model wo checkpoint

nnUNetTrainer__nnUNetResEncUNetMPlans__3d_fullres/fold_all/debug.json

@@ -0,0 +1,53 @@
+{
+    "_best_ema": "0.8639167944091766",
+    "batch_size": "2",
+    "configuration_manager": "{'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [128, 128, 128], 'median_image_size_in_voxels': [235.0, 240.0, 240.0], 'spacing': [1.5, 1.5, 1.5], 'normalization_schemes': ['CTNormalization'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}, 'deep_supervision': True}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}",
+    "configuration_name": "3d_fullres",
+    "cudnn_version": 8902,
+    "current_epoch": "800",
+    "dataloader_train": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x7fe33e837130>",
+    "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D object at 0x7fe33e8373d0>",
+    "dataloader_train.num_processes": "16",
+    "dataloader_train.transform": "Compose ( [SpatialTransform( independent_scale_for_each_axis = False, p_rot_per_sample = 0.2, p_scale_per_sample = 0.2, p_el_per_sample = 0, data_key = 'data', label_key = 'seg', patch_size = [128, 128, 128], patch_center_dist_from_border = None, do_elastic_deform = False, alpha = (0, 0), sigma = (0, 0), do_rotation = True, angle_x = (-0.5235987755982988, 0.5235987755982988), angle_y = (-0.5235987755982988, 0.5235987755982988), angle_z = (-0.5235987755982988, 0.5235987755982988), do_scale = True, scale = (0.7, 1.4), border_mode_data = 'constant', border_cval_data = 0, order_data = 3, border_mode_seg = 'constant', border_cval_seg = -1, order_seg = 1, random_crop = False, p_rot_per_axis = 1, p_independent_scale_per_axis = 1 ), GaussianNoiseTransform( p_per_sample = 0.1, data_key = 'data', noise_variance = (0, 0.1), p_per_channel = 1, per_channel = False ), GaussianBlurTransform( p_per_sample = 0.2, different_sigma_per_channel = True, p_per_channel = 0.5, data_key = 'data', blur_sigma = (0.5, 1.0), different_sigma_per_axis = False, p_isotropic = 0 ), BrightnessMultiplicativeTransform( p_per_sample = 0.15, data_key = 'data', multiplier_range = (0.75, 1.25), per_channel = True ), ContrastAugmentationTransform( p_per_sample = 0.15, data_key = 'data', contrast_range = (0.75, 1.25), preserve_range = True, per_channel = True, p_per_channel = 1 ), SimulateLowResolutionTransform( order_upsample = 3, order_downsample = 0, channels = None, per_channel = True, p_per_channel = 0.5, p_per_sample = 0.25, data_key = 'data', zoom_range = (0.5, 1), ignore_axes = None ), GammaTransform( p_per_sample = 0.1, retain_stats = True, per_channel = True, data_key = 'data', gamma_range = (0.7, 1.5), invert_image = True ), GammaTransform( p_per_sample = 0.3, retain_stats = True, per_channel = True, data_key = 'data', gamma_range = (0.7, 1.5), invert_image = False ), MirrorTransform( p_per_sample = 1, data_key = 'data', label_key = 'seg', axes = (0, 1, 2) ), RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), DownsampleSegForDSTransform2( axes = None, output_key = 'target', input_key = 'target', order = 0, ds_scales = [[1.0, 1.0, 1.0], [0.5, 0.5, 0.5], [0.25, 0.25, 0.25], [0.125, 0.125, 0.125], [0.0625, 0.0625, 0.0625]] ), NumpyToTensor( keys = ['data', 'target'], cast_to = 'float' )] )",
+    "dataloader_val": "<nnunetv2.training.data_augmentation.custom_transforms.limited_length_multithreaded_augmenter.LimitedLenWrapper object at 0x7fe33e836fb0>",
+    "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader_3d.nnUNetDataLoader3D object at 0x7fe33e837340>",
+    "dataloader_val.num_processes": "8",
+    "dataloader_val.transform": "Compose ( [RemoveLabelTransform( output_key = 'seg', input_key = 'seg', replace_with = 0, remove_label = -1 ), RenameTransform( delete_old = True, out_key = 'target', in_key = 'seg' ), DownsampleSegForDSTransform2( axes = None, output_key = 'target', input_key = 'target', order = 0, ds_scales = [[1.0, 1.0, 1.0], [0.5, 0.5, 0.5], [0.25, 0.25, 0.25], [0.125, 0.125, 0.125], [0.0625, 0.0625, 0.0625]] ), NumpyToTensor( keys = ['data', 'target'], cast_to = 'float' )] )",
+    "dataset_json": "{'channel_names': {'0': 'CT'}, 'labels': {'background': 0, 'T1': 1, 'T2': 2, 'T3': 3, 'T4': 4, 'T5': 5, 'T6': 6, 'T7': 7, 'T8': 8, 'T9': 9, 'T10': 10, 'T11': 11, 'T12': 12, 'L1': 13, 'L2': 14, 'L3': 15, 'L4': 16, 'L5': 17, 'L6': 18, 'sacrum': 19, 'coccyx': 20, 'T13': 21}, 'numTraining': 1216, 'file_ending': '.nii.gz', 'dataset_name': 'VertebralBodies Thorax/Abdomen/Sacrum', 'reference': 'https://huggingface.co/datasets/fhofmann/VertebralBodiesCT-Labels', 'release': '2024-08-29', 'license': 'CC BY-SA 4.0', 'description': '1216 CT scans, with segmentations of the vertebral bodies in the thorax, abdomen, and sacrum. CT scans and initial labels derive from the VerSe (https://github.com/anjany/verse, license CC BY-SA 4.0) and TotalSegmentator (license CC BY 4.0 DEED) dataset. Files were processed, and labels were adapted as described in the README.md. A validation set is available separately.'}",
+    "device": "cuda:0",
+    "disable_checkpointing": "False",
+    "enable_deep_supervision": "True",
+    "fold": "all",
+    "folder_with_segs_from_previous_stage": "None",
+    "gpu_name": "NVIDIA A100-SXM4-40GB",
+    "grad_scaler": "<torch.cuda.amp.grad_scaler.GradScaler object at 0x7fe27970c0a0>",
+    "hostname": "srvdrai1",
+    "inference_allowed_mirroring_axes": "(0, 1, 2)",
+    "initial_lr": "0.01",
+    "is_cascaded": "False",
+    "is_ddp": "False",
+    "label_manager": "<nnunetv2.utilities.label_handling.label_handling.LabelManager object at 0x7fe27970c250>",
+    "local_rank": "0",
+    "log_file": "/workspace/data/nnUNet_results/Dataset601_VertebralBodies/nnUNetTrainer__nnUNetResEncUNetMPlans__3d_fullres/fold_all/training_log_2024_8_30_07_58_08.txt",
+    "logger": "<nnunetv2.training.logging.nnunet_logger.nnUNetLogger object at 0x7fe27970c100>",
+    "loss": "DeepSupervisionWrapper(\n  (loss): DC_and_CE_loss(\n    (ce): RobustCrossEntropyLoss()\n    (dc): MemoryEfficientSoftDiceLoss()\n  )\n)",
+    "lr_scheduler": "<nnunetv2.training.lr_scheduler.polylr.PolyLRScheduler object at 0x7fe27a935090>",
