 78.9062 (78.9062)  acc5: 93.7500 (93.7500)  time: 7.7845  data: 7.5740  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6627 (1.5346)  acc1: 71.2240 (71.4560)  acc5: 90.2344 (90.8320)  time: 1.0482  data: 0.8768  max mem: 41794
Test: Total time: 0:00:09 (1.0582 s / it)
* Acc@1 71.866 Acc@5 91.064 loss 1.531
Accuracy of the model on the 50000 test images: 71.9%
Max accuracy: 71.87%
Epoch: [123]  [  0/312]  eta: 1:22:50  lr: 0.002807  min_lr: 0.002807  loss: 3.0725 (3.0725)  weight_decay: 0.0500 (0.0500)  time: 15.9313  data: 13.0514  max mem: 41794
Epoch: [123]  [ 10/312]  eta: 0:11:10  lr: 0.002806  min_lr: 0.002806  loss: 3.7576 (3.4295)  weight_decay: 0.0500 (0.0500)  time: 2.2202  data: 1.3000  max mem: 41794
Epoch: [123]  [ 20/312]  eta: 0:07:09  lr: 0.002806  min_lr: 0.002806  loss: 3.7576 (3.5214)  weight_decay: 0.0500 (0.0500)  time: 0.7486  data: 0.1354  max mem: 41794
Epoch: [123]  [ 30/312]  eta: 0:05:28  lr: 0.002805  min_lr: 0.002805  loss: 3.6998 (3.4214)  weight_decay: 0.0500 (0.0500)  time: 0.5819  data: 0.0734  max mem: 41794
Epoch: [123]  [ 40/312]  eta: 0:04:44  lr: 0.002804  min_lr: 0.002804  loss: 3.2852 (3.4008)  weight_decay: 0.0500 (0.0500)  time: 0.5993  data: 0.0989  max mem: 41794
Epoch: [123]  [ 50/312]  eta: 0:04:18  lr: 0.002804  min_lr: 0.002804  loss: 3.5565 (3.4488)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.2125  max mem: 41794
Epoch: [123]  [ 60/312]  eta: 0:03:47  lr: 0.002803  min_lr: 0.002803  loss: 3.7840 (3.4778)  weight_decay: 0.0500 (0.0500)  time: 0.6127  data: 0.1143  max mem: 41794
Epoch: [123]  [ 70/312]  eta: 0:03:37  lr: 0.002802  min_lr: 0.002802  loss: 3.6451 (3.4558)  weight_decay: 0.0500 (0.0500)  time: 0.6786  data: 0.1863  max mem: 41794
Epoch: [123]  [ 80/312]  eta: 0:03:24  lr: 0.002802  min_lr: 0.002802  loss: 3.5020 (3.4524)  weight_decay: 0.0500 (0.0500)  time: 0.8054  data: 0.3116  max mem: 41794
Epoch: [123]  [ 90/312]  eta: 0:03:07  lr: 0.002801  min_lr: 0.002801  loss: 3.5020 (3.4557)  weight_decay: 0.0500 (0.0500)  time: 0.6607  data: 0.1687  max mem: 41794
Epoch: [123]  [100/312]  eta: 0:03:01  lr: 0.002800  min_lr: 0.002800  loss: 3.5281 (3.4686)  weight_decay: 0.0500 (0.0500)  time: 0.7553  data: 0.2620  max mem: 41794
Epoch: [123]  [110/312]  eta: 0:02:46  lr: 0.002800  min_lr: 0.002800  loss: 3.4558 (3.4584)  weight_decay: 0.0500 (0.0500)  time: 0.7115  data: 0.2193  max mem: 41794
Epoch: [123]  [120/312]  eta: 0:02:38  lr: 0.002799  min_lr: 0.002799  loss: 3.3226 (3.4528)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.1843  max mem: 41794
Epoch: [123]  [130/312]  eta: 0:02:29  lr: 0.002798  min_lr: 0.002798  loss: 3.7546 (3.4721)  weight_decay: 0.0500 (0.0500)  time: 0.8105  data: 0.2823  max mem: 41794
Epoch: [123]  [140/312]  eta: 0:02:19  lr: 0.002798  min_lr: 0.002798  loss: 3.8454 (3.4861)  weight_decay: 0.0500 (0.0500)  time: 0.7026  data: 0.1778  max mem: 41794
Epoch: [123]  [150/312]  eta: 0:02:12  lr: 0.002797  min_lr: 0.002797  loss: 3.5921 (3.4862)  weight_decay: 0.0500 (0.0500)  time: 0.7898  data: 0.2097  max mem: 41794
Epoch: [123]  [160/312]  eta: 0:02:02  lr: 0.002796  min_lr: 0.002796  loss: 3.4747 (3.4809)  weight_decay: 0.0500 (0.0500)  time: 0.7951  data: 0.1406  max mem: 41794
Epoch: [123]  [170/312]  eta: 0:01:53  lr: 0.002796  min_lr: 0.002796  loss: 3.2583 (3.4710)  weight_decay: 0.0500 (0.0500)  time: 0.6505  data: 0.0824  max mem: 41794
Epoch: [123]  [180/312]  eta: 0:01:44  lr: 0.002795  min_lr: 0.002795  loss: 3.3786 (3.4707)  weight_decay: 0.0500 (0.0500)  time: 0.6981  data: 0.1525  max mem: 41794
Epoch: [123]  [190/312]  eta: 0:01:35  lr: 0.002794  min_lr: 0.002794  loss: 3.4166 (3.4658)  weight_decay: 0.0500 (0.0500)  time: 0.6348  data: 0.0819  max mem: 41794
Epoch: [123]  [200/312]  eta: 0:01:28  lr: 0.002794  min_lr: 0.002794  loss: 3.6264 (3.4731)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1196  max mem: 41794
Epoch: [123]  [210/312]  eta: 0:01:19  lr: 0.002793  min_lr: 0.002793  loss: 3.6572 (3.4648)  weight_decay: 0.0500 (0.0500)  time: 0.8214  data: 0.1499  max mem: 41794
Epoch: [123]  [220/312]  eta: 0:01:11  lr: 0.002792  min_lr: 0.002792  loss: 3.5345 (3.4629)  weight_decay: 0.0500 (0.0500)  time: 0.6709  data: 0.0829  max mem: 41794
Epoch: [123]  [230/312]  eta: 0:01:03  lr: 0.002792  min_lr: 0.002792  loss: 3.6683 (3.4663)  weight_decay: 0.0500 (0.0500)  time: 0.7444  data: 0.1334  max mem: 41794
Epoch: [123]  [240/312]  eta: 0:00:55  lr: 0.002791  min_lr: 0.002791  loss: 3.7395 (3.4664)  weight_decay: 0.0500 (0.0500)  time: 0.7952  data: 0.0985  max mem: 41794
Epoch: [123]  [250/312]  eta: 0:00:47  lr: 0.002790  min_lr: 0.002790  loss: 3.6645 (3.4688)  weight_decay: 0.0500 (0.0500)  time: 0.6747  data: 0.0955  max mem: 41794
Epoch: [123]  [260/312]  eta: 0:00:40  lr: 0.002790  min_lr: 0.002790  loss: 3.6645 (3.4727)  weight_decay: 0.0500 (0.0500)  time: 0.7490  data: 0.1472  max mem: 41794
Epoch: [123]  [270/312]  eta: 0:00:32  lr: 0.002789  min_lr: 0.002789  loss: 3.5375 (3.4675)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.0681  max mem: 41794
Epoch: [123]  [280/312]  eta: 0:00:24  lr: 0.002788  min_lr: 0.002788  loss: 3.5199 (3.4700)  weight_decay: 0.0500 (0.0500)  time: 0.6870  data: 0.0653  max mem: 41794
Epoch: [123]  [290/312]  eta: 0:00:16  lr: 0.002788  min_lr: 0.002788  loss: 3.4782 (3.4668)  weight_decay: 0.0500 (0.0500)  time: 0.8220  data: 0.0650  max mem: 41794
Epoch: [123]  [300/312]  eta: 0:00:09  lr: 0.002787  min_lr: 0.002787  loss: 3.5905 (3.4678)  weight_decay: 0.0500 (0.0500)  time: 0.6709  data: 0.0513  max mem: 41794
Epoch: [123]  [310/312]  eta: 0:00:01  lr: 0.002786  min_lr: 0.002786  loss: 3.5905 (3.4699)  weight_decay: 0.0500 (0.0500)  time: 0.5176  data: 0.0512  max mem: 41794
Epoch: [123]  [311/312]  eta: 0:00:00  lr: 0.002786  min_lr: 0.002786  loss: 3.5917 (3.4703)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0001  max mem: 41794
Epoch: [123] Total time: 0:03:56 (0.7586 s / it)
Averaged stats: lr: 0.002786  min_lr: 0.002786  loss: 3.5917 (3.5022)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.3000 (1.3000)  acc1: 77.0833 (77.0833)  acc5: 93.8802 (93.8802)  time: 8.1681  data: 7.9560  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5820 (1.5401)  acc1: 71.4844 (70.9600)  acc5: 90.5660 (90.3840)  time: 1.0832  data: 0.9034  max mem: 41794
Test: Total time: 0:00:09 (1.0939 s / it)
* Acc@1 71.170 Acc@5 90.494 loss 1.531
Accuracy of the model on the 50000 test images: 71.2%
Max accuracy: 71.87%
Epoch: [124]  [  0/312]  eta: 1:18:12  lr: 0.002786  min_lr: 0.002786  loss: 2.6507 (2.6507)  weight_decay: 0.0500 (0.0500)  time: 15.0396  data: 12.4467  max mem: 41794
Epoch: [124]  [ 10/312]  eta: 0:10:43  lr: 0.002786  min_lr: 0.002786  loss: 3.4244 (3.4511)  weight_decay: 0.0500 (0.0500)  time: 2.1307  data: 1.2596  max mem: 41794
Epoch: [124]  [ 20/312]  eta: 0:07:30  lr: 0.002785  min_lr: 0.002785  loss: 3.4992 (3.4979)  weight_decay: 0.0500 (0.0500)  time: 0.8691  data: 0.1430  max mem: 41794
Epoch: [124]  [ 30/312]  eta: 0:05:44  lr: 0.002784  min_lr: 0.002784  loss: 3.6596 (3.4986)  weight_decay: 0.0500 (0.0500)  time: 0.7201  data: 0.0779  max mem: 41794
Epoch: [124]  [ 40/312]  eta: 0:04:55  lr: 0.002784  min_lr: 0.002784  loss: 3.6518 (3.4762)  weight_decay: 0.0500 (0.0500)  time: 0.6031  data: 0.0074  max mem: 41794
Epoch: [124]  [ 50/312]  eta: 0:04:25  lr: 0.002783  min_lr: 0.002783  loss: 3.6761 (3.5416)  weight_decay: 0.0500 (0.0500)  time: 0.6881  data: 0.0317  max mem: 41794
Epoch: [124]  [ 60/312]  eta: 0:03:57  lr: 0.002782  min_lr: 0.002782  loss: 3.7093 (3.5383)  weight_decay: 0.0500 (0.0500)  time: 0.6499  data: 0.0543  max mem: 41794
Epoch: [124]  [ 70/312]  eta: 0:03:47  lr: 0.002782  min_lr: 0.002782  loss: 3.5639 (3.5243)  weight_decay: 0.0500 (0.0500)  time: 0.7518  data: 0.1072  max mem: 41794
Epoch: [124]  [ 80/312]  eta: 0:03:34  lr: 0.002781  min_lr: 0.002781  loss: 3.3012 (3.4801)  weight_decay: 0.0500 (0.0500)  time: 0.8731  data: 0.0829  max mem: 41794
Epoch: [124]  [ 90/312]  eta: 0:03:18  lr: 0.002780  min_lr: 0.002780  loss: 3.3748 (3.4788)  weight_decay: 0.0500 (0.0500)  time: 0.7298  data: 0.0708  max mem: 41794
Epoch: [124]  [100/312]  eta: 0:03:10  lr: 0.002780  min_lr: 0.002780  loss: 3.5793 (3.4725)  weight_decay: 0.0500 (0.0500)  time: 0.7807  data: 0.1266  max mem: 41794
Epoch: [124]  [110/312]  eta: 0:02:54  lr: 0.002779  min_lr: 0.002779  loss: 3.6920 (3.4836)  weight_decay: 0.0500 (0.0500)  time: 0.7250  data: 0.0565  max mem: 41794
Epoch: [124]  [120/312]  eta: 0:02:44  lr: 0.002778  min_lr: 0.002778  loss: 3.7655 (3.4785)  weight_decay: 0.0500 (0.0500)  time: 0.6473  data: 0.0290  max mem: 41794
Epoch: [124]  [130/312]  eta: 0:02:35  lr: 0.002778  min_lr: 0.002778  loss: 3.2095 (3.4540)  weight_decay: 0.0500 (0.0500)  time: 0.8018  data: 0.0577  max mem: 41794
Epoch: [124]  [140/312]  eta: 0:02:23  lr: 0.002777  min_lr: 0.002777  loss: 3.0432 (3.4279)  weight_decay: 0.0500 (0.0500)  time: 0.7127  data: 0.0826  max mem: 41794
Epoch: [124]  [150/312]  eta: 0:02:14  lr: 0.002776  min_lr: 0.002776  loss: 3.0939 (3.4251)  weight_decay: 0.0500 (0.0500)  time: 0.6694  data: 0.0723  max mem: 41794
Epoch: [124]  [160/312]  eta: 0:02:05  lr: 0.002776  min_lr: 0.002776  loss: 3.4275 (3.4190)  weight_decay: 0.0500 (0.0500)  time: 0.7644  data: 0.0397  max mem: 41794
Epoch: [124]  [170/312]  eta: 0:01:55  lr: 0.002775  min_lr: 0.002775  loss: 3.4489 (3.4125)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.0615  max mem: 41794
Epoch: [124]  [180/312]  eta: 0:01:47  lr: 0.002774  min_lr: 0.002774  loss: 3.6830 (3.4285)  weight_decay: 0.0500 (0.0500)  time: 0.7295  data: 0.1203  max mem: 41794
Epoch: [124]  [190/312]  eta: 0:01:37  lr: 0.002774  min_lr: 0.002774  loss: 3.8137 (3.4407)  weight_decay: 0.0500 (0.0500)  time: 0.6964  data: 0.0800  max mem: 41794
Epoch: [124]  [200/312]  eta: 0:01:29  lr: 0.002773  min_lr: 0.002773  loss: 3.8137 (3.4542)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.0690  max mem: 41794
Epoch: [124]  [210/312]  eta: 0:01:21  lr: 0.002772  min_lr: 0.002772  loss: 3.8396 (3.4672)  weight_decay: 0.0500 (0.0500)  time: 0.8061  data: 0.1201  max mem: 41794
Epoch: [124]  [220/312]  eta: 0:01:12  lr: 0.002772  min_lr: 0.002772  loss: 3.6469 (3.4781)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.0767  max mem: 41794
Epoch: [124]  [230/312]  eta: 0:01:04  lr: 0.002771  min_lr: 0.002771  loss: 3.6469 (3.4795)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.1289  max mem: 41794
Epoch: [124]  [240/312]  eta: 0:00:56  lr: 0.002770  min_lr: 0.002770  loss: 3.5415 (3.4722)  weight_decay: 0.0500 (0.0500)  time: 0.8179  data: 0.1745  max mem: 41794
Epoch: [124]  [250/312]  eta: 0:00:48  lr: 0.002770  min_lr: 0.002770  loss: 3.2279 (3.4686)  weight_decay: 0.0500 (0.0500)  time: 0.6740  data: 0.1143  max mem: 41794
Epoch: [124]  [260/312]  eta: 0:00:40  lr: 0.002769  min_lr: 0.002769  loss: 3.4363 (3.4683)  weight_decay: 0.0500 (0.0500)  time: 0.7495  data: 0.1697  max mem: 41794
Epoch: [124]  [270/312]  eta: 0:00:32  lr: 0.002768  min_lr: 0.002768  loss: 3.9260 (3.4858)  weight_decay: 0.0500 (0.0500)  time: 0.7101  data: 0.1268  max mem: 41794
Epoch: [124]  [280/312]  eta: 0:00:24  lr: 0.002768  min_lr: 0.002768  loss: 3.8252 (3.4775)  weight_decay: 0.0500 (0.0500)  time: 0.6592  data: 0.1002  max mem: 41794
Epoch: [124]  [290/312]  eta: 0:00:17  lr: 0.002767  min_lr: 0.002767  loss: 3.3236 (3.4792)  weight_decay: 0.0500 (0.0500)  time: 0.8259  data: 0.1746  max mem: 41794
Epoch: [124]  [300/312]  eta: 0:00:09  lr: 0.002766  min_lr: 0.002766  loss: 3.7590 (3.4878)  weight_decay: 0.0500 (0.0500)  time: 0.6492  data: 0.0748  max mem: 41794
Epoch: [124]  [310/312]  eta: 0:00:01  lr: 0.002766  min_lr: 0.002766  loss: 3.5652 (3.4859)  weight_decay: 0.0500 (0.0500)  time: 0.4628  data: 0.0001  max mem: 41794
Epoch: [124]  [311/312]  eta: 0:00:00  lr: 0.002766  min_lr: 0.002766  loss: 3.4944 (3.4848)  weight_decay: 0.0500 (0.0500)  time: 0.4629  data: 0.0001  max mem: 41794
Epoch: [124] Total time: 0:03:58 (0.7657 s / it)
Averaged stats: lr: 0.002766  min_lr: 0.002766  loss: 3.4944 (3.5022)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.0644 (1.0644)  acc1: 78.3854 (78.3854)  acc5: 93.7500 (93.7500)  time: 8.4312  data: 8.2238  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4807 (1.4255)  acc1: 70.3125 (71.0560)  acc5: 90.3646 (90.9120)  time: 1.0980  data: 0.9138  max mem: 41794
Test: Total time: 0:00:10 (1.1352 s / it)
* Acc@1 71.290 Acc@5 90.850 loss 1.425
Accuracy of the model on the 50000 test images: 71.3%
Max accuracy: 71.87%
Epoch: [125]  [  0/312]  eta: 1:23:42  lr: 0.002766  min_lr: 0.002766  loss: 3.9021 (3.9021)  weight_decay: 0.0500 (0.0500)  time: 16.0972  data: 14.0032  max mem: 41794
Epoch: [125]  [ 10/312]  eta: 0:10:50  lr: 0.002765  min_lr: 0.002765  loss: 3.4329 (3.3936)  weight_decay: 0.0500 (0.0500)  time: 2.1541  data: 1.3501  max mem: 41794
Epoch: [125]  [ 20/312]  eta: 0:07:27  lr: 0.002764  min_lr: 0.002764  loss: 3.1872 (3.3038)  weight_decay: 0.0500 (0.0500)  time: 0.8047  data: 0.1267  max mem: 41794
Epoch: [125]  [ 30/312]  eta: 0:05:42  lr: 0.002764  min_lr: 0.002764  loss: 3.5198 (3.3828)  weight_decay: 0.0500 (0.0500)  time: 0.6972  data: 0.0941  max mem: 41794
Epoch: [125]  [ 40/312]  eta: 0:04:52  lr: 0.002763  min_lr: 0.002763  loss: 3.6483 (3.4034)  weight_decay: 0.0500 (0.0500)  time: 0.5949  data: 0.0173  max mem: 41794
Epoch: [125]  [ 50/312]  eta: 0:04:28  lr: 0.002762  min_lr: 0.002762  loss: 3.7626 (3.4937)  weight_decay: 0.0500 (0.0500)  time: 0.7305  data: 0.1362  max mem: 41794
Epoch: [125]  [ 60/312]  eta: 0:03:56  lr: 0.002762  min_lr: 0.002762  loss: 3.6962 (3.4736)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.1291  max mem: 41794
Epoch: [125]  [ 70/312]  eta: 0:03:45  lr: 0.002761  min_lr: 0.002761  loss: 3.6943 (3.5003)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.1351  max mem: 41794
Epoch: [125]  [ 80/312]  eta: 0:03:31  lr: 0.002760  min_lr: 0.002760  loss: 3.7136 (3.5266)  weight_decay: 0.0500 (0.0500)  time: 0.8312  data: 0.1902  max mem: 41794
Epoch: [125]  [ 90/312]  eta: 0:03:15  lr: 0.002760  min_lr: 0.002760  loss: 3.7464 (3.5401)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.1199  max mem: 41794
Epoch: [125]  [100/312]  eta: 0:03:06  lr: 0.002759  min_lr: 0.002759  loss: 3.6481 (3.5202)  weight_decay: 0.0500 (0.0500)  time: 0.7426  data: 0.1578  max mem: 41794
Epoch: [125]  [110/312]  eta: 0:02:50  lr: 0.002758  min_lr: 0.002758  loss: 3.6481 (3.5400)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.0937  max mem: 41794
Epoch: [125]  [120/312]  eta: 0:02:42  lr: 0.002758  min_lr: 0.002758  loss: 3.7398 (3.5393)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.0602  max mem: 41794
Epoch: [125]  [130/312]  eta: 0:02:34  lr: 0.002757  min_lr: 0.002757  loss: 3.6746 (3.5339)  weight_decay: 0.0500 (0.0500)  time: 0.8583  data: 0.1221  max mem: 41794
Epoch: [125]  [140/312]  eta: 0:02:22  lr: 0.002756  min_lr: 0.002756  loss: 3.5720 (3.5218)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.0750  max mem: 41794
Epoch: [125]  [150/312]  eta: 0:02:13  lr: 0.002756  min_lr: 0.002756  loss: 3.2932 (3.5053)  weight_decay: 0.0500 (0.0500)  time: 0.6666  data: 0.0735  max mem: 41794
Epoch: [125]  [160/312]  eta: 0:02:04  lr: 0.002755  min_lr: 0.002755  loss: 3.5276 (3.4990)  weight_decay: 0.0500 (0.0500)  time: 0.7695  data: 0.1153  max mem: 41794
Epoch: [125]  [170/312]  eta: 0:01:54  lr: 0.002754  min_lr: 0.002754  loss: 3.4308 (3.4974)  weight_decay: 0.0500 (0.0500)  time: 0.7005  data: 0.0865  max mem: 41794
Epoch: [125]  [180/312]  eta: 0:01:47  lr: 0.002754  min_lr: 0.002754  loss: 3.7006 (3.5182)  weight_decay: 0.0500 (0.0500)  time: 0.7988  data: 0.0928  max mem: 41794
Epoch: [125]  [190/312]  eta: 0:01:37  lr: 0.002753  min_lr: 0.002753  loss: 3.5492 (3.5160)  weight_decay: 0.0500 (0.0500)  time: 0.7369  data: 0.0659  max mem: 41794
Epoch: [125]  [200/312]  eta: 0:01:29  lr: 0.002752  min_lr: 0.002752  loss: 3.5297 (3.5235)  weight_decay: 0.0500 (0.0500)  time: 0.6157  data: 0.0371  max mem: 41794
Epoch: [125]  [210/312]  eta: 0:01:20  lr: 0.002752  min_lr: 0.002752  loss: 3.6617 (3.5210)  weight_decay: 0.0500 (0.0500)  time: 0.7161  data: 0.0646  max mem: 41794
Epoch: [125]  [220/312]  eta: 0:01:12  lr: 0.002751  min_lr: 0.002751  loss: 3.6617 (3.5245)  weight_decay: 0.0500 (0.0500)  time: 0.6390  data: 0.0565  max mem: 41794
Epoch: [125]  [230/312]  eta: 0:01:04  lr: 0.002750  min_lr: 0.002750  loss: 3.6299 (3.5239)  weight_decay: 0.0500 (0.0500)  time: 0.7298  data: 0.1160  max mem: 41794
Epoch: [125]  [240/312]  eta: 0:00:56  lr: 0.002750  min_lr: 0.002750  loss: 3.5899 (3.5190)  weight_decay: 0.0500 (0.0500)  time: 0.7385  data: 0.1385  max mem: 41794
Epoch: [125]  [250/312]  eta: 0:00:48  lr: 0.002749  min_lr: 0.002749  loss: 3.4965 (3.5149)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.0896  max mem: 41794
Epoch: [125]  [260/312]  eta: 0:00:40  lr: 0.002748  min_lr: 0.002748  loss: 3.7521 (3.5236)  weight_decay: 0.0500 (0.0500)  time: 0.8241  data: 0.1128  max mem: 41794
Epoch: [125]  [270/312]  eta: 0:00:32  lr: 0.002748  min_lr: 0.002748  loss: 3.6023 (3.5186)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.0700  max mem: 41794
Epoch: [125]  [280/312]  eta: 0:00:24  lr: 0.002747  min_lr: 0.002747  loss: 3.4621 (3.5169)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.0664  max mem: 41794
Epoch: [125]  [290/312]  eta: 0:00:16  lr: 0.002746  min_lr: 0.002746  loss: 3.3408 (3.5106)  weight_decay: 0.0500 (0.0500)  time: 0.7614  data: 0.0664  max mem: 41794
Epoch: [125]  [300/312]  eta: 0:00:09  lr: 0.002746  min_lr: 0.002746  loss: 3.2431 (3.5049)  weight_decay: 0.0500 (0.0500)  time: 0.5823  data: 0.0249  max mem: 41794
Epoch: [125]  [310/312]  eta: 0:00:01  lr: 0.002745  min_lr: 0.002745  loss: 3.4809 (3.5126)  weight_decay: 0.0500 (0.0500)  time: 0.4894  data: 0.0247  max mem: 41794
Epoch: [125]  [311/312]  eta: 0:00:00  lr: 0.002745  min_lr: 0.002745  loss: 3.4809 (3.5099)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [125] Total time: 0:03:56 (0.7594 s / it)
Averaged stats: lr: 0.002745  min_lr: 0.002745  loss: 3.4809 (3.5269)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:21  loss: 1.2876 (1.2876)  acc1: 77.0833 (77.0833)  acc5: 92.5781 (92.5781)  time: 9.0370  data: 8.8265  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5661 (1.5040)  acc1: 68.4896 (70.8640)  acc5: 90.2344 (90.5600)  time: 1.1521  data: 0.9808  max mem: 41794
Test: Total time: 0:00:10 (1.1936 s / it)
* Acc@1 71.098 Acc@5 90.822 loss 1.499
Accuracy of the model on the 50000 test images: 71.1%
Max accuracy: 71.87%
Epoch: [126]  [  0/312]  eta: 1:23:38  lr: 0.002745  min_lr: 0.002745  loss: 3.3794 (3.3794)  weight_decay: 0.0500 (0.0500)  time: 16.0861  data: 13.1075  max mem: 41794
Epoch: [126]  [ 10/312]  eta: 0:11:22  lr: 0.002744  min_lr: 0.002744  loss: 3.5893 (3.6146)  weight_decay: 0.0500 (0.0500)  time: 2.2615  data: 1.3215  max mem: 41794
Epoch: [126]  [ 20/312]  eta: 0:07:37  lr: 0.002744  min_lr: 0.002744  loss: 3.4101 (3.3970)  weight_decay: 0.0500 (0.0500)  time: 0.8402  data: 0.1370  max mem: 41794
Epoch: [126]  [ 30/312]  eta: 0:05:49  lr: 0.002743  min_lr: 0.002743  loss: 3.1862 (3.3634)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.0786  max mem: 41794
Epoch: [126]  [ 40/312]  eta: 0:04:48  lr: 0.002742  min_lr: 0.002742  loss: 3.4767 (3.4449)  weight_decay: 0.0500 (0.0500)  time: 0.5323  data: 0.0161  max mem: 41794
Epoch: [126]  [ 50/312]  eta: 0:04:23  lr: 0.002742  min_lr: 0.002742  loss: 3.7638 (3.4806)  weight_decay: 0.0500 (0.0500)  time: 0.6417  data: 0.0262  max mem: 41794
Epoch: [126]  [ 60/312]  eta: 0:03:54  lr: 0.002741  min_lr: 0.002741  loss: 3.5064 (3.4649)  weight_decay: 0.0500 (0.0500)  time: 0.6603  data: 0.0552  max mem: 41794
Epoch: [126]  [ 70/312]  eta: 0:03:42  lr: 0.002740  min_lr: 0.002740  loss: 3.6466 (3.4758)  weight_decay: 0.0500 (0.0500)  time: 0.7067  data: 0.1066  max mem: 41794
Epoch: [126]  [ 80/312]  eta: 0:03:27  lr: 0.002740  min_lr: 0.002740  loss: 3.7071 (3.4695)  weight_decay: 0.0500 (0.0500)  time: 0.7857  data: 0.0794  max mem: 41794
Epoch: [126]  [ 90/312]  eta: 0:03:13  lr: 0.002739  min_lr: 0.002739  loss: 3.8144 (3.5182)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.0559  max mem: 41794
Epoch: [126]  [100/312]  eta: 0:03:02  lr: 0.002738  min_lr: 0.002738  loss: 3.8281 (3.5089)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.0913  max mem: 41794
Epoch: [126]  [110/312]  eta: 0:02:47  lr: 0.002738  min_lr: 0.002738  loss: 3.6760 (3.5219)  weight_decay: 0.0500 (0.0500)  time: 0.6466  data: 0.0405  max mem: 41794
Epoch: [126]  [120/312]  eta: 0:02:42  lr: 0.002737  min_lr: 0.002737  loss: 3.6955 (3.5200)  weight_decay: 0.0500 (0.0500)  time: 0.7728  data: 0.0913  max mem: 41794
Epoch: [126]  [130/312]  eta: 0:02:33  lr: 0.002736  min_lr: 0.002736  loss: 3.4414 (3.5080)  weight_decay: 0.0500 (0.0500)  time: 0.9231  data: 0.1302  max mem: 41794
Epoch: [126]  [140/312]  eta: 0:02:21  lr: 0.002736  min_lr: 0.002736  loss: 3.5991 (3.5135)  weight_decay: 0.0500 (0.0500)  time: 0.6506  data: 0.0395  max mem: 41794
Epoch: [126]  [150/312]  eta: 0:02:13  lr: 0.002735  min_lr: 0.002735  loss: 3.6051 (3.5102)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.0645  max mem: 41794
Epoch: [126]  [160/312]  eta: 0:02:04  lr: 0.002734  min_lr: 0.002734  loss: 3.6040 (3.5107)  weight_decay: 0.0500 (0.0500)  time: 0.8149  data: 0.0705  max mem: 41794
Epoch: [126]  [170/312]  eta: 0:01:54  lr: 0.002734  min_lr: 0.002734  loss: 3.7254 (3.5203)  weight_decay: 0.0500 (0.0500)  time: 0.6722  data: 0.0543  max mem: 41794
Epoch: [126]  [180/312]  eta: 0:01:46  lr: 0.002733  min_lr: 0.002733  loss: 3.6655 (3.5098)  weight_decay: 0.0500 (0.0500)  time: 0.6975  data: 0.1061  max mem: 41794
Epoch: [126]  [190/312]  eta: 0:01:36  lr: 0.002732  min_lr: 0.002732  loss: 3.6527 (3.5144)  weight_decay: 0.0500 (0.0500)  time: 0.6568  data: 0.0597  max mem: 41794
Epoch: [126]  [200/312]  eta: 0:01:29  lr: 0.002732  min_lr: 0.002732  loss: 3.7217 (3.5143)  weight_decay: 0.0500 (0.0500)  time: 0.7119  data: 0.0803  max mem: 41794
Epoch: [126]  [210/312]  eta: 0:01:21  lr: 0.002731  min_lr: 0.002731  loss: 3.7682 (3.5303)  weight_decay: 0.0500 (0.0500)  time: 0.8500  data: 0.0994  max mem: 41794
Epoch: [126]  [220/312]  eta: 0:01:12  lr: 0.002730  min_lr: 0.002730  loss: 3.8008 (3.5265)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.0573  max mem: 41794
Epoch: [126]  [230/312]  eta: 0:01:04  lr: 0.002730  min_lr: 0.002730  loss: 3.4063 (3.5174)  weight_decay: 0.0500 (0.0500)  time: 0.7114  data: 0.0708  max mem: 41794
Epoch: [126]  [240/312]  eta: 0:00:56  lr: 0.002729  min_lr: 0.002729  loss: 3.1751 (3.5079)  weight_decay: 0.0500 (0.0500)  time: 0.8426  data: 0.0420  max mem: 41794
Epoch: [126]  [250/312]  eta: 0:00:48  lr: 0.002728  min_lr: 0.002728  loss: 3.5881 (3.5043)  weight_decay: 0.0500 (0.0500)  time: 0.7027  data: 0.0212  max mem: 41794
Epoch: [126]  [260/312]  eta: 0:00:40  lr: 0.002728  min_lr: 0.002728  loss: 3.5847 (3.5069)  weight_decay: 0.0500 (0.0500)  time: 0.7098  data: 0.0184  max mem: 41794
Epoch: [126]  [270/312]  eta: 0:00:32  lr: 0.002727  min_lr: 0.002727  loss: 3.6823 (3.5157)  weight_decay: 0.0500 (0.0500)  time: 0.6738  data: 0.0084  max mem: 41794
Epoch: [126]  [280/312]  eta: 0:00:24  lr: 0.002726  min_lr: 0.002726  loss: 3.7962 (3.5221)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.0092  max mem: 41794
Epoch: [126]  [290/312]  eta: 0:00:17  lr: 0.002726  min_lr: 0.002726  loss: 3.7471 (3.5224)  weight_decay: 0.0500 (0.0500)  time: 0.8349  data: 0.0060  max mem: 41794
Epoch: [126]  [300/312]  eta: 0:00:09  lr: 0.002725  min_lr: 0.002725  loss: 3.5991 (3.5176)  weight_decay: 0.0500 (0.0500)  time: 0.6443  data: 0.0002  max mem: 41794
Epoch: [126]  [310/312]  eta: 0:00:01  lr: 0.002724  min_lr: 0.002724  loss: 3.5991 (3.5232)  weight_decay: 0.0500 (0.0500)  time: 0.4675  data: 0.0001  max mem: 41794
Epoch: [126]  [311/312]  eta: 0:00:00  lr: 0.002724  min_lr: 0.002724  loss: 3.6201 (3.5242)  weight_decay: 0.0500 (0.0500)  time: 0.4677  data: 0.0001  max mem: 41794
Epoch: [126] Total time: 0:03:58 (0.7641 s / it)
Averaged stats: lr: 0.002724  min_lr: 0.002724  loss: 3.6201 (3.5209)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.2720 (1.2720)  acc1: 77.3438 (77.3438)  acc5: 94.2708 (94.2708)  time: 7.9491  data: 7.7366  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.7295 (1.5903)  acc1: 68.2292 (69.8880)  acc5: 90.1042 (90.2880)  time: 1.0329  data: 0.8597  max mem: 41794
Test: Total time: 0:00:09 (1.0432 s / it)
* Acc@1 70.812 Acc@5 90.476 loss 1.588
Accuracy of the model on the 50000 test images: 70.8%
Max accuracy: 71.87%
Epoch: [127]  [  0/312]  eta: 1:17:41  lr: 0.002724  min_lr: 0.002724  loss: 3.3315 (3.3315)  weight_decay: 0.0500 (0.0500)  time: 14.9393  data: 13.3403  max mem: 41794
Epoch: [127]  [ 10/312]  eta: 0:11:27  lr: 0.002723  min_lr: 0.002723  loss: 3.3315 (3.4734)  weight_decay: 0.0500 (0.0500)  time: 2.2762  data: 1.3524  max mem: 41794
Epoch: [127]  [ 20/312]  eta: 0:07:42  lr: 0.002723  min_lr: 0.002723  loss: 3.6539 (3.5166)  weight_decay: 0.0500 (0.0500)  time: 0.9148  data: 0.1030  max mem: 41794
Epoch: [127]  [ 30/312]  eta: 0:05:51  lr: 0.002722  min_lr: 0.002722  loss: 3.7142 (3.5110)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.0297  max mem: 41794
Epoch: [127]  [ 40/312]  eta: 0:04:50  lr: 0.002721  min_lr: 0.002721  loss: 3.4212 (3.4712)  weight_decay: 0.0500 (0.0500)  time: 0.5265  data: 0.0113  max mem: 41794
Epoch: [127]  [ 50/312]  eta: 0:04:19  lr: 0.002721  min_lr: 0.002721  loss: 3.5624 (3.4881)  weight_decay: 0.0500 (0.0500)  time: 0.5928  data: 0.0146  max mem: 41794
Epoch: [127]  [ 60/312]  eta: 0:03:49  lr: 0.002720  min_lr: 0.002720  loss: 3.7451 (3.5129)  weight_decay: 0.0500 (0.0500)  time: 0.5880  data: 0.0117  max mem: 41794
Epoch: [127]  [ 70/312]  eta: 0:03:40  lr: 0.002719  min_lr: 0.002719  loss: 3.6844 (3.5181)  weight_decay: 0.0500 (0.0500)  time: 0.7068  data: 0.0337  max mem: 41794
Epoch: [127]  [ 80/312]  eta: 0:03:26  lr: 0.002719  min_lr: 0.002719  loss: 3.6095 (3.5025)  weight_decay: 0.0500 (0.0500)  time: 0.8222  data: 0.0573  max mem: 41794
Epoch: [127]  [ 90/312]  eta: 0:03:13  lr: 0.002718  min_lr: 0.002718  loss: 3.5545 (3.5017)  weight_decay: 0.0500 (0.0500)  time: 0.7255  data: 0.0683  max mem: 41794
Epoch: [127]  [100/312]  eta: 0:03:02  lr: 0.002717  min_lr: 0.002717  loss: 3.6938 (3.5247)  weight_decay: 0.0500 (0.0500)  time: 0.7452  data: 0.0857  max mem: 41794
Epoch: [127]  [110/312]  eta: 0:02:47  lr: 0.002717  min_lr: 0.002717  loss: 3.6938 (3.5253)  weight_decay: 0.0500 (0.0500)  time: 0.6331  data: 0.0462  max mem: 41794
Epoch: [127]  [120/312]  eta: 0:02:39  lr: 0.002716  min_lr: 0.002716  loss: 3.6623 (3.5193)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.0613  max mem: 41794
Epoch: [127]  [130/312]  eta: 0:02:31  lr: 0.002715  min_lr: 0.002715  loss: 3.7360 (3.5211)  weight_decay: 0.0500 (0.0500)  time: 0.8579  data: 0.0930  max mem: 41794
Epoch: [127]  [140/312]  eta: 0:02:19  lr: 0.002715  min_lr: 0.002715  loss: 3.7568 (3.5177)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.0503  max mem: 41794
Epoch: [127]  [150/312]  eta: 0:02:12  lr: 0.002714  min_lr: 0.002714  loss: 3.5836 (3.5155)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.1036  max mem: 41794
Epoch: [127]  [160/312]  eta: 0:02:03  lr: 0.002713  min_lr: 0.002713  loss: 3.4429 (3.5015)  weight_decay: 0.0500 (0.0500)  time: 0.8206  data: 0.1342  max mem: 41794
Epoch: [127]  [170/312]  eta: 0:01:53  lr: 0.002713  min_lr: 0.002713  loss: 3.4193 (3.4925)  weight_decay: 0.0500 (0.0500)  time: 0.6772  data: 0.0797  max mem: 41794
Epoch: [127]  [180/312]  eta: 0:01:45  lr: 0.002712  min_lr: 0.002712  loss: 3.4268 (3.4865)  weight_decay: 0.0500 (0.0500)  time: 0.7277  data: 0.1262  max mem: 41794
Epoch: [127]  [190/312]  eta: 0:01:35  lr: 0.002711  min_lr: 0.002711  loss: 3.6735 (3.5042)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0956  max mem: 41794
Epoch: [127]  [200/312]  eta: 0:01:28  lr: 0.002711  min_lr: 0.002711  loss: 3.7130 (3.4998)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.0538  max mem: 41794
Epoch: [127]  [210/312]  eta: 0:01:20  lr: 0.002710  min_lr: 0.002710  loss: 3.6382 (3.5028)  weight_decay: 0.0500 (0.0500)  time: 0.8185  data: 0.0907  max mem: 41794
Epoch: [127]  [220/312]  eta: 0:01:11  lr: 0.002709  min_lr: 0.002709  loss: 3.6159 (3.4910)  weight_decay: 0.0500 (0.0500)  time: 0.6497  data: 0.0519  max mem: 41794
Epoch: [127]  [230/312]  eta: 0:01:04  lr: 0.002709  min_lr: 0.002709  loss: 3.4645 (3.4957)  weight_decay: 0.0500 (0.0500)  time: 0.7247  data: 0.0841  max mem: 41794
Epoch: [127]  [240/312]  eta: 0:00:56  lr: 0.002708  min_lr: 0.002708  loss: 3.4645 (3.4968)  weight_decay: 0.0500 (0.0500)  time: 0.8299  data: 0.0965  max mem: 41794
Epoch: [127]  [250/312]  eta: 0:00:48  lr: 0.002707  min_lr: 0.002707  loss: 3.6810 (3.5079)  weight_decay: 0.0500 (0.0500)  time: 0.6635  data: 0.0482  max mem: 41794
Epoch: [127]  [260/312]  eta: 0:00:40  lr: 0.002707  min_lr: 0.002707  loss: 3.5748 (3.4971)  weight_decay: 0.0500 (0.0500)  time: 0.6943  data: 0.0819  max mem: 41794
Epoch: [127]  [270/312]  eta: 0:00:32  lr: 0.002706  min_lr: 0.002706  loss: 3.2665 (3.4907)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.0623  max mem: 41794
Epoch: [127]  [280/312]  eta: 0:00:24  lr: 0.002705  min_lr: 0.002705  loss: 3.2855 (3.4895)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.0853  max mem: 41794
Epoch: [127]  [290/312]  eta: 0:00:17  lr: 0.002705  min_lr: 0.002705  loss: 3.7142 (3.4919)  weight_decay: 0.0500 (0.0500)  time: 0.8694  data: 0.1936  max mem: 41794
Epoch: [127]  [300/312]  eta: 0:00:09  lr: 0.002704  min_lr: 0.002704  loss: 3.6721 (3.4954)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.1098  max mem: 41794
Epoch: [127]  [310/312]  eta: 0:00:01  lr: 0.002703  min_lr: 0.002703  loss: 3.4921 (3.4909)  weight_decay: 0.0500 (0.0500)  time: 0.4678  data: 0.0001  max mem: 41794
Epoch: [127]  [311/312]  eta: 0:00:00  lr: 0.002703  min_lr: 0.002703  loss: 3.4921 (3.4884)  weight_decay: 0.0500 (0.0500)  time: 0.4675  data: 0.0001  max mem: 41794
Epoch: [127] Total time: 0:03:56 (0.7589 s / it)
Averaged stats: lr: 0.002703  min_lr: 0.002703  loss: 3.4921 (3.4905)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.2515 (1.2515)  acc1: 79.0365 (79.0365)  acc5: 94.9219 (94.9219)  time: 8.4365  data: 8.2245  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6761 (1.5212)  acc1: 69.6615 (71.4080)  acc5: 90.6250 (91.1680)  time: 1.0859  data: 0.9139  max mem: 41794
Test: Total time: 0:00:10 (1.1215 s / it)
* Acc@1 71.574 Acc@5 91.134 loss 1.532
Accuracy of the model on the 50000 test images: 71.6%
Max accuracy: 71.87%
Epoch: [128]  [  0/312]  eta: 1:24:59  lr: 0.002703  min_lr: 0.002703  loss: 3.9138 (3.9138)  weight_decay: 0.0500 (0.0500)  time: 16.3451  data: 15.8821  max mem: 41794
Epoch: [128]  [ 10/312]  eta: 0:11:02  lr: 0.002702  min_lr: 0.002702  loss: 3.9384 (3.8024)  weight_decay: 0.0500 (0.0500)  time: 2.1936  data: 1.4445  max mem: 41794
Epoch: [128]  [ 20/312]  eta: 0:07:18  lr: 0.002702  min_lr: 0.002702  loss: 3.8475 (3.6660)  weight_decay: 0.0500 (0.0500)  time: 0.7596  data: 0.1208  max mem: 41794
Epoch: [128]  [ 30/312]  eta: 0:05:34  lr: 0.002701  min_lr: 0.002701  loss: 3.4045 (3.5608)  weight_decay: 0.0500 (0.0500)  time: 0.6324  data: 0.1257  max mem: 41794
Epoch: [128]  [ 40/312]  eta: 0:04:46  lr: 0.002700  min_lr: 0.002700  loss: 3.4045 (3.5268)  weight_decay: 0.0500 (0.0500)  time: 0.5837  data: 0.0412  max mem: 41794
Epoch: [128]  [ 50/312]  eta: 0:04:22  lr: 0.002700  min_lr: 0.002700  loss: 3.6605 (3.5704)  weight_decay: 0.0500 (0.0500)  time: 0.7115  data: 0.1328  max mem: 41794
Epoch: [128]  [ 60/312]  eta: 0:03:51  lr: 0.002699  min_lr: 0.002699  loss: 3.6346 (3.5858)  weight_decay: 0.0500 (0.0500)  time: 0.6390  data: 0.0977  max mem: 41794
Epoch: [128]  [ 70/312]  eta: 0:03:39  lr: 0.002698  min_lr: 0.002698  loss: 3.6346 (3.5701)  weight_decay: 0.0500 (0.0500)  time: 0.6692  data: 0.1263  max mem: 41794
Epoch: [128]  [ 80/312]  eta: 0:03:27  lr: 0.002698  min_lr: 0.002698  loss: 3.3242 (3.5485)  weight_decay: 0.0500 (0.0500)  time: 0.8207  data: 0.2812  max mem: 41794
Epoch: [128]  [ 90/312]  eta: 0:03:09  lr: 0.002697  min_lr: 0.002697  loss: 3.0987 (3.5089)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1561  max mem: 41794
Epoch: [128]  [100/312]  eta: 0:03:03  lr: 0.002696  min_lr: 0.002696  loss: 3.2190 (3.5035)  weight_decay: 0.0500 (0.0500)  time: 0.7401  data: 0.2312  max mem: 41794
Epoch: [128]  [110/312]  eta: 0:02:47  lr: 0.002696  min_lr: 0.002696  loss: 3.3835 (3.4879)  weight_decay: 0.0500 (0.0500)  time: 0.7355  data: 0.2312  max mem: 41794
Epoch: [128]  [120/312]  eta: 0:02:39  lr: 0.002695  min_lr: 0.002695  loss: 3.3835 (3.4842)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.1738  max mem: 41794
Epoch: [128]  [130/312]  eta: 0:02:33  lr: 0.002694  min_lr: 0.002694  loss: 3.5982 (3.4907)  weight_decay: 0.0500 (0.0500)  time: 0.9000  data: 0.3990  max mem: 41794
Epoch: [128]  [140/312]  eta: 0:02:20  lr: 0.002694  min_lr: 0.002694  loss: 3.7165 (3.4997)  weight_decay: 0.0500 (0.0500)  time: 0.7239  data: 0.2257  max mem: 41794
Epoch: [128]  [150/312]  eta: 0:02:12  lr: 0.002693  min_lr: 0.002693  loss: 3.7020 (3.4945)  weight_decay: 0.0500 (0.0500)  time: 0.6741  data: 0.1733  max mem: 41794
Epoch: [128]  [160/312]  eta: 0:02:04  lr: 0.002692  min_lr: 0.002692  loss: 3.7176 (3.5101)  weight_decay: 0.0500 (0.0500)  time: 0.8210  data: 0.3251  max mem: 41794
Epoch: [128]  [170/312]  eta: 0:01:53  lr: 0.002692  min_lr: 0.002692  loss: 3.7808 (3.5000)  weight_decay: 0.0500 (0.0500)  time: 0.6483  data: 0.1586  max mem: 41794
Epoch: [128]  [180/312]  eta: 0:01:46  lr: 0.002691  min_lr: 0.002691  loss: 3.5864 (3.5037)  weight_decay: 0.0500 (0.0500)  time: 0.7287  data: 0.2276  max mem: 41794
Epoch: [128]  [190/312]  eta: 0:01:36  lr: 0.002690  min_lr: 0.002690  loss: 3.4893 (3.4907)  weight_decay: 0.0500 (0.0500)  time: 0.7252  data: 0.2213  max mem: 41794
Epoch: [128]  [200/312]  eta: 0:01:29  lr: 0.002690  min_lr: 0.002690  loss: 3.3864 (3.4898)  weight_decay: 0.0500 (0.0500)  time: 0.6903  data: 0.1971  max mem: 41794
Epoch: [128]  [210/312]  eta: 0:01:21  lr: 0.002689  min_lr: 0.002689  loss: 3.6483 (3.5019)  weight_decay: 0.0500 (0.0500)  time: 0.8549  data: 0.3663  max mem: 41794
Epoch: [128]  [220/312]  eta: 0:01:12  lr: 0.002688  min_lr: 0.002688  loss: 3.6608 (3.5005)  weight_decay: 0.0500 (0.0500)  time: 0.6632  data: 0.1697  max mem: 41794
Epoch: [128]  [230/312]  eta: 0:01:04  lr: 0.002688  min_lr: 0.002688  loss: 3.4556 (3.4998)  weight_decay: 0.0500 (0.0500)  time: 0.6864  data: 0.1876  max mem: 41794
Epoch: [128]  [240/312]  eta: 0:00:56  lr: 0.002687  min_lr: 0.002687  loss: 3.5448 (3.5005)  weight_decay: 0.0500 (0.0500)  time: 0.8436  data: 0.3486  max mem: 41794
Epoch: [128]  [250/312]  eta: 0:00:48  lr: 0.002686  min_lr: 0.002686  loss: 3.4235 (3.4968)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.1840  max mem: 41794
Epoch: [128]  [260/312]  eta: 0:00:40  lr: 0.002686  min_lr: 0.002686  loss: 3.4235 (3.4973)  weight_decay: 0.0500 (0.0500)  time: 0.7095  data: 0.2134  max mem: 41794
Epoch: [128]  [270/312]  eta: 0:00:32  lr: 0.002685  min_lr: 0.002685  loss: 3.4820 (3.5002)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1911  max mem: 41794
Epoch: [128]  [280/312]  eta: 0:00:24  lr: 0.002684  min_lr: 0.002684  loss: 3.6560 (3.4978)  weight_decay: 0.0500 (0.0500)  time: 0.6571  data: 0.1620  max mem: 41794
Epoch: [128]  [290/312]  eta: 0:00:17  lr: 0.002684  min_lr: 0.002684  loss: 3.5803 (3.4967)  weight_decay: 0.0500 (0.0500)  time: 0.8123  data: 0.3228  max mem: 41794
Epoch: [128]  [300/312]  eta: 0:00:09  lr: 0.002683  min_lr: 0.002683  loss: 3.4619 (3.4927)  weight_decay: 0.0500 (0.0500)  time: 0.6393  data: 0.1612  max mem: 41794
Epoch: [128]  [310/312]  eta: 0:00:01  lr: 0.002682  min_lr: 0.002682  loss: 3.4137 (3.4889)  weight_decay: 0.0500 (0.0500)  time: 0.4644  data: 0.0001  max mem: 41794
Epoch: [128]  [311/312]  eta: 0:00:00  lr: 0.002682  min_lr: 0.002682  loss: 3.4137 (3.4909)  weight_decay: 0.0500 (0.0500)  time: 0.4644  data: 0.0001  max mem: 41794
Epoch: [128] Total time: 0:03:57 (0.7604 s / it)
Averaged stats: lr: 0.002682  min_lr: 0.002682  loss: 3.4137 (3.4957)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:09  loss: 1.1188 (1.1188)  acc1: 77.6042 (77.6042)  acc5: 94.4010 (94.4010)  time: 7.7312  data: 7.5205  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5304 (1.4083)  acc1: 69.7917 (71.2480)  acc5: 89.7135 (90.6240)  time: 1.0603  data: 0.8890  max mem: 41794
Test: Total time: 0:00:09 (1.0885 s / it)
* Acc@1 71.494 Acc@5 90.934 loss 1.395
Accuracy of the model on the 50000 test images: 71.5%
Max accuracy: 71.87%
Epoch: [129]  [  0/312]  eta: 1:24:12  lr: 0.002682  min_lr: 0.002682  loss: 3.8424 (3.8424)  weight_decay: 0.0500 (0.0500)  time: 16.1935  data: 13.5007  max mem: 41794
Epoch: [129]  [ 10/312]  eta: 0:10:53  lr: 0.002681  min_lr: 0.002681  loss: 3.6830 (3.5764)  weight_decay: 0.0500 (0.0500)  time: 2.1655  data: 1.3654  max mem: 41794
Epoch: [129]  [ 20/312]  eta: 0:07:25  lr: 0.002681  min_lr: 0.002681  loss: 3.7155 (3.6859)  weight_decay: 0.0500 (0.0500)  time: 0.7912  data: 0.1133  max mem: 41794
Epoch: [129]  [ 30/312]  eta: 0:05:39  lr: 0.002680  min_lr: 0.002680  loss: 3.7155 (3.6676)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.0377  max mem: 41794
Epoch: [129]  [ 40/312]  eta: 0:04:58  lr: 0.002679  min_lr: 0.002679  loss: 3.6751 (3.7073)  weight_decay: 0.0500 (0.0500)  time: 0.6456  data: 0.0512  max mem: 41794
Epoch: [129]  [ 50/312]  eta: 0:04:37  lr: 0.002679  min_lr: 0.002679  loss: 3.6850 (3.6960)  weight_decay: 0.0500 (0.0500)  time: 0.8296  data: 0.1488  max mem: 41794
Epoch: [129]  [ 60/312]  eta: 0:04:03  lr: 0.002678  min_lr: 0.002678  loss: 3.7198 (3.6973)  weight_decay: 0.0500 (0.0500)  time: 0.6956  data: 0.0983  max mem: 41794
Epoch: [129]  [ 70/312]  eta: 0:03:45  lr: 0.002677  min_lr: 0.002677  loss: 3.6812 (3.6677)  weight_decay: 0.0500 (0.0500)  time: 0.6081  data: 0.1037  max mem: 41794
Epoch: [129]  [ 80/312]  eta: 0:03:34  lr: 0.002677  min_lr: 0.002677  loss: 3.3364 (3.6318)  weight_decay: 0.0500 (0.0500)  time: 0.8041  data: 0.2397  max mem: 41794
Epoch: [129]  [ 90/312]  eta: 0:03:15  lr: 0.002676  min_lr: 0.002676  loss: 3.4685 (3.5975)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1591  max mem: 41794
Epoch: [129]  [100/312]  eta: 0:03:03  lr: 0.002675  min_lr: 0.002675  loss: 3.4685 (3.5723)  weight_decay: 0.0500 (0.0500)  time: 0.6358  data: 0.1359  max mem: 41794
Epoch: [129]  [110/312]  eta: 0:02:48  lr: 0.002675  min_lr: 0.002675  loss: 3.5061 (3.5636)  weight_decay: 0.0500 (0.0500)  time: 0.6148  data: 0.1134  max mem: 41794
Epoch: [129]  [120/312]  eta: 0:02:41  lr: 0.002674  min_lr: 0.002674  loss: 3.5405 (3.5604)  weight_decay: 0.0500 (0.0500)  time: 0.7121  data: 0.1945  max mem: 41794
Epoch: [129]  [130/312]  eta: 0:02:35  lr: 0.002673  min_lr: 0.002673  loss: 3.5595 (3.5594)  weight_decay: 0.0500 (0.0500)  time: 0.9649  data: 0.3699  max mem: 41794
Epoch: [129]  [140/312]  eta: 0:02:22  lr: 0.002673  min_lr: 0.002673  loss: 3.6135 (3.5713)  weight_decay: 0.0500 (0.0500)  time: 0.7447  data: 0.1761  max mem: 41794
Epoch: [129]  [150/312]  eta: 0:02:14  lr: 0.002672  min_lr: 0.002672  loss: 3.7167 (3.5649)  weight_decay: 0.0500 (0.0500)  time: 0.6903  data: 0.1633  max mem: 41794
Epoch: [129]  [160/312]  eta: 0:02:05  lr: 0.002671  min_lr: 0.002671  loss: 3.3414 (3.5404)  weight_decay: 0.0500 (0.0500)  time: 0.7918  data: 0.2251  max mem: 41794
Epoch: [129]  [170/312]  eta: 0:01:55  lr: 0.002671  min_lr: 0.002671  loss: 3.3414 (3.5358)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.1412  max mem: 41794
Epoch: [129]  [180/312]  eta: 0:01:47  lr: 0.002670  min_lr: 0.002670  loss: 3.4701 (3.5388)  weight_decay: 0.0500 (0.0500)  time: 0.7512  data: 0.2211  max mem: 41794
Epoch: [129]  [190/312]  eta: 0:01:37  lr: 0.002669  min_lr: 0.002669  loss: 3.5668 (3.5296)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.1445  max mem: 41794
Epoch: [129]  [200/312]  eta: 0:01:29  lr: 0.002668  min_lr: 0.002668  loss: 3.6796 (3.5358)  weight_decay: 0.0500 (0.0500)  time: 0.6605  data: 0.1630  max mem: 41794
Epoch: [129]  [210/312]  eta: 0:01:21  lr: 0.002668  min_lr: 0.002668  loss: 3.6990 (3.5454)  weight_decay: 0.0500 (0.0500)  time: 0.8498  data: 0.3470  max mem: 41794
Epoch: [129]  [220/312]  eta: 0:01:12  lr: 0.002667  min_lr: 0.002667  loss: 3.8134 (3.5559)  weight_decay: 0.0500 (0.0500)  time: 0.6878  data: 0.1868  max mem: 41794
Epoch: [129]  [230/312]  eta: 0:01:04  lr: 0.002666  min_lr: 0.002666  loss: 3.6184 (3.5582)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.1740  max mem: 41794
Epoch: [129]  [240/312]  eta: 0:00:56  lr: 0.002666  min_lr: 0.002666  loss: 3.6184 (3.5653)  weight_decay: 0.0500 (0.0500)  time: 0.7666  data: 0.2755  max mem: 41794
Epoch: [129]  [250/312]  eta: 0:00:48  lr: 0.002665  min_lr: 0.002665  loss: 3.5722 (3.5610)  weight_decay: 0.0500 (0.0500)  time: 0.6641  data: 0.1727  max mem: 41794
Epoch: [129]  [260/312]  eta: 0:00:40  lr: 0.002664  min_lr: 0.002664  loss: 3.5722 (3.5617)  weight_decay: 0.0500 (0.0500)  time: 0.7580  data: 0.2733  max mem: 41794
Epoch: [129]  [270/312]  eta: 0:00:32  lr: 0.002664  min_lr: 0.002664  loss: 3.7151 (3.5622)  weight_decay: 0.0500 (0.0500)  time: 0.6946  data: 0.2028  max mem: 41794
Epoch: [129]  [280/312]  eta: 0:00:24  lr: 0.002663  min_lr: 0.002663  loss: 3.5751 (3.5648)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.1905  max mem: 41794
Epoch: [129]  [290/312]  eta: 0:00:17  lr: 0.002662  min_lr: 0.002662  loss: 3.4757 (3.5581)  weight_decay: 0.0500 (0.0500)  time: 0.7737  data: 0.2797  max mem: 41794
Epoch: [129]  [300/312]  eta: 0:00:09  lr: 0.002662  min_lr: 0.002662  loss: 3.3197 (3.5473)  weight_decay: 0.0500 (0.0500)  time: 0.5798  data: 0.0896  max mem: 41794
Epoch: [129]  [310/312]  eta: 0:00:01  lr: 0.002661  min_lr: 0.002661  loss: 3.2172 (3.5398)  weight_decay: 0.0500 (0.0500)  time: 0.4650  data: 0.0001  max mem: 41794
Epoch: [129]  [311/312]  eta: 0:00:00  lr: 0.002661  min_lr: 0.002661  loss: 3.2172 (3.5384)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [129] Total time: 0:03:57 (0.7607 s / it)
Averaged stats: lr: 0.002661  min_lr: 0.002661  loss: 3.2172 (3.5018)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.1447 (1.1447)  acc1: 76.6927 (76.6927)  acc5: 93.6198 (93.6198)  time: 8.5594  data: 8.3475  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4288 (1.3760)  acc1: 69.0104 (70.3040)  acc5: 90.6250 (90.8160)  time: 1.0996  data: 0.9276  max mem: 41794
Test: Total time: 0:00:10 (1.1121 s / it)
* Acc@1 70.742 Acc@5 90.818 loss 1.373
Accuracy of the model on the 50000 test images: 70.7%
Max accuracy: 71.87%
Epoch: [130]  [  0/312]  eta: 1:22:34  lr: 0.002661  min_lr: 0.002661  loss: 3.2729 (3.2729)  weight_decay: 0.0500 (0.0500)  time: 15.8809  data: 13.8313  max mem: 41794
Epoch: [130]  [ 10/312]  eta: 0:10:28  lr: 0.002660  min_lr: 0.002660  loss: 3.7508 (3.6921)  weight_decay: 0.0500 (0.0500)  time: 2.0808  data: 1.2581  max mem: 41794
Epoch: [130]  [ 20/312]  eta: 0:07:37  lr: 0.002660  min_lr: 0.002660  loss: 3.6757 (3.6591)  weight_decay: 0.0500 (0.0500)  time: 0.8528  data: 0.0969  max mem: 41794
Epoch: [130]  [ 30/312]  eta: 0:05:51  lr: 0.002659  min_lr: 0.002659  loss: 3.5471 (3.5631)  weight_decay: 0.0500 (0.0500)  time: 0.7881  data: 0.1043  max mem: 41794
Epoch: [130]  [ 40/312]  eta: 0:04:57  lr: 0.002658  min_lr: 0.002658  loss: 3.5348 (3.5225)  weight_decay: 0.0500 (0.0500)  time: 0.5963  data: 0.0340  max mem: 41794
Epoch: [130]  [ 50/312]  eta: 0:04:21  lr: 0.002657  min_lr: 0.002657  loss: 3.6129 (3.4979)  weight_decay: 0.0500 (0.0500)  time: 0.6104  data: 0.0339  max mem: 41794
Epoch: [130]  [ 60/312]  eta: 0:03:50  lr: 0.002657  min_lr: 0.002657  loss: 3.5979 (3.4926)  weight_decay: 0.0500 (0.0500)  time: 0.5440  data: 0.0080  max mem: 41794
Epoch: [130]  [ 70/312]  eta: 0:03:39  lr: 0.002656  min_lr: 0.002656  loss: 3.5310 (3.4743)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.0685  max mem: 41794
Epoch: [130]  [ 80/312]  eta: 0:03:27  lr: 0.002655  min_lr: 0.002655  loss: 3.5467 (3.4744)  weight_decay: 0.0500 (0.0500)  time: 0.8437  data: 0.0986  max mem: 41794
Epoch: [130]  [ 90/312]  eta: 0:03:13  lr: 0.002655  min_lr: 0.002655  loss: 3.6025 (3.4693)  weight_decay: 0.0500 (0.0500)  time: 0.7514  data: 0.1097  max mem: 41794
Epoch: [130]  [100/312]  eta: 0:03:04  lr: 0.002654  min_lr: 0.002654  loss: 3.5519 (3.4706)  weight_decay: 0.0500 (0.0500)  time: 0.7543  data: 0.1300  max mem: 41794
Epoch: [130]  [110/312]  eta: 0:02:48  lr: 0.002653  min_lr: 0.002653  loss: 3.6221 (3.4941)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.0524  max mem: 41794
Epoch: [130]  [120/312]  eta: 0:02:41  lr: 0.002653  min_lr: 0.002653  loss: 3.4689 (3.4805)  weight_decay: 0.0500 (0.0500)  time: 0.7092  data: 0.0843  max mem: 41794
Epoch: [130]  [130/312]  eta: 0:02:31  lr: 0.002652  min_lr: 0.002652  loss: 3.2532 (3.4624)  weight_decay: 0.0500 (0.0500)  time: 0.8347  data: 0.1014  max mem: 41794
Epoch: [130]  [140/312]  eta: 0:02:20  lr: 0.002651  min_lr: 0.002651  loss: 3.3840 (3.4526)  weight_decay: 0.0500 (0.0500)  time: 0.6684  data: 0.0627  max mem: 41794
Epoch: [130]  [150/312]  eta: 0:02:12  lr: 0.002651  min_lr: 0.002651  loss: 3.6086 (3.4536)  weight_decay: 0.0500 (0.0500)  time: 0.7222  data: 0.1169  max mem: 41794
Epoch: [130]  [160/312]  eta: 0:02:03  lr: 0.002650  min_lr: 0.002650  loss: 3.6026 (3.4664)  weight_decay: 0.0500 (0.0500)  time: 0.7944  data: 0.0752  max mem: 41794
Epoch: [130]  [170/312]  eta: 0:01:54  lr: 0.002649  min_lr: 0.002649  loss: 3.6026 (3.4721)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.0632  max mem: 41794
Epoch: [130]  [180/312]  eta: 0:01:46  lr: 0.002649  min_lr: 0.002649  loss: 3.4347 (3.4658)  weight_decay: 0.0500 (0.0500)  time: 0.7399  data: 0.1320  max mem: 41794
Epoch: [130]  [190/312]  eta: 0:01:36  lr: 0.002648  min_lr: 0.002648  loss: 3.6859 (3.4706)  weight_decay: 0.0500 (0.0500)  time: 0.6809  data: 0.0714  max mem: 41794
Epoch: [130]  [200/312]  eta: 0:01:28  lr: 0.002647  min_lr: 0.002647  loss: 3.5891 (3.4649)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0741  max mem: 41794
Epoch: [130]  [210/312]  eta: 0:01:21  lr: 0.002647  min_lr: 0.002647  loss: 3.3481 (3.4570)  weight_decay: 0.0500 (0.0500)  time: 0.8587  data: 0.1350  max mem: 41794
Epoch: [130]  [220/312]  eta: 0:01:12  lr: 0.002646  min_lr: 0.002646  loss: 3.3646 (3.4557)  weight_decay: 0.0500 (0.0500)  time: 0.7205  data: 0.0943  max mem: 41794
Epoch: [130]  [230/312]  eta: 0:01:04  lr: 0.002645  min_lr: 0.002645  loss: 3.3949 (3.4508)  weight_decay: 0.0500 (0.0500)  time: 0.6927  data: 0.1033  max mem: 41794
Epoch: [130]  [240/312]  eta: 0:00:56  lr: 0.002645  min_lr: 0.002645  loss: 3.3884 (3.4483)  weight_decay: 0.0500 (0.0500)  time: 0.7890  data: 0.0738  max mem: 41794
Epoch: [130]  [250/312]  eta: 0:00:48  lr: 0.002644  min_lr: 0.002644  loss: 3.5939 (3.4524)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.0579  max mem: 41794
Epoch: [130]  [260/312]  eta: 0:00:40  lr: 0.002643  min_lr: 0.002643  loss: 3.5965 (3.4530)  weight_decay: 0.0500 (0.0500)  time: 0.7323  data: 0.1336  max mem: 41794
Epoch: [130]  [270/312]  eta: 0:00:32  lr: 0.002643  min_lr: 0.002643  loss: 3.3548 (3.4439)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.0796  max mem: 41794
Epoch: [130]  [280/312]  eta: 0:00:24  lr: 0.002642  min_lr: 0.002642  loss: 3.1901 (3.4374)  weight_decay: 0.0500 (0.0500)  time: 0.6633  data: 0.0700  max mem: 41794
Epoch: [130]  [290/312]  eta: 0:00:17  lr: 0.002641  min_lr: 0.002641  loss: 3.3975 (3.4355)  weight_decay: 0.0500 (0.0500)  time: 0.8143  data: 0.1094  max mem: 41794
Epoch: [130]  [300/312]  eta: 0:00:09  lr: 0.002640  min_lr: 0.002640  loss: 3.5008 (3.4426)  weight_decay: 0.0500 (0.0500)  time: 0.6574  data: 0.0633  max mem: 41794
Epoch: [130]  [310/312]  eta: 0:00:01  lr: 0.002640  min_lr: 0.002640  loss: 3.7472 (3.4501)  weight_decay: 0.0500 (0.0500)  time: 0.4865  data: 0.0236  max mem: 41794
Epoch: [130]  [311/312]  eta: 0:00:00  lr: 0.002640  min_lr: 0.002640  loss: 3.7880 (3.4517)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0023  max mem: 41794
Epoch: [130] Total time: 0:03:57 (0.7622 s / it)
Averaged stats: lr: 0.002640  min_lr: 0.002640  loss: 3.7880 (3.4960)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.3663 (1.3663)  acc1: 77.3438 (77.3438)  acc5: 93.6198 (93.6198)  time: 8.1785  data: 7.9696  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6985 (1.5936)  acc1: 69.1406 (70.4960)  acc5: 91.0156 (90.6880)  time: 1.0567  data: 0.8856  max mem: 41794
Test: Total time: 0:00:09 (1.0671 s / it)
* Acc@1 71.468 Acc@5 90.684 loss 1.595
Accuracy of the model on the 50000 test images: 71.5%
Max accuracy: 71.87%
Epoch: [131]  [  0/312]  eta: 1:19:31  lr: 0.002640  min_lr: 0.002640  loss: 2.7127 (2.7127)  weight_decay: 0.0500 (0.0500)  time: 15.2936  data: 11.9213  max mem: 41794
Epoch: [131]  [ 10/312]  eta: 0:10:55  lr: 0.002639  min_lr: 0.002639  loss: 3.3464 (3.3198)  weight_decay: 0.0500 (0.0500)  time: 2.1717  data: 1.3284  max mem: 41794
Epoch: [131]  [ 20/312]  eta: 0:07:35  lr: 0.002638  min_lr: 0.002638  loss: 3.3464 (3.2564)  weight_decay: 0.0500 (0.0500)  time: 0.8726  data: 0.2216  max mem: 41794
Epoch: [131]  [ 30/312]  eta: 0:05:48  lr: 0.002638  min_lr: 0.002638  loss: 3.5994 (3.3960)  weight_decay: 0.0500 (0.0500)  time: 0.7184  data: 0.0971  max mem: 41794
Epoch: [131]  [ 40/312]  eta: 0:04:55  lr: 0.002637  min_lr: 0.002637  loss: 3.6570 (3.3503)  weight_decay: 0.0500 (0.0500)  time: 0.5879  data: 0.0171  max mem: 41794
Epoch: [131]  [ 50/312]  eta: 0:04:29  lr: 0.002636  min_lr: 0.002636  loss: 3.3514 (3.3895)  weight_decay: 0.0500 (0.0500)  time: 0.7077  data: 0.0309  max mem: 41794
Epoch: [131]  [ 60/312]  eta: 0:03:56  lr: 0.002636  min_lr: 0.002636  loss: 3.4206 (3.3906)  weight_decay: 0.0500 (0.0500)  time: 0.6395  data: 0.0254  max mem: 41794
Epoch: [131]  [ 70/312]  eta: 0:03:45  lr: 0.002635  min_lr: 0.002635  loss: 3.2523 (3.3647)  weight_decay: 0.0500 (0.0500)  time: 0.6903  data: 0.0472  max mem: 41794
Epoch: [131]  [ 80/312]  eta: 0:03:32  lr: 0.002634  min_lr: 0.002634  loss: 3.3799 (3.3699)  weight_decay: 0.0500 (0.0500)  time: 0.8454  data: 0.0517  max mem: 41794
Epoch: [131]  [ 90/312]  eta: 0:03:13  lr: 0.002634  min_lr: 0.002634  loss: 3.4580 (3.3793)  weight_decay: 0.0500 (0.0500)  time: 0.6615  data: 0.0248  max mem: 41794
Epoch: [131]  [100/312]  eta: 0:03:07  lr: 0.002633  min_lr: 0.002633  loss: 3.4580 (3.3885)  weight_decay: 0.0500 (0.0500)  time: 0.7489  data: 0.0553  max mem: 41794
Epoch: [131]  [110/312]  eta: 0:02:51  lr: 0.002632  min_lr: 0.002632  loss: 3.5545 (3.4016)  weight_decay: 0.0500 (0.0500)  time: 0.7374  data: 0.0367  max mem: 41794
Epoch: [131]  [120/312]  eta: 0:02:43  lr: 0.002631  min_lr: 0.002631  loss: 3.6832 (3.4161)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.0007  max mem: 41794
Epoch: [131]  [130/312]  eta: 0:02:34  lr: 0.002631  min_lr: 0.002631  loss: 3.6349 (3.4202)  weight_decay: 0.0500 (0.0500)  time: 0.8638  data: 0.0022  max mem: 41794
Epoch: [131]  [140/312]  eta: 0:02:21  lr: 0.002630  min_lr: 0.002630  loss: 3.6349 (3.4310)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.0045  max mem: 41794
Epoch: [131]  [150/312]  eta: 0:02:13  lr: 0.002629  min_lr: 0.002629  loss: 3.6611 (3.4399)  weight_decay: 0.0500 (0.0500)  time: 0.6688  data: 0.0044  max mem: 41794
Epoch: [131]  [160/312]  eta: 0:02:05  lr: 0.002629  min_lr: 0.002629  loss: 3.6619 (3.4532)  weight_decay: 0.0500 (0.0500)  time: 0.8201  data: 0.0021  max mem: 41794
Epoch: [131]  [170/312]  eta: 0:01:55  lr: 0.002628  min_lr: 0.002628  loss: 3.6256 (3.4510)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.0163  max mem: 41794
Epoch: [131]  [180/312]  eta: 0:01:47  lr: 0.002627  min_lr: 0.002627  loss: 3.6012 (3.4564)  weight_decay: 0.0500 (0.0500)  time: 0.7447  data: 0.0207  max mem: 41794
Epoch: [131]  [190/312]  eta: 0:01:37  lr: 0.002627  min_lr: 0.002627  loss: 3.4462 (3.4538)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.0051  max mem: 41794
Epoch: [131]  [200/312]  eta: 0:01:29  lr: 0.002626  min_lr: 0.002626  loss: 3.5106 (3.4648)  weight_decay: 0.0500 (0.0500)  time: 0.6533  data: 0.0110  max mem: 41794
Epoch: [131]  [210/312]  eta: 0:01:22  lr: 0.002625  min_lr: 0.002625  loss: 3.7339 (3.4692)  weight_decay: 0.0500 (0.0500)  time: 0.8610  data: 0.0352  max mem: 41794
Epoch: [131]  [220/312]  eta: 0:01:12  lr: 0.002625  min_lr: 0.002625  loss: 3.8099 (3.4757)  weight_decay: 0.0500 (0.0500)  time: 0.7070  data: 0.0248  max mem: 41794
Epoch: [131]  [230/312]  eta: 0:01:05  lr: 0.002624  min_lr: 0.002624  loss: 3.7237 (3.4838)  weight_decay: 0.0500 (0.0500)  time: 0.6918  data: 0.0341  max mem: 41794
Epoch: [131]  [240/312]  eta: 0:00:56  lr: 0.002623  min_lr: 0.002623  loss: 3.7237 (3.4902)  weight_decay: 0.0500 (0.0500)  time: 0.7814  data: 0.0357  max mem: 41794
Epoch: [131]  [250/312]  eta: 0:00:48  lr: 0.002623  min_lr: 0.002623  loss: 3.7772 (3.4942)  weight_decay: 0.0500 (0.0500)  time: 0.6189  data: 0.0322  max mem: 41794
Epoch: [131]  [260/312]  eta: 0:00:40  lr: 0.002622  min_lr: 0.002622  loss: 3.7262 (3.4984)  weight_decay: 0.0500 (0.0500)  time: 0.7347  data: 0.0514  max mem: 41794
Epoch: [131]  [270/312]  eta: 0:00:32  lr: 0.002621  min_lr: 0.002621  loss: 3.8010 (3.5128)  weight_decay: 0.0500 (0.0500)  time: 0.7252  data: 0.0429  max mem: 41794
Epoch: [131]  [280/312]  eta: 0:00:25  lr: 0.002621  min_lr: 0.002621  loss: 3.8985 (3.5196)  weight_decay: 0.0500 (0.0500)  time: 0.7422  data: 0.0677  max mem: 41794
Epoch: [131]  [290/312]  eta: 0:00:17  lr: 0.002620  min_lr: 0.002620  loss: 3.7051 (3.5167)  weight_decay: 0.0500 (0.0500)  time: 0.8069  data: 0.0480  max mem: 41794
Epoch: [131]  [300/312]  eta: 0:00:09  lr: 0.002619  min_lr: 0.002619  loss: 3.7051 (3.5151)  weight_decay: 0.0500 (0.0500)  time: 0.5682  data: 0.0021  max mem: 41794
Epoch: [131]  [310/312]  eta: 0:00:01  lr: 0.002619  min_lr: 0.002619  loss: 3.5088 (3.5163)  weight_decay: 0.0500 (0.0500)  time: 0.4627  data: 0.0001  max mem: 41794
Epoch: [131]  [311/312]  eta: 0:00:00  lr: 0.002618  min_lr: 0.002618  loss: 3.5088 (3.5169)  weight_decay: 0.0500 (0.0500)  time: 0.4624  data: 0.0001  max mem: 41794
Epoch: [131] Total time: 0:03:58 (0.7654 s / it)
Averaged stats: lr: 0.002618  min_lr: 0.002618  loss: 3.5088 (3.4749)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:08  loss: 1.2547 (1.2547)  acc1: 75.2604 (75.2604)  acc5: 93.0990 (93.0990)  time: 7.6343  data: 7.4282  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5447 (1.4870)  acc1: 69.7917 (70.5280)  acc5: 91.0156 (90.7200)  time: 1.0440  data: 0.8703  max mem: 41794
Test: Total time: 0:00:09 (1.0564 s / it)
* Acc@1 71.106 Acc@5 90.754 loss 1.473
Accuracy of the model on the 50000 test images: 71.1%
Max accuracy: 71.87%
Epoch: [132]  [  0/312]  eta: 1:23:16  lr: 0.002618  min_lr: 0.002618  loss: 2.6323 (2.6323)  weight_decay: 0.0500 (0.0500)  time: 16.0129  data: 13.0120  max mem: 41794
Epoch: [132]  [ 10/312]  eta: 0:11:34  lr: 0.002618  min_lr: 0.002618  loss: 3.5796 (3.4223)  weight_decay: 0.0500 (0.0500)  time: 2.3001  data: 1.3787  max mem: 41794
Epoch: [132]  [ 20/312]  eta: 0:07:31  lr: 0.002617  min_lr: 0.002617  loss: 3.5796 (3.4834)  weight_decay: 0.0500 (0.0500)  time: 0.8212  data: 0.1119  max mem: 41794
Epoch: [132]  [ 30/312]  eta: 0:05:43  lr: 0.002616  min_lr: 0.002616  loss: 3.7197 (3.4963)  weight_decay: 0.0500 (0.0500)  time: 0.6255  data: 0.0203  max mem: 41794
Epoch: [132]  [ 40/312]  eta: 0:04:44  lr: 0.002616  min_lr: 0.002616  loss: 3.3299 (3.4339)  weight_decay: 0.0500 (0.0500)  time: 0.5219  data: 0.0177  max mem: 41794
Epoch: [132]  [ 50/312]  eta: 0:04:22  lr: 0.002615  min_lr: 0.002615  loss: 3.3299 (3.4127)  weight_decay: 0.0500 (0.0500)  time: 0.6615  data: 0.1001  max mem: 41794
Epoch: [132]  [ 60/312]  eta: 0:03:51  lr: 0.002614  min_lr: 0.002614  loss: 3.4007 (3.4010)  weight_decay: 0.0500 (0.0500)  time: 0.6547  data: 0.1003  max mem: 41794
Epoch: [132]  [ 70/312]  eta: 0:03:35  lr: 0.002614  min_lr: 0.002614  loss: 3.3806 (3.3920)  weight_decay: 0.0500 (0.0500)  time: 0.6096  data: 0.1039  max mem: 41794
Epoch: [132]  [ 80/312]  eta: 0:03:23  lr: 0.002613  min_lr: 0.002613  loss: 3.5437 (3.3900)  weight_decay: 0.0500 (0.0500)  time: 0.7622  data: 0.2575  max mem: 41794
Epoch: [132]  [ 90/312]  eta: 0:03:06  lr: 0.002612  min_lr: 0.002612  loss: 3.5953 (3.4208)  weight_decay: 0.0500 (0.0500)  time: 0.6559  data: 0.1616  max mem: 41794
Epoch: [132]  [100/312]  eta: 0:02:59  lr: 0.002612  min_lr: 0.002612  loss: 3.7555 (3.4635)  weight_decay: 0.0500 (0.0500)  time: 0.7075  data: 0.1925  max mem: 41794
Epoch: [132]  [110/312]  eta: 0:02:44  lr: 0.002611  min_lr: 0.002611  loss: 3.8012 (3.4786)  weight_decay: 0.0500 (0.0500)  time: 0.7007  data: 0.1870  max mem: 41794
Epoch: [132]  [120/312]  eta: 0:02:37  lr: 0.002610  min_lr: 0.002610  loss: 3.7435 (3.4786)  weight_decay: 0.0500 (0.0500)  time: 0.6876  data: 0.1517  max mem: 41794
Epoch: [132]  [130/312]  eta: 0:02:30  lr: 0.002609  min_lr: 0.002609  loss: 3.6232 (3.4755)  weight_decay: 0.0500 (0.0500)  time: 0.8828  data: 0.3058  max mem: 41794
Epoch: [132]  [140/312]  eta: 0:02:17  lr: 0.002609  min_lr: 0.002609  loss: 3.5923 (3.4682)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.1596  max mem: 41794
Epoch: [132]  [150/312]  eta: 0:02:11  lr: 0.002608  min_lr: 0.002608  loss: 3.5847 (3.4641)  weight_decay: 0.0500 (0.0500)  time: 0.7135  data: 0.1454  max mem: 41794
Epoch: [132]  [160/312]  eta: 0:02:01  lr: 0.002607  min_lr: 0.002607  loss: 3.4564 (3.4480)  weight_decay: 0.0500 (0.0500)  time: 0.8021  data: 0.1862  max mem: 41794
Epoch: [132]  [170/312]  eta: 0:01:52  lr: 0.002607  min_lr: 0.002607  loss: 3.4072 (3.4413)  weight_decay: 0.0500 (0.0500)  time: 0.6388  data: 0.0965  max mem: 41794
Epoch: [132]  [180/312]  eta: 0:01:44  lr: 0.002606  min_lr: 0.002606  loss: 3.6108 (3.4537)  weight_decay: 0.0500 (0.0500)  time: 0.7234  data: 0.1488  max mem: 41794
Epoch: [132]  [190/312]  eta: 0:01:34  lr: 0.002605  min_lr: 0.002605  loss: 3.7796 (3.4629)  weight_decay: 0.0500 (0.0500)  time: 0.6708  data: 0.1001  max mem: 41794
Epoch: [132]  [200/312]  eta: 0:01:27  lr: 0.002605  min_lr: 0.002605  loss: 3.6450 (3.4679)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.1107  max mem: 41794
Epoch: [132]  [210/312]  eta: 0:01:20  lr: 0.002604  min_lr: 0.002604  loss: 3.5836 (3.4722)  weight_decay: 0.0500 (0.0500)  time: 0.8900  data: 0.1928  max mem: 41794
Epoch: [132]  [220/312]  eta: 0:01:11  lr: 0.002603  min_lr: 0.002603  loss: 3.4598 (3.4732)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.0844  max mem: 41794
Epoch: [132]  [230/312]  eta: 0:01:03  lr: 0.002603  min_lr: 0.002603  loss: 3.5316 (3.4777)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.0666  max mem: 41794
Epoch: [132]  [240/312]  eta: 0:00:55  lr: 0.002602  min_lr: 0.002602  loss: 3.3556 (3.4649)  weight_decay: 0.0500 (0.0500)  time: 0.7687  data: 0.0666  max mem: 41794
Epoch: [132]  [250/312]  eta: 0:00:47  lr: 0.002601  min_lr: 0.002601  loss: 3.3556 (3.4671)  weight_decay: 0.0500 (0.0500)  time: 0.6770  data: 0.0885  max mem: 41794
Epoch: [132]  [260/312]  eta: 0:00:40  lr: 0.002601  min_lr: 0.002601  loss: 3.5372 (3.4754)  weight_decay: 0.0500 (0.0500)  time: 0.7830  data: 0.1749  max mem: 41794
Epoch: [132]  [270/312]  eta: 0:00:32  lr: 0.002600  min_lr: 0.002600  loss: 3.6673 (3.4727)  weight_decay: 0.0500 (0.0500)  time: 0.7010  data: 0.0871  max mem: 41794
Epoch: [132]  [280/312]  eta: 0:00:24  lr: 0.002599  min_lr: 0.002599  loss: 3.5015 (3.4624)  weight_decay: 0.0500 (0.0500)  time: 0.6588  data: 0.0638  max mem: 41794
Epoch: [132]  [290/312]  eta: 0:00:16  lr: 0.002599  min_lr: 0.002599  loss: 3.5015 (3.4632)  weight_decay: 0.0500 (0.0500)  time: 0.8668  data: 0.1457  max mem: 41794
Epoch: [132]  [300/312]  eta: 0:00:09  lr: 0.002598  min_lr: 0.002598  loss: 3.5635 (3.4661)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.0823  max mem: 41794
Epoch: [132]  [310/312]  eta: 0:00:01  lr: 0.002597  min_lr: 0.002597  loss: 3.5926 (3.4656)  weight_decay: 0.0500 (0.0500)  time: 0.4714  data: 0.0001  max mem: 41794
Epoch: [132]  [311/312]  eta: 0:00:00  lr: 0.002597  min_lr: 0.002597  loss: 3.5926 (3.4671)  weight_decay: 0.0500 (0.0500)  time: 0.4689  data: 0.0001  max mem: 41794
Epoch: [132] Total time: 0:03:56 (0.7589 s / it)
Averaged stats: lr: 0.002597  min_lr: 0.002597  loss: 3.5926 (3.4982)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.2374 (1.2374)  acc1: 77.9948 (77.9948)  acc5: 93.8802 (93.8802)  time: 8.0694  data: 7.8675  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5554 (1.5087)  acc1: 70.8333 (71.7280)  acc5: 90.7552 (91.3120)  time: 1.0446  data: 0.8743  max mem: 41794
Test: Total time: 0:00:09 (1.0579 s / it)
* Acc@1 72.130 Acc@5 91.246 loss 1.501
Accuracy of the model on the 50000 test images: 72.1%
Max accuracy: 72.13%
Epoch: [133]  [  0/312]  eta: 1:19:48  lr: 0.002597  min_lr: 0.002597  loss: 2.3565 (2.3565)  weight_decay: 0.0500 (0.0500)  time: 15.3489  data: 14.0952  max mem: 41794
Epoch: [133]  [ 10/312]  eta: 0:10:54  lr: 0.002596  min_lr: 0.002596  loss: 3.6023 (3.2661)  weight_decay: 0.0500 (0.0500)  time: 2.1681  data: 1.3957  max mem: 41794
Epoch: [133]  [ 20/312]  eta: 0:07:22  lr: 0.002596  min_lr: 0.002596  loss: 3.6646 (3.4086)  weight_decay: 0.0500 (0.0500)  time: 0.8248  data: 0.1521  max mem: 41794
Epoch: [133]  [ 30/312]  eta: 0:05:36  lr: 0.002595  min_lr: 0.002595  loss: 3.6301 (3.3388)  weight_decay: 0.0500 (0.0500)  time: 0.6560  data: 0.0895  max mem: 41794
Epoch: [133]  [ 40/312]  eta: 0:04:41  lr: 0.002594  min_lr: 0.002594  loss: 3.5520 (3.3712)  weight_decay: 0.0500 (0.0500)  time: 0.5256  data: 0.0146  max mem: 41794
Epoch: [133]  [ 50/312]  eta: 0:04:22  lr: 0.002594  min_lr: 0.002594  loss: 3.5027 (3.3704)  weight_decay: 0.0500 (0.0500)  time: 0.7085  data: 0.1231  max mem: 41794
Epoch: [133]  [ 60/312]  eta: 0:03:51  lr: 0.002593  min_lr: 0.002593  loss: 3.5098 (3.3851)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.1091  max mem: 41794
Epoch: [133]  [ 70/312]  eta: 0:03:44  lr: 0.002592  min_lr: 0.002592  loss: 3.6694 (3.4495)  weight_decay: 0.0500 (0.0500)  time: 0.7352  data: 0.0652  max mem: 41794
Epoch: [133]  [ 80/312]  eta: 0:03:29  lr: 0.002592  min_lr: 0.002592  loss: 3.6150 (3.4635)  weight_decay: 0.0500 (0.0500)  time: 0.8561  data: 0.0719  max mem: 41794
Epoch: [133]  [ 90/312]  eta: 0:03:14  lr: 0.002591  min_lr: 0.002591  loss: 3.5885 (3.4796)  weight_decay: 0.0500 (0.0500)  time: 0.6908  data: 0.0628  max mem: 41794
Epoch: [133]  [100/312]  eta: 0:03:06  lr: 0.002590  min_lr: 0.002590  loss: 3.6632 (3.4859)  weight_decay: 0.0500 (0.0500)  time: 0.7855  data: 0.1122  max mem: 41794
Epoch: [133]  [110/312]  eta: 0:02:50  lr: 0.002589  min_lr: 0.002589  loss: 3.4615 (3.4539)  weight_decay: 0.0500 (0.0500)  time: 0.7143  data: 0.0587  max mem: 41794
Epoch: [133]  [120/312]  eta: 0:02:40  lr: 0.002589  min_lr: 0.002589  loss: 3.3183 (3.4570)  weight_decay: 0.0500 (0.0500)  time: 0.6259  data: 0.0518  max mem: 41794
Epoch: [133]  [130/312]  eta: 0:02:32  lr: 0.002588  min_lr: 0.002588  loss: 3.6448 (3.4594)  weight_decay: 0.0500 (0.0500)  time: 0.8057  data: 0.1593  max mem: 41794
Epoch: [133]  [140/312]  eta: 0:02:20  lr: 0.002587  min_lr: 0.002587  loss: 3.4582 (3.4458)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.1114  max mem: 41794
Epoch: [133]  [150/312]  eta: 0:02:12  lr: 0.002587  min_lr: 0.002587  loss: 3.3150 (3.4278)  weight_decay: 0.0500 (0.0500)  time: 0.6568  data: 0.1313  max mem: 41794
Epoch: [133]  [160/312]  eta: 0:02:03  lr: 0.002586  min_lr: 0.002586  loss: 3.3173 (3.4344)  weight_decay: 0.0500 (0.0500)  time: 0.7630  data: 0.1777  max mem: 41794
Epoch: [133]  [170/312]  eta: 0:01:54  lr: 0.002585  min_lr: 0.002585  loss: 3.6823 (3.4328)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.1492  max mem: 41794
Epoch: [133]  [180/312]  eta: 0:01:47  lr: 0.002585  min_lr: 0.002585  loss: 3.6823 (3.4359)  weight_decay: 0.0500 (0.0500)  time: 0.8378  data: 0.2339  max mem: 41794
Epoch: [133]  [190/312]  eta: 0:01:37  lr: 0.002584  min_lr: 0.002584  loss: 3.5806 (3.4295)  weight_decay: 0.0500 (0.0500)  time: 0.7357  data: 0.1330  max mem: 41794
Epoch: [133]  [200/312]  eta: 0:01:29  lr: 0.002583  min_lr: 0.002583  loss: 3.4532 (3.4385)  weight_decay: 0.0500 (0.0500)  time: 0.6495  data: 0.0935  max mem: 41794
Epoch: [133]  [210/312]  eta: 0:01:21  lr: 0.002583  min_lr: 0.002583  loss: 3.4532 (3.4319)  weight_decay: 0.0500 (0.0500)  time: 0.8098  data: 0.2020  max mem: 41794
Epoch: [133]  [220/312]  eta: 0:01:12  lr: 0.002582  min_lr: 0.002582  loss: 3.3714 (3.4307)  weight_decay: 0.0500 (0.0500)  time: 0.6564  data: 0.1092  max mem: 41794
Epoch: [133]  [230/312]  eta: 0:01:04  lr: 0.002581  min_lr: 0.002581  loss: 3.3794 (3.4292)  weight_decay: 0.0500 (0.0500)  time: 0.7139  data: 0.1300  max mem: 41794
Epoch: [133]  [240/312]  eta: 0:00:56  lr: 0.002581  min_lr: 0.002581  loss: 3.5504 (3.4278)  weight_decay: 0.0500 (0.0500)  time: 0.8200  data: 0.1315  max mem: 41794
Epoch: [133]  [250/312]  eta: 0:00:48  lr: 0.002580  min_lr: 0.002580  loss: 3.5925 (3.4294)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.0953  max mem: 41794
Epoch: [133]  [260/312]  eta: 0:00:40  lr: 0.002579  min_lr: 0.002579  loss: 3.5762 (3.4207)  weight_decay: 0.0500 (0.0500)  time: 0.7970  data: 0.2108  max mem: 41794
Epoch: [133]  [270/312]  eta: 0:00:32  lr: 0.002578  min_lr: 0.002578  loss: 3.4032 (3.4219)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1178  max mem: 41794
Epoch: [133]  [280/312]  eta: 0:00:24  lr: 0.002578  min_lr: 0.002578  loss: 3.4339 (3.4167)  weight_decay: 0.0500 (0.0500)  time: 0.6733  data: 0.0989  max mem: 41794
Epoch: [133]  [290/312]  eta: 0:00:17  lr: 0.002577  min_lr: 0.002577  loss: 3.6258 (3.4222)  weight_decay: 0.0500 (0.0500)  time: 0.7992  data: 0.1719  max mem: 41794
Epoch: [133]  [300/312]  eta: 0:00:09  lr: 0.002576  min_lr: 0.002576  loss: 3.6258 (3.4259)  weight_decay: 0.0500 (0.0500)  time: 0.6131  data: 0.0735  max mem: 41794
Epoch: [133]  [310/312]  eta: 0:00:01  lr: 0.002576  min_lr: 0.002576  loss: 3.4356 (3.4215)  weight_decay: 0.0500 (0.0500)  time: 0.4623  data: 0.0001  max mem: 41794
Epoch: [133]  [311/312]  eta: 0:00:00  lr: 0.002576  min_lr: 0.002576  loss: 3.4607 (3.4216)  weight_decay: 0.0500 (0.0500)  time: 0.4612  data: 0.0001  max mem: 41794
Epoch: [133] Total time: 0:03:58 (0.7635 s / it)
Averaged stats: lr: 0.002576  min_lr: 0.002576  loss: 3.4607 (3.4607)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.1379 (1.1379)  acc1: 77.6042 (77.6042)  acc5: 92.7083 (92.7083)  time: 8.6560  data: 8.4581  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4249 (1.3521)  acc1: 69.6615 (71.2640)  acc5: 91.2760 (90.8160)  time: 1.1176  data: 0.9399  max mem: 41794
Test: Total time: 0:00:10 (1.1654 s / it)
* Acc@1 71.482 Acc@5 90.958 loss 1.341
Accuracy of the model on the 50000 test images: 71.5%
Max accuracy: 72.13%
Epoch: [134]  [  0/312]  eta: 1:20:02  lr: 0.002576  min_lr: 0.002576  loss: 4.1825 (4.1825)  weight_decay: 0.0500 (0.0500)  time: 15.3928  data: 11.6502  max mem: 41794
Epoch: [134]  [ 10/312]  eta: 0:11:31  lr: 0.002575  min_lr: 0.002575  loss: 3.6192 (3.3857)  weight_decay: 0.0500 (0.0500)  time: 2.2900  data: 1.2854  max mem: 41794
Epoch: [134]  [ 20/312]  eta: 0:07:16  lr: 0.002574  min_lr: 0.002574  loss: 3.4864 (3.3074)  weight_decay: 0.0500 (0.0500)  time: 0.7988  data: 0.1328  max mem: 41794
Epoch: [134]  [ 30/312]  eta: 0:05:32  lr: 0.002573  min_lr: 0.002573  loss: 2.9201 (3.2536)  weight_decay: 0.0500 (0.0500)  time: 0.5686  data: 0.0126  max mem: 41794
Epoch: [134]  [ 40/312]  eta: 0:04:37  lr: 0.002573  min_lr: 0.002573  loss: 3.3229 (3.3217)  weight_decay: 0.0500 (0.0500)  time: 0.5216  data: 0.0077  max mem: 41794
Epoch: [134]  [ 50/312]  eta: 0:04:18  lr: 0.002572  min_lr: 0.002572  loss: 3.5970 (3.3469)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.0758  max mem: 41794
Epoch: [134]  [ 60/312]  eta: 0:03:48  lr: 0.002571  min_lr: 0.002571  loss: 3.4828 (3.3541)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.0738  max mem: 41794
Epoch: [134]  [ 70/312]  eta: 0:03:41  lr: 0.002571  min_lr: 0.002571  loss: 3.5180 (3.3734)  weight_decay: 0.0500 (0.0500)  time: 0.7286  data: 0.0929  max mem: 41794
Epoch: [134]  [ 80/312]  eta: 0:03:25  lr: 0.002570  min_lr: 0.002570  loss: 3.5316 (3.3858)  weight_decay: 0.0500 (0.0500)  time: 0.8238  data: 0.1318  max mem: 41794
Epoch: [134]  [ 90/312]  eta: 0:03:08  lr: 0.002569  min_lr: 0.002569  loss: 3.5989 (3.3939)  weight_decay: 0.0500 (0.0500)  time: 0.6087  data: 0.0552  max mem: 41794
Epoch: [134]  [100/312]  eta: 0:03:00  lr: 0.002569  min_lr: 0.002569  loss: 3.8029 (3.4402)  weight_decay: 0.0500 (0.0500)  time: 0.7158  data: 0.1583  max mem: 41794
Epoch: [134]  [110/312]  eta: 0:02:45  lr: 0.002568  min_lr: 0.002568  loss: 3.8029 (3.4381)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.1482  max mem: 41794
Epoch: [134]  [120/312]  eta: 0:02:37  lr: 0.002567  min_lr: 0.002567  loss: 3.4169 (3.4325)  weight_decay: 0.0500 (0.0500)  time: 0.6683  data: 0.1714  max mem: 41794
Epoch: [134]  [130/312]  eta: 0:02:30  lr: 0.002567  min_lr: 0.002567  loss: 3.3862 (3.4239)  weight_decay: 0.0500 (0.0500)  time: 0.8450  data: 0.3531  max mem: 41794
Epoch: [134]  [140/312]  eta: 0:02:18  lr: 0.002566  min_lr: 0.002566  loss: 3.3862 (3.4334)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1910  max mem: 41794
Epoch: [134]  [150/312]  eta: 0:02:10  lr: 0.002565  min_lr: 0.002565  loss: 3.6061 (3.4527)  weight_decay: 0.0500 (0.0500)  time: 0.6905  data: 0.1845  max mem: 41794
Epoch: [134]  [160/312]  eta: 0:02:03  lr: 0.002565  min_lr: 0.002565  loss: 3.6673 (3.4573)  weight_decay: 0.0500 (0.0500)  time: 0.8687  data: 0.3675  max mem: 41794
Epoch: [134]  [170/312]  eta: 0:01:52  lr: 0.002564  min_lr: 0.002564  loss: 3.6397 (3.4535)  weight_decay: 0.0500 (0.0500)  time: 0.6946  data: 0.1926  max mem: 41794
Epoch: [134]  [180/312]  eta: 0:01:45  lr: 0.002563  min_lr: 0.002563  loss: 3.6779 (3.4760)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.1533  max mem: 41794
Epoch: [134]  [190/312]  eta: 0:01:35  lr: 0.002562  min_lr: 0.002562  loss: 3.7865 (3.4730)  weight_decay: 0.0500 (0.0500)  time: 0.6575  data: 0.1500  max mem: 41794
Epoch: [134]  [200/312]  eta: 0:01:27  lr: 0.002562  min_lr: 0.002562  loss: 3.5065 (3.4785)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.1577  max mem: 41794
Epoch: [134]  [210/312]  eta: 0:01:20  lr: 0.002561  min_lr: 0.002561  loss: 3.5065 (3.4703)  weight_decay: 0.0500 (0.0500)  time: 0.8551  data: 0.3078  max mem: 41794
Epoch: [134]  [220/312]  eta: 0:01:11  lr: 0.002560  min_lr: 0.002560  loss: 3.4715 (3.4684)  weight_decay: 0.0500 (0.0500)  time: 0.6772  data: 0.1521  max mem: 41794
Epoch: [134]  [230/312]  eta: 0:01:03  lr: 0.002560  min_lr: 0.002560  loss: 3.5280 (3.4788)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.1449  max mem: 41794
Epoch: [134]  [240/312]  eta: 0:00:55  lr: 0.002559  min_lr: 0.002559  loss: 3.6736 (3.4815)  weight_decay: 0.0500 (0.0500)  time: 0.8202  data: 0.2924  max mem: 41794
Epoch: [134]  [250/312]  eta: 0:00:47  lr: 0.002558  min_lr: 0.002558  loss: 3.6506 (3.4845)  weight_decay: 0.0500 (0.0500)  time: 0.7038  data: 0.1499  max mem: 41794
Epoch: [134]  [260/312]  eta: 0:00:40  lr: 0.002558  min_lr: 0.002558  loss: 3.5494 (3.4851)  weight_decay: 0.0500 (0.0500)  time: 0.7167  data: 0.1147  max mem: 41794
Epoch: [134]  [270/312]  eta: 0:00:32  lr: 0.002557  min_lr: 0.002557  loss: 3.2964 (3.4815)  weight_decay: 0.0500 (0.0500)  time: 0.6565  data: 0.1145  max mem: 41794
Epoch: [134]  [280/312]  eta: 0:00:24  lr: 0.002556  min_lr: 0.002556  loss: 3.5971 (3.4793)  weight_decay: 0.0500 (0.0500)  time: 0.7857  data: 0.1266  max mem: 41794
Epoch: [134]  [290/312]  eta: 0:00:17  lr: 0.002556  min_lr: 0.002556  loss: 3.6374 (3.4723)  weight_decay: 0.0500 (0.0500)  time: 0.9448  data: 0.1569  max mem: 41794
Epoch: [134]  [300/312]  eta: 0:00:09  lr: 0.002555  min_lr: 0.002555  loss: 3.4187 (3.4712)  weight_decay: 0.0500 (0.0500)  time: 0.6400  data: 0.0323  max mem: 41794
Epoch: [134]  [310/312]  eta: 0:00:01  lr: 0.002554  min_lr: 0.002554  loss: 3.5458 (3.4807)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [134]  [311/312]  eta: 0:00:00  lr: 0.002554  min_lr: 0.002554  loss: 3.4908 (3.4787)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0001  max mem: 41794
Epoch: [134] Total time: 0:03:57 (0.7610 s / it)
Averaged stats: lr: 0.002554  min_lr: 0.002554  loss: 3.4908 (3.4862)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3175 (1.3175)  acc1: 75.3906 (75.3906)  acc5: 94.1406 (94.1406)  time: 8.4644  data: 8.2525  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6754 (1.5326)  acc1: 70.1823 (70.8800)  acc5: 91.1458 (91.2000)  time: 1.0886  data: 0.9170  max mem: 41794
Test: Total time: 0:00:10 (1.1300 s / it)
* Acc@1 71.510 Acc@5 91.090 loss 1.534
Accuracy of the model on the 50000 test images: 71.5%
Max accuracy: 72.13%
Epoch: [135]  [  0/312]  eta: 1:22:04  lr: 0.002554  min_lr: 0.002554  loss: 4.1152 (4.1152)  weight_decay: 0.0500 (0.0500)  time: 15.7847  data: 11.6057  max mem: 41794
Epoch: [135]  [ 10/312]  eta: 0:11:39  lr: 0.002553  min_lr: 0.002553  loss: 3.3445 (3.3412)  weight_decay: 0.0500 (0.0500)  time: 2.3166  data: 1.2288  max mem: 41794
Epoch: [135]  [ 20/312]  eta: 0:07:25  lr: 0.002553  min_lr: 0.002553  loss: 3.4431 (3.5108)  weight_decay: 0.0500 (0.0500)  time: 0.8133  data: 0.1115  max mem: 41794
Epoch: [135]  [ 30/312]  eta: 0:05:38  lr: 0.002552  min_lr: 0.002552  loss: 3.6243 (3.4948)  weight_decay: 0.0500 (0.0500)  time: 0.5890  data: 0.0285  max mem: 41794
Epoch: [135]  [ 40/312]  eta: 0:04:40  lr: 0.002551  min_lr: 0.002551  loss: 3.4485 (3.4393)  weight_decay: 0.0500 (0.0500)  time: 0.5109  data: 0.0164  max mem: 41794
Epoch: [135]  [ 50/312]  eta: 0:04:11  lr: 0.002551  min_lr: 0.002551  loss: 3.4485 (3.4293)  weight_decay: 0.0500 (0.0500)  time: 0.5892  data: 0.0699  max mem: 41794
Epoch: [135]  [ 60/312]  eta: 0:03:44  lr: 0.002550  min_lr: 0.002550  loss: 3.6609 (3.4725)  weight_decay: 0.0500 (0.0500)  time: 0.6007  data: 0.0752  max mem: 41794
Epoch: [135]  [ 70/312]  eta: 0:03:33  lr: 0.002549  min_lr: 0.002549  loss: 3.7606 (3.4751)  weight_decay: 0.0500 (0.0500)  time: 0.6747  data: 0.1732  max mem: 41794
Epoch: [135]  [ 80/312]  eta: 0:03:24  lr: 0.002549  min_lr: 0.002549  loss: 3.8227 (3.5066)  weight_decay: 0.0500 (0.0500)  time: 0.8497  data: 0.3557  max mem: 41794
Epoch: [135]  [ 90/312]  eta: 0:03:06  lr: 0.002548  min_lr: 0.002548  loss: 3.6230 (3.4849)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.1961  max mem: 41794
Epoch: [135]  [100/312]  eta: 0:02:57  lr: 0.002547  min_lr: 0.002547  loss: 3.4658 (3.4962)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.1599  max mem: 41794
Epoch: [135]  [110/312]  eta: 0:02:43  lr: 0.002546  min_lr: 0.002546  loss: 3.5728 (3.4983)  weight_decay: 0.0500 (0.0500)  time: 0.6692  data: 0.1582  max mem: 41794
Epoch: [135]  [120/312]  eta: 0:02:36  lr: 0.002546  min_lr: 0.002546  loss: 3.6738 (3.5069)  weight_decay: 0.0500 (0.0500)  time: 0.6942  data: 0.1744  max mem: 41794
Epoch: [135]  [130/312]  eta: 0:02:28  lr: 0.002545  min_lr: 0.002545  loss: 3.5058 (3.4932)  weight_decay: 0.0500 (0.0500)  time: 0.8480  data: 0.3302  max mem: 41794
Epoch: [135]  [140/312]  eta: 0:02:16  lr: 0.002544  min_lr: 0.002544  loss: 3.6047 (3.5196)  weight_decay: 0.0500 (0.0500)  time: 0.6585  data: 0.1589  max mem: 41794
Epoch: [135]  [150/312]  eta: 0:02:09  lr: 0.002544  min_lr: 0.002544  loss: 3.5849 (3.4925)  weight_decay: 0.0500 (0.0500)  time: 0.7054  data: 0.1910  max mem: 41794
Epoch: [135]  [160/312]  eta: 0:02:01  lr: 0.002543  min_lr: 0.002543  loss: 3.1733 (3.4901)  weight_decay: 0.0500 (0.0500)  time: 0.8535  data: 0.3365  max mem: 41794
Epoch: [135]  [170/312]  eta: 0:01:51  lr: 0.002542  min_lr: 0.002542  loss: 3.6343 (3.4773)  weight_decay: 0.0500 (0.0500)  time: 0.6581  data: 0.1586  max mem: 41794
Epoch: [135]  [180/312]  eta: 0:01:44  lr: 0.002542  min_lr: 0.002542  loss: 3.6343 (3.4768)  weight_decay: 0.0500 (0.0500)  time: 0.7163  data: 0.2173  max mem: 41794
Epoch: [135]  [190/312]  eta: 0:01:34  lr: 0.002541  min_lr: 0.002541  loss: 3.7081 (3.4843)  weight_decay: 0.0500 (0.0500)  time: 0.6991  data: 0.2096  max mem: 41794
Epoch: [135]  [200/312]  eta: 0:01:27  lr: 0.002540  min_lr: 0.002540  loss: 3.5823 (3.4918)  weight_decay: 0.0500 (0.0500)  time: 0.6673  data: 0.1831  max mem: 41794
Epoch: [135]  [210/312]  eta: 0:01:19  lr: 0.002540  min_lr: 0.002540  loss: 3.7753 (3.5042)  weight_decay: 0.0500 (0.0500)  time: 0.8599  data: 0.3688  max mem: 41794
Epoch: [135]  [220/312]  eta: 0:01:10  lr: 0.002539  min_lr: 0.002539  loss: 3.6995 (3.5047)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.1911  max mem: 41794
Epoch: [135]  [230/312]  eta: 0:01:03  lr: 0.002538  min_lr: 0.002538  loss: 3.6041 (3.5016)  weight_decay: 0.0500 (0.0500)  time: 0.6706  data: 0.1696  max mem: 41794
Epoch: [135]  [240/312]  eta: 0:00:55  lr: 0.002537  min_lr: 0.002537  loss: 3.6068 (3.4993)  weight_decay: 0.0500 (0.0500)  time: 0.8571  data: 0.3649  max mem: 41794
Epoch: [135]  [250/312]  eta: 0:00:47  lr: 0.002537  min_lr: 0.002537  loss: 3.5975 (3.4962)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1961  max mem: 41794
Epoch: [135]  [260/312]  eta: 0:00:40  lr: 0.002536  min_lr: 0.002536  loss: 3.5918 (3.4912)  weight_decay: 0.0500 (0.0500)  time: 0.6808  data: 0.1845  max mem: 41794
Epoch: [135]  [270/312]  eta: 0:00:31  lr: 0.002535  min_lr: 0.002535  loss: 3.5119 (3.4884)  weight_decay: 0.0500 (0.0500)  time: 0.6819  data: 0.1845  max mem: 41794
Epoch: [135]  [280/312]  eta: 0:00:24  lr: 0.002535  min_lr: 0.002535  loss: 3.5742 (3.4928)  weight_decay: 0.0500 (0.0500)  time: 0.6823  data: 0.1895  max mem: 41794
Epoch: [135]  [290/312]  eta: 0:00:16  lr: 0.002534  min_lr: 0.002534  loss: 3.7421 (3.4986)  weight_decay: 0.0500 (0.0500)  time: 0.8440  data: 0.3467  max mem: 41794
Epoch: [135]  [300/312]  eta: 0:00:09  lr: 0.002533  min_lr: 0.002533  loss: 3.5908 (3.4957)  weight_decay: 0.0500 (0.0500)  time: 0.6527  data: 0.1575  max mem: 41794
Epoch: [135]  [310/312]  eta: 0:00:01  lr: 0.002533  min_lr: 0.002533  loss: 3.5350 (3.4957)  weight_decay: 0.0500 (0.0500)  time: 0.4722  data: 0.0001  max mem: 41794
Epoch: [135]  [311/312]  eta: 0:00:00  lr: 0.002533  min_lr: 0.002533  loss: 3.5350 (3.4961)  weight_decay: 0.0500 (0.0500)  time: 0.4699  data: 0.0001  max mem: 41794
Epoch: [135] Total time: 0:03:54 (0.7529 s / it)
Averaged stats: lr: 0.002533  min_lr: 0.002533  loss: 3.5350 (3.4729)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.2617 (1.2617)  acc1: 77.7344 (77.7344)  acc5: 93.8802 (93.8802)  time: 8.6311  data: 8.4232  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4775 (1.4094)  acc1: 70.5729 (71.6000)  acc5: 91.9271 (91.0880)  time: 1.1097  data: 0.9360  max mem: 41794
Test: Total time: 0:00:10 (1.1402 s / it)
* Acc@1 71.624 Acc@5 91.154 loss 1.408
Accuracy of the model on the 50000 test images: 71.6%
Max accuracy: 72.13%
Epoch: [136]  [  0/312]  eta: 1:18:27  lr: 0.002532  min_lr: 0.002532  loss: 3.8159 (3.8159)  weight_decay: 0.0500 (0.0500)  time: 15.0891  data: 13.3324  max mem: 41794
Epoch: [136]  [ 10/312]  eta: 0:11:18  lr: 0.002532  min_lr: 0.002532  loss: 3.3971 (3.3639)  weight_decay: 0.0500 (0.0500)  time: 2.2464  data: 1.2615  max mem: 41794
Epoch: [136]  [ 20/312]  eta: 0:07:31  lr: 0.002531  min_lr: 0.002531  loss: 3.5184 (3.4624)  weight_decay: 0.0500 (0.0500)  time: 0.8708  data: 0.1008  max mem: 41794
Epoch: [136]  [ 30/312]  eta: 0:05:44  lr: 0.002530  min_lr: 0.002530  loss: 3.5989 (3.4589)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.0806  max mem: 41794
Epoch: [136]  [ 40/312]  eta: 0:04:55  lr: 0.002530  min_lr: 0.002530  loss: 3.5996 (3.4472)  weight_decay: 0.0500 (0.0500)  time: 0.5984  data: 0.0539  max mem: 41794
Epoch: [136]  [ 50/312]  eta: 0:04:35  lr: 0.002529  min_lr: 0.002529  loss: 3.5599 (3.4727)  weight_decay: 0.0500 (0.0500)  time: 0.7866  data: 0.1639  max mem: 41794
Epoch: [136]  [ 60/312]  eta: 0:04:02  lr: 0.002528  min_lr: 0.002528  loss: 3.7027 (3.4821)  weight_decay: 0.0500 (0.0500)  time: 0.7074  data: 0.1173  max mem: 41794
Epoch: [136]  [ 70/312]  eta: 0:03:49  lr: 0.002528  min_lr: 0.002528  loss: 3.7756 (3.5120)  weight_decay: 0.0500 (0.0500)  time: 0.6837  data: 0.1028  max mem: 41794
Epoch: [136]  [ 80/312]  eta: 0:03:30  lr: 0.002527  min_lr: 0.002527  loss: 3.7756 (3.5165)  weight_decay: 0.0500 (0.0500)  time: 0.7396  data: 0.1129  max mem: 41794
Epoch: [136]  [ 90/312]  eta: 0:03:16  lr: 0.002526  min_lr: 0.002526  loss: 3.7339 (3.5188)  weight_decay: 0.0500 (0.0500)  time: 0.6659  data: 0.0822  max mem: 41794
Epoch: [136]  [100/312]  eta: 0:03:07  lr: 0.002526  min_lr: 0.002526  loss: 3.3751 (3.5049)  weight_decay: 0.0500 (0.0500)  time: 0.7946  data: 0.1657  max mem: 41794
Epoch: [136]  [110/312]  eta: 0:02:51  lr: 0.002525  min_lr: 0.002525  loss: 3.5809 (3.5127)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.0942  max mem: 41794
Epoch: [136]  [120/312]  eta: 0:02:43  lr: 0.002524  min_lr: 0.002524  loss: 3.6074 (3.5023)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.0755  max mem: 41794
Epoch: [136]  [130/312]  eta: 0:02:35  lr: 0.002523  min_lr: 0.002523  loss: 2.9467 (3.4552)  weight_decay: 0.0500 (0.0500)  time: 0.8873  data: 0.1211  max mem: 41794
Epoch: [136]  [140/312]  eta: 0:02:22  lr: 0.002523  min_lr: 0.002523  loss: 3.6271 (3.4766)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.0462  max mem: 41794
Epoch: [136]  [150/312]  eta: 0:02:14  lr: 0.002522  min_lr: 0.002522  loss: 3.6334 (3.4724)  weight_decay: 0.0500 (0.0500)  time: 0.6943  data: 0.0995  max mem: 41794
Epoch: [136]  [160/312]  eta: 0:02:04  lr: 0.002521  min_lr: 0.002521  loss: 3.5773 (3.4720)  weight_decay: 0.0500 (0.0500)  time: 0.7402  data: 0.1023  max mem: 41794
Epoch: [136]  [170/312]  eta: 0:01:55  lr: 0.002521  min_lr: 0.002521  loss: 3.6195 (3.4631)  weight_decay: 0.0500 (0.0500)  time: 0.6687  data: 0.0848  max mem: 41794
Epoch: [136]  [180/312]  eta: 0:01:47  lr: 0.002520  min_lr: 0.002520  loss: 3.4006 (3.4535)  weight_decay: 0.0500 (0.0500)  time: 0.8165  data: 0.1769  max mem: 41794
Epoch: [136]  [190/312]  eta: 0:01:37  lr: 0.002519  min_lr: 0.002519  loss: 3.4235 (3.4561)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.0970  max mem: 41794
Epoch: [136]  [200/312]  eta: 0:01:29  lr: 0.002519  min_lr: 0.002519  loss: 3.6025 (3.4535)  weight_decay: 0.0500 (0.0500)  time: 0.6713  data: 0.0662  max mem: 41794
Epoch: [136]  [210/312]  eta: 0:01:22  lr: 0.002518  min_lr: 0.002518  loss: 3.8056 (3.4600)  weight_decay: 0.0500 (0.0500)  time: 0.8740  data: 0.1131  max mem: 41794
Epoch: [136]  [220/312]  eta: 0:01:13  lr: 0.002517  min_lr: 0.002517  loss: 3.8102 (3.4723)  weight_decay: 0.0500 (0.0500)  time: 0.7010  data: 0.0491  max mem: 41794
Epoch: [136]  [230/312]  eta: 0:01:05  lr: 0.002516  min_lr: 0.002516  loss: 3.6871 (3.4622)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.0576  max mem: 41794
Epoch: [136]  [240/312]  eta: 0:00:56  lr: 0.002516  min_lr: 0.002516  loss: 3.4253 (3.4552)  weight_decay: 0.0500 (0.0500)  time: 0.7187  data: 0.0621  max mem: 41794
Epoch: [136]  [250/312]  eta: 0:00:48  lr: 0.002515  min_lr: 0.002515  loss: 3.6458 (3.4539)  weight_decay: 0.0500 (0.0500)  time: 0.6685  data: 0.0619  max mem: 41794
Epoch: [136]  [260/312]  eta: 0:00:41  lr: 0.002514  min_lr: 0.002514  loss: 3.2878 (3.4488)  weight_decay: 0.0500 (0.0500)  time: 0.8187  data: 0.0882  max mem: 41794
Epoch: [136]  [270/312]  eta: 0:00:32  lr: 0.002514  min_lr: 0.002514  loss: 3.2895 (3.4476)  weight_decay: 0.0500 (0.0500)  time: 0.7255  data: 0.0804  max mem: 41794
Epoch: [136]  [280/312]  eta: 0:00:25  lr: 0.002513  min_lr: 0.002513  loss: 3.2132 (3.4428)  weight_decay: 0.0500 (0.0500)  time: 0.6835  data: 0.0906  max mem: 41794
Epoch: [136]  [290/312]  eta: 0:00:17  lr: 0.002512  min_lr: 0.002512  loss: 3.4911 (3.4464)  weight_decay: 0.0500 (0.0500)  time: 0.7685  data: 0.0437  max mem: 41794
Epoch: [136]  [300/312]  eta: 0:00:09  lr: 0.002512  min_lr: 0.002512  loss: 3.5964 (3.4507)  weight_decay: 0.0500 (0.0500)  time: 0.6233  data: 0.0112  max mem: 41794
Epoch: [136]  [310/312]  eta: 0:00:01  lr: 0.002511  min_lr: 0.002511  loss: 3.7391 (3.4582)  weight_decay: 0.0500 (0.0500)  time: 0.4719  data: 0.0089  max mem: 41794
Epoch: [136]  [311/312]  eta: 0:00:00  lr: 0.002511  min_lr: 0.002511  loss: 3.7391 (3.4585)  weight_decay: 0.0500 (0.0500)  time: 0.4714  data: 0.0089  max mem: 41794
Epoch: [136] Total time: 0:03:59 (0.7678 s / it)
Averaged stats: lr: 0.002511  min_lr: 0.002511  loss: 3.7391 (3.4700)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.3130 (1.3130)  acc1: 77.0833 (77.0833)  acc5: 94.1406 (94.1406)  time: 8.4990  data: 8.2870  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5805 (1.4712)  acc1: 70.4427 (71.1360)  acc5: 91.4062 (91.3440)  time: 1.0924  data: 0.9209  max mem: 41794
Test: Total time: 0:00:10 (1.1155 s / it)
* Acc@1 71.566 Acc@5 91.342 loss 1.467
Accuracy of the model on the 50000 test images: 71.6%
Max accuracy: 72.13%
Epoch: [137]  [  0/312]  eta: 1:24:51  lr: 0.002511  min_lr: 0.002511  loss: 3.9615 (3.9615)  weight_decay: 0.0500 (0.0500)  time: 16.3184  data: 15.3052  max mem: 41794
Epoch: [137]  [ 10/312]  eta: 0:11:42  lr: 0.002510  min_lr: 0.002510  loss: 3.4872 (3.3202)  weight_decay: 0.0500 (0.0500)  time: 2.3268  data: 1.4182  max mem: 41794
Epoch: [137]  [ 20/312]  eta: 0:07:40  lr: 0.002509  min_lr: 0.002509  loss: 3.4872 (3.4028)  weight_decay: 0.0500 (0.0500)  time: 0.8389  data: 0.1272  max mem: 41794
Epoch: [137]  [ 30/312]  eta: 0:05:51  lr: 0.002509  min_lr: 0.002509  loss: 3.5307 (3.3978)  weight_decay: 0.0500 (0.0500)  time: 0.6531  data: 0.1155  max mem: 41794
Epoch: [137]  [ 40/312]  eta: 0:04:50  lr: 0.002508  min_lr: 0.002508  loss: 3.4267 (3.3707)  weight_decay: 0.0500 (0.0500)  time: 0.5372  data: 0.0127  max mem: 41794
Epoch: [137]  [ 50/312]  eta: 0:04:21  lr: 0.002507  min_lr: 0.002507  loss: 3.3973 (3.3747)  weight_decay: 0.0500 (0.0500)  time: 0.6114  data: 0.1064  max mem: 41794
Epoch: [137]  [ 60/312]  eta: 0:03:50  lr: 0.002507  min_lr: 0.002507  loss: 3.5313 (3.4228)  weight_decay: 0.0500 (0.0500)  time: 0.6024  data: 0.0987  max mem: 41794
Epoch: [137]  [ 70/312]  eta: 0:03:38  lr: 0.002506  min_lr: 0.002506  loss: 3.5287 (3.4001)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.1477  max mem: 41794
Epoch: [137]  [ 80/312]  eta: 0:03:28  lr: 0.002505  min_lr: 0.002505  loss: 3.5480 (3.4384)  weight_decay: 0.0500 (0.0500)  time: 0.8408  data: 0.3335  max mem: 41794
Epoch: [137]  [ 90/312]  eta: 0:03:10  lr: 0.002505  min_lr: 0.002505  loss: 3.5940 (3.4319)  weight_decay: 0.0500 (0.0500)  time: 0.6959  data: 0.1925  max mem: 41794
Epoch: [137]  [100/312]  eta: 0:03:01  lr: 0.002504  min_lr: 0.002504  loss: 3.5430 (3.4405)  weight_decay: 0.0500 (0.0500)  time: 0.6955  data: 0.1904  max mem: 41794
Epoch: [137]  [110/312]  eta: 0:02:46  lr: 0.002503  min_lr: 0.002503  loss: 3.5273 (3.4331)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.1873  max mem: 41794
Epoch: [137]  [120/312]  eta: 0:02:38  lr: 0.002502  min_lr: 0.002502  loss: 3.7460 (3.4498)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.1751  max mem: 41794
Epoch: [137]  [130/312]  eta: 0:02:31  lr: 0.002502  min_lr: 0.002502  loss: 3.7603 (3.4614)  weight_decay: 0.0500 (0.0500)  time: 0.8602  data: 0.3560  max mem: 41794
Epoch: [137]  [140/312]  eta: 0:02:18  lr: 0.002501  min_lr: 0.002501  loss: 3.7066 (3.4529)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.1829  max mem: 41794
Epoch: [137]  [150/312]  eta: 0:02:11  lr: 0.002500  min_lr: 0.002500  loss: 3.0602 (3.4297)  weight_decay: 0.0500 (0.0500)  time: 0.6809  data: 0.1660  max mem: 41794
Epoch: [137]  [160/312]  eta: 0:02:03  lr: 0.002500  min_lr: 0.002500  loss: 3.4185 (3.4419)  weight_decay: 0.0500 (0.0500)  time: 0.8472  data: 0.3428  max mem: 41794
Epoch: [137]  [170/312]  eta: 0:01:52  lr: 0.002499  min_lr: 0.002499  loss: 3.6336 (3.4509)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.1902  max mem: 41794
Epoch: [137]  [180/312]  eta: 0:01:45  lr: 0.002498  min_lr: 0.002498  loss: 3.5093 (3.4385)  weight_decay: 0.0500 (0.0500)  time: 0.6874  data: 0.1875  max mem: 41794
Epoch: [137]  [190/312]  eta: 0:01:35  lr: 0.002498  min_lr: 0.002498  loss: 3.1852 (3.4336)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.1776  max mem: 41794
Epoch: [137]  [200/312]  eta: 0:01:28  lr: 0.002497  min_lr: 0.002497  loss: 3.0201 (3.4168)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.1961  max mem: 41794
Epoch: [137]  [210/312]  eta: 0:01:20  lr: 0.002496  min_lr: 0.002496  loss: 3.1296 (3.4226)  weight_decay: 0.0500 (0.0500)  time: 0.8588  data: 0.3529  max mem: 41794
Epoch: [137]  [220/312]  eta: 0:01:11  lr: 0.002495  min_lr: 0.002495  loss: 3.4763 (3.4168)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.1603  max mem: 41794
Epoch: [137]  [230/312]  eta: 0:01:04  lr: 0.002495  min_lr: 0.002495  loss: 3.4763 (3.4189)  weight_decay: 0.0500 (0.0500)  time: 0.7129  data: 0.1390  max mem: 41794
Epoch: [137]  [240/312]  eta: 0:00:56  lr: 0.002494  min_lr: 0.002494  loss: 3.6088 (3.4253)  weight_decay: 0.0500 (0.0500)  time: 0.8421  data: 0.2263  max mem: 41794
Epoch: [137]  [250/312]  eta: 0:00:48  lr: 0.002493  min_lr: 0.002493  loss: 3.6088 (3.4148)  weight_decay: 0.0500 (0.0500)  time: 0.6813  data: 0.1213  max mem: 41794
Epoch: [137]  [260/312]  eta: 0:00:40  lr: 0.002493  min_lr: 0.002493  loss: 3.4247 (3.4185)  weight_decay: 0.0500 (0.0500)  time: 0.7422  data: 0.1515  max mem: 41794
Epoch: [137]  [270/312]  eta: 0:00:32  lr: 0.002492  min_lr: 0.002492  loss: 3.4247 (3.4232)  weight_decay: 0.0500 (0.0500)  time: 0.6969  data: 0.1207  max mem: 41794
Epoch: [137]  [280/312]  eta: 0:00:24  lr: 0.002491  min_lr: 0.002491  loss: 3.4314 (3.4203)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.0854  max mem: 41794
Epoch: [137]  [290/312]  eta: 0:00:17  lr: 0.002491  min_lr: 0.002491  loss: 3.6005 (3.4293)  weight_decay: 0.0500 (0.0500)  time: 0.8838  data: 0.1416  max mem: 41794
Epoch: [137]  [300/312]  eta: 0:00:09  lr: 0.002490  min_lr: 0.002490  loss: 3.6045 (3.4322)  weight_decay: 0.0500 (0.0500)  time: 0.6964  data: 0.0688  max mem: 41794
Epoch: [137]  [310/312]  eta: 0:00:01  lr: 0.002489  min_lr: 0.002489  loss: 3.5069 (3.4325)  weight_decay: 0.0500 (0.0500)  time: 0.4782  data: 0.0098  max mem: 41794
Epoch: [137]  [311/312]  eta: 0:00:00  lr: 0.002489  min_lr: 0.002489  loss: 3.5069 (3.4322)  weight_decay: 0.0500 (0.0500)  time: 0.4766  data: 0.0098  max mem: 41794
Epoch: [137] Total time: 0:03:58 (0.7639 s / it)
Averaged stats: lr: 0.002489  min_lr: 0.002489  loss: 3.5069 (3.4590)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.3452 (1.3452)  acc1: 76.8229 (76.8229)  acc5: 92.5781 (92.5781)  time: 8.6216  data: 8.4114  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5442 (1.4968)  acc1: 68.8802 (71.1680)  acc5: 90.2344 (90.5280)  time: 1.1059  data: 0.9347  max mem: 41794
Test: Total time: 0:00:10 (1.1168 s / it)
* Acc@1 71.332 Acc@5 90.718 loss 1.490
Accuracy of the model on the 50000 test images: 71.3%
Max accuracy: 72.13%
Epoch: [138]  [  0/312]  eta: 1:20:25  lr: 0.002489  min_lr: 0.002489  loss: 3.6097 (3.6097)  weight_decay: 0.0500 (0.0500)  time: 15.4650  data: 13.6397  max mem: 41794
Epoch: [138]  [ 10/312]  eta: 0:11:09  lr: 0.002488  min_lr: 0.002488  loss: 3.6379 (3.4857)  weight_decay: 0.0500 (0.0500)  time: 2.2152  data: 1.3394  max mem: 41794
Epoch: [138]  [ 20/312]  eta: 0:07:22  lr: 0.002488  min_lr: 0.002488  loss: 3.6711 (3.5860)  weight_decay: 0.0500 (0.0500)  time: 0.8190  data: 0.1246  max mem: 41794
Epoch: [138]  [ 30/312]  eta: 0:05:37  lr: 0.002487  min_lr: 0.002487  loss: 3.7168 (3.6210)  weight_decay: 0.0500 (0.0500)  time: 0.6364  data: 0.0757  max mem: 41794
Epoch: [138]  [ 40/312]  eta: 0:04:45  lr: 0.002486  min_lr: 0.002486  loss: 3.5595 (3.6200)  weight_decay: 0.0500 (0.0500)  time: 0.5579  data: 0.0283  max mem: 41794
Epoch: [138]  [ 50/312]  eta: 0:04:23  lr: 0.002486  min_lr: 0.002486  loss: 3.5595 (3.5775)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.1303  max mem: 41794
Epoch: [138]  [ 60/312]  eta: 0:03:52  lr: 0.002485  min_lr: 0.002485  loss: 3.6457 (3.5730)  weight_decay: 0.0500 (0.0500)  time: 0.6601  data: 0.1081  max mem: 41794
Epoch: [138]  [ 70/312]  eta: 0:03:44  lr: 0.002484  min_lr: 0.002484  loss: 3.3254 (3.5311)  weight_decay: 0.0500 (0.0500)  time: 0.7194  data: 0.1424  max mem: 41794
Epoch: [138]  [ 80/312]  eta: 0:03:31  lr: 0.002483  min_lr: 0.002483  loss: 3.2775 (3.4759)  weight_decay: 0.0500 (0.0500)  time: 0.8747  data: 0.2020  max mem: 41794
Epoch: [138]  [ 90/312]  eta: 0:03:14  lr: 0.002483  min_lr: 0.002483  loss: 3.4959 (3.4903)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1127  max mem: 41794
Epoch: [138]  [100/312]  eta: 0:03:06  lr: 0.002482  min_lr: 0.002482  loss: 3.5495 (3.4685)  weight_decay: 0.0500 (0.0500)  time: 0.7463  data: 0.1525  max mem: 41794
Epoch: [138]  [110/312]  eta: 0:02:50  lr: 0.002481  min_lr: 0.002481  loss: 3.4329 (3.4538)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.1001  max mem: 41794
Epoch: [138]  [120/312]  eta: 0:02:42  lr: 0.002481  min_lr: 0.002481  loss: 3.5932 (3.4795)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.0772  max mem: 41794
Epoch: [138]  [130/312]  eta: 0:02:34  lr: 0.002480  min_lr: 0.002480  loss: 3.7589 (3.4880)  weight_decay: 0.0500 (0.0500)  time: 0.8625  data: 0.1564  max mem: 41794
Epoch: [138]  [140/312]  eta: 0:02:21  lr: 0.002479  min_lr: 0.002479  loss: 3.6789 (3.4836)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.0798  max mem: 41794
Epoch: [138]  [150/312]  eta: 0:02:13  lr: 0.002479  min_lr: 0.002479  loss: 3.5504 (3.4856)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.0959  max mem: 41794
Epoch: [138]  [160/312]  eta: 0:02:04  lr: 0.002478  min_lr: 0.002478  loss: 3.6165 (3.4933)  weight_decay: 0.0500 (0.0500)  time: 0.7845  data: 0.1076  max mem: 41794
Epoch: [138]  [170/312]  eta: 0:01:54  lr: 0.002477  min_lr: 0.002477  loss: 3.6165 (3.4888)  weight_decay: 0.0500 (0.0500)  time: 0.7066  data: 0.1087  max mem: 41794
Epoch: [138]  [180/312]  eta: 0:01:47  lr: 0.002476  min_lr: 0.002476  loss: 3.6642 (3.5019)  weight_decay: 0.0500 (0.0500)  time: 0.7648  data: 0.1929  max mem: 41794
Epoch: [138]  [190/312]  eta: 0:01:36  lr: 0.002476  min_lr: 0.002476  loss: 3.8100 (3.5101)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.0964  max mem: 41794
Epoch: [138]  [200/312]  eta: 0:01:29  lr: 0.002475  min_lr: 0.002475  loss: 3.8252 (3.5238)  weight_decay: 0.0500 (0.0500)  time: 0.6516  data: 0.0978  max mem: 41794
Epoch: [138]  [210/312]  eta: 0:01:21  lr: 0.002474  min_lr: 0.002474  loss: 3.7408 (3.5185)  weight_decay: 0.0500 (0.0500)  time: 0.8106  data: 0.2007  max mem: 41794
Epoch: [138]  [220/312]  eta: 0:01:11  lr: 0.002474  min_lr: 0.002474  loss: 3.5793 (3.5183)  weight_decay: 0.0500 (0.0500)  time: 0.6556  data: 0.1050  max mem: 41794
Epoch: [138]  [230/312]  eta: 0:01:04  lr: 0.002473  min_lr: 0.002473  loss: 3.6681 (3.5238)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.1310  max mem: 41794
Epoch: [138]  [240/312]  eta: 0:00:56  lr: 0.002472  min_lr: 0.002472  loss: 3.7306 (3.5332)  weight_decay: 0.0500 (0.0500)  time: 0.7859  data: 0.1648  max mem: 41794
Epoch: [138]  [250/312]  eta: 0:00:48  lr: 0.002472  min_lr: 0.002472  loss: 3.5775 (3.5148)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.1116  max mem: 41794
Epoch: [138]  [260/312]  eta: 0:00:40  lr: 0.002471  min_lr: 0.002471  loss: 3.4863 (3.5182)  weight_decay: 0.0500 (0.0500)  time: 0.7512  data: 0.1914  max mem: 41794
Epoch: [138]  [270/312]  eta: 0:00:32  lr: 0.002470  min_lr: 0.002470  loss: 3.6724 (3.5154)  weight_decay: 0.0500 (0.0500)  time: 0.6807  data: 0.1171  max mem: 41794
Epoch: [138]  [280/312]  eta: 0:00:24  lr: 0.002470  min_lr: 0.002470  loss: 3.5416 (3.5152)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.0901  max mem: 41794
Epoch: [138]  [290/312]  eta: 0:00:17  lr: 0.002469  min_lr: 0.002469  loss: 3.5375 (3.5187)  weight_decay: 0.0500 (0.0500)  time: 0.8626  data: 0.1679  max mem: 41794
Epoch: [138]  [300/312]  eta: 0:00:09  lr: 0.002468  min_lr: 0.002468  loss: 3.5665 (3.5132)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.0795  max mem: 41794
Epoch: [138]  [310/312]  eta: 0:00:01  lr: 0.002467  min_lr: 0.002467  loss: 3.4308 (3.5104)  weight_decay: 0.0500 (0.0500)  time: 0.4667  data: 0.0001  max mem: 41794
Epoch: [138]  [311/312]  eta: 0:00:00  lr: 0.002467  min_lr: 0.002467  loss: 3.3131 (3.5098)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0001  max mem: 41794
Epoch: [138] Total time: 0:03:57 (0.7625 s / it)
Averaged stats: lr: 0.002467  min_lr: 0.002467  loss: 3.3131 (3.4799)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 1.3444 (1.3444)  acc1: 75.7812 (75.7812)  acc5: 94.5312 (94.5312)  time: 7.8612  data: 7.6617  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6339 (1.5474)  acc1: 68.0990 (70.1920)  acc5: 90.8854 (90.4320)  time: 1.0748  data: 0.9046  max mem: 41794
Test: Total time: 0:00:09 (1.0855 s / it)
* Acc@1 70.316 Acc@5 90.230 loss 1.539
Accuracy of the model on the 50000 test images: 70.3%
Max accuracy: 72.13%
Epoch: [139]  [  0/312]  eta: 1:21:13  lr: 0.002467  min_lr: 0.002467  loss: 4.2201 (4.2201)  weight_decay: 0.0500 (0.0500)  time: 15.6215  data: 12.3695  max mem: 41794
Epoch: [139]  [ 10/312]  eta: 0:10:58  lr: 0.002467  min_lr: 0.002467  loss: 3.6260 (3.4633)  weight_decay: 0.0500 (0.0500)  time: 2.1813  data: 1.3524  max mem: 41794
Epoch: [139]  [ 20/312]  eta: 0:07:24  lr: 0.002466  min_lr: 0.002466  loss: 3.6260 (3.5615)  weight_decay: 0.0500 (0.0500)  time: 0.8161  data: 0.2064  max mem: 41794
Epoch: [139]  [ 30/312]  eta: 0:05:37  lr: 0.002465  min_lr: 0.002465  loss: 3.6557 (3.5091)  weight_decay: 0.0500 (0.0500)  time: 0.6540  data: 0.0890  max mem: 41794
Epoch: [139]  [ 40/312]  eta: 0:04:43  lr: 0.002464  min_lr: 0.002464  loss: 3.5300 (3.4994)  weight_decay: 0.0500 (0.0500)  time: 0.5399  data: 0.0474  max mem: 41794
Epoch: [139]  [ 50/312]  eta: 0:04:22  lr: 0.002464  min_lr: 0.002464  loss: 3.6157 (3.5016)  weight_decay: 0.0500 (0.0500)  time: 0.7010  data: 0.2075  max mem: 41794
Epoch: [139]  [ 60/312]  eta: 0:03:51  lr: 0.002463  min_lr: 0.002463  loss: 3.6902 (3.5347)  weight_decay: 0.0500 (0.0500)  time: 0.6612  data: 0.1684  max mem: 41794
Epoch: [139]  [ 70/312]  eta: 0:03:40  lr: 0.002462  min_lr: 0.002462  loss: 3.6719 (3.5357)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1921  max mem: 41794
Epoch: [139]  [ 80/312]  eta: 0:03:31  lr: 0.002462  min_lr: 0.002462  loss: 3.4533 (3.5328)  weight_decay: 0.0500 (0.0500)  time: 0.8855  data: 0.3890  max mem: 41794
Epoch: [139]  [ 90/312]  eta: 0:03:11  lr: 0.002461  min_lr: 0.002461  loss: 3.4725 (3.5352)  weight_decay: 0.0500 (0.0500)  time: 0.6917  data: 0.1975  max mem: 41794
Epoch: [139]  [100/312]  eta: 0:03:04  lr: 0.002460  min_lr: 0.002460  loss: 3.5663 (3.5219)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.2033  max mem: 41794
Epoch: [139]  [110/312]  eta: 0:02:48  lr: 0.002460  min_lr: 0.002460  loss: 3.5249 (3.5205)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.2033  max mem: 41794
Epoch: [139]  [120/312]  eta: 0:02:41  lr: 0.002459  min_lr: 0.002459  loss: 3.4393 (3.4995)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.2019  max mem: 41794
Epoch: [139]  [130/312]  eta: 0:02:33  lr: 0.002458  min_lr: 0.002458  loss: 3.3070 (3.4758)  weight_decay: 0.0500 (0.0500)  time: 0.9050  data: 0.4145  max mem: 41794
Epoch: [139]  [140/312]  eta: 0:02:20  lr: 0.002457  min_lr: 0.002457  loss: 3.3667 (3.4781)  weight_decay: 0.0500 (0.0500)  time: 0.6993  data: 0.2132  max mem: 41794
Epoch: [139]  [150/312]  eta: 0:02:12  lr: 0.002457  min_lr: 0.002457  loss: 3.4398 (3.4668)  weight_decay: 0.0500 (0.0500)  time: 0.6633  data: 0.1773  max mem: 41794
Epoch: [139]  [160/312]  eta: 0:02:04  lr: 0.002456  min_lr: 0.002456  loss: 3.4448 (3.4719)  weight_decay: 0.0500 (0.0500)  time: 0.8281  data: 0.3364  max mem: 41794
Epoch: [139]  [170/312]  eta: 0:01:53  lr: 0.002455  min_lr: 0.002455  loss: 3.5614 (3.4653)  weight_decay: 0.0500 (0.0500)  time: 0.6498  data: 0.1597  max mem: 41794
Epoch: [139]  [180/312]  eta: 0:01:46  lr: 0.002455  min_lr: 0.002455  loss: 3.4930 (3.4686)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1830  max mem: 41794
Epoch: [139]  [190/312]  eta: 0:01:36  lr: 0.002454  min_lr: 0.002454  loss: 3.6315 (3.4731)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.1830  max mem: 41794
Epoch: [139]  [200/312]  eta: 0:01:28  lr: 0.002453  min_lr: 0.002453  loss: 3.4526 (3.4687)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.2012  max mem: 41794
Epoch: [139]  [210/312]  eta: 0:01:21  lr: 0.002453  min_lr: 0.002453  loss: 3.2917 (3.4585)  weight_decay: 0.0500 (0.0500)  time: 0.8591  data: 0.3691  max mem: 41794
Epoch: [139]  [220/312]  eta: 0:01:11  lr: 0.002452  min_lr: 0.002452  loss: 3.4764 (3.4600)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.1685  max mem: 41794
Epoch: [139]  [230/312]  eta: 0:01:04  lr: 0.002451  min_lr: 0.002451  loss: 3.6914 (3.4645)  weight_decay: 0.0500 (0.0500)  time: 0.6850  data: 0.1931  max mem: 41794
Epoch: [139]  [240/312]  eta: 0:00:56  lr: 0.002450  min_lr: 0.002450  loss: 3.6692 (3.4672)  weight_decay: 0.0500 (0.0500)  time: 0.8664  data: 0.3701  max mem: 41794
Epoch: [139]  [250/312]  eta: 0:00:48  lr: 0.002450  min_lr: 0.002450  loss: 3.5418 (3.4647)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.1793  max mem: 41794
Epoch: [139]  [260/312]  eta: 0:00:40  lr: 0.002449  min_lr: 0.002449  loss: 3.1810 (3.4603)  weight_decay: 0.0500 (0.0500)  time: 0.6587  data: 0.1653  max mem: 41794
Epoch: [139]  [270/312]  eta: 0:00:32  lr: 0.002448  min_lr: 0.002448  loss: 3.1434 (3.4525)  weight_decay: 0.0500 (0.0500)  time: 0.6551  data: 0.1636  max mem: 41794
Epoch: [139]  [280/312]  eta: 0:00:24  lr: 0.002448  min_lr: 0.002448  loss: 3.5510 (3.4607)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.1721  max mem: 41794
Epoch: [139]  [290/312]  eta: 0:00:16  lr: 0.002447  min_lr: 0.002447  loss: 3.5510 (3.4593)  weight_decay: 0.0500 (0.0500)  time: 0.8319  data: 0.3219  max mem: 41794
Epoch: [139]  [300/312]  eta: 0:00:09  lr: 0.002446  min_lr: 0.002446  loss: 3.4254 (3.4630)  weight_decay: 0.0500 (0.0500)  time: 0.6613  data: 0.1598  max mem: 41794
Epoch: [139]  [310/312]  eta: 0:00:01  lr: 0.002446  min_lr: 0.002446  loss: 3.4254 (3.4601)  weight_decay: 0.0500 (0.0500)  time: 0.4939  data: 0.0098  max mem: 41794
Epoch: [139]  [311/312]  eta: 0:00:00  lr: 0.002446  min_lr: 0.002446  loss: 3.4254 (3.4613)  weight_decay: 0.0500 (0.0500)  time: 0.4757  data: 0.0098  max mem: 41794
Epoch: [139] Total time: 0:03:56 (0.7590 s / it)
Averaged stats: lr: 0.002446  min_lr: 0.002446  loss: 3.4254 (3.4680)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1937 (1.1937)  acc1: 78.1250 (78.1250)  acc5: 94.4010 (94.4010)  time: 8.4610  data: 8.2553  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4661 (1.3970)  acc1: 71.2240 (72.1920)  acc5: 90.7552 (91.2000)  time: 1.1016  data: 0.9173  max mem: 41794
Test: Total time: 0:00:10 (1.1558 s / it)
* Acc@1 72.228 Acc@5 91.204 loss 1.395
Accuracy of the model on the 50000 test images: 72.2%
Max accuracy: 72.23%
Epoch: [140]  [  0/312]  eta: 1:20:18  lr: 0.002445  min_lr: 0.002445  loss: 2.8701 (2.8701)  weight_decay: 0.0500 (0.0500)  time: 15.4454  data: 13.1879  max mem: 41794
Epoch: [140]  [ 10/312]  eta: 0:12:20  lr: 0.002445  min_lr: 0.002445  loss: 3.7553 (3.6769)  weight_decay: 0.0500 (0.0500)  time: 2.4512  data: 1.3153  max mem: 41794
Epoch: [140]  [ 20/312]  eta: 0:07:34  lr: 0.002444  min_lr: 0.002444  loss: 3.7314 (3.6201)  weight_decay: 0.0500 (0.0500)  time: 0.8629  data: 0.0685  max mem: 41794
Epoch: [140]  [ 30/312]  eta: 0:05:43  lr: 0.002443  min_lr: 0.002443  loss: 3.6189 (3.5279)  weight_decay: 0.0500 (0.0500)  time: 0.5414  data: 0.0048  max mem: 41794
Epoch: [140]  [ 40/312]  eta: 0:04:45  lr: 0.002443  min_lr: 0.002443  loss: 3.6366 (3.5619)  weight_decay: 0.0500 (0.0500)  time: 0.5166  data: 0.0054  max mem: 41794
Epoch: [140]  [ 50/312]  eta: 0:04:08  lr: 0.002442  min_lr: 0.002442  loss: 3.6421 (3.5012)  weight_decay: 0.0500 (0.0500)  time: 0.5261  data: 0.0072  max mem: 41794
Epoch: [140]  [ 60/312]  eta: 0:03:40  lr: 0.002441  min_lr: 0.002441  loss: 3.6399 (3.4872)  weight_decay: 0.0500 (0.0500)  time: 0.5194  data: 0.0049  max mem: 41794
Epoch: [140]  [ 70/312]  eta: 0:03:21  lr: 0.002441  min_lr: 0.002441  loss: 3.6551 (3.4807)  weight_decay: 0.0500 (0.0500)  time: 0.5398  data: 0.0313  max mem: 41794
Epoch: [140]  [ 80/312]  eta: 0:03:11  lr: 0.002440  min_lr: 0.002440  loss: 3.6140 (3.4820)  weight_decay: 0.0500 (0.0500)  time: 0.6633  data: 0.1511  max mem: 41794
Epoch: [140]  [ 90/312]  eta: 0:03:00  lr: 0.002439  min_lr: 0.002439  loss: 3.5574 (3.4415)  weight_decay: 0.0500 (0.0500)  time: 0.7361  data: 0.2073  max mem: 41794
Epoch: [140]  [100/312]  eta: 0:02:50  lr: 0.002438  min_lr: 0.002438  loss: 3.4442 (3.4356)  weight_decay: 0.0500 (0.0500)  time: 0.7363  data: 0.1969  max mem: 41794
Epoch: [140]  [110/312]  eta: 0:02:41  lr: 0.002438  min_lr: 0.002438  loss: 3.6540 (3.4417)  weight_decay: 0.0500 (0.0500)  time: 0.7474  data: 0.1917  max mem: 41794
Epoch: [140]  [120/312]  eta: 0:02:31  lr: 0.002437  min_lr: 0.002437  loss: 3.6107 (3.4128)  weight_decay: 0.0500 (0.0500)  time: 0.7138  data: 0.1355  max mem: 41794
Epoch: [140]  [130/312]  eta: 0:02:24  lr: 0.002436  min_lr: 0.002436  loss: 3.1522 (3.4066)  weight_decay: 0.0500 (0.0500)  time: 0.7563  data: 0.1154  max mem: 41794
Epoch: [140]  [140/312]  eta: 0:02:14  lr: 0.002436  min_lr: 0.002436  loss: 3.6687 (3.4239)  weight_decay: 0.0500 (0.0500)  time: 0.7433  data: 0.1235  max mem: 41794
Epoch: [140]  [150/312]  eta: 0:02:07  lr: 0.002435  min_lr: 0.002435  loss: 3.6936 (3.4362)  weight_decay: 0.0500 (0.0500)  time: 0.7330  data: 0.1376  max mem: 41794
Epoch: [140]  [160/312]  eta: 0:01:58  lr: 0.002434  min_lr: 0.002434  loss: 3.5388 (3.4319)  weight_decay: 0.0500 (0.0500)  time: 0.7254  data: 0.1499  max mem: 41794
Epoch: [140]  [170/312]  eta: 0:01:49  lr: 0.002434  min_lr: 0.002434  loss: 3.6235 (3.4432)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.1298  max mem: 41794
Epoch: [140]  [180/312]  eta: 0:01:42  lr: 0.002433  min_lr: 0.002433  loss: 3.6162 (3.4431)  weight_decay: 0.0500 (0.0500)  time: 0.7436  data: 0.1451  max mem: 41794
Epoch: [140]  [190/312]  eta: 0:01:33  lr: 0.002432  min_lr: 0.002432  loss: 3.3863 (3.4357)  weight_decay: 0.0500 (0.0500)  time: 0.7454  data: 0.1946  max mem: 41794
Epoch: [140]  [200/312]  eta: 0:01:25  lr: 0.002431  min_lr: 0.002431  loss: 3.3863 (3.4352)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.1549  max mem: 41794
Epoch: [140]  [210/312]  eta: 0:01:18  lr: 0.002431  min_lr: 0.002431  loss: 3.4659 (3.4299)  weight_decay: 0.0500 (0.0500)  time: 0.7346  data: 0.1541  max mem: 41794
Epoch: [140]  [220/312]  eta: 0:01:10  lr: 0.002430  min_lr: 0.002430  loss: 3.5331 (3.4279)  weight_decay: 0.0500 (0.0500)  time: 0.7140  data: 0.1741  max mem: 41794
Epoch: [140]  [230/312]  eta: 0:01:02  lr: 0.002429  min_lr: 0.002429  loss: 3.6258 (3.4374)  weight_decay: 0.0500 (0.0500)  time: 0.7114  data: 0.1803  max mem: 41794
Epoch: [140]  [240/312]  eta: 0:00:54  lr: 0.002429  min_lr: 0.002429  loss: 3.6838 (3.4463)  weight_decay: 0.0500 (0.0500)  time: 0.7548  data: 0.2316  max mem: 41794
Epoch: [140]  [250/312]  eta: 0:00:46  lr: 0.002428  min_lr: 0.002428  loss: 3.6838 (3.4529)  weight_decay: 0.0500 (0.0500)  time: 0.7000  data: 0.1560  max mem: 41794
Epoch: [140]  [260/312]  eta: 0:00:39  lr: 0.002427  min_lr: 0.002427  loss: 3.5845 (3.4511)  weight_decay: 0.0500 (0.0500)  time: 0.7196  data: 0.1382  max mem: 41794
Epoch: [140]  [270/312]  eta: 0:00:31  lr: 0.002426  min_lr: 0.002426  loss: 3.5845 (3.4514)  weight_decay: 0.0500 (0.0500)  time: 0.7982  data: 0.2428  max mem: 41794
Epoch: [140]  [280/312]  eta: 0:00:24  lr: 0.002426  min_lr: 0.002426  loss: 3.5974 (3.4457)  weight_decay: 0.0500 (0.0500)  time: 0.7342  data: 0.1858  max mem: 41794
Epoch: [140]  [290/312]  eta: 0:00:16  lr: 0.002425  min_lr: 0.002425  loss: 3.6799 (3.4544)  weight_decay: 0.0500 (0.0500)  time: 0.7348  data: 0.1006  max mem: 41794
Epoch: [140]  [300/312]  eta: 0:00:09  lr: 0.002424  min_lr: 0.002424  loss: 3.6799 (3.4607)  weight_decay: 0.0500 (0.0500)  time: 0.7124  data: 0.1119  max mem: 41794
Epoch: [140]  [310/312]  eta: 0:00:01  lr: 0.002424  min_lr: 0.002424  loss: 3.5586 (3.4585)  weight_decay: 0.0500 (0.0500)  time: 0.5434  data: 0.0601  max mem: 41794
Epoch: [140]  [311/312]  eta: 0:00:00  lr: 0.002424  min_lr: 0.002424  loss: 3.5586 (3.4590)  weight_decay: 0.0500 (0.0500)  time: 0.5365  data: 0.0601  max mem: 41794
Epoch: [140] Total time: 0:03:53 (0.7494 s / it)
Averaged stats: lr: 0.002424  min_lr: 0.002424  loss: 3.5586 (3.4868)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.0381 (1.0381)  acc1: 78.1250 (78.1250)  acc5: 95.0521 (95.0521)  time: 8.1357  data: 7.9223  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3291 (1.3026)  acc1: 71.7448 (71.3600)  acc5: 92.0573 (91.5680)  time: 1.0526  data: 0.8803  max mem: 41794
Test: Total time: 0:00:09 (1.0720 s / it)
* Acc@1 71.748 Acc@5 91.212 loss 1.309
Accuracy of the model on the 50000 test images: 71.7%
Max accuracy: 72.23%
Epoch: [141]  [  0/312]  eta: 1:19:53  lr: 0.002424  min_lr: 0.002424  loss: 3.6669 (3.6669)  weight_decay: 0.0500 (0.0500)  time: 15.3628  data: 14.0128  max mem: 41794
Epoch: [141]  [ 10/312]  eta: 0:10:21  lr: 0.002423  min_lr: 0.002423  loss: 3.6669 (3.6327)  weight_decay: 0.0500 (0.0500)  time: 2.0584  data: 1.3864  max mem: 41794
Epoch: [141]  [ 20/312]  eta: 0:07:44  lr: 0.002422  min_lr: 0.002422  loss: 3.4234 (3.5070)  weight_decay: 0.0500 (0.0500)  time: 0.9039  data: 0.2066  max mem: 41794
Epoch: [141]  [ 30/312]  eta: 0:05:54  lr: 0.002421  min_lr: 0.002421  loss: 3.2947 (3.3976)  weight_decay: 0.0500 (0.0500)  time: 0.8174  data: 0.1457  max mem: 41794
Epoch: [141]  [ 40/312]  eta: 0:05:13  lr: 0.002421  min_lr: 0.002421  loss: 3.4625 (3.4626)  weight_decay: 0.0500 (0.0500)  time: 0.6945  data: 0.0337  max mem: 41794
Epoch: [141]  [ 50/312]  eta: 0:04:30  lr: 0.002420  min_lr: 0.002420  loss: 3.6529 (3.5034)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.0421  max mem: 41794
Epoch: [141]  [ 60/312]  eta: 0:03:58  lr: 0.002419  min_lr: 0.002419  loss: 3.5837 (3.5058)  weight_decay: 0.0500 (0.0500)  time: 0.5145  data: 0.0096  max mem: 41794
Epoch: [141]  [ 70/312]  eta: 0:03:42  lr: 0.002419  min_lr: 0.002419  loss: 3.4920 (3.4848)  weight_decay: 0.0500 (0.0500)  time: 0.6285  data: 0.0425  max mem: 41794
Epoch: [141]  [ 80/312]  eta: 0:03:29  lr: 0.002418  min_lr: 0.002418  loss: 3.4920 (3.4905)  weight_decay: 0.0500 (0.0500)  time: 0.7745  data: 0.0721  max mem: 41794
Epoch: [141]  [ 90/312]  eta: 0:03:13  lr: 0.002417  min_lr: 0.002417  loss: 3.6800 (3.4912)  weight_decay: 0.0500 (0.0500)  time: 0.6971  data: 0.0527  max mem: 41794
Epoch: [141]  [100/312]  eta: 0:03:04  lr: 0.002417  min_lr: 0.002417  loss: 3.6626 (3.4733)  weight_decay: 0.0500 (0.0500)  time: 0.7426  data: 0.0664  max mem: 41794
Epoch: [141]  [110/312]  eta: 0:02:49  lr: 0.002416  min_lr: 0.002416  loss: 3.5363 (3.4725)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.0440  max mem: 41794
Epoch: [141]  [120/312]  eta: 0:02:41  lr: 0.002415  min_lr: 0.002415  loss: 3.6906 (3.4947)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.0432  max mem: 41794
Epoch: [141]  [130/312]  eta: 0:02:33  lr: 0.002414  min_lr: 0.002414  loss: 3.7038 (3.5072)  weight_decay: 0.0500 (0.0500)  time: 0.8758  data: 0.1100  max mem: 41794
Epoch: [141]  [140/312]  eta: 0:02:20  lr: 0.002414  min_lr: 0.002414  loss: 3.6835 (3.5083)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0687  max mem: 41794
Epoch: [141]  [150/312]  eta: 0:02:13  lr: 0.002413  min_lr: 0.002413  loss: 3.5681 (3.4994)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.0941  max mem: 41794
Epoch: [141]  [160/312]  eta: 0:02:03  lr: 0.002412  min_lr: 0.002412  loss: 3.5681 (3.5029)  weight_decay: 0.0500 (0.0500)  time: 0.7549  data: 0.0929  max mem: 41794
Epoch: [141]  [170/312]  eta: 0:01:53  lr: 0.002412  min_lr: 0.002412  loss: 3.7057 (3.5128)  weight_decay: 0.0500 (0.0500)  time: 0.6126  data: 0.0364  max mem: 41794
Epoch: [141]  [180/312]  eta: 0:01:45  lr: 0.002411  min_lr: 0.002411  loss: 3.7057 (3.5132)  weight_decay: 0.0500 (0.0500)  time: 0.7345  data: 0.0923  max mem: 41794
Epoch: [141]  [190/312]  eta: 0:01:35  lr: 0.002410  min_lr: 0.002410  loss: 3.7132 (3.5160)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.0566  max mem: 41794
Epoch: [141]  [200/312]  eta: 0:01:28  lr: 0.002409  min_lr: 0.002409  loss: 3.7661 (3.5268)  weight_decay: 0.0500 (0.0500)  time: 0.7002  data: 0.0712  max mem: 41794
Epoch: [141]  [210/312]  eta: 0:01:20  lr: 0.002409  min_lr: 0.002409  loss: 3.8296 (3.5362)  weight_decay: 0.0500 (0.0500)  time: 0.8704  data: 0.1286  max mem: 41794
Epoch: [141]  [220/312]  eta: 0:01:11  lr: 0.002408  min_lr: 0.002408  loss: 3.5676 (3.5261)  weight_decay: 0.0500 (0.0500)  time: 0.6545  data: 0.0580  max mem: 41794
Epoch: [141]  [230/312]  eta: 0:01:04  lr: 0.002407  min_lr: 0.002407  loss: 3.4749 (3.5218)  weight_decay: 0.0500 (0.0500)  time: 0.6591  data: 0.0517  max mem: 41794
Epoch: [141]  [240/312]  eta: 0:00:56  lr: 0.002407  min_lr: 0.002407  loss: 3.5351 (3.5212)  weight_decay: 0.0500 (0.0500)  time: 0.7753  data: 0.0531  max mem: 41794
Epoch: [141]  [250/312]  eta: 0:00:47  lr: 0.002406  min_lr: 0.002406  loss: 3.6950 (3.5285)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.0660  max mem: 41794
Epoch: [141]  [260/312]  eta: 0:00:40  lr: 0.002405  min_lr: 0.002405  loss: 3.7126 (3.5293)  weight_decay: 0.0500 (0.0500)  time: 0.7736  data: 0.1500  max mem: 41794
Epoch: [141]  [270/312]  eta: 0:00:32  lr: 0.002405  min_lr: 0.002405  loss: 3.5222 (3.5251)  weight_decay: 0.0500 (0.0500)  time: 0.7053  data: 0.0861  max mem: 41794
Epoch: [141]  [280/312]  eta: 0:00:24  lr: 0.002404  min_lr: 0.002404  loss: 3.5077 (3.5291)  weight_decay: 0.0500 (0.0500)  time: 0.6776  data: 0.0887  max mem: 41794
Epoch: [141]  [290/312]  eta: 0:00:17  lr: 0.002403  min_lr: 0.002403  loss: 3.4754 (3.5202)  weight_decay: 0.0500 (0.0500)  time: 0.8593  data: 0.1528  max mem: 41794
Epoch: [141]  [300/312]  eta: 0:00:09  lr: 0.002402  min_lr: 0.002402  loss: 3.3393 (3.5135)  weight_decay: 0.0500 (0.0500)  time: 0.6684  data: 0.0645  max mem: 41794
Epoch: [141]  [310/312]  eta: 0:00:01  lr: 0.002402  min_lr: 0.002402  loss: 3.5535 (3.5174)  weight_decay: 0.0500 (0.0500)  time: 0.4671  data: 0.0001  max mem: 41794
Epoch: [141]  [311/312]  eta: 0:00:00  lr: 0.002402  min_lr: 0.002402  loss: 3.6185 (3.5177)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [141] Total time: 0:03:57 (0.7615 s / it)
Averaged stats: lr: 0.002402  min_lr: 0.002402  loss: 3.6185 (3.4819)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1461 (1.1461)  acc1: 76.8229 (76.8229)  acc5: 92.8385 (92.8385)  time: 8.4919  data: 8.2894  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3326 (1.3264)  acc1: 70.9635 (71.4400)  acc5: 92.0573 (91.5360)  time: 1.0965  data: 0.9211  max mem: 41794
Test: Total time: 0:00:10 (1.1428 s / it)
* Acc@1 71.522 Acc@5 91.244 loss 1.333
Accuracy of the model on the 50000 test images: 71.5%
Max accuracy: 72.23%
Epoch: [142]  [  0/312]  eta: 1:26:20  lr: 0.002402  min_lr: 0.002402  loss: 2.4320 (2.4320)  weight_decay: 0.0500 (0.0500)  time: 16.6044  data: 14.4240  max mem: 41794
Epoch: [142]  [ 10/312]  eta: 0:11:31  lr: 0.002401  min_lr: 0.002401  loss: 3.5464 (3.3925)  weight_decay: 0.0500 (0.0500)  time: 2.2906  data: 1.3585  max mem: 41794
Epoch: [142]  [ 20/312]  eta: 0:07:26  lr: 0.002400  min_lr: 0.002400  loss: 3.4619 (3.3518)  weight_decay: 0.0500 (0.0500)  time: 0.7746  data: 0.0915  max mem: 41794
Epoch: [142]  [ 30/312]  eta: 0:05:39  lr: 0.002399  min_lr: 0.002399  loss: 3.4619 (3.4209)  weight_decay: 0.0500 (0.0500)  time: 0.6056  data: 0.0688  max mem: 41794
Epoch: [142]  [ 40/312]  eta: 0:04:44  lr: 0.002399  min_lr: 0.002399  loss: 3.5853 (3.4105)  weight_decay: 0.0500 (0.0500)  time: 0.5365  data: 0.0133  max mem: 41794
Epoch: [142]  [ 50/312]  eta: 0:04:18  lr: 0.002398  min_lr: 0.002398  loss: 3.6559 (3.4900)  weight_decay: 0.0500 (0.0500)  time: 0.6539  data: 0.1162  max mem: 41794
Epoch: [142]  [ 60/312]  eta: 0:03:48  lr: 0.002397  min_lr: 0.002397  loss: 3.6840 (3.5028)  weight_decay: 0.0500 (0.0500)  time: 0.6242  data: 0.1080  max mem: 41794
Epoch: [142]  [ 70/312]  eta: 0:03:37  lr: 0.002397  min_lr: 0.002397  loss: 3.5544 (3.4868)  weight_decay: 0.0500 (0.0500)  time: 0.6761  data: 0.1477  max mem: 41794
Epoch: [142]  [ 80/312]  eta: 0:03:25  lr: 0.002396  min_lr: 0.002396  loss: 3.5625 (3.5079)  weight_decay: 0.0500 (0.0500)  time: 0.8198  data: 0.2973  max mem: 41794
Epoch: [142]  [ 90/312]  eta: 0:03:08  lr: 0.002395  min_lr: 0.002395  loss: 3.6333 (3.5078)  weight_decay: 0.0500 (0.0500)  time: 0.6769  data: 0.1538  max mem: 41794
Epoch: [142]  [100/312]  eta: 0:03:01  lr: 0.002395  min_lr: 0.002395  loss: 3.5424 (3.4953)  weight_decay: 0.0500 (0.0500)  time: 0.7499  data: 0.1957  max mem: 41794
Epoch: [142]  [110/312]  eta: 0:02:46  lr: 0.002394  min_lr: 0.002394  loss: 3.3753 (3.4638)  weight_decay: 0.0500 (0.0500)  time: 0.7130  data: 0.1953  max mem: 41794
Epoch: [142]  [120/312]  eta: 0:02:39  lr: 0.002393  min_lr: 0.002393  loss: 3.2340 (3.4504)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.1832  max mem: 41794
Epoch: [142]  [130/312]  eta: 0:02:30  lr: 0.002392  min_lr: 0.002392  loss: 3.5535 (3.4586)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.3016  max mem: 41794
Epoch: [142]  [140/312]  eta: 0:02:19  lr: 0.002392  min_lr: 0.002392  loss: 3.5128 (3.4544)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.1563  max mem: 41794
Epoch: [142]  [150/312]  eta: 0:02:12  lr: 0.002391  min_lr: 0.002391  loss: 3.4454 (3.4392)  weight_decay: 0.0500 (0.0500)  time: 0.7491  data: 0.1952  max mem: 41794
Epoch: [142]  [160/312]  eta: 0:02:03  lr: 0.002390  min_lr: 0.002390  loss: 3.4755 (3.4486)  weight_decay: 0.0500 (0.0500)  time: 0.8279  data: 0.2655  max mem: 41794
Epoch: [142]  [170/312]  eta: 0:01:53  lr: 0.002390  min_lr: 0.002390  loss: 3.5705 (3.4433)  weight_decay: 0.0500 (0.0500)  time: 0.6448  data: 0.1147  max mem: 41794
Epoch: [142]  [180/312]  eta: 0:01:45  lr: 0.002389  min_lr: 0.002389  loss: 3.4487 (3.4384)  weight_decay: 0.0500 (0.0500)  time: 0.7335  data: 0.1648  max mem: 41794
Epoch: [142]  [190/312]  eta: 0:01:35  lr: 0.002388  min_lr: 0.002388  loss: 3.6236 (3.4482)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.1583  max mem: 41794
Epoch: [142]  [200/312]  eta: 0:01:28  lr: 0.002387  min_lr: 0.002387  loss: 3.6100 (3.4442)  weight_decay: 0.0500 (0.0500)  time: 0.6692  data: 0.1478  max mem: 41794
Epoch: [142]  [210/312]  eta: 0:01:20  lr: 0.002387  min_lr: 0.002387  loss: 3.4596 (3.4387)  weight_decay: 0.0500 (0.0500)  time: 0.8433  data: 0.2821  max mem: 41794
Epoch: [142]  [220/312]  eta: 0:01:11  lr: 0.002386  min_lr: 0.002386  loss: 3.4149 (3.4328)  weight_decay: 0.0500 (0.0500)  time: 0.6694  data: 0.1432  max mem: 41794
Epoch: [142]  [230/312]  eta: 0:01:04  lr: 0.002385  min_lr: 0.002385  loss: 3.1840 (3.4246)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.1647  max mem: 41794
Epoch: [142]  [240/312]  eta: 0:00:56  lr: 0.002385  min_lr: 0.002385  loss: 3.6517 (3.4355)  weight_decay: 0.0500 (0.0500)  time: 0.8386  data: 0.3121  max mem: 41794
Epoch: [142]  [250/312]  eta: 0:00:48  lr: 0.002384  min_lr: 0.002384  loss: 3.7061 (3.4325)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.1605  max mem: 41794
Epoch: [142]  [260/312]  eta: 0:00:40  lr: 0.002383  min_lr: 0.002383  loss: 3.4454 (3.4301)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1761  max mem: 41794
Epoch: [142]  [270/312]  eta: 0:00:32  lr: 0.002383  min_lr: 0.002383  loss: 3.5483 (3.4297)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.1706  max mem: 41794
Epoch: [142]  [280/312]  eta: 0:00:24  lr: 0.002382  min_lr: 0.002382  loss: 3.6908 (3.4333)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.1847  max mem: 41794
Epoch: [142]  [290/312]  eta: 0:00:17  lr: 0.002381  min_lr: 0.002381  loss: 3.5495 (3.4374)  weight_decay: 0.0500 (0.0500)  time: 0.8434  data: 0.3470  max mem: 41794
Epoch: [142]  [300/312]  eta: 0:00:09  lr: 0.002380  min_lr: 0.002380  loss: 3.5410 (3.4387)  weight_decay: 0.0500 (0.0500)  time: 0.6516  data: 0.1628  max mem: 41794
Epoch: [142]  [310/312]  eta: 0:00:01  lr: 0.002380  min_lr: 0.002380  loss: 3.6316 (3.4443)  weight_decay: 0.0500 (0.0500)  time: 0.4683  data: 0.0001  max mem: 41794
Epoch: [142]  [311/312]  eta: 0:00:00  lr: 0.002380  min_lr: 0.002380  loss: 3.6316 (3.4427)  weight_decay: 0.0500 (0.0500)  time: 0.4667  data: 0.0001  max mem: 41794
Epoch: [142] Total time: 0:03:57 (0.7601 s / it)
Averaged stats: lr: 0.002380  min_lr: 0.002380  loss: 3.6316 (3.4530)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1822 (1.1822)  acc1: 78.3854 (78.3854)  acc5: 93.6198 (93.6198)  time: 8.4687  data: 8.2704  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5462 (1.4574)  acc1: 70.1823 (71.5680)  acc5: 90.6250 (90.9760)  time: 1.0961  data: 0.9190  max mem: 41794
Test: Total time: 0:00:10 (1.1423 s / it)
* Acc@1 71.696 Acc@5 91.104 loss 1.453
Accuracy of the model on the 50000 test images: 71.7%
Max accuracy: 72.23%
Epoch: [143]  [  0/312]  eta: 1:20:47  lr: 0.002380  min_lr: 0.002380  loss: 3.0232 (3.0232)  weight_decay: 0.0500 (0.0500)  time: 15.5361  data: 14.0388  max mem: 41794
Epoch: [143]  [ 10/312]  eta: 0:11:13  lr: 0.002379  min_lr: 0.002379  loss: 3.8144 (3.5134)  weight_decay: 0.0500 (0.0500)  time: 2.2310  data: 1.3158  max mem: 41794
Epoch: [143]  [ 20/312]  eta: 0:07:17  lr: 0.002378  min_lr: 0.002378  loss: 3.6625 (3.4161)  weight_decay: 0.0500 (0.0500)  time: 0.7960  data: 0.0637  max mem: 41794
Epoch: [143]  [ 30/312]  eta: 0:05:34  lr: 0.002377  min_lr: 0.002377  loss: 3.6625 (3.4877)  weight_decay: 0.0500 (0.0500)  time: 0.6107  data: 0.0513  max mem: 41794
Epoch: [143]  [ 40/312]  eta: 0:04:39  lr: 0.002377  min_lr: 0.002377  loss: 3.6763 (3.4645)  weight_decay: 0.0500 (0.0500)  time: 0.5350  data: 0.0158  max mem: 41794
Epoch: [143]  [ 50/312]  eta: 0:04:19  lr: 0.002376  min_lr: 0.002376  loss: 3.5704 (3.4812)  weight_decay: 0.0500 (0.0500)  time: 0.6926  data: 0.1409  max mem: 41794
Epoch: [143]  [ 60/312]  eta: 0:03:51  lr: 0.002375  min_lr: 0.002375  loss: 3.4104 (3.4346)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.1535  max mem: 41794
Epoch: [143]  [ 70/312]  eta: 0:03:41  lr: 0.002375  min_lr: 0.002375  loss: 3.2636 (3.4368)  weight_decay: 0.0500 (0.0500)  time: 0.7172  data: 0.1718  max mem: 41794
Epoch: [143]  [ 80/312]  eta: 0:03:29  lr: 0.002374  min_lr: 0.002374  loss: 3.3871 (3.4154)  weight_decay: 0.0500 (0.0500)  time: 0.8615  data: 0.3227  max mem: 41794
Epoch: [143]  [ 90/312]  eta: 0:03:10  lr: 0.002373  min_lr: 0.002373  loss: 3.4029 (3.4377)  weight_decay: 0.0500 (0.0500)  time: 0.6655  data: 0.1734  max mem: 41794
Epoch: [143]  [100/312]  eta: 0:03:03  lr: 0.002373  min_lr: 0.002373  loss: 3.6896 (3.4640)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.2062  max mem: 41794
Epoch: [143]  [110/312]  eta: 0:02:48  lr: 0.002372  min_lr: 0.002372  loss: 3.7773 (3.4769)  weight_decay: 0.0500 (0.0500)  time: 0.7034  data: 0.2031  max mem: 41794
Epoch: [143]  [120/312]  eta: 0:02:39  lr: 0.002371  min_lr: 0.002371  loss: 3.6966 (3.4839)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1645  max mem: 41794
Epoch: [143]  [130/312]  eta: 0:02:31  lr: 0.002370  min_lr: 0.002370  loss: 3.5998 (3.4860)  weight_decay: 0.0500 (0.0500)  time: 0.8400  data: 0.3426  max mem: 41794
Epoch: [143]  [140/312]  eta: 0:02:19  lr: 0.002370  min_lr: 0.002370  loss: 3.5998 (3.4841)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.1787  max mem: 41794
Epoch: [143]  [150/312]  eta: 0:02:11  lr: 0.002369  min_lr: 0.002369  loss: 3.4967 (3.4745)  weight_decay: 0.0500 (0.0500)  time: 0.6704  data: 0.1774  max mem: 41794
Epoch: [143]  [160/312]  eta: 0:02:04  lr: 0.002368  min_lr: 0.002368  loss: 3.4967 (3.4743)  weight_decay: 0.0500 (0.0500)  time: 0.8758  data: 0.3796  max mem: 41794
Epoch: [143]  [170/312]  eta: 0:01:53  lr: 0.002368  min_lr: 0.002368  loss: 3.5889 (3.4765)  weight_decay: 0.0500 (0.0500)  time: 0.6924  data: 0.2028  max mem: 41794
Epoch: [143]  [180/312]  eta: 0:01:46  lr: 0.002367  min_lr: 0.002367  loss: 3.7009 (3.4769)  weight_decay: 0.0500 (0.0500)  time: 0.6945  data: 0.2043  max mem: 41794
Epoch: [143]  [190/312]  eta: 0:01:36  lr: 0.002366  min_lr: 0.002366  loss: 3.6724 (3.4806)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.2043  max mem: 41794
Epoch: [143]  [200/312]  eta: 0:01:28  lr: 0.002365  min_lr: 0.002365  loss: 3.6347 (3.4708)  weight_decay: 0.0500 (0.0500)  time: 0.6671  data: 0.1748  max mem: 41794
Epoch: [143]  [210/312]  eta: 0:01:20  lr: 0.002365  min_lr: 0.002365  loss: 3.5382 (3.4665)  weight_decay: 0.0500 (0.0500)  time: 0.8415  data: 0.3481  max mem: 41794
Epoch: [143]  [220/312]  eta: 0:01:11  lr: 0.002364  min_lr: 0.002364  loss: 3.5121 (3.4652)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.1739  max mem: 41794
Epoch: [143]  [230/312]  eta: 0:01:04  lr: 0.002363  min_lr: 0.002363  loss: 3.3915 (3.4579)  weight_decay: 0.0500 (0.0500)  time: 0.6927  data: 0.1908  max mem: 41794
Epoch: [143]  [240/312]  eta: 0:00:56  lr: 0.002363  min_lr: 0.002363  loss: 3.2876 (3.4531)  weight_decay: 0.0500 (0.0500)  time: 0.8720  data: 0.3776  max mem: 41794
Epoch: [143]  [250/312]  eta: 0:00:48  lr: 0.002362  min_lr: 0.002362  loss: 3.3604 (3.4418)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1874  max mem: 41794
Epoch: [143]  [260/312]  eta: 0:00:40  lr: 0.002361  min_lr: 0.002361  loss: 3.3604 (3.4419)  weight_decay: 0.0500 (0.0500)  time: 0.6990  data: 0.1987  max mem: 41794
Epoch: [143]  [270/312]  eta: 0:00:32  lr: 0.002360  min_lr: 0.002360  loss: 3.5440 (3.4457)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.1987  max mem: 41794
Epoch: [143]  [280/312]  eta: 0:00:24  lr: 0.002360  min_lr: 0.002360  loss: 3.5440 (3.4443)  weight_decay: 0.0500 (0.0500)  time: 0.6594  data: 0.1636  max mem: 41794
Epoch: [143]  [290/312]  eta: 0:00:17  lr: 0.002359  min_lr: 0.002359  loss: 3.3593 (3.4401)  weight_decay: 0.0500 (0.0500)  time: 0.8222  data: 0.3225  max mem: 41794
Epoch: [143]  [300/312]  eta: 0:00:09  lr: 0.002358  min_lr: 0.002358  loss: 3.7220 (3.4420)  weight_decay: 0.0500 (0.0500)  time: 0.6482  data: 0.1592  max mem: 41794
Epoch: [143]  [310/312]  eta: 0:00:01  lr: 0.002358  min_lr: 0.002358  loss: 3.5846 (3.4385)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [143]  [311/312]  eta: 0:00:00  lr: 0.002358  min_lr: 0.002358  loss: 3.5846 (3.4388)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [143] Total time: 0:03:56 (0.7588 s / it)
Averaged stats: lr: 0.002358  min_lr: 0.002358  loss: 3.5846 (3.4369)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 1.1677 (1.1677)  acc1: 78.6458 (78.6458)  acc5: 94.1406 (94.1406)  time: 7.8103  data: 7.6107  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5067 (1.4205)  acc1: 69.4010 (71.7760)  acc5: 90.3646 (90.9600)  time: 1.0382  data: 0.8681  max mem: 41794
Test: Total time: 0:00:09 (1.0486 s / it)
* Acc@1 72.006 Acc@5 91.070 loss 1.414
Accuracy of the model on the 50000 test images: 72.0%
Max accuracy: 72.23%
Epoch: [144]  [  0/312]  eta: 1:14:26  lr: 0.002358  min_lr: 0.002358  loss: 3.6384 (3.6384)  weight_decay: 0.0500 (0.0500)  time: 14.3162  data: 11.9726  max mem: 41794
Epoch: [144]  [ 10/312]  eta: 0:10:09  lr: 0.002357  min_lr: 0.002357  loss: 3.7003 (3.4014)  weight_decay: 0.0500 (0.0500)  time: 2.0193  data: 1.3123  max mem: 41794
Epoch: [144]  [ 20/312]  eta: 0:07:09  lr: 0.002356  min_lr: 0.002356  loss: 3.7120 (3.5188)  weight_decay: 0.0500 (0.0500)  time: 0.8298  data: 0.2059  max mem: 41794
Epoch: [144]  [ 30/312]  eta: 0:05:26  lr: 0.002355  min_lr: 0.002355  loss: 3.6282 (3.5276)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.0831  max mem: 41794
Epoch: [144]  [ 40/312]  eta: 0:04:51  lr: 0.002355  min_lr: 0.002355  loss: 3.6282 (3.5102)  weight_decay: 0.0500 (0.0500)  time: 0.6523  data: 0.1081  max mem: 41794
Epoch: [144]  [ 50/312]  eta: 0:04:26  lr: 0.002354  min_lr: 0.002354  loss: 3.6383 (3.4951)  weight_decay: 0.0500 (0.0500)  time: 0.8006  data: 0.2537  max mem: 41794
Epoch: [144]  [ 60/312]  eta: 0:03:54  lr: 0.002353  min_lr: 0.002353  loss: 3.3381 (3.4483)  weight_decay: 0.0500 (0.0500)  time: 0.6357  data: 0.1477  max mem: 41794
Epoch: [144]  [ 70/312]  eta: 0:03:46  lr: 0.002353  min_lr: 0.002353  loss: 3.4837 (3.4736)  weight_decay: 0.0500 (0.0500)  time: 0.7332  data: 0.1680  max mem: 41794
Epoch: [144]  [ 80/312]  eta: 0:03:31  lr: 0.002352  min_lr: 0.002352  loss: 3.6141 (3.4876)  weight_decay: 0.0500 (0.0500)  time: 0.8514  data: 0.2109  max mem: 41794
Epoch: [144]  [ 90/312]  eta: 0:03:20  lr: 0.002351  min_lr: 0.002351  loss: 3.4843 (3.4833)  weight_decay: 0.0500 (0.0500)  time: 0.7711  data: 0.1166  max mem: 41794
Epoch: [144]  [100/312]  eta: 0:03:07  lr: 0.002350  min_lr: 0.002350  loss: 3.6053 (3.4811)  weight_decay: 0.0500 (0.0500)  time: 0.7796  data: 0.1213  max mem: 41794
Epoch: [144]  [110/312]  eta: 0:02:51  lr: 0.002350  min_lr: 0.002350  loss: 3.6053 (3.4905)  weight_decay: 0.0500 (0.0500)  time: 0.6129  data: 0.0499  max mem: 41794
Epoch: [144]  [120/312]  eta: 0:02:42  lr: 0.002349  min_lr: 0.002349  loss: 3.5257 (3.4892)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0708  max mem: 41794
Epoch: [144]  [130/312]  eta: 0:02:33  lr: 0.002348  min_lr: 0.002348  loss: 3.4445 (3.4784)  weight_decay: 0.0500 (0.0500)  time: 0.8138  data: 0.2029  max mem: 41794
Epoch: [144]  [140/312]  eta: 0:02:20  lr: 0.002348  min_lr: 0.002348  loss: 3.4284 (3.4755)  weight_decay: 0.0500 (0.0500)  time: 0.6419  data: 0.1330  max mem: 41794
Epoch: [144]  [150/312]  eta: 0:02:13  lr: 0.002347  min_lr: 0.002347  loss: 3.5948 (3.4894)  weight_decay: 0.0500 (0.0500)  time: 0.6786  data: 0.1824  max mem: 41794
Epoch: [144]  [160/312]  eta: 0:02:04  lr: 0.002346  min_lr: 0.002346  loss: 3.6762 (3.4999)  weight_decay: 0.0500 (0.0500)  time: 0.8152  data: 0.3219  max mem: 41794
Epoch: [144]  [170/312]  eta: 0:01:54  lr: 0.002345  min_lr: 0.002345  loss: 3.6827 (3.5116)  weight_decay: 0.0500 (0.0500)  time: 0.6892  data: 0.1401  max mem: 41794
Epoch: [144]  [180/312]  eta: 0:01:46  lr: 0.002345  min_lr: 0.002345  loss: 3.5483 (3.5057)  weight_decay: 0.0500 (0.0500)  time: 0.7242  data: 0.1277  max mem: 41794
Epoch: [144]  [190/312]  eta: 0:01:36  lr: 0.002344  min_lr: 0.002344  loss: 3.6012 (3.5050)  weight_decay: 0.0500 (0.0500)  time: 0.6655  data: 0.1277  max mem: 41794
Epoch: [144]  [200/312]  eta: 0:01:29  lr: 0.002343  min_lr: 0.002343  loss: 3.5998 (3.4989)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1408  max mem: 41794
Epoch: [144]  [210/312]  eta: 0:01:21  lr: 0.002343  min_lr: 0.002343  loss: 3.5861 (3.4987)  weight_decay: 0.0500 (0.0500)  time: 0.9159  data: 0.2253  max mem: 41794
Epoch: [144]  [220/312]  eta: 0:01:12  lr: 0.002342  min_lr: 0.002342  loss: 3.7037 (3.5079)  weight_decay: 0.0500 (0.0500)  time: 0.7602  data: 0.1218  max mem: 41794
Epoch: [144]  [230/312]  eta: 0:01:04  lr: 0.002341  min_lr: 0.002341  loss: 3.6237 (3.4957)  weight_decay: 0.0500 (0.0500)  time: 0.6759  data: 0.1084  max mem: 41794
Epoch: [144]  [240/312]  eta: 0:00:56  lr: 0.002341  min_lr: 0.002341  loss: 3.5253 (3.4962)  weight_decay: 0.0500 (0.0500)  time: 0.6892  data: 0.1182  max mem: 41794
Epoch: [144]  [250/312]  eta: 0:00:48  lr: 0.002340  min_lr: 0.002340  loss: 3.6331 (3.5005)  weight_decay: 0.0500 (0.0500)  time: 0.6865  data: 0.0817  max mem: 41794
Epoch: [144]  [260/312]  eta: 0:00:40  lr: 0.002339  min_lr: 0.002339  loss: 3.5992 (3.5002)  weight_decay: 0.0500 (0.0500)  time: 0.7665  data: 0.1174  max mem: 41794
Epoch: [144]  [270/312]  eta: 0:00:32  lr: 0.002338  min_lr: 0.002338  loss: 3.4009 (3.4910)  weight_decay: 0.0500 (0.0500)  time: 0.6251  data: 0.0829  max mem: 41794
Epoch: [144]  [280/312]  eta: 0:00:24  lr: 0.002338  min_lr: 0.002338  loss: 3.5333 (3.4917)  weight_decay: 0.0500 (0.0500)  time: 0.6578  data: 0.1596  max mem: 41794
Epoch: [144]  [290/312]  eta: 0:00:17  lr: 0.002337  min_lr: 0.002337  loss: 3.5835 (3.4884)  weight_decay: 0.0500 (0.0500)  time: 0.8096  data: 0.2966  max mem: 41794
Epoch: [144]  [300/312]  eta: 0:00:09  lr: 0.002336  min_lr: 0.002336  loss: 3.4232 (3.4870)  weight_decay: 0.0500 (0.0500)  time: 0.6384  data: 0.1374  max mem: 41794
Epoch: [144]  [310/312]  eta: 0:00:01  lr: 0.002336  min_lr: 0.002336  loss: 3.5805 (3.4870)  weight_decay: 0.0500 (0.0500)  time: 0.4683  data: 0.0001  max mem: 41794
Epoch: [144]  [311/312]  eta: 0:00:00  lr: 0.002335  min_lr: 0.002335  loss: 3.4333 (3.4868)  weight_decay: 0.0500 (0.0500)  time: 0.4687  data: 0.0001  max mem: 41794
Epoch: [144] Total time: 0:03:57 (0.7611 s / it)
Averaged stats: lr: 0.002335  min_lr: 0.002335  loss: 3.4333 (3.4452)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1908 (1.1908)  acc1: 79.4271 (79.4271)  acc5: 94.4010 (94.4010)  time: 8.4457  data: 8.2384  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5069 (1.4057)  acc1: 71.8750 (72.6880)  acc5: 91.2760 (91.5840)  time: 1.0887  data: 0.9155  max mem: 41794
Test: Total time: 0:00:09 (1.1098 s / it)
* Acc@1 72.838 Acc@5 91.636 loss 1.401
Accuracy of the model on the 50000 test images: 72.8%
Max accuracy: 72.84%
Epoch: [145]  [  0/312]  eta: 1:22:10  lr: 0.002335  min_lr: 0.002335  loss: 2.3327 (2.3327)  weight_decay: 0.0500 (0.0500)  time: 15.8037  data: 12.2449  max mem: 41794
Epoch: [145]  [ 10/312]  eta: 0:11:06  lr: 0.002335  min_lr: 0.002335  loss: 2.7645 (3.0584)  weight_decay: 0.0500 (0.0500)  time: 2.2074  data: 1.3396  max mem: 41794
Epoch: [145]  [ 20/312]  eta: 0:07:16  lr: 0.002334  min_lr: 0.002334  loss: 3.4804 (3.2904)  weight_decay: 0.0500 (0.0500)  time: 0.7780  data: 0.1618  max mem: 41794
Epoch: [145]  [ 30/312]  eta: 0:05:34  lr: 0.002333  min_lr: 0.002333  loss: 3.6154 (3.3783)  weight_decay: 0.0500 (0.0500)  time: 0.6248  data: 0.0403  max mem: 41794
Epoch: [145]  [ 40/312]  eta: 0:04:51  lr: 0.002333  min_lr: 0.002333  loss: 3.6608 (3.4032)  weight_decay: 0.0500 (0.0500)  time: 0.6318  data: 0.1189  max mem: 41794
Epoch: [145]  [ 50/312]  eta: 0:04:27  lr: 0.002332  min_lr: 0.002332  loss: 3.6256 (3.4307)  weight_decay: 0.0500 (0.0500)  time: 0.7693  data: 0.2657  max mem: 41794
Epoch: [145]  [ 60/312]  eta: 0:03:55  lr: 0.002331  min_lr: 0.002331  loss: 3.7522 (3.4753)  weight_decay: 0.0500 (0.0500)  time: 0.6498  data: 0.1500  max mem: 41794
Epoch: [145]  [ 70/312]  eta: 0:03:44  lr: 0.002330  min_lr: 0.002330  loss: 3.8112 (3.5138)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.1981  max mem: 41794
Epoch: [145]  [ 80/312]  eta: 0:03:34  lr: 0.002330  min_lr: 0.002330  loss: 3.6497 (3.5035)  weight_decay: 0.0500 (0.0500)  time: 0.8997  data: 0.3980  max mem: 41794
Epoch: [145]  [ 90/312]  eta: 0:03:15  lr: 0.002329  min_lr: 0.002329  loss: 3.3130 (3.4594)  weight_decay: 0.0500 (0.0500)  time: 0.6981  data: 0.2005  max mem: 41794
Epoch: [145]  [100/312]  eta: 0:03:05  lr: 0.002328  min_lr: 0.002328  loss: 3.3593 (3.4568)  weight_decay: 0.0500 (0.0500)  time: 0.6601  data: 0.1610  max mem: 41794
Epoch: [145]  [110/312]  eta: 0:02:49  lr: 0.002328  min_lr: 0.002328  loss: 3.5435 (3.4531)  weight_decay: 0.0500 (0.0500)  time: 0.6617  data: 0.1609  max mem: 41794
Epoch: [145]  [120/312]  eta: 0:02:41  lr: 0.002327  min_lr: 0.002327  loss: 3.5169 (3.4408)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1961  max mem: 41794
Epoch: [145]  [130/312]  eta: 0:02:33  lr: 0.002326  min_lr: 0.002326  loss: 3.1625 (3.4298)  weight_decay: 0.0500 (0.0500)  time: 0.8740  data: 0.3808  max mem: 41794
Epoch: [145]  [140/312]  eta: 0:02:20  lr: 0.002326  min_lr: 0.002326  loss: 3.0239 (3.3986)  weight_decay: 0.0500 (0.0500)  time: 0.6779  data: 0.1853  max mem: 41794
Epoch: [145]  [150/312]  eta: 0:02:13  lr: 0.002325  min_lr: 0.002325  loss: 3.4209 (3.4255)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.1825  max mem: 41794
Epoch: [145]  [160/312]  eta: 0:02:05  lr: 0.002324  min_lr: 0.002324  loss: 3.7900 (3.4365)  weight_decay: 0.0500 (0.0500)  time: 0.8562  data: 0.3648  max mem: 41794
Epoch: [145]  [170/312]  eta: 0:01:54  lr: 0.002323  min_lr: 0.002323  loss: 3.6216 (3.4328)  weight_decay: 0.0500 (0.0500)  time: 0.6680  data: 0.1830  max mem: 41794
Epoch: [145]  [180/312]  eta: 0:01:46  lr: 0.002323  min_lr: 0.002323  loss: 3.4832 (3.4354)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.1950  max mem: 41794
Epoch: [145]  [190/312]  eta: 0:01:36  lr: 0.002322  min_lr: 0.002322  loss: 3.5853 (3.4427)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.1950  max mem: 41794
Epoch: [145]  [200/312]  eta: 0:01:29  lr: 0.002321  min_lr: 0.002321  loss: 3.6575 (3.4418)  weight_decay: 0.0500 (0.0500)  time: 0.6823  data: 0.1935  max mem: 41794
Epoch: [145]  [210/312]  eta: 0:01:21  lr: 0.002321  min_lr: 0.002321  loss: 3.4949 (3.4360)  weight_decay: 0.0500 (0.0500)  time: 0.8648  data: 0.3760  max mem: 41794
Epoch: [145]  [220/312]  eta: 0:01:12  lr: 0.002320  min_lr: 0.002320  loss: 3.4949 (3.4349)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.1832  max mem: 41794
Epoch: [145]  [230/312]  eta: 0:01:04  lr: 0.002319  min_lr: 0.002319  loss: 3.4541 (3.4329)  weight_decay: 0.0500 (0.0500)  time: 0.7045  data: 0.2046  max mem: 41794
Epoch: [145]  [240/312]  eta: 0:00:56  lr: 0.002318  min_lr: 0.002318  loss: 3.4541 (3.4377)  weight_decay: 0.0500 (0.0500)  time: 0.8208  data: 0.3222  max mem: 41794
Epoch: [145]  [250/312]  eta: 0:00:48  lr: 0.002318  min_lr: 0.002318  loss: 3.7123 (3.4462)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.1886  max mem: 41794
Epoch: [145]  [260/312]  eta: 0:00:40  lr: 0.002317  min_lr: 0.002317  loss: 3.6498 (3.4390)  weight_decay: 0.0500 (0.0500)  time: 0.7474  data: 0.2514  max mem: 41794
Epoch: [145]  [270/312]  eta: 0:00:32  lr: 0.002316  min_lr: 0.002316  loss: 3.4253 (3.4388)  weight_decay: 0.0500 (0.0500)  time: 0.6790  data: 0.1811  max mem: 41794
Epoch: [145]  [280/312]  eta: 0:00:24  lr: 0.002316  min_lr: 0.002316  loss: 3.3105 (3.4343)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1830  max mem: 41794
Epoch: [145]  [290/312]  eta: 0:00:17  lr: 0.002315  min_lr: 0.002315  loss: 3.0293 (3.4276)  weight_decay: 0.0500 (0.0500)  time: 0.8036  data: 0.2715  max mem: 41794
Epoch: [145]  [300/312]  eta: 0:00:09  lr: 0.002314  min_lr: 0.002314  loss: 3.1012 (3.4185)  weight_decay: 0.0500 (0.0500)  time: 0.6092  data: 0.0889  max mem: 41794
Epoch: [145]  [310/312]  eta: 0:00:01  lr: 0.002313  min_lr: 0.002313  loss: 3.3036 (3.4220)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [145]  [311/312]  eta: 0:00:00  lr: 0.002313  min_lr: 0.002313  loss: 3.3289 (3.4228)  weight_decay: 0.0500 (0.0500)  time: 0.4641  data: 0.0001  max mem: 41794
Epoch: [145] Total time: 0:03:57 (0.7620 s / it)
Averaged stats: lr: 0.002313  min_lr: 0.002313  loss: 3.3289 (3.4552)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.1544 (1.1544)  acc1: 78.5156 (78.5156)  acc5: 94.6615 (94.6615)  time: 7.9326  data: 7.7206  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5191 (1.4416)  acc1: 70.4427 (72.0160)  acc5: 90.6250 (91.4400)  time: 1.0552  data: 0.8817  max mem: 41794
Test: Total time: 0:00:09 (1.0714 s / it)
* Acc@1 72.312 Acc@5 91.520 loss 1.429
Accuracy of the model on the 50000 test images: 72.3%
Max accuracy: 72.84%
Epoch: [146]  [  0/312]  eta: 1:17:24  lr: 0.002313  min_lr: 0.002313  loss: 4.1959 (4.1959)  weight_decay: 0.0500 (0.0500)  time: 14.8865  data: 13.5897  max mem: 41794
Epoch: [146]  [ 10/312]  eta: 0:10:14  lr: 0.002313  min_lr: 0.002313  loss: 3.7217 (3.5774)  weight_decay: 0.0500 (0.0500)  time: 2.0337  data: 1.2928  max mem: 41794
Epoch: [146]  [ 20/312]  eta: 0:07:20  lr: 0.002312  min_lr: 0.002312  loss: 3.3301 (3.3527)  weight_decay: 0.0500 (0.0500)  time: 0.8391  data: 0.1355  max mem: 41794
Epoch: [146]  [ 30/312]  eta: 0:05:34  lr: 0.002311  min_lr: 0.002311  loss: 3.2016 (3.3945)  weight_decay: 0.0500 (0.0500)  time: 0.7203  data: 0.1097  max mem: 41794
Epoch: [146]  [ 40/312]  eta: 0:04:56  lr: 0.002310  min_lr: 0.002310  loss: 3.6733 (3.4362)  weight_decay: 0.0500 (0.0500)  time: 0.6479  data: 0.0686  max mem: 41794
Epoch: [146]  [ 50/312]  eta: 0:04:27  lr: 0.002310  min_lr: 0.002310  loss: 3.6533 (3.4011)  weight_decay: 0.0500 (0.0500)  time: 0.7679  data: 0.1166  max mem: 41794
Epoch: [146]  [ 60/312]  eta: 0:03:56  lr: 0.002309  min_lr: 0.002309  loss: 3.4764 (3.4226)  weight_decay: 0.0500 (0.0500)  time: 0.6330  data: 0.0664  max mem: 41794
Epoch: [146]  [ 70/312]  eta: 0:03:46  lr: 0.002308  min_lr: 0.002308  loss: 3.4151 (3.4058)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.1449  max mem: 41794
Epoch: [146]  [ 80/312]  eta: 0:03:28  lr: 0.002308  min_lr: 0.002308  loss: 3.3680 (3.3939)  weight_decay: 0.0500 (0.0500)  time: 0.7784  data: 0.1634  max mem: 41794
Epoch: [146]  [ 90/312]  eta: 0:03:15  lr: 0.002307  min_lr: 0.002307  loss: 3.5144 (3.3848)  weight_decay: 0.0500 (0.0500)  time: 0.6963  data: 0.1175  max mem: 41794
Epoch: [146]  [100/312]  eta: 0:03:07  lr: 0.002306  min_lr: 0.002306  loss: 3.3229 (3.3610)  weight_decay: 0.0500 (0.0500)  time: 0.8202  data: 0.2002  max mem: 41794
Epoch: [146]  [110/312]  eta: 0:02:51  lr: 0.002305  min_lr: 0.002305  loss: 3.2841 (3.3767)  weight_decay: 0.0500 (0.0500)  time: 0.7038  data: 0.1175  max mem: 41794
Epoch: [146]  [120/312]  eta: 0:02:43  lr: 0.002305  min_lr: 0.002305  loss: 3.7296 (3.3998)  weight_decay: 0.0500 (0.0500)  time: 0.6908  data: 0.1163  max mem: 41794
Epoch: [146]  [130/312]  eta: 0:02:36  lr: 0.002304  min_lr: 0.002304  loss: 3.5786 (3.4103)  weight_decay: 0.0500 (0.0500)  time: 0.9138  data: 0.2041  max mem: 41794
Epoch: [146]  [140/312]  eta: 0:02:23  lr: 0.002303  min_lr: 0.002303  loss: 3.5024 (3.4114)  weight_decay: 0.0500 (0.0500)  time: 0.7410  data: 0.1059  max mem: 41794
Epoch: [146]  [150/312]  eta: 0:02:14  lr: 0.002303  min_lr: 0.002303  loss: 3.6879 (3.4317)  weight_decay: 0.0500 (0.0500)  time: 0.6315  data: 0.0753  max mem: 41794
Epoch: [146]  [160/312]  eta: 0:02:04  lr: 0.002302  min_lr: 0.002302  loss: 3.4025 (3.4130)  weight_decay: 0.0500 (0.0500)  time: 0.7224  data: 0.1270  max mem: 41794
Epoch: [146]  [170/312]  eta: 0:01:54  lr: 0.002301  min_lr: 0.002301  loss: 3.4025 (3.4301)  weight_decay: 0.0500 (0.0500)  time: 0.6465  data: 0.0895  max mem: 41794
Epoch: [146]  [180/312]  eta: 0:01:46  lr: 0.002301  min_lr: 0.002301  loss: 3.6815 (3.4265)  weight_decay: 0.0500 (0.0500)  time: 0.7004  data: 0.1438  max mem: 41794
Epoch: [146]  [190/312]  eta: 0:01:36  lr: 0.002300  min_lr: 0.002300  loss: 3.5547 (3.4254)  weight_decay: 0.0500 (0.0500)  time: 0.6531  data: 0.1206  max mem: 41794
Epoch: [146]  [200/312]  eta: 0:01:29  lr: 0.002299  min_lr: 0.002299  loss: 3.3807 (3.4187)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.1413  max mem: 41794
Epoch: [146]  [210/312]  eta: 0:01:21  lr: 0.002298  min_lr: 0.002298  loss: 3.4035 (3.4204)  weight_decay: 0.0500 (0.0500)  time: 0.8599  data: 0.2578  max mem: 41794
Epoch: [146]  [220/312]  eta: 0:01:12  lr: 0.002298  min_lr: 0.002298  loss: 3.5214 (3.4245)  weight_decay: 0.0500 (0.0500)  time: 0.7396  data: 0.1171  max mem: 41794
Epoch: [146]  [230/312]  eta: 0:01:04  lr: 0.002297  min_lr: 0.002297  loss: 3.5648 (3.4277)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.0846  max mem: 41794
Epoch: [146]  [240/312]  eta: 0:00:56  lr: 0.002296  min_lr: 0.002296  loss: 3.4528 (3.4217)  weight_decay: 0.0500 (0.0500)  time: 0.7637  data: 0.1795  max mem: 41794
Epoch: [146]  [250/312]  eta: 0:00:48  lr: 0.002296  min_lr: 0.002296  loss: 3.4017 (3.4144)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.0956  max mem: 41794
Epoch: [146]  [260/312]  eta: 0:00:40  lr: 0.002295  min_lr: 0.002295  loss: 3.2295 (3.4081)  weight_decay: 0.0500 (0.0500)  time: 0.7399  data: 0.1152  max mem: 41794
Epoch: [146]  [270/312]  eta: 0:00:32  lr: 0.002294  min_lr: 0.002294  loss: 3.4493 (3.4095)  weight_decay: 0.0500 (0.0500)  time: 0.6908  data: 0.1167  max mem: 41794
Epoch: [146]  [280/312]  eta: 0:00:24  lr: 0.002293  min_lr: 0.002293  loss: 3.6769 (3.4138)  weight_decay: 0.0500 (0.0500)  time: 0.6542  data: 0.1035  max mem: 41794
Epoch: [146]  [290/312]  eta: 0:00:17  lr: 0.002293  min_lr: 0.002293  loss: 3.6769 (3.4100)  weight_decay: 0.0500 (0.0500)  time: 0.7592  data: 0.2018  max mem: 41794
Epoch: [146]  [300/312]  eta: 0:00:09  lr: 0.002292  min_lr: 0.002292  loss: 3.4449 (3.4132)  weight_decay: 0.0500 (0.0500)  time: 0.5987  data: 0.1001  max mem: 41794
Epoch: [146]  [310/312]  eta: 0:00:01  lr: 0.002291  min_lr: 0.002291  loss: 3.4449 (3.4060)  weight_decay: 0.0500 (0.0500)  time: 0.4728  data: 0.0001  max mem: 41794
Epoch: [146]  [311/312]  eta: 0:00:00  lr: 0.002291  min_lr: 0.002291  loss: 3.4449 (3.4027)  weight_decay: 0.0500 (0.0500)  time: 0.4663  data: 0.0001  max mem: 41794
Epoch: [146] Total time: 0:03:56 (0.7591 s / it)
Averaged stats: lr: 0.002291  min_lr: 0.002291  loss: 3.4449 (3.4411)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0030 (1.0030)  acc1: 78.3854 (78.3854)  acc5: 94.2708 (94.2708)  time: 8.7851  data: 8.5855  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2894 (1.2074)  acc1: 71.8750 (72.5440)  acc5: 91.7969 (91.6000)  time: 1.1295  data: 0.9540  max mem: 41794
Test: Total time: 0:00:10 (1.1664 s / it)
* Acc@1 72.568 Acc@5 91.598 loss 1.208
Accuracy of the model on the 50000 test images: 72.6%
Max accuracy: 72.84%
Epoch: [147]  [  0/312]  eta: 1:20:29  lr: 0.002291  min_lr: 0.002291  loss: 2.6594 (2.6594)  weight_decay: 0.0500 (0.0500)  time: 15.4781  data: 13.5686  max mem: 41794
Epoch: [147]  [ 10/312]  eta: 0:10:55  lr: 0.002290  min_lr: 0.002290  loss: 3.0934 (3.0926)  weight_decay: 0.0500 (0.0500)  time: 2.1720  data: 1.4341  max mem: 41794
Epoch: [147]  [ 20/312]  eta: 0:07:26  lr: 0.002290  min_lr: 0.002290  loss: 3.4365 (3.2939)  weight_decay: 0.0500 (0.0500)  time: 0.8300  data: 0.1712  max mem: 41794
Epoch: [147]  [ 30/312]  eta: 0:05:40  lr: 0.002289  min_lr: 0.002289  loss: 3.6293 (3.3199)  weight_decay: 0.0500 (0.0500)  time: 0.6743  data: 0.0635  max mem: 41794
Epoch: [147]  [ 40/312]  eta: 0:04:48  lr: 0.002288  min_lr: 0.002288  loss: 3.4807 (3.3204)  weight_decay: 0.0500 (0.0500)  time: 0.5730  data: 0.0703  max mem: 41794
Epoch: [147]  [ 50/312]  eta: 0:04:25  lr: 0.002288  min_lr: 0.002288  loss: 3.4286 (3.3205)  weight_decay: 0.0500 (0.0500)  time: 0.7131  data: 0.2175  max mem: 41794
Epoch: [147]  [ 60/312]  eta: 0:03:53  lr: 0.002287  min_lr: 0.002287  loss: 3.5245 (3.3621)  weight_decay: 0.0500 (0.0500)  time: 0.6454  data: 0.1502  max mem: 41794
Epoch: [147]  [ 70/312]  eta: 0:03:43  lr: 0.002286  min_lr: 0.002286  loss: 3.7224 (3.3921)  weight_decay: 0.0500 (0.0500)  time: 0.6950  data: 0.2064  max mem: 41794
Epoch: [147]  [ 80/312]  eta: 0:03:31  lr: 0.002285  min_lr: 0.002285  loss: 3.6859 (3.4176)  weight_decay: 0.0500 (0.0500)  time: 0.8649  data: 0.3302  max mem: 41794
Epoch: [147]  [ 90/312]  eta: 0:03:16  lr: 0.002285  min_lr: 0.002285  loss: 3.6425 (3.4227)  weight_decay: 0.0500 (0.0500)  time: 0.7549  data: 0.1714  max mem: 41794
Epoch: [147]  [100/312]  eta: 0:03:07  lr: 0.002284  min_lr: 0.002284  loss: 3.5740 (3.4264)  weight_decay: 0.0500 (0.0500)  time: 0.7704  data: 0.1803  max mem: 41794
Epoch: [147]  [110/312]  eta: 0:02:51  lr: 0.002283  min_lr: 0.002283  loss: 3.6387 (3.4381)  weight_decay: 0.0500 (0.0500)  time: 0.6712  data: 0.1332  max mem: 41794
Epoch: [147]  [120/312]  eta: 0:02:42  lr: 0.002283  min_lr: 0.002283  loss: 3.6132 (3.4266)  weight_decay: 0.0500 (0.0500)  time: 0.6553  data: 0.1283  max mem: 41794
Epoch: [147]  [130/312]  eta: 0:02:33  lr: 0.002282  min_lr: 0.002282  loss: 3.6132 (3.4305)  weight_decay: 0.0500 (0.0500)  time: 0.8331  data: 0.2880  max mem: 41794
Epoch: [147]  [140/312]  eta: 0:02:21  lr: 0.002281  min_lr: 0.002281  loss: 3.6694 (3.4361)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1720  max mem: 41794
Epoch: [147]  [150/312]  eta: 0:02:13  lr: 0.002280  min_lr: 0.002280  loss: 3.6697 (3.4336)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1700  max mem: 41794
Epoch: [147]  [160/312]  eta: 0:02:05  lr: 0.002280  min_lr: 0.002280  loss: 3.6732 (3.4406)  weight_decay: 0.0500 (0.0500)  time: 0.8363  data: 0.2431  max mem: 41794
Epoch: [147]  [170/312]  eta: 0:01:55  lr: 0.002279  min_lr: 0.002279  loss: 3.5699 (3.4544)  weight_decay: 0.0500 (0.0500)  time: 0.7134  data: 0.1479  max mem: 41794
Epoch: [147]  [180/312]  eta: 0:01:47  lr: 0.002278  min_lr: 0.002278  loss: 3.4139 (3.4355)  weight_decay: 0.0500 (0.0500)  time: 0.7299  data: 0.2011  max mem: 41794
Epoch: [147]  [190/312]  eta: 0:01:37  lr: 0.002278  min_lr: 0.002278  loss: 3.4123 (3.4406)  weight_decay: 0.0500 (0.0500)  time: 0.6759  data: 0.1384  max mem: 41794
Epoch: [147]  [200/312]  eta: 0:01:29  lr: 0.002277  min_lr: 0.002277  loss: 3.5502 (3.4367)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1112  max mem: 41794
Epoch: [147]  [210/312]  eta: 0:01:22  lr: 0.002276  min_lr: 0.002276  loss: 3.4020 (3.4319)  weight_decay: 0.0500 (0.0500)  time: 0.9011  data: 0.2518  max mem: 41794
Epoch: [147]  [220/312]  eta: 0:01:12  lr: 0.002275  min_lr: 0.002275  loss: 3.4253 (3.4345)  weight_decay: 0.0500 (0.0500)  time: 0.7162  data: 0.1424  max mem: 41794
Epoch: [147]  [230/312]  eta: 0:01:05  lr: 0.002275  min_lr: 0.002275  loss: 3.6219 (3.4346)  weight_decay: 0.0500 (0.0500)  time: 0.6869  data: 0.0942  max mem: 41794
Epoch: [147]  [240/312]  eta: 0:00:56  lr: 0.002274  min_lr: 0.002274  loss: 3.5986 (3.4325)  weight_decay: 0.0500 (0.0500)  time: 0.7510  data: 0.0992  max mem: 41794
Epoch: [147]  [250/312]  eta: 0:00:48  lr: 0.002273  min_lr: 0.002273  loss: 3.7036 (3.4376)  weight_decay: 0.0500 (0.0500)  time: 0.6436  data: 0.0847  max mem: 41794
Epoch: [147]  [260/312]  eta: 0:00:41  lr: 0.002273  min_lr: 0.002273  loss: 3.7407 (3.4390)  weight_decay: 0.0500 (0.0500)  time: 0.7904  data: 0.1842  max mem: 41794
Epoch: [147]  [270/312]  eta: 0:00:32  lr: 0.002272  min_lr: 0.002272  loss: 3.5332 (3.4367)  weight_decay: 0.0500 (0.0500)  time: 0.7101  data: 0.1069  max mem: 41794
Epoch: [147]  [280/312]  eta: 0:00:24  lr: 0.002271  min_lr: 0.002271  loss: 3.6435 (3.4467)  weight_decay: 0.0500 (0.0500)  time: 0.6658  data: 0.0801  max mem: 41794
Epoch: [147]  [290/312]  eta: 0:00:17  lr: 0.002270  min_lr: 0.002270  loss: 3.6882 (3.4500)  weight_decay: 0.0500 (0.0500)  time: 0.7767  data: 0.1173  max mem: 41794
Epoch: [147]  [300/312]  eta: 0:00:09  lr: 0.002270  min_lr: 0.002270  loss: 3.3374 (3.4494)  weight_decay: 0.0500 (0.0500)  time: 0.5963  data: 0.0396  max mem: 41794
Epoch: [147]  [310/312]  eta: 0:00:01  lr: 0.002269  min_lr: 0.002269  loss: 3.7480 (3.4574)  weight_decay: 0.0500 (0.0500)  time: 0.4644  data: 0.0015  max mem: 41794
Epoch: [147]  [311/312]  eta: 0:00:00  lr: 0.002269  min_lr: 0.002269  loss: 3.7480 (3.4553)  weight_decay: 0.0500 (0.0500)  time: 0.4637  data: 0.0001  max mem: 41794
Epoch: [147] Total time: 0:03:58 (0.7635 s / it)
Averaged stats: lr: 0.002269  min_lr: 0.002269  loss: 3.7480 (3.4442)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.1118 (1.1118)  acc1: 77.9948 (77.9948)  acc5: 94.5312 (94.5312)  time: 8.6344  data: 8.4296  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4735 (1.3749)  acc1: 69.4010 (71.1520)  acc5: 91.1458 (91.0240)  time: 1.1104  data: 0.9367  max mem: 41794
Test: Total time: 0:00:10 (1.1521 s / it)
* Acc@1 71.542 Acc@5 90.934 loss 1.369
Accuracy of the model on the 50000 test images: 71.5%
Max accuracy: 72.84%
Epoch: [148]  [  0/312]  eta: 1:26:40  lr: 0.002269  min_lr: 0.002269  loss: 3.4554 (3.4554)  weight_decay: 0.0500 (0.0500)  time: 16.6667  data: 15.6096  max mem: 41794
Epoch: [148]  [ 10/312]  eta: 0:11:34  lr: 0.002268  min_lr: 0.002268  loss: 3.4554 (3.4599)  weight_decay: 0.0500 (0.0500)  time: 2.2981  data: 1.4236  max mem: 41794
Epoch: [148]  [ 20/312]  eta: 0:07:38  lr: 0.002267  min_lr: 0.002267  loss: 3.6608 (3.6128)  weight_decay: 0.0500 (0.0500)  time: 0.8150  data: 0.0795  max mem: 41794
Epoch: [148]  [ 30/312]  eta: 0:05:48  lr: 0.002267  min_lr: 0.002267  loss: 3.5347 (3.4969)  weight_decay: 0.0500 (0.0500)  time: 0.6542  data: 0.0913  max mem: 41794
Epoch: [148]  [ 40/312]  eta: 0:04:54  lr: 0.002266  min_lr: 0.002266  loss: 3.3367 (3.4883)  weight_decay: 0.0500 (0.0500)  time: 0.5676  data: 0.0320  max mem: 41794
Epoch: [148]  [ 50/312]  eta: 0:04:30  lr: 0.002265  min_lr: 0.002265  loss: 3.7110 (3.4872)  weight_decay: 0.0500 (0.0500)  time: 0.7127  data: 0.1435  max mem: 41794
Epoch: [148]  [ 60/312]  eta: 0:03:58  lr: 0.002265  min_lr: 0.002265  loss: 3.6360 (3.4692)  weight_decay: 0.0500 (0.0500)  time: 0.6668  data: 0.1261  max mem: 41794
Epoch: [148]  [ 70/312]  eta: 0:03:46  lr: 0.002264  min_lr: 0.002264  loss: 3.3657 (3.4415)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.1273  max mem: 41794
Epoch: [148]  [ 80/312]  eta: 0:03:30  lr: 0.002263  min_lr: 0.002263  loss: 3.3247 (3.4548)  weight_decay: 0.0500 (0.0500)  time: 0.7989  data: 0.1959  max mem: 41794
Epoch: [148]  [ 90/312]  eta: 0:03:15  lr: 0.002262  min_lr: 0.002262  loss: 3.4648 (3.4610)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.1405  max mem: 41794
Epoch: [148]  [100/312]  eta: 0:03:08  lr: 0.002262  min_lr: 0.002262  loss: 3.4976 (3.4500)  weight_decay: 0.0500 (0.0500)  time: 0.8036  data: 0.2134  max mem: 41794
Epoch: [148]  [110/312]  eta: 0:02:52  lr: 0.002261  min_lr: 0.002261  loss: 3.6132 (3.4616)  weight_decay: 0.0500 (0.0500)  time: 0.7353  data: 0.1441  max mem: 41794
Epoch: [148]  [120/312]  eta: 0:02:43  lr: 0.002260  min_lr: 0.002260  loss: 3.6409 (3.4712)  weight_decay: 0.0500 (0.0500)  time: 0.6532  data: 0.1025  max mem: 41794
Epoch: [148]  [130/312]  eta: 0:02:34  lr: 0.002260  min_lr: 0.002260  loss: 3.3211 (3.4503)  weight_decay: 0.0500 (0.0500)  time: 0.8327  data: 0.2141  max mem: 41794
Epoch: [148]  [140/312]  eta: 0:02:21  lr: 0.002259  min_lr: 0.002259  loss: 3.6236 (3.4695)  weight_decay: 0.0500 (0.0500)  time: 0.6738  data: 0.1141  max mem: 41794
Epoch: [148]  [150/312]  eta: 0:02:14  lr: 0.002258  min_lr: 0.002258  loss: 3.7126 (3.4688)  weight_decay: 0.0500 (0.0500)  time: 0.6870  data: 0.1354  max mem: 41794
Epoch: [148]  [160/312]  eta: 0:02:04  lr: 0.002257  min_lr: 0.002257  loss: 3.7162 (3.4782)  weight_decay: 0.0500 (0.0500)  time: 0.7709  data: 0.1355  max mem: 41794
Epoch: [148]  [170/312]  eta: 0:01:55  lr: 0.002257  min_lr: 0.002257  loss: 3.6930 (3.4785)  weight_decay: 0.0500 (0.0500)  time: 0.7014  data: 0.1265  max mem: 41794
Epoch: [148]  [180/312]  eta: 0:01:47  lr: 0.002256  min_lr: 0.002256  loss: 3.5785 (3.4696)  weight_decay: 0.0500 (0.0500)  time: 0.7977  data: 0.2576  max mem: 41794
Epoch: [148]  [190/312]  eta: 0:01:37  lr: 0.002255  min_lr: 0.002255  loss: 3.6117 (3.4805)  weight_decay: 0.0500 (0.0500)  time: 0.6715  data: 0.1317  max mem: 41794
Epoch: [148]  [200/312]  eta: 0:01:29  lr: 0.002255  min_lr: 0.002255  loss: 3.6007 (3.4782)  weight_decay: 0.0500 (0.0500)  time: 0.6552  data: 0.1106  max mem: 41794
Epoch: [148]  [210/312]  eta: 0:01:21  lr: 0.002254  min_lr: 0.002254  loss: 3.4427 (3.4733)  weight_decay: 0.0500 (0.0500)  time: 0.8321  data: 0.2481  max mem: 41794
Epoch: [148]  [220/312]  eta: 0:01:12  lr: 0.002253  min_lr: 0.002253  loss: 3.4136 (3.4759)  weight_decay: 0.0500 (0.0500)  time: 0.6786  data: 0.1382  max mem: 41794
Epoch: [148]  [230/312]  eta: 0:01:04  lr: 0.002253  min_lr: 0.002253  loss: 3.7008 (3.4810)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.1283  max mem: 41794
Epoch: [148]  [240/312]  eta: 0:00:56  lr: 0.002252  min_lr: 0.002252  loss: 3.7008 (3.4852)  weight_decay: 0.0500 (0.0500)  time: 0.7611  data: 0.1337  max mem: 41794
Epoch: [148]  [250/312]  eta: 0:00:48  lr: 0.002251  min_lr: 0.002251  loss: 3.5971 (3.4786)  weight_decay: 0.0500 (0.0500)  time: 0.7030  data: 0.1186  max mem: 41794
Epoch: [148]  [260/312]  eta: 0:00:40  lr: 0.002250  min_lr: 0.002250  loss: 3.1957 (3.4684)  weight_decay: 0.0500 (0.0500)  time: 0.7718  data: 0.2290  max mem: 41794
Epoch: [148]  [270/312]  eta: 0:00:32  lr: 0.002250  min_lr: 0.002250  loss: 3.1957 (3.4644)  weight_decay: 0.0500 (0.0500)  time: 0.6612  data: 0.1166  max mem: 41794
Epoch: [148]  [280/312]  eta: 0:00:24  lr: 0.002249  min_lr: 0.002249  loss: 3.6103 (3.4708)  weight_decay: 0.0500 (0.0500)  time: 0.6778  data: 0.1132  max mem: 41794
Epoch: [148]  [290/312]  eta: 0:00:17  lr: 0.002248  min_lr: 0.002248  loss: 3.7065 (3.4737)  weight_decay: 0.0500 (0.0500)  time: 0.8425  data: 0.2143  max mem: 41794
Epoch: [148]  [300/312]  eta: 0:00:09  lr: 0.002248  min_lr: 0.002248  loss: 3.4012 (3.4672)  weight_decay: 0.0500 (0.0500)  time: 0.6482  data: 0.1028  max mem: 41794
Epoch: [148]  [310/312]  eta: 0:00:01  lr: 0.002247  min_lr: 0.002247  loss: 3.3932 (3.4625)  weight_decay: 0.0500 (0.0500)  time: 0.4628  data: 0.0013  max mem: 41794
Epoch: [148]  [311/312]  eta: 0:00:00  lr: 0.002247  min_lr: 0.002247  loss: 3.4108 (3.4637)  weight_decay: 0.0500 (0.0500)  time: 0.4631  data: 0.0013  max mem: 41794
Epoch: [148] Total time: 0:03:58 (0.7659 s / it)
Averaged stats: lr: 0.002247  min_lr: 0.002247  loss: 3.4108 (3.4369)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.0942 (1.0942)  acc1: 79.1667 (79.1667)  acc5: 94.6615 (94.6615)  time: 8.9219  data: 8.7138  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4575 (1.3420)  acc1: 71.7448 (72.3680)  acc5: 91.7969 (92.2080)  time: 1.1416  data: 0.9700  max mem: 41794
Test: Total time: 0:00:10 (1.1521 s / it)
* Acc@1 72.368 Acc@5 91.720 loss 1.343
Accuracy of the model on the 50000 test images: 72.4%
Max accuracy: 72.84%
Epoch: [149]  [  0/312]  eta: 1:24:22  lr: 0.002247  min_lr: 0.002247  loss: 3.7760 (3.7760)  weight_decay: 0.0500 (0.0500)  time: 16.2254  data: 15.3356  max mem: 41794
Epoch: [149]  [ 10/312]  eta: 0:12:03  lr: 0.002246  min_lr: 0.002246  loss: 3.2627 (3.2210)  weight_decay: 0.0500 (0.0500)  time: 2.3956  data: 1.3990  max mem: 41794
Epoch: [149]  [ 20/312]  eta: 0:07:28  lr: 0.002245  min_lr: 0.002245  loss: 3.0409 (3.2629)  weight_decay: 0.0500 (0.0500)  time: 0.8014  data: 0.0459  max mem: 41794
Epoch: [149]  [ 30/312]  eta: 0:05:40  lr: 0.002245  min_lr: 0.002245  loss: 3.6217 (3.3207)  weight_decay: 0.0500 (0.0500)  time: 0.5558  data: 0.0478  max mem: 41794
Epoch: [149]  [ 40/312]  eta: 0:04:43  lr: 0.002244  min_lr: 0.002244  loss: 3.6503 (3.3336)  weight_decay: 0.0500 (0.0500)  time: 0.5242  data: 0.0050  max mem: 41794
Epoch: [149]  [ 50/312]  eta: 0:04:06  lr: 0.002243  min_lr: 0.002243  loss: 3.6392 (3.3503)  weight_decay: 0.0500 (0.0500)  time: 0.5295  data: 0.0068  max mem: 41794
Epoch: [149]  [ 60/312]  eta: 0:03:40  lr: 0.002242  min_lr: 0.002242  loss: 3.6217 (3.3643)  weight_decay: 0.0500 (0.0500)  time: 0.5266  data: 0.0081  max mem: 41794
Epoch: [149]  [ 70/312]  eta: 0:03:27  lr: 0.002242  min_lr: 0.002242  loss: 3.4220 (3.3376)  weight_decay: 0.0500 (0.0500)  time: 0.6354  data: 0.0971  max mem: 41794
Epoch: [149]  [ 80/312]  eta: 0:03:18  lr: 0.002241  min_lr: 0.002241  loss: 3.4220 (3.3681)  weight_decay: 0.0500 (0.0500)  time: 0.8042  data: 0.2491  max mem: 41794
Epoch: [149]  [ 90/312]  eta: 0:03:02  lr: 0.002240  min_lr: 0.002240  loss: 3.6865 (3.3883)  weight_decay: 0.0500 (0.0500)  time: 0.7098  data: 0.1764  max mem: 41794
Epoch: [149]  [100/312]  eta: 0:02:54  lr: 0.002240  min_lr: 0.002240  loss: 3.4923 (3.3855)  weight_decay: 0.0500 (0.0500)  time: 0.6946  data: 0.1737  max mem: 41794
Epoch: [149]  [110/312]  eta: 0:02:42  lr: 0.002239  min_lr: 0.002239  loss: 3.5877 (3.3877)  weight_decay: 0.0500 (0.0500)  time: 0.7140  data: 0.2027  max mem: 41794
Epoch: [149]  [120/312]  eta: 0:02:33  lr: 0.002238  min_lr: 0.002238  loss: 3.6643 (3.4033)  weight_decay: 0.0500 (0.0500)  time: 0.6813  data: 0.1653  max mem: 41794
Epoch: [149]  [130/312]  eta: 0:02:26  lr: 0.002237  min_lr: 0.002237  loss: 3.5816 (3.3828)  weight_decay: 0.0500 (0.0500)  time: 0.8049  data: 0.2672  max mem: 41794
Epoch: [149]  [140/312]  eta: 0:02:16  lr: 0.002237  min_lr: 0.002237  loss: 3.2884 (3.3808)  weight_decay: 0.0500 (0.0500)  time: 0.7370  data: 0.2179  max mem: 41794
Epoch: [149]  [150/312]  eta: 0:02:07  lr: 0.002236  min_lr: 0.002236  loss: 3.5708 (3.4042)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.1372  max mem: 41794
Epoch: [149]  [160/312]  eta: 0:02:00  lr: 0.002235  min_lr: 0.002235  loss: 3.7075 (3.4138)  weight_decay: 0.0500 (0.0500)  time: 0.8199  data: 0.1986  max mem: 41794
Epoch: [149]  [170/312]  eta: 0:01:50  lr: 0.002235  min_lr: 0.002235  loss: 3.6394 (3.4004)  weight_decay: 0.0500 (0.0500)  time: 0.7200  data: 0.1562  max mem: 41794
Epoch: [149]  [180/312]  eta: 0:01:43  lr: 0.002234  min_lr: 0.002234  loss: 3.4032 (3.4061)  weight_decay: 0.0500 (0.0500)  time: 0.7099  data: 0.1102  max mem: 41794
Epoch: [149]  [190/312]  eta: 0:01:35  lr: 0.002233  min_lr: 0.002233  loss: 3.5713 (3.4085)  weight_decay: 0.0500 (0.0500)  time: 0.7803  data: 0.1029  max mem: 41794
Epoch: [149]  [200/312]  eta: 0:01:26  lr: 0.002232  min_lr: 0.002232  loss: 3.4570 (3.3993)  weight_decay: 0.0500 (0.0500)  time: 0.6603  data: 0.0811  max mem: 41794
Epoch: [149]  [210/312]  eta: 0:01:19  lr: 0.002232  min_lr: 0.002232  loss: 3.3429 (3.3982)  weight_decay: 0.0500 (0.0500)  time: 0.7344  data: 0.1594  max mem: 41794
Epoch: [149]  [220/312]  eta: 0:01:10  lr: 0.002231  min_lr: 0.002231  loss: 3.1512 (3.3834)  weight_decay: 0.0500 (0.0500)  time: 0.7226  data: 0.1284  max mem: 41794
Epoch: [149]  [230/312]  eta: 0:01:03  lr: 0.002230  min_lr: 0.002230  loss: 3.5103 (3.3948)  weight_decay: 0.0500 (0.0500)  time: 0.6995  data: 0.0866  max mem: 41794
Epoch: [149]  [240/312]  eta: 0:00:55  lr: 0.002230  min_lr: 0.002230  loss: 3.6274 (3.3972)  weight_decay: 0.0500 (0.0500)  time: 0.8458  data: 0.1510  max mem: 41794
Epoch: [149]  [250/312]  eta: 0:00:47  lr: 0.002229  min_lr: 0.002229  loss: 3.3510 (3.3905)  weight_decay: 0.0500 (0.0500)  time: 0.7041  data: 0.1066  max mem: 41794
Epoch: [149]  [260/312]  eta: 0:00:39  lr: 0.002228  min_lr: 0.002228  loss: 2.9473 (3.3779)  weight_decay: 0.0500 (0.0500)  time: 0.6837  data: 0.0923  max mem: 41794
Epoch: [149]  [270/312]  eta: 0:00:32  lr: 0.002227  min_lr: 0.002227  loss: 3.1475 (3.3772)  weight_decay: 0.0500 (0.0500)  time: 0.8084  data: 0.0932  max mem: 41794
Epoch: [149]  [280/312]  eta: 0:00:24  lr: 0.002227  min_lr: 0.002227  loss: 3.5576 (3.3860)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.0692  max mem: 41794
Epoch: [149]  [290/312]  eta: 0:00:16  lr: 0.002226  min_lr: 0.002226  loss: 3.7673 (3.3970)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.1313  max mem: 41794
Epoch: [149]  [300/312]  eta: 0:00:09  lr: 0.002225  min_lr: 0.002225  loss: 3.7564 (3.4047)  weight_decay: 0.0500 (0.0500)  time: 0.6810  data: 0.0781  max mem: 41794
Epoch: [149]  [310/312]  eta: 0:00:01  lr: 0.002225  min_lr: 0.002225  loss: 3.7564 (3.4117)  weight_decay: 0.0500 (0.0500)  time: 0.4789  data: 0.0002  max mem: 41794
Epoch: [149]  [311/312]  eta: 0:00:00  lr: 0.002224  min_lr: 0.002224  loss: 3.6955 (3.4119)  weight_decay: 0.0500 (0.0500)  time: 0.4768  data: 0.0001  max mem: 41794
Epoch: [149] Total time: 0:03:55 (0.7532 s / it)
Averaged stats: lr: 0.002224  min_lr: 0.002224  loss: 3.6955 (3.4492)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.2414 (1.2414)  acc1: 80.5990 (80.5990)  acc5: 94.2708 (94.2708)  time: 8.5560  data: 8.3467  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6157 (1.5194)  acc1: 70.4427 (72.0320)  acc5: 90.6250 (91.4720)  time: 1.0987  data: 0.9275  max mem: 41794
Test: Total time: 0:00:10 (1.1216 s / it)
* Acc@1 72.462 Acc@5 91.282 loss 1.527
Accuracy of the model on the 50000 test images: 72.5%
Max accuracy: 72.84%
Epoch: [150]  [  0/312]  eta: 1:18:05  lr: 0.002224  min_lr: 0.002224  loss: 2.5549 (2.5549)  weight_decay: 0.0500 (0.0500)  time: 15.0179  data: 12.6621  max mem: 41794
Epoch: [150]  [ 10/312]  eta: 0:10:15  lr: 0.002224  min_lr: 0.002224  loss: 3.4434 (3.2719)  weight_decay: 0.0500 (0.0500)  time: 2.0388  data: 1.2860  max mem: 41794
Epoch: [150]  [ 20/312]  eta: 0:07:14  lr: 0.002223  min_lr: 0.002223  loss: 3.5055 (3.4194)  weight_decay: 0.0500 (0.0500)  time: 0.8108  data: 0.1557  max mem: 41794
Epoch: [150]  [ 30/312]  eta: 0:05:33  lr: 0.002222  min_lr: 0.002222  loss: 3.6098 (3.5085)  weight_decay: 0.0500 (0.0500)  time: 0.7142  data: 0.0818  max mem: 41794
Epoch: [150]  [ 40/312]  eta: 0:04:53  lr: 0.002222  min_lr: 0.002222  loss: 3.6723 (3.5098)  weight_decay: 0.0500 (0.0500)  time: 0.6520  data: 0.1343  max mem: 41794
Epoch: [150]  [ 50/312]  eta: 0:04:33  lr: 0.002221  min_lr: 0.002221  loss: 3.4613 (3.4650)  weight_decay: 0.0500 (0.0500)  time: 0.8248  data: 0.3195  max mem: 41794
Epoch: [150]  [ 60/312]  eta: 0:03:59  lr: 0.002220  min_lr: 0.002220  loss: 3.5523 (3.4767)  weight_decay: 0.0500 (0.0500)  time: 0.6888  data: 0.1858  max mem: 41794
Epoch: [150]  [ 70/312]  eta: 0:03:47  lr: 0.002219  min_lr: 0.002219  loss: 3.6035 (3.4501)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.1928  max mem: 41794
Epoch: [150]  [ 80/312]  eta: 0:03:36  lr: 0.002219  min_lr: 0.002219  loss: 3.4751 (3.4293)  weight_decay: 0.0500 (0.0500)  time: 0.8741  data: 0.3701  max mem: 41794
Epoch: [150]  [ 90/312]  eta: 0:03:16  lr: 0.002218  min_lr: 0.002218  loss: 3.5900 (3.4592)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.1779  max mem: 41794
Epoch: [150]  [100/312]  eta: 0:03:07  lr: 0.002217  min_lr: 0.002217  loss: 3.6116 (3.4423)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.1869  max mem: 41794
Epoch: [150]  [110/312]  eta: 0:02:51  lr: 0.002217  min_lr: 0.002217  loss: 3.2761 (3.4280)  weight_decay: 0.0500 (0.0500)  time: 0.6961  data: 0.1870  max mem: 41794
Epoch: [150]  [120/312]  eta: 0:02:43  lr: 0.002216  min_lr: 0.002216  loss: 3.2761 (3.4173)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1833  max mem: 41794
Epoch: [150]  [130/312]  eta: 0:02:35  lr: 0.002215  min_lr: 0.002215  loss: 3.6191 (3.4181)  weight_decay: 0.0500 (0.0500)  time: 0.8684  data: 0.3757  max mem: 41794
Epoch: [150]  [140/312]  eta: 0:02:22  lr: 0.002214  min_lr: 0.002214  loss: 3.6944 (3.4420)  weight_decay: 0.0500 (0.0500)  time: 0.6861  data: 0.1931  max mem: 41794
Epoch: [150]  [150/312]  eta: 0:02:14  lr: 0.002214  min_lr: 0.002214  loss: 3.7273 (3.4512)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.1862  max mem: 41794
Epoch: [150]  [160/312]  eta: 0:02:06  lr: 0.002213  min_lr: 0.002213  loss: 3.5767 (3.4644)  weight_decay: 0.0500 (0.0500)  time: 0.8595  data: 0.3571  max mem: 41794
Epoch: [150]  [170/312]  eta: 0:01:55  lr: 0.002212  min_lr: 0.002212  loss: 3.5767 (3.4658)  weight_decay: 0.0500 (0.0500)  time: 0.6652  data: 0.1715  max mem: 41794
Epoch: [150]  [180/312]  eta: 0:01:47  lr: 0.002212  min_lr: 0.002212  loss: 3.4040 (3.4521)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.1849  max mem: 41794
Epoch: [150]  [190/312]  eta: 0:01:37  lr: 0.002211  min_lr: 0.002211  loss: 3.1045 (3.4470)  weight_decay: 0.0500 (0.0500)  time: 0.6760  data: 0.1849  max mem: 41794
Epoch: [150]  [200/312]  eta: 0:01:29  lr: 0.002210  min_lr: 0.002210  loss: 3.6684 (3.4523)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1925  max mem: 41794
Epoch: [150]  [210/312]  eta: 0:01:22  lr: 0.002209  min_lr: 0.002209  loss: 3.6826 (3.4653)  weight_decay: 0.0500 (0.0500)  time: 0.8780  data: 0.3807  max mem: 41794
Epoch: [150]  [220/312]  eta: 0:01:12  lr: 0.002209  min_lr: 0.002209  loss: 3.6862 (3.4694)  weight_decay: 0.0500 (0.0500)  time: 0.6956  data: 0.1888  max mem: 41794
Epoch: [150]  [230/312]  eta: 0:01:04  lr: 0.002208  min_lr: 0.002208  loss: 3.5284 (3.4646)  weight_decay: 0.0500 (0.0500)  time: 0.6701  data: 0.1707  max mem: 41794
Epoch: [150]  [240/312]  eta: 0:00:57  lr: 0.002207  min_lr: 0.002207  loss: 3.5681 (3.4746)  weight_decay: 0.0500 (0.0500)  time: 0.8494  data: 0.3564  max mem: 41794
Epoch: [150]  [250/312]  eta: 0:00:48  lr: 0.002207  min_lr: 0.002207  loss: 3.5759 (3.4725)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.1864  max mem: 41794
Epoch: [150]  [260/312]  eta: 0:00:40  lr: 0.002206  min_lr: 0.002206  loss: 3.3324 (3.4587)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.1877  max mem: 41794
Epoch: [150]  [270/312]  eta: 0:00:32  lr: 0.002205  min_lr: 0.002205  loss: 3.1248 (3.4530)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1877  max mem: 41794
Epoch: [150]  [280/312]  eta: 0:00:24  lr: 0.002204  min_lr: 0.002204  loss: 3.5100 (3.4576)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1687  max mem: 41794
Epoch: [150]  [290/312]  eta: 0:00:17  lr: 0.002204  min_lr: 0.002204  loss: 3.5646 (3.4570)  weight_decay: 0.0500 (0.0500)  time: 0.7812  data: 0.2819  max mem: 41794
Epoch: [150]  [300/312]  eta: 0:00:09  lr: 0.002203  min_lr: 0.002203  loss: 3.5097 (3.4563)  weight_decay: 0.0500 (0.0500)  time: 0.5994  data: 0.1135  max mem: 41794
Epoch: [150]  [310/312]  eta: 0:00:01  lr: 0.002202  min_lr: 0.002202  loss: 3.5183 (3.4564)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [150]  [311/312]  eta: 0:00:00  lr: 0.002202  min_lr: 0.002202  loss: 3.5339 (3.4583)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0001  max mem: 41794
Epoch: [150] Total time: 0:03:57 (0.7618 s / it)
Averaged stats: lr: 0.002202  min_lr: 0.002202  loss: 3.5339 (3.4333)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.1225 (1.1225)  acc1: 79.8177 (79.8177)  acc5: 95.3125 (95.3125)  time: 8.6712  data: 8.4696  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5054 (1.4750)  acc1: 71.4844 (71.9200)  acc5: 91.1458 (91.2000)  time: 1.1166  data: 0.9411  max mem: 41794
Test: Total time: 0:00:10 (1.1538 s / it)
* Acc@1 71.832 Acc@5 91.120 loss 1.488
Accuracy of the model on the 50000 test images: 71.8%
Max accuracy: 72.84%
Epoch: [151]  [  0/312]  eta: 1:21:17  lr: 0.002202  min_lr: 0.002202  loss: 2.9623 (2.9623)  weight_decay: 0.0500 (0.0500)  time: 15.6335  data: 11.6899  max mem: 41794
Epoch: [151]  [ 10/312]  eta: 0:11:16  lr: 0.002201  min_lr: 0.002201  loss: 3.1708 (3.1901)  weight_decay: 0.0500 (0.0500)  time: 2.2388  data: 1.2953  max mem: 41794
Epoch: [151]  [ 20/312]  eta: 0:07:33  lr: 0.002201  min_lr: 0.002201  loss: 3.0979 (3.1949)  weight_decay: 0.0500 (0.0500)  time: 0.8500  data: 0.1929  max mem: 41794
Epoch: [151]  [ 30/312]  eta: 0:05:46  lr: 0.002200  min_lr: 0.002200  loss: 3.5405 (3.3435)  weight_decay: 0.0500 (0.0500)  time: 0.6747  data: 0.0735  max mem: 41794
Epoch: [151]  [ 40/312]  eta: 0:04:49  lr: 0.002199  min_lr: 0.002199  loss: 3.5305 (3.3436)  weight_decay: 0.0500 (0.0500)  time: 0.5525  data: 0.0089  max mem: 41794
Epoch: [151]  [ 50/312]  eta: 0:04:26  lr: 0.002198  min_lr: 0.002198  loss: 3.5558 (3.4136)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.0651  max mem: 41794
Epoch: [151]  [ 60/312]  eta: 0:03:54  lr: 0.002198  min_lr: 0.002198  loss: 3.5558 (3.3962)  weight_decay: 0.0500 (0.0500)  time: 0.6572  data: 0.0667  max mem: 41794
Epoch: [151]  [ 70/312]  eta: 0:03:45  lr: 0.002197  min_lr: 0.002197  loss: 3.4962 (3.4207)  weight_decay: 0.0500 (0.0500)  time: 0.7215  data: 0.1346  max mem: 41794
Epoch: [151]  [ 80/312]  eta: 0:03:30  lr: 0.002196  min_lr: 0.002196  loss: 3.3959 (3.3810)  weight_decay: 0.0500 (0.0500)  time: 0.8328  data: 0.1485  max mem: 41794
Epoch: [151]  [ 90/312]  eta: 0:03:16  lr: 0.002196  min_lr: 0.002196  loss: 3.2822 (3.3814)  weight_decay: 0.0500 (0.0500)  time: 0.7142  data: 0.1245  max mem: 41794
Epoch: [151]  [100/312]  eta: 0:03:08  lr: 0.002195  min_lr: 0.002195  loss: 3.5531 (3.3789)  weight_decay: 0.0500 (0.0500)  time: 0.8066  data: 0.2102  max mem: 41794
Epoch: [151]  [110/312]  eta: 0:02:52  lr: 0.002194  min_lr: 0.002194  loss: 3.6110 (3.4004)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.1019  max mem: 41794
Epoch: [151]  [120/312]  eta: 0:02:44  lr: 0.002193  min_lr: 0.002193  loss: 3.6110 (3.3999)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.1339  max mem: 41794
Epoch: [151]  [130/312]  eta: 0:02:35  lr: 0.002193  min_lr: 0.002193  loss: 3.4602 (3.4127)  weight_decay: 0.0500 (0.0500)  time: 0.8817  data: 0.2463  max mem: 41794
Epoch: [151]  [140/312]  eta: 0:02:22  lr: 0.002192  min_lr: 0.002192  loss: 3.4747 (3.4093)  weight_decay: 0.0500 (0.0500)  time: 0.6714  data: 0.1131  max mem: 41794
Epoch: [151]  [150/312]  eta: 0:02:15  lr: 0.002191  min_lr: 0.002191  loss: 3.6091 (3.4219)  weight_decay: 0.0500 (0.0500)  time: 0.6965  data: 0.1203  max mem: 41794
Epoch: [151]  [160/312]  eta: 0:02:05  lr: 0.002191  min_lr: 0.002191  loss: 3.6745 (3.4319)  weight_decay: 0.0500 (0.0500)  time: 0.8038  data: 0.1203  max mem: 41794
Epoch: [151]  [170/312]  eta: 0:01:56  lr: 0.002190  min_lr: 0.002190  loss: 3.5206 (3.4288)  weight_decay: 0.0500 (0.0500)  time: 0.7055  data: 0.0985  max mem: 41794
Epoch: [151]  [180/312]  eta: 0:01:48  lr: 0.002189  min_lr: 0.002189  loss: 3.5229 (3.4425)  weight_decay: 0.0500 (0.0500)  time: 0.7680  data: 0.1758  max mem: 41794
Epoch: [151]  [190/312]  eta: 0:01:38  lr: 0.002188  min_lr: 0.002188  loss: 3.3487 (3.4266)  weight_decay: 0.0500 (0.0500)  time: 0.6741  data: 0.0780  max mem: 41794
Epoch: [151]  [200/312]  eta: 0:01:30  lr: 0.002188  min_lr: 0.002188  loss: 3.3212 (3.4250)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.0902  max mem: 41794
Epoch: [151]  [210/312]  eta: 0:01:22  lr: 0.002187  min_lr: 0.002187  loss: 3.4436 (3.4222)  weight_decay: 0.0500 (0.0500)  time: 0.8449  data: 0.1838  max mem: 41794
Epoch: [151]  [220/312]  eta: 0:01:13  lr: 0.002186  min_lr: 0.002186  loss: 3.5672 (3.4298)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.0943  max mem: 41794
Epoch: [151]  [230/312]  eta: 0:01:05  lr: 0.002186  min_lr: 0.002186  loss: 3.6213 (3.4311)  weight_decay: 0.0500 (0.0500)  time: 0.7189  data: 0.1156  max mem: 41794
Epoch: [151]  [240/312]  eta: 0:00:57  lr: 0.002185  min_lr: 0.002185  loss: 3.6190 (3.4319)  weight_decay: 0.0500 (0.0500)  time: 0.8110  data: 0.1172  max mem: 41794
Epoch: [151]  [250/312]  eta: 0:00:48  lr: 0.002184  min_lr: 0.002184  loss: 3.6190 (3.4371)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.0783  max mem: 41794
Epoch: [151]  [260/312]  eta: 0:00:41  lr: 0.002183  min_lr: 0.002183  loss: 3.6638 (3.4346)  weight_decay: 0.0500 (0.0500)  time: 0.7942  data: 0.2017  max mem: 41794
Epoch: [151]  [270/312]  eta: 0:00:32  lr: 0.002183  min_lr: 0.002183  loss: 3.3201 (3.4346)  weight_decay: 0.0500 (0.0500)  time: 0.7299  data: 0.1261  max mem: 41794
Epoch: [151]  [280/312]  eta: 0:00:25  lr: 0.002182  min_lr: 0.002182  loss: 3.3078 (3.4335)  weight_decay: 0.0500 (0.0500)  time: 0.6261  data: 0.0807  max mem: 41794
Epoch: [151]  [290/312]  eta: 0:00:17  lr: 0.002181  min_lr: 0.002181  loss: 3.4378 (3.4380)  weight_decay: 0.0500 (0.0500)  time: 0.7805  data: 0.1576  max mem: 41794
Epoch: [151]  [300/312]  eta: 0:00:09  lr: 0.002181  min_lr: 0.002181  loss: 3.6961 (3.4433)  weight_decay: 0.0500 (0.0500)  time: 0.6467  data: 0.0778  max mem: 41794
Epoch: [151]  [310/312]  eta: 0:00:01  lr: 0.002180  min_lr: 0.002180  loss: 3.6857 (3.4463)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [151]  [311/312]  eta: 0:00:00  lr: 0.002180  min_lr: 0.002180  loss: 3.6857 (3.4472)  weight_decay: 0.0500 (0.0500)  time: 0.4624  data: 0.0001  max mem: 41794
Epoch: [151] Total time: 0:04:00 (0.7704 s / it)
Averaged stats: lr: 0.002180  min_lr: 0.002180  loss: 3.6857 (3.4337)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:24  loss: 1.4901 (1.4901)  acc1: 77.8646 (77.8646)  acc5: 94.2708 (94.2708)  time: 9.4361  data: 9.2251  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6586 (1.6334)  acc1: 71.4844 (72.3520)  acc5: 91.7969 (91.7280)  time: 1.1964  data: 1.0251  max mem: 41794
Test: Total time: 0:00:10 (1.2118 s / it)
* Acc@1 72.564 Acc@5 91.604 loss 1.643
Accuracy of the model on the 50000 test images: 72.6%
Max accuracy: 72.84%
Epoch: [152]  [  0/312]  eta: 1:22:34  lr: 0.002180  min_lr: 0.002180  loss: 3.8165 (3.8165)  weight_decay: 0.0500 (0.0500)  time: 15.8794  data: 11.7706  max mem: 41794
Epoch: [152]  [ 10/312]  eta: 0:11:14  lr: 0.002179  min_lr: 0.002179  loss: 3.5138 (3.5612)  weight_decay: 0.0500 (0.0500)  time: 2.2350  data: 1.3957  max mem: 41794
Epoch: [152]  [ 20/312]  eta: 0:07:24  lr: 0.002178  min_lr: 0.002178  loss: 3.5138 (3.5120)  weight_decay: 0.0500 (0.0500)  time: 0.8051  data: 0.1899  max mem: 41794
Epoch: [152]  [ 30/312]  eta: 0:05:37  lr: 0.002178  min_lr: 0.002178  loss: 3.4980 (3.4289)  weight_decay: 0.0500 (0.0500)  time: 0.6258  data: 0.0142  max mem: 41794
Epoch: [152]  [ 40/312]  eta: 0:04:43  lr: 0.002177  min_lr: 0.002177  loss: 3.4190 (3.4003)  weight_decay: 0.0500 (0.0500)  time: 0.5375  data: 0.0085  max mem: 41794
Epoch: [152]  [ 50/312]  eta: 0:04:21  lr: 0.002176  min_lr: 0.002176  loss: 3.7086 (3.4718)  weight_decay: 0.0500 (0.0500)  time: 0.6917  data: 0.0568  max mem: 41794
Epoch: [152]  [ 60/312]  eta: 0:03:51  lr: 0.002175  min_lr: 0.002175  loss: 3.7484 (3.4614)  weight_decay: 0.0500 (0.0500)  time: 0.6680  data: 0.0693  max mem: 41794
Epoch: [152]  [ 70/312]  eta: 0:03:40  lr: 0.002175  min_lr: 0.002175  loss: 3.4549 (3.4588)  weight_decay: 0.0500 (0.0500)  time: 0.6854  data: 0.1442  max mem: 41794
Epoch: [152]  [ 80/312]  eta: 0:03:29  lr: 0.002174  min_lr: 0.002174  loss: 3.2498 (3.4377)  weight_decay: 0.0500 (0.0500)  time: 0.8512  data: 0.2597  max mem: 41794
Epoch: [152]  [ 90/312]  eta: 0:03:10  lr: 0.002173  min_lr: 0.002173  loss: 3.2473 (3.4082)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.1333  max mem: 41794
Epoch: [152]  [100/312]  eta: 0:03:01  lr: 0.002173  min_lr: 0.002173  loss: 3.5593 (3.4208)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.1294  max mem: 41794
Epoch: [152]  [110/312]  eta: 0:02:46  lr: 0.002172  min_lr: 0.002172  loss: 3.7357 (3.4388)  weight_decay: 0.0500 (0.0500)  time: 0.6688  data: 0.1293  max mem: 41794
Epoch: [152]  [120/312]  eta: 0:02:38  lr: 0.002171  min_lr: 0.002171  loss: 3.5468 (3.4449)  weight_decay: 0.0500 (0.0500)  time: 0.6757  data: 0.1623  max mem: 41794
Epoch: [152]  [130/312]  eta: 0:02:31  lr: 0.002170  min_lr: 0.002170  loss: 3.5447 (3.4398)  weight_decay: 0.0500 (0.0500)  time: 0.8666  data: 0.3578  max mem: 41794
Epoch: [152]  [140/312]  eta: 0:02:18  lr: 0.002170  min_lr: 0.002170  loss: 3.5862 (3.4405)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.1980  max mem: 41794
Epoch: [152]  [150/312]  eta: 0:02:11  lr: 0.002169  min_lr: 0.002169  loss: 3.5478 (3.4298)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1891  max mem: 41794
Epoch: [152]  [160/312]  eta: 0:02:03  lr: 0.002168  min_lr: 0.002168  loss: 3.5478 (3.4341)  weight_decay: 0.0500 (0.0500)  time: 0.8575  data: 0.3566  max mem: 41794
Epoch: [152]  [170/312]  eta: 0:01:52  lr: 0.002168  min_lr: 0.002168  loss: 3.5676 (3.4344)  weight_decay: 0.0500 (0.0500)  time: 0.6688  data: 0.1703  max mem: 41794
Epoch: [152]  [180/312]  eta: 0:01:45  lr: 0.002167  min_lr: 0.002167  loss: 3.5769 (3.4384)  weight_decay: 0.0500 (0.0500)  time: 0.7128  data: 0.2114  max mem: 41794
Epoch: [152]  [190/312]  eta: 0:01:35  lr: 0.002166  min_lr: 0.002166  loss: 3.5423 (3.4365)  weight_decay: 0.0500 (0.0500)  time: 0.7060  data: 0.2111  max mem: 41794
Epoch: [152]  [200/312]  eta: 0:01:28  lr: 0.002165  min_lr: 0.002165  loss: 3.5143 (3.4421)  weight_decay: 0.0500 (0.0500)  time: 0.6592  data: 0.1678  max mem: 41794
Epoch: [152]  [210/312]  eta: 0:01:20  lr: 0.002165  min_lr: 0.002165  loss: 3.3906 (3.4385)  weight_decay: 0.0500 (0.0500)  time: 0.8448  data: 0.3381  max mem: 41794
Epoch: [152]  [220/312]  eta: 0:01:11  lr: 0.002164  min_lr: 0.002164  loss: 3.5477 (3.4453)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.1709  max mem: 41794
Epoch: [152]  [230/312]  eta: 0:01:04  lr: 0.002163  min_lr: 0.002163  loss: 3.6176 (3.4394)  weight_decay: 0.0500 (0.0500)  time: 0.6788  data: 0.1626  max mem: 41794
Epoch: [152]  [240/312]  eta: 0:00:56  lr: 0.002163  min_lr: 0.002163  loss: 3.5567 (3.4426)  weight_decay: 0.0500 (0.0500)  time: 0.8655  data: 0.3284  max mem: 41794
Epoch: [152]  [250/312]  eta: 0:00:48  lr: 0.002162  min_lr: 0.002162  loss: 3.6253 (3.4488)  weight_decay: 0.0500 (0.0500)  time: 0.6915  data: 0.1683  max mem: 41794
Epoch: [152]  [260/312]  eta: 0:00:40  lr: 0.002161  min_lr: 0.002161  loss: 3.6253 (3.4420)  weight_decay: 0.0500 (0.0500)  time: 0.6588  data: 0.1490  max mem: 41794
Epoch: [152]  [270/312]  eta: 0:00:32  lr: 0.002160  min_lr: 0.002160  loss: 3.6397 (3.4480)  weight_decay: 0.0500 (0.0500)  time: 0.6419  data: 0.1471  max mem: 41794
Epoch: [152]  [280/312]  eta: 0:00:24  lr: 0.002160  min_lr: 0.002160  loss: 3.6397 (3.4482)  weight_decay: 0.0500 (0.0500)  time: 0.7191  data: 0.1816  max mem: 41794
Epoch: [152]  [290/312]  eta: 0:00:17  lr: 0.002159  min_lr: 0.002159  loss: 3.7211 (3.4550)  weight_decay: 0.0500 (0.0500)  time: 0.8838  data: 0.3424  max mem: 41794
Epoch: [152]  [300/312]  eta: 0:00:09  lr: 0.002158  min_lr: 0.002158  loss: 3.7211 (3.4571)  weight_decay: 0.0500 (0.0500)  time: 0.6440  data: 0.1612  max mem: 41794
Epoch: [152]  [310/312]  eta: 0:00:01  lr: 0.002158  min_lr: 0.002158  loss: 3.5372 (3.4514)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [152]  [311/312]  eta: 0:00:00  lr: 0.002157  min_lr: 0.002157  loss: 3.5372 (3.4531)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [152] Total time: 0:03:56 (0.7584 s / it)
Averaged stats: lr: 0.002157  min_lr: 0.002157  loss: 3.5372 (3.4367)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.2600 (1.2600)  acc1: 80.3385 (80.3385)  acc5: 94.5312 (94.5312)  time: 8.1326  data: 7.9299  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5169 (1.4492)  acc1: 72.2656 (72.6720)  acc5: 91.5365 (91.8080)  time: 1.0517  data: 0.8812  max mem: 41794
Test: Total time: 0:00:09 (1.0621 s / it)
* Acc@1 72.742 Acc@5 91.742 loss 1.459
Accuracy of the model on the 50000 test images: 72.7%
Max accuracy: 72.84%
Epoch: [153]  [  0/312]  eta: 1:18:56  lr: 0.002157  min_lr: 0.002157  loss: 3.2619 (3.2619)  weight_decay: 0.0500 (0.0500)  time: 15.1798  data: 14.7083  max mem: 41794
Epoch: [153]  [ 10/312]  eta: 0:11:28  lr: 0.002157  min_lr: 0.002157  loss: 3.5751 (3.4800)  weight_decay: 0.0500 (0.0500)  time: 2.2788  data: 1.3380  max mem: 41794
Epoch: [153]  [ 20/312]  eta: 0:07:32  lr: 0.002156  min_lr: 0.002156  loss: 3.6248 (3.5707)  weight_decay: 0.0500 (0.0500)  time: 0.8689  data: 0.0830  max mem: 41794
Epoch: [153]  [ 30/312]  eta: 0:05:43  lr: 0.002155  min_lr: 0.002155  loss: 3.6592 (3.5558)  weight_decay: 0.0500 (0.0500)  time: 0.6326  data: 0.0878  max mem: 41794
Epoch: [153]  [ 40/312]  eta: 0:04:47  lr: 0.002155  min_lr: 0.002155  loss: 3.5344 (3.5590)  weight_decay: 0.0500 (0.0500)  time: 0.5357  data: 0.0056  max mem: 41794
Epoch: [153]  [ 50/312]  eta: 0:04:21  lr: 0.002154  min_lr: 0.002154  loss: 3.5344 (3.5528)  weight_decay: 0.0500 (0.0500)  time: 0.6576  data: 0.0367  max mem: 41794
Epoch: [153]  [ 60/312]  eta: 0:03:50  lr: 0.002153  min_lr: 0.002153  loss: 3.6632 (3.5893)  weight_decay: 0.0500 (0.0500)  time: 0.6282  data: 0.0367  max mem: 41794
Epoch: [153]  [ 70/312]  eta: 0:03:43  lr: 0.002152  min_lr: 0.002152  loss: 3.5999 (3.5425)  weight_decay: 0.0500 (0.0500)  time: 0.7298  data: 0.1158  max mem: 41794
Epoch: [153]  [ 80/312]  eta: 0:03:27  lr: 0.002152  min_lr: 0.002152  loss: 3.2759 (3.5121)  weight_decay: 0.0500 (0.0500)  time: 0.8230  data: 0.1159  max mem: 41794
Epoch: [153]  [ 90/312]  eta: 0:03:12  lr: 0.002151  min_lr: 0.002151  loss: 3.4703 (3.5204)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.0889  max mem: 41794
Epoch: [153]  [100/312]  eta: 0:03:04  lr: 0.002150  min_lr: 0.002150  loss: 3.4885 (3.5043)  weight_decay: 0.0500 (0.0500)  time: 0.7916  data: 0.1995  max mem: 41794
Epoch: [153]  [110/312]  eta: 0:02:49  lr: 0.002149  min_lr: 0.002149  loss: 3.3939 (3.4943)  weight_decay: 0.0500 (0.0500)  time: 0.7104  data: 0.1113  max mem: 41794
Epoch: [153]  [120/312]  eta: 0:02:41  lr: 0.002149  min_lr: 0.002149  loss: 3.4887 (3.4951)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.0718  max mem: 41794
Epoch: [153]  [130/312]  eta: 0:02:33  lr: 0.002148  min_lr: 0.002148  loss: 3.4887 (3.4944)  weight_decay: 0.0500 (0.0500)  time: 0.8757  data: 0.1528  max mem: 41794
Epoch: [153]  [140/312]  eta: 0:02:20  lr: 0.002147  min_lr: 0.002147  loss: 3.6872 (3.5046)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.0815  max mem: 41794
Epoch: [153]  [150/312]  eta: 0:02:13  lr: 0.002147  min_lr: 0.002147  loss: 3.7051 (3.5004)  weight_decay: 0.0500 (0.0500)  time: 0.7090  data: 0.0794  max mem: 41794
Epoch: [153]  [160/312]  eta: 0:02:05  lr: 0.002146  min_lr: 0.002146  loss: 3.3947 (3.4888)  weight_decay: 0.0500 (0.0500)  time: 0.8455  data: 0.0794  max mem: 41794
Epoch: [153]  [170/312]  eta: 0:01:55  lr: 0.002145  min_lr: 0.002145  loss: 3.6334 (3.4985)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.0667  max mem: 41794
Epoch: [153]  [180/312]  eta: 0:01:46  lr: 0.002144  min_lr: 0.002144  loss: 3.6200 (3.4965)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.1122  max mem: 41794
Epoch: [153]  [190/312]  eta: 0:01:36  lr: 0.002144  min_lr: 0.002144  loss: 3.2616 (3.4819)  weight_decay: 0.0500 (0.0500)  time: 0.6428  data: 0.0461  max mem: 41794
Epoch: [153]  [200/312]  eta: 0:01:29  lr: 0.002143  min_lr: 0.002143  loss: 3.3800 (3.4849)  weight_decay: 0.0500 (0.0500)  time: 0.7130  data: 0.0796  max mem: 41794
Epoch: [153]  [210/312]  eta: 0:01:21  lr: 0.002142  min_lr: 0.002142  loss: 3.5052 (3.4826)  weight_decay: 0.0500 (0.0500)  time: 0.8826  data: 0.1368  max mem: 41794
Epoch: [153]  [220/312]  eta: 0:01:12  lr: 0.002142  min_lr: 0.002142  loss: 3.4536 (3.4770)  weight_decay: 0.0500 (0.0500)  time: 0.6656  data: 0.0593  max mem: 41794
Epoch: [153]  [230/312]  eta: 0:01:04  lr: 0.002141  min_lr: 0.002141  loss: 3.4655 (3.4741)  weight_decay: 0.0500 (0.0500)  time: 0.6787  data: 0.0664  max mem: 41794
Epoch: [153]  [240/312]  eta: 0:00:56  lr: 0.002140  min_lr: 0.002140  loss: 3.2446 (3.4626)  weight_decay: 0.0500 (0.0500)  time: 0.7913  data: 0.0649  max mem: 41794
Epoch: [153]  [250/312]  eta: 0:00:48  lr: 0.002139  min_lr: 0.002139  loss: 3.3034 (3.4630)  weight_decay: 0.0500 (0.0500)  time: 0.7104  data: 0.0999  max mem: 41794
Epoch: [153]  [260/312]  eta: 0:00:40  lr: 0.002139  min_lr: 0.002139  loss: 3.5991 (3.4678)  weight_decay: 0.0500 (0.0500)  time: 0.7739  data: 0.1731  max mem: 41794
Epoch: [153]  [270/312]  eta: 0:00:32  lr: 0.002138  min_lr: 0.002138  loss: 3.6740 (3.4730)  weight_decay: 0.0500 (0.0500)  time: 0.6773  data: 0.0756  max mem: 41794
Epoch: [153]  [280/312]  eta: 0:00:24  lr: 0.002137  min_lr: 0.002137  loss: 3.7103 (3.4787)  weight_decay: 0.0500 (0.0500)  time: 0.7016  data: 0.1057  max mem: 41794
Epoch: [153]  [290/312]  eta: 0:00:17  lr: 0.002137  min_lr: 0.002137  loss: 3.5380 (3.4762)  weight_decay: 0.0500 (0.0500)  time: 0.8531  data: 0.1580  max mem: 41794
Epoch: [153]  [300/312]  eta: 0:00:09  lr: 0.002136  min_lr: 0.002136  loss: 3.5110 (3.4789)  weight_decay: 0.0500 (0.0500)  time: 0.6329  data: 0.0544  max mem: 41794
Epoch: [153]  [310/312]  eta: 0:00:01  lr: 0.002135  min_lr: 0.002135  loss: 3.6190 (3.4794)  weight_decay: 0.0500 (0.0500)  time: 0.4606  data: 0.0001  max mem: 41794
Epoch: [153]  [311/312]  eta: 0:00:00  lr: 0.002135  min_lr: 0.002135  loss: 3.6520 (3.4804)  weight_decay: 0.0500 (0.0500)  time: 0.4606  data: 0.0001  max mem: 41794
Epoch: [153] Total time: 0:03:59 (0.7674 s / it)
Averaged stats: lr: 0.002135  min_lr: 0.002135  loss: 3.6520 (3.4296)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.3063 (1.3063)  acc1: 81.1198 (81.1198)  acc5: 95.1823 (95.1823)  time: 8.3097  data: 8.0963  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5441 (1.5064)  acc1: 71.8750 (73.2640)  acc5: 91.2760 (91.5840)  time: 1.0719  data: 0.8997  max mem: 41794
Test: Total time: 0:00:09 (1.1002 s / it)
* Acc@1 73.112 Acc@5 91.880 loss 1.498
Accuracy of the model on the 50000 test images: 73.1%
Max accuracy: 73.11%
Epoch: [154]  [  0/312]  eta: 1:24:04  lr: 0.002135  min_lr: 0.002135  loss: 3.5955 (3.5955)  weight_decay: 0.0500 (0.0500)  time: 16.1670  data: 15.6857  max mem: 41794
Epoch: [154]  [ 10/312]  eta: 0:12:00  lr: 0.002134  min_lr: 0.002134  loss: 3.5955 (3.3967)  weight_decay: 0.0500 (0.0500)  time: 2.3859  data: 1.4268  max mem: 41794
Epoch: [154]  [ 20/312]  eta: 0:07:22  lr: 0.002134  min_lr: 0.002134  loss: 3.6579 (3.3864)  weight_decay: 0.0500 (0.0500)  time: 0.7844  data: 0.0149  max mem: 41794
Epoch: [154]  [ 30/312]  eta: 0:05:37  lr: 0.002133  min_lr: 0.002133  loss: 3.4325 (3.3365)  weight_decay: 0.0500 (0.0500)  time: 0.5402  data: 0.0187  max mem: 41794
Epoch: [154]  [ 40/312]  eta: 0:04:38  lr: 0.002132  min_lr: 0.002132  loss: 3.1526 (3.3298)  weight_decay: 0.0500 (0.0500)  time: 0.5090  data: 0.0059  max mem: 41794
Epoch: [154]  [ 50/312]  eta: 0:04:02  lr: 0.002131  min_lr: 0.002131  loss: 3.0900 (3.2540)  weight_decay: 0.0500 (0.0500)  time: 0.5039  data: 0.0070  max mem: 41794
Epoch: [154]  [ 60/312]  eta: 0:03:35  lr: 0.002131  min_lr: 0.002131  loss: 3.1724 (3.2867)  weight_decay: 0.0500 (0.0500)  time: 0.5109  data: 0.0073  max mem: 41794
Epoch: [154]  [ 70/312]  eta: 0:03:23  lr: 0.002130  min_lr: 0.002130  loss: 3.4525 (3.2914)  weight_decay: 0.0500 (0.0500)  time: 0.6236  data: 0.0812  max mem: 41794
Epoch: [154]  [ 80/312]  eta: 0:03:14  lr: 0.002129  min_lr: 0.002129  loss: 3.4525 (3.3286)  weight_decay: 0.0500 (0.0500)  time: 0.7873  data: 0.1931  max mem: 41794
Epoch: [154]  [ 90/312]  eta: 0:03:00  lr: 0.002129  min_lr: 0.002129  loss: 3.5130 (3.3278)  weight_decay: 0.0500 (0.0500)  time: 0.7287  data: 0.1338  max mem: 41794
Epoch: [154]  [100/312]  eta: 0:02:52  lr: 0.002128  min_lr: 0.002128  loss: 3.4646 (3.3237)  weight_decay: 0.0500 (0.0500)  time: 0.7040  data: 0.0578  max mem: 41794
Epoch: [154]  [110/312]  eta: 0:02:43  lr: 0.002127  min_lr: 0.002127  loss: 3.6007 (3.3551)  weight_decay: 0.0500 (0.0500)  time: 0.7737  data: 0.0603  max mem: 41794
Epoch: [154]  [120/312]  eta: 0:02:32  lr: 0.002126  min_lr: 0.002126  loss: 3.6125 (3.3699)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.0529  max mem: 41794
Epoch: [154]  [130/312]  eta: 0:02:25  lr: 0.002126  min_lr: 0.002126  loss: 3.5101 (3.3561)  weight_decay: 0.0500 (0.0500)  time: 0.7586  data: 0.0588  max mem: 41794
Epoch: [154]  [140/312]  eta: 0:02:17  lr: 0.002125  min_lr: 0.002125  loss: 3.5101 (3.3720)  weight_decay: 0.0500 (0.0500)  time: 0.8228  data: 0.0476  max mem: 41794
Epoch: [154]  [150/312]  eta: 0:02:07  lr: 0.002124  min_lr: 0.002124  loss: 3.5618 (3.3536)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.0298  max mem: 41794
Epoch: [154]  [160/312]  eta: 0:02:00  lr: 0.002124  min_lr: 0.002124  loss: 3.3871 (3.3633)  weight_decay: 0.0500 (0.0500)  time: 0.7343  data: 0.0254  max mem: 41794
Epoch: [154]  [170/312]  eta: 0:01:50  lr: 0.002123  min_lr: 0.002123  loss: 3.5114 (3.3775)  weight_decay: 0.0500 (0.0500)  time: 0.7150  data: 0.0180  max mem: 41794
Epoch: [154]  [180/312]  eta: 0:01:42  lr: 0.002122  min_lr: 0.002122  loss: 3.7053 (3.3856)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.0190  max mem: 41794
Epoch: [154]  [190/312]  eta: 0:01:35  lr: 0.002121  min_lr: 0.002121  loss: 3.7136 (3.4058)  weight_decay: 0.0500 (0.0500)  time: 0.8030  data: 0.0436  max mem: 41794
Epoch: [154]  [200/312]  eta: 0:01:26  lr: 0.002121  min_lr: 0.002121  loss: 3.7187 (3.4147)  weight_decay: 0.0500 (0.0500)  time: 0.6915  data: 0.0366  max mem: 41794
Epoch: [154]  [210/312]  eta: 0:01:19  lr: 0.002120  min_lr: 0.002120  loss: 3.5398 (3.4093)  weight_decay: 0.0500 (0.0500)  time: 0.7307  data: 0.0421  max mem: 41794
Epoch: [154]  [220/312]  eta: 0:01:11  lr: 0.002119  min_lr: 0.002119  loss: 3.5130 (3.4075)  weight_decay: 0.0500 (0.0500)  time: 0.8190  data: 0.0399  max mem: 41794
Epoch: [154]  [230/312]  eta: 0:01:02  lr: 0.002118  min_lr: 0.002118  loss: 3.4004 (3.4084)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0154  max mem: 41794
Epoch: [154]  [240/312]  eta: 0:00:55  lr: 0.002118  min_lr: 0.002118  loss: 3.7107 (3.4258)  weight_decay: 0.0500 (0.0500)  time: 0.7295  data: 0.0424  max mem: 41794
Epoch: [154]  [250/312]  eta: 0:00:47  lr: 0.002117  min_lr: 0.002117  loss: 3.6602 (3.4209)  weight_decay: 0.0500 (0.0500)  time: 0.7131  data: 0.0709  max mem: 41794
Epoch: [154]  [260/312]  eta: 0:00:39  lr: 0.002116  min_lr: 0.002116  loss: 3.1388 (3.4122)  weight_decay: 0.0500 (0.0500)  time: 0.6833  data: 0.0447  max mem: 41794
Epoch: [154]  [270/312]  eta: 0:00:32  lr: 0.002116  min_lr: 0.002116  loss: 3.1273 (3.4037)  weight_decay: 0.0500 (0.0500)  time: 0.8229  data: 0.0334  max mem: 41794
Epoch: [154]  [280/312]  eta: 0:00:24  lr: 0.002115  min_lr: 0.002115  loss: 3.3968 (3.4024)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.0478  max mem: 41794
Epoch: [154]  [290/312]  eta: 0:00:16  lr: 0.002114  min_lr: 0.002114  loss: 3.5848 (3.4129)  weight_decay: 0.0500 (0.0500)  time: 0.6803  data: 0.0221  max mem: 41794
Epoch: [154]  [300/312]  eta: 0:00:09  lr: 0.002113  min_lr: 0.002113  loss: 3.6487 (3.4111)  weight_decay: 0.0500 (0.0500)  time: 0.7071  data: 0.0211  max mem: 41794
Epoch: [154]  [310/312]  eta: 0:00:01  lr: 0.002113  min_lr: 0.002113  loss: 3.5702 (3.4109)  weight_decay: 0.0500 (0.0500)  time: 0.5261  data: 0.0210  max mem: 41794
Epoch: [154]  [311/312]  eta: 0:00:00  lr: 0.002113  min_lr: 0.002113  loss: 3.5972 (3.4122)  weight_decay: 0.0500 (0.0500)  time: 0.5250  data: 0.0210  max mem: 41794
Epoch: [154] Total time: 0:03:54 (0.7515 s / it)
Averaged stats: lr: 0.002113  min_lr: 0.002113  loss: 3.5972 (3.4132)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.2599 (1.2599)  acc1: 79.6875 (79.6875)  acc5: 94.6615 (94.6615)  time: 8.5931  data: 8.3860  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.6188 (1.5307)  acc1: 72.6562 (72.0000)  acc5: 90.8854 (91.0080)  time: 1.1038  data: 0.9318  max mem: 41794
Test: Total time: 0:00:10 (1.1144 s / it)
* Acc@1 72.382 Acc@5 91.434 loss 1.518
Accuracy of the model on the 50000 test images: 72.4%
Max accuracy: 73.11%
Epoch: [155]  [  0/312]  eta: 1:21:42  lr: 0.002113  min_lr: 0.002113  loss: 3.6000 (3.6000)  weight_decay: 0.0500 (0.0500)  time: 15.7126  data: 13.0126  max mem: 41794
Epoch: [155]  [ 10/312]  eta: 0:10:09  lr: 0.002112  min_lr: 0.002112  loss: 3.6900 (3.5619)  weight_decay: 0.0500 (0.0500)  time: 2.0173  data: 1.3374  max mem: 41794
Epoch: [155]  [ 20/312]  eta: 0:07:13  lr: 0.002111  min_lr: 0.002111  loss: 3.6597 (3.5487)  weight_decay: 0.0500 (0.0500)  time: 0.7738  data: 0.1937  max mem: 41794
Epoch: [155]  [ 30/312]  eta: 0:05:30  lr: 0.002110  min_lr: 0.002110  loss: 3.5325 (3.5250)  weight_decay: 0.0500 (0.0500)  time: 0.7068  data: 0.1127  max mem: 41794
Epoch: [155]  [ 40/312]  eta: 0:05:00  lr: 0.002110  min_lr: 0.002110  loss: 3.7263 (3.5883)  weight_decay: 0.0500 (0.0500)  time: 0.7038  data: 0.0866  max mem: 41794
Epoch: [155]  [ 50/312]  eta: 0:04:29  lr: 0.002109  min_lr: 0.002109  loss: 3.7205 (3.5294)  weight_decay: 0.0500 (0.0500)  time: 0.8106  data: 0.1438  max mem: 41794
Epoch: [155]  [ 60/312]  eta: 0:03:56  lr: 0.002108  min_lr: 0.002108  loss: 3.5545 (3.5266)  weight_decay: 0.0500 (0.0500)  time: 0.6034  data: 0.0615  max mem: 41794
Epoch: [155]  [ 70/312]  eta: 0:03:43  lr: 0.002108  min_lr: 0.002108  loss: 3.5545 (3.4902)  weight_decay: 0.0500 (0.0500)  time: 0.6570  data: 0.1382  max mem: 41794
Epoch: [155]  [ 80/312]  eta: 0:03:30  lr: 0.002107  min_lr: 0.002107  loss: 3.6751 (3.5341)  weight_decay: 0.0500 (0.0500)  time: 0.8087  data: 0.2439  max mem: 41794
Epoch: [155]  [ 90/312]  eta: 0:03:12  lr: 0.002106  min_lr: 0.002106  loss: 3.6749 (3.5308)  weight_decay: 0.0500 (0.0500)  time: 0.6628  data: 0.1319  max mem: 41794
Epoch: [155]  [100/312]  eta: 0:03:06  lr: 0.002105  min_lr: 0.002105  loss: 3.5469 (3.5155)  weight_decay: 0.0500 (0.0500)  time: 0.7753  data: 0.1829  max mem: 41794
Epoch: [155]  [110/312]  eta: 0:02:51  lr: 0.002105  min_lr: 0.002105  loss: 3.3727 (3.4848)  weight_decay: 0.0500 (0.0500)  time: 0.7554  data: 0.1574  max mem: 41794
Epoch: [155]  [120/312]  eta: 0:02:43  lr: 0.002104  min_lr: 0.002104  loss: 3.6145 (3.4986)  weight_decay: 0.0500 (0.0500)  time: 0.6997  data: 0.1174  max mem: 41794
Epoch: [155]  [130/312]  eta: 0:02:32  lr: 0.002103  min_lr: 0.002103  loss: 3.6216 (3.4953)  weight_decay: 0.0500 (0.0500)  time: 0.8016  data: 0.2049  max mem: 41794
Epoch: [155]  [140/312]  eta: 0:02:21  lr: 0.002103  min_lr: 0.002103  loss: 3.3144 (3.4896)  weight_decay: 0.0500 (0.0500)  time: 0.6668  data: 0.0897  max mem: 41794
Epoch: [155]  [150/312]  eta: 0:02:14  lr: 0.002102  min_lr: 0.002102  loss: 3.4550 (3.4930)  weight_decay: 0.0500 (0.0500)  time: 0.7810  data: 0.0802  max mem: 41794
Epoch: [155]  [160/312]  eta: 0:02:04  lr: 0.002101  min_lr: 0.002101  loss: 3.4348 (3.4824)  weight_decay: 0.0500 (0.0500)  time: 0.7818  data: 0.0860  max mem: 41794
Epoch: [155]  [170/312]  eta: 0:01:56  lr: 0.002100  min_lr: 0.002100  loss: 3.3262 (3.4745)  weight_decay: 0.0500 (0.0500)  time: 0.7042  data: 0.0423  max mem: 41794
Epoch: [155]  [180/312]  eta: 0:01:48  lr: 0.002100  min_lr: 0.002100  loss: 3.3262 (3.4642)  weight_decay: 0.0500 (0.0500)  time: 0.8141  data: 0.0600  max mem: 41794
Epoch: [155]  [190/312]  eta: 0:01:38  lr: 0.002099  min_lr: 0.002099  loss: 3.2309 (3.4507)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.0277  max mem: 41794
Epoch: [155]  [200/312]  eta: 0:01:30  lr: 0.002098  min_lr: 0.002098  loss: 3.6423 (3.4651)  weight_decay: 0.0500 (0.0500)  time: 0.7220  data: 0.0316  max mem: 41794
Epoch: [155]  [210/312]  eta: 0:01:22  lr: 0.002098  min_lr: 0.002098  loss: 3.6789 (3.4586)  weight_decay: 0.0500 (0.0500)  time: 0.8265  data: 0.0314  max mem: 41794
Epoch: [155]  [220/312]  eta: 0:01:13  lr: 0.002097  min_lr: 0.002097  loss: 3.4222 (3.4554)  weight_decay: 0.0500 (0.0500)  time: 0.6423  data: 0.0041  max mem: 41794
Epoch: [155]  [230/312]  eta: 0:01:05  lr: 0.002096  min_lr: 0.002096  loss: 3.4717 (3.4529)  weight_decay: 0.0500 (0.0500)  time: 0.7228  data: 0.0587  max mem: 41794
Epoch: [155]  [240/312]  eta: 0:00:57  lr: 0.002095  min_lr: 0.002095  loss: 3.5270 (3.4466)  weight_decay: 0.0500 (0.0500)  time: 0.7867  data: 0.0587  max mem: 41794
Epoch: [155]  [250/312]  eta: 0:00:48  lr: 0.002095  min_lr: 0.002095  loss: 3.5595 (3.4475)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.0446  max mem: 41794
Epoch: [155]  [260/312]  eta: 0:00:40  lr: 0.002094  min_lr: 0.002094  loss: 3.7364 (3.4521)  weight_decay: 0.0500 (0.0500)  time: 0.7005  data: 0.0465  max mem: 41794
Epoch: [155]  [270/312]  eta: 0:00:32  lr: 0.002093  min_lr: 0.002093  loss: 3.7364 (3.4577)  weight_decay: 0.0500 (0.0500)  time: 0.6228  data: 0.0263  max mem: 41794
Epoch: [155]  [280/312]  eta: 0:00:24  lr: 0.002093  min_lr: 0.002093  loss: 3.4224 (3.4508)  weight_decay: 0.0500 (0.0500)  time: 0.6646  data: 0.0678  max mem: 41794
Epoch: [155]  [290/312]  eta: 0:00:17  lr: 0.002092  min_lr: 0.002092  loss: 3.3569 (3.4544)  weight_decay: 0.0500 (0.0500)  time: 0.8271  data: 0.0994  max mem: 41794
Epoch: [155]  [300/312]  eta: 0:00:09  lr: 0.002091  min_lr: 0.002091  loss: 3.4743 (3.4467)  weight_decay: 0.0500 (0.0500)  time: 0.6622  data: 0.0543  max mem: 41794
Epoch: [155]  [310/312]  eta: 0:00:01  lr: 0.002090  min_lr: 0.002090  loss: 3.4743 (3.4382)  weight_decay: 0.0500 (0.0500)  time: 0.4627  data: 0.0001  max mem: 41794
Epoch: [155]  [311/312]  eta: 0:00:00  lr: 0.002090  min_lr: 0.002090  loss: 3.1100 (3.4369)  weight_decay: 0.0500 (0.0500)  time: 0.4631  data: 0.0001  max mem: 41794
Epoch: [155] Total time: 0:03:59 (0.7664 s / it)
Averaged stats: lr: 0.002090  min_lr: 0.002090  loss: 3.1100 (3.4261)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0991 (1.0991)  acc1: 79.1667 (79.1667)  acc5: 95.0521 (95.0521)  time: 8.7833  data: 8.5711  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4918 (1.3696)  acc1: 72.0052 (72.6240)  acc5: 90.6250 (91.2480)  time: 1.1240  data: 0.9524  max mem: 41794
Test: Total time: 0:00:10 (1.1362 s / it)
* Acc@1 72.394 Acc@5 91.316 loss 1.368
Accuracy of the model on the 50000 test images: 72.4%
Max accuracy: 73.11%
Epoch: [156]  [  0/312]  eta: 1:25:05  lr: 0.002090  min_lr: 0.002090  loss: 3.7318 (3.7318)  weight_decay: 0.0500 (0.0500)  time: 16.3642  data: 12.0799  max mem: 41794
Epoch: [156]  [ 10/312]  eta: 0:11:04  lr: 0.002089  min_lr: 0.002089  loss: 3.5723 (3.6962)  weight_decay: 0.0500 (0.0500)  time: 2.1994  data: 1.2955  max mem: 41794
Epoch: [156]  [ 20/312]  eta: 0:07:35  lr: 0.002089  min_lr: 0.002089  loss: 3.5460 (3.5072)  weight_decay: 0.0500 (0.0500)  time: 0.8192  data: 0.1824  max mem: 41794
Epoch: [156]  [ 30/312]  eta: 0:05:45  lr: 0.002088  min_lr: 0.002088  loss: 3.1157 (3.4169)  weight_decay: 0.0500 (0.0500)  time: 0.6917  data: 0.0743  max mem: 41794
Epoch: [156]  [ 40/312]  eta: 0:04:50  lr: 0.002087  min_lr: 0.002087  loss: 3.0855 (3.3956)  weight_decay: 0.0500 (0.0500)  time: 0.5534  data: 0.0241  max mem: 41794
Epoch: [156]  [ 50/312]  eta: 0:04:24  lr: 0.002087  min_lr: 0.002087  loss: 3.3125 (3.3717)  weight_decay: 0.0500 (0.0500)  time: 0.6703  data: 0.0448  max mem: 41794
Epoch: [156]  [ 60/312]  eta: 0:03:58  lr: 0.002086  min_lr: 0.002086  loss: 3.3956 (3.3731)  weight_decay: 0.0500 (0.0500)  time: 0.7015  data: 0.0989  max mem: 41794
Epoch: [156]  [ 70/312]  eta: 0:03:45  lr: 0.002085  min_lr: 0.002085  loss: 3.4169 (3.3828)  weight_decay: 0.0500 (0.0500)  time: 0.7390  data: 0.1898  max mem: 41794
Epoch: [156]  [ 80/312]  eta: 0:03:29  lr: 0.002084  min_lr: 0.002084  loss: 3.6505 (3.4058)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.1526  max mem: 41794
Epoch: [156]  [ 90/312]  eta: 0:03:14  lr: 0.002084  min_lr: 0.002084  loss: 3.7683 (3.4355)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1337  max mem: 41794
Epoch: [156]  [100/312]  eta: 0:03:04  lr: 0.002083  min_lr: 0.002083  loss: 3.4885 (3.4196)  weight_decay: 0.0500 (0.0500)  time: 0.7340  data: 0.2264  max mem: 41794
Epoch: [156]  [110/312]  eta: 0:02:48  lr: 0.002082  min_lr: 0.002082  loss: 3.4606 (3.4268)  weight_decay: 0.0500 (0.0500)  time: 0.6398  data: 0.1339  max mem: 41794
Epoch: [156]  [120/312]  eta: 0:02:40  lr: 0.002082  min_lr: 0.002082  loss: 3.4472 (3.4083)  weight_decay: 0.0500 (0.0500)  time: 0.6822  data: 0.1911  max mem: 41794
Epoch: [156]  [130/312]  eta: 0:02:32  lr: 0.002081  min_lr: 0.002081  loss: 3.4961 (3.4191)  weight_decay: 0.0500 (0.0500)  time: 0.8736  data: 0.3808  max mem: 41794
Epoch: [156]  [140/312]  eta: 0:02:20  lr: 0.002080  min_lr: 0.002080  loss: 3.6785 (3.4339)  weight_decay: 0.0500 (0.0500)  time: 0.7006  data: 0.2107  max mem: 41794
Epoch: [156]  [150/312]  eta: 0:02:12  lr: 0.002079  min_lr: 0.002079  loss: 3.4961 (3.4277)  weight_decay: 0.0500 (0.0500)  time: 0.6809  data: 0.1907  max mem: 41794
Epoch: [156]  [160/312]  eta: 0:02:04  lr: 0.002079  min_lr: 0.002079  loss: 3.3382 (3.4131)  weight_decay: 0.0500 (0.0500)  time: 0.8198  data: 0.3332  max mem: 41794
Epoch: [156]  [170/312]  eta: 0:01:53  lr: 0.002078  min_lr: 0.002078  loss: 3.2908 (3.3980)  weight_decay: 0.0500 (0.0500)  time: 0.6684  data: 0.1827  max mem: 41794
Epoch: [156]  [180/312]  eta: 0:01:46  lr: 0.002077  min_lr: 0.002077  loss: 3.3131 (3.3956)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1990  max mem: 41794
Epoch: [156]  [190/312]  eta: 0:01:36  lr: 0.002077  min_lr: 0.002077  loss: 3.4451 (3.3962)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1799  max mem: 41794
Epoch: [156]  [200/312]  eta: 0:01:28  lr: 0.002076  min_lr: 0.002076  loss: 3.2474 (3.3939)  weight_decay: 0.0500 (0.0500)  time: 0.6619  data: 0.1695  max mem: 41794
Epoch: [156]  [210/312]  eta: 0:01:20  lr: 0.002075  min_lr: 0.002075  loss: 3.5708 (3.4079)  weight_decay: 0.0500 (0.0500)  time: 0.8101  data: 0.3227  max mem: 41794
Epoch: [156]  [220/312]  eta: 0:01:11  lr: 0.002074  min_lr: 0.002074  loss: 3.7234 (3.4201)  weight_decay: 0.0500 (0.0500)  time: 0.6660  data: 0.1812  max mem: 41794
Epoch: [156]  [230/312]  eta: 0:01:04  lr: 0.002074  min_lr: 0.002074  loss: 3.7642 (3.4235)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.2093  max mem: 41794
Epoch: [156]  [240/312]  eta: 0:00:56  lr: 0.002073  min_lr: 0.002073  loss: 3.7694 (3.4349)  weight_decay: 0.0500 (0.0500)  time: 0.8660  data: 0.3600  max mem: 41794
Epoch: [156]  [250/312]  eta: 0:00:48  lr: 0.002072  min_lr: 0.002072  loss: 3.7654 (3.4438)  weight_decay: 0.0500 (0.0500)  time: 0.6879  data: 0.1901  max mem: 41794
Epoch: [156]  [260/312]  eta: 0:00:40  lr: 0.002072  min_lr: 0.002072  loss: 3.4706 (3.4399)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.1938  max mem: 41794
Epoch: [156]  [270/312]  eta: 0:00:32  lr: 0.002071  min_lr: 0.002071  loss: 3.5904 (3.4463)  weight_decay: 0.0500 (0.0500)  time: 0.7124  data: 0.1824  max mem: 41794
Epoch: [156]  [280/312]  eta: 0:00:24  lr: 0.002070  min_lr: 0.002070  loss: 3.5608 (3.4341)  weight_decay: 0.0500 (0.0500)  time: 0.6684  data: 0.1458  max mem: 41794
Epoch: [156]  [290/312]  eta: 0:00:17  lr: 0.002069  min_lr: 0.002069  loss: 3.3683 (3.4382)  weight_decay: 0.0500 (0.0500)  time: 0.8257  data: 0.2664  max mem: 41794
Epoch: [156]  [300/312]  eta: 0:00:09  lr: 0.002069  min_lr: 0.002069  loss: 3.5918 (3.4393)  weight_decay: 0.0500 (0.0500)  time: 0.6585  data: 0.1378  max mem: 41794
Epoch: [156]  [310/312]  eta: 0:00:01  lr: 0.002068  min_lr: 0.002068  loss: 3.6407 (3.4452)  weight_decay: 0.0500 (0.0500)  time: 0.4820  data: 0.0166  max mem: 41794
Epoch: [156]  [311/312]  eta: 0:00:00  lr: 0.002068  min_lr: 0.002068  loss: 3.6484 (3.4478)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [156] Total time: 0:03:57 (0.7608 s / it)
Averaged stats: lr: 0.002068  min_lr: 0.002068  loss: 3.6484 (3.4079)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.0914 (1.0914)  acc1: 78.7760 (78.7760)  acc5: 94.1406 (94.1406)  time: 8.9478  data: 8.7357  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3669 (1.2750)  acc1: 71.6146 (72.5920)  acc5: 91.1458 (91.2160)  time: 1.1422  data: 0.9707  max mem: 41794
Test: Total time: 0:00:10 (1.1737 s / it)
* Acc@1 72.530 Acc@5 91.464 loss 1.266
Accuracy of the model on the 50000 test images: 72.5%
Max accuracy: 73.11%
Epoch: [157]  [  0/312]  eta: 1:21:13  lr: 0.002068  min_lr: 0.002068  loss: 3.6957 (3.6957)  weight_decay: 0.0500 (0.0500)  time: 15.6197  data: 11.8287  max mem: 41794
Epoch: [157]  [ 10/312]  eta: 0:11:33  lr: 0.002067  min_lr: 0.002067  loss: 3.4411 (3.2885)  weight_decay: 0.0500 (0.0500)  time: 2.2972  data: 1.3737  max mem: 41794
Epoch: [157]  [ 20/312]  eta: 0:07:38  lr: 0.002066  min_lr: 0.002066  loss: 3.3224 (3.2877)  weight_decay: 0.0500 (0.0500)  time: 0.8664  data: 0.2058  max mem: 41794
Epoch: [157]  [ 30/312]  eta: 0:05:47  lr: 0.002066  min_lr: 0.002066  loss: 3.5841 (3.4267)  weight_decay: 0.0500 (0.0500)  time: 0.6481  data: 0.0420  max mem: 41794
Epoch: [157]  [ 40/312]  eta: 0:04:47  lr: 0.002065  min_lr: 0.002065  loss: 3.6488 (3.4169)  weight_decay: 0.0500 (0.0500)  time: 0.5190  data: 0.0007  max mem: 41794
Epoch: [157]  [ 50/312]  eta: 0:04:20  lr: 0.002064  min_lr: 0.002064  loss: 3.3729 (3.3850)  weight_decay: 0.0500 (0.0500)  time: 0.6198  data: 0.0598  max mem: 41794
Epoch: [157]  [ 60/312]  eta: 0:03:53  lr: 0.002063  min_lr: 0.002063  loss: 3.3708 (3.3713)  weight_decay: 0.0500 (0.0500)  time: 0.6552  data: 0.1015  max mem: 41794
Epoch: [157]  [ 70/312]  eta: 0:03:43  lr: 0.002063  min_lr: 0.002063  loss: 3.4551 (3.3933)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.1889  max mem: 41794
Epoch: [157]  [ 80/312]  eta: 0:03:26  lr: 0.002062  min_lr: 0.002062  loss: 3.5983 (3.3966)  weight_decay: 0.0500 (0.0500)  time: 0.7823  data: 0.1644  max mem: 41794
Epoch: [157]  [ 90/312]  eta: 0:03:12  lr: 0.002061  min_lr: 0.002061  loss: 3.4011 (3.3746)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1075  max mem: 41794
Epoch: [157]  [100/312]  eta: 0:03:02  lr: 0.002061  min_lr: 0.002061  loss: 3.3676 (3.3807)  weight_decay: 0.0500 (0.0500)  time: 0.7488  data: 0.2049  max mem: 41794
Epoch: [157]  [110/312]  eta: 0:02:47  lr: 0.002060  min_lr: 0.002060  loss: 3.6094 (3.4046)  weight_decay: 0.0500 (0.0500)  time: 0.6603  data: 0.1186  max mem: 41794
Epoch: [157]  [120/312]  eta: 0:02:39  lr: 0.002059  min_lr: 0.002059  loss: 3.6326 (3.4051)  weight_decay: 0.0500 (0.0500)  time: 0.6678  data: 0.1429  max mem: 41794
Epoch: [157]  [130/312]  eta: 0:02:30  lr: 0.002058  min_lr: 0.002058  loss: 3.6197 (3.4256)  weight_decay: 0.0500 (0.0500)  time: 0.8148  data: 0.2772  max mem: 41794
Epoch: [157]  [140/312]  eta: 0:02:18  lr: 0.002058  min_lr: 0.002058  loss: 3.5766 (3.4081)  weight_decay: 0.0500 (0.0500)  time: 0.6383  data: 0.1382  max mem: 41794
Epoch: [157]  [150/312]  eta: 0:02:11  lr: 0.002057  min_lr: 0.002057  loss: 3.3068 (3.4028)  weight_decay: 0.0500 (0.0500)  time: 0.7049  data: 0.2126  max mem: 41794
Epoch: [157]  [160/312]  eta: 0:02:02  lr: 0.002056  min_lr: 0.002056  loss: 3.4310 (3.3998)  weight_decay: 0.0500 (0.0500)  time: 0.8103  data: 0.2878  max mem: 41794
Epoch: [157]  [170/312]  eta: 0:01:52  lr: 0.002056  min_lr: 0.002056  loss: 3.5631 (3.3958)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.1575  max mem: 41794
Epoch: [157]  [180/312]  eta: 0:01:45  lr: 0.002055  min_lr: 0.002055  loss: 3.6060 (3.4040)  weight_decay: 0.0500 (0.0500)  time: 0.7368  data: 0.1999  max mem: 41794
Epoch: [157]  [190/312]  eta: 0:01:35  lr: 0.002054  min_lr: 0.002054  loss: 3.5777 (3.4084)  weight_decay: 0.0500 (0.0500)  time: 0.6587  data: 0.1183  max mem: 41794
Epoch: [157]  [200/312]  eta: 0:01:28  lr: 0.002053  min_lr: 0.002053  loss: 3.5787 (3.4128)  weight_decay: 0.0500 (0.0500)  time: 0.7135  data: 0.1690  max mem: 41794
Epoch: [157]  [210/312]  eta: 0:01:20  lr: 0.002053  min_lr: 0.002053  loss: 3.6441 (3.4230)  weight_decay: 0.0500 (0.0500)  time: 0.8857  data: 0.3283  max mem: 41794
Epoch: [157]  [220/312]  eta: 0:01:11  lr: 0.002052  min_lr: 0.002052  loss: 3.5828 (3.4201)  weight_decay: 0.0500 (0.0500)  time: 0.6657  data: 0.1617  max mem: 41794
Epoch: [157]  [230/312]  eta: 0:01:04  lr: 0.002051  min_lr: 0.002051  loss: 3.6175 (3.4285)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.1781  max mem: 41794
Epoch: [157]  [240/312]  eta: 0:00:56  lr: 0.002051  min_lr: 0.002051  loss: 3.6631 (3.4337)  weight_decay: 0.0500 (0.0500)  time: 0.8100  data: 0.2809  max mem: 41794
Epoch: [157]  [250/312]  eta: 0:00:47  lr: 0.002050  min_lr: 0.002050  loss: 3.5239 (3.4284)  weight_decay: 0.0500 (0.0500)  time: 0.6573  data: 0.1523  max mem: 41794
Epoch: [157]  [260/312]  eta: 0:00:40  lr: 0.002049  min_lr: 0.002049  loss: 3.3023 (3.4189)  weight_decay: 0.0500 (0.0500)  time: 0.7361  data: 0.2448  max mem: 41794
Epoch: [157]  [270/312]  eta: 0:00:32  lr: 0.002048  min_lr: 0.002048  loss: 3.3370 (3.4166)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.1977  max mem: 41794
Epoch: [157]  [280/312]  eta: 0:00:24  lr: 0.002048  min_lr: 0.002048  loss: 3.5042 (3.4225)  weight_decay: 0.0500 (0.0500)  time: 0.6741  data: 0.1813  max mem: 41794
Epoch: [157]  [290/312]  eta: 0:00:16  lr: 0.002047  min_lr: 0.002047  loss: 3.5838 (3.4294)  weight_decay: 0.0500 (0.0500)  time: 0.8523  data: 0.3633  max mem: 41794
Epoch: [157]  [300/312]  eta: 0:00:09  lr: 0.002046  min_lr: 0.002046  loss: 3.5838 (3.4230)  weight_decay: 0.0500 (0.0500)  time: 0.6665  data: 0.1824  max mem: 41794
Epoch: [157]  [310/312]  eta: 0:00:01  lr: 0.002046  min_lr: 0.002046  loss: 3.5865 (3.4241)  weight_decay: 0.0500 (0.0500)  time: 0.4687  data: 0.0001  max mem: 41794
Epoch: [157]  [311/312]  eta: 0:00:00  lr: 0.002045  min_lr: 0.002045  loss: 3.5127 (3.4237)  weight_decay: 0.0500 (0.0500)  time: 0.4677  data: 0.0001  max mem: 41794
Epoch: [157] Total time: 0:03:56 (0.7583 s / it)
Averaged stats: lr: 0.002045  min_lr: 0.002045  loss: 3.5127 (3.4158)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.1618 (1.1618)  acc1: 79.1667 (79.1667)  acc5: 94.6615 (94.6615)  time: 8.1636  data: 7.9504  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4332 (1.3515)  acc1: 72.0052 (72.9600)  acc5: 91.9271 (92.1280)  time: 1.0550  data: 0.8835  max mem: 41794
Test: Total time: 0:00:09 (1.0675 s / it)
* Acc@1 73.258 Acc@5 92.062 loss 1.341
Accuracy of the model on the 50000 test images: 73.3%
Max accuracy: 73.26%
Epoch: [158]  [  0/312]  eta: 1:23:49  lr: 0.002045  min_lr: 0.002045  loss: 3.9584 (3.9584)  weight_decay: 0.0500 (0.0500)  time: 16.1206  data: 15.6310  max mem: 41794
Epoch: [158]  [ 10/312]  eta: 0:11:07  lr: 0.002045  min_lr: 0.002045  loss: 3.6146 (3.6320)  weight_decay: 0.0500 (0.0500)  time: 2.2114  data: 1.4217  max mem: 41794
Epoch: [158]  [ 20/312]  eta: 0:07:20  lr: 0.002044  min_lr: 0.002044  loss: 3.3149 (3.3751)  weight_decay: 0.0500 (0.0500)  time: 0.7763  data: 0.1052  max mem: 41794
Epoch: [158]  [ 30/312]  eta: 0:05:37  lr: 0.002043  min_lr: 0.002043  loss: 3.1142 (3.3335)  weight_decay: 0.0500 (0.0500)  time: 0.6376  data: 0.1114  max mem: 41794
Epoch: [158]  [ 40/312]  eta: 0:04:42  lr: 0.002042  min_lr: 0.002042  loss: 3.4437 (3.3611)  weight_decay: 0.0500 (0.0500)  time: 0.5465  data: 0.0289  max mem: 41794
Epoch: [158]  [ 50/312]  eta: 0:04:18  lr: 0.002042  min_lr: 0.002042  loss: 3.5365 (3.4030)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.1503  max mem: 41794
Epoch: [158]  [ 60/312]  eta: 0:03:48  lr: 0.002041  min_lr: 0.002041  loss: 3.6228 (3.4291)  weight_decay: 0.0500 (0.0500)  time: 0.6348  data: 0.1283  max mem: 41794
Epoch: [158]  [ 70/312]  eta: 0:03:37  lr: 0.002040  min_lr: 0.002040  loss: 3.6228 (3.4407)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.1553  max mem: 41794
Epoch: [158]  [ 80/312]  eta: 0:03:27  lr: 0.002040  min_lr: 0.002040  loss: 3.6360 (3.4582)  weight_decay: 0.0500 (0.0500)  time: 0.8519  data: 0.3196  max mem: 41794
Epoch: [158]  [ 90/312]  eta: 0:03:08  lr: 0.002039  min_lr: 0.002039  loss: 3.6360 (3.4701)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.1650  max mem: 41794
Epoch: [158]  [100/312]  eta: 0:03:00  lr: 0.002038  min_lr: 0.002038  loss: 3.6266 (3.4969)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1637  max mem: 41794
Epoch: [158]  [110/312]  eta: 0:02:45  lr: 0.002037  min_lr: 0.002037  loss: 3.6080 (3.4917)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.1637  max mem: 41794
Epoch: [158]  [120/312]  eta: 0:02:38  lr: 0.002037  min_lr: 0.002037  loss: 3.5072 (3.4784)  weight_decay: 0.0500 (0.0500)  time: 0.6845  data: 0.1465  max mem: 41794
Epoch: [158]  [130/312]  eta: 0:02:30  lr: 0.002036  min_lr: 0.002036  loss: 3.5387 (3.4693)  weight_decay: 0.0500 (0.0500)  time: 0.8697  data: 0.3028  max mem: 41794
Epoch: [158]  [140/312]  eta: 0:02:18  lr: 0.002035  min_lr: 0.002035  loss: 3.3441 (3.4609)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.1570  max mem: 41794
Epoch: [158]  [150/312]  eta: 0:02:11  lr: 0.002035  min_lr: 0.002035  loss: 3.3854 (3.4571)  weight_decay: 0.0500 (0.0500)  time: 0.6844  data: 0.1358  max mem: 41794
Epoch: [158]  [160/312]  eta: 0:02:03  lr: 0.002034  min_lr: 0.002034  loss: 3.6299 (3.4676)  weight_decay: 0.0500 (0.0500)  time: 0.8517  data: 0.2578  max mem: 41794
Epoch: [158]  [170/312]  eta: 0:01:53  lr: 0.002033  min_lr: 0.002033  loss: 3.6299 (3.4658)  weight_decay: 0.0500 (0.0500)  time: 0.6984  data: 0.1587  max mem: 41794
Epoch: [158]  [180/312]  eta: 0:01:45  lr: 0.002032  min_lr: 0.002032  loss: 3.4906 (3.4552)  weight_decay: 0.0500 (0.0500)  time: 0.6743  data: 0.1684  max mem: 41794
Epoch: [158]  [190/312]  eta: 0:01:35  lr: 0.002032  min_lr: 0.002032  loss: 3.5202 (3.4421)  weight_decay: 0.0500 (0.0500)  time: 0.6366  data: 0.1323  max mem: 41794
Epoch: [158]  [200/312]  eta: 0:01:28  lr: 0.002031  min_lr: 0.002031  loss: 3.2997 (3.4347)  weight_decay: 0.0500 (0.0500)  time: 0.6982  data: 0.2071  max mem: 41794
Epoch: [158]  [210/312]  eta: 0:01:20  lr: 0.002030  min_lr: 0.002030  loss: 3.4171 (3.4340)  weight_decay: 0.0500 (0.0500)  time: 0.8618  data: 0.3671  max mem: 41794
Epoch: [158]  [220/312]  eta: 0:01:11  lr: 0.002030  min_lr: 0.002030  loss: 3.6551 (3.4435)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.1610  max mem: 41794
Epoch: [158]  [230/312]  eta: 0:01:04  lr: 0.002029  min_lr: 0.002029  loss: 3.7075 (3.4365)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.2083  max mem: 41794
Epoch: [158]  [240/312]  eta: 0:00:56  lr: 0.002028  min_lr: 0.002028  loss: 3.0677 (3.4265)  weight_decay: 0.0500 (0.0500)  time: 0.8511  data: 0.3503  max mem: 41794
Epoch: [158]  [250/312]  eta: 0:00:47  lr: 0.002027  min_lr: 0.002027  loss: 3.5385 (3.4269)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.1728  max mem: 41794
Epoch: [158]  [260/312]  eta: 0:00:40  lr: 0.002027  min_lr: 0.002027  loss: 3.6090 (3.4231)  weight_decay: 0.0500 (0.0500)  time: 0.7425  data: 0.2099  max mem: 41794
Epoch: [158]  [270/312]  eta: 0:00:32  lr: 0.002026  min_lr: 0.002026  loss: 3.4419 (3.4160)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.1802  max mem: 41794
Epoch: [158]  [280/312]  eta: 0:00:24  lr: 0.002025  min_lr: 0.002025  loss: 3.4867 (3.4169)  weight_decay: 0.0500 (0.0500)  time: 0.6708  data: 0.1426  max mem: 41794
Epoch: [158]  [290/312]  eta: 0:00:17  lr: 0.002025  min_lr: 0.002025  loss: 3.4535 (3.4126)  weight_decay: 0.0500 (0.0500)  time: 0.8490  data: 0.3224  max mem: 41794
Epoch: [158]  [300/312]  eta: 0:00:09  lr: 0.002024  min_lr: 0.002024  loss: 3.4023 (3.4170)  weight_decay: 0.0500 (0.0500)  time: 0.6635  data: 0.1802  max mem: 41794
Epoch: [158]  [310/312]  eta: 0:00:01  lr: 0.002023  min_lr: 0.002023  loss: 3.2523 (3.4092)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [158]  [311/312]  eta: 0:00:00  lr: 0.002023  min_lr: 0.002023  loss: 3.2523 (3.4101)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [158] Total time: 0:03:56 (0.7592 s / it)
Averaged stats: lr: 0.002023  min_lr: 0.002023  loss: 3.2523 (3.3895)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.9961 (0.9961)  acc1: 79.0365 (79.0365)  acc5: 95.0521 (95.0521)  time: 8.3181  data: 8.1058  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3172 (1.2481)  acc1: 73.1771 (73.5040)  acc5: 91.5365 (92.1120)  time: 1.0725  data: 0.9008  max mem: 41794
Test: Total time: 0:00:09 (1.0826 s / it)
* Acc@1 73.208 Acc@5 91.844 loss 1.257
Accuracy of the model on the 50000 test images: 73.2%
Max accuracy: 73.26%
Epoch: [159]  [  0/312]  eta: 1:26:00  lr: 0.002023  min_lr: 0.002023  loss: 4.1113 (4.1113)  weight_decay: 0.0500 (0.0500)  time: 16.5404  data: 11.3514  max mem: 41794
Epoch: [159]  [ 10/312]  eta: 0:11:23  lr: 0.002022  min_lr: 0.002022  loss: 3.4035 (3.4023)  weight_decay: 0.0500 (0.0500)  time: 2.2628  data: 1.3061  max mem: 41794
Epoch: [159]  [ 20/312]  eta: 0:07:35  lr: 0.002021  min_lr: 0.002021  loss: 3.3875 (3.3737)  weight_decay: 0.0500 (0.0500)  time: 0.8119  data: 0.2005  max mem: 41794
Epoch: [159]  [ 30/312]  eta: 0:05:46  lr: 0.002021  min_lr: 0.002021  loss: 3.3085 (3.3230)  weight_decay: 0.0500 (0.0500)  time: 0.6610  data: 0.0522  max mem: 41794
Epoch: [159]  [ 40/312]  eta: 0:04:50  lr: 0.002020  min_lr: 0.002020  loss: 3.6300 (3.3634)  weight_decay: 0.0500 (0.0500)  time: 0.5511  data: 0.0028  max mem: 41794
Epoch: [159]  [ 50/312]  eta: 0:04:28  lr: 0.002019  min_lr: 0.002019  loss: 3.7159 (3.3664)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.0081  max mem: 41794
Epoch: [159]  [ 60/312]  eta: 0:03:59  lr: 0.002019  min_lr: 0.002019  loss: 3.5098 (3.3432)  weight_decay: 0.0500 (0.0500)  time: 0.7115  data: 0.0082  max mem: 41794
Epoch: [159]  [ 70/312]  eta: 0:03:45  lr: 0.002018  min_lr: 0.002018  loss: 3.5038 (3.3544)  weight_decay: 0.0500 (0.0500)  time: 0.6915  data: 0.0375  max mem: 41794
Epoch: [159]  [ 80/312]  eta: 0:03:32  lr: 0.002017  min_lr: 0.002017  loss: 3.5038 (3.3844)  weight_decay: 0.0500 (0.0500)  time: 0.8154  data: 0.0434  max mem: 41794
Epoch: [159]  [ 90/312]  eta: 0:03:16  lr: 0.002016  min_lr: 0.002016  loss: 3.4900 (3.3802)  weight_decay: 0.0500 (0.0500)  time: 0.7213  data: 0.0411  max mem: 41794
Epoch: [159]  [100/312]  eta: 0:03:06  lr: 0.002016  min_lr: 0.002016  loss: 3.4201 (3.3624)  weight_decay: 0.0500 (0.0500)  time: 0.7143  data: 0.0563  max mem: 41794
Epoch: [159]  [110/312]  eta: 0:02:50  lr: 0.002015  min_lr: 0.002015  loss: 3.0408 (3.3492)  weight_decay: 0.0500 (0.0500)  time: 0.6539  data: 0.0218  max mem: 41794
Epoch: [159]  [120/312]  eta: 0:02:41  lr: 0.002014  min_lr: 0.002014  loss: 3.1162 (3.3478)  weight_decay: 0.0500 (0.0500)  time: 0.6680  data: 0.0401  max mem: 41794
Epoch: [159]  [130/312]  eta: 0:02:34  lr: 0.002014  min_lr: 0.002014  loss: 3.5445 (3.3714)  weight_decay: 0.0500 (0.0500)  time: 0.8774  data: 0.0798  max mem: 41794
Epoch: [159]  [140/312]  eta: 0:02:21  lr: 0.002013  min_lr: 0.002013  loss: 3.5764 (3.3695)  weight_decay: 0.0500 (0.0500)  time: 0.7202  data: 0.0499  max mem: 41794
Epoch: [159]  [150/312]  eta: 0:02:13  lr: 0.002012  min_lr: 0.002012  loss: 3.1787 (3.3573)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0261  max mem: 41794
Epoch: [159]  [160/312]  eta: 0:02:04  lr: 0.002011  min_lr: 0.002011  loss: 3.3793 (3.3723)  weight_decay: 0.0500 (0.0500)  time: 0.7921  data: 0.0166  max mem: 41794
Epoch: [159]  [170/312]  eta: 0:01:54  lr: 0.002011  min_lr: 0.002011  loss: 3.6759 (3.3603)  weight_decay: 0.0500 (0.0500)  time: 0.6718  data: 0.0411  max mem: 41794
Epoch: [159]  [180/312]  eta: 0:01:46  lr: 0.002010  min_lr: 0.002010  loss: 3.5089 (3.3680)  weight_decay: 0.0500 (0.0500)  time: 0.7137  data: 0.0957  max mem: 41794
Epoch: [159]  [190/312]  eta: 0:01:36  lr: 0.002009  min_lr: 0.002009  loss: 3.5349 (3.3692)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.0553  max mem: 41794
Epoch: [159]  [200/312]  eta: 0:01:29  lr: 0.002009  min_lr: 0.002009  loss: 3.5349 (3.3723)  weight_decay: 0.0500 (0.0500)  time: 0.7040  data: 0.0560  max mem: 41794
Epoch: [159]  [210/312]  eta: 0:01:21  lr: 0.002008  min_lr: 0.002008  loss: 3.6065 (3.3835)  weight_decay: 0.0500 (0.0500)  time: 0.8265  data: 0.0687  max mem: 41794
Epoch: [159]  [220/312]  eta: 0:01:12  lr: 0.002007  min_lr: 0.002007  loss: 3.6480 (3.3892)  weight_decay: 0.0500 (0.0500)  time: 0.6509  data: 0.0433  max mem: 41794
Epoch: [159]  [230/312]  eta: 0:01:04  lr: 0.002006  min_lr: 0.002006  loss: 3.5437 (3.3886)  weight_decay: 0.0500 (0.0500)  time: 0.7186  data: 0.0825  max mem: 41794
Epoch: [159]  [240/312]  eta: 0:00:56  lr: 0.002006  min_lr: 0.002006  loss: 3.4830 (3.3890)  weight_decay: 0.0500 (0.0500)  time: 0.7764  data: 0.0706  max mem: 41794
Epoch: [159]  [250/312]  eta: 0:00:48  lr: 0.002005  min_lr: 0.002005  loss: 3.3339 (3.3797)  weight_decay: 0.0500 (0.0500)  time: 0.6386  data: 0.0482  max mem: 41794
Epoch: [159]  [260/312]  eta: 0:00:40  lr: 0.002004  min_lr: 0.002004  loss: 3.4082 (3.3882)  weight_decay: 0.0500 (0.0500)  time: 0.7340  data: 0.1091  max mem: 41794
Epoch: [159]  [270/312]  eta: 0:00:32  lr: 0.002004  min_lr: 0.002004  loss: 3.5873 (3.3978)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.0796  max mem: 41794
Epoch: [159]  [280/312]  eta: 0:00:24  lr: 0.002003  min_lr: 0.002003  loss: 3.5494 (3.3992)  weight_decay: 0.0500 (0.0500)  time: 0.6812  data: 0.0569  max mem: 41794
Epoch: [159]  [290/312]  eta: 0:00:17  lr: 0.002002  min_lr: 0.002002  loss: 3.6218 (3.4052)  weight_decay: 0.0500 (0.0500)  time: 0.8500  data: 0.1017  max mem: 41794
Epoch: [159]  [300/312]  eta: 0:00:09  lr: 0.002001  min_lr: 0.002001  loss: 3.6986 (3.4080)  weight_decay: 0.0500 (0.0500)  time: 0.6715  data: 0.0612  max mem: 41794
Epoch: [159]  [310/312]  eta: 0:00:01  lr: 0.002001  min_lr: 0.002001  loss: 3.6706 (3.4149)  weight_decay: 0.0500 (0.0500)  time: 0.4829  data: 0.0161  max mem: 41794
Epoch: [159]  [311/312]  eta: 0:00:00  lr: 0.002001  min_lr: 0.002001  loss: 3.6516 (3.4124)  weight_decay: 0.0500 (0.0500)  time: 0.4811  data: 0.0161  max mem: 41794
Epoch: [159] Total time: 0:03:58 (0.7631 s / it)
Averaged stats: lr: 0.002001  min_lr: 0.002001  loss: 3.6516 (3.4074)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 1.1200 (1.1200)  acc1: 79.5573 (79.5573)  acc5: 93.6198 (93.6198)  time: 7.8464  data: 7.6348  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4180 (1.3772)  acc1: 72.7865 (72.8480)  acc5: 91.7969 (91.5840)  time: 1.0392  data: 0.8673  max mem: 41794
Test: Total time: 0:00:09 (1.0517 s / it)
* Acc@1 72.878 Acc@5 91.594 loss 1.368
Accuracy of the model on the 50000 test images: 72.9%
Max accuracy: 73.26%
Epoch: [160]  [  0/312]  eta: 1:24:24  lr: 0.002001  min_lr: 0.002001  loss: 2.9719 (2.9719)  weight_decay: 0.0500 (0.0500)  time: 16.2321  data: 15.7704  max mem: 41794
Epoch: [160]  [ 10/312]  eta: 0:10:57  lr: 0.002000  min_lr: 0.002000  loss: 3.5003 (3.5295)  weight_decay: 0.0500 (0.0500)  time: 2.1779  data: 1.4343  max mem: 41794
Epoch: [160]  [ 20/312]  eta: 0:07:20  lr: 0.001999  min_lr: 0.001999  loss: 3.6640 (3.6158)  weight_decay: 0.0500 (0.0500)  time: 0.7714  data: 0.1302  max mem: 41794
Epoch: [160]  [ 30/312]  eta: 0:05:35  lr: 0.001998  min_lr: 0.001998  loss: 3.8043 (3.6066)  weight_decay: 0.0500 (0.0500)  time: 0.6451  data: 0.1355  max mem: 41794
Epoch: [160]  [ 40/312]  eta: 0:04:46  lr: 0.001998  min_lr: 0.001998  loss: 3.5840 (3.5221)  weight_decay: 0.0500 (0.0500)  time: 0.5792  data: 0.0661  max mem: 41794
Epoch: [160]  [ 50/312]  eta: 0:04:24  lr: 0.001997  min_lr: 0.001997  loss: 3.2875 (3.4615)  weight_decay: 0.0500 (0.0500)  time: 0.7292  data: 0.1939  max mem: 41794
Epoch: [160]  [ 60/312]  eta: 0:03:53  lr: 0.001996  min_lr: 0.001996  loss: 3.5591 (3.4887)  weight_decay: 0.0500 (0.0500)  time: 0.6610  data: 0.1345  max mem: 41794
Epoch: [160]  [ 70/312]  eta: 0:03:42  lr: 0.001995  min_lr: 0.001995  loss: 3.6288 (3.5047)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1828  max mem: 41794
Epoch: [160]  [ 80/312]  eta: 0:03:29  lr: 0.001995  min_lr: 0.001995  loss: 3.4956 (3.4527)  weight_decay: 0.0500 (0.0500)  time: 0.8395  data: 0.3393  max mem: 41794
Epoch: [160]  [ 90/312]  eta: 0:03:11  lr: 0.001994  min_lr: 0.001994  loss: 3.1670 (3.4255)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1605  max mem: 41794
Epoch: [160]  [100/312]  eta: 0:03:02  lr: 0.001993  min_lr: 0.001993  loss: 3.2421 (3.4023)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.1794  max mem: 41794
Epoch: [160]  [110/312]  eta: 0:02:47  lr: 0.001993  min_lr: 0.001993  loss: 3.3976 (3.4085)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1766  max mem: 41794
Epoch: [160]  [120/312]  eta: 0:02:40  lr: 0.001992  min_lr: 0.001992  loss: 3.3976 (3.3735)  weight_decay: 0.0500 (0.0500)  time: 0.7033  data: 0.2038  max mem: 41794
Epoch: [160]  [130/312]  eta: 0:02:33  lr: 0.001991  min_lr: 0.001991  loss: 3.3795 (3.3807)  weight_decay: 0.0500 (0.0500)  time: 0.9271  data: 0.4232  max mem: 41794
Epoch: [160]  [140/312]  eta: 0:02:20  lr: 0.001990  min_lr: 0.001990  loss: 3.6998 (3.3933)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.2200  max mem: 41794
Epoch: [160]  [150/312]  eta: 0:02:12  lr: 0.001990  min_lr: 0.001990  loss: 3.6389 (3.4097)  weight_decay: 0.0500 (0.0500)  time: 0.6618  data: 0.1657  max mem: 41794
Epoch: [160]  [160/312]  eta: 0:02:05  lr: 0.001989  min_lr: 0.001989  loss: 3.5906 (3.4106)  weight_decay: 0.0500 (0.0500)  time: 0.8531  data: 0.3564  max mem: 41794
Epoch: [160]  [170/312]  eta: 0:01:54  lr: 0.001988  min_lr: 0.001988  loss: 3.4777 (3.4058)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.1913  max mem: 41794
Epoch: [160]  [180/312]  eta: 0:01:46  lr: 0.001988  min_lr: 0.001988  loss: 3.4593 (3.4058)  weight_decay: 0.0500 (0.0500)  time: 0.7130  data: 0.1824  max mem: 41794
Epoch: [160]  [190/312]  eta: 0:01:36  lr: 0.001987  min_lr: 0.001987  loss: 3.2785 (3.3850)  weight_decay: 0.0500 (0.0500)  time: 0.7117  data: 0.1824  max mem: 41794
Epoch: [160]  [200/312]  eta: 0:01:29  lr: 0.001986  min_lr: 0.001986  loss: 3.3538 (3.3932)  weight_decay: 0.0500 (0.0500)  time: 0.7002  data: 0.1681  max mem: 41794
Epoch: [160]  [210/312]  eta: 0:01:21  lr: 0.001985  min_lr: 0.001985  loss: 3.3538 (3.3761)  weight_decay: 0.0500 (0.0500)  time: 0.8684  data: 0.3025  max mem: 41794
Epoch: [160]  [220/312]  eta: 0:01:12  lr: 0.001985  min_lr: 0.001985  loss: 3.1899 (3.3769)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.1366  max mem: 41794
Epoch: [160]  [230/312]  eta: 0:01:04  lr: 0.001984  min_lr: 0.001984  loss: 3.5096 (3.3731)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.1420  max mem: 41794
Epoch: [160]  [240/312]  eta: 0:00:57  lr: 0.001983  min_lr: 0.001983  loss: 3.5899 (3.3779)  weight_decay: 0.0500 (0.0500)  time: 0.8799  data: 0.3402  max mem: 41794
Epoch: [160]  [250/312]  eta: 0:00:48  lr: 0.001983  min_lr: 0.001983  loss: 3.6626 (3.3788)  weight_decay: 0.0500 (0.0500)  time: 0.6997  data: 0.2004  max mem: 41794
Epoch: [160]  [260/312]  eta: 0:00:40  lr: 0.001982  min_lr: 0.001982  loss: 3.2823 (3.3775)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.1781  max mem: 41794
Epoch: [160]  [270/312]  eta: 0:00:32  lr: 0.001981  min_lr: 0.001981  loss: 3.3421 (3.3834)  weight_decay: 0.0500 (0.0500)  time: 0.6953  data: 0.1781  max mem: 41794
Epoch: [160]  [280/312]  eta: 0:00:24  lr: 0.001980  min_lr: 0.001980  loss: 3.5235 (3.3791)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.1876  max mem: 41794
Epoch: [160]  [290/312]  eta: 0:00:17  lr: 0.001980  min_lr: 0.001980  loss: 3.5480 (3.3853)  weight_decay: 0.0500 (0.0500)  time: 0.8227  data: 0.3201  max mem: 41794
Epoch: [160]  [300/312]  eta: 0:00:09  lr: 0.001979  min_lr: 0.001979  loss: 3.5100 (3.3856)  weight_decay: 0.0500 (0.0500)  time: 0.6235  data: 0.1328  max mem: 41794
Epoch: [160]  [310/312]  eta: 0:00:01  lr: 0.001978  min_lr: 0.001978  loss: 3.4364 (3.3851)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [160]  [311/312]  eta: 0:00:00  lr: 0.001978  min_lr: 0.001978  loss: 3.4364 (3.3859)  weight_decay: 0.0500 (0.0500)  time: 0.4639  data: 0.0001  max mem: 41794
Epoch: [160] Total time: 0:03:58 (0.7647 s / it)
Averaged stats: lr: 0.001978  min_lr: 0.001978  loss: 3.4364 (3.3888)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.0496 (1.0496)  acc1: 79.0365 (79.0365)  acc5: 94.2708 (94.2708)  time: 8.7101  data: 8.5024  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3166 (1.2475)  acc1: 74.7396 (73.4720)  acc5: 91.7969 (91.9360)  time: 1.1158  data: 0.9448  max mem: 41794
Test: Total time: 0:00:10 (1.1336 s / it)
* Acc@1 73.806 Acc@5 92.062 loss 1.242
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 73.81%
Epoch: [161]  [  0/312]  eta: 1:20:11  lr: 0.001978  min_lr: 0.001978  loss: 2.9571 (2.9571)  weight_decay: 0.0500 (0.0500)  time: 15.4208  data: 14.9428  max mem: 41794
Epoch: [161]  [ 10/312]  eta: 0:10:35  lr: 0.001977  min_lr: 0.001977  loss: 3.5816 (3.3002)  weight_decay: 0.0500 (0.0500)  time: 2.1045  data: 1.3590  max mem: 41794
Epoch: [161]  [ 20/312]  eta: 0:07:08  lr: 0.001977  min_lr: 0.001977  loss: 3.5816 (3.3207)  weight_decay: 0.0500 (0.0500)  time: 0.7698  data: 0.1247  max mem: 41794
Epoch: [161]  [ 30/312]  eta: 0:05:26  lr: 0.001976  min_lr: 0.001976  loss: 3.3993 (3.3156)  weight_decay: 0.0500 (0.0500)  time: 0.6388  data: 0.1249  max mem: 41794
Epoch: [161]  [ 40/312]  eta: 0:04:48  lr: 0.001975  min_lr: 0.001975  loss: 3.4825 (3.3651)  weight_decay: 0.0500 (0.0500)  time: 0.6319  data: 0.1292  max mem: 41794
Epoch: [161]  [ 50/312]  eta: 0:04:29  lr: 0.001974  min_lr: 0.001974  loss: 3.5628 (3.3688)  weight_decay: 0.0500 (0.0500)  time: 0.8249  data: 0.2704  max mem: 41794
Epoch: [161]  [ 60/312]  eta: 0:03:58  lr: 0.001974  min_lr: 0.001974  loss: 3.5628 (3.4061)  weight_decay: 0.0500 (0.0500)  time: 0.7092  data: 0.1421  max mem: 41794
Epoch: [161]  [ 70/312]  eta: 0:03:47  lr: 0.001973  min_lr: 0.001973  loss: 3.5386 (3.3865)  weight_decay: 0.0500 (0.0500)  time: 0.7230  data: 0.1251  max mem: 41794
Epoch: [161]  [ 80/312]  eta: 0:03:33  lr: 0.001972  min_lr: 0.001972  loss: 2.9664 (3.3631)  weight_decay: 0.0500 (0.0500)  time: 0.8471  data: 0.1842  max mem: 41794
Epoch: [161]  [ 90/312]  eta: 0:03:18  lr: 0.001972  min_lr: 0.001972  loss: 2.9664 (3.3536)  weight_decay: 0.0500 (0.0500)  time: 0.7178  data: 0.0597  max mem: 41794
Epoch: [161]  [100/312]  eta: 0:03:09  lr: 0.001971  min_lr: 0.001971  loss: 3.4827 (3.3759)  weight_decay: 0.0500 (0.0500)  time: 0.7875  data: 0.0378  max mem: 41794
Epoch: [161]  [110/312]  eta: 0:02:53  lr: 0.001970  min_lr: 0.001970  loss: 3.5844 (3.3913)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.0406  max mem: 41794
Epoch: [161]  [120/312]  eta: 0:02:44  lr: 0.001969  min_lr: 0.001969  loss: 3.4892 (3.3975)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.0442  max mem: 41794
Epoch: [161]  [130/312]  eta: 0:02:34  lr: 0.001969  min_lr: 0.001969  loss: 3.4166 (3.3855)  weight_decay: 0.0500 (0.0500)  time: 0.8001  data: 0.0604  max mem: 41794
Epoch: [161]  [140/312]  eta: 0:02:22  lr: 0.001968  min_lr: 0.001968  loss: 3.5378 (3.3947)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.0532  max mem: 41794
Epoch: [161]  [150/312]  eta: 0:02:14  lr: 0.001967  min_lr: 0.001967  loss: 3.6710 (3.4036)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.0851  max mem: 41794
Epoch: [161]  [160/312]  eta: 0:02:05  lr: 0.001967  min_lr: 0.001967  loss: 3.5492 (3.4059)  weight_decay: 0.0500 (0.0500)  time: 0.7771  data: 0.0515  max mem: 41794
Epoch: [161]  [170/312]  eta: 0:01:54  lr: 0.001966  min_lr: 0.001966  loss: 3.6994 (3.4298)  weight_decay: 0.0500 (0.0500)  time: 0.6482  data: 0.0431  max mem: 41794
Epoch: [161]  [180/312]  eta: 0:01:47  lr: 0.001965  min_lr: 0.001965  loss: 3.7523 (3.4406)  weight_decay: 0.0500 (0.0500)  time: 0.7432  data: 0.0979  max mem: 41794
Epoch: [161]  [190/312]  eta: 0:01:37  lr: 0.001964  min_lr: 0.001964  loss: 3.6155 (3.4475)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.0569  max mem: 41794
Epoch: [161]  [200/312]  eta: 0:01:29  lr: 0.001964  min_lr: 0.001964  loss: 3.5816 (3.4569)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.0448  max mem: 41794
Epoch: [161]  [210/312]  eta: 0:01:21  lr: 0.001963  min_lr: 0.001963  loss: 3.6856 (3.4615)  weight_decay: 0.0500 (0.0500)  time: 0.8426  data: 0.0772  max mem: 41794
Epoch: [161]  [220/312]  eta: 0:01:12  lr: 0.001962  min_lr: 0.001962  loss: 3.6415 (3.4653)  weight_decay: 0.0500 (0.0500)  time: 0.7016  data: 0.0360  max mem: 41794
Epoch: [161]  [230/312]  eta: 0:01:05  lr: 0.001962  min_lr: 0.001962  loss: 3.6739 (3.4739)  weight_decay: 0.0500 (0.0500)  time: 0.6964  data: 0.0278  max mem: 41794
Epoch: [161]  [240/312]  eta: 0:00:57  lr: 0.001961  min_lr: 0.001961  loss: 3.6739 (3.4746)  weight_decay: 0.0500 (0.0500)  time: 0.8138  data: 0.0365  max mem: 41794
Epoch: [161]  [250/312]  eta: 0:00:48  lr: 0.001960  min_lr: 0.001960  loss: 3.5688 (3.4679)  weight_decay: 0.0500 (0.0500)  time: 0.7188  data: 0.0242  max mem: 41794
Epoch: [161]  [260/312]  eta: 0:00:41  lr: 0.001959  min_lr: 0.001959  loss: 3.3500 (3.4600)  weight_decay: 0.0500 (0.0500)  time: 0.7589  data: 0.0339  max mem: 41794
Epoch: [161]  [270/312]  eta: 0:00:32  lr: 0.001959  min_lr: 0.001959  loss: 3.5881 (3.4630)  weight_decay: 0.0500 (0.0500)  time: 0.7070  data: 0.0272  max mem: 41794
Epoch: [161]  [280/312]  eta: 0:00:24  lr: 0.001958  min_lr: 0.001958  loss: 3.5881 (3.4590)  weight_decay: 0.0500 (0.0500)  time: 0.6288  data: 0.0329  max mem: 41794
Epoch: [161]  [290/312]  eta: 0:00:17  lr: 0.001957  min_lr: 0.001957  loss: 3.2251 (3.4526)  weight_decay: 0.0500 (0.0500)  time: 0.7070  data: 0.0260  max mem: 41794
Epoch: [161]  [300/312]  eta: 0:00:09  lr: 0.001956  min_lr: 0.001956  loss: 3.3765 (3.4514)  weight_decay: 0.0500 (0.0500)  time: 0.5861  data: 0.0098  max mem: 41794
Epoch: [161]  [310/312]  eta: 0:00:01  lr: 0.001956  min_lr: 0.001956  loss: 3.6450 (3.4590)  weight_decay: 0.0500 (0.0500)  time: 0.4742  data: 0.0097  max mem: 41794
Epoch: [161]  [311/312]  eta: 0:00:00  lr: 0.001956  min_lr: 0.001956  loss: 3.6987 (3.4602)  weight_decay: 0.0500 (0.0500)  time: 0.4731  data: 0.0097  max mem: 41794
Epoch: [161] Total time: 0:03:57 (0.7627 s / it)
Averaged stats: lr: 0.001956  min_lr: 0.001956  loss: 3.6987 (3.4018)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.2022 (1.2022)  acc1: 79.4271 (79.4271)  acc5: 95.4427 (95.4427)  time: 8.6706  data: 8.4587  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5074 (1.4431)  acc1: 72.0052 (72.2720)  acc5: 91.5365 (91.5040)  time: 1.1114  data: 0.9399  max mem: 41794
Test: Total time: 0:00:10 (1.1498 s / it)
* Acc@1 72.550 Acc@5 91.546 loss 1.450
Accuracy of the model on the 50000 test images: 72.6%
Max accuracy: 73.81%
Epoch: [162]  [  0/312]  eta: 1:21:33  lr: 0.001956  min_lr: 0.001956  loss: 3.6427 (3.6427)  weight_decay: 0.0500 (0.0500)  time: 15.6857  data: 12.7256  max mem: 41794
Epoch: [162]  [ 10/312]  eta: 0:10:33  lr: 0.001955  min_lr: 0.001955  loss: 3.6174 (3.5946)  weight_decay: 0.0500 (0.0500)  time: 2.0968  data: 1.3723  max mem: 41794
Epoch: [162]  [ 20/312]  eta: 0:07:30  lr: 0.001954  min_lr: 0.001954  loss: 3.5721 (3.4642)  weight_decay: 0.0500 (0.0500)  time: 0.8347  data: 0.1948  max mem: 41794
Epoch: [162]  [ 30/312]  eta: 0:05:42  lr: 0.001953  min_lr: 0.001953  loss: 3.4726 (3.5166)  weight_decay: 0.0500 (0.0500)  time: 0.7304  data: 0.0766  max mem: 41794
Epoch: [162]  [ 40/312]  eta: 0:04:58  lr: 0.001953  min_lr: 0.001953  loss: 3.6532 (3.4850)  weight_decay: 0.0500 (0.0500)  time: 0.6336  data: 0.0034  max mem: 41794
Epoch: [162]  [ 50/312]  eta: 0:04:32  lr: 0.001952  min_lr: 0.001952  loss: 3.6011 (3.4576)  weight_decay: 0.0500 (0.0500)  time: 0.7686  data: 0.0473  max mem: 41794
Epoch: [162]  [ 60/312]  eta: 0:03:59  lr: 0.001951  min_lr: 0.001951  loss: 3.1693 (3.4148)  weight_decay: 0.0500 (0.0500)  time: 0.6501  data: 0.0446  max mem: 41794
Epoch: [162]  [ 70/312]  eta: 0:03:47  lr: 0.001951  min_lr: 0.001951  loss: 3.2820 (3.4003)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.0514  max mem: 41794
Epoch: [162]  [ 80/312]  eta: 0:03:35  lr: 0.001950  min_lr: 0.001950  loss: 3.3238 (3.3674)  weight_decay: 0.0500 (0.0500)  time: 0.8642  data: 0.0931  max mem: 41794
Epoch: [162]  [ 90/312]  eta: 0:03:17  lr: 0.001949  min_lr: 0.001949  loss: 2.9658 (3.3342)  weight_decay: 0.0500 (0.0500)  time: 0.7163  data: 0.0438  max mem: 41794
Epoch: [162]  [100/312]  eta: 0:03:06  lr: 0.001948  min_lr: 0.001948  loss: 3.5634 (3.3636)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.0691  max mem: 41794
Epoch: [162]  [110/312]  eta: 0:02:50  lr: 0.001948  min_lr: 0.001948  loss: 3.1803 (3.3251)  weight_decay: 0.0500 (0.0500)  time: 0.6277  data: 0.0676  max mem: 41794
Epoch: [162]  [120/312]  eta: 0:02:41  lr: 0.001947  min_lr: 0.001947  loss: 3.1803 (3.3449)  weight_decay: 0.0500 (0.0500)  time: 0.6675  data: 0.0700  max mem: 41794
Epoch: [162]  [130/312]  eta: 0:02:32  lr: 0.001946  min_lr: 0.001946  loss: 3.7072 (3.3601)  weight_decay: 0.0500 (0.0500)  time: 0.8201  data: 0.1978  max mem: 41794
Epoch: [162]  [140/312]  eta: 0:02:20  lr: 0.001946  min_lr: 0.001946  loss: 3.7155 (3.3643)  weight_decay: 0.0500 (0.0500)  time: 0.6503  data: 0.1285  max mem: 41794
Epoch: [162]  [150/312]  eta: 0:02:13  lr: 0.001945  min_lr: 0.001945  loss: 3.6971 (3.3853)  weight_decay: 0.0500 (0.0500)  time: 0.6981  data: 0.1984  max mem: 41794
Epoch: [162]  [160/312]  eta: 0:02:05  lr: 0.001944  min_lr: 0.001944  loss: 3.6093 (3.3918)  weight_decay: 0.0500 (0.0500)  time: 0.8805  data: 0.3847  max mem: 41794
Epoch: [162]  [170/312]  eta: 0:01:54  lr: 0.001943  min_lr: 0.001943  loss: 3.5945 (3.3984)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.1869  max mem: 41794
Epoch: [162]  [180/312]  eta: 0:01:47  lr: 0.001943  min_lr: 0.001943  loss: 3.6896 (3.4087)  weight_decay: 0.0500 (0.0500)  time: 0.7186  data: 0.2198  max mem: 41794
Epoch: [162]  [190/312]  eta: 0:01:37  lr: 0.001942  min_lr: 0.001942  loss: 3.6515 (3.4147)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.2214  max mem: 41794
Epoch: [162]  [200/312]  eta: 0:01:29  lr: 0.001941  min_lr: 0.001941  loss: 3.6359 (3.4228)  weight_decay: 0.0500 (0.0500)  time: 0.6761  data: 0.1791  max mem: 41794
Epoch: [162]  [210/312]  eta: 0:01:21  lr: 0.001941  min_lr: 0.001941  loss: 3.5373 (3.4113)  weight_decay: 0.0500 (0.0500)  time: 0.8540  data: 0.3566  max mem: 41794
Epoch: [162]  [220/312]  eta: 0:01:12  lr: 0.001940  min_lr: 0.001940  loss: 3.2279 (3.3991)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.1797  max mem: 41794
Epoch: [162]  [230/312]  eta: 0:01:04  lr: 0.001939  min_lr: 0.001939  loss: 3.3357 (3.3956)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1906  max mem: 41794
Epoch: [162]  [240/312]  eta: 0:00:57  lr: 0.001938  min_lr: 0.001938  loss: 3.3055 (3.3806)  weight_decay: 0.0500 (0.0500)  time: 0.8691  data: 0.3706  max mem: 41794
Epoch: [162]  [250/312]  eta: 0:00:48  lr: 0.001938  min_lr: 0.001938  loss: 3.3055 (3.3802)  weight_decay: 0.0500 (0.0500)  time: 0.6803  data: 0.1806  max mem: 41794
Epoch: [162]  [260/312]  eta: 0:00:40  lr: 0.001937  min_lr: 0.001937  loss: 3.3462 (3.3779)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1816  max mem: 41794
Epoch: [162]  [270/312]  eta: 0:00:32  lr: 0.001936  min_lr: 0.001936  loss: 3.5588 (3.3839)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1816  max mem: 41794
Epoch: [162]  [280/312]  eta: 0:00:24  lr: 0.001936  min_lr: 0.001936  loss: 3.5494 (3.3786)  weight_decay: 0.0500 (0.0500)  time: 0.6758  data: 0.1779  max mem: 41794
Epoch: [162]  [290/312]  eta: 0:00:17  lr: 0.001935  min_lr: 0.001935  loss: 3.5732 (3.3864)  weight_decay: 0.0500 (0.0500)  time: 0.8454  data: 0.3515  max mem: 41794
Epoch: [162]  [300/312]  eta: 0:00:09  lr: 0.001934  min_lr: 0.001934  loss: 3.6701 (3.3918)  weight_decay: 0.0500 (0.0500)  time: 0.6554  data: 0.1741  max mem: 41794
Epoch: [162]  [310/312]  eta: 0:00:01  lr: 0.001933  min_lr: 0.001933  loss: 3.7583 (3.3994)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [162]  [311/312]  eta: 0:00:00  lr: 0.001933  min_lr: 0.001933  loss: 3.6632 (3.3957)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [162] Total time: 0:03:58 (0.7648 s / it)
Averaged stats: lr: 0.001933  min_lr: 0.001933  loss: 3.6632 (3.4119)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 1.0655 (1.0655)  acc1: 80.2083 (80.2083)  acc5: 95.4427 (95.4427)  time: 8.7729  data: 8.5722  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4469 (1.3452)  acc1: 72.1354 (73.4880)  acc5: 91.6667 (91.9040)  time: 1.1283  data: 0.9526  max mem: 41794
Test: Total time: 0:00:10 (1.1702 s / it)
* Acc@1 73.824 Acc@5 92.106 loss 1.336
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 73.82%
Epoch: [163]  [  0/312]  eta: 1:22:06  lr: 0.001933  min_lr: 0.001933  loss: 2.9842 (2.9842)  weight_decay: 0.0500 (0.0500)  time: 15.7886  data: 15.2899  max mem: 41794
Epoch: [163]  [ 10/312]  eta: 0:11:00  lr: 0.001932  min_lr: 0.001932  loss: 3.8361 (3.6873)  weight_decay: 0.0500 (0.0500)  time: 2.1881  data: 1.3907  max mem: 41794
Epoch: [163]  [ 20/312]  eta: 0:07:27  lr: 0.001932  min_lr: 0.001932  loss: 3.6795 (3.6233)  weight_decay: 0.0500 (0.0500)  time: 0.8185  data: 0.0198  max mem: 41794
Epoch: [163]  [ 30/312]  eta: 0:05:48  lr: 0.001931  min_lr: 0.001931  loss: 3.6238 (3.5639)  weight_decay: 0.0500 (0.0500)  time: 0.7134  data: 0.0595  max mem: 41794
Epoch: [163]  [ 40/312]  eta: 0:04:48  lr: 0.001930  min_lr: 0.001930  loss: 3.6873 (3.5716)  weight_decay: 0.0500 (0.0500)  time: 0.5665  data: 0.0404  max mem: 41794
Epoch: [163]  [ 50/312]  eta: 0:04:20  lr: 0.001930  min_lr: 0.001930  loss: 3.5523 (3.5414)  weight_decay: 0.0500 (0.0500)  time: 0.6195  data: 0.0697  max mem: 41794
Epoch: [163]  [ 60/312]  eta: 0:03:49  lr: 0.001929  min_lr: 0.001929  loss: 3.5285 (3.5536)  weight_decay: 0.0500 (0.0500)  time: 0.6084  data: 0.0713  max mem: 41794
Epoch: [163]  [ 70/312]  eta: 0:03:41  lr: 0.001928  min_lr: 0.001928  loss: 3.6168 (3.5530)  weight_decay: 0.0500 (0.0500)  time: 0.7177  data: 0.1371  max mem: 41794
Epoch: [163]  [ 80/312]  eta: 0:03:29  lr: 0.001927  min_lr: 0.001927  loss: 3.5362 (3.5261)  weight_decay: 0.0500 (0.0500)  time: 0.8824  data: 0.1916  max mem: 41794
Epoch: [163]  [ 90/312]  eta: 0:03:15  lr: 0.001927  min_lr: 0.001927  loss: 3.5362 (3.5177)  weight_decay: 0.0500 (0.0500)  time: 0.7439  data: 0.1295  max mem: 41794
Epoch: [163]  [100/312]  eta: 0:03:06  lr: 0.001926  min_lr: 0.001926  loss: 3.5807 (3.5197)  weight_decay: 0.0500 (0.0500)  time: 0.7671  data: 0.1831  max mem: 41794
Epoch: [163]  [110/312]  eta: 0:02:50  lr: 0.001925  min_lr: 0.001925  loss: 3.5934 (3.5008)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.1135  max mem: 41794
Epoch: [163]  [120/312]  eta: 0:02:40  lr: 0.001925  min_lr: 0.001925  loss: 3.5632 (3.5098)  weight_decay: 0.0500 (0.0500)  time: 0.6336  data: 0.0982  max mem: 41794
Epoch: [163]  [130/312]  eta: 0:02:33  lr: 0.001924  min_lr: 0.001924  loss: 3.3087 (3.4749)  weight_decay: 0.0500 (0.0500)  time: 0.8319  data: 0.2428  max mem: 41794
Epoch: [163]  [140/312]  eta: 0:02:20  lr: 0.001923  min_lr: 0.001923  loss: 3.2284 (3.4543)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.1485  max mem: 41794
Epoch: [163]  [150/312]  eta: 0:02:14  lr: 0.001922  min_lr: 0.001922  loss: 3.4052 (3.4483)  weight_decay: 0.0500 (0.0500)  time: 0.7617  data: 0.1455  max mem: 41794
Epoch: [163]  [160/312]  eta: 0:02:04  lr: 0.001922  min_lr: 0.001922  loss: 3.4711 (3.4396)  weight_decay: 0.0500 (0.0500)  time: 0.8314  data: 0.1587  max mem: 41794
Epoch: [163]  [170/312]  eta: 0:01:54  lr: 0.001921  min_lr: 0.001921  loss: 3.5014 (3.4367)  weight_decay: 0.0500 (0.0500)  time: 0.6132  data: 0.0709  max mem: 41794
Epoch: [163]  [180/312]  eta: 0:01:45  lr: 0.001920  min_lr: 0.001920  loss: 3.5627 (3.4446)  weight_decay: 0.0500 (0.0500)  time: 0.6591  data: 0.1475  max mem: 41794
Epoch: [163]  [190/312]  eta: 0:01:35  lr: 0.001920  min_lr: 0.001920  loss: 3.5627 (3.4439)  weight_decay: 0.0500 (0.0500)  time: 0.6091  data: 0.0927  max mem: 41794
Epoch: [163]  [200/312]  eta: 0:01:28  lr: 0.001919  min_lr: 0.001919  loss: 3.5788 (3.4437)  weight_decay: 0.0500 (0.0500)  time: 0.7282  data: 0.1649  max mem: 41794
Epoch: [163]  [210/312]  eta: 0:01:21  lr: 0.001918  min_lr: 0.001918  loss: 3.5788 (3.4441)  weight_decay: 0.0500 (0.0500)  time: 0.9256  data: 0.2962  max mem: 41794
Epoch: [163]  [220/312]  eta: 0:01:12  lr: 0.001917  min_lr: 0.001917  loss: 3.1638 (3.4322)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1342  max mem: 41794
Epoch: [163]  [230/312]  eta: 0:01:04  lr: 0.001917  min_lr: 0.001917  loss: 3.2684 (3.4327)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.1009  max mem: 41794
Epoch: [163]  [240/312]  eta: 0:00:56  lr: 0.001916  min_lr: 0.001916  loss: 3.5811 (3.4386)  weight_decay: 0.0500 (0.0500)  time: 0.8312  data: 0.1832  max mem: 41794
Epoch: [163]  [250/312]  eta: 0:00:48  lr: 0.001915  min_lr: 0.001915  loss: 3.5771 (3.4397)  weight_decay: 0.0500 (0.0500)  time: 0.6622  data: 0.0972  max mem: 41794
Epoch: [163]  [260/312]  eta: 0:00:40  lr: 0.001915  min_lr: 0.001915  loss: 3.5102 (3.4388)  weight_decay: 0.0500 (0.0500)  time: 0.6945  data: 0.1220  max mem: 41794
Epoch: [163]  [270/312]  eta: 0:00:32  lr: 0.001914  min_lr: 0.001914  loss: 3.6025 (3.4357)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1077  max mem: 41794
Epoch: [163]  [280/312]  eta: 0:00:24  lr: 0.001913  min_lr: 0.001913  loss: 3.3833 (3.4198)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.0921  max mem: 41794
Epoch: [163]  [290/312]  eta: 0:00:17  lr: 0.001912  min_lr: 0.001912  loss: 3.2755 (3.4187)  weight_decay: 0.0500 (0.0500)  time: 0.8587  data: 0.1971  max mem: 41794
Epoch: [163]  [300/312]  eta: 0:00:09  lr: 0.001912  min_lr: 0.001912  loss: 3.5689 (3.4211)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.1054  max mem: 41794
Epoch: [163]  [310/312]  eta: 0:00:01  lr: 0.001911  min_lr: 0.001911  loss: 3.5689 (3.4253)  weight_decay: 0.0500 (0.0500)  time: 0.4704  data: 0.0001  max mem: 41794
Epoch: [163]  [311/312]  eta: 0:00:00  lr: 0.001911  min_lr: 0.001911  loss: 3.5364 (3.4227)  weight_decay: 0.0500 (0.0500)  time: 0.4679  data: 0.0001  max mem: 41794
Epoch: [163] Total time: 0:03:57 (0.7626 s / it)
Averaged stats: lr: 0.001911  min_lr: 0.001911  loss: 3.5364 (3.3814)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1257 (1.1257)  acc1: 79.5573 (79.5573)  acc5: 94.6615 (94.6615)  time: 8.2561  data: 8.0511  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4248 (1.3453)  acc1: 72.3958 (73.7600)  acc5: 91.4062 (91.9520)  time: 1.0655  data: 0.8946  max mem: 41794
Test: Total time: 0:00:09 (1.0761 s / it)
* Acc@1 73.830 Acc@5 91.904 loss 1.341
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 73.83%
Epoch: [164]  [  0/312]  eta: 1:17:40  lr: 0.001911  min_lr: 0.001911  loss: 3.6209 (3.6209)  weight_decay: 0.0500 (0.0500)  time: 14.9388  data: 13.4506  max mem: 41794
Epoch: [164]  [ 10/312]  eta: 0:09:47  lr: 0.001910  min_lr: 0.001910  loss: 3.3603 (3.1940)  weight_decay: 0.0500 (0.0500)  time: 1.9456  data: 1.3618  max mem: 41794
Epoch: [164]  [ 20/312]  eta: 0:07:05  lr: 0.001909  min_lr: 0.001909  loss: 3.3926 (3.3151)  weight_decay: 0.0500 (0.0500)  time: 0.7824  data: 0.2145  max mem: 41794
Epoch: [164]  [ 30/312]  eta: 0:05:26  lr: 0.001909  min_lr: 0.001909  loss: 3.6089 (3.3575)  weight_decay: 0.0500 (0.0500)  time: 0.7240  data: 0.1399  max mem: 41794
Epoch: [164]  [ 40/312]  eta: 0:04:47  lr: 0.001908  min_lr: 0.001908  loss: 3.4688 (3.3068)  weight_decay: 0.0500 (0.0500)  time: 0.6396  data: 0.1346  max mem: 41794
Epoch: [164]  [ 50/312]  eta: 0:04:28  lr: 0.001907  min_lr: 0.001907  loss: 3.2647 (3.2732)  weight_decay: 0.0500 (0.0500)  time: 0.8235  data: 0.3154  max mem: 41794
Epoch: [164]  [ 60/312]  eta: 0:03:55  lr: 0.001906  min_lr: 0.001906  loss: 3.2092 (3.2436)  weight_decay: 0.0500 (0.0500)  time: 0.6868  data: 0.1830  max mem: 41794
Epoch: [164]  [ 70/312]  eta: 0:03:45  lr: 0.001906  min_lr: 0.001906  loss: 3.0003 (3.2206)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1546  max mem: 41794
Epoch: [164]  [ 80/312]  eta: 0:03:32  lr: 0.001905  min_lr: 0.001905  loss: 2.9738 (3.1978)  weight_decay: 0.0500 (0.0500)  time: 0.8533  data: 0.2656  max mem: 41794
Epoch: [164]  [ 90/312]  eta: 0:03:15  lr: 0.001904  min_lr: 0.001904  loss: 3.5513 (3.2218)  weight_decay: 0.0500 (0.0500)  time: 0.7046  data: 0.1421  max mem: 41794
Epoch: [164]  [100/312]  eta: 0:03:06  lr: 0.001904  min_lr: 0.001904  loss: 3.5784 (3.2510)  weight_decay: 0.0500 (0.0500)  time: 0.7288  data: 0.1401  max mem: 41794
Epoch: [164]  [110/312]  eta: 0:02:50  lr: 0.001903  min_lr: 0.001903  loss: 3.5168 (3.2550)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.1095  max mem: 41794
Epoch: [164]  [120/312]  eta: 0:02:41  lr: 0.001902  min_lr: 0.001902  loss: 3.4203 (3.2637)  weight_decay: 0.0500 (0.0500)  time: 0.6378  data: 0.1399  max mem: 41794
Epoch: [164]  [130/312]  eta: 0:02:32  lr: 0.001901  min_lr: 0.001901  loss: 3.5470 (3.3007)  weight_decay: 0.0500 (0.0500)  time: 0.8148  data: 0.3113  max mem: 41794
Epoch: [164]  [140/312]  eta: 0:02:20  lr: 0.001901  min_lr: 0.001901  loss: 3.4923 (3.2913)  weight_decay: 0.0500 (0.0500)  time: 0.6663  data: 0.1736  max mem: 41794
Epoch: [164]  [150/312]  eta: 0:02:13  lr: 0.001900  min_lr: 0.001900  loss: 3.2210 (3.2774)  weight_decay: 0.0500 (0.0500)  time: 0.7008  data: 0.2002  max mem: 41794
Epoch: [164]  [160/312]  eta: 0:02:05  lr: 0.001899  min_lr: 0.001899  loss: 3.5547 (3.2868)  weight_decay: 0.0500 (0.0500)  time: 0.9249  data: 0.3492  max mem: 41794
Epoch: [164]  [170/312]  eta: 0:01:55  lr: 0.001899  min_lr: 0.001899  loss: 3.5547 (3.2922)  weight_decay: 0.0500 (0.0500)  time: 0.7736  data: 0.1513  max mem: 41794
Epoch: [164]  [180/312]  eta: 0:01:47  lr: 0.001898  min_lr: 0.001898  loss: 3.5229 (3.3031)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.1084  max mem: 41794
Epoch: [164]  [190/312]  eta: 0:01:37  lr: 0.001897  min_lr: 0.001897  loss: 3.6569 (3.3149)  weight_decay: 0.0500 (0.0500)  time: 0.6306  data: 0.1103  max mem: 41794
Epoch: [164]  [200/312]  eta: 0:01:29  lr: 0.001896  min_lr: 0.001896  loss: 3.5580 (3.3204)  weight_decay: 0.0500 (0.0500)  time: 0.6481  data: 0.1271  max mem: 41794
Epoch: [164]  [210/312]  eta: 0:01:21  lr: 0.001896  min_lr: 0.001896  loss: 3.4195 (3.3123)  weight_decay: 0.0500 (0.0500)  time: 0.8169  data: 0.2675  max mem: 41794
Epoch: [164]  [220/312]  eta: 0:01:12  lr: 0.001895  min_lr: 0.001895  loss: 3.5075 (3.3248)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.1431  max mem: 41794
Epoch: [164]  [230/312]  eta: 0:01:04  lr: 0.001894  min_lr: 0.001894  loss: 3.4643 (3.3270)  weight_decay: 0.0500 (0.0500)  time: 0.6879  data: 0.1268  max mem: 41794
Epoch: [164]  [240/312]  eta: 0:00:56  lr: 0.001894  min_lr: 0.001894  loss: 3.3736 (3.3272)  weight_decay: 0.0500 (0.0500)  time: 0.8326  data: 0.2825  max mem: 41794
Epoch: [164]  [250/312]  eta: 0:00:48  lr: 0.001893  min_lr: 0.001893  loss: 3.5688 (3.3398)  weight_decay: 0.0500 (0.0500)  time: 0.6667  data: 0.1564  max mem: 41794
Epoch: [164]  [260/312]  eta: 0:00:40  lr: 0.001892  min_lr: 0.001892  loss: 3.6222 (3.3483)  weight_decay: 0.0500 (0.0500)  time: 0.7340  data: 0.1652  max mem: 41794
Epoch: [164]  [270/312]  eta: 0:00:32  lr: 0.001891  min_lr: 0.001891  loss: 3.5870 (3.3495)  weight_decay: 0.0500 (0.0500)  time: 0.7214  data: 0.1666  max mem: 41794
Epoch: [164]  [280/312]  eta: 0:00:24  lr: 0.001891  min_lr: 0.001891  loss: 3.5286 (3.3550)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.1348  max mem: 41794
Epoch: [164]  [290/312]  eta: 0:00:17  lr: 0.001890  min_lr: 0.001890  loss: 3.3659 (3.3532)  weight_decay: 0.0500 (0.0500)  time: 0.7805  data: 0.2393  max mem: 41794
Epoch: [164]  [300/312]  eta: 0:00:09  lr: 0.001889  min_lr: 0.001889  loss: 3.4556 (3.3589)  weight_decay: 0.0500 (0.0500)  time: 0.5994  data: 0.1071  max mem: 41794
Epoch: [164]  [310/312]  eta: 0:00:01  lr: 0.001889  min_lr: 0.001889  loss: 3.4335 (3.3526)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0010  max mem: 41794
Epoch: [164]  [311/312]  eta: 0:00:00  lr: 0.001888  min_lr: 0.001888  loss: 3.4335 (3.3489)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0010  max mem: 41794
Epoch: [164] Total time: 0:03:57 (0.7601 s / it)
Averaged stats: lr: 0.001888  min_lr: 0.001888  loss: 3.4335 (3.3968)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 0.9142 (0.9142)  acc1: 82.2917 (82.2917)  acc5: 95.3125 (95.3125)  time: 9.1242  data: 8.9122  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1880 (1.1636)  acc1: 72.9167 (74.1600)  acc5: 93.0990 (92.5440)  time: 1.1625  data: 0.9903  max mem: 41794
Test: Total time: 0:00:10 (1.1765 s / it)
* Acc@1 73.794 Acc@5 92.294 loss 1.170
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 73.83%
Epoch: [165]  [  0/312]  eta: 1:22:02  lr: 0.001888  min_lr: 0.001888  loss: 2.9639 (2.9639)  weight_decay: 0.0500 (0.0500)  time: 15.7783  data: 14.7464  max mem: 41794
Epoch: [165]  [ 10/312]  eta: 0:10:57  lr: 0.001888  min_lr: 0.001888  loss: 3.0509 (3.1397)  weight_decay: 0.0500 (0.0500)  time: 2.1772  data: 1.3452  max mem: 41794
Epoch: [165]  [ 20/312]  eta: 0:07:20  lr: 0.001887  min_lr: 0.001887  loss: 3.4532 (3.3182)  weight_decay: 0.0500 (0.0500)  time: 0.7965  data: 0.0348  max mem: 41794
Epoch: [165]  [ 30/312]  eta: 0:05:35  lr: 0.001886  min_lr: 0.001886  loss: 3.5608 (3.3398)  weight_decay: 0.0500 (0.0500)  time: 0.6441  data: 0.0326  max mem: 41794
Epoch: [165]  [ 40/312]  eta: 0:04:48  lr: 0.001886  min_lr: 0.001886  loss: 3.4127 (3.3692)  weight_decay: 0.0500 (0.0500)  time: 0.5906  data: 0.0207  max mem: 41794
Epoch: [165]  [ 50/312]  eta: 0:04:27  lr: 0.001885  min_lr: 0.001885  loss: 3.6230 (3.4338)  weight_decay: 0.0500 (0.0500)  time: 0.7643  data: 0.1159  max mem: 41794
Epoch: [165]  [ 60/312]  eta: 0:03:55  lr: 0.001884  min_lr: 0.001884  loss: 3.6774 (3.4525)  weight_decay: 0.0500 (0.0500)  time: 0.6747  data: 0.0958  max mem: 41794
Epoch: [165]  [ 70/312]  eta: 0:03:44  lr: 0.001883  min_lr: 0.001883  loss: 3.2468 (3.4141)  weight_decay: 0.0500 (0.0500)  time: 0.6814  data: 0.1038  max mem: 41794
Epoch: [165]  [ 80/312]  eta: 0:03:32  lr: 0.001883  min_lr: 0.001883  loss: 3.5416 (3.4322)  weight_decay: 0.0500 (0.0500)  time: 0.8664  data: 0.2107  max mem: 41794
Epoch: [165]  [ 90/312]  eta: 0:03:13  lr: 0.001882  min_lr: 0.001882  loss: 3.5416 (3.3894)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.1075  max mem: 41794
Epoch: [165]  [100/312]  eta: 0:03:05  lr: 0.001881  min_lr: 0.001881  loss: 3.2475 (3.3935)  weight_decay: 0.0500 (0.0500)  time: 0.6969  data: 0.1200  max mem: 41794
Epoch: [165]  [110/312]  eta: 0:02:49  lr: 0.001880  min_lr: 0.001880  loss: 3.4322 (3.3950)  weight_decay: 0.0500 (0.0500)  time: 0.6953  data: 0.1216  max mem: 41794
Epoch: [165]  [120/312]  eta: 0:02:41  lr: 0.001880  min_lr: 0.001880  loss: 3.5334 (3.4109)  weight_decay: 0.0500 (0.0500)  time: 0.6773  data: 0.0971  max mem: 41794
Epoch: [165]  [130/312]  eta: 0:02:33  lr: 0.001879  min_lr: 0.001879  loss: 3.5866 (3.4227)  weight_decay: 0.0500 (0.0500)  time: 0.8525  data: 0.1970  max mem: 41794
Epoch: [165]  [140/312]  eta: 0:02:20  lr: 0.001878  min_lr: 0.001878  loss: 3.6118 (3.4322)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1022  max mem: 41794
Epoch: [165]  [150/312]  eta: 0:02:12  lr: 0.001878  min_lr: 0.001878  loss: 3.4100 (3.4165)  weight_decay: 0.0500 (0.0500)  time: 0.6448  data: 0.1062  max mem: 41794
Epoch: [165]  [160/312]  eta: 0:02:04  lr: 0.001877  min_lr: 0.001877  loss: 3.5869 (3.4336)  weight_decay: 0.0500 (0.0500)  time: 0.8276  data: 0.2069  max mem: 41794
Epoch: [165]  [170/312]  eta: 0:01:54  lr: 0.001876  min_lr: 0.001876  loss: 3.5869 (3.4269)  weight_decay: 0.0500 (0.0500)  time: 0.7149  data: 0.1448  max mem: 41794
Epoch: [165]  [180/312]  eta: 0:01:46  lr: 0.001875  min_lr: 0.001875  loss: 3.2458 (3.4134)  weight_decay: 0.0500 (0.0500)  time: 0.7221  data: 0.1649  max mem: 41794
Epoch: [165]  [190/312]  eta: 0:01:36  lr: 0.001875  min_lr: 0.001875  loss: 3.5184 (3.4219)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1214  max mem: 41794
Epoch: [165]  [200/312]  eta: 0:01:28  lr: 0.001874  min_lr: 0.001874  loss: 3.5448 (3.4225)  weight_decay: 0.0500 (0.0500)  time: 0.6456  data: 0.1075  max mem: 41794
Epoch: [165]  [210/312]  eta: 0:01:21  lr: 0.001873  min_lr: 0.001873  loss: 3.5301 (3.4210)  weight_decay: 0.0500 (0.0500)  time: 0.8598  data: 0.2859  max mem: 41794
Epoch: [165]  [220/312]  eta: 0:01:12  lr: 0.001873  min_lr: 0.001873  loss: 3.2426 (3.4065)  weight_decay: 0.0500 (0.0500)  time: 0.7094  data: 0.1791  max mem: 41794
Epoch: [165]  [230/312]  eta: 0:01:04  lr: 0.001872  min_lr: 0.001872  loss: 3.2740 (3.4087)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1769  max mem: 41794
Epoch: [165]  [240/312]  eta: 0:00:56  lr: 0.001871  min_lr: 0.001871  loss: 3.3261 (3.4039)  weight_decay: 0.0500 (0.0500)  time: 0.7811  data: 0.2466  max mem: 41794
Epoch: [165]  [250/312]  eta: 0:00:48  lr: 0.001870  min_lr: 0.001870  loss: 3.0892 (3.3943)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.1641  max mem: 41794
Epoch: [165]  [260/312]  eta: 0:00:40  lr: 0.001870  min_lr: 0.001870  loss: 3.5963 (3.4024)  weight_decay: 0.0500 (0.0500)  time: 0.7814  data: 0.2510  max mem: 41794
Epoch: [165]  [270/312]  eta: 0:00:32  lr: 0.001869  min_lr: 0.001869  loss: 3.6522 (3.4064)  weight_decay: 0.0500 (0.0500)  time: 0.6850  data: 0.1571  max mem: 41794
Epoch: [165]  [280/312]  eta: 0:00:24  lr: 0.001868  min_lr: 0.001868  loss: 3.4850 (3.4019)  weight_decay: 0.0500 (0.0500)  time: 0.7072  data: 0.1401  max mem: 41794
Epoch: [165]  [290/312]  eta: 0:00:17  lr: 0.001868  min_lr: 0.001868  loss: 2.9807 (3.3871)  weight_decay: 0.0500 (0.0500)  time: 0.9022  data: 0.2443  max mem: 41794
Epoch: [165]  [300/312]  eta: 0:00:09  lr: 0.001867  min_lr: 0.001867  loss: 2.9807 (3.3873)  weight_decay: 0.0500 (0.0500)  time: 0.6757  data: 0.1045  max mem: 41794
Epoch: [165]  [310/312]  eta: 0:00:01  lr: 0.001866  min_lr: 0.001866  loss: 3.5139 (3.3914)  weight_decay: 0.0500 (0.0500)  time: 0.4639  data: 0.0001  max mem: 41794
Epoch: [165]  [311/312]  eta: 0:00:00  lr: 0.001866  min_lr: 0.001866  loss: 3.4561 (3.3879)  weight_decay: 0.0500 (0.0500)  time: 0.4631  data: 0.0001  max mem: 41794
Epoch: [165] Total time: 0:03:59 (0.7667 s / it)
Averaged stats: lr: 0.001866  min_lr: 0.001866  loss: 3.4561 (3.3922)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.0785 (1.0785)  acc1: 80.9896 (80.9896)  acc5: 94.5312 (94.5312)  time: 8.0586  data: 7.8592  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3627 (1.3015)  acc1: 74.4792 (73.7120)  acc5: 92.1875 (91.8720)  time: 1.0641  data: 0.8922  max mem: 41794
Test: Total time: 0:00:09 (1.0773 s / it)
* Acc@1 73.784 Acc@5 92.086 loss 1.294
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 73.83%
Epoch: [166]  [  0/312]  eta: 1:24:05  lr: 0.001866  min_lr: 0.001866  loss: 3.3682 (3.3682)  weight_decay: 0.0500 (0.0500)  time: 16.1709  data: 15.6184  max mem: 41794
Epoch: [166]  [ 10/312]  eta: 0:12:06  lr: 0.001865  min_lr: 0.001865  loss: 3.3682 (3.3266)  weight_decay: 0.0500 (0.0500)  time: 2.4065  data: 1.4225  max mem: 41794
Epoch: [166]  [ 20/312]  eta: 0:07:33  lr: 0.001865  min_lr: 0.001865  loss: 3.4458 (3.3886)  weight_decay: 0.0500 (0.0500)  time: 0.8233  data: 0.0462  max mem: 41794
Epoch: [166]  [ 30/312]  eta: 0:05:42  lr: 0.001864  min_lr: 0.001864  loss: 3.4458 (3.3887)  weight_decay: 0.0500 (0.0500)  time: 0.5603  data: 0.0466  max mem: 41794
Epoch: [166]  [ 40/312]  eta: 0:04:43  lr: 0.001863  min_lr: 0.001863  loss: 3.3322 (3.3611)  weight_decay: 0.0500 (0.0500)  time: 0.5069  data: 0.0042  max mem: 41794
Epoch: [166]  [ 50/312]  eta: 0:04:06  lr: 0.001862  min_lr: 0.001862  loss: 3.4359 (3.3600)  weight_decay: 0.0500 (0.0500)  time: 0.5106  data: 0.0090  max mem: 41794
Epoch: [166]  [ 60/312]  eta: 0:03:38  lr: 0.001862  min_lr: 0.001862  loss: 3.4359 (3.3445)  weight_decay: 0.0500 (0.0500)  time: 0.5115  data: 0.0070  max mem: 41794
Epoch: [166]  [ 70/312]  eta: 0:03:22  lr: 0.001861  min_lr: 0.001861  loss: 3.3905 (3.3411)  weight_decay: 0.0500 (0.0500)  time: 0.5834  data: 0.0296  max mem: 41794
Epoch: [166]  [ 80/312]  eta: 0:03:14  lr: 0.001860  min_lr: 0.001860  loss: 3.3923 (3.3522)  weight_decay: 0.0500 (0.0500)  time: 0.7373  data: 0.0824  max mem: 41794
Epoch: [166]  [ 90/312]  eta: 0:03:02  lr: 0.001860  min_lr: 0.001860  loss: 3.3923 (3.3383)  weight_decay: 0.0500 (0.0500)  time: 0.7567  data: 0.0633  max mem: 41794
Epoch: [166]  [100/312]  eta: 0:02:52  lr: 0.001859  min_lr: 0.001859  loss: 3.5719 (3.3775)  weight_decay: 0.0500 (0.0500)  time: 0.7315  data: 0.0456  max mem: 41794
Epoch: [166]  [110/312]  eta: 0:02:42  lr: 0.001858  min_lr: 0.001858  loss: 3.6328 (3.3804)  weight_decay: 0.0500 (0.0500)  time: 0.7187  data: 0.0358  max mem: 41794
Epoch: [166]  [120/312]  eta: 0:02:32  lr: 0.001857  min_lr: 0.001857  loss: 3.6197 (3.3957)  weight_decay: 0.0500 (0.0500)  time: 0.6895  data: 0.0327  max mem: 41794
Epoch: [166]  [130/312]  eta: 0:02:26  lr: 0.001857  min_lr: 0.001857  loss: 3.6291 (3.3996)  weight_decay: 0.0500 (0.0500)  time: 0.8039  data: 0.0507  max mem: 41794
Epoch: [166]  [140/312]  eta: 0:02:16  lr: 0.001856  min_lr: 0.001856  loss: 3.6291 (3.4159)  weight_decay: 0.0500 (0.0500)  time: 0.7783  data: 0.0236  max mem: 41794
Epoch: [166]  [150/312]  eta: 0:02:07  lr: 0.001855  min_lr: 0.001855  loss: 3.5518 (3.4196)  weight_decay: 0.0500 (0.0500)  time: 0.7006  data: 0.0325  max mem: 41794
Epoch: [166]  [160/312]  eta: 0:01:59  lr: 0.001855  min_lr: 0.001855  loss: 3.4884 (3.4163)  weight_decay: 0.0500 (0.0500)  time: 0.7501  data: 0.0600  max mem: 41794
Epoch: [166]  [170/312]  eta: 0:01:50  lr: 0.001854  min_lr: 0.001854  loss: 3.4741 (3.4224)  weight_decay: 0.0500 (0.0500)  time: 0.7212  data: 0.0492  max mem: 41794
Epoch: [166]  [180/312]  eta: 0:01:43  lr: 0.001853  min_lr: 0.001853  loss: 3.4741 (3.4193)  weight_decay: 0.0500 (0.0500)  time: 0.7461  data: 0.0770  max mem: 41794
Epoch: [166]  [190/312]  eta: 0:01:34  lr: 0.001852  min_lr: 0.001852  loss: 3.5627 (3.4156)  weight_decay: 0.0500 (0.0500)  time: 0.7204  data: 0.0860  max mem: 41794
Epoch: [166]  [200/312]  eta: 0:01:26  lr: 0.001852  min_lr: 0.001852  loss: 3.2238 (3.4099)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.0886  max mem: 41794
Epoch: [166]  [210/312]  eta: 0:01:19  lr: 0.001851  min_lr: 0.001851  loss: 3.2238 (3.4095)  weight_decay: 0.0500 (0.0500)  time: 0.8266  data: 0.0854  max mem: 41794
Epoch: [166]  [220/312]  eta: 0:01:10  lr: 0.001850  min_lr: 0.001850  loss: 3.3858 (3.4103)  weight_decay: 0.0500 (0.0500)  time: 0.7033  data: 0.0480  max mem: 41794
Epoch: [166]  [230/312]  eta: 0:01:03  lr: 0.001849  min_lr: 0.001849  loss: 3.6136 (3.4111)  weight_decay: 0.0500 (0.0500)  time: 0.7172  data: 0.0515  max mem: 41794
Epoch: [166]  [240/312]  eta: 0:00:55  lr: 0.001849  min_lr: 0.001849  loss: 3.4135 (3.4036)  weight_decay: 0.0500 (0.0500)  time: 0.7983  data: 0.0324  max mem: 41794
Epoch: [166]  [250/312]  eta: 0:00:47  lr: 0.001848  min_lr: 0.001848  loss: 3.5513 (3.4127)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.0191  max mem: 41794
Epoch: [166]  [260/312]  eta: 0:00:39  lr: 0.001847  min_lr: 0.001847  loss: 3.5216 (3.3992)  weight_decay: 0.0500 (0.0500)  time: 0.7292  data: 0.0517  max mem: 41794
Epoch: [166]  [270/312]  eta: 0:00:31  lr: 0.001847  min_lr: 0.001847  loss: 3.2726 (3.3990)  weight_decay: 0.0500 (0.0500)  time: 0.6666  data: 0.0396  max mem: 41794
Epoch: [166]  [280/312]  eta: 0:00:24  lr: 0.001846  min_lr: 0.001846  loss: 3.4124 (3.3940)  weight_decay: 0.0500 (0.0500)  time: 0.7051  data: 0.0321  max mem: 41794
Epoch: [166]  [290/312]  eta: 0:00:16  lr: 0.001845  min_lr: 0.001845  loss: 3.5283 (3.3972)  weight_decay: 0.0500 (0.0500)  time: 0.8210  data: 0.0334  max mem: 41794
Epoch: [166]  [300/312]  eta: 0:00:09  lr: 0.001844  min_lr: 0.001844  loss: 3.5971 (3.3942)  weight_decay: 0.0500 (0.0500)  time: 0.6540  data: 0.0219  max mem: 41794
Epoch: [166]  [310/312]  eta: 0:00:01  lr: 0.001844  min_lr: 0.001844  loss: 3.5250 (3.3959)  weight_decay: 0.0500 (0.0500)  time: 0.5016  data: 0.0204  max mem: 41794
Epoch: [166]  [311/312]  eta: 0:00:00  lr: 0.001844  min_lr: 0.001844  loss: 3.5250 (3.3939)  weight_decay: 0.0500 (0.0500)  time: 0.4804  data: 0.0001  max mem: 41794
Epoch: [166] Total time: 0:03:55 (0.7533 s / it)
Averaged stats: lr: 0.001844  min_lr: 0.001844  loss: 3.5250 (3.3810)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.1212 (1.1212)  acc1: 81.1198 (81.1198)  acc5: 94.6615 (94.6615)  time: 8.5153  data: 8.3025  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3633 (1.3226)  acc1: 74.6094 (73.9200)  acc5: 92.9688 (92.2080)  time: 1.0941  data: 0.9226  max mem: 41794
Test: Total time: 0:00:10 (1.1149 s / it)
* Acc@1 73.964 Acc@5 92.330 loss 1.328
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 73.96%
Epoch: [167]  [  0/312]  eta: 1:21:40  lr: 0.001844  min_lr: 0.001844  loss: 3.4935 (3.4935)  weight_decay: 0.0500 (0.0500)  time: 15.7079  data: 13.5001  max mem: 41794
Epoch: [167]  [ 10/312]  eta: 0:11:09  lr: 0.001843  min_lr: 0.001843  loss: 3.4935 (3.4435)  weight_decay: 0.0500 (0.0500)  time: 2.2169  data: 1.2972  max mem: 41794
Epoch: [167]  [ 20/312]  eta: 0:07:27  lr: 0.001842  min_lr: 0.001842  loss: 3.4851 (3.5391)  weight_decay: 0.0500 (0.0500)  time: 0.8228  data: 0.0421  max mem: 41794
Epoch: [167]  [ 30/312]  eta: 0:05:40  lr: 0.001841  min_lr: 0.001841  loss: 3.6818 (3.5895)  weight_decay: 0.0500 (0.0500)  time: 0.6513  data: 0.0096  max mem: 41794
Epoch: [167]  [ 40/312]  eta: 0:04:50  lr: 0.001841  min_lr: 0.001841  loss: 3.6503 (3.5588)  weight_decay: 0.0500 (0.0500)  time: 0.5848  data: 0.0459  max mem: 41794
Epoch: [167]  [ 50/312]  eta: 0:04:28  lr: 0.001840  min_lr: 0.001840  loss: 3.4992 (3.5486)  weight_decay: 0.0500 (0.0500)  time: 0.7453  data: 0.1520  max mem: 41794
Epoch: [167]  [ 60/312]  eta: 0:03:56  lr: 0.001839  min_lr: 0.001839  loss: 3.5283 (3.5527)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1124  max mem: 41794
Epoch: [167]  [ 70/312]  eta: 0:03:45  lr: 0.001839  min_lr: 0.001839  loss: 3.4950 (3.5406)  weight_decay: 0.0500 (0.0500)  time: 0.6968  data: 0.1427  max mem: 41794