+    "my_init_kwargs": "{'plans': {'dataset_name': 'Dataset601_VertebralBodies', 'plans_name': 'nnUNetResEncUNetMPlans', 'original_median_spacing_after_transp': [1.5, 1.5, 1.5], 'original_median_shape_after_transp': [239, 252, 252], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 48, 'patch_size': [256, 256], 'median_image_size_in_voxels': [240.0, 240.0], 'spacing': [1.5, 1.5], 'normalization_schemes': ['CTNormalization'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [128, 128, 128], 'median_image_size_in_voxels': [235.0, 240.0, 240.0], 'spacing': [1.5, 1.5, 1.5], 'normalization_schemes': ['CTNormalization'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncM', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 10076.0, 'mean': 251.18613451837237, 'median': 216.0, 'min': -2048.0, 'percentile_00_5': -158.0, 'percentile_99_5': 1100.0, 'std': 205.1041738973246}}}, 'configuration': '3d_fullres', 'fold': 'all', 'dataset_json': {'channel_names': {'0': 'CT'}, 'labels': {'background': 0, 'T1': 1, 'T2': 2, 'T3': 3, 'T4': 4, 'T5': 5, 'T6': 6, 'T7': 7, 'T8': 8, 'T9': 9, 'T10': 10, 'T11': 11, 'T12': 12, 'L1': 13, 'L2': 14, 'L3': 15, 'L4': 16, 'L5': 17, 'L6': 18, 'sacrum': 19, 'coccyx': 20, 'T13': 21}, 'numTraining': 1216, 'file_ending': '.nii.gz', 'dataset_name': 'VertebralBodies Thorax/Abdomen/Sacrum', 'reference': 'https://huggingface.co/datasets/fhofmann/VertebralBodiesCT-Labels', 'release': '2024-08-29', 'license': 'CC BY-SA 4.0', 'description': '1216 CT scans, with segmentations of the vertebral bodies in the thorax, abdomen, and sacrum. CT scans and initial labels derive from the VerSe (https://github.com/anjany/verse, license CC BY-SA 4.0) and TotalSegmentator (license CC BY 4.0 DEED) dataset. Files were processed, and labels were adapted as described in the README.md. A validation set is available separately.'}, 'unpack_dataset': False, 'device': device(type='cuda')}",
+    "network": "OptimizedModule",
+    "num_epochs": "1000",
+    "num_input_channels": "1",
+    "num_iterations_per_epoch": "250",
+    "num_val_iterations_per_epoch": "50",
+    "optimizer": "SGD (\nParameter Group 0\n    dampening: 0\n    differentiable: False\n    foreach: None\n    fused: None\n    initial_lr: 0.01\n    lr: 0.0023598068186038313\n    maximize: False\n    momentum: 0.99\n    nesterov: True\n    weight_decay: 3e-05\n)",
+    "output_folder": "/workspace/data/nnUNet_results/Dataset601_VertebralBodies/nnUNetTrainer__nnUNetResEncUNetMPlans__3d_fullres/fold_all",
+    "output_folder_base": "/workspace/data/nnUNet_results/Dataset601_VertebralBodies/nnUNetTrainer__nnUNetResEncUNetMPlans__3d_fullres",
+    "oversample_foreground_percent": "0.33",
+    "plans_manager": "{'dataset_name': 'Dataset601_VertebralBodies', 'plans_name': 'nnUNetResEncUNetMPlans', 'original_median_spacing_after_transp': [1.5, 1.5, 1.5], 'original_median_shape_after_transp': [239, 252, 252], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'configurations': {'2d': {'data_identifier': 'nnUNetPlans_2d', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 48, 'patch_size': [256, 256], 'median_image_size_in_voxels': [240.0, 240.0], 'spacing': [1.5, 1.5], 'normalization_schemes': ['CTNormalization'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 7, 'features_per_stage': [32, 64, 128, 256, 512, 512, 512], 'conv_op': 'torch.nn.modules.conv.Conv2d', 'kernel_sizes': [[3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm2d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': True}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [128, 128, 128], 'median_image_size_in_voxels': [235.0, 240.0, 240.0], 'spacing': [1.5, 1.5, 1.5], 'normalization_schemes': ['CTNormalization'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False}}, 'experiment_planner_used': 'nnUNetPlannerResEncM', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 10076.0, 'mean': 251.18613451837237, 'median': 216.0, 'min': -2048.0, 'percentile_00_5': -158.0, 'percentile_99_5': 1100.0, 'std': 205.1041738973246}}}",
+    "preprocessed_dataset_folder": "/workspace/data2/nnUNet_preprocessed/Dataset601_VertebralBodies/nnUNetPlans_3d_fullres",
+    "preprocessed_dataset_folder_base": "/workspace/data2/nnUNet_preprocessed/Dataset601_VertebralBodies",
+    "save_every": "50",
+    "torch_version": "2.3.0+cu121",
+    "unpack_dataset": "False",
+    "was_initialized": "True",
+    "weight_decay": "3e-05"
+}

nnUNetTrainer__nnUNetResEncUNetMPlans__3d_fullres/fold_all/logs/training_log_2024_8_30_07_58_08.txt

@@ -0,0 +1,1432 @@
+
+#######################################################################
+Please cite the following paper when using nnU-Net:
+Isensee, F., Jaeger, P. F., Kohl, S. A., Petersen, J., & Maier-Hein, K. H. (2021). nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 18(2), 203-211.
+#######################################################################
+ 
+2024-08-30 07:58:11.563206: Using torch.compile... 
+2024-08-30 07:58:18.089942: do_dummy_2d_data_aug: False 
+
+This is the configuration used by this training:
+Configuration name: 3d_fullres
+ {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [128, 128, 128], 'median_image_size_in_voxels': [235.0, 240.0, 240.0], 'spacing': [1.5, 1.5, 1.5], 'normalization_schemes': ['CTNormalization'], 'use_mask_for_norm': [False], 'resampling_fn_data': 'resample_data_or_seg_to_shape', 'resampling_fn_seg': 'resample_data_or_seg_to_shape', 'resampling_fn_data_kwargs': {'is_seg': False, 'order': 3, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_seg_kwargs': {'is_seg': True, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'resampling_fn_probabilities': 'resample_data_or_seg_to_shape', 'resampling_fn_probabilities_kwargs': {'is_seg': False, 'order': 1, 'order_z': 0, 'force_separate_z': None}, 'architecture': {'network_class_name': 'dynamic_network_architectures.architectures.unet.ResidualEncoderUNet', 'arch_kwargs': {'n_stages': 6, 'features_per_stage': [32, 64, 128, 256, 320, 320], 'conv_op': 'torch.nn.modules.conv.Conv3d', 'kernel_sizes': [[3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 1, 1, 1, 1], 'conv_bias': True, 'norm_op': 'torch.nn.modules.instancenorm.InstanceNorm3d', 'norm_op_kwargs': {'eps': 1e-05, 'affine': True}, 'dropout_op': None, 'dropout_op_kwargs': None, 'nonlin': 'torch.nn.LeakyReLU', 'nonlin_kwargs': {'inplace': True}, 'deep_supervision': True}, '_kw_requires_import': ['conv_op', 'norm_op', 'dropout_op', 'nonlin']}, 'batch_dice': False} 
+ 
+These are the global plan.json settings:
+ {'dataset_name': 'Dataset601_VertebralBodies', 'plans_name': 'nnUNetResEncUNetMPlans', 'original_median_spacing_after_transp': [1.5, 1.5, 1.5], 'original_median_shape_after_transp': [239, 252, 252], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'experiment_planner_used': 'nnUNetPlannerResEncM', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 10076.0, 'mean': 251.18613451837237, 'median': 216.0, 'min': -2048.0, 'percentile_00_5': -158.0, 'percentile_99_5': 1100.0, 'std': 205.1041738973246}}} 
+ 
+2024-08-30 07:58:21.674600: Unable to plot network architecture: nnUNet_compile is enabled! 