Epoch: [167]  [ 80/312]  eta: 0:03:32  lr: 0.001838  min_lr: 0.001838  loss: 3.3405 (3.5051)  weight_decay: 0.0500 (0.0500)  time: 0.8351  data: 0.2763  max mem: 41794
Epoch: [167]  [ 90/312]  eta: 0:03:12  lr: 0.001837  min_lr: 0.001837  loss: 3.3030 (3.4786)  weight_decay: 0.0500 (0.0500)  time: 0.6378  data: 0.1400  max mem: 41794
Epoch: [167]  [100/312]  eta: 0:03:04  lr: 0.001836  min_lr: 0.001836  loss: 3.3548 (3.4667)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.1878  max mem: 41794
Epoch: [167]  [110/312]  eta: 0:02:48  lr: 0.001836  min_lr: 0.001836  loss: 3.5220 (3.4574)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.1820  max mem: 41794
Epoch: [167]  [120/312]  eta: 0:02:40  lr: 0.001835  min_lr: 0.001835  loss: 3.5735 (3.4632)  weight_decay: 0.0500 (0.0500)  time: 0.6672  data: 0.1663  max mem: 41794
Epoch: [167]  [130/312]  eta: 0:02:32  lr: 0.001834  min_lr: 0.001834  loss: 3.5735 (3.4704)  weight_decay: 0.0500 (0.0500)  time: 0.8624  data: 0.3017  max mem: 41794
Epoch: [167]  [140/312]  eta: 0:02:19  lr: 0.001834  min_lr: 0.001834  loss: 3.6251 (3.4712)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1360  max mem: 41794
Epoch: [167]  [150/312]  eta: 0:02:12  lr: 0.001833  min_lr: 0.001833  loss: 3.6368 (3.4779)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.1336  max mem: 41794
Epoch: [167]  [160/312]  eta: 0:02:04  lr: 0.001832  min_lr: 0.001832  loss: 3.5266 (3.4564)  weight_decay: 0.0500 (0.0500)  time: 0.8876  data: 0.2684  max mem: 41794
Epoch: [167]  [170/312]  eta: 0:01:54  lr: 0.001831  min_lr: 0.001831  loss: 3.3035 (3.4482)  weight_decay: 0.0500 (0.0500)  time: 0.7089  data: 0.1368  max mem: 41794
Epoch: [167]  [180/312]  eta: 0:01:46  lr: 0.001831  min_lr: 0.001831  loss: 3.3158 (3.4455)  weight_decay: 0.0500 (0.0500)  time: 0.6784  data: 0.0985  max mem: 41794
Epoch: [167]  [190/312]  eta: 0:01:36  lr: 0.001830  min_lr: 0.001830  loss: 3.4186 (3.4384)  weight_decay: 0.0500 (0.0500)  time: 0.6762  data: 0.0971  max mem: 41794
Epoch: [167]  [200/312]  eta: 0:01:29  lr: 0.001829  min_lr: 0.001829  loss: 3.4363 (3.4394)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.1121  max mem: 41794
Epoch: [167]  [210/312]  eta: 0:01:21  lr: 0.001829  min_lr: 0.001829  loss: 3.5891 (3.4400)  weight_decay: 0.0500 (0.0500)  time: 0.8548  data: 0.2183  max mem: 41794
Epoch: [167]  [220/312]  eta: 0:01:11  lr: 0.001828  min_lr: 0.001828  loss: 3.4885 (3.4415)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.1068  max mem: 41794
Epoch: [167]  [230/312]  eta: 0:01:04  lr: 0.001827  min_lr: 0.001827  loss: 3.5501 (3.4471)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.1210  max mem: 41794
Epoch: [167]  [240/312]  eta: 0:00:57  lr: 0.001826  min_lr: 0.001826  loss: 3.5148 (3.4343)  weight_decay: 0.0500 (0.0500)  time: 0.8969  data: 0.2398  max mem: 41794
Epoch: [167]  [250/312]  eta: 0:00:48  lr: 0.001826  min_lr: 0.001826  loss: 2.9667 (3.4207)  weight_decay: 0.0500 (0.0500)  time: 0.7068  data: 0.1203  max mem: 41794
Epoch: [167]  [260/312]  eta: 0:00:40  lr: 0.001825  min_lr: 0.001825  loss: 3.5064 (3.4234)  weight_decay: 0.0500 (0.0500)  time: 0.6632  data: 0.0883  max mem: 41794
Epoch: [167]  [270/312]  eta: 0:00:32  lr: 0.001824  min_lr: 0.001824  loss: 3.5145 (3.4252)  weight_decay: 0.0500 (0.0500)  time: 0.6560  data: 0.0875  max mem: 41794
Epoch: [167]  [280/312]  eta: 0:00:24  lr: 0.001824  min_lr: 0.001824  loss: 3.5585 (3.4314)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.0960  max mem: 41794
Epoch: [167]  [290/312]  eta: 0:00:17  lr: 0.001823  min_lr: 0.001823  loss: 3.5972 (3.4356)  weight_decay: 0.0500 (0.0500)  time: 0.8501  data: 0.1817  max mem: 41794
Epoch: [167]  [300/312]  eta: 0:00:09  lr: 0.001822  min_lr: 0.001822  loss: 3.5972 (3.4315)  weight_decay: 0.0500 (0.0500)  time: 0.6302  data: 0.0861  max mem: 41794
Epoch: [167]  [310/312]  eta: 0:00:01  lr: 0.001821  min_lr: 0.001821  loss: 3.5977 (3.4394)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [167]  [311/312]  eta: 0:00:00  lr: 0.001821  min_lr: 0.001821  loss: 3.5977 (3.4398)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [167] Total time: 0:03:57 (0.7620 s / it)
Averaged stats: lr: 0.001821  min_lr: 0.001821  loss: 3.5977 (3.4016)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0715 (1.0715)  acc1: 78.5156 (78.5156)  acc5: 94.4010 (94.4010)  time: 8.8555  data: 8.6501  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4503 (1.3245)  acc1: 72.6562 (72.8800)  acc5: 90.7552 (91.7280)  time: 1.1345  data: 0.9612  max mem: 41794
Test: Total time: 0:00:10 (1.1610 s / it)
* Acc@1 73.452 Acc@5 91.846 loss 1.314
Accuracy of the model on the 50000 test images: 73.5%
Max accuracy: 73.96%
Epoch: [168]  [  0/312]  eta: 1:22:50  lr: 0.001821  min_lr: 0.001821  loss: 3.4577 (3.4577)  weight_decay: 0.0500 (0.0500)  time: 15.9305  data: 13.0515  max mem: 41794
Epoch: [168]  [ 10/312]  eta: 0:11:56  lr: 0.001821  min_lr: 0.001821  loss: 3.4697 (3.3634)  weight_decay: 0.0500 (0.0500)  time: 2.3715  data: 1.3532  max mem: 41794
Epoch: [168]  [ 20/312]  eta: 0:08:01  lr: 0.001820  min_lr: 0.001820  loss: 3.4225 (3.3023)  weight_decay: 0.0500 (0.0500)  time: 0.9345  data: 0.1563  max mem: 41794
Epoch: [168]  [ 30/312]  eta: 0:06:05  lr: 0.001819  min_lr: 0.001819  loss: 3.4291 (3.3460)  weight_decay: 0.0500 (0.0500)  time: 0.7059  data: 0.0718  max mem: 41794
Epoch: [168]  [ 40/312]  eta: 0:04:59  lr: 0.001818  min_lr: 0.001818  loss: 3.4515 (3.3497)  weight_decay: 0.0500 (0.0500)  time: 0.5292  data: 0.0075  max mem: 41794
Epoch: [168]  [ 50/312]  eta: 0:04:19  lr: 0.001818  min_lr: 0.001818  loss: 3.6040 (3.3969)  weight_decay: 0.0500 (0.0500)  time: 0.5135  data: 0.0127  max mem: 41794
Epoch: [168]  [ 60/312]  eta: 0:03:49  lr: 0.001817  min_lr: 0.001817  loss: 3.6141 (3.4049)  weight_decay: 0.0500 (0.0500)  time: 0.5214  data: 0.0128  max mem: 41794
Epoch: [168]  [ 70/312]  eta: 0:03:35  lr: 0.001816  min_lr: 0.001816  loss: 3.3715 (3.3616)  weight_decay: 0.0500 (0.0500)  time: 0.6328  data: 0.1213  max mem: 41794
Epoch: [168]  [ 80/312]  eta: 0:03:23  lr: 0.001816  min_lr: 0.001816  loss: 3.2281 (3.3687)  weight_decay: 0.0500 (0.0500)  time: 0.7745  data: 0.2746  max mem: 41794
Epoch: [168]  [ 90/312]  eta: 0:03:06  lr: 0.001815  min_lr: 0.001815  loss: 3.1494 (3.3512)  weight_decay: 0.0500 (0.0500)  time: 0.6616  data: 0.1540  max mem: 41794
Epoch: [168]  [100/312]  eta: 0:02:56  lr: 0.001814  min_lr: 0.001814  loss: 3.3562 (3.3566)  weight_decay: 0.0500 (0.0500)  time: 0.6574  data: 0.1497  max mem: 41794
Epoch: [168]  [110/312]  eta: 0:02:42  lr: 0.001813  min_lr: 0.001813  loss: 3.5453 (3.3581)  weight_decay: 0.0500 (0.0500)  time: 0.6534  data: 0.1497  max mem: 41794
Epoch: [168]  [120/312]  eta: 0:02:35  lr: 0.001813  min_lr: 0.001813  loss: 3.5117 (3.3656)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1695  max mem: 41794
Epoch: [168]  [130/312]  eta: 0:02:28  lr: 0.001812  min_lr: 0.001812  loss: 3.5636 (3.3872)  weight_decay: 0.0500 (0.0500)  time: 0.8601  data: 0.3541  max mem: 41794
Epoch: [168]  [140/312]  eta: 0:02:16  lr: 0.001811  min_lr: 0.001811  loss: 3.6611 (3.4016)  weight_decay: 0.0500 (0.0500)  time: 0.6979  data: 0.1928  max mem: 41794
Epoch: [168]  [150/312]  eta: 0:02:09  lr: 0.001811  min_lr: 0.001811  loss: 3.6238 (3.4039)  weight_decay: 0.0500 (0.0500)  time: 0.6903  data: 0.1505  max mem: 41794
Epoch: [168]  [160/312]  eta: 0:02:01  lr: 0.001810  min_lr: 0.001810  loss: 3.6465 (3.4155)  weight_decay: 0.0500 (0.0500)  time: 0.8258  data: 0.2928  max mem: 41794
Epoch: [168]  [170/312]  eta: 0:01:51  lr: 0.001809  min_lr: 0.001809  loss: 3.5622 (3.4232)  weight_decay: 0.0500 (0.0500)  time: 0.6642  data: 0.1505  max mem: 41794
Epoch: [168]  [180/312]  eta: 0:01:43  lr: 0.001808  min_lr: 0.001808  loss: 3.5146 (3.4139)  weight_decay: 0.0500 (0.0500)  time: 0.6839  data: 0.1709  max mem: 41794
Epoch: [168]  [190/312]  eta: 0:01:34  lr: 0.001808  min_lr: 0.001808  loss: 3.4953 (3.4118)  weight_decay: 0.0500 (0.0500)  time: 0.6731  data: 0.1710  max mem: 41794
Epoch: [168]  [200/312]  eta: 0:01:26  lr: 0.001807  min_lr: 0.001807  loss: 3.5679 (3.4190)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1893  max mem: 41794
Epoch: [168]  [210/312]  eta: 0:01:19  lr: 0.001806  min_lr: 0.001806  loss: 3.4157 (3.4084)  weight_decay: 0.0500 (0.0500)  time: 0.8858  data: 0.3806  max mem: 41794
Epoch: [168]  [220/312]  eta: 0:01:10  lr: 0.001806  min_lr: 0.001806  loss: 3.3794 (3.4096)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.1919  max mem: 41794
Epoch: [168]  [230/312]  eta: 0:01:03  lr: 0.001805  min_lr: 0.001805  loss: 3.4154 (3.4050)  weight_decay: 0.0500 (0.0500)  time: 0.6973  data: 0.1633  max mem: 41794
Epoch: [168]  [240/312]  eta: 0:00:55  lr: 0.001804  min_lr: 0.001804  loss: 3.1869 (3.3952)  weight_decay: 0.0500 (0.0500)  time: 0.8565  data: 0.3081  max mem: 41794
Epoch: [168]  [250/312]  eta: 0:00:47  lr: 0.001803  min_lr: 0.001803  loss: 3.0477 (3.3813)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.1460  max mem: 41794
Epoch: [168]  [260/312]  eta: 0:00:40  lr: 0.001803  min_lr: 0.001803  loss: 3.0477 (3.3703)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.1523  max mem: 41794
Epoch: [168]  [270/312]  eta: 0:00:31  lr: 0.001802  min_lr: 0.001802  loss: 3.2428 (3.3672)  weight_decay: 0.0500 (0.0500)  time: 0.6777  data: 0.1558  max mem: 41794
Epoch: [168]  [280/312]  eta: 0:00:24  lr: 0.001801  min_lr: 0.001801  loss: 3.5838 (3.3784)  weight_decay: 0.0500 (0.0500)  time: 0.6880  data: 0.1701  max mem: 41794
Epoch: [168]  [290/312]  eta: 0:00:16  lr: 0.001801  min_lr: 0.001801  loss: 3.5838 (3.3771)  weight_decay: 0.0500 (0.0500)  time: 0.8849  data: 0.3398  max mem: 41794
Epoch: [168]  [300/312]  eta: 0:00:09  lr: 0.001800  min_lr: 0.001800  loss: 3.5351 (3.3809)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.1742  max mem: 41794
Epoch: [168]  [310/312]  eta: 0:00:01  lr: 0.001799  min_lr: 0.001799  loss: 3.5150 (3.3820)  weight_decay: 0.0500 (0.0500)  time: 0.4732  data: 0.0001  max mem: 41794
Epoch: [168]  [311/312]  eta: 0:00:00  lr: 0.001799  min_lr: 0.001799  loss: 3.5336 (3.3825)  weight_decay: 0.0500 (0.0500)  time: 0.4734  data: 0.0001  max mem: 41794
Epoch: [168] Total time: 0:03:55 (0.7563 s / it)
Averaged stats: lr: 0.001799  min_lr: 0.001799  loss: 3.5336 (3.3611)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1982 (1.1982)  acc1: 81.1198 (81.1198)  acc5: 94.7917 (94.7917)  time: 8.2483  data: 8.0354  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.5161 (1.4510)  acc1: 73.3073 (73.8080)  acc5: 91.2760 (91.9520)  time: 1.0955  data: 0.9216  max mem: 41794
Test: Total time: 0:00:09 (1.1063 s / it)
* Acc@1 73.484 Acc@5 91.982 loss 1.460
Accuracy of the model on the 50000 test images: 73.5%
Max accuracy: 73.96%
Epoch: [169]  [  0/312]  eta: 1:16:37  lr: 0.001799  min_lr: 0.001799  loss: 3.5584 (3.5584)  weight_decay: 0.0500 (0.0500)  time: 14.7344  data: 11.8766  max mem: 41794
Epoch: [169]  [ 10/312]  eta: 0:11:32  lr: 0.001798  min_lr: 0.001798  loss: 3.4553 (3.2423)  weight_decay: 0.0500 (0.0500)  time: 2.2944  data: 1.3023  max mem: 41794
Epoch: [169]  [ 20/312]  eta: 0:07:16  lr: 0.001798  min_lr: 0.001798  loss: 3.4553 (3.2960)  weight_decay: 0.0500 (0.0500)  time: 0.8346  data: 0.1283  max mem: 41794
Epoch: [169]  [ 30/312]  eta: 0:05:32  lr: 0.001797  min_lr: 0.001797  loss: 3.6360 (3.4112)  weight_decay: 0.0500 (0.0500)  time: 0.5660  data: 0.0082  max mem: 41794
Epoch: [169]  [ 40/312]  eta: 0:04:39  lr: 0.001796  min_lr: 0.001796  loss: 3.4205 (3.3891)  weight_decay: 0.0500 (0.0500)  time: 0.5333  data: 0.0027  max mem: 41794
Epoch: [169]  [ 50/312]  eta: 0:04:18  lr: 0.001795  min_lr: 0.001795  loss: 3.1766 (3.3067)  weight_decay: 0.0500 (0.0500)  time: 0.6870  data: 0.0450  max mem: 41794
Epoch: [169]  [ 60/312]  eta: 0:03:50  lr: 0.001795  min_lr: 0.001795  loss: 3.0873 (3.2970)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0611  max mem: 41794
Epoch: [169]  [ 70/312]  eta: 0:03:39  lr: 0.001794  min_lr: 0.001794  loss: 3.4880 (3.3292)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.0682  max mem: 41794
Epoch: [169]  [ 80/312]  eta: 0:03:27  lr: 0.001793  min_lr: 0.001793  loss: 3.4880 (3.3289)  weight_decay: 0.0500 (0.0500)  time: 0.8348  data: 0.0521  max mem: 41794
Epoch: [169]  [ 90/312]  eta: 0:03:11  lr: 0.001792  min_lr: 0.001792  loss: 3.2310 (3.3119)  weight_decay: 0.0500 (0.0500)  time: 0.7030  data: 0.0461  max mem: 41794
Epoch: [169]  [100/312]  eta: 0:03:03  lr: 0.001792  min_lr: 0.001792  loss: 3.1812 (3.3146)  weight_decay: 0.0500 (0.0500)  time: 0.7544  data: 0.1125  max mem: 41794
Epoch: [169]  [110/312]  eta: 0:02:48  lr: 0.001791  min_lr: 0.001791  loss: 3.1620 (3.3120)  weight_decay: 0.0500 (0.0500)  time: 0.6960  data: 0.0671  max mem: 41794
Epoch: [169]  [120/312]  eta: 0:02:41  lr: 0.001790  min_lr: 0.001790  loss: 3.1800 (3.2984)  weight_decay: 0.0500 (0.0500)  time: 0.7098  data: 0.0714  max mem: 41794
Epoch: [169]  [130/312]  eta: 0:02:33  lr: 0.001790  min_lr: 0.001790  loss: 3.5126 (3.3185)  weight_decay: 0.0500 (0.0500)  time: 0.9017  data: 0.1237  max mem: 41794
Epoch: [169]  [140/312]  eta: 0:02:21  lr: 0.001789  min_lr: 0.001789  loss: 3.5224 (3.3223)  weight_decay: 0.0500 (0.0500)  time: 0.7167  data: 0.0545  max mem: 41794
Epoch: [169]  [150/312]  eta: 0:02:12  lr: 0.001788  min_lr: 0.001788  loss: 3.5571 (3.3289)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.0504  max mem: 41794
Epoch: [169]  [160/312]  eta: 0:02:04  lr: 0.001787  min_lr: 0.001787  loss: 3.5857 (3.3385)  weight_decay: 0.0500 (0.0500)  time: 0.7891  data: 0.0487  max mem: 41794
Epoch: [169]  [170/312]  eta: 0:01:54  lr: 0.001787  min_lr: 0.001787  loss: 3.3728 (3.3263)  weight_decay: 0.0500 (0.0500)  time: 0.7105  data: 0.0594  max mem: 41794
Epoch: [169]  [180/312]  eta: 0:01:46  lr: 0.001786  min_lr: 0.001786  loss: 3.1311 (3.3107)  weight_decay: 0.0500 (0.0500)  time: 0.7306  data: 0.0953  max mem: 41794
Epoch: [169]  [190/312]  eta: 0:01:36  lr: 0.001785  min_lr: 0.001785  loss: 3.2559 (3.3178)  weight_decay: 0.0500 (0.0500)  time: 0.6689  data: 0.0379  max mem: 41794
Epoch: [169]  [200/312]  eta: 0:01:29  lr: 0.001785  min_lr: 0.001785  loss: 3.5514 (3.3268)  weight_decay: 0.0500 (0.0500)  time: 0.6824  data: 0.0399  max mem: 41794
Epoch: [169]  [210/312]  eta: 0:01:21  lr: 0.001784  min_lr: 0.001784  loss: 3.4953 (3.3307)  weight_decay: 0.0500 (0.0500)  time: 0.8177  data: 0.0685  max mem: 41794
Epoch: [169]  [220/312]  eta: 0:01:12  lr: 0.001783  min_lr: 0.001783  loss: 3.4792 (3.3363)  weight_decay: 0.0500 (0.0500)  time: 0.6580  data: 0.0317  max mem: 41794
Epoch: [169]  [230/312]  eta: 0:01:04  lr: 0.001782  min_lr: 0.001782  loss: 3.4108 (3.3283)  weight_decay: 0.0500 (0.0500)  time: 0.7196  data: 0.0538  max mem: 41794
Epoch: [169]  [240/312]  eta: 0:00:56  lr: 0.001782  min_lr: 0.001782  loss: 3.2718 (3.3266)  weight_decay: 0.0500 (0.0500)  time: 0.8071  data: 0.0553  max mem: 41794
Epoch: [169]  [250/312]  eta: 0:00:48  lr: 0.001781  min_lr: 0.001781  loss: 3.4520 (3.3275)  weight_decay: 0.0500 (0.0500)  time: 0.6553  data: 0.0342  max mem: 41794
Epoch: [169]  [260/312]  eta: 0:00:40  lr: 0.001780  min_lr: 0.001780  loss: 3.6629 (3.3334)  weight_decay: 0.0500 (0.0500)  time: 0.7310  data: 0.0818  max mem: 41794
Epoch: [169]  [270/312]  eta: 0:00:32  lr: 0.001780  min_lr: 0.001780  loss: 3.6605 (3.3371)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.0509  max mem: 41794
Epoch: [169]  [280/312]  eta: 0:00:24  lr: 0.001779  min_lr: 0.001779  loss: 3.4943 (3.3393)  weight_decay: 0.0500 (0.0500)  time: 0.6417  data: 0.0438  max mem: 41794
Epoch: [169]  [290/312]  eta: 0:00:17  lr: 0.001778  min_lr: 0.001778  loss: 3.4943 (3.3429)  weight_decay: 0.0500 (0.0500)  time: 0.8471  data: 0.1066  max mem: 41794
Epoch: [169]  [300/312]  eta: 0:00:09  lr: 0.001777  min_lr: 0.001777  loss: 3.4691 (3.3467)  weight_decay: 0.0500 (0.0500)  time: 0.6964  data: 0.0633  max mem: 41794
Epoch: [169]  [310/312]  eta: 0:00:01  lr: 0.001777  min_lr: 0.001777  loss: 3.4908 (3.3533)  weight_decay: 0.0500 (0.0500)  time: 0.4680  data: 0.0001  max mem: 41794
Epoch: [169]  [311/312]  eta: 0:00:00  lr: 0.001777  min_lr: 0.001777  loss: 3.5777 (3.3546)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [169] Total time: 0:03:57 (0.7627 s / it)
Averaged stats: lr: 0.001777  min_lr: 0.001777  loss: 3.5777 (3.3664)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1111 (1.1111)  acc1: 80.8594 (80.8594)  acc5: 95.9635 (95.9635)  time: 8.2646  data: 8.0625  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4115 (1.3843)  acc1: 72.0052 (73.6320)  acc5: 92.9688 (92.1120)  time: 1.0860  data: 0.9136  max mem: 41794
Test: Total time: 0:00:09 (1.0955 s / it)
* Acc@1 73.526 Acc@5 91.984 loss 1.383
Accuracy of the model on the 50000 test images: 73.5%
Max accuracy: 73.96%
Epoch: [170]  [  0/312]  eta: 1:14:12  lr: 0.001777  min_lr: 0.001777  loss: 2.3037 (2.3037)  weight_decay: 0.0500 (0.0500)  time: 14.2706  data: 13.7932  max mem: 41794
Epoch: [170]  [ 10/312]  eta: 0:11:29  lr: 0.001776  min_lr: 0.001776  loss: 3.4617 (3.2322)  weight_decay: 0.0500 (0.0500)  time: 2.2826  data: 1.2575  max mem: 41794
Epoch: [170]  [ 20/312]  eta: 0:07:18  lr: 0.001775  min_lr: 0.001775  loss: 3.5604 (3.3538)  weight_decay: 0.0500 (0.0500)  time: 0.8645  data: 0.0715  max mem: 41794
Epoch: [170]  [ 30/312]  eta: 0:05:35  lr: 0.001774  min_lr: 0.001774  loss: 3.4526 (3.3113)  weight_decay: 0.0500 (0.0500)  time: 0.5892  data: 0.0716  max mem: 41794
Epoch: [170]  [ 40/312]  eta: 0:04:39  lr: 0.001774  min_lr: 0.001774  loss: 3.3612 (3.3207)  weight_decay: 0.0500 (0.0500)  time: 0.5278  data: 0.0115  max mem: 41794
Epoch: [170]  [ 50/312]  eta: 0:04:21  lr: 0.001773  min_lr: 0.001773  loss: 3.3906 (3.3063)  weight_decay: 0.0500 (0.0500)  time: 0.6967  data: 0.1068  max mem: 41794
Epoch: [170]  [ 60/312]  eta: 0:03:52  lr: 0.001772  min_lr: 0.001772  loss: 3.5670 (3.3287)  weight_decay: 0.0500 (0.0500)  time: 0.7045  data: 0.0977  max mem: 41794
Epoch: [170]  [ 70/312]  eta: 0:03:43  lr: 0.001772  min_lr: 0.001772  loss: 3.5670 (3.3249)  weight_decay: 0.0500 (0.0500)  time: 0.7408  data: 0.1158  max mem: 41794
Epoch: [170]  [ 80/312]  eta: 0:03:30  lr: 0.001771  min_lr: 0.001771  loss: 3.6073 (3.3587)  weight_decay: 0.0500 (0.0500)  time: 0.8574  data: 0.2094  max mem: 41794
Epoch: [170]  [ 90/312]  eta: 0:03:12  lr: 0.001770  min_lr: 0.001770  loss: 3.5307 (3.3453)  weight_decay: 0.0500 (0.0500)  time: 0.6660  data: 0.0942  max mem: 41794
Epoch: [170]  [100/312]  eta: 0:03:04  lr: 0.001769  min_lr: 0.001769  loss: 3.2928 (3.3514)  weight_decay: 0.0500 (0.0500)  time: 0.7248  data: 0.1225  max mem: 41794
Epoch: [170]  [110/312]  eta: 0:02:48  lr: 0.001769  min_lr: 0.001769  loss: 3.2928 (3.3538)  weight_decay: 0.0500 (0.0500)  time: 0.6946  data: 0.1224  max mem: 41794
Epoch: [170]  [120/312]  eta: 0:02:40  lr: 0.001768  min_lr: 0.001768  loss: 3.5747 (3.3770)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.0966  max mem: 41794
Epoch: [170]  [130/312]  eta: 0:02:33  lr: 0.001767  min_lr: 0.001767  loss: 3.5951 (3.3632)  weight_decay: 0.0500 (0.0500)  time: 0.8674  data: 0.1938  max mem: 41794
Epoch: [170]  [140/312]  eta: 0:02:21  lr: 0.001767  min_lr: 0.001767  loss: 3.1443 (3.3450)  weight_decay: 0.0500 (0.0500)  time: 0.7155  data: 0.0978  max mem: 41794
Epoch: [170]  [150/312]  eta: 0:02:12  lr: 0.001766  min_lr: 0.001766  loss: 3.4845 (3.3538)  weight_decay: 0.0500 (0.0500)  time: 0.6436  data: 0.0721  max mem: 41794
Epoch: [170]  [160/312]  eta: 0:02:03  lr: 0.001765  min_lr: 0.001765  loss: 3.4845 (3.3445)  weight_decay: 0.0500 (0.0500)  time: 0.7459  data: 0.1772  max mem: 41794
Epoch: [170]  [170/312]  eta: 0:01:53  lr: 0.001764  min_lr: 0.001764  loss: 3.1869 (3.3314)  weight_decay: 0.0500 (0.0500)  time: 0.7075  data: 0.1057  max mem: 41794
Epoch: [170]  [180/312]  eta: 0:01:46  lr: 0.001764  min_lr: 0.001764  loss: 3.3416 (3.3332)  weight_decay: 0.0500 (0.0500)  time: 0.7394  data: 0.0927  max mem: 41794
Epoch: [170]  [190/312]  eta: 0:01:36  lr: 0.001763  min_lr: 0.001763  loss: 3.6426 (3.3377)  weight_decay: 0.0500 (0.0500)  time: 0.6607  data: 0.0927  max mem: 41794
Epoch: [170]  [200/312]  eta: 0:01:28  lr: 0.001762  min_lr: 0.001762  loss: 3.3283 (3.3301)  weight_decay: 0.0500 (0.0500)  time: 0.7110  data: 0.1177  max mem: 41794
Epoch: [170]  [210/312]  eta: 0:01:21  lr: 0.001762  min_lr: 0.001762  loss: 3.2661 (3.3291)  weight_decay: 0.0500 (0.0500)  time: 0.8690  data: 0.2036  max mem: 41794
Epoch: [170]  [220/312]  eta: 0:01:12  lr: 0.001761  min_lr: 0.001761  loss: 3.6003 (3.3302)  weight_decay: 0.0500 (0.0500)  time: 0.6683  data: 0.0866  max mem: 41794
Epoch: [170]  [230/312]  eta: 0:01:04  lr: 0.001760  min_lr: 0.001760  loss: 3.2941 (3.3208)  weight_decay: 0.0500 (0.0500)  time: 0.7143  data: 0.0982  max mem: 41794
Epoch: [170]  [240/312]  eta: 0:00:56  lr: 0.001759  min_lr: 0.001759  loss: 3.4535 (3.3321)  weight_decay: 0.0500 (0.0500)  time: 0.8500  data: 0.1761  max mem: 41794
Epoch: [170]  [250/312]  eta: 0:00:48  lr: 0.001759  min_lr: 0.001759  loss: 3.5612 (3.3295)  weight_decay: 0.0500 (0.0500)  time: 0.7054  data: 0.0785  max mem: 41794
Epoch: [170]  [260/312]  eta: 0:00:40  lr: 0.001758  min_lr: 0.001758  loss: 3.3951 (3.3304)  weight_decay: 0.0500 (0.0500)  time: 0.7188  data: 0.0721  max mem: 41794
Epoch: [170]  [270/312]  eta: 0:00:32  lr: 0.001757  min_lr: 0.001757  loss: 3.3951 (3.3308)  weight_decay: 0.0500 (0.0500)  time: 0.6593  data: 0.0727  max mem: 41794
Epoch: [170]  [280/312]  eta: 0:00:24  lr: 0.001757  min_lr: 0.001757  loss: 3.3797 (3.3248)  weight_decay: 0.0500 (0.0500)  time: 0.6939  data: 0.0883  max mem: 41794
Epoch: [170]  [290/312]  eta: 0:00:17  lr: 0.001756  min_lr: 0.001756  loss: 3.5835 (3.3276)  weight_decay: 0.0500 (0.0500)  time: 0.8146  data: 0.1953  max mem: 41794
Epoch: [170]  [300/312]  eta: 0:00:09  lr: 0.001755  min_lr: 0.001755  loss: 3.4962 (3.3303)  weight_decay: 0.0500 (0.0500)  time: 0.6191  data: 0.1080  max mem: 41794
Epoch: [170]  [310/312]  eta: 0:00:01  lr: 0.001754  min_lr: 0.001754  loss: 3.3132 (3.3285)  weight_decay: 0.0500 (0.0500)  time: 0.4782  data: 0.0001  max mem: 41794
Epoch: [170]  [311/312]  eta: 0:00:00  lr: 0.001754  min_lr: 0.001754  loss: 3.3132 (3.3261)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.0001  max mem: 41794
Epoch: [170] Total time: 0:03:57 (0.7618 s / it)
Averaged stats: lr: 0.001754  min_lr: 0.001754  loss: 3.3132 (3.3682)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.0300 (1.0300)  acc1: 80.4688 (80.4688)  acc5: 95.3125 (95.3125)  time: 8.9348  data: 8.7265  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3479 (1.2720)  acc1: 72.9167 (73.6320)  acc5: 91.5365 (92.3840)  time: 1.1412  data: 0.9697  max mem: 41794
Test: Total time: 0:00:10 (1.1528 s / it)
* Acc@1 73.970 Acc@5 92.390 loss 1.266
Accuracy of the model on the 50000 test images: 74.0%
Max accuracy: 73.97%
Epoch: [171]  [  0/312]  eta: 1:22:26  lr: 0.001754  min_lr: 0.001754  loss: 3.7644 (3.7644)  weight_decay: 0.0500 (0.0500)  time: 15.8539  data: 11.9598  max mem: 41794
Epoch: [171]  [ 10/312]  eta: 0:10:43  lr: 0.001754  min_lr: 0.001754  loss: 3.6963 (3.5519)  weight_decay: 0.0500 (0.0500)  time: 2.1318  data: 1.3186  max mem: 41794
Epoch: [171]  [ 20/312]  eta: 0:07:37  lr: 0.001753  min_lr: 0.001753  loss: 3.5039 (3.4805)  weight_decay: 0.0500 (0.0500)  time: 0.8540  data: 0.2281  max mem: 41794
Epoch: [171]  [ 30/312]  eta: 0:05:49  lr: 0.001752  min_lr: 0.001752  loss: 3.2690 (3.3332)  weight_decay: 0.0500 (0.0500)  time: 0.7482  data: 0.1032  max mem: 41794
Epoch: [171]  [ 40/312]  eta: 0:04:59  lr: 0.001751  min_lr: 0.001751  loss: 3.1521 (3.3068)  weight_decay: 0.0500 (0.0500)  time: 0.6068  data: 0.0103  max mem: 41794
Epoch: [171]  [ 50/312]  eta: 0:04:30  lr: 0.001751  min_lr: 0.001751  loss: 3.2700 (3.3097)  weight_decay: 0.0500 (0.0500)  time: 0.7168  data: 0.0439  max mem: 41794
Epoch: [171]  [ 60/312]  eta: 0:03:58  lr: 0.001750  min_lr: 0.001750  loss: 3.5265 (3.3310)  weight_decay: 0.0500 (0.0500)  time: 0.6378  data: 0.0378  max mem: 41794
Epoch: [171]  [ 70/312]  eta: 0:03:47  lr: 0.001749  min_lr: 0.001749  loss: 3.5265 (3.3315)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.0872  max mem: 41794
Epoch: [171]  [ 80/312]  eta: 0:03:31  lr: 0.001749  min_lr: 0.001749  loss: 3.2708 (3.3272)  weight_decay: 0.0500 (0.0500)  time: 0.7997  data: 0.0923  max mem: 41794
Epoch: [171]  [ 90/312]  eta: 0:03:15  lr: 0.001748  min_lr: 0.001748  loss: 3.4466 (3.3267)  weight_decay: 0.0500 (0.0500)  time: 0.6572  data: 0.0662  max mem: 41794
Epoch: [171]  [100/312]  eta: 0:03:05  lr: 0.001747  min_lr: 0.001747  loss: 3.5398 (3.3341)  weight_decay: 0.0500 (0.0500)  time: 0.7323  data: 0.1322  max mem: 41794
Epoch: [171]  [110/312]  eta: 0:02:49  lr: 0.001747  min_lr: 0.001747  loss: 3.1188 (3.3144)  weight_decay: 0.0500 (0.0500)  time: 0.6709  data: 0.0732  max mem: 41794
Epoch: [171]  [120/312]  eta: 0:02:43  lr: 0.001746  min_lr: 0.001746  loss: 3.2059 (3.3246)  weight_decay: 0.0500 (0.0500)  time: 0.7212  data: 0.1048  max mem: 41794
Epoch: [171]  [130/312]  eta: 0:02:36  lr: 0.001745  min_lr: 0.001745  loss: 3.3615 (3.3098)  weight_decay: 0.0500 (0.0500)  time: 0.9645  data: 0.1597  max mem: 41794
Epoch: [171]  [140/312]  eta: 0:02:24  lr: 0.001744  min_lr: 0.001744  loss: 3.3721 (3.3276)  weight_decay: 0.0500 (0.0500)  time: 0.7814  data: 0.0953  max mem: 41794
Epoch: [171]  [150/312]  eta: 0:02:15  lr: 0.001744  min_lr: 0.001744  loss: 3.7066 (3.3338)  weight_decay: 0.0500 (0.0500)  time: 0.6676  data: 0.0671  max mem: 41794
Epoch: [171]  [160/312]  eta: 0:02:04  lr: 0.001743  min_lr: 0.001743  loss: 3.4514 (3.3369)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0297  max mem: 41794
Epoch: [171]  [170/312]  eta: 0:01:55  lr: 0.001742  min_lr: 0.001742  loss: 3.3813 (3.3412)  weight_decay: 0.0500 (0.0500)  time: 0.6527  data: 0.0486  max mem: 41794
Epoch: [171]  [180/312]  eta: 0:01:47  lr: 0.001742  min_lr: 0.001742  loss: 3.5649 (3.3529)  weight_decay: 0.0500 (0.0500)  time: 0.7748  data: 0.1384  max mem: 41794
Epoch: [171]  [190/312]  eta: 0:01:37  lr: 0.001741  min_lr: 0.001741  loss: 3.5649 (3.3479)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.0927  max mem: 41794
Epoch: [171]  [200/312]  eta: 0:01:29  lr: 0.001740  min_lr: 0.001740  loss: 3.4572 (3.3551)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.0973  max mem: 41794
Epoch: [171]  [210/312]  eta: 0:01:22  lr: 0.001739  min_lr: 0.001739  loss: 3.5125 (3.3540)  weight_decay: 0.0500 (0.0500)  time: 0.8761  data: 0.1551  max mem: 41794
Epoch: [171]  [220/312]  eta: 0:01:12  lr: 0.001739  min_lr: 0.001739  loss: 3.5948 (3.3587)  weight_decay: 0.0500 (0.0500)  time: 0.7103  data: 0.0621  max mem: 41794
Epoch: [171]  [230/312]  eta: 0:01:05  lr: 0.001738  min_lr: 0.001738  loss: 3.5098 (3.3610)  weight_decay: 0.0500 (0.0500)  time: 0.7117  data: 0.0607  max mem: 41794
Epoch: [171]  [240/312]  eta: 0:00:57  lr: 0.001737  min_lr: 0.001737  loss: 3.6096 (3.3754)  weight_decay: 0.0500 (0.0500)  time: 0.7832  data: 0.0642  max mem: 41794
Epoch: [171]  [250/312]  eta: 0:00:48  lr: 0.001737  min_lr: 0.001737  loss: 3.6358 (3.3783)  weight_decay: 0.0500 (0.0500)  time: 0.6931  data: 0.0367  max mem: 41794
Epoch: [171]  [260/312]  eta: 0:00:41  lr: 0.001736  min_lr: 0.001736  loss: 3.4369 (3.3774)  weight_decay: 0.0500 (0.0500)  time: 0.8230  data: 0.0721  max mem: 41794
Epoch: [171]  [270/312]  eta: 0:00:32  lr: 0.001735  min_lr: 0.001735  loss: 3.3582 (3.3697)  weight_decay: 0.0500 (0.0500)  time: 0.7095  data: 0.0451  max mem: 41794
Epoch: [171]  [280/312]  eta: 0:00:25  lr: 0.001734  min_lr: 0.001734  loss: 3.3341 (3.3682)  weight_decay: 0.0500 (0.0500)  time: 0.6638  data: 0.0210  max mem: 41794
Epoch: [171]  [290/312]  eta: 0:00:17  lr: 0.001734  min_lr: 0.001734  loss: 3.2390 (3.3577)  weight_decay: 0.0500 (0.0500)  time: 0.7682  data: 0.0265  max mem: 41794
Epoch: [171]  [300/312]  eta: 0:00:09  lr: 0.001733  min_lr: 0.001733  loss: 3.3492 (3.3605)  weight_decay: 0.0500 (0.0500)  time: 0.5920  data: 0.0078  max mem: 41794
Epoch: [171]  [310/312]  eta: 0:00:01  lr: 0.001732  min_lr: 0.001732  loss: 3.2631 (3.3494)  weight_decay: 0.0500 (0.0500)  time: 0.4629  data: 0.0001  max mem: 41794
Epoch: [171]  [311/312]  eta: 0:00:00  lr: 0.001732  min_lr: 0.001732  loss: 3.2064 (3.3490)  weight_decay: 0.0500 (0.0500)  time: 0.4619  data: 0.0001  max mem: 41794
Epoch: [171] Total time: 0:03:59 (0.7680 s / it)
Averaged stats: lr: 0.001732  min_lr: 0.001732  loss: 3.2064 (3.3749)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.0348 (1.0348)  acc1: 80.3385 (80.3385)  acc5: 95.5729 (95.5729)  time: 8.6543  data: 8.4541  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3344 (1.2832)  acc1: 74.3490 (74.0960)  acc5: 92.1875 (92.4480)  time: 1.1131  data: 0.9394  max mem: 41794
Test: Total time: 0:00:10 (1.1485 s / it)
* Acc@1 74.476 Acc@5 92.476 loss 1.278
Accuracy of the model on the 50000 test images: 74.5%
Max accuracy: 74.48%
Epoch: [172]  [  0/312]  eta: 1:19:38  lr: 0.001732  min_lr: 0.001732  loss: 3.4589 (3.4589)  weight_decay: 0.0500 (0.0500)  time: 15.3142  data: 13.0722  max mem: 41794
Epoch: [172]  [ 10/312]  eta: 0:10:45  lr: 0.001731  min_lr: 0.001731  loss: 3.3846 (3.2292)  weight_decay: 0.0500 (0.0500)  time: 2.1380  data: 1.3739  max mem: 41794
Epoch: [172]  [ 20/312]  eta: 0:07:18  lr: 0.001731  min_lr: 0.001731  loss: 3.4686 (3.3422)  weight_decay: 0.0500 (0.0500)  time: 0.8125  data: 0.1904  max mem: 41794
Epoch: [172]  [ 30/312]  eta: 0:05:34  lr: 0.001730  min_lr: 0.001730  loss: 3.5269 (3.3388)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.0886  max mem: 41794
Epoch: [172]  [ 40/312]  eta: 0:04:46  lr: 0.001729  min_lr: 0.001729  loss: 3.5269 (3.4056)  weight_decay: 0.0500 (0.0500)  time: 0.5785  data: 0.0571  max mem: 41794
Epoch: [172]  [ 50/312]  eta: 0:04:26  lr: 0.001729  min_lr: 0.001729  loss: 3.5043 (3.3795)  weight_decay: 0.0500 (0.0500)  time: 0.7503  data: 0.2266  max mem: 41794
Epoch: [172]  [ 60/312]  eta: 0:03:53  lr: 0.001728  min_lr: 0.001728  loss: 3.5210 (3.3816)  weight_decay: 0.0500 (0.0500)  time: 0.6731  data: 0.1701  max mem: 41794
Epoch: [172]  [ 70/312]  eta: 0:03:49  lr: 0.001727  min_lr: 0.001727  loss: 3.5409 (3.3652)  weight_decay: 0.0500 (0.0500)  time: 0.7742  data: 0.2248  max mem: 41794
Epoch: [172]  [ 80/312]  eta: 0:03:32  lr: 0.001726  min_lr: 0.001726  loss: 3.5399 (3.3795)  weight_decay: 0.0500 (0.0500)  time: 0.8850  data: 0.2632  max mem: 41794
Epoch: [172]  [ 90/312]  eta: 0:03:18  lr: 0.001726  min_lr: 0.001726  loss: 3.4002 (3.3712)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.1444  max mem: 41794
Epoch: [172]  [100/312]  eta: 0:03:09  lr: 0.001725  min_lr: 0.001725  loss: 3.4002 (3.3713)  weight_decay: 0.0500 (0.0500)  time: 0.8031  data: 0.2396  max mem: 41794
Epoch: [172]  [110/312]  eta: 0:02:53  lr: 0.001724  min_lr: 0.001724  loss: 3.3204 (3.3595)  weight_decay: 0.0500 (0.0500)  time: 0.7052  data: 0.1342  max mem: 41794
Epoch: [172]  [120/312]  eta: 0:02:44  lr: 0.001724  min_lr: 0.001724  loss: 3.4366 (3.3691)  weight_decay: 0.0500 (0.0500)  time: 0.6550  data: 0.1163  max mem: 41794
Epoch: [172]  [130/312]  eta: 0:02:36  lr: 0.001723  min_lr: 0.001723  loss: 3.4366 (3.3534)  weight_decay: 0.0500 (0.0500)  time: 0.8685  data: 0.2701  max mem: 41794
Epoch: [172]  [140/312]  eta: 0:02:23  lr: 0.001722  min_lr: 0.001722  loss: 3.3125 (3.3589)  weight_decay: 0.0500 (0.0500)  time: 0.7154  data: 0.1544  max mem: 41794
Epoch: [172]  [150/312]  eta: 0:02:15  lr: 0.001721  min_lr: 0.001721  loss: 3.4499 (3.3577)  weight_decay: 0.0500 (0.0500)  time: 0.6617  data: 0.1279  max mem: 41794
Epoch: [172]  [160/312]  eta: 0:02:04  lr: 0.001721  min_lr: 0.001721  loss: 3.5200 (3.3729)  weight_decay: 0.0500 (0.0500)  time: 0.7067  data: 0.1516  max mem: 41794
Epoch: [172]  [170/312]  eta: 0:01:55  lr: 0.001720  min_lr: 0.001720  loss: 3.5361 (3.3768)  weight_decay: 0.0500 (0.0500)  time: 0.6482  data: 0.1327  max mem: 41794
Epoch: [172]  [180/312]  eta: 0:01:48  lr: 0.001719  min_lr: 0.001719  loss: 3.6451 (3.3869)  weight_decay: 0.0500 (0.0500)  time: 0.8135  data: 0.2632  max mem: 41794
Epoch: [172]  [190/312]  eta: 0:01:37  lr: 0.001719  min_lr: 0.001719  loss: 3.6402 (3.3886)  weight_decay: 0.0500 (0.0500)  time: 0.7168  data: 0.1548  max mem: 41794
Epoch: [172]  [200/312]  eta: 0:01:29  lr: 0.001718  min_lr: 0.001718  loss: 3.5687 (3.3808)  weight_decay: 0.0500 (0.0500)  time: 0.6387  data: 0.1041  max mem: 41794
Epoch: [172]  [210/312]  eta: 0:01:21  lr: 0.001717  min_lr: 0.001717  loss: 3.4508 (3.3804)  weight_decay: 0.0500 (0.0500)  time: 0.8050  data: 0.2497  max mem: 41794
Epoch: [172]  [220/312]  eta: 0:01:12  lr: 0.001716  min_lr: 0.001716  loss: 3.7207 (3.3885)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.1462  max mem: 41794
Epoch: [172]  [230/312]  eta: 0:01:04  lr: 0.001716  min_lr: 0.001716  loss: 3.5035 (3.3906)  weight_decay: 0.0500 (0.0500)  time: 0.6315  data: 0.1332  max mem: 41794
Epoch: [172]  [240/312]  eta: 0:00:56  lr: 0.001715  min_lr: 0.001715  loss: 3.4491 (3.3943)  weight_decay: 0.0500 (0.0500)  time: 0.7915  data: 0.2891  max mem: 41794
Epoch: [172]  [250/312]  eta: 0:00:48  lr: 0.001714  min_lr: 0.001714  loss: 3.3464 (3.3856)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1942  max mem: 41794
Epoch: [172]  [260/312]  eta: 0:00:40  lr: 0.001714  min_lr: 0.001714  loss: 3.2687 (3.3822)  weight_decay: 0.0500 (0.0500)  time: 0.7034  data: 0.2010  max mem: 41794
Epoch: [172]  [270/312]  eta: 0:00:32  lr: 0.001713  min_lr: 0.001713  loss: 3.4819 (3.3803)  weight_decay: 0.0500 (0.0500)  time: 0.6665  data: 0.1633  max mem: 41794
Epoch: [172]  [280/312]  eta: 0:00:24  lr: 0.001712  min_lr: 0.001712  loss: 3.5401 (3.3869)  weight_decay: 0.0500 (0.0500)  time: 0.6994  data: 0.2056  max mem: 41794
Epoch: [172]  [290/312]  eta: 0:00:17  lr: 0.001711  min_lr: 0.001711  loss: 3.5037 (3.3882)  weight_decay: 0.0500 (0.0500)  time: 0.8387  data: 0.3444  max mem: 41794
Epoch: [172]  [300/312]  eta: 0:00:09  lr: 0.001711  min_lr: 0.001711  loss: 3.3239 (3.3885)  weight_decay: 0.0500 (0.0500)  time: 0.6245  data: 0.1392  max mem: 41794
Epoch: [172]  [310/312]  eta: 0:00:01  lr: 0.001710  min_lr: 0.001710  loss: 3.2767 (3.3812)  weight_decay: 0.0500 (0.0500)  time: 0.4644  data: 0.0001  max mem: 41794
Epoch: [172]  [311/312]  eta: 0:00:00  lr: 0.001710  min_lr: 0.001710  loss: 3.3175 (3.3810)  weight_decay: 0.0500 (0.0500)  time: 0.4632  data: 0.0001  max mem: 41794
Epoch: [172] Total time: 0:03:57 (0.7622 s / it)
Averaged stats: lr: 0.001710  min_lr: 0.001710  loss: 3.3175 (3.3574)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.1288 (1.1288)  acc1: 78.7760 (78.7760)  acc5: 95.1823 (95.1823)  time: 7.9188  data: 7.7170  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3841 (1.3180)  acc1: 72.6562 (73.6960)  acc5: 92.3177 (92.4960)  time: 1.0340  data: 0.8576  max mem: 41794
Test: Total time: 0:00:09 (1.0478 s / it)
* Acc@1 73.870 Acc@5 92.348 loss 1.320
Accuracy of the model on the 50000 test images: 73.9%
Max accuracy: 74.48%
Epoch: [173]  [  0/312]  eta: 1:24:45  lr: 0.001710  min_lr: 0.001710  loss: 4.0064 (4.0064)  weight_decay: 0.0500 (0.0500)  time: 16.2987  data: 13.3757  max mem: 41794
Epoch: [173]  [ 10/312]  eta: 0:10:31  lr: 0.001709  min_lr: 0.001709  loss: 3.3655 (3.2281)  weight_decay: 0.0500 (0.0500)  time: 2.0905  data: 1.2188  max mem: 41794
Epoch: [173]  [ 20/312]  eta: 0:07:18  lr: 0.001708  min_lr: 0.001708  loss: 3.4605 (3.2640)  weight_decay: 0.0500 (0.0500)  time: 0.7606  data: 0.0429  max mem: 41794
Epoch: [173]  [ 30/312]  eta: 0:05:37  lr: 0.001708  min_lr: 0.001708  loss: 3.5520 (3.2541)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.0611  max mem: 41794
Epoch: [173]  [ 40/312]  eta: 0:04:58  lr: 0.001707  min_lr: 0.001707  loss: 3.5185 (3.2622)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.0627  max mem: 41794
Epoch: [173]  [ 50/312]  eta: 0:04:31  lr: 0.001706  min_lr: 0.001706  loss: 3.5876 (3.3048)  weight_decay: 0.0500 (0.0500)  time: 0.7849  data: 0.0679  max mem: 41794
Epoch: [173]  [ 60/312]  eta: 0:03:58  lr: 0.001706  min_lr: 0.001706  loss: 3.6010 (3.3467)  weight_decay: 0.0500 (0.0500)  time: 0.6347  data: 0.0252  max mem: 41794
Epoch: [173]  [ 70/312]  eta: 0:03:45  lr: 0.001705  min_lr: 0.001705  loss: 3.4912 (3.3240)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.0993  max mem: 41794
Epoch: [173]  [ 80/312]  eta: 0:03:31  lr: 0.001704  min_lr: 0.001704  loss: 3.3287 (3.3490)  weight_decay: 0.0500 (0.0500)  time: 0.7992  data: 0.1747  max mem: 41794
Epoch: [173]  [ 90/312]  eta: 0:03:14  lr: 0.001703  min_lr: 0.001703  loss: 3.6135 (3.3657)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1019  max mem: 41794
Epoch: [173]  [100/312]  eta: 0:03:05  lr: 0.001703  min_lr: 0.001703  loss: 3.5402 (3.3654)  weight_decay: 0.0500 (0.0500)  time: 0.7337  data: 0.1056  max mem: 41794
Epoch: [173]  [110/312]  eta: 0:02:49  lr: 0.001702  min_lr: 0.001702  loss: 3.5250 (3.3792)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.0798  max mem: 41794
Epoch: [173]  [120/312]  eta: 0:02:42  lr: 0.001701  min_lr: 0.001701  loss: 3.5250 (3.3727)  weight_decay: 0.0500 (0.0500)  time: 0.6943  data: 0.1003  max mem: 41794
Epoch: [173]  [130/312]  eta: 0:02:34  lr: 0.001701  min_lr: 0.001701  loss: 3.1750 (3.3421)  weight_decay: 0.0500 (0.0500)  time: 0.8745  data: 0.1994  max mem: 41794
Epoch: [173]  [140/312]  eta: 0:02:21  lr: 0.001700  min_lr: 0.001700  loss: 3.3346 (3.3545)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.0997  max mem: 41794
Epoch: [173]  [150/312]  eta: 0:02:14  lr: 0.001699  min_lr: 0.001699  loss: 3.2531 (3.3302)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.1256  max mem: 41794
Epoch: [173]  [160/312]  eta: 0:02:05  lr: 0.001699  min_lr: 0.001699  loss: 3.2921 (3.3302)  weight_decay: 0.0500 (0.0500)  time: 0.8316  data: 0.2121  max mem: 41794
Epoch: [173]  [170/312]  eta: 0:01:55  lr: 0.001698  min_lr: 0.001698  loss: 3.4926 (3.3255)  weight_decay: 0.0500 (0.0500)  time: 0.6958  data: 0.1180  max mem: 41794
Epoch: [173]  [180/312]  eta: 0:01:47  lr: 0.001697  min_lr: 0.001697  loss: 3.3717 (3.3304)  weight_decay: 0.0500 (0.0500)  time: 0.7442  data: 0.1436  max mem: 41794
Epoch: [173]  [190/312]  eta: 0:01:37  lr: 0.001696  min_lr: 0.001696  loss: 3.3024 (3.3166)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.1128  max mem: 41794
Epoch: [173]  [200/312]  eta: 0:01:29  lr: 0.001696  min_lr: 0.001696  loss: 3.4199 (3.3297)  weight_decay: 0.0500 (0.0500)  time: 0.6626  data: 0.1167  max mem: 41794
Epoch: [173]  [210/312]  eta: 0:01:22  lr: 0.001695  min_lr: 0.001695  loss: 3.6122 (3.3335)  weight_decay: 0.0500 (0.0500)  time: 0.8891  data: 0.2570  max mem: 41794
Epoch: [173]  [220/312]  eta: 0:01:12  lr: 0.001694  min_lr: 0.001694  loss: 3.5057 (3.3375)  weight_decay: 0.0500 (0.0500)  time: 0.7238  data: 0.1409  max mem: 41794
Epoch: [173]  [230/312]  eta: 0:01:05  lr: 0.001694  min_lr: 0.001694  loss: 3.4942 (3.3444)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.1468  max mem: 41794
Epoch: [173]  [240/312]  eta: 0:00:57  lr: 0.001693  min_lr: 0.001693  loss: 3.4988 (3.3447)  weight_decay: 0.0500 (0.0500)  time: 0.8232  data: 0.1715  max mem: 41794
Epoch: [173]  [250/312]  eta: 0:00:49  lr: 0.001692  min_lr: 0.001692  loss: 3.4801 (3.3453)  weight_decay: 0.0500 (0.0500)  time: 0.7296  data: 0.1308  max mem: 41794
Epoch: [173]  [260/312]  eta: 0:00:41  lr: 0.001691  min_lr: 0.001691  loss: 3.4943 (3.3489)  weight_decay: 0.0500 (0.0500)  time: 0.7644  data: 0.2110  max mem: 41794
Epoch: [173]  [270/312]  eta: 0:00:32  lr: 0.001691  min_lr: 0.001691  loss: 3.5561 (3.3495)  weight_decay: 0.0500 (0.0500)  time: 0.6632  data: 0.1125  max mem: 41794
Epoch: [173]  [280/312]  eta: 0:00:25  lr: 0.001690  min_lr: 0.001690  loss: 3.5472 (3.3533)  weight_decay: 0.0500 (0.0500)  time: 0.6678  data: 0.1211  max mem: 41794
Epoch: [173]  [290/312]  eta: 0:00:17  lr: 0.001689  min_lr: 0.001689  loss: 3.5595 (3.3620)  weight_decay: 0.0500 (0.0500)  time: 0.7905  data: 0.1845  max mem: 41794
Epoch: [173]  [300/312]  eta: 0:00:09  lr: 0.001689  min_lr: 0.001689  loss: 3.5326 (3.3628)  weight_decay: 0.0500 (0.0500)  time: 0.6081  data: 0.0707  max mem: 41794
Epoch: [173]  [310/312]  eta: 0:00:01  lr: 0.001688  min_lr: 0.001688  loss: 3.4474 (3.3615)  weight_decay: 0.0500 (0.0500)  time: 0.4617  data: 0.0001  max mem: 41794
Epoch: [173]  [311/312]  eta: 0:00:00  lr: 0.001688  min_lr: 0.001688  loss: 3.4474 (3.3614)  weight_decay: 0.0500 (0.0500)  time: 0.4611  data: 0.0001  max mem: 41794
Epoch: [173] Total time: 0:03:59 (0.7682 s / it)
Averaged stats: lr: 0.001688  min_lr: 0.001688  loss: 3.4474 (3.3647)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.1188 (1.1188)  acc1: 80.8594 (80.8594)  acc5: 95.1823 (95.1823)  time: 8.3860  data: 8.1791  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3774 (1.3220)  acc1: 75.1302 (74.6880)  acc5: 93.0990 (92.4320)  time: 1.0811  data: 0.9089  max mem: 41794
Test: Total time: 0:00:10 (1.1235 s / it)
* Acc@1 74.530 Acc@5 92.526 loss 1.320
Accuracy of the model on the 50000 test images: 74.5%
Max accuracy: 74.53%
Epoch: [174]  [  0/312]  eta: 1:24:30  lr: 0.001688  min_lr: 0.001688  loss: 3.2270 (3.2270)  weight_decay: 0.0500 (0.0500)  time: 16.2501  data: 15.7649  max mem: 41794
Epoch: [174]  [ 10/312]  eta: 0:11:35  lr: 0.001687  min_lr: 0.001687  loss: 3.4483 (3.3789)  weight_decay: 0.0500 (0.0500)  time: 2.3023  data: 1.4339  max mem: 41794
Epoch: [174]  [ 20/312]  eta: 0:07:28  lr: 0.001686  min_lr: 0.001686  loss: 3.4483 (3.3462)  weight_decay: 0.0500 (0.0500)  time: 0.7994  data: 0.0928  max mem: 41794
Epoch: [174]  [ 30/312]  eta: 0:05:42  lr: 0.001686  min_lr: 0.001686  loss: 3.3225 (3.3388)  weight_decay: 0.0500 (0.0500)  time: 0.6181  data: 0.0972  max mem: 41794
Epoch: [174]  [ 40/312]  eta: 0:04:44  lr: 0.001685  min_lr: 0.001685  loss: 3.4752 (3.3528)  weight_decay: 0.0500 (0.0500)  time: 0.5304  data: 0.0098  max mem: 41794
Epoch: [174]  [ 50/312]  eta: 0:04:16  lr: 0.001684  min_lr: 0.001684  loss: 3.4752 (3.3368)  weight_decay: 0.0500 (0.0500)  time: 0.6148  data: 0.1006  max mem: 41794
Epoch: [174]  [ 60/312]  eta: 0:03:47  lr: 0.001683  min_lr: 0.001683  loss: 3.0752 (3.2998)  weight_decay: 0.0500 (0.0500)  time: 0.6149  data: 0.0973  max mem: 41794
Epoch: [174]  [ 70/312]  eta: 0:03:35  lr: 0.001683  min_lr: 0.001683  loss: 3.1281 (3.2643)  weight_decay: 0.0500 (0.0500)  time: 0.6574  data: 0.1469  max mem: 41794
Epoch: [174]  [ 80/312]  eta: 0:03:27  lr: 0.001682  min_lr: 0.001682  loss: 3.3166 (3.2522)  weight_decay: 0.0500 (0.0500)  time: 0.8623  data: 0.3543  max mem: 41794
Epoch: [174]  [ 90/312]  eta: 0:03:09  lr: 0.001681  min_lr: 0.001681  loss: 3.4357 (3.2700)  weight_decay: 0.0500 (0.0500)  time: 0.7248  data: 0.2135  max mem: 41794
Epoch: [174]  [100/312]  eta: 0:03:00  lr: 0.001681  min_lr: 0.001681  loss: 3.6016 (3.3064)  weight_decay: 0.0500 (0.0500)  time: 0.6875  data: 0.1800  max mem: 41794
Epoch: [174]  [110/312]  eta: 0:02:46  lr: 0.001680  min_lr: 0.001680  loss: 3.3605 (3.3114)  weight_decay: 0.0500 (0.0500)  time: 0.6839  data: 0.1804  max mem: 41794
Epoch: [174]  [120/312]  eta: 0:02:38  lr: 0.001679  min_lr: 0.001679  loss: 3.2806 (3.2725)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.1720  max mem: 41794
Epoch: [174]  [130/312]  eta: 0:02:30  lr: 0.001678  min_lr: 0.001678  loss: 3.0054 (3.2784)  weight_decay: 0.0500 (0.0500)  time: 0.8537  data: 0.3592  max mem: 41794
Epoch: [174]  [140/312]  eta: 0:02:18  lr: 0.001678  min_lr: 0.001678  loss: 3.3601 (3.2813)  weight_decay: 0.0500 (0.0500)  time: 0.6913  data: 0.1948  max mem: 41794
Epoch: [174]  [150/312]  eta: 0:02:10  lr: 0.001677  min_lr: 0.001677  loss: 3.3893 (3.2984)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.1603  max mem: 41794
Epoch: [174]  [160/312]  eta: 0:02:03  lr: 0.001676  min_lr: 0.001676  loss: 3.2237 (3.2825)  weight_decay: 0.0500 (0.0500)  time: 0.8425  data: 0.3304  max mem: 41794
Epoch: [174]  [170/312]  eta: 0:01:52  lr: 0.001676  min_lr: 0.001676  loss: 3.1427 (3.2845)  weight_decay: 0.0500 (0.0500)  time: 0.6808  data: 0.1764  max mem: 41794
Epoch: [174]  [180/312]  eta: 0:01:44  lr: 0.001675  min_lr: 0.001675  loss: 3.5559 (3.2865)  weight_decay: 0.0500 (0.0500)  time: 0.6726  data: 0.1704  max mem: 41794
Epoch: [174]  [190/312]  eta: 0:01:35  lr: 0.001674  min_lr: 0.001674  loss: 3.5873 (3.3040)  weight_decay: 0.0500 (0.0500)  time: 0.6681  data: 0.1706  max mem: 41794
Epoch: [174]  [200/312]  eta: 0:01:27  lr: 0.001674  min_lr: 0.001674  loss: 3.5098 (3.2974)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.1612  max mem: 41794
Epoch: [174]  [210/312]  eta: 0:01:20  lr: 0.001673  min_lr: 0.001673  loss: 3.2538 (3.3040)  weight_decay: 0.0500 (0.0500)  time: 0.8661  data: 0.3257  max mem: 41794
Epoch: [174]  [220/312]  eta: 0:01:11  lr: 0.001672  min_lr: 0.001672  loss: 3.5612 (3.3115)  weight_decay: 0.0500 (0.0500)  time: 0.6757  data: 0.1686  max mem: 41794
Epoch: [174]  [230/312]  eta: 0:01:03  lr: 0.001671  min_lr: 0.001671  loss: 3.5612 (3.3109)  weight_decay: 0.0500 (0.0500)  time: 0.7046  data: 0.1759  max mem: 41794
Epoch: [174]  [240/312]  eta: 0:00:56  lr: 0.001671  min_lr: 0.001671  loss: 3.4573 (3.3143)  weight_decay: 0.0500 (0.0500)  time: 0.8821  data: 0.3422  max mem: 41794
Epoch: [174]  [250/312]  eta: 0:00:47  lr: 0.001670  min_lr: 0.001670  loss: 3.3145 (3.3086)  weight_decay: 0.0500 (0.0500)  time: 0.6804  data: 0.1689  max mem: 41794
Epoch: [174]  [260/312]  eta: 0:00:40  lr: 0.001669  min_lr: 0.001669  loss: 3.4065 (3.3088)  weight_decay: 0.0500 (0.0500)  time: 0.6809  data: 0.1691  max mem: 41794
Epoch: [174]  [270/312]  eta: 0:00:32  lr: 0.001669  min_lr: 0.001669  loss: 3.5048 (3.3152)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.1682  max mem: 41794
Epoch: [174]  [280/312]  eta: 0:00:24  lr: 0.001668  min_lr: 0.001668  loss: 3.5832 (3.3239)  weight_decay: 0.0500 (0.0500)  time: 0.7183  data: 0.1893  max mem: 41794
Epoch: [174]  [290/312]  eta: 0:00:17  lr: 0.001667  min_lr: 0.001667  loss: 3.5474 (3.3275)  weight_decay: 0.0500 (0.0500)  time: 0.9009  data: 0.3671  max mem: 41794
Epoch: [174]  [300/312]  eta: 0:00:09  lr: 0.001666  min_lr: 0.001666  loss: 3.5365 (3.3322)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.1794  max mem: 41794
Epoch: [174]  [310/312]  eta: 0:00:01  lr: 0.001666  min_lr: 0.001666  loss: 3.5126 (3.3396)  weight_decay: 0.0500 (0.0500)  time: 0.4739  data: 0.0001  max mem: 41794
Epoch: [174]  [311/312]  eta: 0:00:00  lr: 0.001666  min_lr: 0.001666  loss: 3.5126 (3.3420)  weight_decay: 0.0500 (0.0500)  time: 0.4679  data: 0.0001  max mem: 41794
Epoch: [174] Total time: 0:03:57 (0.7606 s / it)
Averaged stats: lr: 0.001666  min_lr: 0.001666  loss: 3.5126 (3.3553)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:08  loss: 1.2514 (1.2514)  acc1: 80.3385 (80.3385)  acc5: 94.9219 (94.9219)  time: 7.5622  data: 7.3646  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4458 (1.3901)  acc1: 74.0885 (74.0320)  acc5: 92.9688 (92.3520)  time: 1.0083  data: 0.8323  max mem: 41794
Test: Total time: 0:00:09 (1.0195 s / it)
* Acc@1 73.824 Acc@5 92.194 loss 1.400
Accuracy of the model on the 50000 test images: 73.8%
Max accuracy: 74.53%
Epoch: [175]  [  0/312]  eta: 1:22:19  lr: 0.001666  min_lr: 0.001666  loss: 3.8089 (3.8089)  weight_decay: 0.0500 (0.0500)  time: 15.8323  data: 11.9378  max mem: 41794
Epoch: [175]  [ 10/312]  eta: 0:10:44  lr: 0.001665  min_lr: 0.001665  loss: 3.2469 (3.1595)  weight_decay: 0.0500 (0.0500)  time: 2.1354  data: 1.3177  max mem: 41794
Epoch: [175]  [ 20/312]  eta: 0:07:23  lr: 0.001664  min_lr: 0.001664  loss: 3.3877 (3.2449)  weight_decay: 0.0500 (0.0500)  time: 0.8027  data: 0.1825  max mem: 41794
Epoch: [175]  [ 30/312]  eta: 0:05:38  lr: 0.001663  min_lr: 0.001663  loss: 3.4700 (3.3239)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.0566  max mem: 41794
Epoch: [175]  [ 40/312]  eta: 0:04:52  lr: 0.001663  min_lr: 0.001663  loss: 3.4641 (3.3062)  weight_decay: 0.0500 (0.0500)  time: 0.6122  data: 0.0418  max mem: 41794
Epoch: [175]  [ 50/312]  eta: 0:04:32  lr: 0.001662  min_lr: 0.001662  loss: 3.1934 (3.2804)  weight_decay: 0.0500 (0.0500)  time: 0.7938  data: 0.1536  max mem: 41794
Epoch: [175]  [ 60/312]  eta: 0:03:59  lr: 0.001661  min_lr: 0.001661  loss: 3.1550 (3.2868)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.1141  max mem: 41794
Epoch: [175]  [ 70/312]  eta: 0:03:50  lr: 0.001661  min_lr: 0.001661  loss: 3.2957 (3.2798)  weight_decay: 0.0500 (0.0500)  time: 0.7324  data: 0.1272  max mem: 41794
Epoch: [175]  [ 80/312]  eta: 0:03:36  lr: 0.001660  min_lr: 0.001660  loss: 3.4225 (3.3082)  weight_decay: 0.0500 (0.0500)  time: 0.8801  data: 0.1325  max mem: 41794
Epoch: [175]  [ 90/312]  eta: 0:03:20  lr: 0.001659  min_lr: 0.001659  loss: 3.4467 (3.3035)  weight_decay: 0.0500 (0.0500)  time: 0.7205  data: 0.0897  max mem: 41794
Epoch: [175]  [100/312]  eta: 0:03:09  lr: 0.001658  min_lr: 0.001658  loss: 3.0376 (3.2733)  weight_decay: 0.0500 (0.0500)  time: 0.7471  data: 0.1471  max mem: 41794
Epoch: [175]  [110/312]  eta: 0:02:54  lr: 0.001658  min_lr: 0.001658  loss: 2.9585 (3.2518)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.0686  max mem: 41794
Epoch: [175]  [120/312]  eta: 0:02:44  lr: 0.001657  min_lr: 0.001657  loss: 3.2482 (3.2671)  weight_decay: 0.0500 (0.0500)  time: 0.6634  data: 0.0705  max mem: 41794
Epoch: [175]  [130/312]  eta: 0:02:35  lr: 0.001656  min_lr: 0.001656  loss: 3.2503 (3.2548)  weight_decay: 0.0500 (0.0500)  time: 0.8087  data: 0.1274  max mem: 41794
Epoch: [175]  [140/312]  eta: 0:02:22  lr: 0.001656  min_lr: 0.001656  loss: 3.2144 (3.2469)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.0644  max mem: 41794
Epoch: [175]  [150/312]  eta: 0:02:15  lr: 0.001655  min_lr: 0.001655  loss: 3.2194 (3.2500)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.0891  max mem: 41794
Epoch: [175]  [160/312]  eta: 0:02:05  lr: 0.001654  min_lr: 0.001654  loss: 3.3975 (3.2528)  weight_decay: 0.0500 (0.0500)  time: 0.8156  data: 0.0875  max mem: 41794
Epoch: [175]  [170/312]  eta: 0:01:56  lr: 0.001654  min_lr: 0.001654  loss: 3.2257 (3.2455)  weight_decay: 0.0500 (0.0500)  time: 0.6905  data: 0.0652  max mem: 41794
Epoch: [175]  [180/312]  eta: 0:01:48  lr: 0.001653  min_lr: 0.001653  loss: 3.2257 (3.2525)  weight_decay: 0.0500 (0.0500)  time: 0.7785  data: 0.1090  max mem: 41794
Epoch: [175]  [190/312]  eta: 0:01:38  lr: 0.001652  min_lr: 0.001652  loss: 3.5753 (3.2573)  weight_decay: 0.0500 (0.0500)  time: 0.7163  data: 0.0445  max mem: 41794
Epoch: [175]  [200/312]  eta: 0:01:30  lr: 0.001651  min_lr: 0.001651  loss: 3.5452 (3.2663)  weight_decay: 0.0500 (0.0500)  time: 0.6667  data: 0.0294  max mem: 41794
Epoch: [175]  [210/312]  eta: 0:01:22  lr: 0.001651  min_lr: 0.001651  loss: 3.4905 (3.2650)  weight_decay: 0.0500 (0.0500)  time: 0.8595  data: 0.0794  max mem: 41794
Epoch: [175]  [220/312]  eta: 0:01:13  lr: 0.001650  min_lr: 0.001650  loss: 3.4253 (3.2702)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.0506  max mem: 41794
Epoch: [175]  [230/312]  eta: 0:01:05  lr: 0.001649  min_lr: 0.001649  loss: 3.4370 (3.2759)  weight_decay: 0.0500 (0.0500)  time: 0.6464  data: 0.0431  max mem: 41794
Epoch: [175]  [240/312]  eta: 0:00:57  lr: 0.001649  min_lr: 0.001649  loss: 3.1696 (3.2689)  weight_decay: 0.0500 (0.0500)  time: 0.7422  data: 0.0431  max mem: 41794
Epoch: [175]  [250/312]  eta: 0:00:48  lr: 0.001648  min_lr: 0.001648  loss: 3.5779 (3.2837)  weight_decay: 0.0500 (0.0500)  time: 0.6680  data: 0.0426  max mem: 41794
Epoch: [175]  [260/312]  eta: 0:00:41  lr: 0.001647  min_lr: 0.001647  loss: 3.5779 (3.2790)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.0971  max mem: 41794
Epoch: [175]  [270/312]  eta: 0:00:32  lr: 0.001646  min_lr: 0.001646  loss: 3.3939 (3.2866)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.0551  max mem: 41794
Epoch: [175]  [280/312]  eta: 0:00:25  lr: 0.001646  min_lr: 0.001646  loss: 3.5752 (3.2969)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.0873  max mem: 41794
Epoch: [175]  [290/312]  eta: 0:00:17  lr: 0.001645  min_lr: 0.001645  loss: 3.5743 (3.2975)  weight_decay: 0.0500 (0.0500)  time: 0.7981  data: 0.1316  max mem: 41794
Epoch: [175]  [300/312]  eta: 0:00:09  lr: 0.001644  min_lr: 0.001644  loss: 3.5967 (3.3109)  weight_decay: 0.0500 (0.0500)  time: 0.6089  data: 0.0447  max mem: 41794
Epoch: [175]  [310/312]  eta: 0:00:01  lr: 0.001644  min_lr: 0.001644  loss: 3.6434 (3.3139)  weight_decay: 0.0500 (0.0500)  time: 0.4734  data: 0.0001  max mem: 41794
Epoch: [175]  [311/312]  eta: 0:00:00  lr: 0.001644  min_lr: 0.001644  loss: 3.6434 (3.3133)  weight_decay: 0.0500 (0.0500)  time: 0.4729  data: 0.0001  max mem: 41794
Epoch: [175] Total time: 0:03:59 (0.7663 s / it)
Averaged stats: lr: 0.001644  min_lr: 0.001644  loss: 3.6434 (3.3457)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.1278 (1.1278)  acc1: 80.9896 (80.9896)  acc5: 95.0521 (95.0521)  time: 8.0707  data: 7.8650  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3207 (1.2557)  acc1: 74.3490 (74.6720)  acc5: 91.6667 (92.3840)  time: 1.0496  data: 0.8740  max mem: 41794
Test: Total time: 0:00:09 (1.0629 s / it)
* Acc@1 74.564 Acc@5 92.594 loss 1.247
Accuracy of the model on the 50000 test images: 74.6%
Max accuracy: 74.56%
Epoch: [176]  [  0/312]  eta: 1:19:47  lr: 0.001643  min_lr: 0.001643  loss: 3.9539 (3.9539)  weight_decay: 0.0500 (0.0500)  time: 15.3435  data: 13.5679  max mem: 41794
Epoch: [176]  [ 10/312]  eta: 0:11:10  lr: 0.001643  min_lr: 0.001643  loss: 3.5388 (3.3253)  weight_decay: 0.0500 (0.0500)  time: 2.2195  data: 1.3518  max mem: 41794
Epoch: [176]  [ 20/312]  eta: 0:07:18  lr: 0.001642  min_lr: 0.001642  loss: 3.2256 (3.3216)  weight_decay: 0.0500 (0.0500)  time: 0.8086  data: 0.1638  max mem: 41794
Epoch: [176]  [ 30/312]  eta: 0:05:35  lr: 0.001641  min_lr: 0.001641  loss: 3.4914 (3.3639)  weight_decay: 0.0500 (0.0500)  time: 0.6247  data: 0.0990  max mem: 41794
Epoch: [176]  [ 40/312]  eta: 0:04:47  lr: 0.001641  min_lr: 0.001641  loss: 3.4908 (3.3094)  weight_decay: 0.0500 (0.0500)  time: 0.5948  data: 0.0839  max mem: 41794
Epoch: [176]  [ 50/312]  eta: 0:04:23  lr: 0.001640  min_lr: 0.001640  loss: 3.4908 (3.3223)  weight_decay: 0.0500 (0.0500)  time: 0.7196  data: 0.2190  max mem: 41794
Epoch: [176]  [ 60/312]  eta: 0:03:52  lr: 0.001639  min_lr: 0.001639  loss: 3.5030 (3.2817)  weight_decay: 0.0500 (0.0500)  time: 0.6435  data: 0.1358  max mem: 41794
Epoch: [176]  [ 70/312]  eta: 0:03:38  lr: 0.001639  min_lr: 0.001639  loss: 3.5605 (3.2900)  weight_decay: 0.0500 (0.0500)  time: 0.6390  data: 0.1399  max mem: 41794
Epoch: [176]  [ 80/312]  eta: 0:03:29  lr: 0.001638  min_lr: 0.001638  loss: 3.5605 (3.2842)  weight_decay: 0.0500 (0.0500)  time: 0.8359  data: 0.3402  max mem: 41794
Epoch: [176]  [ 90/312]  eta: 0:03:10  lr: 0.001637  min_lr: 0.001637  loss: 3.3220 (3.2780)  weight_decay: 0.0500 (0.0500)  time: 0.6950  data: 0.2009  max mem: 41794
Epoch: [176]  [100/312]  eta: 0:03:02  lr: 0.001636  min_lr: 0.001636  loss: 3.2463 (3.2743)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.1980  max mem: 41794
Epoch: [176]  [110/312]  eta: 0:02:46  lr: 0.001636  min_lr: 0.001636  loss: 3.1932 (3.2927)  weight_decay: 0.0500 (0.0500)  time: 0.6881  data: 0.1992  max mem: 41794
Epoch: [176]  [120/312]  eta: 0:02:38  lr: 0.001635  min_lr: 0.001635  loss: 3.7023 (3.3038)  weight_decay: 0.0500 (0.0500)  time: 0.6459  data: 0.1572  max mem: 41794
Epoch: [176]  [130/312]  eta: 0:02:32  lr: 0.001634  min_lr: 0.001634  loss: 3.4315 (3.2913)  weight_decay: 0.0500 (0.0500)  time: 0.8946  data: 0.4019  max mem: 41794
Epoch: [176]  [140/312]  eta: 0:02:19  lr: 0.001634  min_lr: 0.001634  loss: 3.4315 (3.2986)  weight_decay: 0.0500 (0.0500)  time: 0.7364  data: 0.2466  max mem: 41794
Epoch: [176]  [150/312]  eta: 0:02:11  lr: 0.001633  min_lr: 0.001633  loss: 3.4566 (3.3074)  weight_decay: 0.0500 (0.0500)  time: 0.6564  data: 0.1595  max mem: 41794
Epoch: [176]  [160/312]  eta: 0:02:03  lr: 0.001632  min_lr: 0.001632  loss: 3.5201 (3.3231)  weight_decay: 0.0500 (0.0500)  time: 0.8318  data: 0.3339  max mem: 41794
Epoch: [176]  [170/312]  eta: 0:01:53  lr: 0.001631  min_lr: 0.001631  loss: 3.5340 (3.3212)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1751  max mem: 41794
Epoch: [176]  [180/312]  eta: 0:01:45  lr: 0.001631  min_lr: 0.001631  loss: 3.4300 (3.3285)  weight_decay: 0.0500 (0.0500)  time: 0.6613  data: 0.1503  max mem: 41794
Epoch: [176]  [190/312]  eta: 0:01:35  lr: 0.001630  min_lr: 0.001630  loss: 3.3910 (3.3264)  weight_decay: 0.0500 (0.0500)  time: 0.6541  data: 0.1504  max mem: 41794
Epoch: [176]  [200/312]  eta: 0:01:27  lr: 0.001629  min_lr: 0.001629  loss: 3.3910 (3.3275)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.1629  max mem: 41794
Epoch: [176]  [210/312]  eta: 0:01:20  lr: 0.001629  min_lr: 0.001629  loss: 3.5323 (3.3361)  weight_decay: 0.0500 (0.0500)  time: 0.8601  data: 0.3379  max mem: 41794
Epoch: [176]  [220/312]  eta: 0:01:11  lr: 0.001628  min_lr: 0.001628  loss: 3.5368 (3.3387)  weight_decay: 0.0500 (0.0500)  time: 0.6762  data: 0.1757  max mem: 41794
Epoch: [176]  [230/312]  eta: 0:01:03  lr: 0.001627  min_lr: 0.001627  loss: 3.4744 (3.3355)  weight_decay: 0.0500 (0.0500)  time: 0.6895  data: 0.1961  max mem: 41794
Epoch: [176]  [240/312]  eta: 0:00:56  lr: 0.001627  min_lr: 0.001627  loss: 3.4463 (3.3318)  weight_decay: 0.0500 (0.0500)  time: 0.8800  data: 0.3817  max mem: 41794
Epoch: [176]  [250/312]  eta: 0:00:47  lr: 0.001626  min_lr: 0.001626  loss: 3.4066 (3.3278)  weight_decay: 0.0500 (0.0500)  time: 0.6851  data: 0.1861  max mem: 41794
Epoch: [176]  [260/312]  eta: 0:00:40  lr: 0.001625  min_lr: 0.001625  loss: 3.5170 (3.3303)  weight_decay: 0.0500 (0.0500)  time: 0.6943  data: 0.1894  max mem: 41794
Epoch: [176]  [270/312]  eta: 0:00:32  lr: 0.001624  min_lr: 0.001624  loss: 3.5170 (3.3314)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.1894  max mem: 41794
Epoch: [176]  [280/312]  eta: 0:00:24  lr: 0.001624  min_lr: 0.001624  loss: 3.6061 (3.3308)  weight_decay: 0.0500 (0.0500)  time: 0.6641  data: 0.1716  max mem: 41794
Epoch: [176]  [290/312]  eta: 0:00:17  lr: 0.001623  min_lr: 0.001623  loss: 3.5649 (3.3347)  weight_decay: 0.0500 (0.0500)  time: 0.8649  data: 0.3726  max mem: 41794
Epoch: [176]  [300/312]  eta: 0:00:09  lr: 0.001622  min_lr: 0.001622  loss: 3.4627 (3.3334)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.2013  max mem: 41794
Epoch: [176]  [310/312]  eta: 0:00:01  lr: 0.001622  min_lr: 0.001622  loss: 3.5134 (3.3393)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [176]  [311/312]  eta: 0:00:00  lr: 0.001621  min_lr: 0.001621  loss: 3.2882 (3.3381)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [176] Total time: 0:03:57 (0.7597 s / it)
Averaged stats: lr: 0.001621  min_lr: 0.001621  loss: 3.2882 (3.3416)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 1.0982 (1.0982)  acc1: 80.9896 (80.9896)  acc5: 95.0521 (95.0521)  time: 7.8431  data: 7.6310  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3498 (1.2993)  acc1: 72.9167 (73.5520)  acc5: 92.7083 (92.3840)  time: 1.0291  data: 0.8546  max mem: 41794
Test: Total time: 0:00:09 (1.0391 s / it)
* Acc@1 74.264 Acc@5 92.520 loss 1.295
Accuracy of the model on the 50000 test images: 74.3%
Max accuracy: 74.56%
Epoch: [177]  [  0/312]  eta: 1:24:39  lr: 0.001621  min_lr: 0.001621  loss: 3.3797 (3.3797)  weight_decay: 0.0500 (0.0500)  time: 16.2818  data: 12.6053  max mem: 41794
Epoch: [177]  [ 10/312]  eta: 0:11:25  lr: 0.001621  min_lr: 0.001621  loss: 3.4189 (3.4120)  weight_decay: 0.0500 (0.0500)  time: 2.2709  data: 1.4434  max mem: 41794
Epoch: [177]  [ 20/312]  eta: 0:07:37  lr: 0.001620  min_lr: 0.001620  loss: 3.1939 (3.2350)  weight_decay: 0.0500 (0.0500)  time: 0.8315  data: 0.2370  max mem: 41794
Epoch: [177]  [ 30/312]  eta: 0:05:48  lr: 0.001619  min_lr: 0.001619  loss: 3.1894 (3.3112)  weight_decay: 0.0500 (0.0500)  time: 0.6665  data: 0.0767  max mem: 41794
Epoch: [177]  [ 40/312]  eta: 0:04:48  lr: 0.001619  min_lr: 0.001619  loss: 3.3975 (3.2487)  weight_decay: 0.0500 (0.0500)  time: 0.5308  data: 0.0036  max mem: 41794
Epoch: [177]  [ 50/312]  eta: 0:04:27  lr: 0.001618  min_lr: 0.001618  loss: 3.2442 (3.2736)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.0749  max mem: 41794
Epoch: [177]  [ 60/312]  eta: 0:03:55  lr: 0.001617  min_lr: 0.001617  loss: 3.4902 (3.2873)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.0749  max mem: 41794
Epoch: [177]  [ 70/312]  eta: 0:03:44  lr: 0.001616  min_lr: 0.001616  loss: 3.1248 (3.2712)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.0691  max mem: 41794
Epoch: [177]  [ 80/312]  eta: 0:03:33  lr: 0.001616  min_lr: 0.001616  loss: 3.1294 (3.2941)  weight_decay: 0.0500 (0.0500)  time: 0.8803  data: 0.0706  max mem: 41794
Epoch: [177]  [ 90/312]  eta: 0:03:17  lr: 0.001615  min_lr: 0.001615  loss: 3.4399 (3.2993)  weight_decay: 0.0500 (0.0500)  time: 0.7543  data: 0.0770  max mem: 41794
Epoch: [177]  [100/312]  eta: 0:03:07  lr: 0.001614  min_lr: 0.001614  loss: 3.4336 (3.2947)  weight_decay: 0.0500 (0.0500)  time: 0.7357  data: 0.1158  max mem: 41794
Epoch: [177]  [110/312]  eta: 0:02:51  lr: 0.001614  min_lr: 0.001614  loss: 3.4629 (3.3100)  weight_decay: 0.0500 (0.0500)  time: 0.6724  data: 0.0442  max mem: 41794
Epoch: [177]  [120/312]  eta: 0:02:44  lr: 0.001613  min_lr: 0.001613  loss: 3.5365 (3.3180)  weight_decay: 0.0500 (0.0500)  time: 0.7229  data: 0.0697  max mem: 41794
Epoch: [177]  [130/312]  eta: 0:02:36  lr: 0.001612  min_lr: 0.001612  loss: 3.5365 (3.3226)  weight_decay: 0.0500 (0.0500)  time: 0.8927  data: 0.1128  max mem: 41794
Epoch: [177]  [140/312]  eta: 0:02:23  lr: 0.001612  min_lr: 0.001612  loss: 3.5812 (3.3313)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.0470  max mem: 41794
Epoch: [177]  [150/312]  eta: 0:02:15  lr: 0.001611  min_lr: 0.001611  loss: 3.5812 (3.3500)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.0512  max mem: 41794
Epoch: [177]  [160/312]  eta: 0:02:05  lr: 0.001610  min_lr: 0.001610  loss: 3.4072 (3.3419)  weight_decay: 0.0500 (0.0500)  time: 0.7975  data: 0.0635  max mem: 41794
Epoch: [177]  [170/312]  eta: 0:01:55  lr: 0.001609  min_lr: 0.001609  loss: 3.2917 (3.3450)  weight_decay: 0.0500 (0.0500)  time: 0.6731  data: 0.0689  max mem: 41794
Epoch: [177]  [180/312]  eta: 0:01:47  lr: 0.001609  min_lr: 0.001609  loss: 3.3096 (3.3363)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.0966  max mem: 41794
Epoch: [177]  [190/312]  eta: 0:01:37  lr: 0.001608  min_lr: 0.001608  loss: 3.3746 (3.3457)  weight_decay: 0.0500 (0.0500)  time: 0.6737  data: 0.0443  max mem: 41794
Epoch: [177]  [200/312]  eta: 0:01:29  lr: 0.001607  min_lr: 0.001607  loss: 3.6282 (3.3522)  weight_decay: 0.0500 (0.0500)  time: 0.6656  data: 0.0451  max mem: 41794
Epoch: [177]  [210/312]  eta: 0:01:21  lr: 0.001607  min_lr: 0.001607  loss: 3.5475 (3.3502)  weight_decay: 0.0500 (0.0500)  time: 0.8039  data: 0.1228  max mem: 41794
Epoch: [177]  [220/312]  eta: 0:01:12  lr: 0.001606  min_lr: 0.001606  loss: 3.4559 (3.3545)  weight_decay: 0.0500 (0.0500)  time: 0.6814  data: 0.0820  max mem: 41794
Epoch: [177]  [230/312]  eta: 0:01:04  lr: 0.001605  min_lr: 0.001605  loss: 3.5339 (3.3549)  weight_decay: 0.0500 (0.0500)  time: 0.7034  data: 0.0773  max mem: 41794
Epoch: [177]  [240/312]  eta: 0:00:57  lr: 0.001604  min_lr: 0.001604  loss: 3.5339 (3.3499)  weight_decay: 0.0500 (0.0500)  time: 0.8263  data: 0.1718  max mem: 41794
Epoch: [177]  [250/312]  eta: 0:00:48  lr: 0.001604  min_lr: 0.001604  loss: 3.5531 (3.3522)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.0951  max mem: 41794
Epoch: [177]  [260/312]  eta: 0:00:40  lr: 0.001603  min_lr: 0.001603  loss: 3.6415 (3.3616)  weight_decay: 0.0500 (0.0500)  time: 0.6733  data: 0.0826  max mem: 41794
Epoch: [177]  [270/312]  eta: 0:00:32  lr: 0.001602  min_lr: 0.001602  loss: 3.5312 (3.3631)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0826  max mem: 41794
Epoch: [177]  [280/312]  eta: 0:00:24  lr: 0.001602  min_lr: 0.001602  loss: 3.3735 (3.3509)  weight_decay: 0.0500 (0.0500)  time: 0.7058  data: 0.0860  max mem: 41794
Epoch: [177]  [290/312]  eta: 0:00:17  lr: 0.001601  min_lr: 0.001601  loss: 3.4647 (3.3568)  weight_decay: 0.0500 (0.0500)  time: 0.8483  data: 0.1547  max mem: 41794
Epoch: [177]  [300/312]  eta: 0:00:09  lr: 0.001600  min_lr: 0.001600  loss: 3.5959 (3.3611)  weight_decay: 0.0500 (0.0500)  time: 0.6281  data: 0.0691  max mem: 41794
Epoch: [177]  [310/312]  eta: 0:00:01  lr: 0.001600  min_lr: 0.001600  loss: 3.6086 (3.3623)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [177]  [311/312]  eta: 0:00:00  lr: 0.001599  min_lr: 0.001599  loss: 3.6602 (3.3647)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [177] Total time: 0:03:58 (0.7650 s / it)
Averaged stats: lr: 0.001599  min_lr: 0.001599  loss: 3.6602 (3.3543)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:07  loss: 1.1586 (1.1586)  acc1: 79.5573 (79.5573)  acc5: 94.6615 (94.6615)  time: 7.5032  data: 7.2910  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4154 (1.3811)  acc1: 72.6562 (73.1520)  acc5: 92.1875 (91.5520)  time: 1.0333  data: 0.8612  max mem: 41794
Test: Total time: 0:00:09 (1.0424 s / it)
* Acc@1 72.966 Acc@5 91.864 loss 1.370
Accuracy of the model on the 50000 test images: 73.0%
Max accuracy: 74.56%
Epoch: [178]  [  0/312]  eta: 1:22:08  lr: 0.001599  min_lr: 0.001599  loss: 3.6704 (3.6704)  weight_decay: 0.0500 (0.0500)  time: 15.7974  data: 13.7851  max mem: 41794
Epoch: [178]  [ 10/312]  eta: 0:11:19  lr: 0.001599  min_lr: 0.001599  loss: 3.3847 (3.1324)  weight_decay: 0.0500 (0.0500)  time: 2.2496  data: 1.3667  max mem: 41794
Epoch: [178]  [ 20/312]  eta: 0:07:27  lr: 0.001598  min_lr: 0.001598  loss: 3.5020 (3.2949)  weight_decay: 0.0500 (0.0500)  time: 0.8208  data: 0.1015  max mem: 41794
Epoch: [178]  [ 30/312]  eta: 0:05:41  lr: 0.001597  min_lr: 0.001597  loss: 3.6630 (3.3702)  weight_decay: 0.0500 (0.0500)  time: 0.6409  data: 0.0462  max mem: 41794
Epoch: [178]  [ 40/312]  eta: 0:04:43  lr: 0.001597  min_lr: 0.001597  loss: 3.5040 (3.3044)  weight_decay: 0.0500 (0.0500)  time: 0.5269  data: 0.0090  max mem: 41794
Epoch: [178]  [ 50/312]  eta: 0:04:24  lr: 0.001596  min_lr: 0.001596  loss: 3.4181 (3.3422)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.0751  max mem: 41794
Epoch: [178]  [ 60/312]  eta: 0:03:53  lr: 0.001595  min_lr: 0.001595  loss: 3.5041 (3.3355)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.0870  max mem: 41794
Epoch: [178]  [ 70/312]  eta: 0:03:43  lr: 0.001594  min_lr: 0.001594  loss: 3.5276 (3.3594)  weight_decay: 0.0500 (0.0500)  time: 0.7004  data: 0.0830  max mem: 41794
Epoch: [178]  [ 80/312]  eta: 0:03:29  lr: 0.001594  min_lr: 0.001594  loss: 3.5276 (3.3573)  weight_decay: 0.0500 (0.0500)  time: 0.8234  data: 0.0852  max mem: 41794
Epoch: [178]  [ 90/312]  eta: 0:03:14  lr: 0.001593  min_lr: 0.001593  loss: 3.4005 (3.3757)  weight_decay: 0.0500 (0.0500)  time: 0.7062  data: 0.0566  max mem: 41794
Epoch: [178]  [100/312]  eta: 0:03:04  lr: 0.001592  min_lr: 0.001592  loss: 3.4846 (3.3782)  weight_decay: 0.0500 (0.0500)  time: 0.7524  data: 0.0912  max mem: 41794
Epoch: [178]  [110/312]  eta: 0:02:49  lr: 0.001592  min_lr: 0.001592  loss: 3.3814 (3.3594)  weight_decay: 0.0500 (0.0500)  time: 0.6685  data: 0.0510  max mem: 41794
Epoch: [178]  [120/312]  eta: 0:02:43  lr: 0.001591  min_lr: 0.001591  loss: 3.1847 (3.3400)  weight_decay: 0.0500 (0.0500)  time: 0.7485  data: 0.0815  max mem: 41794
Epoch: [178]  [130/312]  eta: 0:02:34  lr: 0.001590  min_lr: 0.001590  loss: 3.2538 (3.3423)  weight_decay: 0.0500 (0.0500)  time: 0.9239  data: 0.0925  max mem: 41794
Epoch: [178]  [140/312]  eta: 0:02:22  lr: 0.001590  min_lr: 0.001590  loss: 3.4192 (3.3316)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.0386  max mem: 41794
Epoch: [178]  [150/312]  eta: 0:02:14  lr: 0.001589  min_lr: 0.001589  loss: 3.4192 (3.3337)  weight_decay: 0.0500 (0.0500)  time: 0.6950  data: 0.0476  max mem: 41794
Epoch: [178]  [160/312]  eta: 0:02:05  lr: 0.001588  min_lr: 0.001588  loss: 3.5729 (3.3500)  weight_decay: 0.0500 (0.0500)  time: 0.7761  data: 0.0245  max mem: 41794
Epoch: [178]  [170/312]  eta: 0:01:55  lr: 0.001587  min_lr: 0.001587  loss: 3.5054 (3.3482)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.0326  max mem: 41794
Epoch: [178]  [180/312]  eta: 0:01:47  lr: 0.001587  min_lr: 0.001587  loss: 3.2477 (3.3348)  weight_decay: 0.0500 (0.0500)  time: 0.7585  data: 0.0441  max mem: 41794
Epoch: [178]  [190/312]  eta: 0:01:37  lr: 0.001586  min_lr: 0.001586  loss: 3.2230 (3.3374)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.0160  max mem: 41794
Epoch: [178]  [200/312]  eta: 0:01:30  lr: 0.001585  min_lr: 0.001585  loss: 3.4527 (3.3289)  weight_decay: 0.0500 (0.0500)  time: 0.7139  data: 0.0340  max mem: 41794
Epoch: [178]  [210/312]  eta: 0:01:22  lr: 0.001585  min_lr: 0.001585  loss: 3.4730 (3.3289)  weight_decay: 0.0500 (0.0500)  time: 0.8505  data: 0.0504  max mem: 41794
Epoch: [178]  [220/312]  eta: 0:01:12  lr: 0.001584  min_lr: 0.001584  loss: 3.4884 (3.3322)  weight_decay: 0.0500 (0.0500)  time: 0.6367  data: 0.0170  max mem: 41794
Epoch: [178]  [230/312]  eta: 0:01:05  lr: 0.001583  min_lr: 0.001583  loss: 3.5299 (3.3271)  weight_decay: 0.0500 (0.0500)  time: 0.6722  data: 0.0525  max mem: 41794
Epoch: [178]  [240/312]  eta: 0:00:56  lr: 0.001583  min_lr: 0.001583  loss: 3.5299 (3.3244)  weight_decay: 0.0500 (0.0500)  time: 0.7827  data: 0.0806  max mem: 41794
Epoch: [178]  [250/312]  eta: 0:00:48  lr: 0.001582  min_lr: 0.001582  loss: 3.5315 (3.3243)  weight_decay: 0.0500 (0.0500)  time: 0.6980  data: 0.0304  max mem: 41794
Epoch: [178]  [260/312]  eta: 0:00:40  lr: 0.001581  min_lr: 0.001581  loss: 3.4494 (3.3216)  weight_decay: 0.0500 (0.0500)  time: 0.7533  data: 0.0424  max mem: 41794
Epoch: [178]  [270/312]  eta: 0:00:32  lr: 0.001580  min_lr: 0.001580  loss: 3.4494 (3.3255)  weight_decay: 0.0500 (0.0500)  time: 0.6692  data: 0.0408  max mem: 41794
Epoch: [178]  [280/312]  eta: 0:00:24  lr: 0.001580  min_lr: 0.001580  loss: 3.4142 (3.3267)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.0538  max mem: 41794
Epoch: [178]  [290/312]  eta: 0:00:17  lr: 0.001579  min_lr: 0.001579  loss: 3.2076 (3.3132)  weight_decay: 0.0500 (0.0500)  time: 0.8170  data: 0.0871  max mem: 41794
Epoch: [178]  [300/312]  eta: 0:00:09  lr: 0.001578  min_lr: 0.001578  loss: 2.9945 (3.3126)  weight_decay: 0.0500 (0.0500)  time: 0.6200  data: 0.0337  max mem: 41794
Epoch: [178]  [310/312]  eta: 0:00:01  lr: 0.001578  min_lr: 0.001578  loss: 3.2668 (3.3062)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0001  max mem: 41794
Epoch: [178]  [311/312]  eta: 0:00:00  lr: 0.001578  min_lr: 0.001578  loss: 3.4590 (3.3067)  weight_decay: 0.0500 (0.0500)  time: 0.4642  data: 0.0001  max mem: 41794
Epoch: [178] Total time: 0:03:58 (0.7654 s / it)
Averaged stats: lr: 0.001578  min_lr: 0.001578  loss: 3.4590 (3.3194)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.0373 (1.0373)  acc1: 80.3385 (80.3385)  acc5: 95.3125 (95.3125)  time: 7.9353  data: 7.7221  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3069 (1.2352)  acc1: 73.3073 (74.0160)  acc5: 92.5781 (92.8000)  time: 1.0299  data: 0.8581  max mem: 41794
Test: Total time: 0:00:09 (1.0432 s / it)
* Acc@1 74.440 Acc@5 92.792 loss 1.238
Accuracy of the model on the 50000 test images: 74.4%
Max accuracy: 74.56%
Epoch: [179]  [  0/312]  eta: 1:21:54  lr: 0.001577  min_lr: 0.001577  loss: 3.6849 (3.6849)  weight_decay: 0.0500 (0.0500)  time: 15.7504  data: 11.8255  max mem: 41794
Epoch: [179]  [ 10/312]  eta: 0:11:40  lr: 0.001577  min_lr: 0.001577  loss: 3.6325 (3.3887)  weight_decay: 0.0500 (0.0500)  time: 2.3188  data: 1.2892  max mem: 41794
Epoch: [179]  [ 20/312]  eta: 0:07:14  lr: 0.001576  min_lr: 0.001576  loss: 3.5670 (3.4737)  weight_decay: 0.0500 (0.0500)  time: 0.7745  data: 0.1181  max mem: 41794
Epoch: [179]  [ 30/312]  eta: 0:05:30  lr: 0.001575  min_lr: 0.001575  loss: 3.4915 (3.3990)  weight_decay: 0.0500 (0.0500)  time: 0.5416  data: 0.0074  max mem: 41794
Epoch: [179]  [ 40/312]  eta: 0:04:34  lr: 0.001575  min_lr: 0.001575  loss: 3.4331 (3.3664)  weight_decay: 0.0500 (0.0500)  time: 0.5081  data: 0.0091  max mem: 41794
Epoch: [179]  [ 50/312]  eta: 0:04:09  lr: 0.001574  min_lr: 0.001574  loss: 3.4342 (3.3926)  weight_decay: 0.0500 (0.0500)  time: 0.6110  data: 0.0053  max mem: 41794
Epoch: [179]  [ 60/312]  eta: 0:03:42  lr: 0.001573  min_lr: 0.001573  loss: 3.3213 (3.3136)  weight_decay: 0.0500 (0.0500)  time: 0.6174  data: 0.0136  max mem: 41794
Epoch: [179]  [ 70/312]  eta: 0:03:33  lr: 0.001573  min_lr: 0.001573  loss: 3.3684 (3.3372)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.0849  max mem: 41794
Epoch: [179]  [ 80/312]  eta: 0:03:23  lr: 0.001572  min_lr: 0.001572  loss: 3.5058 (3.3553)  weight_decay: 0.0500 (0.0500)  time: 0.8675  data: 0.1184  max mem: 41794
Epoch: [179]  [ 90/312]  eta: 0:03:08  lr: 0.001571  min_lr: 0.001571  loss: 3.4385 (3.3560)  weight_decay: 0.0500 (0.0500)  time: 0.7244  data: 0.0903  max mem: 41794
Epoch: [179]  [100/312]  eta: 0:02:59  lr: 0.001570  min_lr: 0.001570  loss: 3.4795 (3.3589)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.0668  max mem: 41794
Epoch: [179]  [110/312]  eta: 0:02:46  lr: 0.001570  min_lr: 0.001570  loss: 3.4845 (3.3710)  weight_decay: 0.0500 (0.0500)  time: 0.7072  data: 0.0539  max mem: 41794
Epoch: [179]  [120/312]  eta: 0:02:37  lr: 0.001569  min_lr: 0.001569  loss: 3.4845 (3.3665)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.0609  max mem: 41794
Epoch: [179]  [130/312]  eta: 0:02:30  lr: 0.001568  min_lr: 0.001568  loss: 3.6715 (3.3871)  weight_decay: 0.0500 (0.0500)  time: 0.8486  data: 0.0326  max mem: 41794
Epoch: [179]  [140/312]  eta: 0:02:18  lr: 0.001568  min_lr: 0.001568  loss: 3.6715 (3.3817)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.0334  max mem: 41794
Epoch: [179]  [150/312]  eta: 0:02:11  lr: 0.001567  min_lr: 0.001567  loss: 3.4193 (3.3822)  weight_decay: 0.0500 (0.0500)  time: 0.7137  data: 0.0703  max mem: 41794
Epoch: [179]  [160/312]  eta: 0:02:02  lr: 0.001566  min_lr: 0.001566  loss: 3.4193 (3.3779)  weight_decay: 0.0500 (0.0500)  time: 0.7761  data: 0.0425  max mem: 41794
Epoch: [179]  [170/312]  eta: 0:01:53  lr: 0.001566  min_lr: 0.001566  loss: 3.4247 (3.3854)  weight_decay: 0.0500 (0.0500)  time: 0.6846  data: 0.0323  max mem: 41794
Epoch: [179]  [180/312]  eta: 0:01:44  lr: 0.001565  min_lr: 0.001565  loss: 3.2095 (3.3660)  weight_decay: 0.0500 (0.0500)  time: 0.7060  data: 0.0323  max mem: 41794
Epoch: [179]  [190/312]  eta: 0:01:34  lr: 0.001564  min_lr: 0.001564  loss: 3.2078 (3.3651)  weight_decay: 0.0500 (0.0500)  time: 0.6241  data: 0.0036  max mem: 41794
Epoch: [179]  [200/312]  eta: 0:01:27  lr: 0.001563  min_lr: 0.001563  loss: 3.4339 (3.3686)  weight_decay: 0.0500 (0.0500)  time: 0.6839  data: 0.0204  max mem: 41794
Epoch: [179]  [210/312]  eta: 0:01:20  lr: 0.001563  min_lr: 0.001563  loss: 3.5087 (3.3751)  weight_decay: 0.0500 (0.0500)  time: 0.8553  data: 0.0174  max mem: 41794
Epoch: [179]  [220/312]  eta: 0:01:11  lr: 0.001562  min_lr: 0.001562  loss: 3.6192 (3.3736)  weight_decay: 0.0500 (0.0500)  time: 0.7088  data: 0.0336  max mem: 41794
Epoch: [179]  [230/312]  eta: 0:01:03  lr: 0.001561  min_lr: 0.001561  loss: 3.2779 (3.3633)  weight_decay: 0.0500 (0.0500)  time: 0.6819  data: 0.0510  max mem: 41794
Epoch: [179]  [240/312]  eta: 0:00:55  lr: 0.001561  min_lr: 0.001561  loss: 3.1987 (3.3619)  weight_decay: 0.0500 (0.0500)  time: 0.7954  data: 0.0180  max mem: 41794
Epoch: [179]  [250/312]  eta: 0:00:47  lr: 0.001560  min_lr: 0.001560  loss: 3.2537 (3.3544)  weight_decay: 0.0500 (0.0500)  time: 0.7361  data: 0.0412  max mem: 41794
Epoch: [179]  [260/312]  eta: 0:00:40  lr: 0.001559  min_lr: 0.001559  loss: 3.4601 (3.3595)  weight_decay: 0.0500 (0.0500)  time: 0.7162  data: 0.0412  max mem: 41794
Epoch: [179]  [270/312]  eta: 0:00:32  lr: 0.001559  min_lr: 0.001559  loss: 3.4601 (3.3553)  weight_decay: 0.0500 (0.0500)  time: 0.6233  data: 0.0007  max mem: 41794
Epoch: [179]  [280/312]  eta: 0:00:24  lr: 0.001558  min_lr: 0.001558  loss: 3.2250 (3.3473)  weight_decay: 0.0500 (0.0500)  time: 0.6828  data: 0.0148  max mem: 41794
Epoch: [179]  [290/312]  eta: 0:00:16  lr: 0.001557  min_lr: 0.001557  loss: 3.2996 (3.3495)  weight_decay: 0.0500 (0.0500)  time: 0.8498  data: 0.0146  max mem: 41794
Epoch: [179]  [300/312]  eta: 0:00:09  lr: 0.001556  min_lr: 0.001556  loss: 3.2996 (3.3426)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0169  max mem: 41794
Epoch: [179]  [310/312]  eta: 0:00:01  lr: 0.001556  min_lr: 0.001556  loss: 3.0022 (3.3382)  weight_decay: 0.0500 (0.0500)  time: 0.4955  data: 0.0168  max mem: 41794
Epoch: [179]  [311/312]  eta: 0:00:00  lr: 0.001556  min_lr: 0.001556  loss: 3.0022 (3.3362)  weight_decay: 0.0500 (0.0500)  time: 0.4831  data: 0.0168  max mem: 41794
Epoch: [179] Total time: 0:03:56 (0.7569 s / it)
Averaged stats: lr: 0.001556  min_lr: 0.001556  loss: 3.0022 (3.3398)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 0.8573 (0.8573)  acc1: 82.8125 (82.8125)  acc5: 95.3125 (95.3125)  time: 8.0164  data: 7.8067  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2025 (1.1300)  acc1: 73.8281 (75.3760)  acc5: 93.4896 (93.0880)  time: 1.0390  data: 0.8675  max mem: 41794
Test: Total time: 0:00:09 (1.0527 s / it)
* Acc@1 75.542 Acc@5 92.972 loss 1.130
Accuracy of the model on the 50000 test images: 75.5%
Max accuracy: 75.54%
Epoch: [180]  [  0/312]  eta: 1:20:05  lr: 0.001556  min_lr: 0.001556  loss: 3.7638 (3.7638)  weight_decay: 0.0500 (0.0500)  time: 15.4012  data: 12.2405  max mem: 41794
Epoch: [180]  [ 10/312]  eta: 0:10:53  lr: 0.001555  min_lr: 0.001555  loss: 3.3031 (3.3708)  weight_decay: 0.0500 (0.0500)  time: 2.1638  data: 1.4010  max mem: 41794
Epoch: [180]  [ 20/312]  eta: 0:07:41  lr: 0.001554  min_lr: 0.001554  loss: 3.2860 (3.2935)  weight_decay: 0.0500 (0.0500)  time: 0.8897  data: 0.2804  max mem: 41794
Epoch: [180]  [ 30/312]  eta: 0:05:53  lr: 0.001553  min_lr: 0.001553  loss: 3.4863 (3.3905)  weight_decay: 0.0500 (0.0500)  time: 0.7507  data: 0.1342  max mem: 41794
Epoch: [180]  [ 40/312]  eta: 0:04:58  lr: 0.001553  min_lr: 0.001553  loss: 3.4770 (3.3583)  weight_decay: 0.0500 (0.0500)  time: 0.5881  data: 0.0714  max mem: 41794
Epoch: [180]  [ 50/312]  eta: 0:04:22  lr: 0.001552  min_lr: 0.001552  loss: 3.3441 (3.3072)  weight_decay: 0.0500 (0.0500)  time: 0.6176  data: 0.0939  max mem: 41794
Epoch: [180]  [ 60/312]  eta: 0:03:51  lr: 0.001551  min_lr: 0.001551  loss: 3.5267 (3.3369)  weight_decay: 0.0500 (0.0500)  time: 0.5583  data: 0.0351  max mem: 41794
Epoch: [180]  [ 70/312]  eta: 0:03:42  lr: 0.001551  min_lr: 0.001551  loss: 3.4533 (3.3062)  weight_decay: 0.0500 (0.0500)  time: 0.7042  data: 0.0971  max mem: 41794
Epoch: [180]  [ 80/312]  eta: 0:03:30  lr: 0.001550  min_lr: 0.001550  loss: 3.4734 (3.3503)  weight_decay: 0.0500 (0.0500)  time: 0.8615  data: 0.0972  max mem: 41794
Epoch: [180]  [ 90/312]  eta: 0:03:15  lr: 0.001549  min_lr: 0.001549  loss: 3.6471 (3.3706)  weight_decay: 0.0500 (0.0500)  time: 0.7381  data: 0.0965  max mem: 41794
Epoch: [180]  [100/312]  eta: 0:03:06  lr: 0.001549  min_lr: 0.001549  loss: 3.3425 (3.3432)  weight_decay: 0.0500 (0.0500)  time: 0.7746  data: 0.1577  max mem: 41794
Epoch: [180]  [110/312]  eta: 0:02:50  lr: 0.001548  min_lr: 0.001548  loss: 3.3773 (3.3478)  weight_decay: 0.0500 (0.0500)  time: 0.6909  data: 0.0619  max mem: 41794
Epoch: [180]  [120/312]  eta: 0:02:42  lr: 0.001547  min_lr: 0.001547  loss: 3.4523 (3.3560)  weight_decay: 0.0500 (0.0500)  time: 0.6916  data: 0.0752  max mem: 41794
Epoch: [180]  [130/312]  eta: 0:02:34  lr: 0.001546  min_lr: 0.001546  loss: 3.5496 (3.3474)  weight_decay: 0.0500 (0.0500)  time: 0.8676  data: 0.1251  max mem: 41794
Epoch: [180]  [140/312]  eta: 0:02:22  lr: 0.001546  min_lr: 0.001546  loss: 3.2826 (3.3536)  weight_decay: 0.0500 (0.0500)  time: 0.7002  data: 0.0780  max mem: 41794
Epoch: [180]  [150/312]  eta: 0:02:14  lr: 0.001545  min_lr: 0.001545  loss: 3.3059 (3.3596)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.0768  max mem: 41794
Epoch: [180]  [160/312]  eta: 0:02:05  lr: 0.001544  min_lr: 0.001544  loss: 3.7926 (3.3892)  weight_decay: 0.0500 (0.0500)  time: 0.8096  data: 0.0507  max mem: 41794
Epoch: [180]  [170/312]  eta: 0:01:55  lr: 0.001544  min_lr: 0.001544  loss: 3.6312 (3.3890)  weight_decay: 0.0500 (0.0500)  time: 0.7197  data: 0.0791  max mem: 41794
Epoch: [180]  [180/312]  eta: 0:01:47  lr: 0.001543  min_lr: 0.001543  loss: 3.5625 (3.3899)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.1139  max mem: 41794
Epoch: [180]  [190/312]  eta: 0:01:37  lr: 0.001542  min_lr: 0.001542  loss: 3.3291 (3.3767)  weight_decay: 0.0500 (0.0500)  time: 0.6272  data: 0.0382  max mem: 41794
Epoch: [180]  [200/312]  eta: 0:01:29  lr: 0.001542  min_lr: 0.001542  loss: 3.3617 (3.3863)  weight_decay: 0.0500 (0.0500)  time: 0.6741  data: 0.0614  max mem: 41794
Epoch: [180]  [210/312]  eta: 0:01:21  lr: 0.001541  min_lr: 0.001541  loss: 3.5207 (3.3858)  weight_decay: 0.0500 (0.0500)  time: 0.8673  data: 0.1378  max mem: 41794
Epoch: [180]  [220/312]  eta: 0:01:12  lr: 0.001540  min_lr: 0.001540  loss: 3.6434 (3.3967)  weight_decay: 0.0500 (0.0500)  time: 0.6913  data: 0.0784  max mem: 41794
Epoch: [180]  [230/312]  eta: 0:01:04  lr: 0.001539  min_lr: 0.001539  loss: 3.6794 (3.3994)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.0690  max mem: 41794
Epoch: [180]  [240/312]  eta: 0:00:56  lr: 0.001539  min_lr: 0.001539  loss: 3.5865 (3.4005)  weight_decay: 0.0500 (0.0500)  time: 0.7793  data: 0.0716  max mem: 41794
Epoch: [180]  [250/312]  eta: 0:00:48  lr: 0.001538  min_lr: 0.001538  loss: 3.4952 (3.3986)  weight_decay: 0.0500 (0.0500)  time: 0.6909  data: 0.1059  max mem: 41794
Epoch: [180]  [260/312]  eta: 0:00:40  lr: 0.001537  min_lr: 0.001537  loss: 3.5671 (3.4013)  weight_decay: 0.0500 (0.0500)  time: 0.7485  data: 0.1939  max mem: 41794
Epoch: [180]  [270/312]  eta: 0:00:32  lr: 0.001537  min_lr: 0.001537  loss: 3.3610 (3.3963)  weight_decay: 0.0500 (0.0500)  time: 0.6563  data: 0.0912  max mem: 41794
Epoch: [180]  [280/312]  eta: 0:00:24  lr: 0.001536  min_lr: 0.001536  loss: 3.2121 (3.3897)  weight_decay: 0.0500 (0.0500)  time: 0.7266  data: 0.1300  max mem: 41794
Epoch: [180]  [290/312]  eta: 0:00:17  lr: 0.001535  min_lr: 0.001535  loss: 3.4527 (3.3925)  weight_decay: 0.0500 (0.0500)  time: 0.8510  data: 0.1897  max mem: 41794
Epoch: [180]  [300/312]  eta: 0:00:09  lr: 0.001535  min_lr: 0.001535  loss: 3.4527 (3.3833)  weight_decay: 0.0500 (0.0500)  time: 0.6106  data: 0.0600  max mem: 41794
Epoch: [180]  [310/312]  eta: 0:00:01  lr: 0.001534  min_lr: 0.001534  loss: 3.4746 (3.3836)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0001  max mem: 41794
Epoch: [180]  [311/312]  eta: 0:00:00  lr: 0.001534  min_lr: 0.001534  loss: 3.4746 (3.3803)  weight_decay: 0.0500 (0.0500)  time: 0.4628  data: 0.0001  max mem: 41794
Epoch: [180] Total time: 0:03:58 (0.7646 s / it)
Averaged stats: lr: 0.001534  min_lr: 0.001534  loss: 3.4746 (3.3349)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.8965 (0.8965)  acc1: 81.5104 (81.5104)  acc5: 94.7917 (94.7917)  time: 8.3075  data: 8.0948  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2700 (1.1444)  acc1: 74.6094 (75.6320)  acc5: 92.3177 (92.6880)  time: 1.0722  data: 0.8995  max mem: 41794
Test: Total time: 0:00:09 (1.0917 s / it)
* Acc@1 75.580 Acc@5 92.904 loss 1.142
Accuracy of the model on the 50000 test images: 75.6%
Max accuracy: 75.58%
Epoch: [181]  [  0/312]  eta: 1:24:45  lr: 0.001534  min_lr: 0.001534  loss: 3.6632 (3.6632)  weight_decay: 0.0500 (0.0500)  time: 16.2992  data: 15.0250  max mem: 41794
Epoch: [181]  [ 10/312]  eta: 0:10:28  lr: 0.001533  min_lr: 0.001533  loss: 3.5037 (3.4444)  weight_decay: 0.0500 (0.0500)  time: 2.0823  data: 1.4301  max mem: 41794
Epoch: [181]  [ 20/312]  eta: 0:07:16  lr: 0.001532  min_lr: 0.001532  loss: 3.4806 (3.4165)  weight_decay: 0.0500 (0.0500)  time: 0.7543  data: 0.1613  max mem: 41794
Epoch: [181]  [ 30/312]  eta: 0:05:32  lr: 0.001532  min_lr: 0.001532  loss: 3.4806 (3.3702)  weight_decay: 0.0500 (0.0500)  time: 0.6839  data: 0.1263  max mem: 41794
Epoch: [181]  [ 40/312]  eta: 0:04:58  lr: 0.001531  min_lr: 0.001531  loss: 3.3610 (3.3894)  weight_decay: 0.0500 (0.0500)  time: 0.6771  data: 0.1760  max mem: 41794
Epoch: [181]  [ 50/312]  eta: 0:04:34  lr: 0.001530  min_lr: 0.001530  loss: 3.5070 (3.3925)  weight_decay: 0.0500 (0.0500)  time: 0.8471  data: 0.3430  max mem: 41794
Epoch: [181]  [ 60/312]  eta: 0:04:01  lr: 0.001530  min_lr: 0.001530  loss: 3.4805 (3.3532)  weight_decay: 0.0500 (0.0500)  time: 0.6757  data: 0.1676  max mem: 41794
Epoch: [181]  [ 70/312]  eta: 0:03:48  lr: 0.001529  min_lr: 0.001529  loss: 3.2871 (3.3399)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.1782  max mem: 41794
Epoch: [181]  [ 80/312]  eta: 0:03:37  lr: 0.001528  min_lr: 0.001528  loss: 3.4594 (3.3548)  weight_decay: 0.0500 (0.0500)  time: 0.8667  data: 0.3601  max mem: 41794
Epoch: [181]  [ 90/312]  eta: 0:03:16  lr: 0.001527  min_lr: 0.001527  loss: 3.4069 (3.3536)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.1825  max mem: 41794
Epoch: [181]  [100/312]  eta: 0:03:07  lr: 0.001527  min_lr: 0.001527  loss: 3.1332 (3.3195)  weight_decay: 0.0500 (0.0500)  time: 0.6672  data: 0.1743  max mem: 41794
Epoch: [181]  [110/312]  eta: 0:02:51  lr: 0.001526  min_lr: 0.001526  loss: 3.1353 (3.3206)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.1744  max mem: 41794
Epoch: [181]  [120/312]  eta: 0:02:43  lr: 0.001525  min_lr: 0.001525  loss: 3.4301 (3.3151)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.2116  max mem: 41794
Epoch: [181]  [130/312]  eta: 0:02:35  lr: 0.001525  min_lr: 0.001525  loss: 3.5084 (3.3243)  weight_decay: 0.0500 (0.0500)  time: 0.8953  data: 0.3931  max mem: 41794
Epoch: [181]  [140/312]  eta: 0:02:22  lr: 0.001524  min_lr: 0.001524  loss: 3.5291 (3.3256)  weight_decay: 0.0500 (0.0500)  time: 0.6766  data: 0.1822  max mem: 41794
Epoch: [181]  [150/312]  eta: 0:02:14  lr: 0.001523  min_lr: 0.001523  loss: 3.4560 (3.3287)  weight_decay: 0.0500 (0.0500)  time: 0.6550  data: 0.1621  max mem: 41794
Epoch: [181]  [160/312]  eta: 0:02:06  lr: 0.001523  min_lr: 0.001523  loss: 3.5073 (3.3279)  weight_decay: 0.0500 (0.0500)  time: 0.8394  data: 0.3321  max mem: 41794
Epoch: [181]  [170/312]  eta: 0:01:54  lr: 0.001522  min_lr: 0.001522  loss: 3.3281 (3.3242)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.1706  max mem: 41794
Epoch: [181]  [180/312]  eta: 0:01:47  lr: 0.001521  min_lr: 0.001521  loss: 3.5157 (3.3342)  weight_decay: 0.0500 (0.0500)  time: 0.6776  data: 0.1884  max mem: 41794
Epoch: [181]  [190/312]  eta: 0:01:37  lr: 0.001520  min_lr: 0.001520  loss: 3.5452 (3.3320)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.1884  max mem: 41794
Epoch: [181]  [200/312]  eta: 0:01:29  lr: 0.001520  min_lr: 0.001520  loss: 3.4049 (3.3333)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.1889  max mem: 41794
Epoch: [181]  [210/312]  eta: 0:01:21  lr: 0.001519  min_lr: 0.001519  loss: 3.4938 (3.3416)  weight_decay: 0.0500 (0.0500)  time: 0.8527  data: 0.3579  max mem: 41794
Epoch: [181]  [220/312]  eta: 0:01:12  lr: 0.001518  min_lr: 0.001518  loss: 3.5322 (3.3395)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.1699  max mem: 41794
Epoch: [181]  [230/312]  eta: 0:01:04  lr: 0.001518  min_lr: 0.001518  loss: 3.4740 (3.3370)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.2100  max mem: 41794
Epoch: [181]  [240/312]  eta: 0:00:57  lr: 0.001517  min_lr: 0.001517  loss: 3.3981 (3.3355)  weight_decay: 0.0500 (0.0500)  time: 0.8534  data: 0.3582  max mem: 41794
Epoch: [181]  [250/312]  eta: 0:00:48  lr: 0.001516  min_lr: 0.001516  loss: 3.4232 (3.3369)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1805  max mem: 41794
Epoch: [181]  [260/312]  eta: 0:00:40  lr: 0.001516  min_lr: 0.001516  loss: 3.4684 (3.3379)  weight_decay: 0.0500 (0.0500)  time: 0.7134  data: 0.2058  max mem: 41794
Epoch: [181]  [270/312]  eta: 0:00:32  lr: 0.001515  min_lr: 0.001515  loss: 3.4511 (3.3347)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1745  max mem: 41794
Epoch: [181]  [280/312]  eta: 0:00:24  lr: 0.001514  min_lr: 0.001514  loss: 3.3113 (3.3338)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.1608  max mem: 41794
Epoch: [181]  [290/312]  eta: 0:00:17  lr: 0.001513  min_lr: 0.001513  loss: 3.3269 (3.3289)  weight_decay: 0.0500 (0.0500)  time: 0.8164  data: 0.2850  max mem: 41794
Epoch: [181]  [300/312]  eta: 0:00:09  lr: 0.001513  min_lr: 0.001513  loss: 3.4718 (3.3335)  weight_decay: 0.0500 (0.0500)  time: 0.6390  data: 0.1246  max mem: 41794
Epoch: [181]  [310/312]  eta: 0:00:01  lr: 0.001512  min_lr: 0.001512  loss: 3.4718 (3.3355)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0001  max mem: 41794
Epoch: [181]  [311/312]  eta: 0:00:00  lr: 0.001512  min_lr: 0.001512  loss: 3.4718 (3.3336)  weight_decay: 0.0500 (0.0500)  time: 0.4671  data: 0.0001  max mem: 41794
Epoch: [181] Total time: 0:03:58 (0.7640 s / it)
Averaged stats: lr: 0.001512  min_lr: 0.001512  loss: 3.4718 (3.3379)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 1.0083 (1.0083)  acc1: 80.4688 (80.4688)  acc5: 95.4427 (95.4427)  time: 8.1249  data: 7.9127  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3533 (1.2747)  acc1: 73.0469 (74.3200)  acc5: 92.4479 (92.5600)  time: 1.0580  data: 0.8793  max mem: 41794
Test: Total time: 0:00:09 (1.0688 s / it)
* Acc@1 74.354 Acc@5 92.468 loss 1.282
Accuracy of the model on the 50000 test images: 74.4%
Max accuracy: 75.58%
Epoch: [182]  [  0/312]  eta: 1:19:59  lr: 0.001512  min_lr: 0.001512  loss: 3.8810 (3.8810)  weight_decay: 0.0500 (0.0500)  time: 15.3821  data: 12.9130  max mem: 41794
Epoch: [182]  [ 10/312]  eta: 0:10:45  lr: 0.001511  min_lr: 0.001511  loss: 3.5419 (3.4545)  weight_decay: 0.0500 (0.0500)  time: 2.1377  data: 1.2447  max mem: 41794
Epoch: [182]  [ 20/312]  eta: 0:07:35  lr: 0.001511  min_lr: 0.001511  loss: 3.3648 (3.3348)  weight_decay: 0.0500 (0.0500)  time: 0.8704  data: 0.0862  max mem: 41794
Epoch: [182]  [ 30/312]  eta: 0:05:49  lr: 0.001510  min_lr: 0.001510  loss: 3.4215 (3.2884)  weight_decay: 0.0500 (0.0500)  time: 0.7434  data: 0.0646  max mem: 41794
Epoch: [182]  [ 40/312]  eta: 0:04:59  lr: 0.001509  min_lr: 0.001509  loss: 3.4729 (3.3436)  weight_decay: 0.0500 (0.0500)  time: 0.6144  data: 0.0313  max mem: 41794
Epoch: [182]  [ 50/312]  eta: 0:04:29  lr: 0.001508  min_lr: 0.001508  loss: 3.5204 (3.3536)  weight_decay: 0.0500 (0.0500)  time: 0.6991  data: 0.0806  max mem: 41794
Epoch: [182]  [ 60/312]  eta: 0:03:57  lr: 0.001508  min_lr: 0.001508  loss: 3.5204 (3.3872)  weight_decay: 0.0500 (0.0500)  time: 0.6156  data: 0.0669  max mem: 41794
Epoch: [182]  [ 70/312]  eta: 0:03:46  lr: 0.001507  min_lr: 0.001507  loss: 3.5872 (3.3818)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.1417  max mem: 41794
Epoch: [182]  [ 80/312]  eta: 0:03:29  lr: 0.001506  min_lr: 0.001506  loss: 3.4648 (3.3730)  weight_decay: 0.0500 (0.0500)  time: 0.7961  data: 0.1907  max mem: 41794
Epoch: [182]  [ 90/312]  eta: 0:03:13  lr: 0.001506  min_lr: 0.001506  loss: 3.2452 (3.3352)  weight_decay: 0.0500 (0.0500)  time: 0.6523  data: 0.1123  max mem: 41794
Epoch: [182]  [100/312]  eta: 0:03:06  lr: 0.001505  min_lr: 0.001505  loss: 3.3713 (3.3443)  weight_decay: 0.0500 (0.0500)  time: 0.7656  data: 0.2138  max mem: 41794
Epoch: [182]  [110/312]  eta: 0:02:50  lr: 0.001504  min_lr: 0.001504  loss: 3.2696 (3.3166)  weight_decay: 0.0500 (0.0500)  time: 0.7106  data: 0.1512  max mem: 41794
Epoch: [182]  [120/312]  eta: 0:02:42  lr: 0.001504  min_lr: 0.001504  loss: 3.4293 (3.3359)  weight_decay: 0.0500 (0.0500)  time: 0.6767  data: 0.1015  max mem: 41794
Epoch: [182]  [130/312]  eta: 0:02:33  lr: 0.001503  min_lr: 0.001503  loss: 3.4989 (3.3403)  weight_decay: 0.0500 (0.0500)  time: 0.8433  data: 0.2572  max mem: 41794
Epoch: [182]  [140/312]  eta: 0:02:20  lr: 0.001502  min_lr: 0.001502  loss: 3.3573 (3.3425)  weight_decay: 0.0500 (0.0500)  time: 0.6709  data: 0.1565  max mem: 41794
Epoch: [182]  [150/312]  eta: 0:02:13  lr: 0.001501  min_lr: 0.001501  loss: 3.2384 (3.3291)  weight_decay: 0.0500 (0.0500)  time: 0.6771  data: 0.1587  max mem: 41794
Epoch: [182]  [160/312]  eta: 0:02:04  lr: 0.001501  min_lr: 0.001501  loss: 3.3416 (3.3300)  weight_decay: 0.0500 (0.0500)  time: 0.8190  data: 0.2997  max mem: 41794
Epoch: [182]  [170/312]  eta: 0:01:54  lr: 0.001500  min_lr: 0.001500  loss: 3.4880 (3.3204)  weight_decay: 0.0500 (0.0500)  time: 0.6921  data: 0.1545  max mem: 41794
Epoch: [182]  [180/312]  eta: 0:01:47  lr: 0.001499  min_lr: 0.001499  loss: 3.4660 (3.3215)  weight_decay: 0.0500 (0.0500)  time: 0.7451  data: 0.1374  max mem: 41794
Epoch: [182]  [190/312]  eta: 0:01:36  lr: 0.001499  min_lr: 0.001499  loss: 3.5512 (3.3221)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.1247  max mem: 41794
Epoch: [182]  [200/312]  eta: 0:01:29  lr: 0.001498  min_lr: 0.001498  loss: 3.6029 (3.3283)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.1194  max mem: 41794
Epoch: [182]  [210/312]  eta: 0:01:22  lr: 0.001497  min_lr: 0.001497  loss: 3.3378 (3.3261)  weight_decay: 0.0500 (0.0500)  time: 0.9005  data: 0.2254  max mem: 41794
Epoch: [182]  [220/312]  eta: 0:01:12  lr: 0.001497  min_lr: 0.001497  loss: 3.2440 (3.3148)  weight_decay: 0.0500 (0.0500)  time: 0.7159  data: 0.1067  max mem: 41794
Epoch: [182]  [230/312]  eta: 0:01:04  lr: 0.001496  min_lr: 0.001496  loss: 3.5021 (3.3234)  weight_decay: 0.0500 (0.0500)  time: 0.6541  data: 0.0875  max mem: 41794
Epoch: [182]  [240/312]  eta: 0:00:56  lr: 0.001495  min_lr: 0.001495  loss: 3.5021 (3.3134)  weight_decay: 0.0500 (0.0500)  time: 0.7437  data: 0.0940  max mem: 41794
Epoch: [182]  [250/312]  eta: 0:00:48  lr: 0.001495  min_lr: 0.001495  loss: 3.3219 (3.3092)  weight_decay: 0.0500 (0.0500)  time: 0.6766  data: 0.0909  max mem: 41794
Epoch: [182]  [260/312]  eta: 0:00:40  lr: 0.001494  min_lr: 0.001494  loss: 3.3974 (3.3057)  weight_decay: 0.0500 (0.0500)  time: 0.7698  data: 0.1840  max mem: 41794
Epoch: [182]  [270/312]  eta: 0:00:32  lr: 0.001493  min_lr: 0.001493  loss: 3.4773 (3.3164)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1003  max mem: 41794
Epoch: [182]  [280/312]  eta: 0:00:24  lr: 0.001492  min_lr: 0.001492  loss: 3.5964 (3.3242)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.0780  max mem: 41794
Epoch: [182]  [290/312]  eta: 0:00:17  lr: 0.001492  min_lr: 0.001492  loss: 3.5479 (3.3155)  weight_decay: 0.0500 (0.0500)  time: 0.7949  data: 0.1163  max mem: 41794
Epoch: [182]  [300/312]  eta: 0:00:09  lr: 0.001491  min_lr: 0.001491  loss: 3.2575 (3.3116)  weight_decay: 0.0500 (0.0500)  time: 0.6148  data: 0.0387  max mem: 41794
Epoch: [182]  [310/312]  eta: 0:00:01  lr: 0.001490  min_lr: 0.001490  loss: 3.2996 (3.3134)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [182]  [311/312]  eta: 0:00:00  lr: 0.001490  min_lr: 0.001490  loss: 3.3156 (3.3148)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [182] Total time: 0:03:57 (0.7619 s / it)
Averaged stats: lr: 0.001490  min_lr: 0.001490  loss: 3.3156 (3.3309)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.0335 (1.0335)  acc1: 81.9010 (81.9010)  acc5: 95.3125 (95.3125)  time: 8.0223  data: 7.8094  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3166 (1.2351)  acc1: 74.0885 (74.3680)  acc5: 92.3177 (92.5120)  time: 1.0771  data: 0.9048  max mem: 41794
Test: Total time: 0:00:09 (1.0860 s / it)
* Acc@1 74.554 Acc@5 92.572 loss 1.230
Accuracy of the model on the 50000 test images: 74.6%
Max accuracy: 75.58%
Epoch: [183]  [  0/312]  eta: 1:26:48  lr: 0.001490  min_lr: 0.001490  loss: 2.7068 (2.7068)  weight_decay: 0.0500 (0.0500)  time: 16.6948  data: 15.0434  max mem: 41794
Epoch: [183]  [ 10/312]  eta: 0:10:28  lr: 0.001490  min_lr: 0.001490  loss: 3.5706 (3.3864)  weight_decay: 0.0500 (0.0500)  time: 2.0817  data: 1.3683  max mem: 41794
Epoch: [183]  [ 20/312]  eta: 0:07:34  lr: 0.001489  min_lr: 0.001489  loss: 3.5178 (3.2911)  weight_decay: 0.0500 (0.0500)  time: 0.7997  data: 0.1195  max mem: 41794
Epoch: [183]  [ 30/312]  eta: 0:05:47  lr: 0.001488  min_lr: 0.001488  loss: 3.5687 (3.4006)  weight_decay: 0.0500 (0.0500)  time: 0.7653  data: 0.1198  max mem: 41794
Epoch: [183]  [ 40/312]  eta: 0:05:07  lr: 0.001487  min_lr: 0.001487  loss: 3.5346 (3.2804)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1023  max mem: 41794
Epoch: [183]  [ 50/312]  eta: 0:04:37  lr: 0.001487  min_lr: 0.001487  loss: 3.4657 (3.3259)  weight_decay: 0.0500 (0.0500)  time: 0.7945  data: 0.1527  max mem: 41794
Epoch: [183]  [ 60/312]  eta: 0:04:05  lr: 0.001486  min_lr: 0.001486  loss: 3.4492 (3.3296)  weight_decay: 0.0500 (0.0500)  time: 0.6506  data: 0.0678  max mem: 41794
Epoch: [183]  [ 70/312]  eta: 0:03:51  lr: 0.001485  min_lr: 0.001485  loss: 3.3460 (3.3335)  weight_decay: 0.0500 (0.0500)  time: 0.6931  data: 0.1193  max mem: 41794
Epoch: [183]  [ 80/312]  eta: 0:03:34  lr: 0.001485  min_lr: 0.001485  loss: 3.2646 (3.3093)  weight_decay: 0.0500 (0.0500)  time: 0.7800  data: 0.1263  max mem: 41794
Epoch: [183]  [ 90/312]  eta: 0:03:18  lr: 0.001484  min_lr: 0.001484  loss: 2.8683 (3.2842)  weight_decay: 0.0500 (0.0500)  time: 0.6770  data: 0.1046  max mem: 41794
Epoch: [183]  [100/312]  eta: 0:03:10  lr: 0.001483  min_lr: 0.001483  loss: 3.0936 (3.2781)  weight_decay: 0.0500 (0.0500)  time: 0.7941  data: 0.1962  max mem: 41794
Epoch: [183]  [110/312]  eta: 0:02:54  lr: 0.001483  min_lr: 0.001483  loss: 3.2787 (3.2904)  weight_decay: 0.0500 (0.0500)  time: 0.7286  data: 0.1179  max mem: 41794
Epoch: [183]  [120/312]  eta: 0:02:44  lr: 0.001482  min_lr: 0.001482  loss: 3.3556 (3.2889)  weight_decay: 0.0500 (0.0500)  time: 0.6414  data: 0.0983  max mem: 41794
Epoch: [183]  [130/312]  eta: 0:02:36  lr: 0.001481  min_lr: 0.001481  loss: 3.1353 (3.2696)  weight_decay: 0.0500 (0.0500)  time: 0.8188  data: 0.2133  max mem: 41794
Epoch: [183]  [140/312]  eta: 0:02:23  lr: 0.001480  min_lr: 0.001480  loss: 3.1278 (3.2612)  weight_decay: 0.0500 (0.0500)  time: 0.6810  data: 0.1179  max mem: 41794
Epoch: [183]  [150/312]  eta: 0:02:15  lr: 0.001480  min_lr: 0.001480  loss: 3.3127 (3.2501)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.1213  max mem: 41794
Epoch: [183]  [160/312]  eta: 0:02:05  lr: 0.001479  min_lr: 0.001479  loss: 3.3689 (3.2575)  weight_decay: 0.0500 (0.0500)  time: 0.8045  data: 0.1524  max mem: 41794
Epoch: [183]  [170/312]  eta: 0:01:56  lr: 0.001478  min_lr: 0.001478  loss: 3.5032 (3.2634)  weight_decay: 0.0500 (0.0500)  time: 0.7246  data: 0.1515  max mem: 41794
Epoch: [183]  [180/312]  eta: 0:01:47  lr: 0.001478  min_lr: 0.001478  loss: 3.5273 (3.2711)  weight_decay: 0.0500 (0.0500)  time: 0.7315  data: 0.1974  max mem: 41794
Epoch: [183]  [190/312]  eta: 0:01:37  lr: 0.001477  min_lr: 0.001477  loss: 3.5019 (3.2761)  weight_decay: 0.0500 (0.0500)  time: 0.6088  data: 0.0777  max mem: 41794
Epoch: [183]  [200/312]  eta: 0:01:29  lr: 0.001476  min_lr: 0.001476  loss: 3.4083 (3.2657)  weight_decay: 0.0500 (0.0500)  time: 0.6569  data: 0.1139  max mem: 41794
Epoch: [183]  [210/312]  eta: 0:01:22  lr: 0.001476  min_lr: 0.001476  loss: 3.4604 (3.2793)  weight_decay: 0.0500 (0.0500)  time: 0.8880  data: 0.2346  max mem: 41794
Epoch: [183]  [220/312]  eta: 0:01:13  lr: 0.001475  min_lr: 0.001475  loss: 3.5857 (3.2887)  weight_decay: 0.0500 (0.0500)  time: 0.7280  data: 0.1212  max mem: 41794
Epoch: [183]  [230/312]  eta: 0:01:05  lr: 0.001474  min_lr: 0.001474  loss: 3.5512 (3.3009)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.1043  max mem: 41794
Epoch: [183]  [240/312]  eta: 0:00:57  lr: 0.001474  min_lr: 0.001474  loss: 3.5348 (3.2973)  weight_decay: 0.0500 (0.0500)  time: 0.7979  data: 0.1095  max mem: 41794
Epoch: [183]  [250/312]  eta: 0:00:49  lr: 0.001473  min_lr: 0.001473  loss: 3.3679 (3.2978)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.0899  max mem: 41794
Epoch: [183]  [260/312]  eta: 0:00:41  lr: 0.001472  min_lr: 0.001472  loss: 3.3324 (3.2917)  weight_decay: 0.0500 (0.0500)  time: 0.7677  data: 0.1658  max mem: 41794
Epoch: [183]  [270/312]  eta: 0:00:32  lr: 0.001471  min_lr: 0.001471  loss: 3.2133 (3.2918)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.0822  max mem: 41794
Epoch: [183]  [280/312]  eta: 0:00:25  lr: 0.001471  min_lr: 0.001471  loss: 3.1807 (3.2895)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.0653  max mem: 41794
Epoch: [183]  [290/312]  eta: 0:00:17  lr: 0.001470  min_lr: 0.001470  loss: 3.4549 (3.2958)  weight_decay: 0.0500 (0.0500)  time: 0.7783  data: 0.1120  max mem: 41794
Epoch: [183]  [300/312]  eta: 0:00:09  lr: 0.001469  min_lr: 0.001469  loss: 3.4210 (3.2859)  weight_decay: 0.0500 (0.0500)  time: 0.6023  data: 0.0476  max mem: 41794
Epoch: [183]  [310/312]  eta: 0:00:01  lr: 0.001469  min_lr: 0.001469  loss: 3.3704 (3.2867)  weight_decay: 0.0500 (0.0500)  time: 0.4618  data: 0.0001  max mem: 41794
Epoch: [183]  [311/312]  eta: 0:00:00  lr: 0.001469  min_lr: 0.001469  loss: 3.3704 (3.2886)  weight_decay: 0.0500 (0.0500)  time: 0.4618  data: 0.0001  max mem: 41794
Epoch: [183] Total time: 0:03:59 (0.7682 s / it)
Averaged stats: lr: 0.001469  min_lr: 0.001469  loss: 3.3704 (3.3064)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.0724 (1.0724)  acc1: 80.8594 (80.8594)  acc5: 95.4427 (95.4427)  time: 8.6411  data: 8.4273  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2591 (1.2465)  acc1: 76.8229 (74.4480)  acc5: 92.1875 (92.7040)  time: 1.1081  data: 0.9364  max mem: 41794
Test: Total time: 0:00:10 (1.1254 s / it)
* Acc@1 74.874 Acc@5 92.726 loss 1.237
Accuracy of the model on the 50000 test images: 74.9%
Max accuracy: 75.58%
Epoch: [184]  [  0/312]  eta: 1:24:44  lr: 0.001469  min_lr: 0.001469  loss: 3.8582 (3.8582)  weight_decay: 0.0500 (0.0500)  time: 16.2949  data: 14.5340  max mem: 41794
Epoch: [184]  [ 10/312]  eta: 0:11:27  lr: 0.001468  min_lr: 0.001468  loss: 3.4583 (3.4644)  weight_decay: 0.0500 (0.0500)  time: 2.2771  data: 1.3319  max mem: 41794
Epoch: [184]  [ 20/312]  eta: 0:07:16  lr: 0.001467  min_lr: 0.001467  loss: 3.3438 (3.3223)  weight_decay: 0.0500 (0.0500)  time: 0.7556  data: 0.0420  max mem: 41794
Epoch: [184]  [ 30/312]  eta: 0:05:32  lr: 0.001466  min_lr: 0.001466  loss: 3.1794 (3.3255)  weight_decay: 0.0500 (0.0500)  time: 0.5748  data: 0.0365  max mem: 41794
Epoch: [184]  [ 40/312]  eta: 0:04:40  lr: 0.001466  min_lr: 0.001466  loss: 3.1919 (3.3305)  weight_decay: 0.0500 (0.0500)  time: 0.5409  data: 0.0061  max mem: 41794
Epoch: [184]  [ 50/312]  eta: 0:04:17  lr: 0.001465  min_lr: 0.001465  loss: 3.4567 (3.3281)  weight_decay: 0.0500 (0.0500)  time: 0.6809  data: 0.1352  max mem: 41794
Epoch: [184]  [ 60/312]  eta: 0:03:47  lr: 0.001464  min_lr: 0.001464  loss: 3.2709 (3.2686)  weight_decay: 0.0500 (0.0500)  time: 0.6390  data: 0.1297  max mem: 41794
Epoch: [184]  [ 70/312]  eta: 0:03:37  lr: 0.001464  min_lr: 0.001464  loss: 3.1690 (3.2675)  weight_decay: 0.0500 (0.0500)  time: 0.6768  data: 0.1533  max mem: 41794
Epoch: [184]  [ 80/312]  eta: 0:03:25  lr: 0.001463  min_lr: 0.001463  loss: 3.1742 (3.2494)  weight_decay: 0.0500 (0.0500)  time: 0.8392  data: 0.2970  max mem: 41794
Epoch: [184]  [ 90/312]  eta: 0:03:08  lr: 0.001462  min_lr: 0.001462  loss: 3.0624 (3.2433)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.1711  max mem: 41794
Epoch: [184]  [100/312]  eta: 0:03:02  lr: 0.001462  min_lr: 0.001462  loss: 3.1400 (3.2417)  weight_decay: 0.0500 (0.0500)  time: 0.7555  data: 0.1880  max mem: 41794
Epoch: [184]  [110/312]  eta: 0:02:47  lr: 0.001461  min_lr: 0.001461  loss: 3.4889 (3.2668)  weight_decay: 0.0500 (0.0500)  time: 0.7258  data: 0.1611  max mem: 41794
Epoch: [184]  [120/312]  eta: 0:02:41  lr: 0.001460  min_lr: 0.001460  loss: 3.5797 (3.2698)  weight_decay: 0.0500 (0.0500)  time: 0.7334  data: 0.1607  max mem: 41794
Epoch: [184]  [130/312]  eta: 0:02:31  lr: 0.001460  min_lr: 0.001460  loss: 3.3804 (3.2755)  weight_decay: 0.0500 (0.0500)  time: 0.8707  data: 0.3066  max mem: 41794
Epoch: [184]  [140/312]  eta: 0:02:19  lr: 0.001459  min_lr: 0.001459  loss: 3.4542 (3.2677)  weight_decay: 0.0500 (0.0500)  time: 0.6329  data: 0.1465  max mem: 41794
Epoch: [184]  [150/312]  eta: 0:02:12  lr: 0.001458  min_lr: 0.001458  loss: 3.4326 (3.2716)  weight_decay: 0.0500 (0.0500)  time: 0.7016  data: 0.1991  max mem: 41794
Epoch: [184]  [160/312]  eta: 0:02:04  lr: 0.001457  min_lr: 0.001457  loss: 3.4761 (3.2763)  weight_decay: 0.0500 (0.0500)  time: 0.8580  data: 0.3537  max mem: 41794
Epoch: [184]  [170/312]  eta: 0:01:53  lr: 0.001457  min_lr: 0.001457  loss: 3.5338 (3.2822)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.1834  max mem: 41794
Epoch: [184]  [180/312]  eta: 0:01:45  lr: 0.001456  min_lr: 0.001456  loss: 3.5317 (3.3018)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.1854  max mem: 41794
Epoch: [184]  [190/312]  eta: 0:01:35  lr: 0.001455  min_lr: 0.001455  loss: 3.5365 (3.3077)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.1571  max mem: 41794
Epoch: [184]  [200/312]  eta: 0:01:28  lr: 0.001455  min_lr: 0.001455  loss: 3.5363 (3.3202)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1991  max mem: 41794
Epoch: [184]  [210/312]  eta: 0:01:20  lr: 0.001454  min_lr: 0.001454  loss: 3.4676 (3.3286)  weight_decay: 0.0500 (0.0500)  time: 0.8643  data: 0.3534  max mem: 41794
Epoch: [184]  [220/312]  eta: 0:01:11  lr: 0.001453  min_lr: 0.001453  loss: 3.4554 (3.3278)  weight_decay: 0.0500 (0.0500)  time: 0.6536  data: 0.1549  max mem: 41794
Epoch: [184]  [230/312]  eta: 0:01:04  lr: 0.001453  min_lr: 0.001453  loss: 3.5067 (3.3341)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.1668  max mem: 41794
Epoch: [184]  [240/312]  eta: 0:00:56  lr: 0.001452  min_lr: 0.001452  loss: 3.5143 (3.3327)  weight_decay: 0.0500 (0.0500)  time: 0.8309  data: 0.2990  max mem: 41794
Epoch: [184]  [250/312]  eta: 0:00:48  lr: 0.001451  min_lr: 0.001451  loss: 3.5080 (3.3321)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1464  max mem: 41794
Epoch: [184]  [260/312]  eta: 0:00:40  lr: 0.001451  min_lr: 0.001451  loss: 3.4897 (3.3309)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.1680  max mem: 41794
Epoch: [184]  [270/312]  eta: 0:00:32  lr: 0.001450  min_lr: 0.001450  loss: 3.4168 (3.3330)  weight_decay: 0.0500 (0.0500)  time: 0.6802  data: 0.1544  max mem: 41794
Epoch: [184]  [280/312]  eta: 0:00:24  lr: 0.001449  min_lr: 0.001449  loss: 3.6012 (3.3407)  weight_decay: 0.0500 (0.0500)  time: 0.6881  data: 0.1546  max mem: 41794
Epoch: [184]  [290/312]  eta: 0:00:17  lr: 0.001448  min_lr: 0.001448  loss: 3.6323 (3.3449)  weight_decay: 0.0500 (0.0500)  time: 0.8881  data: 0.3268  max mem: 41794
Epoch: [184]  [300/312]  eta: 0:00:09  lr: 0.001448  min_lr: 0.001448  loss: 3.3100 (3.3378)  weight_decay: 0.0500 (0.0500)  time: 0.6869  data: 0.1726  max mem: 41794
Epoch: [184]  [310/312]  eta: 0:00:01  lr: 0.001447  min_lr: 0.001447  loss: 3.1241 (3.3309)  weight_decay: 0.0500 (0.0500)  time: 0.4683  data: 0.0001  max mem: 41794
Epoch: [184]  [311/312]  eta: 0:00:00  lr: 0.001447  min_lr: 0.001447  loss: 3.1241 (3.3322)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [184] Total time: 0:03:57 (0.7621 s / it)
Averaged stats: lr: 0.001447  min_lr: 0.001447  loss: 3.1241 (3.3092)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1227 (1.1227)  acc1: 80.5990 (80.5990)  acc5: 94.7917 (94.7917)  time: 8.3202  data: 8.1112  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3524 (1.2421)  acc1: 74.0885 (74.8960)  acc5: 92.1875 (92.8480)  time: 1.0725  data: 0.9013  max mem: 41794
Test: Total time: 0:00:09 (1.0887 s / it)
* Acc@1 75.012 Acc@5 92.974 loss 1.241
Accuracy of the model on the 50000 test images: 75.0%
Max accuracy: 75.58%
Epoch: [185]  [  0/312]  eta: 1:21:54  lr: 0.001447  min_lr: 0.001447  loss: 2.5208 (2.5208)  weight_decay: 0.0500 (0.0500)  time: 15.7518  data: 14.9038  max mem: 41794
Epoch: [185]  [ 10/312]  eta: 0:10:24  lr: 0.001446  min_lr: 0.001446  loss: 3.5015 (3.3081)  weight_decay: 0.0500 (0.0500)  time: 2.0679  data: 1.3556  max mem: 41794
Epoch: [185]  [ 20/312]  eta: 0:07:05  lr: 0.001446  min_lr: 0.001446  loss: 3.3807 (3.2069)  weight_decay: 0.0500 (0.0500)  time: 0.7432  data: 0.1432  max mem: 41794
Epoch: [185]  [ 30/312]  eta: 0:05:25  lr: 0.001445  min_lr: 0.001445  loss: 3.3807 (3.2826)  weight_decay: 0.0500 (0.0500)  time: 0.6545  data: 0.1432  max mem: 41794
Epoch: [185]  [ 40/312]  eta: 0:04:51  lr: 0.001444  min_lr: 0.001444  loss: 3.4198 (3.2705)  weight_decay: 0.0500 (0.0500)  time: 0.6644  data: 0.1279  max mem: 41794
Epoch: [185]  [ 50/312]  eta: 0:04:29  lr: 0.001444  min_lr: 0.001444  loss: 3.3987 (3.3118)  weight_decay: 0.0500 (0.0500)  time: 0.8332  data: 0.2683  max mem: 41794
Epoch: [185]  [ 60/312]  eta: 0:03:57  lr: 0.001443  min_lr: 0.001443  loss: 3.3654 (3.3098)  weight_decay: 0.0500 (0.0500)  time: 0.6743  data: 0.1412  max mem: 41794
Epoch: [185]  [ 70/312]  eta: 0:03:46  lr: 0.001442  min_lr: 0.001442  loss: 3.3805 (3.3308)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.1752  max mem: 41794
Epoch: [185]  [ 80/312]  eta: 0:03:31  lr: 0.001441  min_lr: 0.001441  loss: 3.5353 (3.3457)  weight_decay: 0.0500 (0.0500)  time: 0.8250  data: 0.2792  max mem: 41794
Epoch: [185]  [ 90/312]  eta: 0:03:14  lr: 0.001441  min_lr: 0.001441  loss: 3.4526 (3.3322)  weight_decay: 0.0500 (0.0500)  time: 0.6675  data: 0.1661  max mem: 41794
Epoch: [185]  [100/312]  eta: 0:03:07  lr: 0.001440  min_lr: 0.001440  loss: 3.4526 (3.3212)  weight_decay: 0.0500 (0.0500)  time: 0.7785  data: 0.2456  max mem: 41794
Epoch: [185]  [110/312]  eta: 0:02:52  lr: 0.001439  min_lr: 0.001439  loss: 3.1272 (3.3063)  weight_decay: 0.0500 (0.0500)  time: 0.7322  data: 0.1858  max mem: 41794
Epoch: [185]  [120/312]  eta: 0:02:44  lr: 0.001439  min_lr: 0.001439  loss: 3.2820 (3.2948)  weight_decay: 0.0500 (0.0500)  time: 0.7078  data: 0.1376  max mem: 41794
Epoch: [185]  [130/312]  eta: 0:02:36  lr: 0.001438  min_lr: 0.001438  loss: 3.3407 (3.3070)  weight_decay: 0.0500 (0.0500)  time: 0.8978  data: 0.2507  max mem: 41794
Epoch: [185]  [140/312]  eta: 0:02:23  lr: 0.001437  min_lr: 0.001437  loss: 3.3262 (3.2927)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.1175  max mem: 41794
Epoch: [185]  [150/312]  eta: 0:02:15  lr: 0.001437  min_lr: 0.001437  loss: 3.1488 (3.2783)  weight_decay: 0.0500 (0.0500)  time: 0.6962  data: 0.1209  max mem: 41794
Epoch: [185]  [160/312]  eta: 0:02:05  lr: 0.001436  min_lr: 0.001436  loss: 2.8754 (3.2571)  weight_decay: 0.0500 (0.0500)  time: 0.7842  data: 0.1396  max mem: 41794
Epoch: [185]  [170/312]  eta: 0:01:55  lr: 0.001435  min_lr: 0.001435  loss: 3.2108 (3.2687)  weight_decay: 0.0500 (0.0500)  time: 0.6540  data: 0.0962  max mem: 41794
Epoch: [185]  [180/312]  eta: 0:01:47  lr: 0.001435  min_lr: 0.001435  loss: 3.4168 (3.2663)  weight_decay: 0.0500 (0.0500)  time: 0.7426  data: 0.1836  max mem: 41794
Epoch: [185]  [190/312]  eta: 0:01:37  lr: 0.001434  min_lr: 0.001434  loss: 3.3676 (3.2641)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.1103  max mem: 41794
Epoch: [185]  [200/312]  eta: 0:01:30  lr: 0.001433  min_lr: 0.001433  loss: 3.3751 (3.2616)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.0805  max mem: 41794
Epoch: [185]  [210/312]  eta: 0:01:22  lr: 0.001432  min_lr: 0.001432  loss: 3.3514 (3.2577)  weight_decay: 0.0500 (0.0500)  time: 0.8680  data: 0.1720  max mem: 41794
Epoch: [185]  [220/312]  eta: 0:01:13  lr: 0.001432  min_lr: 0.001432  loss: 3.4008 (3.2603)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.0936  max mem: 41794
Epoch: [185]  [230/312]  eta: 0:01:05  lr: 0.001431  min_lr: 0.001431  loss: 3.5319 (3.2713)  weight_decay: 0.0500 (0.0500)  time: 0.6906  data: 0.0959  max mem: 41794
Epoch: [185]  [240/312]  eta: 0:00:57  lr: 0.001430  min_lr: 0.001430  loss: 3.5319 (3.2761)  weight_decay: 0.0500 (0.0500)  time: 0.7890  data: 0.1249  max mem: 41794
Epoch: [185]  [250/312]  eta: 0:00:48  lr: 0.001430  min_lr: 0.001430  loss: 3.4253 (3.2791)  weight_decay: 0.0500 (0.0500)  time: 0.6653  data: 0.0976  max mem: 41794
Epoch: [185]  [260/312]  eta: 0:00:41  lr: 0.001429  min_lr: 0.001429  loss: 3.3920 (3.2786)  weight_decay: 0.0500 (0.0500)  time: 0.7708  data: 0.1699  max mem: 41794
Epoch: [185]  [270/312]  eta: 0:00:32  lr: 0.001428  min_lr: 0.001428  loss: 3.3075 (3.2778)  weight_decay: 0.0500 (0.0500)  time: 0.7088  data: 0.1020  max mem: 41794
Epoch: [185]  [280/312]  eta: 0:00:25  lr: 0.001428  min_lr: 0.001428  loss: 3.1635 (3.2747)  weight_decay: 0.0500 (0.0500)  time: 0.6987  data: 0.0847  max mem: 41794
Epoch: [185]  [290/312]  eta: 0:00:17  lr: 0.001427  min_lr: 0.001427  loss: 3.2237 (3.2788)  weight_decay: 0.0500 (0.0500)  time: 0.7721  data: 0.1171  max mem: 41794
Epoch: [185]  [300/312]  eta: 0:00:09  lr: 0.001426  min_lr: 0.001426  loss: 3.2896 (3.2799)  weight_decay: 0.0500 (0.0500)  time: 0.5594  data: 0.0328  max mem: 41794
Epoch: [185]  [310/312]  eta: 0:00:01  lr: 0.001426  min_lr: 0.001426  loss: 3.3000 (3.2798)  weight_decay: 0.0500 (0.0500)  time: 0.4615  data: 0.0001  max mem: 41794
Epoch: [185]  [311/312]  eta: 0:00:00  lr: 0.001426  min_lr: 0.001426  loss: 3.3000 (3.2813)  weight_decay: 0.0500 (0.0500)  time: 0.4609  data: 0.0001  max mem: 41794
Epoch: [185] Total time: 0:03:59 (0.7662 s / it)
Averaged stats: lr: 0.001426  min_lr: 0.001426  loss: 3.3000 (3.3168)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.0125 (1.0125)  acc1: 80.9896 (80.9896)  acc5: 95.7031 (95.7031)  time: 8.4051  data: 8.2033  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4328 (1.3082)  acc1: 73.3073 (75.3440)  acc5: 92.1875 (92.7520)  time: 1.0883  data: 0.9116  max mem: 41794
Test: Total time: 0:00:09 (1.1016 s / it)
* Acc@1 75.374 Acc@5 92.858 loss 1.304
Accuracy of the model on the 50000 test images: 75.4%
Max accuracy: 75.58%
Epoch: [186]  [  0/312]  eta: 1:21:58  lr: 0.001425  min_lr: 0.001425  loss: 3.9352 (3.9352)  weight_decay: 0.0500 (0.0500)  time: 15.7643  data: 13.2396  max mem: 41794
Epoch: [186]  [ 10/312]  eta: 0:10:59  lr: 0.001425  min_lr: 0.001425  loss: 3.4886 (3.5280)  weight_decay: 0.0500 (0.0500)  time: 2.1832  data: 1.3195  max mem: 41794
Epoch: [186]  [ 20/312]  eta: 0:07:13  lr: 0.001424  min_lr: 0.001424  loss: 3.5411 (3.5489)  weight_decay: 0.0500 (0.0500)  time: 0.7716  data: 0.1268  max mem: 41794
Epoch: [186]  [ 30/312]  eta: 0:05:30  lr: 0.001423  min_lr: 0.001423  loss: 3.5411 (3.5104)  weight_decay: 0.0500 (0.0500)  time: 0.6144  data: 0.0635  max mem: 41794
Epoch: [186]  [ 40/312]  eta: 0:04:49  lr: 0.001423  min_lr: 0.001423  loss: 3.4909 (3.4919)  weight_decay: 0.0500 (0.0500)  time: 0.6195  data: 0.0113  max mem: 41794
Epoch: [186]  [ 50/312]  eta: 0:04:33  lr: 0.001422  min_lr: 0.001422  loss: 3.2752 (3.4339)  weight_decay: 0.0500 (0.0500)  time: 0.8482  data: 0.0352  max mem: 41794
Epoch: [186]  [ 60/312]  eta: 0:04:01  lr: 0.001421  min_lr: 0.001421  loss: 3.2085 (3.3916)  weight_decay: 0.0500 (0.0500)  time: 0.7406  data: 0.0392  max mem: 41794
Epoch: [186]  [ 70/312]  eta: 0:03:46  lr: 0.001421  min_lr: 0.001421  loss: 3.3839 (3.4151)  weight_decay: 0.0500 (0.0500)  time: 0.6617  data: 0.0342  max mem: 41794
Epoch: [186]  [ 80/312]  eta: 0:03:33  lr: 0.001420  min_lr: 0.001420  loss: 3.5473 (3.4177)  weight_decay: 0.0500 (0.0500)  time: 0.8104  data: 0.0219  max mem: 41794
Epoch: [186]  [ 90/312]  eta: 0:03:16  lr: 0.001419  min_lr: 0.001419  loss: 3.2233 (3.3818)  weight_decay: 0.0500 (0.0500)  time: 0.7123  data: 0.0332  max mem: 41794
Epoch: [186]  [100/312]  eta: 0:03:07  lr: 0.001419  min_lr: 0.001419  loss: 3.2959 (3.3953)  weight_decay: 0.0500 (0.0500)  time: 0.7276  data: 0.0655  max mem: 41794
Epoch: [186]  [110/312]  eta: 0:02:51  lr: 0.001418  min_lr: 0.001418  loss: 3.4133 (3.3823)  weight_decay: 0.0500 (0.0500)  time: 0.6757  data: 0.0364  max mem: 41794
Epoch: [186]  [120/312]  eta: 0:02:43  lr: 0.001417  min_lr: 0.001417  loss: 2.9069 (3.3305)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.0493  max mem: 41794
Epoch: [186]  [130/312]  eta: 0:02:35  lr: 0.001416  min_lr: 0.001416  loss: 3.0750 (3.3272)  weight_decay: 0.0500 (0.0500)  time: 0.8899  data: 0.0960  max mem: 41794
Epoch: [186]  [140/312]  eta: 0:02:22  lr: 0.001416  min_lr: 0.001416  loss: 3.2955 (3.3175)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.0486  max mem: 41794
Epoch: [186]  [150/312]  eta: 0:02:14  lr: 0.001415  min_lr: 0.001415  loss: 3.2352 (3.3161)  weight_decay: 0.0500 (0.0500)  time: 0.6489  data: 0.0448  max mem: 41794
Epoch: [186]  [160/312]  eta: 0:02:05  lr: 0.001414  min_lr: 0.001414  loss: 3.5206 (3.3172)  weight_decay: 0.0500 (0.0500)  time: 0.7891  data: 0.0448  max mem: 41794
Epoch: [186]  [170/312]  eta: 0:01:55  lr: 0.001414  min_lr: 0.001414  loss: 3.5449 (3.3291)  weight_decay: 0.0500 (0.0500)  time: 0.7123  data: 0.0778  max mem: 41794
Epoch: [186]  [180/312]  eta: 0:01:47  lr: 0.001413  min_lr: 0.001413  loss: 3.5368 (3.3202)  weight_decay: 0.0500 (0.0500)  time: 0.7377  data: 0.1208  max mem: 41794
Epoch: [186]  [190/312]  eta: 0:01:37  lr: 0.001412  min_lr: 0.001412  loss: 3.0810 (3.3188)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0437  max mem: 41794
Epoch: [186]  [200/312]  eta: 0:01:29  lr: 0.001412  min_lr: 0.001412  loss: 3.3937 (3.3189)  weight_decay: 0.0500 (0.0500)  time: 0.6595  data: 0.0567  max mem: 41794
Epoch: [186]  [210/312]  eta: 0:01:21  lr: 0.001411  min_lr: 0.001411  loss: 3.4690 (3.3207)  weight_decay: 0.0500 (0.0500)  time: 0.8518  data: 0.1287  max mem: 41794
Epoch: [186]  [220/312]  eta: 0:01:12  lr: 0.001410  min_lr: 0.001410  loss: 3.4331 (3.3160)  weight_decay: 0.0500 (0.0500)  time: 0.6888  data: 0.0726  max mem: 41794
Epoch: [186]  [230/312]  eta: 0:01:04  lr: 0.001410  min_lr: 0.001410  loss: 3.4331 (3.3167)  weight_decay: 0.0500 (0.0500)  time: 0.6595  data: 0.0753  max mem: 41794
Epoch: [186]  [240/312]  eta: 0:00:56  lr: 0.001409  min_lr: 0.001409  loss: 3.4962 (3.3198)  weight_decay: 0.0500 (0.0500)  time: 0.7577  data: 0.0766  max mem: 41794
Epoch: [186]  [250/312]  eta: 0:00:48  lr: 0.001408  min_lr: 0.001408  loss: 3.6009 (3.3254)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.0718  max mem: 41794
Epoch: [186]  [260/312]  eta: 0:00:40  lr: 0.001408  min_lr: 0.001408  loss: 3.2288 (3.3088)  weight_decay: 0.0500 (0.0500)  time: 0.7638  data: 0.1186  max mem: 41794
Epoch: [186]  [270/312]  eta: 0:00:32  lr: 0.001407  min_lr: 0.001407  loss: 3.3000 (3.3159)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.0524  max mem: 41794
Epoch: [186]  [280/312]  eta: 0:00:24  lr: 0.001406  min_lr: 0.001406  loss: 3.5868 (3.3175)  weight_decay: 0.0500 (0.0500)  time: 0.7036  data: 0.0704  max mem: 41794
Epoch: [186]  [290/312]  eta: 0:00:17  lr: 0.001406  min_lr: 0.001406  loss: 3.5107 (3.3260)  weight_decay: 0.0500 (0.0500)  time: 0.7858  data: 0.0892  max mem: 41794
Epoch: [186]  [300/312]  eta: 0:00:09  lr: 0.001405  min_lr: 0.001405  loss: 3.3998 (3.3231)  weight_decay: 0.0500 (0.0500)  time: 0.5988  data: 0.0229  max mem: 41794
Epoch: [186]  [310/312]  eta: 0:00:01  lr: 0.001404  min_lr: 0.001404  loss: 3.2666 (3.3246)  weight_decay: 0.0500 (0.0500)  time: 0.4671  data: 0.0001  max mem: 41794
Epoch: [186]  [311/312]  eta: 0:00:00  lr: 0.001404  min_lr: 0.001404  loss: 3.2666 (3.3236)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.0001  max mem: 41794
Epoch: [186] Total time: 0:03:58 (0.7632 s / it)
Averaged stats: lr: 0.001404  min_lr: 0.001404  loss: 3.2666 (3.3157)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.1561 (1.1561)  acc1: 81.6406 (81.6406)  acc5: 95.4427 (95.4427)  time: 8.2349  data: 8.0220  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3761 (1.3278)  acc1: 73.8281 (74.5440)  acc5: 92.1875 (92.7040)  time: 1.0864  data: 0.9083  max mem: 41794
Test: Total time: 0:00:09 (1.0960 s / it)
* Acc@1 75.130 Acc@5 92.876 loss 1.316
Accuracy of the model on the 50000 test images: 75.1%
Max accuracy: 75.58%
Epoch: [187]  [  0/312]  eta: 1:24:24  lr: 0.001404  min_lr: 0.001404  loss: 3.2298 (3.2298)  weight_decay: 0.0500 (0.0500)  time: 16.2325  data: 14.0702  max mem: 41794
Epoch: [187]  [ 10/312]  eta: 0:11:13  lr: 0.001403  min_lr: 0.001403  loss: 3.2298 (3.1480)  weight_decay: 0.0500 (0.0500)  time: 2.2302  data: 1.3664  max mem: 41794
Epoch: [187]  [ 20/312]  eta: 0:07:29  lr: 0.001403  min_lr: 0.001403  loss: 3.2355 (3.1685)  weight_decay: 0.0500 (0.0500)  time: 0.8040  data: 0.0966  max mem: 41794
Epoch: [187]  [ 30/312]  eta: 0:05:40  lr: 0.001402  min_lr: 0.001402  loss: 3.2355 (3.1627)  weight_decay: 0.0500 (0.0500)  time: 0.6474  data: 0.0497  max mem: 41794
Epoch: [187]  [ 40/312]  eta: 0:04:44  lr: 0.001401  min_lr: 0.001401  loss: 3.2387 (3.1923)  weight_decay: 0.0500 (0.0500)  time: 0.5281  data: 0.0014  max mem: 41794
Epoch: [187]  [ 50/312]  eta: 0:04:24  lr: 0.001401  min_lr: 0.001401  loss: 3.1715 (3.1655)  weight_decay: 0.0500 (0.0500)  time: 0.7022  data: 0.0785  max mem: 41794
Epoch: [187]  [ 60/312]  eta: 0:03:53  lr: 0.001400  min_lr: 0.001400  loss: 2.9617 (3.1667)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.0898  max mem: 41794
Epoch: [187]  [ 70/312]  eta: 0:03:43  lr: 0.001399  min_lr: 0.001399  loss: 3.4035 (3.1705)  weight_decay: 0.0500 (0.0500)  time: 0.7028  data: 0.1323  max mem: 41794
Epoch: [187]  [ 80/312]  eta: 0:03:31  lr: 0.001399  min_lr: 0.001399  loss: 3.1471 (3.1669)  weight_decay: 0.0500 (0.0500)  time: 0.8560  data: 0.2267  max mem: 41794
Epoch: [187]  [ 90/312]  eta: 0:03:12  lr: 0.001398  min_lr: 0.001398  loss: 3.5053 (3.2272)  weight_decay: 0.0500 (0.0500)  time: 0.6646  data: 0.1185  max mem: 41794
Epoch: [187]  [100/312]  eta: 0:03:03  lr: 0.001397  min_lr: 0.001397  loss: 3.6753 (3.2474)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.1110  max mem: 41794
Epoch: [187]  [110/312]  eta: 0:02:47  lr: 0.001396  min_lr: 0.001396  loss: 3.3953 (3.2368)  weight_decay: 0.0500 (0.0500)  time: 0.6694  data: 0.0987  max mem: 41794
Epoch: [187]  [120/312]  eta: 0:02:41  lr: 0.001396  min_lr: 0.001396  loss: 3.3953 (3.2631)  weight_decay: 0.0500 (0.0500)  time: 0.7207  data: 0.1311  max mem: 41794
Epoch: [187]  [130/312]  eta: 0:02:33  lr: 0.001395  min_lr: 0.001395  loss: 3.6081 (3.2811)  weight_decay: 0.0500 (0.0500)  time: 0.9250  data: 0.2335  max mem: 41794
Epoch: [187]  [140/312]  eta: 0:02:21  lr: 0.001394  min_lr: 0.001394  loss: 3.4932 (3.2750)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.1043  max mem: 41794
Epoch: [187]  [150/312]  eta: 0:02:13  lr: 0.001394  min_lr: 0.001394  loss: 3.5282 (3.2886)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.1018  max mem: 41794
Epoch: [187]  [160/312]  eta: 0:02:05  lr: 0.001393  min_lr: 0.001393  loss: 3.5282 (3.2949)  weight_decay: 0.0500 (0.0500)  time: 0.8623  data: 0.1934  max mem: 41794
Epoch: [187]  [170/312]  eta: 0:01:54  lr: 0.001392  min_lr: 0.001392  loss: 3.3971 (3.2978)  weight_decay: 0.0500 (0.0500)  time: 0.6817  data: 0.0951  max mem: 41794
Epoch: [187]  [180/312]  eta: 0:01:46  lr: 0.001392  min_lr: 0.001392  loss: 3.2212 (3.2874)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.0947  max mem: 41794
Epoch: [187]  [190/312]  eta: 0:01:36  lr: 0.001391  min_lr: 0.001391  loss: 3.2212 (3.2862)  weight_decay: 0.0500 (0.0500)  time: 0.6439  data: 0.0943  max mem: 41794
Epoch: [187]  [200/312]  eta: 0:01:29  lr: 0.001390  min_lr: 0.001390  loss: 3.5136 (3.2853)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.1181  max mem: 41794
Epoch: [187]  [210/312]  eta: 0:01:21  lr: 0.001390  min_lr: 0.001390  loss: 3.3196 (3.2730)  weight_decay: 0.0500 (0.0500)  time: 0.8645  data: 0.2297  max mem: 41794
Epoch: [187]  [220/312]  eta: 0:01:12  lr: 0.001389  min_lr: 0.001389  loss: 3.1606 (3.2680)  weight_decay: 0.0500 (0.0500)  time: 0.6683  data: 0.1133  max mem: 41794
Epoch: [187]  [230/312]  eta: 0:01:04  lr: 0.001388  min_lr: 0.001388  loss: 3.1627 (3.2678)  weight_decay: 0.0500 (0.0500)  time: 0.6886  data: 0.1105  max mem: 41794
Epoch: [187]  [240/312]  eta: 0:00:56  lr: 0.001388  min_lr: 0.001388  loss: 3.3730 (3.2677)  weight_decay: 0.0500 (0.0500)  time: 0.8421  data: 0.2077  max mem: 41794
Epoch: [187]  [250/312]  eta: 0:00:48  lr: 0.001387  min_lr: 0.001387  loss: 3.3336 (3.2623)  weight_decay: 0.0500 (0.0500)  time: 0.6985  data: 0.0993  max mem: 41794
Epoch: [187]  [260/312]  eta: 0:00:40  lr: 0.001386  min_lr: 0.001386  loss: 3.0088 (3.2570)  weight_decay: 0.0500 (0.0500)  time: 0.7567  data: 0.0676  max mem: 41794
Epoch: [187]  [270/312]  eta: 0:00:32  lr: 0.001385  min_lr: 0.001385  loss: 3.3279 (3.2638)  weight_decay: 0.0500 (0.0500)  time: 0.7095  data: 0.0662  max mem: 41794
Epoch: [187]  [280/312]  eta: 0:00:24  lr: 0.001385  min_lr: 0.001385  loss: 3.3279 (3.2505)  weight_decay: 0.0500 (0.0500)  time: 0.6519  data: 0.0382  max mem: 41794
Epoch: [187]  [290/312]  eta: 0:00:17  lr: 0.001384  min_lr: 0.001384  loss: 3.0300 (3.2524)  weight_decay: 0.0500 (0.0500)  time: 0.7916  data: 0.0642  max mem: 41794
Epoch: [187]  [300/312]  eta: 0:00:09  lr: 0.001383  min_lr: 0.001383  loss: 3.3786 (3.2534)  weight_decay: 0.0500 (0.0500)  time: 0.6331  data: 0.0355  max mem: 41794
Epoch: [187]  [310/312]  eta: 0:00:01  lr: 0.001383  min_lr: 0.001383  loss: 3.2533 (3.2521)  weight_decay: 0.0500 (0.0500)  time: 0.4739  data: 0.0092  max mem: 41794
Epoch: [187]  [311/312]  eta: 0:00:00  lr: 0.001383  min_lr: 0.001383  loss: 3.2533 (3.2536)  weight_decay: 0.0500 (0.0500)  time: 0.4725  data: 0.0092  max mem: 41794
Epoch: [187] Total time: 0:03:58 (0.7638 s / it)
Averaged stats: lr: 0.001383  min_lr: 0.001383  loss: 3.2533 (3.3026)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.8922 (0.8922)  acc1: 81.5104 (81.5104)  acc5: 95.4427 (95.4427)  time: 8.2268  data: 8.0142  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2517 (1.1664)  acc1: 73.1771 (74.6400)  acc5: 93.4896 (92.5920)  time: 1.0621  data: 0.8906  max mem: 41794
Test: Total time: 0:00:09 (1.0746 s / it)
* Acc@1 74.928 Acc@5 92.718 loss 1.151
Accuracy of the model on the 50000 test images: 74.9%
Max accuracy: 75.58%
Epoch: [188]  [  0/312]  eta: 1:23:10  lr: 0.001383  min_lr: 0.001383  loss: 3.4391 (3.4391)  weight_decay: 0.0500 (0.0500)  time: 15.9939  data: 13.4341  max mem: 41794
Epoch: [188]  [ 10/312]  eta: 0:10:59  lr: 0.001382  min_lr: 0.001382  loss: 3.4304 (3.3238)  weight_decay: 0.0500 (0.0500)  time: 2.1829  data: 1.2974  max mem: 41794
Epoch: [188]  [ 20/312]  eta: 0:07:12  lr: 0.001381  min_lr: 0.001381  loss: 3.4304 (3.4078)  weight_decay: 0.0500 (0.0500)  time: 0.7555  data: 0.1351  max mem: 41794
Epoch: [188]  [ 30/312]  eta: 0:05:29  lr: 0.001381  min_lr: 0.001381  loss: 3.5100 (3.3824)  weight_decay: 0.0500 (0.0500)  time: 0.6124  data: 0.0971  max mem: 41794
Epoch: [188]  [ 40/312]  eta: 0:04:44  lr: 0.001380  min_lr: 0.001380  loss: 3.4858 (3.3619)  weight_decay: 0.0500 (0.0500)  time: 0.5901  data: 0.0735  max mem: 41794
Epoch: [188]  [ 50/312]  eta: 0:04:23  lr: 0.001379  min_lr: 0.001379  loss: 3.4814 (3.3577)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.1784  max mem: 41794
Epoch: [188]  [ 60/312]  eta: 0:03:51  lr: 0.001379  min_lr: 0.001379  loss: 3.3669 (3.3518)  weight_decay: 0.0500 (0.0500)  time: 0.6612  data: 0.1091  max mem: 41794
Epoch: [188]  [ 70/312]  eta: 0:03:43  lr: 0.001378  min_lr: 0.001378  loss: 3.3473 (3.3301)  weight_decay: 0.0500 (0.0500)  time: 0.7186  data: 0.1093  max mem: 41794
Epoch: [188]  [ 80/312]  eta: 0:03:32  lr: 0.001377  min_lr: 0.001377  loss: 3.2896 (3.3097)  weight_decay: 0.0500 (0.0500)  time: 0.9049  data: 0.1553  max mem: 41794
Epoch: [188]  [ 90/312]  eta: 0:03:15  lr: 0.001376  min_lr: 0.001376  loss: 3.2896 (3.3150)  weight_decay: 0.0500 (0.0500)  time: 0.7248  data: 0.0977  max mem: 41794
Epoch: [188]  [100/312]  eta: 0:03:07  lr: 0.001376  min_lr: 0.001376  loss: 3.4881 (3.3346)  weight_decay: 0.0500 (0.0500)  time: 0.7432  data: 0.1392  max mem: 41794
Epoch: [188]  [110/312]  eta: 0:02:51  lr: 0.001375  min_lr: 0.001375  loss: 3.5420 (3.3352)  weight_decay: 0.0500 (0.0500)  time: 0.7043  data: 0.0915  max mem: 41794
Epoch: [188]  [120/312]  eta: 0:02:43  lr: 0.001374  min_lr: 0.001374  loss: 3.4068 (3.3405)  weight_decay: 0.0500 (0.0500)  time: 0.7022  data: 0.0817  max mem: 41794
Epoch: [188]  [130/312]  eta: 0:02:33  lr: 0.001374  min_lr: 0.001374  loss: 3.4068 (3.3517)  weight_decay: 0.0500 (0.0500)  time: 0.8019  data: 0.0840  max mem: 41794
Epoch: [188]  [140/312]  eta: 0:02:22  lr: 0.001373  min_lr: 0.001373  loss: 3.5562 (3.3588)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.0713  max mem: 41794
Epoch: [188]  [150/312]  eta: 0:02:14  lr: 0.001372  min_lr: 0.001372  loss: 3.5943 (3.3616)  weight_decay: 0.0500 (0.0500)  time: 0.7644  data: 0.1835  max mem: 41794
Epoch: [188]  [160/312]  eta: 0:02:04  lr: 0.001372  min_lr: 0.001372  loss: 3.2914 (3.3446)  weight_decay: 0.0500 (0.0500)  time: 0.7972  data: 0.1532  max mem: 41794
Epoch: [188]  [170/312]  eta: 0:01:54  lr: 0.001371  min_lr: 0.001371  loss: 3.1526 (3.3402)  weight_decay: 0.0500 (0.0500)  time: 0.6385  data: 0.0875  max mem: 41794
Epoch: [188]  [180/312]  eta: 0:01:47  lr: 0.001370  min_lr: 0.001370  loss: 3.1481 (3.3244)  weight_decay: 0.0500 (0.0500)  time: 0.7183  data: 0.1527  max mem: 41794
Epoch: [188]  [190/312]  eta: 0:01:36  lr: 0.001370  min_lr: 0.001370  loss: 3.2955 (3.3224)  weight_decay: 0.0500 (0.0500)  time: 0.6655  data: 0.1013  max mem: 41794
Epoch: [188]  [200/312]  eta: 0:01:28  lr: 0.001369  min_lr: 0.001369  loss: 3.3083 (3.3194)  weight_decay: 0.0500 (0.0500)  time: 0.6339  data: 0.0747  max mem: 41794
Epoch: [188]  [210/312]  eta: 0:01:20  lr: 0.001368  min_lr: 0.001368  loss: 3.3664 (3.3242)  weight_decay: 0.0500 (0.0500)  time: 0.7951  data: 0.1574  max mem: 41794
Epoch: [188]  [220/312]  eta: 0:01:11  lr: 0.001368  min_lr: 0.001368  loss: 3.3596 (3.3311)  weight_decay: 0.0500 (0.0500)  time: 0.6507  data: 0.0841  max mem: 41794
Epoch: [188]  [230/312]  eta: 0:01:04  lr: 0.001367  min_lr: 0.001367  loss: 3.3611 (3.3362)  weight_decay: 0.0500 (0.0500)  time: 0.7293  data: 0.1264  max mem: 41794
Epoch: [188]  [240/312]  eta: 0:00:57  lr: 0.001366  min_lr: 0.001366  loss: 3.2734 (3.3296)  weight_decay: 0.0500 (0.0500)  time: 0.9240  data: 0.2261  max mem: 41794
Epoch: [188]  [250/312]  eta: 0:00:48  lr: 0.001366  min_lr: 0.001366  loss: 3.1563 (3.3228)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1003  max mem: 41794
Epoch: [188]  [260/312]  eta: 0:00:40  lr: 0.001365  min_lr: 0.001365  loss: 3.1265 (3.3138)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.0796  max mem: 41794
Epoch: [188]  [270/312]  eta: 0:00:32  lr: 0.001364  min_lr: 0.001364  loss: 3.3123 (3.3212)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.0796  max mem: 41794
Epoch: [188]  [280/312]  eta: 0:00:24  lr: 0.001364  min_lr: 0.001364  loss: 3.5011 (3.3184)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1279  max mem: 41794
Epoch: [188]  [290/312]  eta: 0:00:17  lr: 0.001363  min_lr: 0.001363  loss: 3.1880 (3.3100)  weight_decay: 0.0500 (0.0500)  time: 0.8211  data: 0.2317  max mem: 41794
Epoch: [188]  [300/312]  eta: 0:00:09  lr: 0.001362  min_lr: 0.001362  loss: 3.3519 (3.3189)  weight_decay: 0.0500 (0.0500)  time: 0.6236  data: 0.1042  max mem: 41794
Epoch: [188]  [310/312]  eta: 0:00:01  lr: 0.001361  min_lr: 0.001361  loss: 3.6334 (3.3255)  weight_decay: 0.0500 (0.0500)  time: 0.4667  data: 0.0001  max mem: 41794
Epoch: [188]  [311/312]  eta: 0:00:00  lr: 0.001361  min_lr: 0.001361  loss: 3.6341 (3.3267)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0001  max mem: 41794
Epoch: [188] Total time: 0:03:57 (0.7617 s / it)
Averaged stats: lr: 0.001361  min_lr: 0.001361  loss: 3.6341 (3.3280)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 0.9841 (0.9841)  acc1: 80.2083 (80.2083)  acc5: 95.1823 (95.1823)  time: 8.9909  data: 8.7827  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3096 (1.2270)  acc1: 74.8698 (75.2160)  acc5: 92.5781 (92.7040)  time: 1.1471  data: 0.9759  max mem: 41794
Test: Total time: 0:00:10 (1.1819 s / it)
* Acc@1 75.380 Acc@5 92.936 loss 1.222
Accuracy of the model on the 50000 test images: 75.4%
Max accuracy: 75.58%
Epoch: [189]  [  0/312]  eta: 1:19:56  lr: 0.001361  min_lr: 0.001361  loss: 3.0080 (3.0080)  weight_decay: 0.0500 (0.0500)  time: 15.3723  data: 13.7798  max mem: 41794
Epoch: [189]  [ 10/312]  eta: 0:10:49  lr: 0.001361  min_lr: 0.001361  loss: 3.0080 (3.1642)  weight_decay: 0.0500 (0.0500)  time: 2.1517  data: 1.3826  max mem: 41794
Epoch: [189]  [ 20/312]  eta: 0:07:08  lr: 0.001360  min_lr: 0.001360  loss: 3.3573 (3.3061)  weight_decay: 0.0500 (0.0500)  time: 0.7730  data: 0.1649  max mem: 41794
Epoch: [189]  [ 30/312]  eta: 0:05:26  lr: 0.001359  min_lr: 0.001359  loss: 3.4227 (3.2530)  weight_decay: 0.0500 (0.0500)  time: 0.6097  data: 0.1001  max mem: 41794
Epoch: [189]  [ 40/312]  eta: 0:04:42  lr: 0.001359  min_lr: 0.001359  loss: 3.3277 (3.2533)  weight_decay: 0.0500 (0.0500)  time: 0.5862  data: 0.0651  max mem: 41794
Epoch: [189]  [ 50/312]  eta: 0:04:25  lr: 0.001358  min_lr: 0.001358  loss: 3.2011 (3.2392)  weight_decay: 0.0500 (0.0500)  time: 0.7915  data: 0.2267  max mem: 41794
Epoch: [189]  [ 60/312]  eta: 0:03:53  lr: 0.001357  min_lr: 0.001357  loss: 3.2011 (3.2389)  weight_decay: 0.0500 (0.0500)  time: 0.7017  data: 0.1685  max mem: 41794
Epoch: [189]  [ 70/312]  eta: 0:03:45  lr: 0.001357  min_lr: 0.001357  loss: 3.3557 (3.2527)  weight_decay: 0.0500 (0.0500)  time: 0.7208  data: 0.1124  max mem: 41794
Epoch: [189]  [ 80/312]  eta: 0:03:33  lr: 0.001356  min_lr: 0.001356  loss: 3.4300 (3.2712)  weight_decay: 0.0500 (0.0500)  time: 0.8998  data: 0.2512  max mem: 41794
Epoch: [189]  [ 90/312]  eta: 0:03:14  lr: 0.001355  min_lr: 0.001355  loss: 3.4300 (3.2409)  weight_decay: 0.0500 (0.0500)  time: 0.6737  data: 0.1394  max mem: 41794
Epoch: [189]  [100/312]  eta: 0:03:04  lr: 0.001355  min_lr: 0.001355  loss: 3.4380 (3.2548)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.1272  max mem: 41794
Epoch: [189]  [110/312]  eta: 0:02:49  lr: 0.001354  min_lr: 0.001354  loss: 3.5665 (3.2807)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.1272  max mem: 41794
Epoch: [189]  [120/312]  eta: 0:02:40  lr: 0.001353  min_lr: 0.001353  loss: 3.5665 (3.2811)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1469  max mem: 41794
Epoch: [189]  [130/312]  eta: 0:02:32  lr: 0.001352  min_lr: 0.001352  loss: 3.3686 (3.2691)  weight_decay: 0.0500 (0.0500)  time: 0.8560  data: 0.3072  max mem: 41794
Epoch: [189]  [140/312]  eta: 0:02:20  lr: 0.001352  min_lr: 0.001352  loss: 3.1642 (3.2737)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1610  max mem: 41794
Epoch: [189]  [150/312]  eta: 0:02:12  lr: 0.001351  min_lr: 0.001351  loss: 3.4011 (3.2872)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.1468  max mem: 41794
Epoch: [189]  [160/312]  eta: 0:02:04  lr: 0.001350  min_lr: 0.001350  loss: 3.4828 (3.2965)  weight_decay: 0.0500 (0.0500)  time: 0.8209  data: 0.3067  max mem: 41794
Epoch: [189]  [170/312]  eta: 0:01:53  lr: 0.001350  min_lr: 0.001350  loss: 3.4482 (3.2878)  weight_decay: 0.0500 (0.0500)  time: 0.6553  data: 0.1606  max mem: 41794
Epoch: [189]  [180/312]  eta: 0:01:45  lr: 0.001349  min_lr: 0.001349  loss: 3.0985 (3.2764)  weight_decay: 0.0500 (0.0500)  time: 0.6888  data: 0.1918  max mem: 41794
Epoch: [189]  [190/312]  eta: 0:01:35  lr: 0.001348  min_lr: 0.001348  loss: 3.3600 (3.2872)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.1917  max mem: 41794
Epoch: [189]  [200/312]  eta: 0:01:28  lr: 0.001348  min_lr: 0.001348  loss: 3.3080 (3.2875)  weight_decay: 0.0500 (0.0500)  time: 0.6536  data: 0.1512  max mem: 41794
Epoch: [189]  [210/312]  eta: 0:01:20  lr: 0.001347  min_lr: 0.001347  loss: 3.3080 (3.2942)  weight_decay: 0.0500 (0.0500)  time: 0.8349  data: 0.2874  max mem: 41794
Epoch: [189]  [220/312]  eta: 0:01:11  lr: 0.001346  min_lr: 0.001346  loss: 3.1713 (3.2829)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1368  max mem: 41794
Epoch: [189]  [230/312]  eta: 0:01:04  lr: 0.001346  min_lr: 0.001346  loss: 3.3426 (3.2891)  weight_decay: 0.0500 (0.0500)  time: 0.6851  data: 0.1532  max mem: 41794
Epoch: [189]  [240/312]  eta: 0:00:56  lr: 0.001345  min_lr: 0.001345  loss: 3.5787 (3.2979)  weight_decay: 0.0500 (0.0500)  time: 0.8349  data: 0.2718  max mem: 41794
Epoch: [189]  [250/312]  eta: 0:00:48  lr: 0.001344  min_lr: 0.001344  loss: 3.5787 (3.3067)  weight_decay: 0.0500 (0.0500)  time: 0.6997  data: 0.1524  max mem: 41794
Epoch: [189]  [260/312]  eta: 0:00:40  lr: 0.001344  min_lr: 0.001344  loss: 3.5068 (3.3075)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.2003  max mem: 41794
Epoch: [189]  [270/312]  eta: 0:00:32  lr: 0.001343  min_lr: 0.001343  loss: 3.2351 (3.3094)  weight_decay: 0.0500 (0.0500)  time: 0.7206  data: 0.1672  max mem: 41794
Epoch: [189]  [280/312]  eta: 0:00:24  lr: 0.001342  min_lr: 0.001342  loss: 3.3064 (3.3040)  weight_decay: 0.0500 (0.0500)  time: 0.6555  data: 0.1039  max mem: 41794
Epoch: [189]  [290/312]  eta: 0:00:16  lr: 0.001342  min_lr: 0.001342  loss: 3.2367 (3.2987)  weight_decay: 0.0500 (0.0500)  time: 0.7998  data: 0.2162  max mem: 41794
Epoch: [189]  [300/312]  eta: 0:00:09  lr: 0.001341  min_lr: 0.001341  loss: 3.2367 (3.3003)  weight_decay: 0.0500 (0.0500)  time: 0.6263  data: 0.1127  max mem: 41794
Epoch: [189]  [310/312]  eta: 0:00:01  lr: 0.001340  min_lr: 0.001340  loss: 3.1937 (3.2932)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0001  max mem: 41794
Epoch: [189]  [311/312]  eta: 0:00:00  lr: 0.001340  min_lr: 0.001340  loss: 3.1937 (3.2938)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [189] Total time: 0:03:56 (0.7584 s / it)
Averaged stats: lr: 0.001340  min_lr: 0.001340  loss: 3.1937 (3.3072)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.0740 (1.0740)  acc1: 81.3802 (81.3802)  acc5: 94.9219 (94.9219)  time: 8.9201  data: 8.7186  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2677 (1.2033)  acc1: 74.4792 (75.2320)  acc5: 92.9688 (92.8320)  time: 1.1469  data: 0.9688  max mem: 41794
Test: Total time: 0:00:10 (1.1868 s / it)
* Acc@1 75.270 Acc@5 93.022 loss 1.190
Accuracy of the model on the 50000 test images: 75.3%
Max accuracy: 75.58%
Epoch: [190]  [  0/312]  eta: 1:19:59  lr: 0.001340  min_lr: 0.001340  loss: 3.4631 (3.4631)  weight_decay: 0.0500 (0.0500)  time: 15.3842  data: 13.1371  max mem: 41794
Epoch: [190]  [ 10/312]  eta: 0:11:14  lr: 0.001339  min_lr: 0.001339  loss: 3.4568 (3.1602)  weight_decay: 0.0500 (0.0500)  time: 2.2345  data: 1.4277  max mem: 41794
Epoch: [190]  [ 20/312]  eta: 0:07:32  lr: 0.001339  min_lr: 0.001339  loss: 3.1322 (3.1414)  weight_decay: 0.0500 (0.0500)  time: 0.8568  data: 0.1977  max mem: 41794
Epoch: [190]  [ 30/312]  eta: 0:05:44  lr: 0.001338  min_lr: 0.001338  loss: 3.3771 (3.2441)  weight_decay: 0.0500 (0.0500)  time: 0.6626  data: 0.0710  max mem: 41794
Epoch: [190]  [ 40/312]  eta: 0:04:49  lr: 0.001337  min_lr: 0.001337  loss: 3.4169 (3.2843)  weight_decay: 0.0500 (0.0500)  time: 0.5531  data: 0.0126  max mem: 41794
Epoch: [190]  [ 50/312]  eta: 0:04:28  lr: 0.001337  min_lr: 0.001337  loss: 3.3612 (3.2850)  weight_decay: 0.0500 (0.0500)  time: 0.7235  data: 0.0799  max mem: 41794
Epoch: [190]  [ 60/312]  eta: 0:03:56  lr: 0.001336  min_lr: 0.001336  loss: 3.3176 (3.2851)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.0692  max mem: 41794
Epoch: [190]  [ 70/312]  eta: 0:03:46  lr: 0.001335  min_lr: 0.001335  loss: 3.5458 (3.3205)  weight_decay: 0.0500 (0.0500)  time: 0.7038  data: 0.1026  max mem: 41794
Epoch: [190]  [ 80/312]  eta: 0:03:30  lr: 0.001335  min_lr: 0.001335  loss: 3.5821 (3.3318)  weight_decay: 0.0500 (0.0500)  time: 0.8104  data: 0.1293  max mem: 41794
Epoch: [190]  [ 90/312]  eta: 0:03:14  lr: 0.001334  min_lr: 0.001334  loss: 3.5790 (3.3303)  weight_decay: 0.0500 (0.0500)  time: 0.6762  data: 0.1017  max mem: 41794
Epoch: [190]  [100/312]  eta: 0:03:06  lr: 0.001333  min_lr: 0.001333  loss: 3.5637 (3.3421)  weight_decay: 0.0500 (0.0500)  time: 0.7629  data: 0.1809  max mem: 41794
Epoch: [190]  [110/312]  eta: 0:02:50  lr: 0.001333  min_lr: 0.001333  loss: 3.5508 (3.3455)  weight_decay: 0.0500 (0.0500)  time: 0.6881  data: 0.1099  max mem: 41794
Epoch: [190]  [120/312]  eta: 0:02:43  lr: 0.001332  min_lr: 0.001332  loss: 3.4475 (3.3434)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.0839  max mem: 41794
Epoch: [190]  [130/312]  eta: 0:02:34  lr: 0.001331  min_lr: 0.001331  loss: 3.4475 (3.3349)  weight_decay: 0.0500 (0.0500)  time: 0.8729  data: 0.1410  max mem: 41794
Epoch: [190]  [140/312]  eta: 0:02:21  lr: 0.001331  min_lr: 0.001331  loss: 3.4651 (3.3428)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.0610  max mem: 41794
Epoch: [190]  [150/312]  eta: 0:02:13  lr: 0.001330  min_lr: 0.001330  loss: 3.4581 (3.3409)  weight_decay: 0.0500 (0.0500)  time: 0.6491  data: 0.0611  max mem: 41794
Epoch: [190]  [160/312]  eta: 0:02:04  lr: 0.001329  min_lr: 0.001329  loss: 3.3196 (3.3385)  weight_decay: 0.0500 (0.0500)  time: 0.7940  data: 0.0611  max mem: 41794
Epoch: [190]  [170/312]  eta: 0:01:55  lr: 0.001329  min_lr: 0.001329  loss: 3.3290 (3.3430)  weight_decay: 0.0500 (0.0500)  time: 0.7398  data: 0.0852  max mem: 41794
Epoch: [190]  [180/312]  eta: 0:01:47  lr: 0.001328  min_lr: 0.001328  loss: 3.3290 (3.3221)  weight_decay: 0.0500 (0.0500)  time: 0.7737  data: 0.1318  max mem: 41794
Epoch: [190]  [190/312]  eta: 0:01:37  lr: 0.001327  min_lr: 0.001327  loss: 3.1992 (3.3139)  weight_decay: 0.0500 (0.0500)  time: 0.6777  data: 0.0473  max mem: 41794
Epoch: [190]  [200/312]  eta: 0:01:29  lr: 0.001327  min_lr: 0.001327  loss: 3.4179 (3.3089)  weight_decay: 0.0500 (0.0500)  time: 0.6601  data: 0.0595  max mem: 41794
Epoch: [190]  [210/312]  eta: 0:01:21  lr: 0.001326  min_lr: 0.001326  loss: 3.4878 (3.3142)  weight_decay: 0.0500 (0.0500)  time: 0.8477  data: 0.1169  max mem: 41794
Epoch: [190]  [220/312]  eta: 0:01:12  lr: 0.001325  min_lr: 0.001325  loss: 3.5033 (3.3174)  weight_decay: 0.0500 (0.0500)  time: 0.6972  data: 0.0580  max mem: 41794
Epoch: [190]  [230/312]  eta: 0:01:04  lr: 0.001325  min_lr: 0.001325  loss: 3.5033 (3.3168)  weight_decay: 0.0500 (0.0500)  time: 0.6438  data: 0.0574  max mem: 41794
Epoch: [190]  [240/312]  eta: 0:00:56  lr: 0.001324  min_lr: 0.001324  loss: 3.4978 (3.3249)  weight_decay: 0.0500 (0.0500)  time: 0.7447  data: 0.0626  max mem: 41794
Epoch: [190]  [250/312]  eta: 0:00:48  lr: 0.001323  min_lr: 0.001323  loss: 3.6140 (3.3328)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.0795  max mem: 41794
Epoch: [190]  [260/312]  eta: 0:00:40  lr: 0.001322  min_lr: 0.001322  loss: 3.3268 (3.3241)  weight_decay: 0.0500 (0.0500)  time: 0.7476  data: 0.1540  max mem: 41794
Epoch: [190]  [270/312]  eta: 0:00:32  lr: 0.001322  min_lr: 0.001322  loss: 3.1928 (3.3208)  weight_decay: 0.0500 (0.0500)  time: 0.6721  data: 0.0804  max mem: 41794
Epoch: [190]  [280/312]  eta: 0:00:24  lr: 0.001321  min_lr: 0.001321  loss: 3.5714 (3.3280)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.0705  max mem: 41794
Epoch: [190]  [290/312]  eta: 0:00:17  lr: 0.001320  min_lr: 0.001320  loss: 3.5714 (3.3242)  weight_decay: 0.0500 (0.0500)  time: 0.8851  data: 0.1073  max mem: 41794
Epoch: [190]  [300/312]  eta: 0:00:09  lr: 0.001320  min_lr: 0.001320  loss: 3.3513 (3.3252)  weight_decay: 0.0500 (0.0500)  time: 0.6589  data: 0.0372  max mem: 41794
Epoch: [190]  [310/312]  eta: 0:00:01  lr: 0.001319  min_lr: 0.001319  loss: 3.3141 (3.3247)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [190]  [311/312]  eta: 0:00:00  lr: 0.001319  min_lr: 0.001319  loss: 3.3141 (3.3255)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [190] Total time: 0:03:59 (0.7662 s / it)
Averaged stats: lr: 0.001319  min_lr: 0.001319  loss: 3.3141 (3.2850)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.9769 (0.9769)  acc1: 82.0312 (82.0312)  acc5: 94.9219 (94.9219)  time: 8.4490  data: 8.2513  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2916 (1.2327)  acc1: 73.5677 (75.2960)  acc5: 92.9688 (93.1680)  time: 1.0870  data: 0.9169  max mem: 41794
Test: Total time: 0:00:09 (1.1013 s / it)
* Acc@1 75.654 Acc@5 93.050 loss 1.232
Accuracy of the model on the 50000 test images: 75.7%
Max accuracy: 75.65%
Epoch: [191]  [  0/312]  eta: 1:24:56  lr: 0.001319  min_lr: 0.001319  loss: 3.2363 (3.2363)  weight_decay: 0.0500 (0.0500)  time: 16.3358  data: 15.8625  max mem: 41794
Epoch: [191]  [ 10/312]  eta: 0:11:03  lr: 0.001318  min_lr: 0.001318  loss: 3.5933 (3.4267)  weight_decay: 0.0500 (0.0500)  time: 2.1969  data: 1.4427  max mem: 41794
Epoch: [191]  [ 20/312]  eta: 0:07:27  lr: 0.001318  min_lr: 0.001318  loss: 3.4692 (3.2456)  weight_decay: 0.0500 (0.0500)  time: 0.7911  data: 0.0236  max mem: 41794
Epoch: [191]  [ 30/312]  eta: 0:05:50  lr: 0.001317  min_lr: 0.001317  loss: 3.4835 (3.3137)  weight_decay: 0.0500 (0.0500)  time: 0.7159  data: 0.0707  max mem: 41794
Epoch: [191]  [ 40/312]  eta: 0:04:53  lr: 0.001316  min_lr: 0.001316  loss: 3.6071 (3.3704)  weight_decay: 0.0500 (0.0500)  time: 0.6021  data: 0.0479  max mem: 41794
Epoch: [191]  [ 50/312]  eta: 0:04:28  lr: 0.001316  min_lr: 0.001316  loss: 3.4304 (3.3204)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.0203  max mem: 41794
Epoch: [191]  [ 60/312]  eta: 0:03:58  lr: 0.001315  min_lr: 0.001315  loss: 3.0894 (3.2874)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.0431  max mem: 41794
Epoch: [191]  [ 70/312]  eta: 0:03:45  lr: 0.001314  min_lr: 0.001314  loss: 3.0861 (3.2499)  weight_decay: 0.0500 (0.0500)  time: 0.7003  data: 0.0934  max mem: 41794
Epoch: [191]  [ 80/312]  eta: 0:03:32  lr: 0.001314  min_lr: 0.001314  loss: 3.1164 (3.2513)  weight_decay: 0.0500 (0.0500)  time: 0.8297  data: 0.1253  max mem: 41794
Epoch: [191]  [ 90/312]  eta: 0:03:15  lr: 0.001313  min_lr: 0.001313  loss: 3.3957 (3.2611)  weight_decay: 0.0500 (0.0500)  time: 0.6935  data: 0.0987  max mem: 41794
Epoch: [191]  [100/312]  eta: 0:03:07  lr: 0.001312  min_lr: 0.001312  loss: 3.0431 (3.2254)  weight_decay: 0.0500 (0.0500)  time: 0.7434  data: 0.1565  max mem: 41794
Epoch: [191]  [110/312]  eta: 0:02:51  lr: 0.001312  min_lr: 0.001312  loss: 3.0085 (3.2184)  weight_decay: 0.0500 (0.0500)  time: 0.7052  data: 0.1141  max mem: 41794
Epoch: [191]  [120/312]  eta: 0:02:44  lr: 0.001311  min_lr: 0.001311  loss: 3.2351 (3.2170)  weight_decay: 0.0500 (0.0500)  time: 0.7170  data: 0.0896  max mem: 41794
Epoch: [191]  [130/312]  eta: 0:02:36  lr: 0.001310  min_lr: 0.001310  loss: 3.2702 (3.2208)  weight_decay: 0.0500 (0.0500)  time: 0.9188  data: 0.1847  max mem: 41794
Epoch: [191]  [140/312]  eta: 0:02:23  lr: 0.001310  min_lr: 0.001310  loss: 3.2702 (3.2079)  weight_decay: 0.0500 (0.0500)  time: 0.7028  data: 0.0970  max mem: 41794
Epoch: [191]  [150/312]  eta: 0:02:15  lr: 0.001309  min_lr: 0.001309  loss: 3.2632 (3.2146)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.0642  max mem: 41794
Epoch: [191]  [160/312]  eta: 0:02:05  lr: 0.001308  min_lr: 0.001308  loss: 3.4614 (3.2293)  weight_decay: 0.0500 (0.0500)  time: 0.7592  data: 0.0638  max mem: 41794
Epoch: [191]  [170/312]  eta: 0:01:56  lr: 0.001307  min_lr: 0.001307  loss: 3.4282 (3.2250)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.0718  max mem: 41794
Epoch: [191]  [180/312]  eta: 0:01:48  lr: 0.001307  min_lr: 0.001307  loss: 3.2519 (3.2213)  weight_decay: 0.0500 (0.0500)  time: 0.7783  data: 0.1397  max mem: 41794
Epoch: [191]  [190/312]  eta: 0:01:37  lr: 0.001306  min_lr: 0.001306  loss: 3.1496 (3.2215)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.0687  max mem: 41794
Epoch: [191]  [200/312]  eta: 0:01:30  lr: 0.001305  min_lr: 0.001305  loss: 3.1831 (3.2181)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.0726  max mem: 41794
Epoch: [191]  [210/312]  eta: 0:01:22  lr: 0.001305  min_lr: 0.001305  loss: 3.4158 (3.2292)  weight_decay: 0.0500 (0.0500)  time: 0.8699  data: 0.1834  max mem: 41794
Epoch: [191]  [220/312]  eta: 0:01:13  lr: 0.001304  min_lr: 0.001304  loss: 3.5613 (3.2430)  weight_decay: 0.0500 (0.0500)  time: 0.7025  data: 0.1114  max mem: 41794
Epoch: [191]  [230/312]  eta: 0:01:05  lr: 0.001303  min_lr: 0.001303  loss: 3.5338 (3.2429)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.1322  max mem: 41794
Epoch: [191]  [240/312]  eta: 0:00:57  lr: 0.001303  min_lr: 0.001303  loss: 3.4565 (3.2461)  weight_decay: 0.0500 (0.0500)  time: 0.7824  data: 0.1339  max mem: 41794
Epoch: [191]  [250/312]  eta: 0:00:49  lr: 0.001302  min_lr: 0.001302  loss: 3.4092 (3.2440)  weight_decay: 0.0500 (0.0500)  time: 0.7138  data: 0.1295  max mem: 41794
Epoch: [191]  [260/312]  eta: 0:00:41  lr: 0.001301  min_lr: 0.001301  loss: 3.2063 (3.2370)  weight_decay: 0.0500 (0.0500)  time: 0.8042  data: 0.2504  max mem: 41794
Epoch: [191]  [270/312]  eta: 0:00:32  lr: 0.001301  min_lr: 0.001301  loss: 3.4723 (3.2468)  weight_decay: 0.0500 (0.0500)  time: 0.6803  data: 0.1232  max mem: 41794
Epoch: [191]  [280/312]  eta: 0:00:25  lr: 0.001300  min_lr: 0.001300  loss: 3.5631 (3.2542)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.1022  max mem: 41794
Epoch: [191]  [290/312]  eta: 0:00:17  lr: 0.001299  min_lr: 0.001299  loss: 3.4189 (3.2521)  weight_decay: 0.0500 (0.0500)  time: 0.8290  data: 0.1752  max mem: 41794
Epoch: [191]  [300/312]  eta: 0:00:09  lr: 0.001299  min_lr: 0.001299  loss: 3.2061 (3.2532)  weight_decay: 0.0500 (0.0500)  time: 0.6369  data: 0.0734  max mem: 41794
Epoch: [191]  [310/312]  eta: 0:00:01  lr: 0.001298  min_lr: 0.001298  loss: 3.4540 (3.2588)  weight_decay: 0.0500 (0.0500)  time: 0.4622  data: 0.0001  max mem: 41794
Epoch: [191]  [311/312]  eta: 0:00:00  lr: 0.001298  min_lr: 0.001298  loss: 3.5006 (3.2607)  weight_decay: 0.0500 (0.0500)  time: 0.4619  data: 0.0001  max mem: 41794
Epoch: [191] Total time: 0:04:00 (0.7719 s / it)
Averaged stats: lr: 0.001298  min_lr: 0.001298  loss: 3.5006 (3.2974)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:23  loss: 1.1205 (1.1205)  acc1: 80.5990 (80.5990)  acc5: 95.0521 (95.0521)  time: 9.2365  data: 9.0242  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3881 (1.2940)  acc1: 75.2604 (74.6240)  acc5: 92.4479 (92.9280)  time: 1.1742  data: 1.0028  max mem: 41794
Test: Total time: 0:00:10 (1.1910 s / it)
* Acc@1 74.866 Acc@5 92.784 loss 1.296
Accuracy of the model on the 50000 test images: 74.9%
Max accuracy: 75.65%
Epoch: [192]  [  0/312]  eta: 1:21:20  lr: 0.001298  min_lr: 0.001298  loss: 2.3910 (2.3910)  weight_decay: 0.0500 (0.0500)  time: 15.6425  data: 12.5778  max mem: 41794
Epoch: [192]  [ 10/312]  eta: 0:11:07  lr: 0.001297  min_lr: 0.001297  loss: 3.1364 (3.0015)  weight_decay: 0.0500 (0.0500)  time: 2.2098  data: 1.4047  max mem: 41794
Epoch: [192]  [ 20/312]  eta: 0:07:24  lr: 0.001297  min_lr: 0.001297  loss: 3.2548 (3.1421)  weight_decay: 0.0500 (0.0500)  time: 0.8157  data: 0.1737  max mem: 41794
Epoch: [192]  [ 30/312]  eta: 0:05:37  lr: 0.001296  min_lr: 0.001296  loss: 3.2548 (3.1228)  weight_decay: 0.0500 (0.0500)  time: 0.6416  data: 0.0380  max mem: 41794
Epoch: [192]  [ 40/312]  eta: 0:04:50  lr: 0.001295  min_lr: 0.001295  loss: 3.2593 (3.2276)  weight_decay: 0.0500 (0.0500)  time: 0.5952  data: 0.0108  max mem: 41794
Epoch: [192]  [ 50/312]  eta: 0:04:29  lr: 0.001295  min_lr: 0.001295  loss: 3.4293 (3.2261)  weight_decay: 0.0500 (0.0500)  time: 0.7648  data: 0.0769  max mem: 41794
Epoch: [192]  [ 60/312]  eta: 0:03:56  lr: 0.001294  min_lr: 0.001294  loss: 3.3897 (3.2482)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0768  max mem: 41794
Epoch: [192]  [ 70/312]  eta: 0:03:47  lr: 0.001293  min_lr: 0.001293  loss: 3.4189 (3.2760)  weight_decay: 0.0500 (0.0500)  time: 0.7229  data: 0.0935  max mem: 41794
Epoch: [192]  [ 80/312]  eta: 0:03:32  lr: 0.001293  min_lr: 0.001293  loss: 3.4809 (3.2700)  weight_decay: 0.0500 (0.0500)  time: 0.8403  data: 0.1119  max mem: 41794
Epoch: [192]  [ 90/312]  eta: 0:03:14  lr: 0.001292  min_lr: 0.001292  loss: 3.4805 (3.2866)  weight_decay: 0.0500 (0.0500)  time: 0.6459  data: 0.0661  max mem: 41794
Epoch: [192]  [100/312]  eta: 0:03:05  lr: 0.001291  min_lr: 0.001291  loss: 3.4955 (3.2990)  weight_decay: 0.0500 (0.0500)  time: 0.7090  data: 0.1381  max mem: 41794
Epoch: [192]  [110/312]  eta: 0:02:49  lr: 0.001291  min_lr: 0.001291  loss: 3.4811 (3.2903)  weight_decay: 0.0500 (0.0500)  time: 0.6691  data: 0.0935  max mem: 41794
Epoch: [192]  [120/312]  eta: 0:02:42  lr: 0.001290  min_lr: 0.001290  loss: 3.3437 (3.2824)  weight_decay: 0.0500 (0.0500)  time: 0.7058  data: 0.0874  max mem: 41794
Epoch: [192]  [130/312]  eta: 0:02:35  lr: 0.001289  min_lr: 0.001289  loss: 3.3850 (3.2868)  weight_decay: 0.0500 (0.0500)  time: 0.9188  data: 0.1863  max mem: 41794
Epoch: [192]  [140/312]  eta: 0:02:22  lr: 0.001289  min_lr: 0.001289  loss: 3.4395 (3.2974)  weight_decay: 0.0500 (0.0500)  time: 0.7248  data: 0.1181  max mem: 41794
Epoch: [192]  [150/312]  eta: 0:02:15  lr: 0.001288  min_lr: 0.001288  loss: 3.5561 (3.3156)  weight_decay: 0.0500 (0.0500)  time: 0.7489  data: 0.1278  max mem: 41794
Epoch: [192]  [160/312]  eta: 0:02:06  lr: 0.001287  min_lr: 0.001287  loss: 3.5693 (3.3169)  weight_decay: 0.0500 (0.0500)  time: 0.8476  data: 0.1092  max mem: 41794
Epoch: [192]  [170/312]  eta: 0:01:56  lr: 0.001286  min_lr: 0.001286  loss: 3.1956 (3.3057)  weight_decay: 0.0500 (0.0500)  time: 0.6661  data: 0.0559  max mem: 41794
Epoch: [192]  [180/312]  eta: 0:01:48  lr: 0.001286  min_lr: 0.001286  loss: 3.2099 (3.3102)  weight_decay: 0.0500 (0.0500)  time: 0.7337  data: 0.1367  max mem: 41794
Epoch: [192]  [190/312]  eta: 0:01:38  lr: 0.001285  min_lr: 0.001285  loss: 3.5126 (3.3152)  weight_decay: 0.0500 (0.0500)  time: 0.6850  data: 0.0814  max mem: 41794
Epoch: [192]  [200/312]  eta: 0:01:30  lr: 0.001284  min_lr: 0.001284  loss: 3.4051 (3.3114)  weight_decay: 0.0500 (0.0500)  time: 0.6776  data: 0.0945  max mem: 41794
Epoch: [192]  [210/312]  eta: 0:01:22  lr: 0.001284  min_lr: 0.001284  loss: 3.3636 (3.3140)  weight_decay: 0.0500 (0.0500)  time: 0.8678  data: 0.1892  max mem: 41794
Epoch: [192]  [220/312]  eta: 0:01:13  lr: 0.001283  min_lr: 0.001283  loss: 3.4644 (3.3202)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.0954  max mem: 41794
Epoch: [192]  [230/312]  eta: 0:01:05  lr: 0.001282  min_lr: 0.001282  loss: 3.4947 (3.3195)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.0846  max mem: 41794
Epoch: [192]  [240/312]  eta: 0:00:57  lr: 0.001282  min_lr: 0.001282  loss: 3.3742 (3.3209)  weight_decay: 0.0500 (0.0500)  time: 0.8084  data: 0.1876  max mem: 41794
Epoch: [192]  [250/312]  eta: 0:00:48  lr: 0.001281  min_lr: 0.001281  loss: 3.3742 (3.3245)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1037  max mem: 41794
Epoch: [192]  [260/312]  eta: 0:00:41  lr: 0.001280  min_lr: 0.001280  loss: 3.3300 (3.3228)  weight_decay: 0.0500 (0.0500)  time: 0.7278  data: 0.1516  max mem: 41794
Epoch: [192]  [270/312]  eta: 0:00:32  lr: 0.001280  min_lr: 0.001280  loss: 3.2312 (3.3161)  weight_decay: 0.0500 (0.0500)  time: 0.7007  data: 0.1536  max mem: 41794
Epoch: [192]  [280/312]  eta: 0:00:24  lr: 0.001279  min_lr: 0.001279  loss: 3.4762 (3.3137)  weight_decay: 0.0500 (0.0500)  time: 0.6397  data: 0.1379  max mem: 41794
Epoch: [192]  [290/312]  eta: 0:00:17  lr: 0.001278  min_lr: 0.001278  loss: 3.3004 (3.3090)  weight_decay: 0.0500 (0.0500)  time: 0.8032  data: 0.2968  max mem: 41794
Epoch: [192]  [300/312]  eta: 0:00:09  lr: 0.001278  min_lr: 0.001278  loss: 3.2675 (3.3103)  weight_decay: 0.0500 (0.0500)  time: 0.6523  data: 0.1614  max mem: 41794
Epoch: [192]  [310/312]  eta: 0:00:01  lr: 0.001277  min_lr: 0.001277  loss: 3.5621 (3.3210)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0001  max mem: 41794
Epoch: [192]  [311/312]  eta: 0:00:00  lr: 0.001277  min_lr: 0.001277  loss: 3.5384 (3.3200)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [192] Total time: 0:03:59 (0.7688 s / it)
Averaged stats: lr: 0.001277  min_lr: 0.001277  loss: 3.5384 (3.2856)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 1.0540 (1.0540)  acc1: 81.3802 (81.3802)  acc5: 95.5729 (95.5729)  time: 8.3407  data: 8.1407  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3754 (1.3022)  acc1: 72.5260 (74.6560)  acc5: 92.7083 (92.6560)  time: 1.0907  data: 0.9194  max mem: 41794
Test: Total time: 0:00:09 (1.1013 s / it)
* Acc@1 74.498 Acc@5 92.496 loss 1.311
Accuracy of the model on the 50000 test images: 74.5%
Max accuracy: 75.65%
Epoch: [193]  [  0/312]  eta: 1:26:02  lr: 0.001277  min_lr: 0.001277  loss: 2.7341 (2.7341)  weight_decay: 0.0500 (0.0500)  time: 16.5470  data: 14.1067  max mem: 41794
Epoch: [193]  [ 10/312]  eta: 0:10:30  lr: 0.001276  min_lr: 0.001276  loss: 3.0583 (3.0671)  weight_decay: 0.0500 (0.0500)  time: 2.0888  data: 1.3216  max mem: 41794
Epoch: [193]  [ 20/312]  eta: 0:07:28  lr: 0.001276  min_lr: 0.001276  loss: 3.1109 (3.2532)  weight_decay: 0.0500 (0.0500)  time: 0.7847  data: 0.1192  max mem: 41794
Epoch: [193]  [ 30/312]  eta: 0:05:41  lr: 0.001275  min_lr: 0.001275  loss: 3.2762 (3.2436)  weight_decay: 0.0500 (0.0500)  time: 0.7281  data: 0.1021  max mem: 41794
Epoch: [193]  [ 40/312]  eta: 0:04:56  lr: 0.001274  min_lr: 0.001274  loss: 3.2595 (3.2479)  weight_decay: 0.0500 (0.0500)  time: 0.6232  data: 0.0758  max mem: 41794
Epoch: [193]  [ 50/312]  eta: 0:04:38  lr: 0.001274  min_lr: 0.001274  loss: 3.4277 (3.2447)  weight_decay: 0.0500 (0.0500)  time: 0.8301  data: 0.1532  max mem: 41794
Epoch: [193]  [ 60/312]  eta: 0:04:03  lr: 0.001273  min_lr: 0.001273  loss: 3.4277 (3.2215)  weight_decay: 0.0500 (0.0500)  time: 0.7169  data: 0.0822  max mem: 41794
Epoch: [193]  [ 70/312]  eta: 0:03:54  lr: 0.001272  min_lr: 0.001272  loss: 3.1531 (3.2024)  weight_decay: 0.0500 (0.0500)  time: 0.7273  data: 0.0467  max mem: 41794
Epoch: [193]  [ 80/312]  eta: 0:03:36  lr: 0.001272  min_lr: 0.001272  loss: 3.3853 (3.2405)  weight_decay: 0.0500 (0.0500)  time: 0.8196  data: 0.0565  max mem: 41794
Epoch: [193]  [ 90/312]  eta: 0:03:21  lr: 0.001271  min_lr: 0.001271  loss: 3.4527 (3.2459)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.0901  max mem: 41794
Epoch: [193]  [100/312]  eta: 0:03:12  lr: 0.001270  min_lr: 0.001270  loss: 3.3225 (3.2348)  weight_decay: 0.0500 (0.0500)  time: 0.8253  data: 0.1694  max mem: 41794
Epoch: [193]  [110/312]  eta: 0:02:56  lr: 0.001270  min_lr: 0.001270  loss: 3.1481 (3.2272)  weight_decay: 0.0500 (0.0500)  time: 0.7281  data: 0.0936  max mem: 41794
Epoch: [193]  [120/312]  eta: 0:02:46  lr: 0.001269  min_lr: 0.001269  loss: 3.2127 (3.2250)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.0801  max mem: 41794
Epoch: [193]  [130/312]  eta: 0:02:37  lr: 0.001268  min_lr: 0.001268  loss: 3.4341 (3.2351)  weight_decay: 0.0500 (0.0500)  time: 0.8067  data: 0.1435  max mem: 41794
Epoch: [193]  [140/312]  eta: 0:02:24  lr: 0.001268  min_lr: 0.001268  loss: 3.4676 (3.2509)  weight_decay: 0.0500 (0.0500)  time: 0.6447  data: 0.0679  max mem: 41794
Epoch: [193]  [150/312]  eta: 0:02:14  lr: 0.001267  min_lr: 0.001267  loss: 3.4676 (3.2582)  weight_decay: 0.0500 (0.0500)  time: 0.6275  data: 0.0892  max mem: 41794
Epoch: [193]  [160/312]  eta: 0:02:05  lr: 0.001266  min_lr: 0.001266  loss: 3.1836 (3.2449)  weight_decay: 0.0500 (0.0500)  time: 0.7500  data: 0.1862  max mem: 41794
Epoch: [193]  [170/312]  eta: 0:01:55  lr: 0.001266  min_lr: 0.001266  loss: 3.2079 (3.2551)  weight_decay: 0.0500 (0.0500)  time: 0.6436  data: 0.1028  max mem: 41794
Epoch: [193]  [180/312]  eta: 0:01:47  lr: 0.001265  min_lr: 0.001265  loss: 3.5401 (3.2570)  weight_decay: 0.0500 (0.0500)  time: 0.7471  data: 0.1876  max mem: 41794
Epoch: [193]  [190/312]  eta: 0:01:37  lr: 0.001264  min_lr: 0.001264  loss: 3.4254 (3.2639)  weight_decay: 0.0500 (0.0500)  time: 0.7116  data: 0.1823  max mem: 41794
Epoch: [193]  [200/312]  eta: 0:01:30  lr: 0.001264  min_lr: 0.001264  loss: 3.3002 (3.2525)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.1652  max mem: 41794
Epoch: [193]  [210/312]  eta: 0:01:22  lr: 0.001263  min_lr: 0.001263  loss: 2.9636 (3.2491)  weight_decay: 0.0500 (0.0500)  time: 0.9147  data: 0.3235  max mem: 41794
Epoch: [193]  [220/312]  eta: 0:01:13  lr: 0.001262  min_lr: 0.001262  loss: 3.5871 (3.2666)  weight_decay: 0.0500 (0.0500)  time: 0.7206  data: 0.1588  max mem: 41794
Epoch: [193]  [230/312]  eta: 0:01:05  lr: 0.001262  min_lr: 0.001262  loss: 3.5923 (3.2703)  weight_decay: 0.0500 (0.0500)  time: 0.6564  data: 0.1456  max mem: 41794
Epoch: [193]  [240/312]  eta: 0:00:57  lr: 0.001261  min_lr: 0.001261  loss: 3.4287 (3.2673)  weight_decay: 0.0500 (0.0500)  time: 0.7876  data: 0.2652  max mem: 41794
Epoch: [193]  [250/312]  eta: 0:00:48  lr: 0.001260  min_lr: 0.001260  loss: 3.4850 (3.2709)  weight_decay: 0.0500 (0.0500)  time: 0.6714  data: 0.1670  max mem: 41794
Epoch: [193]  [260/312]  eta: 0:00:41  lr: 0.001260  min_lr: 0.001260  loss: 3.4850 (3.2674)  weight_decay: 0.0500 (0.0500)  time: 0.7342  data: 0.2374  max mem: 41794
Epoch: [193]  [270/312]  eta: 0:00:32  lr: 0.001259  min_lr: 0.001259  loss: 3.4166 (3.2651)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.1907  max mem: 41794
Epoch: [193]  [280/312]  eta: 0:00:25  lr: 0.001258  min_lr: 0.001258  loss: 3.0319 (3.2592)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.1704  max mem: 41794
Epoch: [193]  [290/312]  eta: 0:00:17  lr: 0.001258  min_lr: 0.001258  loss: 3.0319 (3.2539)  weight_decay: 0.0500 (0.0500)  time: 0.8189  data: 0.2915  max mem: 41794
Epoch: [193]  [300/312]  eta: 0:00:09  lr: 0.001257  min_lr: 0.001257  loss: 3.3684 (3.2632)  weight_decay: 0.0500 (0.0500)  time: 0.6301  data: 0.1214  max mem: 41794
Epoch: [193]  [310/312]  eta: 0:00:01  lr: 0.001256  min_lr: 0.001256  loss: 3.5171 (3.2653)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [193]  [311/312]  eta: 0:00:00  lr: 0.001256  min_lr: 0.001256  loss: 3.5171 (3.2668)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0001  max mem: 41794
Epoch: [193] Total time: 0:03:59 (0.7684 s / it)
Averaged stats: lr: 0.001256  min_lr: 0.001256  loss: 3.5171 (3.2978)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.9860 (0.9860)  acc1: 82.6823 (82.6823)  acc5: 95.1823 (95.1823)  time: 8.6220  data: 8.4113  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2866 (1.2356)  acc1: 75.1302 (74.7360)  acc5: 93.3594 (92.9120)  time: 1.1185  data: 0.9347  max mem: 41794
Test: Total time: 0:00:10 (1.1739 s / it)
* Acc@1 75.286 Acc@5 93.142 loss 1.227
Accuracy of the model on the 50000 test images: 75.3%
Max accuracy: 75.65%
Epoch: [194]  [  0/312]  eta: 1:20:52  lr: 0.001256  min_lr: 0.001256  loss: 3.5650 (3.5650)  weight_decay: 0.0500 (0.0500)  time: 15.5529  data: 15.0538  max mem: 41794
Epoch: [194]  [ 10/312]  eta: 0:11:10  lr: 0.001255  min_lr: 0.001255  loss: 3.6127 (3.5202)  weight_decay: 0.0500 (0.0500)  time: 2.2188  data: 1.3765  max mem: 41794
Epoch: [194]  [ 20/312]  eta: 0:07:20  lr: 0.001255  min_lr: 0.001255  loss: 3.6347 (3.5773)  weight_decay: 0.0500 (0.0500)  time: 0.8056  data: 0.0686  max mem: 41794
Epoch: [194]  [ 30/312]  eta: 0:05:36  lr: 0.001254  min_lr: 0.001254  loss: 3.6100 (3.5062)  weight_decay: 0.0500 (0.0500)  time: 0.6280  data: 0.0725  max mem: 41794
Epoch: [194]  [ 40/312]  eta: 0:04:42  lr: 0.001253  min_lr: 0.001253  loss: 3.2123 (3.3757)  weight_decay: 0.0500 (0.0500)  time: 0.5435  data: 0.0213  max mem: 41794
Epoch: [194]  [ 50/312]  eta: 0:04:20  lr: 0.001253  min_lr: 0.001253  loss: 3.0577 (3.3336)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.1343  max mem: 41794
Epoch: [194]  [ 60/312]  eta: 0:03:49  lr: 0.001252  min_lr: 0.001252  loss: 3.3467 (3.3675)  weight_decay: 0.0500 (0.0500)  time: 0.6523  data: 0.1216  max mem: 41794
Epoch: [194]  [ 70/312]  eta: 0:03:39  lr: 0.001251  min_lr: 0.001251  loss: 3.4935 (3.3623)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.1271  max mem: 41794
Epoch: [194]  [ 80/312]  eta: 0:03:29  lr: 0.001251  min_lr: 0.001251  loss: 3.4353 (3.3592)  weight_decay: 0.0500 (0.0500)  time: 0.8744  data: 0.2319  max mem: 41794
Epoch: [194]  [ 90/312]  eta: 0:03:10  lr: 0.001250  min_lr: 0.001250  loss: 3.5248 (3.3675)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.1098  max mem: 41794
Epoch: [194]  [100/312]  eta: 0:03:02  lr: 0.001249  min_lr: 0.001249  loss: 3.3029 (3.3440)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.1398  max mem: 41794
Epoch: [194]  [110/312]  eta: 0:02:46  lr: 0.001249  min_lr: 0.001249  loss: 3.3029 (3.3548)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1354  max mem: 41794
Epoch: [194]  [120/312]  eta: 0:02:39  lr: 0.001248  min_lr: 0.001248  loss: 3.5463 (3.3572)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.1055  max mem: 41794
Epoch: [194]  [130/312]  eta: 0:02:31  lr: 0.001247  min_lr: 0.001247  loss: 3.4052 (3.3537)  weight_decay: 0.0500 (0.0500)  time: 0.8717  data: 0.2074  max mem: 41794
Epoch: [194]  [140/312]  eta: 0:02:19  lr: 0.001247  min_lr: 0.001247  loss: 3.3838 (3.3492)  weight_decay: 0.0500 (0.0500)  time: 0.6817  data: 0.1060  max mem: 41794
Epoch: [194]  [150/312]  eta: 0:02:11  lr: 0.001246  min_lr: 0.001246  loss: 3.4719 (3.3489)  weight_decay: 0.0500 (0.0500)  time: 0.6878  data: 0.0987  max mem: 41794
Epoch: [194]  [160/312]  eta: 0:02:03  lr: 0.001245  min_lr: 0.001245  loss: 3.4039 (3.3538)  weight_decay: 0.0500 (0.0500)  time: 0.8604  data: 0.1730  max mem: 41794
Epoch: [194]  [170/312]  eta: 0:01:53  lr: 0.001245  min_lr: 0.001245  loss: 3.4353 (3.3507)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.0797  max mem: 41794
Epoch: [194]  [180/312]  eta: 0:01:45  lr: 0.001244  min_lr: 0.001244  loss: 3.4332 (3.3393)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.1034  max mem: 41794
Epoch: [194]  [190/312]  eta: 0:01:35  lr: 0.001243  min_lr: 0.001243  loss: 3.3085 (3.3387)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.1020  max mem: 41794
Epoch: [194]  [200/312]  eta: 0:01:28  lr: 0.001243  min_lr: 0.001243  loss: 3.4149 (3.3316)  weight_decay: 0.0500 (0.0500)  time: 0.6620  data: 0.1382  max mem: 41794
Epoch: [194]  [210/312]  eta: 0:01:20  lr: 0.001242  min_lr: 0.001242  loss: 3.4620 (3.3406)  weight_decay: 0.0500 (0.0500)  time: 0.8390  data: 0.2848  max mem: 41794
Epoch: [194]  [220/312]  eta: 0:01:11  lr: 0.001241  min_lr: 0.001241  loss: 3.5397 (3.3459)  weight_decay: 0.0500 (0.0500)  time: 0.6700  data: 0.1471  max mem: 41794
Epoch: [194]  [230/312]  eta: 0:01:04  lr: 0.001241  min_lr: 0.001241  loss: 3.5397 (3.3534)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1633  max mem: 41794
Epoch: [194]  [240/312]  eta: 0:00:56  lr: 0.001240  min_lr: 0.001240  loss: 3.3626 (3.3370)  weight_decay: 0.0500 (0.0500)  time: 0.8799  data: 0.3094  max mem: 41794
Epoch: [194]  [250/312]  eta: 0:00:48  lr: 0.001239  min_lr: 0.001239  loss: 3.0573 (3.3342)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1468  max mem: 41794
Epoch: [194]  [260/312]  eta: 0:00:40  lr: 0.001239  min_lr: 0.001239  loss: 3.3647 (3.3314)  weight_decay: 0.0500 (0.0500)  time: 0.6586  data: 0.1301  max mem: 41794
Epoch: [194]  [270/312]  eta: 0:00:32  lr: 0.001238  min_lr: 0.001238  loss: 3.5305 (3.3335)  weight_decay: 0.0500 (0.0500)  time: 0.6560  data: 0.1301  max mem: 41794
Epoch: [194]  [280/312]  eta: 0:00:24  lr: 0.001237  min_lr: 0.001237  loss: 3.5264 (3.3354)  weight_decay: 0.0500 (0.0500)  time: 0.7011  data: 0.1453  max mem: 41794
Epoch: [194]  [290/312]  eta: 0:00:16  lr: 0.001237  min_lr: 0.001237  loss: 3.4439 (3.3308)  weight_decay: 0.0500 (0.0500)  time: 0.8513  data: 0.2701  max mem: 41794
Epoch: [194]  [300/312]  eta: 0:00:09  lr: 0.001236  min_lr: 0.001236  loss: 3.4481 (3.3337)  weight_decay: 0.0500 (0.0500)  time: 0.6342  data: 0.1251  max mem: 41794
Epoch: [194]  [310/312]  eta: 0:00:01  lr: 0.001235  min_lr: 0.001235  loss: 3.4803 (3.3408)  weight_decay: 0.0500 (0.0500)  time: 0.4667  data: 0.0001  max mem: 41794
Epoch: [194]  [311/312]  eta: 0:00:00  lr: 0.001235  min_lr: 0.001235  loss: 3.4803 (3.3429)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [194] Total time: 0:03:56 (0.7571 s / it)
Averaged stats: lr: 0.001235  min_lr: 0.001235  loss: 3.4803 (3.2871)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.0888 (1.0888)  acc1: 80.9896 (80.9896)  acc5: 95.5729 (95.5729)  time: 8.2357  data: 8.0323  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.4640 (1.3545)  acc1: 74.2188 (74.9440)  acc5: 93.0990 (93.0720)  time: 1.0790  data: 0.9083  max mem: 41794
Test: Total time: 0:00:09 (1.0904 s / it)
* Acc@1 75.348 Acc@5 93.090 loss 1.357
Accuracy of the model on the 50000 test images: 75.3%
Max accuracy: 75.65%
Epoch: [195]  [  0/312]  eta: 1:18:16  lr: 0.001235  min_lr: 0.001235  loss: 2.1952 (2.1952)  weight_decay: 0.0500 (0.0500)  time: 15.0540  data: 12.4569  max mem: 41794
Epoch: [195]  [ 10/312]  eta: 0:10:16  lr: 0.001235  min_lr: 0.001235  loss: 3.4482 (3.2610)  weight_decay: 0.0500 (0.0500)  time: 2.0416  data: 1.3685  max mem: 41794
Epoch: [195]  [ 20/312]  eta: 0:07:39  lr: 0.001234  min_lr: 0.001234  loss: 3.2322 (3.1180)  weight_decay: 0.0500 (0.0500)  time: 0.9006  data: 0.2557  max mem: 41794
Epoch: [195]  [ 30/312]  eta: 0:05:51  lr: 0.001233  min_lr: 0.001233  loss: 3.1576 (3.1399)  weight_decay: 0.0500 (0.0500)  time: 0.8087  data: 0.1325  max mem: 41794
Epoch: [195]  [ 40/312]  eta: 0:05:08  lr: 0.001233  min_lr: 0.001233  loss: 3.3200 (3.1537)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.0715  max mem: 41794
Epoch: [195]  [ 50/312]  eta: 0:04:31  lr: 0.001232  min_lr: 0.001232  loss: 3.2752 (3.1818)  weight_decay: 0.0500 (0.0500)  time: 0.7136  data: 0.1034  max mem: 41794
Epoch: [195]  [ 60/312]  eta: 0:03:59  lr: 0.001231  min_lr: 0.001231  loss: 3.3533 (3.1814)  weight_decay: 0.0500 (0.0500)  time: 0.5713  data: 0.0413  max mem: 41794
Epoch: [195]  [ 70/312]  eta: 0:03:46  lr: 0.001231  min_lr: 0.001231  loss: 3.4810 (3.2041)  weight_decay: 0.0500 (0.0500)  time: 0.6758  data: 0.0833  max mem: 41794
Epoch: [195]  [ 80/312]  eta: 0:03:32  lr: 0.001230  min_lr: 0.001230  loss: 3.4371 (3.2133)  weight_decay: 0.0500 (0.0500)  time: 0.8109  data: 0.0809  max mem: 41794
Epoch: [195]  [ 90/312]  eta: 0:03:17  lr: 0.001229  min_lr: 0.001229  loss: 3.3748 (3.2302)  weight_decay: 0.0500 (0.0500)  time: 0.7241  data: 0.0979  max mem: 41794
Epoch: [195]  [100/312]  eta: 0:03:10  lr: 0.001229  min_lr: 0.001229  loss: 3.2123 (3.2139)  weight_decay: 0.0500 (0.0500)  time: 0.8294  data: 0.1785  max mem: 41794
Epoch: [195]  [110/312]  eta: 0:02:54  lr: 0.001228  min_lr: 0.001228  loss: 3.2443 (3.2236)  weight_decay: 0.0500 (0.0500)  time: 0.7508  data: 0.0841  max mem: 41794
Epoch: [195]  [120/312]  eta: 0:02:45  lr: 0.001227  min_lr: 0.001227  loss: 3.2443 (3.2218)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.0740  max mem: 41794
Epoch: [195]  [130/312]  eta: 0:02:36  lr: 0.001227  min_lr: 0.001227  loss: 3.1202 (3.2269)  weight_decay: 0.0500 (0.0500)  time: 0.8324  data: 0.1521  max mem: 41794
Epoch: [195]  [140/312]  eta: 0:02:23  lr: 0.001226  min_lr: 0.001226  loss: 3.5296 (3.2439)  weight_decay: 0.0500 (0.0500)  time: 0.6710  data: 0.0815  max mem: 41794
Epoch: [195]  [150/312]  eta: 0:02:15  lr: 0.001225  min_lr: 0.001225  loss: 3.4303 (3.2572)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.0878  max mem: 41794
Epoch: [195]  [160/312]  eta: 0:02:05  lr: 0.001225  min_lr: 0.001225  loss: 3.3356 (3.2546)  weight_decay: 0.0500 (0.0500)  time: 0.7930  data: 0.0893  max mem: 41794
Epoch: [195]  [170/312]  eta: 0:01:56  lr: 0.001224  min_lr: 0.001224  loss: 3.1165 (3.2448)  weight_decay: 0.0500 (0.0500)  time: 0.6891  data: 0.0935  max mem: 41794
Epoch: [195]  [180/312]  eta: 0:01:48  lr: 0.001223  min_lr: 0.001223  loss: 3.2484 (3.2493)  weight_decay: 0.0500 (0.0500)  time: 0.7871  data: 0.1881  max mem: 41794
Epoch: [195]  [190/312]  eta: 0:01:38  lr: 0.001223  min_lr: 0.001223  loss: 3.3903 (3.2497)  weight_decay: 0.0500 (0.0500)  time: 0.7042  data: 0.0967  max mem: 41794
Epoch: [195]  [200/312]  eta: 0:01:30  lr: 0.001222  min_lr: 0.001222  loss: 3.5954 (3.2715)  weight_decay: 0.0500 (0.0500)  time: 0.6520  data: 0.0782  max mem: 41794
Epoch: [195]  [210/312]  eta: 0:01:22  lr: 0.001221  min_lr: 0.001221  loss: 3.5920 (3.2636)  weight_decay: 0.0500 (0.0500)  time: 0.8425  data: 0.1846  max mem: 41794
Epoch: [195]  [220/312]  eta: 0:01:13  lr: 0.001221  min_lr: 0.001221  loss: 3.2082 (3.2680)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.1071  max mem: 41794
Epoch: [195]  [230/312]  eta: 0:01:05  lr: 0.001220  min_lr: 0.001220  loss: 3.3280 (3.2618)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.1232  max mem: 41794
Epoch: [195]  [240/312]  eta: 0:00:57  lr: 0.001219  min_lr: 0.001219  loss: 3.1628 (3.2574)  weight_decay: 0.0500 (0.0500)  time: 0.7711  data: 0.1275  max mem: 41794
Epoch: [195]  [250/312]  eta: 0:00:48  lr: 0.001219  min_lr: 0.001219  loss: 3.0055 (3.2509)  weight_decay: 0.0500 (0.0500)  time: 0.6671  data: 0.0998  max mem: 41794
Epoch: [195]  [260/312]  eta: 0:00:41  lr: 0.001218  min_lr: 0.001218  loss: 3.2324 (3.2491)  weight_decay: 0.0500 (0.0500)  time: 0.7591  data: 0.1967  max mem: 41794
Epoch: [195]  [270/312]  eta: 0:00:32  lr: 0.001217  min_lr: 0.001217  loss: 3.3363 (3.2534)  weight_decay: 0.0500 (0.0500)  time: 0.6688  data: 0.1018  max mem: 41794
Epoch: [195]  [280/312]  eta: 0:00:25  lr: 0.001217  min_lr: 0.001217  loss: 3.4048 (3.2628)  weight_decay: 0.0500 (0.0500)  time: 0.6545  data: 0.1151  max mem: 41794
Epoch: [195]  [290/312]  eta: 0:00:17  lr: 0.001216  min_lr: 0.001216  loss: 3.4048 (3.2624)  weight_decay: 0.0500 (0.0500)  time: 0.7650  data: 0.2085  max mem: 41794
Epoch: [195]  [300/312]  eta: 0:00:09  lr: 0.001215  min_lr: 0.001215  loss: 3.2699 (3.2584)  weight_decay: 0.0500 (0.0500)  time: 0.5972  data: 0.0937  max mem: 41794
Epoch: [195]  [310/312]  eta: 0:00:01  lr: 0.001215  min_lr: 0.001215  loss: 3.5090 (3.2590)  weight_decay: 0.0500 (0.0500)  time: 0.4669  data: 0.0001  max mem: 41794
Epoch: [195]  [311/312]  eta: 0:00:00  lr: 0.001215  min_lr: 0.001215  loss: 3.5121 (3.2603)  weight_decay: 0.0500 (0.0500)  time: 0.4667  data: 0.0001  max mem: 41794
Epoch: [195] Total time: 0:03:58 (0.7657 s / it)
Averaged stats: lr: 0.001215  min_lr: 0.001215  loss: 3.5121 (3.2852)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.9978 (0.9978)  acc1: 82.4219 (82.4219)  acc5: 96.0938 (96.0938)  time: 8.7807  data: 8.5720  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2345 (1.2061)  acc1: 74.7396 (75.3440)  acc5: 93.8802 (93.4560)  time: 1.1244  data: 0.9525  max mem: 41794
Test: Total time: 0:00:10 (1.1618 s / it)
* Acc@1 75.674 Acc@5 93.302 loss 1.203
Accuracy of the model on the 50000 test images: 75.7%
Max accuracy: 75.67%
Epoch: [196]  [  0/312]  eta: 1:22:02  lr: 0.001215  min_lr: 0.001215  loss: 2.5073 (2.5073)  weight_decay: 0.0500 (0.0500)  time: 15.7773  data: 12.0004  max mem: 41794
Epoch: [196]  [ 10/312]  eta: 0:10:55  lr: 0.001214  min_lr: 0.001214  loss: 3.4648 (3.2391)  weight_decay: 0.0500 (0.0500)  time: 2.1720  data: 1.3598  max mem: 41794
Epoch: [196]  [ 20/312]  eta: 0:07:28  lr: 0.001213  min_lr: 0.001213  loss: 3.1587 (3.0577)  weight_decay: 0.0500 (0.0500)  time: 0.8247  data: 0.2115  max mem: 41794
Epoch: [196]  [ 30/312]  eta: 0:05:42  lr: 0.001213  min_lr: 0.001213  loss: 3.3061 (3.1906)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.0640  max mem: 41794
Epoch: [196]  [ 40/312]  eta: 0:04:47  lr: 0.001212  min_lr: 0.001212  loss: 3.3868 (3.1436)  weight_decay: 0.0500 (0.0500)  time: 0.5512  data: 0.0082  max mem: 41794
Epoch: [196]  [ 50/312]  eta: 0:04:30  lr: 0.001211  min_lr: 0.001211  loss: 2.9544 (3.1338)  weight_decay: 0.0500 (0.0500)  time: 0.7499  data: 0.1303  max mem: 41794
Epoch: [196]  [ 60/312]  eta: 0:03:57  lr: 0.001211  min_lr: 0.001211  loss: 3.2036 (3.1496)  weight_decay: 0.0500 (0.0500)  time: 0.7152  data: 0.1229  max mem: 41794
Epoch: [196]  [ 70/312]  eta: 0:03:45  lr: 0.001210  min_lr: 0.001210  loss: 3.3354 (3.1915)  weight_decay: 0.0500 (0.0500)  time: 0.6729  data: 0.1117  max mem: 41794
Epoch: [196]  [ 80/312]  eta: 0:03:29  lr: 0.001209  min_lr: 0.001209  loss: 3.4682 (3.2233)  weight_decay: 0.0500 (0.0500)  time: 0.7817  data: 0.1337  max mem: 41794
Epoch: [196]  [ 90/312]  eta: 0:03:13  lr: 0.001209  min_lr: 0.001209  loss: 3.3870 (3.2352)  weight_decay: 0.0500 (0.0500)  time: 0.6638  data: 0.0868  max mem: 41794
Epoch: [196]  [100/312]  eta: 0:03:04  lr: 0.001208  min_lr: 0.001208  loss: 3.3812 (3.2366)  weight_decay: 0.0500 (0.0500)  time: 0.7419  data: 0.1479  max mem: 41794
Epoch: [196]  [110/312]  eta: 0:02:49  lr: 0.001207  min_lr: 0.001207  loss: 3.2934 (3.2517)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.0838  max mem: 41794
Epoch: [196]  [120/312]  eta: 0:02:42  lr: 0.001207  min_lr: 0.001207  loss: 3.4988 (3.2656)  weight_decay: 0.0500 (0.0500)  time: 0.7127  data: 0.0840  max mem: 41794
Epoch: [196]  [130/312]  eta: 0:02:34  lr: 0.001206  min_lr: 0.001206  loss: 3.1771 (3.2549)  weight_decay: 0.0500 (0.0500)  time: 0.8986  data: 0.1599  max mem: 41794
Epoch: [196]  [140/312]  eta: 0:02:21  lr: 0.001205  min_lr: 0.001205  loss: 3.0894 (3.2460)  weight_decay: 0.0500 (0.0500)  time: 0.6864  data: 0.0765  max mem: 41794
Epoch: [196]  [150/312]  eta: 0:02:13  lr: 0.001205  min_lr: 0.001205  loss: 3.1548 (3.2506)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0769  max mem: 41794
Epoch: [196]  [160/312]  eta: 0:02:04  lr: 0.001204  min_lr: 0.001204  loss: 3.3748 (3.2612)  weight_decay: 0.0500 (0.0500)  time: 0.7908  data: 0.1009  max mem: 41794
Epoch: [196]  [170/312]  eta: 0:01:55  lr: 0.001203  min_lr: 0.001203  loss: 3.2870 (3.2539)  weight_decay: 0.0500 (0.0500)  time: 0.7191  data: 0.1187  max mem: 41794
Epoch: [196]  [180/312]  eta: 0:01:47  lr: 0.001203  min_lr: 0.001203  loss: 3.1493 (3.2473)  weight_decay: 0.0500 (0.0500)  time: 0.7826  data: 0.1734  max mem: 41794
Epoch: [196]  [190/312]  eta: 0:01:37  lr: 0.001202  min_lr: 0.001202  loss: 3.3869 (3.2503)  weight_decay: 0.0500 (0.0500)  time: 0.6939  data: 0.0793  max mem: 41794
Epoch: [196]  [200/312]  eta: 0:01:29  lr: 0.001201  min_lr: 0.001201  loss: 3.4150 (3.2514)  weight_decay: 0.0500 (0.0500)  time: 0.6891  data: 0.0953  max mem: 41794
Epoch: [196]  [210/312]  eta: 0:01:21  lr: 0.001201  min_lr: 0.001201  loss: 3.3998 (3.2442)  weight_decay: 0.0500 (0.0500)  time: 0.8491  data: 0.1980  max mem: 41794
Epoch: [196]  [220/312]  eta: 0:01:12  lr: 0.001200  min_lr: 0.001200  loss: 3.3590 (3.2460)  weight_decay: 0.0500 (0.0500)  time: 0.6654  data: 0.1033  max mem: 41794
Epoch: [196]  [230/312]  eta: 0:01:04  lr: 0.001199  min_lr: 0.001199  loss: 3.4102 (3.2540)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.1120  max mem: 41794
Epoch: [196]  [240/312]  eta: 0:00:56  lr: 0.001199  min_lr: 0.001199  loss: 3.3921 (3.2515)  weight_decay: 0.0500 (0.0500)  time: 0.7409  data: 0.1266  max mem: 41794
Epoch: [196]  [250/312]  eta: 0:00:48  lr: 0.001198  min_lr: 0.001198  loss: 3.3141 (3.2446)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.1387  max mem: 41794
Epoch: [196]  [260/312]  eta: 0:00:40  lr: 0.001197  min_lr: 0.001197  loss: 3.2491 (3.2443)  weight_decay: 0.0500 (0.0500)  time: 0.7672  data: 0.2397  max mem: 41794
Epoch: [196]  [270/312]  eta: 0:00:32  lr: 0.001197  min_lr: 0.001197  loss: 3.1974 (3.2401)  weight_decay: 0.0500 (0.0500)  time: 0.6393  data: 0.1162  max mem: 41794
Epoch: [196]  [280/312]  eta: 0:00:24  lr: 0.001196  min_lr: 0.001196  loss: 3.1974 (3.2447)  weight_decay: 0.0500 (0.0500)  time: 0.6702  data: 0.1416  max mem: 41794
Epoch: [196]  [290/312]  eta: 0:00:17  lr: 0.001196  min_lr: 0.001196  loss: 3.4266 (3.2471)  weight_decay: 0.0500 (0.0500)  time: 0.8254  data: 0.2468  max mem: 41794
Epoch: [196]  [300/312]  eta: 0:00:09  lr: 0.001195  min_lr: 0.001195  loss: 3.4777 (3.2464)  weight_decay: 0.0500 (0.0500)  time: 0.6356  data: 0.1055  max mem: 41794
Epoch: [196]  [310/312]  eta: 0:00:01  lr: 0.001194  min_lr: 0.001194  loss: 3.2933 (3.2432)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [196]  [311/312]  eta: 0:00:00  lr: 0.001194  min_lr: 0.001194  loss: 3.3044 (3.2444)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [196] Total time: 0:03:57 (0.7626 s / it)
Averaged stats: lr: 0.001194  min_lr: 0.001194  loss: 3.3044 (3.2551)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.9573 (0.9573)  acc1: 81.9010 (81.9010)  acc5: 95.5729 (95.5729)  time: 8.3225  data: 8.1120  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2352 (1.1622)  acc1: 74.2188 (75.2960)  acc5: 92.5781 (93.1040)  time: 1.0726  data: 0.9014  max mem: 41794
Test: Total time: 0:00:09 (1.0942 s / it)
* Acc@1 75.628 Acc@5 93.192 loss 1.164
Accuracy of the model on the 50000 test images: 75.6%
Max accuracy: 75.67%
Epoch: [197]  [  0/312]  eta: 1:19:44  lr: 0.001194  min_lr: 0.001194  loss: 3.2219 (3.2219)  weight_decay: 0.0500 (0.0500)  time: 15.3358  data: 12.1578  max mem: 41794
Epoch: [197]  [ 10/312]  eta: 0:10:40  lr: 0.001193  min_lr: 0.001193  loss: 3.2219 (3.1866)  weight_decay: 0.0500 (0.0500)  time: 2.1199  data: 1.3314  max mem: 41794
Epoch: [197]  [ 20/312]  eta: 0:07:07  lr: 0.001193  min_lr: 0.001193  loss: 3.3860 (3.2369)  weight_decay: 0.0500 (0.0500)  time: 0.7712  data: 0.1818  max mem: 41794
Epoch: [197]  [ 30/312]  eta: 0:05:25  lr: 0.001192  min_lr: 0.001192  loss: 3.1341 (3.1573)  weight_decay: 0.0500 (0.0500)  time: 0.6237  data: 0.0577  max mem: 41794
Epoch: [197]  [ 40/312]  eta: 0:04:41  lr: 0.001191  min_lr: 0.001191  loss: 2.8904 (3.1311)  weight_decay: 0.0500 (0.0500)  time: 0.5863  data: 0.0622  max mem: 41794
Epoch: [197]  [ 50/312]  eta: 0:04:20  lr: 0.001191  min_lr: 0.001191  loss: 3.3297 (3.1569)  weight_decay: 0.0500 (0.0500)  time: 0.7451  data: 0.1675  max mem: 41794
Epoch: [197]  [ 60/312]  eta: 0:03:51  lr: 0.001190  min_lr: 0.001190  loss: 3.4153 (3.1754)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.1077  max mem: 41794
Epoch: [197]  [ 70/312]  eta: 0:03:44  lr: 0.001189  min_lr: 0.001189  loss: 3.3944 (3.1926)  weight_decay: 0.0500 (0.0500)  time: 0.7549  data: 0.1649  max mem: 41794
Epoch: [197]  [ 80/312]  eta: 0:03:30  lr: 0.001189  min_lr: 0.001189  loss: 3.3347 (3.2118)  weight_decay: 0.0500 (0.0500)  time: 0.8756  data: 0.2522  max mem: 41794
Epoch: [197]  [ 90/312]  eta: 0:03:13  lr: 0.001188  min_lr: 0.001188  loss: 3.2450 (3.2034)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.1302  max mem: 41794
Epoch: [197]  [100/312]  eta: 0:03:06  lr: 0.001187  min_lr: 0.001187  loss: 3.2890 (3.2069)  weight_decay: 0.0500 (0.0500)  time: 0.7744  data: 0.1708  max mem: 41794
Epoch: [197]  [110/312]  eta: 0:02:51  lr: 0.001187  min_lr: 0.001187  loss: 3.4000 (3.2097)  weight_decay: 0.0500 (0.0500)  time: 0.7282  data: 0.1304  max mem: 41794
Epoch: [197]  [120/312]  eta: 0:02:43  lr: 0.001186  min_lr: 0.001186  loss: 3.3765 (3.2102)  weight_decay: 0.0500 (0.0500)  time: 0.7194  data: 0.1377  max mem: 41794
Epoch: [197]  [130/312]  eta: 0:02:35  lr: 0.001186  min_lr: 0.001186  loss: 3.3093 (3.2099)  weight_decay: 0.0500 (0.0500)  time: 0.8799  data: 0.2078  max mem: 41794
Epoch: [197]  [140/312]  eta: 0:02:22  lr: 0.001185  min_lr: 0.001185  loss: 3.2577 (3.2065)  weight_decay: 0.0500 (0.0500)  time: 0.6666  data: 0.0722  max mem: 41794
Epoch: [197]  [150/312]  eta: 0:02:13  lr: 0.001184  min_lr: 0.001184  loss: 3.2516 (3.2130)  weight_decay: 0.0500 (0.0500)  time: 0.6358  data: 0.0445  max mem: 41794
Epoch: [197]  [160/312]  eta: 0:02:04  lr: 0.001184  min_lr: 0.001184  loss: 3.2516 (3.2174)  weight_decay: 0.0500 (0.0500)  time: 0.7551  data: 0.1322  max mem: 41794
Epoch: [197]  [170/312]  eta: 0:01:54  lr: 0.001183  min_lr: 0.001183  loss: 3.2498 (3.2167)  weight_decay: 0.0500 (0.0500)  time: 0.6577  data: 0.1058  max mem: 41794
Epoch: [197]  [180/312]  eta: 0:01:46  lr: 0.001182  min_lr: 0.001182  loss: 3.2507 (3.2152)  weight_decay: 0.0500 (0.0500)  time: 0.7425  data: 0.1628  max mem: 41794
Epoch: [197]  [190/312]  eta: 0:01:36  lr: 0.001182  min_lr: 0.001182  loss: 3.3530 (3.2248)  weight_decay: 0.0500 (0.0500)  time: 0.6990  data: 0.1480  max mem: 41794
Epoch: [197]  [200/312]  eta: 0:01:29  lr: 0.001181  min_lr: 0.001181  loss: 3.4656 (3.2348)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.1312  max mem: 41794
Epoch: [197]  [210/312]  eta: 0:01:21  lr: 0.001180  min_lr: 0.001180  loss: 3.4034 (3.2296)  weight_decay: 0.0500 (0.0500)  time: 0.8890  data: 0.2633  max mem: 41794
Epoch: [197]  [220/312]  eta: 0:01:12  lr: 0.001180  min_lr: 0.001180  loss: 3.5077 (3.2406)  weight_decay: 0.0500 (0.0500)  time: 0.7041  data: 0.1341  max mem: 41794
Epoch: [197]  [230/312]  eta: 0:01:04  lr: 0.001179  min_lr: 0.001179  loss: 3.5496 (3.2447)  weight_decay: 0.0500 (0.0500)  time: 0.6835  data: 0.1405  max mem: 41794
Epoch: [197]  [240/312]  eta: 0:00:56  lr: 0.001178  min_lr: 0.001178  loss: 3.5299 (3.2486)  weight_decay: 0.0500 (0.0500)  time: 0.8143  data: 0.2545  max mem: 41794
Epoch: [197]  [250/312]  eta: 0:00:48  lr: 0.001178  min_lr: 0.001178  loss: 3.3815 (3.2510)  weight_decay: 0.0500 (0.0500)  time: 0.6365  data: 0.1277  max mem: 41794
Epoch: [197]  [260/312]  eta: 0:00:40  lr: 0.001177  min_lr: 0.001177  loss: 3.3815 (3.2477)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.2096  max mem: 41794
Epoch: [197]  [270/312]  eta: 0:00:32  lr: 0.001176  min_lr: 0.001176  loss: 3.3491 (3.2514)  weight_decay: 0.0500 (0.0500)  time: 0.6918  data: 0.1964  max mem: 41794
Epoch: [197]  [280/312]  eta: 0:00:24  lr: 0.001176  min_lr: 0.001176  loss: 3.3873 (3.2558)  weight_decay: 0.0500 (0.0500)  time: 0.7055  data: 0.2093  max mem: 41794
Epoch: [197]  [290/312]  eta: 0:00:17  lr: 0.001175  min_lr: 0.001175  loss: 3.4323 (3.2604)  weight_decay: 0.0500 (0.0500)  time: 0.8338  data: 0.3361  max mem: 41794
Epoch: [197]  [300/312]  eta: 0:00:09  lr: 0.001174  min_lr: 0.001174  loss: 3.4390 (3.2571)  weight_decay: 0.0500 (0.0500)  time: 0.6145  data: 0.1271  max mem: 41794
Epoch: [197]  [310/312]  eta: 0:00:01  lr: 0.001174  min_lr: 0.001174  loss: 3.1115 (3.2445)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [197]  [311/312]  eta: 0:00:00  lr: 0.001174  min_lr: 0.001174  loss: 3.1115 (3.2457)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0001  max mem: 41794
Epoch: [197] Total time: 0:03:58 (0.7634 s / it)
Averaged stats: lr: 0.001174  min_lr: 0.001174  loss: 3.1115 (3.2719)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 0.9333 (0.9333)  acc1: 80.0781 (80.0781)  acc5: 95.3125 (95.3125)  time: 8.7377  data: 8.5250  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0968 (1.0975)  acc1: 76.8229 (75.7760)  acc5: 93.2292 (93.1840)  time: 1.1191  data: 0.9473  max mem: 41794
Test: Total time: 0:00:10 (1.1305 s / it)
* Acc@1 76.192 Acc@5 93.304 loss 1.093
Accuracy of the model on the 50000 test images: 76.2%
Max accuracy: 76.19%
Epoch: [198]  [  0/312]  eta: 1:18:04  lr: 0.001174  min_lr: 0.001174  loss: 3.5207 (3.5207)  weight_decay: 0.0500 (0.0500)  time: 15.0154  data: 13.9235  max mem: 41794
Epoch: [198]  [ 10/312]  eta: 0:09:42  lr: 0.001173  min_lr: 0.001173  loss: 3.4868 (3.5389)  weight_decay: 0.0500 (0.0500)  time: 1.9292  data: 1.2664  max mem: 41794
Epoch: [198]  [ 20/312]  eta: 0:07:01  lr: 0.001172  min_lr: 0.001172  loss: 3.4087 (3.3090)  weight_decay: 0.0500 (0.0500)  time: 0.7640  data: 0.1185  max mem: 41794
Epoch: [198]  [ 30/312]  eta: 0:05:23  lr: 0.001172  min_lr: 0.001172  loss: 3.4215 (3.3392)  weight_decay: 0.0500 (0.0500)  time: 0.7150  data: 0.1185  max mem: 41794
Epoch: [198]  [ 40/312]  eta: 0:04:50  lr: 0.001171  min_lr: 0.001171  loss: 3.4015 (3.2997)  weight_decay: 0.0500 (0.0500)  time: 0.6740  data: 0.1392  max mem: 41794
Epoch: [198]  [ 50/312]  eta: 0:04:34  lr: 0.001170  min_lr: 0.001170  loss: 3.2420 (3.2966)  weight_decay: 0.0500 (0.0500)  time: 0.8926  data: 0.2987  max mem: 41794
Epoch: [198]  [ 60/312]  eta: 0:04:01  lr: 0.001170  min_lr: 0.001170  loss: 3.3315 (3.2595)  weight_decay: 0.0500 (0.0500)  time: 0.7363  data: 0.1607  max mem: 41794
Epoch: [198]  [ 70/312]  eta: 0:03:49  lr: 0.001169  min_lr: 0.001169  loss: 3.2609 (3.2325)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.1152  max mem: 41794
Epoch: [198]  [ 80/312]  eta: 0:03:34  lr: 0.001168  min_lr: 0.001168  loss: 3.0842 (3.2148)  weight_decay: 0.0500 (0.0500)  time: 0.8234  data: 0.1976  max mem: 41794
Epoch: [198]  [ 90/312]  eta: 0:03:16  lr: 0.001168  min_lr: 0.001168  loss: 3.3248 (3.2370)  weight_decay: 0.0500 (0.0500)  time: 0.6567  data: 0.1193  max mem: 41794
Epoch: [198]  [100/312]  eta: 0:03:07  lr: 0.001167  min_lr: 0.001167  loss: 3.3620 (3.2402)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1820  max mem: 41794
Epoch: [198]  [110/312]  eta: 0:02:51  lr: 0.001166  min_lr: 0.001166  loss: 3.4268 (3.2543)  weight_decay: 0.0500 (0.0500)  time: 0.6903  data: 0.1468  max mem: 41794
Epoch: [198]  [120/312]  eta: 0:02:42  lr: 0.001166  min_lr: 0.001166  loss: 3.5185 (3.2664)  weight_decay: 0.0500 (0.0500)  time: 0.6531  data: 0.1022  max mem: 41794
Epoch: [198]  [130/312]  eta: 0:02:35  lr: 0.001165  min_lr: 0.001165  loss: 3.3128 (3.2576)  weight_decay: 0.0500 (0.0500)  time: 0.8644  data: 0.2577  max mem: 41794
Epoch: [198]  [140/312]  eta: 0:02:22  lr: 0.001164  min_lr: 0.001164  loss: 3.3128 (3.2757)  weight_decay: 0.0500 (0.0500)  time: 0.7042  data: 0.1566  max mem: 41794
Epoch: [198]  [150/312]  eta: 0:02:14  lr: 0.001164  min_lr: 0.001164  loss: 3.4414 (3.2882)  weight_decay: 0.0500 (0.0500)  time: 0.6692  data: 0.1627  max mem: 41794
Epoch: [198]  [160/312]  eta: 0:02:05  lr: 0.001163  min_lr: 0.001163  loss: 3.4414 (3.2984)  weight_decay: 0.0500 (0.0500)  time: 0.8397  data: 0.3300  max mem: 41794
Epoch: [198]  [170/312]  eta: 0:01:54  lr: 0.001162  min_lr: 0.001162  loss: 3.5260 (3.3004)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1678  max mem: 41794
Epoch: [198]  [180/312]  eta: 0:01:47  lr: 0.001162  min_lr: 0.001162  loss: 3.5379 (3.3016)  weight_decay: 0.0500 (0.0500)  time: 0.6835  data: 0.1618  max mem: 41794
Epoch: [198]  [190/312]  eta: 0:01:36  lr: 0.001161  min_lr: 0.001161  loss: 3.4026 (3.2981)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.1618  max mem: 41794
Epoch: [198]  [200/312]  eta: 0:01:29  lr: 0.001161  min_lr: 0.001161  loss: 3.3282 (3.2916)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.1827  max mem: 41794
Epoch: [198]  [210/312]  eta: 0:01:21  lr: 0.001160  min_lr: 0.001160  loss: 3.2750 (3.2803)  weight_decay: 0.0500 (0.0500)  time: 0.8522  data: 0.3499  max mem: 41794
Epoch: [198]  [220/312]  eta: 0:01:12  lr: 0.001159  min_lr: 0.001159  loss: 3.3768 (3.2825)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.1678  max mem: 41794
Epoch: [198]  [230/312]  eta: 0:01:04  lr: 0.001159  min_lr: 0.001159  loss: 3.4742 (3.2885)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.1558  max mem: 41794
Epoch: [198]  [240/312]  eta: 0:00:56  lr: 0.001158  min_lr: 0.001158  loss: 3.4613 (3.2920)  weight_decay: 0.0500 (0.0500)  time: 0.7628  data: 0.2572  max mem: 41794
Epoch: [198]  [250/312]  eta: 0:00:48  lr: 0.001157  min_lr: 0.001157  loss: 3.4186 (3.2900)  weight_decay: 0.0500 (0.0500)  time: 0.6991  data: 0.1498  max mem: 41794
Epoch: [198]  [260/312]  eta: 0:00:40  lr: 0.001157  min_lr: 0.001157  loss: 3.2561 (3.2926)  weight_decay: 0.0500 (0.0500)  time: 0.7899  data: 0.1847  max mem: 41794
Epoch: [198]  [270/312]  eta: 0:00:32  lr: 0.001156  min_lr: 0.001156  loss: 3.4319 (3.2885)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.1398  max mem: 41794
Epoch: [198]  [280/312]  eta: 0:00:24  lr: 0.001155  min_lr: 0.001155  loss: 3.4714 (3.2924)  weight_decay: 0.0500 (0.0500)  time: 0.6728  data: 0.1221  max mem: 41794
Epoch: [198]  [290/312]  eta: 0:00:17  lr: 0.001155  min_lr: 0.001155  loss: 3.3271 (3.2864)  weight_decay: 0.0500 (0.0500)  time: 0.7850  data: 0.1959  max mem: 41794
Epoch: [198]  [300/312]  eta: 0:00:09  lr: 0.001154  min_lr: 0.001154  loss: 3.2494 (3.2854)  weight_decay: 0.0500 (0.0500)  time: 0.6018  data: 0.0769  max mem: 41794
Epoch: [198]  [310/312]  eta: 0:00:01  lr: 0.001153  min_lr: 0.001153  loss: 3.4360 (3.2833)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [198]  [311/312]  eta: 0:00:00  lr: 0.001153  min_lr: 0.001153  loss: 3.2481 (3.2812)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [198] Total time: 0:03:58 (0.7629 s / it)
Averaged stats: lr: 0.001153  min_lr: 0.001153  loss: 3.2481 (3.2694)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.0099 (1.0099)  acc1: 80.3385 (80.3385)  acc5: 94.4010 (94.4010)  time: 8.9076  data: 8.7061  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2883 (1.2383)  acc1: 74.4792 (74.7840)  acc5: 93.6198 (92.8960)  time: 1.1429  data: 0.9674  max mem: 41794
Test: Total time: 0:00:10 (1.1850 s / it)
* Acc@1 75.206 Acc@5 92.904 loss 1.238
Accuracy of the model on the 50000 test images: 75.2%
Max accuracy: 76.19%
Epoch: [199]  [  0/312]  eta: 1:24:29  lr: 0.001153  min_lr: 0.001153  loss: 3.5271 (3.5271)  weight_decay: 0.0500 (0.0500)  time: 16.2500  data: 13.4939  max mem: 41794
Epoch: [199]  [ 10/312]  eta: 0:10:25  lr: 0.001153  min_lr: 0.001153  loss: 3.3169 (3.0800)  weight_decay: 0.0500 (0.0500)  time: 2.0697  data: 1.2825  max mem: 41794
Epoch: [199]  [ 20/312]  eta: 0:07:14  lr: 0.001152  min_lr: 0.001152  loss: 3.0501 (3.0506)  weight_decay: 0.0500 (0.0500)  time: 0.7507  data: 0.0952  max mem: 41794
Epoch: [199]  [ 30/312]  eta: 0:05:35  lr: 0.001151  min_lr: 0.001151  loss: 3.0501 (3.0859)  weight_decay: 0.0500 (0.0500)  time: 0.7050  data: 0.0949  max mem: 41794
Epoch: [199]  [ 40/312]  eta: 0:04:51  lr: 0.001151  min_lr: 0.001151  loss: 3.2448 (3.1118)  weight_decay: 0.0500 (0.0500)  time: 0.6351  data: 0.1374  max mem: 41794
Epoch: [199]  [ 50/312]  eta: 0:04:30  lr: 0.001150  min_lr: 0.001150  loss: 3.2448 (3.1206)  weight_decay: 0.0500 (0.0500)  time: 0.7914  data: 0.2660  max mem: 41794
Epoch: [199]  [ 60/312]  eta: 0:03:57  lr: 0.001149  min_lr: 0.001149  loss: 3.2909 (3.1482)  weight_decay: 0.0500 (0.0500)  time: 0.6804  data: 0.1592  max mem: 41794
Epoch: [199]  [ 70/312]  eta: 0:03:50  lr: 0.001149  min_lr: 0.001149  loss: 3.2909 (3.1618)  weight_decay: 0.0500 (0.0500)  time: 0.7418  data: 0.1743  max mem: 41794
Epoch: [199]  [ 80/312]  eta: 0:03:33  lr: 0.001148  min_lr: 0.001148  loss: 3.3302 (3.1797)  weight_decay: 0.0500 (0.0500)  time: 0.8545  data: 0.2711  max mem: 41794
Epoch: [199]  [ 90/312]  eta: 0:03:18  lr: 0.001147  min_lr: 0.001147  loss: 3.4577 (3.2167)  weight_decay: 0.0500 (0.0500)  time: 0.6912  data: 0.1066  max mem: 41794
Epoch: [199]  [100/312]  eta: 0:03:07  lr: 0.001147  min_lr: 0.001147  loss: 3.4607 (3.2336)  weight_decay: 0.0500 (0.0500)  time: 0.7446  data: 0.1493  max mem: 41794
Epoch: [199]  [110/312]  eta: 0:02:51  lr: 0.001146  min_lr: 0.001146  loss: 3.3351 (3.2385)  weight_decay: 0.0500 (0.0500)  time: 0.6561  data: 0.1401  max mem: 41794
Epoch: [199]  [120/312]  eta: 0:02:43  lr: 0.001145  min_lr: 0.001145  loss: 3.2389 (3.2246)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1561  max mem: 41794
Epoch: [199]  [130/312]  eta: 0:02:35  lr: 0.001145  min_lr: 0.001145  loss: 3.2899 (3.2154)  weight_decay: 0.0500 (0.0500)  time: 0.8544  data: 0.3425  max mem: 41794
Epoch: [199]  [140/312]  eta: 0:02:21  lr: 0.001144  min_lr: 0.001144  loss: 3.2932 (3.2165)  weight_decay: 0.0500 (0.0500)  time: 0.6662  data: 0.1871  max mem: 41794
Epoch: [199]  [150/312]  eta: 0:02:14  lr: 0.001143  min_lr: 0.001143  loss: 3.3526 (3.2128)  weight_decay: 0.0500 (0.0500)  time: 0.7092  data: 0.2314  max mem: 41794
Epoch: [199]  [160/312]  eta: 0:02:05  lr: 0.001143  min_lr: 0.001143  loss: 3.2638 (3.2202)  weight_decay: 0.0500 (0.0500)  time: 0.8556  data: 0.3562  max mem: 41794
Epoch: [199]  [170/312]  eta: 0:01:55  lr: 0.001142  min_lr: 0.001142  loss: 3.1270 (3.2056)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1623  max mem: 41794
Epoch: [199]  [180/312]  eta: 0:01:48  lr: 0.001142  min_lr: 0.001142  loss: 3.0633 (3.2072)  weight_decay: 0.0500 (0.0500)  time: 0.7377  data: 0.2087  max mem: 41794
Epoch: [199]  [190/312]  eta: 0:01:37  lr: 0.001141  min_lr: 0.001141  loss: 3.1003 (3.1969)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.1720  max mem: 41794
Epoch: [199]  [200/312]  eta: 0:01:29  lr: 0.001140  min_lr: 0.001140  loss: 3.1691 (3.2019)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.1473  max mem: 41794
Epoch: [199]  [210/312]  eta: 0:01:22  lr: 0.001140  min_lr: 0.001140  loss: 3.4383 (3.2075)  weight_decay: 0.0500 (0.0500)  time: 0.8599  data: 0.3512  max mem: 41794
Epoch: [199]  [220/312]  eta: 0:01:13  lr: 0.001139  min_lr: 0.001139  loss: 3.3630 (3.2114)  weight_decay: 0.0500 (0.0500)  time: 0.7064  data: 0.2045  max mem: 41794
Epoch: [199]  [230/312]  eta: 0:01:05  lr: 0.001138  min_lr: 0.001138  loss: 3.3713 (3.2133)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1692  max mem: 41794
Epoch: [199]  [240/312]  eta: 0:00:57  lr: 0.001138  min_lr: 0.001138  loss: 3.4509 (3.2141)  weight_decay: 0.0500 (0.0500)  time: 0.9006  data: 0.3867  max mem: 41794
Epoch: [199]  [250/312]  eta: 0:00:49  lr: 0.001137  min_lr: 0.001137  loss: 3.4573 (3.2182)  weight_decay: 0.0500 (0.0500)  time: 0.7211  data: 0.2182  max mem: 41794
Epoch: [199]  [260/312]  eta: 0:00:41  lr: 0.001136  min_lr: 0.001136  loss: 3.4902 (3.2176)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.1677  max mem: 41794
Epoch: [199]  [270/312]  eta: 0:00:32  lr: 0.001136  min_lr: 0.001136  loss: 3.3892 (3.2202)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.1677  max mem: 41794
Epoch: [199]  [280/312]  eta: 0:00:25  lr: 0.001135  min_lr: 0.001135  loss: 3.4348 (3.2291)  weight_decay: 0.0500 (0.0500)  time: 0.6805  data: 0.1834  max mem: 41794
Epoch: [199]  [290/312]  eta: 0:00:17  lr: 0.001134  min_lr: 0.001134  loss: 3.2591 (3.2171)  weight_decay: 0.0500 (0.0500)  time: 0.7902  data: 0.2976  max mem: 41794
Epoch: [199]  [300/312]  eta: 0:00:09  lr: 0.001134  min_lr: 0.001134  loss: 3.1403 (3.2194)  weight_decay: 0.0500 (0.0500)  time: 0.5955  data: 0.1145  max mem: 41794
Epoch: [199]  [310/312]  eta: 0:00:01  lr: 0.001133  min_lr: 0.001133  loss: 3.3111 (3.2223)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [199]  [311/312]  eta: 0:00:00  lr: 0.001133  min_lr: 0.001133  loss: 3.3263 (3.2226)  weight_decay: 0.0500 (0.0500)  time: 0.4625  data: 0.0001  max mem: 41794
Epoch: [199] Total time: 0:03:59 (0.7673 s / it)
Averaged stats: lr: 0.001133  min_lr: 0.001133  loss: 3.3263 (3.2715)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 1.0982 (1.0982)  acc1: 80.2083 (80.2083)  acc5: 94.9219 (94.9219)  time: 7.9542  data: 7.7448  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3449 (1.2427)  acc1: 75.3906 (75.4720)  acc5: 93.3594 (93.3280)  time: 1.0519  data: 0.8744  max mem: 41794
Test: Total time: 0:00:09 (1.0626 s / it)
* Acc@1 75.864 Acc@5 93.258 loss 1.244
Accuracy of the model on the 50000 test images: 75.9%
Max accuracy: 76.19%
Epoch: [200]  [  0/312]  eta: 1:22:18  lr: 0.001133  min_lr: 0.001133  loss: 3.7743 (3.7743)  weight_decay: 0.0500 (0.0500)  time: 15.8280  data: 12.5029  max mem: 41794
Epoch: [200]  [ 10/312]  eta: 0:12:01  lr: 0.001132  min_lr: 0.001132  loss: 3.2289 (3.1347)  weight_decay: 0.0500 (0.0500)  time: 2.3877  data: 1.3286  max mem: 41794
Epoch: [200]  [ 20/312]  eta: 0:07:27  lr: 0.001132  min_lr: 0.001132  loss: 3.2289 (3.1492)  weight_decay: 0.0500 (0.0500)  time: 0.8167  data: 0.1127  max mem: 41794
Epoch: [200]  [ 30/312]  eta: 0:05:40  lr: 0.001131  min_lr: 0.001131  loss: 3.3297 (3.1660)  weight_decay: 0.0500 (0.0500)  time: 0.5588  data: 0.0130  max mem: 41794
Epoch: [200]  [ 40/312]  eta: 0:04:42  lr: 0.001130  min_lr: 0.001130  loss: 3.2972 (3.1974)  weight_decay: 0.0500 (0.0500)  time: 0.5224  data: 0.0120  max mem: 41794
Epoch: [200]  [ 50/312]  eta: 0:04:09  lr: 0.001130  min_lr: 0.001130  loss: 3.4612 (3.2522)  weight_decay: 0.0500 (0.0500)  time: 0.5587  data: 0.0066  max mem: 41794
Epoch: [200]  [ 60/312]  eta: 0:03:41  lr: 0.001129  min_lr: 0.001129  loss: 3.4612 (3.2624)  weight_decay: 0.0500 (0.0500)  time: 0.5510  data: 0.0057  max mem: 41794
Epoch: [200]  [ 70/312]  eta: 0:03:34  lr: 0.001128  min_lr: 0.001128  loss: 3.3645 (3.2647)  weight_decay: 0.0500 (0.0500)  time: 0.7122  data: 0.0601  max mem: 41794
Epoch: [200]  [ 80/312]  eta: 0:03:22  lr: 0.001128  min_lr: 0.001128  loss: 3.3111 (3.2472)  weight_decay: 0.0500 (0.0500)  time: 0.8555  data: 0.0699  max mem: 41794
Epoch: [200]  [ 90/312]  eta: 0:03:07  lr: 0.001127  min_lr: 0.001127  loss: 3.1578 (3.2336)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.0299  max mem: 41794
Epoch: [200]  [100/312]  eta: 0:02:58  lr: 0.001126  min_lr: 0.001126  loss: 3.3423 (3.2380)  weight_decay: 0.0500 (0.0500)  time: 0.7159  data: 0.0233  max mem: 41794
Epoch: [200]  [110/312]  eta: 0:02:44  lr: 0.001126  min_lr: 0.001126  loss: 3.3973 (3.2480)  weight_decay: 0.0500 (0.0500)  time: 0.6675  data: 0.0172  max mem: 41794
Epoch: [200]  [120/312]  eta: 0:02:37  lr: 0.001125  min_lr: 0.001125  loss: 3.0968 (3.2084)  weight_decay: 0.0500 (0.0500)  time: 0.7075  data: 0.0492  max mem: 41794
Epoch: [200]  [130/312]  eta: 0:02:29  lr: 0.001125  min_lr: 0.001125  loss: 3.0966 (3.2120)  weight_decay: 0.0500 (0.0500)  time: 0.8529  data: 0.0710  max mem: 41794
Epoch: [200]  [140/312]  eta: 0:02:17  lr: 0.001124  min_lr: 0.001124  loss: 3.5397 (3.2336)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.0308  max mem: 41794
Epoch: [200]  [150/312]  eta: 0:02:09  lr: 0.001123  min_lr: 0.001123  loss: 3.5460 (3.2400)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.0410  max mem: 41794
Epoch: [200]  [160/312]  eta: 0:02:01  lr: 0.001123  min_lr: 0.001123  loss: 3.2253 (3.2314)  weight_decay: 0.0500 (0.0500)  time: 0.8061  data: 0.0691  max mem: 41794
Epoch: [200]  [170/312]  eta: 0:01:52  lr: 0.001122  min_lr: 0.001122  loss: 3.1016 (3.2298)  weight_decay: 0.0500 (0.0500)  time: 0.7051  data: 0.0494  max mem: 41794
Epoch: [200]  [180/312]  eta: 0:01:44  lr: 0.001121  min_lr: 0.001121  loss: 3.3537 (3.2384)  weight_decay: 0.0500 (0.0500)  time: 0.7013  data: 0.0511  max mem: 41794
Epoch: [200]  [190/312]  eta: 0:01:35  lr: 0.001121  min_lr: 0.001121  loss: 3.0652 (3.2313)  weight_decay: 0.0500 (0.0500)  time: 0.6919  data: 0.0537  max mem: 41794
Epoch: [200]  [200/312]  eta: 0:01:27  lr: 0.001120  min_lr: 0.001120  loss: 3.0652 (3.2261)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.0508  max mem: 41794
Epoch: [200]  [210/312]  eta: 0:01:19  lr: 0.001119  min_lr: 0.001119  loss: 3.3607 (3.2368)  weight_decay: 0.0500 (0.0500)  time: 0.7445  data: 0.0686  max mem: 41794
Epoch: [200]  [220/312]  eta: 0:01:11  lr: 0.001119  min_lr: 0.001119  loss: 3.3607 (3.2303)  weight_decay: 0.0500 (0.0500)  time: 0.7105  data: 0.0902  max mem: 41794
Epoch: [200]  [230/312]  eta: 0:01:03  lr: 0.001118  min_lr: 0.001118  loss: 3.2518 (3.2262)  weight_decay: 0.0500 (0.0500)  time: 0.7150  data: 0.1058  max mem: 41794
Epoch: [200]  [240/312]  eta: 0:00:55  lr: 0.001117  min_lr: 0.001117  loss: 3.3056 (3.2340)  weight_decay: 0.0500 (0.0500)  time: 0.7568  data: 0.1054  max mem: 41794
Epoch: [200]  [250/312]  eta: 0:00:47  lr: 0.001117  min_lr: 0.001117  loss: 3.3731 (3.2250)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.0797  max mem: 41794
Epoch: [200]  [260/312]  eta: 0:00:39  lr: 0.001116  min_lr: 0.001116  loss: 3.3731 (3.2326)  weight_decay: 0.0500 (0.0500)  time: 0.7304  data: 0.1068  max mem: 41794
Epoch: [200]  [270/312]  eta: 0:00:32  lr: 0.001115  min_lr: 0.001115  loss: 3.5415 (3.2421)  weight_decay: 0.0500 (0.0500)  time: 0.7336  data: 0.0952  max mem: 41794
Epoch: [200]  [280/312]  eta: 0:00:24  lr: 0.001115  min_lr: 0.001115  loss: 3.4113 (3.2369)  weight_decay: 0.0500 (0.0500)  time: 0.6759  data: 0.0792  max mem: 41794
Epoch: [200]  [290/312]  eta: 0:00:16  lr: 0.001114  min_lr: 0.001114  loss: 3.4115 (3.2467)  weight_decay: 0.0500 (0.0500)  time: 0.8395  data: 0.1406  max mem: 41794
Epoch: [200]  [300/312]  eta: 0:00:09  lr: 0.001114  min_lr: 0.001114  loss: 3.4158 (3.2506)  weight_decay: 0.0500 (0.0500)  time: 0.7306  data: 0.1036  max mem: 41794
Epoch: [200]  [310/312]  eta: 0:00:01  lr: 0.001113  min_lr: 0.001113  loss: 3.3093 (3.2519)  weight_decay: 0.0500 (0.0500)  time: 0.4958  data: 0.0226  max mem: 41794
Epoch: [200]  [311/312]  eta: 0:00:00  lr: 0.001113  min_lr: 0.001113  loss: 3.3093 (3.2546)  weight_decay: 0.0500 (0.0500)  time: 0.4712  data: 0.0002  max mem: 41794
Epoch: [200] Total time: 0:03:56 (0.7567 s / it)
Averaged stats: lr: 0.001113  min_lr: 0.001113  loss: 3.3093 (3.2704)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 1.0592 (1.0592)  acc1: 81.7708 (81.7708)  acc5: 95.0521 (95.0521)  time: 8.7971  data: 8.5922  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3603 (1.2619)  acc1: 75.0000 (76.0800)  acc5: 92.1875 (92.9760)  time: 1.1277  data: 0.9547  max mem: 41794
Test: Total time: 0:00:10 (1.1506 s / it)
* Acc@1 76.028 Acc@5 93.174 loss 1.262
Accuracy of the model on the 50000 test images: 76.0%
Max accuracy: 76.19%
Epoch: [201]  [  0/312]  eta: 1:23:39  lr: 0.001113  min_lr: 0.001113  loss: 3.2756 (3.2756)  weight_decay: 0.0500 (0.0500)  time: 16.0869  data: 14.9878  max mem: 41794
Epoch: [201]  [ 10/312]  eta: 0:10:27  lr: 0.001112  min_lr: 0.001112  loss: 3.2756 (3.2466)  weight_decay: 0.0500 (0.0500)  time: 2.0787  data: 1.3670  max mem: 41794
Epoch: [201]  [ 20/312]  eta: 0:07:14  lr: 0.001112  min_lr: 0.001112  loss: 3.2143 (3.2716)  weight_decay: 0.0500 (0.0500)  time: 0.7581  data: 0.1511  max mem: 41794
Epoch: [201]  [ 30/312]  eta: 0:05:31  lr: 0.001111  min_lr: 0.001111  loss: 3.3315 (3.2531)  weight_decay: 0.0500 (0.0500)  time: 0.6773  data: 0.1490  max mem: 41794
Epoch: [201]  [ 40/312]  eta: 0:04:49  lr: 0.001110  min_lr: 0.001110  loss: 3.2736 (3.2206)  weight_decay: 0.0500 (0.0500)  time: 0.6209  data: 0.1209  max mem: 41794
Epoch: [201]  [ 50/312]  eta: 0:04:25  lr: 0.001110  min_lr: 0.001110  loss: 3.1188 (3.1903)  weight_decay: 0.0500 (0.0500)  time: 0.7677  data: 0.2581  max mem: 41794
Epoch: [201]  [ 60/312]  eta: 0:03:54  lr: 0.001109  min_lr: 0.001109  loss: 3.2426 (3.1993)  weight_decay: 0.0500 (0.0500)  time: 0.6498  data: 0.1380  max mem: 41794
Epoch: [201]  [ 70/312]  eta: 0:03:42  lr: 0.001108  min_lr: 0.001108  loss: 3.3411 (3.2295)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1895  max mem: 41794
Epoch: [201]  [ 80/312]  eta: 0:03:33  lr: 0.001108  min_lr: 0.001108  loss: 3.5149 (3.2428)  weight_decay: 0.0500 (0.0500)  time: 0.8949  data: 0.4027  max mem: 41794
Epoch: [201]  [ 90/312]  eta: 0:03:13  lr: 0.001107  min_lr: 0.001107  loss: 3.1465 (3.2313)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.2143  max mem: 41794
Epoch: [201]  [100/312]  eta: 0:03:05  lr: 0.001106  min_lr: 0.001106  loss: 3.3970 (3.2469)  weight_decay: 0.0500 (0.0500)  time: 0.7003  data: 0.2135  max mem: 41794
Epoch: [201]  [110/312]  eta: 0:02:50  lr: 0.001106  min_lr: 0.001106  loss: 3.6090 (3.2604)  weight_decay: 0.0500 (0.0500)  time: 0.7068  data: 0.2129  max mem: 41794
Epoch: [201]  [120/312]  eta: 0:02:43  lr: 0.001105  min_lr: 0.001105  loss: 3.3773 (3.2534)  weight_decay: 0.0500 (0.0500)  time: 0.7252  data: 0.2371  max mem: 41794
Epoch: [201]  [130/312]  eta: 0:02:34  lr: 0.001104  min_lr: 0.001104  loss: 3.2533 (3.2431)  weight_decay: 0.0500 (0.0500)  time: 0.8980  data: 0.4031  max mem: 41794
Epoch: [201]  [140/312]  eta: 0:02:21  lr: 0.001104  min_lr: 0.001104  loss: 3.1225 (3.2284)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.1667  max mem: 41794
Epoch: [201]  [150/312]  eta: 0:02:13  lr: 0.001103  min_lr: 0.001103  loss: 3.4028 (3.2505)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.1670  max mem: 41794
Epoch: [201]  [160/312]  eta: 0:02:05  lr: 0.001102  min_lr: 0.001102  loss: 3.5649 (3.2602)  weight_decay: 0.0500 (0.0500)  time: 0.8415  data: 0.3470  max mem: 41794
Epoch: [201]  [170/312]  eta: 0:01:54  lr: 0.001102  min_lr: 0.001102  loss: 3.5056 (3.2668)  weight_decay: 0.0500 (0.0500)  time: 0.6758  data: 0.1806  max mem: 41794
Epoch: [201]  [180/312]  eta: 0:01:47  lr: 0.001101  min_lr: 0.001101  loss: 3.4475 (3.2691)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.1626  max mem: 41794
Epoch: [201]  [190/312]  eta: 0:01:37  lr: 0.001101  min_lr: 0.001101  loss: 3.4475 (3.2825)  weight_decay: 0.0500 (0.0500)  time: 0.7092  data: 0.1629  max mem: 41794
Epoch: [201]  [200/312]  eta: 0:01:29  lr: 0.001100  min_lr: 0.001100  loss: 3.4549 (3.2853)  weight_decay: 0.0500 (0.0500)  time: 0.6929  data: 0.1461  max mem: 41794
Epoch: [201]  [210/312]  eta: 0:01:22  lr: 0.001099  min_lr: 0.001099  loss: 3.3972 (3.2778)  weight_decay: 0.0500 (0.0500)  time: 0.8846  data: 0.2916  max mem: 41794
Epoch: [201]  [220/312]  eta: 0:01:12  lr: 0.001099  min_lr: 0.001099  loss: 3.3248 (3.2803)  weight_decay: 0.0500 (0.0500)  time: 0.6861  data: 0.1464  max mem: 41794
Epoch: [201]  [230/312]  eta: 0:01:05  lr: 0.001098  min_lr: 0.001098  loss: 3.2734 (3.2797)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1393  max mem: 41794
Epoch: [201]  [240/312]  eta: 0:00:57  lr: 0.001097  min_lr: 0.001097  loss: 3.2734 (3.2819)  weight_decay: 0.0500 (0.0500)  time: 0.7991  data: 0.2351  max mem: 41794
Epoch: [201]  [250/312]  eta: 0:00:48  lr: 0.001097  min_lr: 0.001097  loss: 3.3585 (3.2774)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.1348  max mem: 41794
Epoch: [201]  [260/312]  eta: 0:00:41  lr: 0.001096  min_lr: 0.001096  loss: 3.3500 (3.2782)  weight_decay: 0.0500 (0.0500)  time: 0.7937  data: 0.1658  max mem: 41794
Epoch: [201]  [270/312]  eta: 0:00:32  lr: 0.001095  min_lr: 0.001095  loss: 3.5395 (3.2858)  weight_decay: 0.0500 (0.0500)  time: 0.7243  data: 0.1313  max mem: 41794
Epoch: [201]  [280/312]  eta: 0:00:25  lr: 0.001095  min_lr: 0.001095  loss: 3.5021 (3.2833)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.0876  max mem: 41794
Epoch: [201]  [290/312]  eta: 0:00:17  lr: 0.001094  min_lr: 0.001094  loss: 3.3040 (3.2804)  weight_decay: 0.0500 (0.0500)  time: 0.7612  data: 0.1134  max mem: 41794
Epoch: [201]  [300/312]  eta: 0:00:09  lr: 0.001094  min_lr: 0.001094  loss: 3.5007 (3.2841)  weight_decay: 0.0500 (0.0500)  time: 0.5778  data: 0.0301  max mem: 41794
Epoch: [201]  [310/312]  eta: 0:00:01  lr: 0.001093  min_lr: 0.001093  loss: 3.4874 (3.2800)  weight_decay: 0.0500 (0.0500)  time: 0.4670  data: 0.0001  max mem: 41794
Epoch: [201]  [311/312]  eta: 0:00:00  lr: 0.001093  min_lr: 0.001093  loss: 3.4874 (3.2759)  weight_decay: 0.0500 (0.0500)  time: 0.4663  data: 0.0001  max mem: 41794
Epoch: [201] Total time: 0:03:58 (0.7655 s / it)
Averaged stats: lr: 0.001093  min_lr: 0.001093  loss: 3.4874 (3.2459)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 0.9377 (0.9377)  acc1: 82.0312 (82.0312)  acc5: 95.1823 (95.1823)  time: 7.8128  data: 7.6038  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1855 (1.1431)  acc1: 75.9115 (75.8720)  acc5: 94.0104 (93.4240)  time: 1.0173  data: 0.8461  max mem: 41794
Test: Total time: 0:00:09 (1.0278 s / it)
* Acc@1 76.190 Acc@5 93.468 loss 1.144
Accuracy of the model on the 50000 test images: 76.2%
Max accuracy: 76.19%
Epoch: [202]  [  0/312]  eta: 1:25:31  lr: 0.001093  min_lr: 0.001093  loss: 3.7892 (3.7892)  weight_decay: 0.0500 (0.0500)  time: 16.4486  data: 15.4407  max mem: 41794
Epoch: [202]  [ 10/312]  eta: 0:11:07  lr: 0.001092  min_lr: 0.001092  loss: 3.5650 (3.2575)  weight_decay: 0.0500 (0.0500)  time: 2.2089  data: 1.4220  max mem: 41794
Epoch: [202]  [ 20/312]  eta: 0:07:18  lr: 0.001091  min_lr: 0.001091  loss: 3.2477 (3.2673)  weight_decay: 0.0500 (0.0500)  time: 0.7548  data: 0.0831  max mem: 41794
Epoch: [202]  [ 30/312]  eta: 0:05:33  lr: 0.001091  min_lr: 0.001091  loss: 3.2477 (3.2503)  weight_decay: 0.0500 (0.0500)  time: 0.6187  data: 0.0734  max mem: 41794
Epoch: [202]  [ 40/312]  eta: 0:04:42  lr: 0.001090  min_lr: 0.001090  loss: 3.3516 (3.2331)  weight_decay: 0.0500 (0.0500)  time: 0.5548  data: 0.0007  max mem: 41794
Epoch: [202]  [ 50/312]  eta: 0:04:19  lr: 0.001090  min_lr: 0.001090  loss: 3.3755 (3.2405)  weight_decay: 0.0500 (0.0500)  time: 0.6952  data: 0.0686  max mem: 41794
Epoch: [202]  [ 60/312]  eta: 0:03:50  lr: 0.001089  min_lr: 0.001089  loss: 3.3902 (3.2526)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.0686  max mem: 41794
Epoch: [202]  [ 70/312]  eta: 0:03:38  lr: 0.001088  min_lr: 0.001088  loss: 3.5531 (3.2981)  weight_decay: 0.0500 (0.0500)  time: 0.6767  data: 0.0685  max mem: 41794
Epoch: [202]  [ 80/312]  eta: 0:03:28  lr: 0.001088  min_lr: 0.001088  loss: 3.5663 (3.3106)  weight_decay: 0.0500 (0.0500)  time: 0.8441  data: 0.1187  max mem: 41794
Epoch: [202]  [ 90/312]  eta: 0:03:11  lr: 0.001087  min_lr: 0.001087  loss: 3.4119 (3.2968)  weight_decay: 0.0500 (0.0500)  time: 0.7097  data: 0.0822  max mem: 41794
Epoch: [202]  [100/312]  eta: 0:03:05  lr: 0.001086  min_lr: 0.001086  loss: 3.4273 (3.3155)  weight_decay: 0.0500 (0.0500)  time: 0.7743  data: 0.1359  max mem: 41794
Epoch: [202]  [110/312]  eta: 0:02:49  lr: 0.001086  min_lr: 0.001086  loss: 3.4178 (3.3106)  weight_decay: 0.0500 (0.0500)  time: 0.7423  data: 0.1046  max mem: 41794
Epoch: [202]  [120/312]  eta: 0:02:41  lr: 0.001085  min_lr: 0.001085  loss: 3.0950 (3.2929)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.0543  max mem: 41794
Epoch: [202]  [130/312]  eta: 0:02:32  lr: 0.001084  min_lr: 0.001084  loss: 2.9901 (3.2724)  weight_decay: 0.0500 (0.0500)  time: 0.8365  data: 0.1281  max mem: 41794
Epoch: [202]  [140/312]  eta: 0:02:20  lr: 0.001084  min_lr: 0.001084  loss: 3.1525 (3.2581)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.0748  max mem: 41794
Epoch: [202]  [150/312]  eta: 0:02:12  lr: 0.001083  min_lr: 0.001083  loss: 3.3686 (3.2589)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.0587  max mem: 41794
Epoch: [202]  [160/312]  eta: 0:02:04  lr: 0.001083  min_lr: 0.001083  loss: 3.4002 (3.2638)  weight_decay: 0.0500 (0.0500)  time: 0.8356  data: 0.1136  max mem: 41794
Epoch: [202]  [170/312]  eta: 0:01:53  lr: 0.001082  min_lr: 0.001082  loss: 3.2461 (3.2537)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.0899  max mem: 41794
Epoch: [202]  [180/312]  eta: 0:01:46  lr: 0.001081  min_lr: 0.001081  loss: 3.2382 (3.2530)  weight_decay: 0.0500 (0.0500)  time: 0.7251  data: 0.1439  max mem: 41794
Epoch: [202]  [190/312]  eta: 0:01:36  lr: 0.001081  min_lr: 0.001081  loss: 3.2819 (3.2534)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.1100  max mem: 41794
Epoch: [202]  [200/312]  eta: 0:01:28  lr: 0.001080  min_lr: 0.001080  loss: 3.4028 (3.2590)  weight_decay: 0.0500 (0.0500)  time: 0.6556  data: 0.0735  max mem: 41794
Epoch: [202]  [210/312]  eta: 0:01:21  lr: 0.001079  min_lr: 0.001079  loss: 3.3777 (3.2578)  weight_decay: 0.0500 (0.0500)  time: 0.8441  data: 0.1806  max mem: 41794
Epoch: [202]  [220/312]  eta: 0:01:11  lr: 0.001079  min_lr: 0.001079  loss: 3.2967 (3.2564)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.1077  max mem: 41794
Epoch: [202]  [230/312]  eta: 0:01:04  lr: 0.001078  min_lr: 0.001078  loss: 3.3066 (3.2561)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1056  max mem: 41794
Epoch: [202]  [240/312]  eta: 0:00:56  lr: 0.001077  min_lr: 0.001077  loss: 3.4422 (3.2627)  weight_decay: 0.0500 (0.0500)  time: 0.8535  data: 0.1972  max mem: 41794
Epoch: [202]  [250/312]  eta: 0:00:48  lr: 0.001077  min_lr: 0.001077  loss: 3.4996 (3.2649)  weight_decay: 0.0500 (0.0500)  time: 0.6786  data: 0.0934  max mem: 41794
Epoch: [202]  [260/312]  eta: 0:00:40  lr: 0.001076  min_lr: 0.001076  loss: 3.3038 (3.2640)  weight_decay: 0.0500 (0.0500)  time: 0.7003  data: 0.0865  max mem: 41794
Epoch: [202]  [270/312]  eta: 0:00:32  lr: 0.001075  min_lr: 0.001075  loss: 3.2518 (3.2595)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.0852  max mem: 41794
Epoch: [202]  [280/312]  eta: 0:00:24  lr: 0.001075  min_lr: 0.001075  loss: 3.4130 (3.2610)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.0625  max mem: 41794
Epoch: [202]  [290/312]  eta: 0:00:17  lr: 0.001074  min_lr: 0.001074  loss: 3.4130 (3.2600)  weight_decay: 0.0500 (0.0500)  time: 0.8341  data: 0.1429  max mem: 41794
Epoch: [202]  [300/312]  eta: 0:00:09  lr: 0.001074  min_lr: 0.001074  loss: 3.4791 (3.2662)  weight_decay: 0.0500 (0.0500)  time: 0.6312  data: 0.0807  max mem: 41794
Epoch: [202]  [310/312]  eta: 0:00:01  lr: 0.001073  min_lr: 0.001073  loss: 3.5230 (3.2717)  weight_decay: 0.0500 (0.0500)  time: 0.4676  data: 0.0001  max mem: 41794
Epoch: [202]  [311/312]  eta: 0:00:00  lr: 0.001073  min_lr: 0.001073  loss: 3.5230 (3.2716)  weight_decay: 0.0500 (0.0500)  time: 0.4667  data: 0.0001  max mem: 41794
Epoch: [202] Total time: 0:03:57 (0.7605 s / it)
Averaged stats: lr: 0.001073  min_lr: 0.001073  loss: 3.5230 (3.2642)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 0.8928 (0.8928)  acc1: 81.7708 (81.7708)  acc5: 95.9635 (95.9635)  time: 8.0741  data: 7.8695  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2860 (1.2156)  acc1: 74.7396 (75.9360)  acc5: 93.4896 (93.4240)  time: 1.0484  data: 0.8751  max mem: 41794
Test: Total time: 0:00:09 (1.0922 s / it)
* Acc@1 76.464 Acc@5 93.430 loss 1.212
Accuracy of the model on the 50000 test images: 76.5%
Max accuracy: 76.46%
Epoch: [203]  [  0/312]  eta: 1:18:03  lr: 0.001073  min_lr: 0.001073  loss: 3.7548 (3.7548)  weight_decay: 0.0500 (0.0500)  time: 15.0108  data: 11.7804  max mem: 41794
Epoch: [203]  [ 10/312]  eta: 0:11:47  lr: 0.001072  min_lr: 0.001072  loss: 3.4023 (3.3981)  weight_decay: 0.0500 (0.0500)  time: 2.3441  data: 1.2988  max mem: 41794
Epoch: [203]  [ 20/312]  eta: 0:07:29  lr: 0.001072  min_lr: 0.001072  loss: 3.3935 (3.2686)  weight_decay: 0.0500 (0.0500)  time: 0.8668  data: 0.1960  max mem: 41794
Epoch: [203]  [ 30/312]  eta: 0:05:44  lr: 0.001071  min_lr: 0.001071  loss: 3.3722 (3.2680)  weight_decay: 0.0500 (0.0500)  time: 0.6016  data: 0.0757  max mem: 41794
Epoch: [203]  [ 40/312]  eta: 0:04:45  lr: 0.001070  min_lr: 0.001070  loss: 3.4832 (3.3098)  weight_decay: 0.0500 (0.0500)  time: 0.5366  data: 0.0131  max mem: 41794
Epoch: [203]  [ 50/312]  eta: 0:04:18  lr: 0.001070  min_lr: 0.001070  loss: 3.4832 (3.2920)  weight_decay: 0.0500 (0.0500)  time: 0.6242  data: 0.1097  max mem: 41794
Epoch: [203]  [ 60/312]  eta: 0:03:49  lr: 0.001069  min_lr: 0.001069  loss: 3.3695 (3.2984)  weight_decay: 0.0500 (0.0500)  time: 0.6182  data: 0.1054  max mem: 41794
Epoch: [203]  [ 70/312]  eta: 0:03:35  lr: 0.001068  min_lr: 0.001068  loss: 3.3252 (3.2785)  weight_decay: 0.0500 (0.0500)  time: 0.6431  data: 0.1386  max mem: 41794
Epoch: [203]  [ 80/312]  eta: 0:03:26  lr: 0.001068  min_lr: 0.001068  loss: 3.3634 (3.2777)  weight_decay: 0.0500 (0.0500)  time: 0.8399  data: 0.3405  max mem: 41794
Epoch: [203]  [ 90/312]  eta: 0:03:08  lr: 0.001067  min_lr: 0.001067  loss: 3.4185 (3.2898)  weight_decay: 0.0500 (0.0500)  time: 0.7038  data: 0.2058  max mem: 41794
Epoch: [203]  [100/312]  eta: 0:03:00  lr: 0.001066  min_lr: 0.001066  loss: 3.3570 (3.2823)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.1862  max mem: 41794
Epoch: [203]  [110/312]  eta: 0:02:45  lr: 0.001066  min_lr: 0.001066  loss: 3.3060 (3.2846)  weight_decay: 0.0500 (0.0500)  time: 0.6812  data: 0.1861  max mem: 41794
Epoch: [203]  [120/312]  eta: 0:02:37  lr: 0.001065  min_lr: 0.001065  loss: 3.4484 (3.2902)  weight_decay: 0.0500 (0.0500)  time: 0.6641  data: 0.1633  max mem: 41794
Epoch: [203]  [130/312]  eta: 0:02:30  lr: 0.001065  min_lr: 0.001065  loss: 3.4058 (3.2793)  weight_decay: 0.0500 (0.0500)  time: 0.8569  data: 0.3399  max mem: 41794
Epoch: [203]  [140/312]  eta: 0:02:18  lr: 0.001064  min_lr: 0.001064  loss: 3.2753 (3.2867)  weight_decay: 0.0500 (0.0500)  time: 0.7023  data: 0.1773  max mem: 41794
Epoch: [203]  [150/312]  eta: 0:02:10  lr: 0.001063  min_lr: 0.001063  loss: 3.2753 (3.2701)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.1514  max mem: 41794
Epoch: [203]  [160/312]  eta: 0:02:02  lr: 0.001063  min_lr: 0.001063  loss: 3.0869 (3.2650)  weight_decay: 0.0500 (0.0500)  time: 0.8049  data: 0.2961  max mem: 41794
Epoch: [203]  [170/312]  eta: 0:01:51  lr: 0.001062  min_lr: 0.001062  loss: 3.3786 (3.2755)  weight_decay: 0.0500 (0.0500)  time: 0.6722  data: 0.1496  max mem: 41794
Epoch: [203]  [180/312]  eta: 0:01:44  lr: 0.001061  min_lr: 0.001061  loss: 3.3946 (3.2676)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.1595  max mem: 41794
Epoch: [203]  [190/312]  eta: 0:01:34  lr: 0.001061  min_lr: 0.001061  loss: 3.2241 (3.2653)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.1625  max mem: 41794
Epoch: [203]  [200/312]  eta: 0:01:27  lr: 0.001060  min_lr: 0.001060  loss: 3.2686 (3.2655)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.1424  max mem: 41794
Epoch: [203]  [210/312]  eta: 0:01:20  lr: 0.001059  min_lr: 0.001059  loss: 3.4271 (3.2700)  weight_decay: 0.0500 (0.0500)  time: 0.8649  data: 0.2681  max mem: 41794
Epoch: [203]  [220/312]  eta: 0:01:11  lr: 0.001059  min_lr: 0.001059  loss: 3.4271 (3.2631)  weight_decay: 0.0500 (0.0500)  time: 0.6588  data: 0.1334  max mem: 41794
Epoch: [203]  [230/312]  eta: 0:01:03  lr: 0.001058  min_lr: 0.001058  loss: 3.1999 (3.2621)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.1564  max mem: 41794
Epoch: [203]  [240/312]  eta: 0:00:55  lr: 0.001058  min_lr: 0.001058  loss: 3.3343 (3.2639)  weight_decay: 0.0500 (0.0500)  time: 0.8095  data: 0.2856  max mem: 41794
Epoch: [203]  [250/312]  eta: 0:00:47  lr: 0.001057  min_lr: 0.001057  loss: 3.2360 (3.2626)  weight_decay: 0.0500 (0.0500)  time: 0.6760  data: 0.1298  max mem: 41794
Epoch: [203]  [260/312]  eta: 0:00:40  lr: 0.001056  min_lr: 0.001056  loss: 3.2126 (3.2549)  weight_decay: 0.0500 (0.0500)  time: 0.7261  data: 0.1454  max mem: 41794
Epoch: [203]  [270/312]  eta: 0:00:31  lr: 0.001056  min_lr: 0.001056  loss: 3.2462 (3.2495)  weight_decay: 0.0500 (0.0500)  time: 0.6726  data: 0.1521  max mem: 41794
Epoch: [203]  [280/312]  eta: 0:00:24  lr: 0.001055  min_lr: 0.001055  loss: 3.3530 (3.2511)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1610  max mem: 41794
Epoch: [203]  [290/312]  eta: 0:00:16  lr: 0.001054  min_lr: 0.001054  loss: 3.3037 (3.2454)  weight_decay: 0.0500 (0.0500)  time: 0.9204  data: 0.3090  max mem: 41794
Epoch: [203]  [300/312]  eta: 0:00:09  lr: 0.001054  min_lr: 0.001054  loss: 3.3037 (3.2479)  weight_decay: 0.0500 (0.0500)  time: 0.7314  data: 0.1552  max mem: 41794
Epoch: [203]  [310/312]  eta: 0:00:01  lr: 0.001053  min_lr: 0.001053  loss: 3.4466 (3.2456)  weight_decay: 0.0500 (0.0500)  time: 0.4704  data: 0.0001  max mem: 41794
Epoch: [203]  [311/312]  eta: 0:00:00  lr: 0.001053  min_lr: 0.001053  loss: 3.4466 (3.2467)  weight_decay: 0.0500 (0.0500)  time: 0.4691  data: 0.0001  max mem: 41794
Epoch: [203] Total time: 0:03:56 (0.7590 s / it)
Averaged stats: lr: 0.001053  min_lr: 0.001053  loss: 3.4466 (3.2422)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 0.9806 (0.9806)  acc1: 81.9010 (81.9010)  acc5: 96.4844 (96.4844)  time: 9.1593  data: 8.9556  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2999 (1.2129)  acc1: 75.3906 (75.9200)  acc5: 93.2292 (93.7120)  time: 1.1694  data: 0.9951  max mem: 41794
Test: Total time: 0:00:10 (1.2112 s / it)
* Acc@1 76.032 Acc@5 93.406 loss 1.219
Accuracy of the model on the 50000 test images: 76.0%
Max accuracy: 76.46%
Epoch: [204]  [  0/312]  eta: 1:23:31  lr: 0.001053  min_lr: 0.001053  loss: 2.6805 (2.6805)  weight_decay: 0.0500 (0.0500)  time: 16.0617  data: 15.2519  max mem: 41794
Epoch: [204]  [ 10/312]  eta: 0:11:08  lr: 0.001052  min_lr: 0.001052  loss: 3.1581 (3.1876)  weight_decay: 0.0500 (0.0500)  time: 2.2138  data: 1.3901  max mem: 41794
Epoch: [204]  [ 20/312]  eta: 0:07:20  lr: 0.001052  min_lr: 0.001052  loss: 3.5412 (3.3944)  weight_decay: 0.0500 (0.0500)  time: 0.7817  data: 0.0984  max mem: 41794
Epoch: [204]  [ 30/312]  eta: 0:05:34  lr: 0.001051  min_lr: 0.001051  loss: 3.5059 (3.3588)  weight_decay: 0.0500 (0.0500)  time: 0.6229  data: 0.0973  max mem: 41794
Epoch: [204]  [ 40/312]  eta: 0:04:49  lr: 0.001050  min_lr: 0.001050  loss: 3.3583 (3.3087)  weight_decay: 0.0500 (0.0500)  time: 0.6005  data: 0.0933  max mem: 41794
Epoch: [204]  [ 50/312]  eta: 0:04:25  lr: 0.001050  min_lr: 0.001050  loss: 3.2519 (3.2821)  weight_decay: 0.0500 (0.0500)  time: 0.7429  data: 0.2305  max mem: 41794
Epoch: [204]  [ 60/312]  eta: 0:03:53  lr: 0.001049  min_lr: 0.001049  loss: 3.3475 (3.2698)  weight_decay: 0.0500 (0.0500)  time: 0.6463  data: 0.1383  max mem: 41794
Epoch: [204]  [ 70/312]  eta: 0:03:45  lr: 0.001049  min_lr: 0.001049  loss: 3.4447 (3.2792)  weight_decay: 0.0500 (0.0500)  time: 0.7232  data: 0.2233  max mem: 41794
Epoch: [204]  [ 80/312]  eta: 0:03:34  lr: 0.001048  min_lr: 0.001048  loss: 3.3995 (3.2742)  weight_decay: 0.0500 (0.0500)  time: 0.9214  data: 0.4167  max mem: 41794
Epoch: [204]  [ 90/312]  eta: 0:03:15  lr: 0.001047  min_lr: 0.001047  loss: 3.4163 (3.2886)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1953  max mem: 41794
Epoch: [204]  [100/312]  eta: 0:03:06  lr: 0.001047  min_lr: 0.001047  loss: 3.3819 (3.2683)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.1974  max mem: 41794
Epoch: [204]  [110/312]  eta: 0:02:50  lr: 0.001046  min_lr: 0.001046  loss: 3.3819 (3.2847)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1961  max mem: 41794
Epoch: [204]  [120/312]  eta: 0:02:42  lr: 0.001045  min_lr: 0.001045  loss: 3.4983 (3.2792)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.1822  max mem: 41794
Epoch: [204]  [130/312]  eta: 0:02:33  lr: 0.001045  min_lr: 0.001045  loss: 3.3833 (3.2730)  weight_decay: 0.0500 (0.0500)  time: 0.8551  data: 0.3545  max mem: 41794
Epoch: [204]  [140/312]  eta: 0:02:21  lr: 0.001044  min_lr: 0.001044  loss: 3.4978 (3.2870)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.1730  max mem: 41794
Epoch: [204]  [150/312]  eta: 0:02:13  lr: 0.001044  min_lr: 0.001044  loss: 3.4978 (3.2897)  weight_decay: 0.0500 (0.0500)  time: 0.6776  data: 0.1828  max mem: 41794
Epoch: [204]  [160/312]  eta: 0:02:05  lr: 0.001043  min_lr: 0.001043  loss: 3.4378 (3.2805)  weight_decay: 0.0500 (0.0500)  time: 0.8552  data: 0.3634  max mem: 41794
Epoch: [204]  [170/312]  eta: 0:01:54  lr: 0.001042  min_lr: 0.001042  loss: 3.4945 (3.2846)  weight_decay: 0.0500 (0.0500)  time: 0.6790  data: 0.1812  max mem: 41794
Epoch: [204]  [180/312]  eta: 0:01:46  lr: 0.001042  min_lr: 0.001042  loss: 3.4596 (3.2788)  weight_decay: 0.0500 (0.0500)  time: 0.6824  data: 0.1764  max mem: 41794
Epoch: [204]  [190/312]  eta: 0:01:36  lr: 0.001041  min_lr: 0.001041  loss: 3.2418 (3.2730)  weight_decay: 0.0500 (0.0500)  time: 0.6673  data: 0.1764  max mem: 41794
Epoch: [204]  [200/312]  eta: 0:01:29  lr: 0.001040  min_lr: 0.001040  loss: 3.2778 (3.2807)  weight_decay: 0.0500 (0.0500)  time: 0.6986  data: 0.2152  max mem: 41794
Epoch: [204]  [210/312]  eta: 0:01:21  lr: 0.001040  min_lr: 0.001040  loss: 3.2724 (3.2632)  weight_decay: 0.0500 (0.0500)  time: 0.8574  data: 0.3661  max mem: 41794
Epoch: [204]  [220/312]  eta: 0:01:12  lr: 0.001039  min_lr: 0.001039  loss: 3.3839 (3.2743)  weight_decay: 0.0500 (0.0500)  time: 0.6426  data: 0.1515  max mem: 41794
Epoch: [204]  [230/312]  eta: 0:01:04  lr: 0.001038  min_lr: 0.001038  loss: 3.5240 (3.2814)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.1968  max mem: 41794
Epoch: [204]  [240/312]  eta: 0:00:57  lr: 0.001038  min_lr: 0.001038  loss: 3.3724 (3.2783)  weight_decay: 0.0500 (0.0500)  time: 0.8755  data: 0.3612  max mem: 41794
Epoch: [204]  [250/312]  eta: 0:00:48  lr: 0.001037  min_lr: 0.001037  loss: 3.1404 (3.2846)  weight_decay: 0.0500 (0.0500)  time: 0.6837  data: 0.1668  max mem: 41794
Epoch: [204]  [260/312]  eta: 0:00:40  lr: 0.001037  min_lr: 0.001037  loss: 3.3472 (3.2866)  weight_decay: 0.0500 (0.0500)  time: 0.6984  data: 0.1737  max mem: 41794
Epoch: [204]  [270/312]  eta: 0:00:32  lr: 0.001036  min_lr: 0.001036  loss: 3.4609 (3.2888)  weight_decay: 0.0500 (0.0500)  time: 0.6959  data: 0.1720  max mem: 41794
Epoch: [204]  [280/312]  eta: 0:00:24  lr: 0.001035  min_lr: 0.001035  loss: 3.4318 (3.2857)  weight_decay: 0.0500 (0.0500)  time: 0.6854  data: 0.1490  max mem: 41794
Epoch: [204]  [290/312]  eta: 0:00:17  lr: 0.001035  min_lr: 0.001035  loss: 3.1886 (3.2819)  weight_decay: 0.0500 (0.0500)  time: 0.8150  data: 0.2772  max mem: 41794
Epoch: [204]  [300/312]  eta: 0:00:09  lr: 0.001034  min_lr: 0.001034  loss: 3.1775 (3.2769)  weight_decay: 0.0500 (0.0500)  time: 0.6229  data: 0.1286  max mem: 41794
Epoch: [204]  [310/312]  eta: 0:00:01  lr: 0.001033  min_lr: 0.001033  loss: 3.2743 (3.2788)  weight_decay: 0.0500 (0.0500)  time: 0.4723  data: 0.0001  max mem: 41794
Epoch: [204]  [311/312]  eta: 0:00:00  lr: 0.001033  min_lr: 0.001033  loss: 3.3498 (3.2809)  weight_decay: 0.0500 (0.0500)  time: 0.4709  data: 0.0001  max mem: 41794
Epoch: [204] Total time: 0:03:58 (0.7629 s / it)
Averaged stats: lr: 0.001033  min_lr: 0.001033  loss: 3.3498 (3.2587)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 1.0235 (1.0235)  acc1: 81.2500 (81.2500)  acc5: 95.7031 (95.7031)  time: 8.9198  data: 8.7068  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3037 (1.2153)  acc1: 75.7812 (75.6000)  acc5: 93.0990 (93.3280)  time: 1.1392  data: 0.9675  max mem: 41794
Test: Total time: 0:00:10 (1.1526 s / it)
* Acc@1 75.650 Acc@5 93.210 loss 1.224
Accuracy of the model on the 50000 test images: 75.7%
Max accuracy: 76.46%
Epoch: [205]  [  0/312]  eta: 1:22:22  lr: 0.001033  min_lr: 0.001033  loss: 3.2928 (3.2928)  weight_decay: 0.0500 (0.0500)  time: 15.8410  data: 14.6791  max mem: 41794
Epoch: [205]  [ 10/312]  eta: 0:11:17  lr: 0.001033  min_lr: 0.001033  loss: 3.2928 (3.1027)  weight_decay: 0.0500 (0.0500)  time: 2.2418  data: 1.3814  max mem: 41794
Epoch: [205]  [ 20/312]  eta: 0:07:24  lr: 0.001032  min_lr: 0.001032  loss: 3.3877 (3.1835)  weight_decay: 0.0500 (0.0500)  time: 0.8077  data: 0.0901  max mem: 41794
Epoch: [205]  [ 30/312]  eta: 0:05:39  lr: 0.001031  min_lr: 0.001031  loss: 3.4426 (3.2331)  weight_decay: 0.0500 (0.0500)  time: 0.6329  data: 0.0668  max mem: 41794
Epoch: [205]  [ 40/312]  eta: 0:04:46  lr: 0.001031  min_lr: 0.001031  loss: 3.2309 (3.2176)  weight_decay: 0.0500 (0.0500)  time: 0.5626  data: 0.0122  max mem: 41794
Epoch: [205]  [ 50/312]  eta: 0:04:26  lr: 0.001030  min_lr: 0.001030  loss: 3.4210 (3.2590)  weight_decay: 0.0500 (0.0500)  time: 0.7235  data: 0.1345  max mem: 41794
Epoch: [205]  [ 60/312]  eta: 0:03:54  lr: 0.001030  min_lr: 0.001030  loss: 3.4820 (3.2664)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.1250  max mem: 41794
Epoch: [205]  [ 70/312]  eta: 0:03:46  lr: 0.001029  min_lr: 0.001029  loss: 3.3007 (3.2557)  weight_decay: 0.0500 (0.0500)  time: 0.7352  data: 0.1293  max mem: 41794
Epoch: [205]  [ 80/312]  eta: 0:03:34  lr: 0.001028  min_lr: 0.001028  loss: 3.1796 (3.2168)  weight_decay: 0.0500 (0.0500)  time: 0.9116  data: 0.1705  max mem: 41794
Epoch: [205]  [ 90/312]  eta: 0:03:17  lr: 0.001028  min_lr: 0.001028  loss: 3.1529 (3.2243)  weight_decay: 0.0500 (0.0500)  time: 0.7294  data: 0.1038  max mem: 41794
Epoch: [205]  [100/312]  eta: 0:03:08  lr: 0.001027  min_lr: 0.001027  loss: 3.2849 (3.2238)  weight_decay: 0.0500 (0.0500)  time: 0.7394  data: 0.1133  max mem: 41794
Epoch: [205]  [110/312]  eta: 0:02:52  lr: 0.001026  min_lr: 0.001026  loss: 3.3875 (3.2326)  weight_decay: 0.0500 (0.0500)  time: 0.6912  data: 0.0526  max mem: 41794
Epoch: [205]  [120/312]  eta: 0:02:43  lr: 0.001026  min_lr: 0.001026  loss: 3.2479 (3.2291)  weight_decay: 0.0500 (0.0500)  time: 0.6678  data: 0.0595  max mem: 41794
Epoch: [205]  [130/312]  eta: 0:02:35  lr: 0.001025  min_lr: 0.001025  loss: 3.2479 (3.2267)  weight_decay: 0.0500 (0.0500)  time: 0.8477  data: 0.1228  max mem: 41794
Epoch: [205]  [140/312]  eta: 0:02:22  lr: 0.001025  min_lr: 0.001025  loss: 3.3382 (3.2344)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.0653  max mem: 41794
Epoch: [205]  [150/312]  eta: 0:02:14  lr: 0.001024  min_lr: 0.001024  loss: 3.4247 (3.2353)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.0527  max mem: 41794
Epoch: [205]  [160/312]  eta: 0:02:06  lr: 0.001023  min_lr: 0.001023  loss: 3.2721 (3.2178)  weight_decay: 0.0500 (0.0500)  time: 0.8763  data: 0.0851  max mem: 41794
Epoch: [205]  [170/312]  eta: 0:01:56  lr: 0.001023  min_lr: 0.001023  loss: 2.9432 (3.2120)  weight_decay: 0.0500 (0.0500)  time: 0.7305  data: 0.0772  max mem: 41794
Epoch: [205]  [180/312]  eta: 0:01:48  lr: 0.001022  min_lr: 0.001022  loss: 2.9892 (3.2078)  weight_decay: 0.0500 (0.0500)  time: 0.7455  data: 0.1032  max mem: 41794
Epoch: [205]  [190/312]  eta: 0:01:38  lr: 0.001021  min_lr: 0.001021  loss: 3.3928 (3.2168)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.0597  max mem: 41794
Epoch: [205]  [200/312]  eta: 0:01:30  lr: 0.001021  min_lr: 0.001021  loss: 3.1223 (3.2051)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.0467  max mem: 41794
Epoch: [205]  [210/312]  eta: 0:01:22  lr: 0.001020  min_lr: 0.001020  loss: 3.1207 (3.2077)  weight_decay: 0.0500 (0.0500)  time: 0.8184  data: 0.0918  max mem: 41794
Epoch: [205]  [220/312]  eta: 0:01:13  lr: 0.001019  min_lr: 0.001019  loss: 3.4506 (3.2210)  weight_decay: 0.0500 (0.0500)  time: 0.6495  data: 0.0583  max mem: 41794
Epoch: [205]  [230/312]  eta: 0:01:05  lr: 0.001019  min_lr: 0.001019  loss: 3.4506 (3.2116)  weight_decay: 0.0500 (0.0500)  time: 0.7344  data: 0.0993  max mem: 41794
Epoch: [205]  [240/312]  eta: 0:00:57  lr: 0.001018  min_lr: 0.001018  loss: 3.2046 (3.2086)  weight_decay: 0.0500 (0.0500)  time: 0.8790  data: 0.1332  max mem: 41794
Epoch: [205]  [250/312]  eta: 0:00:49  lr: 0.001018  min_lr: 0.001018  loss: 3.3531 (3.2089)  weight_decay: 0.0500 (0.0500)  time: 0.6492  data: 0.0467  max mem: 41794
Epoch: [205]  [260/312]  eta: 0:00:41  lr: 0.001017  min_lr: 0.001017  loss: 3.3531 (3.2134)  weight_decay: 0.0500 (0.0500)  time: 0.6804  data: 0.0857  max mem: 41794
Epoch: [205]  [270/312]  eta: 0:00:32  lr: 0.001016  min_lr: 0.001016  loss: 3.3674 (3.2154)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0855  max mem: 41794
Epoch: [205]  [280/312]  eta: 0:00:25  lr: 0.001016  min_lr: 0.001016  loss: 3.3454 (3.2150)  weight_decay: 0.0500 (0.0500)  time: 0.7034  data: 0.1004  max mem: 41794
Epoch: [205]  [290/312]  eta: 0:00:17  lr: 0.001015  min_lr: 0.001015  loss: 3.3743 (3.2196)  weight_decay: 0.0500 (0.0500)  time: 0.8194  data: 0.1646  max mem: 41794
Epoch: [205]  [300/312]  eta: 0:00:09  lr: 0.001014  min_lr: 0.001014  loss: 3.4472 (3.2171)  weight_decay: 0.0500 (0.0500)  time: 0.6008  data: 0.0646  max mem: 41794
Epoch: [205]  [310/312]  eta: 0:00:01  lr: 0.001014  min_lr: 0.001014  loss: 3.1765 (3.2128)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [205]  [311/312]  eta: 0:00:00  lr: 0.001014  min_lr: 0.001014  loss: 3.1306 (3.2125)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [205] Total time: 0:04:00 (0.7701 s / it)
Averaged stats: lr: 0.001014  min_lr: 0.001014  loss: 3.1306 (3.2382)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.9530 (0.9530)  acc1: 83.0729 (83.0729)  acc5: 96.0938 (96.0938)  time: 8.5756  data: 8.3632  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2379 (1.1707)  acc1: 76.1719 (76.4480)  acc5: 93.8802 (93.6800)  time: 1.1093  data: 0.9330  max mem: 41794
Test: Total time: 0:00:10 (1.1195 s / it)
* Acc@1 76.492 Acc@5 93.584 loss 1.176
Accuracy of the model on the 50000 test images: 76.5%
Max accuracy: 76.49%
Epoch: [206]  [  0/312]  eta: 1:20:45  lr: 0.001014  min_lr: 0.001014  loss: 3.3063 (3.3063)  weight_decay: 0.0500 (0.0500)  time: 15.5294  data: 15.0307  max mem: 41794
Epoch: [206]  [ 10/312]  eta: 0:11:04  lr: 0.001013  min_lr: 0.001013  loss: 3.1436 (3.1182)  weight_decay: 0.0500 (0.0500)  time: 2.2008  data: 1.3671  max mem: 41794
Epoch: [206]  [ 20/312]  eta: 0:07:29  lr: 0.001012  min_lr: 0.001012  loss: 3.1436 (3.2189)  weight_decay: 0.0500 (0.0500)  time: 0.8406  data: 0.0840  max mem: 41794
Epoch: [206]  [ 30/312]  eta: 0:05:42  lr: 0.001012  min_lr: 0.001012  loss: 3.4720 (3.2913)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.0839  max mem: 41794
Epoch: [206]  [ 40/312]  eta: 0:04:45  lr: 0.001011  min_lr: 0.001011  loss: 3.5300 (3.3146)  weight_decay: 0.0500 (0.0500)  time: 0.5344  data: 0.0073  max mem: 41794
Epoch: [206]  [ 50/312]  eta: 0:04:25  lr: 0.001011  min_lr: 0.001011  loss: 3.2605 (3.2584)  weight_decay: 0.0500 (0.0500)  time: 0.7064  data: 0.1102  max mem: 41794
Epoch: [206]  [ 60/312]  eta: 0:03:53  lr: 0.001010  min_lr: 0.001010  loss: 3.2605 (3.2862)  weight_decay: 0.0500 (0.0500)  time: 0.6759  data: 0.1036  max mem: 41794
Epoch: [206]  [ 70/312]  eta: 0:03:42  lr: 0.001009  min_lr: 0.001009  loss: 3.3510 (3.2840)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.1112  max mem: 41794
Epoch: [206]  [ 80/312]  eta: 0:03:28  lr: 0.001009  min_lr: 0.001009  loss: 3.2022 (3.2732)  weight_decay: 0.0500 (0.0500)  time: 0.8153  data: 0.1379  max mem: 41794
Epoch: [206]  [ 90/312]  eta: 0:03:13  lr: 0.001008  min_lr: 0.001008  loss: 3.3165 (3.2574)  weight_decay: 0.0500 (0.0500)  time: 0.7013  data: 0.1064  max mem: 41794
Epoch: [206]  [100/312]  eta: 0:03:06  lr: 0.001007  min_lr: 0.001007  loss: 3.3165 (3.2601)  weight_decay: 0.0500 (0.0500)  time: 0.7920  data: 0.2291  max mem: 41794
Epoch: [206]  [110/312]  eta: 0:02:50  lr: 0.001007  min_lr: 0.001007  loss: 3.3463 (3.2652)  weight_decay: 0.0500 (0.0500)  time: 0.7207  data: 0.1501  max mem: 41794
Epoch: [206]  [120/312]  eta: 0:02:42  lr: 0.001006  min_lr: 0.001006  loss: 3.2696 (3.2556)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1259  max mem: 41794
Epoch: [206]  [130/312]  eta: 0:02:35  lr: 0.001006  min_lr: 0.001006  loss: 3.1386 (3.2476)  weight_decay: 0.0500 (0.0500)  time: 0.8905  data: 0.2733  max mem: 41794
Epoch: [206]  [140/312]  eta: 0:02:22  lr: 0.001005  min_lr: 0.001005  loss: 3.3313 (3.2588)  weight_decay: 0.0500 (0.0500)  time: 0.7053  data: 0.1500  max mem: 41794
Epoch: [206]  [150/312]  eta: 0:02:14  lr: 0.001004  min_lr: 0.001004  loss: 3.5178 (3.2675)  weight_decay: 0.0500 (0.0500)  time: 0.6691  data: 0.1110  max mem: 41794
Epoch: [206]  [160/312]  eta: 0:02:04  lr: 0.001004  min_lr: 0.001004  loss: 3.3936 (3.2487)  weight_decay: 0.0500 (0.0500)  time: 0.7458  data: 0.1523  max mem: 41794
Epoch: [206]  [170/312]  eta: 0:01:54  lr: 0.001003  min_lr: 0.001003  loss: 3.1922 (3.2426)  weight_decay: 0.0500 (0.0500)  time: 0.6751  data: 0.1502  max mem: 41794
Epoch: [206]  [180/312]  eta: 0:01:47  lr: 0.001002  min_lr: 0.001002  loss: 3.2614 (3.2486)  weight_decay: 0.0500 (0.0500)  time: 0.7897  data: 0.2574  max mem: 41794
Epoch: [206]  [190/312]  eta: 0:01:37  lr: 0.001002  min_lr: 0.001002  loss: 3.4064 (3.2487)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.1512  max mem: 41794
Epoch: [206]  [200/312]  eta: 0:01:29  lr: 0.001001  min_lr: 0.001001  loss: 3.3313 (3.2517)  weight_decay: 0.0500 (0.0500)  time: 0.6634  data: 0.1167  max mem: 41794
Epoch: [206]  [210/312]  eta: 0:01:21  lr: 0.001001  min_lr: 0.001001  loss: 3.4603 (3.2579)  weight_decay: 0.0500 (0.0500)  time: 0.8599  data: 0.2478  max mem: 41794
Epoch: [206]  [220/312]  eta: 0:01:12  lr: 0.001000  min_lr: 0.001000  loss: 3.4722 (3.2561)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.1317  max mem: 41794
Epoch: [206]  [230/312]  eta: 0:01:05  lr: 0.000999  min_lr: 0.000999  loss: 3.2517 (3.2506)  weight_decay: 0.0500 (0.0500)  time: 0.6926  data: 0.1256  max mem: 41794
Epoch: [206]  [240/312]  eta: 0:00:57  lr: 0.000999  min_lr: 0.000999  loss: 3.1368 (3.2403)  weight_decay: 0.0500 (0.0500)  time: 0.8463  data: 0.2224  max mem: 41794
Epoch: [206]  [250/312]  eta: 0:00:48  lr: 0.000998  min_lr: 0.000998  loss: 3.2189 (3.2446)  weight_decay: 0.0500 (0.0500)  time: 0.7254  data: 0.1717  max mem: 41794
Epoch: [206]  [260/312]  eta: 0:00:41  lr: 0.000998  min_lr: 0.000998  loss: 3.2125 (3.2304)  weight_decay: 0.0500 (0.0500)  time: 0.7883  data: 0.2215  max mem: 41794
Epoch: [206]  [270/312]  eta: 0:00:32  lr: 0.000997  min_lr: 0.000997  loss: 3.2485 (3.2337)  weight_decay: 0.0500 (0.0500)  time: 0.7153  data: 0.1473  max mem: 41794
Epoch: [206]  [280/312]  eta: 0:00:25  lr: 0.000996  min_lr: 0.000996  loss: 3.2485 (3.2287)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.0783  max mem: 41794
Epoch: [206]  [290/312]  eta: 0:00:17  lr: 0.000996  min_lr: 0.000996  loss: 3.1740 (3.2310)  weight_decay: 0.0500 (0.0500)  time: 0.8120  data: 0.1369  max mem: 41794
Epoch: [206]  [300/312]  eta: 0:00:09  lr: 0.000995  min_lr: 0.000995  loss: 3.4305 (3.2307)  weight_decay: 0.0500 (0.0500)  time: 0.6112  data: 0.0590  max mem: 41794
Epoch: [206]  [310/312]  eta: 0:00:01  lr: 0.000994  min_lr: 0.000994  loss: 3.3731 (3.2373)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [206]  [311/312]  eta: 0:00:00  lr: 0.000994  min_lr: 0.000994  loss: 3.3731 (3.2367)  weight_decay: 0.0500 (0.0500)  time: 0.4631  data: 0.0001  max mem: 41794
Epoch: [206] Total time: 0:04:00 (0.7696 s / it)
Averaged stats: lr: 0.000994  min_lr: 0.000994  loss: 3.3731 (3.2174)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 1.0599 (1.0599)  acc1: 81.5104 (81.5104)  acc5: 96.8750 (96.8750)  time: 8.3020  data: 8.0892  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2972 (1.2027)  acc1: 76.0417 (76.6880)  acc5: 93.4896 (93.6640)  time: 1.0817  data: 0.9023  max mem: 41794
Test: Total time: 0:00:09 (1.0976 s / it)
* Acc@1 76.624 Acc@5 93.680 loss 1.210
Accuracy of the model on the 50000 test images: 76.6%
Max accuracy: 76.62%
Epoch: [207]  [  0/312]  eta: 1:23:19  lr: 0.000994  min_lr: 0.000994  loss: 2.9115 (2.9115)  weight_decay: 0.0500 (0.0500)  time: 16.0230  data: 12.8960  max mem: 41794
Epoch: [207]  [ 10/312]  eta: 0:11:23  lr: 0.000994  min_lr: 0.000994  loss: 3.0640 (3.0650)  weight_decay: 0.0500 (0.0500)  time: 2.2627  data: 1.3790  max mem: 41794
Epoch: [207]  [ 20/312]  eta: 0:07:33  lr: 0.000993  min_lr: 0.000993  loss: 3.2605 (3.2669)  weight_decay: 0.0500 (0.0500)  time: 0.8287  data: 0.1797  max mem: 41794
Epoch: [207]  [ 30/312]  eta: 0:05:45  lr: 0.000992  min_lr: 0.000992  loss: 3.4981 (3.3678)  weight_decay: 0.0500 (0.0500)  time: 0.6563  data: 0.0720  max mem: 41794
Epoch: [207]  [ 40/312]  eta: 0:04:45  lr: 0.000992  min_lr: 0.000992  loss: 3.4426 (3.3048)  weight_decay: 0.0500 (0.0500)  time: 0.5256  data: 0.0088  max mem: 41794
Epoch: [207]  [ 50/312]  eta: 0:04:12  lr: 0.000991  min_lr: 0.000991  loss: 3.3150 (3.2892)  weight_decay: 0.0500 (0.0500)  time: 0.5556  data: 0.0031  max mem: 41794
Epoch: [207]  [ 60/312]  eta: 0:03:45  lr: 0.000991  min_lr: 0.000991  loss: 3.3063 (3.2677)  weight_decay: 0.0500 (0.0500)  time: 0.5697  data: 0.0176  max mem: 41794
Epoch: [207]  [ 70/312]  eta: 0:03:38  lr: 0.000990  min_lr: 0.000990  loss: 3.2488 (3.2516)  weight_decay: 0.0500 (0.0500)  time: 0.7437  data: 0.1548  max mem: 41794
Epoch: [207]  [ 80/312]  eta: 0:03:23  lr: 0.000989  min_lr: 0.000989  loss: 3.2020 (3.2010)  weight_decay: 0.0500 (0.0500)  time: 0.8349  data: 0.1983  max mem: 41794
Epoch: [207]  [ 90/312]  eta: 0:03:09  lr: 0.000989  min_lr: 0.000989  loss: 3.0344 (3.2061)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1372  max mem: 41794
Epoch: [207]  [100/312]  eta: 0:03:00  lr: 0.000988  min_lr: 0.000988  loss: 3.3665 (3.2158)  weight_decay: 0.0500 (0.0500)  time: 0.7360  data: 0.2256  max mem: 41794
Epoch: [207]  [110/312]  eta: 0:02:45  lr: 0.000987  min_lr: 0.000987  loss: 3.3297 (3.2074)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.1496  max mem: 41794
Epoch: [207]  [120/312]  eta: 0:02:38  lr: 0.000987  min_lr: 0.000987  loss: 3.3036 (3.2149)  weight_decay: 0.0500 (0.0500)  time: 0.7036  data: 0.1839  max mem: 41794
Epoch: [207]  [130/312]  eta: 0:02:32  lr: 0.000986  min_lr: 0.000986  loss: 3.3429 (3.2172)  weight_decay: 0.0500 (0.0500)  time: 0.9351  data: 0.4203  max mem: 41794
Epoch: [207]  [140/312]  eta: 0:02:19  lr: 0.000986  min_lr: 0.000986  loss: 3.3429 (3.2107)  weight_decay: 0.0500 (0.0500)  time: 0.7335  data: 0.2374  max mem: 41794
Epoch: [207]  [150/312]  eta: 0:02:13  lr: 0.000985  min_lr: 0.000985  loss: 3.1420 (3.2061)  weight_decay: 0.0500 (0.0500)  time: 0.7247  data: 0.1884  max mem: 41794
Epoch: [207]  [160/312]  eta: 0:02:02  lr: 0.000984  min_lr: 0.000984  loss: 3.3776 (3.2298)  weight_decay: 0.0500 (0.0500)  time: 0.7730  data: 0.2041  max mem: 41794
Epoch: [207]  [170/312]  eta: 0:01:53  lr: 0.000984  min_lr: 0.000984  loss: 3.5085 (3.2216)  weight_decay: 0.0500 (0.0500)  time: 0.6213  data: 0.0966  max mem: 41794
Epoch: [207]  [180/312]  eta: 0:01:45  lr: 0.000983  min_lr: 0.000983  loss: 3.1075 (3.2112)  weight_decay: 0.0500 (0.0500)  time: 0.7485  data: 0.2394  max mem: 41794
Epoch: [207]  [190/312]  eta: 0:01:35  lr: 0.000982  min_lr: 0.000982  loss: 3.0954 (3.2074)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.1596  max mem: 41794
Epoch: [207]  [200/312]  eta: 0:01:28  lr: 0.000982  min_lr: 0.000982  loss: 3.2593 (3.2097)  weight_decay: 0.0500 (0.0500)  time: 0.7090  data: 0.1615  max mem: 41794
Epoch: [207]  [210/312]  eta: 0:01:20  lr: 0.000981  min_lr: 0.000981  loss: 3.3707 (3.2096)  weight_decay: 0.0500 (0.0500)  time: 0.8662  data: 0.2990  max mem: 41794
Epoch: [207]  [220/312]  eta: 0:01:11  lr: 0.000981  min_lr: 0.000981  loss: 3.0251 (3.2003)  weight_decay: 0.0500 (0.0500)  time: 0.6598  data: 0.1381  max mem: 41794
Epoch: [207]  [230/312]  eta: 0:01:04  lr: 0.000980  min_lr: 0.000980  loss: 3.2714 (3.1977)  weight_decay: 0.0500 (0.0500)  time: 0.6781  data: 0.1565  max mem: 41794
Epoch: [207]  [240/312]  eta: 0:00:56  lr: 0.000979  min_lr: 0.000979  loss: 3.3061 (3.1939)  weight_decay: 0.0500 (0.0500)  time: 0.8051  data: 0.2627  max mem: 41794
Epoch: [207]  [250/312]  eta: 0:00:48  lr: 0.000979  min_lr: 0.000979  loss: 3.2373 (3.1917)  weight_decay: 0.0500 (0.0500)  time: 0.7026  data: 0.1797  max mem: 41794
Epoch: [207]  [260/312]  eta: 0:00:40  lr: 0.000978  min_lr: 0.000978  loss: 3.2093 (3.1935)  weight_decay: 0.0500 (0.0500)  time: 0.7645  data: 0.2344  max mem: 41794
Epoch: [207]  [270/312]  eta: 0:00:32  lr: 0.000978  min_lr: 0.000978  loss: 3.1385 (3.1933)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.1633  max mem: 41794
Epoch: [207]  [280/312]  eta: 0:00:24  lr: 0.000977  min_lr: 0.000977  loss: 3.3658 (3.2014)  weight_decay: 0.0500 (0.0500)  time: 0.6810  data: 0.1412  max mem: 41794
Epoch: [207]  [290/312]  eta: 0:00:17  lr: 0.000976  min_lr: 0.000976  loss: 3.4720 (3.2079)  weight_decay: 0.0500 (0.0500)  time: 0.9330  data: 0.2701  max mem: 41794
Epoch: [207]  [300/312]  eta: 0:00:09  lr: 0.000976  min_lr: 0.000976  loss: 3.4231 (3.2071)  weight_decay: 0.0500 (0.0500)  time: 0.7379  data: 0.1310  max mem: 41794
Epoch: [207]  [310/312]  eta: 0:00:01  lr: 0.000975  min_lr: 0.000975  loss: 3.4231 (3.2123)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [207]  [311/312]  eta: 0:00:00  lr: 0.000975  min_lr: 0.000975  loss: 3.4763 (3.2144)  weight_decay: 0.0500 (0.0500)  time: 0.4650  data: 0.0001  max mem: 41794
Epoch: [207] Total time: 0:03:59 (0.7667 s / it)
Averaged stats: lr: 0.000975  min_lr: 0.000975  loss: 3.4763 (3.2269)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 0.9615 (0.9615)  acc1: 82.4219 (82.4219)  acc5: 95.5729 (95.5729)  time: 9.1274  data: 8.9155  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3151 (1.2108)  acc1: 75.1302 (75.2800)  acc5: 92.9688 (93.2480)  time: 1.1626  data: 0.9907  max mem: 41794
Test: Total time: 0:00:10 (1.1785 s / it)
* Acc@1 76.214 Acc@5 93.440 loss 1.200
Accuracy of the model on the 50000 test images: 76.2%
Max accuracy: 76.62%
Epoch: [208]  [  0/312]  eta: 1:17:05  lr: 0.000975  min_lr: 0.000975  loss: 3.1728 (3.1728)  weight_decay: 0.0500 (0.0500)  time: 14.8263  data: 13.2659  max mem: 41794
Epoch: [208]  [ 10/312]  eta: 0:10:12  lr: 0.000974  min_lr: 0.000974  loss: 3.4475 (3.3716)  weight_decay: 0.0500 (0.0500)  time: 2.0286  data: 1.3106  max mem: 41794
Epoch: [208]  [ 20/312]  eta: 0:07:15  lr: 0.000974  min_lr: 0.000974  loss: 3.4475 (3.2663)  weight_decay: 0.0500 (0.0500)  time: 0.8244  data: 0.1905  max mem: 41794
Epoch: [208]  [ 30/312]  eta: 0:05:31  lr: 0.000973  min_lr: 0.000973  loss: 3.2582 (3.2584)  weight_decay: 0.0500 (0.0500)  time: 0.7060  data: 0.1406  max mem: 41794
Epoch: [208]  [ 40/312]  eta: 0:04:51  lr: 0.000972  min_lr: 0.000972  loss: 3.4722 (3.3017)  weight_decay: 0.0500 (0.0500)  time: 0.6280  data: 0.0689  max mem: 41794
Epoch: [208]  [ 50/312]  eta: 0:04:29  lr: 0.000972  min_lr: 0.000972  loss: 3.3958 (3.2512)  weight_decay: 0.0500 (0.0500)  time: 0.8041  data: 0.2374  max mem: 41794
Epoch: [208]  [ 60/312]  eta: 0:03:57  lr: 0.000971  min_lr: 0.000971  loss: 3.3235 (3.2646)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1766  max mem: 41794
Epoch: [208]  [ 70/312]  eta: 0:03:46  lr: 0.000971  min_lr: 0.000971  loss: 3.3235 (3.2367)  weight_decay: 0.0500 (0.0500)  time: 0.6979  data: 0.1617  max mem: 41794
Epoch: [208]  [ 80/312]  eta: 0:03:31  lr: 0.000970  min_lr: 0.000970  loss: 3.0742 (3.2430)  weight_decay: 0.0500 (0.0500)  time: 0.8155  data: 0.2816  max mem: 41794
Epoch: [208]  [ 90/312]  eta: 0:03:18  lr: 0.000969  min_lr: 0.000969  loss: 3.4024 (3.2418)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.1207  max mem: 41794
Epoch: [208]  [100/312]  eta: 0:03:10  lr: 0.000969  min_lr: 0.000969  loss: 3.4024 (3.2474)  weight_decay: 0.0500 (0.0500)  time: 0.8389  data: 0.0820  max mem: 41794
Epoch: [208]  [110/312]  eta: 0:02:54  lr: 0.000968  min_lr: 0.000968  loss: 3.2969 (3.2433)  weight_decay: 0.0500 (0.0500)  time: 0.7084  data: 0.0820  max mem: 41794
Epoch: [208]  [120/312]  eta: 0:02:45  lr: 0.000968  min_lr: 0.000968  loss: 2.8987 (3.2197)  weight_decay: 0.0500 (0.0500)  time: 0.7022  data: 0.0535  max mem: 41794
Epoch: [208]  [130/312]  eta: 0:02:36  lr: 0.000967  min_lr: 0.000967  loss: 3.2799 (3.2344)  weight_decay: 0.0500 (0.0500)  time: 0.8478  data: 0.1135  max mem: 41794
Epoch: [208]  [140/312]  eta: 0:02:23  lr: 0.000966  min_lr: 0.000966  loss: 3.4694 (3.2177)  weight_decay: 0.0500 (0.0500)  time: 0.6449  data: 0.0605  max mem: 41794
Epoch: [208]  [150/312]  eta: 0:02:15  lr: 0.000966  min_lr: 0.000966  loss: 3.2473 (3.2204)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.1072  max mem: 41794
Epoch: [208]  [160/312]  eta: 0:02:06  lr: 0.000965  min_lr: 0.000965  loss: 3.2473 (3.2207)  weight_decay: 0.0500 (0.0500)  time: 0.8306  data: 0.2283  max mem: 41794
Epoch: [208]  [170/312]  eta: 0:01:55  lr: 0.000964  min_lr: 0.000964  loss: 3.2488 (3.2294)  weight_decay: 0.0500 (0.0500)  time: 0.6636  data: 0.1232  max mem: 41794
Epoch: [208]  [180/312]  eta: 0:01:47  lr: 0.000964  min_lr: 0.000964  loss: 3.0969 (3.2171)  weight_decay: 0.0500 (0.0500)  time: 0.6892  data: 0.1394  max mem: 41794
Epoch: [208]  [190/312]  eta: 0:01:37  lr: 0.000963  min_lr: 0.000963  loss: 3.3744 (3.2261)  weight_decay: 0.0500 (0.0500)  time: 0.6661  data: 0.1380  max mem: 41794
Epoch: [208]  [200/312]  eta: 0:01:29  lr: 0.000963  min_lr: 0.000963  loss: 3.3744 (3.2238)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.1410  max mem: 41794
Epoch: [208]  [210/312]  eta: 0:01:22  lr: 0.000962  min_lr: 0.000962  loss: 3.4340 (3.2312)  weight_decay: 0.0500 (0.0500)  time: 0.8736  data: 0.3057  max mem: 41794
Epoch: [208]  [220/312]  eta: 0:01:12  lr: 0.000961  min_lr: 0.000961  loss: 3.5640 (3.2376)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.1653  max mem: 41794
Epoch: [208]  [230/312]  eta: 0:01:05  lr: 0.000961  min_lr: 0.000961  loss: 3.3630 (3.2275)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.1686  max mem: 41794
Epoch: [208]  [240/312]  eta: 0:00:57  lr: 0.000960  min_lr: 0.000960  loss: 2.8052 (3.2209)  weight_decay: 0.0500 (0.0500)  time: 0.8499  data: 0.3101  max mem: 41794
Epoch: [208]  [250/312]  eta: 0:00:48  lr: 0.000960  min_lr: 0.000960  loss: 3.1180 (3.2189)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.1422  max mem: 41794
Epoch: [208]  [260/312]  eta: 0:00:41  lr: 0.000959  min_lr: 0.000959  loss: 3.3048 (3.2205)  weight_decay: 0.0500 (0.0500)  time: 0.7233  data: 0.1116  max mem: 41794
Epoch: [208]  [270/312]  eta: 0:00:32  lr: 0.000958  min_lr: 0.000958  loss: 3.3334 (3.2233)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.1116  max mem: 41794
Epoch: [208]  [280/312]  eta: 0:00:25  lr: 0.000958  min_lr: 0.000958  loss: 3.4955 (3.2312)  weight_decay: 0.0500 (0.0500)  time: 0.6710  data: 0.1166  max mem: 41794
Epoch: [208]  [290/312]  eta: 0:00:17  lr: 0.000957  min_lr: 0.000957  loss: 3.4977 (3.2410)  weight_decay: 0.0500 (0.0500)  time: 0.7998  data: 0.1960  max mem: 41794
Epoch: [208]  [300/312]  eta: 0:00:09  lr: 0.000956  min_lr: 0.000956  loss: 3.5001 (3.2441)  weight_decay: 0.0500 (0.0500)  time: 0.6133  data: 0.0797  max mem: 41794
Epoch: [208]  [310/312]  eta: 0:00:01  lr: 0.000956  min_lr: 0.000956  loss: 3.5029 (3.2543)  weight_decay: 0.0500 (0.0500)  time: 0.4671  data: 0.0001  max mem: 41794
Epoch: [208]  [311/312]  eta: 0:00:00  lr: 0.000956  min_lr: 0.000956  loss: 3.5029 (3.2545)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [208] Total time: 0:03:59 (0.7672 s / it)
Averaged stats: lr: 0.000956  min_lr: 0.000956  loss: 3.5029 (3.2255)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.2208 (1.2208)  acc1: 80.4688 (80.4688)  acc5: 95.5729 (95.5729)  time: 8.5149  data: 8.3024  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3871 (1.3401)  acc1: 76.1719 (75.3440)  acc5: 92.8385 (93.2160)  time: 1.0984  data: 0.9226  max mem: 41794
Test: Total time: 0:00:10 (1.1151 s / it)
* Acc@1 75.968 Acc@5 93.436 loss 1.327
Accuracy of the model on the 50000 test images: 76.0%
Max accuracy: 76.62%
Epoch: [209]  [  0/312]  eta: 1:25:07  lr: 0.000956  min_lr: 0.000956  loss: 3.8405 (3.8405)  weight_decay: 0.0500 (0.0500)  time: 16.3704  data: 15.8965  max mem: 41794
Epoch: [209]  [ 10/312]  eta: 0:11:50  lr: 0.000955  min_lr: 0.000955  loss: 3.1375 (3.0988)  weight_decay: 0.0500 (0.0500)  time: 2.3516  data: 1.4459  max mem: 41794
Epoch: [209]  [ 20/312]  eta: 0:07:53  lr: 0.000955  min_lr: 0.000955  loss: 3.3910 (3.2527)  weight_decay: 0.0500 (0.0500)  time: 0.8832  data: 0.0169  max mem: 41794
Epoch: [209]  [ 30/312]  eta: 0:05:59  lr: 0.000954  min_lr: 0.000954  loss: 3.3910 (3.1972)  weight_decay: 0.0500 (0.0500)  time: 0.6819  data: 0.0273  max mem: 41794
Epoch: [209]  [ 40/312]  eta: 0:04:55  lr: 0.000953  min_lr: 0.000953  loss: 3.1713 (3.1892)  weight_decay: 0.0500 (0.0500)  time: 0.5265  data: 0.0130  max mem: 41794
Epoch: [209]  [ 50/312]  eta: 0:04:19  lr: 0.000953  min_lr: 0.000953  loss: 3.2326 (3.2108)  weight_decay: 0.0500 (0.0500)  time: 0.5536  data: 0.0117  max mem: 41794
Epoch: [209]  [ 60/312]  eta: 0:03:49  lr: 0.000952  min_lr: 0.000952  loss: 3.2814 (3.2432)  weight_decay: 0.0500 (0.0500)  time: 0.5531  data: 0.0111  max mem: 41794
Epoch: [209]  [ 70/312]  eta: 0:03:36  lr: 0.000951  min_lr: 0.000951  loss: 3.3798 (3.2564)  weight_decay: 0.0500 (0.0500)  time: 0.6526  data: 0.0731  max mem: 41794
Epoch: [209]  [ 80/312]  eta: 0:03:23  lr: 0.000951  min_lr: 0.000951  loss: 3.3798 (3.2446)  weight_decay: 0.0500 (0.0500)  time: 0.7746  data: 0.1966  max mem: 41794
Epoch: [209]  [ 90/312]  eta: 0:03:09  lr: 0.000950  min_lr: 0.000950  loss: 3.3755 (3.2444)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1254  max mem: 41794
Epoch: [209]  [100/312]  eta: 0:03:02  lr: 0.000950  min_lr: 0.000950  loss: 3.3342 (3.2197)  weight_decay: 0.0500 (0.0500)  time: 0.7835  data: 0.1379  max mem: 41794
Epoch: [209]  [110/312]  eta: 0:02:47  lr: 0.000949  min_lr: 0.000949  loss: 3.2391 (3.2280)  weight_decay: 0.0500 (0.0500)  time: 0.7125  data: 0.1379  max mem: 41794
Epoch: [209]  [120/312]  eta: 0:02:38  lr: 0.000948  min_lr: 0.000948  loss: 3.3476 (3.2144)  weight_decay: 0.0500 (0.0500)  time: 0.6517  data: 0.1235  max mem: 41794
Epoch: [209]  [130/312]  eta: 0:02:29  lr: 0.000948  min_lr: 0.000948  loss: 3.1795 (3.1981)  weight_decay: 0.0500 (0.0500)  time: 0.7984  data: 0.2713  max mem: 41794
Epoch: [209]  [140/312]  eta: 0:02:18  lr: 0.000947  min_lr: 0.000947  loss: 3.4881 (3.2113)  weight_decay: 0.0500 (0.0500)  time: 0.6803  data: 0.1500  max mem: 41794
Epoch: [209]  [150/312]  eta: 0:02:11  lr: 0.000947  min_lr: 0.000947  loss: 3.4881 (3.2080)  weight_decay: 0.0500 (0.0500)  time: 0.7327  data: 0.1902  max mem: 41794
Epoch: [209]  [160/312]  eta: 0:02:03  lr: 0.000946  min_lr: 0.000946  loss: 3.0750 (3.1863)  weight_decay: 0.0500 (0.0500)  time: 0.8493  data: 0.3480  max mem: 41794
Epoch: [209]  [170/312]  eta: 0:01:52  lr: 0.000945  min_lr: 0.000945  loss: 3.3613 (3.2002)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.1625  max mem: 41794
Epoch: [209]  [180/312]  eta: 0:01:45  lr: 0.000945  min_lr: 0.000945  loss: 3.4924 (3.1983)  weight_decay: 0.0500 (0.0500)  time: 0.7055  data: 0.1629  max mem: 41794
Epoch: [209]  [190/312]  eta: 0:01:35  lr: 0.000944  min_lr: 0.000944  loss: 3.2291 (3.2046)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.1603  max mem: 41794
Epoch: [209]  [200/312]  eta: 0:01:28  lr: 0.000944  min_lr: 0.000944  loss: 3.2942 (3.2119)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.1607  max mem: 41794
Epoch: [209]  [210/312]  eta: 0:01:20  lr: 0.000943  min_lr: 0.000943  loss: 3.5226 (3.2235)  weight_decay: 0.0500 (0.0500)  time: 0.8723  data: 0.3078  max mem: 41794
Epoch: [209]  [220/312]  eta: 0:01:11  lr: 0.000942  min_lr: 0.000942  loss: 3.5226 (3.2289)  weight_decay: 0.0500 (0.0500)  time: 0.6811  data: 0.1524  max mem: 41794
Epoch: [209]  [230/312]  eta: 0:01:04  lr: 0.000942  min_lr: 0.000942  loss: 3.5154 (3.2387)  weight_decay: 0.0500 (0.0500)  time: 0.6878  data: 0.1625  max mem: 41794
Epoch: [209]  [240/312]  eta: 0:00:56  lr: 0.000941  min_lr: 0.000941  loss: 3.3588 (3.2416)  weight_decay: 0.0500 (0.0500)  time: 0.8564  data: 0.3401  max mem: 41794
Epoch: [209]  [250/312]  eta: 0:00:48  lr: 0.000940  min_lr: 0.000940  loss: 3.3754 (3.2470)  weight_decay: 0.0500 (0.0500)  time: 0.6913  data: 0.1845  max mem: 41794
Epoch: [209]  [260/312]  eta: 0:00:40  lr: 0.000940  min_lr: 0.000940  loss: 3.5135 (3.2534)  weight_decay: 0.0500 (0.0500)  time: 0.6958  data: 0.1849  max mem: 41794
Epoch: [209]  [270/312]  eta: 0:00:32  lr: 0.000939  min_lr: 0.000939  loss: 3.3065 (3.2464)  weight_decay: 0.0500 (0.0500)  time: 0.6747  data: 0.1833  max mem: 41794
Epoch: [209]  [280/312]  eta: 0:00:24  lr: 0.000939  min_lr: 0.000939  loss: 3.4382 (3.2505)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.1882  max mem: 41794
Epoch: [209]  [290/312]  eta: 0:00:17  lr: 0.000938  min_lr: 0.000938  loss: 3.4555 (3.2521)  weight_decay: 0.0500 (0.0500)  time: 0.9129  data: 0.4120  max mem: 41794
Epoch: [209]  [300/312]  eta: 0:00:09  lr: 0.000937  min_lr: 0.000937  loss: 3.4393 (3.2589)  weight_decay: 0.0500 (0.0500)  time: 0.7128  data: 0.2242  max mem: 41794
Epoch: [209]  [310/312]  eta: 0:00:01  lr: 0.000937  min_lr: 0.000937  loss: 3.4744 (3.2615)  weight_decay: 0.0500 (0.0500)  time: 0.4682  data: 0.0001  max mem: 41794
Epoch: [209]  [311/312]  eta: 0:00:00  lr: 0.000937  min_lr: 0.000937  loss: 3.4393 (3.2601)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [209] Total time: 0:03:58 (0.7633 s / it)
Averaged stats: lr: 0.000937  min_lr: 0.000937  loss: 3.4393 (3.2276)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 0.9600 (0.9600)  acc1: 81.1198 (81.1198)  acc5: 96.3542 (96.3542)  time: 8.9109  data: 8.7022  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2388 (1.1727)  acc1: 76.1719 (76.3360)  acc5: 93.6198 (93.8400)  time: 1.1380  data: 0.9670  max mem: 41794
Test: Total time: 0:00:10 (1.1827 s / it)
* Acc@1 76.642 Acc@5 93.654 loss 1.176
Accuracy of the model on the 50000 test images: 76.6%
Max accuracy: 76.64%
Epoch: [210]  [  0/312]  eta: 1:16:58  lr: 0.000937  min_lr: 0.000937  loss: 3.8594 (3.8594)  weight_decay: 0.0500 (0.0500)  time: 14.8036  data: 13.4190  max mem: 41794
Epoch: [210]  [ 10/312]  eta: 0:11:39  lr: 0.000936  min_lr: 0.000936  loss: 3.4566 (3.3025)  weight_decay: 0.0500 (0.0500)  time: 2.3164  data: 1.4191  max mem: 41794
Epoch: [210]  [ 20/312]  eta: 0:07:29  lr: 0.000935  min_lr: 0.000935  loss: 3.4793 (3.4171)  weight_decay: 0.0500 (0.0500)  time: 0.8766  data: 0.1629  max mem: 41794
Epoch: [210]  [ 30/312]  eta: 0:05:43  lr: 0.000935  min_lr: 0.000935  loss: 3.4740 (3.3388)  weight_decay: 0.0500 (0.0500)  time: 0.6127  data: 0.0549  max mem: 41794
Epoch: [210]  [ 40/312]  eta: 0:04:44  lr: 0.000934  min_lr: 0.000934  loss: 3.3629 (3.3366)  weight_decay: 0.0500 (0.0500)  time: 0.5259  data: 0.0041  max mem: 41794
Epoch: [210]  [ 50/312]  eta: 0:04:16  lr: 0.000934  min_lr: 0.000934  loss: 3.3633 (3.3351)  weight_decay: 0.0500 (0.0500)  time: 0.6104  data: 0.0800  max mem: 41794
Epoch: [210]  [ 60/312]  eta: 0:03:47  lr: 0.000933  min_lr: 0.000933  loss: 3.4582 (3.3103)  weight_decay: 0.0500 (0.0500)  time: 0.6048  data: 0.0777  max mem: 41794
Epoch: [210]  [ 70/312]  eta: 0:03:35  lr: 0.000932  min_lr: 0.000932  loss: 3.1848 (3.2754)  weight_decay: 0.0500 (0.0500)  time: 0.6620  data: 0.1471  max mem: 41794
Epoch: [210]  [ 80/312]  eta: 0:03:22  lr: 0.000932  min_lr: 0.000932  loss: 3.0761 (3.2467)  weight_decay: 0.0500 (0.0500)  time: 0.7904  data: 0.2844  max mem: 41794
Epoch: [210]  [ 90/312]  eta: 0:03:06  lr: 0.000931  min_lr: 0.000931  loss: 3.2481 (3.2567)  weight_decay: 0.0500 (0.0500)  time: 0.6633  data: 0.1740  max mem: 41794
Epoch: [210]  [100/312]  eta: 0:02:59  lr: 0.000931  min_lr: 0.000931  loss: 3.2382 (3.2378)  weight_decay: 0.0500 (0.0500)  time: 0.7292  data: 0.2354  max mem: 41794
Epoch: [210]  [110/312]  eta: 0:02:44  lr: 0.000930  min_lr: 0.000930  loss: 3.0641 (3.2216)  weight_decay: 0.0500 (0.0500)  time: 0.6927  data: 0.2007  max mem: 41794
Epoch: [210]  [120/312]  eta: 0:02:37  lr: 0.000929  min_lr: 0.000929  loss: 3.1400 (3.2199)  weight_decay: 0.0500 (0.0500)  time: 0.6950  data: 0.1891  max mem: 41794
Epoch: [210]  [130/312]  eta: 0:02:29  lr: 0.000929  min_lr: 0.000929  loss: 3.2774 (3.2064)  weight_decay: 0.0500 (0.0500)  time: 0.8592  data: 0.3501  max mem: 41794
Epoch: [210]  [140/312]  eta: 0:02:17  lr: 0.000928  min_lr: 0.000928  loss: 3.2064 (3.1929)  weight_decay: 0.0500 (0.0500)  time: 0.6664  data: 0.1630  max mem: 41794
Epoch: [210]  [150/312]  eta: 0:02:10  lr: 0.000928  min_lr: 0.000928  loss: 3.4905 (3.2122)  weight_decay: 0.0500 (0.0500)  time: 0.7249  data: 0.2210  max mem: 41794
Epoch: [210]  [160/312]  eta: 0:02:02  lr: 0.000927  min_lr: 0.000927  loss: 3.4319 (3.2132)  weight_decay: 0.0500 (0.0500)  time: 0.8741  data: 0.3757  max mem: 41794
Epoch: [210]  [170/312]  eta: 0:01:52  lr: 0.000926  min_lr: 0.000926  loss: 3.2140 (3.2119)  weight_decay: 0.0500 (0.0500)  time: 0.6919  data: 0.1875  max mem: 41794
Epoch: [210]  [180/312]  eta: 0:01:45  lr: 0.000926  min_lr: 0.000926  loss: 3.1652 (3.2033)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.2012  max mem: 41794
Epoch: [210]  [190/312]  eta: 0:01:35  lr: 0.000925  min_lr: 0.000925  loss: 3.2215 (3.1992)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1690  max mem: 41794
Epoch: [210]  [200/312]  eta: 0:01:28  lr: 0.000925  min_lr: 0.000925  loss: 3.4002 (3.2027)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.1877  max mem: 41794
Epoch: [210]  [210/312]  eta: 0:01:20  lr: 0.000924  min_lr: 0.000924  loss: 3.3824 (3.2048)  weight_decay: 0.0500 (0.0500)  time: 0.8734  data: 0.3738  max mem: 41794
Epoch: [210]  [220/312]  eta: 0:01:11  lr: 0.000923  min_lr: 0.000923  loss: 3.0750 (3.1920)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.1910  max mem: 41794
Epoch: [210]  [230/312]  eta: 0:01:04  lr: 0.000923  min_lr: 0.000923  loss: 3.0200 (3.1937)  weight_decay: 0.0500 (0.0500)  time: 0.6945  data: 0.1900  max mem: 41794
Epoch: [210]  [240/312]  eta: 0:00:56  lr: 0.000922  min_lr: 0.000922  loss: 3.4774 (3.2024)  weight_decay: 0.0500 (0.0500)  time: 0.8648  data: 0.3596  max mem: 41794
Epoch: [210]  [250/312]  eta: 0:00:48  lr: 0.000922  min_lr: 0.000922  loss: 3.5034 (3.2113)  weight_decay: 0.0500 (0.0500)  time: 0.7049  data: 0.2059  max mem: 41794
Epoch: [210]  [260/312]  eta: 0:00:40  lr: 0.000921  min_lr: 0.000921  loss: 3.3119 (3.1919)  weight_decay: 0.0500 (0.0500)  time: 0.7164  data: 0.2202  max mem: 41794
Epoch: [210]  [270/312]  eta: 0:00:32  lr: 0.000920  min_lr: 0.000920  loss: 3.3243 (3.2028)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.1893  max mem: 41794
Epoch: [210]  [280/312]  eta: 0:00:24  lr: 0.000920  min_lr: 0.000920  loss: 3.4850 (3.2085)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.1986  max mem: 41794
Epoch: [210]  [290/312]  eta: 0:00:17  lr: 0.000919  min_lr: 0.000919  loss: 3.4042 (3.2070)  weight_decay: 0.0500 (0.0500)  time: 0.9003  data: 0.4041  max mem: 41794
Epoch: [210]  [300/312]  eta: 0:00:09  lr: 0.000918  min_lr: 0.000918  loss: 3.0664 (3.2000)  weight_decay: 0.0500 (0.0500)  time: 0.7070  data: 0.2223  max mem: 41794
Epoch: [210]  [310/312]  eta: 0:00:01  lr: 0.000918  min_lr: 0.000918  loss: 2.8481 (3.1913)  weight_decay: 0.0500 (0.0500)  time: 0.4821  data: 0.0161  max mem: 41794
Epoch: [210]  [311/312]  eta: 0:00:00  lr: 0.000918  min_lr: 0.000918  loss: 2.6945 (3.1881)  weight_decay: 0.0500 (0.0500)  time: 0.4663  data: 0.0001  max mem: 41794
Epoch: [210] Total time: 0:03:58 (0.7650 s / it)
Averaged stats: lr: 0.000918  min_lr: 0.000918  loss: 2.6945 (3.2109)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.9726 (0.9726)  acc1: 82.8125 (82.8125)  acc5: 95.3125 (95.3125)  time: 8.6043  data: 8.3987  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1839 (1.1262)  acc1: 75.1302 (76.1440)  acc5: 94.7917 (93.7120)  time: 1.1154  data: 0.9333  max mem: 41794
Test: Total time: 0:00:10 (1.1621 s / it)
* Acc@1 76.578 Acc@5 93.526 loss 1.128
Accuracy of the model on the 50000 test images: 76.6%
Max accuracy: 76.64%
Epoch: [211]  [  0/312]  eta: 1:26:34  lr: 0.000918  min_lr: 0.000918  loss: 2.6902 (2.6902)  weight_decay: 0.0500 (0.0500)  time: 16.6486  data: 14.5755  max mem: 41794
Epoch: [211]  [ 10/312]  eta: 0:10:24  lr: 0.000917  min_lr: 0.000917  loss: 3.1469 (3.0972)  weight_decay: 0.0500 (0.0500)  time: 2.0664  data: 1.3257  max mem: 41794
Epoch: [211]  [ 20/312]  eta: 0:07:38  lr: 0.000917  min_lr: 0.000917  loss: 3.3547 (3.2227)  weight_decay: 0.0500 (0.0500)  time: 0.8146  data: 0.0435  max mem: 41794
Epoch: [211]  [ 30/312]  eta: 0:05:53  lr: 0.000916  min_lr: 0.000916  loss: 3.3878 (3.2313)  weight_decay: 0.0500 (0.0500)  time: 0.8039  data: 0.0599  max mem: 41794
Epoch: [211]  [ 40/312]  eta: 0:05:03  lr: 0.000915  min_lr: 0.000915  loss: 3.3878 (3.2362)  weight_decay: 0.0500 (0.0500)  time: 0.6371  data: 0.0184  max mem: 41794
Epoch: [211]  [ 50/312]  eta: 0:04:29  lr: 0.000915  min_lr: 0.000915  loss: 3.4097 (3.2551)  weight_decay: 0.0500 (0.0500)  time: 0.6813  data: 0.0068  max mem: 41794
Epoch: [211]  [ 60/312]  eta: 0:03:57  lr: 0.000914  min_lr: 0.000914  loss: 3.2168 (3.2298)  weight_decay: 0.0500 (0.0500)  time: 0.5862  data: 0.0100  max mem: 41794
Epoch: [211]  [ 70/312]  eta: 0:03:44  lr: 0.000914  min_lr: 0.000914  loss: 3.1887 (3.2343)  weight_decay: 0.0500 (0.0500)  time: 0.6658  data: 0.0874  max mem: 41794
Epoch: [211]  [ 80/312]  eta: 0:03:31  lr: 0.000913  min_lr: 0.000913  loss: 3.3091 (3.2614)  weight_decay: 0.0500 (0.0500)  time: 0.8220  data: 0.1286  max mem: 41794
Epoch: [211]  [ 90/312]  eta: 0:03:15  lr: 0.000912  min_lr: 0.000912  loss: 3.4946 (3.2665)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.0775  max mem: 41794
Epoch: [211]  [100/312]  eta: 0:03:06  lr: 0.000912  min_lr: 0.000912  loss: 3.3722 (3.2568)  weight_decay: 0.0500 (0.0500)  time: 0.7545  data: 0.0906  max mem: 41794
Epoch: [211]  [110/312]  eta: 0:02:51  lr: 0.000911  min_lr: 0.000911  loss: 3.3850 (3.2732)  weight_decay: 0.0500 (0.0500)  time: 0.7037  data: 0.0631  max mem: 41794
Epoch: [211]  [120/312]  eta: 0:02:42  lr: 0.000911  min_lr: 0.000911  loss: 3.4467 (3.2589)  weight_decay: 0.0500 (0.0500)  time: 0.6777  data: 0.0551  max mem: 41794
Epoch: [211]  [130/312]  eta: 0:02:33  lr: 0.000910  min_lr: 0.000910  loss: 3.1280 (3.2472)  weight_decay: 0.0500 (0.0500)  time: 0.8267  data: 0.1712  max mem: 41794
Epoch: [211]  [140/312]  eta: 0:02:20  lr: 0.000909  min_lr: 0.000909  loss: 3.0588 (3.2328)  weight_decay: 0.0500 (0.0500)  time: 0.6478  data: 0.1206  max mem: 41794
Epoch: [211]  [150/312]  eta: 0:02:12  lr: 0.000909  min_lr: 0.000909  loss: 2.9908 (3.2130)  weight_decay: 0.0500 (0.0500)  time: 0.6669  data: 0.1591  max mem: 41794
Epoch: [211]  [160/312]  eta: 0:02:04  lr: 0.000908  min_lr: 0.000908  loss: 3.2656 (3.2299)  weight_decay: 0.0500 (0.0500)  time: 0.8352  data: 0.2683  max mem: 41794
Epoch: [211]  [170/312]  eta: 0:01:54  lr: 0.000908  min_lr: 0.000908  loss: 3.4073 (3.2278)  weight_decay: 0.0500 (0.0500)  time: 0.6760  data: 0.1100  max mem: 41794
Epoch: [211]  [180/312]  eta: 0:01:46  lr: 0.000907  min_lr: 0.000907  loss: 3.3282 (3.2323)  weight_decay: 0.0500 (0.0500)  time: 0.7013  data: 0.1361  max mem: 41794
Epoch: [211]  [190/312]  eta: 0:01:36  lr: 0.000906  min_lr: 0.000906  loss: 3.3282 (3.2263)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.1362  max mem: 41794