+2024-08-30 07:58:22.131281:  
+2024-08-30 07:58:22.132290: Epoch 800 
+2024-08-30 07:58:22.133175: Current learning rate: 0.00235 
+2024-08-30 08:03:45.428643: train_loss -0.8298 
+2024-08-30 08:03:45.430047: val_loss -0.8349 
+2024-08-30 08:03:45.431003: Pseudo dice [0.9243, 0.9449, 0.9335, 0.8995, 0.8846, 0.9162, 0.9295, 0.9348, 0.9324, 0.913, 0.8889, 0.8969, 0.9247, 0.9191, 0.9173, 0.8506, 0.7589, 0.0, 0.9421, nan, 0.0] 
+2024-08-30 08:03:45.431701: Epoch time: 323.3 s 
+2024-08-30 08:03:46.966835:  
+2024-08-30 08:03:46.967671: Epoch 801 
+2024-08-30 08:03:46.968291: Current learning rate: 0.00234 
+2024-08-30 08:04:38.504457: train_loss -0.8275 
+2024-08-30 08:04:38.505691: val_loss -0.841 
+2024-08-30 08:04:38.506451: Pseudo dice [0.898, 0.9391, 0.937, 0.8957, 0.8836, 0.8873, 0.8928, 0.9155, 0.9086, 0.9039, 0.8952, 0.9038, 0.9531, 0.9499, 0.9427, 0.9347, 0.8904, 0.0, 0.9463, nan, 0.0] 
+2024-08-30 08:04:38.507064: Epoch time: 51.54 s 
+2024-08-30 08:04:39.909204:  
+2024-08-30 08:04:39.909985: Epoch 802 
+2024-08-30 08:04:39.910569: Current learning rate: 0.00233 
+2024-08-30 08:05:32.788320: train_loss -0.8287 
+2024-08-30 08:05:32.791261: val_loss -0.8546 
+2024-08-30 08:05:32.792677: Pseudo dice [0.9208, 0.946, 0.9223, 0.9167, 0.9083, 0.9257, 0.9309, 0.9319, 0.9459, 0.9684, 0.9716, 0.9611, 0.9578, 0.9272, 0.9065, 0.8822, 0.8567, 0.0, 0.9434, nan, nan] 
+2024-08-30 08:05:32.793562: Epoch time: 52.88 s 
+2024-08-30 08:05:34.988766:  
+2024-08-30 08:05:34.989611: Epoch 803 
+2024-08-30 08:05:34.990308: Current learning rate: 0.00232 
+2024-08-30 08:06:26.532500: train_loss -0.8275 
+2024-08-30 08:06:26.533911: val_loss -0.8135 
+2024-08-30 08:06:26.535242: Pseudo dice [0.8438, 0.8301, 0.7534, 0.7937, 0.832, 0.7874, 0.7203, 0.7308, 0.7882, 0.8375, 0.8646, 0.8654, 0.8509, 0.8684, 0.8871, 0.9275, 0.9166, 0.0, 0.9492, nan, nan] 
+2024-08-30 08:06:26.536349: Epoch time: 51.55 s 
+2024-08-30 08:06:28.116220:  
+2024-08-30 08:06:28.116964: Epoch 804 
+2024-08-30 08:06:28.117596: Current learning rate: 0.00231 
+2024-08-30 08:07:22.739134: train_loss -0.826 
+2024-08-30 08:07:22.741840: val_loss -0.8538 
+2024-08-30 08:07:22.742763: Pseudo dice [0.9266, 0.9505, 0.9481, 0.9558, 0.9594, 0.9632, 0.9063, 0.9122, 0.9563, 0.9658, 0.9565, 0.9547, 0.9174, 0.8873, 0.9158, 0.9345, 0.8662, 0.0, 0.9411, nan, nan] 
+2024-08-30 08:07:22.743593: Epoch time: 54.63 s 
+2024-08-30 08:07:24.305591:  
+2024-08-30 08:07:24.306453: Epoch 805 
+2024-08-30 08:07:24.307409: Current learning rate: 0.0023 
+2024-08-30 08:08:16.513894: train_loss -0.8196 
+2024-08-30 08:08:16.515238: val_loss -0.8443 
+2024-08-30 08:08:16.516107: Pseudo dice [0.9205, 0.9388, 0.943, 0.9425, 0.9243, 0.9212, 0.9322, 0.9344, 0.9331, 0.9398, 0.9214, 0.9125, 0.9076, 0.8794, 0.8772, 0.8927, 0.843, 0.0, 0.9436, nan, nan] 
+2024-08-30 08:08:16.516857: Epoch time: 52.21 s 
+2024-08-30 08:08:17.939694:  
+2024-08-30 08:08:17.940573: Epoch 806 
+2024-08-30 08:08:17.941231: Current learning rate: 0.00229 
+2024-08-30 08:09:10.630731: train_loss -0.8309 
+2024-08-30 08:09:10.632896: val_loss -0.8427 
+2024-08-30 08:09:10.634222: Pseudo dice [0.921, 0.9139, 0.8767, 0.8661, 0.8639, 0.8846, 0.8672, 0.8613, 0.8769, 0.8845, 0.9237, 0.95, 0.9552, 0.9689, 0.9725, 0.9647, 0.9313, 0.0, 0.935, nan, nan] 
+2024-08-30 08:09:10.635642: Epoch time: 52.69 s 
+2024-08-30 08:09:12.215455:  
+2024-08-30 08:09:12.216362: Epoch 807 
+2024-08-30 08:09:12.216984: Current learning rate: 0.00228 
+2024-08-30 08:10:03.502030: train_loss -0.8369 
+2024-08-30 08:10:03.503799: val_loss -0.8569 
+2024-08-30 08:10:03.504809: Pseudo dice [0.919, 0.9469, 0.933, 0.9408, 0.9519, 0.9526, 0.9455, 0.9432, 0.9465, 0.9597, 0.9574, 0.9595, 0.963, 0.952, 0.9375, 0.9212, 0.8982, 0.0, 0.9518, nan, nan] 
+2024-08-30 08:10:03.505649: Epoch time: 51.29 s 
+2024-08-30 08:10:04.943223:  
+2024-08-30 08:10:04.944008: Epoch 808 
+2024-08-30 08:10:04.944632: Current learning rate: 0.00226 
+2024-08-30 08:10:55.502361: train_loss -0.8244 
+2024-08-30 08:10:55.504170: val_loss -0.8443 
+2024-08-30 08:10:55.504948: Pseudo dice [0.9041, 0.914, 0.9259, 0.9342, 0.9182, 0.9123, 0.9433, 0.9522, 0.9473, 0.9455, 0.9306, 0.9368, 0.9317, 0.9557, 0.9111, 0.8272, 0.7577, 0.0, 0.945, nan, 0.0] 
+2024-08-30 08:10:55.505560: Epoch time: 50.56 s 
+2024-08-30 08:10:56.940577:  
+2024-08-30 08:10:56.941448: Epoch 809 
+2024-08-30 08:10:56.942101: Current learning rate: 0.00225 
+2024-08-30 08:11:49.197581: train_loss -0.8294 
+2024-08-30 08:11:49.199519: val_loss -0.8529 
+2024-08-30 08:11:49.201134: Pseudo dice [0.9261, 0.9319, 0.933, 0.9313, 0.9415, 0.9552, 0.9307, 0.8648, 0.8206, 0.8718, 0.9199, 0.9415, 0.9534, 0.9488, 0.9468, 0.9375, 0.9104, 0.0, 0.9443, nan, 0.0] 
+2024-08-30 08:11:49.203180: Epoch time: 52.26 s 
+2024-08-30 08:11:50.628483:  
+2024-08-30 08:11:50.629328: Epoch 810 
+2024-08-30 08:11:50.629950: Current learning rate: 0.00224 
+2024-08-30 08:12:44.200685: train_loss -0.8284 
+2024-08-30 08:12:44.202734: val_loss -0.8515 
+2024-08-30 08:12:44.203762: Pseudo dice [0.9285, 0.9179, 0.9271, 0.9394, 0.8995, 0.9022, 0.9364, 0.912, 0.8835, 0.9053, 0.9436, 0.9638, 0.9683, 0.9266, 0.8912, 0.9094, 0.8941, 0.0, 0.9413, nan, nan] 
+2024-08-30 08:12:44.204494: Epoch time: 53.58 s 
+2024-08-30 08:12:45.877465:  
+2024-08-30 08:12:45.878678: Epoch 811 
+2024-08-30 08:12:45.879470: Current learning rate: 0.00223 
+2024-08-30 08:13:35.151080: train_loss -0.818 
+2024-08-30 08:13:35.152261: val_loss -0.8408 
+2024-08-30 08:13:35.153176: Pseudo dice [0.9395, 0.924, 0.8925, 0.8974, 0.8862, 0.8433, 0.8485, 0.8944, 0.8938, 0.8996, 0.9513, 0.9481, 0.9603, 0.9192, 0.8756, 0.8428, 0.8499, nan, 0.945, nan, nan] 
+2024-08-30 08:13:35.153873: Epoch time: 49.28 s 
+2024-08-30 08:13:36.626055:  
+2024-08-30 08:13:36.626868: Epoch 812 