Epoch: [211]  [200/312]  eta: 0:01:29  lr: 0.000906  min_lr: 0.000906  loss: 3.3880 (3.2260)  weight_decay: 0.0500 (0.0500)  time: 0.7078  data: 0.0994  max mem: 41794
Epoch: [211]  [210/312]  eta: 0:01:21  lr: 0.000905  min_lr: 0.000905  loss: 3.4358 (3.2270)  weight_decay: 0.0500 (0.0500)  time: 0.8558  data: 0.1801  max mem: 41794
Epoch: [211]  [220/312]  eta: 0:01:12  lr: 0.000905  min_lr: 0.000905  loss: 3.4274 (3.2310)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.0814  max mem: 41794
Epoch: [211]  [230/312]  eta: 0:01:04  lr: 0.000904  min_lr: 0.000904  loss: 3.3786 (3.2252)  weight_decay: 0.0500 (0.0500)  time: 0.7317  data: 0.0660  max mem: 41794
Epoch: [211]  [240/312]  eta: 0:00:56  lr: 0.000903  min_lr: 0.000903  loss: 3.2558 (3.2244)  weight_decay: 0.0500 (0.0500)  time: 0.8282  data: 0.1255  max mem: 41794
Epoch: [211]  [250/312]  eta: 0:00:48  lr: 0.000903  min_lr: 0.000903  loss: 3.4289 (3.2367)  weight_decay: 0.0500 (0.0500)  time: 0.6509  data: 0.0601  max mem: 41794
Epoch: [211]  [260/312]  eta: 0:00:40  lr: 0.000902  min_lr: 0.000902  loss: 3.4289 (3.2379)  weight_decay: 0.0500 (0.0500)  time: 0.7395  data: 0.0768  max mem: 41794
Epoch: [211]  [270/312]  eta: 0:00:32  lr: 0.000902  min_lr: 0.000902  loss: 3.3767 (3.2420)  weight_decay: 0.0500 (0.0500)  time: 0.7106  data: 0.0785  max mem: 41794
Epoch: [211]  [280/312]  eta: 0:00:24  lr: 0.000901  min_lr: 0.000901  loss: 3.4753 (3.2511)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.0389  max mem: 41794
Epoch: [211]  [290/312]  eta: 0:00:17  lr: 0.000900  min_lr: 0.000900  loss: 3.2708 (3.2443)  weight_decay: 0.0500 (0.0500)  time: 0.8051  data: 0.0786  max mem: 41794
Epoch: [211]  [300/312]  eta: 0:00:09  lr: 0.000900  min_lr: 0.000900  loss: 3.2575 (3.2485)  weight_decay: 0.0500 (0.0500)  time: 0.6317  data: 0.0417  max mem: 41794
Epoch: [211]  [310/312]  eta: 0:00:01  lr: 0.000899  min_lr: 0.000899  loss: 3.3262 (3.2498)  weight_decay: 0.0500 (0.0500)  time: 0.4957  data: 0.0001  max mem: 41794
Epoch: [211]  [311/312]  eta: 0:00:00  lr: 0.000899  min_lr: 0.000899  loss: 3.3262 (3.2505)  weight_decay: 0.0500 (0.0500)  time: 0.4641  data: 0.0001  max mem: 41794
Epoch: [211] Total time: 0:03:58 (0.7656 s / it)
Averaged stats: lr: 0.000899  min_lr: 0.000899  loss: 3.3262 (3.2160)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 0.9383 (0.9383)  acc1: 82.4219 (82.4219)  acc5: 96.2240 (96.2240)  time: 8.7151  data: 8.5163  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2140 (1.1220)  acc1: 76.3021 (76.5920)  acc5: 93.4896 (93.8560)  time: 1.1236  data: 0.9464  max mem: 41794
Test: Total time: 0:00:10 (1.1691 s / it)
* Acc@1 76.940 Acc@5 93.898 loss 1.123
Accuracy of the model on the 50000 test images: 76.9%
Max accuracy: 76.94%
Epoch: [212]  [  0/312]  eta: 1:21:23  lr: 0.000899  min_lr: 0.000899  loss: 3.5962 (3.5962)  weight_decay: 0.0500 (0.0500)  time: 15.6524  data: 13.1747  max mem: 41794
Epoch: [212]  [ 10/312]  eta: 0:10:45  lr: 0.000898  min_lr: 0.000898  loss: 3.3147 (3.2437)  weight_decay: 0.0500 (0.0500)  time: 2.1375  data: 1.3791  max mem: 41794
Epoch: [212]  [ 20/312]  eta: 0:07:21  lr: 0.000898  min_lr: 0.000898  loss: 3.3147 (3.2795)  weight_decay: 0.0500 (0.0500)  time: 0.8066  data: 0.1900  max mem: 41794
Epoch: [212]  [ 30/312]  eta: 0:05:37  lr: 0.000897  min_lr: 0.000897  loss: 3.3644 (3.2633)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.0934  max mem: 41794
Epoch: [212]  [ 40/312]  eta: 0:04:45  lr: 0.000897  min_lr: 0.000897  loss: 3.3190 (3.2186)  weight_decay: 0.0500 (0.0500)  time: 0.5634  data: 0.0185  max mem: 41794
Epoch: [212]  [ 50/312]  eta: 0:04:24  lr: 0.000896  min_lr: 0.000896  loss: 3.3190 (3.2220)  weight_decay: 0.0500 (0.0500)  time: 0.7226  data: 0.0979  max mem: 41794
Epoch: [212]  [ 60/312]  eta: 0:03:54  lr: 0.000895  min_lr: 0.000895  loss: 3.4384 (3.2326)  weight_decay: 0.0500 (0.0500)  time: 0.6909  data: 0.1030  max mem: 41794
Epoch: [212]  [ 70/312]  eta: 0:03:40  lr: 0.000895  min_lr: 0.000895  loss: 3.4240 (3.2324)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.1229  max mem: 41794
Epoch: [212]  [ 80/312]  eta: 0:03:29  lr: 0.000894  min_lr: 0.000894  loss: 3.3141 (3.2007)  weight_decay: 0.0500 (0.0500)  time: 0.8155  data: 0.2289  max mem: 41794
Epoch: [212]  [ 90/312]  eta: 0:03:11  lr: 0.000894  min_lr: 0.000894  loss: 3.3141 (3.2280)  weight_decay: 0.0500 (0.0500)  time: 0.6862  data: 0.1450  max mem: 41794
Epoch: [212]  [100/312]  eta: 0:03:02  lr: 0.000893  min_lr: 0.000893  loss: 3.4094 (3.2439)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.1626  max mem: 41794
Epoch: [212]  [110/312]  eta: 0:02:47  lr: 0.000892  min_lr: 0.000892  loss: 3.3066 (3.2353)  weight_decay: 0.0500 (0.0500)  time: 0.6729  data: 0.1442  max mem: 41794
Epoch: [212]  [120/312]  eta: 0:02:40  lr: 0.000892  min_lr: 0.000892  loss: 3.3066 (3.2209)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1681  max mem: 41794
Epoch: [212]  [130/312]  eta: 0:02:31  lr: 0.000891  min_lr: 0.000891  loss: 3.3229 (3.2132)  weight_decay: 0.0500 (0.0500)  time: 0.8728  data: 0.2920  max mem: 41794
Epoch: [212]  [140/312]  eta: 0:02:19  lr: 0.000891  min_lr: 0.000891  loss: 3.1364 (3.2090)  weight_decay: 0.0500 (0.0500)  time: 0.6703  data: 0.1245  max mem: 41794
Epoch: [212]  [150/312]  eta: 0:02:11  lr: 0.000890  min_lr: 0.000890  loss: 3.1184 (3.2030)  weight_decay: 0.0500 (0.0500)  time: 0.6642  data: 0.1559  max mem: 41794
Epoch: [212]  [160/312]  eta: 0:02:03  lr: 0.000889  min_lr: 0.000889  loss: 3.3820 (3.2124)  weight_decay: 0.0500 (0.0500)  time: 0.8198  data: 0.2712  max mem: 41794
Epoch: [212]  [170/312]  eta: 0:01:53  lr: 0.000889  min_lr: 0.000889  loss: 3.1817 (3.1877)  weight_decay: 0.0500 (0.0500)  time: 0.7329  data: 0.1995  max mem: 41794
Epoch: [212]  [180/312]  eta: 0:01:47  lr: 0.000888  min_lr: 0.000888  loss: 2.9245 (3.1797)  weight_decay: 0.0500 (0.0500)  time: 0.8033  data: 0.2127  max mem: 41794
Epoch: [212]  [190/312]  eta: 0:01:36  lr: 0.000888  min_lr: 0.000888  loss: 3.1227 (3.1813)  weight_decay: 0.0500 (0.0500)  time: 0.7239  data: 0.1291  max mem: 41794
Epoch: [212]  [200/312]  eta: 0:01:29  lr: 0.000887  min_lr: 0.000887  loss: 3.2592 (3.1862)  weight_decay: 0.0500 (0.0500)  time: 0.6690  data: 0.0864  max mem: 41794
Epoch: [212]  [210/312]  eta: 0:01:21  lr: 0.000886  min_lr: 0.000886  loss: 3.3083 (3.1914)  weight_decay: 0.0500 (0.0500)  time: 0.8379  data: 0.1990  max mem: 41794
Epoch: [212]  [220/312]  eta: 0:01:12  lr: 0.000886  min_lr: 0.000886  loss: 3.4359 (3.2022)  weight_decay: 0.0500 (0.0500)  time: 0.6653  data: 0.1147  max mem: 41794
Epoch: [212]  [230/312]  eta: 0:01:04  lr: 0.000885  min_lr: 0.000885  loss: 3.5091 (3.2078)  weight_decay: 0.0500 (0.0500)  time: 0.6786  data: 0.1475  max mem: 41794
Epoch: [212]  [240/312]  eta: 0:00:56  lr: 0.000885  min_lr: 0.000885  loss: 3.4630 (3.2016)  weight_decay: 0.0500 (0.0500)  time: 0.8673  data: 0.2819  max mem: 41794
Epoch: [212]  [250/312]  eta: 0:00:48  lr: 0.000884  min_lr: 0.000884  loss: 3.3440 (3.2103)  weight_decay: 0.0500 (0.0500)  time: 0.6835  data: 0.1365  max mem: 41794
Epoch: [212]  [260/312]  eta: 0:00:40  lr: 0.000883  min_lr: 0.000883  loss: 3.4013 (3.2107)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.1294  max mem: 41794
Epoch: [212]  [270/312]  eta: 0:00:32  lr: 0.000883  min_lr: 0.000883  loss: 3.3712 (3.2149)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.1293  max mem: 41794
Epoch: [212]  [280/312]  eta: 0:00:24  lr: 0.000882  min_lr: 0.000882  loss: 3.3148 (3.2060)  weight_decay: 0.0500 (0.0500)  time: 0.7057  data: 0.1518  max mem: 41794
Epoch: [212]  [290/312]  eta: 0:00:17  lr: 0.000882  min_lr: 0.000882  loss: 3.3907 (3.2156)  weight_decay: 0.0500 (0.0500)  time: 0.8260  data: 0.2708  max mem: 41794
Epoch: [212]  [300/312]  eta: 0:00:09  lr: 0.000881  min_lr: 0.000881  loss: 3.4351 (3.2144)  weight_decay: 0.0500 (0.0500)  time: 0.6025  data: 0.1193  max mem: 41794
Epoch: [212]  [310/312]  eta: 0:00:01  lr: 0.000880  min_lr: 0.000880  loss: 3.2477 (3.2055)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [212]  [311/312]  eta: 0:00:00  lr: 0.000880  min_lr: 0.000880  loss: 3.2477 (3.2069)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [212] Total time: 0:03:57 (0.7613 s / it)
Averaged stats: lr: 0.000880  min_lr: 0.000880  loss: 3.2477 (3.1948)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.9631 (0.9631)  acc1: 83.0729 (83.0729)  acc5: 95.4427 (95.4427)  time: 8.8698  data: 8.6573  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3182 (1.2416)  acc1: 76.5625 (76.6560)  acc5: 94.0104 (93.6640)  time: 1.1340  data: 0.9620  max mem: 41794
Test: Total time: 0:00:10 (1.1729 s / it)
* Acc@1 76.854 Acc@5 93.660 loss 1.238
Accuracy of the model on the 50000 test images: 76.9%
Max accuracy: 76.94%
Epoch: [213]  [  0/312]  eta: 1:21:31  lr: 0.000880  min_lr: 0.000880  loss: 3.8427 (3.8427)  weight_decay: 0.0500 (0.0500)  time: 15.6782  data: 11.7429  max mem: 41794
Epoch: [213]  [ 10/312]  eta: 0:10:31  lr: 0.000880  min_lr: 0.000880  loss: 3.4395 (3.2706)  weight_decay: 0.0500 (0.0500)  time: 2.0926  data: 1.2389  max mem: 41794
Epoch: [213]  [ 20/312]  eta: 0:07:22  lr: 0.000879  min_lr: 0.000879  loss: 3.1430 (3.1081)  weight_decay: 0.0500 (0.0500)  time: 0.8069  data: 0.1376  max mem: 41794
Epoch: [213]  [ 30/312]  eta: 0:05:40  lr: 0.000879  min_lr: 0.000879  loss: 3.1027 (3.1459)  weight_decay: 0.0500 (0.0500)  time: 0.7188  data: 0.0680  max mem: 41794
Epoch: [213]  [ 40/312]  eta: 0:05:00  lr: 0.000878  min_lr: 0.000878  loss: 3.2209 (3.0935)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.0671  max mem: 41794
Epoch: [213]  [ 50/312]  eta: 0:04:30  lr: 0.000877  min_lr: 0.000877  loss: 3.2209 (3.1308)  weight_decay: 0.0500 (0.0500)  time: 0.7634  data: 0.0759  max mem: 41794
Epoch: [213]  [ 60/312]  eta: 0:04:01  lr: 0.000877  min_lr: 0.000877  loss: 3.3986 (3.1600)  weight_decay: 0.0500 (0.0500)  time: 0.6527  data: 0.0340  max mem: 41794
Epoch: [213]  [ 70/312]  eta: 0:03:47  lr: 0.000876  min_lr: 0.000876  loss: 3.3572 (3.1822)  weight_decay: 0.0500 (0.0500)  time: 0.7090  data: 0.0511  max mem: 41794
Epoch: [213]  [ 80/312]  eta: 0:03:35  lr: 0.000876  min_lr: 0.000876  loss: 3.3416 (3.1763)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.0527  max mem: 41794
Epoch: [213]  [ 90/312]  eta: 0:03:18  lr: 0.000875  min_lr: 0.000875  loss: 3.1169 (3.1819)  weight_decay: 0.0500 (0.0500)  time: 0.7183  data: 0.0613  max mem: 41794
Epoch: [213]  [100/312]  eta: 0:03:08  lr: 0.000874  min_lr: 0.000874  loss: 3.1852 (3.1849)  weight_decay: 0.0500 (0.0500)  time: 0.7315  data: 0.1009  max mem: 41794
Epoch: [213]  [110/312]  eta: 0:02:52  lr: 0.000874  min_lr: 0.000874  loss: 3.3832 (3.1874)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.0419  max mem: 41794
Epoch: [213]  [120/312]  eta: 0:02:45  lr: 0.000873  min_lr: 0.000873  loss: 3.2727 (3.1958)  weight_decay: 0.0500 (0.0500)  time: 0.7317  data: 0.0593  max mem: 41794
Epoch: [213]  [130/312]  eta: 0:02:36  lr: 0.000873  min_lr: 0.000873  loss: 3.4084 (3.2027)  weight_decay: 0.0500 (0.0500)  time: 0.8727  data: 0.1101  max mem: 41794
Epoch: [213]  [140/312]  eta: 0:02:23  lr: 0.000872  min_lr: 0.000872  loss: 3.3369 (3.1853)  weight_decay: 0.0500 (0.0500)  time: 0.6452  data: 0.0515  max mem: 41794
Epoch: [213]  [150/312]  eta: 0:02:14  lr: 0.000871  min_lr: 0.000871  loss: 2.9287 (3.1807)  weight_decay: 0.0500 (0.0500)  time: 0.6607  data: 0.0471  max mem: 41794
Epoch: [213]  [160/312]  eta: 0:02:05  lr: 0.000871  min_lr: 0.000871  loss: 3.2703 (3.1856)  weight_decay: 0.0500 (0.0500)  time: 0.7865  data: 0.0535  max mem: 41794
Epoch: [213]  [170/312]  eta: 0:01:55  lr: 0.000870  min_lr: 0.000870  loss: 2.9841 (3.1663)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.0403  max mem: 41794
Epoch: [213]  [180/312]  eta: 0:01:48  lr: 0.000870  min_lr: 0.000870  loss: 3.1279 (3.1659)  weight_decay: 0.0500 (0.0500)  time: 0.7685  data: 0.0799  max mem: 41794
Epoch: [213]  [190/312]  eta: 0:01:38  lr: 0.000869  min_lr: 0.000869  loss: 3.2533 (3.1678)  weight_decay: 0.0500 (0.0500)  time: 0.7046  data: 0.0483  max mem: 41794
Epoch: [213]  [200/312]  eta: 0:01:30  lr: 0.000868  min_lr: 0.000868  loss: 3.3652 (3.1765)  weight_decay: 0.0500 (0.0500)  time: 0.6573  data: 0.0574  max mem: 41794
Epoch: [213]  [210/312]  eta: 0:01:22  lr: 0.000868  min_lr: 0.000868  loss: 3.3163 (3.1744)  weight_decay: 0.0500 (0.0500)  time: 0.8281  data: 0.1190  max mem: 41794
Epoch: [213]  [220/312]  eta: 0:01:12  lr: 0.000867  min_lr: 0.000867  loss: 3.2765 (3.1818)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.0639  max mem: 41794
Epoch: [213]  [230/312]  eta: 0:01:05  lr: 0.000867  min_lr: 0.000867  loss: 3.3811 (3.1879)  weight_decay: 0.0500 (0.0500)  time: 0.6927  data: 0.0834  max mem: 41794
Epoch: [213]  [240/312]  eta: 0:00:56  lr: 0.000866  min_lr: 0.000866  loss: 3.3749 (3.1817)  weight_decay: 0.0500 (0.0500)  time: 0.7755  data: 0.0874  max mem: 41794
Epoch: [213]  [250/312]  eta: 0:00:48  lr: 0.000865  min_lr: 0.000865  loss: 3.3641 (3.1922)  weight_decay: 0.0500 (0.0500)  time: 0.6988  data: 0.0671  max mem: 41794
Epoch: [213]  [260/312]  eta: 0:00:41  lr: 0.000865  min_lr: 0.000865  loss: 3.4861 (3.1943)  weight_decay: 0.0500 (0.0500)  time: 0.8299  data: 0.1302  max mem: 41794
Epoch: [213]  [270/312]  eta: 0:00:32  lr: 0.000864  min_lr: 0.000864  loss: 3.4979 (3.1964)  weight_decay: 0.0500 (0.0500)  time: 0.7144  data: 0.0693  max mem: 41794
Epoch: [213]  [280/312]  eta: 0:00:25  lr: 0.000864  min_lr: 0.000864  loss: 3.2573 (3.1977)  weight_decay: 0.0500 (0.0500)  time: 0.6659  data: 0.0503  max mem: 41794
Epoch: [213]  [290/312]  eta: 0:00:17  lr: 0.000863  min_lr: 0.000863  loss: 3.2238 (3.1959)  weight_decay: 0.0500 (0.0500)  time: 0.7692  data: 0.0813  max mem: 41794
Epoch: [213]  [300/312]  eta: 0:00:09  lr: 0.000863  min_lr: 0.000863  loss: 3.1877 (3.1936)  weight_decay: 0.0500 (0.0500)  time: 0.5929  data: 0.0330  max mem: 41794
Epoch: [213]  [310/312]  eta: 0:00:01  lr: 0.000862  min_lr: 0.000862  loss: 3.2749 (3.1982)  weight_decay: 0.0500 (0.0500)  time: 0.4632  data: 0.0001  max mem: 41794
Epoch: [213]  [311/312]  eta: 0:00:00  lr: 0.000862  min_lr: 0.000862  loss: 3.4487 (3.1991)  weight_decay: 0.0500 (0.0500)  time: 0.4637  data: 0.0001  max mem: 41794
Epoch: [213] Total time: 0:03:59 (0.7676 s / it)
Averaged stats: lr: 0.000862  min_lr: 0.000862  loss: 3.4487 (3.2013)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 1.0789 (1.0789)  acc1: 84.3750 (84.3750)  acc5: 96.2240 (96.2240)  time: 8.4974  data: 8.2886  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3986 (1.3097)  acc1: 74.8698 (76.2560)  acc5: 93.3594 (93.3920)  time: 1.1058  data: 0.9348  max mem: 41794
Test: Total time: 0:00:10 (1.1213 s / it)
* Acc@1 76.458 Acc@5 93.474 loss 1.314
Accuracy of the model on the 50000 test images: 76.5%
Max accuracy: 76.94%
Epoch: [214]  [  0/312]  eta: 1:24:11  lr: 0.000862  min_lr: 0.000862  loss: 3.5108 (3.5108)  weight_decay: 0.0500 (0.0500)  time: 16.1913  data: 15.4696  max mem: 41794
Epoch: [214]  [ 10/312]  eta: 0:11:45  lr: 0.000861  min_lr: 0.000861  loss: 3.3816 (3.2297)  weight_decay: 0.0500 (0.0500)  time: 2.3363  data: 1.4164  max mem: 41794
Epoch: [214]  [ 20/312]  eta: 0:07:29  lr: 0.000861  min_lr: 0.000861  loss: 3.3816 (3.2409)  weight_decay: 0.0500 (0.0500)  time: 0.8075  data: 0.0184  max mem: 41794
Epoch: [214]  [ 30/312]  eta: 0:05:43  lr: 0.000860  min_lr: 0.000860  loss: 3.3436 (3.2175)  weight_decay: 0.0500 (0.0500)  time: 0.6016  data: 0.0207  max mem: 41794
Epoch: [214]  [ 40/312]  eta: 0:04:46  lr: 0.000859  min_lr: 0.000859  loss: 3.2044 (3.2441)  weight_decay: 0.0500 (0.0500)  time: 0.5413  data: 0.0098  max mem: 41794
Epoch: [214]  [ 50/312]  eta: 0:04:24  lr: 0.000859  min_lr: 0.000859  loss: 3.3293 (3.2264)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.0106  max mem: 41794
Epoch: [214]  [ 60/312]  eta: 0:03:55  lr: 0.000858  min_lr: 0.000858  loss: 3.4986 (3.2645)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.0355  max mem: 41794
Epoch: [214]  [ 70/312]  eta: 0:03:40  lr: 0.000858  min_lr: 0.000858  loss: 3.4986 (3.2798)  weight_decay: 0.0500 (0.0500)  time: 0.6585  data: 0.0673  max mem: 41794
Epoch: [214]  [ 80/312]  eta: 0:03:26  lr: 0.000857  min_lr: 0.000857  loss: 3.4618 (3.2956)  weight_decay: 0.0500 (0.0500)  time: 0.7459  data: 0.0726  max mem: 41794
Epoch: [214]  [ 90/312]  eta: 0:03:11  lr: 0.000857  min_lr: 0.000857  loss: 3.5147 (3.3143)  weight_decay: 0.0500 (0.0500)  time: 0.6888  data: 0.0805  max mem: 41794
Epoch: [214]  [100/312]  eta: 0:03:03  lr: 0.000856  min_lr: 0.000856  loss: 3.3039 (3.3042)  weight_decay: 0.0500 (0.0500)  time: 0.7755  data: 0.1462  max mem: 41794
Epoch: [214]  [110/312]  eta: 0:02:48  lr: 0.000855  min_lr: 0.000855  loss: 3.3085 (3.2892)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.0982  max mem: 41794
Epoch: [214]  [120/312]  eta: 0:02:41  lr: 0.000855  min_lr: 0.000855  loss: 3.4256 (3.2789)  weight_decay: 0.0500 (0.0500)  time: 0.7084  data: 0.0694  max mem: 41794
Epoch: [214]  [130/312]  eta: 0:02:32  lr: 0.000854  min_lr: 0.000854  loss: 3.3800 (3.2573)  weight_decay: 0.0500 (0.0500)  time: 0.8579  data: 0.0963  max mem: 41794
Epoch: [214]  [140/312]  eta: 0:02:20  lr: 0.000854  min_lr: 0.000854  loss: 2.8883 (3.2362)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.0679  max mem: 41794
Epoch: [214]  [150/312]  eta: 0:02:12  lr: 0.000853  min_lr: 0.000853  loss: 3.1517 (3.2361)  weight_decay: 0.0500 (0.0500)  time: 0.7145  data: 0.0873  max mem: 41794
Epoch: [214]  [160/312]  eta: 0:02:04  lr: 0.000852  min_lr: 0.000852  loss: 3.3726 (3.2487)  weight_decay: 0.0500 (0.0500)  time: 0.8052  data: 0.0623  max mem: 41794
Epoch: [214]  [170/312]  eta: 0:01:54  lr: 0.000852  min_lr: 0.000852  loss: 3.3726 (3.2461)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.0626  max mem: 41794
Epoch: [214]  [180/312]  eta: 0:01:47  lr: 0.000851  min_lr: 0.000851  loss: 3.2921 (3.2487)  weight_decay: 0.0500 (0.0500)  time: 0.7685  data: 0.1460  max mem: 41794
Epoch: [214]  [190/312]  eta: 0:01:36  lr: 0.000851  min_lr: 0.000851  loss: 3.4566 (3.2434)  weight_decay: 0.0500 (0.0500)  time: 0.7153  data: 0.1022  max mem: 41794
Epoch: [214]  [200/312]  eta: 0:01:29  lr: 0.000850  min_lr: 0.000850  loss: 3.4667 (3.2407)  weight_decay: 0.0500 (0.0500)  time: 0.6723  data: 0.0790  max mem: 41794
Epoch: [214]  [210/312]  eta: 0:01:21  lr: 0.000849  min_lr: 0.000849  loss: 3.2877 (3.2380)  weight_decay: 0.0500 (0.0500)  time: 0.8711  data: 0.1790  max mem: 41794
Epoch: [214]  [220/312]  eta: 0:01:12  lr: 0.000849  min_lr: 0.000849  loss: 3.2877 (3.2429)  weight_decay: 0.0500 (0.0500)  time: 0.6984  data: 0.1034  max mem: 41794
Epoch: [214]  [230/312]  eta: 0:01:05  lr: 0.000848  min_lr: 0.000848  loss: 3.2157 (3.2300)  weight_decay: 0.0500 (0.0500)  time: 0.7102  data: 0.0869  max mem: 41794
Epoch: [214]  [240/312]  eta: 0:00:56  lr: 0.000848  min_lr: 0.000848  loss: 2.9791 (3.2238)  weight_decay: 0.0500 (0.0500)  time: 0.7579  data: 0.0869  max mem: 41794
Epoch: [214]  [250/312]  eta: 0:00:48  lr: 0.000847  min_lr: 0.000847  loss: 3.1867 (3.2284)  weight_decay: 0.0500 (0.0500)  time: 0.6179  data: 0.0778  max mem: 41794
Epoch: [214]  [260/312]  eta: 0:00:40  lr: 0.000846  min_lr: 0.000846  loss: 3.2696 (3.2199)  weight_decay: 0.0500 (0.0500)  time: 0.7592  data: 0.2070  max mem: 41794
Epoch: [214]  [270/312]  eta: 0:00:32  lr: 0.000846  min_lr: 0.000846  loss: 3.1553 (3.2171)  weight_decay: 0.0500 (0.0500)  time: 0.6891  data: 0.1298  max mem: 41794
Epoch: [214]  [280/312]  eta: 0:00:24  lr: 0.000845  min_lr: 0.000845  loss: 3.1962 (3.2164)  weight_decay: 0.0500 (0.0500)  time: 0.7051  data: 0.1230  max mem: 41794
Epoch: [214]  [290/312]  eta: 0:00:17  lr: 0.000845  min_lr: 0.000845  loss: 3.3709 (3.2132)  weight_decay: 0.0500 (0.0500)  time: 0.8766  data: 0.2748  max mem: 41794
Epoch: [214]  [300/312]  eta: 0:00:09  lr: 0.000844  min_lr: 0.000844  loss: 3.4377 (3.2207)  weight_decay: 0.0500 (0.0500)  time: 0.6625  data: 0.1522  max mem: 41794
Epoch: [214]  [310/312]  eta: 0:00:01  lr: 0.000844  min_lr: 0.000844  loss: 3.4488 (3.2219)  weight_decay: 0.0500 (0.0500)  time: 0.4710  data: 0.0001  max mem: 41794
Epoch: [214]  [311/312]  eta: 0:00:00  lr: 0.000844  min_lr: 0.000844  loss: 3.4488 (3.2220)  weight_decay: 0.0500 (0.0500)  time: 0.4696  data: 0.0001  max mem: 41794
Epoch: [214] Total time: 0:03:59 (0.7668 s / it)
Averaged stats: lr: 0.000844  min_lr: 0.000844  loss: 3.4488 (3.2000)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 0.9707 (0.9707)  acc1: 83.5938 (83.5938)  acc5: 95.4427 (95.4427)  time: 8.6719  data: 8.4661  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2779 (1.2121)  acc1: 75.9115 (76.5440)  acc5: 93.0990 (93.3920)  time: 1.1133  data: 0.9408  max mem: 41794
Test: Total time: 0:00:10 (1.1586 s / it)
* Acc@1 76.796 Acc@5 93.616 loss 1.207
Accuracy of the model on the 50000 test images: 76.8%
Max accuracy: 76.94%
Epoch: [215]  [  0/312]  eta: 1:24:38  lr: 0.000843  min_lr: 0.000843  loss: 2.5092 (2.5092)  weight_decay: 0.0500 (0.0500)  time: 16.2776  data: 12.2783  max mem: 41794
Epoch: [215]  [ 10/312]  eta: 0:11:21  lr: 0.000843  min_lr: 0.000843  loss: 3.0423 (2.9201)  weight_decay: 0.0500 (0.0500)  time: 2.2562  data: 1.4012  max mem: 41794
Epoch: [215]  [ 20/312]  eta: 0:07:37  lr: 0.000842  min_lr: 0.000842  loss: 3.1644 (3.1004)  weight_decay: 0.0500 (0.0500)  time: 0.8298  data: 0.2324  max mem: 41794
Epoch: [215]  [ 30/312]  eta: 0:05:48  lr: 0.000842  min_lr: 0.000842  loss: 2.8590 (2.9967)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.0823  max mem: 41794
Epoch: [215]  [ 40/312]  eta: 0:04:50  lr: 0.000841  min_lr: 0.000841  loss: 2.7273 (3.0201)  weight_decay: 0.0500 (0.0500)  time: 0.5487  data: 0.0069  max mem: 41794
Epoch: [215]  [ 50/312]  eta: 0:04:31  lr: 0.000841  min_lr: 0.000841  loss: 3.2389 (3.0715)  weight_decay: 0.0500 (0.0500)  time: 0.7218  data: 0.0567  max mem: 41794
Epoch: [215]  [ 60/312]  eta: 0:03:58  lr: 0.000840  min_lr: 0.000840  loss: 3.3660 (3.0681)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.0581  max mem: 41794
Epoch: [215]  [ 70/312]  eta: 0:03:45  lr: 0.000839  min_lr: 0.000839  loss: 3.3660 (3.1003)  weight_decay: 0.0500 (0.0500)  time: 0.6681  data: 0.0420  max mem: 41794
Epoch: [215]  [ 80/312]  eta: 0:03:34  lr: 0.000839  min_lr: 0.000839  loss: 3.2208 (3.1052)  weight_decay: 0.0500 (0.0500)  time: 0.8641  data: 0.0426  max mem: 41794
Epoch: [215]  [ 90/312]  eta: 0:03:17  lr: 0.000838  min_lr: 0.000838  loss: 3.2208 (3.1239)  weight_decay: 0.0500 (0.0500)  time: 0.7376  data: 0.0537  max mem: 41794
Epoch: [215]  [100/312]  eta: 0:03:07  lr: 0.000838  min_lr: 0.000838  loss: 3.4253 (3.1492)  weight_decay: 0.0500 (0.0500)  time: 0.7280  data: 0.0740  max mem: 41794
Epoch: [215]  [110/312]  eta: 0:02:52  lr: 0.000837  min_lr: 0.000837  loss: 3.4471 (3.1737)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.0293  max mem: 41794
Epoch: [215]  [120/312]  eta: 0:02:43  lr: 0.000836  min_lr: 0.000836  loss: 3.1073 (3.1458)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.0385  max mem: 41794
Epoch: [215]  [130/312]  eta: 0:02:35  lr: 0.000836  min_lr: 0.000836  loss: 2.9860 (3.1512)  weight_decay: 0.0500 (0.0500)  time: 0.8765  data: 0.0738  max mem: 41794
Epoch: [215]  [140/312]  eta: 0:02:22  lr: 0.000835  min_lr: 0.000835  loss: 3.2118 (3.1596)  weight_decay: 0.0500 (0.0500)  time: 0.6900  data: 0.0439  max mem: 41794
Epoch: [215]  [150/312]  eta: 0:02:14  lr: 0.000835  min_lr: 0.000835  loss: 3.0989 (3.1578)  weight_decay: 0.0500 (0.0500)  time: 0.6521  data: 0.0347  max mem: 41794
Epoch: [215]  [160/312]  eta: 0:02:05  lr: 0.000834  min_lr: 0.000834  loss: 3.2056 (3.1603)  weight_decay: 0.0500 (0.0500)  time: 0.8124  data: 0.0398  max mem: 41794
Epoch: [215]  [170/312]  eta: 0:01:55  lr: 0.000833  min_lr: 0.000833  loss: 3.2596 (3.1464)  weight_decay: 0.0500 (0.0500)  time: 0.7191  data: 0.0705  max mem: 41794
Epoch: [215]  [180/312]  eta: 0:01:47  lr: 0.000833  min_lr: 0.000833  loss: 3.0111 (3.1395)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.0918  max mem: 41794
Epoch: [215]  [190/312]  eta: 0:01:37  lr: 0.000832  min_lr: 0.000832  loss: 3.2048 (3.1395)  weight_decay: 0.0500 (0.0500)  time: 0.6441  data: 0.0285  max mem: 41794
Epoch: [215]  [200/312]  eta: 0:01:29  lr: 0.000832  min_lr: 0.000832  loss: 3.2463 (3.1475)  weight_decay: 0.0500 (0.0500)  time: 0.6570  data: 0.0512  max mem: 41794
Epoch: [215]  [210/312]  eta: 0:01:21  lr: 0.000831  min_lr: 0.000831  loss: 3.3943 (3.1633)  weight_decay: 0.0500 (0.0500)  time: 0.8158  data: 0.1265  max mem: 41794
Epoch: [215]  [220/312]  eta: 0:01:12  lr: 0.000831  min_lr: 0.000831  loss: 3.3752 (3.1596)  weight_decay: 0.0500 (0.0500)  time: 0.6542  data: 0.0759  max mem: 41794
Epoch: [215]  [230/312]  eta: 0:01:04  lr: 0.000830  min_lr: 0.000830  loss: 3.1595 (3.1612)  weight_decay: 0.0500 (0.0500)  time: 0.6953  data: 0.0714  max mem: 41794
Epoch: [215]  [240/312]  eta: 0:00:57  lr: 0.000829  min_lr: 0.000829  loss: 3.1096 (3.1500)  weight_decay: 0.0500 (0.0500)  time: 0.8608  data: 0.1128  max mem: 41794
Epoch: [215]  [250/312]  eta: 0:00:48  lr: 0.000829  min_lr: 0.000829  loss: 2.9601 (3.1517)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.0674  max mem: 41794
Epoch: [215]  [260/312]  eta: 0:00:40  lr: 0.000828  min_lr: 0.000828  loss: 3.0547 (3.1540)  weight_decay: 0.0500 (0.0500)  time: 0.7153  data: 0.0862  max mem: 41794
Epoch: [215]  [270/312]  eta: 0:00:32  lr: 0.000828  min_lr: 0.000828  loss: 3.3629 (3.1619)  weight_decay: 0.0500 (0.0500)  time: 0.6985  data: 0.0609  max mem: 41794
Epoch: [215]  [280/312]  eta: 0:00:24  lr: 0.000827  min_lr: 0.000827  loss: 3.4446 (3.1677)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.0643  max mem: 41794
Epoch: [215]  [290/312]  eta: 0:00:17  lr: 0.000827  min_lr: 0.000827  loss: 3.2507 (3.1662)  weight_decay: 0.0500 (0.0500)  time: 0.8249  data: 0.1185  max mem: 41794
Epoch: [215]  [300/312]  eta: 0:00:09  lr: 0.000826  min_lr: 0.000826  loss: 3.1078 (3.1591)  weight_decay: 0.0500 (0.0500)  time: 0.6279  data: 0.0546  max mem: 41794
Epoch: [215]  [310/312]  eta: 0:00:01  lr: 0.000825  min_lr: 0.000825  loss: 3.0849 (3.1639)  weight_decay: 0.0500 (0.0500)  time: 0.4680  data: 0.0001  max mem: 41794
Epoch: [215]  [311/312]  eta: 0:00:00  lr: 0.000825  min_lr: 0.000825  loss: 3.0849 (3.1640)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [215] Total time: 0:03:58 (0.7651 s / it)
Averaged stats: lr: 0.000825  min_lr: 0.000825  loss: 3.0849 (3.1901)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 0.9381 (0.9381)  acc1: 82.5521 (82.5521)  acc5: 95.5729 (95.5729)  time: 7.9612  data: 7.7485  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1848 (1.1370)  acc1: 77.7344 (77.2000)  acc5: 94.4010 (93.8400)  time: 1.0705  data: 0.8912  max mem: 41794
Test: Total time: 0:00:09 (1.0809 s / it)
* Acc@1 77.382 Acc@5 93.840 loss 1.125
Accuracy of the model on the 50000 test images: 77.4%
Max accuracy: 77.38%
Epoch: [216]  [  0/312]  eta: 1:23:53  lr: 0.000825  min_lr: 0.000825  loss: 3.7010 (3.7010)  weight_decay: 0.0500 (0.0500)  time: 16.1335  data: 15.6532  max mem: 41794
Epoch: [216]  [ 10/312]  eta: 0:12:33  lr: 0.000825  min_lr: 0.000825  loss: 3.2177 (3.0563)  weight_decay: 0.0500 (0.0500)  time: 2.4960  data: 1.4294  max mem: 41794
Epoch: [216]  [ 20/312]  eta: 0:07:37  lr: 0.000824  min_lr: 0.000824  loss: 3.2724 (3.1855)  weight_decay: 0.0500 (0.0500)  time: 0.8395  data: 0.0056  max mem: 41794
Epoch: [216]  [ 30/312]  eta: 0:05:45  lr: 0.000823  min_lr: 0.000823  loss: 3.3410 (3.1583)  weight_decay: 0.0500 (0.0500)  time: 0.5269  data: 0.0058  max mem: 41794
Epoch: [216]  [ 40/312]  eta: 0:04:45  lr: 0.000823  min_lr: 0.000823  loss: 3.2342 (3.1524)  weight_decay: 0.0500 (0.0500)  time: 0.5080  data: 0.0058  max mem: 41794
Epoch: [216]  [ 50/312]  eta: 0:04:07  lr: 0.000822  min_lr: 0.000822  loss: 3.3095 (3.1804)  weight_decay: 0.0500 (0.0500)  time: 0.5128  data: 0.0041  max mem: 41794
Epoch: [216]  [ 60/312]  eta: 0:03:40  lr: 0.000822  min_lr: 0.000822  loss: 3.3905 (3.1532)  weight_decay: 0.0500 (0.0500)  time: 0.5087  data: 0.0023  max mem: 41794
Epoch: [216]  [ 70/312]  eta: 0:03:19  lr: 0.000821  min_lr: 0.000821  loss: 2.8882 (3.1502)  weight_decay: 0.0500 (0.0500)  time: 0.5186  data: 0.0033  max mem: 41794
Epoch: [216]  [ 80/312]  eta: 0:03:13  lr: 0.000821  min_lr: 0.000821  loss: 3.2783 (3.1574)  weight_decay: 0.0500 (0.0500)  time: 0.7078  data: 0.0565  max mem: 41794
Epoch: [216]  [ 90/312]  eta: 0:03:00  lr: 0.000820  min_lr: 0.000820  loss: 3.2233 (3.1402)  weight_decay: 0.0500 (0.0500)  time: 0.7745  data: 0.1165  max mem: 41794
Epoch: [216]  [100/312]  eta: 0:02:51  lr: 0.000819  min_lr: 0.000819  loss: 3.1674 (3.1284)  weight_decay: 0.0500 (0.0500)  time: 0.7196  data: 0.0820  max mem: 41794
Epoch: [216]  [110/312]  eta: 0:02:44  lr: 0.000819  min_lr: 0.000819  loss: 3.3262 (3.1543)  weight_decay: 0.0500 (0.0500)  time: 0.8069  data: 0.0636  max mem: 41794
Epoch: [216]  [120/312]  eta: 0:02:32  lr: 0.000818  min_lr: 0.000818  loss: 3.1613 (3.1457)  weight_decay: 0.0500 (0.0500)  time: 0.7296  data: 0.0652  max mem: 41794
Epoch: [216]  [130/312]  eta: 0:02:26  lr: 0.000818  min_lr: 0.000818  loss: 3.1613 (3.1478)  weight_decay: 0.0500 (0.0500)  time: 0.7604  data: 0.0498  max mem: 41794
Epoch: [216]  [140/312]  eta: 0:02:18  lr: 0.000817  min_lr: 0.000817  loss: 3.0306 (3.1235)  weight_decay: 0.0500 (0.0500)  time: 0.8652  data: 0.0650  max mem: 41794
Epoch: [216]  [150/312]  eta: 0:02:08  lr: 0.000817  min_lr: 0.000817  loss: 3.0306 (3.1272)  weight_decay: 0.0500 (0.0500)  time: 0.7184  data: 0.0361  max mem: 41794
Epoch: [216]  [160/312]  eta: 0:02:01  lr: 0.000816  min_lr: 0.000816  loss: 3.0888 (3.1099)  weight_decay: 0.0500 (0.0500)  time: 0.7285  data: 0.0112  max mem: 41794
Epoch: [216]  [170/312]  eta: 0:01:51  lr: 0.000815  min_lr: 0.000815  loss: 3.1078 (3.1219)  weight_decay: 0.0500 (0.0500)  time: 0.7491  data: 0.0189  max mem: 41794
Epoch: [216]  [180/312]  eta: 0:01:43  lr: 0.000815  min_lr: 0.000815  loss: 3.3501 (3.1355)  weight_decay: 0.0500 (0.0500)  time: 0.6793  data: 0.0309  max mem: 41794
Epoch: [216]  [190/312]  eta: 0:01:35  lr: 0.000814  min_lr: 0.000814  loss: 3.3157 (3.1381)  weight_decay: 0.0500 (0.0500)  time: 0.7709  data: 0.0403  max mem: 41794
Epoch: [216]  [200/312]  eta: 0:01:27  lr: 0.000814  min_lr: 0.000814  loss: 3.3296 (3.1449)  weight_decay: 0.0500 (0.0500)  time: 0.7930  data: 0.0230  max mem: 41794
Epoch: [216]  [210/312]  eta: 0:01:19  lr: 0.000813  min_lr: 0.000813  loss: 3.4280 (3.1435)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.0336  max mem: 41794
Epoch: [216]  [220/312]  eta: 0:01:11  lr: 0.000812  min_lr: 0.000812  loss: 3.2546 (3.1474)  weight_decay: 0.0500 (0.0500)  time: 0.7124  data: 0.0446  max mem: 41794
Epoch: [216]  [230/312]  eta: 0:01:03  lr: 0.000812  min_lr: 0.000812  loss: 3.3760 (3.1514)  weight_decay: 0.0500 (0.0500)  time: 0.6985  data: 0.0421  max mem: 41794
Epoch: [216]  [240/312]  eta: 0:00:55  lr: 0.000811  min_lr: 0.000811  loss: 3.4444 (3.1556)  weight_decay: 0.0500 (0.0500)  time: 0.7079  data: 0.0537  max mem: 41794
Epoch: [216]  [250/312]  eta: 0:00:47  lr: 0.000811  min_lr: 0.000811  loss: 3.3999 (3.1589)  weight_decay: 0.0500 (0.0500)  time: 0.7860  data: 0.0485  max mem: 41794
Epoch: [216]  [260/312]  eta: 0:00:39  lr: 0.000810  min_lr: 0.000810  loss: 3.3962 (3.1565)  weight_decay: 0.0500 (0.0500)  time: 0.7277  data: 0.0688  max mem: 41794
Epoch: [216]  [270/312]  eta: 0:00:32  lr: 0.000810  min_lr: 0.000810  loss: 3.4265 (3.1627)  weight_decay: 0.0500 (0.0500)  time: 0.7547  data: 0.1050  max mem: 41794
Epoch: [216]  [280/312]  eta: 0:00:24  lr: 0.000809  min_lr: 0.000809  loss: 3.4442 (3.1744)  weight_decay: 0.0500 (0.0500)  time: 0.7606  data: 0.0672  max mem: 41794
Epoch: [216]  [290/312]  eta: 0:00:16  lr: 0.000808  min_lr: 0.000808  loss: 3.4442 (3.1785)  weight_decay: 0.0500 (0.0500)  time: 0.7395  data: 0.0567  max mem: 41794
Epoch: [216]  [300/312]  eta: 0:00:09  lr: 0.000808  min_lr: 0.000808  loss: 3.1569 (3.1733)  weight_decay: 0.0500 (0.0500)  time: 0.6581  data: 0.0552  max mem: 41794
Epoch: [216]  [310/312]  eta: 0:00:01  lr: 0.000807  min_lr: 0.000807  loss: 3.2656 (3.1796)  weight_decay: 0.0500 (0.0500)  time: 0.4990  data: 0.0093  max mem: 41794
Epoch: [216]  [311/312]  eta: 0:00:00  lr: 0.000807  min_lr: 0.000807  loss: 3.4521 (3.1806)  weight_decay: 0.0500 (0.0500)  time: 0.4973  data: 0.0093  max mem: 41794
Epoch: [216] Total time: 0:03:56 (0.7581 s / it)
Averaged stats: lr: 0.000807  min_lr: 0.000807  loss: 3.4521 (3.1928)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 1.0902 (1.0902)  acc1: 83.8542 (83.8542)  acc5: 95.8333 (95.8333)  time: 8.6118  data: 8.4127  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3558 (1.2888)  acc1: 75.2604 (76.5440)  acc5: 94.1406 (93.8720)  time: 1.1116  data: 0.9348  max mem: 41794
Test: Total time: 0:00:10 (1.1582 s / it)
* Acc@1 76.914 Acc@5 93.824 loss 1.279
Accuracy of the model on the 50000 test images: 76.9%
Max accuracy: 77.38%
Epoch: [217]  [  0/312]  eta: 1:22:12  lr: 0.000807  min_lr: 0.000807  loss: 3.6558 (3.6558)  weight_decay: 0.0500 (0.0500)  time: 15.8094  data: 11.3436  max mem: 41794
Epoch: [217]  [ 10/312]  eta: 0:11:21  lr: 0.000807  min_lr: 0.000807  loss: 3.4860 (3.2576)  weight_decay: 0.0500 (0.0500)  time: 2.2554  data: 1.3461  max mem: 41794
Epoch: [217]  [ 20/312]  eta: 0:07:28  lr: 0.000806  min_lr: 0.000806  loss: 3.1877 (3.0660)  weight_decay: 0.0500 (0.0500)  time: 0.8217  data: 0.2464  max mem: 41794
Epoch: [217]  [ 30/312]  eta: 0:05:42  lr: 0.000805  min_lr: 0.000805  loss: 3.1431 (3.1392)  weight_decay: 0.0500 (0.0500)  time: 0.6441  data: 0.0847  max mem: 41794
Epoch: [217]  [ 40/312]  eta: 0:04:50  lr: 0.000805  min_lr: 0.000805  loss: 3.3121 (3.1128)  weight_decay: 0.0500 (0.0500)  time: 0.5751  data: 0.0395  max mem: 41794
Epoch: [217]  [ 50/312]  eta: 0:04:34  lr: 0.000804  min_lr: 0.000804  loss: 3.3860 (3.1575)  weight_decay: 0.0500 (0.0500)  time: 0.7841  data: 0.1442  max mem: 41794
Epoch: [217]  [ 60/312]  eta: 0:04:00  lr: 0.000804  min_lr: 0.000804  loss: 3.3860 (3.1667)  weight_decay: 0.0500 (0.0500)  time: 0.7268  data: 0.1166  max mem: 41794
Epoch: [217]  [ 70/312]  eta: 0:03:47  lr: 0.000803  min_lr: 0.000803  loss: 3.0089 (3.1379)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.0864  max mem: 41794
Epoch: [217]  [ 80/312]  eta: 0:03:30  lr: 0.000803  min_lr: 0.000803  loss: 2.9390 (3.1273)  weight_decay: 0.0500 (0.0500)  time: 0.7597  data: 0.0955  max mem: 41794
Epoch: [217]  [ 90/312]  eta: 0:03:14  lr: 0.000802  min_lr: 0.000802  loss: 2.9390 (3.1367)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.0844  max mem: 41794
Epoch: [217]  [100/312]  eta: 0:03:06  lr: 0.000801  min_lr: 0.000801  loss: 3.5320 (3.1785)  weight_decay: 0.0500 (0.0500)  time: 0.7789  data: 0.2019  max mem: 41794
Epoch: [217]  [110/312]  eta: 0:02:51  lr: 0.000801  min_lr: 0.000801  loss: 3.4297 (3.1861)  weight_decay: 0.0500 (0.0500)  time: 0.7206  data: 0.1273  max mem: 41794
Epoch: [217]  [120/312]  eta: 0:02:43  lr: 0.000800  min_lr: 0.000800  loss: 3.2056 (3.1820)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1069  max mem: 41794
Epoch: [217]  [130/312]  eta: 0:02:35  lr: 0.000800  min_lr: 0.000800  loss: 3.2056 (3.1770)  weight_decay: 0.0500 (0.0500)  time: 0.9001  data: 0.2378  max mem: 41794
Epoch: [217]  [140/312]  eta: 0:02:23  lr: 0.000799  min_lr: 0.000799  loss: 3.0545 (3.1744)  weight_decay: 0.0500 (0.0500)  time: 0.7213  data: 0.1315  max mem: 41794
Epoch: [217]  [150/312]  eta: 0:02:14  lr: 0.000799  min_lr: 0.000799  loss: 3.0545 (3.1505)  weight_decay: 0.0500 (0.0500)  time: 0.6481  data: 0.0732  max mem: 41794
Epoch: [217]  [160/312]  eta: 0:02:05  lr: 0.000798  min_lr: 0.000798  loss: 3.0186 (3.1419)  weight_decay: 0.0500 (0.0500)  time: 0.7694  data: 0.0749  max mem: 41794
Epoch: [217]  [170/312]  eta: 0:01:55  lr: 0.000797  min_lr: 0.000797  loss: 3.2129 (3.1454)  weight_decay: 0.0500 (0.0500)  time: 0.7109  data: 0.0982  max mem: 41794
Epoch: [217]  [180/312]  eta: 0:01:48  lr: 0.000797  min_lr: 0.000797  loss: 3.3414 (3.1559)  weight_decay: 0.0500 (0.0500)  time: 0.8226  data: 0.2052  max mem: 41794
Epoch: [217]  [190/312]  eta: 0:01:38  lr: 0.000796  min_lr: 0.000796  loss: 3.3414 (3.1628)  weight_decay: 0.0500 (0.0500)  time: 0.7466  data: 0.1114  max mem: 41794
Epoch: [217]  [200/312]  eta: 0:01:30  lr: 0.000796  min_lr: 0.000796  loss: 3.3243 (3.1680)  weight_decay: 0.0500 (0.0500)  time: 0.6309  data: 0.0766  max mem: 41794
Epoch: [217]  [210/312]  eta: 0:01:22  lr: 0.000795  min_lr: 0.000795  loss: 3.3145 (3.1667)  weight_decay: 0.0500 (0.0500)  time: 0.8352  data: 0.1866  max mem: 41794
Epoch: [217]  [220/312]  eta: 0:01:13  lr: 0.000794  min_lr: 0.000794  loss: 3.3145 (3.1679)  weight_decay: 0.0500 (0.0500)  time: 0.7154  data: 0.1128  max mem: 41794
Epoch: [217]  [230/312]  eta: 0:01:05  lr: 0.000794  min_lr: 0.000794  loss: 3.0293 (3.1556)  weight_decay: 0.0500 (0.0500)  time: 0.6807  data: 0.0893  max mem: 41794
Epoch: [217]  [240/312]  eta: 0:00:57  lr: 0.000793  min_lr: 0.000793  loss: 3.1915 (3.1564)  weight_decay: 0.0500 (0.0500)  time: 0.7569  data: 0.0940  max mem: 41794
Epoch: [217]  [250/312]  eta: 0:00:48  lr: 0.000793  min_lr: 0.000793  loss: 3.3169 (3.1596)  weight_decay: 0.0500 (0.0500)  time: 0.6687  data: 0.1049  max mem: 41794
Epoch: [217]  [260/312]  eta: 0:00:41  lr: 0.000792  min_lr: 0.000792  loss: 3.3169 (3.1620)  weight_decay: 0.0500 (0.0500)  time: 0.7814  data: 0.2156  max mem: 41794
Epoch: [217]  [270/312]  eta: 0:00:32  lr: 0.000792  min_lr: 0.000792  loss: 3.3189 (3.1584)  weight_decay: 0.0500 (0.0500)  time: 0.6970  data: 0.1224  max mem: 41794
Epoch: [217]  [280/312]  eta: 0:00:25  lr: 0.000791  min_lr: 0.000791  loss: 3.4205 (3.1670)  weight_decay: 0.0500 (0.0500)  time: 0.6674  data: 0.1050  max mem: 41794
Epoch: [217]  [290/312]  eta: 0:00:17  lr: 0.000790  min_lr: 0.000790  loss: 3.3017 (3.1599)  weight_decay: 0.0500 (0.0500)  time: 0.7883  data: 0.1655  max mem: 41794
Epoch: [217]  [300/312]  eta: 0:00:09  lr: 0.000790  min_lr: 0.000790  loss: 3.2622 (3.1627)  weight_decay: 0.0500 (0.0500)  time: 0.6130  data: 0.0672  max mem: 41794
Epoch: [217]  [310/312]  eta: 0:00:01  lr: 0.000789  min_lr: 0.000789  loss: 3.3452 (3.1628)  weight_decay: 0.0500 (0.0500)  time: 0.4624  data: 0.0001  max mem: 41794
Epoch: [217]  [311/312]  eta: 0:00:00  lr: 0.000789  min_lr: 0.000789  loss: 3.3452 (3.1612)  weight_decay: 0.0500 (0.0500)  time: 0.4622  data: 0.0001  max mem: 41794
Epoch: [217] Total time: 0:03:59 (0.7684 s / it)
Averaged stats: lr: 0.000789  min_lr: 0.000789  loss: 3.3452 (3.1852)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.9986 (0.9986)  acc1: 83.0729 (83.0729)  acc5: 95.3125 (95.3125)  time: 8.3290  data: 8.1164  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2717 (1.1751)  acc1: 75.6510 (77.0560)  acc5: 93.7500 (93.8240)  time: 1.0735  data: 0.9019  max mem: 41794
Test: Total time: 0:00:09 (1.0890 s / it)
* Acc@1 77.188 Acc@5 93.842 loss 1.183
Accuracy of the model on the 50000 test images: 77.2%
Max accuracy: 77.38%
Epoch: [218]  [  0/312]  eta: 1:23:14  lr: 0.000789  min_lr: 0.000789  loss: 2.5323 (2.5323)  weight_decay: 0.0500 (0.0500)  time: 16.0086  data: 13.5039  max mem: 41794
Epoch: [218]  [ 10/312]  eta: 0:11:46  lr: 0.000789  min_lr: 0.000789  loss: 3.3589 (3.0935)  weight_decay: 0.0500 (0.0500)  time: 2.3382  data: 1.3735  max mem: 41794
Epoch: [218]  [ 20/312]  eta: 0:07:41  lr: 0.000788  min_lr: 0.000788  loss: 3.3878 (3.2273)  weight_decay: 0.0500 (0.0500)  time: 0.8573  data: 0.1274  max mem: 41794
Epoch: [218]  [ 30/312]  eta: 0:05:49  lr: 0.000788  min_lr: 0.000788  loss: 3.1320 (3.1386)  weight_decay: 0.0500 (0.0500)  time: 0.6329  data: 0.0557  max mem: 41794
Epoch: [218]  [ 40/312]  eta: 0:04:47  lr: 0.000787  min_lr: 0.000787  loss: 2.9854 (3.1277)  weight_decay: 0.0500 (0.0500)  time: 0.5111  data: 0.0088  max mem: 41794
Epoch: [218]  [ 50/312]  eta: 0:04:13  lr: 0.000786  min_lr: 0.000786  loss: 3.1578 (3.1293)  weight_decay: 0.0500 (0.0500)  time: 0.5509  data: 0.0007  max mem: 41794
Epoch: [218]  [ 60/312]  eta: 0:03:45  lr: 0.000786  min_lr: 0.000786  loss: 3.0831 (3.1254)  weight_decay: 0.0500 (0.0500)  time: 0.5586  data: 0.0051  max mem: 41794
Epoch: [218]  [ 70/312]  eta: 0:03:33  lr: 0.000785  min_lr: 0.000785  loss: 3.0904 (3.1290)  weight_decay: 0.0500 (0.0500)  time: 0.6577  data: 0.1458  max mem: 41794
Epoch: [218]  [ 80/312]  eta: 0:03:22  lr: 0.000785  min_lr: 0.000785  loss: 3.1815 (3.1432)  weight_decay: 0.0500 (0.0500)  time: 0.8161  data: 0.3073  max mem: 41794
Epoch: [218]  [ 90/312]  eta: 0:03:06  lr: 0.000784  min_lr: 0.000784  loss: 3.4519 (3.1743)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1777  max mem: 41794
Epoch: [218]  [100/312]  eta: 0:02:58  lr: 0.000784  min_lr: 0.000784  loss: 3.4394 (3.1830)  weight_decay: 0.0500 (0.0500)  time: 0.7150  data: 0.1688  max mem: 41794
Epoch: [218]  [110/312]  eta: 0:02:44  lr: 0.000783  min_lr: 0.000783  loss: 3.1905 (3.1684)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1577  max mem: 41794
Epoch: [218]  [120/312]  eta: 0:02:36  lr: 0.000782  min_lr: 0.000782  loss: 3.3317 (3.1956)  weight_decay: 0.0500 (0.0500)  time: 0.6835  data: 0.1761  max mem: 41794
Epoch: [218]  [130/312]  eta: 0:02:29  lr: 0.000782  min_lr: 0.000782  loss: 3.3135 (3.1805)  weight_decay: 0.0500 (0.0500)  time: 0.8725  data: 0.3462  max mem: 41794
Epoch: [218]  [140/312]  eta: 0:02:17  lr: 0.000781  min_lr: 0.000781  loss: 3.2351 (3.1793)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.1730  max mem: 41794
Epoch: [218]  [150/312]  eta: 0:02:10  lr: 0.000781  min_lr: 0.000781  loss: 3.2264 (3.1768)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.1753  max mem: 41794
Epoch: [218]  [160/312]  eta: 0:02:02  lr: 0.000780  min_lr: 0.000780  loss: 3.3724 (3.1960)  weight_decay: 0.0500 (0.0500)  time: 0.8413  data: 0.3230  max mem: 41794
Epoch: [218]  [170/312]  eta: 0:01:52  lr: 0.000780  min_lr: 0.000780  loss: 3.4259 (3.2052)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.1665  max mem: 41794
Epoch: [218]  [180/312]  eta: 0:01:45  lr: 0.000779  min_lr: 0.000779  loss: 3.3971 (3.2137)  weight_decay: 0.0500 (0.0500)  time: 0.7484  data: 0.1921  max mem: 41794
Epoch: [218]  [190/312]  eta: 0:01:35  lr: 0.000778  min_lr: 0.000778  loss: 3.3654 (3.2131)  weight_decay: 0.0500 (0.0500)  time: 0.7179  data: 0.1760  max mem: 41794
Epoch: [218]  [200/312]  eta: 0:01:27  lr: 0.000778  min_lr: 0.000778  loss: 3.2959 (3.2148)  weight_decay: 0.0500 (0.0500)  time: 0.6451  data: 0.1354  max mem: 41794
Epoch: [218]  [210/312]  eta: 0:01:20  lr: 0.000777  min_lr: 0.000777  loss: 3.2831 (3.2168)  weight_decay: 0.0500 (0.0500)  time: 0.8186  data: 0.3020  max mem: 41794
Epoch: [218]  [220/312]  eta: 0:01:11  lr: 0.000777  min_lr: 0.000777  loss: 3.2831 (3.2213)  weight_decay: 0.0500 (0.0500)  time: 0.6991  data: 0.1673  max mem: 41794
Epoch: [218]  [230/312]  eta: 0:01:03  lr: 0.000776  min_lr: 0.000776  loss: 3.3563 (3.2225)  weight_decay: 0.0500 (0.0500)  time: 0.6965  data: 0.1725  max mem: 41794
Epoch: [218]  [240/312]  eta: 0:00:56  lr: 0.000776  min_lr: 0.000776  loss: 3.3696 (3.2261)  weight_decay: 0.0500 (0.0500)  time: 0.8193  data: 0.3078  max mem: 41794
Epoch: [218]  [250/312]  eta: 0:00:47  lr: 0.000775  min_lr: 0.000775  loss: 3.4062 (3.2294)  weight_decay: 0.0500 (0.0500)  time: 0.7122  data: 0.1861  max mem: 41794
Epoch: [218]  [260/312]  eta: 0:00:40  lr: 0.000774  min_lr: 0.000774  loss: 3.3485 (3.2290)  weight_decay: 0.0500 (0.0500)  time: 0.7300  data: 0.2186  max mem: 41794
Epoch: [218]  [270/312]  eta: 0:00:32  lr: 0.000774  min_lr: 0.000774  loss: 3.1109 (3.2207)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.1684  max mem: 41794
Epoch: [218]  [280/312]  eta: 0:00:24  lr: 0.000773  min_lr: 0.000773  loss: 3.1109 (3.2169)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1797  max mem: 41794
Epoch: [218]  [290/312]  eta: 0:00:17  lr: 0.000773  min_lr: 0.000773  loss: 3.3348 (3.2198)  weight_decay: 0.0500 (0.0500)  time: 0.8709  data: 0.3604  max mem: 41794
Epoch: [218]  [300/312]  eta: 0:00:09  lr: 0.000772  min_lr: 0.000772  loss: 3.5112 (3.2267)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1811  max mem: 41794
Epoch: [218]  [310/312]  eta: 0:00:01  lr: 0.000772  min_lr: 0.000772  loss: 3.4718 (3.2289)  weight_decay: 0.0500 (0.0500)  time: 0.4714  data: 0.0001  max mem: 41794
Epoch: [218]  [311/312]  eta: 0:00:00  lr: 0.000772  min_lr: 0.000772  loss: 3.4633 (3.2284)  weight_decay: 0.0500 (0.0500)  time: 0.4685  data: 0.0001  max mem: 41794
Epoch: [218] Total time: 0:03:56 (0.7592 s / it)
Averaged stats: lr: 0.000772  min_lr: 0.000772  loss: 3.4633 (3.1894)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 0.9894 (0.9894)  acc1: 83.5938 (83.5938)  acc5: 96.2240 (96.2240)  time: 8.6953  data: 8.4922  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3733 (1.2712)  acc1: 76.3021 (76.9280)  acc5: 93.6198 (93.9040)  time: 1.1178  data: 0.9436  max mem: 41794
Test: Total time: 0:00:10 (1.1429 s / it)
* Acc@1 77.140 Acc@5 93.864 loss 1.272
Accuracy of the model on the 50000 test images: 77.1%
Max accuracy: 77.38%
Epoch: [219]  [  0/312]  eta: 1:23:12  lr: 0.000771  min_lr: 0.000771  loss: 3.4537 (3.4537)  weight_decay: 0.0500 (0.0500)  time: 16.0032  data: 15.5124  max mem: 41794
Epoch: [219]  [ 10/312]  eta: 0:10:45  lr: 0.000771  min_lr: 0.000771  loss: 3.2874 (3.2427)  weight_decay: 0.0500 (0.0500)  time: 2.1376  data: 1.4108  max mem: 41794
Epoch: [219]  [ 20/312]  eta: 0:07:04  lr: 0.000770  min_lr: 0.000770  loss: 3.2874 (3.3300)  weight_decay: 0.0500 (0.0500)  time: 0.7247  data: 0.0792  max mem: 41794
Epoch: [219]  [ 30/312]  eta: 0:05:23  lr: 0.000770  min_lr: 0.000770  loss: 3.3388 (3.2920)  weight_decay: 0.0500 (0.0500)  time: 0.6041  data: 0.0793  max mem: 41794
Epoch: [219]  [ 40/312]  eta: 0:04:45  lr: 0.000769  min_lr: 0.000769  loss: 3.3704 (3.2966)  weight_decay: 0.0500 (0.0500)  time: 0.6291  data: 0.0585  max mem: 41794
Epoch: [219]  [ 50/312]  eta: 0:04:26  lr: 0.000769  min_lr: 0.000769  loss: 3.4106 (3.3091)  weight_decay: 0.0500 (0.0500)  time: 0.8177  data: 0.1871  max mem: 41794
Epoch: [219]  [ 60/312]  eta: 0:03:54  lr: 0.000768  min_lr: 0.000768  loss: 3.4070 (3.3016)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1292  max mem: 41794
Epoch: [219]  [ 70/312]  eta: 0:03:45  lr: 0.000767  min_lr: 0.000767  loss: 3.3793 (3.2975)  weight_decay: 0.0500 (0.0500)  time: 0.7046  data: 0.0874  max mem: 41794
Epoch: [219]  [ 80/312]  eta: 0:03:32  lr: 0.000767  min_lr: 0.000767  loss: 3.3532 (3.2990)  weight_decay: 0.0500 (0.0500)  time: 0.8756  data: 0.1927  max mem: 41794
Epoch: [219]  [ 90/312]  eta: 0:03:13  lr: 0.000766  min_lr: 0.000766  loss: 3.3217 (3.2815)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.1059  max mem: 41794
Epoch: [219]  [100/312]  eta: 0:03:05  lr: 0.000766  min_lr: 0.000766  loss: 3.1364 (3.2753)  weight_decay: 0.0500 (0.0500)  time: 0.7078  data: 0.0939  max mem: 41794
Epoch: [219]  [110/312]  eta: 0:02:50  lr: 0.000765  min_lr: 0.000765  loss: 3.4039 (3.2995)  weight_decay: 0.0500 (0.0500)  time: 0.7138  data: 0.1010  max mem: 41794
Epoch: [219]  [120/312]  eta: 0:02:43  lr: 0.000765  min_lr: 0.000765  loss: 3.4849 (3.2978)  weight_decay: 0.0500 (0.0500)  time: 0.7289  data: 0.1131  max mem: 41794
Epoch: [219]  [130/312]  eta: 0:02:34  lr: 0.000764  min_lr: 0.000764  loss: 3.3226 (3.2785)  weight_decay: 0.0500 (0.0500)  time: 0.8887  data: 0.1332  max mem: 41794
Epoch: [219]  [140/312]  eta: 0:02:22  lr: 0.000764  min_lr: 0.000764  loss: 3.4395 (3.2886)  weight_decay: 0.0500 (0.0500)  time: 0.6896  data: 0.0581  max mem: 41794
Epoch: [219]  [150/312]  eta: 0:02:14  lr: 0.000763  min_lr: 0.000763  loss: 3.5201 (3.2798)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.1206  max mem: 41794
Epoch: [219]  [160/312]  eta: 0:02:05  lr: 0.000762  min_lr: 0.000762  loss: 3.4132 (3.2755)  weight_decay: 0.0500 (0.0500)  time: 0.8204  data: 0.2261  max mem: 41794
Epoch: [219]  [170/312]  eta: 0:01:54  lr: 0.000762  min_lr: 0.000762  loss: 3.0819 (3.2550)  weight_decay: 0.0500 (0.0500)  time: 0.6494  data: 0.1364  max mem: 41794
Epoch: [219]  [180/312]  eta: 0:01:46  lr: 0.000761  min_lr: 0.000761  loss: 3.1883 (3.2493)  weight_decay: 0.0500 (0.0500)  time: 0.6690  data: 0.1455  max mem: 41794
Epoch: [219]  [190/312]  eta: 0:01:36  lr: 0.000761  min_lr: 0.000761  loss: 3.2402 (3.2353)  weight_decay: 0.0500 (0.0500)  time: 0.6532  data: 0.1455  max mem: 41794
Epoch: [219]  [200/312]  eta: 0:01:29  lr: 0.000760  min_lr: 0.000760  loss: 3.3262 (3.2437)  weight_decay: 0.0500 (0.0500)  time: 0.6833  data: 0.1885  max mem: 41794
Epoch: [219]  [210/312]  eta: 0:01:21  lr: 0.000760  min_lr: 0.000760  loss: 3.3679 (3.2423)  weight_decay: 0.0500 (0.0500)  time: 0.8534  data: 0.3436  max mem: 41794
Epoch: [219]  [220/312]  eta: 0:01:12  lr: 0.000759  min_lr: 0.000759  loss: 3.4085 (3.2486)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1558  max mem: 41794
Epoch: [219]  [230/312]  eta: 0:01:05  lr: 0.000758  min_lr: 0.000758  loss: 3.4085 (3.2492)  weight_decay: 0.0500 (0.0500)  time: 0.7281  data: 0.1558  max mem: 41794
Epoch: [219]  [240/312]  eta: 0:00:57  lr: 0.000758  min_lr: 0.000758  loss: 3.5079 (3.2593)  weight_decay: 0.0500 (0.0500)  time: 0.8549  data: 0.2890  max mem: 41794
Epoch: [219]  [250/312]  eta: 0:00:48  lr: 0.000757  min_lr: 0.000757  loss: 3.2467 (3.2473)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1352  max mem: 41794
Epoch: [219]  [260/312]  eta: 0:00:40  lr: 0.000757  min_lr: 0.000757  loss: 2.9522 (3.2449)  weight_decay: 0.0500 (0.0500)  time: 0.7299  data: 0.1243  max mem: 41794
Epoch: [219]  [270/312]  eta: 0:00:32  lr: 0.000756  min_lr: 0.000756  loss: 3.2903 (3.2458)  weight_decay: 0.0500 (0.0500)  time: 0.6803  data: 0.1229  max mem: 41794
Epoch: [219]  [280/312]  eta: 0:00:24  lr: 0.000756  min_lr: 0.000756  loss: 3.2903 (3.2396)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.1190  max mem: 41794
Epoch: [219]  [290/312]  eta: 0:00:17  lr: 0.000755  min_lr: 0.000755  loss: 3.2160 (3.2405)  weight_decay: 0.0500 (0.0500)  time: 0.8512  data: 0.2366  max mem: 41794
Epoch: [219]  [300/312]  eta: 0:00:09  lr: 0.000755  min_lr: 0.000755  loss: 3.4117 (3.2428)  weight_decay: 0.0500 (0.0500)  time: 0.6628  data: 0.1180  max mem: 41794
Epoch: [219]  [310/312]  eta: 0:00:01  lr: 0.000754  min_lr: 0.000754  loss: 3.1761 (3.2354)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [219]  [311/312]  eta: 0:00:00  lr: 0.000754  min_lr: 0.000754  loss: 3.0034 (3.2345)  weight_decay: 0.0500 (0.0500)  time: 0.4640  data: 0.0001  max mem: 41794
Epoch: [219] Total time: 0:03:59 (0.7670 s / it)
Averaged stats: lr: 0.000754  min_lr: 0.000754  loss: 3.0034 (3.1855)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 0.8682 (0.8682)  acc1: 82.8125 (82.8125)  acc5: 95.9635 (95.9635)  time: 8.2079  data: 8.0070  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2052 (1.1203)  acc1: 77.8646 (77.4880)  acc5: 94.6615 (93.6800)  time: 1.0666  data: 0.8898  max mem: 41794
Test: Total time: 0:00:09 (1.1106 s / it)
* Acc@1 77.474 Acc@5 94.024 loss 1.120
Accuracy of the model on the 50000 test images: 77.5%
Max accuracy: 77.47%
Epoch: [220]  [  0/312]  eta: 1:20:38  lr: 0.000754  min_lr: 0.000754  loss: 3.4466 (3.4466)  weight_decay: 0.0500 (0.0500)  time: 15.5093  data: 14.7285  max mem: 41794
Epoch: [220]  [ 10/312]  eta: 0:11:11  lr: 0.000753  min_lr: 0.000753  loss: 3.4466 (3.3368)  weight_decay: 0.0500 (0.0500)  time: 2.2249  data: 1.3397  max mem: 41794
Epoch: [220]  [ 20/312]  eta: 0:07:19  lr: 0.000753  min_lr: 0.000753  loss: 3.4787 (3.3627)  weight_decay: 0.0500 (0.0500)  time: 0.8058  data: 0.1047  max mem: 41794
Epoch: [220]  [ 30/312]  eta: 0:05:36  lr: 0.000752  min_lr: 0.000752  loss: 3.4141 (3.3253)  weight_decay: 0.0500 (0.0500)  time: 0.6265  data: 0.1046  max mem: 41794
Epoch: [220]  [ 40/312]  eta: 0:04:45  lr: 0.000752  min_lr: 0.000752  loss: 3.3720 (3.2042)  weight_decay: 0.0500 (0.0500)  time: 0.5718  data: 0.0535  max mem: 41794
Epoch: [220]  [ 50/312]  eta: 0:04:24  lr: 0.000751  min_lr: 0.000751  loss: 3.1660 (3.1947)  weight_decay: 0.0500 (0.0500)  time: 0.7271  data: 0.2246  max mem: 41794
Epoch: [220]  [ 60/312]  eta: 0:03:53  lr: 0.000750  min_lr: 0.000750  loss: 3.1828 (3.1682)  weight_decay: 0.0500 (0.0500)  time: 0.6689  data: 0.1717  max mem: 41794
Epoch: [220]  [ 70/312]  eta: 0:03:42  lr: 0.000750  min_lr: 0.000750  loss: 3.2867 (3.1581)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1847  max mem: 41794
Epoch: [220]  [ 80/312]  eta: 0:03:30  lr: 0.000749  min_lr: 0.000749  loss: 3.4291 (3.2066)  weight_decay: 0.0500 (0.0500)  time: 0.8454  data: 0.3452  max mem: 41794
Epoch: [220]  [ 90/312]  eta: 0:03:11  lr: 0.000749  min_lr: 0.000749  loss: 3.5334 (3.2208)  weight_decay: 0.0500 (0.0500)  time: 0.6641  data: 0.1716  max mem: 41794
Epoch: [220]  [100/312]  eta: 0:03:03  lr: 0.000748  min_lr: 0.000748  loss: 3.2939 (3.2019)  weight_decay: 0.0500 (0.0500)  time: 0.7094  data: 0.2156  max mem: 41794
Epoch: [220]  [110/312]  eta: 0:02:48  lr: 0.000748  min_lr: 0.000748  loss: 3.2133 (3.2100)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.2052  max mem: 41794
Epoch: [220]  [120/312]  eta: 0:02:40  lr: 0.000747  min_lr: 0.000747  loss: 3.2328 (3.1961)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.1843  max mem: 41794
Epoch: [220]  [130/312]  eta: 0:02:32  lr: 0.000747  min_lr: 0.000747  loss: 3.1151 (3.1866)  weight_decay: 0.0500 (0.0500)  time: 0.8807  data: 0.3784  max mem: 41794
Epoch: [220]  [140/312]  eta: 0:02:20  lr: 0.000746  min_lr: 0.000746  loss: 3.0819 (3.1738)  weight_decay: 0.0500 (0.0500)  time: 0.6939  data: 0.1948  max mem: 41794
Epoch: [220]  [150/312]  eta: 0:02:12  lr: 0.000745  min_lr: 0.000745  loss: 3.1393 (3.1704)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.1788  max mem: 41794
Epoch: [220]  [160/312]  eta: 0:02:04  lr: 0.000745  min_lr: 0.000745  loss: 3.2367 (3.1644)  weight_decay: 0.0500 (0.0500)  time: 0.8620  data: 0.3554  max mem: 41794
Epoch: [220]  [170/312]  eta: 0:01:53  lr: 0.000744  min_lr: 0.000744  loss: 3.2697 (3.1631)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1773  max mem: 41794
Epoch: [220]  [180/312]  eta: 0:01:46  lr: 0.000744  min_lr: 0.000744  loss: 3.2697 (3.1655)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.1848  max mem: 41794
Epoch: [220]  [190/312]  eta: 0:01:36  lr: 0.000743  min_lr: 0.000743  loss: 3.3349 (3.1801)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.1848  max mem: 41794
Epoch: [220]  [200/312]  eta: 0:01:28  lr: 0.000743  min_lr: 0.000743  loss: 3.3240 (3.1796)  weight_decay: 0.0500 (0.0500)  time: 0.6737  data: 0.1766  max mem: 41794
Epoch: [220]  [210/312]  eta: 0:01:21  lr: 0.000742  min_lr: 0.000742  loss: 3.2432 (3.1715)  weight_decay: 0.0500 (0.0500)  time: 0.8794  data: 0.3812  max mem: 41794
Epoch: [220]  [220/312]  eta: 0:01:12  lr: 0.000742  min_lr: 0.000742  loss: 3.2754 (3.1691)  weight_decay: 0.0500 (0.0500)  time: 0.7051  data: 0.2052  max mem: 41794
Epoch: [220]  [230/312]  eta: 0:01:04  lr: 0.000741  min_lr: 0.000741  loss: 3.2464 (3.1686)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.1902  max mem: 41794
Epoch: [220]  [240/312]  eta: 0:00:57  lr: 0.000740  min_lr: 0.000740  loss: 3.0310 (3.1560)  weight_decay: 0.0500 (0.0500)  time: 0.8807  data: 0.3672  max mem: 41794
Epoch: [220]  [250/312]  eta: 0:00:48  lr: 0.000740  min_lr: 0.000740  loss: 3.2744 (3.1673)  weight_decay: 0.0500 (0.0500)  time: 0.7028  data: 0.1894  max mem: 41794
Epoch: [220]  [260/312]  eta: 0:00:40  lr: 0.000739  min_lr: 0.000739  loss: 3.4350 (3.1692)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.2105  max mem: 41794
Epoch: [220]  [270/312]  eta: 0:00:32  lr: 0.000739  min_lr: 0.000739  loss: 3.3207 (3.1632)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.1987  max mem: 41794
Epoch: [220]  [280/312]  eta: 0:00:25  lr: 0.000738  min_lr: 0.000738  loss: 3.3207 (3.1675)  weight_decay: 0.0500 (0.0500)  time: 0.7000  data: 0.1987  max mem: 41794
Epoch: [220]  [290/312]  eta: 0:00:17  lr: 0.000738  min_lr: 0.000738  loss: 3.4355 (3.1713)  weight_decay: 0.0500 (0.0500)  time: 0.8359  data: 0.3190  max mem: 41794
Epoch: [220]  [300/312]  eta: 0:00:09  lr: 0.000737  min_lr: 0.000737  loss: 3.4301 (3.1790)  weight_decay: 0.0500 (0.0500)  time: 0.6211  data: 0.1207  max mem: 41794
Epoch: [220]  [310/312]  eta: 0:00:01  lr: 0.000736  min_lr: 0.000736  loss: 3.3445 (3.1758)  weight_decay: 0.0500 (0.0500)  time: 0.4650  data: 0.0001  max mem: 41794
Epoch: [220]  [311/312]  eta: 0:00:00  lr: 0.000736  min_lr: 0.000736  loss: 3.3894 (3.1772)  weight_decay: 0.0500 (0.0500)  time: 0.4641  data: 0.0001  max mem: 41794
Epoch: [220] Total time: 0:03:59 (0.7663 s / it)
Averaged stats: lr: 0.000736  min_lr: 0.000736  loss: 3.3894 (3.1705)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.9717 (0.9717)  acc1: 82.6823 (82.6823)  acc5: 95.8333 (95.8333)  time: 8.5921  data: 8.3791  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2012 (1.1175)  acc1: 76.6927 (77.2160)  acc5: 93.0990 (94.0960)  time: 1.1043  data: 0.9324  max mem: 41794
Test: Total time: 0:00:10 (1.1460 s / it)
* Acc@1 77.624 Acc@5 94.050 loss 1.118
Accuracy of the model on the 50000 test images: 77.6%
Max accuracy: 77.62%
Epoch: [221]  [  0/312]  eta: 1:23:33  lr: 0.000736  min_lr: 0.000736  loss: 2.4510 (2.4510)  weight_decay: 0.0500 (0.0500)  time: 16.0694  data: 15.6056  max mem: 41794
Epoch: [221]  [ 10/312]  eta: 0:10:44  lr: 0.000736  min_lr: 0.000736  loss: 3.3858 (3.2780)  weight_decay: 0.0500 (0.0500)  time: 2.1348  data: 1.4205  max mem: 41794
Epoch: [221]  [ 20/312]  eta: 0:07:15  lr: 0.000735  min_lr: 0.000735  loss: 3.3878 (3.3228)  weight_decay: 0.0500 (0.0500)  time: 0.7629  data: 0.1429  max mem: 41794
Epoch: [221]  [ 30/312]  eta: 0:05:31  lr: 0.000735  min_lr: 0.000735  loss: 3.2767 (3.2555)  weight_decay: 0.0500 (0.0500)  time: 0.6452  data: 0.1436  max mem: 41794
Epoch: [221]  [ 40/312]  eta: 0:04:50  lr: 0.000734  min_lr: 0.000734  loss: 3.1235 (3.2368)  weight_decay: 0.0500 (0.0500)  time: 0.6206  data: 0.1282  max mem: 41794
Epoch: [221]  [ 50/312]  eta: 0:04:29  lr: 0.000734  min_lr: 0.000734  loss: 3.0546 (3.2420)  weight_decay: 0.0500 (0.0500)  time: 0.8065  data: 0.3015  max mem: 41794
Epoch: [221]  [ 60/312]  eta: 0:03:57  lr: 0.000733  min_lr: 0.000733  loss: 3.3716 (3.2423)  weight_decay: 0.0500 (0.0500)  time: 0.6906  data: 0.1753  max mem: 41794
Epoch: [221]  [ 70/312]  eta: 0:03:43  lr: 0.000732  min_lr: 0.000732  loss: 3.2995 (3.2211)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1476  max mem: 41794
Epoch: [221]  [ 80/312]  eta: 0:03:30  lr: 0.000732  min_lr: 0.000732  loss: 3.1697 (3.2230)  weight_decay: 0.0500 (0.0500)  time: 0.7981  data: 0.2862  max mem: 41794
Epoch: [221]  [ 90/312]  eta: 0:03:13  lr: 0.000731  min_lr: 0.000731  loss: 2.9713 (3.1779)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1392  max mem: 41794
Epoch: [221]  [100/312]  eta: 0:03:05  lr: 0.000731  min_lr: 0.000731  loss: 3.1087 (3.1975)  weight_decay: 0.0500 (0.0500)  time: 0.7417  data: 0.1356  max mem: 41794
Epoch: [221]  [110/312]  eta: 0:02:49  lr: 0.000730  min_lr: 0.000730  loss: 3.4310 (3.2264)  weight_decay: 0.0500 (0.0500)  time: 0.6962  data: 0.1355  max mem: 41794
Epoch: [221]  [120/312]  eta: 0:02:41  lr: 0.000730  min_lr: 0.000730  loss: 3.4310 (3.2081)  weight_decay: 0.0500 (0.0500)  time: 0.6779  data: 0.0946  max mem: 41794
Epoch: [221]  [130/312]  eta: 0:02:33  lr: 0.000729  min_lr: 0.000729  loss: 2.9818 (3.1891)  weight_decay: 0.0500 (0.0500)  time: 0.8686  data: 0.1873  max mem: 41794
Epoch: [221]  [140/312]  eta: 0:02:21  lr: 0.000729  min_lr: 0.000729  loss: 2.9196 (3.1695)  weight_decay: 0.0500 (0.0500)  time: 0.6862  data: 0.0933  max mem: 41794
Epoch: [221]  [150/312]  eta: 0:02:14  lr: 0.000728  min_lr: 0.000728  loss: 3.0578 (3.1687)  weight_decay: 0.0500 (0.0500)  time: 0.7247  data: 0.1109  max mem: 41794
Epoch: [221]  [160/312]  eta: 0:02:04  lr: 0.000727  min_lr: 0.000727  loss: 3.1649 (3.1673)  weight_decay: 0.0500 (0.0500)  time: 0.7924  data: 0.1122  max mem: 41794
Epoch: [221]  [170/312]  eta: 0:01:54  lr: 0.000727  min_lr: 0.000727  loss: 3.0344 (3.1603)  weight_decay: 0.0500 (0.0500)  time: 0.6673  data: 0.0712  max mem: 41794
Epoch: [221]  [180/312]  eta: 0:01:47  lr: 0.000726  min_lr: 0.000726  loss: 3.1243 (3.1563)  weight_decay: 0.0500 (0.0500)  time: 0.8131  data: 0.1397  max mem: 41794
Epoch: [221]  [190/312]  eta: 0:01:37  lr: 0.000726  min_lr: 0.000726  loss: 3.1469 (3.1520)  weight_decay: 0.0500 (0.0500)  time: 0.7156  data: 0.0720  max mem: 41794
Epoch: [221]  [200/312]  eta: 0:01:30  lr: 0.000725  min_lr: 0.000725  loss: 3.2799 (3.1566)  weight_decay: 0.0500 (0.0500)  time: 0.7056  data: 0.0511  max mem: 41794
Epoch: [221]  [210/312]  eta: 0:01:22  lr: 0.000725  min_lr: 0.000725  loss: 3.2982 (3.1590)  weight_decay: 0.0500 (0.0500)  time: 0.8583  data: 0.0573  max mem: 41794
Epoch: [221]  [220/312]  eta: 0:01:13  lr: 0.000724  min_lr: 0.000724  loss: 3.2751 (3.1612)  weight_decay: 0.0500 (0.0500)  time: 0.7228  data: 0.0397  max mem: 41794
Epoch: [221]  [230/312]  eta: 0:01:05  lr: 0.000724  min_lr: 0.000724  loss: 3.2751 (3.1675)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.0320  max mem: 41794
Epoch: [221]  [240/312]  eta: 0:00:56  lr: 0.000723  min_lr: 0.000723  loss: 3.2678 (3.1625)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.0037  max mem: 41794
Epoch: [221]  [250/312]  eta: 0:00:48  lr: 0.000722  min_lr: 0.000722  loss: 3.2096 (3.1639)  weight_decay: 0.0500 (0.0500)  time: 0.6240  data: 0.0037  max mem: 41794
Epoch: [221]  [260/312]  eta: 0:00:40  lr: 0.000722  min_lr: 0.000722  loss: 3.2096 (3.1619)  weight_decay: 0.0500 (0.0500)  time: 0.7193  data: 0.0042  max mem: 41794
Epoch: [221]  [270/312]  eta: 0:00:32  lr: 0.000721  min_lr: 0.000721  loss: 3.2967 (3.1693)  weight_decay: 0.0500 (0.0500)  time: 0.6776  data: 0.0080  max mem: 41794
Epoch: [221]  [280/312]  eta: 0:00:24  lr: 0.000721  min_lr: 0.000721  loss: 3.4274 (3.1697)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.0053  max mem: 41794
Epoch: [221]  [290/312]  eta: 0:00:17  lr: 0.000720  min_lr: 0.000720  loss: 3.2225 (3.1600)  weight_decay: 0.0500 (0.0500)  time: 0.8361  data: 0.0235  max mem: 41794
Epoch: [221]  [300/312]  eta: 0:00:09  lr: 0.000720  min_lr: 0.000720  loss: 3.1142 (3.1591)  weight_decay: 0.0500 (0.0500)  time: 0.6218  data: 0.0224  max mem: 41794
Epoch: [221]  [310/312]  eta: 0:00:01  lr: 0.000719  min_lr: 0.000719  loss: 3.1090 (3.1551)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [221]  [311/312]  eta: 0:00:00  lr: 0.000719  min_lr: 0.000719  loss: 3.1090 (3.1566)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [221] Total time: 0:03:58 (0.7637 s / it)
Averaged stats: lr: 0.000719  min_lr: 0.000719  loss: 3.1090 (3.1722)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 0.9479 (0.9479)  acc1: 84.2448 (84.2448)  acc5: 96.2240 (96.2240)  time: 8.0104  data: 7.8044  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2820 (1.1727)  acc1: 77.6042 (77.5840)  acc5: 93.6198 (93.9680)  time: 1.0450  data: 0.8672  max mem: 41794
Test: Total time: 0:00:09 (1.0584 s / it)
* Acc@1 77.218 Acc@5 93.842 loss 1.179
Accuracy of the model on the 50000 test images: 77.2%
Max accuracy: 77.62%
Epoch: [222]  [  0/312]  eta: 1:19:58  lr: 0.000719  min_lr: 0.000719  loss: 3.2560 (3.2560)  weight_decay: 0.0500 (0.0500)  time: 15.3783  data: 13.3122  max mem: 41794
Epoch: [222]  [ 10/312]  eta: 0:11:21  lr: 0.000719  min_lr: 0.000719  loss: 3.2560 (3.1388)  weight_decay: 0.0500 (0.0500)  time: 2.2577  data: 1.2931  max mem: 41794
Epoch: [222]  [ 20/312]  eta: 0:07:30  lr: 0.000718  min_lr: 0.000718  loss: 3.1631 (3.1091)  weight_decay: 0.0500 (0.0500)  time: 0.8512  data: 0.1014  max mem: 41794
Epoch: [222]  [ 30/312]  eta: 0:05:43  lr: 0.000717  min_lr: 0.000717  loss: 3.2255 (3.1804)  weight_decay: 0.0500 (0.0500)  time: 0.6438  data: 0.0618  max mem: 41794
Epoch: [222]  [ 40/312]  eta: 0:04:44  lr: 0.000717  min_lr: 0.000717  loss: 3.4318 (3.1738)  weight_decay: 0.0500 (0.0500)  time: 0.5213  data: 0.0063  max mem: 41794
Epoch: [222]  [ 50/312]  eta: 0:04:20  lr: 0.000716  min_lr: 0.000716  loss: 3.1340 (3.1422)  weight_decay: 0.0500 (0.0500)  time: 0.6538  data: 0.0302  max mem: 41794
Epoch: [222]  [ 60/312]  eta: 0:03:49  lr: 0.000716  min_lr: 0.000716  loss: 3.1880 (3.1602)  weight_decay: 0.0500 (0.0500)  time: 0.6414  data: 0.0306  max mem: 41794
Epoch: [222]  [ 70/312]  eta: 0:03:39  lr: 0.000715  min_lr: 0.000715  loss: 3.3212 (3.1627)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.0782  max mem: 41794
Epoch: [222]  [ 80/312]  eta: 0:03:28  lr: 0.000715  min_lr: 0.000715  loss: 3.3336 (3.1743)  weight_decay: 0.0500 (0.0500)  time: 0.8549  data: 0.1052  max mem: 41794
Epoch: [222]  [ 90/312]  eta: 0:03:10  lr: 0.000714  min_lr: 0.000714  loss: 3.2250 (3.1788)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.0485  max mem: 41794
Epoch: [222]  [100/312]  eta: 0:03:04  lr: 0.000714  min_lr: 0.000714  loss: 3.3600 (3.1867)  weight_decay: 0.0500 (0.0500)  time: 0.7553  data: 0.0653  max mem: 41794
Epoch: [222]  [110/312]  eta: 0:02:49  lr: 0.000713  min_lr: 0.000713  loss: 3.3885 (3.1838)  weight_decay: 0.0500 (0.0500)  time: 0.7392  data: 0.0548  max mem: 41794
Epoch: [222]  [120/312]  eta: 0:02:40  lr: 0.000712  min_lr: 0.000712  loss: 3.2558 (3.1878)  weight_decay: 0.0500 (0.0500)  time: 0.6711  data: 0.0181  max mem: 41794
Epoch: [222]  [130/312]  eta: 0:02:32  lr: 0.000712  min_lr: 0.000712  loss: 3.2558 (3.1885)  weight_decay: 0.0500 (0.0500)  time: 0.8288  data: 0.0231  max mem: 41794
Epoch: [222]  [140/312]  eta: 0:02:20  lr: 0.000711  min_lr: 0.000711  loss: 3.0772 (3.1707)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0203  max mem: 41794
Epoch: [222]  [150/312]  eta: 0:02:12  lr: 0.000711  min_lr: 0.000711  loss: 3.0089 (3.1654)  weight_decay: 0.0500 (0.0500)  time: 0.7181  data: 0.0308  max mem: 41794
Epoch: [222]  [160/312]  eta: 0:02:05  lr: 0.000710  min_lr: 0.000710  loss: 3.3723 (3.1833)  weight_decay: 0.0500 (0.0500)  time: 0.8984  data: 0.0274  max mem: 41794
Epoch: [222]  [170/312]  eta: 0:01:54  lr: 0.000710  min_lr: 0.000710  loss: 3.4617 (3.1980)  weight_decay: 0.0500 (0.0500)  time: 0.7188  data: 0.0221  max mem: 41794
Epoch: [222]  [180/312]  eta: 0:01:46  lr: 0.000709  min_lr: 0.000709  loss: 3.4272 (3.2029)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.0206  max mem: 41794
Epoch: [222]  [190/312]  eta: 0:01:36  lr: 0.000709  min_lr: 0.000709  loss: 3.4030 (3.2031)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.0020  max mem: 41794
Epoch: [222]  [200/312]  eta: 0:01:29  lr: 0.000708  min_lr: 0.000708  loss: 3.3273 (3.2052)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.0187  max mem: 41794
Epoch: [222]  [210/312]  eta: 0:01:21  lr: 0.000708  min_lr: 0.000708  loss: 3.2970 (3.1945)  weight_decay: 0.0500 (0.0500)  time: 0.8634  data: 0.0411  max mem: 41794
Epoch: [222]  [220/312]  eta: 0:01:12  lr: 0.000707  min_lr: 0.000707  loss: 3.2445 (3.1951)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.0258  max mem: 41794
Epoch: [222]  [230/312]  eta: 0:01:04  lr: 0.000706  min_lr: 0.000706  loss: 3.1552 (3.1896)  weight_decay: 0.0500 (0.0500)  time: 0.6726  data: 0.0136  max mem: 41794
Epoch: [222]  [240/312]  eta: 0:00:56  lr: 0.000706  min_lr: 0.000706  loss: 3.0457 (3.1810)  weight_decay: 0.0500 (0.0500)  time: 0.8323  data: 0.0122  max mem: 41794
Epoch: [222]  [250/312]  eta: 0:00:48  lr: 0.000705  min_lr: 0.000705  loss: 3.0923 (3.1785)  weight_decay: 0.0500 (0.0500)  time: 0.6900  data: 0.0307  max mem: 41794
Epoch: [222]  [260/312]  eta: 0:00:40  lr: 0.000705  min_lr: 0.000705  loss: 3.2646 (3.1785)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.0471  max mem: 41794
Epoch: [222]  [270/312]  eta: 0:00:32  lr: 0.000704  min_lr: 0.000704  loss: 3.2155 (3.1783)  weight_decay: 0.0500 (0.0500)  time: 0.6790  data: 0.0171  max mem: 41794
Epoch: [222]  [280/312]  eta: 0:00:24  lr: 0.000704  min_lr: 0.000704  loss: 3.2560 (3.1799)  weight_decay: 0.0500 (0.0500)  time: 0.7001  data: 0.0242  max mem: 41794
Epoch: [222]  [290/312]  eta: 0:00:17  lr: 0.000703  min_lr: 0.000703  loss: 3.3706 (3.1794)  weight_decay: 0.0500 (0.0500)  time: 0.8943  data: 0.0284  max mem: 41794
Epoch: [222]  [300/312]  eta: 0:00:09  lr: 0.000703  min_lr: 0.000703  loss: 3.1281 (3.1732)  weight_decay: 0.0500 (0.0500)  time: 0.6778  data: 0.0046  max mem: 41794
Epoch: [222]  [310/312]  eta: 0:00:01  lr: 0.000702  min_lr: 0.000702  loss: 2.9575 (3.1659)  weight_decay: 0.0500 (0.0500)  time: 0.4642  data: 0.0001  max mem: 41794
Epoch: [222]  [311/312]  eta: 0:00:00  lr: 0.000702  min_lr: 0.000702  loss: 2.9970 (3.1656)  weight_decay: 0.0500 (0.0500)  time: 0.4643  data: 0.0001  max mem: 41794
Epoch: [222] Total time: 0:03:59 (0.7673 s / it)
Averaged stats: lr: 0.000702  min_lr: 0.000702  loss: 2.9970 (3.1921)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 0.9335 (0.9335)  acc1: 83.9844 (83.9844)  acc5: 96.0938 (96.0938)  time: 8.4181  data: 8.2094  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1558 (1.0787)  acc1: 77.9948 (77.5520)  acc5: 94.2708 (94.3200)  time: 1.0832  data: 0.9122  max mem: 41794
Test: Total time: 0:00:09 (1.0942 s / it)
* Acc@1 77.774 Acc@5 94.218 loss 1.075
Accuracy of the model on the 50000 test images: 77.8%
Max accuracy: 77.77%
Epoch: [223]  [  0/312]  eta: 1:15:11  lr: 0.000702  min_lr: 0.000702  loss: 2.8895 (2.8895)  weight_decay: 0.0500 (0.0500)  time: 14.4592  data: 11.8697  max mem: 41794
Epoch: [223]  [ 10/312]  eta: 0:10:09  lr: 0.000701  min_lr: 0.000701  loss: 2.8895 (2.9577)  weight_decay: 0.0500 (0.0500)  time: 2.0178  data: 1.2621  max mem: 41794
Epoch: [223]  [ 20/312]  eta: 0:07:29  lr: 0.000701  min_lr: 0.000701  loss: 3.2740 (3.0872)  weight_decay: 0.0500 (0.0500)  time: 0.8922  data: 0.1265  max mem: 41794
Epoch: [223]  [ 30/312]  eta: 0:05:49  lr: 0.000700  min_lr: 0.000700  loss: 3.1766 (3.0651)  weight_decay: 0.0500 (0.0500)  time: 0.8122  data: 0.0687  max mem: 41794
Epoch: [223]  [ 40/312]  eta: 0:05:06  lr: 0.000700  min_lr: 0.000700  loss: 3.2744 (3.1224)  weight_decay: 0.0500 (0.0500)  time: 0.6928  data: 0.0719  max mem: 41794
Epoch: [223]  [ 50/312]  eta: 0:04:26  lr: 0.000699  min_lr: 0.000699  loss: 3.2512 (3.0856)  weight_decay: 0.0500 (0.0500)  time: 0.6709  data: 0.0502  max mem: 41794
Epoch: [223]  [ 60/312]  eta: 0:03:58  lr: 0.000699  min_lr: 0.000699  loss: 3.0019 (3.0952)  weight_decay: 0.0500 (0.0500)  time: 0.5833  data: 0.0332  max mem: 41794
Epoch: [223]  [ 70/312]  eta: 0:03:48  lr: 0.000698  min_lr: 0.000698  loss: 3.1075 (3.0920)  weight_decay: 0.0500 (0.0500)  time: 0.7579  data: 0.1067  max mem: 41794
Epoch: [223]  [ 80/312]  eta: 0:03:29  lr: 0.000698  min_lr: 0.000698  loss: 3.1173 (3.0870)  weight_decay: 0.0500 (0.0500)  time: 0.7609  data: 0.1478  max mem: 41794
Epoch: [223]  [ 90/312]  eta: 0:03:14  lr: 0.000697  min_lr: 0.000697  loss: 3.1590 (3.1129)  weight_decay: 0.0500 (0.0500)  time: 0.6388  data: 0.1176  max mem: 41794
Epoch: [223]  [100/312]  eta: 0:03:05  lr: 0.000696  min_lr: 0.000696  loss: 3.2384 (3.1025)  weight_decay: 0.0500 (0.0500)  time: 0.7730  data: 0.1976  max mem: 41794
Epoch: [223]  [110/312]  eta: 0:02:50  lr: 0.000696  min_lr: 0.000696  loss: 3.2162 (3.1169)  weight_decay: 0.0500 (0.0500)  time: 0.6862  data: 0.1333  max mem: 41794
Epoch: [223]  [120/312]  eta: 0:02:42  lr: 0.000695  min_lr: 0.000695  loss: 3.2417 (3.1172)  weight_decay: 0.0500 (0.0500)  time: 0.6993  data: 0.1430  max mem: 41794
Epoch: [223]  [130/312]  eta: 0:02:33  lr: 0.000695  min_lr: 0.000695  loss: 3.2573 (3.1297)  weight_decay: 0.0500 (0.0500)  time: 0.8442  data: 0.2865  max mem: 41794
Epoch: [223]  [140/312]  eta: 0:02:20  lr: 0.000694  min_lr: 0.000694  loss: 3.4106 (3.1380)  weight_decay: 0.0500 (0.0500)  time: 0.6350  data: 0.1442  max mem: 41794
Epoch: [223]  [150/312]  eta: 0:02:12  lr: 0.000694  min_lr: 0.000694  loss: 3.1487 (3.1301)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.1871  max mem: 41794
Epoch: [223]  [160/312]  eta: 0:02:04  lr: 0.000693  min_lr: 0.000693  loss: 3.2649 (3.1484)  weight_decay: 0.0500 (0.0500)  time: 0.8602  data: 0.3620  max mem: 41794
Epoch: [223]  [170/312]  eta: 0:01:53  lr: 0.000693  min_lr: 0.000693  loss: 3.4238 (3.1563)  weight_decay: 0.0500 (0.0500)  time: 0.6660  data: 0.1755  max mem: 41794
Epoch: [223]  [180/312]  eta: 0:01:46  lr: 0.000692  min_lr: 0.000692  loss: 3.2918 (3.1593)  weight_decay: 0.0500 (0.0500)  time: 0.6794  data: 0.1912  max mem: 41794
Epoch: [223]  [190/312]  eta: 0:01:36  lr: 0.000692  min_lr: 0.000692  loss: 3.4382 (3.1810)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.1912  max mem: 41794
Epoch: [223]  [200/312]  eta: 0:01:28  lr: 0.000691  min_lr: 0.000691  loss: 3.4382 (3.1751)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.1818  max mem: 41794
Epoch: [223]  [210/312]  eta: 0:01:21  lr: 0.000690  min_lr: 0.000690  loss: 3.1209 (3.1638)  weight_decay: 0.0500 (0.0500)  time: 0.8674  data: 0.3671  max mem: 41794
Epoch: [223]  [220/312]  eta: 0:01:12  lr: 0.000690  min_lr: 0.000690  loss: 3.2473 (3.1672)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.1863  max mem: 41794
Epoch: [223]  [230/312]  eta: 0:01:04  lr: 0.000689  min_lr: 0.000689  loss: 3.3433 (3.1741)  weight_decay: 0.0500 (0.0500)  time: 0.6772  data: 0.1854  max mem: 41794
Epoch: [223]  [240/312]  eta: 0:00:56  lr: 0.000689  min_lr: 0.000689  loss: 3.3433 (3.1655)  weight_decay: 0.0500 (0.0500)  time: 0.8657  data: 0.3719  max mem: 41794
Epoch: [223]  [250/312]  eta: 0:00:48  lr: 0.000688  min_lr: 0.000688  loss: 3.1518 (3.1630)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1908  max mem: 41794
Epoch: [223]  [260/312]  eta: 0:00:40  lr: 0.000688  min_lr: 0.000688  loss: 3.3386 (3.1663)  weight_decay: 0.0500 (0.0500)  time: 0.6681  data: 0.1742  max mem: 41794
Epoch: [223]  [270/312]  eta: 0:00:32  lr: 0.000687  min_lr: 0.000687  loss: 3.3893 (3.1735)  weight_decay: 0.0500 (0.0500)  time: 0.6666  data: 0.1710  max mem: 41794
Epoch: [223]  [280/312]  eta: 0:00:24  lr: 0.000687  min_lr: 0.000687  loss: 3.2582 (3.1656)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.1565  max mem: 41794
Epoch: [223]  [290/312]  eta: 0:00:17  lr: 0.000686  min_lr: 0.000686  loss: 3.1904 (3.1673)  weight_decay: 0.0500 (0.0500)  time: 0.8341  data: 0.3190  max mem: 41794
Epoch: [223]  [300/312]  eta: 0:00:09  lr: 0.000686  min_lr: 0.000686  loss: 3.1341 (3.1482)  weight_decay: 0.0500 (0.0500)  time: 0.6571  data: 0.1628  max mem: 41794
Epoch: [223]  [310/312]  eta: 0:00:01  lr: 0.000685  min_lr: 0.000685  loss: 2.7473 (3.1496)  weight_decay: 0.0500 (0.0500)  time: 0.4679  data: 0.0001  max mem: 41794
Epoch: [223]  [311/312]  eta: 0:00:00  lr: 0.000685  min_lr: 0.000685  loss: 2.7473 (3.1485)  weight_decay: 0.0500 (0.0500)  time: 0.4679  data: 0.0001  max mem: 41794
Epoch: [223] Total time: 0:03:57 (0.7600 s / it)
Averaged stats: lr: 0.000685  min_lr: 0.000685  loss: 2.7473 (3.1773)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.8579 (0.8579)  acc1: 83.3333 (83.3333)  acc5: 96.2240 (96.2240)  time: 8.8582  data: 8.6575  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1679 (1.0631)  acc1: 78.7760 (77.7920)  acc5: 94.2708 (94.2720)  time: 1.1374  data: 0.9620  max mem: 41794
Test: Total time: 0:00:10 (1.1774 s / it)
* Acc@1 77.974 Acc@5 94.202 loss 1.067
Accuracy of the model on the 50000 test images: 78.0%
Max accuracy: 77.97%
Epoch: [224]  [  0/312]  eta: 1:21:36  lr: 0.000685  min_lr: 0.000685  loss: 3.4882 (3.4882)  weight_decay: 0.0500 (0.0500)  time: 15.6934  data: 15.2203  max mem: 41794
Epoch: [224]  [ 10/312]  eta: 0:11:04  lr: 0.000684  min_lr: 0.000684  loss: 3.2895 (3.1627)  weight_decay: 0.0500 (0.0500)  time: 2.2010  data: 1.4040  max mem: 41794
Epoch: [224]  [ 20/312]  eta: 0:07:13  lr: 0.000684  min_lr: 0.000684  loss: 3.2895 (3.2066)  weight_decay: 0.0500 (0.0500)  time: 0.7732  data: 0.0801  max mem: 41794
Epoch: [224]  [ 30/312]  eta: 0:05:30  lr: 0.000683  min_lr: 0.000683  loss: 3.3296 (3.1889)  weight_decay: 0.0500 (0.0500)  time: 0.6046  data: 0.0729  max mem: 41794
Epoch: [224]  [ 40/312]  eta: 0:04:41  lr: 0.000683  min_lr: 0.000683  loss: 3.3817 (3.2075)  weight_decay: 0.0500 (0.0500)  time: 0.5600  data: 0.0099  max mem: 41794
Epoch: [224]  [ 50/312]  eta: 0:04:20  lr: 0.000682  min_lr: 0.000682  loss: 3.4008 (3.2464)  weight_decay: 0.0500 (0.0500)  time: 0.7178  data: 0.0659  max mem: 41794
Epoch: [224]  [ 60/312]  eta: 0:03:49  lr: 0.000682  min_lr: 0.000682  loss: 3.1870 (3.2049)  weight_decay: 0.0500 (0.0500)  time: 0.6587  data: 0.0621  max mem: 41794
Epoch: [224]  [ 70/312]  eta: 0:03:38  lr: 0.000681  min_lr: 0.000681  loss: 2.9560 (3.2030)  weight_decay: 0.0500 (0.0500)  time: 0.6734  data: 0.1225  max mem: 41794
Epoch: [224]  [ 80/312]  eta: 0:03:28  lr: 0.000681  min_lr: 0.000681  loss: 3.1690 (3.1780)  weight_decay: 0.0500 (0.0500)  time: 0.8554  data: 0.2253  max mem: 41794
Epoch: [224]  [ 90/312]  eta: 0:03:11  lr: 0.000680  min_lr: 0.000680  loss: 3.0800 (3.1692)  weight_decay: 0.0500 (0.0500)  time: 0.7084  data: 0.1334  max mem: 41794
Epoch: [224]  [100/312]  eta: 0:03:02  lr: 0.000680  min_lr: 0.000680  loss: 2.9241 (3.1512)  weight_decay: 0.0500 (0.0500)  time: 0.7228  data: 0.1331  max mem: 41794
Epoch: [224]  [110/312]  eta: 0:02:47  lr: 0.000679  min_lr: 0.000679  loss: 2.9753 (3.1466)  weight_decay: 0.0500 (0.0500)  time: 0.6881  data: 0.1048  max mem: 41794
Epoch: [224]  [120/312]  eta: 0:02:39  lr: 0.000678  min_lr: 0.000678  loss: 2.8700 (3.1209)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.0945  max mem: 41794
Epoch: [224]  [130/312]  eta: 0:02:32  lr: 0.000678  min_lr: 0.000678  loss: 3.3402 (3.1345)  weight_decay: 0.0500 (0.0500)  time: 0.8757  data: 0.1810  max mem: 41794
Epoch: [224]  [140/312]  eta: 0:02:19  lr: 0.000677  min_lr: 0.000677  loss: 3.2886 (3.1226)  weight_decay: 0.0500 (0.0500)  time: 0.6851  data: 0.0884  max mem: 41794
Epoch: [224]  [150/312]  eta: 0:02:12  lr: 0.000677  min_lr: 0.000677  loss: 3.0537 (3.1101)  weight_decay: 0.0500 (0.0500)  time: 0.6773  data: 0.0797  max mem: 41794
Epoch: [224]  [160/312]  eta: 0:02:03  lr: 0.000676  min_lr: 0.000676  loss: 3.1998 (3.1229)  weight_decay: 0.0500 (0.0500)  time: 0.8355  data: 0.1480  max mem: 41794
Epoch: [224]  [170/312]  eta: 0:01:53  lr: 0.000676  min_lr: 0.000676  loss: 3.2529 (3.1195)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.0727  max mem: 41794
Epoch: [224]  [180/312]  eta: 0:01:46  lr: 0.000675  min_lr: 0.000675  loss: 3.1111 (3.1231)  weight_decay: 0.0500 (0.0500)  time: 0.7164  data: 0.0278  max mem: 41794
Epoch: [224]  [190/312]  eta: 0:01:36  lr: 0.000675  min_lr: 0.000675  loss: 3.4323 (3.1406)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0253  max mem: 41794
Epoch: [224]  [200/312]  eta: 0:01:28  lr: 0.000674  min_lr: 0.000674  loss: 3.4290 (3.1442)  weight_decay: 0.0500 (0.0500)  time: 0.6620  data: 0.0366  max mem: 41794
Epoch: [224]  [210/312]  eta: 0:01:20  lr: 0.000674  min_lr: 0.000674  loss: 3.0442 (3.1278)  weight_decay: 0.0500 (0.0500)  time: 0.8554  data: 0.0512  max mem: 41794
Epoch: [224]  [220/312]  eta: 0:01:11  lr: 0.000673  min_lr: 0.000673  loss: 2.9602 (3.1272)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.0153  max mem: 41794
Epoch: [224]  [230/312]  eta: 0:01:04  lr: 0.000673  min_lr: 0.000673  loss: 3.1750 (3.1330)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.0040  max mem: 41794
Epoch: [224]  [240/312]  eta: 0:00:56  lr: 0.000672  min_lr: 0.000672  loss: 3.2000 (3.1332)  weight_decay: 0.0500 (0.0500)  time: 0.8489  data: 0.0209  max mem: 41794
Epoch: [224]  [250/312]  eta: 0:00:48  lr: 0.000671  min_lr: 0.000671  loss: 2.9411 (3.1248)  weight_decay: 0.0500 (0.0500)  time: 0.6861  data: 0.0190  max mem: 41794
Epoch: [224]  [260/312]  eta: 0:00:40  lr: 0.000671  min_lr: 0.000671  loss: 3.0672 (3.1305)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.0022  max mem: 41794
Epoch: [224]  [270/312]  eta: 0:00:32  lr: 0.000670  min_lr: 0.000670  loss: 3.4729 (3.1376)  weight_decay: 0.0500 (0.0500)  time: 0.6685  data: 0.0006  max mem: 41794
Epoch: [224]  [280/312]  eta: 0:00:24  lr: 0.000670  min_lr: 0.000670  loss: 3.3104 (3.1362)  weight_decay: 0.0500 (0.0500)  time: 0.6874  data: 0.0046  max mem: 41794
Epoch: [224]  [290/312]  eta: 0:00:17  lr: 0.000669  min_lr: 0.000669  loss: 3.0339 (3.1287)  weight_decay: 0.0500 (0.0500)  time: 0.8521  data: 0.0044  max mem: 41794
Epoch: [224]  [300/312]  eta: 0:00:09  lr: 0.000669  min_lr: 0.000669  loss: 3.1954 (3.1338)  weight_decay: 0.0500 (0.0500)  time: 0.6538  data: 0.0002  max mem: 41794
Epoch: [224]  [310/312]  eta: 0:00:01  lr: 0.000668  min_lr: 0.000668  loss: 3.4097 (3.1396)  weight_decay: 0.0500 (0.0500)  time: 0.4694  data: 0.0001  max mem: 41794
Epoch: [224]  [311/312]  eta: 0:00:00  lr: 0.000668  min_lr: 0.000668  loss: 3.4097 (3.1411)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [224] Total time: 0:03:57 (0.7611 s / it)
Averaged stats: lr: 0.000668  min_lr: 0.000668  loss: 3.4097 (3.1586)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:07  loss: 1.0275 (1.0275)  acc1: 82.1615 (82.1615)  acc5: 95.8333 (95.8333)  time: 7.5125  data: 7.3017  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2162 (1.1832)  acc1: 76.9531 (77.7120)  acc5: 93.7500 (93.9360)  time: 1.0648  data: 0.8838  max mem: 41794
Test: Total time: 0:00:09 (1.0748 s / it)
* Acc@1 77.734 Acc@5 94.052 loss 1.194
Accuracy of the model on the 50000 test images: 77.7%
Max accuracy: 77.97%
Epoch: [225]  [  0/312]  eta: 1:24:52  lr: 0.000668  min_lr: 0.000668  loss: 3.0712 (3.0712)  weight_decay: 0.0500 (0.0500)  time: 16.3229  data: 12.2270  max mem: 41794
Epoch: [225]  [ 10/312]  eta: 0:11:23  lr: 0.000668  min_lr: 0.000668  loss: 3.2687 (3.2064)  weight_decay: 0.0500 (0.0500)  time: 2.2648  data: 1.2938  max mem: 41794
Epoch: [225]  [ 20/312]  eta: 0:07:38  lr: 0.000667  min_lr: 0.000667  loss: 3.2687 (3.1313)  weight_decay: 0.0500 (0.0500)  time: 0.8323  data: 0.1540  max mem: 41794
Epoch: [225]  [ 30/312]  eta: 0:05:48  lr: 0.000667  min_lr: 0.000667  loss: 3.0576 (3.0645)  weight_decay: 0.0500 (0.0500)  time: 0.6675  data: 0.0566  max mem: 41794
Epoch: [225]  [ 40/312]  eta: 0:04:51  lr: 0.000666  min_lr: 0.000666  loss: 3.3014 (3.1290)  weight_decay: 0.0500 (0.0500)  time: 0.5496  data: 0.0095  max mem: 41794
Epoch: [225]  [ 50/312]  eta: 0:04:28  lr: 0.000665  min_lr: 0.000665  loss: 3.3106 (3.1080)  weight_decay: 0.0500 (0.0500)  time: 0.7019  data: 0.0537  max mem: 41794
Epoch: [225]  [ 60/312]  eta: 0:03:56  lr: 0.000665  min_lr: 0.000665  loss: 3.1118 (3.0912)  weight_decay: 0.0500 (0.0500)  time: 0.6673  data: 0.0488  max mem: 41794
Epoch: [225]  [ 70/312]  eta: 0:03:42  lr: 0.000664  min_lr: 0.000664  loss: 3.3196 (3.1241)  weight_decay: 0.0500 (0.0500)  time: 0.6541  data: 0.0513  max mem: 41794
Epoch: [225]  [ 80/312]  eta: 0:03:29  lr: 0.000664  min_lr: 0.000664  loss: 3.2711 (3.0935)  weight_decay: 0.0500 (0.0500)  time: 0.7889  data: 0.0564  max mem: 41794
Epoch: [225]  [ 90/312]  eta: 0:03:15  lr: 0.000663  min_lr: 0.000663  loss: 3.2549 (3.1147)  weight_decay: 0.0500 (0.0500)  time: 0.7367  data: 0.1069  max mem: 41794
Epoch: [225]  [100/312]  eta: 0:03:05  lr: 0.000663  min_lr: 0.000663  loss: 3.4182 (3.1353)  weight_decay: 0.0500 (0.0500)  time: 0.7743  data: 0.2029  max mem: 41794
Epoch: [225]  [110/312]  eta: 0:02:50  lr: 0.000662  min_lr: 0.000662  loss: 3.0553 (3.1226)  weight_decay: 0.0500 (0.0500)  time: 0.6705  data: 0.1031  max mem: 41794
Epoch: [225]  [120/312]  eta: 0:02:41  lr: 0.000662  min_lr: 0.000662  loss: 3.0456 (3.1244)  weight_decay: 0.0500 (0.0500)  time: 0.6573  data: 0.1184  max mem: 41794
Epoch: [225]  [130/312]  eta: 0:02:32  lr: 0.000661  min_lr: 0.000661  loss: 3.2912 (3.1300)  weight_decay: 0.0500 (0.0500)  time: 0.8227  data: 0.2502  max mem: 41794
Epoch: [225]  [140/312]  eta: 0:02:20  lr: 0.000661  min_lr: 0.000661  loss: 3.3277 (3.1433)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.1324  max mem: 41794
Epoch: [225]  [150/312]  eta: 0:02:12  lr: 0.000660  min_lr: 0.000660  loss: 3.3918 (3.1567)  weight_decay: 0.0500 (0.0500)  time: 0.6667  data: 0.1386  max mem: 41794
Epoch: [225]  [160/312]  eta: 0:02:04  lr: 0.000660  min_lr: 0.000660  loss: 3.3918 (3.1555)  weight_decay: 0.0500 (0.0500)  time: 0.8292  data: 0.2447  max mem: 41794
Epoch: [225]  [170/312]  eta: 0:01:54  lr: 0.000659  min_lr: 0.000659  loss: 3.2280 (3.1524)  weight_decay: 0.0500 (0.0500)  time: 0.6987  data: 0.1341  max mem: 41794
Epoch: [225]  [180/312]  eta: 0:01:46  lr: 0.000659  min_lr: 0.000659  loss: 3.3268 (3.1693)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1419  max mem: 41794
Epoch: [225]  [190/312]  eta: 0:01:36  lr: 0.000658  min_lr: 0.000658  loss: 3.4569 (3.1770)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.1145  max mem: 41794
Epoch: [225]  [200/312]  eta: 0:01:29  lr: 0.000657  min_lr: 0.000657  loss: 3.3886 (3.1745)  weight_decay: 0.0500 (0.0500)  time: 0.7162  data: 0.1324  max mem: 41794
Epoch: [225]  [210/312]  eta: 0:01:22  lr: 0.000657  min_lr: 0.000657  loss: 3.3059 (3.1830)  weight_decay: 0.0500 (0.0500)  time: 0.9542  data: 0.2505  max mem: 41794
Epoch: [225]  [220/312]  eta: 0:01:12  lr: 0.000656  min_lr: 0.000656  loss: 3.3623 (3.1861)  weight_decay: 0.0500 (0.0500)  time: 0.7381  data: 0.1192  max mem: 41794
Epoch: [225]  [230/312]  eta: 0:01:05  lr: 0.000656  min_lr: 0.000656  loss: 3.2460 (3.1815)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.0751  max mem: 41794
Epoch: [225]  [240/312]  eta: 0:00:56  lr: 0.000655  min_lr: 0.000655  loss: 3.0974 (3.1774)  weight_decay: 0.0500 (0.0500)  time: 0.7808  data: 0.0747  max mem: 41794
Epoch: [225]  [250/312]  eta: 0:00:48  lr: 0.000655  min_lr: 0.000655  loss: 3.1970 (3.1805)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.0791  max mem: 41794
Epoch: [225]  [260/312]  eta: 0:00:40  lr: 0.000654  min_lr: 0.000654  loss: 3.3071 (3.1811)  weight_decay: 0.0500 (0.0500)  time: 0.7335  data: 0.1229  max mem: 41794
Epoch: [225]  [270/312]  eta: 0:00:32  lr: 0.000654  min_lr: 0.000654  loss: 3.3884 (3.1780)  weight_decay: 0.0500 (0.0500)  time: 0.6588  data: 0.0444  max mem: 41794
Epoch: [225]  [280/312]  eta: 0:00:25  lr: 0.000653  min_lr: 0.000653  loss: 3.3668 (3.1785)  weight_decay: 0.0500 (0.0500)  time: 0.7385  data: 0.0681  max mem: 41794
Epoch: [225]  [290/312]  eta: 0:00:17  lr: 0.000653  min_lr: 0.000653  loss: 3.2739 (3.1734)  weight_decay: 0.0500 (0.0500)  time: 0.8460  data: 0.0779  max mem: 41794
Epoch: [225]  [300/312]  eta: 0:00:09  lr: 0.000652  min_lr: 0.000652  loss: 3.0913 (3.1703)  weight_decay: 0.0500 (0.0500)  time: 0.5937  data: 0.0102  max mem: 41794
Epoch: [225]  [310/312]  eta: 0:00:01  lr: 0.000652  min_lr: 0.000652  loss: 3.2483 (3.1715)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [225]  [311/312]  eta: 0:00:00  lr: 0.000652  min_lr: 0.000652  loss: 3.2483 (3.1704)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [225] Total time: 0:03:59 (0.7663 s / it)
Averaged stats: lr: 0.000652  min_lr: 0.000652  loss: 3.2483 (3.1737)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 0.8603 (0.8603)  acc1: 83.4635 (83.4635)  acc5: 95.7031 (95.7031)  time: 8.7102  data: 8.4972  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1821 (1.0800)  acc1: 76.8229 (77.3440)  acc5: 94.0104 (94.0960)  time: 1.1159  data: 0.9442  max mem: 41794
Test: Total time: 0:00:10 (1.1351 s / it)
* Acc@1 77.712 Acc@5 94.104 loss 1.081
Accuracy of the model on the 50000 test images: 77.7%
Max accuracy: 77.97%
Epoch: [226]  [  0/312]  eta: 1:26:28  lr: 0.000651  min_lr: 0.000651  loss: 3.2794 (3.2794)  weight_decay: 0.0500 (0.0500)  time: 16.6288  data: 13.8482  max mem: 41794
Epoch: [226]  [ 10/312]  eta: 0:11:32  lr: 0.000651  min_lr: 0.000651  loss: 3.2957 (3.2451)  weight_decay: 0.0500 (0.0500)  time: 2.2934  data: 1.3916  max mem: 41794
Epoch: [226]  [ 20/312]  eta: 0:07:39  lr: 0.000650  min_lr: 0.000650  loss: 3.2957 (3.2856)  weight_decay: 0.0500 (0.0500)  time: 0.8200  data: 0.0907  max mem: 41794
Epoch: [226]  [ 30/312]  eta: 0:05:48  lr: 0.000650  min_lr: 0.000650  loss: 3.4152 (3.2970)  weight_decay: 0.0500 (0.0500)  time: 0.6553  data: 0.0181  max mem: 41794
Epoch: [226]  [ 40/312]  eta: 0:04:48  lr: 0.000649  min_lr: 0.000649  loss: 3.2766 (3.2201)  weight_decay: 0.0500 (0.0500)  time: 0.5218  data: 0.0023  max mem: 41794
Epoch: [226]  [ 50/312]  eta: 0:04:23  lr: 0.000649  min_lr: 0.000649  loss: 3.2100 (3.2261)  weight_decay: 0.0500 (0.0500)  time: 0.6486  data: 0.0585  max mem: 41794
Epoch: [226]  [ 60/312]  eta: 0:03:54  lr: 0.000648  min_lr: 0.000648  loss: 3.4221 (3.2330)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.0804  max mem: 41794
Epoch: [226]  [ 70/312]  eta: 0:03:46  lr: 0.000648  min_lr: 0.000648  loss: 3.1946 (3.1975)  weight_decay: 0.0500 (0.0500)  time: 0.7607  data: 0.1097  max mem: 41794
Epoch: [226]  [ 80/312]  eta: 0:03:28  lr: 0.000647  min_lr: 0.000647  loss: 3.1756 (3.1982)  weight_decay: 0.0500 (0.0500)  time: 0.8076  data: 0.1123  max mem: 41794
Epoch: [226]  [ 90/312]  eta: 0:03:16  lr: 0.000647  min_lr: 0.000647  loss: 3.4070 (3.2222)  weight_decay: 0.0500 (0.0500)  time: 0.7063  data: 0.0526  max mem: 41794
Epoch: [226]  [100/312]  eta: 0:03:07  lr: 0.000646  min_lr: 0.000646  loss: 3.4070 (3.2178)  weight_decay: 0.0500 (0.0500)  time: 0.8313  data: 0.0275  max mem: 41794
Epoch: [226]  [110/312]  eta: 0:02:53  lr: 0.000646  min_lr: 0.000646  loss: 3.3239 (3.2101)  weight_decay: 0.0500 (0.0500)  time: 0.7218  data: 0.0328  max mem: 41794
Epoch: [226]  [120/312]  eta: 0:02:43  lr: 0.000645  min_lr: 0.000645  loss: 3.2581 (3.2028)  weight_decay: 0.0500 (0.0500)  time: 0.6930  data: 0.0419  max mem: 41794
Epoch: [226]  [130/312]  eta: 0:02:34  lr: 0.000645  min_lr: 0.000645  loss: 3.3518 (3.1999)  weight_decay: 0.0500 (0.0500)  time: 0.8012  data: 0.0110  max mem: 41794
Epoch: [226]  [140/312]  eta: 0:02:21  lr: 0.000644  min_lr: 0.000644  loss: 3.2776 (3.1984)  weight_decay: 0.0500 (0.0500)  time: 0.6387  data: 0.0018  max mem: 41794
Epoch: [226]  [150/312]  eta: 0:02:14  lr: 0.000644  min_lr: 0.000644  loss: 3.2776 (3.2085)  weight_decay: 0.0500 (0.0500)  time: 0.7196  data: 0.0249  max mem: 41794
Epoch: [226]  [160/312]  eta: 0:02:04  lr: 0.000643  min_lr: 0.000643  loss: 3.3528 (3.2133)  weight_decay: 0.0500 (0.0500)  time: 0.7902  data: 0.0255  max mem: 41794
Epoch: [226]  [170/312]  eta: 0:01:55  lr: 0.000642  min_lr: 0.000642  loss: 3.3445 (3.2118)  weight_decay: 0.0500 (0.0500)  time: 0.6537  data: 0.0116  max mem: 41794
Epoch: [226]  [180/312]  eta: 0:01:47  lr: 0.000642  min_lr: 0.000642  loss: 3.2133 (3.2115)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.0497  max mem: 41794
Epoch: [226]  [190/312]  eta: 0:01:37  lr: 0.000641  min_lr: 0.000641  loss: 3.4321 (3.2127)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.0425  max mem: 41794
Epoch: [226]  [200/312]  eta: 0:01:29  lr: 0.000641  min_lr: 0.000641  loss: 3.3894 (3.2131)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.0402  max mem: 41794
Epoch: [226]  [210/312]  eta: 0:01:22  lr: 0.000640  min_lr: 0.000640  loss: 3.3715 (3.2142)  weight_decay: 0.0500 (0.0500)  time: 0.8942  data: 0.0832  max mem: 41794
Epoch: [226]  [220/312]  eta: 0:01:12  lr: 0.000640  min_lr: 0.000640  loss: 3.3844 (3.2161)  weight_decay: 0.0500 (0.0500)  time: 0.7295  data: 0.0494  max mem: 41794
Epoch: [226]  [230/312]  eta: 0:01:05  lr: 0.000639  min_lr: 0.000639  loss: 3.3146 (3.2110)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.0620  max mem: 41794
Epoch: [226]  [240/312]  eta: 0:00:56  lr: 0.000639  min_lr: 0.000639  loss: 2.9231 (3.2002)  weight_decay: 0.0500 (0.0500)  time: 0.7744  data: 0.0622  max mem: 41794
Epoch: [226]  [250/312]  eta: 0:00:48  lr: 0.000638  min_lr: 0.000638  loss: 3.1505 (3.1977)  weight_decay: 0.0500 (0.0500)  time: 0.6768  data: 0.0209  max mem: 41794
Epoch: [226]  [260/312]  eta: 0:00:41  lr: 0.000638  min_lr: 0.000638  loss: 3.2472 (3.1917)  weight_decay: 0.0500 (0.0500)  time: 0.7862  data: 0.0553  max mem: 41794
Epoch: [226]  [270/312]  eta: 0:00:32  lr: 0.000637  min_lr: 0.000637  loss: 3.1489 (3.1890)  weight_decay: 0.0500 (0.0500)  time: 0.6869  data: 0.0409  max mem: 41794
Epoch: [226]  [280/312]  eta: 0:00:25  lr: 0.000637  min_lr: 0.000637  loss: 3.0197 (3.1765)  weight_decay: 0.0500 (0.0500)  time: 0.7095  data: 0.0590  max mem: 41794
Epoch: [226]  [290/312]  eta: 0:00:17  lr: 0.000636  min_lr: 0.000636  loss: 2.5180 (3.1674)  weight_decay: 0.0500 (0.0500)  time: 0.8167  data: 0.0943  max mem: 41794
Epoch: [226]  [300/312]  eta: 0:00:09  lr: 0.000636  min_lr: 0.000636  loss: 3.1090 (3.1697)  weight_decay: 0.0500 (0.0500)  time: 0.5979  data: 0.0376  max mem: 41794
Epoch: [226]  [310/312]  eta: 0:00:01  lr: 0.000635  min_lr: 0.000635  loss: 3.1092 (3.1656)  weight_decay: 0.0500 (0.0500)  time: 0.4694  data: 0.0001  max mem: 41794
Epoch: [226]  [311/312]  eta: 0:00:00  lr: 0.000635  min_lr: 0.000635  loss: 3.1092 (3.1663)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0001  max mem: 41794
Epoch: [226] Total time: 0:03:59 (0.7688 s / it)
Averaged stats: lr: 0.000635  min_lr: 0.000635  loss: 3.1092 (3.1647)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.9652 (0.9652)  acc1: 83.0729 (83.0729)  acc5: 96.4844 (96.4844)  time: 8.6497  data: 8.4427  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2266 (1.1459)  acc1: 76.3021 (77.4720)  acc5: 94.4010 (94.1760)  time: 1.1097  data: 0.9381  max mem: 41794
Test: Total time: 0:00:10 (1.1214 s / it)
* Acc@1 77.556 Acc@5 94.108 loss 1.158
Accuracy of the model on the 50000 test images: 77.6%
Max accuracy: 77.97%
Epoch: [227]  [  0/312]  eta: 1:23:41  lr: 0.000635  min_lr: 0.000635  loss: 3.7929 (3.7929)  weight_decay: 0.0500 (0.0500)  time: 16.0944  data: 13.6211  max mem: 41794
Epoch: [227]  [ 10/312]  eta: 0:10:51  lr: 0.000634  min_lr: 0.000634  loss: 3.3401 (3.2846)  weight_decay: 0.0500 (0.0500)  time: 2.1573  data: 1.2390  max mem: 41794
Epoch: [227]  [ 20/312]  eta: 0:07:21  lr: 0.000634  min_lr: 0.000634  loss: 2.9833 (3.1083)  weight_decay: 0.0500 (0.0500)  time: 0.7817  data: 0.0430  max mem: 41794
Epoch: [227]  [ 30/312]  eta: 0:05:37  lr: 0.000633  min_lr: 0.000633  loss: 3.2920 (3.1804)  weight_decay: 0.0500 (0.0500)  time: 0.6704  data: 0.0431  max mem: 41794
Epoch: [227]  [ 40/312]  eta: 0:04:59  lr: 0.000633  min_lr: 0.000633  loss: 3.3564 (3.2062)  weight_decay: 0.0500 (0.0500)  time: 0.6743  data: 0.0449  max mem: 41794
Epoch: [227]  [ 50/312]  eta: 0:04:40  lr: 0.000632  min_lr: 0.000632  loss: 3.3129 (3.1780)  weight_decay: 0.0500 (0.0500)  time: 0.8717  data: 0.0967  max mem: 41794
Epoch: [227]  [ 60/312]  eta: 0:04:07  lr: 0.000632  min_lr: 0.000632  loss: 3.0179 (3.1352)  weight_decay: 0.0500 (0.0500)  time: 0.7371  data: 0.0647  max mem: 41794
Epoch: [227]  [ 70/312]  eta: 0:03:55  lr: 0.000631  min_lr: 0.000631  loss: 3.0562 (3.1408)  weight_decay: 0.0500 (0.0500)  time: 0.7266  data: 0.0571  max mem: 41794
Epoch: [227]  [ 80/312]  eta: 0:03:40  lr: 0.000631  min_lr: 0.000631  loss: 3.3647 (3.1388)  weight_decay: 0.0500 (0.0500)  time: 0.8463  data: 0.0492  max mem: 41794
Epoch: [227]  [ 90/312]  eta: 0:03:20  lr: 0.000630  min_lr: 0.000630  loss: 3.4530 (3.1641)  weight_decay: 0.0500 (0.0500)  time: 0.6485  data: 0.0048  max mem: 41794
Epoch: [227]  [100/312]  eta: 0:03:10  lr: 0.000630  min_lr: 0.000630  loss: 3.1544 (3.1487)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.0212  max mem: 41794
Epoch: [227]  [110/312]  eta: 0:02:54  lr: 0.000629  min_lr: 0.000629  loss: 3.0488 (3.1339)  weight_decay: 0.0500 (0.0500)  time: 0.6875  data: 0.0279  max mem: 41794
Epoch: [227]  [120/312]  eta: 0:02:45  lr: 0.000629  min_lr: 0.000629  loss: 3.2862 (3.1548)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.0111  max mem: 41794
Epoch: [227]  [130/312]  eta: 0:02:37  lr: 0.000628  min_lr: 0.000628  loss: 3.3986 (3.1745)  weight_decay: 0.0500 (0.0500)  time: 0.8633  data: 0.0127  max mem: 41794
Epoch: [227]  [140/312]  eta: 0:02:24  lr: 0.000628  min_lr: 0.000628  loss: 3.4087 (3.1906)  weight_decay: 0.0500 (0.0500)  time: 0.7224  data: 0.0227  max mem: 41794
Epoch: [227]  [150/312]  eta: 0:02:14  lr: 0.000627  min_lr: 0.000627  loss: 3.4088 (3.1953)  weight_decay: 0.0500 (0.0500)  time: 0.6320  data: 0.0208  max mem: 41794
Epoch: [227]  [160/312]  eta: 0:02:06  lr: 0.000627  min_lr: 0.000627  loss: 3.2451 (3.1891)  weight_decay: 0.0500 (0.0500)  time: 0.7732  data: 0.0094  max mem: 41794
Epoch: [227]  [170/312]  eta: 0:01:55  lr: 0.000626  min_lr: 0.000626  loss: 3.1104 (3.1789)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.0046  max mem: 41794
Epoch: [227]  [180/312]  eta: 0:01:48  lr: 0.000626  min_lr: 0.000626  loss: 3.1104 (3.1710)  weight_decay: 0.0500 (0.0500)  time: 0.7423  data: 0.0245  max mem: 41794
Epoch: [227]  [190/312]  eta: 0:01:38  lr: 0.000625  min_lr: 0.000625  loss: 3.1933 (3.1765)  weight_decay: 0.0500 (0.0500)  time: 0.7153  data: 0.0249  max mem: 41794
Epoch: [227]  [200/312]  eta: 0:01:30  lr: 0.000625  min_lr: 0.000625  loss: 3.3594 (3.1838)  weight_decay: 0.0500 (0.0500)  time: 0.6615  data: 0.0144  max mem: 41794
Epoch: [227]  [210/312]  eta: 0:01:22  lr: 0.000624  min_lr: 0.000624  loss: 3.3491 (3.1855)  weight_decay: 0.0500 (0.0500)  time: 0.8232  data: 0.0162  max mem: 41794
Epoch: [227]  [220/312]  eta: 0:01:13  lr: 0.000624  min_lr: 0.000624  loss: 3.3826 (3.1927)  weight_decay: 0.0500 (0.0500)  time: 0.6930  data: 0.0333  max mem: 41794
Epoch: [227]  [230/312]  eta: 0:01:05  lr: 0.000623  min_lr: 0.000623  loss: 3.4925 (3.1977)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.0395  max mem: 41794
Epoch: [227]  [240/312]  eta: 0:00:57  lr: 0.000622  min_lr: 0.000622  loss: 3.2378 (3.1887)  weight_decay: 0.0500 (0.0500)  time: 0.8053  data: 0.0125  max mem: 41794
Epoch: [227]  [250/312]  eta: 0:00:48  lr: 0.000622  min_lr: 0.000622  loss: 3.0810 (3.1796)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.0329  max mem: 41794
Epoch: [227]  [260/312]  eta: 0:00:41  lr: 0.000621  min_lr: 0.000621  loss: 3.1735 (3.1771)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.0397  max mem: 41794
Epoch: [227]  [270/312]  eta: 0:00:32  lr: 0.000621  min_lr: 0.000621  loss: 3.0205 (3.1689)  weight_decay: 0.0500 (0.0500)  time: 0.6459  data: 0.0095  max mem: 41794
Epoch: [227]  [280/312]  eta: 0:00:25  lr: 0.000620  min_lr: 0.000620  loss: 3.0205 (3.1654)  weight_decay: 0.0500 (0.0500)  time: 0.7255  data: 0.0309  max mem: 41794
Epoch: [227]  [290/312]  eta: 0:00:17  lr: 0.000620  min_lr: 0.000620  loss: 3.2070 (3.1630)  weight_decay: 0.0500 (0.0500)  time: 0.7941  data: 0.0306  max mem: 41794
Epoch: [227]  [300/312]  eta: 0:00:09  lr: 0.000619  min_lr: 0.000619  loss: 3.2687 (3.1693)  weight_decay: 0.0500 (0.0500)  time: 0.5546  data: 0.0002  max mem: 41794
Epoch: [227]  [310/312]  eta: 0:00:01  lr: 0.000619  min_lr: 0.000619  loss: 3.3706 (3.1663)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [227]  [311/312]  eta: 0:00:00  lr: 0.000619  min_lr: 0.000619  loss: 3.3706 (3.1658)  weight_decay: 0.0500 (0.0500)  time: 0.4638  data: 0.0001  max mem: 41794
Epoch: [227] Total time: 0:03:59 (0.7661 s / it)
Averaged stats: lr: 0.000619  min_lr: 0.000619  loss: 3.3706 (3.1513)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.9852 (0.9852)  acc1: 83.8542 (83.8542)  acc5: 96.2240 (96.2240)  time: 8.6548  data: 8.4428  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2330 (1.1557)  acc1: 77.3438 (78.0800)  acc5: 94.4010 (94.3200)  time: 1.1097  data: 0.9382  max mem: 41794
Test: Total time: 0:00:10 (1.1343 s / it)
* Acc@1 77.966 Acc@5 94.258 loss 1.163
Accuracy of the model on the 50000 test images: 78.0%
Max accuracy: 77.97%
Epoch: [228]  [  0/312]  eta: 1:13:06  lr: 0.000619  min_lr: 0.000619  loss: 3.4908 (3.4908)  weight_decay: 0.0500 (0.0500)  time: 14.0589  data: 12.9620  max mem: 41794
Epoch: [228]  [ 10/312]  eta: 0:11:01  lr: 0.000618  min_lr: 0.000618  loss: 2.9799 (3.0334)  weight_decay: 0.0500 (0.0500)  time: 2.1904  data: 1.2982  max mem: 41794
Epoch: [228]  [ 20/312]  eta: 0:07:12  lr: 0.000618  min_lr: 0.000618  loss: 3.2272 (3.1265)  weight_decay: 0.0500 (0.0500)  time: 0.8510  data: 0.1370  max mem: 41794
Epoch: [228]  [ 30/312]  eta: 0:05:30  lr: 0.000617  min_lr: 0.000617  loss: 3.2410 (3.1312)  weight_decay: 0.0500 (0.0500)  time: 0.6091  data: 0.0786  max mem: 41794
Epoch: [228]  [ 40/312]  eta: 0:04:44  lr: 0.000617  min_lr: 0.000617  loss: 3.1332 (3.1354)  weight_decay: 0.0500 (0.0500)  time: 0.5929  data: 0.0895  max mem: 41794
Epoch: [228]  [ 50/312]  eta: 0:04:24  lr: 0.000616  min_lr: 0.000616  loss: 3.2113 (3.1615)  weight_decay: 0.0500 (0.0500)  time: 0.7641  data: 0.2188  max mem: 41794
Epoch: [228]  [ 60/312]  eta: 0:03:53  lr: 0.000616  min_lr: 0.000616  loss: 3.2113 (3.1128)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.1371  max mem: 41794
Epoch: [228]  [ 70/312]  eta: 0:03:45  lr: 0.000615  min_lr: 0.000615  loss: 2.9258 (3.1140)  weight_decay: 0.0500 (0.0500)  time: 0.7336  data: 0.1440  max mem: 41794
Epoch: [228]  [ 80/312]  eta: 0:03:36  lr: 0.000615  min_lr: 0.000615  loss: 3.1413 (3.1147)  weight_decay: 0.0500 (0.0500)  time: 0.9494  data: 0.2620  max mem: 41794
Epoch: [228]  [ 90/312]  eta: 0:03:17  lr: 0.000614  min_lr: 0.000614  loss: 3.1456 (3.1185)  weight_decay: 0.0500 (0.0500)  time: 0.7348  data: 0.1187  max mem: 41794
Epoch: [228]  [100/312]  eta: 0:03:08  lr: 0.000614  min_lr: 0.000614  loss: 3.2585 (3.1302)  weight_decay: 0.0500 (0.0500)  time: 0.7176  data: 0.0867  max mem: 41794
Epoch: [228]  [110/312]  eta: 0:02:52  lr: 0.000613  min_lr: 0.000613  loss: 3.4085 (3.1617)  weight_decay: 0.0500 (0.0500)  time: 0.6900  data: 0.0867  max mem: 41794
Epoch: [228]  [120/312]  eta: 0:02:42  lr: 0.000612  min_lr: 0.000612  loss: 3.3605 (3.1478)  weight_decay: 0.0500 (0.0500)  time: 0.6424  data: 0.0777  max mem: 41794
Epoch: [228]  [130/312]  eta: 0:02:36  lr: 0.000612  min_lr: 0.000612  loss: 3.2751 (3.1490)  weight_decay: 0.0500 (0.0500)  time: 0.8878  data: 0.1816  max mem: 41794
Epoch: [228]  [140/312]  eta: 0:02:23  lr: 0.000611  min_lr: 0.000611  loss: 3.3768 (3.1607)  weight_decay: 0.0500 (0.0500)  time: 0.7413  data: 0.1047  max mem: 41794
Epoch: [228]  [150/312]  eta: 0:02:14  lr: 0.000611  min_lr: 0.000611  loss: 3.3654 (3.1463)  weight_decay: 0.0500 (0.0500)  time: 0.6426  data: 0.0639  max mem: 41794
Epoch: [228]  [160/312]  eta: 0:02:04  lr: 0.000610  min_lr: 0.000610  loss: 3.0505 (3.1406)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.1444  max mem: 41794
Epoch: [228]  [170/312]  eta: 0:01:55  lr: 0.000610  min_lr: 0.000610  loss: 3.3118 (3.1440)  weight_decay: 0.0500 (0.0500)  time: 0.6592  data: 0.0813  max mem: 41794
Epoch: [228]  [180/312]  eta: 0:01:47  lr: 0.000609  min_lr: 0.000609  loss: 3.5170 (3.1568)  weight_decay: 0.0500 (0.0500)  time: 0.7503  data: 0.1161  max mem: 41794
Epoch: [228]  [190/312]  eta: 0:01:36  lr: 0.000609  min_lr: 0.000609  loss: 3.0778 (3.1349)  weight_decay: 0.0500 (0.0500)  time: 0.6578  data: 0.1160  max mem: 41794
Epoch: [228]  [200/312]  eta: 0:01:29  lr: 0.000608  min_lr: 0.000608  loss: 2.8777 (3.1267)  weight_decay: 0.0500 (0.0500)  time: 0.6616  data: 0.1160  max mem: 41794
Epoch: [228]  [210/312]  eta: 0:01:21  lr: 0.000608  min_lr: 0.000608  loss: 2.9867 (3.1187)  weight_decay: 0.0500 (0.0500)  time: 0.8177  data: 0.2377  max mem: 41794
Epoch: [228]  [220/312]  eta: 0:01:11  lr: 0.000607  min_lr: 0.000607  loss: 3.2090 (3.1125)  weight_decay: 0.0500 (0.0500)  time: 0.6451  data: 0.1224  max mem: 41794
Epoch: [228]  [230/312]  eta: 0:01:04  lr: 0.000607  min_lr: 0.000607  loss: 3.2500 (3.1124)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.1655  max mem: 41794
Epoch: [228]  [240/312]  eta: 0:00:56  lr: 0.000606  min_lr: 0.000606  loss: 3.1178 (3.1069)  weight_decay: 0.0500 (0.0500)  time: 0.8759  data: 0.3379  max mem: 41794
Epoch: [228]  [250/312]  eta: 0:00:48  lr: 0.000606  min_lr: 0.000606  loss: 3.1993 (3.1105)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1730  max mem: 41794
Epoch: [228]  [260/312]  eta: 0:00:40  lr: 0.000605  min_lr: 0.000605  loss: 3.2239 (3.1166)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.1425  max mem: 41794
Epoch: [228]  [270/312]  eta: 0:00:32  lr: 0.000605  min_lr: 0.000605  loss: 3.3213 (3.1188)  weight_decay: 0.0500 (0.0500)  time: 0.7007  data: 0.1425  max mem: 41794
Epoch: [228]  [280/312]  eta: 0:00:24  lr: 0.000604  min_lr: 0.000604  loss: 3.3221 (3.1273)  weight_decay: 0.0500 (0.0500)  time: 0.7157  data: 0.1164  max mem: 41794
Epoch: [228]  [290/312]  eta: 0:00:17  lr: 0.000604  min_lr: 0.000604  loss: 3.1809 (3.1155)  weight_decay: 0.0500 (0.0500)  time: 0.8296  data: 0.2230  max mem: 41794
Epoch: [228]  [300/312]  eta: 0:00:09  lr: 0.000603  min_lr: 0.000603  loss: 3.0338 (3.1188)  weight_decay: 0.0500 (0.0500)  time: 0.6002  data: 0.1069  max mem: 41794
Epoch: [228]  [310/312]  eta: 0:00:01  lr: 0.000603  min_lr: 0.000603  loss: 3.3911 (3.1308)  weight_decay: 0.0500 (0.0500)  time: 0.4663  data: 0.0001  max mem: 41794
Epoch: [228]  [311/312]  eta: 0:00:00  lr: 0.000603  min_lr: 0.000603  loss: 3.3492 (3.1311)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [228] Total time: 0:03:57 (0.7628 s / it)
Averaged stats: lr: 0.000603  min_lr: 0.000603  loss: 3.3492 (3.1416)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 1.0772 (1.0772)  acc1: 83.8542 (83.8542)  acc5: 96.0938 (96.0938)  time: 8.0269  data: 7.8252  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.3977 (1.3051)  acc1: 77.4740 (77.5200)  acc5: 94.1406 (93.9360)  time: 1.0776  data: 0.9064  max mem: 41794
Test: Total time: 0:00:09 (1.0882 s / it)
* Acc@1 77.572 Acc@5 94.092 loss 1.306
Accuracy of the model on the 50000 test images: 77.6%
Max accuracy: 77.97%
Epoch: [229]  [  0/312]  eta: 1:22:09  lr: 0.000603  min_lr: 0.000603  loss: 2.7427 (2.7427)  weight_decay: 0.0500 (0.0500)  time: 15.7987  data: 15.3215  max mem: 41794
Epoch: [229]  [ 10/312]  eta: 0:11:16  lr: 0.000602  min_lr: 0.000602  loss: 3.1765 (3.1695)  weight_decay: 0.0500 (0.0500)  time: 2.2400  data: 1.3971  max mem: 41794
Epoch: [229]  [ 20/312]  eta: 0:07:16  lr: 0.000602  min_lr: 0.000602  loss: 3.3179 (3.1546)  weight_decay: 0.0500 (0.0500)  time: 0.7798  data: 0.0552  max mem: 41794
Epoch: [229]  [ 30/312]  eta: 0:05:31  lr: 0.000601  min_lr: 0.000601  loss: 3.3478 (3.1814)  weight_decay: 0.0500 (0.0500)  time: 0.5926  data: 0.0597  max mem: 41794
Epoch: [229]  [ 40/312]  eta: 0:04:43  lr: 0.000601  min_lr: 0.000601  loss: 3.3801 (3.1913)  weight_decay: 0.0500 (0.0500)  time: 0.5641  data: 0.0727  max mem: 41794
Epoch: [229]  [ 50/312]  eta: 0:04:19  lr: 0.000600  min_lr: 0.000600  loss: 3.4050 (3.1785)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.2008  max mem: 41794
Epoch: [229]  [ 60/312]  eta: 0:03:48  lr: 0.000600  min_lr: 0.000600  loss: 3.3028 (3.1552)  weight_decay: 0.0500 (0.0500)  time: 0.6354  data: 0.1379  max mem: 41794
Epoch: [229]  [ 70/312]  eta: 0:03:38  lr: 0.000599  min_lr: 0.000599  loss: 2.9510 (3.1142)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1792  max mem: 41794
Epoch: [229]  [ 80/312]  eta: 0:03:27  lr: 0.000598  min_lr: 0.000598  loss: 2.9470 (3.1081)  weight_decay: 0.0500 (0.0500)  time: 0.8542  data: 0.3546  max mem: 41794
Epoch: [229]  [ 90/312]  eta: 0:03:09  lr: 0.000598  min_lr: 0.000598  loss: 2.9470 (3.0945)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.1824  max mem: 41794
Epoch: [229]  [100/312]  eta: 0:03:00  lr: 0.000597  min_lr: 0.000597  loss: 3.0215 (3.0966)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1674  max mem: 41794
Epoch: [229]  [110/312]  eta: 0:02:45  lr: 0.000597  min_lr: 0.000597  loss: 3.0927 (3.0948)  weight_decay: 0.0500 (0.0500)  time: 0.6675  data: 0.1638  max mem: 41794
Epoch: [229]  [120/312]  eta: 0:02:38  lr: 0.000596  min_lr: 0.000596  loss: 3.1923 (3.1004)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1754  max mem: 41794
Epoch: [229]  [130/312]  eta: 0:02:30  lr: 0.000596  min_lr: 0.000596  loss: 3.1918 (3.0971)  weight_decay: 0.0500 (0.0500)  time: 0.8753  data: 0.3305  max mem: 41794
Epoch: [229]  [140/312]  eta: 0:02:18  lr: 0.000595  min_lr: 0.000595  loss: 3.1542 (3.0971)  weight_decay: 0.0500 (0.0500)  time: 0.6812  data: 0.1557  max mem: 41794
Epoch: [229]  [150/312]  eta: 0:02:11  lr: 0.000595  min_lr: 0.000595  loss: 3.2857 (3.0981)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.1207  max mem: 41794
Epoch: [229]  [160/312]  eta: 0:02:02  lr: 0.000594  min_lr: 0.000594  loss: 3.2252 (3.1030)  weight_decay: 0.0500 (0.0500)  time: 0.7952  data: 0.2081  max mem: 41794
Epoch: [229]  [170/312]  eta: 0:01:52  lr: 0.000594  min_lr: 0.000594  loss: 3.2758 (3.1180)  weight_decay: 0.0500 (0.0500)  time: 0.6610  data: 0.0905  max mem: 41794
Epoch: [229]  [180/312]  eta: 0:01:44  lr: 0.000593  min_lr: 0.000593  loss: 3.5222 (3.1258)  weight_decay: 0.0500 (0.0500)  time: 0.7221  data: 0.1173  max mem: 41794
Epoch: [229]  [190/312]  eta: 0:01:34  lr: 0.000593  min_lr: 0.000593  loss: 3.3012 (3.1298)  weight_decay: 0.0500 (0.0500)  time: 0.6658  data: 0.1148  max mem: 41794
Epoch: [229]  [200/312]  eta: 0:01:27  lr: 0.000592  min_lr: 0.000592  loss: 3.1989 (3.1151)  weight_decay: 0.0500 (0.0500)  time: 0.6978  data: 0.1472  max mem: 41794
Epoch: [229]  [210/312]  eta: 0:01:20  lr: 0.000592  min_lr: 0.000592  loss: 2.9777 (3.1145)  weight_decay: 0.0500 (0.0500)  time: 0.8968  data: 0.2212  max mem: 41794
Epoch: [229]  [220/312]  eta: 0:01:11  lr: 0.000591  min_lr: 0.000591  loss: 3.2169 (3.1165)  weight_decay: 0.0500 (0.0500)  time: 0.7176  data: 0.1001  max mem: 41794
Epoch: [229]  [230/312]  eta: 0:01:04  lr: 0.000591  min_lr: 0.000591  loss: 3.2864 (3.1244)  weight_decay: 0.0500 (0.0500)  time: 0.7015  data: 0.0565  max mem: 41794
Epoch: [229]  [240/312]  eta: 0:00:55  lr: 0.000590  min_lr: 0.000590  loss: 3.3843 (3.1266)  weight_decay: 0.0500 (0.0500)  time: 0.7244  data: 0.0682  max mem: 41794
Epoch: [229]  [250/312]  eta: 0:00:48  lr: 0.000590  min_lr: 0.000590  loss: 3.4421 (3.1368)  weight_decay: 0.0500 (0.0500)  time: 0.7254  data: 0.0683  max mem: 41794
Epoch: [229]  [260/312]  eta: 0:00:40  lr: 0.000589  min_lr: 0.000589  loss: 3.3727 (3.1422)  weight_decay: 0.0500 (0.0500)  time: 0.8273  data: 0.0358  max mem: 41794
Epoch: [229]  [270/312]  eta: 0:00:32  lr: 0.000589  min_lr: 0.000589  loss: 3.2555 (3.1394)  weight_decay: 0.0500 (0.0500)  time: 0.6608  data: 0.0218  max mem: 41794
Epoch: [229]  [280/312]  eta: 0:00:24  lr: 0.000588  min_lr: 0.000588  loss: 3.2187 (3.1414)  weight_decay: 0.0500 (0.0500)  time: 0.7259  data: 0.0663  max mem: 41794
Epoch: [229]  [290/312]  eta: 0:00:17  lr: 0.000588  min_lr: 0.000588  loss: 3.3879 (3.1389)  weight_decay: 0.0500 (0.0500)  time: 0.8335  data: 0.0497  max mem: 41794
Epoch: [229]  [300/312]  eta: 0:00:09  lr: 0.000587  min_lr: 0.000587  loss: 3.3977 (3.1416)  weight_decay: 0.0500 (0.0500)  time: 0.6091  data: 0.0002  max mem: 41794
Epoch: [229]  [310/312]  eta: 0:00:01  lr: 0.000587  min_lr: 0.000587  loss: 3.4266 (3.1495)  weight_decay: 0.0500 (0.0500)  time: 0.4680  data: 0.0001  max mem: 41794
Epoch: [229]  [311/312]  eta: 0:00:00  lr: 0.000587  min_lr: 0.000587  loss: 3.4161 (3.1493)  weight_decay: 0.0500 (0.0500)  time: 0.4665  data: 0.0001  max mem: 41794
Epoch: [229] Total time: 0:03:57 (0.7603 s / it)
Averaged stats: lr: 0.000587  min_lr: 0.000587  loss: 3.4161 (3.1413)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 0.8575 (0.8575)  acc1: 84.8958 (84.8958)  acc5: 97.2656 (97.2656)  time: 8.6829  data: 8.4738  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2239 (1.1228)  acc1: 76.5625 (77.8560)  acc5: 94.1406 (94.4800)  time: 1.1127  data: 0.9416  max mem: 41794
Test: Total time: 0:00:10 (1.1291 s / it)
* Acc@1 77.896 Acc@5 94.282 loss 1.125
Accuracy of the model on the 50000 test images: 77.9%
Max accuracy: 77.97%
Epoch: [230]  [  0/312]  eta: 1:23:04  lr: 0.000587  min_lr: 0.000587  loss: 2.4639 (2.4639)  weight_decay: 0.0500 (0.0500)  time: 15.9759  data: 15.2747  max mem: 41794
Epoch: [230]  [ 10/312]  eta: 0:11:07  lr: 0.000586  min_lr: 0.000586  loss: 3.3268 (3.1528)  weight_decay: 0.0500 (0.0500)  time: 2.2098  data: 1.3893  max mem: 41794
Epoch: [230]  [ 20/312]  eta: 0:07:31  lr: 0.000586  min_lr: 0.000586  loss: 3.3485 (3.1512)  weight_decay: 0.0500 (0.0500)  time: 0.8238  data: 0.0804  max mem: 41794
Epoch: [230]  [ 30/312]  eta: 0:05:43  lr: 0.000585  min_lr: 0.000585  loss: 2.9055 (3.0457)  weight_decay: 0.0500 (0.0500)  time: 0.6733  data: 0.0904  max mem: 41794
Epoch: [230]  [ 40/312]  eta: 0:04:49  lr: 0.000585  min_lr: 0.000585  loss: 3.0584 (3.0357)  weight_decay: 0.0500 (0.0500)  time: 0.5587  data: 0.0259  max mem: 41794
Epoch: [230]  [ 50/312]  eta: 0:04:26  lr: 0.000584  min_lr: 0.000584  loss: 3.1580 (3.0543)  weight_decay: 0.0500 (0.0500)  time: 0.7005  data: 0.1200  max mem: 41794
Epoch: [230]  [ 60/312]  eta: 0:03:54  lr: 0.000584  min_lr: 0.000584  loss: 3.2479 (3.0476)  weight_decay: 0.0500 (0.0500)  time: 0.6616  data: 0.1047  max mem: 41794
Epoch: [230]  [ 70/312]  eta: 0:03:46  lr: 0.000583  min_lr: 0.000583  loss: 3.3164 (3.0675)  weight_decay: 0.0500 (0.0500)  time: 0.7387  data: 0.1132  max mem: 41794
Epoch: [230]  [ 80/312]  eta: 0:03:31  lr: 0.000583  min_lr: 0.000583  loss: 3.2684 (3.0764)  weight_decay: 0.0500 (0.0500)  time: 0.8519  data: 0.1343  max mem: 41794
Epoch: [230]  [ 90/312]  eta: 0:03:15  lr: 0.000582  min_lr: 0.000582  loss: 3.2349 (3.0892)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0823  max mem: 41794
Epoch: [230]  [100/312]  eta: 0:03:07  lr: 0.000582  min_lr: 0.000582  loss: 3.2349 (3.0815)  weight_decay: 0.0500 (0.0500)  time: 0.7603  data: 0.1637  max mem: 41794
Epoch: [230]  [110/312]  eta: 0:02:51  lr: 0.000581  min_lr: 0.000581  loss: 3.2685 (3.0805)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.1030  max mem: 41794
Epoch: [230]  [120/312]  eta: 0:02:43  lr: 0.000581  min_lr: 0.000581  loss: 3.2631 (3.0775)  weight_decay: 0.0500 (0.0500)  time: 0.7012  data: 0.1029  max mem: 41794
Epoch: [230]  [130/312]  eta: 0:02:33  lr: 0.000580  min_lr: 0.000580  loss: 3.0299 (3.0610)  weight_decay: 0.0500 (0.0500)  time: 0.8385  data: 0.1427  max mem: 41794
Epoch: [230]  [140/312]  eta: 0:02:22  lr: 0.000580  min_lr: 0.000580  loss: 2.9936 (3.0599)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.0863  max mem: 41794
Epoch: [230]  [150/312]  eta: 0:02:14  lr: 0.000579  min_lr: 0.000579  loss: 2.9873 (3.0542)  weight_decay: 0.0500 (0.0500)  time: 0.7497  data: 0.1435  max mem: 41794
Epoch: [230]  [160/312]  eta: 0:02:04  lr: 0.000579  min_lr: 0.000579  loss: 3.2286 (3.0546)  weight_decay: 0.0500 (0.0500)  time: 0.7647  data: 0.1010  max mem: 41794
Epoch: [230]  [170/312]  eta: 0:01:54  lr: 0.000578  min_lr: 0.000578  loss: 3.2286 (3.0517)  weight_decay: 0.0500 (0.0500)  time: 0.6285  data: 0.0788  max mem: 41794
Epoch: [230]  [180/312]  eta: 0:01:46  lr: 0.000578  min_lr: 0.000578  loss: 3.1510 (3.0547)  weight_decay: 0.0500 (0.0500)  time: 0.7341  data: 0.1783  max mem: 41794
Epoch: [230]  [190/312]  eta: 0:01:36  lr: 0.000577  min_lr: 0.000577  loss: 3.3992 (3.0676)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.1036  max mem: 41794
Epoch: [230]  [200/312]  eta: 0:01:29  lr: 0.000577  min_lr: 0.000577  loss: 3.4205 (3.0672)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1429  max mem: 41794
Epoch: [230]  [210/312]  eta: 0:01:21  lr: 0.000576  min_lr: 0.000576  loss: 3.3522 (3.0807)  weight_decay: 0.0500 (0.0500)  time: 0.8319  data: 0.2807  max mem: 41794
Epoch: [230]  [220/312]  eta: 0:01:12  lr: 0.000575  min_lr: 0.000575  loss: 3.4318 (3.0869)  weight_decay: 0.0500 (0.0500)  time: 0.6367  data: 0.1385  max mem: 41794
Epoch: [230]  [230/312]  eta: 0:01:04  lr: 0.000575  min_lr: 0.000575  loss: 3.3425 (3.0964)  weight_decay: 0.0500 (0.0500)  time: 0.7017  data: 0.1750  max mem: 41794
Epoch: [230]  [240/312]  eta: 0:00:56  lr: 0.000574  min_lr: 0.000574  loss: 3.4238 (3.1002)  weight_decay: 0.0500 (0.0500)  time: 0.8645  data: 0.2478  max mem: 41794
Epoch: [230]  [250/312]  eta: 0:00:48  lr: 0.000574  min_lr: 0.000574  loss: 3.3814 (3.1036)  weight_decay: 0.0500 (0.0500)  time: 0.7333  data: 0.1523  max mem: 41794
Epoch: [230]  [260/312]  eta: 0:00:41  lr: 0.000573  min_lr: 0.000573  loss: 3.3377 (3.1094)  weight_decay: 0.0500 (0.0500)  time: 0.8184  data: 0.2086  max mem: 41794
Epoch: [230]  [270/312]  eta: 0:00:32  lr: 0.000573  min_lr: 0.000573  loss: 3.1981 (3.1083)  weight_decay: 0.0500 (0.0500)  time: 0.7459  data: 0.1301  max mem: 41794
Epoch: [230]  [280/312]  eta: 0:00:25  lr: 0.000572  min_lr: 0.000572  loss: 3.1391 (3.1039)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.0920  max mem: 41794
Epoch: [230]  [290/312]  eta: 0:00:17  lr: 0.000572  min_lr: 0.000572  loss: 3.2639 (3.1053)  weight_decay: 0.0500 (0.0500)  time: 0.7987  data: 0.1410  max mem: 41794
Epoch: [230]  [300/312]  eta: 0:00:09  lr: 0.000571  min_lr: 0.000571  loss: 3.0563 (3.0990)  weight_decay: 0.0500 (0.0500)  time: 0.6016  data: 0.0498  max mem: 41794
Epoch: [230]  [310/312]  eta: 0:00:01  lr: 0.000571  min_lr: 0.000571  loss: 3.1255 (3.1012)  weight_decay: 0.0500 (0.0500)  time: 0.4626  data: 0.0001  max mem: 41794
Epoch: [230]  [311/312]  eta: 0:00:00  lr: 0.000571  min_lr: 0.000571  loss: 3.1255 (3.1020)  weight_decay: 0.0500 (0.0500)  time: 0.4616  data: 0.0001  max mem: 41794
Epoch: [230] Total time: 0:03:59 (0.7681 s / it)
Averaged stats: lr: 0.000571  min_lr: 0.000571  loss: 3.1255 (3.1301)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 0.9396 (0.9396)  acc1: 83.9844 (83.9844)  acc5: 96.3542 (96.3542)  time: 8.0237  data: 7.8111  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2299 (1.1228)  acc1: 76.8229 (77.6640)  acc5: 94.1406 (94.2240)  time: 1.0650  data: 0.8934  max mem: 41794
Test: Total time: 0:00:09 (1.0867 s / it)
* Acc@1 78.160 Acc@5 94.276 loss 1.116
Accuracy of the model on the 50000 test images: 78.2%
Max accuracy: 78.16%
Epoch: [231]  [  0/312]  eta: 1:11:39  lr: 0.000571  min_lr: 0.000571  loss: 3.3449 (3.3449)  weight_decay: 0.0500 (0.0500)  time: 13.7793  data: 12.9906  max mem: 41794
Epoch: [231]  [ 10/312]  eta: 0:11:06  lr: 0.000570  min_lr: 0.000570  loss: 3.3985 (3.2917)  weight_decay: 0.0500 (0.0500)  time: 2.2078  data: 1.3070  max mem: 41794
Epoch: [231]  [ 20/312]  eta: 0:07:24  lr: 0.000570  min_lr: 0.000570  loss: 3.3746 (3.2668)  weight_decay: 0.0500 (0.0500)  time: 0.9089  data: 0.1069  max mem: 41794
Epoch: [231]  [ 30/312]  eta: 0:05:40  lr: 0.000569  min_lr: 0.000569  loss: 3.2719 (3.2254)  weight_decay: 0.0500 (0.0500)  time: 0.6546  data: 0.0488  max mem: 41794
Epoch: [231]  [ 40/312]  eta: 0:04:43  lr: 0.000569  min_lr: 0.000569  loss: 3.3672 (3.2671)  weight_decay: 0.0500 (0.0500)  time: 0.5395  data: 0.0205  max mem: 41794
Epoch: [231]  [ 50/312]  eta: 0:04:18  lr: 0.000568  min_lr: 0.000568  loss: 3.3863 (3.2499)  weight_decay: 0.0500 (0.0500)  time: 0.6490  data: 0.0870  max mem: 41794
Epoch: [231]  [ 60/312]  eta: 0:03:50  lr: 0.000568  min_lr: 0.000568  loss: 3.2830 (3.2170)  weight_decay: 0.0500 (0.0500)  time: 0.6478  data: 0.0794  max mem: 41794
Epoch: [231]  [ 70/312]  eta: 0:03:42  lr: 0.000567  min_lr: 0.000567  loss: 3.1951 (3.1723)  weight_decay: 0.0500 (0.0500)  time: 0.7432  data: 0.1086  max mem: 41794
Epoch: [231]  [ 80/312]  eta: 0:03:23  lr: 0.000567  min_lr: 0.000567  loss: 3.0911 (3.1620)  weight_decay: 0.0500 (0.0500)  time: 0.7708  data: 0.1242  max mem: 41794
Epoch: [231]  [ 90/312]  eta: 0:03:10  lr: 0.000566  min_lr: 0.000566  loss: 3.2608 (3.1547)  weight_decay: 0.0500 (0.0500)  time: 0.6497  data: 0.1012  max mem: 41794
Epoch: [231]  [100/312]  eta: 0:03:33  lr: 0.000566  min_lr: 0.000566  loss: 2.9783 (3.1322)  weight_decay: 0.0500 (0.0500)  time: 1.5334  data: 0.6620  max mem: 41794
Epoch: [231]  [110/312]  eta: 0:03:13  lr: 0.000565  min_lr: 0.000565  loss: 3.0988 (3.1293)  weight_decay: 0.0500 (0.0500)  time: 1.4110  data: 0.5782  max mem: 41794
Epoch: [231]  [120/312]  eta: 0:03:36  lr: 0.000565  min_lr: 0.000565  loss: 2.8742 (3.1018)  weight_decay: 0.0500 (0.0500)  time: 1.7292  data: 0.8897  max mem: 41794
Epoch: [231]  [130/312]  eta: 0:03:33  lr: 0.000564  min_lr: 0.000564  loss: 2.8535 (3.1107)  weight_decay: 0.0500 (0.0500)  time: 2.3557  data: 1.1143  max mem: 41794
Epoch: [231]  [140/312]  eta: 0:03:13  lr: 0.000564  min_lr: 0.000564  loss: 3.2258 (3.1254)  weight_decay: 0.0500 (0.0500)  time: 1.0983  data: 0.2253  max mem: 41794
Epoch: [231]  [150/312]  eta: 0:03:13  lr: 0.000563  min_lr: 0.000563  loss: 3.2729 (3.1338)  weight_decay: 0.0500 (0.0500)  time: 1.3539  data: 0.4967  max mem: 41794
Epoch: [231]  [160/312]  eta: 0:03:06  lr: 0.000563  min_lr: 0.000563  loss: 3.2729 (3.1427)  weight_decay: 0.0500 (0.0500)  time: 1.9543  data: 0.9778  max mem: 41794
Epoch: [231]  [170/312]  eta: 0:02:49  lr: 0.000562  min_lr: 0.000562  loss: 3.4706 (3.1404)  weight_decay: 0.0500 (0.0500)  time: 1.1486  data: 0.5249  max mem: 41794
Epoch: [231]  [180/312]  eta: 0:02:43  lr: 0.000562  min_lr: 0.000562  loss: 3.1330 (3.1290)  weight_decay: 0.0500 (0.0500)  time: 1.3236  data: 0.2084  max mem: 41794
Epoch: [231]  [190/312]  eta: 0:02:29  lr: 0.000561  min_lr: 0.000561  loss: 3.1741 (3.1374)  weight_decay: 0.0500 (0.0500)  time: 1.5392  data: 0.1653  max mem: 41794
Epoch: [231]  [200/312]  eta: 0:02:23  lr: 0.000561  min_lr: 0.000561  loss: 3.3955 (3.1443)  weight_decay: 0.0500 (0.0500)  time: 1.6574  data: 0.3171  max mem: 41794
Epoch: [231]  [210/312]  eta: 0:02:14  lr: 0.000560  min_lr: 0.000560  loss: 3.3955 (3.1537)  weight_decay: 0.0500 (0.0500)  time: 2.1645  data: 0.6765  max mem: 41794
Epoch: [231]  [220/312]  eta: 0:01:57  lr: 0.000560  min_lr: 0.000560  loss: 3.3613 (3.1587)  weight_decay: 0.0500 (0.0500)  time: 1.2642  data: 0.3599  max mem: 41794
Epoch: [231]  [230/312]  eta: 0:01:45  lr: 0.000559  min_lr: 0.000559  loss: 3.3301 (3.1588)  weight_decay: 0.0500 (0.0500)  time: 0.9348  data: 0.2655  max mem: 41794
Epoch: [231]  [240/312]  eta: 0:01:30  lr: 0.000559  min_lr: 0.000559  loss: 3.2791 (3.1603)  weight_decay: 0.0500 (0.0500)  time: 1.0170  data: 0.2656  max mem: 41794
Epoch: [231]  [250/312]  eta: 0:01:16  lr: 0.000558  min_lr: 0.000558  loss: 3.0726 (3.1558)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.0274  max mem: 41794
Epoch: [231]  [260/312]  eta: 0:01:03  lr: 0.000558  min_lr: 0.000558  loss: 2.8396 (3.1499)  weight_decay: 0.0500 (0.0500)  time: 0.8203  data: 0.0578  max mem: 41794
Epoch: [231]  [270/312]  eta: 0:00:50  lr: 0.000557  min_lr: 0.000557  loss: 2.9519 (3.1477)  weight_decay: 0.0500 (0.0500)  time: 0.7095  data: 0.0310  max mem: 41794
Epoch: [231]  [280/312]  eta: 0:00:37  lr: 0.000557  min_lr: 0.000557  loss: 3.0856 (3.1469)  weight_decay: 0.0500 (0.0500)  time: 0.6492  data: 0.0185  max mem: 41794
Epoch: [231]  [290/312]  eta: 0:00:25  lr: 0.000556  min_lr: 0.000556  loss: 3.0856 (3.1440)  weight_decay: 0.0500 (0.0500)  time: 0.8280  data: 0.0183  max mem: 41794
Epoch: [231]  [300/312]  eta: 0:00:13  lr: 0.000556  min_lr: 0.000556  loss: 3.0856 (3.1384)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.0105  max mem: 41794
Epoch: [231]  [310/312]  eta: 0:00:02  lr: 0.000555  min_lr: 0.000555  loss: 3.0856 (3.1358)  weight_decay: 0.0500 (0.0500)  time: 0.4762  data: 0.0104  max mem: 41794
Epoch: [231]  [311/312]  eta: 0:00:01  lr: 0.000555  min_lr: 0.000555  loss: 3.3301 (3.1366)  weight_decay: 0.0500 (0.0500)  time: 0.4663  data: 0.0001  max mem: 41794
Epoch: [231] Total time: 0:05:52 (1.1302 s / it)
Averaged stats: lr: 0.000555  min_lr: 0.000555  loss: 3.3301 (3.1314)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 0.8482 (0.8482)  acc1: 83.5938 (83.5938)  acc5: 96.7448 (96.7448)  time: 8.8997  data: 8.6923  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1811 (1.0793)  acc1: 77.6042 (78.2240)  acc5: 93.8802 (94.2880)  time: 1.1370  data: 0.9659  max mem: 41794
Test: Total time: 0:00:10 (1.1537 s / it)
* Acc@1 78.066 Acc@5 94.286 loss 1.086
Accuracy of the model on the 50000 test images: 78.1%
Max accuracy: 78.16%
Epoch: [232]  [  0/312]  eta: 1:22:01  lr: 0.000555  min_lr: 0.000555  loss: 3.3124 (3.3124)  weight_decay: 0.0500 (0.0500)  time: 15.7733  data: 11.9774  max mem: 41794
Epoch: [232]  [ 10/312]  eta: 0:11:28  lr: 0.000555  min_lr: 0.000555  loss: 3.4637 (3.3909)  weight_decay: 0.0500 (0.0500)  time: 2.2795  data: 1.2010  max mem: 41794
Epoch: [232]  [ 20/312]  eta: 0:07:31  lr: 0.000554  min_lr: 0.000554  loss: 3.3745 (3.1788)  weight_decay: 0.0500 (0.0500)  time: 0.8341  data: 0.0846  max mem: 41794
Epoch: [232]  [ 30/312]  eta: 0:05:46  lr: 0.000554  min_lr: 0.000554  loss: 3.0603 (3.1630)  weight_decay: 0.0500 (0.0500)  time: 0.6499  data: 0.0253  max mem: 41794
Epoch: [232]  [ 40/312]  eta: 0:04:46  lr: 0.000553  min_lr: 0.000553  loss: 3.1195 (3.1067)  weight_decay: 0.0500 (0.0500)  time: 0.5391  data: 0.0049  max mem: 41794
Epoch: [232]  [ 50/312]  eta: 0:04:18  lr: 0.000553  min_lr: 0.000553  loss: 3.1580 (3.1016)  weight_decay: 0.0500 (0.0500)  time: 0.6131  data: 0.0076  max mem: 41794
Epoch: [232]  [ 60/312]  eta: 0:03:48  lr: 0.000552  min_lr: 0.000552  loss: 3.1580 (3.0928)  weight_decay: 0.0500 (0.0500)  time: 0.6073  data: 0.0054  max mem: 41794
Epoch: [232]  [ 70/312]  eta: 0:03:40  lr: 0.000552  min_lr: 0.000552  loss: 3.3422 (3.1178)  weight_decay: 0.0500 (0.0500)  time: 0.7226  data: 0.0695  max mem: 41794
Epoch: [232]  [ 80/312]  eta: 0:03:25  lr: 0.000551  min_lr: 0.000551  loss: 3.3189 (3.1236)  weight_decay: 0.0500 (0.0500)  time: 0.8265  data: 0.0695  max mem: 41794
Epoch: [232]  [ 90/312]  eta: 0:03:11  lr: 0.000551  min_lr: 0.000551  loss: 3.1725 (3.1265)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.0464  max mem: 41794
Epoch: [232]  [100/312]  eta: 0:03:03  lr: 0.000550  min_lr: 0.000550  loss: 3.1580 (3.1065)  weight_decay: 0.0500 (0.0500)  time: 0.7638  data: 0.1334  max mem: 41794
Epoch: [232]  [110/312]  eta: 0:02:47  lr: 0.000550  min_lr: 0.000550  loss: 2.9161 (3.0935)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.0901  max mem: 41794
Epoch: [232]  [120/312]  eta: 0:02:40  lr: 0.000549  min_lr: 0.000549  loss: 3.1198 (3.1020)  weight_decay: 0.0500 (0.0500)  time: 0.6974  data: 0.0901  max mem: 41794
Epoch: [232]  [130/312]  eta: 0:02:32  lr: 0.000549  min_lr: 0.000549  loss: 3.1283 (3.0740)  weight_decay: 0.0500 (0.0500)  time: 0.8758  data: 0.1644  max mem: 41794
Epoch: [232]  [140/312]  eta: 0:02:19  lr: 0.000548  min_lr: 0.000548  loss: 3.0022 (3.0748)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.0773  max mem: 41794
Epoch: [232]  [150/312]  eta: 0:02:11  lr: 0.000548  min_lr: 0.000548  loss: 3.2089 (3.0619)  weight_decay: 0.0500 (0.0500)  time: 0.6676  data: 0.0743  max mem: 41794
Epoch: [232]  [160/312]  eta: 0:02:02  lr: 0.000547  min_lr: 0.000547  loss: 3.1228 (3.0678)  weight_decay: 0.0500 (0.0500)  time: 0.7686  data: 0.0744  max mem: 41794
Epoch: [232]  [170/312]  eta: 0:01:53  lr: 0.000547  min_lr: 0.000547  loss: 3.1523 (3.0766)  weight_decay: 0.0500 (0.0500)  time: 0.6872  data: 0.0882  max mem: 41794
Epoch: [232]  [180/312]  eta: 0:01:46  lr: 0.000546  min_lr: 0.000546  loss: 3.0508 (3.0628)  weight_decay: 0.0500 (0.0500)  time: 0.7841  data: 0.1778  max mem: 41794
Epoch: [232]  [190/312]  eta: 0:01:36  lr: 0.000546  min_lr: 0.000546  loss: 2.9514 (3.0585)  weight_decay: 0.0500 (0.0500)  time: 0.6952  data: 0.0902  max mem: 41794
Epoch: [232]  [200/312]  eta: 0:01:29  lr: 0.000545  min_lr: 0.000545  loss: 3.1401 (3.0612)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.0946  max mem: 41794
Epoch: [232]  [210/312]  eta: 0:01:21  lr: 0.000545  min_lr: 0.000545  loss: 3.2221 (3.0648)  weight_decay: 0.0500 (0.0500)  time: 0.8504  data: 0.1564  max mem: 41794
Epoch: [232]  [220/312]  eta: 0:01:11  lr: 0.000544  min_lr: 0.000544  loss: 3.0344 (3.0549)  weight_decay: 0.0500 (0.0500)  time: 0.6355  data: 0.0649  max mem: 41794
Epoch: [232]  [230/312]  eta: 0:01:04  lr: 0.000544  min_lr: 0.000544  loss: 2.8609 (3.0503)  weight_decay: 0.0500 (0.0500)  time: 0.6918  data: 0.0728  max mem: 41794
Epoch: [232]  [240/312]  eta: 0:00:56  lr: 0.000543  min_lr: 0.000543  loss: 3.0238 (3.0497)  weight_decay: 0.0500 (0.0500)  time: 0.8232  data: 0.0717  max mem: 41794
Epoch: [232]  [250/312]  eta: 0:00:48  lr: 0.000543  min_lr: 0.000543  loss: 3.1302 (3.0545)  weight_decay: 0.0500 (0.0500)  time: 0.6867  data: 0.0573  max mem: 41794
Epoch: [232]  [260/312]  eta: 0:00:40  lr: 0.000542  min_lr: 0.000542  loss: 3.3924 (3.0672)  weight_decay: 0.0500 (0.0500)  time: 0.7870  data: 0.1166  max mem: 41794
Epoch: [232]  [270/312]  eta: 0:00:32  lr: 0.000542  min_lr: 0.000542  loss: 3.3924 (3.0714)  weight_decay: 0.0500 (0.0500)  time: 0.7294  data: 0.0641  max mem: 41794
Epoch: [232]  [280/312]  eta: 0:00:24  lr: 0.000541  min_lr: 0.000541  loss: 3.3293 (3.0767)  weight_decay: 0.0500 (0.0500)  time: 0.6479  data: 0.0435  max mem: 41794
Epoch: [232]  [290/312]  eta: 0:00:17  lr: 0.000541  min_lr: 0.000541  loss: 3.2308 (3.0684)  weight_decay: 0.0500 (0.0500)  time: 0.8511  data: 0.0646  max mem: 41794
Epoch: [232]  [300/312]  eta: 0:00:09  lr: 0.000540  min_lr: 0.000540  loss: 3.2308 (3.0718)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.0242  max mem: 41794
Epoch: [232]  [310/312]  eta: 0:00:01  lr: 0.000540  min_lr: 0.000540  loss: 3.1887 (3.0680)  weight_decay: 0.0500 (0.0500)  time: 0.4630  data: 0.0001  max mem: 41794
Epoch: [232]  [311/312]  eta: 0:00:00  lr: 0.000540  min_lr: 0.000540  loss: 3.1887 (3.0693)  weight_decay: 0.0500 (0.0500)  time: 0.4629  data: 0.0001  max mem: 41794
Epoch: [232] Total time: 0:03:59 (0.7663 s / it)
Averaged stats: lr: 0.000540  min_lr: 0.000540  loss: 3.1887 (3.1172)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 0.9947 (0.9947)  acc1: 83.4635 (83.4635)  acc5: 96.3542 (96.3542)  time: 8.0439  data: 7.8416  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2113 (1.1761)  acc1: 77.7344 (77.5040)  acc5: 94.4010 (94.2880)  time: 1.1153  data: 0.9448  max mem: 41794
Test: Total time: 0:00:10 (1.1429 s / it)
* Acc@1 78.040 Acc@5 94.368 loss 1.177
Accuracy of the model on the 50000 test images: 78.0%
Max accuracy: 78.16%
Epoch: [233]  [  0/312]  eta: 1:22:19  lr: 0.000540  min_lr: 0.000540  loss: 3.7465 (3.7465)  weight_decay: 0.0500 (0.0500)  time: 15.8308  data: 13.0033  max mem: 41794
Epoch: [233]  [ 10/312]  eta: 0:11:05  lr: 0.000539  min_lr: 0.000539  loss: 3.4574 (3.4125)  weight_decay: 0.0500 (0.0500)  time: 2.2050  data: 1.3472  max mem: 41794
Epoch: [233]  [ 20/312]  eta: 0:07:17  lr: 0.000539  min_lr: 0.000539  loss: 3.3556 (3.3154)  weight_decay: 0.0500 (0.0500)  time: 0.7833  data: 0.1832  max mem: 41794
Epoch: [233]  [ 30/312]  eta: 0:05:33  lr: 0.000538  min_lr: 0.000538  loss: 3.3556 (3.2667)  weight_decay: 0.0500 (0.0500)  time: 0.6177  data: 0.0954  max mem: 41794
Epoch: [233]  [ 40/312]  eta: 0:04:42  lr: 0.000538  min_lr: 0.000538  loss: 3.3666 (3.2554)  weight_decay: 0.0500 (0.0500)  time: 0.5521  data: 0.0501  max mem: 41794
Epoch: [233]  [ 50/312]  eta: 0:04:20  lr: 0.000537  min_lr: 0.000537  loss: 3.4294 (3.2173)  weight_decay: 0.0500 (0.0500)  time: 0.7021  data: 0.2038  max mem: 41794
Epoch: [233]  [ 60/312]  eta: 0:03:49  lr: 0.000537  min_lr: 0.000537  loss: 3.1759 (3.2021)  weight_decay: 0.0500 (0.0500)  time: 0.6508  data: 0.1571  max mem: 41794
Epoch: [233]  [ 70/312]  eta: 0:03:40  lr: 0.000536  min_lr: 0.000536  loss: 3.1712 (3.1902)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.1870  max mem: 41794
Epoch: [233]  [ 80/312]  eta: 0:03:29  lr: 0.000536  min_lr: 0.000536  loss: 3.1137 (3.1569)  weight_decay: 0.0500 (0.0500)  time: 0.8718  data: 0.3352  max mem: 41794
Epoch: [233]  [ 90/312]  eta: 0:03:10  lr: 0.000535  min_lr: 0.000535  loss: 2.7878 (3.1294)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1593  max mem: 41794
Epoch: [233]  [100/312]  eta: 0:03:04  lr: 0.000535  min_lr: 0.000535  loss: 3.3126 (3.1430)  weight_decay: 0.0500 (0.0500)  time: 0.7477  data: 0.2178  max mem: 41794
Epoch: [233]  [110/312]  eta: 0:02:48  lr: 0.000534  min_lr: 0.000534  loss: 3.0440 (3.1274)  weight_decay: 0.0500 (0.0500)  time: 0.7343  data: 0.2073  max mem: 41794
Epoch: [233]  [120/312]  eta: 0:02:41  lr: 0.000534  min_lr: 0.000534  loss: 3.0090 (3.1303)  weight_decay: 0.0500 (0.0500)  time: 0.7034  data: 0.1538  max mem: 41794
Epoch: [233]  [130/312]  eta: 0:02:34  lr: 0.000533  min_lr: 0.000533  loss: 3.3747 (3.1338)  weight_decay: 0.0500 (0.0500)  time: 0.9251  data: 0.2966  max mem: 41794
Epoch: [233]  [140/312]  eta: 0:02:21  lr: 0.000533  min_lr: 0.000533  loss: 3.3520 (3.1351)  weight_decay: 0.0500 (0.0500)  time: 0.7158  data: 0.1465  max mem: 41794
Epoch: [233]  [150/312]  eta: 0:02:13  lr: 0.000533  min_lr: 0.000533  loss: 3.3333 (3.1426)  weight_decay: 0.0500 (0.0500)  time: 0.6696  data: 0.1435  max mem: 41794
Epoch: [233]  [160/312]  eta: 0:02:05  lr: 0.000532  min_lr: 0.000532  loss: 3.2018 (3.1341)  weight_decay: 0.0500 (0.0500)  time: 0.8132  data: 0.2127  max mem: 41794
Epoch: [233]  [170/312]  eta: 0:01:54  lr: 0.000532  min_lr: 0.000532  loss: 3.1082 (3.1371)  weight_decay: 0.0500 (0.0500)  time: 0.6712  data: 0.1040  max mem: 41794
Epoch: [233]  [180/312]  eta: 0:01:46  lr: 0.000531  min_lr: 0.000531  loss: 3.3534 (3.1383)  weight_decay: 0.0500 (0.0500)  time: 0.6775  data: 0.1385  max mem: 41794
Epoch: [233]  [190/312]  eta: 0:01:36  lr: 0.000531  min_lr: 0.000531  loss: 3.2825 (3.1495)  weight_decay: 0.0500 (0.0500)  time: 0.6450  data: 0.1074  max mem: 41794
Epoch: [233]  [200/312]  eta: 0:01:29  lr: 0.000530  min_lr: 0.000530  loss: 3.4079 (3.1644)  weight_decay: 0.0500 (0.0500)  time: 0.6917  data: 0.1541  max mem: 41794
Epoch: [233]  [210/312]  eta: 0:01:21  lr: 0.000530  min_lr: 0.000530  loss: 3.2668 (3.1515)  weight_decay: 0.0500 (0.0500)  time: 0.8695  data: 0.3161  max mem: 41794
Epoch: [233]  [220/312]  eta: 0:01:12  lr: 0.000529  min_lr: 0.000529  loss: 3.2668 (3.1632)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.1627  max mem: 41794
Epoch: [233]  [230/312]  eta: 0:01:04  lr: 0.000529  min_lr: 0.000529  loss: 3.3896 (3.1546)  weight_decay: 0.0500 (0.0500)  time: 0.6895  data: 0.1962  max mem: 41794
Epoch: [233]  [240/312]  eta: 0:00:57  lr: 0.000528  min_lr: 0.000528  loss: 3.1478 (3.1515)  weight_decay: 0.0500 (0.0500)  time: 0.8883  data: 0.3999  max mem: 41794