+2024-08-30 08:13:36.627525: Current learning rate: 0.00222 
+2024-08-30 08:14:25.278524: train_loss -0.8262 
+2024-08-30 08:14:25.281066: val_loss -0.841 
+2024-08-30 08:14:25.282288: Pseudo dice [0.9162, 0.9452, 0.956, 0.9518, 0.9345, 0.912, 0.9154, 0.9162, 0.8872, 0.874, 0.8917, 0.8981, 0.9605, 0.9475, 0.9276, 0.877, 0.7966, 0.0, 0.942, nan, 0.0] 
+2024-08-30 08:14:25.283130: Epoch time: 48.66 s 
+2024-08-30 08:14:26.765371:  
+2024-08-30 08:14:26.766247: Epoch 813 
+2024-08-30 08:14:26.766987: Current learning rate: 0.00221 
+2024-08-30 08:15:17.590965: train_loss -0.8314 
+2024-08-30 08:15:17.593155: val_loss -0.8355 
+2024-08-30 08:15:17.594711: Pseudo dice [0.9184, 0.9248, 0.9114, 0.9097, 0.8947, 0.8811, 0.8787, 0.8975, 0.8757, 0.881, 0.8705, 0.9022, 0.8961, 0.8504, 0.8486, 0.8488, 0.8674, 0.0, 0.9501, nan, nan] 
+2024-08-30 08:15:17.596956: Epoch time: 50.83 s 
+2024-08-30 08:15:19.065178:  
+2024-08-30 08:15:19.066002: Epoch 814 
+2024-08-30 08:15:19.066709: Current learning rate: 0.0022 
+2024-08-30 08:16:10.943189: train_loss -0.8306 
+2024-08-30 08:16:10.955420: val_loss -0.8514 
+2024-08-30 08:16:10.956509: Pseudo dice [0.9252, 0.9536, 0.952, 0.9485, 0.9561, 0.9566, 0.9676, 0.9678, 0.9677, 0.9643, 0.95, 0.9457, 0.9073, 0.9278, 0.9319, 0.905, 0.8595, 0.0, 0.9426, nan, 0.0] 
+2024-08-30 08:16:10.957301: Epoch time: 51.88 s 
+2024-08-30 08:16:13.075094:  
+2024-08-30 08:16:13.075996: Epoch 815 
+2024-08-30 08:16:13.076652: Current learning rate: 0.00219 
+2024-08-30 08:17:06.350408: train_loss -0.8157 
+2024-08-30 08:17:06.351511: val_loss -0.842 
+2024-08-30 08:17:06.352343: Pseudo dice [0.9262, 0.9019, 0.8707, 0.8943, 0.9377, 0.9371, 0.9258, 0.9275, 0.9172, 0.9048, 0.9104, 0.9174, 0.9221, 0.9076, 0.8824, 0.8531, 0.824, 0.0, 0.9371, nan, nan] 
+2024-08-30 08:17:06.353109: Epoch time: 53.28 s 
+2024-08-30 08:17:07.755994:  
+2024-08-30 08:17:07.757254: Epoch 816 
+2024-08-30 08:17:07.757989: Current learning rate: 0.00218 
+2024-08-30 08:17:58.611602: train_loss -0.8396 
+2024-08-30 08:17:58.613509: val_loss -0.8431 
+2024-08-30 08:17:58.614527: Pseudo dice [0.9309, 0.9434, 0.924, 0.9131, 0.9369, 0.9263, 0.9033, 0.9149, 0.9345, 0.9132, 0.864, 0.8876, 0.9373, 0.9518, 0.9265, 0.8795, 0.8378, 0.0, 0.9452, nan, nan] 
+2024-08-30 08:17:58.615318: Epoch time: 50.86 s 
+2024-08-30 08:18:00.178677:  
+2024-08-30 08:18:00.179557: Epoch 817 
+2024-08-30 08:18:00.180188: Current learning rate: 0.00217 
+2024-08-30 08:18:54.053931: train_loss -0.8233 
+2024-08-30 08:18:54.055542: val_loss -0.8204 
+2024-08-30 08:18:54.056728: Pseudo dice [0.8693, 0.8951, 0.8895, 0.8801, 0.8989, 0.9228, 0.9474, 0.9307, 0.9131, 0.9114, 0.9073, 0.9106, 0.8527, 0.7768, 0.7465, 0.7157, 0.7106, 0.0, 0.9297, nan, nan] 
+2024-08-30 08:18:54.057687: Epoch time: 53.88 s 
+2024-08-30 08:18:55.619385:  
+2024-08-30 08:18:55.620272: Epoch 818 
+2024-08-30 08:18:55.620930: Current learning rate: 0.00216 
+2024-08-30 08:19:46.511590: train_loss -0.8308 
+2024-08-30 08:19:46.513474: val_loss -0.8539 
+2024-08-30 08:19:46.514338: Pseudo dice [0.9501, 0.9587, 0.9606, 0.9649, 0.9644, 0.9682, 0.958, 0.9286, 0.8767, 0.8543, 0.8755, 0.8759, 0.9143, 0.8994, 0.8829, 0.9086, 0.9096, 0.0, 0.9338, nan, 0.0] 
+2024-08-30 08:19:46.515176: Epoch time: 50.9 s 
+2024-08-30 08:19:48.165145:  
+2024-08-30 08:19:48.166027: Epoch 819 
+2024-08-30 08:19:48.166734: Current learning rate: 0.00215 
+2024-08-30 08:20:38.473231: train_loss -0.8336 
+2024-08-30 08:20:38.475305: val_loss -0.8618 
+2024-08-30 08:20:38.476350: Pseudo dice [0.9202, 0.9409, 0.9411, 0.9425, 0.9466, 0.9563, 0.9615, 0.9633, 0.9232, 0.896, 0.9343, 0.956, 0.9608, 0.969, 0.9541, 0.9523, 0.9448, nan, 0.949, nan, nan] 
+2024-08-30 08:20:38.477642: Epoch time: 50.31 s 
+2024-08-30 08:20:40.186954:  
+2024-08-30 08:20:40.187847: Epoch 820 
+2024-08-30 08:20:40.188530: Current learning rate: 0.00214 
+2024-08-30 08:21:33.444569: train_loss -0.8376 
+2024-08-30 08:21:33.446877: val_loss -0.8386 
+2024-08-30 08:21:33.448053: Pseudo dice [0.9123, 0.9334, 0.9372, 0.9498, 0.9616, 0.9515, 0.9048, 0.8704, 0.901, 0.9199, 0.9388, 0.8919, 0.9033, 0.9588, 0.8961, 0.8513, 0.8759, 0.0, 0.9404, nan, nan] 
+2024-08-30 08:21:33.448708: Epoch time: 53.26 s 
+2024-08-30 08:21:35.020370:  
+2024-08-30 08:21:35.021193: Epoch 821 
+2024-08-30 08:21:35.021801: Current learning rate: 0.00213 
+2024-08-30 08:22:28.024297: train_loss -0.8313 
+2024-08-30 08:22:28.025396: val_loss -0.835 
+2024-08-30 08:22:28.026192: Pseudo dice [0.9202, 0.9456, 0.9307, 0.928, 0.9179, 0.9048, 0.9096, 0.9235, 0.9275, 0.94, 0.9337, 0.9078, 0.8945, 0.8613, 0.8258, 0.7934, 0.7485, 0.0, 0.943, nan, nan] 
+2024-08-30 08:22:28.027001: Epoch time: 53.01 s 
+2024-08-30 08:22:29.338894:  
+2024-08-30 08:22:29.339920: Epoch 822 
+2024-08-30 08:22:29.340533: Current learning rate: 0.00212 
+2024-08-30 08:23:23.282223: train_loss -0.8291 
+2024-08-30 08:23:23.285022: val_loss -0.8318 
+2024-08-30 08:23:23.285893: Pseudo dice [0.9273, 0.9355, 0.9379, 0.931, 0.9015, 0.8874, 0.8629, 0.8341, 0.7981, 0.8074, 0.8294, 0.8404, 0.8596, 0.8865, 0.9295, 0.9218, 0.8733, 0.0, 0.9337, nan, nan] 
+2024-08-30 08:23:23.286742: Epoch time: 53.95 s 
+2024-08-30 08:23:24.822444:  
+2024-08-30 08:23:24.823317: Epoch 823 
+2024-08-30 08:23:24.823978: Current learning rate: 0.0021 
+2024-08-30 08:24:16.636767: train_loss -0.8283 
+2024-08-30 08:24:16.637908: val_loss -0.8405 
+2024-08-30 08:24:16.638814: Pseudo dice [0.8966, 0.869, 0.819, 0.8487, 0.843, 0.8508, 0.8709, 0.8922, 0.9166, 0.9319, 0.9118, 0.8867, 0.9029, 0.9309, 0.9442, 0.9578, 0.9509, nan, 0.9552, nan, nan] 
+2024-08-30 08:24:16.639580: Epoch time: 51.82 s 
+2024-08-30 08:24:17.981040:  
+2024-08-30 08:24:17.981936: Epoch 824 
+2024-08-30 08:24:17.982702: Current learning rate: 0.00209 
+2024-08-30 08:25:09.459312: train_loss -0.8332 
+2024-08-30 08:25:09.461983: val_loss -0.847 
+2024-08-30 08:25:09.463172: Pseudo dice [0.9129, 0.9362, 0.9282, 0.9401, 0.9414, 0.9529, 0.9507, 0.9306, 0.9334, 0.9195, 0.9052, 0.885, 0.9312, 0.9253, 0.9305, 0.9056, 0.8777, 0.0, 0.9459, nan, 0.0] 
+2024-08-30 08:25:09.464164: Epoch time: 51.48 s 
+2024-08-30 08:25:11.123589:  
+2024-08-30 08:25:11.124481: Epoch 825 
+2024-08-30 08:25:11.125131: Current learning rate: 0.00208 
+2024-08-30 08:26:03.669369: train_loss -0.8402 
+2024-08-30 08:26:03.671208: val_loss -0.8566 
+2024-08-30 08:26:03.672660: Pseudo dice [0.9124, 0.9442, 0.942, 0.952, 0.9593, 0.9669, 0.9522, 0.9513, 0.9681, 0.9742, 0.9675, 0.9643, 0.9618, 0.9555, 0.9259, 0.9126, 0.8724, 0.0, 0.9548, nan, nan] 
+2024-08-30 08:26:03.673769: Epoch time: 52.55 s 
+2024-08-30 08:26:05.188530:  
+2024-08-30 08:26:05.189483: Epoch 826 
+2024-08-30 08:26:05.190189: Current learning rate: 0.00207 
+2024-08-30 08:26:56.577958: train_loss -0.8289 
+2024-08-30 08:26:56.580436: val_loss -0.8455 
+2024-08-30 08:26:56.581439: Pseudo dice [0.8892, 0.9066, 0.9185, 0.953, 0.9587, 0.9408, 0.914, 0.8965, 0.9056, 0.9023, 0.8851, 0.9108, 0.9464, 0.908, 0.9074, 0.8988, 0.8317, 0.0, 0.9439, nan, nan] 
+2024-08-30 08:26:56.582162: Epoch time: 51.39 s 
+2024-08-30 08:26:57.894692:  
+2024-08-30 08:26:57.895434: Epoch 827 
+2024-08-30 08:26:57.896052: Current learning rate: 0.00206 
+2024-08-30 08:27:47.155619: train_loss -0.8429 
+2024-08-30 08:27:47.157499: val_loss -0.8626 
+2024-08-30 08:27:47.158466: Pseudo dice [0.9454, 0.9514, 0.9409, 0.9206, 0.913, 0.9173, 0.9452, 0.9572, 0.9547, 0.9567, 0.9601, 0.9514, 0.9524, 0.9657, 0.967, 0.9685, 0.9502, nan, 0.947, nan, nan] 
+2024-08-30 08:27:47.159390: Epoch time: 49.26 s 
+2024-08-30 08:27:47.160301: Yayy! New best EMA pseudo Dice: 0.8698 
+2024-08-30 08:27:52.292960:  
+2024-08-30 08:27:52.293909: Epoch 828 
+2024-08-30 08:27:52.294642: Current learning rate: 0.00205 
+2024-08-30 08:28:43.239846: train_loss -0.8371 
+2024-08-30 08:28:43.241592: val_loss -0.8352 
+2024-08-30 08:28:43.243298: Pseudo dice [0.9016, 0.9249, 0.9414, 0.9493, 0.9637, 0.9653, 0.963, 0.9665, 0.9624, 0.9481, 0.9269, 0.9263, 0.8967, 0.8463, 0.7995, 0.7497, 0.7564, 0.0, 0.9564, nan, 0.0] 
+2024-08-30 08:28:43.244061: Epoch time: 50.95 s 
+2024-08-30 08:28:44.624213:  
+2024-08-30 08:28:44.625808: Epoch 829 
+2024-08-30 08:28:44.626519: Current learning rate: 0.00204 
+2024-08-30 08:29:37.756772: train_loss -0.8299 
+2024-08-30 08:29:37.757998: val_loss -0.8568 
+2024-08-30 08:29:37.758874: Pseudo dice [0.9323, 0.9466, 0.942, 0.9373, 0.9468, 0.9594, 0.966, 0.9621, 0.9544, 0.9604, 0.9596, 0.9597, 0.9711, 0.9682, 0.9462, 0.8969, 0.8514, 0.0, 0.9407, nan, nan] 
+2024-08-30 08:29:37.759763: Epoch time: 53.14 s 
+2024-08-30 08:29:39.056917:  
+2024-08-30 08:29:39.057681: Epoch 830 
+2024-08-30 08:29:39.058367: Current learning rate: 0.00203 
+2024-08-30 08:30:35.667782: train_loss -0.8141 
+2024-08-30 08:30:35.669878: val_loss -0.8573 
+2024-08-30 08:30:35.670662: Pseudo dice [0.9336, 0.9498, 0.9528, 0.9497, 0.9628, 0.9605, 0.9664, 0.9729, 0.9727, 0.9722, 0.9661, 0.9445, 0.9273, 0.9296, 0.9067, 0.8506, 0.8598, 0.0, 0.9463, nan, nan] 
+2024-08-30 08:30:35.671288: Epoch time: 56.61 s 
+2024-08-30 08:30:35.671895: Yayy! New best EMA pseudo Dice: 0.8699 
+2024-08-30 08:30:40.406667:  
+2024-08-30 08:30:40.407565: Epoch 831 
+2024-08-30 08:30:40.408188: Current learning rate: 0.00202 
+2024-08-30 08:31:33.459675: train_loss -0.8337 
+2024-08-30 08:31:33.460820: val_loss -0.8456 
+2024-08-30 08:31:33.461634: Pseudo dice [0.9263, 0.9389, 0.9259, 0.9389, 0.9516, 0.9527, 0.9041, 0.8772, 0.9082, 0.9303, 0.9191, 0.9362, 0.9529, 0.9466, 0.9279, 0.8727, 0.8546, 0.0, 0.9462, nan, nan] 
+2024-08-30 08:31:33.462340: Epoch time: 53.06 s 
+2024-08-30 08:31:33.463093: Yayy! New best EMA pseudo Dice: 0.8703 
+2024-08-30 08:31:37.837598:  
+2024-08-30 08:31:37.838884: Epoch 832 
+2024-08-30 08:31:37.839515: Current learning rate: 0.00201 
+2024-08-30 08:32:29.919479: train_loss -0.8371 
+2024-08-30 08:32:29.921590: val_loss -0.8504 
+2024-08-30 08:32:29.922781: Pseudo dice [0.9288, 0.9495, 0.9564, 0.958, 0.9473, 0.9519, 0.9621, 0.9535, 0.9327, 0.9095, 0.8838, 0.8829, 0.9102, 0.8988, 0.8911, 0.9062, 0.9134, nan, 0.945, nan, nan] 
+2024-08-30 08:32:29.923557: Epoch time: 52.08 s 
+2024-08-30 08:32:29.924380: Yayy! New best EMA pseudo Dice: 0.876 
+2024-08-30 08:32:34.433934:  
+2024-08-30 08:32:34.434817: Epoch 833 
+2024-08-30 08:32:34.435482: Current learning rate: 0.002 
+2024-08-30 08:33:25.896383: train_loss -0.8359 
+2024-08-30 08:33:25.898402: val_loss -0.8484 
+2024-08-30 08:33:25.899250: Pseudo dice [0.9355, 0.9488, 0.9456, 0.9063, 0.908, 0.9442, 0.9221, 0.9163, 0.9263, 0.9353, 0.9369, 0.9373, 0.9159, 0.9038, 0.9198, 0.8978, 0.9145, 0.0, 0.9548, nan, nan] 
+2024-08-30 08:33:25.900124: Epoch time: 51.47 s 
+2024-08-30 08:33:25.900837: Yayy! New best EMA pseudo Dice: 0.8761 
+2024-08-30 08:33:30.282899:  
+2024-08-30 08:33:30.283942: Epoch 834 
+2024-08-30 08:33:30.284630: Current learning rate: 0.00199 
+2024-08-30 08:34:22.623142: train_loss -0.8318 
+2024-08-30 08:34:22.624439: val_loss -0.85 
+2024-08-30 08:34:22.625240: Pseudo dice [0.9323, 0.9508, 0.9449, 0.9214, 0.8995, 0.8596, 0.8281, 0.8324, 0.8634, 0.909, 0.9332, 0.9516, 0.9706, 0.9711, 0.9661, 0.94, 0.9221, 0.0, 0.9507, nan, nan] 
+2024-08-30 08:34:22.626162: Epoch time: 52.34 s 
+2024-08-30 08:34:23.944872:  
+2024-08-30 08:34:23.945734: Epoch 835 
+2024-08-30 08:34:23.946381: Current learning rate: 0.00198 
+2024-08-30 08:35:17.867495: train_loss -0.8387 
+2024-08-30 08:35:17.870377: val_loss -0.8609 