Epoch: [233]  [250/312]  eta: 0:00:48  lr: 0.000528  min_lr: 0.000528  loss: 3.1794 (3.1525)  weight_decay: 0.0500 (0.0500)  time: 0.6942  data: 0.2043  max mem: 41794
Epoch: [233]  [260/312]  eta: 0:00:40  lr: 0.000527  min_lr: 0.000527  loss: 3.2746 (3.1517)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1800  max mem: 41794
Epoch: [233]  [270/312]  eta: 0:00:32  lr: 0.000527  min_lr: 0.000527  loss: 3.1295 (3.1499)  weight_decay: 0.0500 (0.0500)  time: 0.6783  data: 0.1800  max mem: 41794
Epoch: [233]  [280/312]  eta: 0:00:24  lr: 0.000526  min_lr: 0.000526  loss: 3.1295 (3.1543)  weight_decay: 0.0500 (0.0500)  time: 0.7008  data: 0.2033  max mem: 41794
Epoch: [233]  [290/312]  eta: 0:00:17  lr: 0.000526  min_lr: 0.000526  loss: 3.4247 (3.1641)  weight_decay: 0.0500 (0.0500)  time: 0.8302  data: 0.3305  max mem: 41794
Epoch: [233]  [300/312]  eta: 0:00:09  lr: 0.000525  min_lr: 0.000525  loss: 3.3932 (3.1683)  weight_decay: 0.0500 (0.0500)  time: 0.6147  data: 0.1276  max mem: 41794
Epoch: [233]  [310/312]  eta: 0:00:01  lr: 0.000525  min_lr: 0.000525  loss: 3.3894 (3.1734)  weight_decay: 0.0500 (0.0500)  time: 0.4653  data: 0.0001  max mem: 41794
Epoch: [233]  [311/312]  eta: 0:00:00  lr: 0.000525  min_lr: 0.000525  loss: 3.3894 (3.1723)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [233] Total time: 0:03:58 (0.7630 s / it)
Averaged stats: lr: 0.000525  min_lr: 0.000525  loss: 3.3894 (3.1328)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 0.9961 (0.9961)  acc1: 83.7240 (83.7240)  acc5: 96.2240 (96.2240)  time: 8.1095  data: 7.8990  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2253 (1.1633)  acc1: 77.2135 (77.6960)  acc5: 94.5312 (94.4800)  time: 1.0903  data: 0.9118  max mem: 41794
Test: Total time: 0:00:09 (1.1036 s / it)
* Acc@1 78.220 Acc@5 94.382 loss 1.162
Accuracy of the model on the 50000 test images: 78.2%
Max accuracy: 78.22%
Epoch: [234]  [  0/312]  eta: 1:22:30  lr: 0.000525  min_lr: 0.000525  loss: 2.4881 (2.4881)  weight_decay: 0.0500 (0.0500)  time: 15.8678  data: 12.9315  max mem: 41794
Epoch: [234]  [ 10/312]  eta: 0:10:24  lr: 0.000524  min_lr: 0.000524  loss: 3.0432 (2.9377)  weight_decay: 0.0500 (0.0500)  time: 2.0677  data: 1.3430  max mem: 41794
Epoch: [234]  [ 20/312]  eta: 0:07:32  lr: 0.000524  min_lr: 0.000524  loss: 3.1171 (3.0101)  weight_decay: 0.0500 (0.0500)  time: 0.8332  data: 0.2231  max mem: 41794
Epoch: [234]  [ 30/312]  eta: 0:05:45  lr: 0.000523  min_lr: 0.000523  loss: 3.1525 (3.1117)  weight_decay: 0.0500 (0.0500)  time: 0.7640  data: 0.1351  max mem: 41794
Epoch: [234]  [ 40/312]  eta: 0:04:54  lr: 0.000523  min_lr: 0.000523  loss: 3.1177 (3.0840)  weight_decay: 0.0500 (0.0500)  time: 0.5962  data: 0.0581  max mem: 41794
Epoch: [234]  [ 50/312]  eta: 0:04:29  lr: 0.000522  min_lr: 0.000522  loss: 2.9575 (3.0467)  weight_decay: 0.0500 (0.0500)  time: 0.7249  data: 0.1336  max mem: 41794
Epoch: [234]  [ 60/312]  eta: 0:03:57  lr: 0.000522  min_lr: 0.000522  loss: 3.3057 (3.1135)  weight_decay: 0.0500 (0.0500)  time: 0.6536  data: 0.0821  max mem: 41794
Epoch: [234]  [ 70/312]  eta: 0:03:46  lr: 0.000521  min_lr: 0.000521  loss: 3.3057 (3.1201)  weight_decay: 0.0500 (0.0500)  time: 0.6981  data: 0.1018  max mem: 41794
Epoch: [234]  [ 80/312]  eta: 0:03:32  lr: 0.000521  min_lr: 0.000521  loss: 3.1037 (3.1243)  weight_decay: 0.0500 (0.0500)  time: 0.8296  data: 0.1054  max mem: 41794
Epoch: [234]  [ 90/312]  eta: 0:03:16  lr: 0.000520  min_lr: 0.000520  loss: 3.3551 (3.1548)  weight_decay: 0.0500 (0.0500)  time: 0.7118  data: 0.0760  max mem: 41794
Epoch: [234]  [100/312]  eta: 0:03:10  lr: 0.000520  min_lr: 0.000520  loss: 3.2152 (3.1344)  weight_decay: 0.0500 (0.0500)  time: 0.8342  data: 0.1267  max mem: 41794
Epoch: [234]  [110/312]  eta: 0:02:54  lr: 0.000519  min_lr: 0.000519  loss: 3.1461 (3.1347)  weight_decay: 0.0500 (0.0500)  time: 0.7655  data: 0.0593  max mem: 41794
Epoch: [234]  [120/312]  eta: 0:02:44  lr: 0.000519  min_lr: 0.000519  loss: 3.3077 (3.1472)  weight_decay: 0.0500 (0.0500)  time: 0.6330  data: 0.0524  max mem: 41794
Epoch: [234]  [130/312]  eta: 0:02:36  lr: 0.000518  min_lr: 0.000518  loss: 3.3965 (3.1667)  weight_decay: 0.0500 (0.0500)  time: 0.8266  data: 0.1293  max mem: 41794
Epoch: [234]  [140/312]  eta: 0:02:23  lr: 0.000518  min_lr: 0.000518  loss: 3.3409 (3.1523)  weight_decay: 0.0500 (0.0500)  time: 0.7119  data: 0.0797  max mem: 41794
Epoch: [234]  [150/312]  eta: 0:02:15  lr: 0.000517  min_lr: 0.000517  loss: 3.2018 (3.1560)  weight_decay: 0.0500 (0.0500)  time: 0.7056  data: 0.0693  max mem: 41794
Epoch: [234]  [160/312]  eta: 0:02:05  lr: 0.000517  min_lr: 0.000517  loss: 3.2018 (3.1553)  weight_decay: 0.0500 (0.0500)  time: 0.7482  data: 0.0751  max mem: 41794
Epoch: [234]  [170/312]  eta: 0:01:56  lr: 0.000516  min_lr: 0.000516  loss: 3.0797 (3.1557)  weight_decay: 0.0500 (0.0500)  time: 0.6583  data: 0.0489  max mem: 41794
Epoch: [234]  [180/312]  eta: 0:01:47  lr: 0.000516  min_lr: 0.000516  loss: 3.2555 (3.1477)  weight_decay: 0.0500 (0.0500)  time: 0.7586  data: 0.1104  max mem: 41794
Epoch: [234]  [190/312]  eta: 0:01:37  lr: 0.000515  min_lr: 0.000515  loss: 3.3679 (3.1582)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.0679  max mem: 41794
Epoch: [234]  [200/312]  eta: 0:01:30  lr: 0.000515  min_lr: 0.000515  loss: 3.3848 (3.1612)  weight_decay: 0.0500 (0.0500)  time: 0.7145  data: 0.1070  max mem: 41794
Epoch: [234]  [210/312]  eta: 0:01:22  lr: 0.000514  min_lr: 0.000514  loss: 3.3730 (3.1653)  weight_decay: 0.0500 (0.0500)  time: 0.8885  data: 0.1829  max mem: 41794
Epoch: [234]  [220/312]  eta: 0:01:13  lr: 0.000514  min_lr: 0.000514  loss: 3.3440 (3.1687)  weight_decay: 0.0500 (0.0500)  time: 0.6790  data: 0.0768  max mem: 41794
Epoch: [234]  [230/312]  eta: 0:01:05  lr: 0.000514  min_lr: 0.000514  loss: 3.1404 (3.1638)  weight_decay: 0.0500 (0.0500)  time: 0.6895  data: 0.1095  max mem: 41794
Epoch: [234]  [240/312]  eta: 0:00:57  lr: 0.000513  min_lr: 0.000513  loss: 3.0990 (3.1549)  weight_decay: 0.0500 (0.0500)  time: 0.7649  data: 0.1122  max mem: 41794
Epoch: [234]  [250/312]  eta: 0:00:48  lr: 0.000513  min_lr: 0.000513  loss: 3.1178 (3.1570)  weight_decay: 0.0500 (0.0500)  time: 0.6789  data: 0.0645  max mem: 41794
Epoch: [234]  [260/312]  eta: 0:00:41  lr: 0.000512  min_lr: 0.000512  loss: 3.1793 (3.1587)  weight_decay: 0.0500 (0.0500)  time: 0.8078  data: 0.1640  max mem: 41794
Epoch: [234]  [270/312]  eta: 0:00:32  lr: 0.000512  min_lr: 0.000512  loss: 3.2989 (3.1614)  weight_decay: 0.0500 (0.0500)  time: 0.7121  data: 0.1051  max mem: 41794
Epoch: [234]  [280/312]  eta: 0:00:25  lr: 0.000511  min_lr: 0.000511  loss: 3.3049 (3.1628)  weight_decay: 0.0500 (0.0500)  time: 0.6430  data: 0.0782  max mem: 41794
Epoch: [234]  [290/312]  eta: 0:00:17  lr: 0.000511  min_lr: 0.000511  loss: 3.0226 (3.1568)  weight_decay: 0.0500 (0.0500)  time: 0.7198  data: 0.1266  max mem: 41794
Epoch: [234]  [300/312]  eta: 0:00:09  lr: 0.000510  min_lr: 0.000510  loss: 3.1543 (3.1599)  weight_decay: 0.0500 (0.0500)  time: 0.5680  data: 0.0508  max mem: 41794
Epoch: [234]  [310/312]  eta: 0:00:01  lr: 0.000510  min_lr: 0.000510  loss: 3.1125 (3.1543)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0001  max mem: 41794
Epoch: [234]  [311/312]  eta: 0:00:00  lr: 0.000510  min_lr: 0.000510  loss: 3.1125 (3.1553)  weight_decay: 0.0500 (0.0500)  time: 0.4628  data: 0.0001  max mem: 41794
Epoch: [234] Total time: 0:03:58 (0.7652 s / it)
Averaged stats: lr: 0.000510  min_lr: 0.000510  loss: 3.1125 (3.1262)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.9430 (0.9430)  acc1: 84.1146 (84.1146)  acc5: 96.3542 (96.3542)  time: 8.3115  data: 8.1123  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2298 (1.1384)  acc1: 79.5573 (78.0800)  acc5: 93.4896 (94.2720)  time: 1.0715  data: 0.9015  max mem: 41794
Test: Total time: 0:00:09 (1.0905 s / it)
* Acc@1 78.242 Acc@5 94.404 loss 1.138
Accuracy of the model on the 50000 test images: 78.2%
Max accuracy: 78.24%
Epoch: [235]  [  0/312]  eta: 1:20:15  lr: 0.000510  min_lr: 0.000510  loss: 3.4646 (3.4646)  weight_decay: 0.0500 (0.0500)  time: 15.4331  data: 14.9510  max mem: 41794
Epoch: [235]  [ 10/312]  eta: 0:11:19  lr: 0.000509  min_lr: 0.000509  loss: 3.2379 (3.0785)  weight_decay: 0.0500 (0.0500)  time: 2.2506  data: 1.3879  max mem: 41794
Epoch: [235]  [ 20/312]  eta: 0:07:10  lr: 0.000509  min_lr: 0.000509  loss: 3.2379 (3.0694)  weight_decay: 0.0500 (0.0500)  time: 0.7780  data: 0.0765  max mem: 41794
Epoch: [235]  [ 30/312]  eta: 0:05:28  lr: 0.000508  min_lr: 0.000508  loss: 3.3172 (3.0979)  weight_decay: 0.0500 (0.0500)  time: 0.5682  data: 0.0610  max mem: 41794
Epoch: [235]  [ 40/312]  eta: 0:04:36  lr: 0.000508  min_lr: 0.000508  loss: 3.3172 (3.1182)  weight_decay: 0.0500 (0.0500)  time: 0.5328  data: 0.0269  max mem: 41794
Epoch: [235]  [ 50/312]  eta: 0:04:09  lr: 0.000507  min_lr: 0.000507  loss: 3.3928 (3.1253)  weight_decay: 0.0500 (0.0500)  time: 0.6196  data: 0.1123  max mem: 41794
Epoch: [235]  [ 60/312]  eta: 0:03:43  lr: 0.000507  min_lr: 0.000507  loss: 3.2694 (3.1164)  weight_decay: 0.0500 (0.0500)  time: 0.6206  data: 0.1201  max mem: 41794
Epoch: [235]  [ 70/312]  eta: 0:03:34  lr: 0.000506  min_lr: 0.000506  loss: 3.2666 (3.1219)  weight_decay: 0.0500 (0.0500)  time: 0.7138  data: 0.2165  max mem: 41794
Epoch: [235]  [ 80/312]  eta: 0:03:22  lr: 0.000506  min_lr: 0.000506  loss: 3.2044 (3.1265)  weight_decay: 0.0500 (0.0500)  time: 0.8336  data: 0.3347  max mem: 41794
Epoch: [235]  [ 90/312]  eta: 0:03:06  lr: 0.000505  min_lr: 0.000505  loss: 3.1591 (3.1140)  weight_decay: 0.0500 (0.0500)  time: 0.6924  data: 0.1995  max mem: 41794
Epoch: [235]  [100/312]  eta: 0:02:58  lr: 0.000505  min_lr: 0.000505  loss: 3.1693 (3.1282)  weight_decay: 0.0500 (0.0500)  time: 0.7266  data: 0.2214  max mem: 41794
Epoch: [235]  [110/312]  eta: 0:02:44  lr: 0.000504  min_lr: 0.000504  loss: 3.1526 (3.1220)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.1747  max mem: 41794
Epoch: [235]  [120/312]  eta: 0:02:37  lr: 0.000504  min_lr: 0.000504  loss: 3.1526 (3.1344)  weight_decay: 0.0500 (0.0500)  time: 0.7085  data: 0.1782  max mem: 41794
Epoch: [235]  [130/312]  eta: 0:02:30  lr: 0.000503  min_lr: 0.000503  loss: 3.1075 (3.1230)  weight_decay: 0.0500 (0.0500)  time: 0.9099  data: 0.3205  max mem: 41794
Epoch: [235]  [140/312]  eta: 0:02:18  lr: 0.000503  min_lr: 0.000503  loss: 3.1843 (3.1326)  weight_decay: 0.0500 (0.0500)  time: 0.7036  data: 0.1430  max mem: 41794
Epoch: [235]  [150/312]  eta: 0:02:11  lr: 0.000502  min_lr: 0.000502  loss: 3.0579 (3.1183)  weight_decay: 0.0500 (0.0500)  time: 0.6985  data: 0.1641  max mem: 41794
Epoch: [235]  [160/312]  eta: 0:02:02  lr: 0.000502  min_lr: 0.000502  loss: 3.0579 (3.1273)  weight_decay: 0.0500 (0.0500)  time: 0.8075  data: 0.2710  max mem: 41794
Epoch: [235]  [170/312]  eta: 0:01:52  lr: 0.000501  min_lr: 0.000501  loss: 3.3420 (3.1362)  weight_decay: 0.0500 (0.0500)  time: 0.6644  data: 0.1481  max mem: 41794
Epoch: [235]  [180/312]  eta: 0:01:45  lr: 0.000501  min_lr: 0.000501  loss: 3.2280 (3.1247)  weight_decay: 0.0500 (0.0500)  time: 0.7233  data: 0.2135  max mem: 41794
Epoch: [235]  [190/312]  eta: 0:01:35  lr: 0.000501  min_lr: 0.000501  loss: 3.2218 (3.1292)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.1905  max mem: 41794
Epoch: [235]  [200/312]  eta: 0:01:28  lr: 0.000500  min_lr: 0.000500  loss: 3.2218 (3.1245)  weight_decay: 0.0500 (0.0500)  time: 0.7019  data: 0.2046  max mem: 41794
Epoch: [235]  [210/312]  eta: 0:01:20  lr: 0.000500  min_lr: 0.000500  loss: 3.1432 (3.1278)  weight_decay: 0.0500 (0.0500)  time: 0.8617  data: 0.3662  max mem: 41794
Epoch: [235]  [220/312]  eta: 0:01:11  lr: 0.000499  min_lr: 0.000499  loss: 3.1432 (3.1270)  weight_decay: 0.0500 (0.0500)  time: 0.6907  data: 0.1927  max mem: 41794
Epoch: [235]  [230/312]  eta: 0:01:04  lr: 0.000499  min_lr: 0.000499  loss: 3.0464 (3.1220)  weight_decay: 0.0500 (0.0500)  time: 0.7047  data: 0.1997  max mem: 41794
Epoch: [235]  [240/312]  eta: 0:00:56  lr: 0.000498  min_lr: 0.000498  loss: 3.0228 (3.1125)  weight_decay: 0.0500 (0.0500)  time: 0.8097  data: 0.3079  max mem: 41794
Epoch: [235]  [250/312]  eta: 0:00:47  lr: 0.000498  min_lr: 0.000498  loss: 3.0578 (3.1170)  weight_decay: 0.0500 (0.0500)  time: 0.6714  data: 0.1802  max mem: 41794
Epoch: [235]  [260/312]  eta: 0:00:40  lr: 0.000497  min_lr: 0.000497  loss: 3.2789 (3.1155)  weight_decay: 0.0500 (0.0500)  time: 0.7274  data: 0.2354  max mem: 41794
Epoch: [235]  [270/312]  eta: 0:00:32  lr: 0.000497  min_lr: 0.000497  loss: 3.2130 (3.1173)  weight_decay: 0.0500 (0.0500)  time: 0.6865  data: 0.1889  max mem: 41794
Epoch: [235]  [280/312]  eta: 0:00:24  lr: 0.000496  min_lr: 0.000496  loss: 3.0992 (3.1096)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.1872  max mem: 41794
Epoch: [235]  [290/312]  eta: 0:00:17  lr: 0.000496  min_lr: 0.000496  loss: 3.2153 (3.1178)  weight_decay: 0.0500 (0.0500)  time: 0.8581  data: 0.3560  max mem: 41794
Epoch: [235]  [300/312]  eta: 0:00:09  lr: 0.000495  min_lr: 0.000495  loss: 3.3991 (3.1170)  weight_decay: 0.0500 (0.0500)  time: 0.6774  data: 0.1834  max mem: 41794
Epoch: [235]  [310/312]  eta: 0:00:01  lr: 0.000495  min_lr: 0.000495  loss: 3.2964 (3.1208)  weight_decay: 0.0500 (0.0500)  time: 0.4699  data: 0.0024  max mem: 41794
Epoch: [235]  [311/312]  eta: 0:00:00  lr: 0.000495  min_lr: 0.000495  loss: 3.3703 (3.1217)  weight_decay: 0.0500 (0.0500)  time: 0.4675  data: 0.0024  max mem: 41794
Epoch: [235] Total time: 0:03:57 (0.7600 s / it)
Averaged stats: lr: 0.000495  min_lr: 0.000495  loss: 3.3703 (3.1343)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.9467 (0.9467)  acc1: 84.6354 (84.6354)  acc5: 96.8750 (96.8750)  time: 8.3141  data: 8.1016  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2068 (1.1322)  acc1: 77.2135 (78.4000)  acc5: 95.3125 (94.6880)  time: 1.0833  data: 0.9003  max mem: 41794
Test: Total time: 0:00:09 (1.0938 s / it)
* Acc@1 78.296 Acc@5 94.520 loss 1.136
Accuracy of the model on the 50000 test images: 78.3%
Max accuracy: 78.30%
Epoch: [236]  [  0/312]  eta: 1:18:45  lr: 0.000495  min_lr: 0.000495  loss: 3.1793 (3.1793)  weight_decay: 0.0500 (0.0500)  time: 15.1470  data: 13.1882  max mem: 41794
Epoch: [236]  [ 10/312]  eta: 0:11:58  lr: 0.000494  min_lr: 0.000494  loss: 3.1793 (3.0187)  weight_decay: 0.0500 (0.0500)  time: 2.3781  data: 1.3596  max mem: 41794
Epoch: [236]  [ 20/312]  eta: 0:07:16  lr: 0.000494  min_lr: 0.000494  loss: 3.1872 (3.0612)  weight_decay: 0.0500 (0.0500)  time: 0.8107  data: 0.0921  max mem: 41794
Epoch: [236]  [ 30/312]  eta: 0:05:31  lr: 0.000493  min_lr: 0.000493  loss: 3.0105 (2.9971)  weight_decay: 0.0500 (0.0500)  time: 0.5159  data: 0.0049  max mem: 41794
Epoch: [236]  [ 40/312]  eta: 0:04:34  lr: 0.000493  min_lr: 0.000493  loss: 3.0105 (3.0518)  weight_decay: 0.0500 (0.0500)  time: 0.5039  data: 0.0047  max mem: 41794
Epoch: [236]  [ 50/312]  eta: 0:04:01  lr: 0.000492  min_lr: 0.000492  loss: 3.2308 (3.0438)  weight_decay: 0.0500 (0.0500)  time: 0.5219  data: 0.0115  max mem: 41794
Epoch: [236]  [ 60/312]  eta: 0:03:35  lr: 0.000492  min_lr: 0.000492  loss: 3.2770 (3.0699)  weight_decay: 0.0500 (0.0500)  time: 0.5346  data: 0.0119  max mem: 41794
Epoch: [236]  [ 70/312]  eta: 0:03:23  lr: 0.000491  min_lr: 0.000491  loss: 3.1722 (3.0375)  weight_decay: 0.0500 (0.0500)  time: 0.6376  data: 0.1258  max mem: 41794
Epoch: [236]  [ 80/312]  eta: 0:03:15  lr: 0.000491  min_lr: 0.000491  loss: 2.9970 (3.0242)  weight_decay: 0.0500 (0.0500)  time: 0.7991  data: 0.2017  max mem: 41794
Epoch: [236]  [ 90/312]  eta: 0:03:02  lr: 0.000490  min_lr: 0.000490  loss: 3.0606 (3.0196)  weight_decay: 0.0500 (0.0500)  time: 0.7571  data: 0.0889  max mem: 41794
Epoch: [236]  [100/312]  eta: 0:02:54  lr: 0.000490  min_lr: 0.000490  loss: 3.3132 (3.0408)  weight_decay: 0.0500 (0.0500)  time: 0.7403  data: 0.0608  max mem: 41794
Epoch: [236]  [110/312]  eta: 0:02:43  lr: 0.000490  min_lr: 0.000490  loss: 3.3132 (3.0492)  weight_decay: 0.0500 (0.0500)  time: 0.7439  data: 0.0566  max mem: 41794
Epoch: [236]  [120/312]  eta: 0:02:34  lr: 0.000489  min_lr: 0.000489  loss: 3.2093 (3.0474)  weight_decay: 0.0500 (0.0500)  time: 0.7105  data: 0.0400  max mem: 41794
Epoch: [236]  [130/312]  eta: 0:02:26  lr: 0.000489  min_lr: 0.000489  loss: 3.3479 (3.0687)  weight_decay: 0.0500 (0.0500)  time: 0.7821  data: 0.0369  max mem: 41794
Epoch: [236]  [140/312]  eta: 0:02:16  lr: 0.000488  min_lr: 0.000488  loss: 3.2925 (3.0649)  weight_decay: 0.0500 (0.0500)  time: 0.7557  data: 0.0179  max mem: 41794
Epoch: [236]  [150/312]  eta: 0:02:07  lr: 0.000488  min_lr: 0.000488  loss: 3.1030 (3.0678)  weight_decay: 0.0500 (0.0500)  time: 0.6936  data: 0.0369  max mem: 41794
Epoch: [236]  [160/312]  eta: 0:02:00  lr: 0.000487  min_lr: 0.000487  loss: 3.1149 (3.0664)  weight_decay: 0.0500 (0.0500)  time: 0.7480  data: 0.0231  max mem: 41794
Epoch: [236]  [170/312]  eta: 0:01:50  lr: 0.000487  min_lr: 0.000487  loss: 3.2609 (3.0808)  weight_decay: 0.0500 (0.0500)  time: 0.7025  data: 0.0322  max mem: 41794
Epoch: [236]  [180/312]  eta: 0:01:43  lr: 0.000486  min_lr: 0.000486  loss: 3.2609 (3.0743)  weight_decay: 0.0500 (0.0500)  time: 0.7179  data: 0.0666  max mem: 41794
Epoch: [236]  [190/312]  eta: 0:01:34  lr: 0.000486  min_lr: 0.000486  loss: 3.0784 (3.0809)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.0370  max mem: 41794
Epoch: [236]  [200/312]  eta: 0:01:26  lr: 0.000485  min_lr: 0.000485  loss: 3.1842 (3.0742)  weight_decay: 0.0500 (0.0500)  time: 0.7035  data: 0.0277  max mem: 41794
Epoch: [236]  [210/312]  eta: 0:01:19  lr: 0.000485  min_lr: 0.000485  loss: 3.2734 (3.0898)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.0686  max mem: 41794
Epoch: [236]  [220/312]  eta: 0:01:10  lr: 0.000484  min_lr: 0.000484  loss: 3.2932 (3.0861)  weight_decay: 0.0500 (0.0500)  time: 0.7409  data: 0.0431  max mem: 41794
Epoch: [236]  [230/312]  eta: 0:01:03  lr: 0.000484  min_lr: 0.000484  loss: 3.2579 (3.0899)  weight_decay: 0.0500 (0.0500)  time: 0.7040  data: 0.0505  max mem: 41794
Epoch: [236]  [240/312]  eta: 0:00:55  lr: 0.000483  min_lr: 0.000483  loss: 3.0727 (3.0810)  weight_decay: 0.0500 (0.0500)  time: 0.7280  data: 0.1102  max mem: 41794
Epoch: [236]  [250/312]  eta: 0:00:47  lr: 0.000483  min_lr: 0.000483  loss: 2.8809 (3.0777)  weight_decay: 0.0500 (0.0500)  time: 0.6987  data: 0.1113  max mem: 41794
Epoch: [236]  [260/312]  eta: 0:00:39  lr: 0.000482  min_lr: 0.000482  loss: 3.1354 (3.0736)  weight_decay: 0.0500 (0.0500)  time: 0.7350  data: 0.1085  max mem: 41794
Epoch: [236]  [270/312]  eta: 0:00:32  lr: 0.000482  min_lr: 0.000482  loss: 3.2774 (3.0757)  weight_decay: 0.0500 (0.0500)  time: 0.7490  data: 0.0753  max mem: 41794
Epoch: [236]  [280/312]  eta: 0:00:24  lr: 0.000482  min_lr: 0.000482  loss: 3.3654 (3.0828)  weight_decay: 0.0500 (0.0500)  time: 0.7236  data: 0.0994  max mem: 41794
Epoch: [236]  [290/312]  eta: 0:00:16  lr: 0.000481  min_lr: 0.000481  loss: 3.3130 (3.0839)  weight_decay: 0.0500 (0.0500)  time: 0.7412  data: 0.1571  max mem: 41794
Epoch: [236]  [300/312]  eta: 0:00:09  lr: 0.000481  min_lr: 0.000481  loss: 3.1004 (3.0820)  weight_decay: 0.0500 (0.0500)  time: 0.6760  data: 0.0961  max mem: 41794
Epoch: [236]  [310/312]  eta: 0:00:01  lr: 0.000480  min_lr: 0.000480  loss: 3.1521 (3.0830)  weight_decay: 0.0500 (0.0500)  time: 0.5406  data: 0.0219  max mem: 41794
Epoch: [236]  [311/312]  eta: 0:00:00  lr: 0.000480  min_lr: 0.000480  loss: 3.1004 (3.0806)  weight_decay: 0.0500 (0.0500)  time: 0.5404  data: 0.0219  max mem: 41794
Epoch: [236] Total time: 0:03:54 (0.7529 s / it)
Averaged stats: lr: 0.000480  min_lr: 0.000480  loss: 3.1004 (3.1167)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.8540 (0.8540)  acc1: 85.4167 (85.4167)  acc5: 96.2240 (96.2240)  time: 8.6643  data: 8.4628  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1556 (1.0748)  acc1: 77.9948 (78.7520)  acc5: 94.4010 (94.0480)  time: 1.1152  data: 0.9404  max mem: 41794
Test: Total time: 0:00:10 (1.1626 s / it)
* Acc@1 78.482 Acc@5 94.440 loss 1.068
Accuracy of the model on the 50000 test images: 78.5%
Max accuracy: 78.48%
Epoch: [237]  [  0/312]  eta: 1:20:38  lr: 0.000480  min_lr: 0.000480  loss: 2.3664 (2.3664)  weight_decay: 0.0500 (0.0500)  time: 15.5071  data: 14.3310  max mem: 41794
Epoch: [237]  [ 10/312]  eta: 0:10:56  lr: 0.000480  min_lr: 0.000480  loss: 2.8283 (2.7954)  weight_decay: 0.0500 (0.0500)  time: 2.1754  data: 1.3036  max mem: 41794
Epoch: [237]  [ 20/312]  eta: 0:07:24  lr: 0.000479  min_lr: 0.000479  loss: 3.2262 (2.9570)  weight_decay: 0.0500 (0.0500)  time: 0.8225  data: 0.0826  max mem: 41794
Epoch: [237]  [ 30/312]  eta: 0:05:38  lr: 0.000479  min_lr: 0.000479  loss: 3.2177 (3.0036)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.0853  max mem: 41794
Epoch: [237]  [ 40/312]  eta: 0:04:53  lr: 0.000478  min_lr: 0.000478  loss: 3.0820 (3.0249)  weight_decay: 0.0500 (0.0500)  time: 0.6116  data: 0.1112  max mem: 41794
Epoch: [237]  [ 50/312]  eta: 0:04:29  lr: 0.000478  min_lr: 0.000478  loss: 3.4230 (3.0907)  weight_decay: 0.0500 (0.0500)  time: 0.7600  data: 0.2574  max mem: 41794
Epoch: [237]  [ 60/312]  eta: 0:03:57  lr: 0.000477  min_lr: 0.000477  loss: 3.4230 (3.0929)  weight_decay: 0.0500 (0.0500)  time: 0.6609  data: 0.1497  max mem: 41794
Epoch: [237]  [ 70/312]  eta: 0:03:43  lr: 0.000477  min_lr: 0.000477  loss: 3.1856 (3.0918)  weight_decay: 0.0500 (0.0500)  time: 0.6569  data: 0.1472  max mem: 41794
Epoch: [237]  [ 80/312]  eta: 0:03:33  lr: 0.000476  min_lr: 0.000476  loss: 3.2427 (3.1154)  weight_decay: 0.0500 (0.0500)  time: 0.8490  data: 0.3411  max mem: 41794
Epoch: [237]  [ 90/312]  eta: 0:03:13  lr: 0.000476  min_lr: 0.000476  loss: 3.3225 (3.1249)  weight_decay: 0.0500 (0.0500)  time: 0.6878  data: 0.1961  max mem: 41794
Epoch: [237]  [100/312]  eta: 0:03:06  lr: 0.000475  min_lr: 0.000475  loss: 3.0775 (3.1096)  weight_decay: 0.0500 (0.0500)  time: 0.7151  data: 0.2159  max mem: 41794
Epoch: [237]  [110/312]  eta: 0:02:50  lr: 0.000475  min_lr: 0.000475  loss: 2.9536 (3.1072)  weight_decay: 0.0500 (0.0500)  time: 0.7167  data: 0.2146  max mem: 41794
Epoch: [237]  [120/312]  eta: 0:02:43  lr: 0.000474  min_lr: 0.000474  loss: 3.2341 (3.1090)  weight_decay: 0.0500 (0.0500)  time: 0.7129  data: 0.2152  max mem: 41794
Epoch: [237]  [130/312]  eta: 0:02:35  lr: 0.000474  min_lr: 0.000474  loss: 3.1664 (3.1077)  weight_decay: 0.0500 (0.0500)  time: 0.9081  data: 0.3987  max mem: 41794
Epoch: [237]  [140/312]  eta: 0:02:22  lr: 0.000474  min_lr: 0.000474  loss: 3.1664 (3.1113)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.1842  max mem: 41794
Epoch: [237]  [150/312]  eta: 0:02:15  lr: 0.000473  min_lr: 0.000473  loss: 3.2787 (3.1182)  weight_decay: 0.0500 (0.0500)  time: 0.7077  data: 0.2095  max mem: 41794
Epoch: [237]  [160/312]  eta: 0:02:07  lr: 0.000473  min_lr: 0.000473  loss: 3.0180 (3.1083)  weight_decay: 0.0500 (0.0500)  time: 0.8934  data: 0.3912  max mem: 41794
Epoch: [237]  [170/312]  eta: 0:01:55  lr: 0.000472  min_lr: 0.000472  loss: 3.2430 (3.1198)  weight_decay: 0.0500 (0.0500)  time: 0.6770  data: 0.1823  max mem: 41794
Epoch: [237]  [180/312]  eta: 0:01:47  lr: 0.000472  min_lr: 0.000472  loss: 3.3741 (3.1379)  weight_decay: 0.0500 (0.0500)  time: 0.6299  data: 0.1375  max mem: 41794
Epoch: [237]  [190/312]  eta: 0:01:37  lr: 0.000471  min_lr: 0.000471  loss: 3.3978 (3.1379)  weight_decay: 0.0500 (0.0500)  time: 0.6302  data: 0.1375  max mem: 41794
Epoch: [237]  [200/312]  eta: 0:01:29  lr: 0.000471  min_lr: 0.000471  loss: 3.0538 (3.1345)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.1760  max mem: 41794
Epoch: [237]  [210/312]  eta: 0:01:21  lr: 0.000470  min_lr: 0.000470  loss: 3.1149 (3.1337)  weight_decay: 0.0500 (0.0500)  time: 0.8599  data: 0.3664  max mem: 41794
Epoch: [237]  [220/312]  eta: 0:01:12  lr: 0.000470  min_lr: 0.000470  loss: 3.3247 (3.1398)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.1910  max mem: 41794
Epoch: [237]  [230/312]  eta: 0:01:04  lr: 0.000469  min_lr: 0.000469  loss: 3.1838 (3.1244)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1939  max mem: 41794
Epoch: [237]  [240/312]  eta: 0:00:57  lr: 0.000469  min_lr: 0.000469  loss: 3.1423 (3.1298)  weight_decay: 0.0500 (0.0500)  time: 0.8698  data: 0.3696  max mem: 41794
Epoch: [237]  [250/312]  eta: 0:00:48  lr: 0.000468  min_lr: 0.000468  loss: 3.1423 (3.1274)  weight_decay: 0.0500 (0.0500)  time: 0.6758  data: 0.1763  max mem: 41794
Epoch: [237]  [260/312]  eta: 0:00:40  lr: 0.000468  min_lr: 0.000468  loss: 3.0851 (3.1268)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1927  max mem: 41794
Epoch: [237]  [270/312]  eta: 0:00:32  lr: 0.000468  min_lr: 0.000468  loss: 3.0291 (3.1151)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.1927  max mem: 41794
Epoch: [237]  [280/312]  eta: 0:00:24  lr: 0.000467  min_lr: 0.000467  loss: 2.6266 (3.1077)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.1886  max mem: 41794
Epoch: [237]  [290/312]  eta: 0:00:17  lr: 0.000467  min_lr: 0.000467  loss: 2.9982 (3.1022)  weight_decay: 0.0500 (0.0500)  time: 0.8249  data: 0.3361  max mem: 41794
Epoch: [237]  [300/312]  eta: 0:00:09  lr: 0.000466  min_lr: 0.000466  loss: 3.0477 (3.0975)  weight_decay: 0.0500 (0.0500)  time: 0.6327  data: 0.1479  max mem: 41794
Epoch: [237]  [310/312]  eta: 0:00:01  lr: 0.000466  min_lr: 0.000466  loss: 3.1209 (3.0998)  weight_decay: 0.0500 (0.0500)  time: 0.4675  data: 0.0001  max mem: 41794
Epoch: [237]  [311/312]  eta: 0:00:00  lr: 0.000466  min_lr: 0.000466  loss: 3.1209 (3.1000)  weight_decay: 0.0500 (0.0500)  time: 0.4647  data: 0.0001  max mem: 41794
Epoch: [237] Total time: 0:03:58 (0.7645 s / it)
Averaged stats: lr: 0.000466  min_lr: 0.000466  loss: 3.1209 (3.1039)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 0.9197 (0.9197)  acc1: 83.8542 (83.8542)  acc5: 96.3542 (96.3542)  time: 7.9716  data: 7.7596  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1769 (1.1084)  acc1: 78.7760 (78.4160)  acc5: 94.2708 (94.2080)  time: 1.0498  data: 0.8784  max mem: 41794
Test: Total time: 0:00:09 (1.0608 s / it)
* Acc@1 78.580 Acc@5 94.362 loss 1.106
Accuracy of the model on the 50000 test images: 78.6%
Max accuracy: 78.58%
Epoch: [238]  [  0/312]  eta: 1:20:11  lr: 0.000466  min_lr: 0.000466  loss: 3.3191 (3.3191)  weight_decay: 0.0500 (0.0500)  time: 15.4219  data: 12.5888  max mem: 41794
Epoch: [238]  [ 10/312]  eta: 0:10:48  lr: 0.000465  min_lr: 0.000465  loss: 3.2820 (3.1422)  weight_decay: 0.0500 (0.0500)  time: 2.1458  data: 1.3885  max mem: 41794
Epoch: [238]  [ 20/312]  eta: 0:07:37  lr: 0.000465  min_lr: 0.000465  loss: 3.2349 (3.1362)  weight_decay: 0.0500 (0.0500)  time: 0.8724  data: 0.2424  max mem: 41794
Epoch: [238]  [ 30/312]  eta: 0:05:48  lr: 0.000464  min_lr: 0.000464  loss: 3.2123 (3.0872)  weight_decay: 0.0500 (0.0500)  time: 0.7377  data: 0.1091  max mem: 41794
Epoch: [238]  [ 40/312]  eta: 0:04:51  lr: 0.000464  min_lr: 0.000464  loss: 3.3231 (3.1224)  weight_decay: 0.0500 (0.0500)  time: 0.5509  data: 0.0280  max mem: 41794
Epoch: [238]  [ 50/312]  eta: 0:04:30  lr: 0.000463  min_lr: 0.000463  loss: 3.3603 (3.1259)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.0972  max mem: 41794
Epoch: [238]  [ 60/312]  eta: 0:03:58  lr: 0.000463  min_lr: 0.000463  loss: 3.2027 (3.1254)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.0704  max mem: 41794
Epoch: [238]  [ 70/312]  eta: 0:03:42  lr: 0.000462  min_lr: 0.000462  loss: 3.3478 (3.1588)  weight_decay: 0.0500 (0.0500)  time: 0.6314  data: 0.0854  max mem: 41794
Epoch: [238]  [ 80/312]  eta: 0:03:30  lr: 0.000462  min_lr: 0.000462  loss: 3.3756 (3.1686)  weight_decay: 0.0500 (0.0500)  time: 0.7858  data: 0.1929  max mem: 41794
Epoch: [238]  [ 90/312]  eta: 0:03:13  lr: 0.000461  min_lr: 0.000461  loss: 3.2753 (3.1534)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.1399  max mem: 41794
Epoch: [238]  [100/312]  eta: 0:03:04  lr: 0.000461  min_lr: 0.000461  loss: 3.4002 (3.1824)  weight_decay: 0.0500 (0.0500)  time: 0.7113  data: 0.1490  max mem: 41794
Epoch: [238]  [110/312]  eta: 0:02:49  lr: 0.000461  min_lr: 0.000461  loss: 3.3195 (3.1767)  weight_decay: 0.0500 (0.0500)  time: 0.6887  data: 0.1335  max mem: 41794
Epoch: [238]  [120/312]  eta: 0:02:40  lr: 0.000460  min_lr: 0.000460  loss: 3.1471 (3.1627)  weight_decay: 0.0500 (0.0500)  time: 0.6870  data: 0.1379  max mem: 41794
Epoch: [238]  [130/312]  eta: 0:02:33  lr: 0.000460  min_lr: 0.000460  loss: 3.3254 (3.1715)  weight_decay: 0.0500 (0.0500)  time: 0.8609  data: 0.2312  max mem: 41794
Epoch: [238]  [140/312]  eta: 0:02:20  lr: 0.000459  min_lr: 0.000459  loss: 3.4149 (3.1806)  weight_decay: 0.0500 (0.0500)  time: 0.7059  data: 0.1328  max mem: 41794
Epoch: [238]  [150/312]  eta: 0:02:13  lr: 0.000459  min_lr: 0.000459  loss: 3.1332 (3.1653)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.1336  max mem: 41794
Epoch: [238]  [160/312]  eta: 0:02:03  lr: 0.000458  min_lr: 0.000458  loss: 3.0399 (3.1516)  weight_decay: 0.0500 (0.0500)  time: 0.7511  data: 0.1267  max mem: 41794
Epoch: [238]  [170/312]  eta: 0:01:53  lr: 0.000458  min_lr: 0.000458  loss: 3.2468 (3.1564)  weight_decay: 0.0500 (0.0500)  time: 0.6575  data: 0.1115  max mem: 41794
Epoch: [238]  [180/312]  eta: 0:01:46  lr: 0.000457  min_lr: 0.000457  loss: 3.1418 (3.1452)  weight_decay: 0.0500 (0.0500)  time: 0.7791  data: 0.2228  max mem: 41794
Epoch: [238]  [190/312]  eta: 0:01:36  lr: 0.000457  min_lr: 0.000457  loss: 3.0518 (3.1356)  weight_decay: 0.0500 (0.0500)  time: 0.6936  data: 0.1277  max mem: 41794
Epoch: [238]  [200/312]  eta: 0:01:29  lr: 0.000456  min_lr: 0.000456  loss: 3.1398 (3.1357)  weight_decay: 0.0500 (0.0500)  time: 0.7164  data: 0.1426  max mem: 41794
Epoch: [238]  [210/312]  eta: 0:01:21  lr: 0.000456  min_lr: 0.000456  loss: 3.2040 (3.1416)  weight_decay: 0.0500 (0.0500)  time: 0.9008  data: 0.2689  max mem: 41794
Epoch: [238]  [220/312]  eta: 0:01:12  lr: 0.000456  min_lr: 0.000456  loss: 3.3191 (3.1467)  weight_decay: 0.0500 (0.0500)  time: 0.6837  data: 0.1271  max mem: 41794
Epoch: [238]  [230/312]  eta: 0:01:04  lr: 0.000455  min_lr: 0.000455  loss: 3.2916 (3.1472)  weight_decay: 0.0500 (0.0500)  time: 0.6777  data: 0.1183  max mem: 41794
Epoch: [238]  [240/312]  eta: 0:00:56  lr: 0.000455  min_lr: 0.000455  loss: 3.1670 (3.1451)  weight_decay: 0.0500 (0.0500)  time: 0.7347  data: 0.1185  max mem: 41794
Epoch: [238]  [250/312]  eta: 0:00:48  lr: 0.000454  min_lr: 0.000454  loss: 3.0560 (3.1359)  weight_decay: 0.0500 (0.0500)  time: 0.6560  data: 0.1045  max mem: 41794
Epoch: [238]  [260/312]  eta: 0:00:40  lr: 0.000454  min_lr: 0.000454  loss: 3.0333 (3.1338)  weight_decay: 0.0500 (0.0500)  time: 0.8093  data: 0.2474  max mem: 41794
Epoch: [238]  [270/312]  eta: 0:00:32  lr: 0.000453  min_lr: 0.000453  loss: 3.1760 (3.1367)  weight_decay: 0.0500 (0.0500)  time: 0.7154  data: 0.1438  max mem: 41794
Epoch: [238]  [280/312]  eta: 0:00:24  lr: 0.000453  min_lr: 0.000453  loss: 3.1962 (3.1377)  weight_decay: 0.0500 (0.0500)  time: 0.6657  data: 0.1201  max mem: 41794
Epoch: [238]  [290/312]  eta: 0:00:17  lr: 0.000452  min_lr: 0.000452  loss: 2.8859 (3.1265)  weight_decay: 0.0500 (0.0500)  time: 0.8300  data: 0.2217  max mem: 41794
Epoch: [238]  [300/312]  eta: 0:00:09  lr: 0.000452  min_lr: 0.000452  loss: 2.8253 (3.1200)  weight_decay: 0.0500 (0.0500)  time: 0.6551  data: 0.1020  max mem: 41794
Epoch: [238]  [310/312]  eta: 0:00:01  lr: 0.000451  min_lr: 0.000451  loss: 3.2479 (3.1236)  weight_decay: 0.0500 (0.0500)  time: 0.4652  data: 0.0001  max mem: 41794
Epoch: [238]  [311/312]  eta: 0:00:00  lr: 0.000451  min_lr: 0.000451  loss: 3.3602 (3.1244)  weight_decay: 0.0500 (0.0500)  time: 0.4627  data: 0.0001  max mem: 41794
Epoch: [238] Total time: 0:03:58 (0.7639 s / it)
Averaged stats: lr: 0.000451  min_lr: 0.000451  loss: 3.3602 (3.1162)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.9696 (0.9696)  acc1: 85.2865 (85.2865)  acc5: 97.0052 (97.0052)  time: 8.8472  data: 8.6374  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2447 (1.1548)  acc1: 77.7344 (78.2400)  acc5: 94.5312 (94.4000)  time: 1.1317  data: 0.9598  max mem: 41794
Test: Total time: 0:00:10 (1.1657 s / it)
* Acc@1 78.558 Acc@5 94.534 loss 1.160
Accuracy of the model on the 50000 test images: 78.6%
Max accuracy: 78.58%
Epoch: [239]  [  0/312]  eta: 1:25:12  lr: 0.000451  min_lr: 0.000451  loss: 3.3758 (3.3758)  weight_decay: 0.0500 (0.0500)  time: 16.3859  data: 14.8199  max mem: 41794
Epoch: [239]  [ 10/312]  eta: 0:10:19  lr: 0.000451  min_lr: 0.000451  loss: 3.3044 (3.0831)  weight_decay: 0.0500 (0.0500)  time: 2.0500  data: 1.3478  max mem: 41794
Epoch: [239]  [ 20/312]  eta: 0:07:15  lr: 0.000450  min_lr: 0.000450  loss: 3.1745 (3.0756)  weight_decay: 0.0500 (0.0500)  time: 0.7455  data: 0.0829  max mem: 41794
Epoch: [239]  [ 30/312]  eta: 0:05:34  lr: 0.000450  min_lr: 0.000450  loss: 3.1745 (3.0979)  weight_decay: 0.0500 (0.0500)  time: 0.7116  data: 0.0943  max mem: 41794
Epoch: [239]  [ 40/312]  eta: 0:04:59  lr: 0.000449  min_lr: 0.000449  loss: 3.3435 (3.1724)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.0999  max mem: 41794
Epoch: [239]  [ 50/312]  eta: 0:04:38  lr: 0.000449  min_lr: 0.000449  loss: 3.3435 (3.1165)  weight_decay: 0.0500 (0.0500)  time: 0.8682  data: 0.2292  max mem: 41794
Epoch: [239]  [ 60/312]  eta: 0:04:04  lr: 0.000449  min_lr: 0.000449  loss: 2.8710 (3.0583)  weight_decay: 0.0500 (0.0500)  time: 0.7027  data: 0.1414  max mem: 41794
Epoch: [239]  [ 70/312]  eta: 0:03:51  lr: 0.000448  min_lr: 0.000448  loss: 3.0616 (3.1090)  weight_decay: 0.0500 (0.0500)  time: 0.6913  data: 0.1273  max mem: 41794
Epoch: [239]  [ 80/312]  eta: 0:03:36  lr: 0.000448  min_lr: 0.000448  loss: 3.3754 (3.1201)  weight_decay: 0.0500 (0.0500)  time: 0.8189  data: 0.1932  max mem: 41794
Epoch: [239]  [ 90/312]  eta: 0:03:17  lr: 0.000447  min_lr: 0.000447  loss: 3.3385 (3.1336)  weight_decay: 0.0500 (0.0500)  time: 0.6565  data: 0.0786  max mem: 41794
Epoch: [239]  [100/312]  eta: 0:03:10  lr: 0.000447  min_lr: 0.000447  loss: 3.2784 (3.1373)  weight_decay: 0.0500 (0.0500)  time: 0.7484  data: 0.1211  max mem: 41794
Epoch: [239]  [110/312]  eta: 0:02:54  lr: 0.000446  min_lr: 0.000446  loss: 3.1953 (3.1426)  weight_decay: 0.0500 (0.0500)  time: 0.7331  data: 0.1144  max mem: 41794
Epoch: [239]  [120/312]  eta: 0:02:46  lr: 0.000446  min_lr: 0.000446  loss: 3.2313 (3.1473)  weight_decay: 0.0500 (0.0500)  time: 0.7037  data: 0.0857  max mem: 41794
Epoch: [239]  [130/312]  eta: 0:02:36  lr: 0.000445  min_lr: 0.000445  loss: 3.2637 (3.1458)  weight_decay: 0.0500 (0.0500)  time: 0.8571  data: 0.1039  max mem: 41794
Epoch: [239]  [140/312]  eta: 0:02:25  lr: 0.000445  min_lr: 0.000445  loss: 3.0423 (3.1375)  weight_decay: 0.0500 (0.0500)  time: 0.7166  data: 0.0739  max mem: 41794
Epoch: [239]  [150/312]  eta: 0:02:16  lr: 0.000445  min_lr: 0.000445  loss: 3.2615 (3.1447)  weight_decay: 0.0500 (0.0500)  time: 0.6964  data: 0.0887  max mem: 41794
Epoch: [239]  [160/312]  eta: 0:02:06  lr: 0.000444  min_lr: 0.000444  loss: 3.3453 (3.1533)  weight_decay: 0.0500 (0.0500)  time: 0.7385  data: 0.0437  max mem: 41794
Epoch: [239]  [170/312]  eta: 0:01:56  lr: 0.000444  min_lr: 0.000444  loss: 3.3681 (3.1679)  weight_decay: 0.0500 (0.0500)  time: 0.6541  data: 0.0412  max mem: 41794
Epoch: [239]  [180/312]  eta: 0:01:48  lr: 0.000443  min_lr: 0.000443  loss: 3.4028 (3.1708)  weight_decay: 0.0500 (0.0500)  time: 0.7775  data: 0.0998  max mem: 41794
Epoch: [239]  [190/312]  eta: 0:01:38  lr: 0.000443  min_lr: 0.000443  loss: 3.2929 (3.1676)  weight_decay: 0.0500 (0.0500)  time: 0.7490  data: 0.0840  max mem: 41794
Epoch: [239]  [200/312]  eta: 0:01:30  lr: 0.000442  min_lr: 0.000442  loss: 3.1282 (3.1594)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.0723  max mem: 41794
Epoch: [239]  [210/312]  eta: 0:01:22  lr: 0.000442  min_lr: 0.000442  loss: 3.1282 (3.1532)  weight_decay: 0.0500 (0.0500)  time: 0.7685  data: 0.0563  max mem: 41794
Epoch: [239]  [220/312]  eta: 0:01:13  lr: 0.000441  min_lr: 0.000441  loss: 3.1805 (3.1535)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.0641  max mem: 41794
Epoch: [239]  [230/312]  eta: 0:01:05  lr: 0.000441  min_lr: 0.000441  loss: 3.2703 (3.1574)  weight_decay: 0.0500 (0.0500)  time: 0.7027  data: 0.1147  max mem: 41794
Epoch: [239]  [240/312]  eta: 0:00:57  lr: 0.000440  min_lr: 0.000440  loss: 3.3496 (3.1647)  weight_decay: 0.0500 (0.0500)  time: 0.7429  data: 0.0605  max mem: 41794
Epoch: [239]  [250/312]  eta: 0:00:48  lr: 0.000440  min_lr: 0.000440  loss: 3.3371 (3.1618)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.0622  max mem: 41794
Epoch: [239]  [260/312]  eta: 0:00:41  lr: 0.000440  min_lr: 0.000440  loss: 3.0177 (3.1519)  weight_decay: 0.0500 (0.0500)  time: 0.7519  data: 0.1299  max mem: 41794
Epoch: [239]  [270/312]  eta: 0:00:32  lr: 0.000439  min_lr: 0.000439  loss: 3.0177 (3.1482)  weight_decay: 0.0500 (0.0500)  time: 0.6760  data: 0.0736  max mem: 41794
Epoch: [239]  [280/312]  eta: 0:00:25  lr: 0.000439  min_lr: 0.000439  loss: 3.1854 (3.1475)  weight_decay: 0.0500 (0.0500)  time: 0.6589  data: 0.0804  max mem: 41794
Epoch: [239]  [290/312]  eta: 0:00:17  lr: 0.000438  min_lr: 0.000438  loss: 3.3966 (3.1530)  weight_decay: 0.0500 (0.0500)  time: 0.7375  data: 0.1090  max mem: 41794
Epoch: [239]  [300/312]  eta: 0:00:09  lr: 0.000438  min_lr: 0.000438  loss: 3.3460 (3.1526)  weight_decay: 0.0500 (0.0500)  time: 0.5877  data: 0.0459  max mem: 41794
Epoch: [239]  [310/312]  eta: 0:00:01  lr: 0.000437  min_lr: 0.000437  loss: 3.2068 (3.1465)  weight_decay: 0.0500 (0.0500)  time: 0.4794  data: 0.0170  max mem: 41794
Epoch: [239]  [311/312]  eta: 0:00:00  lr: 0.000437  min_lr: 0.000437  loss: 3.2133 (3.1469)  weight_decay: 0.0500 (0.0500)  time: 0.4630  data: 0.0001  max mem: 41794
Epoch: [239] Total time: 0:03:58 (0.7640 s / it)
Averaged stats: lr: 0.000437  min_lr: 0.000437  loss: 3.2133 (3.1066)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.8116 (0.8116)  acc1: 85.5469 (85.5469)  acc5: 96.7448 (96.7448)  time: 8.3024  data: 8.0945  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1262 (1.0435)  acc1: 77.7344 (78.4800)  acc5: 94.7917 (94.6400)  time: 1.0894  data: 0.9183  max mem: 41794
Test: Total time: 0:00:09 (1.1006 s / it)
* Acc@1 78.546 Acc@5 94.596 loss 1.046
Accuracy of the model on the 50000 test images: 78.5%
Max accuracy: 78.58%
Epoch: [240]  [  0/312]  eta: 1:12:11  lr: 0.000437  min_lr: 0.000437  loss: 2.7207 (2.7207)  weight_decay: 0.0500 (0.0500)  time: 13.8827  data: 11.6027  max mem: 41794
Epoch: [240]  [ 10/312]  eta: 0:10:00  lr: 0.000437  min_lr: 0.000437  loss: 3.1792 (3.1856)  weight_decay: 0.0500 (0.0500)  time: 1.9887  data: 1.2884  max mem: 41794
Epoch: [240]  [ 20/312]  eta: 0:07:21  lr: 0.000436  min_lr: 0.000436  loss: 3.2375 (3.1660)  weight_decay: 0.0500 (0.0500)  time: 0.8945  data: 0.2022  max mem: 41794
Epoch: [240]  [ 30/312]  eta: 0:05:38  lr: 0.000436  min_lr: 0.000436  loss: 3.3093 (3.1711)  weight_decay: 0.0500 (0.0500)  time: 0.7692  data: 0.0758  max mem: 41794
Epoch: [240]  [ 40/312]  eta: 0:04:58  lr: 0.000435  min_lr: 0.000435  loss: 3.1952 (3.1389)  weight_decay: 0.0500 (0.0500)  time: 0.6619  data: 0.0439  max mem: 41794
Epoch: [240]  [ 50/312]  eta: 0:04:34  lr: 0.000435  min_lr: 0.000435  loss: 3.0969 (3.1369)  weight_decay: 0.0500 (0.0500)  time: 0.8053  data: 0.0959  max mem: 41794
Epoch: [240]  [ 60/312]  eta: 0:04:02  lr: 0.000435  min_lr: 0.000435  loss: 3.2702 (3.1522)  weight_decay: 0.0500 (0.0500)  time: 0.6874  data: 0.0544  max mem: 41794
Epoch: [240]  [ 70/312]  eta: 0:03:51  lr: 0.000434  min_lr: 0.000434  loss: 3.2702 (3.1422)  weight_decay: 0.0500 (0.0500)  time: 0.7272  data: 0.0710  max mem: 41794
Epoch: [240]  [ 80/312]  eta: 0:03:33  lr: 0.000434  min_lr: 0.000434  loss: 3.1688 (3.1290)  weight_decay: 0.0500 (0.0500)  time: 0.7889  data: 0.0710  max mem: 41794
Epoch: [240]  [ 90/312]  eta: 0:03:17  lr: 0.000433  min_lr: 0.000433  loss: 3.1397 (3.1214)  weight_decay: 0.0500 (0.0500)  time: 0.6516  data: 0.0459  max mem: 41794
Epoch: [240]  [100/312]  eta: 0:03:08  lr: 0.000433  min_lr: 0.000433  loss: 3.1825 (3.1113)  weight_decay: 0.0500 (0.0500)  time: 0.7517  data: 0.0903  max mem: 41794
Epoch: [240]  [110/312]  eta: 0:02:52  lr: 0.000432  min_lr: 0.000432  loss: 3.2641 (3.1106)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.0450  max mem: 41794
Epoch: [240]  [120/312]  eta: 0:02:44  lr: 0.000432  min_lr: 0.000432  loss: 3.2641 (3.1082)  weight_decay: 0.0500 (0.0500)  time: 0.7109  data: 0.0317  max mem: 41794
Epoch: [240]  [130/312]  eta: 0:02:35  lr: 0.000431  min_lr: 0.000431  loss: 3.0423 (3.0973)  weight_decay: 0.0500 (0.0500)  time: 0.8809  data: 0.0498  max mem: 41794
Epoch: [240]  [140/312]  eta: 0:02:22  lr: 0.000431  min_lr: 0.000431  loss: 2.9645 (3.0881)  weight_decay: 0.0500 (0.0500)  time: 0.6681  data: 0.0188  max mem: 41794
Epoch: [240]  [150/312]  eta: 0:02:14  lr: 0.000431  min_lr: 0.000431  loss: 3.0724 (3.0945)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0437  max mem: 41794
Epoch: [240]  [160/312]  eta: 0:02:05  lr: 0.000430  min_lr: 0.000430  loss: 3.3035 (3.0948)  weight_decay: 0.0500 (0.0500)  time: 0.7901  data: 0.0931  max mem: 41794
Epoch: [240]  [170/312]  eta: 0:01:55  lr: 0.000430  min_lr: 0.000430  loss: 3.3400 (3.1014)  weight_decay: 0.0500 (0.0500)  time: 0.6654  data: 0.0818  max mem: 41794
Epoch: [240]  [180/312]  eta: 0:01:47  lr: 0.000429  min_lr: 0.000429  loss: 3.3400 (3.1012)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.1073  max mem: 41794
Epoch: [240]  [190/312]  eta: 0:01:36  lr: 0.000429  min_lr: 0.000429  loss: 3.3368 (3.0964)  weight_decay: 0.0500 (0.0500)  time: 0.6269  data: 0.0755  max mem: 41794
Epoch: [240]  [200/312]  eta: 0:01:29  lr: 0.000428  min_lr: 0.000428  loss: 2.9267 (3.0874)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.1418  max mem: 41794
Epoch: [240]  [210/312]  eta: 0:01:21  lr: 0.000428  min_lr: 0.000428  loss: 3.0291 (3.0949)  weight_decay: 0.0500 (0.0500)  time: 0.8411  data: 0.2737  max mem: 41794
Epoch: [240]  [220/312]  eta: 0:01:12  lr: 0.000427  min_lr: 0.000427  loss: 3.1863 (3.0918)  weight_decay: 0.0500 (0.0500)  time: 0.6813  data: 0.1326  max mem: 41794
Epoch: [240]  [230/312]  eta: 0:01:04  lr: 0.000427  min_lr: 0.000427  loss: 3.2288 (3.1027)  weight_decay: 0.0500 (0.0500)  time: 0.6916  data: 0.1319  max mem: 41794
Epoch: [240]  [240/312]  eta: 0:00:56  lr: 0.000427  min_lr: 0.000427  loss: 3.3363 (3.1026)  weight_decay: 0.0500 (0.0500)  time: 0.8192  data: 0.2678  max mem: 41794
Epoch: [240]  [250/312]  eta: 0:00:48  lr: 0.000426  min_lr: 0.000426  loss: 3.2952 (3.1029)  weight_decay: 0.0500 (0.0500)  time: 0.7216  data: 0.1576  max mem: 41794
Epoch: [240]  [260/312]  eta: 0:00:40  lr: 0.000426  min_lr: 0.000426  loss: 3.1885 (3.1026)  weight_decay: 0.0500 (0.0500)  time: 0.7234  data: 0.1406  max mem: 41794
Epoch: [240]  [270/312]  eta: 0:00:32  lr: 0.000425  min_lr: 0.000425  loss: 3.1885 (3.0970)  weight_decay: 0.0500 (0.0500)  time: 0.6657  data: 0.1196  max mem: 41794
Epoch: [240]  [280/312]  eta: 0:00:24  lr: 0.000425  min_lr: 0.000425  loss: 3.1088 (3.0933)  weight_decay: 0.0500 (0.0500)  time: 0.6942  data: 0.1648  max mem: 41794
Epoch: [240]  [290/312]  eta: 0:00:17  lr: 0.000424  min_lr: 0.000424  loss: 3.1362 (3.0961)  weight_decay: 0.0500 (0.0500)  time: 0.8268  data: 0.3034  max mem: 41794
Epoch: [240]  [300/312]  eta: 0:00:09  lr: 0.000424  min_lr: 0.000424  loss: 3.2472 (3.0972)  weight_decay: 0.0500 (0.0500)  time: 0.6269  data: 0.1390  max mem: 41794
Epoch: [240]  [310/312]  eta: 0:00:01  lr: 0.000423  min_lr: 0.000423  loss: 3.2472 (3.0987)  weight_decay: 0.0500 (0.0500)  time: 0.4642  data: 0.0001  max mem: 41794
Epoch: [240]  [311/312]  eta: 0:00:00  lr: 0.000423  min_lr: 0.000423  loss: 3.2644 (3.0993)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [240] Total time: 0:03:58 (0.7638 s / it)
Averaged stats: lr: 0.000423  min_lr: 0.000423  loss: 3.2644 (3.1009)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.9210 (0.9210)  acc1: 84.2448 (84.2448)  acc5: 96.6146 (96.6146)  time: 8.5174  data: 8.3053  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1429 (1.0938)  acc1: 77.8646 (78.5920)  acc5: 94.5312 (94.7840)  time: 1.0945  data: 0.9229  max mem: 41794
Test: Total time: 0:00:10 (1.1316 s / it)
* Acc@1 78.588 Acc@5 94.600 loss 1.096
Accuracy of the model on the 50000 test images: 78.6%
Max accuracy: 78.59%
Epoch: [241]  [  0/312]  eta: 1:22:08  lr: 0.000423  min_lr: 0.000423  loss: 2.0433 (2.0433)  weight_decay: 0.0500 (0.0500)  time: 15.7958  data: 15.2951  max mem: 41794
Epoch: [241]  [ 10/312]  eta: 0:10:19  lr: 0.000423  min_lr: 0.000423  loss: 3.4306 (3.2619)  weight_decay: 0.0500 (0.0500)  time: 2.0515  data: 1.3911  max mem: 41794
Epoch: [241]  [ 20/312]  eta: 0:07:12  lr: 0.000422  min_lr: 0.000422  loss: 3.3503 (3.0785)  weight_decay: 0.0500 (0.0500)  time: 0.7667  data: 0.1611  max mem: 41794
Epoch: [241]  [ 30/312]  eta: 0:05:29  lr: 0.000422  min_lr: 0.000422  loss: 3.1256 (3.1304)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1616  max mem: 41794
Epoch: [241]  [ 40/312]  eta: 0:04:49  lr: 0.000422  min_lr: 0.000422  loss: 3.3871 (3.1878)  weight_decay: 0.0500 (0.0500)  time: 0.6240  data: 0.1241  max mem: 41794
Epoch: [241]  [ 50/312]  eta: 0:04:24  lr: 0.000421  min_lr: 0.000421  loss: 3.3092 (3.1294)  weight_decay: 0.0500 (0.0500)  time: 0.7587  data: 0.2194  max mem: 41794
Epoch: [241]  [ 60/312]  eta: 0:03:52  lr: 0.000421  min_lr: 0.000421  loss: 3.2152 (3.1311)  weight_decay: 0.0500 (0.0500)  time: 0.6355  data: 0.0966  max mem: 41794
Epoch: [241]  [ 70/312]  eta: 0:03:43  lr: 0.000420  min_lr: 0.000420  loss: 3.0973 (3.1058)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.0929  max mem: 41794
Epoch: [241]  [ 80/312]  eta: 0:03:31  lr: 0.000420  min_lr: 0.000420  loss: 2.9720 (3.1026)  weight_decay: 0.0500 (0.0500)  time: 0.8827  data: 0.1712  max mem: 41794
Epoch: [241]  [ 90/312]  eta: 0:03:15  lr: 0.000419  min_lr: 0.000419  loss: 3.2280 (3.1027)  weight_decay: 0.0500 (0.0500)  time: 0.7378  data: 0.1416  max mem: 41794
Epoch: [241]  [100/312]  eta: 0:03:06  lr: 0.000419  min_lr: 0.000419  loss: 3.3132 (3.1078)  weight_decay: 0.0500 (0.0500)  time: 0.7551  data: 0.1434  max mem: 41794
Epoch: [241]  [110/312]  eta: 0:02:51  lr: 0.000418  min_lr: 0.000418  loss: 3.3468 (3.1193)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.0961  max mem: 41794
Epoch: [241]  [120/312]  eta: 0:02:43  lr: 0.000418  min_lr: 0.000418  loss: 3.3432 (3.1266)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.0938  max mem: 41794
Epoch: [241]  [130/312]  eta: 0:02:32  lr: 0.000418  min_lr: 0.000418  loss: 3.3804 (3.1274)  weight_decay: 0.0500 (0.0500)  time: 0.7848  data: 0.1032  max mem: 41794
Epoch: [241]  [140/312]  eta: 0:02:22  lr: 0.000417  min_lr: 0.000417  loss: 3.2422 (3.1292)  weight_decay: 0.0500 (0.0500)  time: 0.6861  data: 0.1000  max mem: 41794
Epoch: [241]  [150/312]  eta: 0:02:14  lr: 0.000417  min_lr: 0.000417  loss: 3.2221 (3.1309)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.1682  max mem: 41794
Epoch: [241]  [160/312]  eta: 0:02:04  lr: 0.000416  min_lr: 0.000416  loss: 3.2706 (3.1241)  weight_decay: 0.0500 (0.0500)  time: 0.7834  data: 0.1045  max mem: 41794
Epoch: [241]  [170/312]  eta: 0:01:54  lr: 0.000416  min_lr: 0.000416  loss: 3.2368 (3.1146)  weight_decay: 0.0500 (0.0500)  time: 0.6375  data: 0.0628  max mem: 41794
Epoch: [241]  [180/312]  eta: 0:01:46  lr: 0.000415  min_lr: 0.000415  loss: 3.3108 (3.1169)  weight_decay: 0.0500 (0.0500)  time: 0.7231  data: 0.1310  max mem: 41794
Epoch: [241]  [190/312]  eta: 0:01:37  lr: 0.000415  min_lr: 0.000415  loss: 3.2682 (3.1045)  weight_decay: 0.0500 (0.0500)  time: 0.7158  data: 0.1168  max mem: 41794
Epoch: [241]  [200/312]  eta: 0:01:29  lr: 0.000415  min_lr: 0.000415  loss: 3.0601 (3.1031)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.1089  max mem: 41794
Epoch: [241]  [210/312]  eta: 0:01:21  lr: 0.000414  min_lr: 0.000414  loss: 3.2151 (3.1092)  weight_decay: 0.0500 (0.0500)  time: 0.7889  data: 0.1075  max mem: 41794
Epoch: [241]  [220/312]  eta: 0:01:12  lr: 0.000414  min_lr: 0.000414  loss: 3.3035 (3.1112)  weight_decay: 0.0500 (0.0500)  time: 0.6554  data: 0.0722  max mem: 41794
Epoch: [241]  [230/312]  eta: 0:01:04  lr: 0.000413  min_lr: 0.000413  loss: 3.3020 (3.1176)  weight_decay: 0.0500 (0.0500)  time: 0.7447  data: 0.1235  max mem: 41794
Epoch: [241]  [240/312]  eta: 0:00:56  lr: 0.000413  min_lr: 0.000413  loss: 3.2298 (3.1189)  weight_decay: 0.0500 (0.0500)  time: 0.8204  data: 0.1157  max mem: 41794
Epoch: [241]  [250/312]  eta: 0:00:48  lr: 0.000412  min_lr: 0.000412  loss: 3.0709 (3.1203)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.0804  max mem: 41794
Epoch: [241]  [260/312]  eta: 0:00:40  lr: 0.000412  min_lr: 0.000412  loss: 3.3644 (3.1259)  weight_decay: 0.0500 (0.0500)  time: 0.7235  data: 0.0942  max mem: 41794
Epoch: [241]  [270/312]  eta: 0:00:32  lr: 0.000411  min_lr: 0.000411  loss: 3.3199 (3.1257)  weight_decay: 0.0500 (0.0500)  time: 0.7371  data: 0.1044  max mem: 41794
Epoch: [241]  [280/312]  eta: 0:00:25  lr: 0.000411  min_lr: 0.000411  loss: 3.2849 (3.1258)  weight_decay: 0.0500 (0.0500)  time: 0.7561  data: 0.1281  max mem: 41794
Epoch: [241]  [290/312]  eta: 0:00:17  lr: 0.000411  min_lr: 0.000411  loss: 3.0465 (3.1192)  weight_decay: 0.0500 (0.0500)  time: 0.7772  data: 0.0662  max mem: 41794
Epoch: [241]  [300/312]  eta: 0:00:09  lr: 0.000410  min_lr: 0.000410  loss: 2.9872 (3.1177)  weight_decay: 0.0500 (0.0500)  time: 0.5696  data: 0.0002  max mem: 41794
Epoch: [241]  [310/312]  eta: 0:00:01  lr: 0.000410  min_lr: 0.000410  loss: 3.2394 (3.1197)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [241]  [311/312]  eta: 0:00:00  lr: 0.000410  min_lr: 0.000410  loss: 3.2394 (3.1192)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [241] Total time: 0:03:58 (0.7633 s / it)
Averaged stats: lr: 0.000410  min_lr: 0.000410  loss: 3.2394 (3.1022)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.8642 (0.8642)  acc1: 84.1146 (84.1146)  acc5: 96.6146 (96.6146)  time: 8.6012  data: 8.4000  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1197 (1.0442)  acc1: 78.9062 (78.2240)  acc5: 94.4010 (94.6560)  time: 1.1093  data: 0.9334  max mem: 41794
Test: Total time: 0:00:10 (1.1574 s / it)
* Acc@1 78.782 Acc@5 94.640 loss 1.045
Accuracy of the model on the 50000 test images: 78.8%
Max accuracy: 78.78%
Epoch: [242]  [  0/312]  eta: 1:21:24  lr: 0.000410  min_lr: 0.000410  loss: 2.5458 (2.5458)  weight_decay: 0.0500 (0.0500)  time: 15.6562  data: 14.4705  max mem: 41794
Epoch: [242]  [ 10/312]  eta: 0:10:55  lr: 0.000409  min_lr: 0.000409  loss: 3.3661 (3.2029)  weight_decay: 0.0500 (0.0500)  time: 2.1717  data: 1.3182  max mem: 41794
Epoch: [242]  [ 20/312]  eta: 0:07:13  lr: 0.000409  min_lr: 0.000409  loss: 3.3602 (3.1664)  weight_decay: 0.0500 (0.0500)  time: 0.7768  data: 0.0968  max mem: 41794
Epoch: [242]  [ 30/312]  eta: 0:05:31  lr: 0.000408  min_lr: 0.000408  loss: 3.2422 (3.1365)  weight_decay: 0.0500 (0.0500)  time: 0.6298  data: 0.0957  max mem: 41794
Epoch: [242]  [ 40/312]  eta: 0:04:44  lr: 0.000408  min_lr: 0.000408  loss: 3.2047 (3.1071)  weight_decay: 0.0500 (0.0500)  time: 0.5838  data: 0.0675  max mem: 41794
Epoch: [242]  [ 50/312]  eta: 0:04:21  lr: 0.000407  min_lr: 0.000407  loss: 3.2047 (3.0797)  weight_decay: 0.0500 (0.0500)  time: 0.7188  data: 0.2192  max mem: 41794
Epoch: [242]  [ 60/312]  eta: 0:03:50  lr: 0.000407  min_lr: 0.000407  loss: 3.1615 (3.0961)  weight_decay: 0.0500 (0.0500)  time: 0.6482  data: 0.1524  max mem: 41794
Epoch: [242]  [ 70/312]  eta: 0:03:39  lr: 0.000407  min_lr: 0.000407  loss: 3.1402 (3.1017)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.1796  max mem: 41794
Epoch: [242]  [ 80/312]  eta: 0:03:30  lr: 0.000406  min_lr: 0.000406  loss: 3.2007 (3.1199)  weight_decay: 0.0500 (0.0500)  time: 0.8865  data: 0.3845  max mem: 41794
Epoch: [242]  [ 90/312]  eta: 0:03:11  lr: 0.000406  min_lr: 0.000406  loss: 3.2882 (3.1435)  weight_decay: 0.0500 (0.0500)  time: 0.7042  data: 0.2057  max mem: 41794
Epoch: [242]  [100/312]  eta: 0:03:03  lr: 0.000405  min_lr: 0.000405  loss: 3.2933 (3.1411)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.1972  max mem: 41794
Epoch: [242]  [110/312]  eta: 0:02:47  lr: 0.000405  min_lr: 0.000405  loss: 3.2441 (3.1486)  weight_decay: 0.0500 (0.0500)  time: 0.6814  data: 0.1973  max mem: 41794
Epoch: [242]  [120/312]  eta: 0:02:40  lr: 0.000404  min_lr: 0.000404  loss: 3.1525 (3.1473)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1965  max mem: 41794
Epoch: [242]  [130/312]  eta: 0:02:31  lr: 0.000404  min_lr: 0.000404  loss: 3.2816 (3.1573)  weight_decay: 0.0500 (0.0500)  time: 0.8462  data: 0.3570  max mem: 41794
Epoch: [242]  [140/312]  eta: 0:02:19  lr: 0.000404  min_lr: 0.000404  loss: 3.2552 (3.1612)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1625  max mem: 41794
Epoch: [242]  [150/312]  eta: 0:02:11  lr: 0.000403  min_lr: 0.000403  loss: 3.1475 (3.1572)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.1863  max mem: 41794
Epoch: [242]  [160/312]  eta: 0:02:04  lr: 0.000403  min_lr: 0.000403  loss: 3.2391 (3.1637)  weight_decay: 0.0500 (0.0500)  time: 0.8911  data: 0.3697  max mem: 41794
Epoch: [242]  [170/312]  eta: 0:01:53  lr: 0.000402  min_lr: 0.000402  loss: 3.3160 (3.1686)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.1880  max mem: 41794
Epoch: [242]  [180/312]  eta: 0:01:46  lr: 0.000402  min_lr: 0.000402  loss: 3.3220 (3.1647)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.1341  max mem: 41794
Epoch: [242]  [190/312]  eta: 0:01:36  lr: 0.000401  min_lr: 0.000401  loss: 3.1809 (3.1561)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.1313  max mem: 41794
Epoch: [242]  [200/312]  eta: 0:01:28  lr: 0.000401  min_lr: 0.000401  loss: 3.1809 (3.1610)  weight_decay: 0.0500 (0.0500)  time: 0.6690  data: 0.1321  max mem: 41794
Epoch: [242]  [210/312]  eta: 0:01:21  lr: 0.000401  min_lr: 0.000401  loss: 3.1670 (3.1557)  weight_decay: 0.0500 (0.0500)  time: 0.8947  data: 0.2577  max mem: 41794
Epoch: [242]  [220/312]  eta: 0:01:12  lr: 0.000400  min_lr: 0.000400  loss: 3.1446 (3.1550)  weight_decay: 0.0500 (0.0500)  time: 0.7164  data: 0.1263  max mem: 41794
Epoch: [242]  [230/312]  eta: 0:01:04  lr: 0.000400  min_lr: 0.000400  loss: 3.2291 (3.1537)  weight_decay: 0.0500 (0.0500)  time: 0.7324  data: 0.1141  max mem: 41794
Epoch: [242]  [240/312]  eta: 0:00:56  lr: 0.000399  min_lr: 0.000399  loss: 3.3552 (3.1604)  weight_decay: 0.0500 (0.0500)  time: 0.8452  data: 0.1214  max mem: 41794
Epoch: [242]  [250/312]  eta: 0:00:48  lr: 0.000399  min_lr: 0.000399  loss: 3.2947 (3.1566)  weight_decay: 0.0500 (0.0500)  time: 0.6880  data: 0.0934  max mem: 41794
Epoch: [242]  [260/312]  eta: 0:00:40  lr: 0.000398  min_lr: 0.000398  loss: 3.2643 (3.1615)  weight_decay: 0.0500 (0.0500)  time: 0.7483  data: 0.1513  max mem: 41794
Epoch: [242]  [270/312]  eta: 0:00:32  lr: 0.000398  min_lr: 0.000398  loss: 3.2312 (3.1544)  weight_decay: 0.0500 (0.0500)  time: 0.6689  data: 0.0660  max mem: 41794
Epoch: [242]  [280/312]  eta: 0:00:24  lr: 0.000398  min_lr: 0.000398  loss: 3.1989 (3.1522)  weight_decay: 0.0500 (0.0500)  time: 0.6782  data: 0.0664  max mem: 41794
Epoch: [242]  [290/312]  eta: 0:00:17  lr: 0.000397  min_lr: 0.000397  loss: 3.1989 (3.1485)  weight_decay: 0.0500 (0.0500)  time: 0.8200  data: 0.1126  max mem: 41794
Epoch: [242]  [300/312]  eta: 0:00:09  lr: 0.000397  min_lr: 0.000397  loss: 3.2088 (3.1498)  weight_decay: 0.0500 (0.0500)  time: 0.6254  data: 0.0466  max mem: 41794
Epoch: [242]  [310/312]  eta: 0:00:01  lr: 0.000396  min_lr: 0.000396  loss: 3.2088 (3.1400)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [242]  [311/312]  eta: 0:00:00  lr: 0.000396  min_lr: 0.000396  loss: 3.1670 (3.1401)  weight_decay: 0.0500 (0.0500)  time: 0.4639  data: 0.0001  max mem: 41794
Epoch: [242] Total time: 0:03:57 (0.7623 s / it)
Averaged stats: lr: 0.000396  min_lr: 0.000396  loss: 3.1670 (3.0836)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 0.9181 (0.9181)  acc1: 84.3750 (84.3750)  acc5: 96.8750 (96.8750)  time: 8.0633  data: 7.8506  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1351 (1.0831)  acc1: 78.2552 (78.6080)  acc5: 94.6615 (94.6880)  time: 1.0474  data: 0.8724  max mem: 41794
Test: Total time: 0:00:09 (1.0584 s / it)
* Acc@1 78.808 Acc@5 94.732 loss 1.085
Accuracy of the model on the 50000 test images: 78.8%
Max accuracy: 78.81%
Epoch: [243]  [  0/312]  eta: 1:23:15  lr: 0.000396  min_lr: 0.000396  loss: 1.8506 (1.8506)  weight_decay: 0.0500 (0.0500)  time: 16.0114  data: 12.4662  max mem: 41794
Epoch: [243]  [ 10/312]  eta: 0:11:15  lr: 0.000396  min_lr: 0.000396  loss: 3.2936 (2.9907)  weight_decay: 0.0500 (0.0500)  time: 2.2366  data: 1.4386  max mem: 41794
Epoch: [243]  [ 20/312]  eta: 0:07:31  lr: 0.000395  min_lr: 0.000395  loss: 3.2936 (3.0722)  weight_decay: 0.0500 (0.0500)  time: 0.8240  data: 0.2676  max mem: 41794
Epoch: [243]  [ 30/312]  eta: 0:05:44  lr: 0.000395  min_lr: 0.000395  loss: 3.2100 (3.0197)  weight_decay: 0.0500 (0.0500)  time: 0.6644  data: 0.1052  max mem: 41794
Epoch: [243]  [ 40/312]  eta: 0:04:47  lr: 0.000394  min_lr: 0.000394  loss: 3.2302 (3.0674)  weight_decay: 0.0500 (0.0500)  time: 0.5426  data: 0.0191  max mem: 41794
Epoch: [243]  [ 50/312]  eta: 0:04:26  lr: 0.000394  min_lr: 0.000394  loss: 3.2302 (3.0773)  weight_decay: 0.0500 (0.0500)  time: 0.7039  data: 0.1027  max mem: 41794
Epoch: [243]  [ 60/312]  eta: 0:03:54  lr: 0.000394  min_lr: 0.000394  loss: 3.2004 (3.0747)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.0895  max mem: 41794
Epoch: [243]  [ 70/312]  eta: 0:03:41  lr: 0.000393  min_lr: 0.000393  loss: 3.0942 (3.0498)  weight_decay: 0.0500 (0.0500)  time: 0.6550  data: 0.0903  max mem: 41794
Epoch: [243]  [ 80/312]  eta: 0:03:24  lr: 0.000393  min_lr: 0.000393  loss: 3.0057 (3.0525)  weight_decay: 0.0500 (0.0500)  time: 0.7246  data: 0.1279  max mem: 41794
Epoch: [243]  [ 90/312]  eta: 0:03:11  lr: 0.000392  min_lr: 0.000392  loss: 3.1156 (3.0602)  weight_decay: 0.0500 (0.0500)  time: 0.6760  data: 0.1248  max mem: 41794
Epoch: [243]  [100/312]  eta: 0:03:04  lr: 0.000392  min_lr: 0.000392  loss: 3.1419 (3.0650)  weight_decay: 0.0500 (0.0500)  time: 0.8422  data: 0.2031  max mem: 41794
Epoch: [243]  [110/312]  eta: 0:02:49  lr: 0.000391  min_lr: 0.000391  loss: 3.0396 (3.0547)  weight_decay: 0.0500 (0.0500)  time: 0.7242  data: 0.1167  max mem: 41794
Epoch: [243]  [120/312]  eta: 0:02:40  lr: 0.000391  min_lr: 0.000391  loss: 2.9981 (3.0527)  weight_decay: 0.0500 (0.0500)  time: 0.6385  data: 0.1101  max mem: 41794
Epoch: [243]  [130/312]  eta: 0:02:32  lr: 0.000391  min_lr: 0.000391  loss: 3.1104 (3.0563)  weight_decay: 0.0500 (0.0500)  time: 0.8203  data: 0.2573  max mem: 41794
Epoch: [243]  [140/312]  eta: 0:02:20  lr: 0.000390  min_lr: 0.000390  loss: 3.2491 (3.0584)  weight_decay: 0.0500 (0.0500)  time: 0.7009  data: 0.1480  max mem: 41794
Epoch: [243]  [150/312]  eta: 0:02:12  lr: 0.000390  min_lr: 0.000390  loss: 2.9912 (3.0459)  weight_decay: 0.0500 (0.0500)  time: 0.7136  data: 0.1159  max mem: 41794
Epoch: [243]  [160/312]  eta: 0:02:02  lr: 0.000389  min_lr: 0.000389  loss: 3.1346 (3.0518)  weight_decay: 0.0500 (0.0500)  time: 0.7491  data: 0.1541  max mem: 41794
Epoch: [243]  [170/312]  eta: 0:01:54  lr: 0.000389  min_lr: 0.000389  loss: 3.1765 (3.0490)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0999  max mem: 41794
Epoch: [243]  [180/312]  eta: 0:01:46  lr: 0.000388  min_lr: 0.000388  loss: 2.9686 (3.0408)  weight_decay: 0.0500 (0.0500)  time: 0.8241  data: 0.1469  max mem: 41794
Epoch: [243]  [190/312]  eta: 0:01:36  lr: 0.000388  min_lr: 0.000388  loss: 2.8578 (3.0359)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.0899  max mem: 41794
Epoch: [243]  [200/312]  eta: 0:01:29  lr: 0.000388  min_lr: 0.000388  loss: 2.8844 (3.0293)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.0877  max mem: 41794
Epoch: [243]  [210/312]  eta: 0:01:21  lr: 0.000387  min_lr: 0.000387  loss: 3.2355 (3.0357)  weight_decay: 0.0500 (0.0500)  time: 0.8458  data: 0.1976  max mem: 41794
Epoch: [243]  [220/312]  eta: 0:01:12  lr: 0.000387  min_lr: 0.000387  loss: 3.2981 (3.0393)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.1160  max mem: 41794
Epoch: [243]  [230/312]  eta: 0:01:04  lr: 0.000386  min_lr: 0.000386  loss: 3.1520 (3.0374)  weight_decay: 0.0500 (0.0500)  time: 0.6981  data: 0.1046  max mem: 41794
Epoch: [243]  [240/312]  eta: 0:00:56  lr: 0.000386  min_lr: 0.000386  loss: 3.0810 (3.0334)  weight_decay: 0.0500 (0.0500)  time: 0.7202  data: 0.1378  max mem: 41794
Epoch: [243]  [250/312]  eta: 0:00:48  lr: 0.000385  min_lr: 0.000385  loss: 3.0285 (3.0321)  weight_decay: 0.0500 (0.0500)  time: 0.6699  data: 0.1112  max mem: 41794
Epoch: [243]  [260/312]  eta: 0:00:40  lr: 0.000385  min_lr: 0.000385  loss: 3.2958 (3.0419)  weight_decay: 0.0500 (0.0500)  time: 0.7782  data: 0.1789  max mem: 41794
Epoch: [243]  [270/312]  eta: 0:00:32  lr: 0.000385  min_lr: 0.000385  loss: 3.3248 (3.0493)  weight_decay: 0.0500 (0.0500)  time: 0.6644  data: 0.1030  max mem: 41794
Epoch: [243]  [280/312]  eta: 0:00:24  lr: 0.000384  min_lr: 0.000384  loss: 3.2444 (3.0505)  weight_decay: 0.0500 (0.0500)  time: 0.7066  data: 0.1440  max mem: 41794
Epoch: [243]  [290/312]  eta: 0:00:17  lr: 0.000384  min_lr: 0.000384  loss: 3.1954 (3.0541)  weight_decay: 0.0500 (0.0500)  time: 0.8633  data: 0.2193  max mem: 41794
Epoch: [243]  [300/312]  eta: 0:00:09  lr: 0.000383  min_lr: 0.000383  loss: 3.1966 (3.0612)  weight_decay: 0.0500 (0.0500)  time: 0.6411  data: 0.0757  max mem: 41794
Epoch: [243]  [310/312]  eta: 0:00:01  lr: 0.000383  min_lr: 0.000383  loss: 3.3473 (3.0668)  weight_decay: 0.0500 (0.0500)  time: 0.4640  data: 0.0001  max mem: 41794
Epoch: [243]  [311/312]  eta: 0:00:00  lr: 0.000383  min_lr: 0.000383  loss: 3.2954 (3.0673)  weight_decay: 0.0500 (0.0500)  time: 0.4642  data: 0.0001  max mem: 41794
Epoch: [243] Total time: 0:03:57 (0.7626 s / it)
Averaged stats: lr: 0.000383  min_lr: 0.000383  loss: 3.2954 (3.0770)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.9318 (0.9318)  acc1: 84.8958 (84.8958)  acc5: 96.4844 (96.4844)  time: 8.2823  data: 8.0739  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1800 (1.1227)  acc1: 78.5156 (78.2240)  acc5: 94.7917 (94.8320)  time: 1.0684  data: 0.8972  max mem: 41794
Test: Total time: 0:00:09 (1.0806 s / it)
* Acc@1 78.786 Acc@5 94.656 loss 1.118
Accuracy of the model on the 50000 test images: 78.8%
Max accuracy: 78.81%
Epoch: [244]  [  0/312]  eta: 1:20:48  lr: 0.000383  min_lr: 0.000383  loss: 3.7306 (3.7306)  weight_decay: 0.0500 (0.0500)  time: 15.5406  data: 15.0679  max mem: 41794
Epoch: [244]  [ 10/312]  eta: 0:11:21  lr: 0.000382  min_lr: 0.000382  loss: 3.2625 (3.2380)  weight_decay: 0.0500 (0.0500)  time: 2.2582  data: 1.3744  max mem: 41794
Epoch: [244]  [ 20/312]  eta: 0:07:22  lr: 0.000382  min_lr: 0.000382  loss: 3.1193 (3.1852)  weight_decay: 0.0500 (0.0500)  time: 0.8133  data: 0.0076  max mem: 41794
Epoch: [244]  [ 30/312]  eta: 0:05:36  lr: 0.000382  min_lr: 0.000382  loss: 3.0803 (3.0583)  weight_decay: 0.0500 (0.0500)  time: 0.6090  data: 0.0054  max mem: 41794
Epoch: [244]  [ 40/312]  eta: 0:04:44  lr: 0.000381  min_lr: 0.000381  loss: 2.9789 (3.0508)  weight_decay: 0.0500 (0.0500)  time: 0.5548  data: 0.0377  max mem: 41794
Epoch: [244]  [ 50/312]  eta: 0:04:21  lr: 0.000381  min_lr: 0.000381  loss: 3.1331 (3.0561)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1524  max mem: 41794
Epoch: [244]  [ 60/312]  eta: 0:03:49  lr: 0.000380  min_lr: 0.000380  loss: 3.1966 (3.0688)  weight_decay: 0.0500 (0.0500)  time: 0.6381  data: 0.1154  max mem: 41794
Epoch: [244]  [ 70/312]  eta: 0:03:39  lr: 0.000380  min_lr: 0.000380  loss: 3.3316 (3.0767)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.1625  max mem: 41794
Epoch: [244]  [ 80/312]  eta: 0:03:27  lr: 0.000379  min_lr: 0.000379  loss: 3.1892 (3.0414)  weight_decay: 0.0500 (0.0500)  time: 0.8319  data: 0.3079  max mem: 41794
Epoch: [244]  [ 90/312]  eta: 0:03:11  lr: 0.000379  min_lr: 0.000379  loss: 2.8815 (3.0568)  weight_decay: 0.0500 (0.0500)  time: 0.7090  data: 0.1465  max mem: 41794
Epoch: [244]  [100/312]  eta: 0:03:02  lr: 0.000379  min_lr: 0.000379  loss: 3.3784 (3.0734)  weight_decay: 0.0500 (0.0500)  time: 0.7340  data: 0.1197  max mem: 41794
Epoch: [244]  [110/312]  eta: 0:02:47  lr: 0.000378  min_lr: 0.000378  loss: 3.2309 (3.0778)  weight_decay: 0.0500 (0.0500)  time: 0.6716  data: 0.1204  max mem: 41794
Epoch: [244]  [120/312]  eta: 0:02:39  lr: 0.000378  min_lr: 0.000378  loss: 3.0143 (3.0650)  weight_decay: 0.0500 (0.0500)  time: 0.6921  data: 0.0794  max mem: 41794
Epoch: [244]  [130/312]  eta: 0:02:33  lr: 0.000377  min_lr: 0.000377  loss: 3.0368 (3.0684)  weight_decay: 0.0500 (0.0500)  time: 0.9248  data: 0.1555  max mem: 41794
Epoch: [244]  [140/312]  eta: 0:02:20  lr: 0.000377  min_lr: 0.000377  loss: 3.2053 (3.0723)  weight_decay: 0.0500 (0.0500)  time: 0.7299  data: 0.0779  max mem: 41794
Epoch: [244]  [150/312]  eta: 0:02:12  lr: 0.000377  min_lr: 0.000377  loss: 3.2053 (3.0777)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.0634  max mem: 41794
Epoch: [244]  [160/312]  eta: 0:02:03  lr: 0.000376  min_lr: 0.000376  loss: 2.9712 (3.0725)  weight_decay: 0.0500 (0.0500)  time: 0.7653  data: 0.1532  max mem: 41794
Epoch: [244]  [170/312]  eta: 0:01:53  lr: 0.000376  min_lr: 0.000376  loss: 3.0728 (3.0661)  weight_decay: 0.0500 (0.0500)  time: 0.6629  data: 0.0905  max mem: 41794
Epoch: [244]  [180/312]  eta: 0:01:47  lr: 0.000375  min_lr: 0.000375  loss: 3.1868 (3.0675)  weight_decay: 0.0500 (0.0500)  time: 0.8040  data: 0.0802  max mem: 41794
Epoch: [244]  [190/312]  eta: 0:01:36  lr: 0.000375  min_lr: 0.000375  loss: 3.1550 (3.0646)  weight_decay: 0.0500 (0.0500)  time: 0.7286  data: 0.0816  max mem: 41794
Epoch: [244]  [200/312]  eta: 0:01:29  lr: 0.000374  min_lr: 0.000374  loss: 3.1387 (3.0640)  weight_decay: 0.0500 (0.0500)  time: 0.6527  data: 0.0674  max mem: 41794
Epoch: [244]  [210/312]  eta: 0:01:21  lr: 0.000374  min_lr: 0.000374  loss: 3.0938 (3.0611)  weight_decay: 0.0500 (0.0500)  time: 0.8255  data: 0.1269  max mem: 41794
Epoch: [244]  [220/312]  eta: 0:01:12  lr: 0.000374  min_lr: 0.000374  loss: 3.1297 (3.0636)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0617  max mem: 41794
Epoch: [244]  [230/312]  eta: 0:01:04  lr: 0.000373  min_lr: 0.000373  loss: 3.1244 (3.0606)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.0835  max mem: 41794
Epoch: [244]  [240/312]  eta: 0:00:56  lr: 0.000373  min_lr: 0.000373  loss: 2.8830 (3.0539)  weight_decay: 0.0500 (0.0500)  time: 0.8098  data: 0.2070  max mem: 41794
Epoch: [244]  [250/312]  eta: 0:00:48  lr: 0.000372  min_lr: 0.000372  loss: 2.9706 (3.0559)  weight_decay: 0.0500 (0.0500)  time: 0.6624  data: 0.1242  max mem: 41794
Epoch: [244]  [260/312]  eta: 0:00:40  lr: 0.000372  min_lr: 0.000372  loss: 3.0118 (3.0499)  weight_decay: 0.0500 (0.0500)  time: 0.7155  data: 0.1210  max mem: 41794
Epoch: [244]  [270/312]  eta: 0:00:32  lr: 0.000372  min_lr: 0.000372  loss: 3.0118 (3.0557)  weight_decay: 0.0500 (0.0500)  time: 0.6729  data: 0.1218  max mem: 41794
Epoch: [244]  [280/312]  eta: 0:00:24  lr: 0.000371  min_lr: 0.000371  loss: 3.4006 (3.0655)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.1149  max mem: 41794
Epoch: [244]  [290/312]  eta: 0:00:17  lr: 0.000371  min_lr: 0.000371  loss: 3.4006 (3.0648)  weight_decay: 0.0500 (0.0500)  time: 0.8673  data: 0.2543  max mem: 41794
Epoch: [244]  [300/312]  eta: 0:00:09  lr: 0.000370  min_lr: 0.000370  loss: 3.2319 (3.0627)  weight_decay: 0.0500 (0.0500)  time: 0.6561  data: 0.1405  max mem: 41794
Epoch: [244]  [310/312]  eta: 0:00:01  lr: 0.000370  min_lr: 0.000370  loss: 3.1440 (3.0639)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [244]  [311/312]  eta: 0:00:00  lr: 0.000370  min_lr: 0.000370  loss: 3.2320 (3.0648)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [244] Total time: 0:03:57 (0.7622 s / it)
Averaged stats: lr: 0.000370  min_lr: 0.000370  loss: 3.2320 (3.0740)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:10  loss: 0.9054 (0.9054)  acc1: 83.5938 (83.5938)  acc5: 96.4844 (96.4844)  time: 7.8004  data: 7.5936  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1870 (1.0984)  acc1: 77.7344 (78.2240)  acc5: 95.1823 (94.7680)  time: 1.0281  data: 0.8569  max mem: 41794
Test: Total time: 0:00:09 (1.0384 s / it)
* Acc@1 78.622 Acc@5 94.700 loss 1.097
Accuracy of the model on the 50000 test images: 78.6%
Max accuracy: 78.81%
Epoch: [245]  [  0/312]  eta: 1:19:58  lr: 0.000370  min_lr: 0.000370  loss: 2.8019 (2.8019)  weight_decay: 0.0500 (0.0500)  time: 15.3787  data: 12.2793  max mem: 41794
Epoch: [245]  [ 10/312]  eta: 0:11:03  lr: 0.000369  min_lr: 0.000369  loss: 3.2395 (3.1744)  weight_decay: 0.0500 (0.0500)  time: 2.1979  data: 1.3248  max mem: 41794
Epoch: [245]  [ 20/312]  eta: 0:07:23  lr: 0.000369  min_lr: 0.000369  loss: 3.1188 (3.0738)  weight_decay: 0.0500 (0.0500)  time: 0.8269  data: 0.1665  max mem: 41794
Epoch: [245]  [ 30/312]  eta: 0:05:38  lr: 0.000369  min_lr: 0.000369  loss: 2.8097 (3.0133)  weight_decay: 0.0500 (0.0500)  time: 0.6519  data: 0.0521  max mem: 41794
Epoch: [245]  [ 40/312]  eta: 0:04:52  lr: 0.000368  min_lr: 0.000368  loss: 3.0504 (3.0219)  weight_decay: 0.0500 (0.0500)  time: 0.6053  data: 0.0197  max mem: 41794
Epoch: [245]  [ 50/312]  eta: 0:04:34  lr: 0.000368  min_lr: 0.000368  loss: 3.2442 (3.0327)  weight_decay: 0.0500 (0.0500)  time: 0.8073  data: 0.0983  max mem: 41794
Epoch: [245]  [ 60/312]  eta: 0:04:00  lr: 0.000367  min_lr: 0.000367  loss: 3.2442 (3.0521)  weight_decay: 0.0500 (0.0500)  time: 0.7137  data: 0.0805  max mem: 41794
Epoch: [245]  [ 70/312]  eta: 0:03:48  lr: 0.000367  min_lr: 0.000367  loss: 3.2323 (3.0629)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.0585  max mem: 41794
Epoch: [245]  [ 80/312]  eta: 0:03:33  lr: 0.000366  min_lr: 0.000366  loss: 3.0899 (3.0561)  weight_decay: 0.0500 (0.0500)  time: 0.8142  data: 0.0616  max mem: 41794
Epoch: [245]  [ 90/312]  eta: 0:03:18  lr: 0.000366  min_lr: 0.000366  loss: 3.0037 (3.0368)  weight_decay: 0.0500 (0.0500)  time: 0.7087  data: 0.0686  max mem: 41794
Epoch: [245]  [100/312]  eta: 0:03:09  lr: 0.000366  min_lr: 0.000366  loss: 3.3181 (3.0799)  weight_decay: 0.0500 (0.0500)  time: 0.7803  data: 0.1324  max mem: 41794
Epoch: [245]  [110/312]  eta: 0:02:53  lr: 0.000365  min_lr: 0.000365  loss: 3.4165 (3.0829)  weight_decay: 0.0500 (0.0500)  time: 0.6902  data: 0.0689  max mem: 41794
Epoch: [245]  [120/312]  eta: 0:02:44  lr: 0.000365  min_lr: 0.000365  loss: 3.2150 (3.0787)  weight_decay: 0.0500 (0.0500)  time: 0.6638  data: 0.0603  max mem: 41794
Epoch: [245]  [130/312]  eta: 0:02:37  lr: 0.000364  min_lr: 0.000364  loss: 3.1376 (3.0682)  weight_decay: 0.0500 (0.0500)  time: 0.8938  data: 0.1472  max mem: 41794
Epoch: [245]  [140/312]  eta: 0:02:23  lr: 0.000364  min_lr: 0.000364  loss: 3.1376 (3.0758)  weight_decay: 0.0500 (0.0500)  time: 0.7281  data: 0.0875  max mem: 41794
Epoch: [245]  [150/312]  eta: 0:02:15  lr: 0.000364  min_lr: 0.000364  loss: 3.2665 (3.0873)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.0661  max mem: 41794
Epoch: [245]  [160/312]  eta: 0:02:06  lr: 0.000363  min_lr: 0.000363  loss: 3.2608 (3.0836)  weight_decay: 0.0500 (0.0500)  time: 0.7773  data: 0.0713  max mem: 41794
Epoch: [245]  [170/312]  eta: 0:01:56  lr: 0.000363  min_lr: 0.000363  loss: 3.1017 (3.0773)  weight_decay: 0.0500 (0.0500)  time: 0.6842  data: 0.0778  max mem: 41794
Epoch: [245]  [180/312]  eta: 0:01:48  lr: 0.000362  min_lr: 0.000362  loss: 3.1987 (3.0864)  weight_decay: 0.0500 (0.0500)  time: 0.7529  data: 0.1538  max mem: 41794
Epoch: [245]  [190/312]  eta: 0:01:38  lr: 0.000362  min_lr: 0.000362  loss: 3.2362 (3.0887)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.0819  max mem: 41794
Epoch: [245]  [200/312]  eta: 0:01:30  lr: 0.000362  min_lr: 0.000362  loss: 3.2833 (3.0874)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.0733  max mem: 41794
Epoch: [245]  [210/312]  eta: 0:01:22  lr: 0.000361  min_lr: 0.000361  loss: 2.9521 (3.0729)  weight_decay: 0.0500 (0.0500)  time: 0.8820  data: 0.1517  max mem: 41794
Epoch: [245]  [220/312]  eta: 0:01:13  lr: 0.000361  min_lr: 0.000361  loss: 3.1156 (3.0751)  weight_decay: 0.0500 (0.0500)  time: 0.7116  data: 0.0804  max mem: 41794
Epoch: [245]  [230/312]  eta: 0:01:05  lr: 0.000360  min_lr: 0.000360  loss: 3.1521 (3.0731)  weight_decay: 0.0500 (0.0500)  time: 0.6540  data: 0.0503  max mem: 41794
Epoch: [245]  [240/312]  eta: 0:00:57  lr: 0.000360  min_lr: 0.000360  loss: 3.2568 (3.0807)  weight_decay: 0.0500 (0.0500)  time: 0.7733  data: 0.0552  max mem: 41794
Epoch: [245]  [250/312]  eta: 0:00:48  lr: 0.000359  min_lr: 0.000359  loss: 3.2230 (3.0864)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.0750  max mem: 41794
Epoch: [245]  [260/312]  eta: 0:00:41  lr: 0.000359  min_lr: 0.000359  loss: 3.2154 (3.0855)  weight_decay: 0.0500 (0.0500)  time: 0.7066  data: 0.1297  max mem: 41794
Epoch: [245]  [270/312]  eta: 0:00:32  lr: 0.000359  min_lr: 0.000359  loss: 3.1227 (3.0812)  weight_decay: 0.0500 (0.0500)  time: 0.6486  data: 0.0658  max mem: 41794
Epoch: [245]  [280/312]  eta: 0:00:24  lr: 0.000358  min_lr: 0.000358  loss: 3.0534 (3.0814)  weight_decay: 0.0500 (0.0500)  time: 0.6735  data: 0.0802  max mem: 41794
Epoch: [245]  [290/312]  eta: 0:00:17  lr: 0.000358  min_lr: 0.000358  loss: 3.1957 (3.0838)  weight_decay: 0.0500 (0.0500)  time: 0.8075  data: 0.1407  max mem: 41794
Epoch: [245]  [300/312]  eta: 0:00:09  lr: 0.000357  min_lr: 0.000357  loss: 3.1957 (3.0790)  weight_decay: 0.0500 (0.0500)  time: 0.6249  data: 0.0651  max mem: 41794
Epoch: [245]  [310/312]  eta: 0:00:01  lr: 0.000357  min_lr: 0.000357  loss: 3.1222 (3.0789)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [245]  [311/312]  eta: 0:00:00  lr: 0.000357  min_lr: 0.000357  loss: 3.1222 (3.0787)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [245] Total time: 0:03:59 (0.7670 s / it)
Averaged stats: lr: 0.000357  min_lr: 0.000357  loss: 3.1222 (3.0878)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:27  loss: 0.8655 (0.8655)  acc1: 84.6354 (84.6354)  acc5: 96.8750 (96.8750)  time: 9.7549  data: 9.5487  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1423 (1.0731)  acc1: 78.7760 (78.5920)  acc5: 94.6615 (94.9600)  time: 1.2338  data: 1.0611  max mem: 41794
Test: Total time: 0:00:11 (1.2694 s / it)
* Acc@1 78.962 Acc@5 94.836 loss 1.073
Accuracy of the model on the 50000 test images: 79.0%
Max accuracy: 78.96%
Epoch: [246]  [  0/312]  eta: 1:19:37  lr: 0.000357  min_lr: 0.000357  loss: 3.3374 (3.3374)  weight_decay: 0.0500 (0.0500)  time: 15.3137  data: 14.6187  max mem: 41794
Epoch: [246]  [ 10/312]  eta: 0:10:53  lr: 0.000356  min_lr: 0.000356  loss: 2.9629 (2.9901)  weight_decay: 0.0500 (0.0500)  time: 2.1655  data: 1.3404  max mem: 41794
Epoch: [246]  [ 20/312]  eta: 0:07:17  lr: 0.000356  min_lr: 0.000356  loss: 3.1087 (3.0766)  weight_decay: 0.0500 (0.0500)  time: 0.8083  data: 0.0842  max mem: 41794
Epoch: [246]  [ 30/312]  eta: 0:05:34  lr: 0.000356  min_lr: 0.000356  loss: 3.1910 (3.0536)  weight_decay: 0.0500 (0.0500)  time: 0.6480  data: 0.0805  max mem: 41794
Epoch: [246]  [ 40/312]  eta: 0:04:53  lr: 0.000355  min_lr: 0.000355  loss: 3.1593 (3.0557)  weight_decay: 0.0500 (0.0500)  time: 0.6403  data: 0.0501  max mem: 41794
Epoch: [246]  [ 50/312]  eta: 0:04:28  lr: 0.000355  min_lr: 0.000355  loss: 3.2529 (3.0812)  weight_decay: 0.0500 (0.0500)  time: 0.7748  data: 0.1015  max mem: 41794
Epoch: [246]  [ 60/312]  eta: 0:03:56  lr: 0.000354  min_lr: 0.000354  loss: 3.2880 (3.0963)  weight_decay: 0.0500 (0.0500)  time: 0.6430  data: 0.0542  max mem: 41794
Epoch: [246]  [ 70/312]  eta: 0:03:44  lr: 0.000354  min_lr: 0.000354  loss: 3.2488 (3.1122)  weight_decay: 0.0500 (0.0500)  time: 0.6868  data: 0.0689  max mem: 41794
Epoch: [246]  [ 80/312]  eta: 0:03:27  lr: 0.000354  min_lr: 0.000354  loss: 3.2488 (3.1182)  weight_decay: 0.0500 (0.0500)  time: 0.7585  data: 0.0988  max mem: 41794
Epoch: [246]  [ 90/312]  eta: 0:03:16  lr: 0.000353  min_lr: 0.000353  loss: 3.2993 (3.1274)  weight_decay: 0.0500 (0.0500)  time: 0.7277  data: 0.0891  max mem: 41794
Epoch: [246]  [100/312]  eta: 0:03:07  lr: 0.000353  min_lr: 0.000353  loss: 3.1815 (3.1092)  weight_decay: 0.0500 (0.0500)  time: 0.8623  data: 0.1220  max mem: 41794
Epoch: [246]  [110/312]  eta: 0:02:52  lr: 0.000352  min_lr: 0.000352  loss: 3.1982 (3.1262)  weight_decay: 0.0500 (0.0500)  time: 0.7011  data: 0.0665  max mem: 41794
Epoch: [246]  [120/312]  eta: 0:02:44  lr: 0.000352  min_lr: 0.000352  loss: 3.2986 (3.1279)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.0512  max mem: 41794
Epoch: [246]  [130/312]  eta: 0:02:38  lr: 0.000352  min_lr: 0.000352  loss: 3.1453 (3.1260)  weight_decay: 0.0500 (0.0500)  time: 0.9709  data: 0.1681  max mem: 41794
Epoch: [246]  [140/312]  eta: 0:02:25  lr: 0.000351  min_lr: 0.000351  loss: 3.1784 (3.1349)  weight_decay: 0.0500 (0.0500)  time: 0.7915  data: 0.1272  max mem: 41794
Epoch: [246]  [150/312]  eta: 0:02:15  lr: 0.000351  min_lr: 0.000351  loss: 3.1784 (3.1297)  weight_decay: 0.0500 (0.0500)  time: 0.6027  data: 0.0483  max mem: 41794
Epoch: [246]  [160/312]  eta: 0:02:04  lr: 0.000350  min_lr: 0.000350  loss: 3.0224 (3.1242)  weight_decay: 0.0500 (0.0500)  time: 0.6389  data: 0.0478  max mem: 41794
Epoch: [246]  [170/312]  eta: 0:01:56  lr: 0.000350  min_lr: 0.000350  loss: 3.2702 (3.1265)  weight_decay: 0.0500 (0.0500)  time: 0.6913  data: 0.0576  max mem: 41794
Epoch: [246]  [180/312]  eta: 0:01:48  lr: 0.000350  min_lr: 0.000350  loss: 3.2702 (3.1242)  weight_decay: 0.0500 (0.0500)  time: 0.8225  data: 0.1241  max mem: 41794
Epoch: [246]  [190/312]  eta: 0:01:38  lr: 0.000349  min_lr: 0.000349  loss: 3.1869 (3.1231)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.0733  max mem: 41794
Epoch: [246]  [200/312]  eta: 0:01:30  lr: 0.000349  min_lr: 0.000349  loss: 3.0271 (3.1168)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.0700  max mem: 41794
Epoch: [246]  [210/312]  eta: 0:01:22  lr: 0.000348  min_lr: 0.000348  loss: 2.9758 (3.1100)  weight_decay: 0.0500 (0.0500)  time: 0.8523  data: 0.1225  max mem: 41794
Epoch: [246]  [220/312]  eta: 0:01:13  lr: 0.000348  min_lr: 0.000348  loss: 3.0344 (3.1072)  weight_decay: 0.0500 (0.0500)  time: 0.6761  data: 0.0564  max mem: 41794
Epoch: [246]  [230/312]  eta: 0:01:05  lr: 0.000348  min_lr: 0.000348  loss: 3.1622 (3.0964)  weight_decay: 0.0500 (0.0500)  time: 0.6994  data: 0.0634  max mem: 41794
Epoch: [246]  [240/312]  eta: 0:00:56  lr: 0.000347  min_lr: 0.000347  loss: 3.0133 (3.0941)  weight_decay: 0.0500 (0.0500)  time: 0.7297  data: 0.0619  max mem: 41794
Epoch: [246]  [250/312]  eta: 0:00:48  lr: 0.000347  min_lr: 0.000347  loss: 3.0652 (3.0921)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.0482  max mem: 41794
Epoch: [246]  [260/312]  eta: 0:00:41  lr: 0.000346  min_lr: 0.000346  loss: 3.1144 (3.0913)  weight_decay: 0.0500 (0.0500)  time: 0.8222  data: 0.0985  max mem: 41794
Epoch: [246]  [270/312]  eta: 0:00:32  lr: 0.000346  min_lr: 0.000346  loss: 3.1210 (3.0914)  weight_decay: 0.0500 (0.0500)  time: 0.6702  data: 0.0527  max mem: 41794
Epoch: [246]  [280/312]  eta: 0:00:25  lr: 0.000346  min_lr: 0.000346  loss: 3.1210 (3.0940)  weight_decay: 0.0500 (0.0500)  time: 0.6557  data: 0.0754  max mem: 41794
Epoch: [246]  [290/312]  eta: 0:00:17  lr: 0.000345  min_lr: 0.000345  loss: 3.2080 (3.0917)  weight_decay: 0.0500 (0.0500)  time: 0.7744  data: 0.1424  max mem: 41794
Epoch: [246]  [300/312]  eta: 0:00:09  lr: 0.000345  min_lr: 0.000345  loss: 3.1320 (3.0941)  weight_decay: 0.0500 (0.0500)  time: 0.6096  data: 0.0674  max mem: 41794
Epoch: [246]  [310/312]  eta: 0:00:01  lr: 0.000344  min_lr: 0.000344  loss: 3.1451 (3.0917)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [246]  [311/312]  eta: 0:00:00  lr: 0.000344  min_lr: 0.000344  loss: 3.1451 (3.0912)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [246] Total time: 0:03:59 (0.7665 s / it)
Averaged stats: lr: 0.000344  min_lr: 0.000344  loss: 3.1451 (3.0862)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:09  loss: 0.9732 (0.9732)  acc1: 84.6354 (84.6354)  acc5: 97.1354 (97.1354)  time: 7.7020  data: 7.4897  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2138 (1.1416)  acc1: 78.3854 (78.4320)  acc5: 94.4010 (94.7360)  time: 1.0917  data: 0.9203  max mem: 41794
Test: Total time: 0:00:09 (1.1016 s / it)
* Acc@1 78.720 Acc@5 94.706 loss 1.141
Accuracy of the model on the 50000 test images: 78.7%
Max accuracy: 78.96%
Epoch: [247]  [  0/312]  eta: 1:19:41  lr: 0.000344  min_lr: 0.000344  loss: 3.5579 (3.5579)  weight_decay: 0.0500 (0.0500)  time: 15.3264  data: 14.1053  max mem: 41794
Epoch: [247]  [ 10/312]  eta: 0:10:19  lr: 0.000344  min_lr: 0.000344  loss: 3.4277 (3.1837)  weight_decay: 0.0500 (0.0500)  time: 2.0510  data: 1.3556  max mem: 41794
Epoch: [247]  [ 20/312]  eta: 0:07:07  lr: 0.000343  min_lr: 0.000343  loss: 3.3481 (3.2078)  weight_decay: 0.0500 (0.0500)  time: 0.7715  data: 0.1494  max mem: 41794
Epoch: [247]  [ 30/312]  eta: 0:05:25  lr: 0.000343  min_lr: 0.000343  loss: 3.2751 (3.1901)  weight_decay: 0.0500 (0.0500)  time: 0.6624  data: 0.1094  max mem: 41794
Epoch: [247]  [ 40/312]  eta: 0:05:00  lr: 0.000343  min_lr: 0.000343  loss: 3.2959 (3.1808)  weight_decay: 0.0500 (0.0500)  time: 0.7246  data: 0.0662  max mem: 41794
Epoch: [247]  [ 50/312]  eta: 0:04:38  lr: 0.000342  min_lr: 0.000342  loss: 3.2352 (3.1743)  weight_decay: 0.0500 (0.0500)  time: 0.9236  data: 0.1192  max mem: 41794
Epoch: [247]  [ 60/312]  eta: 0:04:05  lr: 0.000342  min_lr: 0.000342  loss: 3.2351 (3.1468)  weight_decay: 0.0500 (0.0500)  time: 0.7115  data: 0.0536  max mem: 41794
Epoch: [247]  [ 70/312]  eta: 0:03:51  lr: 0.000341  min_lr: 0.000341  loss: 2.9906 (3.0963)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.0437  max mem: 41794
Epoch: [247]  [ 80/312]  eta: 0:03:37  lr: 0.000341  min_lr: 0.000341  loss: 3.2329 (3.1056)  weight_decay: 0.0500 (0.0500)  time: 0.8208  data: 0.0817  max mem: 41794
Epoch: [247]  [ 90/312]  eta: 0:03:18  lr: 0.000341  min_lr: 0.000341  loss: 3.2790 (3.1167)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.0387  max mem: 41794
Epoch: [247]  [100/312]  eta: 0:03:10  lr: 0.000340  min_lr: 0.000340  loss: 3.3059 (3.1300)  weight_decay: 0.0500 (0.0500)  time: 0.7422  data: 0.0554  max mem: 41794
Epoch: [247]  [110/312]  eta: 0:02:54  lr: 0.000340  min_lr: 0.000340  loss: 3.3334 (3.1427)  weight_decay: 0.0500 (0.0500)  time: 0.7404  data: 0.0632  max mem: 41794
Epoch: [247]  [120/312]  eta: 0:02:44  lr: 0.000339  min_lr: 0.000339  loss: 3.3093 (3.1489)  weight_decay: 0.0500 (0.0500)  time: 0.6591  data: 0.0462  max mem: 41794
Epoch: [247]  [130/312]  eta: 0:02:37  lr: 0.000339  min_lr: 0.000339  loss: 3.2963 (3.1512)  weight_decay: 0.0500 (0.0500)  time: 0.8682  data: 0.1083  max mem: 41794
Epoch: [247]  [140/312]  eta: 0:02:24  lr: 0.000339  min_lr: 0.000339  loss: 3.3032 (3.1541)  weight_decay: 0.0500 (0.0500)  time: 0.7345  data: 0.0706  max mem: 41794
Epoch: [247]  [150/312]  eta: 0:02:16  lr: 0.000338  min_lr: 0.000338  loss: 3.3363 (3.1646)  weight_decay: 0.0500 (0.0500)  time: 0.6823  data: 0.0618  max mem: 41794
Epoch: [247]  [160/312]  eta: 0:02:06  lr: 0.000338  min_lr: 0.000338  loss: 3.2705 (3.1490)  weight_decay: 0.0500 (0.0500)  time: 0.7828  data: 0.0830  max mem: 41794
Epoch: [247]  [170/312]  eta: 0:01:56  lr: 0.000337  min_lr: 0.000337  loss: 3.2899 (3.1588)  weight_decay: 0.0500 (0.0500)  time: 0.6516  data: 0.0634  max mem: 41794
Epoch: [247]  [180/312]  eta: 0:01:48  lr: 0.000337  min_lr: 0.000337  loss: 3.3378 (3.1591)  weight_decay: 0.0500 (0.0500)  time: 0.7142  data: 0.1337  max mem: 41794
Epoch: [247]  [190/312]  eta: 0:01:38  lr: 0.000337  min_lr: 0.000337  loss: 3.2891 (3.1532)  weight_decay: 0.0500 (0.0500)  time: 0.6738  data: 0.0924  max mem: 41794
Epoch: [247]  [200/312]  eta: 0:01:30  lr: 0.000336  min_lr: 0.000336  loss: 3.2891 (3.1509)  weight_decay: 0.0500 (0.0500)  time: 0.6980  data: 0.0987  max mem: 41794
Epoch: [247]  [210/312]  eta: 0:01:22  lr: 0.000336  min_lr: 0.000336  loss: 3.0651 (3.1416)  weight_decay: 0.0500 (0.0500)  time: 0.8743  data: 0.1687  max mem: 41794
Epoch: [247]  [220/312]  eta: 0:01:13  lr: 0.000335  min_lr: 0.000335  loss: 3.0651 (3.1380)  weight_decay: 0.0500 (0.0500)  time: 0.6721  data: 0.0709  max mem: 41794
Epoch: [247]  [230/312]  eta: 0:01:05  lr: 0.000335  min_lr: 0.000335  loss: 3.2303 (3.1287)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.0508  max mem: 41794
Epoch: [247]  [240/312]  eta: 0:00:57  lr: 0.000335  min_lr: 0.000335  loss: 3.1618 (3.1288)  weight_decay: 0.0500 (0.0500)  time: 0.8239  data: 0.0718  max mem: 41794
Epoch: [247]  [250/312]  eta: 0:00:48  lr: 0.000334  min_lr: 0.000334  loss: 3.1618 (3.1220)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0633  max mem: 41794
Epoch: [247]  [260/312]  eta: 0:00:41  lr: 0.000334  min_lr: 0.000334  loss: 3.2731 (3.1277)  weight_decay: 0.0500 (0.0500)  time: 0.7180  data: 0.0988  max mem: 41794
Epoch: [247]  [270/312]  eta: 0:00:32  lr: 0.000333  min_lr: 0.000333  loss: 3.3109 (3.1259)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.0571  max mem: 41794
Epoch: [247]  [280/312]  eta: 0:00:25  lr: 0.000333  min_lr: 0.000333  loss: 3.2424 (3.1309)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.0846  max mem: 41794
Epoch: [247]  [290/312]  eta: 0:00:17  lr: 0.000333  min_lr: 0.000333  loss: 3.2896 (3.1357)  weight_decay: 0.0500 (0.0500)  time: 0.8025  data: 0.1322  max mem: 41794
Epoch: [247]  [300/312]  eta: 0:00:09  lr: 0.000332  min_lr: 0.000332  loss: 3.2559 (3.1309)  weight_decay: 0.0500 (0.0500)  time: 0.6116  data: 0.0480  max mem: 41794
Epoch: [247]  [310/312]  eta: 0:00:01  lr: 0.000332  min_lr: 0.000332  loss: 3.1750 (3.1290)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [247]  [311/312]  eta: 0:00:00  lr: 0.000332  min_lr: 0.000332  loss: 3.1628 (3.1263)  weight_decay: 0.0500 (0.0500)  time: 0.4647  data: 0.0001  max mem: 41794
Epoch: [247] Total time: 0:03:59 (0.7680 s / it)
Averaged stats: lr: 0.000332  min_lr: 0.000332  loss: 3.1628 (3.0632)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.8683 (0.8683)  acc1: 85.2865 (85.2865)  acc5: 96.2240 (96.2240)  time: 8.5260  data: 8.3239  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1168 (1.0529)  acc1: 77.7344 (78.5440)  acc5: 94.5312 (94.7840)  time: 1.1003  data: 0.9250  max mem: 41794
Test: Total time: 0:00:10 (1.1450 s / it)
* Acc@1 78.850 Acc@5 94.710 loss 1.050
Accuracy of the model on the 50000 test images: 78.9%
Max accuracy: 78.96%
Epoch: [248]  [  0/312]  eta: 1:25:40  lr: 0.000332  min_lr: 0.000332  loss: 2.8120 (2.8120)  weight_decay: 0.0500 (0.0500)  time: 16.4745  data: 14.3345  max mem: 41794
Epoch: [248]  [ 10/312]  eta: 0:11:22  lr: 0.000331  min_lr: 0.000331  loss: 3.1294 (3.1954)  weight_decay: 0.0500 (0.0500)  time: 2.2603  data: 1.3951  max mem: 41794
Epoch: [248]  [ 20/312]  eta: 0:07:37  lr: 0.000331  min_lr: 0.000331  loss: 3.1428 (3.1889)  weight_decay: 0.0500 (0.0500)  time: 0.8205  data: 0.1267  max mem: 41794
Epoch: [248]  [ 30/312]  eta: 0:05:47  lr: 0.000331  min_lr: 0.000331  loss: 3.2598 (3.1892)  weight_decay: 0.0500 (0.0500)  time: 0.6685  data: 0.0853  max mem: 41794
Epoch: [248]  [ 40/312]  eta: 0:04:51  lr: 0.000330  min_lr: 0.000330  loss: 3.2925 (3.1958)  weight_decay: 0.0500 (0.0500)  time: 0.5550  data: 0.0152  max mem: 41794
Epoch: [248]  [ 50/312]  eta: 0:04:28  lr: 0.000330  min_lr: 0.000330  loss: 3.3383 (3.2046)  weight_decay: 0.0500 (0.0500)  time: 0.7036  data: 0.1223  max mem: 41794
Epoch: [248]  [ 60/312]  eta: 0:03:56  lr: 0.000329  min_lr: 0.000329  loss: 3.0447 (3.1337)  weight_decay: 0.0500 (0.0500)  time: 0.6598  data: 0.1165  max mem: 41794
Epoch: [248]  [ 70/312]  eta: 0:03:44  lr: 0.000329  min_lr: 0.000329  loss: 2.6842 (3.0934)  weight_decay: 0.0500 (0.0500)  time: 0.6787  data: 0.1506  max mem: 41794
Epoch: [248]  [ 80/312]  eta: 0:03:32  lr: 0.000329  min_lr: 0.000329  loss: 3.0638 (3.0890)  weight_decay: 0.0500 (0.0500)  time: 0.8583  data: 0.2961  max mem: 41794
Epoch: [248]  [ 90/312]  eta: 0:03:14  lr: 0.000328  min_lr: 0.000328  loss: 3.0810 (3.0736)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.1761  max mem: 41794
Epoch: [248]  [100/312]  eta: 0:03:05  lr: 0.000328  min_lr: 0.000328  loss: 3.0810 (3.0912)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.1778  max mem: 41794
Epoch: [248]  [110/312]  eta: 0:02:49  lr: 0.000327  min_lr: 0.000327  loss: 3.1028 (3.0873)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1479  max mem: 41794
Epoch: [248]  [120/312]  eta: 0:02:41  lr: 0.000327  min_lr: 0.000327  loss: 2.9954 (3.0727)  weight_decay: 0.0500 (0.0500)  time: 0.6936  data: 0.1683  max mem: 41794
Epoch: [248]  [130/312]  eta: 0:02:33  lr: 0.000327  min_lr: 0.000327  loss: 3.3127 (3.0745)  weight_decay: 0.0500 (0.0500)  time: 0.8649  data: 0.3076  max mem: 41794
Epoch: [248]  [140/312]  eta: 0:02:20  lr: 0.000326  min_lr: 0.000326  loss: 3.2252 (3.0698)  weight_decay: 0.0500 (0.0500)  time: 0.6624  data: 0.1400  max mem: 41794
Epoch: [248]  [150/312]  eta: 0:02:12  lr: 0.000326  min_lr: 0.000326  loss: 3.0311 (3.0667)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1502  max mem: 41794
Epoch: [248]  [160/312]  eta: 0:02:04  lr: 0.000325  min_lr: 0.000325  loss: 2.7342 (3.0423)  weight_decay: 0.0500 (0.0500)  time: 0.8288  data: 0.2575  max mem: 41794
Epoch: [248]  [170/312]  eta: 0:01:54  lr: 0.000325  min_lr: 0.000325  loss: 2.7748 (3.0486)  weight_decay: 0.0500 (0.0500)  time: 0.6857  data: 0.1333  max mem: 41794
Epoch: [248]  [180/312]  eta: 0:01:46  lr: 0.000325  min_lr: 0.000325  loss: 3.1511 (3.0484)  weight_decay: 0.0500 (0.0500)  time: 0.7273  data: 0.1675  max mem: 41794
Epoch: [248]  [190/312]  eta: 0:01:36  lr: 0.000324  min_lr: 0.000324  loss: 3.1210 (3.0497)  weight_decay: 0.0500 (0.0500)  time: 0.6777  data: 0.1422  max mem: 41794
Epoch: [248]  [200/312]  eta: 0:01:29  lr: 0.000324  min_lr: 0.000324  loss: 3.1210 (3.0419)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.1554  max mem: 41794
Epoch: [248]  [210/312]  eta: 0:01:22  lr: 0.000323  min_lr: 0.000323  loss: 3.3290 (3.0477)  weight_decay: 0.0500 (0.0500)  time: 0.9244  data: 0.3102  max mem: 41794
Epoch: [248]  [220/312]  eta: 0:01:12  lr: 0.000323  min_lr: 0.000323  loss: 3.3045 (3.0467)  weight_decay: 0.0500 (0.0500)  time: 0.7168  data: 0.1554  max mem: 41794
Epoch: [248]  [230/312]  eta: 0:01:05  lr: 0.000323  min_lr: 0.000323  loss: 3.2728 (3.0491)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.1490  max mem: 41794
Epoch: [248]  [240/312]  eta: 0:00:56  lr: 0.000322  min_lr: 0.000322  loss: 3.2986 (3.0543)  weight_decay: 0.0500 (0.0500)  time: 0.7845  data: 0.1900  max mem: 41794
Epoch: [248]  [250/312]  eta: 0:00:48  lr: 0.000322  min_lr: 0.000322  loss: 3.0239 (3.0482)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.1258  max mem: 41794
Epoch: [248]  [260/312]  eta: 0:00:41  lr: 0.000322  min_lr: 0.000322  loss: 3.0344 (3.0471)  weight_decay: 0.0500 (0.0500)  time: 0.7920  data: 0.2359  max mem: 41794
Epoch: [248]  [270/312]  eta: 0:00:32  lr: 0.000321  min_lr: 0.000321  loss: 3.2276 (3.0431)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1517  max mem: 41794
Epoch: [248]  [280/312]  eta: 0:00:25  lr: 0.000321  min_lr: 0.000321  loss: 3.1499 (3.0379)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.1154  max mem: 41794
Epoch: [248]  [290/312]  eta: 0:00:17  lr: 0.000320  min_lr: 0.000320  loss: 3.1462 (3.0423)  weight_decay: 0.0500 (0.0500)  time: 0.8111  data: 0.1983  max mem: 41794
Epoch: [248]  [300/312]  eta: 0:00:09  lr: 0.000320  min_lr: 0.000320  loss: 3.1462 (3.0441)  weight_decay: 0.0500 (0.0500)  time: 0.6146  data: 0.0832  max mem: 41794
Epoch: [248]  [310/312]  eta: 0:00:01  lr: 0.000320  min_lr: 0.000320  loss: 2.9537 (3.0405)  weight_decay: 0.0500 (0.0500)  time: 0.4674  data: 0.0001  max mem: 41794
Epoch: [248]  [311/312]  eta: 0:00:00  lr: 0.000320  min_lr: 0.000320  loss: 3.0157 (3.0408)  weight_decay: 0.0500 (0.0500)  time: 0.4663  data: 0.0001  max mem: 41794
Epoch: [248] Total time: 0:03:59 (0.7664 s / it)
Averaged stats: lr: 0.000320  min_lr: 0.000320  loss: 3.0157 (3.0687)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:09  loss: 0.8873 (0.8873)  acc1: 84.8958 (84.8958)  acc5: 96.3542 (96.3542)  time: 7.7085  data: 7.4965  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1103 (1.0620)  acc1: 77.6042 (78.3040)  acc5: 95.1823 (94.7360)  time: 1.0435  data: 0.8708  max mem: 41794
Test: Total time: 0:00:09 (1.0546 s / it)
* Acc@1 79.204 Acc@5 94.832 loss 1.056
Accuracy of the model on the 50000 test images: 79.2%
Max accuracy: 79.20%
Epoch: [249]  [  0/312]  eta: 1:19:29  lr: 0.000320  min_lr: 0.000320  loss: 2.0343 (2.0343)  weight_decay: 0.0500 (0.0500)  time: 15.2874  data: 12.4422  max mem: 41794
Epoch: [249]  [ 10/312]  eta: 0:10:47  lr: 0.000319  min_lr: 0.000319  loss: 2.6030 (2.6605)  weight_decay: 0.0500 (0.0500)  time: 2.1449  data: 1.3971  max mem: 41794
Epoch: [249]  [ 20/312]  eta: 0:07:38  lr: 0.000319  min_lr: 0.000319  loss: 2.7824 (2.8108)  weight_decay: 0.0500 (0.0500)  time: 0.8843  data: 0.2767  max mem: 41794
Epoch: [249]  [ 30/312]  eta: 0:05:48  lr: 0.000318  min_lr: 0.000318  loss: 2.9525 (2.8463)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.1307  max mem: 41794
Epoch: [249]  [ 40/312]  eta: 0:04:57  lr: 0.000318  min_lr: 0.000318  loss: 3.0388 (2.8622)  weight_decay: 0.0500 (0.0500)  time: 0.5950  data: 0.0803  max mem: 41794
Epoch: [249]  [ 50/312]  eta: 0:04:32  lr: 0.000318  min_lr: 0.000318  loss: 3.0388 (2.8684)  weight_decay: 0.0500 (0.0500)  time: 0.7380  data: 0.2004  max mem: 41794
Epoch: [249]  [ 60/312]  eta: 0:03:59  lr: 0.000317  min_lr: 0.000317  loss: 2.7648 (2.8922)  weight_decay: 0.0500 (0.0500)  time: 0.6532  data: 0.1209  max mem: 41794
Epoch: [249]  [ 70/312]  eta: 0:03:46  lr: 0.000317  min_lr: 0.000317  loss: 3.1663 (2.9213)  weight_decay: 0.0500 (0.0500)  time: 0.6772  data: 0.1151  max mem: 41794
Epoch: [249]  [ 80/312]  eta: 0:03:28  lr: 0.000316  min_lr: 0.000316  loss: 3.3004 (2.9613)  weight_decay: 0.0500 (0.0500)  time: 0.7358  data: 0.1412  max mem: 41794
Epoch: [249]  [ 90/312]  eta: 0:03:14  lr: 0.000316  min_lr: 0.000316  loss: 3.3013 (2.9931)  weight_decay: 0.0500 (0.0500)  time: 0.6518  data: 0.1256  max mem: 41794
Epoch: [249]  [100/312]  eta: 0:03:06  lr: 0.000316  min_lr: 0.000316  loss: 3.2051 (2.9744)  weight_decay: 0.0500 (0.0500)  time: 0.8188  data: 0.2443  max mem: 41794
Epoch: [249]  [110/312]  eta: 0:02:51  lr: 0.000315  min_lr: 0.000315  loss: 3.1647 (3.0026)  weight_decay: 0.0500 (0.0500)  time: 0.7208  data: 0.1455  max mem: 41794
Epoch: [249]  [120/312]  eta: 0:02:43  lr: 0.000315  min_lr: 0.000315  loss: 3.2342 (3.0084)  weight_decay: 0.0500 (0.0500)  time: 0.7102  data: 0.1369  max mem: 41794
Epoch: [249]  [130/312]  eta: 0:02:34  lr: 0.000314  min_lr: 0.000314  loss: 2.9727 (3.0105)  weight_decay: 0.0500 (0.0500)  time: 0.8748  data: 0.2475  max mem: 41794
Epoch: [249]  [140/312]  eta: 0:02:22  lr: 0.000314  min_lr: 0.000314  loss: 2.9257 (3.0027)  weight_decay: 0.0500 (0.0500)  time: 0.6593  data: 0.1113  max mem: 41794
Epoch: [249]  [150/312]  eta: 0:02:14  lr: 0.000314  min_lr: 0.000314  loss: 2.9373 (3.0049)  weight_decay: 0.0500 (0.0500)  time: 0.6888  data: 0.1164  max mem: 41794
Epoch: [249]  [160/312]  eta: 0:02:05  lr: 0.000313  min_lr: 0.000313  loss: 2.9092 (3.0006)  weight_decay: 0.0500 (0.0500)  time: 0.8415  data: 0.2618  max mem: 41794
Epoch: [249]  [170/312]  eta: 0:01:55  lr: 0.000313  min_lr: 0.000313  loss: 3.1760 (3.0120)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.1460  max mem: 41794
Epoch: [249]  [180/312]  eta: 0:01:47  lr: 0.000313  min_lr: 0.000313  loss: 3.2954 (3.0256)  weight_decay: 0.0500 (0.0500)  time: 0.7132  data: 0.0999  max mem: 41794
Epoch: [249]  [190/312]  eta: 0:01:37  lr: 0.000312  min_lr: 0.000312  loss: 3.1875 (3.0238)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.0999  max mem: 41794
Epoch: [249]  [200/312]  eta: 0:01:29  lr: 0.000312  min_lr: 0.000312  loss: 3.0524 (3.0219)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1137  max mem: 41794
Epoch: [249]  [210/312]  eta: 0:01:21  lr: 0.000311  min_lr: 0.000311  loss: 2.8843 (3.0153)  weight_decay: 0.0500 (0.0500)  time: 0.8467  data: 0.2436  max mem: 41794
Epoch: [249]  [220/312]  eta: 0:01:12  lr: 0.000311  min_lr: 0.000311  loss: 3.0937 (3.0158)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1412  max mem: 41794
Epoch: [249]  [230/312]  eta: 0:01:05  lr: 0.000311  min_lr: 0.000311  loss: 3.3418 (3.0287)  weight_decay: 0.0500 (0.0500)  time: 0.7149  data: 0.1443  max mem: 41794
Epoch: [249]  [240/312]  eta: 0:00:57  lr: 0.000310  min_lr: 0.000310  loss: 3.3818 (3.0407)  weight_decay: 0.0500 (0.0500)  time: 0.8314  data: 0.2145  max mem: 41794
Epoch: [249]  [250/312]  eta: 0:00:48  lr: 0.000310  min_lr: 0.000310  loss: 3.2981 (3.0383)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.1093  max mem: 41794
Epoch: [249]  [260/312]  eta: 0:00:41  lr: 0.000309  min_lr: 0.000309  loss: 3.2242 (3.0416)  weight_decay: 0.0500 (0.0500)  time: 0.7313  data: 0.1145  max mem: 41794
Epoch: [249]  [270/312]  eta: 0:00:32  lr: 0.000309  min_lr: 0.000309  loss: 3.2242 (3.0415)  weight_decay: 0.0500 (0.0500)  time: 0.6610  data: 0.0867  max mem: 41794
Epoch: [249]  [280/312]  eta: 0:00:24  lr: 0.000309  min_lr: 0.000309  loss: 3.1172 (3.0372)  weight_decay: 0.0500 (0.0500)  time: 0.6754  data: 0.0976  max mem: 41794
Epoch: [249]  [290/312]  eta: 0:00:17  lr: 0.000308  min_lr: 0.000308  loss: 3.2229 (3.0403)  weight_decay: 0.0500 (0.0500)  time: 0.8465  data: 0.1745  max mem: 41794
Epoch: [249]  [300/312]  eta: 0:00:09  lr: 0.000308  min_lr: 0.000308  loss: 3.1881 (3.0403)  weight_decay: 0.0500 (0.0500)  time: 0.6590  data: 0.0795  max mem: 41794
Epoch: [249]  [310/312]  eta: 0:00:01  lr: 0.000308  min_lr: 0.000308  loss: 3.1795 (3.0365)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0024  max mem: 41794
Epoch: [249]  [311/312]  eta: 0:00:00  lr: 0.000308  min_lr: 0.000308  loss: 2.9480 (3.0353)  weight_decay: 0.0500 (0.0500)  time: 0.4631  data: 0.0001  max mem: 41794
Epoch: [249] Total time: 0:03:59 (0.7680 s / it)
Averaged stats: lr: 0.000308  min_lr: 0.000308  loss: 2.9480 (3.0544)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:08  loss: 0.8448 (0.8448)  acc1: 85.4167 (85.4167)  acc5: 96.4844 (96.4844)  time: 7.6134  data: 7.4001  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1145 (1.0502)  acc1: 77.8646 (78.7200)  acc5: 95.0521 (94.8640)  time: 1.0853  data: 0.9135  max mem: 41794
Test: Total time: 0:00:09 (1.0953 s / it)
* Acc@1 79.080 Acc@5 94.770 loss 1.047
Accuracy of the model on the 50000 test images: 79.1%
Max accuracy: 79.20%
Epoch: [250]  [  0/312]  eta: 1:23:48  lr: 0.000307  min_lr: 0.000307  loss: 2.6029 (2.6029)  weight_decay: 0.0500 (0.0500)  time: 16.1175  data: 11.9108  max mem: 41794
Epoch: [250]  [ 10/312]  eta: 0:11:13  lr: 0.000307  min_lr: 0.000307  loss: 3.0264 (2.9548)  weight_decay: 0.0500 (0.0500)  time: 2.2297  data: 1.3013  max mem: 41794
Epoch: [250]  [ 20/312]  eta: 0:07:39  lr: 0.000307  min_lr: 0.000307  loss: 3.0320 (2.9537)  weight_decay: 0.0500 (0.0500)  time: 0.8474  data: 0.1680  max mem: 41794
Epoch: [250]  [ 30/312]  eta: 0:05:49  lr: 0.000306  min_lr: 0.000306  loss: 3.1869 (2.9541)  weight_decay: 0.0500 (0.0500)  time: 0.6924  data: 0.0585  max mem: 41794
Epoch: [250]  [ 40/312]  eta: 0:04:51  lr: 0.000306  min_lr: 0.000306  loss: 3.2852 (3.0257)  weight_decay: 0.0500 (0.0500)  time: 0.5420  data: 0.0129  max mem: 41794
Epoch: [250]  [ 50/312]  eta: 0:04:25  lr: 0.000306  min_lr: 0.000306  loss: 3.2852 (3.0173)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.0138  max mem: 41794
Epoch: [250]  [ 60/312]  eta: 0:03:56  lr: 0.000305  min_lr: 0.000305  loss: 3.2842 (3.0616)  weight_decay: 0.0500 (0.0500)  time: 0.6617  data: 0.0389  max mem: 41794
Epoch: [250]  [ 70/312]  eta: 0:03:46  lr: 0.000305  min_lr: 0.000305  loss: 3.1829 (3.0458)  weight_decay: 0.0500 (0.0500)  time: 0.7465  data: 0.1425  max mem: 41794
Epoch: [250]  [ 80/312]  eta: 0:03:34  lr: 0.000304  min_lr: 0.000304  loss: 3.1336 (3.0265)  weight_decay: 0.0500 (0.0500)  time: 0.8869  data: 0.1420  max mem: 41794
Epoch: [250]  [ 90/312]  eta: 0:03:16  lr: 0.000304  min_lr: 0.000304  loss: 3.1548 (3.0312)  weight_decay: 0.0500 (0.0500)  time: 0.7079  data: 0.0767  max mem: 41794
Epoch: [250]  [100/312]  eta: 0:03:09  lr: 0.000304  min_lr: 0.000304  loss: 3.2381 (3.0548)  weight_decay: 0.0500 (0.0500)  time: 0.7588  data: 0.1082  max mem: 41794
Epoch: [250]  [110/312]  eta: 0:02:53  lr: 0.000303  min_lr: 0.000303  loss: 3.2196 (3.0579)  weight_decay: 0.0500 (0.0500)  time: 0.7215  data: 0.0644  max mem: 41794
Epoch: [250]  [120/312]  eta: 0:02:44  lr: 0.000303  min_lr: 0.000303  loss: 3.1307 (3.0470)  weight_decay: 0.0500 (0.0500)  time: 0.6745  data: 0.0537  max mem: 41794
Epoch: [250]  [130/312]  eta: 0:02:34  lr: 0.000303  min_lr: 0.000303  loss: 3.2503 (3.0578)  weight_decay: 0.0500 (0.0500)  time: 0.7975  data: 0.1206  max mem: 41794
Epoch: [250]  [140/312]  eta: 0:02:21  lr: 0.000302  min_lr: 0.000302  loss: 3.3135 (3.0658)  weight_decay: 0.0500 (0.0500)  time: 0.6237  data: 0.0733  max mem: 41794
Epoch: [250]  [150/312]  eta: 0:02:13  lr: 0.000302  min_lr: 0.000302  loss: 3.0959 (3.0606)  weight_decay: 0.0500 (0.0500)  time: 0.6648  data: 0.1136  max mem: 41794
Epoch: [250]  [160/312]  eta: 0:02:04  lr: 0.000301  min_lr: 0.000301  loss: 3.2149 (3.0759)  weight_decay: 0.0500 (0.0500)  time: 0.8085  data: 0.2285  max mem: 41794
Epoch: [250]  [170/312]  eta: 0:01:53  lr: 0.000301  min_lr: 0.000301  loss: 3.2507 (3.0826)  weight_decay: 0.0500 (0.0500)  time: 0.6317  data: 0.1155  max mem: 41794
Epoch: [250]  [180/312]  eta: 0:01:46  lr: 0.000301  min_lr: 0.000301  loss: 3.1980 (3.0877)  weight_decay: 0.0500 (0.0500)  time: 0.7091  data: 0.1434  max mem: 41794
Epoch: [250]  [190/312]  eta: 0:01:36  lr: 0.000300  min_lr: 0.000300  loss: 3.2374 (3.0883)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1447  max mem: 41794
Epoch: [250]  [200/312]  eta: 0:01:28  lr: 0.000300  min_lr: 0.000300  loss: 3.2797 (3.0835)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.1079  max mem: 41794
Epoch: [250]  [210/312]  eta: 0:01:21  lr: 0.000299  min_lr: 0.000299  loss: 2.8230 (3.0707)  weight_decay: 0.0500 (0.0500)  time: 0.8490  data: 0.2009  max mem: 41794
Epoch: [250]  [220/312]  eta: 0:01:12  lr: 0.000299  min_lr: 0.000299  loss: 3.1443 (3.0789)  weight_decay: 0.0500 (0.0500)  time: 0.6772  data: 0.0949  max mem: 41794
Epoch: [250]  [230/312]  eta: 0:01:04  lr: 0.000299  min_lr: 0.000299  loss: 3.2624 (3.0836)  weight_decay: 0.0500 (0.0500)  time: 0.7028  data: 0.1012  max mem: 41794
Epoch: [250]  [240/312]  eta: 0:00:56  lr: 0.000298  min_lr: 0.000298  loss: 3.1604 (3.0805)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.1952  max mem: 41794
Epoch: [250]  [250/312]  eta: 0:00:48  lr: 0.000298  min_lr: 0.000298  loss: 2.9247 (3.0649)  weight_decay: 0.0500 (0.0500)  time: 0.6384  data: 0.0946  max mem: 41794
Epoch: [250]  [260/312]  eta: 0:00:40  lr: 0.000298  min_lr: 0.000298  loss: 2.9750 (3.0655)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.1433  max mem: 41794
Epoch: [250]  [270/312]  eta: 0:00:32  lr: 0.000297  min_lr: 0.000297  loss: 3.2926 (3.0640)  weight_decay: 0.0500 (0.0500)  time: 0.6770  data: 0.1447  max mem: 41794
Epoch: [250]  [280/312]  eta: 0:00:24  lr: 0.000297  min_lr: 0.000297  loss: 2.9399 (3.0558)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.1643  max mem: 41794
Epoch: [250]  [290/312]  eta: 0:00:17  lr: 0.000296  min_lr: 0.000296  loss: 2.9347 (3.0572)  weight_decay: 0.0500 (0.0500)  time: 0.8472  data: 0.2997  max mem: 41794
Epoch: [250]  [300/312]  eta: 0:00:09  lr: 0.000296  min_lr: 0.000296  loss: 3.2130 (3.0648)  weight_decay: 0.0500 (0.0500)  time: 0.6653  data: 0.1373  max mem: 41794
Epoch: [250]  [310/312]  eta: 0:00:01  lr: 0.000296  min_lr: 0.000296  loss: 3.2130 (3.0617)  weight_decay: 0.0500 (0.0500)  time: 0.4710  data: 0.0001  max mem: 41794
Epoch: [250]  [311/312]  eta: 0:00:00  lr: 0.000296  min_lr: 0.000296  loss: 3.2799 (3.0624)  weight_decay: 0.0500 (0.0500)  time: 0.4694  data: 0.0001  max mem: 41794
Epoch: [250] Total time: 0:03:57 (0.7616 s / it)
Averaged stats: lr: 0.000296  min_lr: 0.000296  loss: 3.2799 (3.0525)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 0.9414 (0.9414)  acc1: 85.1562 (85.1562)  acc5: 96.6146 (96.6146)  time: 8.9734  data: 8.7600  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1484 (1.0835)  acc1: 79.4271 (78.8640)  acc5: 94.4010 (94.8640)  time: 1.1450  data: 0.9734  max mem: 41794
Test: Total time: 0:00:10 (1.1861 s / it)
* Acc@1 79.112 Acc@5 94.850 loss 1.082
Accuracy of the model on the 50000 test images: 79.1%
Max accuracy: 79.20%
Epoch: [251]  [  0/312]  eta: 1:25:10  lr: 0.000296  min_lr: 0.000296  loss: 3.6578 (3.6578)  weight_decay: 0.0500 (0.0500)  time: 16.3804  data: 14.5172  max mem: 41794
Epoch: [251]  [ 10/312]  eta: 0:11:06  lr: 0.000295  min_lr: 0.000295  loss: 3.3527 (3.2390)  weight_decay: 0.0500 (0.0500)  time: 2.2069  data: 1.3611  max mem: 41794
Epoch: [251]  [ 20/312]  eta: 0:07:36  lr: 0.000295  min_lr: 0.000295  loss: 3.1870 (3.1696)  weight_decay: 0.0500 (0.0500)  time: 0.8223  data: 0.0499  max mem: 41794
Epoch: [251]  [ 30/312]  eta: 0:05:46  lr: 0.000295  min_lr: 0.000295  loss: 3.1328 (3.0827)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.0275  max mem: 41794
Epoch: [251]  [ 40/312]  eta: 0:04:50  lr: 0.000294  min_lr: 0.000294  loss: 3.0183 (3.0306)  weight_decay: 0.0500 (0.0500)  time: 0.5482  data: 0.0019  max mem: 41794
Epoch: [251]  [ 50/312]  eta: 0:04:17  lr: 0.000294  min_lr: 0.000294  loss: 2.9021 (3.0308)  weight_decay: 0.0500 (0.0500)  time: 0.6037  data: 0.0439  max mem: 41794
Epoch: [251]  [ 60/312]  eta: 0:03:53  lr: 0.000293  min_lr: 0.000293  loss: 3.2621 (3.0631)  weight_decay: 0.0500 (0.0500)  time: 0.6380  data: 0.0428  max mem: 41794
Epoch: [251]  [ 70/312]  eta: 0:03:42  lr: 0.000293  min_lr: 0.000293  loss: 3.2661 (3.0825)  weight_decay: 0.0500 (0.0500)  time: 0.7522  data: 0.1175  max mem: 41794
Epoch: [251]  [ 80/312]  eta: 0:03:30  lr: 0.000293  min_lr: 0.000293  loss: 3.0195 (3.0600)  weight_decay: 0.0500 (0.0500)  time: 0.8526  data: 0.1900  max mem: 41794
Epoch: [251]  [ 90/312]  eta: 0:03:13  lr: 0.000292  min_lr: 0.000292  loss: 2.8049 (3.0414)  weight_decay: 0.0500 (0.0500)  time: 0.7010  data: 0.1072  max mem: 41794
Epoch: [251]  [100/312]  eta: 0:03:04  lr: 0.000292  min_lr: 0.000292  loss: 3.0938 (3.0479)  weight_decay: 0.0500 (0.0500)  time: 0.7193  data: 0.1126  max mem: 41794
Epoch: [251]  [110/312]  eta: 0:02:49  lr: 0.000292  min_lr: 0.000292  loss: 3.2610 (3.0502)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.0786  max mem: 41794
Epoch: [251]  [120/312]  eta: 0:02:41  lr: 0.000291  min_lr: 0.000291  loss: 2.7391 (3.0089)  weight_decay: 0.0500 (0.0500)  time: 0.7009  data: 0.0827  max mem: 41794
Epoch: [251]  [130/312]  eta: 0:02:33  lr: 0.000291  min_lr: 0.000291  loss: 2.7391 (3.0090)  weight_decay: 0.0500 (0.0500)  time: 0.8885  data: 0.1838  max mem: 41794
Epoch: [251]  [140/312]  eta: 0:02:21  lr: 0.000290  min_lr: 0.000290  loss: 3.2344 (3.0218)  weight_decay: 0.0500 (0.0500)  time: 0.6966  data: 0.1018  max mem: 41794
Epoch: [251]  [150/312]  eta: 0:02:14  lr: 0.000290  min_lr: 0.000290  loss: 3.1778 (3.0171)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1176  max mem: 41794
Epoch: [251]  [160/312]  eta: 0:02:05  lr: 0.000290  min_lr: 0.000290  loss: 3.0347 (3.0198)  weight_decay: 0.0500 (0.0500)  time: 0.8330  data: 0.1602  max mem: 41794
Epoch: [251]  [170/312]  eta: 0:01:55  lr: 0.000289  min_lr: 0.000289  loss: 3.2770 (3.0226)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1067  max mem: 41794
Epoch: [251]  [180/312]  eta: 0:01:47  lr: 0.000289  min_lr: 0.000289  loss: 3.2260 (3.0202)  weight_decay: 0.0500 (0.0500)  time: 0.7416  data: 0.1335  max mem: 41794
Epoch: [251]  [190/312]  eta: 0:01:37  lr: 0.000289  min_lr: 0.000289  loss: 2.9904 (3.0175)  weight_decay: 0.0500 (0.0500)  time: 0.6767  data: 0.0736  max mem: 41794
Epoch: [251]  [200/312]  eta: 0:01:29  lr: 0.000288  min_lr: 0.000288  loss: 3.0775 (3.0266)  weight_decay: 0.0500 (0.0500)  time: 0.6833  data: 0.1061  max mem: 41794
Epoch: [251]  [210/312]  eta: 0:01:22  lr: 0.000288  min_lr: 0.000288  loss: 3.2594 (3.0197)  weight_decay: 0.0500 (0.0500)  time: 0.8811  data: 0.1963  max mem: 41794
Epoch: [251]  [220/312]  eta: 0:01:12  lr: 0.000287  min_lr: 0.000287  loss: 3.2352 (3.0320)  weight_decay: 0.0500 (0.0500)  time: 0.6922  data: 0.0943  max mem: 41794
Epoch: [251]  [230/312]  eta: 0:01:05  lr: 0.000287  min_lr: 0.000287  loss: 3.2302 (3.0314)  weight_decay: 0.0500 (0.0500)  time: 0.6623  data: 0.0782  max mem: 41794
Epoch: [251]  [240/312]  eta: 0:00:56  lr: 0.000287  min_lr: 0.000287  loss: 3.0463 (3.0341)  weight_decay: 0.0500 (0.0500)  time: 0.7846  data: 0.0817  max mem: 41794
Epoch: [251]  [250/312]  eta: 0:00:48  lr: 0.000286  min_lr: 0.000286  loss: 3.0027 (3.0287)  weight_decay: 0.0500 (0.0500)  time: 0.6982  data: 0.0785  max mem: 41794
Epoch: [251]  [260/312]  eta: 0:00:40  lr: 0.000286  min_lr: 0.000286  loss: 2.9897 (3.0314)  weight_decay: 0.0500 (0.0500)  time: 0.7419  data: 0.1288  max mem: 41794
Epoch: [251]  [270/312]  eta: 0:00:32  lr: 0.000286  min_lr: 0.000286  loss: 3.2650 (3.0398)  weight_decay: 0.0500 (0.0500)  time: 0.6659  data: 0.0584  max mem: 41794
Epoch: [251]  [280/312]  eta: 0:00:24  lr: 0.000285  min_lr: 0.000285  loss: 3.3594 (3.0499)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.0760  max mem: 41794
Epoch: [251]  [290/312]  eta: 0:00:17  lr: 0.000285  min_lr: 0.000285  loss: 3.2354 (3.0494)  weight_decay: 0.0500 (0.0500)  time: 0.8140  data: 0.1174  max mem: 41794
Epoch: [251]  [300/312]  eta: 0:00:09  lr: 0.000284  min_lr: 0.000284  loss: 3.2009 (3.0460)  weight_decay: 0.0500 (0.0500)  time: 0.6332  data: 0.0458  max mem: 41794
Epoch: [251]  [310/312]  eta: 0:00:01  lr: 0.000284  min_lr: 0.000284  loss: 3.2009 (3.0481)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.0001  max mem: 41794
Epoch: [251]  [311/312]  eta: 0:00:00  lr: 0.000284  min_lr: 0.000284  loss: 3.1413 (3.0463)  weight_decay: 0.0500 (0.0500)  time: 0.4650  data: 0.0001  max mem: 41794
Epoch: [251] Total time: 0:03:58 (0.7652 s / it)
Averaged stats: lr: 0.000284  min_lr: 0.000284  loss: 3.1413 (3.0594)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 0.8307 (0.8307)  acc1: 85.5469 (85.5469)  acc5: 95.9635 (95.9635)  time: 8.1719  data: 7.9601  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1200 (1.0669)  acc1: 77.0833 (78.2080)  acc5: 94.9219 (94.7040)  time: 1.0635  data: 0.8846  max mem: 41794
Test: Total time: 0:00:09 (1.0740 s / it)
* Acc@1 79.076 Acc@5 94.682 loss 1.057
Accuracy of the model on the 50000 test images: 79.1%
Max accuracy: 79.20%
Epoch: [252]  [  0/312]  eta: 1:25:02  lr: 0.000284  min_lr: 0.000284  loss: 2.6575 (2.6575)  weight_decay: 0.0500 (0.0500)  time: 16.3557  data: 11.8632  max mem: 41794
Epoch: [252]  [ 10/312]  eta: 0:11:39  lr: 0.000284  min_lr: 0.000284  loss: 3.2765 (3.1172)  weight_decay: 0.0500 (0.0500)  time: 2.3173  data: 1.3407  max mem: 41794
Epoch: [252]  [ 20/312]  eta: 0:07:18  lr: 0.000283  min_lr: 0.000283  loss: 3.0478 (3.0337)  weight_decay: 0.0500 (0.0500)  time: 0.7599  data: 0.1474  max mem: 41794
Epoch: [252]  [ 30/312]  eta: 0:05:34  lr: 0.000283  min_lr: 0.000283  loss: 3.1088 (3.1067)  weight_decay: 0.0500 (0.0500)  time: 0.5633  data: 0.0057  max mem: 41794
Epoch: [252]  [ 40/312]  eta: 0:04:42  lr: 0.000283  min_lr: 0.000283  loss: 3.2955 (3.1173)  weight_decay: 0.0500 (0.0500)  time: 0.5533  data: 0.0073  max mem: 41794
Epoch: [252]  [ 50/312]  eta: 0:04:21  lr: 0.000282  min_lr: 0.000282  loss: 3.2110 (3.0764)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.0209  max mem: 41794
Epoch: [252]  [ 60/312]  eta: 0:03:51  lr: 0.000282  min_lr: 0.000282  loss: 3.0678 (3.1030)  weight_decay: 0.0500 (0.0500)  time: 0.6706  data: 0.0166  max mem: 41794
Epoch: [252]  [ 70/312]  eta: 0:03:40  lr: 0.000281  min_lr: 0.000281  loss: 3.0567 (3.0843)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.0810  max mem: 41794
Epoch: [252]  [ 80/312]  eta: 0:03:28  lr: 0.000281  min_lr: 0.000281  loss: 3.0733 (3.0992)  weight_decay: 0.0500 (0.0500)  time: 0.8374  data: 0.1694  max mem: 41794
Epoch: [252]  [ 90/312]  eta: 0:03:09  lr: 0.000281  min_lr: 0.000281  loss: 3.2160 (3.0838)  weight_decay: 0.0500 (0.0500)  time: 0.6528  data: 0.0891  max mem: 41794
Epoch: [252]  [100/312]  eta: 0:03:01  lr: 0.000280  min_lr: 0.000280  loss: 3.1112 (3.0876)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.1115  max mem: 41794
Epoch: [252]  [110/312]  eta: 0:02:46  lr: 0.000280  min_lr: 0.000280  loss: 3.2336 (3.0916)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1127  max mem: 41794
Epoch: [252]  [120/312]  eta: 0:02:39  lr: 0.000280  min_lr: 0.000280  loss: 3.2336 (3.0885)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1160  max mem: 41794
Epoch: [252]  [130/312]  eta: 0:02:30  lr: 0.000279  min_lr: 0.000279  loss: 3.2106 (3.0913)  weight_decay: 0.0500 (0.0500)  time: 0.8436  data: 0.2280  max mem: 41794
Epoch: [252]  [140/312]  eta: 0:02:18  lr: 0.000279  min_lr: 0.000279  loss: 3.2763 (3.0936)  weight_decay: 0.0500 (0.0500)  time: 0.6603  data: 0.1165  max mem: 41794
Epoch: [252]  [150/312]  eta: 0:02:11  lr: 0.000279  min_lr: 0.000279  loss: 3.1087 (3.0818)  weight_decay: 0.0500 (0.0500)  time: 0.6955  data: 0.1708  max mem: 41794
Epoch: [252]  [160/312]  eta: 0:02:03  lr: 0.000278  min_lr: 0.000278  loss: 3.0172 (3.0752)  weight_decay: 0.0500 (0.0500)  time: 0.8882  data: 0.3352  max mem: 41794
Epoch: [252]  [170/312]  eta: 0:01:53  lr: 0.000278  min_lr: 0.000278  loss: 2.8408 (3.0639)  weight_decay: 0.0500 (0.0500)  time: 0.6928  data: 0.1721  max mem: 41794
Epoch: [252]  [180/312]  eta: 0:01:45  lr: 0.000277  min_lr: 0.000277  loss: 2.8350 (3.0522)  weight_decay: 0.0500 (0.0500)  time: 0.6946  data: 0.1418  max mem: 41794
Epoch: [252]  [190/312]  eta: 0:01:35  lr: 0.000277  min_lr: 0.000277  loss: 3.0863 (3.0576)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.1399  max mem: 41794
Epoch: [252]  [200/312]  eta: 0:01:28  lr: 0.000277  min_lr: 0.000277  loss: 3.2243 (3.0527)  weight_decay: 0.0500 (0.0500)  time: 0.6866  data: 0.1603  max mem: 41794
Epoch: [252]  [210/312]  eta: 0:01:21  lr: 0.000276  min_lr: 0.000276  loss: 2.9483 (3.0507)  weight_decay: 0.0500 (0.0500)  time: 0.8849  data: 0.3053  max mem: 41794
Epoch: [252]  [220/312]  eta: 0:01:11  lr: 0.000276  min_lr: 0.000276  loss: 3.1304 (3.0539)  weight_decay: 0.0500 (0.0500)  time: 0.6930  data: 0.1495  max mem: 41794
Epoch: [252]  [230/312]  eta: 0:01:04  lr: 0.000276  min_lr: 0.000276  loss: 3.1868 (3.0528)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.1149  max mem: 41794
Epoch: [252]  [240/312]  eta: 0:00:56  lr: 0.000275  min_lr: 0.000275  loss: 3.1641 (3.0600)  weight_decay: 0.0500 (0.0500)  time: 0.8604  data: 0.2519  max mem: 41794
Epoch: [252]  [250/312]  eta: 0:00:48  lr: 0.000275  min_lr: 0.000275  loss: 3.1460 (3.0586)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1392  max mem: 41794
Epoch: [252]  [260/312]  eta: 0:00:40  lr: 0.000275  min_lr: 0.000275  loss: 2.9461 (3.0542)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.1165  max mem: 41794
Epoch: [252]  [270/312]  eta: 0:00:32  lr: 0.000274  min_lr: 0.000274  loss: 2.8701 (3.0436)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.1164  max mem: 41794
Epoch: [252]  [280/312]  eta: 0:00:24  lr: 0.000274  min_lr: 0.000274  loss: 3.1758 (3.0508)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.1214  max mem: 41794
Epoch: [252]  [290/312]  eta: 0:00:17  lr: 0.000273  min_lr: 0.000273  loss: 3.2691 (3.0531)  weight_decay: 0.0500 (0.0500)  time: 0.8502  data: 0.2215  max mem: 41794
Epoch: [252]  [300/312]  eta: 0:00:09  lr: 0.000273  min_lr: 0.000273  loss: 3.2617 (3.0621)  weight_decay: 0.0500 (0.0500)  time: 0.6665  data: 0.1005  max mem: 41794
Epoch: [252]  [310/312]  eta: 0:00:01  lr: 0.000273  min_lr: 0.000273  loss: 3.2430 (3.0560)  weight_decay: 0.0500 (0.0500)  time: 0.4694  data: 0.0001  max mem: 41794
Epoch: [252]  [311/312]  eta: 0:00:00  lr: 0.000273  min_lr: 0.000273  loss: 3.0047 (3.0555)  weight_decay: 0.0500 (0.0500)  time: 0.4693  data: 0.0001  max mem: 41794
Epoch: [252] Total time: 0:03:57 (0.7608 s / it)
Averaged stats: lr: 0.000273  min_lr: 0.000273  loss: 3.0047 (3.0580)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 0.8179 (0.8179)  acc1: 85.1562 (85.1562)  acc5: 97.1354 (97.1354)  time: 8.1249  data: 7.9175  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1316 (1.0613)  acc1: 78.1250 (78.7200)  acc5: 94.6615 (94.9760)  time: 1.0701  data: 0.8992  max mem: 41794
Test: Total time: 0:00:09 (1.0819 s / it)
* Acc@1 79.166 Acc@5 94.814 loss 1.053
Accuracy of the model on the 50000 test images: 79.2%
Max accuracy: 79.20%
Epoch: [253]  [  0/312]  eta: 1:22:04  lr: 0.000273  min_lr: 0.000273  loss: 3.4902 (3.4902)  weight_decay: 0.0500 (0.0500)  time: 15.7839  data: 14.0774  max mem: 41794
Epoch: [253]  [ 10/312]  eta: 0:10:51  lr: 0.000272  min_lr: 0.000272  loss: 2.8906 (2.9551)  weight_decay: 0.0500 (0.0500)  time: 2.1575  data: 1.3656  max mem: 41794
Epoch: [253]  [ 20/312]  eta: 0:07:08  lr: 0.000272  min_lr: 0.000272  loss: 3.0419 (3.0744)  weight_decay: 0.0500 (0.0500)  time: 0.7507  data: 0.1372  max mem: 41794
Epoch: [253]  [ 30/312]  eta: 0:05:27  lr: 0.000272  min_lr: 0.000272  loss: 3.0921 (3.0206)  weight_decay: 0.0500 (0.0500)  time: 0.6157  data: 0.0904  max mem: 41794
Epoch: [253]  [ 40/312]  eta: 0:04:43  lr: 0.000271  min_lr: 0.000271  loss: 2.7036 (2.9182)  weight_decay: 0.0500 (0.0500)  time: 0.5971  data: 0.0879  max mem: 41794
Epoch: [253]  [ 50/312]  eta: 0:04:18  lr: 0.000271  min_lr: 0.000271  loss: 2.7036 (2.9633)  weight_decay: 0.0500 (0.0500)  time: 0.7144  data: 0.2156  max mem: 41794
Epoch: [253]  [ 60/312]  eta: 0:03:50  lr: 0.000270  min_lr: 0.000270  loss: 3.1742 (2.9735)  weight_decay: 0.0500 (0.0500)  time: 0.6546  data: 0.1286  max mem: 41794
Epoch: [253]  [ 70/312]  eta: 0:03:40  lr: 0.000270  min_lr: 0.000270  loss: 3.1401 (2.9991)  weight_decay: 0.0500 (0.0500)  time: 0.7148  data: 0.1825  max mem: 41794
Epoch: [253]  [ 80/312]  eta: 0:03:28  lr: 0.000270  min_lr: 0.000270  loss: 3.0743 (2.9974)  weight_decay: 0.0500 (0.0500)  time: 0.8467  data: 0.3462  max mem: 41794
Epoch: [253]  [ 90/312]  eta: 0:03:10  lr: 0.000269  min_lr: 0.000269  loss: 3.1515 (3.0132)  weight_decay: 0.0500 (0.0500)  time: 0.6686  data: 0.1649  max mem: 41794
Epoch: [253]  [100/312]  eta: 0:03:02  lr: 0.000269  min_lr: 0.000269  loss: 3.1515 (3.0146)  weight_decay: 0.0500 (0.0500)  time: 0.7065  data: 0.1829  max mem: 41794
Epoch: [253]  [110/312]  eta: 0:02:47  lr: 0.000269  min_lr: 0.000269  loss: 3.1971 (3.0351)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.1826  max mem: 41794
Epoch: [253]  [120/312]  eta: 0:02:40  lr: 0.000268  min_lr: 0.000268  loss: 3.0814 (3.0181)  weight_decay: 0.0500 (0.0500)  time: 0.7147  data: 0.2180  max mem: 41794
Epoch: [253]  [130/312]  eta: 0:02:32  lr: 0.000268  min_lr: 0.000268  loss: 3.0718 (3.0326)  weight_decay: 0.0500 (0.0500)  time: 0.9067  data: 0.4032  max mem: 41794
Epoch: [253]  [140/312]  eta: 0:02:20  lr: 0.000268  min_lr: 0.000268  loss: 3.2277 (3.0320)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.1857  max mem: 41794
Epoch: [253]  [150/312]  eta: 0:02:12  lr: 0.000267  min_lr: 0.000267  loss: 3.2049 (3.0196)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.1867  max mem: 41794
Epoch: [253]  [160/312]  eta: 0:02:04  lr: 0.000267  min_lr: 0.000267  loss: 2.8541 (3.0143)  weight_decay: 0.0500 (0.0500)  time: 0.8607  data: 0.3587  max mem: 41794
Epoch: [253]  [170/312]  eta: 0:01:53  lr: 0.000267  min_lr: 0.000267  loss: 3.0338 (3.0083)  weight_decay: 0.0500 (0.0500)  time: 0.6647  data: 0.1726  max mem: 41794
Epoch: [253]  [180/312]  eta: 0:01:46  lr: 0.000266  min_lr: 0.000266  loss: 2.7694 (2.9975)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.1800  max mem: 41794
Epoch: [253]  [190/312]  eta: 0:01:36  lr: 0.000266  min_lr: 0.000266  loss: 3.1852 (3.0058)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.1800  max mem: 41794
Epoch: [253]  [200/312]  eta: 0:01:28  lr: 0.000265  min_lr: 0.000265  loss: 3.1994 (3.0167)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.1807  max mem: 41794
Epoch: [253]  [210/312]  eta: 0:01:21  lr: 0.000265  min_lr: 0.000265  loss: 3.2882 (3.0086)  weight_decay: 0.0500 (0.0500)  time: 0.8683  data: 0.3535  max mem: 41794
Epoch: [253]  [220/312]  eta: 0:01:11  lr: 0.000265  min_lr: 0.000265  loss: 2.9295 (3.0107)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1734  max mem: 41794
Epoch: [253]  [230/312]  eta: 0:01:04  lr: 0.000264  min_lr: 0.000264  loss: 3.2562 (3.0207)  weight_decay: 0.0500 (0.0500)  time: 0.7131  data: 0.1832  max mem: 41794
Epoch: [253]  [240/312]  eta: 0:00:56  lr: 0.000264  min_lr: 0.000264  loss: 3.1952 (3.0217)  weight_decay: 0.0500 (0.0500)  time: 0.8621  data: 0.3284  max mem: 41794
Epoch: [253]  [250/312]  eta: 0:00:48  lr: 0.000264  min_lr: 0.000264  loss: 3.0747 (3.0239)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.1458  max mem: 41794
Epoch: [253]  [260/312]  eta: 0:00:40  lr: 0.000263  min_lr: 0.000263  loss: 3.2885 (3.0294)  weight_decay: 0.0500 (0.0500)  time: 0.7134  data: 0.1332  max mem: 41794
Epoch: [253]  [270/312]  eta: 0:00:32  lr: 0.000263  min_lr: 0.000263  loss: 3.2578 (3.0324)  weight_decay: 0.0500 (0.0500)  time: 0.6617  data: 0.1350  max mem: 41794
Epoch: [253]  [280/312]  eta: 0:00:24  lr: 0.000263  min_lr: 0.000263  loss: 3.1656 (3.0346)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.1537  max mem: 41794
Epoch: [253]  [290/312]  eta: 0:00:17  lr: 0.000262  min_lr: 0.000262  loss: 3.2089 (3.0448)  weight_decay: 0.0500 (0.0500)  time: 0.8169  data: 0.2603  max mem: 41794
Epoch: [253]  [300/312]  eta: 0:00:09  lr: 0.000262  min_lr: 0.000262  loss: 3.2994 (3.0504)  weight_decay: 0.0500 (0.0500)  time: 0.6188  data: 0.1087  max mem: 41794
Epoch: [253]  [310/312]  eta: 0:00:01  lr: 0.000262  min_lr: 0.000262  loss: 3.3731 (3.0607)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [253]  [311/312]  eta: 0:00:00  lr: 0.000261  min_lr: 0.000261  loss: 3.4583 (3.0630)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [253] Total time: 0:03:57 (0.7602 s / it)
Averaged stats: lr: 0.000261  min_lr: 0.000261  loss: 3.4583 (3.0539)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:11  loss: 0.9670 (0.9670)  acc1: 84.7656 (84.7656)  acc5: 96.7448 (96.7448)  time: 7.8916  data: 7.6887  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2569 (1.1734)  acc1: 78.3854 (78.2880)  acc5: 94.7917 (94.6560)  time: 1.0822  data: 0.9111  max mem: 41794
Test: Total time: 0:00:10 (1.1132 s / it)
* Acc@1 78.884 Acc@5 94.634 loss 1.171
Accuracy of the model on the 50000 test images: 78.9%
Max accuracy: 79.20%
Epoch: [254]  [  0/312]  eta: 1:22:48  lr: 0.000261  min_lr: 0.000261  loss: 3.5601 (3.5601)  weight_decay: 0.0500 (0.0500)  time: 15.9233  data: 12.7319  max mem: 41794
Epoch: [254]  [ 10/312]  eta: 0:11:21  lr: 0.000261  min_lr: 0.000261  loss: 3.4129 (3.1800)  weight_decay: 0.0500 (0.0500)  time: 2.2573  data: 1.3818  max mem: 41794
Epoch: [254]  [ 20/312]  eta: 0:07:28  lr: 0.000261  min_lr: 0.000261  loss: 3.1775 (3.0727)  weight_decay: 0.0500 (0.0500)  time: 0.8156  data: 0.1560  max mem: 41794
Epoch: [254]  [ 30/312]  eta: 0:05:42  lr: 0.000260  min_lr: 0.000260  loss: 3.1927 (3.0832)  weight_decay: 0.0500 (0.0500)  time: 0.6387  data: 0.0369  max mem: 41794
Epoch: [254]  [ 40/312]  eta: 0:04:44  lr: 0.000260  min_lr: 0.000260  loss: 3.1780 (3.0806)  weight_decay: 0.0500 (0.0500)  time: 0.5316  data: 0.0046  max mem: 41794
Epoch: [254]  [ 50/312]  eta: 0:04:20  lr: 0.000260  min_lr: 0.000260  loss: 3.0484 (3.0344)  weight_decay: 0.0500 (0.0500)  time: 0.6568  data: 0.0443  max mem: 41794
Epoch: [254]  [ 60/312]  eta: 0:03:51  lr: 0.000259  min_lr: 0.000259  loss: 3.0726 (3.0550)  weight_decay: 0.0500 (0.0500)  time: 0.6583  data: 0.0642  max mem: 41794
Epoch: [254]  [ 70/312]  eta: 0:03:42  lr: 0.000259  min_lr: 0.000259  loss: 3.2379 (3.0619)  weight_decay: 0.0500 (0.0500)  time: 0.7327  data: 0.0822  max mem: 41794
Epoch: [254]  [ 80/312]  eta: 0:03:25  lr: 0.000259  min_lr: 0.000259  loss: 3.1633 (3.0687)  weight_decay: 0.0500 (0.0500)  time: 0.7903  data: 0.0668  max mem: 41794
Epoch: [254]  [ 90/312]  eta: 0:03:09  lr: 0.000258  min_lr: 0.000258  loss: 3.1120 (3.0551)  weight_decay: 0.0500 (0.0500)  time: 0.6216  data: 0.0317  max mem: 41794
Epoch: [254]  [100/312]  eta: 0:03:02  lr: 0.000258  min_lr: 0.000258  loss: 3.1120 (3.0543)  weight_decay: 0.0500 (0.0500)  time: 0.7543  data: 0.0964  max mem: 41794
Epoch: [254]  [110/312]  eta: 0:02:47  lr: 0.000258  min_lr: 0.000258  loss: 3.2330 (3.0664)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.0698  max mem: 41794
Epoch: [254]  [120/312]  eta: 0:02:39  lr: 0.000257  min_lr: 0.000257  loss: 3.2363 (3.0846)  weight_decay: 0.0500 (0.0500)  time: 0.6673  data: 0.0568  max mem: 41794
Epoch: [254]  [130/312]  eta: 0:02:32  lr: 0.000257  min_lr: 0.000257  loss: 3.2519 (3.0918)  weight_decay: 0.0500 (0.0500)  time: 0.8778  data: 0.1245  max mem: 41794
Epoch: [254]  [140/312]  eta: 0:02:20  lr: 0.000257  min_lr: 0.000257  loss: 2.9910 (3.0693)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.1133  max mem: 41794
Epoch: [254]  [150/312]  eta: 0:02:12  lr: 0.000256  min_lr: 0.000256  loss: 2.8856 (3.0646)  weight_decay: 0.0500 (0.0500)  time: 0.7243  data: 0.0937  max mem: 41794
Epoch: [254]  [160/312]  eta: 0:02:02  lr: 0.000256  min_lr: 0.000256  loss: 2.9722 (3.0468)  weight_decay: 0.0500 (0.0500)  time: 0.7499  data: 0.0727  max mem: 41794
Epoch: [254]  [170/312]  eta: 0:01:53  lr: 0.000255  min_lr: 0.000255  loss: 3.0079 (3.0584)  weight_decay: 0.0500 (0.0500)  time: 0.6448  data: 0.0585  max mem: 41794
Epoch: [254]  [180/312]  eta: 0:01:45  lr: 0.000255  min_lr: 0.000255  loss: 3.3066 (3.0673)  weight_decay: 0.0500 (0.0500)  time: 0.7260  data: 0.1241  max mem: 41794
Epoch: [254]  [190/312]  eta: 0:01:35  lr: 0.000255  min_lr: 0.000255  loss: 3.2038 (3.0664)  weight_decay: 0.0500 (0.0500)  time: 0.6452  data: 0.0931  max mem: 41794
Epoch: [254]  [200/312]  eta: 0:01:28  lr: 0.000254  min_lr: 0.000254  loss: 3.3291 (3.0743)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.1199  max mem: 41794
Epoch: [254]  [210/312]  eta: 0:01:20  lr: 0.000254  min_lr: 0.000254  loss: 3.2744 (3.0832)  weight_decay: 0.0500 (0.0500)  time: 0.8054  data: 0.2083  max mem: 41794
Epoch: [254]  [220/312]  eta: 0:01:11  lr: 0.000254  min_lr: 0.000254  loss: 3.2744 (3.0815)  weight_decay: 0.0500 (0.0500)  time: 0.6149  data: 0.0919  max mem: 41794
Epoch: [254]  [230/312]  eta: 0:01:03  lr: 0.000253  min_lr: 0.000253  loss: 3.1778 (3.0787)  weight_decay: 0.0500 (0.0500)  time: 0.7106  data: 0.1892  max mem: 41794
Epoch: [254]  [240/312]  eta: 0:00:56  lr: 0.000253  min_lr: 0.000253  loss: 3.2311 (3.0819)  weight_decay: 0.0500 (0.0500)  time: 0.8348  data: 0.3189  max mem: 41794
Epoch: [254]  [250/312]  eta: 0:00:47  lr: 0.000253  min_lr: 0.000253  loss: 3.2311 (3.0766)  weight_decay: 0.0500 (0.0500)  time: 0.6964  data: 0.1451  max mem: 41794
Epoch: [254]  [260/312]  eta: 0:00:40  lr: 0.000252  min_lr: 0.000252  loss: 3.0981 (3.0726)  weight_decay: 0.0500 (0.0500)  time: 0.7451  data: 0.1214  max mem: 41794
Epoch: [254]  [270/312]  eta: 0:00:32  lr: 0.000252  min_lr: 0.000252  loss: 3.2722 (3.0784)  weight_decay: 0.0500 (0.0500)  time: 0.6766  data: 0.1067  max mem: 41794
Epoch: [254]  [280/312]  eta: 0:00:24  lr: 0.000252  min_lr: 0.000252  loss: 3.1472 (3.0737)  weight_decay: 0.0500 (0.0500)  time: 0.6946  data: 0.1175  max mem: 41794
Epoch: [254]  [290/312]  eta: 0:00:17  lr: 0.000251  min_lr: 0.000251  loss: 2.9873 (3.0649)  weight_decay: 0.0500 (0.0500)  time: 0.8827  data: 0.2958  max mem: 41794
Epoch: [254]  [300/312]  eta: 0:00:09  lr: 0.000251  min_lr: 0.000251  loss: 2.8635 (3.0595)  weight_decay: 0.0500 (0.0500)  time: 0.6731  data: 0.1787  max mem: 41794
Epoch: [254]  [310/312]  eta: 0:00:01  lr: 0.000251  min_lr: 0.000251  loss: 2.9349 (3.0507)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [254]  [311/312]  eta: 0:00:00  lr: 0.000251  min_lr: 0.000251  loss: 2.9643 (3.0529)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [254] Total time: 0:03:56 (0.7591 s / it)
Averaged stats: lr: 0.000251  min_lr: 0.000251  loss: 2.9643 (3.0428)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.8241 (0.8241)  acc1: 86.3281 (86.3281)  acc5: 96.4844 (96.4844)  time: 8.6486  data: 8.4429  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1086 (1.0241)  acc1: 78.3854 (78.9120)  acc5: 94.6615 (94.8640)  time: 1.1103  data: 0.9382  max mem: 41794
Test: Total time: 0:00:10 (1.1538 s / it)
* Acc@1 79.372 Acc@5 94.890 loss 1.019
Accuracy of the model on the 50000 test images: 79.4%
Max accuracy: 79.37%
Epoch: [255]  [  0/312]  eta: 1:20:24  lr: 0.000250  min_lr: 0.000250  loss: 3.6580 (3.6580)  weight_decay: 0.0500 (0.0500)  time: 15.4634  data: 13.4381  max mem: 41794
Epoch: [255]  [ 10/312]  eta: 0:10:56  lr: 0.000250  min_lr: 0.000250  loss: 3.3098 (3.2611)  weight_decay: 0.0500 (0.0500)  time: 2.1742  data: 1.4462  max mem: 41794
Epoch: [255]  [ 20/312]  eta: 0:07:14  lr: 0.000250  min_lr: 0.000250  loss: 3.1445 (3.1147)  weight_decay: 0.0500 (0.0500)  time: 0.7909  data: 0.2407  max mem: 41794
Epoch: [255]  [ 30/312]  eta: 0:05:30  lr: 0.000249  min_lr: 0.000249  loss: 3.0780 (3.0670)  weight_decay: 0.0500 (0.0500)  time: 0.6198  data: 0.1192  max mem: 41794
Epoch: [255]  [ 40/312]  eta: 0:04:41  lr: 0.000249  min_lr: 0.000249  loss: 2.8862 (3.0363)  weight_decay: 0.0500 (0.0500)  time: 0.5585  data: 0.0643  max mem: 41794
Epoch: [255]  [ 50/312]  eta: 0:04:18  lr: 0.000249  min_lr: 0.000249  loss: 3.0007 (3.0096)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.2055  max mem: 41794
Epoch: [255]  [ 60/312]  eta: 0:03:47  lr: 0.000248  min_lr: 0.000248  loss: 3.1373 (3.0167)  weight_decay: 0.0500 (0.0500)  time: 0.6348  data: 0.1435  max mem: 41794
Epoch: [255]  [ 70/312]  eta: 0:03:40  lr: 0.000248  min_lr: 0.000248  loss: 3.1515 (3.0318)  weight_decay: 0.0500 (0.0500)  time: 0.7134  data: 0.2050  max mem: 41794
Epoch: [255]  [ 80/312]  eta: 0:03:29  lr: 0.000248  min_lr: 0.000248  loss: 2.8106 (3.0090)  weight_decay: 0.0500 (0.0500)  time: 0.8974  data: 0.3394  max mem: 41794
Epoch: [255]  [ 90/312]  eta: 0:03:11  lr: 0.000247  min_lr: 0.000247  loss: 2.7726 (3.0174)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1470  max mem: 41794
Epoch: [255]  [100/312]  eta: 0:03:04  lr: 0.000247  min_lr: 0.000247  loss: 3.1724 (3.0040)  weight_decay: 0.0500 (0.0500)  time: 0.7485  data: 0.1690  max mem: 41794
Epoch: [255]  [110/312]  eta: 0:02:49  lr: 0.000247  min_lr: 0.000247  loss: 3.0177 (2.9933)  weight_decay: 0.0500 (0.0500)  time: 0.7427  data: 0.1571  max mem: 41794
Epoch: [255]  [120/312]  eta: 0:02:41  lr: 0.000246  min_lr: 0.000246  loss: 2.9424 (2.9862)  weight_decay: 0.0500 (0.0500)  time: 0.6986  data: 0.0967  max mem: 41794
Epoch: [255]  [130/312]  eta: 0:02:35  lr: 0.000246  min_lr: 0.000246  loss: 3.1842 (2.9996)  weight_decay: 0.0500 (0.0500)  time: 0.9235  data: 0.1773  max mem: 41794
Epoch: [255]  [140/312]  eta: 0:02:22  lr: 0.000246  min_lr: 0.000246  loss: 3.1113 (2.9891)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.0862  max mem: 41794
Epoch: [255]  [150/312]  eta: 0:02:14  lr: 0.000245  min_lr: 0.000245  loss: 2.9083 (2.9910)  weight_decay: 0.0500 (0.0500)  time: 0.6798  data: 0.0563  max mem: 41794
Epoch: [255]  [160/312]  eta: 0:02:05  lr: 0.000245  min_lr: 0.000245  loss: 3.1559 (2.9901)  weight_decay: 0.0500 (0.0500)  time: 0.8058  data: 0.0546  max mem: 41794
Epoch: [255]  [170/312]  eta: 0:01:55  lr: 0.000245  min_lr: 0.000245  loss: 2.8488 (2.9823)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.0643  max mem: 41794
Epoch: [255]  [180/312]  eta: 0:01:47  lr: 0.000244  min_lr: 0.000244  loss: 3.0575 (2.9979)  weight_decay: 0.0500 (0.0500)  time: 0.7385  data: 0.1314  max mem: 41794
Epoch: [255]  [190/312]  eta: 0:01:37  lr: 0.000244  min_lr: 0.000244  loss: 3.1452 (3.0031)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.0726  max mem: 41794
Epoch: [255]  [200/312]  eta: 0:01:29  lr: 0.000244  min_lr: 0.000244  loss: 3.1452 (3.0070)  weight_decay: 0.0500 (0.0500)  time: 0.6330  data: 0.0407  max mem: 41794
Epoch: [255]  [210/312]  eta: 0:01:21  lr: 0.000243  min_lr: 0.000243  loss: 2.7557 (2.9901)  weight_decay: 0.0500 (0.0500)  time: 0.8332  data: 0.1071  max mem: 41794
Epoch: [255]  [220/312]  eta: 0:01:12  lr: 0.000243  min_lr: 0.000243  loss: 2.9603 (2.9896)  weight_decay: 0.0500 (0.0500)  time: 0.7077  data: 0.0701  max mem: 41794
Epoch: [255]  [230/312]  eta: 0:01:04  lr: 0.000243  min_lr: 0.000243  loss: 3.0363 (2.9870)  weight_decay: 0.0500 (0.0500)  time: 0.6392  data: 0.0131  max mem: 41794
Epoch: [255]  [240/312]  eta: 0:00:56  lr: 0.000242  min_lr: 0.000242  loss: 3.0363 (2.9895)  weight_decay: 0.0500 (0.0500)  time: 0.8093  data: 0.0736  max mem: 41794
Epoch: [255]  [250/312]  eta: 0:00:48  lr: 0.000242  min_lr: 0.000242  loss: 3.3263 (3.0096)  weight_decay: 0.0500 (0.0500)  time: 0.6702  data: 0.0628  max mem: 41794
Epoch: [255]  [260/312]  eta: 0:00:40  lr: 0.000242  min_lr: 0.000242  loss: 3.3453 (3.0080)  weight_decay: 0.0500 (0.0500)  time: 0.6792  data: 0.0793  max mem: 41794
Epoch: [255]  [270/312]  eta: 0:00:32  lr: 0.000241  min_lr: 0.000241  loss: 2.9874 (3.0050)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.0792  max mem: 41794
Epoch: [255]  [280/312]  eta: 0:00:24  lr: 0.000241  min_lr: 0.000241  loss: 2.9300 (3.0003)  weight_decay: 0.0500 (0.0500)  time: 0.7168  data: 0.0982  max mem: 41794
Epoch: [255]  [290/312]  eta: 0:00:17  lr: 0.000240  min_lr: 0.000240  loss: 3.0408 (3.0053)  weight_decay: 0.0500 (0.0500)  time: 0.8090  data: 0.1192  max mem: 41794
Epoch: [255]  [300/312]  eta: 0:00:09  lr: 0.000240  min_lr: 0.000240  loss: 3.2130 (3.0100)  weight_decay: 0.0500 (0.0500)  time: 0.5806  data: 0.0260  max mem: 41794