+2024-08-30 08:35:17.871281: Pseudo dice [0.9046, 0.9497, 0.9472, 0.9573, 0.9583, 0.9423, 0.9298, 0.9285, 0.953, 0.9676, 0.9662, 0.9654, 0.9695, 0.9583, 0.961, 0.9583, 0.902, 0.0, 0.9462, nan, nan] 
+2024-08-30 08:35:17.872017: Epoch time: 53.93 s 
+2024-08-30 08:35:17.872632: Yayy! New best EMA pseudo Dice: 0.8778 
+2024-08-30 08:35:22.239060:  
+2024-08-30 08:35:22.239956: Epoch 836 
+2024-08-30 08:35:22.240781: Current learning rate: 0.00196 
+2024-08-30 08:36:15.614010: train_loss -0.8287 
+2024-08-30 08:36:15.616992: val_loss -0.8443 
+2024-08-30 08:36:15.617964: Pseudo dice [0.9094, 0.9418, 0.9388, 0.9396, 0.9294, 0.9082, 0.9189, 0.9234, 0.877, 0.8867, 0.8986, 0.8993, 0.9003, 0.8602, 0.8569, 0.9135, 0.8982, 0.0, 0.9551, nan, nan] 
+2024-08-30 08:36:15.619199: Epoch time: 53.38 s 
+2024-08-30 08:36:16.960143:  
+2024-08-30 08:36:16.961052: Epoch 837 
+2024-08-30 08:36:16.961776: Current learning rate: 0.00195 
+2024-08-30 08:37:07.388469: train_loss -0.8272 
+2024-08-30 08:37:07.390649: val_loss -0.8489 
+2024-08-30 08:37:07.391516: Pseudo dice [0.9166, 0.9045, 0.8891, 0.8701, 0.8837, 0.8848, 0.9035, 0.905, 0.9083, 0.915, 0.93, 0.9457, 0.9486, 0.9387, 0.9536, 0.9627, 0.9111, 0.0, 0.9435, nan, nan] 
+2024-08-30 08:37:07.392443: Epoch time: 50.43 s 
+2024-08-30 08:37:08.861180:  
+2024-08-30 08:37:08.862068: Epoch 838 
+2024-08-30 08:37:08.862830: Current learning rate: 0.00194 
+2024-08-30 08:38:05.148122: train_loss -0.8258 
+2024-08-30 08:38:05.150485: val_loss -0.8404 
+2024-08-30 08:38:05.151386: Pseudo dice [0.8883, 0.8898, 0.9005, 0.8151, 0.8185, 0.8683, 0.8654, 0.8936, 0.9216, 0.9188, 0.9192, 0.9373, 0.958, 0.9664, 0.9683, 0.9513, 0.9398, 0.0, 0.9475, nan, nan] 
+2024-08-30 08:38:05.152148: Epoch time: 56.29 s 
+2024-08-30 08:38:06.483434:  
+2024-08-30 08:38:06.484345: Epoch 839 
+2024-08-30 08:38:06.485116: Current learning rate: 0.00193 
+2024-08-30 08:38:57.670839: train_loss -0.8387 
+2024-08-30 08:38:57.673256: val_loss -0.8605 
+2024-08-30 08:38:57.674401: Pseudo dice [0.9394, 0.9507, 0.9524, 0.9487, 0.954, 0.9456, 0.9253, 0.9307, 0.9418, 0.9422, 0.9399, 0.9486, 0.9598, 0.9701, 0.9737, 0.9696, 0.933, nan, 0.9466, nan, 0.0] 
+2024-08-30 08:38:57.675169: Epoch time: 51.19 s 
+2024-08-30 08:38:59.269151:  
+2024-08-30 08:38:59.270121: Epoch 840 
+2024-08-30 08:38:59.270879: Current learning rate: 0.00192 
+2024-08-30 08:39:53.401448: train_loss -0.841 
+2024-08-30 08:39:53.402682: val_loss -0.8495 
+2024-08-30 08:39:53.403647: Pseudo dice [0.9174, 0.9366, 0.9184, 0.9149, 0.9108, 0.9191, 0.9439, 0.9464, 0.9191, 0.9199, 0.9207, 0.9107, 0.9452, 0.9343, 0.9256, 0.9306, 0.924, 0.0, 0.9462, nan, 0.0] 
+2024-08-30 08:39:53.404493: Epoch time: 54.14 s 
+2024-08-30 08:39:54.918965:  
+2024-08-30 08:39:54.919803: Epoch 841 
+2024-08-30 08:39:54.920514: Current learning rate: 0.00191 
+2024-08-30 08:40:47.035358: train_loss -0.8342 
+2024-08-30 08:40:47.037726: val_loss -0.8393 
+2024-08-30 08:40:47.038593: Pseudo dice [0.9156, 0.9198, 0.9055, 0.9006, 0.8871, 0.8714, 0.8915, 0.8892, 0.8902, 0.8827, 0.8617, 0.8779, 0.9057, 0.8994, 0.8778, 0.8781, 0.8957, 0.0, 0.9567, nan, nan] 
+2024-08-30 08:40:47.039367: Epoch time: 52.12 s 
+2024-08-30 08:40:48.478051:  
+2024-08-30 08:40:48.478997: Epoch 842 
+2024-08-30 08:40:48.479679: Current learning rate: 0.0019 
+2024-08-30 08:41:39.741175: train_loss -0.8354 
+2024-08-30 08:41:39.742464: val_loss -0.8438 
+2024-08-30 08:41:39.743416: Pseudo dice [0.9415, 0.9181, 0.883, 0.8168, 0.8267, 0.9008, 0.9337, 0.9525, 0.9629, 0.9413, 0.925, 0.9168, 0.886, 0.8656, 0.8794, 0.9152, 0.9164, 0.0, 0.9501, nan, 0.0] 
+2024-08-30 08:41:39.744118: Epoch time: 51.27 s 
+2024-08-30 08:41:41.238216:  
+2024-08-30 08:41:41.239219: Epoch 843 
+2024-08-30 08:41:41.239991: Current learning rate: 0.00189 
+2024-08-30 08:42:34.509820: train_loss -0.8206 
+2024-08-30 08:42:34.511800: val_loss -0.8529 
+2024-08-30 08:42:34.512738: Pseudo dice [0.9251, 0.9411, 0.947, 0.9551, 0.9312, 0.9193, 0.9308, 0.9373, 0.9253, 0.9274, 0.9375, 0.9481, 0.9566, 0.9615, 0.9748, 0.9434, 0.826, 0.0, 0.9267, nan, nan] 
+2024-08-30 08:42:34.513470: Epoch time: 53.27 s 
+2024-08-30 08:42:35.918376:  
+2024-08-30 08:42:35.919334: Epoch 844 
+2024-08-30 08:42:35.920022: Current learning rate: 0.00188 
+2024-08-30 08:43:29.541205: train_loss -0.8317 
+2024-08-30 08:43:29.542472: val_loss -0.8121 
+2024-08-30 08:43:29.543409: Pseudo dice [0.8903, 0.9135, 0.8928, 0.8693, 0.854, 0.8603, 0.8552, 0.8442, 0.8344, 0.8207, 0.8267, 0.8474, 0.9029, 0.9041, 0.8716, 0.8192, 0.7569, 0.0, 0.9575, nan, 0.0] 
+2024-08-30 08:43:29.544298: Epoch time: 53.63 s 
+2024-08-30 08:43:30.901514:  
+2024-08-30 08:43:30.902413: Epoch 845 
+2024-08-30 08:43:30.903183: Current learning rate: 0.00187 
+2024-08-30 08:44:23.643970: train_loss -0.8304 
+2024-08-30 08:44:23.645729: val_loss -0.8566 
+2024-08-30 08:44:23.646530: Pseudo dice [0.9207, 0.9391, 0.8981, 0.8785, 0.902, 0.893, 0.891, 0.9151, 0.939, 0.9532, 0.9421, 0.921, 0.9327, 0.9647, 0.9618, 0.9327, 0.8771, 0.0, 0.936, nan, nan] 
+2024-08-30 08:44:23.647211: Epoch time: 52.75 s 
+2024-08-30 08:44:24.951501:  
+2024-08-30 08:44:24.952410: Epoch 846 
+2024-08-30 08:44:24.953138: Current learning rate: 0.00186 
+2024-08-30 08:45:15.575314: train_loss -0.8288 
+2024-08-30 08:45:15.577284: val_loss -0.8433 
+2024-08-30 08:45:15.578287: Pseudo dice [0.9362, 0.9328, 0.9073, 0.8303, 0.8414, 0.8708, 0.8822, 0.9292, 0.9562, 0.9539, 0.9443, 0.9305, 0.9291, 0.8906, 0.8432, 0.8253, 0.8535, 0.0, 0.9521, nan, nan] 
+2024-08-30 08:45:15.579273: Epoch time: 50.63 s 
+2024-08-30 08:45:17.240562:  
+2024-08-30 08:45:17.241265: Epoch 847 
+2024-08-30 08:45:17.241756: Current learning rate: 0.00185 
+2024-08-30 08:46:09.978026: train_loss -0.8326 
+2024-08-30 08:46:09.991204: val_loss -0.8519 