Epoch: [255]  [310/312]  eta: 0:00:01  lr: 0.000240  min_lr: 0.000240  loss: 3.1385 (3.0063)  weight_decay: 0.0500 (0.0500)  time: 0.4683  data: 0.0048  max mem: 41794
Epoch: [255]  [311/312]  eta: 0:00:00  lr: 0.000240  min_lr: 0.000240  loss: 3.1385 (3.0070)  weight_decay: 0.0500 (0.0500)  time: 0.4686  data: 0.0048  max mem: 41794
Epoch: [255] Total time: 0:03:57 (0.7601 s / it)
Averaged stats: lr: 0.000240  min_lr: 0.000240  loss: 3.1385 (3.0294)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 0.8453 (0.8453)  acc1: 84.5052 (84.5052)  acc5: 96.7448 (96.7448)  time: 8.9964  data: 8.7924  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1256 (1.0551)  acc1: 77.7344 (79.1520)  acc5: 95.3125 (95.0880)  time: 1.1512  data: 0.9770  max mem: 41794
Test: Total time: 0:00:10 (1.1947 s / it)
* Acc@1 79.446 Acc@5 94.942 loss 1.058
Accuracy of the model on the 50000 test images: 79.4%
Max accuracy: 79.45%
Epoch: [256]  [  0/312]  eta: 1:25:23  lr: 0.000240  min_lr: 0.000240  loss: 3.1214 (3.1214)  weight_decay: 0.0500 (0.0500)  time: 16.4210  data: 15.9430  max mem: 41794
Epoch: [256]  [ 10/312]  eta: 0:11:16  lr: 0.000239  min_lr: 0.000239  loss: 3.0982 (2.9874)  weight_decay: 0.0500 (0.0500)  time: 2.2409  data: 1.4499  max mem: 41794
Epoch: [256]  [ 20/312]  eta: 0:07:21  lr: 0.000239  min_lr: 0.000239  loss: 3.0352 (2.9893)  weight_decay: 0.0500 (0.0500)  time: 0.7681  data: 0.0992  max mem: 41794
Epoch: [256]  [ 30/312]  eta: 0:05:36  lr: 0.000239  min_lr: 0.000239  loss: 3.0352 (3.0183)  weight_decay: 0.0500 (0.0500)  time: 0.6164  data: 0.1035  max mem: 41794
Epoch: [256]  [ 40/312]  eta: 0:04:45  lr: 0.000238  min_lr: 0.000238  loss: 3.0334 (2.9833)  weight_decay: 0.0500 (0.0500)  time: 0.5590  data: 0.0378  max mem: 41794
Epoch: [256]  [ 50/312]  eta: 0:04:24  lr: 0.000238  min_lr: 0.000238  loss: 3.0334 (2.9885)  weight_decay: 0.0500 (0.0500)  time: 0.7238  data: 0.1291  max mem: 41794
Epoch: [256]  [ 60/312]  eta: 0:03:52  lr: 0.000238  min_lr: 0.000238  loss: 3.2478 (3.0107)  weight_decay: 0.0500 (0.0500)  time: 0.6650  data: 0.0962  max mem: 41794
Epoch: [256]  [ 70/312]  eta: 0:03:39  lr: 0.000237  min_lr: 0.000237  loss: 3.2365 (3.0165)  weight_decay: 0.0500 (0.0500)  time: 0.6475  data: 0.1223  max mem: 41794
Epoch: [256]  [ 80/312]  eta: 0:03:27  lr: 0.000237  min_lr: 0.000237  loss: 3.2894 (3.0455)  weight_decay: 0.0500 (0.0500)  time: 0.8068  data: 0.2673  max mem: 41794
Epoch: [256]  [ 90/312]  eta: 0:03:10  lr: 0.000237  min_lr: 0.000237  loss: 3.2564 (3.0424)  weight_decay: 0.0500 (0.0500)  time: 0.6893  data: 0.1456  max mem: 41794
Epoch: [256]  [100/312]  eta: 0:03:02  lr: 0.000236  min_lr: 0.000236  loss: 3.2436 (3.0587)  weight_decay: 0.0500 (0.0500)  time: 0.7371  data: 0.1282  max mem: 41794
Epoch: [256]  [110/312]  eta: 0:02:47  lr: 0.000236  min_lr: 0.000236  loss: 3.3477 (3.0652)  weight_decay: 0.0500 (0.0500)  time: 0.6874  data: 0.1282  max mem: 41794
Epoch: [256]  [120/312]  eta: 0:02:40  lr: 0.000236  min_lr: 0.000236  loss: 3.2537 (3.0584)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.0952  max mem: 41794
Epoch: [256]  [130/312]  eta: 0:02:31  lr: 0.000235  min_lr: 0.000235  loss: 3.1633 (3.0646)  weight_decay: 0.0500 (0.0500)  time: 0.8411  data: 0.1634  max mem: 41794
Epoch: [256]  [140/312]  eta: 0:02:18  lr: 0.000235  min_lr: 0.000235  loss: 3.1248 (3.0626)  weight_decay: 0.0500 (0.0500)  time: 0.6198  data: 0.0689  max mem: 41794
Epoch: [256]  [150/312]  eta: 0:02:11  lr: 0.000235  min_lr: 0.000235  loss: 3.1724 (3.0720)  weight_decay: 0.0500 (0.0500)  time: 0.7075  data: 0.1065  max mem: 41794
Epoch: [256]  [160/312]  eta: 0:02:03  lr: 0.000234  min_lr: 0.000234  loss: 3.0444 (3.0559)  weight_decay: 0.0500 (0.0500)  time: 0.8471  data: 0.2110  max mem: 41794
Epoch: [256]  [170/312]  eta: 0:01:53  lr: 0.000234  min_lr: 0.000234  loss: 2.7466 (3.0467)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1068  max mem: 41794
Epoch: [256]  [180/312]  eta: 0:01:45  lr: 0.000234  min_lr: 0.000234  loss: 3.0967 (3.0513)  weight_decay: 0.0500 (0.0500)  time: 0.7025  data: 0.1056  max mem: 41794
Epoch: [256]  [190/312]  eta: 0:01:35  lr: 0.000233  min_lr: 0.000233  loss: 3.0671 (3.0374)  weight_decay: 0.0500 (0.0500)  time: 0.6694  data: 0.1069  max mem: 41794
Epoch: [256]  [200/312]  eta: 0:01:28  lr: 0.000233  min_lr: 0.000233  loss: 3.0961 (3.0370)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.1092  max mem: 41794
Epoch: [256]  [210/312]  eta: 0:01:21  lr: 0.000233  min_lr: 0.000233  loss: 3.0725 (3.0351)  weight_decay: 0.0500 (0.0500)  time: 0.8907  data: 0.2020  max mem: 41794
Epoch: [256]  [220/312]  eta: 0:01:11  lr: 0.000232  min_lr: 0.000232  loss: 3.2495 (3.0449)  weight_decay: 0.0500 (0.0500)  time: 0.6929  data: 0.0979  max mem: 41794
Epoch: [256]  [230/312]  eta: 0:01:04  lr: 0.000232  min_lr: 0.000232  loss: 3.0942 (3.0255)  weight_decay: 0.0500 (0.0500)  time: 0.6669  data: 0.0651  max mem: 41794
Epoch: [256]  [240/312]  eta: 0:00:56  lr: 0.000232  min_lr: 0.000232  loss: 2.8547 (3.0374)  weight_decay: 0.0500 (0.0500)  time: 0.8147  data: 0.1413  max mem: 41794
Epoch: [256]  [250/312]  eta: 0:00:48  lr: 0.000231  min_lr: 0.000231  loss: 3.1795 (3.0373)  weight_decay: 0.0500 (0.0500)  time: 0.6935  data: 0.0820  max mem: 41794
Epoch: [256]  [260/312]  eta: 0:00:40  lr: 0.000231  min_lr: 0.000231  loss: 3.0367 (3.0398)  weight_decay: 0.0500 (0.0500)  time: 0.7209  data: 0.0905  max mem: 41794
Epoch: [256]  [270/312]  eta: 0:00:32  lr: 0.000231  min_lr: 0.000231  loss: 3.0938 (3.0368)  weight_decay: 0.0500 (0.0500)  time: 0.6691  data: 0.0868  max mem: 41794
Epoch: [256]  [280/312]  eta: 0:00:24  lr: 0.000230  min_lr: 0.000230  loss: 3.0600 (3.0324)  weight_decay: 0.0500 (0.0500)  time: 0.6616  data: 0.1229  max mem: 41794
Epoch: [256]  [290/312]  eta: 0:00:17  lr: 0.000230  min_lr: 0.000230  loss: 3.1194 (3.0398)  weight_decay: 0.0500 (0.0500)  time: 0.8390  data: 0.2296  max mem: 41794
Epoch: [256]  [300/312]  eta: 0:00:09  lr: 0.000230  min_lr: 0.000230  loss: 3.0655 (3.0331)  weight_decay: 0.0500 (0.0500)  time: 0.6614  data: 0.1071  max mem: 41794
Epoch: [256]  [310/312]  eta: 0:00:01  lr: 0.000229  min_lr: 0.000229  loss: 2.6803 (3.0210)  weight_decay: 0.0500 (0.0500)  time: 0.4678  data: 0.0001  max mem: 41794
Epoch: [256]  [311/312]  eta: 0:00:00  lr: 0.000229  min_lr: 0.000229  loss: 2.7177 (3.0221)  weight_decay: 0.0500 (0.0500)  time: 0.4671  data: 0.0001  max mem: 41794
Epoch: [256] Total time: 0:03:56 (0.7594 s / it)
Averaged stats: lr: 0.000229  min_lr: 0.000229  loss: 2.7177 (3.0466)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 0.8257 (0.8257)  acc1: 85.9375 (85.9375)  acc5: 96.6146 (96.6146)  time: 8.7397  data: 8.5266  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1035 (1.0160)  acc1: 80.0781 (79.3920)  acc5: 95.8333 (95.2160)  time: 1.1192  data: 0.9475  max mem: 41794
Test: Total time: 0:00:10 (1.1328 s / it)
* Acc@1 79.486 Acc@5 95.058 loss 1.015
Accuracy of the model on the 50000 test images: 79.5%
Max accuracy: 79.49%
Epoch: [257]  [  0/312]  eta: 1:19:03  lr: 0.000229  min_lr: 0.000229  loss: 3.4961 (3.4961)  weight_decay: 0.0500 (0.0500)  time: 15.2045  data: 14.7239  max mem: 41794
Epoch: [257]  [ 10/312]  eta: 0:10:24  lr: 0.000229  min_lr: 0.000229  loss: 2.8289 (3.0039)  weight_decay: 0.0500 (0.0500)  time: 2.0678  data: 1.3393  max mem: 41794
Epoch: [257]  [ 20/312]  eta: 0:07:04  lr: 0.000229  min_lr: 0.000229  loss: 2.9600 (3.0501)  weight_decay: 0.0500 (0.0500)  time: 0.7670  data: 0.1308  max mem: 41794
Epoch: [257]  [ 30/312]  eta: 0:05:24  lr: 0.000228  min_lr: 0.000228  loss: 3.2469 (3.0548)  weight_decay: 0.0500 (0.0500)  time: 0.6474  data: 0.1308  max mem: 41794
Epoch: [257]  [ 40/312]  eta: 0:04:43  lr: 0.000228  min_lr: 0.000228  loss: 3.2529 (3.0638)  weight_decay: 0.0500 (0.0500)  time: 0.6115  data: 0.1083  max mem: 41794
Epoch: [257]  [ 50/312]  eta: 0:04:23  lr: 0.000228  min_lr: 0.000228  loss: 3.2763 (3.0713)  weight_decay: 0.0500 (0.0500)  time: 0.7798  data: 0.2746  max mem: 41794
Epoch: [257]  [ 60/312]  eta: 0:03:52  lr: 0.000227  min_lr: 0.000227  loss: 3.2552 (3.0655)  weight_decay: 0.0500 (0.0500)  time: 0.6703  data: 0.1669  max mem: 41794
Epoch: [257]  [ 70/312]  eta: 0:03:43  lr: 0.000227  min_lr: 0.000227  loss: 3.2039 (3.0806)  weight_decay: 0.0500 (0.0500)  time: 0.7139  data: 0.1675  max mem: 41794
Epoch: [257]  [ 80/312]  eta: 0:03:31  lr: 0.000227  min_lr: 0.000227  loss: 3.2039 (3.0904)  weight_decay: 0.0500 (0.0500)  time: 0.8896  data: 0.3348  max mem: 41794
Epoch: [257]  [ 90/312]  eta: 0:03:12  lr: 0.000226  min_lr: 0.000226  loss: 3.0291 (3.0781)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1679  max mem: 41794
Epoch: [257]  [100/312]  eta: 0:03:05  lr: 0.000226  min_lr: 0.000226  loss: 3.0420 (3.0722)  weight_decay: 0.0500 (0.0500)  time: 0.7096  data: 0.2069  max mem: 41794
Epoch: [257]  [110/312]  eta: 0:02:49  lr: 0.000226  min_lr: 0.000226  loss: 3.1797 (3.0638)  weight_decay: 0.0500 (0.0500)  time: 0.7068  data: 0.2069  max mem: 41794
Epoch: [257]  [120/312]  eta: 0:02:42  lr: 0.000225  min_lr: 0.000225  loss: 3.2086 (3.0814)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.2086  max mem: 41794
Epoch: [257]  [130/312]  eta: 0:02:33  lr: 0.000225  min_lr: 0.000225  loss: 3.2647 (3.0716)  weight_decay: 0.0500 (0.0500)  time: 0.8679  data: 0.3763  max mem: 41794
Epoch: [257]  [140/312]  eta: 0:02:20  lr: 0.000225  min_lr: 0.000225  loss: 3.1310 (3.0776)  weight_decay: 0.0500 (0.0500)  time: 0.6565  data: 0.1683  max mem: 41794
Epoch: [257]  [150/312]  eta: 0:02:12  lr: 0.000224  min_lr: 0.000224  loss: 3.2731 (3.0935)  weight_decay: 0.0500 (0.0500)  time: 0.6680  data: 0.1798  max mem: 41794
Epoch: [257]  [160/312]  eta: 0:02:04  lr: 0.000224  min_lr: 0.000224  loss: 3.3141 (3.0890)  weight_decay: 0.0500 (0.0500)  time: 0.8245  data: 0.3291  max mem: 41794
Epoch: [257]  [170/312]  eta: 0:01:54  lr: 0.000224  min_lr: 0.000224  loss: 3.0888 (3.0736)  weight_decay: 0.0500 (0.0500)  time: 0.6920  data: 0.1992  max mem: 41794
Epoch: [257]  [180/312]  eta: 0:01:46  lr: 0.000223  min_lr: 0.000223  loss: 2.9291 (3.0587)  weight_decay: 0.0500 (0.0500)  time: 0.7285  data: 0.2348  max mem: 41794
Epoch: [257]  [190/312]  eta: 0:01:36  lr: 0.000223  min_lr: 0.000223  loss: 2.8016 (3.0542)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1854  max mem: 41794
Epoch: [257]  [200/312]  eta: 0:01:29  lr: 0.000223  min_lr: 0.000223  loss: 2.7712 (3.0511)  weight_decay: 0.0500 (0.0500)  time: 0.6885  data: 0.1774  max mem: 41794
Epoch: [257]  [210/312]  eta: 0:01:21  lr: 0.000222  min_lr: 0.000222  loss: 3.1864 (3.0613)  weight_decay: 0.0500 (0.0500)  time: 0.8798  data: 0.3331  max mem: 41794
Epoch: [257]  [220/312]  eta: 0:01:12  lr: 0.000222  min_lr: 0.000222  loss: 3.1685 (3.0546)  weight_decay: 0.0500 (0.0500)  time: 0.6923  data: 0.1563  max mem: 41794
Epoch: [257]  [230/312]  eta: 0:01:04  lr: 0.000222  min_lr: 0.000222  loss: 3.0599 (3.0568)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.1610  max mem: 41794
Epoch: [257]  [240/312]  eta: 0:00:56  lr: 0.000221  min_lr: 0.000221  loss: 3.2033 (3.0540)  weight_decay: 0.0500 (0.0500)  time: 0.8316  data: 0.3084  max mem: 41794
Epoch: [257]  [250/312]  eta: 0:00:48  lr: 0.000221  min_lr: 0.000221  loss: 3.0222 (3.0498)  weight_decay: 0.0500 (0.0500)  time: 0.6716  data: 0.1481  max mem: 41794
Epoch: [257]  [260/312]  eta: 0:00:40  lr: 0.000221  min_lr: 0.000221  loss: 3.1114 (3.0469)  weight_decay: 0.0500 (0.0500)  time: 0.6906  data: 0.1457  max mem: 41794
Epoch: [257]  [270/312]  eta: 0:00:32  lr: 0.000220  min_lr: 0.000220  loss: 3.1114 (3.0432)  weight_decay: 0.0500 (0.0500)  time: 0.6585  data: 0.1456  max mem: 41794
Epoch: [257]  [280/312]  eta: 0:00:24  lr: 0.000220  min_lr: 0.000220  loss: 3.0459 (3.0396)  weight_decay: 0.0500 (0.0500)  time: 0.7074  data: 0.1663  max mem: 41794
Epoch: [257]  [290/312]  eta: 0:00:17  lr: 0.000220  min_lr: 0.000220  loss: 3.1348 (3.0412)  weight_decay: 0.0500 (0.0500)  time: 0.8479  data: 0.3029  max mem: 41794
Epoch: [257]  [300/312]  eta: 0:00:09  lr: 0.000219  min_lr: 0.000219  loss: 3.0094 (3.0335)  weight_decay: 0.0500 (0.0500)  time: 0.6213  data: 0.1370  max mem: 41794
Epoch: [257]  [310/312]  eta: 0:00:01  lr: 0.000219  min_lr: 0.000219  loss: 2.9898 (3.0331)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [257]  [311/312]  eta: 0:00:00  lr: 0.000219  min_lr: 0.000219  loss: 3.1164 (3.0338)  weight_decay: 0.0500 (0.0500)  time: 0.4645  data: 0.0001  max mem: 41794
Epoch: [257] Total time: 0:03:57 (0.7625 s / it)
Averaged stats: lr: 0.000219  min_lr: 0.000219  loss: 3.1164 (3.0320)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 0.8748 (0.8748)  acc1: 84.8958 (84.8958)  acc5: 95.9635 (95.9635)  time: 8.4087  data: 8.1991  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1319 (1.0709)  acc1: 79.0365 (79.1840)  acc5: 94.5312 (94.9280)  time: 1.0963  data: 0.9245  max mem: 41794
Test: Total time: 0:00:09 (1.1106 s / it)
* Acc@1 79.476 Acc@5 94.852 loss 1.069
Accuracy of the model on the 50000 test images: 79.5%
Max accuracy: 79.49%
Epoch: [258]  [  0/312]  eta: 1:23:18  lr: 0.000219  min_lr: 0.000219  loss: 3.4721 (3.4721)  weight_decay: 0.0500 (0.0500)  time: 16.0201  data: 12.6472  max mem: 41794
Epoch: [258]  [ 10/312]  eta: 0:10:42  lr: 0.000219  min_lr: 0.000219  loss: 3.1944 (2.9711)  weight_decay: 0.0500 (0.0500)  time: 2.1280  data: 1.3685  max mem: 41794
Epoch: [258]  [ 20/312]  eta: 0:07:40  lr: 0.000218  min_lr: 0.000218  loss: 3.2360 (3.0860)  weight_decay: 0.0500 (0.0500)  time: 0.8562  data: 0.2296  max mem: 41794
Epoch: [258]  [ 30/312]  eta: 0:05:52  lr: 0.000218  min_lr: 0.000218  loss: 3.2864 (3.0989)  weight_decay: 0.0500 (0.0500)  time: 0.7693  data: 0.1143  max mem: 41794
Epoch: [258]  [ 40/312]  eta: 0:05:03  lr: 0.000218  min_lr: 0.000218  loss: 3.0884 (3.0847)  weight_decay: 0.0500 (0.0500)  time: 0.6306  data: 0.0472  max mem: 41794
Epoch: [258]  [ 50/312]  eta: 0:04:36  lr: 0.000217  min_lr: 0.000217  loss: 2.9469 (3.0296)  weight_decay: 0.0500 (0.0500)  time: 0.7491  data: 0.1245  max mem: 41794
Epoch: [258]  [ 60/312]  eta: 0:04:02  lr: 0.000217  min_lr: 0.000217  loss: 2.8247 (3.0247)  weight_decay: 0.0500 (0.0500)  time: 0.6439  data: 0.0826  max mem: 41794
Epoch: [258]  [ 70/312]  eta: 0:03:48  lr: 0.000217  min_lr: 0.000217  loss: 3.2854 (3.0495)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0948  max mem: 41794
Epoch: [258]  [ 80/312]  eta: 0:03:32  lr: 0.000216  min_lr: 0.000216  loss: 3.2528 (3.0394)  weight_decay: 0.0500 (0.0500)  time: 0.7719  data: 0.1251  max mem: 41794
Epoch: [258]  [ 90/312]  eta: 0:03:17  lr: 0.000216  min_lr: 0.000216  loss: 3.2146 (3.0559)  weight_decay: 0.0500 (0.0500)  time: 0.6930  data: 0.1288  max mem: 41794
Epoch: [258]  [100/312]  eta: 0:03:09  lr: 0.000216  min_lr: 0.000216  loss: 3.2398 (3.0515)  weight_decay: 0.0500 (0.0500)  time: 0.8080  data: 0.2241  max mem: 41794
Epoch: [258]  [110/312]  eta: 0:02:53  lr: 0.000215  min_lr: 0.000215  loss: 3.1591 (3.0547)  weight_decay: 0.0500 (0.0500)  time: 0.7115  data: 0.1285  max mem: 41794
Epoch: [258]  [120/312]  eta: 0:02:44  lr: 0.000215  min_lr: 0.000215  loss: 3.1199 (3.0648)  weight_decay: 0.0500 (0.0500)  time: 0.6805  data: 0.1083  max mem: 41794
Epoch: [258]  [130/312]  eta: 0:02:35  lr: 0.000215  min_lr: 0.000215  loss: 3.1199 (3.0560)  weight_decay: 0.0500 (0.0500)  time: 0.8446  data: 0.1918  max mem: 41794
Epoch: [258]  [140/312]  eta: 0:02:22  lr: 0.000214  min_lr: 0.000214  loss: 3.0474 (3.0456)  weight_decay: 0.0500 (0.0500)  time: 0.6567  data: 0.0864  max mem: 41794
Epoch: [258]  [150/312]  eta: 0:02:15  lr: 0.000214  min_lr: 0.000214  loss: 3.0876 (3.0498)  weight_decay: 0.0500 (0.0500)  time: 0.7108  data: 0.1207  max mem: 41794
Epoch: [258]  [160/312]  eta: 0:02:05  lr: 0.000214  min_lr: 0.000214  loss: 3.3821 (3.0686)  weight_decay: 0.0500 (0.0500)  time: 0.7959  data: 0.1221  max mem: 41794
Epoch: [258]  [170/312]  eta: 0:01:55  lr: 0.000213  min_lr: 0.000213  loss: 3.3324 (3.0551)  weight_decay: 0.0500 (0.0500)  time: 0.6634  data: 0.0918  max mem: 41794
Epoch: [258]  [180/312]  eta: 0:01:47  lr: 0.000213  min_lr: 0.000213  loss: 2.8262 (3.0520)  weight_decay: 0.0500 (0.0500)  time: 0.7439  data: 0.1835  max mem: 41794
Epoch: [258]  [190/312]  eta: 0:01:37  lr: 0.000213  min_lr: 0.000213  loss: 3.0304 (3.0483)  weight_decay: 0.0500 (0.0500)  time: 0.6565  data: 0.0938  max mem: 41794
Epoch: [258]  [200/312]  eta: 0:01:30  lr: 0.000212  min_lr: 0.000212  loss: 3.1125 (3.0487)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.0975  max mem: 41794
Epoch: [258]  [210/312]  eta: 0:01:21  lr: 0.000212  min_lr: 0.000212  loss: 2.8870 (3.0368)  weight_decay: 0.0500 (0.0500)  time: 0.8437  data: 0.1519  max mem: 41794
Epoch: [258]  [220/312]  eta: 0:01:12  lr: 0.000212  min_lr: 0.000212  loss: 2.8770 (3.0321)  weight_decay: 0.0500 (0.0500)  time: 0.6627  data: 0.0875  max mem: 41794
Epoch: [258]  [230/312]  eta: 0:01:05  lr: 0.000211  min_lr: 0.000211  loss: 3.1417 (3.0311)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.1321  max mem: 41794
Epoch: [258]  [240/312]  eta: 0:00:56  lr: 0.000211  min_lr: 0.000211  loss: 3.1417 (3.0297)  weight_decay: 0.0500 (0.0500)  time: 0.7705  data: 0.1607  max mem: 41794
Epoch: [258]  [250/312]  eta: 0:00:48  lr: 0.000211  min_lr: 0.000211  loss: 3.1913 (3.0385)  weight_decay: 0.0500 (0.0500)  time: 0.6663  data: 0.1233  max mem: 41794
Epoch: [258]  [260/312]  eta: 0:00:40  lr: 0.000211  min_lr: 0.000211  loss: 3.2029 (3.0354)  weight_decay: 0.0500 (0.0500)  time: 0.7214  data: 0.1958  max mem: 41794
Epoch: [258]  [270/312]  eta: 0:00:32  lr: 0.000210  min_lr: 0.000210  loss: 2.9036 (3.0255)  weight_decay: 0.0500 (0.0500)  time: 0.6764  data: 0.1461  max mem: 41794
Epoch: [258]  [280/312]  eta: 0:00:24  lr: 0.000210  min_lr: 0.000210  loss: 2.8534 (3.0236)  weight_decay: 0.0500 (0.0500)  time: 0.6909  data: 0.1670  max mem: 41794
Epoch: [258]  [290/312]  eta: 0:00:17  lr: 0.000210  min_lr: 0.000210  loss: 3.1260 (3.0223)  weight_decay: 0.0500 (0.0500)  time: 0.8269  data: 0.2685  max mem: 41794
Epoch: [258]  [300/312]  eta: 0:00:09  lr: 0.000209  min_lr: 0.000209  loss: 3.1719 (3.0234)  weight_decay: 0.0500 (0.0500)  time: 0.6355  data: 0.1139  max mem: 41794
Epoch: [258]  [310/312]  eta: 0:00:01  lr: 0.000209  min_lr: 0.000209  loss: 3.2128 (3.0250)  weight_decay: 0.0500 (0.0500)  time: 0.4702  data: 0.0002  max mem: 41794
Epoch: [258]  [311/312]  eta: 0:00:00  lr: 0.000209  min_lr: 0.000209  loss: 3.2128 (3.0254)  weight_decay: 0.0500 (0.0500)  time: 0.4696  data: 0.0002  max mem: 41794
Epoch: [258] Total time: 0:03:58 (0.7657 s / it)
Averaged stats: lr: 0.000209  min_lr: 0.000209  loss: 3.2128 (3.0417)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.8158 (0.8158)  acc1: 84.8958 (84.8958)  acc5: 96.6146 (96.6146)  time: 8.5152  data: 8.3024  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1146 (1.0413)  acc1: 78.2552 (79.0240)  acc5: 95.5729 (95.0400)  time: 1.0942  data: 0.9226  max mem: 41794
Test: Total time: 0:00:10 (1.1333 s / it)
* Acc@1 79.572 Acc@5 94.942 loss 1.034
Accuracy of the model on the 50000 test images: 79.6%
Max accuracy: 79.57%
Epoch: [259]  [  0/312]  eta: 1:18:58  lr: 0.000209  min_lr: 0.000209  loss: 3.2744 (3.2744)  weight_decay: 0.0500 (0.0500)  time: 15.1866  data: 13.7377  max mem: 41794
Epoch: [259]  [ 10/312]  eta: 0:10:41  lr: 0.000209  min_lr: 0.000209  loss: 3.0729 (3.0740)  weight_decay: 0.0500 (0.0500)  time: 2.1254  data: 1.3666  max mem: 41794
Epoch: [259]  [ 20/312]  eta: 0:07:17  lr: 0.000208  min_lr: 0.000208  loss: 3.0729 (3.1171)  weight_decay: 0.0500 (0.0500)  time: 0.8142  data: 0.1406  max mem: 41794
Epoch: [259]  [ 30/312]  eta: 0:05:32  lr: 0.000208  min_lr: 0.000208  loss: 3.0113 (3.1095)  weight_decay: 0.0500 (0.0500)  time: 0.6610  data: 0.0763  max mem: 41794
Epoch: [259]  [ 40/312]  eta: 0:04:46  lr: 0.000208  min_lr: 0.000208  loss: 3.0113 (3.0983)  weight_decay: 0.0500 (0.0500)  time: 0.5884  data: 0.0409  max mem: 41794
Epoch: [259]  [ 50/312]  eta: 0:04:28  lr: 0.000207  min_lr: 0.000207  loss: 3.1275 (3.0889)  weight_decay: 0.0500 (0.0500)  time: 0.7812  data: 0.1743  max mem: 41794
Epoch: [259]  [ 60/312]  eta: 0:03:56  lr: 0.000207  min_lr: 0.000207  loss: 3.2425 (3.0993)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.1345  max mem: 41794
Epoch: [259]  [ 70/312]  eta: 0:03:47  lr: 0.000207  min_lr: 0.000207  loss: 3.1393 (3.0584)  weight_decay: 0.0500 (0.0500)  time: 0.7200  data: 0.1204  max mem: 41794
Epoch: [259]  [ 80/312]  eta: 0:03:31  lr: 0.000206  min_lr: 0.000206  loss: 3.1393 (3.0884)  weight_decay: 0.0500 (0.0500)  time: 0.8267  data: 0.1212  max mem: 41794
Epoch: [259]  [ 90/312]  eta: 0:03:16  lr: 0.000206  min_lr: 0.000206  loss: 3.1781 (3.0867)  weight_decay: 0.0500 (0.0500)  time: 0.7058  data: 0.1014  max mem: 41794
Epoch: [259]  [100/312]  eta: 0:03:08  lr: 0.000206  min_lr: 0.000206  loss: 3.1693 (3.0929)  weight_decay: 0.0500 (0.0500)  time: 0.8033  data: 0.1814  max mem: 41794
Epoch: [259]  [110/312]  eta: 0:02:52  lr: 0.000205  min_lr: 0.000205  loss: 3.3161 (3.1031)  weight_decay: 0.0500 (0.0500)  time: 0.7136  data: 0.0824  max mem: 41794
Epoch: [259]  [120/312]  eta: 0:02:44  lr: 0.000205  min_lr: 0.000205  loss: 3.2619 (3.1061)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.0807  max mem: 41794
Epoch: [259]  [130/312]  eta: 0:02:35  lr: 0.000205  min_lr: 0.000205  loss: 3.2124 (3.1030)  weight_decay: 0.0500 (0.0500)  time: 0.8524  data: 0.1308  max mem: 41794
Epoch: [259]  [140/312]  eta: 0:02:23  lr: 0.000204  min_lr: 0.000204  loss: 3.2079 (3.1054)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.0720  max mem: 41794
Epoch: [259]  [150/312]  eta: 0:02:14  lr: 0.000204  min_lr: 0.000204  loss: 3.3600 (3.1176)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.1147  max mem: 41794
Epoch: [259]  [160/312]  eta: 0:02:04  lr: 0.000204  min_lr: 0.000204  loss: 3.3605 (3.1187)  weight_decay: 0.0500 (0.0500)  time: 0.7425  data: 0.1118  max mem: 41794
Epoch: [259]  [170/312]  eta: 0:01:55  lr: 0.000203  min_lr: 0.000203  loss: 3.2352 (3.1202)  weight_decay: 0.0500 (0.0500)  time: 0.6682  data: 0.1094  max mem: 41794
Epoch: [259]  [180/312]  eta: 0:01:47  lr: 0.000203  min_lr: 0.000203  loss: 3.1920 (3.1150)  weight_decay: 0.0500 (0.0500)  time: 0.7417  data: 0.1859  max mem: 41794
Epoch: [259]  [190/312]  eta: 0:01:37  lr: 0.000203  min_lr: 0.000203  loss: 3.2117 (3.1177)  weight_decay: 0.0500 (0.0500)  time: 0.6501  data: 0.0950  max mem: 41794
Epoch: [259]  [200/312]  eta: 0:01:29  lr: 0.000203  min_lr: 0.000203  loss: 3.2901 (3.1192)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.1222  max mem: 41794
Epoch: [259]  [210/312]  eta: 0:01:22  lr: 0.000202  min_lr: 0.000202  loss: 3.2440 (3.1236)  weight_decay: 0.0500 (0.0500)  time: 0.8905  data: 0.1929  max mem: 41794
Epoch: [259]  [220/312]  eta: 0:01:13  lr: 0.000202  min_lr: 0.000202  loss: 3.0009 (3.1053)  weight_decay: 0.0500 (0.0500)  time: 0.7706  data: 0.1246  max mem: 41794
Epoch: [259]  [230/312]  eta: 0:01:05  lr: 0.000202  min_lr: 0.000202  loss: 3.0009 (3.1068)  weight_decay: 0.0500 (0.0500)  time: 0.7027  data: 0.1092  max mem: 41794
Epoch: [259]  [240/312]  eta: 0:00:57  lr: 0.000201  min_lr: 0.000201  loss: 3.2140 (3.1021)  weight_decay: 0.0500 (0.0500)  time: 0.7840  data: 0.0718  max mem: 41794
Epoch: [259]  [250/312]  eta: 0:00:48  lr: 0.000201  min_lr: 0.000201  loss: 2.8778 (3.0901)  weight_decay: 0.0500 (0.0500)  time: 0.7176  data: 0.0943  max mem: 41794
Epoch: [259]  [260/312]  eta: 0:00:41  lr: 0.000201  min_lr: 0.000201  loss: 3.0362 (3.0911)  weight_decay: 0.0500 (0.0500)  time: 0.7376  data: 0.1287  max mem: 41794
Epoch: [259]  [270/312]  eta: 0:00:32  lr: 0.000200  min_lr: 0.000200  loss: 3.2671 (3.0973)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.0630  max mem: 41794
Epoch: [259]  [280/312]  eta: 0:00:25  lr: 0.000200  min_lr: 0.000200  loss: 3.1016 (3.0864)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.0948  max mem: 41794
Epoch: [259]  [290/312]  eta: 0:00:17  lr: 0.000200  min_lr: 0.000200  loss: 2.9673 (3.0824)  weight_decay: 0.0500 (0.0500)  time: 0.7840  data: 0.1174  max mem: 41794
Epoch: [259]  [300/312]  eta: 0:00:09  lr: 0.000199  min_lr: 0.000199  loss: 3.0895 (3.0785)  weight_decay: 0.0500 (0.0500)  time: 0.5878  data: 0.0349  max mem: 41794
Epoch: [259]  [310/312]  eta: 0:00:01  lr: 0.000199  min_lr: 0.000199  loss: 3.0687 (3.0739)  weight_decay: 0.0500 (0.0500)  time: 0.4626  data: 0.0001  max mem: 41794
Epoch: [259]  [311/312]  eta: 0:00:00  lr: 0.000199  min_lr: 0.000199  loss: 3.0565 (3.0706)  weight_decay: 0.0500 (0.0500)  time: 0.4618  data: 0.0001  max mem: 41794
Epoch: [259] Total time: 0:03:59 (0.7671 s / it)
Averaged stats: lr: 0.000199  min_lr: 0.000199  loss: 3.0565 (3.0433)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.8358 (0.8358)  acc1: 85.9375 (85.9375)  acc5: 96.3542 (96.3542)  time: 8.8567  data: 8.6442  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0969 (1.0249)  acc1: 79.0365 (79.3920)  acc5: 95.0521 (95.1200)  time: 1.1325  data: 0.9605  max mem: 41794
Test: Total time: 0:00:10 (1.1554 s / it)
* Acc@1 79.542 Acc@5 94.976 loss 1.020
Accuracy of the model on the 50000 test images: 79.5%
Max accuracy: 79.57%
Epoch: [260]  [  0/312]  eta: 1:24:31  lr: 0.000199  min_lr: 0.000199  loss: 3.2290 (3.2290)  weight_decay: 0.0500 (0.0500)  time: 16.2553  data: 13.2499  max mem: 41794
Epoch: [260]  [ 10/312]  eta: 0:10:48  lr: 0.000199  min_lr: 0.000199  loss: 3.2602 (3.1106)  weight_decay: 0.0500 (0.0500)  time: 2.1482  data: 1.4228  max mem: 41794
Epoch: [260]  [ 20/312]  eta: 0:07:15  lr: 0.000198  min_lr: 0.000198  loss: 3.2453 (3.0973)  weight_decay: 0.0500 (0.0500)  time: 0.7536  data: 0.2330  max mem: 41794
Epoch: [260]  [ 30/312]  eta: 0:05:32  lr: 0.000198  min_lr: 0.000198  loss: 3.2202 (3.0907)  weight_decay: 0.0500 (0.0500)  time: 0.6458  data: 0.1167  max mem: 41794
Epoch: [260]  [ 40/312]  eta: 0:04:52  lr: 0.000198  min_lr: 0.000198  loss: 3.2202 (3.0977)  weight_decay: 0.0500 (0.0500)  time: 0.6409  data: 0.1226  max mem: 41794
Epoch: [260]  [ 50/312]  eta: 0:04:29  lr: 0.000197  min_lr: 0.000197  loss: 3.1712 (3.0663)  weight_decay: 0.0500 (0.0500)  time: 0.7966  data: 0.2871  max mem: 41794
Epoch: [260]  [ 60/312]  eta: 0:03:57  lr: 0.000197  min_lr: 0.000197  loss: 3.1503 (3.0781)  weight_decay: 0.0500 (0.0500)  time: 0.6630  data: 0.1686  max mem: 41794
Epoch: [260]  [ 70/312]  eta: 0:03:45  lr: 0.000197  min_lr: 0.000197  loss: 3.1503 (3.0899)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1851  max mem: 41794
Epoch: [260]  [ 80/312]  eta: 0:03:36  lr: 0.000197  min_lr: 0.000197  loss: 2.9920 (3.0443)  weight_decay: 0.0500 (0.0500)  time: 0.9103  data: 0.4075  max mem: 41794
Epoch: [260]  [ 90/312]  eta: 0:03:16  lr: 0.000196  min_lr: 0.000196  loss: 2.9124 (3.0395)  weight_decay: 0.0500 (0.0500)  time: 0.7149  data: 0.2229  max mem: 41794
Epoch: [260]  [100/312]  eta: 0:03:06  lr: 0.000196  min_lr: 0.000196  loss: 3.0410 (3.0189)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.1754  max mem: 41794
Epoch: [260]  [110/312]  eta: 0:02:50  lr: 0.000196  min_lr: 0.000196  loss: 2.9221 (3.0075)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1773  max mem: 41794
Epoch: [260]  [120/312]  eta: 0:02:42  lr: 0.000195  min_lr: 0.000195  loss: 3.0202 (3.0209)  weight_decay: 0.0500 (0.0500)  time: 0.6744  data: 0.1947  max mem: 41794
Epoch: [260]  [130/312]  eta: 0:02:34  lr: 0.000195  min_lr: 0.000195  loss: 3.0838 (3.0050)  weight_decay: 0.0500 (0.0500)  time: 0.8778  data: 0.3863  max mem: 41794
Epoch: [260]  [140/312]  eta: 0:02:21  lr: 0.000195  min_lr: 0.000195  loss: 3.0838 (3.0156)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.1940  max mem: 41794
Epoch: [260]  [150/312]  eta: 0:02:14  lr: 0.000194  min_lr: 0.000194  loss: 3.2391 (3.0136)  weight_decay: 0.0500 (0.0500)  time: 0.6951  data: 0.2090  max mem: 41794
Epoch: [260]  [160/312]  eta: 0:02:06  lr: 0.000194  min_lr: 0.000194  loss: 2.9812 (3.0098)  weight_decay: 0.0500 (0.0500)  time: 0.8804  data: 0.3804  max mem: 41794
Epoch: [260]  [170/312]  eta: 0:01:55  lr: 0.000194  min_lr: 0.000194  loss: 2.9812 (3.0095)  weight_decay: 0.0500 (0.0500)  time: 0.6673  data: 0.1721  max mem: 41794
Epoch: [260]  [180/312]  eta: 0:01:47  lr: 0.000193  min_lr: 0.000193  loss: 3.2036 (3.0083)  weight_decay: 0.0500 (0.0500)  time: 0.6943  data: 0.2108  max mem: 41794
Epoch: [260]  [190/312]  eta: 0:01:37  lr: 0.000193  min_lr: 0.000193  loss: 3.2052 (3.0131)  weight_decay: 0.0500 (0.0500)  time: 0.6980  data: 0.2133  max mem: 41794
Epoch: [260]  [200/312]  eta: 0:01:29  lr: 0.000193  min_lr: 0.000193  loss: 2.9345 (3.0071)  weight_decay: 0.0500 (0.0500)  time: 0.6763  data: 0.1893  max mem: 41794
Epoch: [260]  [210/312]  eta: 0:01:22  lr: 0.000193  min_lr: 0.000193  loss: 2.8266 (3.0021)  weight_decay: 0.0500 (0.0500)  time: 0.8835  data: 0.3878  max mem: 41794
Epoch: [260]  [220/312]  eta: 0:01:12  lr: 0.000192  min_lr: 0.000192  loss: 3.1697 (3.0098)  weight_decay: 0.0500 (0.0500)  time: 0.6971  data: 0.2017  max mem: 41794
Epoch: [260]  [230/312]  eta: 0:01:05  lr: 0.000192  min_lr: 0.000192  loss: 3.1641 (3.0059)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.1936  max mem: 41794
Epoch: [260]  [240/312]  eta: 0:00:57  lr: 0.000192  min_lr: 0.000192  loss: 3.1765 (3.0109)  weight_decay: 0.0500 (0.0500)  time: 0.8490  data: 0.3543  max mem: 41794
Epoch: [260]  [250/312]  eta: 0:00:48  lr: 0.000191  min_lr: 0.000191  loss: 3.2420 (3.0118)  weight_decay: 0.0500 (0.0500)  time: 0.7107  data: 0.2177  max mem: 41794
Epoch: [260]  [260/312]  eta: 0:00:41  lr: 0.000191  min_lr: 0.000191  loss: 3.1141 (3.0123)  weight_decay: 0.0500 (0.0500)  time: 0.7414  data: 0.2551  max mem: 41794
Epoch: [260]  [270/312]  eta: 0:00:32  lr: 0.000191  min_lr: 0.000191  loss: 3.0300 (3.0061)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.2001  max mem: 41794
Epoch: [260]  [280/312]  eta: 0:00:25  lr: 0.000190  min_lr: 0.000190  loss: 3.1168 (3.0066)  weight_decay: 0.0500 (0.0500)  time: 0.7092  data: 0.2148  max mem: 41794
Epoch: [260]  [290/312]  eta: 0:00:17  lr: 0.000190  min_lr: 0.000190  loss: 3.0449 (3.0001)  weight_decay: 0.0500 (0.0500)  time: 0.7688  data: 0.2663  max mem: 41794
Epoch: [260]  [300/312]  eta: 0:00:09  lr: 0.000190  min_lr: 0.000190  loss: 2.9471 (2.9994)  weight_decay: 0.0500 (0.0500)  time: 0.5460  data: 0.0533  max mem: 41794
Epoch: [260]  [310/312]  eta: 0:00:01  lr: 0.000189  min_lr: 0.000189  loss: 3.0518 (3.0013)  weight_decay: 0.0500 (0.0500)  time: 0.4680  data: 0.0001  max mem: 41794
Epoch: [260]  [311/312]  eta: 0:00:00  lr: 0.000189  min_lr: 0.000189  loss: 3.0518 (3.0023)  weight_decay: 0.0500 (0.0500)  time: 0.4676  data: 0.0001  max mem: 41794
Epoch: [260] Total time: 0:03:59 (0.7666 s / it)
Averaged stats: lr: 0.000189  min_lr: 0.000189  loss: 3.0518 (3.0220)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.8803 (0.8803)  acc1: 85.9375 (85.9375)  acc5: 96.2240 (96.2240)  time: 8.5416  data: 8.3408  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1399 (1.0692)  acc1: 78.2552 (79.1680)  acc5: 95.0521 (95.0560)  time: 1.1066  data: 0.9268  max mem: 41794
Test: Total time: 0:00:10 (1.1551 s / it)
* Acc@1 79.650 Acc@5 95.044 loss 1.065
Accuracy of the model on the 50000 test images: 79.7%
Max accuracy: 79.65%
Epoch: [261]  [  0/312]  eta: 1:19:52  lr: 0.000189  min_lr: 0.000189  loss: 3.3231 (3.3231)  weight_decay: 0.0500 (0.0500)  time: 15.3602  data: 13.7140  max mem: 41794
Epoch: [261]  [ 10/312]  eta: 0:11:06  lr: 0.000189  min_lr: 0.000189  loss: 3.0816 (2.9749)  weight_decay: 0.0500 (0.0500)  time: 2.2084  data: 1.4205  max mem: 41794
Epoch: [261]  [ 20/312]  eta: 0:07:37  lr: 0.000189  min_lr: 0.000189  loss: 3.1801 (3.1004)  weight_decay: 0.0500 (0.0500)  time: 0.8760  data: 0.1788  max mem: 41794
Epoch: [261]  [ 30/312]  eta: 0:05:51  lr: 0.000188  min_lr: 0.000188  loss: 3.2676 (3.1106)  weight_decay: 0.0500 (0.0500)  time: 0.7182  data: 0.1013  max mem: 41794
Epoch: [261]  [ 40/312]  eta: 0:04:49  lr: 0.000188  min_lr: 0.000188  loss: 3.2514 (3.0876)  weight_decay: 0.0500 (0.0500)  time: 0.5384  data: 0.0209  max mem: 41794
Epoch: [261]  [ 50/312]  eta: 0:04:18  lr: 0.000188  min_lr: 0.000188  loss: 3.2131 (3.0658)  weight_decay: 0.0500 (0.0500)  time: 0.5818  data: 0.0662  max mem: 41794
Epoch: [261]  [ 60/312]  eta: 0:03:49  lr: 0.000188  min_lr: 0.000188  loss: 3.2131 (3.0450)  weight_decay: 0.0500 (0.0500)  time: 0.5899  data: 0.0771  max mem: 41794
Epoch: [261]  [ 70/312]  eta: 0:03:38  lr: 0.000187  min_lr: 0.000187  loss: 2.8029 (2.9970)  weight_decay: 0.0500 (0.0500)  time: 0.6879  data: 0.1518  max mem: 41794
Epoch: [261]  [ 80/312]  eta: 0:03:24  lr: 0.000187  min_lr: 0.000187  loss: 2.8904 (2.9891)  weight_decay: 0.0500 (0.0500)  time: 0.8046  data: 0.2333  max mem: 41794
Epoch: [261]  [ 90/312]  eta: 0:03:09  lr: 0.000187  min_lr: 0.000187  loss: 3.0869 (3.0000)  weight_decay: 0.0500 (0.0500)  time: 0.6833  data: 0.1584  max mem: 41794
Epoch: [261]  [100/312]  eta: 0:03:00  lr: 0.000186  min_lr: 0.000186  loss: 3.2148 (2.9945)  weight_decay: 0.0500 (0.0500)  time: 0.7297  data: 0.2180  max mem: 41794
Epoch: [261]  [110/312]  eta: 0:02:45  lr: 0.000186  min_lr: 0.000186  loss: 2.9063 (2.9784)  weight_decay: 0.0500 (0.0500)  time: 0.6711  data: 0.1550  max mem: 41794
Epoch: [261]  [120/312]  eta: 0:02:37  lr: 0.000186  min_lr: 0.000186  loss: 2.8887 (2.9753)  weight_decay: 0.0500 (0.0500)  time: 0.6688  data: 0.1759  max mem: 41794
Epoch: [261]  [130/312]  eta: 0:02:30  lr: 0.000185  min_lr: 0.000185  loss: 3.2203 (2.9903)  weight_decay: 0.0500 (0.0500)  time: 0.8507  data: 0.3608  max mem: 41794
Epoch: [261]  [140/312]  eta: 0:02:18  lr: 0.000185  min_lr: 0.000185  loss: 3.2203 (2.9892)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1869  max mem: 41794
Epoch: [261]  [150/312]  eta: 0:02:10  lr: 0.000185  min_lr: 0.000185  loss: 3.0072 (2.9896)  weight_decay: 0.0500 (0.0500)  time: 0.6629  data: 0.1602  max mem: 41794
Epoch: [261]  [160/312]  eta: 0:02:02  lr: 0.000185  min_lr: 0.000185  loss: 2.9620 (2.9921)  weight_decay: 0.0500 (0.0500)  time: 0.8042  data: 0.3104  max mem: 41794
Epoch: [261]  [170/312]  eta: 0:01:52  lr: 0.000184  min_lr: 0.000184  loss: 3.1396 (2.9936)  weight_decay: 0.0500 (0.0500)  time: 0.7129  data: 0.1729  max mem: 41794
Epoch: [261]  [180/312]  eta: 0:01:45  lr: 0.000184  min_lr: 0.000184  loss: 3.1396 (2.9923)  weight_decay: 0.0500 (0.0500)  time: 0.7449  data: 0.1473  max mem: 41794
Epoch: [261]  [190/312]  eta: 0:01:35  lr: 0.000184  min_lr: 0.000184  loss: 3.1391 (2.9917)  weight_decay: 0.0500 (0.0500)  time: 0.6724  data: 0.1267  max mem: 41794
Epoch: [261]  [200/312]  eta: 0:01:27  lr: 0.000183  min_lr: 0.000183  loss: 3.1718 (3.0060)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.1428  max mem: 41794
Epoch: [261]  [210/312]  eta: 0:01:20  lr: 0.000183  min_lr: 0.000183  loss: 3.1179 (2.9978)  weight_decay: 0.0500 (0.0500)  time: 0.8652  data: 0.2677  max mem: 41794
Epoch: [261]  [220/312]  eta: 0:01:11  lr: 0.000183  min_lr: 0.000183  loss: 2.8540 (2.9945)  weight_decay: 0.0500 (0.0500)  time: 0.7024  data: 0.1539  max mem: 41794
Epoch: [261]  [230/312]  eta: 0:01:04  lr: 0.000182  min_lr: 0.000182  loss: 3.2448 (3.0100)  weight_decay: 0.0500 (0.0500)  time: 0.7049  data: 0.1361  max mem: 41794
Epoch: [261]  [240/312]  eta: 0:00:56  lr: 0.000182  min_lr: 0.000182  loss: 3.2889 (3.0110)  weight_decay: 0.0500 (0.0500)  time: 0.8550  data: 0.1986  max mem: 41794
Epoch: [261]  [250/312]  eta: 0:00:47  lr: 0.000182  min_lr: 0.000182  loss: 3.2749 (3.0156)  weight_decay: 0.0500 (0.0500)  time: 0.6777  data: 0.0930  max mem: 41794
Epoch: [261]  [260/312]  eta: 0:00:40  lr: 0.000182  min_lr: 0.000182  loss: 3.2519 (3.0152)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.0928  max mem: 41794
Epoch: [261]  [270/312]  eta: 0:00:32  lr: 0.000181  min_lr: 0.000181  loss: 3.0571 (3.0113)  weight_decay: 0.0500 (0.0500)  time: 0.6997  data: 0.0922  max mem: 41794
Epoch: [261]  [280/312]  eta: 0:00:24  lr: 0.000181  min_lr: 0.000181  loss: 3.0520 (3.0123)  weight_decay: 0.0500 (0.0500)  time: 0.6856  data: 0.0851  max mem: 41794
Epoch: [261]  [290/312]  eta: 0:00:17  lr: 0.000181  min_lr: 0.000181  loss: 3.0562 (3.0110)  weight_decay: 0.0500 (0.0500)  time: 0.8568  data: 0.1986  max mem: 41794
Epoch: [261]  [300/312]  eta: 0:00:09  lr: 0.000180  min_lr: 0.000180  loss: 3.0562 (3.0129)  weight_decay: 0.0500 (0.0500)  time: 0.6581  data: 0.1147  max mem: 41794
Epoch: [261]  [310/312]  eta: 0:00:01  lr: 0.000180  min_lr: 0.000180  loss: 3.1451 (3.0177)  weight_decay: 0.0500 (0.0500)  time: 0.4672  data: 0.0001  max mem: 41794
Epoch: [261]  [311/312]  eta: 0:00:00  lr: 0.000180  min_lr: 0.000180  loss: 3.1272 (3.0161)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0001  max mem: 41794
Epoch: [261] Total time: 0:03:57 (0.7597 s / it)
Averaged stats: lr: 0.000180  min_lr: 0.000180  loss: 3.1272 (3.0302)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.8885 (0.8885)  acc1: 85.6771 (85.6771)  acc5: 96.2240 (96.2240)  time: 8.3221  data: 8.1149  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1655 (1.0721)  acc1: 79.5573 (79.7280)  acc5: 95.0521 (95.0240)  time: 1.1219  data: 0.9402  max mem: 41794
Test: Total time: 0:00:10 (1.1588 s / it)
* Acc@1 79.784 Acc@5 95.020 loss 1.070
Accuracy of the model on the 50000 test images: 79.8%
Max accuracy: 79.78%
Epoch: [262]  [  0/312]  eta: 1:16:01  lr: 0.000180  min_lr: 0.000180  loss: 2.7955 (2.7955)  weight_decay: 0.0500 (0.0500)  time: 14.6218  data: 12.5592  max mem: 41794
Epoch: [262]  [ 10/312]  eta: 0:10:53  lr: 0.000180  min_lr: 0.000180  loss: 3.1448 (2.8827)  weight_decay: 0.0500 (0.0500)  time: 2.1647  data: 1.3907  max mem: 41794
Epoch: [262]  [ 20/312]  eta: 0:07:30  lr: 0.000179  min_lr: 0.000179  loss: 3.1136 (2.8926)  weight_decay: 0.0500 (0.0500)  time: 0.8872  data: 0.2109  max mem: 41794
Epoch: [262]  [ 30/312]  eta: 0:05:41  lr: 0.000179  min_lr: 0.000179  loss: 3.1136 (2.9568)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.0760  max mem: 41794
Epoch: [262]  [ 40/312]  eta: 0:04:51  lr: 0.000179  min_lr: 0.000179  loss: 3.1842 (2.9721)  weight_decay: 0.0500 (0.0500)  time: 0.5813  data: 0.0804  max mem: 41794
Epoch: [262]  [ 50/312]  eta: 0:04:24  lr: 0.000179  min_lr: 0.000179  loss: 3.1842 (2.9880)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1919  max mem: 41794
Epoch: [262]  [ 60/312]  eta: 0:03:52  lr: 0.000178  min_lr: 0.000178  loss: 3.2339 (3.0083)  weight_decay: 0.0500 (0.0500)  time: 0.6201  data: 0.1138  max mem: 41794
Epoch: [262]  [ 70/312]  eta: 0:03:43  lr: 0.000178  min_lr: 0.000178  loss: 3.2544 (3.0306)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.2054  max mem: 41794
Epoch: [262]  [ 80/312]  eta: 0:03:31  lr: 0.000178  min_lr: 0.000178  loss: 3.1812 (3.0289)  weight_decay: 0.0500 (0.0500)  time: 0.8731  data: 0.3687  max mem: 41794
Epoch: [262]  [ 90/312]  eta: 0:03:14  lr: 0.000177  min_lr: 0.000177  loss: 3.0982 (3.0226)  weight_decay: 0.0500 (0.0500)  time: 0.7165  data: 0.2181  max mem: 41794
Epoch: [262]  [100/312]  eta: 0:03:06  lr: 0.000177  min_lr: 0.000177  loss: 3.0719 (3.0282)  weight_decay: 0.0500 (0.0500)  time: 0.7463  data: 0.2405  max mem: 41794
Epoch: [262]  [110/312]  eta: 0:02:50  lr: 0.000177  min_lr: 0.000177  loss: 3.2158 (3.0425)  weight_decay: 0.0500 (0.0500)  time: 0.6916  data: 0.1866  max mem: 41794
Epoch: [262]  [120/312]  eta: 0:02:42  lr: 0.000176  min_lr: 0.000176  loss: 3.1717 (3.0424)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1994  max mem: 41794
Epoch: [262]  [130/312]  eta: 0:02:35  lr: 0.000176  min_lr: 0.000176  loss: 3.1330 (3.0483)  weight_decay: 0.0500 (0.0500)  time: 0.8993  data: 0.4074  max mem: 41794
Epoch: [262]  [140/312]  eta: 0:02:22  lr: 0.000176  min_lr: 0.000176  loss: 2.9019 (3.0324)  weight_decay: 0.0500 (0.0500)  time: 0.7070  data: 0.2089  max mem: 41794
Epoch: [262]  [150/312]  eta: 0:02:14  lr: 0.000176  min_lr: 0.000176  loss: 2.9019 (3.0260)  weight_decay: 0.0500 (0.0500)  time: 0.6611  data: 0.1639  max mem: 41794
Epoch: [262]  [160/312]  eta: 0:02:05  lr: 0.000175  min_lr: 0.000175  loss: 3.2371 (3.0432)  weight_decay: 0.0500 (0.0500)  time: 0.8111  data: 0.3156  max mem: 41794
Epoch: [262]  [170/312]  eta: 0:01:55  lr: 0.000175  min_lr: 0.000175  loss: 3.3346 (3.0376)  weight_decay: 0.0500 (0.0500)  time: 0.6878  data: 0.1903  max mem: 41794
Epoch: [262]  [180/312]  eta: 0:01:47  lr: 0.000175  min_lr: 0.000175  loss: 2.9860 (3.0274)  weight_decay: 0.0500 (0.0500)  time: 0.7163  data: 0.2194  max mem: 41794
Epoch: [262]  [190/312]  eta: 0:01:37  lr: 0.000174  min_lr: 0.000174  loss: 3.1894 (3.0349)  weight_decay: 0.0500 (0.0500)  time: 0.6818  data: 0.1813  max mem: 41794
Epoch: [262]  [200/312]  eta: 0:01:29  lr: 0.000174  min_lr: 0.000174  loss: 3.1749 (3.0284)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1730  max mem: 41794
Epoch: [262]  [210/312]  eta: 0:01:21  lr: 0.000174  min_lr: 0.000174  loss: 3.0422 (3.0327)  weight_decay: 0.0500 (0.0500)  time: 0.8625  data: 0.3600  max mem: 41794
Epoch: [262]  [220/312]  eta: 0:01:12  lr: 0.000174  min_lr: 0.000174  loss: 3.2300 (3.0320)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1875  max mem: 41794
Epoch: [262]  [230/312]  eta: 0:01:04  lr: 0.000173  min_lr: 0.000173  loss: 3.2723 (3.0392)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.1684  max mem: 41794
Epoch: [262]  [240/312]  eta: 0:00:57  lr: 0.000173  min_lr: 0.000173  loss: 3.2191 (3.0427)  weight_decay: 0.0500 (0.0500)  time: 0.8464  data: 0.3108  max mem: 41794
Epoch: [262]  [250/312]  eta: 0:00:48  lr: 0.000173  min_lr: 0.000173  loss: 3.2915 (3.0533)  weight_decay: 0.0500 (0.0500)  time: 0.6806  data: 0.1577  max mem: 41794
Epoch: [262]  [260/312]  eta: 0:00:40  lr: 0.000172  min_lr: 0.000172  loss: 3.2825 (3.0566)  weight_decay: 0.0500 (0.0500)  time: 0.6933  data: 0.1827  max mem: 41794
Epoch: [262]  [270/312]  eta: 0:00:32  lr: 0.000172  min_lr: 0.000172  loss: 3.2354 (3.0541)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.1692  max mem: 41794
Epoch: [262]  [280/312]  eta: 0:00:24  lr: 0.000172  min_lr: 0.000172  loss: 3.2555 (3.0638)  weight_decay: 0.0500 (0.0500)  time: 0.6906  data: 0.1896  max mem: 41794
Epoch: [262]  [290/312]  eta: 0:00:17  lr: 0.000171  min_lr: 0.000171  loss: 3.2150 (3.0620)  weight_decay: 0.0500 (0.0500)  time: 0.7875  data: 0.2827  max mem: 41794
Epoch: [262]  [300/312]  eta: 0:00:09  lr: 0.000171  min_lr: 0.000171  loss: 3.0814 (3.0648)  weight_decay: 0.0500 (0.0500)  time: 0.5850  data: 0.0946  max mem: 41794
Epoch: [262]  [310/312]  eta: 0:00:01  lr: 0.000171  min_lr: 0.000171  loss: 3.3338 (3.0684)  weight_decay: 0.0500 (0.0500)  time: 0.4665  data: 0.0001  max mem: 41794
Epoch: [262]  [311/312]  eta: 0:00:00  lr: 0.000171  min_lr: 0.000171  loss: 3.3338 (3.0683)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [262] Total time: 0:03:57 (0.7622 s / it)
Averaged stats: lr: 0.000171  min_lr: 0.000171  loss: 3.3338 (3.0164)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 0.9719 (0.9719)  acc1: 85.9375 (85.9375)  acc5: 96.2240 (96.2240)  time: 9.1334  data: 8.9209  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2109 (1.1464)  acc1: 78.6458 (79.4560)  acc5: 95.0521 (95.1360)  time: 1.1631  data: 0.9913  max mem: 41794
Test: Total time: 0:00:10 (1.1759 s / it)
* Acc@1 79.852 Acc@5 95.078 loss 1.141
Accuracy of the model on the 50000 test images: 79.9%
Max accuracy: 79.85%
Epoch: [263]  [  0/312]  eta: 1:19:22  lr: 0.000171  min_lr: 0.000171  loss: 2.9100 (2.9100)  weight_decay: 0.0500 (0.0500)  time: 15.2638  data: 13.3866  max mem: 41794
Epoch: [263]  [ 10/312]  eta: 0:11:25  lr: 0.000171  min_lr: 0.000171  loss: 3.2030 (3.0907)  weight_decay: 0.0500 (0.0500)  time: 2.2685  data: 1.3150  max mem: 41794
Epoch: [263]  [ 20/312]  eta: 0:07:23  lr: 0.000170  min_lr: 0.000170  loss: 3.0948 (3.1151)  weight_decay: 0.0500 (0.0500)  time: 0.8321  data: 0.0735  max mem: 41794
Epoch: [263]  [ 30/312]  eta: 0:05:43  lr: 0.000170  min_lr: 0.000170  loss: 3.2672 (3.1941)  weight_decay: 0.0500 (0.0500)  time: 0.6423  data: 0.0516  max mem: 41794
Epoch: [263]  [ 40/312]  eta: 0:04:46  lr: 0.000170  min_lr: 0.000170  loss: 3.2672 (3.1404)  weight_decay: 0.0500 (0.0500)  time: 0.5615  data: 0.0357  max mem: 41794
Epoch: [263]  [ 50/312]  eta: 0:04:15  lr: 0.000169  min_lr: 0.000169  loss: 3.0722 (3.1476)  weight_decay: 0.0500 (0.0500)  time: 0.5954  data: 0.0711  max mem: 41794
Epoch: [263]  [ 60/312]  eta: 0:03:46  lr: 0.000169  min_lr: 0.000169  loss: 3.0521 (3.0814)  weight_decay: 0.0500 (0.0500)  time: 0.5787  data: 0.0677  max mem: 41794
Epoch: [263]  [ 70/312]  eta: 0:03:37  lr: 0.000169  min_lr: 0.000169  loss: 3.1357 (3.1176)  weight_decay: 0.0500 (0.0500)  time: 0.7006  data: 0.2010  max mem: 41794
Epoch: [263]  [ 80/312]  eta: 0:03:26  lr: 0.000169  min_lr: 0.000169  loss: 3.2439 (3.0725)  weight_decay: 0.0500 (0.0500)  time: 0.8647  data: 0.3661  max mem: 41794
Epoch: [263]  [ 90/312]  eta: 0:03:10  lr: 0.000168  min_lr: 0.000168  loss: 3.0261 (3.0806)  weight_decay: 0.0500 (0.0500)  time: 0.7142  data: 0.2110  max mem: 41794
Epoch: [263]  [100/312]  eta: 0:03:02  lr: 0.000168  min_lr: 0.000168  loss: 3.0034 (3.0487)  weight_decay: 0.0500 (0.0500)  time: 0.7408  data: 0.2392  max mem: 41794
Epoch: [263]  [110/312]  eta: 0:02:47  lr: 0.000168  min_lr: 0.000168  loss: 2.8161 (3.0468)  weight_decay: 0.0500 (0.0500)  time: 0.6914  data: 0.1939  max mem: 41794
Epoch: [263]  [120/312]  eta: 0:02:39  lr: 0.000167  min_lr: 0.000167  loss: 3.0821 (3.0460)  weight_decay: 0.0500 (0.0500)  time: 0.6740  data: 0.1777  max mem: 41794
Epoch: [263]  [130/312]  eta: 0:02:31  lr: 0.000167  min_lr: 0.000167  loss: 3.0143 (3.0395)  weight_decay: 0.0500 (0.0500)  time: 0.8695  data: 0.3805  max mem: 41794
Epoch: [263]  [140/312]  eta: 0:02:19  lr: 0.000167  min_lr: 0.000167  loss: 3.0547 (3.0421)  weight_decay: 0.0500 (0.0500)  time: 0.6905  data: 0.2034  max mem: 41794
Epoch: [263]  [150/312]  eta: 0:02:11  lr: 0.000167  min_lr: 0.000167  loss: 3.0628 (3.0415)  weight_decay: 0.0500 (0.0500)  time: 0.6884  data: 0.1914  max mem: 41794
Epoch: [263]  [160/312]  eta: 0:02:04  lr: 0.000166  min_lr: 0.000166  loss: 3.0464 (3.0355)  weight_decay: 0.0500 (0.0500)  time: 0.8703  data: 0.3639  max mem: 41794
Epoch: [263]  [170/312]  eta: 0:01:53  lr: 0.000166  min_lr: 0.000166  loss: 2.9512 (3.0317)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.1918  max mem: 41794
Epoch: [263]  [180/312]  eta: 0:01:45  lr: 0.000166  min_lr: 0.000166  loss: 3.2121 (3.0397)  weight_decay: 0.0500 (0.0500)  time: 0.6821  data: 0.1780  max mem: 41794
Epoch: [263]  [190/312]  eta: 0:01:35  lr: 0.000165  min_lr: 0.000165  loss: 3.0745 (3.0291)  weight_decay: 0.0500 (0.0500)  time: 0.6603  data: 0.1594  max mem: 41794
Epoch: [263]  [200/312]  eta: 0:01:28  lr: 0.000165  min_lr: 0.000165  loss: 3.0172 (3.0256)  weight_decay: 0.0500 (0.0500)  time: 0.6805  data: 0.1823  max mem: 41794
Epoch: [263]  [210/312]  eta: 0:01:21  lr: 0.000165  min_lr: 0.000165  loss: 3.0789 (3.0275)  weight_decay: 0.0500 (0.0500)  time: 0.8707  data: 0.3749  max mem: 41794
Epoch: [263]  [220/312]  eta: 0:01:11  lr: 0.000165  min_lr: 0.000165  loss: 3.0738 (3.0187)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1933  max mem: 41794
Epoch: [263]  [230/312]  eta: 0:01:04  lr: 0.000164  min_lr: 0.000164  loss: 3.0324 (3.0149)  weight_decay: 0.0500 (0.0500)  time: 0.6936  data: 0.1971  max mem: 41794
Epoch: [263]  [240/312]  eta: 0:00:56  lr: 0.000164  min_lr: 0.000164  loss: 3.2737 (3.0264)  weight_decay: 0.0500 (0.0500)  time: 0.8594  data: 0.3642  max mem: 41794
Epoch: [263]  [250/312]  eta: 0:00:48  lr: 0.000164  min_lr: 0.000164  loss: 3.2812 (3.0276)  weight_decay: 0.0500 (0.0500)  time: 0.6829  data: 0.1854  max mem: 41794
Epoch: [263]  [260/312]  eta: 0:00:40  lr: 0.000163  min_lr: 0.000163  loss: 3.1274 (3.0253)  weight_decay: 0.0500 (0.0500)  time: 0.6883  data: 0.1894  max mem: 41794
Epoch: [263]  [270/312]  eta: 0:00:32  lr: 0.000163  min_lr: 0.000163  loss: 3.0619 (3.0195)  weight_decay: 0.0500 (0.0500)  time: 0.6718  data: 0.1717  max mem: 41794
Epoch: [263]  [280/312]  eta: 0:00:24  lr: 0.000163  min_lr: 0.000163  loss: 2.9665 (3.0165)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.1645  max mem: 41794
Epoch: [263]  [290/312]  eta: 0:00:17  lr: 0.000163  min_lr: 0.000163  loss: 2.9388 (3.0174)  weight_decay: 0.0500 (0.0500)  time: 0.8313  data: 0.3336  max mem: 41794
Epoch: [263]  [300/312]  eta: 0:00:09  lr: 0.000162  min_lr: 0.000162  loss: 3.0022 (3.0138)  weight_decay: 0.0500 (0.0500)  time: 0.6545  data: 0.1695  max mem: 41794
Epoch: [263]  [310/312]  eta: 0:00:01  lr: 0.000162  min_lr: 0.000162  loss: 3.2531 (3.0171)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0001  max mem: 41794
Epoch: [263]  [311/312]  eta: 0:00:00  lr: 0.000162  min_lr: 0.000162  loss: 3.2111 (3.0159)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [263] Total time: 0:03:57 (0.7596 s / it)
Averaged stats: lr: 0.000162  min_lr: 0.000162  loss: 3.2111 (3.0205)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.8646 (0.8646)  acc1: 85.8073 (85.8073)  acc5: 96.3542 (96.3542)  time: 8.4587  data: 8.2575  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1296 (1.0470)  acc1: 78.9062 (79.2000)  acc5: 95.0521 (95.1360)  time: 1.0938  data: 0.9176  max mem: 41794
Test: Total time: 0:00:09 (1.1094 s / it)
* Acc@1 79.666 Acc@5 95.014 loss 1.042
Accuracy of the model on the 50000 test images: 79.7%
Max accuracy: 79.85%
Epoch: [264]  [  0/312]  eta: 1:18:03  lr: 0.000162  min_lr: 0.000162  loss: 2.7220 (2.7220)  weight_decay: 0.0500 (0.0500)  time: 15.0106  data: 14.2457  max mem: 41794
Epoch: [264]  [ 10/312]  eta: 0:10:25  lr: 0.000162  min_lr: 0.000162  loss: 2.9771 (2.9886)  weight_decay: 0.0500 (0.0500)  time: 2.0720  data: 1.2976  max mem: 41794
Epoch: [264]  [ 20/312]  eta: 0:06:57  lr: 0.000161  min_lr: 0.000161  loss: 3.1898 (3.0725)  weight_decay: 0.0500 (0.0500)  time: 0.7491  data: 0.0368  max mem: 41794
Epoch: [264]  [ 30/312]  eta: 0:05:20  lr: 0.000161  min_lr: 0.000161  loss: 3.1257 (2.9239)  weight_decay: 0.0500 (0.0500)  time: 0.6238  data: 0.0545  max mem: 41794
Epoch: [264]  [ 40/312]  eta: 0:04:45  lr: 0.000161  min_lr: 0.000161  loss: 2.9311 (2.9532)  weight_decay: 0.0500 (0.0500)  time: 0.6535  data: 0.0564  max mem: 41794
Epoch: [264]  [ 50/312]  eta: 0:04:27  lr: 0.000160  min_lr: 0.000160  loss: 3.1413 (3.0155)  weight_decay: 0.0500 (0.0500)  time: 0.8387  data: 0.1311  max mem: 41794
Epoch: [264]  [ 60/312]  eta: 0:03:56  lr: 0.000160  min_lr: 0.000160  loss: 3.3082 (3.0261)  weight_decay: 0.0500 (0.0500)  time: 0.7124  data: 0.1102  max mem: 41794
Epoch: [264]  [ 70/312]  eta: 0:03:45  lr: 0.000160  min_lr: 0.000160  loss: 3.2771 (3.0243)  weight_decay: 0.0500 (0.0500)  time: 0.7119  data: 0.1440  max mem: 41794
Epoch: [264]  [ 80/312]  eta: 0:03:30  lr: 0.000160  min_lr: 0.000160  loss: 3.2274 (3.0391)  weight_decay: 0.0500 (0.0500)  time: 0.8168  data: 0.2183  max mem: 41794
Epoch: [264]  [ 90/312]  eta: 0:03:14  lr: 0.000159  min_lr: 0.000159  loss: 3.3018 (3.0528)  weight_decay: 0.0500 (0.0500)  time: 0.6815  data: 0.0910  max mem: 41794
Epoch: [264]  [100/312]  eta: 0:03:05  lr: 0.000159  min_lr: 0.000159  loss: 3.1721 (3.0430)  weight_decay: 0.0500 (0.0500)  time: 0.7315  data: 0.1122  max mem: 41794
Epoch: [264]  [110/312]  eta: 0:02:49  lr: 0.000159  min_lr: 0.000159  loss: 3.2115 (3.0458)  weight_decay: 0.0500 (0.0500)  time: 0.6571  data: 0.1122  max mem: 41794
Epoch: [264]  [120/312]  eta: 0:02:41  lr: 0.000159  min_lr: 0.000159  loss: 2.8265 (3.0310)  weight_decay: 0.0500 (0.0500)  time: 0.6638  data: 0.1146  max mem: 41794
Epoch: [264]  [130/312]  eta: 0:02:33  lr: 0.000158  min_lr: 0.000158  loss: 2.9177 (3.0409)  weight_decay: 0.0500 (0.0500)  time: 0.8690  data: 0.2150  max mem: 41794
Epoch: [264]  [140/312]  eta: 0:02:20  lr: 0.000158  min_lr: 0.000158  loss: 3.2303 (3.0516)  weight_decay: 0.0500 (0.0500)  time: 0.6919  data: 0.1009  max mem: 41794
Epoch: [264]  [150/312]  eta: 0:02:13  lr: 0.000158  min_lr: 0.000158  loss: 3.2242 (3.0535)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.0946  max mem: 41794
Epoch: [264]  [160/312]  eta: 0:02:03  lr: 0.000157  min_lr: 0.000157  loss: 3.1457 (3.0446)  weight_decay: 0.0500 (0.0500)  time: 0.7711  data: 0.1863  max mem: 41794
Epoch: [264]  [170/312]  eta: 0:01:54  lr: 0.000157  min_lr: 0.000157  loss: 3.2771 (3.0484)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.0923  max mem: 41794
Epoch: [264]  [180/312]  eta: 0:01:46  lr: 0.000157  min_lr: 0.000157  loss: 3.3158 (3.0659)  weight_decay: 0.0500 (0.0500)  time: 0.7735  data: 0.0779  max mem: 41794
Epoch: [264]  [190/312]  eta: 0:01:36  lr: 0.000157  min_lr: 0.000157  loss: 3.3065 (3.0648)  weight_decay: 0.0500 (0.0500)  time: 0.6693  data: 0.0792  max mem: 41794
Epoch: [264]  [200/312]  eta: 0:01:29  lr: 0.000156  min_lr: 0.000156  loss: 3.0726 (3.0582)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.0751  max mem: 41794
Epoch: [264]  [210/312]  eta: 0:01:21  lr: 0.000156  min_lr: 0.000156  loss: 3.1799 (3.0564)  weight_decay: 0.0500 (0.0500)  time: 0.8744  data: 0.1245  max mem: 41794
Epoch: [264]  [220/312]  eta: 0:01:12  lr: 0.000156  min_lr: 0.000156  loss: 3.1841 (3.0585)  weight_decay: 0.0500 (0.0500)  time: 0.6756  data: 0.0514  max mem: 41794
Epoch: [264]  [230/312]  eta: 0:01:04  lr: 0.000155  min_lr: 0.000155  loss: 3.2323 (3.0614)  weight_decay: 0.0500 (0.0500)  time: 0.6636  data: 0.0412  max mem: 41794
Epoch: [264]  [240/312]  eta: 0:00:56  lr: 0.000155  min_lr: 0.000155  loss: 3.2131 (3.0632)  weight_decay: 0.0500 (0.0500)  time: 0.7325  data: 0.0787  max mem: 41794
Epoch: [264]  [250/312]  eta: 0:00:48  lr: 0.000155  min_lr: 0.000155  loss: 3.2776 (3.0715)  weight_decay: 0.0500 (0.0500)  time: 0.7022  data: 0.0529  max mem: 41794
Epoch: [264]  [260/312]  eta: 0:00:40  lr: 0.000155  min_lr: 0.000155  loss: 3.2066 (3.0709)  weight_decay: 0.0500 (0.0500)  time: 0.8079  data: 0.0415  max mem: 41794
Epoch: [264]  [270/312]  eta: 0:00:32  lr: 0.000154  min_lr: 0.000154  loss: 3.0384 (3.0620)  weight_decay: 0.0500 (0.0500)  time: 0.6859  data: 0.0383  max mem: 41794
Epoch: [264]  [280/312]  eta: 0:00:24  lr: 0.000154  min_lr: 0.000154  loss: 3.1753 (3.0638)  weight_decay: 0.0500 (0.0500)  time: 0.7146  data: 0.0304  max mem: 41794
Epoch: [264]  [290/312]  eta: 0:00:17  lr: 0.000154  min_lr: 0.000154  loss: 3.1686 (3.0594)  weight_decay: 0.0500 (0.0500)  time: 0.8404  data: 0.0187  max mem: 41794
Epoch: [264]  [300/312]  eta: 0:00:09  lr: 0.000154  min_lr: 0.000154  loss: 2.8836 (3.0446)  weight_decay: 0.0500 (0.0500)  time: 0.6204  data: 0.0002  max mem: 41794
Epoch: [264]  [310/312]  eta: 0:00:01  lr: 0.000153  min_lr: 0.000153  loss: 3.0695 (3.0506)  weight_decay: 0.0500 (0.0500)  time: 0.4630  data: 0.0001  max mem: 41794
Epoch: [264]  [311/312]  eta: 0:00:00  lr: 0.000153  min_lr: 0.000153  loss: 3.1820 (3.0512)  weight_decay: 0.0500 (0.0500)  time: 0.4621  data: 0.0001  max mem: 41794
Epoch: [264] Total time: 0:03:58 (0.7637 s / it)
Averaged stats: lr: 0.000153  min_lr: 0.000153  loss: 3.1820 (3.0098)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.8749 (0.8749)  acc1: 85.8073 (85.8073)  acc5: 96.4844 (96.4844)  time: 8.5679  data: 8.3584  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1572 (1.0811)  acc1: 79.2969 (79.5520)  acc5: 95.3125 (95.1200)  time: 1.1129  data: 0.9288  max mem: 41794
Test: Total time: 0:00:10 (1.1288 s / it)
* Acc@1 79.944 Acc@5 95.154 loss 1.078
Accuracy of the model on the 50000 test images: 79.9%
Max accuracy: 79.94%
Epoch: [265]  [  0/312]  eta: 1:21:08  lr: 0.000153  min_lr: 0.000153  loss: 3.3845 (3.3845)  weight_decay: 0.0500 (0.0500)  time: 15.6041  data: 15.1210  max mem: 41794
Epoch: [265]  [ 10/312]  eta: 0:11:11  lr: 0.000153  min_lr: 0.000153  loss: 3.2408 (2.9796)  weight_decay: 0.0500 (0.0500)  time: 2.2250  data: 1.3756  max mem: 41794
Epoch: [265]  [ 20/312]  eta: 0:07:09  lr: 0.000153  min_lr: 0.000153  loss: 3.2401 (3.0409)  weight_decay: 0.0500 (0.0500)  time: 0.7633  data: 0.0596  max mem: 41794
Epoch: [265]  [ 30/312]  eta: 0:05:27  lr: 0.000152  min_lr: 0.000152  loss: 3.1932 (3.0383)  weight_decay: 0.0500 (0.0500)  time: 0.5758  data: 0.0594  max mem: 41794
Epoch: [265]  [ 40/312]  eta: 0:04:35  lr: 0.000152  min_lr: 0.000152  loss: 3.1297 (3.0484)  weight_decay: 0.0500 (0.0500)  time: 0.5304  data: 0.0216  max mem: 41794
Epoch: [265]  [ 50/312]  eta: 0:04:17  lr: 0.000152  min_lr: 0.000152  loss: 3.1232 (3.0251)  weight_decay: 0.0500 (0.0500)  time: 0.7100  data: 0.2107  max mem: 41794
Epoch: [265]  [ 60/312]  eta: 0:03:47  lr: 0.000152  min_lr: 0.000152  loss: 2.9959 (3.0059)  weight_decay: 0.0500 (0.0500)  time: 0.6837  data: 0.1897  max mem: 41794
Epoch: [265]  [ 70/312]  eta: 0:03:37  lr: 0.000151  min_lr: 0.000151  loss: 2.9596 (2.9921)  weight_decay: 0.0500 (0.0500)  time: 0.6879  data: 0.1834  max mem: 41794
Epoch: [265]  [ 80/312]  eta: 0:03:27  lr: 0.000151  min_lr: 0.000151  loss: 3.1912 (3.0340)  weight_decay: 0.0500 (0.0500)  time: 0.8649  data: 0.3647  max mem: 41794
Epoch: [265]  [ 90/312]  eta: 0:03:09  lr: 0.000151  min_lr: 0.000151  loss: 3.2518 (3.0344)  weight_decay: 0.0500 (0.0500)  time: 0.6843  data: 0.1863  max mem: 41794
Epoch: [265]  [100/312]  eta: 0:03:01  lr: 0.000150  min_lr: 0.000150  loss: 3.1584 (3.0185)  weight_decay: 0.0500 (0.0500)  time: 0.7054  data: 0.1838  max mem: 41794
Epoch: [265]  [110/312]  eta: 0:02:46  lr: 0.000150  min_lr: 0.000150  loss: 3.1308 (3.0240)  weight_decay: 0.0500 (0.0500)  time: 0.6949  data: 0.1795  max mem: 41794
Epoch: [265]  [120/312]  eta: 0:02:39  lr: 0.000150  min_lr: 0.000150  loss: 3.1308 (3.0166)  weight_decay: 0.0500 (0.0500)  time: 0.7136  data: 0.1719  max mem: 41794
Epoch: [265]  [130/312]  eta: 0:02:32  lr: 0.000150  min_lr: 0.000150  loss: 3.2670 (3.0173)  weight_decay: 0.0500 (0.0500)  time: 0.9199  data: 0.3047  max mem: 41794
Epoch: [265]  [140/312]  eta: 0:02:20  lr: 0.000149  min_lr: 0.000149  loss: 3.1312 (2.9941)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.1352  max mem: 41794
Epoch: [265]  [150/312]  eta: 0:02:12  lr: 0.000149  min_lr: 0.000149  loss: 3.1312 (3.0112)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.1295  max mem: 41794
Epoch: [265]  [160/312]  eta: 0:02:03  lr: 0.000149  min_lr: 0.000149  loss: 3.2191 (3.0111)  weight_decay: 0.0500 (0.0500)  time: 0.8203  data: 0.2538  max mem: 41794
Epoch: [265]  [170/312]  eta: 0:01:54  lr: 0.000149  min_lr: 0.000149  loss: 3.2191 (3.0237)  weight_decay: 0.0500 (0.0500)  time: 0.6993  data: 0.1267  max mem: 41794
Epoch: [265]  [180/312]  eta: 0:01:47  lr: 0.000148  min_lr: 0.000148  loss: 3.1578 (3.0271)  weight_decay: 0.0500 (0.0500)  time: 0.7747  data: 0.1248  max mem: 41794
Epoch: [265]  [190/312]  eta: 0:01:36  lr: 0.000148  min_lr: 0.000148  loss: 3.1398 (3.0216)  weight_decay: 0.0500 (0.0500)  time: 0.7002  data: 0.1248  max mem: 41794
Epoch: [265]  [200/312]  eta: 0:01:29  lr: 0.000148  min_lr: 0.000148  loss: 3.1721 (3.0254)  weight_decay: 0.0500 (0.0500)  time: 0.6690  data: 0.0990  max mem: 41794
Epoch: [265]  [210/312]  eta: 0:01:21  lr: 0.000147  min_lr: 0.000147  loss: 3.1147 (3.0186)  weight_decay: 0.0500 (0.0500)  time: 0.8534  data: 0.1988  max mem: 41794
Epoch: [265]  [220/312]  eta: 0:01:12  lr: 0.000147  min_lr: 0.000147  loss: 3.1147 (3.0272)  weight_decay: 0.0500 (0.0500)  time: 0.6850  data: 0.1009  max mem: 41794
Epoch: [265]  [230/312]  eta: 0:01:04  lr: 0.000147  min_lr: 0.000147  loss: 3.2119 (3.0312)  weight_decay: 0.0500 (0.0500)  time: 0.6656  data: 0.0951  max mem: 41794
Epoch: [265]  [240/312]  eta: 0:00:56  lr: 0.000147  min_lr: 0.000147  loss: 2.9688 (3.0183)  weight_decay: 0.0500 (0.0500)  time: 0.7997  data: 0.2032  max mem: 41794
Epoch: [265]  [250/312]  eta: 0:00:48  lr: 0.000146  min_lr: 0.000146  loss: 3.1848 (3.0264)  weight_decay: 0.0500 (0.0500)  time: 0.6996  data: 0.1091  max mem: 41794
Epoch: [265]  [260/312]  eta: 0:00:40  lr: 0.000146  min_lr: 0.000146  loss: 3.2743 (3.0314)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.0603  max mem: 41794
Epoch: [265]  [270/312]  eta: 0:00:32  lr: 0.000146  min_lr: 0.000146  loss: 3.1423 (3.0314)  weight_decay: 0.0500 (0.0500)  time: 0.6704  data: 0.0626  max mem: 41794
Epoch: [265]  [280/312]  eta: 0:00:24  lr: 0.000146  min_lr: 0.000146  loss: 3.1300 (3.0277)  weight_decay: 0.0500 (0.0500)  time: 0.6806  data: 0.0672  max mem: 41794
Epoch: [265]  [290/312]  eta: 0:00:17  lr: 0.000145  min_lr: 0.000145  loss: 3.1023 (3.0275)  weight_decay: 0.0500 (0.0500)  time: 0.8346  data: 0.1364  max mem: 41794
Epoch: [265]  [300/312]  eta: 0:00:09  lr: 0.000145  min_lr: 0.000145  loss: 3.1023 (3.0257)  weight_decay: 0.0500 (0.0500)  time: 0.6422  data: 0.0719  max mem: 41794
Epoch: [265]  [310/312]  eta: 0:00:01  lr: 0.000145  min_lr: 0.000145  loss: 3.0928 (3.0240)  weight_decay: 0.0500 (0.0500)  time: 0.4676  data: 0.0001  max mem: 41794
Epoch: [265]  [311/312]  eta: 0:00:00  lr: 0.000145  min_lr: 0.000145  loss: 2.9875 (3.0234)  weight_decay: 0.0500 (0.0500)  time: 0.4659  data: 0.0001  max mem: 41794
Epoch: [265] Total time: 0:03:57 (0.7627 s / it)
Averaged stats: lr: 0.000145  min_lr: 0.000145  loss: 2.9875 (3.0166)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.8278 (0.8278)  acc1: 86.4583 (86.4583)  acc5: 96.7448 (96.7448)  time: 8.8393  data: 8.6291  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1141 (1.0362)  acc1: 78.6458 (79.6000)  acc5: 96.0938 (95.2800)  time: 1.1301  data: 0.9589  max mem: 41794
Test: Total time: 0:00:10 (1.1588 s / it)
* Acc@1 80.052 Acc@5 95.106 loss 1.025
Accuracy of the model on the 50000 test images: 80.1%
Max accuracy: 80.05%
Epoch: [266]  [  0/312]  eta: 1:24:50  lr: 0.000145  min_lr: 0.000145  loss: 3.2362 (3.2362)  weight_decay: 0.0500 (0.0500)  time: 16.3157  data: 11.6142  max mem: 41794
Epoch: [266]  [ 10/312]  eta: 0:11:25  lr: 0.000144  min_lr: 0.000144  loss: 3.1999 (3.1125)  weight_decay: 0.0500 (0.0500)  time: 2.2696  data: 1.3794  max mem: 41794
Epoch: [266]  [ 20/312]  eta: 0:07:15  lr: 0.000144  min_lr: 0.000144  loss: 3.0448 (2.9925)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.2437  max mem: 41794
Epoch: [266]  [ 30/312]  eta: 0:05:33  lr: 0.000144  min_lr: 0.000144  loss: 3.0448 (2.9853)  weight_decay: 0.0500 (0.0500)  time: 0.5830  data: 0.0734  max mem: 41794
Epoch: [266]  [ 40/312]  eta: 0:04:36  lr: 0.000144  min_lr: 0.000144  loss: 3.0585 (2.9723)  weight_decay: 0.0500 (0.0500)  time: 0.5181  data: 0.0093  max mem: 41794
Epoch: [266]  [ 50/312]  eta: 0:04:11  lr: 0.000143  min_lr: 0.000143  loss: 3.0695 (2.9734)  weight_decay: 0.0500 (0.0500)  time: 0.6179  data: 0.0615  max mem: 41794
Epoch: [266]  [ 60/312]  eta: 0:03:43  lr: 0.000143  min_lr: 0.000143  loss: 2.9838 (2.9515)  weight_decay: 0.0500 (0.0500)  time: 0.6177  data: 0.0667  max mem: 41794
Epoch: [266]  [ 70/312]  eta: 0:03:30  lr: 0.000143  min_lr: 0.000143  loss: 2.8147 (2.9157)  weight_decay: 0.0500 (0.0500)  time: 0.6369  data: 0.1382  max mem: 41794
Epoch: [266]  [ 80/312]  eta: 0:03:19  lr: 0.000143  min_lr: 0.000143  loss: 3.1270 (2.9565)  weight_decay: 0.0500 (0.0500)  time: 0.7796  data: 0.2532  max mem: 41794
Epoch: [266]  [ 90/312]  eta: 0:03:04  lr: 0.000142  min_lr: 0.000142  loss: 3.1975 (2.9704)  weight_decay: 0.0500 (0.0500)  time: 0.6999  data: 0.1735  max mem: 41794
Epoch: [266]  [100/312]  eta: 0:02:58  lr: 0.000142  min_lr: 0.000142  loss: 3.1846 (2.9812)  weight_decay: 0.0500 (0.0500)  time: 0.7594  data: 0.2178  max mem: 41794
Epoch: [266]  [110/312]  eta: 0:02:43  lr: 0.000142  min_lr: 0.000142  loss: 3.1846 (2.9703)  weight_decay: 0.0500 (0.0500)  time: 0.7071  data: 0.1665  max mem: 41794
Epoch: [266]  [120/312]  eta: 0:02:36  lr: 0.000142  min_lr: 0.000142  loss: 3.0385 (2.9708)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.1332  max mem: 41794
Epoch: [266]  [130/312]  eta: 0:02:29  lr: 0.000141  min_lr: 0.000141  loss: 3.1419 (2.9857)  weight_decay: 0.0500 (0.0500)  time: 0.8699  data: 0.2546  max mem: 41794
Epoch: [266]  [140/312]  eta: 0:02:17  lr: 0.000141  min_lr: 0.000141  loss: 3.2277 (2.9930)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.1220  max mem: 41794
Epoch: [266]  [150/312]  eta: 0:02:10  lr: 0.000141  min_lr: 0.000141  loss: 3.1276 (2.9978)  weight_decay: 0.0500 (0.0500)  time: 0.7400  data: 0.1319  max mem: 41794
Epoch: [266]  [160/312]  eta: 0:02:01  lr: 0.000140  min_lr: 0.000140  loss: 3.1276 (3.0031)  weight_decay: 0.0500 (0.0500)  time: 0.8157  data: 0.1487  max mem: 41794
Epoch: [266]  [170/312]  eta: 0:01:52  lr: 0.000140  min_lr: 0.000140  loss: 3.1435 (3.0004)  weight_decay: 0.0500 (0.0500)  time: 0.6631  data: 0.0787  max mem: 41794
Epoch: [266]  [180/312]  eta: 0:01:44  lr: 0.000140  min_lr: 0.000140  loss: 3.1916 (3.0110)  weight_decay: 0.0500 (0.0500)  time: 0.7537  data: 0.1450  max mem: 41794
Epoch: [266]  [190/312]  eta: 0:01:34  lr: 0.000140  min_lr: 0.000140  loss: 3.0051 (2.9940)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.0851  max mem: 41794
Epoch: [266]  [200/312]  eta: 0:01:27  lr: 0.000139  min_lr: 0.000139  loss: 3.0385 (3.0018)  weight_decay: 0.0500 (0.0500)  time: 0.6501  data: 0.0969  max mem: 41794
Epoch: [266]  [210/312]  eta: 0:01:20  lr: 0.000139  min_lr: 0.000139  loss: 3.1799 (3.0049)  weight_decay: 0.0500 (0.0500)  time: 0.8400  data: 0.2172  max mem: 41794
Epoch: [266]  [220/312]  eta: 0:01:10  lr: 0.000139  min_lr: 0.000139  loss: 3.2460 (3.0143)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.1252  max mem: 41794
Epoch: [266]  [230/312]  eta: 0:01:03  lr: 0.000139  min_lr: 0.000139  loss: 3.2520 (3.0213)  weight_decay: 0.0500 (0.0500)  time: 0.6436  data: 0.1156  max mem: 41794
Epoch: [266]  [240/312]  eta: 0:00:55  lr: 0.000138  min_lr: 0.000138  loss: 3.0688 (3.0216)  weight_decay: 0.0500 (0.0500)  time: 0.7939  data: 0.2332  max mem: 41794
Epoch: [266]  [250/312]  eta: 0:00:47  lr: 0.000138  min_lr: 0.000138  loss: 3.1742 (3.0241)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.1659  max mem: 41794
Epoch: [266]  [260/312]  eta: 0:00:40  lr: 0.000138  min_lr: 0.000138  loss: 2.8751 (3.0108)  weight_decay: 0.0500 (0.0500)  time: 0.7284  data: 0.2134  max mem: 41794
Epoch: [266]  [270/312]  eta: 0:00:31  lr: 0.000138  min_lr: 0.000138  loss: 3.0177 (3.0201)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.1689  max mem: 41794
Epoch: [266]  [280/312]  eta: 0:00:24  lr: 0.000137  min_lr: 0.000137  loss: 3.2503 (3.0232)  weight_decay: 0.0500 (0.0500)  time: 0.7000  data: 0.1724  max mem: 41794
Epoch: [266]  [290/312]  eta: 0:00:16  lr: 0.000137  min_lr: 0.000137  loss: 3.1937 (3.0256)  weight_decay: 0.0500 (0.0500)  time: 0.8702  data: 0.3303  max mem: 41794
Epoch: [266]  [300/312]  eta: 0:00:09  lr: 0.000137  min_lr: 0.000137  loss: 3.1523 (3.0197)  weight_decay: 0.0500 (0.0500)  time: 0.6571  data: 0.1584  max mem: 41794
Epoch: [266]  [310/312]  eta: 0:00:01  lr: 0.000137  min_lr: 0.000137  loss: 2.9046 (3.0189)  weight_decay: 0.0500 (0.0500)  time: 0.4725  data: 0.0001  max mem: 41794
Epoch: [266]  [311/312]  eta: 0:00:00  lr: 0.000137  min_lr: 0.000137  loss: 2.9289 (3.0186)  weight_decay: 0.0500 (0.0500)  time: 0.4699  data: 0.0001  max mem: 41794
Epoch: [266] Total time: 0:03:55 (0.7536 s / it)
Averaged stats: lr: 0.000137  min_lr: 0.000137  loss: 2.9289 (3.0128)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.8361 (0.8361)  acc1: 85.5469 (85.5469)  acc5: 96.6146 (96.6146)  time: 8.6116  data: 8.4003  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1023 (1.0272)  acc1: 79.2969 (79.5680)  acc5: 95.1823 (95.2640)  time: 1.1048  data: 0.9335  max mem: 41794
Test: Total time: 0:00:10 (1.1452 s / it)
* Acc@1 79.932 Acc@5 95.072 loss 1.020
Accuracy of the model on the 50000 test images: 79.9%
Max accuracy: 80.05%
Epoch: [267]  [  0/312]  eta: 1:18:32  lr: 0.000136  min_lr: 0.000136  loss: 3.3012 (3.3012)  weight_decay: 0.0500 (0.0500)  time: 15.1041  data: 11.8398  max mem: 41794
Epoch: [267]  [ 10/312]  eta: 0:10:57  lr: 0.000136  min_lr: 0.000136  loss: 3.1790 (3.0852)  weight_decay: 0.0500 (0.0500)  time: 2.1763  data: 1.2823  max mem: 41794
Epoch: [267]  [ 20/312]  eta: 0:07:26  lr: 0.000136  min_lr: 0.000136  loss: 3.1790 (3.0578)  weight_decay: 0.0500 (0.0500)  time: 0.8518  data: 0.1708  max mem: 41794
Epoch: [267]  [ 30/312]  eta: 0:05:42  lr: 0.000136  min_lr: 0.000136  loss: 3.0029 (2.9448)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.0632  max mem: 41794
Epoch: [267]  [ 40/312]  eta: 0:04:51  lr: 0.000135  min_lr: 0.000135  loss: 2.8263 (2.9666)  weight_decay: 0.0500 (0.0500)  time: 0.5870  data: 0.0118  max mem: 41794
Epoch: [267]  [ 50/312]  eta: 0:04:30  lr: 0.000135  min_lr: 0.000135  loss: 2.9894 (2.9753)  weight_decay: 0.0500 (0.0500)  time: 0.7510  data: 0.0565  max mem: 41794
Epoch: [267]  [ 60/312]  eta: 0:03:58  lr: 0.000135  min_lr: 0.000135  loss: 2.9894 (2.9584)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.0506  max mem: 41794
Epoch: [267]  [ 70/312]  eta: 0:03:45  lr: 0.000135  min_lr: 0.000135  loss: 2.9938 (2.9625)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.0481  max mem: 41794
Epoch: [267]  [ 80/312]  eta: 0:03:30  lr: 0.000134  min_lr: 0.000134  loss: 2.9792 (2.9409)  weight_decay: 0.0500 (0.0500)  time: 0.7907  data: 0.0515  max mem: 41794
Epoch: [267]  [ 90/312]  eta: 0:03:14  lr: 0.000134  min_lr: 0.000134  loss: 3.1337 (2.9570)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.0403  max mem: 41794
Epoch: [267]  [100/312]  eta: 0:03:05  lr: 0.000134  min_lr: 0.000134  loss: 3.1528 (2.9536)  weight_decay: 0.0500 (0.0500)  time: 0.7389  data: 0.0869  max mem: 41794
Epoch: [267]  [110/312]  eta: 0:02:49  lr: 0.000134  min_lr: 0.000134  loss: 3.0944 (2.9754)  weight_decay: 0.0500 (0.0500)  time: 0.6732  data: 0.0506  max mem: 41794
Epoch: [267]  [120/312]  eta: 0:02:41  lr: 0.000133  min_lr: 0.000133  loss: 3.0770 (2.9811)  weight_decay: 0.0500 (0.0500)  time: 0.6807  data: 0.0577  max mem: 41794
Epoch: [267]  [130/312]  eta: 0:02:33  lr: 0.000133  min_lr: 0.000133  loss: 3.0418 (2.9690)  weight_decay: 0.0500 (0.0500)  time: 0.8628  data: 0.0875  max mem: 41794
Epoch: [267]  [140/312]  eta: 0:02:21  lr: 0.000133  min_lr: 0.000133  loss: 2.7439 (2.9610)  weight_decay: 0.0500 (0.0500)  time: 0.7054  data: 0.0304  max mem: 41794
Epoch: [267]  [150/312]  eta: 0:02:14  lr: 0.000133  min_lr: 0.000133  loss: 3.0189 (2.9686)  weight_decay: 0.0500 (0.0500)  time: 0.7487  data: 0.0736  max mem: 41794
Epoch: [267]  [160/312]  eta: 0:02:04  lr: 0.000132  min_lr: 0.000132  loss: 3.2721 (2.9923)  weight_decay: 0.0500 (0.0500)  time: 0.8187  data: 0.1228  max mem: 41794
Epoch: [267]  [170/312]  eta: 0:01:55  lr: 0.000132  min_lr: 0.000132  loss: 3.2827 (2.9939)  weight_decay: 0.0500 (0.0500)  time: 0.6841  data: 0.0500  max mem: 41794
Epoch: [267]  [180/312]  eta: 0:01:47  lr: 0.000132  min_lr: 0.000132  loss: 3.1378 (3.0008)  weight_decay: 0.0500 (0.0500)  time: 0.7553  data: 0.0059  max mem: 41794
Epoch: [267]  [190/312]  eta: 0:01:37  lr: 0.000132  min_lr: 0.000132  loss: 3.1122 (2.9983)  weight_decay: 0.0500 (0.0500)  time: 0.6652  data: 0.0060  max mem: 41794
Epoch: [267]  [200/312]  eta: 0:01:29  lr: 0.000131  min_lr: 0.000131  loss: 3.1809 (3.0088)  weight_decay: 0.0500 (0.0500)  time: 0.6894  data: 0.0021  max mem: 41794
Epoch: [267]  [210/312]  eta: 0:01:21  lr: 0.000131  min_lr: 0.000131  loss: 3.1807 (2.9994)  weight_decay: 0.0500 (0.0500)  time: 0.8294  data: 0.0020  max mem: 41794
Epoch: [267]  [220/312]  eta: 0:01:12  lr: 0.000131  min_lr: 0.000131  loss: 2.7632 (2.9915)  weight_decay: 0.0500 (0.0500)  time: 0.6738  data: 0.0006  max mem: 41794
Epoch: [267]  [230/312]  eta: 0:01:04  lr: 0.000131  min_lr: 0.000131  loss: 2.9653 (2.9975)  weight_decay: 0.0500 (0.0500)  time: 0.6990  data: 0.0038  max mem: 41794
Epoch: [267]  [240/312]  eta: 0:00:56  lr: 0.000130  min_lr: 0.000130  loss: 2.9653 (2.9920)  weight_decay: 0.0500 (0.0500)  time: 0.7718  data: 0.0053  max mem: 41794
Epoch: [267]  [250/312]  eta: 0:00:48  lr: 0.000130  min_lr: 0.000130  loss: 2.9959 (2.9931)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.0068  max mem: 41794
Epoch: [267]  [260/312]  eta: 0:00:40  lr: 0.000130  min_lr: 0.000130  loss: 3.2047 (3.0001)  weight_decay: 0.0500 (0.0500)  time: 0.7674  data: 0.0069  max mem: 41794
Epoch: [267]  [270/312]  eta: 0:00:32  lr: 0.000130  min_lr: 0.000130  loss: 3.3385 (3.0026)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.0032  max mem: 41794
Epoch: [267]  [280/312]  eta: 0:00:24  lr: 0.000129  min_lr: 0.000129  loss: 3.2914 (3.0044)  weight_decay: 0.0500 (0.0500)  time: 0.6583  data: 0.0100  max mem: 41794
Epoch: [267]  [290/312]  eta: 0:00:17  lr: 0.000129  min_lr: 0.000129  loss: 3.2345 (3.0135)  weight_decay: 0.0500 (0.0500)  time: 0.7919  data: 0.0089  max mem: 41794
Epoch: [267]  [300/312]  eta: 0:00:09  lr: 0.000129  min_lr: 0.000129  loss: 3.1714 (3.0101)  weight_decay: 0.0500 (0.0500)  time: 0.6180  data: 0.0002  max mem: 41794
Epoch: [267]  [310/312]  eta: 0:00:01  lr: 0.000129  min_lr: 0.000129  loss: 3.1964 (3.0150)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [267]  [311/312]  eta: 0:00:00  lr: 0.000129  min_lr: 0.000129  loss: 3.1964 (3.0161)  weight_decay: 0.0500 (0.0500)  time: 0.4648  data: 0.0001  max mem: 41794
Epoch: [267] Total time: 0:03:58 (0.7639 s / it)
Averaged stats: lr: 0.000129  min_lr: 0.000129  loss: 3.1964 (3.0014)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:23  loss: 0.8750 (0.8750)  acc1: 85.4167 (85.4167)  acc5: 96.8750 (96.8750)  time: 9.2402  data: 9.0423  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1614 (1.0769)  acc1: 78.2552 (79.4560)  acc5: 95.3125 (95.1520)  time: 1.1824  data: 1.0048  max mem: 41794
Test: Total time: 0:00:11 (1.2223 s / it)
* Acc@1 79.896 Acc@5 95.138 loss 1.074
Accuracy of the model on the 50000 test images: 79.9%
Max accuracy: 80.05%
Epoch: [268]  [  0/312]  eta: 1:19:48  lr: 0.000128  min_lr: 0.000128  loss: 2.7752 (2.7752)  weight_decay: 0.0500 (0.0500)  time: 15.3469  data: 13.5398  max mem: 41794
Epoch: [268]  [ 10/312]  eta: 0:11:01  lr: 0.000128  min_lr: 0.000128  loss: 3.2260 (3.1468)  weight_decay: 0.0500 (0.0500)  time: 2.1903  data: 1.4015  max mem: 41794
Epoch: [268]  [ 20/312]  eta: 0:07:13  lr: 0.000128  min_lr: 0.000128  loss: 3.1357 (3.0274)  weight_decay: 0.0500 (0.0500)  time: 0.7911  data: 0.1793  max mem: 41794
Epoch: [268]  [ 30/312]  eta: 0:05:30  lr: 0.000128  min_lr: 0.000128  loss: 2.9745 (2.9799)  weight_decay: 0.0500 (0.0500)  time: 0.6130  data: 0.0857  max mem: 41794
Epoch: [268]  [ 40/312]  eta: 0:04:42  lr: 0.000127  min_lr: 0.000127  loss: 2.9861 (2.9727)  weight_decay: 0.0500 (0.0500)  time: 0.5703  data: 0.0317  max mem: 41794
Epoch: [268]  [ 50/312]  eta: 0:04:19  lr: 0.000127  min_lr: 0.000127  loss: 3.1846 (3.0168)  weight_decay: 0.0500 (0.0500)  time: 0.7081  data: 0.1847  max mem: 41794
Epoch: [268]  [ 60/312]  eta: 0:03:49  lr: 0.000127  min_lr: 0.000127  loss: 3.1846 (3.0474)  weight_decay: 0.0500 (0.0500)  time: 0.6517  data: 0.1537  max mem: 41794
Epoch: [268]  [ 70/312]  eta: 0:03:44  lr: 0.000127  min_lr: 0.000127  loss: 3.1226 (3.0494)  weight_decay: 0.0500 (0.0500)  time: 0.7631  data: 0.2566  max mem: 41794
Epoch: [268]  [ 80/312]  eta: 0:03:30  lr: 0.000126  min_lr: 0.000126  loss: 3.1482 (3.0627)  weight_decay: 0.0500 (0.0500)  time: 0.8986  data: 0.3968  max mem: 41794
Epoch: [268]  [ 90/312]  eta: 0:03:13  lr: 0.000126  min_lr: 0.000126  loss: 3.1532 (3.0386)  weight_decay: 0.0500 (0.0500)  time: 0.6712  data: 0.1807  max mem: 41794
Epoch: [268]  [100/312]  eta: 0:03:04  lr: 0.000126  min_lr: 0.000126  loss: 3.0276 (3.0295)  weight_decay: 0.0500 (0.0500)  time: 0.7245  data: 0.2390  max mem: 41794
Epoch: [268]  [110/312]  eta: 0:02:49  lr: 0.000126  min_lr: 0.000126  loss: 3.0276 (3.0167)  weight_decay: 0.0500 (0.0500)  time: 0.6835  data: 0.1991  max mem: 41794
Epoch: [268]  [120/312]  eta: 0:02:40  lr: 0.000125  min_lr: 0.000125  loss: 2.8907 (3.0054)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.1807  max mem: 41794
Epoch: [268]  [130/312]  eta: 0:02:33  lr: 0.000125  min_lr: 0.000125  loss: 3.0076 (3.0156)  weight_decay: 0.0500 (0.0500)  time: 0.8718  data: 0.3827  max mem: 41794
Epoch: [268]  [140/312]  eta: 0:02:20  lr: 0.000125  min_lr: 0.000125  loss: 3.2201 (3.0216)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.2026  max mem: 41794
Epoch: [268]  [150/312]  eta: 0:02:13  lr: 0.000125  min_lr: 0.000125  loss: 3.1951 (3.0208)  weight_decay: 0.0500 (0.0500)  time: 0.6861  data: 0.1958  max mem: 41794
Epoch: [268]  [160/312]  eta: 0:02:04  lr: 0.000124  min_lr: 0.000124  loss: 3.1395 (3.0254)  weight_decay: 0.0500 (0.0500)  time: 0.8569  data: 0.3625  max mem: 41794
Epoch: [268]  [170/312]  eta: 0:01:53  lr: 0.000124  min_lr: 0.000124  loss: 3.1345 (3.0290)  weight_decay: 0.0500 (0.0500)  time: 0.6620  data: 0.1673  max mem: 41794
Epoch: [268]  [180/312]  eta: 0:01:46  lr: 0.000124  min_lr: 0.000124  loss: 2.9833 (3.0233)  weight_decay: 0.0500 (0.0500)  time: 0.6928  data: 0.1960  max mem: 41794
Epoch: [268]  [190/312]  eta: 0:01:36  lr: 0.000124  min_lr: 0.000124  loss: 3.2292 (3.0270)  weight_decay: 0.0500 (0.0500)  time: 0.6944  data: 0.1960  max mem: 41794
Epoch: [268]  [200/312]  eta: 0:01:29  lr: 0.000123  min_lr: 0.000123  loss: 3.2995 (3.0386)  weight_decay: 0.0500 (0.0500)  time: 0.6855  data: 0.1702  max mem: 41794
Epoch: [268]  [210/312]  eta: 0:01:21  lr: 0.000123  min_lr: 0.000123  loss: 3.2207 (3.0401)  weight_decay: 0.0500 (0.0500)  time: 0.8507  data: 0.3318  max mem: 41794
Epoch: [268]  [220/312]  eta: 0:01:12  lr: 0.000123  min_lr: 0.000123  loss: 2.9801 (3.0281)  weight_decay: 0.0500 (0.0500)  time: 0.6599  data: 0.1622  max mem: 41794
Epoch: [268]  [230/312]  eta: 0:01:04  lr: 0.000123  min_lr: 0.000123  loss: 2.9127 (3.0318)  weight_decay: 0.0500 (0.0500)  time: 0.7040  data: 0.1803  max mem: 41794
Epoch: [268]  [240/312]  eta: 0:00:56  lr: 0.000123  min_lr: 0.000123  loss: 3.1719 (3.0254)  weight_decay: 0.0500 (0.0500)  time: 0.8486  data: 0.2669  max mem: 41794
Epoch: [268]  [250/312]  eta: 0:00:48  lr: 0.000122  min_lr: 0.000122  loss: 3.1033 (3.0207)  weight_decay: 0.0500 (0.0500)  time: 0.6936  data: 0.1484  max mem: 41794
Epoch: [268]  [260/312]  eta: 0:00:40  lr: 0.000122  min_lr: 0.000122  loss: 3.1550 (3.0284)  weight_decay: 0.0500 (0.0500)  time: 0.7584  data: 0.1824  max mem: 41794
Epoch: [268]  [270/312]  eta: 0:00:32  lr: 0.000122  min_lr: 0.000122  loss: 3.3329 (3.0330)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.1212  max mem: 41794
Epoch: [268]  [280/312]  eta: 0:00:24  lr: 0.000122  min_lr: 0.000122  loss: 3.1558 (3.0336)  weight_decay: 0.0500 (0.0500)  time: 0.6704  data: 0.0921  max mem: 41794
Epoch: [268]  [290/312]  eta: 0:00:17  lr: 0.000121  min_lr: 0.000121  loss: 2.9634 (3.0252)  weight_decay: 0.0500 (0.0500)  time: 0.8064  data: 0.1594  max mem: 41794
Epoch: [268]  [300/312]  eta: 0:00:09  lr: 0.000121  min_lr: 0.000121  loss: 2.9634 (3.0187)  weight_decay: 0.0500 (0.0500)  time: 0.6199  data: 0.0677  max mem: 41794
Epoch: [268]  [310/312]  eta: 0:00:01  lr: 0.000121  min_lr: 0.000121  loss: 3.0484 (3.0164)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0001  max mem: 41794
Epoch: [268]  [311/312]  eta: 0:00:00  lr: 0.000121  min_lr: 0.000121  loss: 3.0484 (3.0175)  weight_decay: 0.0500 (0.0500)  time: 0.4640  data: 0.0001  max mem: 41794
Epoch: [268] Total time: 0:03:57 (0.7627 s / it)
Averaged stats: lr: 0.000121  min_lr: 0.000121  loss: 3.0484 (3.0004)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:20  loss: 0.8483 (0.8483)  acc1: 86.0677 (86.0677)  acc5: 97.0052 (97.0052)  time: 8.9684  data: 8.7587  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1156 (1.0423)  acc1: 78.9062 (79.8400)  acc5: 95.0521 (95.1360)  time: 1.1444  data: 0.9733  max mem: 41794
Test: Total time: 0:00:10 (1.1604 s / it)
* Acc@1 79.896 Acc@5 95.108 loss 1.038
Accuracy of the model on the 50000 test images: 79.9%
Max accuracy: 80.05%
Epoch: [269]  [  0/312]  eta: 1:21:58  lr: 0.000121  min_lr: 0.000121  loss: 3.1388 (3.1388)  weight_decay: 0.0500 (0.0500)  time: 15.7656  data: 15.2686  max mem: 41794
Epoch: [269]  [ 10/312]  eta: 0:11:05  lr: 0.000120  min_lr: 0.000120  loss: 3.1778 (3.1318)  weight_decay: 0.0500 (0.0500)  time: 2.2025  data: 1.3888  max mem: 41794
Epoch: [269]  [ 20/312]  eta: 0:07:12  lr: 0.000120  min_lr: 0.000120  loss: 3.0972 (2.9717)  weight_decay: 0.0500 (0.0500)  time: 0.7669  data: 0.0987  max mem: 41794
Epoch: [269]  [ 30/312]  eta: 0:05:29  lr: 0.000120  min_lr: 0.000120  loss: 2.9581 (2.9446)  weight_decay: 0.0500 (0.0500)  time: 0.6009  data: 0.1081  max mem: 41794
Epoch: [269]  [ 40/312]  eta: 0:04:45  lr: 0.000120  min_lr: 0.000120  loss: 2.9630 (2.9324)  weight_decay: 0.0500 (0.0500)  time: 0.5954  data: 0.0962  max mem: 41794
Epoch: [269]  [ 50/312]  eta: 0:04:22  lr: 0.000120  min_lr: 0.000120  loss: 2.9831 (2.9156)  weight_decay: 0.0500 (0.0500)  time: 0.7384  data: 0.2303  max mem: 41794
Epoch: [269]  [ 60/312]  eta: 0:03:50  lr: 0.000119  min_lr: 0.000119  loss: 2.9810 (2.9360)  weight_decay: 0.0500 (0.0500)  time: 0.6431  data: 0.1442  max mem: 41794
Epoch: [269]  [ 70/312]  eta: 0:03:41  lr: 0.000119  min_lr: 0.000119  loss: 3.1268 (2.9556)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.2109  max mem: 41794
Epoch: [269]  [ 80/312]  eta: 0:03:31  lr: 0.000119  min_lr: 0.000119  loss: 3.1268 (2.9657)  weight_decay: 0.0500 (0.0500)  time: 0.8962  data: 0.3966  max mem: 41794
Epoch: [269]  [ 90/312]  eta: 0:03:12  lr: 0.000119  min_lr: 0.000119  loss: 3.2858 (2.9954)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.1863  max mem: 41794
Epoch: [269]  [100/312]  eta: 0:03:04  lr: 0.000118  min_lr: 0.000118  loss: 3.0972 (2.9785)  weight_decay: 0.0500 (0.0500)  time: 0.7137  data: 0.2077  max mem: 41794
Epoch: [269]  [110/312]  eta: 0:02:49  lr: 0.000118  min_lr: 0.000118  loss: 3.1376 (2.9919)  weight_decay: 0.0500 (0.0500)  time: 0.7055  data: 0.2077  max mem: 41794
Epoch: [269]  [120/312]  eta: 0:02:41  lr: 0.000118  min_lr: 0.000118  loss: 3.2038 (3.0050)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.1986  max mem: 41794
Epoch: [269]  [130/312]  eta: 0:02:34  lr: 0.000118  min_lr: 0.000118  loss: 3.0688 (2.9984)  weight_decay: 0.0500 (0.0500)  time: 0.8944  data: 0.3902  max mem: 41794
Epoch: [269]  [140/312]  eta: 0:02:21  lr: 0.000117  min_lr: 0.000117  loss: 3.0688 (2.9877)  weight_decay: 0.0500 (0.0500)  time: 0.6899  data: 0.1922  max mem: 41794
Epoch: [269]  [150/312]  eta: 0:02:13  lr: 0.000117  min_lr: 0.000117  loss: 3.0745 (2.9935)  weight_decay: 0.0500 (0.0500)  time: 0.6697  data: 0.1784  max mem: 41794
Epoch: [269]  [160/312]  eta: 0:02:05  lr: 0.000117  min_lr: 0.000117  loss: 3.3220 (3.0063)  weight_decay: 0.0500 (0.0500)  time: 0.8622  data: 0.3666  max mem: 41794
Epoch: [269]  [170/312]  eta: 0:01:54  lr: 0.000117  min_lr: 0.000117  loss: 3.3444 (3.0222)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1888  max mem: 41794
Epoch: [269]  [180/312]  eta: 0:01:46  lr: 0.000116  min_lr: 0.000116  loss: 3.0646 (3.0172)  weight_decay: 0.0500 (0.0500)  time: 0.6661  data: 0.1709  max mem: 41794
Epoch: [269]  [190/312]  eta: 0:01:36  lr: 0.000116  min_lr: 0.000116  loss: 2.9218 (3.0136)  weight_decay: 0.0500 (0.0500)  time: 0.6663  data: 0.1709  max mem: 41794
Epoch: [269]  [200/312]  eta: 0:01:28  lr: 0.000116  min_lr: 0.000116  loss: 3.2569 (3.0149)  weight_decay: 0.0500 (0.0500)  time: 0.6596  data: 0.1640  max mem: 41794
Epoch: [269]  [210/312]  eta: 0:01:21  lr: 0.000116  min_lr: 0.000116  loss: 3.1077 (3.0105)  weight_decay: 0.0500 (0.0500)  time: 0.8494  data: 0.3397  max mem: 41794
Epoch: [269]  [220/312]  eta: 0:01:12  lr: 0.000115  min_lr: 0.000115  loss: 3.0152 (3.0108)  weight_decay: 0.0500 (0.0500)  time: 0.6816  data: 0.1764  max mem: 41794
Epoch: [269]  [230/312]  eta: 0:01:04  lr: 0.000115  min_lr: 0.000115  loss: 3.1138 (3.0144)  weight_decay: 0.0500 (0.0500)  time: 0.6739  data: 0.1748  max mem: 41794
Epoch: [269]  [240/312]  eta: 0:00:56  lr: 0.000115  min_lr: 0.000115  loss: 3.0954 (3.0090)  weight_decay: 0.0500 (0.0500)  time: 0.8358  data: 0.3248  max mem: 41794
Epoch: [269]  [250/312]  eta: 0:00:48  lr: 0.000115  min_lr: 0.000115  loss: 3.1411 (3.0169)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.1507  max mem: 41794
Epoch: [269]  [260/312]  eta: 0:00:40  lr: 0.000114  min_lr: 0.000114  loss: 3.1512 (3.0179)  weight_decay: 0.0500 (0.0500)  time: 0.6752  data: 0.1720  max mem: 41794
Epoch: [269]  [270/312]  eta: 0:00:32  lr: 0.000114  min_lr: 0.000114  loss: 2.9842 (3.0154)  weight_decay: 0.0500 (0.0500)  time: 0.6749  data: 0.1720  max mem: 41794
Epoch: [269]  [280/312]  eta: 0:00:24  lr: 0.000114  min_lr: 0.000114  loss: 3.0992 (3.0201)  weight_decay: 0.0500 (0.0500)  time: 0.6737  data: 0.1497  max mem: 41794
Epoch: [269]  [290/312]  eta: 0:00:17  lr: 0.000114  min_lr: 0.000114  loss: 3.0963 (3.0096)  weight_decay: 0.0500 (0.0500)  time: 0.8698  data: 0.3176  max mem: 41794
Epoch: [269]  [300/312]  eta: 0:00:09  lr: 0.000113  min_lr: 0.000113  loss: 3.0385 (3.0153)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.1683  max mem: 41794
Epoch: [269]  [310/312]  eta: 0:00:01  lr: 0.000113  min_lr: 0.000113  loss: 3.2163 (3.0174)  weight_decay: 0.0500 (0.0500)  time: 0.4681  data: 0.0001  max mem: 41794
Epoch: [269]  [311/312]  eta: 0:00:00  lr: 0.000113  min_lr: 0.000113  loss: 3.2163 (3.0195)  weight_decay: 0.0500 (0.0500)  time: 0.4661  data: 0.0001  max mem: 41794
Epoch: [269] Total time: 0:03:57 (0.7612 s / it)
Averaged stats: lr: 0.000113  min_lr: 0.000113  loss: 3.2163 (3.0023)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 0.9483 (0.9483)  acc1: 85.9375 (85.9375)  acc5: 96.6146 (96.6146)  time: 8.1890  data: 7.9891  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2085 (1.1322)  acc1: 78.3854 (79.6640)  acc5: 95.4427 (95.2000)  time: 1.1053  data: 0.9341  max mem: 41794
Test: Total time: 0:00:10 (1.1182 s / it)
* Acc@1 79.958 Acc@5 95.068 loss 1.128
Accuracy of the model on the 50000 test images: 80.0%
Max accuracy: 80.05%
Epoch: [270]  [  0/312]  eta: 1:23:32  lr: 0.000113  min_lr: 0.000113  loss: 3.1954 (3.1954)  weight_decay: 0.0500 (0.0500)  time: 16.0663  data: 14.0022  max mem: 41794
Epoch: [270]  [ 10/312]  eta: 0:10:55  lr: 0.000113  min_lr: 0.000113  loss: 3.0454 (2.9500)  weight_decay: 0.0500 (0.0500)  time: 2.1698  data: 1.3787  max mem: 41794
Epoch: [270]  [ 20/312]  eta: 0:07:25  lr: 0.000113  min_lr: 0.000113  loss: 2.9513 (2.9097)  weight_decay: 0.0500 (0.0500)  time: 0.8002  data: 0.1099  max mem: 41794
Epoch: [270]  [ 30/312]  eta: 0:05:39  lr: 0.000112  min_lr: 0.000112  loss: 3.0706 (2.9542)  weight_decay: 0.0500 (0.0500)  time: 0.6725  data: 0.0588  max mem: 41794
Epoch: [270]  [ 40/312]  eta: 0:04:47  lr: 0.000112  min_lr: 0.000112  loss: 3.1773 (2.9804)  weight_decay: 0.0500 (0.0500)  time: 0.5622  data: 0.0338  max mem: 41794
Epoch: [270]  [ 50/312]  eta: 0:04:29  lr: 0.000112  min_lr: 0.000112  loss: 3.1938 (2.9913)  weight_decay: 0.0500 (0.0500)  time: 0.7560  data: 0.1418  max mem: 41794
Epoch: [270]  [ 60/312]  eta: 0:03:56  lr: 0.000112  min_lr: 0.000112  loss: 3.1985 (2.9839)  weight_decay: 0.0500 (0.0500)  time: 0.6998  data: 0.1154  max mem: 41794
Epoch: [270]  [ 70/312]  eta: 0:03:45  lr: 0.000112  min_lr: 0.000112  loss: 3.0321 (3.0031)  weight_decay: 0.0500 (0.0500)  time: 0.6817  data: 0.1007  max mem: 41794
Epoch: [270]  [ 80/312]  eta: 0:03:33  lr: 0.000111  min_lr: 0.000111  loss: 3.1031 (3.0122)  weight_decay: 0.0500 (0.0500)  time: 0.8552  data: 0.2444  max mem: 41794
Epoch: [270]  [ 90/312]  eta: 0:03:14  lr: 0.000111  min_lr: 0.000111  loss: 3.2731 (3.0511)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1442  max mem: 41794
Epoch: [270]  [100/312]  eta: 0:03:07  lr: 0.000111  min_lr: 0.000111  loss: 3.2120 (3.0445)  weight_decay: 0.0500 (0.0500)  time: 0.7503  data: 0.1304  max mem: 41794
Epoch: [270]  [110/312]  eta: 0:02:51  lr: 0.000111  min_lr: 0.000111  loss: 3.0681 (3.0340)  weight_decay: 0.0500 (0.0500)  time: 0.7381  data: 0.1304  max mem: 41794
Epoch: [270]  [120/312]  eta: 0:02:45  lr: 0.000110  min_lr: 0.000110  loss: 3.0293 (3.0348)  weight_decay: 0.0500 (0.0500)  time: 0.7412  data: 0.0853  max mem: 41794
Epoch: [270]  [130/312]  eta: 0:02:36  lr: 0.000110  min_lr: 0.000110  loss: 3.0597 (3.0381)  weight_decay: 0.0500 (0.0500)  time: 0.9144  data: 0.1086  max mem: 41794
Epoch: [270]  [140/312]  eta: 0:02:24  lr: 0.000110  min_lr: 0.000110  loss: 3.1213 (3.0213)  weight_decay: 0.0500 (0.0500)  time: 0.7189  data: 0.0704  max mem: 41794
Epoch: [270]  [150/312]  eta: 0:02:16  lr: 0.000110  min_lr: 0.000110  loss: 3.0641 (3.0225)  weight_decay: 0.0500 (0.0500)  time: 0.7050  data: 0.0836  max mem: 41794
Epoch: [270]  [160/312]  eta: 0:02:05  lr: 0.000109  min_lr: 0.000109  loss: 3.0450 (3.0160)  weight_decay: 0.0500 (0.0500)  time: 0.7165  data: 0.0424  max mem: 41794
Epoch: [270]  [170/312]  eta: 0:01:56  lr: 0.000109  min_lr: 0.000109  loss: 3.0016 (3.0159)  weight_decay: 0.0500 (0.0500)  time: 0.6872  data: 0.0554  max mem: 41794
Epoch: [270]  [180/312]  eta: 0:01:48  lr: 0.000109  min_lr: 0.000109  loss: 3.2654 (3.0274)  weight_decay: 0.0500 (0.0500)  time: 0.8136  data: 0.1059  max mem: 41794
Epoch: [270]  [190/312]  eta: 0:01:38  lr: 0.000109  min_lr: 0.000109  loss: 3.1195 (3.0130)  weight_decay: 0.0500 (0.0500)  time: 0.6880  data: 0.0563  max mem: 41794
Epoch: [270]  [200/312]  eta: 0:01:30  lr: 0.000109  min_lr: 0.000109  loss: 2.9322 (3.0068)  weight_decay: 0.0500 (0.0500)  time: 0.6750  data: 0.0461  max mem: 41794
Epoch: [270]  [210/312]  eta: 0:01:22  lr: 0.000108  min_lr: 0.000108  loss: 2.9504 (2.9994)  weight_decay: 0.0500 (0.0500)  time: 0.7943  data: 0.0537  max mem: 41794
Epoch: [270]  [220/312]  eta: 0:01:13  lr: 0.000108  min_lr: 0.000108  loss: 3.0714 (3.0027)  weight_decay: 0.0500 (0.0500)  time: 0.6629  data: 0.0515  max mem: 41794
Epoch: [270]  [230/312]  eta: 0:01:05  lr: 0.000108  min_lr: 0.000108  loss: 3.1235 (3.0044)  weight_decay: 0.0500 (0.0500)  time: 0.7028  data: 0.0992  max mem: 41794
Epoch: [270]  [240/312]  eta: 0:00:56  lr: 0.000108  min_lr: 0.000108  loss: 3.2243 (3.0044)  weight_decay: 0.0500 (0.0500)  time: 0.7351  data: 0.0783  max mem: 41794
Epoch: [270]  [250/312]  eta: 0:00:48  lr: 0.000107  min_lr: 0.000107  loss: 3.1327 (3.0037)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.0719  max mem: 41794
Epoch: [270]  [260/312]  eta: 0:00:41  lr: 0.000107  min_lr: 0.000107  loss: 3.1863 (3.0156)  weight_decay: 0.0500 (0.0500)  time: 0.7643  data: 0.1080  max mem: 41794
Epoch: [270]  [270/312]  eta: 0:00:32  lr: 0.000107  min_lr: 0.000107  loss: 3.2956 (3.0204)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.0621  max mem: 41794
Epoch: [270]  [280/312]  eta: 0:00:24  lr: 0.000107  min_lr: 0.000107  loss: 2.9473 (3.0050)  weight_decay: 0.0500 (0.0500)  time: 0.6759  data: 0.0817  max mem: 41794
Epoch: [270]  [290/312]  eta: 0:00:17  lr: 0.000106  min_lr: 0.000106  loss: 2.7826 (3.0094)  weight_decay: 0.0500 (0.0500)  time: 0.7798  data: 0.1148  max mem: 41794
Epoch: [270]  [300/312]  eta: 0:00:09  lr: 0.000106  min_lr: 0.000106  loss: 3.1102 (3.0042)  weight_decay: 0.0500 (0.0500)  time: 0.5937  data: 0.0393  max mem: 41794
Epoch: [270]  [310/312]  eta: 0:00:01  lr: 0.000106  min_lr: 0.000106  loss: 3.1102 (3.0040)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0029  max mem: 41794
Epoch: [270]  [311/312]  eta: 0:00:00  lr: 0.000106  min_lr: 0.000106  loss: 2.9453 (3.0035)  weight_decay: 0.0500 (0.0500)  time: 0.4656  data: 0.0029  max mem: 41794
Epoch: [270] Total time: 0:03:58 (0.7648 s / it)
Averaged stats: lr: 0.000106  min_lr: 0.000106  loss: 2.9453 (2.9955)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 0.8576 (0.8576)  acc1: 86.0677 (86.0677)  acc5: 96.8750 (96.8750)  time: 8.3544  data: 8.1542  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1352 (1.0598)  acc1: 78.3854 (79.4240)  acc5: 95.5729 (95.3440)  time: 1.0791  data: 0.9061  max mem: 41794
Test: Total time: 0:00:09 (1.0953 s / it)
* Acc@1 79.898 Acc@5 95.184 loss 1.055
Accuracy of the model on the 50000 test images: 79.9%
Max accuracy: 80.05%
Epoch: [271]  [  0/312]  eta: 1:18:42  lr: 0.000106  min_lr: 0.000106  loss: 2.4200 (2.4200)  weight_decay: 0.0500 (0.0500)  time: 15.1367  data: 12.6121  max mem: 41794
Epoch: [271]  [ 10/312]  eta: 0:10:27  lr: 0.000106  min_lr: 0.000106  loss: 3.0228 (2.9768)  weight_decay: 0.0500 (0.0500)  time: 2.0793  data: 1.4115  max mem: 41794
Epoch: [271]  [ 20/312]  eta: 0:07:39  lr: 0.000105  min_lr: 0.000105  loss: 3.2320 (2.9778)  weight_decay: 0.0500 (0.0500)  time: 0.8942  data: 0.2641  max mem: 41794
Epoch: [271]  [ 30/312]  eta: 0:05:53  lr: 0.000105  min_lr: 0.000105  loss: 3.2320 (2.9759)  weight_decay: 0.0500 (0.0500)  time: 0.7995  data: 0.1248  max mem: 41794
Epoch: [271]  [ 40/312]  eta: 0:04:57  lr: 0.000105  min_lr: 0.000105  loss: 3.1446 (2.9924)  weight_decay: 0.0500 (0.0500)  time: 0.5916  data: 0.0462  max mem: 41794
Epoch: [271]  [ 50/312]  eta: 0:04:28  lr: 0.000105  min_lr: 0.000105  loss: 3.1198 (3.0198)  weight_decay: 0.0500 (0.0500)  time: 0.6726  data: 0.1147  max mem: 41794
Epoch: [271]  [ 60/312]  eta: 0:03:56  lr: 0.000105  min_lr: 0.000105  loss: 3.1128 (3.0202)  weight_decay: 0.0500 (0.0500)  time: 0.6204  data: 0.0751  max mem: 41794
Epoch: [271]  [ 70/312]  eta: 0:03:44  lr: 0.000104  min_lr: 0.000104  loss: 2.8578 (2.9869)  weight_decay: 0.0500 (0.0500)  time: 0.6766  data: 0.1155  max mem: 41794
Epoch: [271]  [ 80/312]  eta: 0:03:30  lr: 0.000104  min_lr: 0.000104  loss: 3.1451 (3.0031)  weight_decay: 0.0500 (0.0500)  time: 0.8067  data: 0.1717  max mem: 41794
Epoch: [271]  [ 90/312]  eta: 0:03:13  lr: 0.000104  min_lr: 0.000104  loss: 3.2333 (3.0184)  weight_decay: 0.0500 (0.0500)  time: 0.6800  data: 0.1126  max mem: 41794
Epoch: [271]  [100/312]  eta: 0:03:05  lr: 0.000104  min_lr: 0.000104  loss: 3.1834 (3.0061)  weight_decay: 0.0500 (0.0500)  time: 0.7514  data: 0.1791  max mem: 41794
Epoch: [271]  [110/312]  eta: 0:02:49  lr: 0.000103  min_lr: 0.000103  loss: 3.1834 (3.0188)  weight_decay: 0.0500 (0.0500)  time: 0.6959  data: 0.1246  max mem: 41794
Epoch: [271]  [120/312]  eta: 0:02:42  lr: 0.000103  min_lr: 0.000103  loss: 3.2505 (3.0302)  weight_decay: 0.0500 (0.0500)  time: 0.6982  data: 0.0984  max mem: 41794
Epoch: [271]  [130/312]  eta: 0:02:33  lr: 0.000103  min_lr: 0.000103  loss: 3.3151 (3.0609)  weight_decay: 0.0500 (0.0500)  time: 0.8497  data: 0.1627  max mem: 41794
Epoch: [271]  [140/312]  eta: 0:02:21  lr: 0.000103  min_lr: 0.000103  loss: 3.3572 (3.0614)  weight_decay: 0.0500 (0.0500)  time: 0.6604  data: 0.0664  max mem: 41794
Epoch: [271]  [150/312]  eta: 0:02:13  lr: 0.000102  min_lr: 0.000102  loss: 3.0904 (3.0573)  weight_decay: 0.0500 (0.0500)  time: 0.6853  data: 0.0906  max mem: 41794
Epoch: [271]  [160/312]  eta: 0:02:04  lr: 0.000102  min_lr: 0.000102  loss: 3.1275 (3.0595)  weight_decay: 0.0500 (0.0500)  time: 0.8227  data: 0.1217  max mem: 41794
Epoch: [271]  [170/312]  eta: 0:01:55  lr: 0.000102  min_lr: 0.000102  loss: 3.2481 (3.0645)  weight_decay: 0.0500 (0.0500)  time: 0.7228  data: 0.1070  max mem: 41794
Epoch: [271]  [180/312]  eta: 0:01:47  lr: 0.000102  min_lr: 0.000102  loss: 2.9241 (3.0455)  weight_decay: 0.0500 (0.0500)  time: 0.7682  data: 0.1419  max mem: 41794
Epoch: [271]  [190/312]  eta: 0:01:37  lr: 0.000102  min_lr: 0.000102  loss: 2.9241 (3.0514)  weight_decay: 0.0500 (0.0500)  time: 0.6938  data: 0.0668  max mem: 41794
Epoch: [271]  [200/312]  eta: 0:01:29  lr: 0.000101  min_lr: 0.000101  loss: 3.0620 (3.0474)  weight_decay: 0.0500 (0.0500)  time: 0.6783  data: 0.0602  max mem: 41794
Epoch: [271]  [210/312]  eta: 0:01:22  lr: 0.000101  min_lr: 0.000101  loss: 3.0122 (3.0410)  weight_decay: 0.0500 (0.0500)  time: 0.8892  data: 0.1738  max mem: 41794
Epoch: [271]  [220/312]  eta: 0:01:12  lr: 0.000101  min_lr: 0.000101  loss: 2.8959 (3.0361)  weight_decay: 0.0500 (0.0500)  time: 0.7089  data: 0.1142  max mem: 41794
Epoch: [271]  [230/312]  eta: 0:01:05  lr: 0.000101  min_lr: 0.000101  loss: 3.0889 (3.0316)  weight_decay: 0.0500 (0.0500)  time: 0.6575  data: 0.0746  max mem: 41794
Epoch: [271]  [240/312]  eta: 0:00:57  lr: 0.000100  min_lr: 0.000100  loss: 3.1734 (3.0318)  weight_decay: 0.0500 (0.0500)  time: 0.7885  data: 0.1024  max mem: 41794
Epoch: [271]  [250/312]  eta: 0:00:48  lr: 0.000100  min_lr: 0.000100  loss: 3.2132 (3.0338)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.1040  max mem: 41794
Epoch: [271]  [260/312]  eta: 0:00:40  lr: 0.000100  min_lr: 0.000100  loss: 3.2363 (3.0374)  weight_decay: 0.0500 (0.0500)  time: 0.7340  data: 0.1552  max mem: 41794
Epoch: [271]  [270/312]  eta: 0:00:32  lr: 0.000100  min_lr: 0.000100  loss: 3.2777 (3.0381)  weight_decay: 0.0500 (0.0500)  time: 0.6591  data: 0.0796  max mem: 41794
Epoch: [271]  [280/312]  eta: 0:00:24  lr: 0.000100  min_lr: 0.000100  loss: 3.2313 (3.0348)  weight_decay: 0.0500 (0.0500)  time: 0.6736  data: 0.0925  max mem: 41794
Epoch: [271]  [290/312]  eta: 0:00:17  lr: 0.000099  min_lr: 0.000099  loss: 3.1782 (3.0405)  weight_decay: 0.0500 (0.0500)  time: 0.8357  data: 0.1747  max mem: 41794
Epoch: [271]  [300/312]  eta: 0:00:09  lr: 0.000099  min_lr: 0.000099  loss: 3.2062 (3.0423)  weight_decay: 0.0500 (0.0500)  time: 0.6440  data: 0.0826  max mem: 41794
Epoch: [271]  [310/312]  eta: 0:00:01  lr: 0.000099  min_lr: 0.000099  loss: 3.0765 (3.0457)  weight_decay: 0.0500 (0.0500)  time: 0.4635  data: 0.0001  max mem: 41794
Epoch: [271]  [311/312]  eta: 0:00:00  lr: 0.000099  min_lr: 0.000099  loss: 3.0765 (3.0467)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [271] Total time: 0:03:59 (0.7668 s / it)
Averaged stats: lr: 0.000099  min_lr: 0.000099  loss: 3.0765 (3.0120)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.9293 (0.9293)  acc1: 85.6771 (85.6771)  acc5: 96.8750 (96.8750)  time: 8.8467  data: 8.6343  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2049 (1.1230)  acc1: 78.3854 (79.5680)  acc5: 95.7031 (95.2640)  time: 1.1316  data: 0.9594  max mem: 41794
Test: Total time: 0:00:10 (1.1440 s / it)
* Acc@1 79.972 Acc@5 95.134 loss 1.119
Accuracy of the model on the 50000 test images: 80.0%
Max accuracy: 80.05%
Epoch: [272]  [  0/312]  eta: 1:22:51  lr: 0.000099  min_lr: 0.000099  loss: 2.9781 (2.9781)  weight_decay: 0.0500 (0.0500)  time: 15.9350  data: 14.8683  max mem: 41794
Epoch: [272]  [ 10/312]  eta: 0:11:29  lr: 0.000099  min_lr: 0.000099  loss: 2.7931 (2.7117)  weight_decay: 0.0500 (0.0500)  time: 2.2825  data: 1.3524  max mem: 41794
Epoch: [272]  [ 20/312]  eta: 0:07:26  lr: 0.000098  min_lr: 0.000098  loss: 2.8500 (2.8893)  weight_decay: 0.0500 (0.0500)  time: 0.8074  data: 0.0137  max mem: 41794
Epoch: [272]  [ 30/312]  eta: 0:05:39  lr: 0.000098  min_lr: 0.000098  loss: 3.2331 (2.9417)  weight_decay: 0.0500 (0.0500)  time: 0.6086  data: 0.0154  max mem: 41794
Epoch: [272]  [ 40/312]  eta: 0:04:41  lr: 0.000098  min_lr: 0.000098  loss: 3.0506 (2.9138)  weight_decay: 0.0500 (0.0500)  time: 0.5149  data: 0.0077  max mem: 41794
Epoch: [272]  [ 50/312]  eta: 0:04:18  lr: 0.000098  min_lr: 0.000098  loss: 3.0828 (2.9747)  weight_decay: 0.0500 (0.0500)  time: 0.6549  data: 0.1614  max mem: 41794
Epoch: [272]  [ 60/312]  eta: 0:03:48  lr: 0.000098  min_lr: 0.000098  loss: 3.2828 (3.0123)  weight_decay: 0.0500 (0.0500)  time: 0.6496  data: 0.1561  max mem: 41794
Epoch: [272]  [ 70/312]  eta: 0:03:37  lr: 0.000097  min_lr: 0.000097  loss: 3.2587 (3.0252)  weight_decay: 0.0500 (0.0500)  time: 0.6654  data: 0.1647  max mem: 41794
Epoch: [272]  [ 80/312]  eta: 0:03:26  lr: 0.000097  min_lr: 0.000097  loss: 3.1555 (3.0350)  weight_decay: 0.0500 (0.0500)  time: 0.8362  data: 0.3310  max mem: 41794
Epoch: [272]  [ 90/312]  eta: 0:03:10  lr: 0.000097  min_lr: 0.000097  loss: 3.1708 (3.0416)  weight_decay: 0.0500 (0.0500)  time: 0.7192  data: 0.2135  max mem: 41794
Epoch: [272]  [100/312]  eta: 0:03:03  lr: 0.000097  min_lr: 0.000097  loss: 3.1889 (3.0333)  weight_decay: 0.0500 (0.0500)  time: 0.7565  data: 0.2498  max mem: 41794
Epoch: [272]  [110/312]  eta: 0:02:47  lr: 0.000096  min_lr: 0.000096  loss: 3.1264 (3.0374)  weight_decay: 0.0500 (0.0500)  time: 0.7018  data: 0.2033  max mem: 41794
Epoch: [272]  [120/312]  eta: 0:02:39  lr: 0.000096  min_lr: 0.000096  loss: 3.0528 (3.0322)  weight_decay: 0.0500 (0.0500)  time: 0.6588  data: 0.1574  max mem: 41794
Epoch: [272]  [130/312]  eta: 0:02:31  lr: 0.000096  min_lr: 0.000096  loss: 2.9576 (3.0161)  weight_decay: 0.0500 (0.0500)  time: 0.8347  data: 0.3243  max mem: 41794
Epoch: [272]  [140/312]  eta: 0:02:18  lr: 0.000096  min_lr: 0.000096  loss: 3.0438 (3.0246)  weight_decay: 0.0500 (0.0500)  time: 0.6662  data: 0.1677  max mem: 41794
Epoch: [272]  [150/312]  eta: 0:02:12  lr: 0.000096  min_lr: 0.000096  loss: 3.0600 (3.0050)  weight_decay: 0.0500 (0.0500)  time: 0.7248  data: 0.2328  max mem: 41794
Epoch: [272]  [160/312]  eta: 0:02:04  lr: 0.000095  min_lr: 0.000095  loss: 2.8943 (3.0062)  weight_decay: 0.0500 (0.0500)  time: 0.9111  data: 0.4241  max mem: 41794
Epoch: [272]  [170/312]  eta: 0:01:53  lr: 0.000095  min_lr: 0.000095  loss: 3.1248 (3.0063)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.1931  max mem: 41794
Epoch: [272]  [180/312]  eta: 0:01:46  lr: 0.000095  min_lr: 0.000095  loss: 3.1248 (3.0111)  weight_decay: 0.0500 (0.0500)  time: 0.6822  data: 0.1786  max mem: 41794
Epoch: [272]  [190/312]  eta: 0:01:36  lr: 0.000095  min_lr: 0.000095  loss: 3.0081 (3.0022)  weight_decay: 0.0500 (0.0500)  time: 0.6765  data: 0.1777  max mem: 41794
Epoch: [272]  [200/312]  eta: 0:01:28  lr: 0.000094  min_lr: 0.000094  loss: 2.9163 (2.9945)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1753  max mem: 41794
Epoch: [272]  [210/312]  eta: 0:01:20  lr: 0.000094  min_lr: 0.000094  loss: 3.0585 (2.9961)  weight_decay: 0.0500 (0.0500)  time: 0.8447  data: 0.3438  max mem: 41794
Epoch: [272]  [220/312]  eta: 0:01:11  lr: 0.000094  min_lr: 0.000094  loss: 3.0207 (2.9865)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1691  max mem: 41794
Epoch: [272]  [230/312]  eta: 0:01:04  lr: 0.000094  min_lr: 0.000094  loss: 2.8606 (2.9819)  weight_decay: 0.0500 (0.0500)  time: 0.6735  data: 0.1797  max mem: 41794
Epoch: [272]  [240/312]  eta: 0:00:56  lr: 0.000094  min_lr: 0.000094  loss: 3.1238 (2.9975)  weight_decay: 0.0500 (0.0500)  time: 0.8404  data: 0.3507  max mem: 41794
Epoch: [272]  [250/312]  eta: 0:00:47  lr: 0.000093  min_lr: 0.000093  loss: 3.3643 (3.0013)  weight_decay: 0.0500 (0.0500)  time: 0.6605  data: 0.1716  max mem: 41794
Epoch: [272]  [260/312]  eta: 0:00:40  lr: 0.000093  min_lr: 0.000093  loss: 3.0900 (2.9895)  weight_decay: 0.0500 (0.0500)  time: 0.6890  data: 0.1936  max mem: 41794
Epoch: [272]  [270/312]  eta: 0:00:32  lr: 0.000093  min_lr: 0.000093  loss: 3.1150 (2.9962)  weight_decay: 0.0500 (0.0500)  time: 0.6932  data: 0.1936  max mem: 41794
Epoch: [272]  [280/312]  eta: 0:00:24  lr: 0.000093  min_lr: 0.000093  loss: 3.2421 (3.0063)  weight_decay: 0.0500 (0.0500)  time: 0.6810  data: 0.1855  max mem: 41794
Epoch: [272]  [290/312]  eta: 0:00:16  lr: 0.000093  min_lr: 0.000093  loss: 3.1433 (3.0035)  weight_decay: 0.0500 (0.0500)  time: 0.8397  data: 0.3469  max mem: 41794
Epoch: [272]  [300/312]  eta: 0:00:09  lr: 0.000092  min_lr: 0.000092  loss: 3.1433 (3.0119)  weight_decay: 0.0500 (0.0500)  time: 0.6485  data: 0.1618  max mem: 41794
Epoch: [272]  [310/312]  eta: 0:00:01  lr: 0.000092  min_lr: 0.000092  loss: 3.1959 (3.0118)  weight_decay: 0.0500 (0.0500)  time: 0.4705  data: 0.0001  max mem: 41794
Epoch: [272]  [311/312]  eta: 0:00:00  lr: 0.000092  min_lr: 0.000092  loss: 3.1959 (3.0134)  weight_decay: 0.0500 (0.0500)  time: 0.4691  data: 0.0001  max mem: 41794
Epoch: [272] Total time: 0:03:56 (0.7584 s / it)
Averaged stats: lr: 0.000092  min_lr: 0.000092  loss: 3.1959 (2.9923)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:14  loss: 0.9302 (0.9302)  acc1: 86.0677 (86.0677)  acc5: 96.4844 (96.4844)  time: 8.2329  data: 8.0206  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1820 (1.1181)  acc1: 79.8177 (79.7920)  acc5: 95.1823 (95.1360)  time: 1.0627  data: 0.8913  max mem: 41794
Test: Total time: 0:00:09 (1.0775 s / it)
* Acc@1 79.938 Acc@5 95.106 loss 1.111
Accuracy of the model on the 50000 test images: 79.9%
Max accuracy: 80.05%
Epoch: [273]  [  0/312]  eta: 1:22:04  lr: 0.000092  min_lr: 0.000092  loss: 3.2529 (3.2529)  weight_decay: 0.0500 (0.0500)  time: 15.7831  data: 11.9931  max mem: 41794
Epoch: [273]  [ 10/312]  eta: 0:11:24  lr: 0.000092  min_lr: 0.000092  loss: 2.9211 (3.0117)  weight_decay: 0.0500 (0.0500)  time: 2.2666  data: 1.2469  max mem: 41794
Epoch: [273]  [ 20/312]  eta: 0:07:16  lr: 0.000092  min_lr: 0.000092  loss: 2.9010 (2.8774)  weight_decay: 0.0500 (0.0500)  time: 0.7804  data: 0.0891  max mem: 41794
Epoch: [273]  [ 30/312]  eta: 0:05:31  lr: 0.000091  min_lr: 0.000091  loss: 2.9185 (2.8990)  weight_decay: 0.0500 (0.0500)  time: 0.5739  data: 0.0112  max mem: 41794
Epoch: [273]  [ 40/312]  eta: 0:04:36  lr: 0.000091  min_lr: 0.000091  loss: 3.1645 (2.9455)  weight_decay: 0.0500 (0.0500)  time: 0.5130  data: 0.0121  max mem: 41794
Epoch: [273]  [ 50/312]  eta: 0:04:14  lr: 0.000091  min_lr: 0.000091  loss: 3.1694 (2.9942)  weight_decay: 0.0500 (0.0500)  time: 0.6579  data: 0.0319  max mem: 41794
Epoch: [273]  [ 60/312]  eta: 0:03:46  lr: 0.000091  min_lr: 0.000091  loss: 3.0545 (2.9923)  weight_decay: 0.0500 (0.0500)  time: 0.6591  data: 0.0524  max mem: 41794
Epoch: [273]  [ 70/312]  eta: 0:03:39  lr: 0.000091  min_lr: 0.000091  loss: 3.0545 (2.9912)  weight_decay: 0.0500 (0.0500)  time: 0.7373  data: 0.0822  max mem: 41794
Epoch: [273]  [ 80/312]  eta: 0:03:26  lr: 0.000090  min_lr: 0.000090  loss: 3.0633 (2.9725)  weight_decay: 0.0500 (0.0500)  time: 0.8585  data: 0.0607  max mem: 41794
Epoch: [273]  [ 90/312]  eta: 0:03:09  lr: 0.000090  min_lr: 0.000090  loss: 3.0714 (2.9677)  weight_decay: 0.0500 (0.0500)  time: 0.6622  data: 0.0367  max mem: 41794
Epoch: [273]  [100/312]  eta: 0:03:02  lr: 0.000090  min_lr: 0.000090  loss: 3.0714 (2.9539)  weight_decay: 0.0500 (0.0500)  time: 0.7476  data: 0.0799  max mem: 41794
Epoch: [273]  [110/312]  eta: 0:02:47  lr: 0.000090  min_lr: 0.000090  loss: 3.1010 (2.9461)  weight_decay: 0.0500 (0.0500)  time: 0.7210  data: 0.0505  max mem: 41794
Epoch: [273]  [120/312]  eta: 0:02:41  lr: 0.000089  min_lr: 0.000089  loss: 3.1010 (2.9488)  weight_decay: 0.0500 (0.0500)  time: 0.7455  data: 0.0272  max mem: 41794
Epoch: [273]  [130/312]  eta: 0:02:33  lr: 0.000089  min_lr: 0.000089  loss: 2.8832 (2.9319)  weight_decay: 0.0500 (0.0500)  time: 0.9344  data: 0.0306  max mem: 41794
Epoch: [273]  [140/312]  eta: 0:02:21  lr: 0.000089  min_lr: 0.000089  loss: 2.9993 (2.9450)  weight_decay: 0.0500 (0.0500)  time: 0.7044  data: 0.0180  max mem: 41794
Epoch: [273]  [150/312]  eta: 0:02:13  lr: 0.000089  min_lr: 0.000089  loss: 3.1073 (2.9513)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.0144  max mem: 41794
Epoch: [273]  [160/312]  eta: 0:02:04  lr: 0.000089  min_lr: 0.000089  loss: 3.1073 (2.9581)  weight_decay: 0.0500 (0.0500)  time: 0.7984  data: 0.0129  max mem: 41794
Epoch: [273]  [170/312]  eta: 0:01:54  lr: 0.000088  min_lr: 0.000088  loss: 3.0219 (2.9574)  weight_decay: 0.0500 (0.0500)  time: 0.6801  data: 0.0201  max mem: 41794
Epoch: [273]  [180/312]  eta: 0:01:47  lr: 0.000088  min_lr: 0.000088  loss: 3.1889 (2.9667)  weight_decay: 0.0500 (0.0500)  time: 0.7498  data: 0.0263  max mem: 41794
Epoch: [273]  [190/312]  eta: 0:01:37  lr: 0.000088  min_lr: 0.000088  loss: 3.1889 (2.9744)  weight_decay: 0.0500 (0.0500)  time: 0.6860  data: 0.0189  max mem: 41794
Epoch: [273]  [200/312]  eta: 0:01:29  lr: 0.000088  min_lr: 0.000088  loss: 3.0414 (2.9654)  weight_decay: 0.0500 (0.0500)  time: 0.6870  data: 0.0128  max mem: 41794
Epoch: [273]  [210/312]  eta: 0:01:22  lr: 0.000088  min_lr: 0.000088  loss: 2.7573 (2.9660)  weight_decay: 0.0500 (0.0500)  time: 0.9337  data: 0.0155  max mem: 41794
Epoch: [273]  [220/312]  eta: 0:01:13  lr: 0.000087  min_lr: 0.000087  loss: 2.9609 (2.9637)  weight_decay: 0.0500 (0.0500)  time: 0.7432  data: 0.0099  max mem: 41794
Epoch: [273]  [230/312]  eta: 0:01:05  lr: 0.000087  min_lr: 0.000087  loss: 3.0803 (2.9700)  weight_decay: 0.0500 (0.0500)  time: 0.7159  data: 0.0064  max mem: 41794
Epoch: [273]  [240/312]  eta: 0:00:57  lr: 0.000087  min_lr: 0.000087  loss: 3.0453 (2.9638)  weight_decay: 0.0500 (0.0500)  time: 0.8005  data: 0.0050  max mem: 41794
Epoch: [273]  [250/312]  eta: 0:00:48  lr: 0.000087  min_lr: 0.000087  loss: 2.9924 (2.9654)  weight_decay: 0.0500 (0.0500)  time: 0.6579  data: 0.0024  max mem: 41794
Epoch: [273]  [260/312]  eta: 0:00:41  lr: 0.000087  min_lr: 0.000087  loss: 3.3688 (2.9742)  weight_decay: 0.0500 (0.0500)  time: 0.7484  data: 0.0024  max mem: 41794
Epoch: [273]  [270/312]  eta: 0:00:32  lr: 0.000086  min_lr: 0.000086  loss: 3.3688 (2.9782)  weight_decay: 0.0500 (0.0500)  time: 0.6779  data: 0.0048  max mem: 41794
Epoch: [273]  [280/312]  eta: 0:00:25  lr: 0.000086  min_lr: 0.000086  loss: 3.1573 (2.9821)  weight_decay: 0.0500 (0.0500)  time: 0.7024  data: 0.0091  max mem: 41794
Epoch: [273]  [290/312]  eta: 0:00:17  lr: 0.000086  min_lr: 0.000086  loss: 3.1877 (2.9847)  weight_decay: 0.0500 (0.0500)  time: 0.8142  data: 0.0061  max mem: 41794
Epoch: [273]  [300/312]  eta: 0:00:09  lr: 0.000086  min_lr: 0.000086  loss: 3.1454 (2.9852)  weight_decay: 0.0500 (0.0500)  time: 0.6010  data: 0.0002  max mem: 41794
Epoch: [273]  [310/312]  eta: 0:00:01  lr: 0.000086  min_lr: 0.000086  loss: 3.0584 (2.9824)  weight_decay: 0.0500 (0.0500)  time: 0.4655  data: 0.0001  max mem: 41794
Epoch: [273]  [311/312]  eta: 0:00:00  lr: 0.000085  min_lr: 0.000085  loss: 3.0622 (2.9828)  weight_decay: 0.0500 (0.0500)  time: 0.4647  data: 0.0001  max mem: 41794
Epoch: [273] Total time: 0:04:00 (0.7698 s / it)
Averaged stats: lr: 0.000085  min_lr: 0.000085  loss: 3.0622 (2.9946)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 0.8598 (0.8598)  acc1: 85.9375 (85.9375)  acc5: 96.3542 (96.3542)  time: 8.2064  data: 7.9936  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1228 (1.0484)  acc1: 79.1667 (79.5200)  acc5: 95.8333 (95.3600)  time: 1.0685  data: 0.8883  max mem: 41794
Test: Total time: 0:00:09 (1.0838 s / it)
* Acc@1 80.032 Acc@5 95.204 loss 1.044
Accuracy of the model on the 50000 test images: 80.0%
Max accuracy: 80.05%
Epoch: [274]  [  0/312]  eta: 1:24:28  lr: 0.000085  min_lr: 0.000085  loss: 3.3090 (3.3090)  weight_decay: 0.0500 (0.0500)  time: 16.2462  data: 11.5940  max mem: 41794
Epoch: [274]  [ 10/312]  eta: 0:11:23  lr: 0.000085  min_lr: 0.000085  loss: 3.1818 (2.9676)  weight_decay: 0.0500 (0.0500)  time: 2.2641  data: 1.3793  max mem: 41794
Epoch: [274]  [ 20/312]  eta: 0:07:21  lr: 0.000085  min_lr: 0.000085  loss: 3.1231 (2.8524)  weight_decay: 0.0500 (0.0500)  time: 0.7764  data: 0.1929  max mem: 41794
Epoch: [274]  [ 30/312]  eta: 0:05:35  lr: 0.000085  min_lr: 0.000085  loss: 2.9342 (2.8580)  weight_decay: 0.0500 (0.0500)  time: 0.5991  data: 0.0166  max mem: 41794
Epoch: [274]  [ 40/312]  eta: 0:04:43  lr: 0.000085  min_lr: 0.000085  loss: 3.1565 (2.9270)  weight_decay: 0.0500 (0.0500)  time: 0.5457  data: 0.0041  max mem: 41794
Epoch: [274]  [ 50/312]  eta: 0:04:22  lr: 0.000084  min_lr: 0.000084  loss: 3.2269 (2.9293)  weight_decay: 0.0500 (0.0500)  time: 0.7084  data: 0.0581  max mem: 41794
Epoch: [274]  [ 60/312]  eta: 0:03:51  lr: 0.000084  min_lr: 0.000084  loss: 3.1405 (2.9442)  weight_decay: 0.0500 (0.0500)  time: 0.6681  data: 0.0573  max mem: 41794
Epoch: [274]  [ 70/312]  eta: 0:03:41  lr: 0.000084  min_lr: 0.000084  loss: 3.1468 (2.9794)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.0505  max mem: 41794
Epoch: [274]  [ 80/312]  eta: 0:03:27  lr: 0.000084  min_lr: 0.000084  loss: 3.3271 (3.0108)  weight_decay: 0.0500 (0.0500)  time: 0.8276  data: 0.0516  max mem: 41794
Epoch: [274]  [ 90/312]  eta: 0:03:11  lr: 0.000084  min_lr: 0.000084  loss: 3.3108 (3.0118)  weight_decay: 0.0500 (0.0500)  time: 0.6772  data: 0.0563  max mem: 41794
Epoch: [274]  [100/312]  eta: 0:03:01  lr: 0.000083  min_lr: 0.000083  loss: 3.1487 (3.0149)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.0654  max mem: 41794
Epoch: [274]  [110/312]  eta: 0:02:47  lr: 0.000083  min_lr: 0.000083  loss: 3.1215 (3.0193)  weight_decay: 0.0500 (0.0500)  time: 0.6593  data: 0.0279  max mem: 41794
Epoch: [274]  [120/312]  eta: 0:02:40  lr: 0.000083  min_lr: 0.000083  loss: 3.1215 (3.0130)  weight_decay: 0.0500 (0.0500)  time: 0.7314  data: 0.0762  max mem: 41794
Epoch: [274]  [130/312]  eta: 0:02:31  lr: 0.000083  min_lr: 0.000083  loss: 2.9154 (2.9980)  weight_decay: 0.0500 (0.0500)  time: 0.8523  data: 0.0754  max mem: 41794
Epoch: [274]  [140/312]  eta: 0:02:19  lr: 0.000083  min_lr: 0.000083  loss: 3.1411 (3.0082)  weight_decay: 0.0500 (0.0500)  time: 0.6392  data: 0.0246  max mem: 41794
Epoch: [274]  [150/312]  eta: 0:02:12  lr: 0.000082  min_lr: 0.000082  loss: 3.1033 (3.0051)  weight_decay: 0.0500 (0.0500)  time: 0.7438  data: 0.0763  max mem: 41794
Epoch: [274]  [160/312]  eta: 0:02:03  lr: 0.000082  min_lr: 0.000082  loss: 3.0891 (3.0074)  weight_decay: 0.0500 (0.0500)  time: 0.8283  data: 0.0713  max mem: 41794
Epoch: [274]  [170/312]  eta: 0:01:54  lr: 0.000082  min_lr: 0.000082  loss: 3.1500 (3.0100)  weight_decay: 0.0500 (0.0500)  time: 0.6776  data: 0.0607  max mem: 41794
Epoch: [274]  [180/312]  eta: 0:01:46  lr: 0.000082  min_lr: 0.000082  loss: 3.1220 (3.0006)  weight_decay: 0.0500 (0.0500)  time: 0.7696  data: 0.1146  max mem: 41794
Epoch: [274]  [190/312]  eta: 0:01:36  lr: 0.000082  min_lr: 0.000082  loss: 2.9277 (2.9914)  weight_decay: 0.0500 (0.0500)  time: 0.6755  data: 0.0577  max mem: 41794
Epoch: [274]  [200/312]  eta: 0:01:29  lr: 0.000081  min_lr: 0.000081  loss: 2.9405 (2.9921)  weight_decay: 0.0500 (0.0500)  time: 0.6993  data: 0.0537  max mem: 41794
Epoch: [274]  [210/312]  eta: 0:01:21  lr: 0.000081  min_lr: 0.000081  loss: 3.1909 (2.9973)  weight_decay: 0.0500 (0.0500)  time: 0.8941  data: 0.1118  max mem: 41794
Epoch: [274]  [220/312]  eta: 0:01:12  lr: 0.000081  min_lr: 0.000081  loss: 3.1674 (2.9971)  weight_decay: 0.0500 (0.0500)  time: 0.6898  data: 0.0588  max mem: 41794
Epoch: [274]  [230/312]  eta: 0:01:04  lr: 0.000081  min_lr: 0.000081  loss: 3.1927 (3.0037)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.0437  max mem: 41794
Epoch: [274]  [240/312]  eta: 0:00:56  lr: 0.000081  min_lr: 0.000081  loss: 3.1566 (3.0039)  weight_decay: 0.0500 (0.0500)  time: 0.7857  data: 0.0457  max mem: 41794
Epoch: [274]  [250/312]  eta: 0:00:48  lr: 0.000080  min_lr: 0.000080  loss: 3.0783 (3.0016)  weight_decay: 0.0500 (0.0500)  time: 0.6535  data: 0.0357  max mem: 41794
Epoch: [274]  [260/312]  eta: 0:00:40  lr: 0.000080  min_lr: 0.000080  loss: 3.2535 (3.0061)  weight_decay: 0.0500 (0.0500)  time: 0.7526  data: 0.1032  max mem: 41794
Epoch: [274]  [270/312]  eta: 0:00:32  lr: 0.000080  min_lr: 0.000080  loss: 3.2535 (3.0113)  weight_decay: 0.0500 (0.0500)  time: 0.6985  data: 0.0717  max mem: 41794
Epoch: [274]  [280/312]  eta: 0:00:24  lr: 0.000080  min_lr: 0.000080  loss: 3.1405 (3.0127)  weight_decay: 0.0500 (0.0500)  time: 0.6972  data: 0.0555  max mem: 41794
Epoch: [274]  [290/312]  eta: 0:00:17  lr: 0.000080  min_lr: 0.000080  loss: 3.1064 (3.0140)  weight_decay: 0.0500 (0.0500)  time: 0.8917  data: 0.0645  max mem: 41794
Epoch: [274]  [300/312]  eta: 0:00:09  lr: 0.000079  min_lr: 0.000079  loss: 3.1689 (3.0109)  weight_decay: 0.0500 (0.0500)  time: 0.6786  data: 0.0109  max mem: 41794
Epoch: [274]  [310/312]  eta: 0:00:01  lr: 0.000079  min_lr: 0.000079  loss: 2.9966 (3.0076)  weight_decay: 0.0500 (0.0500)  time: 0.4654  data: 0.0001  max mem: 41794
Epoch: [274]  [311/312]  eta: 0:00:00  lr: 0.000079  min_lr: 0.000079  loss: 2.9966 (3.0083)  weight_decay: 0.0500 (0.0500)  time: 0.4650  data: 0.0001  max mem: 41794
Epoch: [274] Total time: 0:03:59 (0.7665 s / it)
Averaged stats: lr: 0.000079  min_lr: 0.000079  loss: 2.9966 (2.9949)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:02:08  loss: 0.8349 (0.8349)  acc1: 85.6771 (85.6771)  acc5: 96.4844 (96.4844)  time: 14.2874  data: 14.0750  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1053 (1.0249)  acc1: 78.9062 (79.7120)  acc5: 95.8333 (95.4080)  time: 1.7752  data: 1.6002  max mem: 41794
Test: Total time: 0:00:16 (1.7894 s / it)
* Acc@1 80.074 Acc@5 95.226 loss 1.021
Accuracy of the model on the 50000 test images: 80.1%
Max accuracy: 80.07%
Epoch: [275]  [  0/312]  eta: 1:39:08  lr: 0.000079  min_lr: 0.000079  loss: 2.0211 (2.0211)  weight_decay: 0.0500 (0.0500)  time: 19.0657  data: 17.1162  max mem: 41794
Epoch: [275]  [ 10/312]  eta: 0:11:33  lr: 0.000079  min_lr: 0.000079  loss: 2.6744 (2.6550)  weight_decay: 0.0500 (0.0500)  time: 2.2971  data: 1.6674  max mem: 41794
Epoch: [275]  [ 20/312]  eta: 0:09:30  lr: 0.000079  min_lr: 0.000079  loss: 3.1695 (2.9061)  weight_decay: 0.0500 (0.0500)  time: 1.0996  data: 0.4758  max mem: 41794
Epoch: [275]  [ 30/312]  eta: 0:07:02  lr: 0.000079  min_lr: 0.000079  loss: 3.2344 (2.9425)  weight_decay: 0.0500 (0.0500)  time: 1.0572  data: 0.4171  max mem: 41794
Epoch: [275]  [ 40/312]  eta: 0:06:09  lr: 0.000078  min_lr: 0.000078  loss: 3.1675 (2.9422)  weight_decay: 0.0500 (0.0500)  time: 0.7356  data: 0.0745  max mem: 41794
Epoch: [275]  [ 50/312]  eta: 0:05:29  lr: 0.000078  min_lr: 0.000078  loss: 3.1974 (3.0091)  weight_decay: 0.0500 (0.0500)  time: 0.8878  data: 0.0995  max mem: 41794
Epoch: [275]  [ 60/312]  eta: 0:04:46  lr: 0.000078  min_lr: 0.000078  loss: 3.2584 (3.0355)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.0321  max mem: 41794
Epoch: [275]  [ 70/312]  eta: 0:04:23  lr: 0.000078  min_lr: 0.000078  loss: 3.1932 (3.0261)  weight_decay: 0.0500 (0.0500)  time: 0.6526  data: 0.0145  max mem: 41794
Epoch: [275]  [ 80/312]  eta: 0:04:04  lr: 0.000078  min_lr: 0.000078  loss: 3.0679 (3.0156)  weight_decay: 0.0500 (0.0500)  time: 0.8016  data: 0.0122  max mem: 41794
Epoch: [275]  [ 90/312]  eta: 0:03:41  lr: 0.000077  min_lr: 0.000077  loss: 2.7749 (2.9894)  weight_decay: 0.0500 (0.0500)  time: 0.6837  data: 0.0187  max mem: 41794
Epoch: [275]  [100/312]  eta: 0:03:27  lr: 0.000077  min_lr: 0.000077  loss: 2.8261 (2.9721)  weight_decay: 0.0500 (0.0500)  time: 0.6628  data: 0.0303  max mem: 41794
Epoch: [275]  [110/312]  eta: 0:03:08  lr: 0.000077  min_lr: 0.000077  loss: 2.9254 (2.9768)  weight_decay: 0.0500 (0.0500)  time: 0.6302  data: 0.0142  max mem: 41794
Epoch: [275]  [120/312]  eta: 0:02:59  lr: 0.000077  min_lr: 0.000077  loss: 2.9824 (2.9740)  weight_decay: 0.0500 (0.0500)  time: 0.7156  data: 0.0167  max mem: 41794
Epoch: [275]  [130/312]  eta: 0:02:49  lr: 0.000077  min_lr: 0.000077  loss: 2.9576 (2.9589)  weight_decay: 0.0500 (0.0500)  time: 0.9125  data: 0.0559  max mem: 41794
Epoch: [275]  [140/312]  eta: 0:02:34  lr: 0.000076  min_lr: 0.000076  loss: 2.9618 (2.9597)  weight_decay: 0.0500 (0.0500)  time: 0.6783  data: 0.0398  max mem: 41794
Epoch: [275]  [150/312]  eta: 0:02:24  lr: 0.000076  min_lr: 0.000076  loss: 3.1722 (2.9677)  weight_decay: 0.0500 (0.0500)  time: 0.6515  data: 0.0208  max mem: 41794
Epoch: [275]  [160/312]  eta: 0:02:15  lr: 0.000076  min_lr: 0.000076  loss: 3.2678 (2.9748)  weight_decay: 0.0500 (0.0500)  time: 0.8652  data: 0.0208  max mem: 41794
Epoch: [275]  [170/312]  eta: 0:02:04  lr: 0.000076  min_lr: 0.000076  loss: 2.8270 (2.9712)  weight_decay: 0.0500 (0.0500)  time: 0.7355  data: 0.0377  max mem: 41794
Epoch: [275]  [180/312]  eta: 0:01:55  lr: 0.000076  min_lr: 0.000076  loss: 3.0751 (2.9830)  weight_decay: 0.0500 (0.0500)  time: 0.7079  data: 0.0678  max mem: 41794
Epoch: [275]  [190/312]  eta: 0:01:44  lr: 0.000075  min_lr: 0.000075  loss: 3.2249 (2.9873)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.0306  max mem: 41794
Epoch: [275]  [200/312]  eta: 0:01:35  lr: 0.000075  min_lr: 0.000075  loss: 3.2577 (2.9873)  weight_decay: 0.0500 (0.0500)  time: 0.6438  data: 0.0063  max mem: 41794
Epoch: [275]  [210/312]  eta: 0:01:29  lr: 0.000075  min_lr: 0.000075  loss: 3.2594 (2.9912)  weight_decay: 0.0500 (0.0500)  time: 1.1289  data: 0.3201  max mem: 41794
Epoch: [275]  [220/312]  eta: 0:01:19  lr: 0.000075  min_lr: 0.000075  loss: 3.1584 (2.9929)  weight_decay: 0.0500 (0.0500)  time: 0.9649  data: 0.3143  max mem: 41794
Epoch: [275]  [230/312]  eta: 0:01:10  lr: 0.000075  min_lr: 0.000075  loss: 3.1550 (2.9853)  weight_decay: 0.0500 (0.0500)  time: 0.6673  data: 0.0535  max mem: 41794
Epoch: [275]  [240/312]  eta: 0:01:01  lr: 0.000074  min_lr: 0.000074  loss: 3.1550 (2.9838)  weight_decay: 0.0500 (0.0500)  time: 0.8338  data: 0.0874  max mem: 41794
Epoch: [275]  [250/312]  eta: 0:00:52  lr: 0.000074  min_lr: 0.000074  loss: 2.8405 (2.9675)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.0553  max mem: 41794
Epoch: [275]  [260/312]  eta: 0:00:43  lr: 0.000074  min_lr: 0.000074  loss: 2.8726 (2.9661)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.0426  max mem: 41794
Epoch: [275]  [270/312]  eta: 0:00:34  lr: 0.000074  min_lr: 0.000074  loss: 3.1596 (2.9687)  weight_decay: 0.0500 (0.0500)  time: 0.6825  data: 0.0218  max mem: 41794
Epoch: [275]  [280/312]  eta: 0:00:26  lr: 0.000074  min_lr: 0.000074  loss: 2.9371 (2.9638)  weight_decay: 0.0500 (0.0500)  time: 0.6551  data: 0.0405  max mem: 41794
Epoch: [275]  [290/312]  eta: 0:00:18  lr: 0.000073  min_lr: 0.000073  loss: 2.9790 (2.9651)  weight_decay: 0.0500 (0.0500)  time: 0.7154  data: 0.0403  max mem: 41794
Epoch: [275]  [300/312]  eta: 0:00:09  lr: 0.000073  min_lr: 0.000073  loss: 3.0758 (2.9677)  weight_decay: 0.0500 (0.0500)  time: 0.5420  data: 0.0002  max mem: 41794
Epoch: [275]  [310/312]  eta: 0:00:01  lr: 0.000073  min_lr: 0.000073  loss: 3.1138 (2.9667)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [275]  [311/312]  eta: 0:00:00  lr: 0.000073  min_lr: 0.000073  loss: 3.0769 (2.9649)  weight_decay: 0.0500 (0.0500)  time: 0.4658  data: 0.0001  max mem: 41794
Epoch: [275] Total time: 0:04:11 (0.8068 s / it)
Averaged stats: lr: 0.000073  min_lr: 0.000073  loss: 3.0769 (2.9752)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:02:00  loss: 0.8185 (0.8185)  acc1: 86.0677 (86.0677)  acc5: 96.4844 (96.4844)  time: 13.3981  data: 13.1986  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0757 (1.0055)  acc1: 78.7760 (79.5680)  acc5: 95.5729 (95.3120)  time: 1.6428  data: 1.4666  max mem: 41794
Test: Total time: 0:00:14 (1.6520 s / it)
* Acc@1 80.206 Acc@5 95.190 loss 1.000
Accuracy of the model on the 50000 test images: 80.2%
Max accuracy: 80.21%
Epoch: [276]  [  0/312]  eta: 1:10:42  lr: 0.000073  min_lr: 0.000073  loss: 3.3779 (3.3779)  weight_decay: 0.0500 (0.0500)  time: 13.5983  data: 10.6770  max mem: 41794
Epoch: [276]  [ 10/312]  eta: 0:09:14  lr: 0.000073  min_lr: 0.000073  loss: 3.1337 (3.0847)  weight_decay: 0.0500 (0.0500)  time: 1.8347  data: 1.1001  max mem: 41794
Epoch: [276]  [ 20/312]  eta: 0:06:36  lr: 0.000073  min_lr: 0.000073  loss: 3.0851 (3.0442)  weight_decay: 0.0500 (0.0500)  time: 0.7441  data: 0.1650  max mem: 41794
Epoch: [276]  [ 30/312]  eta: 0:05:04  lr: 0.000072  min_lr: 0.000072  loss: 2.8357 (2.9990)  weight_decay: 0.0500 (0.0500)  time: 0.6654  data: 0.0942  max mem: 41794
Epoch: [276]  [ 40/312]  eta: 0:04:27  lr: 0.000072  min_lr: 0.000072  loss: 2.8970 (2.9778)  weight_decay: 0.0500 (0.0500)  time: 0.5917  data: 0.0500  max mem: 41794
Epoch: [276]  [ 50/312]  eta: 0:04:08  lr: 0.000072  min_lr: 0.000072  loss: 3.2445 (2.9713)  weight_decay: 0.0500 (0.0500)  time: 0.7442  data: 0.1733  max mem: 41794
Epoch: [276]  [ 60/312]  eta: 0:03:39  lr: 0.000072  min_lr: 0.000072  loss: 2.7188 (2.9274)  weight_decay: 0.0500 (0.0500)  time: 0.6371  data: 0.1239  max mem: 41794
Epoch: [276]  [ 70/312]  eta: 0:03:33  lr: 0.000072  min_lr: 0.000072  loss: 3.0604 (2.9561)  weight_decay: 0.0500 (0.0500)  time: 0.7112  data: 0.1372  max mem: 41794
Epoch: [276]  [ 80/312]  eta: 0:03:22  lr: 0.000072  min_lr: 0.000072  loss: 3.1443 (2.9478)  weight_decay: 0.0500 (0.0500)  time: 0.8850  data: 0.1959  max mem: 41794
Epoch: [276]  [ 90/312]  eta: 0:03:04  lr: 0.000071  min_lr: 0.000071  loss: 3.0635 (2.9684)  weight_decay: 0.0500 (0.0500)  time: 0.6510  data: 0.0593  max mem: 41794
Epoch: [276]  [100/312]  eta: 0:02:57  lr: 0.000071  min_lr: 0.000071  loss: 3.0774 (2.9655)  weight_decay: 0.0500 (0.0500)  time: 0.6851  data: 0.0620  max mem: 41794
Epoch: [276]  [110/312]  eta: 0:02:42  lr: 0.000071  min_lr: 0.000071  loss: 3.0325 (2.9519)  weight_decay: 0.0500 (0.0500)  time: 0.6848  data: 0.0620  max mem: 41794
Epoch: [276]  [120/312]  eta: 0:02:33  lr: 0.000071  min_lr: 0.000071  loss: 2.9056 (2.9387)  weight_decay: 0.0500 (0.0500)  time: 0.6165  data: 0.0406  max mem: 41794
Epoch: [276]  [130/312]  eta: 0:02:25  lr: 0.000071  min_lr: 0.000071  loss: 2.8631 (2.9389)  weight_decay: 0.0500 (0.0500)  time: 0.7849  data: 0.1127  max mem: 41794
Epoch: [276]  [140/312]  eta: 0:02:13  lr: 0.000070  min_lr: 0.000070  loss: 2.8631 (2.9299)  weight_decay: 0.0500 (0.0500)  time: 0.6510  data: 0.0727  max mem: 41794
Epoch: [276]  [150/312]  eta: 0:02:07  lr: 0.000070  min_lr: 0.000070  loss: 3.1548 (2.9521)  weight_decay: 0.0500 (0.0500)  time: 0.6984  data: 0.0572  max mem: 41794
Epoch: [276]  [160/312]  eta: 0:01:59  lr: 0.000070  min_lr: 0.000070  loss: 3.1132 (2.9310)  weight_decay: 0.0500 (0.0500)  time: 0.8629  data: 0.1045  max mem: 41794
Epoch: [276]  [170/312]  eta: 0:01:49  lr: 0.000070  min_lr: 0.000070  loss: 2.6935 (2.9245)  weight_decay: 0.0500 (0.0500)  time: 0.6737  data: 0.0491  max mem: 41794
Epoch: [276]  [180/312]  eta: 0:01:43  lr: 0.000070  min_lr: 0.000070  loss: 3.0309 (2.9272)  weight_decay: 0.0500 (0.0500)  time: 0.7248  data: 0.0635  max mem: 41794
Epoch: [276]  [190/312]  eta: 0:01:33  lr: 0.000069  min_lr: 0.000069  loss: 3.1547 (2.9351)  weight_decay: 0.0500 (0.0500)  time: 0.7060  data: 0.0661  max mem: 41794
Epoch: [276]  [200/312]  eta: 0:01:26  lr: 0.000069  min_lr: 0.000069  loss: 3.2019 (2.9513)  weight_decay: 0.0500 (0.0500)  time: 0.6778  data: 0.0363  max mem: 41794
Epoch: [276]  [210/312]  eta: 0:01:19  lr: 0.000069  min_lr: 0.000069  loss: 3.2185 (2.9553)  weight_decay: 0.0500 (0.0500)  time: 0.8631  data: 0.0650  max mem: 41794
Epoch: [276]  [220/312]  eta: 0:01:10  lr: 0.000069  min_lr: 0.000069  loss: 3.1259 (2.9576)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.0339  max mem: 41794
Epoch: [276]  [230/312]  eta: 0:01:03  lr: 0.000069  min_lr: 0.000069  loss: 3.0851 (2.9632)  weight_decay: 0.0500 (0.0500)  time: 0.6832  data: 0.0292  max mem: 41794
Epoch: [276]  [240/312]  eta: 0:00:55  lr: 0.000069  min_lr: 0.000069  loss: 3.0851 (2.9684)  weight_decay: 0.0500 (0.0500)  time: 0.8169  data: 0.0546  max mem: 41794
Epoch: [276]  [250/312]  eta: 0:00:47  lr: 0.000068  min_lr: 0.000068  loss: 2.9227 (2.9546)  weight_decay: 0.0500 (0.0500)  time: 0.6851  data: 0.0280  max mem: 41794
Epoch: [276]  [260/312]  eta: 0:00:39  lr: 0.000068  min_lr: 0.000068  loss: 2.7767 (2.9587)  weight_decay: 0.0500 (0.0500)  time: 0.7252  data: 0.0019  max mem: 41794
Epoch: [276]  [270/312]  eta: 0:00:31  lr: 0.000068  min_lr: 0.000068  loss: 3.2319 (2.9596)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.0037  max mem: 41794
Epoch: [276]  [280/312]  eta: 0:00:24  lr: 0.000068  min_lr: 0.000068  loss: 3.2308 (2.9635)  weight_decay: 0.0500 (0.0500)  time: 0.6651  data: 0.0235  max mem: 41794
Epoch: [276]  [290/312]  eta: 0:00:16  lr: 0.000068  min_lr: 0.000068  loss: 3.1381 (2.9614)  weight_decay: 0.0500 (0.0500)  time: 0.7551  data: 0.0203  max mem: 41794
Epoch: [276]  [300/312]  eta: 0:00:08  lr: 0.000067  min_lr: 0.000067  loss: 2.9006 (2.9597)  weight_decay: 0.0500 (0.0500)  time: 0.5846  data: 0.0002  max mem: 41794
Epoch: [276]  [310/312]  eta: 0:00:01  lr: 0.000067  min_lr: 0.000067  loss: 3.0553 (2.9668)  weight_decay: 0.0500 (0.0500)  time: 0.4632  data: 0.0001  max mem: 41794
Epoch: [276]  [311/312]  eta: 0:00:00  lr: 0.000067  min_lr: 0.000067  loss: 3.2010 (2.9681)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0001  max mem: 41794
Epoch: [276] Total time: 0:03:51 (0.7420 s / it)
Averaged stats: lr: 0.000067  min_lr: 0.000067  loss: 3.2010 (3.0001)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:03:43  loss: 0.9056 (0.9056)  acc1: 85.8073 (85.8073)  acc5: 96.4844 (96.4844)  time: 24.8798  data: 24.6687  max mem: 41794
Test:  [8/9]  eta: 0:00:02  loss: 1.1904 (1.1026)  acc1: 79.8177 (79.7600)  acc5: 95.1823 (95.1360)  time: 2.9125  data: 2.7410  max mem: 41794
Test: Total time: 0:00:26 (2.9407 s / it)
* Acc@1 80.118 Acc@5 95.198 loss 1.101
Accuracy of the model on the 50000 test images: 80.1%
Max accuracy: 80.21%
Epoch: [277]  [  0/312]  eta: 1:44:33  lr: 0.000067  min_lr: 0.000067  loss: 3.1940 (3.1940)  weight_decay: 0.0500 (0.0500)  time: 20.1059  data: 19.3929  max mem: 41794
Epoch: [277]  [ 10/312]  eta: 0:12:03  lr: 0.000067  min_lr: 0.000067  loss: 2.8527 (2.9381)  weight_decay: 0.0500 (0.0500)  time: 2.3971  data: 1.7634  max mem: 41794
Epoch: [277]  [ 20/312]  eta: 0:08:50  lr: 0.000067  min_lr: 0.000067  loss: 3.1422 (3.0804)  weight_decay: 0.0500 (0.0500)  time: 0.9019  data: 0.3574  max mem: 41794
Epoch: [277]  [ 30/312]  eta: 0:06:29  lr: 0.000067  min_lr: 0.000067  loss: 3.2083 (3.0127)  weight_decay: 0.0500 (0.0500)  time: 0.8206  data: 0.3574  max mem: 41794
Epoch: [277]  [ 40/312]  eta: 0:05:53  lr: 0.000066  min_lr: 0.000066  loss: 3.0406 (3.0131)  weight_decay: 0.0500 (0.0500)  time: 0.7574  data: 0.2853  max mem: 41794
Epoch: [277]  [ 50/312]  eta: 0:05:23  lr: 0.000066  min_lr: 0.000066  loss: 2.9664 (2.9847)  weight_decay: 0.0500 (0.0500)  time: 1.0086  data: 0.5358  max mem: 41794
Epoch: [277]  [ 60/312]  eta: 0:04:39  lr: 0.000066  min_lr: 0.000066  loss: 2.9934 (2.9933)  weight_decay: 0.0500 (0.0500)  time: 0.7138  data: 0.2508  max mem: 41794
Epoch: [277]  [ 70/312]  eta: 0:04:37  lr: 0.000066  min_lr: 0.000066  loss: 2.9089 (2.9468)  weight_decay: 0.0500 (0.0500)  time: 0.9290  data: 0.4578  max mem: 41794
Epoch: [277]  [ 80/312]  eta: 0:04:30  lr: 0.000066  min_lr: 0.000066  loss: 2.5880 (2.9092)  weight_decay: 0.0500 (0.0500)  time: 1.3522  data: 0.8775  max mem: 41794
Epoch: [277]  [ 90/312]  eta: 0:04:02  lr: 0.000066  min_lr: 0.000066  loss: 2.8561 (2.9202)  weight_decay: 0.0500 (0.0500)  time: 0.8934  data: 0.4201  max mem: 41794
Epoch: [277]  [100/312]  eta: 0:03:57  lr: 0.000065  min_lr: 0.000065  loss: 3.0874 (2.9282)  weight_decay: 0.0500 (0.0500)  time: 0.9229  data: 0.4522  max mem: 41794
Epoch: [277]  [110/312]  eta: 0:03:34  lr: 0.000065  min_lr: 0.000065  loss: 3.0602 (2.9213)  weight_decay: 0.0500 (0.0500)  time: 0.9145  data: 0.4521  max mem: 41794
Epoch: [277]  [120/312]  eta: 0:03:25  lr: 0.000065  min_lr: 0.000065  loss: 3.1287 (2.9411)  weight_decay: 0.0500 (0.0500)  time: 0.8183  data: 0.3531  max mem: 41794
Epoch: [277]  [130/312]  eta: 0:03:28  lr: 0.000065  min_lr: 0.000065  loss: 3.1666 (2.9235)  weight_decay: 0.0500 (0.0500)  time: 1.6062  data: 1.1403  max mem: 41794
Epoch: [277]  [140/312]  eta: 0:03:08  lr: 0.000065  min_lr: 0.000065  loss: 2.9545 (2.9189)  weight_decay: 0.0500 (0.0500)  time: 1.2498  data: 0.7876  max mem: 41794
Epoch: [277]  [150/312]  eta: 0:02:59  lr: 0.000064  min_lr: 0.000064  loss: 3.0689 (2.9238)  weight_decay: 0.0500 (0.0500)  time: 0.8837  data: 0.4219  max mem: 41794
Epoch: [277]  [160/312]  eta: 0:02:48  lr: 0.000064  min_lr: 0.000064  loss: 3.0689 (2.9328)  weight_decay: 0.0500 (0.0500)  time: 1.2160  data: 0.7523  max mem: 41794
Epoch: [277]  [170/312]  eta: 0:02:32  lr: 0.000064  min_lr: 0.000064  loss: 3.0376 (2.9249)  weight_decay: 0.0500 (0.0500)  time: 0.7941  data: 0.3308  max mem: 41794
Epoch: [277]  [180/312]  eta: 0:02:22  lr: 0.000064  min_lr: 0.000064  loss: 3.0367 (2.9276)  weight_decay: 0.0500 (0.0500)  time: 0.8303  data: 0.3670  max mem: 41794
Epoch: [277]  [190/312]  eta: 0:02:07  lr: 0.000064  min_lr: 0.000064  loss: 3.1394 (2.9312)  weight_decay: 0.0500 (0.0500)  time: 0.8316  data: 0.3670  max mem: 41794
Epoch: [277]  [200/312]  eta: 0:01:57  lr: 0.000064  min_lr: 0.000064  loss: 3.1452 (2.9363)  weight_decay: 0.0500 (0.0500)  time: 0.7807  data: 0.3176  max mem: 41794
Epoch: [277]  [210/312]  eta: 0:01:47  lr: 0.000063  min_lr: 0.000063  loss: 3.0333 (2.9286)  weight_decay: 0.0500 (0.0500)  time: 1.1658  data: 0.7028  max mem: 41794
Epoch: [277]  [220/312]  eta: 0:01:34  lr: 0.000063  min_lr: 0.000063  loss: 3.1918 (2.9464)  weight_decay: 0.0500 (0.0500)  time: 0.8487  data: 0.3856  max mem: 41794
Epoch: [277]  [230/312]  eta: 0:01:25  lr: 0.000063  min_lr: 0.000063  loss: 3.1246 (2.9336)  weight_decay: 0.0500 (0.0500)  time: 0.8439  data: 0.3773  max mem: 41794
Epoch: [277]  [240/312]  eta: 0:01:14  lr: 0.000063  min_lr: 0.000063  loss: 2.7605 (2.9351)  weight_decay: 0.0500 (0.0500)  time: 1.0872  data: 0.6056  max mem: 41794
Epoch: [277]  [250/312]  eta: 0:01:02  lr: 0.000063  min_lr: 0.000063  loss: 2.9923 (2.9329)  weight_decay: 0.0500 (0.0500)  time: 0.7181  data: 0.2288  max mem: 41794
Epoch: [277]  [260/312]  eta: 0:00:52  lr: 0.000063  min_lr: 0.000063  loss: 2.9923 (2.9355)  weight_decay: 0.0500 (0.0500)  time: 0.6651  data: 0.1800  max mem: 41794
Epoch: [277]  [270/312]  eta: 0:00:41  lr: 0.000062  min_lr: 0.000062  loss: 2.8364 (2.9281)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.1800  max mem: 41794
Epoch: [277]  [280/312]  eta: 0:00:31  lr: 0.000062  min_lr: 0.000062  loss: 2.8172 (2.9278)  weight_decay: 0.0500 (0.0500)  time: 0.7130  data: 0.2171  max mem: 41794
Epoch: [277]  [290/312]  eta: 0:00:21  lr: 0.000062  min_lr: 0.000062  loss: 2.9882 (2.9345)  weight_decay: 0.0500 (0.0500)  time: 0.7967  data: 0.3066  max mem: 41794
Epoch: [277]  [300/312]  eta: 0:00:11  lr: 0.000062  min_lr: 0.000062  loss: 3.2181 (2.9329)  weight_decay: 0.0500 (0.0500)  time: 0.5692  data: 0.0899  max mem: 41794
Epoch: [277]  [310/312]  eta: 0:00:01  lr: 0.000062  min_lr: 0.000062  loss: 3.1376 (2.9376)  weight_decay: 0.0500 (0.0500)  time: 0.4668  data: 0.0001  max mem: 41794
Epoch: [277]  [311/312]  eta: 0:00:00  lr: 0.000062  min_lr: 0.000062  loss: 3.0164 (2.9357)  weight_decay: 0.0500 (0.0500)  time: 0.4662  data: 0.0001  max mem: 41794
Epoch: [277] Total time: 0:04:55 (0.9471 s / it)