+2024-08-30 08:46:09.992387: Pseudo dice [0.926, 0.8973, 0.8754, 0.911, 0.9288, 0.9329, 0.9321, 0.926, 0.9219, 0.921, 0.948, 0.963, 0.9617, 0.9694, 0.963, 0.9507, 0.9403, 0.0, 0.9505, nan, nan] 
+2024-08-30 08:46:09.993431: Epoch time: 52.74 s 
+2024-08-30 08:46:11.367785:  
+2024-08-30 08:46:11.368587: Epoch 848 
+2024-08-30 08:46:11.369292: Current learning rate: 0.00184 
+2024-08-30 08:47:03.955757: train_loss -0.838 
+2024-08-30 08:47:03.956916: val_loss -0.8315 
+2024-08-30 08:47:03.957751: Pseudo dice [0.9125, 0.9179, 0.8678, 0.8423, 0.8697, 0.8744, 0.8542, 0.8571, 0.8618, 0.8344, 0.8276, 0.8869, 0.9626, 0.9507, 0.936, 0.9326, 0.9405, 0.0, 0.9413, nan, 0.0] 
+2024-08-30 08:47:03.958437: Epoch time: 52.59 s 
+2024-08-30 08:47:05.276408:  
+2024-08-30 08:47:05.277321: Epoch 849 
+2024-08-30 08:47:05.278058: Current learning rate: 0.00182 
+2024-08-30 08:47:59.892816: train_loss -0.8387 
+2024-08-30 08:47:59.895495: val_loss -0.8549 
+2024-08-30 08:47:59.896668: Pseudo dice [0.9251, 0.9427, 0.9384, 0.9354, 0.9379, 0.9505, 0.959, 0.9614, 0.9629, 0.958, 0.9555, 0.9562, 0.9562, 0.936, 0.8962, 0.8824, 0.8624, 0.0, 0.9507, nan, nan] 
+2024-08-30 08:47:59.897888: Epoch time: 54.62 s 
+2024-08-30 08:48:03.877543:  
+2024-08-30 08:48:03.878408: Epoch 850 
+2024-08-30 08:48:03.879123: Current learning rate: 0.00181 
+2024-08-30 08:48:55.590902: train_loss -0.8367 
+2024-08-30 08:48:55.592170: val_loss -0.8651 
+2024-08-30 08:48:55.593011: Pseudo dice [0.9056, 0.927, 0.9329, 0.9429, 0.9531, 0.9613, 0.9652, 0.9692, 0.9625, 0.9509, 0.9524, 0.9603, 0.9667, 0.9757, 0.9683, 0.957, 0.9337, nan, 0.9366, nan, nan] 
+2024-08-30 08:48:55.593708: Epoch time: 51.72 s 
+2024-08-30 08:48:57.592999:  
+2024-08-30 08:48:57.593977: Epoch 851 
+2024-08-30 08:48:57.594691: Current learning rate: 0.0018 
+2024-08-30 08:49:48.381647: train_loss -0.8374 
+2024-08-30 08:49:48.383764: val_loss -0.8453 
+2024-08-30 08:49:48.384715: Pseudo dice [0.895, 0.8924, 0.8888, 0.8955, 0.8853, 0.8824, 0.9294, 0.9467, 0.9421, 0.9434, 0.9375, 0.9262, 0.9169, 0.9239, 0.9239, 0.929, 0.9112, 0.0, 0.9424, nan, nan] 
+2024-08-30 08:49:48.385490: Epoch time: 50.79 s 
+2024-08-30 08:49:49.953361:  
+2024-08-30 08:49:49.954305: Epoch 852 
+2024-08-30 08:49:49.955045: Current learning rate: 0.00179 
+2024-08-30 08:50:41.454666: train_loss -0.8347 
+2024-08-30 08:50:41.456775: val_loss -0.8491 
+2024-08-30 08:50:41.458367: Pseudo dice [0.9167, 0.9272, 0.8905, 0.9042, 0.9136, 0.9085, 0.9005, 0.9117, 0.9186, 0.9246, 0.9365, 0.9549, 0.9551, 0.9315, 0.9235, 0.904, 0.885, 0.0, 0.9512, nan, 0.0] 
+2024-08-30 08:50:41.459208: Epoch time: 51.5 s 
+2024-08-30 08:50:42.787482:  
+2024-08-30 08:50:42.788740: Epoch 853 
+2024-08-30 08:50:42.789418: Current learning rate: 0.00178 
+2024-08-30 08:51:35.049179: train_loss -0.8385 
+2024-08-30 08:51:35.051316: val_loss -0.8338 
+2024-08-30 08:51:35.052195: Pseudo dice [0.9174, 0.9189, 0.9197, 0.9246, 0.9394, 0.941, 0.9031, 0.8926, 0.8781, 0.8344, 0.7994, 0.8644, 0.9192, 0.944, 0.9667, 0.9473, 0.8746, 0.0, 0.9445, nan, nan] 
+2024-08-30 08:51:35.053322: Epoch time: 52.26 s 
+2024-08-30 08:51:36.377242:  
+2024-08-30 08:51:36.378135: Epoch 854 
+2024-08-30 08:51:36.378981: Current learning rate: 0.00177 
+2024-08-30 08:52:28.232911: train_loss -0.8387 
+2024-08-30 08:52:28.234886: val_loss -0.8301 
+2024-08-30 08:52:28.235937: Pseudo dice [0.9305, 0.9367, 0.9232, 0.9403, 0.9553, 0.9441, 0.9246, 0.855, 0.8164, 0.826, 0.8148, 0.8342, 0.9149, 0.91, 0.9103, 0.9004, 0.9058, 0.0, 0.9433, nan, 0.0] 
+2024-08-30 08:52:28.236702: Epoch time: 51.86 s 
+2024-08-30 08:52:29.565022:  
+2024-08-30 08:52:29.565928: Epoch 855 
+2024-08-30 08:52:29.566649: Current learning rate: 0.00176 
+2024-08-30 08:53:22.500609: train_loss -0.8342 
+2024-08-30 08:53:22.502524: val_loss -0.856 
+2024-08-30 08:53:22.503296: Pseudo dice [0.8973, 0.9256, 0.9276, 0.9283, 0.9204, 0.9272, 0.9373, 0.9462, 0.9605, 0.9675, 0.9693, 0.9664, 0.9648, 0.9482, 0.9414, 0.9262, 0.8635, 0.0, 0.9429, nan, nan] 
+2024-08-30 08:53:22.504044: Epoch time: 52.94 s 
+2024-08-30 08:53:23.820922:  
+2024-08-30 08:53:23.821906: Epoch 856 
+2024-08-30 08:53:23.822630: Current learning rate: 0.00175 
+2024-08-30 08:54:16.268810: train_loss -0.8411 
+2024-08-30 08:54:16.270410: val_loss -0.8504 
+2024-08-30 08:54:16.271315: Pseudo dice [0.9122, 0.9143, 0.9029, 0.8923, 0.8937, 0.9052, 0.9128, 0.9075, 0.9167, 0.9291, 0.9404, 0.9495, 0.9684, 0.9705, 0.9591, 0.9309, 0.9091, 0.0, 0.9387, nan, 0.0] 
+2024-08-30 08:54:16.272061: Epoch time: 52.45 s 
+2024-08-30 08:54:17.691141:  
+2024-08-30 08:54:17.692030: Epoch 857 
+2024-08-30 08:54:17.692727: Current learning rate: 0.00174 
+2024-08-30 08:55:09.571531: train_loss -0.832 
+2024-08-30 08:55:09.573950: val_loss -0.849 
+2024-08-30 08:55:09.575038: Pseudo dice [0.9242, 0.9466, 0.9406, 0.9254, 0.9234, 0.9233, 0.9262, 0.9339, 0.9336, 0.9337, 0.9398, 0.9177, 0.9436, 0.9119, 0.8793, 0.833, 0.7875, nan, 0.9264, nan, 0.0] 
+2024-08-30 08:55:09.575829: Epoch time: 51.88 s 
+2024-08-30 08:55:10.855058:  
+2024-08-30 08:55:10.855939: Epoch 858 
+2024-08-30 08:55:10.856600: Current learning rate: 0.00173 
+2024-08-30 08:55:59.863614: train_loss -0.831 
+2024-08-30 08:55:59.865005: val_loss -0.8535 
+2024-08-30 08:55:59.867006: Pseudo dice [0.9404, 0.9531, 0.9496, 0.9549, 0.9598, 0.9671, 0.9671, 0.9523, 0.9575, 0.958, 0.9564, 0.933, 0.9251, 0.9185, 0.92, 0.9268, 0.8361, 0.0, 0.9322, nan, nan] 
+2024-08-30 08:55:59.867766: Epoch time: 49.01 s 
+2024-08-30 08:56:01.408063:  
+2024-08-30 08:56:01.409021: Epoch 859 
+2024-08-30 08:56:01.409722: Current learning rate: 0.00172 
+2024-08-30 08:56:51.798097: train_loss -0.843 
+2024-08-30 08:56:51.800311: val_loss -0.8369 
+2024-08-30 08:56:51.801406: Pseudo dice [0.9064, 0.9308, 0.9259, 0.9113, 0.888, 0.8785, 0.8644, 0.8599, 0.8873, 0.9033, 0.918, 0.9246, 0.9489, 0.9225, 0.9154, 0.9183, 0.8533, 0.0, 0.9459, nan, 0.0] 