Averaged stats: lr: 0.000062  min_lr: 0.000062  loss: 3.0164 (2.9866)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 0.8407 (0.8407)  acc1: 85.5469 (85.5469)  acc5: 96.8750 (96.8750)  time: 8.3577  data: 8.1554  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1025 (1.0281)  acc1: 79.5573 (79.7440)  acc5: 95.5729 (95.4080)  time: 1.1143  data: 0.9438  max mem: 41794
Test: Total time: 0:00:10 (1.1249 s / it)
* Acc@1 80.190 Acc@5 95.224 loss 1.025
Accuracy of the model on the 50000 test images: 80.2%
Max accuracy: 80.21%
Epoch: [278]  [  0/312]  eta: 1:19:48  lr: 0.000062  min_lr: 0.000062  loss: 2.0274 (2.0274)  weight_decay: 0.0500 (0.0500)  time: 15.3475  data: 12.7870  max mem: 41794
Epoch: [278]  [ 10/312]  eta: 0:10:19  lr: 0.000061  min_lr: 0.000061  loss: 3.2609 (2.9931)  weight_decay: 0.0500 (0.0500)  time: 2.0515  data: 1.2757  max mem: 41794
Epoch: [278]  [ 20/312]  eta: 0:07:13  lr: 0.000061  min_lr: 0.000061  loss: 3.1712 (2.9446)  weight_decay: 0.0500 (0.0500)  time: 0.7905  data: 0.1323  max mem: 41794
Epoch: [278]  [ 30/312]  eta: 0:05:28  lr: 0.000061  min_lr: 0.000061  loss: 2.9278 (2.9478)  weight_decay: 0.0500 (0.0500)  time: 0.6795  data: 0.0746  max mem: 41794
Epoch: [278]  [ 40/312]  eta: 0:04:52  lr: 0.000061  min_lr: 0.000061  loss: 2.9278 (2.8974)  weight_decay: 0.0500 (0.0500)  time: 0.6476  data: 0.0906  max mem: 41794
Epoch: [278]  [ 50/312]  eta: 0:04:29  lr: 0.000061  min_lr: 0.000061  loss: 3.0982 (2.9461)  weight_decay: 0.0500 (0.0500)  time: 0.8143  data: 0.2122  max mem: 41794
Epoch: [278]  [ 60/312]  eta: 0:03:56  lr: 0.000061  min_lr: 0.000061  loss: 3.1950 (2.9728)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.1264  max mem: 41794
Epoch: [278]  [ 70/312]  eta: 0:03:43  lr: 0.000060  min_lr: 0.000060  loss: 2.9784 (2.9245)  weight_decay: 0.0500 (0.0500)  time: 0.6544  data: 0.1503  max mem: 41794
Epoch: [278]  [ 80/312]  eta: 0:03:30  lr: 0.000060  min_lr: 0.000060  loss: 2.9457 (2.9540)  weight_decay: 0.0500 (0.0500)  time: 0.8076  data: 0.3026  max mem: 41794
Epoch: [278]  [ 90/312]  eta: 0:03:11  lr: 0.000060  min_lr: 0.000060  loss: 3.1400 (2.9628)  weight_decay: 0.0500 (0.0500)  time: 0.6406  data: 0.1529  max mem: 41794
Epoch: [278]  [100/312]  eta: 0:03:03  lr: 0.000060  min_lr: 0.000060  loss: 3.1216 (2.9729)  weight_decay: 0.0500 (0.0500)  time: 0.6969  data: 0.1915  max mem: 41794
Epoch: [278]  [110/312]  eta: 0:02:47  lr: 0.000060  min_lr: 0.000060  loss: 3.2211 (2.9870)  weight_decay: 0.0500 (0.0500)  time: 0.6925  data: 0.1915  max mem: 41794
Epoch: [278]  [120/312]  eta: 0:02:41  lr: 0.000060  min_lr: 0.000060  loss: 2.8772 (2.9625)  weight_decay: 0.0500 (0.0500)  time: 0.7048  data: 0.2086  max mem: 41794
Epoch: [278]  [130/312]  eta: 0:02:33  lr: 0.000059  min_lr: 0.000059  loss: 2.8195 (2.9610)  weight_decay: 0.0500 (0.0500)  time: 0.9303  data: 0.3750  max mem: 41794
Epoch: [278]  [140/312]  eta: 0:02:21  lr: 0.000059  min_lr: 0.000059  loss: 3.1948 (2.9630)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.1671  max mem: 41794
Epoch: [278]  [150/312]  eta: 0:02:13  lr: 0.000059  min_lr: 0.000059  loss: 3.0132 (2.9692)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.1464  max mem: 41794
Epoch: [278]  [160/312]  eta: 0:02:05  lr: 0.000059  min_lr: 0.000059  loss: 3.1568 (2.9790)  weight_decay: 0.0500 (0.0500)  time: 0.8642  data: 0.3287  max mem: 41794
Epoch: [278]  [170/312]  eta: 0:01:54  lr: 0.000059  min_lr: 0.000059  loss: 3.2315 (2.9928)  weight_decay: 0.0500 (0.0500)  time: 0.6691  data: 0.1829  max mem: 41794
Epoch: [278]  [180/312]  eta: 0:01:46  lr: 0.000058  min_lr: 0.000058  loss: 3.1869 (2.9962)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.1618  max mem: 41794
Epoch: [278]  [190/312]  eta: 0:01:36  lr: 0.000058  min_lr: 0.000058  loss: 3.1856 (3.0036)  weight_decay: 0.0500 (0.0500)  time: 0.6901  data: 0.1619  max mem: 41794
Epoch: [278]  [200/312]  eta: 0:01:29  lr: 0.000058  min_lr: 0.000058  loss: 3.1856 (3.0110)  weight_decay: 0.0500 (0.0500)  time: 0.6882  data: 0.1733  max mem: 41794
Epoch: [278]  [210/312]  eta: 0:01:21  lr: 0.000058  min_lr: 0.000058  loss: 3.1992 (3.0075)  weight_decay: 0.0500 (0.0500)  time: 0.8656  data: 0.3427  max mem: 41794
Epoch: [278]  [220/312]  eta: 0:01:12  lr: 0.000058  min_lr: 0.000058  loss: 2.9118 (2.9989)  weight_decay: 0.0500 (0.0500)  time: 0.6730  data: 0.1701  max mem: 41794
Epoch: [278]  [230/312]  eta: 0:01:04  lr: 0.000058  min_lr: 0.000058  loss: 2.7534 (2.9943)  weight_decay: 0.0500 (0.0500)  time: 0.6990  data: 0.1958  max mem: 41794
Epoch: [278]  [240/312]  eta: 0:00:57  lr: 0.000057  min_lr: 0.000057  loss: 3.1503 (3.0033)  weight_decay: 0.0500 (0.0500)  time: 0.8695  data: 0.3626  max mem: 41794
Epoch: [278]  [250/312]  eta: 0:00:48  lr: 0.000057  min_lr: 0.000057  loss: 3.2120 (3.0112)  weight_decay: 0.0500 (0.0500)  time: 0.6678  data: 0.1674  max mem: 41794
Epoch: [278]  [260/312]  eta: 0:00:40  lr: 0.000057  min_lr: 0.000057  loss: 3.2276 (3.0109)  weight_decay: 0.0500 (0.0500)  time: 0.6916  data: 0.1896  max mem: 41794
Epoch: [278]  [270/312]  eta: 0:00:32  lr: 0.000057  min_lr: 0.000057  loss: 3.0056 (3.0061)  weight_decay: 0.0500 (0.0500)  time: 0.6822  data: 0.1895  max mem: 41794
Epoch: [278]  [280/312]  eta: 0:00:24  lr: 0.000057  min_lr: 0.000057  loss: 2.7884 (2.9939)  weight_decay: 0.0500 (0.0500)  time: 0.6821  data: 0.1945  max mem: 41794
Epoch: [278]  [290/312]  eta: 0:00:17  lr: 0.000057  min_lr: 0.000057  loss: 2.7884 (2.9895)  weight_decay: 0.0500 (0.0500)  time: 0.8256  data: 0.3329  max mem: 41794
Epoch: [278]  [300/312]  eta: 0:00:09  lr: 0.000056  min_lr: 0.000056  loss: 3.1121 (2.9956)  weight_decay: 0.0500 (0.0500)  time: 0.6208  data: 0.1389  max mem: 41794
Epoch: [278]  [310/312]  eta: 0:00:01  lr: 0.000056  min_lr: 0.000056  loss: 3.1625 (2.9965)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [278]  [311/312]  eta: 0:00:00  lr: 0.000056  min_lr: 0.000056  loss: 3.1839 (2.9977)  weight_decay: 0.0500 (0.0500)  time: 0.4637  data: 0.0001  max mem: 41794
Epoch: [278] Total time: 0:03:58 (0.7634 s / it)
Averaged stats: lr: 0.000056  min_lr: 0.000056  loss: 3.1839 (2.9959)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.8802 (0.8802)  acc1: 85.9375 (85.9375)  acc5: 96.6146 (96.6146)  time: 8.5797  data: 8.3667  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1574 (1.0830)  acc1: 78.7760 (79.6640)  acc5: 95.5729 (95.2960)  time: 1.1143  data: 0.9427  max mem: 41794
Test: Total time: 0:00:10 (1.1255 s / it)
* Acc@1 80.156 Acc@5 95.234 loss 1.082
Accuracy of the model on the 50000 test images: 80.2%
Max accuracy: 80.21%
Epoch: [279]  [  0/312]  eta: 1:22:03  lr: 0.000056  min_lr: 0.000056  loss: 2.7166 (2.7166)  weight_decay: 0.0500 (0.0500)  time: 15.7804  data: 14.1820  max mem: 41794
Epoch: [279]  [ 10/312]  eta: 0:10:00  lr: 0.000056  min_lr: 0.000056  loss: 2.7166 (2.6906)  weight_decay: 0.0500 (0.0500)  time: 1.9869  data: 1.3893  max mem: 41794
Epoch: [279]  [ 20/312]  eta: 0:07:10  lr: 0.000056  min_lr: 0.000056  loss: 2.9752 (2.8847)  weight_decay: 0.0500 (0.0500)  time: 0.7587  data: 0.1809  max mem: 41794
Epoch: [279]  [ 30/312]  eta: 0:05:29  lr: 0.000056  min_lr: 0.000056  loss: 3.0217 (2.8928)  weight_decay: 0.0500 (0.0500)  time: 0.7193  data: 0.1263  max mem: 41794
Epoch: [279]  [ 40/312]  eta: 0:05:00  lr: 0.000056  min_lr: 0.000056  loss: 2.9309 (2.8928)  weight_decay: 0.0500 (0.0500)  time: 0.7185  data: 0.1595  max mem: 41794
Epoch: [279]  [ 50/312]  eta: 0:04:40  lr: 0.000055  min_lr: 0.000055  loss: 2.9412 (2.8962)  weight_decay: 0.0500 (0.0500)  time: 0.9147  data: 0.2696  max mem: 41794
Epoch: [279]  [ 60/312]  eta: 0:04:06  lr: 0.000055  min_lr: 0.000055  loss: 2.8788 (2.8896)  weight_decay: 0.0500 (0.0500)  time: 0.7151  data: 0.1108  max mem: 41794
Epoch: [279]  [ 70/312]  eta: 0:03:50  lr: 0.000055  min_lr: 0.000055  loss: 3.1405 (2.9299)  weight_decay: 0.0500 (0.0500)  time: 0.6562  data: 0.0689  max mem: 41794
Epoch: [279]  [ 80/312]  eta: 0:03:35  lr: 0.000055  min_lr: 0.000055  loss: 3.3244 (2.9380)  weight_decay: 0.0500 (0.0500)  time: 0.7768  data: 0.0741  max mem: 41794
Epoch: [279]  [ 90/312]  eta: 0:03:17  lr: 0.000055  min_lr: 0.000055  loss: 2.8980 (2.9124)  weight_decay: 0.0500 (0.0500)  time: 0.6682  data: 0.0595  max mem: 41794
Epoch: [279]  [100/312]  eta: 0:03:07  lr: 0.000055  min_lr: 0.000055  loss: 2.8635 (2.9145)  weight_decay: 0.0500 (0.0500)  time: 0.7093  data: 0.1116  max mem: 41794
Epoch: [279]  [110/312]  eta: 0:02:51  lr: 0.000054  min_lr: 0.000054  loss: 2.9685 (2.9188)  weight_decay: 0.0500 (0.0500)  time: 0.6641  data: 0.0612  max mem: 41794
Epoch: [279]  [120/312]  eta: 0:02:44  lr: 0.000054  min_lr: 0.000054  loss: 2.9970 (2.9131)  weight_decay: 0.0500 (0.0500)  time: 0.7144  data: 0.0868  max mem: 41794
Epoch: [279]  [130/312]  eta: 0:02:36  lr: 0.000054  min_lr: 0.000054  loss: 3.0735 (2.9270)  weight_decay: 0.0500 (0.0500)  time: 0.9246  data: 0.1812  max mem: 41794
Epoch: [279]  [140/312]  eta: 0:02:23  lr: 0.000054  min_lr: 0.000054  loss: 3.1838 (2.9482)  weight_decay: 0.0500 (0.0500)  time: 0.7073  data: 0.0982  max mem: 41794
Epoch: [279]  [150/312]  eta: 0:02:14  lr: 0.000054  min_lr: 0.000054  loss: 3.1957 (2.9573)  weight_decay: 0.0500 (0.0500)  time: 0.6461  data: 0.0505  max mem: 41794
Epoch: [279]  [160/312]  eta: 0:02:06  lr: 0.000054  min_lr: 0.000054  loss: 2.9151 (2.9458)  weight_decay: 0.0500 (0.0500)  time: 0.7940  data: 0.0584  max mem: 41794
Epoch: [279]  [170/312]  eta: 0:01:56  lr: 0.000053  min_lr: 0.000053  loss: 2.7803 (2.9449)  weight_decay: 0.0500 (0.0500)  time: 0.7175  data: 0.0831  max mem: 41794
Epoch: [279]  [180/312]  eta: 0:01:49  lr: 0.000053  min_lr: 0.000053  loss: 2.9959 (2.9473)  weight_decay: 0.0500 (0.0500)  time: 0.7998  data: 0.0819  max mem: 41794
Epoch: [279]  [190/312]  eta: 0:01:39  lr: 0.000053  min_lr: 0.000053  loss: 3.0187 (2.9541)  weight_decay: 0.0500 (0.0500)  time: 0.7638  data: 0.0264  max mem: 41794
Epoch: [279]  [200/312]  eta: 0:01:31  lr: 0.000053  min_lr: 0.000053  loss: 3.2525 (2.9623)  weight_decay: 0.0500 (0.0500)  time: 0.6967  data: 0.0386  max mem: 41794
Epoch: [279]  [210/312]  eta: 0:01:22  lr: 0.000053  min_lr: 0.000053  loss: 3.2340 (2.9693)  weight_decay: 0.0500 (0.0500)  time: 0.7737  data: 0.0196  max mem: 41794
Epoch: [279]  [220/312]  eta: 0:01:13  lr: 0.000053  min_lr: 0.000053  loss: 3.2002 (2.9811)  weight_decay: 0.0500 (0.0500)  time: 0.6148  data: 0.0058  max mem: 41794
Epoch: [279]  [230/312]  eta: 0:01:05  lr: 0.000052  min_lr: 0.000052  loss: 3.2009 (2.9732)  weight_decay: 0.0500 (0.0500)  time: 0.6827  data: 0.0629  max mem: 41794
Epoch: [279]  [240/312]  eta: 0:00:57  lr: 0.000052  min_lr: 0.000052  loss: 3.2009 (2.9737)  weight_decay: 0.0500 (0.0500)  time: 0.7623  data: 0.0944  max mem: 41794
Epoch: [279]  [250/312]  eta: 0:00:48  lr: 0.000052  min_lr: 0.000052  loss: 3.1332 (2.9708)  weight_decay: 0.0500 (0.0500)  time: 0.6873  data: 0.0613  max mem: 41794
Epoch: [279]  [260/312]  eta: 0:00:41  lr: 0.000052  min_lr: 0.000052  loss: 3.1262 (2.9774)  weight_decay: 0.0500 (0.0500)  time: 0.8083  data: 0.0802  max mem: 41794
Epoch: [279]  [270/312]  eta: 0:00:32  lr: 0.000052  min_lr: 0.000052  loss: 3.1522 (2.9787)  weight_decay: 0.0500 (0.0500)  time: 0.7141  data: 0.0561  max mem: 41794
Epoch: [279]  [280/312]  eta: 0:00:25  lr: 0.000052  min_lr: 0.000052  loss: 3.0504 (2.9817)  weight_decay: 0.0500 (0.0500)  time: 0.7008  data: 0.0740  max mem: 41794
Epoch: [279]  [290/312]  eta: 0:00:17  lr: 0.000051  min_lr: 0.000051  loss: 3.0826 (2.9843)  weight_decay: 0.0500 (0.0500)  time: 0.8063  data: 0.0996  max mem: 41794
Epoch: [279]  [300/312]  eta: 0:00:09  lr: 0.000051  min_lr: 0.000051  loss: 3.1217 (2.9811)  weight_decay: 0.0500 (0.0500)  time: 0.5985  data: 0.0260  max mem: 41794
Epoch: [279]  [310/312]  eta: 0:00:01  lr: 0.000051  min_lr: 0.000051  loss: 3.1217 (2.9864)  weight_decay: 0.0500 (0.0500)  time: 0.4628  data: 0.0001  max mem: 41794
Epoch: [279]  [311/312]  eta: 0:00:00  lr: 0.000051  min_lr: 0.000051  loss: 3.0977 (2.9851)  weight_decay: 0.0500 (0.0500)  time: 0.4623  data: 0.0001  max mem: 41794
Epoch: [279] Total time: 0:04:00 (0.7710 s / it)
Averaged stats: lr: 0.000051  min_lr: 0.000051  loss: 3.0977 (2.9846)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:13  loss: 0.8779 (0.8779)  acc1: 85.0260 (85.0260)  acc5: 96.7448 (96.7448)  time: 8.1190  data: 7.9073  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1537 (1.0716)  acc1: 79.0365 (79.5840)  acc5: 95.5729 (95.4080)  time: 1.1038  data: 0.9323  max mem: 41794
Test: Total time: 0:00:10 (1.1231 s / it)
* Acc@1 80.138 Acc@5 95.248 loss 1.068
Accuracy of the model on the 50000 test images: 80.1%
Max accuracy: 80.21%
Epoch: [280]  [  0/312]  eta: 1:13:36  lr: 0.000051  min_lr: 0.000051  loss: 3.5844 (3.5844)  weight_decay: 0.0500 (0.0500)  time: 14.1547  data: 11.9217  max mem: 41794
Epoch: [280]  [ 10/312]  eta: 0:10:46  lr: 0.000051  min_lr: 0.000051  loss: 3.3271 (3.1418)  weight_decay: 0.0500 (0.0500)  time: 2.1408  data: 1.3608  max mem: 41794
Epoch: [280]  [ 20/312]  eta: 0:07:55  lr: 0.000051  min_lr: 0.000051  loss: 3.0747 (3.0309)  weight_decay: 0.0500 (0.0500)  time: 1.0032  data: 0.2368  max mem: 41794
Epoch: [280]  [ 30/312]  eta: 0:06:03  lr: 0.000051  min_lr: 0.000051  loss: 3.0012 (3.0214)  weight_decay: 0.0500 (0.0500)  time: 0.8182  data: 0.0891  max mem: 41794
Epoch: [280]  [ 40/312]  eta: 0:04:59  lr: 0.000050  min_lr: 0.000050  loss: 3.0012 (2.9891)  weight_decay: 0.0500 (0.0500)  time: 0.5432  data: 0.0138  max mem: 41794
Epoch: [280]  [ 50/312]  eta: 0:04:22  lr: 0.000050  min_lr: 0.000050  loss: 3.0133 (2.9681)  weight_decay: 0.0500 (0.0500)  time: 0.5604  data: 0.0254  max mem: 41794
Epoch: [280]  [ 60/312]  eta: 0:03:51  lr: 0.000050  min_lr: 0.000050  loss: 3.0190 (2.9710)  weight_decay: 0.0500 (0.0500)  time: 0.5472  data: 0.0188  max mem: 41794
Epoch: [280]  [ 70/312]  eta: 0:03:40  lr: 0.000050  min_lr: 0.000050  loss: 3.0600 (2.9749)  weight_decay: 0.0500 (0.0500)  time: 0.6790  data: 0.0583  max mem: 41794
Epoch: [280]  [ 80/312]  eta: 0:03:25  lr: 0.000050  min_lr: 0.000050  loss: 3.1225 (2.9448)  weight_decay: 0.0500 (0.0500)  time: 0.7836  data: 0.0621  max mem: 41794
Epoch: [280]  [ 90/312]  eta: 0:03:12  lr: 0.000050  min_lr: 0.000050  loss: 2.9982 (2.9660)  weight_decay: 0.0500 (0.0500)  time: 0.7061  data: 0.0631  max mem: 41794
Epoch: [280]  [100/312]  eta: 0:03:02  lr: 0.000050  min_lr: 0.000050  loss: 3.1284 (2.9760)  weight_decay: 0.0500 (0.0500)  time: 0.7639  data: 0.0937  max mem: 41794
Epoch: [280]  [110/312]  eta: 0:02:47  lr: 0.000049  min_lr: 0.000049  loss: 2.9523 (2.9587)  weight_decay: 0.0500 (0.0500)  time: 0.6590  data: 0.0373  max mem: 41794
Epoch: [280]  [120/312]  eta: 0:02:39  lr: 0.000049  min_lr: 0.000049  loss: 2.9250 (2.9529)  weight_decay: 0.0500 (0.0500)  time: 0.6717  data: 0.0414  max mem: 41794
Epoch: [280]  [130/312]  eta: 0:02:30  lr: 0.000049  min_lr: 0.000049  loss: 3.1150 (2.9607)  weight_decay: 0.0500 (0.0500)  time: 0.8063  data: 0.0627  max mem: 41794
Epoch: [280]  [140/312]  eta: 0:02:18  lr: 0.000049  min_lr: 0.000049  loss: 3.2617 (2.9717)  weight_decay: 0.0500 (0.0500)  time: 0.6409  data: 0.0233  max mem: 41794
Epoch: [280]  [150/312]  eta: 0:02:11  lr: 0.000049  min_lr: 0.000049  loss: 2.9564 (2.9467)  weight_decay: 0.0500 (0.0500)  time: 0.7188  data: 0.0762  max mem: 41794
Epoch: [280]  [160/312]  eta: 0:02:01  lr: 0.000049  min_lr: 0.000049  loss: 2.6962 (2.9477)  weight_decay: 0.0500 (0.0500)  time: 0.7933  data: 0.1148  max mem: 41794
Epoch: [280]  [170/312]  eta: 0:01:52  lr: 0.000048  min_lr: 0.000048  loss: 3.1878 (2.9607)  weight_decay: 0.0500 (0.0500)  time: 0.6315  data: 0.0586  max mem: 41794
Epoch: [280]  [180/312]  eta: 0:01:44  lr: 0.000048  min_lr: 0.000048  loss: 3.0964 (2.9600)  weight_decay: 0.0500 (0.0500)  time: 0.6794  data: 0.1471  max mem: 41794
Epoch: [280]  [190/312]  eta: 0:01:34  lr: 0.000048  min_lr: 0.000048  loss: 3.0964 (2.9617)  weight_decay: 0.0500 (0.0500)  time: 0.6319  data: 0.1289  max mem: 41794
Epoch: [280]  [200/312]  eta: 0:01:27  lr: 0.000048  min_lr: 0.000048  loss: 3.0073 (2.9621)  weight_decay: 0.0500 (0.0500)  time: 0.6852  data: 0.1847  max mem: 41794
Epoch: [280]  [210/312]  eta: 0:01:19  lr: 0.000048  min_lr: 0.000048  loss: 3.0262 (2.9690)  weight_decay: 0.0500 (0.0500)  time: 0.8750  data: 0.3800  max mem: 41794
Epoch: [280]  [220/312]  eta: 0:01:10  lr: 0.000048  min_lr: 0.000048  loss: 3.1725 (2.9785)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.1959  max mem: 41794
Epoch: [280]  [230/312]  eta: 0:01:03  lr: 0.000048  min_lr: 0.000048  loss: 3.1770 (2.9768)  weight_decay: 0.0500 (0.0500)  time: 0.6854  data: 0.1821  max mem: 41794
Epoch: [280]  [240/312]  eta: 0:00:55  lr: 0.000047  min_lr: 0.000047  loss: 3.0371 (2.9738)  weight_decay: 0.0500 (0.0500)  time: 0.8484  data: 0.3452  max mem: 41794
Epoch: [280]  [250/312]  eta: 0:00:47  lr: 0.000047  min_lr: 0.000047  loss: 3.0862 (2.9787)  weight_decay: 0.0500 (0.0500)  time: 0.6623  data: 0.1649  max mem: 41794
Epoch: [280]  [260/312]  eta: 0:00:40  lr: 0.000047  min_lr: 0.000047  loss: 3.0474 (2.9701)  weight_decay: 0.0500 (0.0500)  time: 0.6820  data: 0.1874  max mem: 41794
Epoch: [280]  [270/312]  eta: 0:00:31  lr: 0.000047  min_lr: 0.000047  loss: 3.1291 (2.9786)  weight_decay: 0.0500 (0.0500)  time: 0.6799  data: 0.1880  max mem: 41794
Epoch: [280]  [280/312]  eta: 0:00:24  lr: 0.000047  min_lr: 0.000047  loss: 3.1291 (2.9738)  weight_decay: 0.0500 (0.0500)  time: 0.6891  data: 0.1921  max mem: 41794
Epoch: [280]  [290/312]  eta: 0:00:16  lr: 0.000047  min_lr: 0.000047  loss: 2.9179 (2.9719)  weight_decay: 0.0500 (0.0500)  time: 0.9005  data: 0.4014  max mem: 41794
Epoch: [280]  [300/312]  eta: 0:00:09  lr: 0.000046  min_lr: 0.000046  loss: 3.0804 (2.9720)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.2116  max mem: 41794
Epoch: [280]  [310/312]  eta: 0:00:01  lr: 0.000046  min_lr: 0.000046  loss: 3.0570 (2.9685)  weight_decay: 0.0500 (0.0500)  time: 0.4700  data: 0.0002  max mem: 41794
Epoch: [280]  [311/312]  eta: 0:00:00  lr: 0.000046  min_lr: 0.000046  loss: 3.0570 (2.9694)  weight_decay: 0.0500 (0.0500)  time: 0.4682  data: 0.0002  max mem: 41794
Epoch: [280] Total time: 0:03:56 (0.7570 s / it)
Averaged stats: lr: 0.000046  min_lr: 0.000046  loss: 3.0570 (2.9874)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:22  loss: 0.8255 (0.8255)  acc1: 86.3281 (86.3281)  acc5: 96.6146 (96.6146)  time: 9.1647  data: 8.9521  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1028 (1.0231)  acc1: 79.1667 (79.9040)  acc5: 95.8333 (95.3760)  time: 1.1668  data: 0.9948  max mem: 41794
Test: Total time: 0:00:10 (1.1817 s / it)
* Acc@1 80.244 Acc@5 95.234 loss 1.018
Accuracy of the model on the 50000 test images: 80.2%
Max accuracy: 80.24%
Epoch: [281]  [  0/312]  eta: 1:13:26  lr: 0.000046  min_lr: 0.000046  loss: 2.9551 (2.9551)  weight_decay: 0.0500 (0.0500)  time: 14.1227  data: 11.7355  max mem: 41794
Epoch: [281]  [ 10/312]  eta: 0:10:39  lr: 0.000046  min_lr: 0.000046  loss: 2.7762 (2.8051)  weight_decay: 0.0500 (0.0500)  time: 2.1171  data: 1.2535  max mem: 41794
Epoch: [281]  [ 20/312]  eta: 0:07:25  lr: 0.000046  min_lr: 0.000046  loss: 2.9502 (2.8774)  weight_decay: 0.0500 (0.0500)  time: 0.8963  data: 0.1699  max mem: 41794
Epoch: [281]  [ 30/312]  eta: 0:05:40  lr: 0.000046  min_lr: 0.000046  loss: 3.2098 (2.9464)  weight_decay: 0.0500 (0.0500)  time: 0.7056  data: 0.0691  max mem: 41794
Epoch: [281]  [ 40/312]  eta: 0:04:44  lr: 0.000046  min_lr: 0.000046  loss: 3.1901 (2.9454)  weight_decay: 0.0500 (0.0500)  time: 0.5452  data: 0.0181  max mem: 41794
Epoch: [281]  [ 50/312]  eta: 0:04:14  lr: 0.000046  min_lr: 0.000046  loss: 3.1262 (2.9475)  weight_decay: 0.0500 (0.0500)  time: 0.6075  data: 0.0556  max mem: 41794
Epoch: [281]  [ 60/312]  eta: 0:03:50  lr: 0.000045  min_lr: 0.000045  loss: 3.0745 (2.9579)  weight_decay: 0.0500 (0.0500)  time: 0.6399  data: 0.1055  max mem: 41794
Epoch: [281]  [ 70/312]  eta: 0:03:40  lr: 0.000045  min_lr: 0.000045  loss: 2.8030 (2.9499)  weight_decay: 0.0500 (0.0500)  time: 0.7599  data: 0.1787  max mem: 41794
Epoch: [281]  [ 80/312]  eta: 0:03:25  lr: 0.000045  min_lr: 0.000045  loss: 3.0362 (2.9674)  weight_decay: 0.0500 (0.0500)  time: 0.8053  data: 0.1202  max mem: 41794
Epoch: [281]  [ 90/312]  eta: 0:03:12  lr: 0.000045  min_lr: 0.000045  loss: 3.2368 (2.9896)  weight_decay: 0.0500 (0.0500)  time: 0.7147  data: 0.1245  max mem: 41794
Epoch: [281]  [100/312]  eta: 0:03:04  lr: 0.000045  min_lr: 0.000045  loss: 3.2368 (2.9897)  weight_decay: 0.0500 (0.0500)  time: 0.8044  data: 0.2168  max mem: 41794
Epoch: [281]  [110/312]  eta: 0:02:49  lr: 0.000045  min_lr: 0.000045  loss: 3.0857 (2.9832)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.1042  max mem: 41794
Epoch: [281]  [120/312]  eta: 0:02:41  lr: 0.000044  min_lr: 0.000044  loss: 3.2554 (2.9823)  weight_decay: 0.0500 (0.0500)  time: 0.6983  data: 0.0871  max mem: 41794
Epoch: [281]  [130/312]  eta: 0:02:33  lr: 0.000044  min_lr: 0.000044  loss: 3.1836 (2.9915)  weight_decay: 0.0500 (0.0500)  time: 0.8720  data: 0.1639  max mem: 41794
Epoch: [281]  [140/312]  eta: 0:02:20  lr: 0.000044  min_lr: 0.000044  loss: 3.1342 (2.9917)  weight_decay: 0.0500 (0.0500)  time: 0.6707  data: 0.0812  max mem: 41794
Epoch: [281]  [150/312]  eta: 0:02:13  lr: 0.000044  min_lr: 0.000044  loss: 2.9213 (2.9790)  weight_decay: 0.0500 (0.0500)  time: 0.7119  data: 0.0956  max mem: 41794
Epoch: [281]  [160/312]  eta: 0:02:04  lr: 0.000044  min_lr: 0.000044  loss: 2.9620 (2.9790)  weight_decay: 0.0500 (0.0500)  time: 0.8331  data: 0.0956  max mem: 41794
Epoch: [281]  [170/312]  eta: 0:01:55  lr: 0.000044  min_lr: 0.000044  loss: 2.9620 (2.9739)  weight_decay: 0.0500 (0.0500)  time: 0.7279  data: 0.1162  max mem: 41794
Epoch: [281]  [180/312]  eta: 0:01:47  lr: 0.000044  min_lr: 0.000044  loss: 3.0776 (2.9781)  weight_decay: 0.0500 (0.0500)  time: 0.7434  data: 0.2036  max mem: 41794
Epoch: [281]  [190/312]  eta: 0:01:37  lr: 0.000043  min_lr: 0.000043  loss: 3.0554 (2.9749)  weight_decay: 0.0500 (0.0500)  time: 0.6300  data: 0.0880  max mem: 41794
Epoch: [281]  [200/312]  eta: 0:01:29  lr: 0.000043  min_lr: 0.000043  loss: 3.1033 (2.9793)  weight_decay: 0.0500 (0.0500)  time: 0.6929  data: 0.1037  max mem: 41794
Epoch: [281]  [210/312]  eta: 0:01:21  lr: 0.000043  min_lr: 0.000043  loss: 3.2623 (2.9887)  weight_decay: 0.0500 (0.0500)  time: 0.8742  data: 0.2048  max mem: 41794
Epoch: [281]  [220/312]  eta: 0:01:12  lr: 0.000043  min_lr: 0.000043  loss: 3.2623 (2.9926)  weight_decay: 0.0500 (0.0500)  time: 0.6802  data: 0.1017  max mem: 41794
Epoch: [281]  [230/312]  eta: 0:01:05  lr: 0.000043  min_lr: 0.000043  loss: 3.2482 (2.9898)  weight_decay: 0.0500 (0.0500)  time: 0.7088  data: 0.1110  max mem: 41794
Epoch: [281]  [240/312]  eta: 0:00:56  lr: 0.000043  min_lr: 0.000043  loss: 2.7447 (2.9767)  weight_decay: 0.0500 (0.0500)  time: 0.7949  data: 0.1110  max mem: 41794
Epoch: [281]  [250/312]  eta: 0:00:48  lr: 0.000043  min_lr: 0.000043  loss: 2.9093 (2.9776)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.0831  max mem: 41794
Epoch: [281]  [260/312]  eta: 0:00:40  lr: 0.000042  min_lr: 0.000042  loss: 3.0834 (2.9798)  weight_decay: 0.0500 (0.0500)  time: 0.7690  data: 0.1750  max mem: 41794
Epoch: [281]  [270/312]  eta: 0:00:32  lr: 0.000042  min_lr: 0.000042  loss: 3.1018 (2.9779)  weight_decay: 0.0500 (0.0500)  time: 0.6916  data: 0.0926  max mem: 41794
Epoch: [281]  [280/312]  eta: 0:00:24  lr: 0.000042  min_lr: 0.000042  loss: 3.0486 (2.9801)  weight_decay: 0.0500 (0.0500)  time: 0.6461  data: 0.0737  max mem: 41794
Epoch: [281]  [290/312]  eta: 0:00:17  lr: 0.000042  min_lr: 0.000042  loss: 3.1583 (2.9861)  weight_decay: 0.0500 (0.0500)  time: 0.8200  data: 0.1635  max mem: 41794
Epoch: [281]  [300/312]  eta: 0:00:09  lr: 0.000042  min_lr: 0.000042  loss: 3.1079 (2.9768)  weight_decay: 0.0500 (0.0500)  time: 0.6593  data: 0.0902  max mem: 41794
Epoch: [281]  [310/312]  eta: 0:00:01  lr: 0.000042  min_lr: 0.000042  loss: 3.1747 (2.9871)  weight_decay: 0.0500 (0.0500)  time: 0.4634  data: 0.0001  max mem: 41794
Epoch: [281]  [311/312]  eta: 0:00:00  lr: 0.000042  min_lr: 0.000042  loss: 3.1990 (2.9881)  weight_decay: 0.0500 (0.0500)  time: 0.4636  data: 0.0001  max mem: 41794
Epoch: [281] Total time: 0:03:58 (0.7655 s / it)
Averaged stats: lr: 0.000042  min_lr: 0.000042  loss: 3.1990 (2.9749)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:17  loss: 0.9390 (0.9390)  acc1: 85.9375 (85.9375)  acc5: 96.7448 (96.7448)  time: 8.6531  data: 8.4408  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.2132 (1.1268)  acc1: 78.7760 (79.5040)  acc5: 95.3125 (95.3120)  time: 1.1096  data: 0.9379  max mem: 41794
Test: Total time: 0:00:10 (1.1510 s / it)
* Acc@1 80.256 Acc@5 95.268 loss 1.121
Accuracy of the model on the 50000 test images: 80.3%
Max accuracy: 80.26%
Epoch: [282]  [  0/312]  eta: 1:19:02  lr: 0.000042  min_lr: 0.000042  loss: 3.3106 (3.3106)  weight_decay: 0.0500 (0.0500)  time: 15.2019  data: 13.4651  max mem: 41794
Epoch: [282]  [ 10/312]  eta: 0:10:17  lr: 0.000041  min_lr: 0.000041  loss: 3.2381 (3.0985)  weight_decay: 0.0500 (0.0500)  time: 2.0454  data: 1.3757  max mem: 41794
Epoch: [282]  [ 20/312]  eta: 0:07:11  lr: 0.000041  min_lr: 0.000041  loss: 3.1010 (2.9549)  weight_decay: 0.0500 (0.0500)  time: 0.7930  data: 0.2021  max mem: 41794
Epoch: [282]  [ 30/312]  eta: 0:05:29  lr: 0.000041  min_lr: 0.000041  loss: 3.0168 (2.9520)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.1191  max mem: 41794
Epoch: [282]  [ 40/312]  eta: 0:04:51  lr: 0.000041  min_lr: 0.000041  loss: 2.9758 (2.8966)  weight_decay: 0.0500 (0.0500)  time: 0.6474  data: 0.0951  max mem: 41794
Epoch: [282]  [ 50/312]  eta: 0:04:32  lr: 0.000041  min_lr: 0.000041  loss: 2.8941 (2.8804)  weight_decay: 0.0500 (0.0500)  time: 0.8449  data: 0.2111  max mem: 41794
Epoch: [282]  [ 60/312]  eta: 0:04:00  lr: 0.000041  min_lr: 0.000041  loss: 2.9621 (2.9051)  weight_decay: 0.0500 (0.0500)  time: 0.7063  data: 0.1167  max mem: 41794
Epoch: [282]  [ 70/312]  eta: 0:03:49  lr: 0.000041  min_lr: 0.000041  loss: 2.9621 (2.9001)  weight_decay: 0.0500 (0.0500)  time: 0.7150  data: 0.1030  max mem: 41794
Epoch: [282]  [ 80/312]  eta: 0:03:39  lr: 0.000040  min_lr: 0.000040  loss: 2.9278 (2.8929)  weight_decay: 0.0500 (0.0500)  time: 0.9235  data: 0.2147  max mem: 41794
Epoch: [282]  [ 90/312]  eta: 0:03:19  lr: 0.000040  min_lr: 0.000040  loss: 3.0647 (2.9139)  weight_decay: 0.0500 (0.0500)  time: 0.7144  data: 0.1121  max mem: 41794
Epoch: [282]  [100/312]  eta: 0:03:08  lr: 0.000040  min_lr: 0.000040  loss: 3.1388 (2.8947)  weight_decay: 0.0500 (0.0500)  time: 0.6498  data: 0.0772  max mem: 41794
Epoch: [282]  [110/312]  eta: 0:02:52  lr: 0.000040  min_lr: 0.000040  loss: 2.8798 (2.9039)  weight_decay: 0.0500 (0.0500)  time: 0.6451  data: 0.0773  max mem: 41794
Epoch: [282]  [120/312]  eta: 0:02:44  lr: 0.000040  min_lr: 0.000040  loss: 2.9958 (2.8996)  weight_decay: 0.0500 (0.0500)  time: 0.7058  data: 0.1287  max mem: 41794
Epoch: [282]  [130/312]  eta: 0:02:34  lr: 0.000040  min_lr: 0.000040  loss: 3.0484 (2.9102)  weight_decay: 0.0500 (0.0500)  time: 0.8444  data: 0.2235  max mem: 41794
Epoch: [282]  [140/312]  eta: 0:02:22  lr: 0.000040  min_lr: 0.000040  loss: 3.0484 (2.9068)  weight_decay: 0.0500 (0.0500)  time: 0.6347  data: 0.0955  max mem: 41794
Epoch: [282]  [150/312]  eta: 0:02:13  lr: 0.000040  min_lr: 0.000040  loss: 2.7611 (2.8955)  weight_decay: 0.0500 (0.0500)  time: 0.6298  data: 0.1125  max mem: 41794
Epoch: [282]  [160/312]  eta: 0:02:04  lr: 0.000039  min_lr: 0.000039  loss: 2.7322 (2.8895)  weight_decay: 0.0500 (0.0500)  time: 0.7942  data: 0.2681  max mem: 41794
Epoch: [282]  [170/312]  eta: 0:01:54  lr: 0.000039  min_lr: 0.000039  loss: 3.1693 (2.9025)  weight_decay: 0.0500 (0.0500)  time: 0.6695  data: 0.1563  max mem: 41794
Epoch: [282]  [180/312]  eta: 0:01:46  lr: 0.000039  min_lr: 0.000039  loss: 3.0729 (2.8981)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.1658  max mem: 41794
Epoch: [282]  [190/312]  eta: 0:01:36  lr: 0.000039  min_lr: 0.000039  loss: 3.1610 (2.9134)  weight_decay: 0.0500 (0.0500)  time: 0.6626  data: 0.1658  max mem: 41794
Epoch: [282]  [200/312]  eta: 0:01:28  lr: 0.000039  min_lr: 0.000039  loss: 3.2075 (2.9154)  weight_decay: 0.0500 (0.0500)  time: 0.6787  data: 0.1888  max mem: 41794
Epoch: [282]  [210/312]  eta: 0:01:21  lr: 0.000039  min_lr: 0.000039  loss: 3.1834 (2.9277)  weight_decay: 0.0500 (0.0500)  time: 0.8835  data: 0.3517  max mem: 41794
Epoch: [282]  [220/312]  eta: 0:01:12  lr: 0.000039  min_lr: 0.000039  loss: 3.1834 (2.9293)  weight_decay: 0.0500 (0.0500)  time: 0.6976  data: 0.1637  max mem: 41794
Epoch: [282]  [230/312]  eta: 0:01:04  lr: 0.000038  min_lr: 0.000038  loss: 2.8786 (2.9235)  weight_decay: 0.0500 (0.0500)  time: 0.7087  data: 0.1252  max mem: 41794
Epoch: [282]  [240/312]  eta: 0:00:56  lr: 0.000038  min_lr: 0.000038  loss: 2.8671 (2.9191)  weight_decay: 0.0500 (0.0500)  time: 0.7714  data: 0.1387  max mem: 41794
Epoch: [282]  [250/312]  eta: 0:00:48  lr: 0.000038  min_lr: 0.000038  loss: 3.0457 (2.9145)  weight_decay: 0.0500 (0.0500)  time: 0.6486  data: 0.0815  max mem: 41794
Epoch: [282]  [260/312]  eta: 0:00:40  lr: 0.000038  min_lr: 0.000038  loss: 2.9300 (2.9190)  weight_decay: 0.0500 (0.0500)  time: 0.8016  data: 0.2012  max mem: 41794
Epoch: [282]  [270/312]  eta: 0:00:32  lr: 0.000038  min_lr: 0.000038  loss: 2.8393 (2.9185)  weight_decay: 0.0500 (0.0500)  time: 0.7195  data: 0.1338  max mem: 41794
Epoch: [282]  [280/312]  eta: 0:00:24  lr: 0.000038  min_lr: 0.000038  loss: 3.0570 (2.9211)  weight_decay: 0.0500 (0.0500)  time: 0.6797  data: 0.1158  max mem: 41794
Epoch: [282]  [290/312]  eta: 0:00:17  lr: 0.000038  min_lr: 0.000038  loss: 3.1652 (2.9293)  weight_decay: 0.0500 (0.0500)  time: 0.7784  data: 0.1560  max mem: 41794
Epoch: [282]  [300/312]  eta: 0:00:09  lr: 0.000037  min_lr: 0.000037  loss: 3.2201 (2.9346)  weight_decay: 0.0500 (0.0500)  time: 0.5872  data: 0.0406  max mem: 41794
Epoch: [282]  [310/312]  eta: 0:00:01  lr: 0.000037  min_lr: 0.000037  loss: 3.0556 (2.9349)  weight_decay: 0.0500 (0.0500)  time: 0.4638  data: 0.0001  max mem: 41794
Epoch: [282]  [311/312]  eta: 0:00:00  lr: 0.000037  min_lr: 0.000037  loss: 3.0192 (2.9322)  weight_decay: 0.0500 (0.0500)  time: 0.4637  data: 0.0001  max mem: 41794
Epoch: [282] Total time: 0:03:57 (0.7600 s / it)
Averaged stats: lr: 0.000037  min_lr: 0.000037  loss: 3.0192 (2.9690)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:08  loss: 0.8392 (0.8392)  acc1: 85.9375 (85.9375)  acc5: 96.3542 (96.3542)  time: 7.6345  data: 7.4226  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1004 (1.0215)  acc1: 79.5573 (79.8560)  acc5: 95.3125 (95.3440)  time: 1.0568  data: 0.8853  max mem: 41794
Test: Total time: 0:00:09 (1.0658 s / it)
* Acc@1 80.262 Acc@5 95.312 loss 1.018
Accuracy of the model on the 50000 test images: 80.3%
Max accuracy: 80.26%
Epoch: [283]  [  0/312]  eta: 1:20:00  lr: 0.000037  min_lr: 0.000037  loss: 2.1075 (2.1075)  weight_decay: 0.0500 (0.0500)  time: 15.3873  data: 14.6953  max mem: 41794
Epoch: [283]  [ 10/312]  eta: 0:11:11  lr: 0.000037  min_lr: 0.000037  loss: 3.0205 (2.9494)  weight_decay: 0.0500 (0.0500)  time: 2.2233  data: 1.3366  max mem: 41794
Epoch: [283]  [ 20/312]  eta: 0:07:20  lr: 0.000037  min_lr: 0.000037  loss: 3.0205 (2.9115)  weight_decay: 0.0500 (0.0500)  time: 0.8151  data: 0.0520  max mem: 41794
Epoch: [283]  [ 30/312]  eta: 0:05:35  lr: 0.000037  min_lr: 0.000037  loss: 3.2048 (2.9932)  weight_decay: 0.0500 (0.0500)  time: 0.6222  data: 0.0520  max mem: 41794
Epoch: [283]  [ 40/312]  eta: 0:04:38  lr: 0.000037  min_lr: 0.000037  loss: 2.9466 (2.9148)  weight_decay: 0.0500 (0.0500)  time: 0.5131  data: 0.0031  max mem: 41794
Epoch: [283]  [ 50/312]  eta: 0:04:15  lr: 0.000037  min_lr: 0.000037  loss: 2.7267 (2.8942)  weight_decay: 0.0500 (0.0500)  time: 0.6452  data: 0.1462  max mem: 41794
Epoch: [283]  [ 60/312]  eta: 0:03:46  lr: 0.000036  min_lr: 0.000036  loss: 3.0416 (2.9311)  weight_decay: 0.0500 (0.0500)  time: 0.6416  data: 0.1437  max mem: 41794
Epoch: [283]  [ 70/312]  eta: 0:03:37  lr: 0.000036  min_lr: 0.000036  loss: 3.1886 (2.9609)  weight_decay: 0.0500 (0.0500)  time: 0.6991  data: 0.1972  max mem: 41794
Epoch: [283]  [ 80/312]  eta: 0:03:24  lr: 0.000036  min_lr: 0.000036  loss: 3.0400 (2.9742)  weight_decay: 0.0500 (0.0500)  time: 0.8271  data: 0.3186  max mem: 41794
Epoch: [283]  [ 90/312]  eta: 0:03:09  lr: 0.000036  min_lr: 0.000036  loss: 3.1636 (3.0048)  weight_decay: 0.0500 (0.0500)  time: 0.6947  data: 0.1910  max mem: 41794
Epoch: [283]  [100/312]  eta: 0:03:03  lr: 0.000036  min_lr: 0.000036  loss: 3.1636 (2.9975)  weight_decay: 0.0500 (0.0500)  time: 0.7933  data: 0.2501  max mem: 41794
Epoch: [283]  [110/312]  eta: 0:02:47  lr: 0.000036  min_lr: 0.000036  loss: 3.2091 (3.0131)  weight_decay: 0.0500 (0.0500)  time: 0.7266  data: 0.1823  max mem: 41794
Epoch: [283]  [120/312]  eta: 0:02:42  lr: 0.000036  min_lr: 0.000036  loss: 3.0791 (2.9978)  weight_decay: 0.0500 (0.0500)  time: 0.7452  data: 0.1536  max mem: 41794
Epoch: [283]  [130/312]  eta: 0:02:34  lr: 0.000036  min_lr: 0.000036  loss: 2.8625 (2.9945)  weight_decay: 0.0500 (0.0500)  time: 0.9336  data: 0.2585  max mem: 41794
Epoch: [283]  [140/312]  eta: 0:02:21  lr: 0.000035  min_lr: 0.000035  loss: 3.2298 (3.0107)  weight_decay: 0.0500 (0.0500)  time: 0.6858  data: 0.1066  max mem: 41794
Epoch: [283]  [150/312]  eta: 0:02:13  lr: 0.000035  min_lr: 0.000035  loss: 3.2056 (3.0045)  weight_decay: 0.0500 (0.0500)  time: 0.6834  data: 0.1223  max mem: 41794
Epoch: [283]  [160/312]  eta: 0:02:03  lr: 0.000035  min_lr: 0.000035  loss: 3.1030 (3.0135)  weight_decay: 0.0500 (0.0500)  time: 0.7621  data: 0.1674  max mem: 41794
Epoch: [283]  [170/312]  eta: 0:01:53  lr: 0.000035  min_lr: 0.000035  loss: 3.1372 (3.0083)  weight_decay: 0.0500 (0.0500)  time: 0.6250  data: 0.1017  max mem: 41794
Epoch: [283]  [180/312]  eta: 0:01:46  lr: 0.000035  min_lr: 0.000035  loss: 3.1688 (3.0180)  weight_decay: 0.0500 (0.0500)  time: 0.7796  data: 0.2028  max mem: 41794
Epoch: [283]  [190/312]  eta: 0:01:36  lr: 0.000035  min_lr: 0.000035  loss: 3.1120 (3.0195)  weight_decay: 0.0500 (0.0500)  time: 0.7296  data: 0.1468  max mem: 41794
Epoch: [283]  [200/312]  eta: 0:01:29  lr: 0.000035  min_lr: 0.000035  loss: 3.0603 (3.0085)  weight_decay: 0.0500 (0.0500)  time: 0.6711  data: 0.0908  max mem: 41794
Epoch: [283]  [210/312]  eta: 0:01:21  lr: 0.000034  min_lr: 0.000034  loss: 3.0868 (3.0154)  weight_decay: 0.0500 (0.0500)  time: 0.8906  data: 0.1970  max mem: 41794
Epoch: [283]  [220/312]  eta: 0:01:12  lr: 0.000034  min_lr: 0.000034  loss: 3.1112 (3.0164)  weight_decay: 0.0500 (0.0500)  time: 0.7181  data: 0.1068  max mem: 41794
Epoch: [283]  [230/312]  eta: 0:01:04  lr: 0.000034  min_lr: 0.000034  loss: 3.0777 (3.0155)  weight_decay: 0.0500 (0.0500)  time: 0.6510  data: 0.0883  max mem: 41794
Epoch: [283]  [240/312]  eta: 0:00:56  lr: 0.000034  min_lr: 0.000034  loss: 3.0276 (3.0097)  weight_decay: 0.0500 (0.0500)  time: 0.7728  data: 0.0883  max mem: 41794
Epoch: [283]  [250/312]  eta: 0:00:48  lr: 0.000034  min_lr: 0.000034  loss: 3.2684 (3.0122)  weight_decay: 0.0500 (0.0500)  time: 0.7258  data: 0.1051  max mem: 41794
Epoch: [283]  [260/312]  eta: 0:00:40  lr: 0.000034  min_lr: 0.000034  loss: 3.1851 (3.0137)  weight_decay: 0.0500 (0.0500)  time: 0.7544  data: 0.1868  max mem: 41794
Epoch: [283]  [270/312]  eta: 0:00:32  lr: 0.000034  min_lr: 0.000034  loss: 3.0037 (3.0063)  weight_decay: 0.0500 (0.0500)  time: 0.6497  data: 0.0839  max mem: 41794
Epoch: [283]  [280/312]  eta: 0:00:24  lr: 0.000034  min_lr: 0.000034  loss: 2.9817 (3.0107)  weight_decay: 0.0500 (0.0500)  time: 0.6791  data: 0.1043  max mem: 41794
Epoch: [283]  [290/312]  eta: 0:00:17  lr: 0.000033  min_lr: 0.000033  loss: 3.0359 (3.0071)  weight_decay: 0.0500 (0.0500)  time: 0.8444  data: 0.1752  max mem: 41794
Epoch: [283]  [300/312]  eta: 0:00:09  lr: 0.000033  min_lr: 0.000033  loss: 3.2402 (3.0117)  weight_decay: 0.0500 (0.0500)  time: 0.6483  data: 0.0729  max mem: 41794
Epoch: [283]  [310/312]  eta: 0:00:01  lr: 0.000033  min_lr: 0.000033  loss: 3.2567 (3.0112)  weight_decay: 0.0500 (0.0500)  time: 0.4627  data: 0.0001  max mem: 41794
Epoch: [283]  [311/312]  eta: 0:00:00  lr: 0.000033  min_lr: 0.000033  loss: 3.2567 (3.0110)  weight_decay: 0.0500 (0.0500)  time: 0.4627  data: 0.0001  max mem: 41794
Epoch: [283] Total time: 0:03:58 (0.7647 s / it)
Averaged stats: lr: 0.000033  min_lr: 0.000033  loss: 3.2567 (2.9658)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:12  loss: 0.8606 (0.8606)  acc1: 86.0677 (86.0677)  acc5: 96.7448 (96.7448)  time: 8.0721  data: 7.8605  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1188 (1.0458)  acc1: 79.0365 (79.8720)  acc5: 95.7031 (95.4720)  time: 1.0808  data: 0.9093  max mem: 41794
Test: Total time: 0:00:10 (1.1186 s / it)
* Acc@1 80.274 Acc@5 95.304 loss 1.043
Accuracy of the model on the 50000 test images: 80.3%
Max accuracy: 80.27%
Epoch: [284]  [  0/312]  eta: 1:13:05  lr: 0.000033  min_lr: 0.000033  loss: 3.0262 (3.0262)  weight_decay: 0.0500 (0.0500)  time: 14.0569  data: 11.7445  max mem: 41794
Epoch: [284]  [ 10/312]  eta: 0:09:58  lr: 0.000033  min_lr: 0.000033  loss: 3.2184 (3.1755)  weight_decay: 0.0500 (0.0500)  time: 1.9821  data: 1.3372  max mem: 41794
Epoch: [284]  [ 20/312]  eta: 0:07:04  lr: 0.000033  min_lr: 0.000033  loss: 3.2032 (3.1549)  weight_decay: 0.0500 (0.0500)  time: 0.8236  data: 0.2572  max mem: 41794
Epoch: [284]  [ 30/312]  eta: 0:05:23  lr: 0.000033  min_lr: 0.000033  loss: 3.1640 (3.1163)  weight_decay: 0.0500 (0.0500)  time: 0.6905  data: 0.1136  max mem: 41794
Epoch: [284]  [ 40/312]  eta: 0:04:50  lr: 0.000033  min_lr: 0.000033  loss: 3.1640 (3.0784)  weight_decay: 0.0500 (0.0500)  time: 0.6616  data: 0.1342  max mem: 41794
Epoch: [284]  [ 50/312]  eta: 0:04:36  lr: 0.000032  min_lr: 0.000032  loss: 3.2391 (3.0852)  weight_decay: 0.0500 (0.0500)  time: 0.9060  data: 0.2907  max mem: 41794
Epoch: [284]  [ 60/312]  eta: 0:04:02  lr: 0.000032  min_lr: 0.000032  loss: 3.0919 (3.0593)  weight_decay: 0.0500 (0.0500)  time: 0.7508  data: 0.1615  max mem: 41794
Epoch: [284]  [ 70/312]  eta: 0:03:49  lr: 0.000032  min_lr: 0.000032  loss: 3.0602 (3.0569)  weight_decay: 0.0500 (0.0500)  time: 0.6780  data: 0.1152  max mem: 41794
Epoch: [284]  [ 80/312]  eta: 0:03:34  lr: 0.000032  min_lr: 0.000032  loss: 3.0653 (3.0419)  weight_decay: 0.0500 (0.0500)  time: 0.7991  data: 0.1314  max mem: 41794
Epoch: [284]  [ 90/312]  eta: 0:03:19  lr: 0.000032  min_lr: 0.000032  loss: 3.1522 (3.0622)  weight_decay: 0.0500 (0.0500)  time: 0.7258  data: 0.1254  max mem: 41794
Epoch: [284]  [100/312]  eta: 0:03:10  lr: 0.000032  min_lr: 0.000032  loss: 3.1522 (3.0332)  weight_decay: 0.0500 (0.0500)  time: 0.8113  data: 0.2212  max mem: 41794
Epoch: [284]  [110/312]  eta: 0:02:54  lr: 0.000032  min_lr: 0.000032  loss: 3.0115 (3.0278)  weight_decay: 0.0500 (0.0500)  time: 0.7163  data: 0.1127  max mem: 41794
Epoch: [284]  [120/312]  eta: 0:02:45  lr: 0.000032  min_lr: 0.000032  loss: 3.1632 (3.0256)  weight_decay: 0.0500 (0.0500)  time: 0.6548  data: 0.0964  max mem: 41794
Epoch: [284]  [130/312]  eta: 0:02:36  lr: 0.000031  min_lr: 0.000031  loss: 3.1152 (3.0053)  weight_decay: 0.0500 (0.0500)  time: 0.8133  data: 0.1893  max mem: 41794
Epoch: [284]  [140/312]  eta: 0:02:23  lr: 0.000031  min_lr: 0.000031  loss: 2.9629 (3.0006)  weight_decay: 0.0500 (0.0500)  time: 0.6612  data: 0.0936  max mem: 41794
Epoch: [284]  [150/312]  eta: 0:02:15  lr: 0.000031  min_lr: 0.000031  loss: 3.0285 (2.9994)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.0868  max mem: 41794
Epoch: [284]  [160/312]  eta: 0:02:05  lr: 0.000031  min_lr: 0.000031  loss: 3.0285 (3.0010)  weight_decay: 0.0500 (0.0500)  time: 0.7868  data: 0.0880  max mem: 41794
Epoch: [284]  [170/312]  eta: 0:01:55  lr: 0.000031  min_lr: 0.000031  loss: 2.8958 (2.9810)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.0743  max mem: 41794
Epoch: [284]  [180/312]  eta: 0:01:48  lr: 0.000031  min_lr: 0.000031  loss: 2.7828 (2.9766)  weight_decay: 0.0500 (0.0500)  time: 0.7787  data: 0.1790  max mem: 41794
Epoch: [284]  [190/312]  eta: 0:01:38  lr: 0.000031  min_lr: 0.000031  loss: 3.0893 (2.9770)  weight_decay: 0.0500 (0.0500)  time: 0.7156  data: 0.1079  max mem: 41794
Epoch: [284]  [200/312]  eta: 0:01:30  lr: 0.000031  min_lr: 0.000031  loss: 3.0893 (2.9729)  weight_decay: 0.0500 (0.0500)  time: 0.6934  data: 0.0631  max mem: 41794
Epoch: [284]  [210/312]  eta: 0:01:21  lr: 0.000030  min_lr: 0.000030  loss: 2.9895 (2.9687)  weight_decay: 0.0500 (0.0500)  time: 0.7788  data: 0.0652  max mem: 41794
Epoch: [284]  [220/312]  eta: 0:01:12  lr: 0.000030  min_lr: 0.000030  loss: 3.0640 (2.9726)  weight_decay: 0.0500 (0.0500)  time: 0.6310  data: 0.0495  max mem: 41794
Epoch: [284]  [230/312]  eta: 0:01:05  lr: 0.000030  min_lr: 0.000030  loss: 3.2213 (2.9796)  weight_decay: 0.0500 (0.0500)  time: 0.7250  data: 0.1275  max mem: 41794
Epoch: [284]  [240/312]  eta: 0:00:56  lr: 0.000030  min_lr: 0.000030  loss: 2.9434 (2.9694)  weight_decay: 0.0500 (0.0500)  time: 0.7388  data: 0.0837  max mem: 41794
Epoch: [284]  [250/312]  eta: 0:00:48  lr: 0.000030  min_lr: 0.000030  loss: 2.8753 (2.9708)  weight_decay: 0.0500 (0.0500)  time: 0.7070  data: 0.0876  max mem: 41794
Epoch: [284]  [260/312]  eta: 0:00:41  lr: 0.000030  min_lr: 0.000030  loss: 2.9560 (2.9719)  weight_decay: 0.0500 (0.0500)  time: 0.7992  data: 0.1132  max mem: 41794
Epoch: [284]  [270/312]  eta: 0:00:32  lr: 0.000030  min_lr: 0.000030  loss: 2.9739 (2.9776)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.0431  max mem: 41794
Epoch: [284]  [280/312]  eta: 0:00:25  lr: 0.000030  min_lr: 0.000030  loss: 3.1027 (2.9805)  weight_decay: 0.0500 (0.0500)  time: 0.6965  data: 0.0781  max mem: 41794
Epoch: [284]  [290/312]  eta: 0:00:17  lr: 0.000030  min_lr: 0.000030  loss: 2.8637 (2.9757)  weight_decay: 0.0500 (0.0500)  time: 0.7588  data: 0.0626  max mem: 41794
Epoch: [284]  [300/312]  eta: 0:00:09  lr: 0.000029  min_lr: 0.000029  loss: 2.9999 (2.9811)  weight_decay: 0.0500 (0.0500)  time: 0.5686  data: 0.0033  max mem: 41794
Epoch: [284]  [310/312]  eta: 0:00:01  lr: 0.000029  min_lr: 0.000029  loss: 3.1655 (2.9796)  weight_decay: 0.0500 (0.0500)  time: 0.4682  data: 0.0032  max mem: 41794
Epoch: [284]  [311/312]  eta: 0:00:00  lr: 0.000029  min_lr: 0.000029  loss: 3.1100 (2.9763)  weight_decay: 0.0500 (0.0500)  time: 0.4673  data: 0.0032  max mem: 41794
Epoch: [284] Total time: 0:03:58 (0.7641 s / it)
Averaged stats: lr: 0.000029  min_lr: 0.000029  loss: 3.1100 (2.9744)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.8523 (0.8523)  acc1: 86.1979 (86.1979)  acc5: 96.7448 (96.7448)  time: 8.4503  data: 8.2506  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1136 (1.0454)  acc1: 78.9062 (79.9840)  acc5: 95.7031 (95.2640)  time: 1.0944  data: 0.9168  max mem: 41794
Test: Total time: 0:00:10 (1.1111 s / it)
* Acc@1 80.320 Acc@5 95.224 loss 1.042
Accuracy of the model on the 50000 test images: 80.3%
Max accuracy: 80.32%
Epoch: [285]  [  0/312]  eta: 1:19:04  lr: 0.000029  min_lr: 0.000029  loss: 2.5166 (2.5166)  weight_decay: 0.0500 (0.0500)  time: 15.2071  data: 12.9249  max mem: 41794
Epoch: [285]  [ 10/312]  eta: 0:10:23  lr: 0.000029  min_lr: 0.000029  loss: 3.1525 (3.0358)  weight_decay: 0.0500 (0.0500)  time: 2.0632  data: 1.3271  max mem: 41794
Epoch: [285]  [ 20/312]  eta: 0:07:22  lr: 0.000029  min_lr: 0.000029  loss: 3.0335 (3.0270)  weight_decay: 0.0500 (0.0500)  time: 0.8326  data: 0.2024  max mem: 41794
Epoch: [285]  [ 30/312]  eta: 0:05:38  lr: 0.000029  min_lr: 0.000029  loss: 3.0057 (2.9678)  weight_decay: 0.0500 (0.0500)  time: 0.7252  data: 0.1208  max mem: 41794
Epoch: [285]  [ 40/312]  eta: 0:04:57  lr: 0.000029  min_lr: 0.000029  loss: 3.0497 (2.9629)  weight_decay: 0.0500 (0.0500)  time: 0.6527  data: 0.0822  max mem: 41794
Epoch: [285]  [ 50/312]  eta: 0:04:32  lr: 0.000029  min_lr: 0.000029  loss: 3.0025 (2.9726)  weight_decay: 0.0500 (0.0500)  time: 0.7927  data: 0.1505  max mem: 41794
Epoch: [285]  [ 60/312]  eta: 0:04:00  lr: 0.000029  min_lr: 0.000029  loss: 3.0459 (2.9791)  weight_decay: 0.0500 (0.0500)  time: 0.6637  data: 0.0706  max mem: 41794
Epoch: [285]  [ 70/312]  eta: 0:03:47  lr: 0.000028  min_lr: 0.000028  loss: 3.1391 (2.9843)  weight_decay: 0.0500 (0.0500)  time: 0.6863  data: 0.0751  max mem: 41794
Epoch: [285]  [ 80/312]  eta: 0:03:33  lr: 0.000028  min_lr: 0.000028  loss: 3.1523 (3.0027)  weight_decay: 0.0500 (0.0500)  time: 0.8143  data: 0.0996  max mem: 41794
Epoch: [285]  [ 90/312]  eta: 0:03:17  lr: 0.000028  min_lr: 0.000028  loss: 3.1584 (2.9892)  weight_decay: 0.0500 (0.0500)  time: 0.7138  data: 0.1062  max mem: 41794
Epoch: [285]  [100/312]  eta: 0:03:07  lr: 0.000028  min_lr: 0.000028  loss: 2.8966 (2.9471)  weight_decay: 0.0500 (0.0500)  time: 0.7435  data: 0.1566  max mem: 41794
Epoch: [285]  [110/312]  eta: 0:02:51  lr: 0.000028  min_lr: 0.000028  loss: 2.8966 (2.9570)  weight_decay: 0.0500 (0.0500)  time: 0.6643  data: 0.0756  max mem: 41794
Epoch: [285]  [120/312]  eta: 0:02:43  lr: 0.000028  min_lr: 0.000028  loss: 3.1342 (2.9544)  weight_decay: 0.0500 (0.0500)  time: 0.6808  data: 0.0994  max mem: 41794
Epoch: [285]  [130/312]  eta: 0:02:33  lr: 0.000028  min_lr: 0.000028  loss: 3.0136 (2.9396)  weight_decay: 0.0500 (0.0500)  time: 0.8173  data: 0.1608  max mem: 41794
Epoch: [285]  [140/312]  eta: 0:02:21  lr: 0.000028  min_lr: 0.000028  loss: 3.0136 (2.9511)  weight_decay: 0.0500 (0.0500)  time: 0.6523  data: 0.0832  max mem: 41794
Epoch: [285]  [150/312]  eta: 0:02:14  lr: 0.000027  min_lr: 0.000027  loss: 2.9513 (2.9304)  weight_decay: 0.0500 (0.0500)  time: 0.7334  data: 0.1526  max mem: 41794
Epoch: [285]  [160/312]  eta: 0:02:04  lr: 0.000027  min_lr: 0.000027  loss: 2.9035 (2.9359)  weight_decay: 0.0500 (0.0500)  time: 0.8042  data: 0.1757  max mem: 41794
Epoch: [285]  [170/312]  eta: 0:01:55  lr: 0.000027  min_lr: 0.000027  loss: 3.2243 (2.9332)  weight_decay: 0.0500 (0.0500)  time: 0.6871  data: 0.1457  max mem: 41794
Epoch: [285]  [180/312]  eta: 0:01:47  lr: 0.000027  min_lr: 0.000027  loss: 2.7507 (2.9211)  weight_decay: 0.0500 (0.0500)  time: 0.7895  data: 0.2455  max mem: 41794
Epoch: [285]  [190/312]  eta: 0:01:37  lr: 0.000027  min_lr: 0.000027  loss: 2.7449 (2.9251)  weight_decay: 0.0500 (0.0500)  time: 0.6897  data: 0.1469  max mem: 41794
Epoch: [285]  [200/312]  eta: 0:01:29  lr: 0.000027  min_lr: 0.000027  loss: 2.9915 (2.9229)  weight_decay: 0.0500 (0.0500)  time: 0.6483  data: 0.1221  max mem: 41794
Epoch: [285]  [210/312]  eta: 0:01:21  lr: 0.000027  min_lr: 0.000027  loss: 3.0796 (2.9321)  weight_decay: 0.0500 (0.0500)  time: 0.8220  data: 0.2823  max mem: 41794
Epoch: [285]  [220/312]  eta: 0:01:12  lr: 0.000027  min_lr: 0.000027  loss: 3.2602 (2.9339)  weight_decay: 0.0500 (0.0500)  time: 0.6748  data: 0.1629  max mem: 41794
Epoch: [285]  [230/312]  eta: 0:01:04  lr: 0.000027  min_lr: 0.000027  loss: 3.0046 (2.9338)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1738  max mem: 41794
Epoch: [285]  [240/312]  eta: 0:00:56  lr: 0.000026  min_lr: 0.000026  loss: 3.0046 (2.9329)  weight_decay: 0.0500 (0.0500)  time: 0.7945  data: 0.3004  max mem: 41794
Epoch: [285]  [250/312]  eta: 0:00:48  lr: 0.000026  min_lr: 0.000026  loss: 2.7485 (2.9173)  weight_decay: 0.0500 (0.0500)  time: 0.6409  data: 0.1462  max mem: 41794
Epoch: [285]  [260/312]  eta: 0:00:40  lr: 0.000026  min_lr: 0.000026  loss: 2.6519 (2.9174)  weight_decay: 0.0500 (0.0500)  time: 0.7281  data: 0.2312  max mem: 41794
Epoch: [285]  [270/312]  eta: 0:00:32  lr: 0.000026  min_lr: 0.000026  loss: 3.1599 (2.9189)  weight_decay: 0.0500 (0.0500)  time: 0.7095  data: 0.2122  max mem: 41794
Epoch: [285]  [280/312]  eta: 0:00:24  lr: 0.000026  min_lr: 0.000026  loss: 3.0099 (2.9191)  weight_decay: 0.0500 (0.0500)  time: 0.6777  data: 0.1847  max mem: 41794
Epoch: [285]  [290/312]  eta: 0:00:17  lr: 0.000026  min_lr: 0.000026  loss: 3.0099 (2.9216)  weight_decay: 0.0500 (0.0500)  time: 0.7948  data: 0.2812  max mem: 41794
Epoch: [285]  [300/312]  eta: 0:00:09  lr: 0.000026  min_lr: 0.000026  loss: 3.0106 (2.9195)  weight_decay: 0.0500 (0.0500)  time: 0.5994  data: 0.0969  max mem: 41794
Epoch: [285]  [310/312]  eta: 0:00:01  lr: 0.000026  min_lr: 0.000026  loss: 2.7573 (2.9177)  weight_decay: 0.0500 (0.0500)  time: 0.4651  data: 0.0001  max mem: 41794
Epoch: [285]  [311/312]  eta: 0:00:00  lr: 0.000026  min_lr: 0.000026  loss: 2.7409 (2.9164)  weight_decay: 0.0500 (0.0500)  time: 0.4646  data: 0.0001  max mem: 41794
Epoch: [285] Total time: 0:03:57 (0.7606 s / it)
Averaged stats: lr: 0.000026  min_lr: 0.000026  loss: 2.7409 (2.9607)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:16  loss: 0.8141 (0.8141)  acc1: 86.4583 (86.4583)  acc5: 96.8750 (96.8750)  time: 8.5425  data: 8.3401  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0810 (1.0007)  acc1: 79.4271 (80.0640)  acc5: 95.4427 (95.4080)  time: 1.1001  data: 0.9268  max mem: 41794
Test: Total time: 0:00:10 (1.1439 s / it)
* Acc@1 80.398 Acc@5 95.322 loss 0.997
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.40%
Epoch: [286]  [  0/312]  eta: 1:17:39  lr: 0.000026  min_lr: 0.000026  loss: 2.3052 (2.3052)  weight_decay: 0.0500 (0.0500)  time: 14.9353  data: 12.9509  max mem: 41794
Epoch: [286]  [ 10/312]  eta: 0:10:30  lr: 0.000026  min_lr: 0.000026  loss: 2.9643 (2.8241)  weight_decay: 0.0500 (0.0500)  time: 2.0884  data: 1.3226  max mem: 41794
Epoch: [286]  [ 20/312]  eta: 0:07:20  lr: 0.000025  min_lr: 0.000025  loss: 3.0209 (2.8366)  weight_decay: 0.0500 (0.0500)  time: 0.8372  data: 0.1289  max mem: 41794
Epoch: [286]  [ 30/312]  eta: 0:05:37  lr: 0.000025  min_lr: 0.000025  loss: 3.0209 (2.8620)  weight_decay: 0.0500 (0.0500)  time: 0.7047  data: 0.0547  max mem: 41794
Epoch: [286]  [ 40/312]  eta: 0:04:48  lr: 0.000025  min_lr: 0.000025  loss: 3.1748 (2.9027)  weight_decay: 0.0500 (0.0500)  time: 0.5911  data: 0.0834  max mem: 41794
Epoch: [286]  [ 50/312]  eta: 0:04:21  lr: 0.000025  min_lr: 0.000025  loss: 3.1575 (2.9479)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.1987  max mem: 41794
Epoch: [286]  [ 60/312]  eta: 0:03:49  lr: 0.000025  min_lr: 0.000025  loss: 3.1346 (2.9625)  weight_decay: 0.0500 (0.0500)  time: 0.6063  data: 0.1214  max mem: 41794
Epoch: [286]  [ 70/312]  eta: 0:03:38  lr: 0.000025  min_lr: 0.000025  loss: 3.1285 (2.9724)  weight_decay: 0.0500 (0.0500)  time: 0.6595  data: 0.1675  max mem: 41794
Epoch: [286]  [ 80/312]  eta: 0:03:27  lr: 0.000025  min_lr: 0.000025  loss: 3.1170 (2.9795)  weight_decay: 0.0500 (0.0500)  time: 0.8360  data: 0.3202  max mem: 41794
Epoch: [286]  [ 90/312]  eta: 0:03:13  lr: 0.000025  min_lr: 0.000025  loss: 3.0245 (2.9671)  weight_decay: 0.0500 (0.0500)  time: 0.7542  data: 0.1906  max mem: 41794
Epoch: [286]  [100/312]  eta: 0:03:03  lr: 0.000025  min_lr: 0.000025  loss: 2.8620 (2.9623)  weight_decay: 0.0500 (0.0500)  time: 0.7565  data: 0.1296  max mem: 41794
Epoch: [286]  [110/312]  eta: 0:02:48  lr: 0.000024  min_lr: 0.000024  loss: 3.1206 (2.9792)  weight_decay: 0.0500 (0.0500)  time: 0.6607  data: 0.0924  max mem: 41794
Epoch: [286]  [120/312]  eta: 0:02:40  lr: 0.000024  min_lr: 0.000024  loss: 3.0538 (2.9620)  weight_decay: 0.0500 (0.0500)  time: 0.6969  data: 0.0935  max mem: 41794
Epoch: [286]  [130/312]  eta: 0:02:33  lr: 0.000024  min_lr: 0.000024  loss: 2.7827 (2.9591)  weight_decay: 0.0500 (0.0500)  time: 0.8901  data: 0.1825  max mem: 41794
Epoch: [286]  [140/312]  eta: 0:02:20  lr: 0.000024  min_lr: 0.000024  loss: 3.0148 (2.9496)  weight_decay: 0.0500 (0.0500)  time: 0.7029  data: 0.1072  max mem: 41794
Epoch: [286]  [150/312]  eta: 0:02:14  lr: 0.000024  min_lr: 0.000024  loss: 3.0513 (2.9531)  weight_decay: 0.0500 (0.0500)  time: 0.7424  data: 0.1826  max mem: 41794
Epoch: [286]  [160/312]  eta: 0:02:05  lr: 0.000024  min_lr: 0.000024  loss: 3.1129 (2.9612)  weight_decay: 0.0500 (0.0500)  time: 0.8960  data: 0.2417  max mem: 41794
Epoch: [286]  [170/312]  eta: 0:01:55  lr: 0.000024  min_lr: 0.000024  loss: 2.9021 (2.9454)  weight_decay: 0.0500 (0.0500)  time: 0.6992  data: 0.1143  max mem: 41794
Epoch: [286]  [180/312]  eta: 0:01:48  lr: 0.000024  min_lr: 0.000024  loss: 3.0290 (2.9583)  weight_decay: 0.0500 (0.0500)  time: 0.7421  data: 0.1340  max mem: 41794
Epoch: [286]  [190/312]  eta: 0:01:37  lr: 0.000024  min_lr: 0.000024  loss: 3.2513 (2.9568)  weight_decay: 0.0500 (0.0500)  time: 0.7149  data: 0.0971  max mem: 41794
Epoch: [286]  [200/312]  eta: 0:01:30  lr: 0.000023  min_lr: 0.000023  loss: 3.2571 (2.9711)  weight_decay: 0.0500 (0.0500)  time: 0.6796  data: 0.0790  max mem: 41794
Epoch: [286]  [210/312]  eta: 0:01:22  lr: 0.000023  min_lr: 0.000023  loss: 3.1804 (2.9736)  weight_decay: 0.0500 (0.0500)  time: 0.8283  data: 0.1324  max mem: 41794
Epoch: [286]  [220/312]  eta: 0:01:12  lr: 0.000023  min_lr: 0.000023  loss: 3.1567 (2.9795)  weight_decay: 0.0500 (0.0500)  time: 0.6639  data: 0.0758  max mem: 41794
Epoch: [286]  [230/312]  eta: 0:01:05  lr: 0.000023  min_lr: 0.000023  loss: 3.1579 (2.9758)  weight_decay: 0.0500 (0.0500)  time: 0.6836  data: 0.1242  max mem: 41794
Epoch: [286]  [240/312]  eta: 0:00:57  lr: 0.000023  min_lr: 0.000023  loss: 3.1673 (2.9796)  weight_decay: 0.0500 (0.0500)  time: 0.8342  data: 0.1903  max mem: 41794
Epoch: [286]  [250/312]  eta: 0:00:48  lr: 0.000023  min_lr: 0.000023  loss: 3.2017 (2.9794)  weight_decay: 0.0500 (0.0500)  time: 0.6600  data: 0.0894  max mem: 41794
Epoch: [286]  [260/312]  eta: 0:00:40  lr: 0.000023  min_lr: 0.000023  loss: 3.0234 (2.9803)  weight_decay: 0.0500 (0.0500)  time: 0.6941  data: 0.1553  max mem: 41794
Epoch: [286]  [270/312]  eta: 0:00:32  lr: 0.000023  min_lr: 0.000023  loss: 2.8833 (2.9823)  weight_decay: 0.0500 (0.0500)  time: 0.6943  data: 0.1543  max mem: 41794
Epoch: [286]  [280/312]  eta: 0:00:24  lr: 0.000023  min_lr: 0.000023  loss: 3.1548 (2.9899)  weight_decay: 0.0500 (0.0500)  time: 0.6658  data: 0.1625  max mem: 41794
Epoch: [286]  [290/312]  eta: 0:00:17  lr: 0.000022  min_lr: 0.000022  loss: 3.1156 (2.9853)  weight_decay: 0.0500 (0.0500)  time: 0.8094  data: 0.3014  max mem: 41794
Epoch: [286]  [300/312]  eta: 0:00:09  lr: 0.000022  min_lr: 0.000022  loss: 2.7232 (2.9770)  weight_decay: 0.0500 (0.0500)  time: 0.6254  data: 0.1393  max mem: 41794
Epoch: [286]  [310/312]  eta: 0:00:01  lr: 0.000022  min_lr: 0.000022  loss: 3.0587 (2.9776)  weight_decay: 0.0500 (0.0500)  time: 0.4657  data: 0.0001  max mem: 41794
Epoch: [286]  [311/312]  eta: 0:00:00  lr: 0.000022  min_lr: 0.000022  loss: 2.9948 (2.9771)  weight_decay: 0.0500 (0.0500)  time: 0.4649  data: 0.0001  max mem: 41794
Epoch: [286] Total time: 0:03:58 (0.7646 s / it)
Averaged stats: lr: 0.000022  min_lr: 0.000022  loss: 2.9948 (2.9721)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:15  loss: 0.8247 (0.8247)  acc1: 85.6771 (85.6771)  acc5: 96.7448 (96.7448)  time: 8.3724  data: 8.1702  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0940 (1.0126)  acc1: 79.4271 (79.7440)  acc5: 95.3125 (95.3920)  time: 1.0832  data: 0.9079  max mem: 41794
Test: Total time: 0:00:09 (1.1084 s / it)
* Acc@1 80.370 Acc@5 95.256 loss 1.010
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.40%
Epoch: [287]  [  0/312]  eta: 1:23:34  lr: 0.000022  min_lr: 0.000022  loss: 3.0436 (3.0436)  weight_decay: 0.0500 (0.0500)  time: 16.0733  data: 11.7619  max mem: 41794
Epoch: [287]  [ 10/312]  eta: 0:11:05  lr: 0.000022  min_lr: 0.000022  loss: 3.0971 (3.0751)  weight_decay: 0.0500 (0.0500)  time: 2.2044  data: 1.3226  max mem: 41794
Epoch: [287]  [ 20/312]  eta: 0:07:31  lr: 0.000022  min_lr: 0.000022  loss: 3.0159 (3.0046)  weight_decay: 0.0500 (0.0500)  time: 0.8213  data: 0.2031  max mem: 41794
Epoch: [287]  [ 30/312]  eta: 0:05:42  lr: 0.000022  min_lr: 0.000022  loss: 2.8653 (2.9637)  weight_decay: 0.0500 (0.0500)  time: 0.6724  data: 0.0751  max mem: 41794
Epoch: [287]  [ 40/312]  eta: 0:04:49  lr: 0.000022  min_lr: 0.000022  loss: 2.9848 (2.9708)  weight_decay: 0.0500 (0.0500)  time: 0.5586  data: 0.0124  max mem: 41794
Epoch: [287]  [ 50/312]  eta: 0:04:36  lr: 0.000022  min_lr: 0.000022  loss: 3.1936 (3.0277)  weight_decay: 0.0500 (0.0500)  time: 0.8021  data: 0.0848  max mem: 41794
Epoch: [287]  [ 60/312]  eta: 0:04:03  lr: 0.000022  min_lr: 0.000022  loss: 3.1955 (3.0360)  weight_decay: 0.0500 (0.0500)  time: 0.7591  data: 0.0867  max mem: 41794
Epoch: [287]  [ 70/312]  eta: 0:03:48  lr: 0.000022  min_lr: 0.000022  loss: 3.1881 (3.0521)  weight_decay: 0.0500 (0.0500)  time: 0.6678  data: 0.0627  max mem: 41794
Epoch: [287]  [ 80/312]  eta: 0:03:34  lr: 0.000021  min_lr: 0.000021  loss: 3.2642 (3.0647)  weight_decay: 0.0500 (0.0500)  time: 0.7934  data: 0.0616  max mem: 41794
Epoch: [287]  [ 90/312]  eta: 0:03:17  lr: 0.000021  min_lr: 0.000021  loss: 2.9719 (3.0174)  weight_decay: 0.0500 (0.0500)  time: 0.6895  data: 0.0629  max mem: 41794
Epoch: [287]  [100/312]  eta: 0:03:08  lr: 0.000021  min_lr: 0.000021  loss: 2.9135 (3.0251)  weight_decay: 0.0500 (0.0500)  time: 0.7472  data: 0.1149  max mem: 41794
Epoch: [287]  [110/312]  eta: 0:02:52  lr: 0.000021  min_lr: 0.000021  loss: 3.2481 (3.0370)  weight_decay: 0.0500 (0.0500)  time: 0.6989  data: 0.0541  max mem: 41794
Epoch: [287]  [120/312]  eta: 0:02:43  lr: 0.000021  min_lr: 0.000021  loss: 3.2113 (3.0377)  weight_decay: 0.0500 (0.0500)  time: 0.6772  data: 0.0657  max mem: 41794
Epoch: [287]  [130/312]  eta: 0:02:36  lr: 0.000021  min_lr: 0.000021  loss: 2.9893 (3.0338)  weight_decay: 0.0500 (0.0500)  time: 0.8750  data: 0.1538  max mem: 41794
Epoch: [287]  [140/312]  eta: 0:02:23  lr: 0.000021  min_lr: 0.000021  loss: 3.0800 (3.0248)  weight_decay: 0.0500 (0.0500)  time: 0.7062  data: 0.0887  max mem: 41794
Epoch: [287]  [150/312]  eta: 0:02:15  lr: 0.000021  min_lr: 0.000021  loss: 2.9589 (3.0075)  weight_decay: 0.0500 (0.0500)  time: 0.6770  data: 0.0904  max mem: 41794
Epoch: [287]  [160/312]  eta: 0:02:05  lr: 0.000021  min_lr: 0.000021  loss: 3.0675 (3.0104)  weight_decay: 0.0500 (0.0500)  time: 0.7685  data: 0.0965  max mem: 41794
Epoch: [287]  [170/312]  eta: 0:01:55  lr: 0.000020  min_lr: 0.000020  loss: 3.0143 (2.9939)  weight_decay: 0.0500 (0.0500)  time: 0.6646  data: 0.0889  max mem: 41794
Epoch: [287]  [180/312]  eta: 0:01:48  lr: 0.000020  min_lr: 0.000020  loss: 3.0873 (2.9990)  weight_decay: 0.0500 (0.0500)  time: 0.7736  data: 0.1547  max mem: 41794
Epoch: [287]  [190/312]  eta: 0:01:37  lr: 0.000020  min_lr: 0.000020  loss: 3.0833 (2.9769)  weight_decay: 0.0500 (0.0500)  time: 0.6958  data: 0.0756  max mem: 41794
Epoch: [287]  [200/312]  eta: 0:01:29  lr: 0.000020  min_lr: 0.000020  loss: 2.9818 (2.9801)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.0612  max mem: 41794
Epoch: [287]  [210/312]  eta: 0:01:22  lr: 0.000020  min_lr: 0.000020  loss: 3.1073 (2.9795)  weight_decay: 0.0500 (0.0500)  time: 0.8744  data: 0.1409  max mem: 41794
Epoch: [287]  [220/312]  eta: 0:01:13  lr: 0.000020  min_lr: 0.000020  loss: 3.2035 (2.9847)  weight_decay: 0.0500 (0.0500)  time: 0.7027  data: 0.0835  max mem: 41794
Epoch: [287]  [230/312]  eta: 0:01:05  lr: 0.000020  min_lr: 0.000020  loss: 3.2565 (2.9879)  weight_decay: 0.0500 (0.0500)  time: 0.6552  data: 0.0571  max mem: 41794
Epoch: [287]  [240/312]  eta: 0:00:57  lr: 0.000020  min_lr: 0.000020  loss: 3.2594 (2.9920)  weight_decay: 0.0500 (0.0500)  time: 0.7977  data: 0.0591  max mem: 41794
Epoch: [287]  [250/312]  eta: 0:00:48  lr: 0.000020  min_lr: 0.000020  loss: 3.2339 (2.9924)  weight_decay: 0.0500 (0.0500)  time: 0.6862  data: 0.0607  max mem: 41794
Epoch: [287]  [260/312]  eta: 0:00:41  lr: 0.000020  min_lr: 0.000020  loss: 3.2240 (2.9932)  weight_decay: 0.0500 (0.0500)  time: 0.7631  data: 0.1172  max mem: 41794
Epoch: [287]  [270/312]  eta: 0:00:32  lr: 0.000020  min_lr: 0.000020  loss: 3.1271 (2.9968)  weight_decay: 0.0500 (0.0500)  time: 0.7151  data: 0.0590  max mem: 41794
Epoch: [287]  [280/312]  eta: 0:00:25  lr: 0.000019  min_lr: 0.000019  loss: 3.0541 (2.9948)  weight_decay: 0.0500 (0.0500)  time: 0.6646  data: 0.0475  max mem: 41794
Epoch: [287]  [290/312]  eta: 0:00:17  lr: 0.000019  min_lr: 0.000019  loss: 3.0582 (2.9956)  weight_decay: 0.0500 (0.0500)  time: 0.7906  data: 0.0661  max mem: 41794
Epoch: [287]  [300/312]  eta: 0:00:09  lr: 0.000019  min_lr: 0.000019  loss: 3.2070 (2.9987)  weight_decay: 0.0500 (0.0500)  time: 0.6123  data: 0.0190  max mem: 41794
Epoch: [287]  [310/312]  eta: 0:00:01  lr: 0.000019  min_lr: 0.000019  loss: 3.0061 (2.9939)  weight_decay: 0.0500 (0.0500)  time: 0.4664  data: 0.0001  max mem: 41794
Epoch: [287]  [311/312]  eta: 0:00:00  lr: 0.000019  min_lr: 0.000019  loss: 3.0061 (2.9949)  weight_decay: 0.0500 (0.0500)  time: 0.4660  data: 0.0001  max mem: 41794
Epoch: [287] Total time: 0:03:59 (0.7677 s / it)
Averaged stats: lr: 0.000019  min_lr: 0.000019  loss: 3.0061 (2.9640)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.8272 (0.8272)  acc1: 86.0677 (86.0677)  acc5: 96.4844 (96.4844)  time: 8.8081  data: 8.6045  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0994 (1.0198)  acc1: 79.6875 (79.9680)  acc5: 95.1823 (95.2960)  time: 1.1279  data: 0.9561  max mem: 41794
Test: Total time: 0:00:10 (1.1510 s / it)
* Acc@1 80.386 Acc@5 95.310 loss 1.016
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.40%
Epoch: [288]  [  0/312]  eta: 1:22:22  lr: 0.000019  min_lr: 0.000019  loss: 1.9616 (1.9616)  weight_decay: 0.0500 (0.0500)  time: 15.8408  data: 11.6775  max mem: 41794
Epoch: [288]  [ 10/312]  eta: 0:11:44  lr: 0.000019  min_lr: 0.000019  loss: 2.8747 (2.7244)  weight_decay: 0.0500 (0.0500)  time: 2.3315  data: 1.2939  max mem: 41794
Epoch: [288]  [ 20/312]  eta: 0:07:27  lr: 0.000019  min_lr: 0.000019  loss: 2.9911 (2.8652)  weight_decay: 0.0500 (0.0500)  time: 0.8181  data: 0.1559  max mem: 41794
Epoch: [288]  [ 30/312]  eta: 0:05:40  lr: 0.000019  min_lr: 0.000019  loss: 2.9733 (2.8412)  weight_decay: 0.0500 (0.0500)  time: 0.5880  data: 0.0299  max mem: 41794
Epoch: [288]  [ 40/312]  eta: 0:04:41  lr: 0.000019  min_lr: 0.000019  loss: 2.8883 (2.9041)  weight_decay: 0.0500 (0.0500)  time: 0.5126  data: 0.0023  max mem: 41794
Epoch: [288]  [ 50/312]  eta: 0:04:17  lr: 0.000019  min_lr: 0.000019  loss: 3.2254 (2.9256)  weight_decay: 0.0500 (0.0500)  time: 0.6374  data: 0.0309  max mem: 41794
Epoch: [288]  [ 60/312]  eta: 0:03:47  lr: 0.000019  min_lr: 0.000019  loss: 3.2142 (2.9468)  weight_decay: 0.0500 (0.0500)  time: 0.6348  data: 0.0320  max mem: 41794
Epoch: [288]  [ 70/312]  eta: 0:03:39  lr: 0.000018  min_lr: 0.000018  loss: 3.1366 (2.9416)  weight_decay: 0.0500 (0.0500)  time: 0.7055  data: 0.0630  max mem: 41794
Epoch: [288]  [ 80/312]  eta: 0:03:24  lr: 0.000018  min_lr: 0.000018  loss: 2.8933 (2.9104)  weight_decay: 0.0500 (0.0500)  time: 0.8203  data: 0.0619  max mem: 41794
Epoch: [288]  [ 90/312]  eta: 0:03:09  lr: 0.000018  min_lr: 0.000018  loss: 2.9432 (2.9307)  weight_decay: 0.0500 (0.0500)  time: 0.6691  data: 0.0569  max mem: 41794
Epoch: [288]  [100/312]  eta: 0:03:02  lr: 0.000018  min_lr: 0.000018  loss: 3.1809 (2.9341)  weight_decay: 0.0500 (0.0500)  time: 0.7808  data: 0.1446  max mem: 41794
Epoch: [288]  [110/312]  eta: 0:02:47  lr: 0.000018  min_lr: 0.000018  loss: 3.1381 (2.9529)  weight_decay: 0.0500 (0.0500)  time: 0.7265  data: 0.0884  max mem: 41794
Epoch: [288]  [120/312]  eta: 0:02:39  lr: 0.000018  min_lr: 0.000018  loss: 3.2612 (2.9720)  weight_decay: 0.0500 (0.0500)  time: 0.6703  data: 0.0627  max mem: 41794
Epoch: [288]  [130/312]  eta: 0:02:31  lr: 0.000018  min_lr: 0.000018  loss: 3.0042 (2.9532)  weight_decay: 0.0500 (0.0500)  time: 0.8356  data: 0.1122  max mem: 41794
Epoch: [288]  [140/312]  eta: 0:02:18  lr: 0.000018  min_lr: 0.000018  loss: 2.6307 (2.9429)  weight_decay: 0.0500 (0.0500)  time: 0.6682  data: 0.0516  max mem: 41794
Epoch: [288]  [150/312]  eta: 0:02:11  lr: 0.000018  min_lr: 0.000018  loss: 2.6846 (2.9362)  weight_decay: 0.0500 (0.0500)  time: 0.6940  data: 0.0499  max mem: 41794
Epoch: [288]  [160/312]  eta: 0:02:02  lr: 0.000018  min_lr: 0.000018  loss: 2.8696 (2.9184)  weight_decay: 0.0500 (0.0500)  time: 0.8152  data: 0.0496  max mem: 41794
Epoch: [288]  [170/312]  eta: 0:01:52  lr: 0.000017  min_lr: 0.000017  loss: 3.1024 (2.9255)  weight_decay: 0.0500 (0.0500)  time: 0.6597  data: 0.0404  max mem: 41794
Epoch: [288]  [180/312]  eta: 0:01:45  lr: 0.000017  min_lr: 0.000017  loss: 2.9393 (2.9166)  weight_decay: 0.0500 (0.0500)  time: 0.7279  data: 0.0815  max mem: 41794
Epoch: [288]  [190/312]  eta: 0:01:35  lr: 0.000017  min_lr: 0.000017  loss: 2.9393 (2.9195)  weight_decay: 0.0500 (0.0500)  time: 0.6919  data: 0.0458  max mem: 41794
Epoch: [288]  [200/312]  eta: 0:01:28  lr: 0.000017  min_lr: 0.000017  loss: 3.1347 (2.9165)  weight_decay: 0.0500 (0.0500)  time: 0.6746  data: 0.0435  max mem: 41794
Epoch: [288]  [210/312]  eta: 0:01:20  lr: 0.000017  min_lr: 0.000017  loss: 3.0639 (2.9188)  weight_decay: 0.0500 (0.0500)  time: 0.8579  data: 0.0856  max mem: 41794
Epoch: [288]  [220/312]  eta: 0:01:11  lr: 0.000017  min_lr: 0.000017  loss: 3.0234 (2.9198)  weight_decay: 0.0500 (0.0500)  time: 0.6904  data: 0.0496  max mem: 41794
Epoch: [288]  [230/312]  eta: 0:01:04  lr: 0.000017  min_lr: 0.000017  loss: 3.0234 (2.9260)  weight_decay: 0.0500 (0.0500)  time: 0.7221  data: 0.0624  max mem: 41794
Epoch: [288]  [240/312]  eta: 0:00:56  lr: 0.000017  min_lr: 0.000017  loss: 3.1110 (2.9200)  weight_decay: 0.0500 (0.0500)  time: 0.8397  data: 0.0584  max mem: 41794
Epoch: [288]  [250/312]  eta: 0:00:48  lr: 0.000017  min_lr: 0.000017  loss: 3.2200 (2.9326)  weight_decay: 0.0500 (0.0500)  time: 0.6495  data: 0.0313  max mem: 41794
Epoch: [288]  [260/312]  eta: 0:00:40  lr: 0.000017  min_lr: 0.000017  loss: 3.2318 (2.9368)  weight_decay: 0.0500 (0.0500)  time: 0.6838  data: 0.0607  max mem: 41794
Epoch: [288]  [270/312]  eta: 0:00:32  lr: 0.000017  min_lr: 0.000017  loss: 3.0762 (2.9383)  weight_decay: 0.0500 (0.0500)  time: 0.6649  data: 0.0318  max mem: 41794
Epoch: [288]  [280/312]  eta: 0:00:24  lr: 0.000016  min_lr: 0.000016  loss: 3.1382 (2.9437)  weight_decay: 0.0500 (0.0500)  time: 0.7060  data: 0.0184  max mem: 41794
Epoch: [288]  [290/312]  eta: 0:00:17  lr: 0.000016  min_lr: 0.000016  loss: 3.0154 (2.9377)  weight_decay: 0.0500 (0.0500)  time: 0.8953  data: 0.0409  max mem: 41794
Epoch: [288]  [300/312]  eta: 0:00:09  lr: 0.000016  min_lr: 0.000016  loss: 3.0154 (2.9324)  weight_decay: 0.0500 (0.0500)  time: 0.6862  data: 0.0246  max mem: 41794
Epoch: [288]  [310/312]  eta: 0:00:01  lr: 0.000016  min_lr: 0.000016  loss: 2.9231 (2.9251)  weight_decay: 0.0500 (0.0500)  time: 0.4707  data: 0.0001  max mem: 41794
Epoch: [288]  [311/312]  eta: 0:00:00  lr: 0.000016  min_lr: 0.000016  loss: 2.7912 (2.9243)  weight_decay: 0.0500 (0.0500)  time: 0.4700  data: 0.0001  max mem: 41794
Epoch: [288] Total time: 0:03:57 (0.7611 s / it)
Averaged stats: lr: 0.000016  min_lr: 0.000016  loss: 2.7912 (2.9737)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:19  loss: 0.7736 (0.7736)  acc1: 86.3281 (86.3281)  acc5: 96.6146 (96.6146)  time: 8.8419  data: 8.6318  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0400 (0.9639)  acc1: 79.9479 (80.0480)  acc5: 95.5729 (95.2960)  time: 1.1305  data: 0.9592  max mem: 41794
Test: Total time: 0:00:10 (1.1698 s / it)
* Acc@1 80.364 Acc@5 95.246 loss 0.959
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.40%
Epoch: [289]  [  0/312]  eta: 1:27:01  lr: 0.000016  min_lr: 0.000016  loss: 3.3795 (3.3795)  weight_decay: 0.0500 (0.0500)  time: 16.7369  data: 14.6579  max mem: 41794
Epoch: [289]  [ 10/312]  eta: 0:10:18  lr: 0.000016  min_lr: 0.000016  loss: 3.2309 (3.0873)  weight_decay: 0.0500 (0.0500)  time: 2.0483  data: 1.3416  max mem: 41794
Epoch: [289]  [ 20/312]  eta: 0:07:08  lr: 0.000016  min_lr: 0.000016  loss: 3.0246 (3.0332)  weight_decay: 0.0500 (0.0500)  time: 0.7038  data: 0.1402  max mem: 41794
Epoch: [289]  [ 30/312]  eta: 0:05:26  lr: 0.000016  min_lr: 0.000016  loss: 3.2054 (3.0410)  weight_decay: 0.0500 (0.0500)  time: 0.6690  data: 0.1358  max mem: 41794
Epoch: [289]  [ 40/312]  eta: 0:04:51  lr: 0.000016  min_lr: 0.000016  loss: 3.2054 (3.0250)  weight_decay: 0.0500 (0.0500)  time: 0.6592  data: 0.1374  max mem: 41794
Epoch: [289]  [ 50/312]  eta: 0:04:29  lr: 0.000016  min_lr: 0.000016  loss: 2.9769 (2.9821)  weight_decay: 0.0500 (0.0500)  time: 0.8284  data: 0.2951  max mem: 41794
Epoch: [289]  [ 60/312]  eta: 0:03:58  lr: 0.000016  min_lr: 0.000016  loss: 2.7922 (2.9671)  weight_decay: 0.0500 (0.0500)  time: 0.6830  data: 0.1585  max mem: 41794
Epoch: [289]  [ 70/312]  eta: 0:03:46  lr: 0.000016  min_lr: 0.000016  loss: 3.1420 (2.9781)  weight_decay: 0.0500 (0.0500)  time: 0.7040  data: 0.1568  max mem: 41794
Epoch: [289]  [ 80/312]  eta: 0:03:35  lr: 0.000016  min_lr: 0.000016  loss: 3.2000 (2.9811)  weight_decay: 0.0500 (0.0500)  time: 0.8748  data: 0.2756  max mem: 41794
Epoch: [289]  [ 90/312]  eta: 0:03:21  lr: 0.000015  min_lr: 0.000015  loss: 3.1250 (2.9850)  weight_decay: 0.0500 (0.0500)  time: 0.8096  data: 0.1209  max mem: 41794
Epoch: [289]  [100/312]  eta: 0:03:13  lr: 0.000015  min_lr: 0.000015  loss: 3.1305 (2.9758)  weight_decay: 0.0500 (0.0500)  time: 0.8546  data: 0.0183  max mem: 41794
Epoch: [289]  [110/312]  eta: 0:02:57  lr: 0.000015  min_lr: 0.000015  loss: 3.0745 (2.9902)  weight_decay: 0.0500 (0.0500)  time: 0.7441  data: 0.0198  max mem: 41794
Epoch: [289]  [120/312]  eta: 0:02:45  lr: 0.000015  min_lr: 0.000015  loss: 3.1362 (2.9967)  weight_decay: 0.0500 (0.0500)  time: 0.6132  data: 0.0063  max mem: 41794
Epoch: [289]  [130/312]  eta: 0:02:34  lr: 0.000015  min_lr: 0.000015  loss: 2.7795 (2.9774)  weight_decay: 0.0500 (0.0500)  time: 0.6870  data: 0.0394  max mem: 41794
Epoch: [289]  [140/312]  eta: 0:02:23  lr: 0.000015  min_lr: 0.000015  loss: 2.7539 (2.9614)  weight_decay: 0.0500 (0.0500)  time: 0.6422  data: 0.0367  max mem: 41794
Epoch: [289]  [150/312]  eta: 0:02:14  lr: 0.000015  min_lr: 0.000015  loss: 3.2195 (2.9739)  weight_decay: 0.0500 (0.0500)  time: 0.7072  data: 0.0688  max mem: 41794
Epoch: [289]  [160/312]  eta: 0:02:05  lr: 0.000015  min_lr: 0.000015  loss: 3.1294 (2.9576)  weight_decay: 0.0500 (0.0500)  time: 0.8000  data: 0.1504  max mem: 41794
Epoch: [289]  [170/312]  eta: 0:01:55  lr: 0.000015  min_lr: 0.000015  loss: 2.6373 (2.9492)  weight_decay: 0.0500 (0.0500)  time: 0.6747  data: 0.0823  max mem: 41794
Epoch: [289]  [180/312]  eta: 0:01:47  lr: 0.000015  min_lr: 0.000015  loss: 3.2280 (2.9598)  weight_decay: 0.0500 (0.0500)  time: 0.7266  data: 0.0817  max mem: 41794
Epoch: [289]  [190/312]  eta: 0:01:37  lr: 0.000015  min_lr: 0.000015  loss: 3.2280 (2.9517)  weight_decay: 0.0500 (0.0500)  time: 0.6911  data: 0.0830  max mem: 41794
Epoch: [289]  [200/312]  eta: 0:01:29  lr: 0.000014  min_lr: 0.000014  loss: 2.9835 (2.9514)  weight_decay: 0.0500 (0.0500)  time: 0.6448  data: 0.1217  max mem: 41794
Epoch: [289]  [210/312]  eta: 0:01:21  lr: 0.000014  min_lr: 0.000014  loss: 3.1462 (2.9558)  weight_decay: 0.0500 (0.0500)  time: 0.8113  data: 0.2532  max mem: 41794
Epoch: [289]  [220/312]  eta: 0:01:12  lr: 0.000014  min_lr: 0.000014  loss: 3.1462 (2.9555)  weight_decay: 0.0500 (0.0500)  time: 0.6876  data: 0.1334  max mem: 41794
Epoch: [289]  [230/312]  eta: 0:01:05  lr: 0.000014  min_lr: 0.000014  loss: 2.9992 (2.9477)  weight_decay: 0.0500 (0.0500)  time: 0.7059  data: 0.1194  max mem: 41794
Epoch: [289]  [240/312]  eta: 0:00:56  lr: 0.000014  min_lr: 0.000014  loss: 3.1582 (2.9601)  weight_decay: 0.0500 (0.0500)  time: 0.7999  data: 0.2404  max mem: 41794
Epoch: [289]  [250/312]  eta: 0:00:48  lr: 0.000014  min_lr: 0.000014  loss: 3.1957 (2.9670)  weight_decay: 0.0500 (0.0500)  time: 0.6436  data: 0.1216  max mem: 41794
Epoch: [289]  [260/312]  eta: 0:00:40  lr: 0.000014  min_lr: 0.000014  loss: 3.1611 (2.9708)  weight_decay: 0.0500 (0.0500)  time: 0.7106  data: 0.1573  max mem: 41794
Epoch: [289]  [270/312]  eta: 0:00:32  lr: 0.000014  min_lr: 0.000014  loss: 3.1611 (2.9708)  weight_decay: 0.0500 (0.0500)  time: 0.6785  data: 0.1573  max mem: 41794
Epoch: [289]  [280/312]  eta: 0:00:24  lr: 0.000014  min_lr: 0.000014  loss: 3.2396 (2.9770)  weight_decay: 0.0500 (0.0500)  time: 0.6840  data: 0.1647  max mem: 41794
Epoch: [289]  [290/312]  eta: 0:00:17  lr: 0.000014  min_lr: 0.000014  loss: 2.8540 (2.9676)  weight_decay: 0.0500 (0.0500)  time: 0.8349  data: 0.3076  max mem: 41794
Epoch: [289]  [300/312]  eta: 0:00:09  lr: 0.000014  min_lr: 0.000014  loss: 2.9242 (2.9696)  weight_decay: 0.0500 (0.0500)  time: 0.6645  data: 0.1434  max mem: 41794
Epoch: [289]  [310/312]  eta: 0:00:01  lr: 0.000014  min_lr: 0.000014  loss: 3.0931 (2.9723)  weight_decay: 0.0500 (0.0500)  time: 0.4996  data: 0.0001  max mem: 41794
Epoch: [289]  [311/312]  eta: 0:00:00  lr: 0.000014  min_lr: 0.000014  loss: 3.1468 (2.9735)  weight_decay: 0.0500 (0.0500)  time: 0.4666  data: 0.0001  max mem: 41794
Epoch: [289] Total time: 0:03:58 (0.7657 s / it)
Averaged stats: lr: 0.000014  min_lr: 0.000014  loss: 3.1468 (2.9751)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:18  loss: 0.8875 (0.8875)  acc1: 86.3281 (86.3281)  acc5: 96.7448 (96.7448)  time: 8.7491  data: 8.5374  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1544 (1.0741)  acc1: 79.5573 (79.8720)  acc5: 95.1823 (95.3440)  time: 1.1202  data: 0.9487  max mem: 41794
Test: Total time: 0:00:10 (1.1652 s / it)
* Acc@1 80.330 Acc@5 95.266 loss 1.070
Accuracy of the model on the 50000 test images: 80.3%
Max accuracy: 80.40%
Epoch: [290]  [  0/312]  eta: 1:21:43  lr: 0.000014  min_lr: 0.000014  loss: 3.2663 (3.2663)  weight_decay: 0.0500 (0.0500)  time: 15.7159  data: 12.3269  max mem: 41794
Epoch: [290]  [ 10/312]  eta: 0:11:13  lr: 0.000013  min_lr: 0.000013  loss: 3.1161 (2.9146)  weight_decay: 0.0500 (0.0500)  time: 2.2289  data: 1.3090  max mem: 41794
Epoch: [290]  [ 20/312]  eta: 0:07:17  lr: 0.000013  min_lr: 0.000013  loss: 3.0806 (2.9666)  weight_decay: 0.0500 (0.0500)  time: 0.7891  data: 0.1290  max mem: 41794
Epoch: [290]  [ 30/312]  eta: 0:05:33  lr: 0.000013  min_lr: 0.000013  loss: 3.0891 (2.9434)  weight_decay: 0.0500 (0.0500)  time: 0.6047  data: 0.0303  max mem: 41794
Epoch: [290]  [ 40/312]  eta: 0:04:44  lr: 0.000013  min_lr: 0.000013  loss: 2.9366 (2.9026)  weight_decay: 0.0500 (0.0500)  time: 0.5694  data: 0.0108  max mem: 41794
Epoch: [290]  [ 50/312]  eta: 0:04:23  lr: 0.000013  min_lr: 0.000013  loss: 3.1300 (2.9490)  weight_decay: 0.0500 (0.0500)  time: 0.7332  data: 0.0967  max mem: 41794
Epoch: [290]  [ 60/312]  eta: 0:03:54  lr: 0.000013  min_lr: 0.000013  loss: 3.2125 (2.9641)  weight_decay: 0.0500 (0.0500)  time: 0.6889  data: 0.1163  max mem: 41794
Epoch: [290]  [ 70/312]  eta: 0:03:43  lr: 0.000013  min_lr: 0.000013  loss: 3.1523 (2.9886)  weight_decay: 0.0500 (0.0500)  time: 0.7186  data: 0.1200  max mem: 41794
Epoch: [290]  [ 80/312]  eta: 0:03:39  lr: 0.000013  min_lr: 0.000013  loss: 3.2364 (2.9909)  weight_decay: 0.0500 (0.0500)  time: 0.9983  data: 0.3849  max mem: 41794
Epoch: [290]  [ 90/312]  eta: 0:03:23  lr: 0.000013  min_lr: 0.000013  loss: 2.9880 (2.9629)  weight_decay: 0.0500 (0.0500)  time: 0.8782  data: 0.3057  max mem: 41794
Epoch: [290]  [100/312]  eta: 0:03:28  lr: 0.000013  min_lr: 0.000013  loss: 2.6431 (2.9341)  weight_decay: 0.0500 (0.0500)  time: 1.1440  data: 0.4628  max mem: 41794
Epoch: [290]  [110/312]  eta: 0:03:10  lr: 0.000013  min_lr: 0.000013  loss: 2.8279 (2.9338)  weight_decay: 0.0500 (0.0500)  time: 1.0611  data: 0.4479  max mem: 41794
Epoch: [290]  [120/312]  eta: 0:02:59  lr: 0.000013  min_lr: 0.000013  loss: 3.0768 (2.9346)  weight_decay: 0.0500 (0.0500)  time: 0.6948  data: 0.1682  max mem: 41794
Epoch: [290]  [130/312]  eta: 0:02:49  lr: 0.000013  min_lr: 0.000013  loss: 3.0886 (2.9362)  weight_decay: 0.0500 (0.0500)  time: 0.8903  data: 0.3388  max mem: 41794
Epoch: [290]  [140/312]  eta: 0:02:35  lr: 0.000012  min_lr: 0.000012  loss: 3.0299 (2.9297)  weight_decay: 0.0500 (0.0500)  time: 0.6849  data: 0.1714  max mem: 41794
Epoch: [290]  [150/312]  eta: 0:02:27  lr: 0.000012  min_lr: 0.000012  loss: 3.0963 (2.9365)  weight_decay: 0.0500 (0.0500)  time: 0.7481  data: 0.2575  max mem: 41794
Epoch: [290]  [160/312]  eta: 0:02:17  lr: 0.000012  min_lr: 0.000012  loss: 3.0153 (2.9277)  weight_decay: 0.0500 (0.0500)  time: 0.9430  data: 0.4451  max mem: 41794
Epoch: [290]  [170/312]  eta: 0:02:05  lr: 0.000012  min_lr: 0.000012  loss: 2.9729 (2.9318)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.1901  max mem: 41794
Epoch: [290]  [180/312]  eta: 0:01:56  lr: 0.000012  min_lr: 0.000012  loss: 3.0621 (2.9269)  weight_decay: 0.0500 (0.0500)  time: 0.6926  data: 0.1905  max mem: 41794
Epoch: [290]  [190/312]  eta: 0:01:45  lr: 0.000012  min_lr: 0.000012  loss: 2.8477 (2.9113)  weight_decay: 0.0500 (0.0500)  time: 0.6921  data: 0.1887  max mem: 41794
Epoch: [290]  [200/312]  eta: 0:01:36  lr: 0.000012  min_lr: 0.000012  loss: 2.8797 (2.9131)  weight_decay: 0.0500 (0.0500)  time: 0.7015  data: 0.1666  max mem: 41794
Epoch: [290]  [210/312]  eta: 0:01:28  lr: 0.000012  min_lr: 0.000012  loss: 2.9005 (2.9139)  weight_decay: 0.0500 (0.0500)  time: 0.9245  data: 0.3880  max mem: 41794
Epoch: [290]  [220/312]  eta: 0:01:18  lr: 0.000012  min_lr: 0.000012  loss: 3.2408 (2.9318)  weight_decay: 0.0500 (0.0500)  time: 0.7172  data: 0.2221  max mem: 41794
Epoch: [290]  [230/312]  eta: 0:01:09  lr: 0.000012  min_lr: 0.000012  loss: 3.2940 (2.9412)  weight_decay: 0.0500 (0.0500)  time: 0.6735  data: 0.1766  max mem: 41794
Epoch: [290]  [240/312]  eta: 0:01:01  lr: 0.000012  min_lr: 0.000012  loss: 3.2834 (2.9505)  weight_decay: 0.0500 (0.0500)  time: 0.9601  data: 0.4636  max mem: 41794
Epoch: [290]  [250/312]  eta: 0:00:52  lr: 0.000012  min_lr: 0.000012  loss: 3.2059 (2.9545)  weight_decay: 0.0500 (0.0500)  time: 0.7851  data: 0.2876  max mem: 41794
Epoch: [290]  [260/312]  eta: 0:00:44  lr: 0.000012  min_lr: 0.000012  loss: 2.9921 (2.9439)  weight_decay: 0.0500 (0.0500)  time: 0.8434  data: 0.3488  max mem: 41794
Epoch: [290]  [270/312]  eta: 0:00:35  lr: 0.000011  min_lr: 0.000011  loss: 2.8496 (2.9432)  weight_decay: 0.0500 (0.0500)  time: 0.8398  data: 0.3499  max mem: 41794
Epoch: [290]  [280/312]  eta: 0:00:27  lr: 0.000011  min_lr: 0.000011  loss: 3.0707 (2.9418)  weight_decay: 0.0500 (0.0500)  time: 0.6698  data: 0.1804  max mem: 41794
Epoch: [290]  [290/312]  eta: 0:00:18  lr: 0.000011  min_lr: 0.000011  loss: 3.0707 (2.9409)  weight_decay: 0.0500 (0.0500)  time: 0.9639  data: 0.4666  max mem: 41794
Epoch: [290]  [300/312]  eta: 0:00:10  lr: 0.000011  min_lr: 0.000011  loss: 3.0783 (2.9446)  weight_decay: 0.0500 (0.0500)  time: 0.7927  data: 0.2878  max mem: 41794
Epoch: [290]  [310/312]  eta: 0:00:01  lr: 0.000011  min_lr: 0.000011  loss: 3.1869 (2.9508)  weight_decay: 0.0500 (0.0500)  time: 0.5296  data: 0.0001  max mem: 41794
Epoch: [290]  [311/312]  eta: 0:00:00  lr: 0.000011  min_lr: 0.000011  loss: 3.1869 (2.9523)  weight_decay: 0.0500 (0.0500)  time: 0.5336  data: 0.0001  max mem: 41794
Epoch: [290] Total time: 0:04:21 (0.8384 s / it)
Averaged stats: lr: 0.000011  min_lr: 0.000011  loss: 3.1869 (2.9654)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:21  loss: 0.8893 (0.8893)  acc1: 85.8073 (85.8073)  acc5: 96.6146 (96.6146)  time: 9.0400  data: 8.7785  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1559 (1.0782)  acc1: 79.5573 (79.7920)  acc5: 95.3125 (95.1360)  time: 1.2043  data: 0.9755  max mem: 41794
Test: Total time: 0:00:10 (1.2173 s / it)
* Acc@1 80.170 Acc@5 95.186 loss 1.075
Accuracy of the model on the 50000 test images: 80.2%
Max accuracy: 80.40%
Epoch: [291]  [  0/312]  eta: 1:23:13  lr: 0.000011  min_lr: 0.000011  loss: 2.8650 (2.8650)  weight_decay: 0.0500 (0.0500)  time: 16.0042  data: 15.4404  max mem: 41794
Epoch: [291]  [ 10/312]  eta: 0:11:29  lr: 0.000011  min_lr: 0.000011  loss: 2.8286 (2.7701)  weight_decay: 0.0500 (0.0500)  time: 2.2828  data: 1.4043  max mem: 41794
Epoch: [291]  [ 20/312]  eta: 0:08:00  lr: 0.000011  min_lr: 0.000011  loss: 2.9705 (2.8476)  weight_decay: 0.0500 (0.0500)  time: 0.9287  data: 0.1088  max mem: 41794
Epoch: [291]  [ 30/312]  eta: 0:06:09  lr: 0.000011  min_lr: 0.000011  loss: 3.1695 (2.8877)  weight_decay: 0.0500 (0.0500)  time: 0.7738  data: 0.1141  max mem: 41794
Epoch: [291]  [ 40/312]  eta: 0:05:16  lr: 0.000011  min_lr: 0.000011  loss: 3.1695 (2.9314)  weight_decay: 0.0500 (0.0500)  time: 0.6535  data: 0.0340  max mem: 41794
Epoch: [291]  [ 50/312]  eta: 0:04:40  lr: 0.000011  min_lr: 0.000011  loss: 3.1498 (2.9111)  weight_decay: 0.0500 (0.0500)  time: 0.7040  data: 0.0395  max mem: 41794
Epoch: [291]  [ 60/312]  eta: 0:04:09  lr: 0.000011  min_lr: 0.000011  loss: 3.1094 (2.9187)  weight_decay: 0.0500 (0.0500)  time: 0.6400  data: 0.0121  max mem: 41794
Epoch: [291]  [ 70/312]  eta: 0:03:57  lr: 0.000011  min_lr: 0.000011  loss: 3.2721 (2.9706)  weight_decay: 0.0500 (0.0500)  time: 0.7537  data: 0.0676  max mem: 41794
Epoch: [291]  [ 80/312]  eta: 0:03:40  lr: 0.000011  min_lr: 0.000011  loss: 3.2749 (2.9825)  weight_decay: 0.0500 (0.0500)  time: 0.8306  data: 0.0690  max mem: 41794
Epoch: [291]  [ 90/312]  eta: 0:03:24  lr: 0.000011  min_lr: 0.000011  loss: 3.2749 (3.0023)  weight_decay: 0.0500 (0.0500)  time: 0.7024  data: 0.0568  max mem: 41794
Epoch: [291]  [100/312]  eta: 0:03:27  lr: 0.000010  min_lr: 0.000010  loss: 3.2401 (3.0007)  weight_decay: 0.0500 (0.0500)  time: 1.0814  data: 0.2740  max mem: 41794
Epoch: [291]  [110/312]  eta: 0:03:10  lr: 0.000010  min_lr: 0.000010  loss: 3.1085 (3.0041)  weight_decay: 0.0500 (0.0500)  time: 1.0338  data: 0.2196  max mem: 41794
Epoch: [291]  [120/312]  eta: 0:03:12  lr: 0.000010  min_lr: 0.000010  loss: 2.9819 (2.9830)  weight_decay: 0.0500 (0.0500)  time: 1.1306  data: 0.2761  max mem: 41794
Epoch: [291]  [130/312]  eta: 0:03:05  lr: 0.000010  min_lr: 0.000010  loss: 3.1137 (2.9889)  weight_decay: 0.0500 (0.0500)  time: 1.4549  data: 0.4424  max mem: 41794
Epoch: [291]  [140/312]  eta: 0:02:50  lr: 0.000010  min_lr: 0.000010  loss: 3.1794 (2.9964)  weight_decay: 0.0500 (0.0500)  time: 0.9065  data: 0.1669  max mem: 41794
Epoch: [291]  [150/312]  eta: 0:02:39  lr: 0.000010  min_lr: 0.000010  loss: 2.9941 (2.9922)  weight_decay: 0.0500 (0.0500)  time: 0.7323  data: 0.0762  max mem: 41794
Epoch: [291]  [160/312]  eta: 0:02:26  lr: 0.000010  min_lr: 0.000010  loss: 3.1295 (2.9882)  weight_decay: 0.0500 (0.0500)  time: 0.7899  data: 0.0762  max mem: 41794
Epoch: [291]  [170/312]  eta: 0:02:14  lr: 0.000010  min_lr: 0.000010  loss: 3.0116 (2.9888)  weight_decay: 0.0500 (0.0500)  time: 0.7166  data: 0.0748  max mem: 41794
Epoch: [291]  [180/312]  eta: 0:02:04  lr: 0.000010  min_lr: 0.000010  loss: 3.0116 (2.9875)  weight_decay: 0.0500 (0.0500)  time: 0.7925  data: 0.1567  max mem: 41794
Epoch: [291]  [190/312]  eta: 0:01:53  lr: 0.000010  min_lr: 0.000010  loss: 3.0182 (2.9829)  weight_decay: 0.0500 (0.0500)  time: 0.7240  data: 0.0826  max mem: 41794
Epoch: [291]  [200/312]  eta: 0:01:44  lr: 0.000010  min_lr: 0.000010  loss: 3.0364 (2.9838)  weight_decay: 0.0500 (0.0500)  time: 0.8149  data: 0.0831  max mem: 41794
Epoch: [291]  [210/312]  eta: 0:01:36  lr: 0.000010  min_lr: 0.000010  loss: 3.0364 (2.9760)  weight_decay: 0.0500 (0.0500)  time: 1.1376  data: 0.2693  max mem: 41794
Epoch: [291]  [220/312]  eta: 0:01:25  lr: 0.000010  min_lr: 0.000010  loss: 2.8756 (2.9679)  weight_decay: 0.0500 (0.0500)  time: 0.9168  data: 0.1890  max mem: 41794
Epoch: [291]  [230/312]  eta: 0:01:15  lr: 0.000010  min_lr: 0.000010  loss: 2.7093 (2.9582)  weight_decay: 0.0500 (0.0500)  time: 0.6847  data: 0.0793  max mem: 41794
Epoch: [291]  [240/312]  eta: 0:01:06  lr: 0.000010  min_lr: 0.000010  loss: 2.8117 (2.9569)  weight_decay: 0.0500 (0.0500)  time: 0.8533  data: 0.1131  max mem: 41794
Epoch: [291]  [250/312]  eta: 0:00:57  lr: 0.000009  min_lr: 0.000009  loss: 3.1378 (2.9589)  weight_decay: 0.0500 (0.0500)  time: 0.8865  data: 0.1626  max mem: 41794
Epoch: [291]  [260/312]  eta: 0:00:47  lr: 0.000009  min_lr: 0.000009  loss: 3.1500 (2.9638)  weight_decay: 0.0500 (0.0500)  time: 0.8394  data: 0.1899  max mem: 41794
Epoch: [291]  [270/312]  eta: 0:00:38  lr: 0.000009  min_lr: 0.000009  loss: 3.0626 (2.9620)  weight_decay: 0.0500 (0.0500)  time: 0.7181  data: 0.0655  max mem: 41794
Epoch: [291]  [280/312]  eta: 0:00:28  lr: 0.000009  min_lr: 0.000009  loss: 3.0626 (2.9634)  weight_decay: 0.0500 (0.0500)  time: 0.7389  data: 0.0732  max mem: 41794
Epoch: [291]  [290/312]  eta: 0:00:19  lr: 0.000009  min_lr: 0.000009  loss: 3.1350 (2.9679)  weight_decay: 0.0500 (0.0500)  time: 0.8808  data: 0.1421  max mem: 41794
Epoch: [291]  [300/312]  eta: 0:00:10  lr: 0.000009  min_lr: 0.000009  loss: 3.2308 (2.9710)  weight_decay: 0.0500 (0.0500)  time: 0.7583  data: 0.0709  max mem: 41794
Epoch: [291]  [310/312]  eta: 0:00:01  lr: 0.000009  min_lr: 0.000009  loss: 3.0444 (2.9643)  weight_decay: 0.0500 (0.0500)  time: 0.6336  data: 0.0001  max mem: 41794
Epoch: [291]  [311/312]  eta: 0:00:00  lr: 0.000009  min_lr: 0.000009  loss: 3.0444 (2.9664)  weight_decay: 0.0500 (0.0500)  time: 0.6344  data: 0.0001  max mem: 41794
Epoch: [291] Total time: 0:04:38 (0.8925 s / it)
Averaged stats: lr: 0.000009  min_lr: 0.000009  loss: 3.0444 (2.9624)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:27  loss: 0.8373 (0.8373)  acc1: 85.9375 (85.9375)  acc5: 96.7448 (96.7448)  time: 9.6772  data: 9.3949  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1062 (1.0259)  acc1: 79.6875 (79.9200)  acc5: 95.8333 (95.3760)  time: 1.2773  data: 1.0440  max mem: 41794
Test: Total time: 0:00:11 (1.2941 s / it)
* Acc@1 80.350 Acc@5 95.274 loss 1.022
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.40%
Epoch: [292]  [  0/312]  eta: 1:28:57  lr: 0.000009  min_lr: 0.000009  loss: 3.5026 (3.5026)  weight_decay: 0.0500 (0.0500)  time: 17.1069  data: 12.2055  max mem: 41794
Epoch: [292]  [ 10/312]  eta: 0:12:21  lr: 0.000009  min_lr: 0.000009  loss: 2.9201 (2.8817)  weight_decay: 0.0500 (0.0500)  time: 2.4550  data: 1.3537  max mem: 41794
Epoch: [292]  [ 20/312]  eta: 0:08:18  lr: 0.000009  min_lr: 0.000009  loss: 2.9201 (2.8730)  weight_decay: 0.0500 (0.0500)  time: 0.9376  data: 0.1734  max mem: 41794
Epoch: [292]  [ 30/312]  eta: 0:06:21  lr: 0.000009  min_lr: 0.000009  loss: 3.0958 (2.9488)  weight_decay: 0.0500 (0.0500)  time: 0.7480  data: 0.0414  max mem: 41794
Epoch: [292]  [ 40/312]  eta: 0:05:20  lr: 0.000009  min_lr: 0.000009  loss: 3.2665 (3.0337)  weight_decay: 0.0500 (0.0500)  time: 0.6248  data: 0.0035  max mem: 41794
Epoch: [292]  [ 50/312]  eta: 0:04:49  lr: 0.000009  min_lr: 0.000009  loss: 3.2665 (3.0606)  weight_decay: 0.0500 (0.0500)  time: 0.7206  data: 0.0197  max mem: 41794
Epoch: [292]  [ 60/312]  eta: 0:04:17  lr: 0.000009  min_lr: 0.000009  loss: 3.1739 (3.0556)  weight_decay: 0.0500 (0.0500)  time: 0.6987  data: 0.0188  max mem: 41794
Epoch: [292]  [ 70/312]  eta: 0:04:02  lr: 0.000009  min_lr: 0.000009  loss: 3.2099 (3.0689)  weight_decay: 0.0500 (0.0500)  time: 0.7329  data: 0.0379  max mem: 41794
Epoch: [292]  [ 80/312]  eta: 0:03:45  lr: 0.000009  min_lr: 0.000009  loss: 3.1206 (3.0249)  weight_decay: 0.0500 (0.0500)  time: 0.8252  data: 0.0379  max mem: 41794
Epoch: [292]  [ 90/312]  eta: 0:03:30  lr: 0.000008  min_lr: 0.000008  loss: 2.8138 (2.9943)  weight_decay: 0.0500 (0.0500)  time: 0.7542  data: 0.0671  max mem: 41794
Epoch: [292]  [100/312]  eta: 0:03:19  lr: 0.000008  min_lr: 0.000008  loss: 2.9399 (3.0046)  weight_decay: 0.0500 (0.0500)  time: 0.8160  data: 0.1067  max mem: 41794
Epoch: [292]  [110/312]  eta: 0:03:04  lr: 0.000008  min_lr: 0.000008  loss: 3.0446 (2.9916)  weight_decay: 0.0500 (0.0500)  time: 0.7545  data: 0.0404  max mem: 41794
Epoch: [292]  [120/312]  eta: 0:02:55  lr: 0.000008  min_lr: 0.000008  loss: 2.9489 (2.9727)  weight_decay: 0.0500 (0.0500)  time: 0.7690  data: 0.0361  max mem: 41794
Epoch: [292]  [130/312]  eta: 0:02:44  lr: 0.000008  min_lr: 0.000008  loss: 2.9489 (2.9723)  weight_decay: 0.0500 (0.0500)  time: 0.8748  data: 0.0437  max mem: 41794
Epoch: [292]  [140/312]  eta: 0:02:32  lr: 0.000008  min_lr: 0.000008  loss: 3.0747 (2.9671)  weight_decay: 0.0500 (0.0500)  time: 0.7129  data: 0.0173  max mem: 41794
Epoch: [292]  [150/312]  eta: 0:02:24  lr: 0.000008  min_lr: 0.000008  loss: 3.0912 (2.9652)  weight_decay: 0.0500 (0.0500)  time: 0.7882  data: 0.0446  max mem: 41794
Epoch: [292]  [160/312]  eta: 0:02:14  lr: 0.000008  min_lr: 0.000008  loss: 3.0912 (2.9720)  weight_decay: 0.0500 (0.0500)  time: 0.8874  data: 0.0396  max mem: 41794
Epoch: [292]  [170/312]  eta: 0:02:04  lr: 0.000008  min_lr: 0.000008  loss: 3.1970 (2.9775)  weight_decay: 0.0500 (0.0500)  time: 0.7432  data: 0.0321  max mem: 41794
Epoch: [292]  [180/312]  eta: 0:01:55  lr: 0.000008  min_lr: 0.000008  loss: 3.0754 (2.9710)  weight_decay: 0.0500 (0.0500)  time: 0.7954  data: 0.0688  max mem: 41794
Epoch: [292]  [190/312]  eta: 0:01:45  lr: 0.000008  min_lr: 0.000008  loss: 3.0215 (2.9679)  weight_decay: 0.0500 (0.0500)  time: 0.7535  data: 0.0413  max mem: 41794
Epoch: [292]  [200/312]  eta: 0:01:36  lr: 0.000008  min_lr: 0.000008  loss: 3.1689 (2.9833)  weight_decay: 0.0500 (0.0500)  time: 0.7371  data: 0.0138  max mem: 41794
Epoch: [292]  [210/312]  eta: 0:01:28  lr: 0.000008  min_lr: 0.000008  loss: 3.1858 (2.9820)  weight_decay: 0.0500 (0.0500)  time: 0.8876  data: 0.0753  max mem: 41794
Epoch: [292]  [220/312]  eta: 0:01:18  lr: 0.000008  min_lr: 0.000008  loss: 3.1858 (2.9857)  weight_decay: 0.0500 (0.0500)  time: 0.7692  data: 0.0621  max mem: 41794
Epoch: [292]  [230/312]  eta: 0:01:10  lr: 0.000008  min_lr: 0.000008  loss: 2.8539 (2.9682)  weight_decay: 0.0500 (0.0500)  time: 0.7698  data: 0.0309  max mem: 41794
Epoch: [292]  [240/312]  eta: 0:01:01  lr: 0.000008  min_lr: 0.000008  loss: 2.8539 (2.9720)  weight_decay: 0.0500 (0.0500)  time: 0.8516  data: 0.0937  max mem: 41794
Epoch: [292]  [250/312]  eta: 0:00:52  lr: 0.000008  min_lr: 0.000008  loss: 2.7233 (2.9547)  weight_decay: 0.0500 (0.0500)  time: 0.7398  data: 0.0634  max mem: 41794
Epoch: [292]  [260/312]  eta: 0:00:44  lr: 0.000007  min_lr: 0.000007  loss: 2.6465 (2.9467)  weight_decay: 0.0500 (0.0500)  time: 0.7622  data: 0.0685  max mem: 41794
Epoch: [292]  [270/312]  eta: 0:00:35  lr: 0.000007  min_lr: 0.000007  loss: 3.0773 (2.9516)  weight_decay: 0.0500 (0.0500)  time: 0.7235  data: 0.0685  max mem: 41794
Epoch: [292]  [280/312]  eta: 0:00:26  lr: 0.000007  min_lr: 0.000007  loss: 3.1425 (2.9520)  weight_decay: 0.0500 (0.0500)  time: 0.7285  data: 0.0832  max mem: 41794
Epoch: [292]  [290/312]  eta: 0:00:18  lr: 0.000007  min_lr: 0.000007  loss: 2.9381 (2.9488)  weight_decay: 0.0500 (0.0500)  time: 0.8704  data: 0.1795  max mem: 41794
Epoch: [292]  [300/312]  eta: 0:00:09  lr: 0.000007  min_lr: 0.000007  loss: 3.1047 (2.9495)  weight_decay: 0.0500 (0.0500)  time: 0.7629  data: 0.0967  max mem: 41794
Epoch: [292]  [310/312]  eta: 0:00:01  lr: 0.000007  min_lr: 0.000007  loss: 3.0256 (2.9477)  weight_decay: 0.0500 (0.0500)  time: 0.6403  data: 0.0001  max mem: 41794
Epoch: [292]  [311/312]  eta: 0:00:00  lr: 0.000007  min_lr: 0.000007  loss: 3.0982 (2.9485)  weight_decay: 0.0500 (0.0500)  time: 0.6423  data: 0.0001  max mem: 41794
Epoch: [292] Total time: 0:04:19 (0.8325 s / it)
Averaged stats: lr: 0.000007  min_lr: 0.000007  loss: 3.0982 (2.9805)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:26  loss: 0.8423 (0.8423)  acc1: 86.3281 (86.3281)  acc5: 96.7448 (96.7448)  time: 9.6401  data: 9.3517  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1132 (1.0336)  acc1: 79.5573 (80.0320)  acc5: 95.8333 (95.3600)  time: 1.2718  data: 1.0392  max mem: 41794
Test: Total time: 0:00:11 (1.2886 s / it)
* Acc@1 80.382 Acc@5 95.308 loss 1.029
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.40%
Epoch: [293]  [  0/312]  eta: 1:26:11  lr: 0.000007  min_lr: 0.000007  loss: 2.4522 (2.4522)  weight_decay: 0.0500 (0.0500)  time: 16.5764  data: 15.5075  max mem: 41794
Epoch: [293]  [ 10/312]  eta: 0:11:42  lr: 0.000007  min_lr: 0.000007  loss: 2.7646 (2.8281)  weight_decay: 0.0500 (0.0500)  time: 2.3253  data: 1.4353  max mem: 41794
Epoch: [293]  [ 20/312]  eta: 0:07:56  lr: 0.000007  min_lr: 0.000007  loss: 2.8609 (2.8965)  weight_decay: 0.0500 (0.0500)  time: 0.8832  data: 0.0954  max mem: 41794
Epoch: [293]  [ 30/312]  eta: 0:06:07  lr: 0.000007  min_lr: 0.000007  loss: 3.0118 (2.9194)  weight_decay: 0.0500 (0.0500)  time: 0.7430  data: 0.0823  max mem: 41794
Epoch: [293]  [ 40/312]  eta: 0:05:19  lr: 0.000007  min_lr: 0.000007  loss: 3.2200 (2.9681)  weight_decay: 0.0500 (0.0500)  time: 0.6924  data: 0.0366  max mem: 41794
Epoch: [293]  [ 50/312]  eta: 0:04:50  lr: 0.000007  min_lr: 0.000007  loss: 3.2200 (2.9898)  weight_decay: 0.0500 (0.0500)  time: 0.8082  data: 0.0889  max mem: 41794
Epoch: [293]  [ 60/312]  eta: 0:04:18  lr: 0.000007  min_lr: 0.000007  loss: 3.2194 (2.9806)  weight_decay: 0.0500 (0.0500)  time: 0.7189  data: 0.0541  max mem: 41794
Epoch: [293]  [ 70/312]  eta: 0:04:04  lr: 0.000007  min_lr: 0.000007  loss: 3.0442 (2.9718)  weight_decay: 0.0500 (0.0500)  time: 0.7497  data: 0.0870  max mem: 41794
Epoch: [293]  [ 80/312]  eta: 0:03:50  lr: 0.000007  min_lr: 0.000007  loss: 2.9780 (2.9676)  weight_decay: 0.0500 (0.0500)  time: 0.9007  data: 0.1734  max mem: 41794
Epoch: [293]  [ 90/312]  eta: 0:03:31  lr: 0.000007  min_lr: 0.000007  loss: 2.9106 (2.9435)  weight_decay: 0.0500 (0.0500)  time: 0.7476  data: 0.0890  max mem: 41794
Epoch: [293]  [100/312]  eta: 0:03:22  lr: 0.000007  min_lr: 0.000007  loss: 2.9949 (2.9409)  weight_decay: 0.0500 (0.0500)  time: 0.7933  data: 0.0565  max mem: 41794
Epoch: [293]  [110/312]  eta: 0:03:06  lr: 0.000007  min_lr: 0.000007  loss: 3.1934 (2.9494)  weight_decay: 0.0500 (0.0500)  time: 0.7842  data: 0.0606  max mem: 41794
Epoch: [293]  [120/312]  eta: 0:02:57  lr: 0.000007  min_lr: 0.000007  loss: 3.1664 (2.9551)  weight_decay: 0.0500 (0.0500)  time: 0.7848  data: 0.0675  max mem: 41794
Epoch: [293]  [130/312]  eta: 0:02:47  lr: 0.000006  min_lr: 0.000006  loss: 3.1664 (2.9704)  weight_decay: 0.0500 (0.0500)  time: 0.9262  data: 0.1060  max mem: 41794
Epoch: [293]  [140/312]  eta: 0:02:35  lr: 0.000006  min_lr: 0.000006  loss: 3.0559 (2.9588)  weight_decay: 0.0500 (0.0500)  time: 0.7642  data: 0.0755  max mem: 41794
Epoch: [293]  [150/312]  eta: 0:02:26  lr: 0.000006  min_lr: 0.000006  loss: 3.0559 (2.9661)  weight_decay: 0.0500 (0.0500)  time: 0.7984  data: 0.1164  max mem: 41794
Epoch: [293]  [160/312]  eta: 0:02:16  lr: 0.000006  min_lr: 0.000006  loss: 3.2744 (2.9794)  weight_decay: 0.0500 (0.0500)  time: 0.8799  data: 0.0880  max mem: 41794
Epoch: [293]  [170/312]  eta: 0:02:06  lr: 0.000006  min_lr: 0.000006  loss: 3.1843 (2.9742)  weight_decay: 0.0500 (0.0500)  time: 0.7970  data: 0.0821  max mem: 41794
Epoch: [293]  [180/312]  eta: 0:01:57  lr: 0.000006  min_lr: 0.000006  loss: 3.1927 (2.9795)  weight_decay: 0.0500 (0.0500)  time: 0.8172  data: 0.1128  max mem: 41794
Epoch: [293]  [190/312]  eta: 0:01:47  lr: 0.000006  min_lr: 0.000006  loss: 3.0358 (2.9665)  weight_decay: 0.0500 (0.0500)  time: 0.7489  data: 0.0512  max mem: 41794
Epoch: [293]  [200/312]  eta: 0:01:38  lr: 0.000006  min_lr: 0.000006  loss: 3.0178 (2.9736)  weight_decay: 0.0500 (0.0500)  time: 0.7357  data: 0.0559  max mem: 41794
Epoch: [293]  [210/312]  eta: 0:01:29  lr: 0.000006  min_lr: 0.000006  loss: 2.8791 (2.9685)  weight_decay: 0.0500 (0.0500)  time: 0.8962  data: 0.0961  max mem: 41794
Epoch: [293]  [220/312]  eta: 0:01:19  lr: 0.000006  min_lr: 0.000006  loss: 2.8791 (2.9714)  weight_decay: 0.0500 (0.0500)  time: 0.7868  data: 0.0712  max mem: 41794
Epoch: [293]  [230/312]  eta: 0:01:10  lr: 0.000006  min_lr: 0.000006  loss: 2.9393 (2.9625)  weight_decay: 0.0500 (0.0500)  time: 0.7082  data: 0.0731  max mem: 41794
Epoch: [293]  [240/312]  eta: 0:01:02  lr: 0.000006  min_lr: 0.000006  loss: 3.1815 (2.9678)  weight_decay: 0.0500 (0.0500)  time: 0.8376  data: 0.1550  max mem: 41794
Epoch: [293]  [250/312]  eta: 0:00:53  lr: 0.000006  min_lr: 0.000006  loss: 3.1827 (2.9676)  weight_decay: 0.0500 (0.0500)  time: 0.8138  data: 0.1375  max mem: 41794
Epoch: [293]  [260/312]  eta: 0:00:44  lr: 0.000006  min_lr: 0.000006  loss: 3.0151 (2.9654)  weight_decay: 0.0500 (0.0500)  time: 0.7883  data: 0.0843  max mem: 41794
Epoch: [293]  [270/312]  eta: 0:00:35  lr: 0.000006  min_lr: 0.000006  loss: 3.0891 (2.9711)  weight_decay: 0.0500 (0.0500)  time: 0.7497  data: 0.0782  max mem: 41794
Epoch: [293]  [280/312]  eta: 0:00:27  lr: 0.000006  min_lr: 0.000006  loss: 3.2123 (2.9662)  weight_decay: 0.0500 (0.0500)  time: 0.8517  data: 0.0949  max mem: 41794
Epoch: [293]  [290/312]  eta: 0:00:18  lr: 0.000006  min_lr: 0.000006  loss: 3.0837 (2.9668)  weight_decay: 0.0500 (0.0500)  time: 0.8863  data: 0.0586  max mem: 41794
Epoch: [293]  [300/312]  eta: 0:00:10  lr: 0.000006  min_lr: 0.000006  loss: 3.1066 (2.9691)  weight_decay: 0.0500 (0.0500)  time: 0.6905  data: 0.0029  max mem: 41794
Epoch: [293]  [310/312]  eta: 0:00:01  lr: 0.000006  min_lr: 0.000006  loss: 2.8102 (2.9613)  weight_decay: 0.0500 (0.0500)  time: 0.6341  data: 0.0029  max mem: 41794
Epoch: [293]  [311/312]  eta: 0:00:00  lr: 0.000006  min_lr: 0.000006  loss: 2.8102 (2.9629)  weight_decay: 0.0500 (0.0500)  time: 0.6348  data: 0.0028  max mem: 41794
Epoch: [293] Total time: 0:04:24 (0.8466 s / it)
Averaged stats: lr: 0.000006  min_lr: 0.000006  loss: 2.8102 (2.9578)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:37  loss: 0.8198 (0.8198)  acc1: 86.3281 (86.3281)  acc5: 97.0052 (97.0052)  time: 10.8297  data: 10.5397  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0884 (1.0088)  acc1: 79.5573 (79.9040)  acc5: 95.3125 (95.3440)  time: 1.4038  data: 1.1712  max mem: 41794
Test: Total time: 0:00:12 (1.4159 s / it)
* Acc@1 80.424 Acc@5 95.332 loss 1.004
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.42%
Epoch: [294]  [  0/312]  eta: 1:27:13  lr: 0.000006  min_lr: 0.000006  loss: 3.2003 (3.2003)  weight_decay: 0.0500 (0.0500)  time: 16.7746  data: 16.1496  max mem: 41794
Epoch: [294]  [ 10/312]  eta: 0:11:40  lr: 0.000005  min_lr: 0.000005  loss: 3.0502 (2.9578)  weight_decay: 0.0500 (0.0500)  time: 2.3190  data: 1.4687  max mem: 41794
Epoch: [294]  [ 20/312]  eta: 0:07:47  lr: 0.000005  min_lr: 0.000005  loss: 2.7179 (2.8298)  weight_decay: 0.0500 (0.0500)  time: 0.8423  data: 0.0825  max mem: 41794
Epoch: [294]  [ 30/312]  eta: 0:06:00  lr: 0.000005  min_lr: 0.000005  loss: 2.9319 (2.8889)  weight_decay: 0.0500 (0.0500)  time: 0.7049  data: 0.0830  max mem: 41794
Epoch: [294]  [ 40/312]  eta: 0:05:16  lr: 0.000005  min_lr: 0.000005  loss: 2.9664 (2.8619)  weight_decay: 0.0500 (0.0500)  time: 0.7031  data: 0.0682  max mem: 41794
Epoch: [294]  [ 50/312]  eta: 0:04:49  lr: 0.000005  min_lr: 0.000005  loss: 2.9084 (2.8731)  weight_decay: 0.0500 (0.0500)  time: 0.8337  data: 0.1923  max mem: 41794
Epoch: [294]  [ 60/312]  eta: 0:04:17  lr: 0.000005  min_lr: 0.000005  loss: 2.9500 (2.8689)  weight_decay: 0.0500 (0.0500)  time: 0.7299  data: 0.1268  max mem: 41794
Epoch: [294]  [ 70/312]  eta: 0:04:01  lr: 0.000005  min_lr: 0.000005  loss: 3.0387 (2.8736)  weight_decay: 0.0500 (0.0500)  time: 0.7349  data: 0.1349  max mem: 41794
Epoch: [294]  [ 80/312]  eta: 0:03:51  lr: 0.000005  min_lr: 0.000005  loss: 3.0683 (2.8623)  weight_decay: 0.0500 (0.0500)  time: 0.9349  data: 0.3291  max mem: 41794
Epoch: [294]  [ 90/312]  eta: 0:03:32  lr: 0.000005  min_lr: 0.000005  loss: 2.8441 (2.8637)  weight_decay: 0.0500 (0.0500)  time: 0.7983  data: 0.1965  max mem: 41794
Epoch: [294]  [100/312]  eta: 0:03:21  lr: 0.000005  min_lr: 0.000005  loss: 2.9686 (2.8635)  weight_decay: 0.0500 (0.0500)  time: 0.7437  data: 0.1211  max mem: 41794
Epoch: [294]  [110/312]  eta: 0:03:05  lr: 0.000005  min_lr: 0.000005  loss: 3.0712 (2.8741)  weight_decay: 0.0500 (0.0500)  time: 0.7468  data: 0.1224  max mem: 41794
Epoch: [294]  [120/312]  eta: 0:02:55  lr: 0.000005  min_lr: 0.000005  loss: 3.0245 (2.8659)  weight_decay: 0.0500 (0.0500)  time: 0.7411  data: 0.1086  max mem: 41794
Epoch: [294]  [130/312]  eta: 0:02:47  lr: 0.000005  min_lr: 0.000005  loss: 2.6590 (2.8495)  weight_decay: 0.0500 (0.0500)  time: 0.9200  data: 0.1708  max mem: 41794
Epoch: [294]  [140/312]  eta: 0:02:34  lr: 0.000005  min_lr: 0.000005  loss: 2.7456 (2.8503)  weight_decay: 0.0500 (0.0500)  time: 0.7957  data: 0.0854  max mem: 41794
Epoch: [294]  [150/312]  eta: 0:02:26  lr: 0.000005  min_lr: 0.000005  loss: 2.9688 (2.8495)  weight_decay: 0.0500 (0.0500)  time: 0.8140  data: 0.1110  max mem: 41794
Epoch: [294]  [160/312]  eta: 0:02:15  lr: 0.000005  min_lr: 0.000005  loss: 3.0059 (2.8532)  weight_decay: 0.0500 (0.0500)  time: 0.8286  data: 0.0900  max mem: 41794
Epoch: [294]  [170/312]  eta: 0:02:05  lr: 0.000005  min_lr: 0.000005  loss: 3.0340 (2.8620)  weight_decay: 0.0500 (0.0500)  time: 0.7399  data: 0.0877  max mem: 41794
Epoch: [294]  [180/312]  eta: 0:01:57  lr: 0.000005  min_lr: 0.000005  loss: 2.9779 (2.8596)  weight_decay: 0.0500 (0.0500)  time: 0.8956  data: 0.1699  max mem: 41794
Epoch: [294]  [190/312]  eta: 0:01:46  lr: 0.000005  min_lr: 0.000005  loss: 2.8685 (2.8623)  weight_decay: 0.0500 (0.0500)  time: 0.7846  data: 0.0842  max mem: 41794
Epoch: [294]  [200/312]  eta: 0:01:37  lr: 0.000005  min_lr: 0.000005  loss: 2.8031 (2.8625)  weight_decay: 0.0500 (0.0500)  time: 0.7367  data: 0.0435  max mem: 41794
Epoch: [294]  [210/312]  eta: 0:01:28  lr: 0.000005  min_lr: 0.000005  loss: 2.9172 (2.8693)  weight_decay: 0.0500 (0.0500)  time: 0.8269  data: 0.0761  max mem: 41794
Epoch: [294]  [220/312]  eta: 0:01:19  lr: 0.000005  min_lr: 0.000005  loss: 3.0512 (2.8741)  weight_decay: 0.0500 (0.0500)  time: 0.7289  data: 0.0715  max mem: 41794
Epoch: [294]  [230/312]  eta: 0:01:10  lr: 0.000004  min_lr: 0.000004  loss: 2.9520 (2.8780)  weight_decay: 0.0500 (0.0500)  time: 0.7837  data: 0.0933  max mem: 41794
Epoch: [294]  [240/312]  eta: 0:01:01  lr: 0.000004  min_lr: 0.000004  loss: 3.0621 (2.8820)  weight_decay: 0.0500 (0.0500)  time: 0.8182  data: 0.0665  max mem: 41794
Epoch: [294]  [250/312]  eta: 0:00:53  lr: 0.000004  min_lr: 0.000004  loss: 3.0621 (2.8843)  weight_decay: 0.0500 (0.0500)  time: 0.8327  data: 0.0866  max mem: 41794
Epoch: [294]  [260/312]  eta: 0:00:44  lr: 0.000004  min_lr: 0.000004  loss: 3.1688 (2.8898)  weight_decay: 0.0500 (0.0500)  time: 0.8983  data: 0.1112  max mem: 41794
Epoch: [294]  [270/312]  eta: 0:00:35  lr: 0.000004  min_lr: 0.000004  loss: 3.1731 (2.8940)  weight_decay: 0.0500 (0.0500)  time: 0.7400  data: 0.0370  max mem: 41794
Epoch: [294]  [280/312]  eta: 0:00:27  lr: 0.000004  min_lr: 0.000004  loss: 3.1981 (2.8988)  weight_decay: 0.0500 (0.0500)  time: 0.7238  data: 0.0604  max mem: 41794
Epoch: [294]  [290/312]  eta: 0:00:18  lr: 0.000004  min_lr: 0.000004  loss: 3.1981 (2.9040)  weight_decay: 0.0500 (0.0500)  time: 0.8143  data: 0.0977  max mem: 41794
Epoch: [294]  [300/312]  eta: 0:00:10  lr: 0.000004  min_lr: 0.000004  loss: 2.7638 (2.8932)  weight_decay: 0.0500 (0.0500)  time: 0.7127  data: 0.0392  max mem: 41794
Epoch: [294]  [310/312]  eta: 0:00:01  lr: 0.000004  min_lr: 0.000004  loss: 2.7814 (2.8916)  weight_decay: 0.0500 (0.0500)  time: 0.6404  data: 0.0001  max mem: 41794
Epoch: [294]  [311/312]  eta: 0:00:00  lr: 0.000004  min_lr: 0.000004  loss: 2.7814 (2.8930)  weight_decay: 0.0500 (0.0500)  time: 0.6388  data: 0.0001  max mem: 41794
Epoch: [294] Total time: 0:04:22 (0.8415 s / it)
Averaged stats: lr: 0.000004  min_lr: 0.000004  loss: 2.7814 (2.9485)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:31  loss: 0.8194 (0.8194)  acc1: 86.0677 (86.0677)  acc5: 96.8750 (96.8750)  time: 10.1233  data: 9.8668  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0865 (1.0080)  acc1: 79.4271 (80.0480)  acc5: 95.8333 (95.4720)  time: 1.3373  data: 1.1080  max mem: 41794
Test: Total time: 0:00:12 (1.3727 s / it)
* Acc@1 80.360 Acc@5 95.314 loss 1.004
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.42%
Epoch: [295]  [  0/312]  eta: 1:27:15  lr: 0.000004  min_lr: 0.000004  loss: 3.0244 (3.0244)  weight_decay: 0.0500 (0.0500)  time: 16.7800  data: 12.8963  max mem: 41794
Epoch: [295]  [ 10/312]  eta: 0:11:59  lr: 0.000004  min_lr: 0.000004  loss: 3.0244 (2.8933)  weight_decay: 0.0500 (0.0500)  time: 2.3837  data: 1.4002  max mem: 41794
Epoch: [295]  [ 20/312]  eta: 0:08:13  lr: 0.000004  min_lr: 0.000004  loss: 2.8056 (2.8840)  weight_decay: 0.0500 (0.0500)  time: 0.9350  data: 0.1887  max mem: 41794
Epoch: [295]  [ 30/312]  eta: 0:06:16  lr: 0.000004  min_lr: 0.000004  loss: 2.9478 (2.9352)  weight_decay: 0.0500 (0.0500)  time: 0.7580  data: 0.0636  max mem: 41794
Epoch: [295]  [ 40/312]  eta: 0:05:20  lr: 0.000004  min_lr: 0.000004  loss: 3.0043 (2.9487)  weight_decay: 0.0500 (0.0500)  time: 0.6418  data: 0.0398  max mem: 41794
Epoch: [295]  [ 50/312]  eta: 0:04:59  lr: 0.000004  min_lr: 0.000004  loss: 3.1099 (2.9711)  weight_decay: 0.0500 (0.0500)  time: 0.8418  data: 0.1657  max mem: 41794
Epoch: [295]  [ 60/312]  eta: 0:04:25  lr: 0.000004  min_lr: 0.000004  loss: 3.1742 (2.9498)  weight_decay: 0.0500 (0.0500)  time: 0.7999  data: 0.1265  max mem: 41794
Epoch: [295]  [ 70/312]  eta: 0:04:09  lr: 0.000004  min_lr: 0.000004  loss: 2.8287 (2.9124)  weight_decay: 0.0500 (0.0500)  time: 0.7496  data: 0.1213  max mem: 41794
Epoch: [295]  [ 80/312]  eta: 0:03:49  lr: 0.000004  min_lr: 0.000004  loss: 2.8287 (2.9148)  weight_decay: 0.0500 (0.0500)  time: 0.7924  data: 0.1242  max mem: 41794
Epoch: [295]  [ 90/312]  eta: 0:03:34  lr: 0.000004  min_lr: 0.000004  loss: 3.0512 (2.9189)  weight_decay: 0.0500 (0.0500)  time: 0.7419  data: 0.0985  max mem: 41794
Epoch: [295]  [100/312]  eta: 0:03:24  lr: 0.000004  min_lr: 0.000004  loss: 3.0512 (2.9227)  weight_decay: 0.0500 (0.0500)  time: 0.8739  data: 0.1969  max mem: 41794
Epoch: [295]  [110/312]  eta: 0:03:08  lr: 0.000004  min_lr: 0.000004  loss: 3.1256 (2.9285)  weight_decay: 0.0500 (0.0500)  time: 0.7803  data: 0.1041  max mem: 41794
Epoch: [295]  [120/312]  eta: 0:02:59  lr: 0.000004  min_lr: 0.000004  loss: 3.1584 (2.9341)  weight_decay: 0.0500 (0.0500)  time: 0.7682  data: 0.0999  max mem: 41794
Epoch: [295]  [130/312]  eta: 0:02:49  lr: 0.000004  min_lr: 0.000004  loss: 3.1763 (2.9536)  weight_decay: 0.0500 (0.0500)  time: 0.9318  data: 0.1971  max mem: 41794
Epoch: [295]  [140/312]  eta: 0:02:36  lr: 0.000004  min_lr: 0.000004  loss: 3.1711 (2.9526)  weight_decay: 0.0500 (0.0500)  time: 0.7655  data: 0.0998  max mem: 41794
Epoch: [295]  [150/312]  eta: 0:02:27  lr: 0.000004  min_lr: 0.000004  loss: 2.9906 (2.9513)  weight_decay: 0.0500 (0.0500)  time: 0.7587  data: 0.0830  max mem: 41794
Epoch: [295]  [160/312]  eta: 0:02:16  lr: 0.000004  min_lr: 0.000004  loss: 3.0500 (2.9582)  weight_decay: 0.0500 (0.0500)  time: 0.8389  data: 0.0870  max mem: 41794
Epoch: [295]  [170/312]  eta: 0:02:06  lr: 0.000003  min_lr: 0.000003  loss: 3.1430 (2.9664)  weight_decay: 0.0500 (0.0500)  time: 0.7302  data: 0.0564  max mem: 41794
Epoch: [295]  [180/312]  eta: 0:01:57  lr: 0.000003  min_lr: 0.000003  loss: 3.0316 (2.9676)  weight_decay: 0.0500 (0.0500)  time: 0.7911  data: 0.1260  max mem: 41794
Epoch: [295]  [190/312]  eta: 0:01:46  lr: 0.000003  min_lr: 0.000003  loss: 3.0962 (2.9676)  weight_decay: 0.0500 (0.0500)  time: 0.7427  data: 0.0742  max mem: 41794
Epoch: [295]  [200/312]  eta: 0:01:38  lr: 0.000003  min_lr: 0.000003  loss: 3.2129 (2.9721)  weight_decay: 0.0500 (0.0500)  time: 0.7727  data: 0.0642  max mem: 41794
Epoch: [295]  [210/312]  eta: 0:01:29  lr: 0.000003  min_lr: 0.000003  loss: 3.1304 (2.9693)  weight_decay: 0.0500 (0.0500)  time: 0.9271  data: 0.1446  max mem: 41794
Epoch: [295]  [220/312]  eta: 0:01:19  lr: 0.000003  min_lr: 0.000003  loss: 3.0903 (2.9705)  weight_decay: 0.0500 (0.0500)  time: 0.7549  data: 0.0829  max mem: 41794
Epoch: [295]  [230/312]  eta: 0:01:11  lr: 0.000003  min_lr: 0.000003  loss: 3.2346 (2.9784)  weight_decay: 0.0500 (0.0500)  time: 0.7655  data: 0.0894  max mem: 41794
Epoch: [295]  [240/312]  eta: 0:01:02  lr: 0.000003  min_lr: 0.000003  loss: 3.1121 (2.9757)  weight_decay: 0.0500 (0.0500)  time: 0.8203  data: 0.0932  max mem: 41794
Epoch: [295]  [250/312]  eta: 0:00:53  lr: 0.000003  min_lr: 0.000003  loss: 2.8367 (2.9603)  weight_decay: 0.0500 (0.0500)  time: 0.7142  data: 0.0557  max mem: 41794
Epoch: [295]  [260/312]  eta: 0:00:44  lr: 0.000003  min_lr: 0.000003  loss: 2.7678 (2.9513)  weight_decay: 0.0500 (0.0500)  time: 0.8093  data: 0.1066  max mem: 41794
Epoch: [295]  [270/312]  eta: 0:00:35  lr: 0.000003  min_lr: 0.000003  loss: 2.9626 (2.9521)  weight_decay: 0.0500 (0.0500)  time: 0.7502  data: 0.0601  max mem: 41794
Epoch: [295]  [280/312]  eta: 0:00:27  lr: 0.000003  min_lr: 0.000003  loss: 3.0889 (2.9597)  weight_decay: 0.0500 (0.0500)  time: 0.7424  data: 0.0748  max mem: 41794
Epoch: [295]  [290/312]  eta: 0:00:18  lr: 0.000003  min_lr: 0.000003  loss: 3.0735 (2.9568)  weight_decay: 0.0500 (0.0500)  time: 0.8689  data: 0.1391  max mem: 41794
Epoch: [295]  [300/312]  eta: 0:00:10  lr: 0.000003  min_lr: 0.000003  loss: 3.0220 (2.9583)  weight_decay: 0.0500 (0.0500)  time: 0.7454  data: 0.0676  max mem: 41794
Epoch: [295]  [310/312]  eta: 0:00:01  lr: 0.000003  min_lr: 0.000003  loss: 3.0239 (2.9608)  weight_decay: 0.0500 (0.0500)  time: 0.6371  data: 0.0001  max mem: 41794
Epoch: [295]  [311/312]  eta: 0:00:00  lr: 0.000003  min_lr: 0.000003  loss: 3.0239 (2.9619)  weight_decay: 0.0500 (0.0500)  time: 0.6358  data: 0.0001  max mem: 41794
Epoch: [295] Total time: 0:04:22 (0.8429 s / it)
Averaged stats: lr: 0.000003  min_lr: 0.000003  loss: 3.0239 (2.9633)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:33  loss: 0.8690 (0.8690)  acc1: 86.0677 (86.0677)  acc5: 96.4844 (96.4844)  time: 10.3847  data: 10.1314  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1452 (1.0647)  acc1: 79.2969 (79.9200)  acc5: 95.5729 (95.2640)  time: 1.3560  data: 1.1275  max mem: 41794
Test: Total time: 0:00:12 (1.3696 s / it)
* Acc@1 80.302 Acc@5 95.266 loss 1.060
Accuracy of the model on the 50000 test images: 80.3%
Max accuracy: 80.42%
Epoch: [296]  [  0/312]  eta: 1:26:26  lr: 0.000003  min_lr: 0.000003  loss: 2.1545 (2.1545)  weight_decay: 0.0500 (0.0500)  time: 16.6248  data: 13.9430  max mem: 41794
Epoch: [296]  [ 10/312]  eta: 0:11:21  lr: 0.000003  min_lr: 0.000003  loss: 3.0614 (2.9221)  weight_decay: 0.0500 (0.0500)  time: 2.2582  data: 1.4751  max mem: 41794
Epoch: [296]  [ 20/312]  eta: 0:08:08  lr: 0.000003  min_lr: 0.000003  loss: 2.9134 (2.7798)  weight_decay: 0.0500 (0.0500)  time: 0.9256  data: 0.2145  max mem: 41794
Epoch: [296]  [ 30/312]  eta: 0:06:17  lr: 0.000003  min_lr: 0.000003  loss: 2.7488 (2.8582)  weight_decay: 0.0500 (0.0500)  time: 0.8302  data: 0.1066  max mem: 41794
Epoch: [296]  [ 40/312]  eta: 0:05:33  lr: 0.000003  min_lr: 0.000003  loss: 3.1366 (2.8818)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.0232  max mem: 41794
Epoch: [296]  [ 50/312]  eta: 0:05:00  lr: 0.000003  min_lr: 0.000003  loss: 3.0534 (2.8928)  weight_decay: 0.0500 (0.0500)  time: 0.8540  data: 0.0203  max mem: 41794
Epoch: [296]  [ 60/312]  eta: 0:04:26  lr: 0.000003  min_lr: 0.000003  loss: 3.0070 (2.9117)  weight_decay: 0.0500 (0.0500)  time: 0.7065  data: 0.0037  max mem: 41794
Epoch: [296]  [ 70/312]  eta: 0:04:09  lr: 0.000003  min_lr: 0.000003  loss: 2.6545 (2.8512)  weight_decay: 0.0500 (0.0500)  time: 0.7351  data: 0.0007  max mem: 41794
Epoch: [296]  [ 80/312]  eta: 0:03:53  lr: 0.000003  min_lr: 0.000003  loss: 2.5806 (2.8461)  weight_decay: 0.0500 (0.0500)  time: 0.8517  data: 0.0227  max mem: 41794
Epoch: [296]  [ 90/312]  eta: 0:03:36  lr: 0.000003  min_lr: 0.000003  loss: 2.9757 (2.8646)  weight_decay: 0.0500 (0.0500)  time: 0.7675  data: 0.0229  max mem: 41794
Epoch: [296]  [100/312]  eta: 0:03:25  lr: 0.000003  min_lr: 0.000003  loss: 3.0312 (2.8705)  weight_decay: 0.0500 (0.0500)  time: 0.8198  data: 0.0187  max mem: 41794
Epoch: [296]  [110/312]  eta: 0:03:09  lr: 0.000003  min_lr: 0.000003  loss: 3.2232 (2.9065)  weight_decay: 0.0500 (0.0500)  time: 0.7648  data: 0.0183  max mem: 41794
Epoch: [296]  [120/312]  eta: 0:02:58  lr: 0.000003  min_lr: 0.000003  loss: 3.2232 (2.9025)  weight_decay: 0.0500 (0.0500)  time: 0.7252  data: 0.0006  max mem: 41794
Epoch: [296]  [130/312]  eta: 0:02:48  lr: 0.000003  min_lr: 0.000003  loss: 2.8632 (2.8950)  weight_decay: 0.0500 (0.0500)  time: 0.8626  data: 0.0253  max mem: 41794
Epoch: [296]  [140/312]  eta: 0:02:35  lr: 0.000003  min_lr: 0.000003  loss: 3.0513 (2.8944)  weight_decay: 0.0500 (0.0500)  time: 0.7366  data: 0.0252  max mem: 41794
Epoch: [296]  [150/312]  eta: 0:02:26  lr: 0.000003  min_lr: 0.000003  loss: 3.1387 (2.8946)  weight_decay: 0.0500 (0.0500)  time: 0.7551  data: 0.0258  max mem: 41794
Epoch: [296]  [160/312]  eta: 0:02:15  lr: 0.000003  min_lr: 0.000003  loss: 2.9956 (2.8987)  weight_decay: 0.0500 (0.0500)  time: 0.8257  data: 0.0258  max mem: 41794
Epoch: [296]  [170/312]  eta: 0:02:05  lr: 0.000003  min_lr: 0.000003  loss: 3.0484 (2.8963)  weight_decay: 0.0500 (0.0500)  time: 0.7340  data: 0.0261  max mem: 41794
Epoch: [296]  [180/312]  eta: 0:01:56  lr: 0.000002  min_lr: 0.000002  loss: 3.0998 (2.9028)  weight_decay: 0.0500 (0.0500)  time: 0.7931  data: 0.0585  max mem: 41794
Epoch: [296]  [190/312]  eta: 0:01:45  lr: 0.000002  min_lr: 0.000002  loss: 2.9295 (2.8918)  weight_decay: 0.0500 (0.0500)  time: 0.7298  data: 0.0353  max mem: 41794
Epoch: [296]  [200/312]  eta: 0:01:37  lr: 0.000002  min_lr: 0.000002  loss: 2.9630 (2.9028)  weight_decay: 0.0500 (0.0500)  time: 0.7331  data: 0.0494  max mem: 41794
Epoch: [296]  [210/312]  eta: 0:01:28  lr: 0.000002  min_lr: 0.000002  loss: 3.1314 (2.9063)  weight_decay: 0.0500 (0.0500)  time: 0.8642  data: 0.1041  max mem: 41794
Epoch: [296]  [220/312]  eta: 0:01:18  lr: 0.000002  min_lr: 0.000002  loss: 3.0270 (2.9063)  weight_decay: 0.0500 (0.0500)  time: 0.7512  data: 0.0578  max mem: 41794
Epoch: [296]  [230/312]  eta: 0:01:10  lr: 0.000002  min_lr: 0.000002  loss: 3.0274 (2.9071)  weight_decay: 0.0500 (0.0500)  time: 0.8137  data: 0.0623  max mem: 41794
Epoch: [296]  [240/312]  eta: 0:01:01  lr: 0.000002  min_lr: 0.000002  loss: 2.8700 (2.9019)  weight_decay: 0.0500 (0.0500)  time: 0.8747  data: 0.0717  max mem: 41794
Epoch: [296]  [250/312]  eta: 0:00:52  lr: 0.000002  min_lr: 0.000002  loss: 2.6275 (2.8873)  weight_decay: 0.0500 (0.0500)  time: 0.7581  data: 0.0235  max mem: 41794
Epoch: [296]  [260/312]  eta: 0:00:44  lr: 0.000002  min_lr: 0.000002  loss: 2.5513 (2.8893)  weight_decay: 0.0500 (0.0500)  time: 0.8574  data: 0.0301  max mem: 41794
Epoch: [296]  [270/312]  eta: 0:00:35  lr: 0.000002  min_lr: 0.000002  loss: 3.1079 (2.8960)  weight_decay: 0.0500 (0.0500)  time: 0.7788  data: 0.0168  max mem: 41794
Epoch: [296]  [280/312]  eta: 0:00:27  lr: 0.000002  min_lr: 0.000002  loss: 3.1248 (2.8987)  weight_decay: 0.0500 (0.0500)  time: 0.7236  data: 0.0169  max mem: 41794
Epoch: [296]  [290/312]  eta: 0:00:18  lr: 0.000002  min_lr: 0.000002  loss: 3.1702 (2.9058)  weight_decay: 0.0500 (0.0500)  time: 0.8453  data: 0.0427  max mem: 41794
Epoch: [296]  [300/312]  eta: 0:00:10  lr: 0.000002  min_lr: 0.000002  loss: 3.0677 (2.9038)  weight_decay: 0.0500 (0.0500)  time: 0.7393  data: 0.0264  max mem: 41794
Epoch: [296]  [310/312]  eta: 0:00:01  lr: 0.000002  min_lr: 0.000002  loss: 3.0591 (2.9066)  weight_decay: 0.0500 (0.0500)  time: 0.6339  data: 0.0001  max mem: 41794
Epoch: [296]  [311/312]  eta: 0:00:00  lr: 0.000002  min_lr: 0.000002  loss: 3.1164 (2.9073)  weight_decay: 0.0500 (0.0500)  time: 0.6340  data: 0.0001  max mem: 41794
Epoch: [296] Total time: 0:04:22 (0.8401 s / it)
Averaged stats: lr: 0.000002  min_lr: 0.000002  loss: 3.1164 (2.9665)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:31  loss: 0.8823 (0.8823)  acc1: 86.1979 (86.1979)  acc5: 96.4844 (96.4844)  time: 10.1580  data: 9.9054  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1451 (1.0698)  acc1: 79.0365 (79.9200)  acc5: 95.8333 (95.3600)  time: 1.3376  data: 1.1007  max mem: 41794
Test: Total time: 0:00:12 (1.3538 s / it)
* Acc@1 80.264 Acc@5 95.308 loss 1.066
Accuracy of the model on the 50000 test images: 80.3%
Max accuracy: 80.42%
Epoch: [297]  [  0/312]  eta: 1:23:00  lr: 0.000002  min_lr: 0.000002  loss: 3.5387 (3.5387)  weight_decay: 0.0500 (0.0500)  time: 15.9641  data: 13.8730  max mem: 41794
Epoch: [297]  [ 10/312]  eta: 0:11:57  lr: 0.000002  min_lr: 0.000002  loss: 3.2513 (3.2486)  weight_decay: 0.0500 (0.0500)  time: 2.3762  data: 1.3522  max mem: 41794
Epoch: [297]  [ 20/312]  eta: 0:07:52  lr: 0.000002  min_lr: 0.000002  loss: 3.1146 (3.0046)  weight_decay: 0.0500 (0.0500)  time: 0.8993  data: 0.0810  max mem: 41794
Epoch: [297]  [ 30/312]  eta: 0:06:05  lr: 0.000002  min_lr: 0.000002  loss: 2.9241 (3.0156)  weight_decay: 0.0500 (0.0500)  time: 0.6994  data: 0.0314  max mem: 41794
Epoch: [297]  [ 40/312]  eta: 0:05:15  lr: 0.000002  min_lr: 0.000002  loss: 3.0587 (3.0103)  weight_decay: 0.0500 (0.0500)  time: 0.6831  data: 0.0771  max mem: 41794
Epoch: [297]  [ 50/312]  eta: 0:04:49  lr: 0.000002  min_lr: 0.000002  loss: 3.0536 (3.0154)  weight_decay: 0.0500 (0.0500)  time: 0.8125  data: 0.2171  max mem: 41794
Epoch: [297]  [ 60/312]  eta: 0:04:17  lr: 0.000002  min_lr: 0.000002  loss: 3.1716 (3.0315)  weight_decay: 0.0500 (0.0500)  time: 0.7380  data: 0.1407  max mem: 41794
Epoch: [297]  [ 70/312]  eta: 0:04:02  lr: 0.000002  min_lr: 0.000002  loss: 3.1371 (3.0231)  weight_decay: 0.0500 (0.0500)  time: 0.7398  data: 0.1380  max mem: 41794
Epoch: [297]  [ 80/312]  eta: 0:03:52  lr: 0.000002  min_lr: 0.000002  loss: 3.1778 (3.0089)  weight_decay: 0.0500 (0.0500)  time: 0.9481  data: 0.3446  max mem: 41794
Epoch: [297]  [ 90/312]  eta: 0:03:33  lr: 0.000002  min_lr: 0.000002  loss: 2.9046 (3.0011)  weight_decay: 0.0500 (0.0500)  time: 0.8099  data: 0.2072  max mem: 41794
Epoch: [297]  [100/312]  eta: 0:03:23  lr: 0.000002  min_lr: 0.000002  loss: 2.9793 (2.9999)  weight_decay: 0.0500 (0.0500)  time: 0.7766  data: 0.1767  max mem: 41794
Epoch: [297]  [110/312]  eta: 0:03:07  lr: 0.000002  min_lr: 0.000002  loss: 3.1129 (2.9960)  weight_decay: 0.0500 (0.0500)  time: 0.7749  data: 0.1771  max mem: 41794
Epoch: [297]  [120/312]  eta: 0:02:57  lr: 0.000002  min_lr: 0.000002  loss: 3.0901 (2.9760)  weight_decay: 0.0500 (0.0500)  time: 0.7499  data: 0.1507  max mem: 41794
Epoch: [297]  [130/312]  eta: 0:02:47  lr: 0.000002  min_lr: 0.000002  loss: 2.7126 (2.9583)  weight_decay: 0.0500 (0.0500)  time: 0.8848  data: 0.2879  max mem: 41794
Epoch: [297]  [140/312]  eta: 0:02:34  lr: 0.000002  min_lr: 0.000002  loss: 2.9856 (2.9625)  weight_decay: 0.0500 (0.0500)  time: 0.7348  data: 0.1382  max mem: 41794
Epoch: [297]  [150/312]  eta: 0:02:25  lr: 0.000002  min_lr: 0.000002  loss: 3.0073 (2.9536)  weight_decay: 0.0500 (0.0500)  time: 0.7548  data: 0.1573  max mem: 41794
Epoch: [297]  [160/312]  eta: 0:02:16  lr: 0.000002  min_lr: 0.000002  loss: 2.9145 (2.9406)  weight_decay: 0.0500 (0.0500)  time: 0.8884  data: 0.2840  max mem: 41794
Epoch: [297]  [170/312]  eta: 0:02:04  lr: 0.000002  min_lr: 0.000002  loss: 2.9678 (2.9469)  weight_decay: 0.0500 (0.0500)  time: 0.7345  data: 0.1273  max mem: 41794
Epoch: [297]  [180/312]  eta: 0:01:56  lr: 0.000002  min_lr: 0.000002  loss: 3.1196 (2.9479)  weight_decay: 0.0500 (0.0500)  time: 0.7658  data: 0.1640  max mem: 41794
Epoch: [297]  [190/312]  eta: 0:01:45  lr: 0.000002  min_lr: 0.000002  loss: 2.9752 (2.9449)  weight_decay: 0.0500 (0.0500)  time: 0.7643  data: 0.1640  max mem: 41794
Epoch: [297]  [200/312]  eta: 0:01:37  lr: 0.000002  min_lr: 0.000002  loss: 3.0037 (2.9425)  weight_decay: 0.0500 (0.0500)  time: 0.7598  data: 0.1641  max mem: 41794
Epoch: [297]  [210/312]  eta: 0:01:28  lr: 0.000002  min_lr: 0.000002  loss: 3.0037 (2.9362)  weight_decay: 0.0500 (0.0500)  time: 0.9205  data: 0.3246  max mem: 41794
Epoch: [297]  [220/312]  eta: 0:01:19  lr: 0.000002  min_lr: 0.000002  loss: 2.9868 (2.9326)  weight_decay: 0.0500 (0.0500)  time: 0.7591  data: 0.1611  max mem: 41794
Epoch: [297]  [230/312]  eta: 0:01:10  lr: 0.000002  min_lr: 0.000002  loss: 3.0752 (2.9367)  weight_decay: 0.0500 (0.0500)  time: 0.7487  data: 0.1539  max mem: 41794
Epoch: [297]  [240/312]  eta: 0:01:02  lr: 0.000002  min_lr: 0.000002  loss: 3.1337 (2.9346)  weight_decay: 0.0500 (0.0500)  time: 0.9029  data: 0.3057  max mem: 41794
Epoch: [297]  [250/312]  eta: 0:00:52  lr: 0.000002  min_lr: 0.000002  loss: 2.8829 (2.9319)  weight_decay: 0.0500 (0.0500)  time: 0.7553  data: 0.1523  max mem: 41794
Epoch: [297]  [260/312]  eta: 0:00:44  lr: 0.000002  min_lr: 0.000002  loss: 3.0837 (2.9403)  weight_decay: 0.0500 (0.0500)  time: 0.7823  data: 0.1836  max mem: 41794
Epoch: [297]  [270/312]  eta: 0:00:35  lr: 0.000002  min_lr: 0.000002  loss: 2.9834 (2.9386)  weight_decay: 0.0500 (0.0500)  time: 0.7799  data: 0.1841  max mem: 41794
Epoch: [297]  [280/312]  eta: 0:00:27  lr: 0.000002  min_lr: 0.000002  loss: 3.1714 (2.9506)  weight_decay: 0.0500 (0.0500)  time: 0.7592  data: 0.1632  max mem: 41794
Epoch: [297]  [290/312]  eta: 0:00:18  lr: 0.000002  min_lr: 0.000002  loss: 3.1842 (2.9568)  weight_decay: 0.0500 (0.0500)  time: 0.8939  data: 0.2912  max mem: 41794
Epoch: [297]  [300/312]  eta: 0:00:10  lr: 0.000002  min_lr: 0.000002  loss: 3.2290 (2.9662)  weight_decay: 0.0500 (0.0500)  time: 0.7525  data: 0.1289  max mem: 41794
Epoch: [297]  [310/312]  eta: 0:00:01  lr: 0.000002  min_lr: 0.000002  loss: 3.0317 (2.9615)  weight_decay: 0.0500 (0.0500)  time: 0.6372  data: 0.0001  max mem: 41794
Epoch: [297]  [311/312]  eta: 0:00:00  lr: 0.000002  min_lr: 0.000002  loss: 3.1116 (2.9626)  weight_decay: 0.0500 (0.0500)  time: 0.6367  data: 0.0001  max mem: 41794
Epoch: [297] Total time: 0:04:22 (0.8426 s / it)
Averaged stats: lr: 0.000002  min_lr: 0.000002  loss: 3.1116 (2.9690)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:33  loss: 0.8554 (0.8554)  acc1: 86.0677 (86.0677)  acc5: 96.7448 (96.7448)  time: 10.3385  data: 10.0848  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1239 (1.0439)  acc1: 79.2969 (80.0320)  acc5: 95.5729 (95.2800)  time: 1.3560  data: 1.1206  max mem: 41794
Test: Total time: 0:00:12 (1.3710 s / it)
* Acc@1 80.328 Acc@5 95.324 loss 1.039
Accuracy of the model on the 50000 test images: 80.3%
Max accuracy: 80.42%
Epoch: [298]  [  0/312]  eta: 1:22:46  lr: 0.000002  min_lr: 0.000002  loss: 3.2036 (3.2036)  weight_decay: 0.0500 (0.0500)  time: 15.9197  data: 13.3067  max mem: 41794
Epoch: [298]  [ 10/312]  eta: 0:11:14  lr: 0.000001  min_lr: 0.000001  loss: 3.1355 (3.0910)  weight_decay: 0.0500 (0.0500)  time: 2.2340  data: 1.3113  max mem: 41794
Epoch: [298]  [ 20/312]  eta: 0:08:18  lr: 0.000001  min_lr: 0.000001  loss: 3.1172 (2.9768)  weight_decay: 0.0500 (0.0500)  time: 0.9963  data: 0.0589  max mem: 41794
Epoch: [298]  [ 30/312]  eta: 0:06:23  lr: 0.000001  min_lr: 0.000001  loss: 3.1172 (2.9872)  weight_decay: 0.0500 (0.0500)  time: 0.8802  data: 0.0163  max mem: 41794
Epoch: [298]  [ 40/312]  eta: 0:05:26  lr: 0.000001  min_lr: 0.000001  loss: 3.1199 (3.0290)  weight_decay: 0.0500 (0.0500)  time: 0.6661  data: 0.0621  max mem: 41794
Epoch: [298]  [ 50/312]  eta: 0:04:45  lr: 0.000001  min_lr: 0.000001  loss: 3.2193 (3.0510)  weight_decay: 0.0500 (0.0500)  time: 0.6696  data: 0.0491  max mem: 41794
Epoch: [298]  [ 60/312]  eta: 0:04:14  lr: 0.000001  min_lr: 0.000001  loss: 3.1914 (3.0159)  weight_decay: 0.0500 (0.0500)  time: 0.6228  data: 0.0035  max mem: 41794
Epoch: [298]  [ 70/312]  eta: 0:03:59  lr: 0.000001  min_lr: 0.000001  loss: 3.0187 (2.9983)  weight_decay: 0.0500 (0.0500)  time: 0.7291  data: 0.1124  max mem: 41794
Epoch: [298]  [ 80/312]  eta: 0:03:49  lr: 0.000001  min_lr: 0.000001  loss: 2.8149 (2.9867)  weight_decay: 0.0500 (0.0500)  time: 0.9198  data: 0.3053  max mem: 41794
Epoch: [298]  [ 90/312]  eta: 0:03:29  lr: 0.000001  min_lr: 0.000001  loss: 2.8149 (2.9670)  weight_decay: 0.0500 (0.0500)  time: 0.7933  data: 0.1963  max mem: 41794
Epoch: [298]  [100/312]  eta: 0:03:17  lr: 0.000001  min_lr: 0.000001  loss: 2.7442 (2.9549)  weight_decay: 0.0500 (0.0500)  time: 0.7100  data: 0.1040  max mem: 41794
Epoch: [298]  [110/312]  eta: 0:03:02  lr: 0.000001  min_lr: 0.000001  loss: 2.7442 (2.9504)  weight_decay: 0.0500 (0.0500)  time: 0.7098  data: 0.1040  max mem: 41794
Epoch: [298]  [120/312]  eta: 0:02:54  lr: 0.000001  min_lr: 0.000001  loss: 3.1129 (2.9642)  weight_decay: 0.0500 (0.0500)  time: 0.7848  data: 0.1867  max mem: 41794
Epoch: [298]  [130/312]  eta: 0:02:44  lr: 0.000001  min_lr: 0.000001  loss: 3.2196 (2.9747)  weight_decay: 0.0500 (0.0500)  time: 0.8999  data: 0.3031  max mem: 41794
Epoch: [298]  [140/312]  eta: 0:02:31  lr: 0.000001  min_lr: 0.000001  loss: 3.2274 (2.9848)  weight_decay: 0.0500 (0.0500)  time: 0.7197  data: 0.1198  max mem: 41794
Epoch: [298]  [150/312]  eta: 0:02:23  lr: 0.000001  min_lr: 0.000001  loss: 3.1753 (2.9937)  weight_decay: 0.0500 (0.0500)  time: 0.7859  data: 0.1851  max mem: 41794
Epoch: [298]  [160/312]  eta: 0:02:14  lr: 0.000001  min_lr: 0.000001  loss: 3.2081 (2.9982)  weight_decay: 0.0500 (0.0500)  time: 0.9057  data: 0.3081  max mem: 41794
Epoch: [298]  [170/312]  eta: 0:02:03  lr: 0.000001  min_lr: 0.000001  loss: 3.1448 (2.9977)  weight_decay: 0.0500 (0.0500)  time: 0.7285  data: 0.1274  max mem: 41794
Epoch: [298]  [180/312]  eta: 0:01:54  lr: 0.000001  min_lr: 0.000001  loss: 3.1377 (3.0039)  weight_decay: 0.0500 (0.0500)  time: 0.7589  data: 0.1519  max mem: 41794
Epoch: [298]  [190/312]  eta: 0:01:44  lr: 0.000001  min_lr: 0.000001  loss: 3.0513 (2.9875)  weight_decay: 0.0500 (0.0500)  time: 0.7549  data: 0.1509  max mem: 41794
Epoch: [298]  [200/312]  eta: 0:01:36  lr: 0.000001  min_lr: 0.000001  loss: 2.9444 (2.9786)  weight_decay: 0.0500 (0.0500)  time: 0.7603  data: 0.1209  max mem: 41794
Epoch: [298]  [210/312]  eta: 0:01:27  lr: 0.000001  min_lr: 0.000001  loss: 3.1224 (2.9809)  weight_decay: 0.0500 (0.0500)  time: 0.9099  data: 0.2441  max mem: 41794
Epoch: [298]  [220/312]  eta: 0:01:18  lr: 0.000001  min_lr: 0.000001  loss: 2.9086 (2.9684)  weight_decay: 0.0500 (0.0500)  time: 0.7494  data: 0.1239  max mem: 41794
Epoch: [298]  [230/312]  eta: 0:01:09  lr: 0.000001  min_lr: 0.000001  loss: 3.0027 (2.9681)  weight_decay: 0.0500 (0.0500)  time: 0.7694  data: 0.1336  max mem: 41794
Epoch: [298]  [240/312]  eta: 0:01:01  lr: 0.000001  min_lr: 0.000001  loss: 3.0877 (2.9659)  weight_decay: 0.0500 (0.0500)  time: 0.8920  data: 0.2434  max mem: 41794
Epoch: [298]  [250/312]  eta: 0:00:52  lr: 0.000001  min_lr: 0.000001  loss: 2.7478 (2.9531)  weight_decay: 0.0500 (0.0500)  time: 0.7404  data: 0.1118  max mem: 41794
Epoch: [298]  [260/312]  eta: 0:00:44  lr: 0.000001  min_lr: 0.000001  loss: 2.8131 (2.9521)  weight_decay: 0.0500 (0.0500)  time: 0.7929  data: 0.1055  max mem: 41794
Epoch: [298]  [270/312]  eta: 0:00:35  lr: 0.000001  min_lr: 0.000001  loss: 3.1377 (2.9444)  weight_decay: 0.0500 (0.0500)  time: 0.7748  data: 0.1041  max mem: 41794
Epoch: [298]  [280/312]  eta: 0:00:27  lr: 0.000001  min_lr: 0.000001  loss: 3.0315 (2.9485)  weight_decay: 0.0500 (0.0500)  time: 0.7948  data: 0.1034  max mem: 41794
Epoch: [298]  [290/312]  eta: 0:00:18  lr: 0.000001  min_lr: 0.000001  loss: 3.1763 (2.9513)  weight_decay: 0.0500 (0.0500)  time: 0.9407  data: 0.1634  max mem: 41794
Epoch: [298]  [300/312]  eta: 0:00:10  lr: 0.000001  min_lr: 0.000001  loss: 3.1571 (2.9520)  weight_decay: 0.0500 (0.0500)  time: 0.7659  data: 0.0603  max mem: 41794
Epoch: [298]  [310/312]  eta: 0:00:01  lr: 0.000001  min_lr: 0.000001  loss: 2.9829 (2.9507)  weight_decay: 0.0500 (0.0500)  time: 0.6373  data: 0.0001  max mem: 41794
Epoch: [298]  [311/312]  eta: 0:00:00  lr: 0.000001  min_lr: 0.000001  loss: 2.9829 (2.9524)  weight_decay: 0.0500 (0.0500)  time: 0.6381  data: 0.0001  max mem: 41794
Epoch: [298] Total time: 0:04:21 (0.8388 s / it)
Averaged stats: lr: 0.000001  min_lr: 0.000001  loss: 2.9829 (2.9696)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:34  loss: 0.8508 (0.8508)  acc1: 85.8073 (85.8073)  acc5: 96.6146 (96.6146)  time: 10.4837  data: 10.2293  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.1150 (1.0431)  acc1: 79.6875 (79.9200)  acc5: 95.7031 (95.2800)  time: 1.3740  data: 1.1367  max mem: 41794
Test: Total time: 0:00:12 (1.3916 s / it)
* Acc@1 80.244 Acc@5 95.274 loss 1.038
Accuracy of the model on the 50000 test images: 80.2%
Max accuracy: 80.42%
Epoch: [299]  [  0/312]  eta: 1:22:54  lr: 0.000001  min_lr: 0.000001  loss: 3.5536 (3.5536)  weight_decay: 0.0500 (0.0500)  time: 15.9449  data: 13.0529  max mem: 41794
Epoch: [299]  [ 10/312]  eta: 0:10:56  lr: 0.000001  min_lr: 0.000001  loss: 2.6877 (2.6664)  weight_decay: 0.0500 (0.0500)  time: 2.1745  data: 1.3700  max mem: 41794
Epoch: [299]  [ 20/312]  eta: 0:07:58  lr: 0.000001  min_lr: 0.000001  loss: 2.6877 (2.7335)  weight_decay: 0.0500 (0.0500)  time: 0.9237  data: 0.1832  max mem: 41794
Epoch: [299]  [ 30/312]  eta: 0:06:08  lr: 0.000001  min_lr: 0.000001  loss: 3.0499 (2.7994)  weight_decay: 0.0500 (0.0500)  time: 0.8303  data: 0.0915  max mem: 41794
Epoch: [299]  [ 40/312]  eta: 0:05:21  lr: 0.000001  min_lr: 0.000001  loss: 3.0652 (2.8449)  weight_decay: 0.0500 (0.0500)  time: 0.6984  data: 0.0234  max mem: 41794
Epoch: [299]  [ 50/312]  eta: 0:04:57  lr: 0.000001  min_lr: 0.000001  loss: 3.0652 (2.8910)  weight_decay: 0.0500 (0.0500)  time: 0.8680  data: 0.0699  max mem: 41794
Epoch: [299]  [ 60/312]  eta: 0:04:23  lr: 0.000001  min_lr: 0.000001  loss: 3.0736 (2.9024)  weight_decay: 0.0500 (0.0500)  time: 0.7751  data: 0.0560  max mem: 41794
Epoch: [299]  [ 70/312]  eta: 0:04:06  lr: 0.000001  min_lr: 0.000001  loss: 2.9874 (2.9160)  weight_decay: 0.0500 (0.0500)  time: 0.7243  data: 0.0265  max mem: 41794
Epoch: [299]  [ 80/312]  eta: 0:03:53  lr: 0.000001  min_lr: 0.000001  loss: 2.9855 (2.9104)  weight_decay: 0.0500 (0.0500)  time: 0.8871  data: 0.0300  max mem: 41794
Epoch: [299]  [ 90/312]  eta: 0:03:34  lr: 0.000001  min_lr: 0.000001  loss: 2.8420 (2.9057)  weight_decay: 0.0500 (0.0500)  time: 0.7766  data: 0.0193  max mem: 41794
Epoch: [299]  [100/312]  eta: 0:03:23  lr: 0.000001  min_lr: 0.000001  loss: 3.0289 (2.9090)  weight_decay: 0.0500 (0.0500)  time: 0.7636  data: 0.0298  max mem: 41794
Epoch: [299]  [110/312]  eta: 0:03:07  lr: 0.000001  min_lr: 0.000001  loss: 2.9066 (2.8914)  weight_decay: 0.0500 (0.0500)  time: 0.7480  data: 0.0147  max mem: 41794
Epoch: [299]  [120/312]  eta: 0:03:00  lr: 0.000001  min_lr: 0.000001  loss: 2.8058 (2.8810)  weight_decay: 0.0500 (0.0500)  time: 0.8527  data: 0.0429  max mem: 41794
Epoch: [299]  [130/312]  eta: 0:02:51  lr: 0.000001  min_lr: 0.000001  loss: 2.9552 (2.8938)  weight_decay: 0.0500 (0.0500)  time: 1.0345  data: 0.0484  max mem: 41794
Epoch: [299]  [140/312]  eta: 0:02:38  lr: 0.000001  min_lr: 0.000001  loss: 3.0131 (2.8913)  weight_decay: 0.0500 (0.0500)  time: 0.7786  data: 0.0073  max mem: 41794
Epoch: [299]  [150/312]  eta: 0:02:27  lr: 0.000001  min_lr: 0.000001  loss: 3.1072 (2.9133)  weight_decay: 0.0500 (0.0500)  time: 0.6986  data: 0.0018  max mem: 41794
Epoch: [299]  [160/312]  eta: 0:02:17  lr: 0.000001  min_lr: 0.000001  loss: 3.2204 (2.9265)  weight_decay: 0.0500 (0.0500)  time: 0.7947  data: 0.0024  max mem: 41794
Epoch: [299]  [170/312]  eta: 0:02:07  lr: 0.000001  min_lr: 0.000001  loss: 3.2152 (2.9449)  weight_decay: 0.0500 (0.0500)  time: 0.7783  data: 0.0025  max mem: 41794
Epoch: [299]  [180/312]  eta: 0:01:59  lr: 0.000001  min_lr: 0.000001  loss: 3.1102 (2.9433)  weight_decay: 0.0500 (0.0500)  time: 0.8987  data: 0.0007  max mem: 41794
Epoch: [299]  [190/312]  eta: 0:01:48  lr: 0.000001  min_lr: 0.000001  loss: 3.0995 (2.9528)  weight_decay: 0.0500 (0.0500)  time: 0.8147  data: 0.0022  max mem: 41794
Epoch: [299]  [200/312]  eta: 0:01:38  lr: 0.000001  min_lr: 0.000001  loss: 3.2912 (2.9616)  weight_decay: 0.0500 (0.0500)  time: 0.6955  data: 0.0021  max mem: 41794
Epoch: [299]  [210/312]  eta: 0:01:29  lr: 0.000001  min_lr: 0.000001  loss: 3.1620 (2.9627)  weight_decay: 0.0500 (0.0500)  time: 0.8384  data: 0.0038  max mem: 41794
Epoch: [299]  [220/312]  eta: 0:01:20  lr: 0.000001  min_lr: 0.000001  loss: 2.8151 (2.9597)  weight_decay: 0.0500 (0.0500)  time: 0.7863  data: 0.0039  max mem: 41794
Epoch: [299]  [230/312]  eta: 0:01:11  lr: 0.000001  min_lr: 0.000001  loss: 2.9757 (2.9594)  weight_decay: 0.0500 (0.0500)  time: 0.7776  data: 0.0120  max mem: 41794
Epoch: [299]  [240/312]  eta: 0:01:02  lr: 0.000001  min_lr: 0.000001  loss: 2.9962 (2.9628)  weight_decay: 0.0500 (0.0500)  time: 0.7947  data: 0.0119  max mem: 41794
Epoch: [299]  [250/312]  eta: 0:00:53  lr: 0.000001  min_lr: 0.000001  loss: 3.0017 (2.9635)  weight_decay: 0.0500 (0.0500)  time: 0.7876  data: 0.0008  max mem: 41794
Epoch: [299]  [260/312]  eta: 0:00:44  lr: 0.000001  min_lr: 0.000001  loss: 2.9642 (2.9598)  weight_decay: 0.0500 (0.0500)  time: 0.8146  data: 0.0028  max mem: 41794
Epoch: [299]  [270/312]  eta: 0:00:35  lr: 0.000001  min_lr: 0.000001  loss: 2.8379 (2.9557)  weight_decay: 0.0500 (0.0500)  time: 0.6937  data: 0.0027  max mem: 41794
Epoch: [299]  [280/312]  eta: 0:00:27  lr: 0.000001  min_lr: 0.000001  loss: 2.9250 (2.9556)  weight_decay: 0.0500 (0.0500)  time: 0.7735  data: 0.0064  max mem: 41794
Epoch: [299]  [290/312]  eta: 0:00:18  lr: 0.000001  min_lr: 0.000001  loss: 2.9609 (2.9561)  weight_decay: 0.0500 (0.0500)  time: 0.8534  data: 0.0062  max mem: 41794
Epoch: [299]  [300/312]  eta: 0:00:10  lr: 0.000001  min_lr: 0.000001  loss: 2.8591 (2.9513)  weight_decay: 0.0500 (0.0500)  time: 0.7032  data: 0.0002  max mem: 41794
Epoch: [299]  [310/312]  eta: 0:00:01  lr: 0.000001  min_lr: 0.000001  loss: 2.8591 (2.9476)  weight_decay: 0.0500 (0.0500)  time: 0.6329  data: 0.0001  max mem: 41794
Epoch: [299]  [311/312]  eta: 0:00:00  lr: 0.000001  min_lr: 0.000001  loss: 2.8591 (2.9484)  weight_decay: 0.0500 (0.0500)  time: 0.6340  data: 0.0001  max mem: 41794
Epoch: [299] Total time: 0:04:24 (0.8463 s / it)
Averaged stats: lr: 0.000001  min_lr: 0.000001  loss: 2.8591 (2.9564)  weight_decay: 0.0500 (0.0500)
Test:  [0/9]  eta: 0:01:38  loss: 0.7938 (0.7938)  acc1: 86.1979 (86.1979)  acc5: 96.8750 (96.8750)  time: 10.9316  data: 10.6724  max mem: 41794
Test:  [8/9]  eta: 0:00:01  loss: 1.0595 (0.9819)  acc1: 79.2969 (80.0320)  acc5: 95.5729 (95.4080)  time: 1.4217  data: 1.1859  max mem: 41794
Test: Total time: 0:00:12 (1.4317 s / it)
* Acc@1 80.426 Acc@5 95.336 loss 0.977
Accuracy of the model on the 50000 test images: 80.4%
Max accuracy: 80.43%
Training time 20:53:18