+2024-08-30 08:56:51.802177: Epoch time: 50.39 s 
+2024-08-30 08:56:53.236082:  
+2024-08-30 08:56:53.237377: Epoch 860 
+2024-08-30 08:56:53.238142: Current learning rate: 0.0017 
+2024-08-30 08:57:48.362655: train_loss -0.8373 
+2024-08-30 08:57:48.363925: val_loss -0.8358 
+2024-08-30 08:57:48.364877: Pseudo dice [0.9434, 0.9557, 0.9561, 0.9474, 0.9004, 0.8748, 0.8762, 0.8653, 0.846, 0.8354, 0.8613, 0.91, 0.9531, 0.9548, 0.9597, 0.9397, 0.8697, 0.0, 0.9262, nan, nan] 
+2024-08-30 08:57:48.365652: Epoch time: 55.13 s 
+2024-08-30 08:57:49.672308:  
+2024-08-30 08:57:49.673516: Epoch 861 
+2024-08-30 08:57:49.674527: Current learning rate: 0.00169 
+2024-08-30 08:58:45.218183: train_loss -0.8346 
+2024-08-30 08:58:45.220060: val_loss -0.8465 
+2024-08-30 08:58:45.220916: Pseudo dice [0.9053, 0.9293, 0.9208, 0.9106, 0.9059, 0.9117, 0.9204, 0.9207, 0.9181, 0.9415, 0.9398, 0.938, 0.9492, 0.9557, 0.9598, 0.9258, 0.8812, 0.0, 0.9454, nan, nan] 
+2024-08-30 08:58:45.221845: Epoch time: 55.55 s 
+2024-08-30 08:58:46.646538:  
+2024-08-30 08:58:46.647485: Epoch 862 
+2024-08-30 08:58:46.648240: Current learning rate: 0.00168 
+2024-08-30 08:59:40.994913: train_loss -0.8318 
+2024-08-30 08:59:40.996721: val_loss -0.851 
+2024-08-30 08:59:40.997763: Pseudo dice [0.9387, 0.9371, 0.9391, 0.9518, 0.9575, 0.9651, 0.9665, 0.9642, 0.9657, 0.9506, 0.946, 0.9581, 0.9337, 0.8996, 0.8819, 0.8758, 0.8834, 0.0, 0.9446, nan, nan] 
+2024-08-30 08:59:40.999017: Epoch time: 54.35 s 
+2024-08-30 08:59:42.270347:  
+2024-08-30 08:59:42.270894: Epoch 863 
+2024-08-30 08:59:42.271382: Current learning rate: 0.00167 
+2024-08-30 09:00:33.523203: train_loss -0.8386 
+2024-08-30 09:00:33.525758: val_loss -0.8467 
+2024-08-30 09:00:33.527037: Pseudo dice [0.9169, 0.9109, 0.8887, 0.9015, 0.9131, 0.9238, 0.9044, 0.9005, 0.8851, 0.8783, 0.9046, 0.9107, 0.9397, 0.9291, 0.9289, 0.9144, 0.8798, 0.0, 0.9498, nan, 0.0] 
+2024-08-30 09:00:33.527949: Epoch time: 51.26 s 
+2024-08-30 09:00:35.809411:  
+2024-08-30 09:00:35.810287: Epoch 864 
+2024-08-30 09:00:35.811031: Current learning rate: 0.00166 
+2024-08-30 09:01:28.219671: train_loss -0.838 
+2024-08-30 09:01:28.221019: val_loss -0.8554 
+2024-08-30 09:01:28.221902: Pseudo dice [0.9124, 0.934, 0.9384, 0.9115, 0.9199, 0.9542, 0.9577, 0.9646, 0.9678, 0.9709, 0.9679, 0.9698, 0.9702, 0.9411, 0.9, 0.8693, 0.8403, 0.0, 0.9338, nan, nan] 
+2024-08-30 09:01:28.222677: Epoch time: 52.41 s 
+2024-08-30 09:01:29.836327:  
+2024-08-30 09:01:29.837203: Epoch 865 
+2024-08-30 09:01:29.837966: Current learning rate: 0.00165 
+2024-08-30 09:02:22.851339: train_loss -0.8418 
+2024-08-30 09:02:22.853962: val_loss -0.8571 
+2024-08-30 09:02:22.854879: Pseudo dice [0.942, 0.9536, 0.952, 0.9326, 0.9033, 0.8973, 0.9031, 0.9102, 0.9125, 0.9062, 0.8857, 0.9213, 0.9625, 0.9712, 0.9754, 0.9669, 0.9257, 0.0, 0.9539, nan, nan] 
+2024-08-30 09:02:22.855863: Epoch time: 53.02 s 
+2024-08-30 09:02:24.152373:  
+2024-08-30 09:02:24.153325: Epoch 866 
+2024-08-30 09:02:24.154001: Current learning rate: 0.00164 
+2024-08-30 09:03:16.440200: train_loss -0.8464 
+2024-08-30 09:03:16.441539: val_loss -0.8543 
+2024-08-30 09:03:16.442665: Pseudo dice [0.9283, 0.9498, 0.9386, 0.9328, 0.953, 0.9616, 0.9685, 0.9518, 0.9429, 0.9621, 0.9631, 0.9694, 0.9678, 0.9614, 0.9301, 0.8632, 0.8204, 0.0, 0.9587, nan, nan] 
+2024-08-30 09:03:16.443727: Epoch time: 52.29 s 
+2024-08-30 09:03:17.824345:  
+2024-08-30 09:03:17.825303: Epoch 867 
+2024-08-30 09:03:17.825968: Current learning rate: 0.00163 
+2024-08-30 09:04:08.592646: train_loss -0.8416 
+2024-08-30 09:04:08.595639: val_loss -0.8524 
+2024-08-30 09:04:08.596529: Pseudo dice [0.9298, 0.9471, 0.9465, 0.9608, 0.9608, 0.9581, 0.9576, 0.9526, 0.9453, 0.9294, 0.9132, 0.9362, 0.9493, 0.9362, 0.9075, 0.8671, 0.7967, 0.0, 0.9418, nan, nan] 
+2024-08-30 09:04:08.597581: Epoch time: 50.77 s 
+2024-08-30 09:04:09.935222:  
+2024-08-30 09:04:09.936138: Epoch 868 
+2024-08-30 09:04:09.936894: Current learning rate: 0.00162 
+2024-08-30 09:05:02.175758: train_loss -0.8382 
+2024-08-30 09:05:02.180358: val_loss -0.8533 
+2024-08-30 09:05:02.181579: Pseudo dice [0.9343, 0.9478, 0.9531, 0.9568, 0.9592, 0.9551, 0.9494, 0.933, 0.9293, 0.9281, 0.9349, 0.9247, 0.9568, 0.9299, 0.9029, 0.9175, 0.8949, 0.0, 0.9369, nan, 0.0] 
+2024-08-30 09:05:02.182430: Epoch time: 52.24 s 
+2024-08-30 09:05:03.815528:  
+2024-08-30 09:05:03.817090: Epoch 869 
+2024-08-30 09:05:03.817828: Current learning rate: 0.00161 
+2024-08-30 09:05:57.417204: train_loss -0.8328 
+2024-08-30 09:05:57.419668: val_loss -0.8604 
+2024-08-30 09:05:57.420626: Pseudo dice [0.9082, 0.949, 0.9422, 0.9247, 0.9396, 0.9548, 0.9273, 0.9226, 0.9415, 0.9536, 0.9577, 0.9531, 0.966, 0.9682, 0.9672, 0.9542, 0.9222, 0.0, 0.9538, nan, nan] 
+2024-08-30 09:05:57.421386: Epoch time: 53.6 s 
+2024-08-30 09:05:58.851858:  
+2024-08-30 09:05:58.852747: Epoch 870 
+2024-08-30 09:05:58.853485: Current learning rate: 0.00159 
+2024-08-30 09:06:52.801990: train_loss -0.8372 
+2024-08-30 09:06:52.803658: val_loss -0.8536 
+2024-08-30 09:06:52.804487: Pseudo dice [0.9369, 0.9397, 0.9356, 0.9546, 0.9619, 0.9644, 0.9407, 0.9462, 0.9642, 0.9676, 0.9406, 0.9255, 0.8748, 0.8859, 0.9038, 0.8989, 0.903, 0.0, 0.9541, nan, 0.0] 
+2024-08-30 09:06:52.805259: Epoch time: 53.95 s 
+2024-08-30 09:06:54.142854:  
+2024-08-30 09:06:54.143848: Epoch 871 
+2024-08-30 09:06:54.144514: Current learning rate: 0.00158 
+2024-08-30 09:07:48.394995: train_loss -0.8409 
+2024-08-30 09:07:48.397168: val_loss -0.8576 
+2024-08-30 09:07:48.398114: Pseudo dice [0.9372, 0.9585, 0.9488, 0.9537, 0.9626, 0.9471, 0.9049, 0.9016, 0.9215, 0.9314, 0.9359, 0.9476, 0.9646, 0.9705, 0.9729, 0.962, 0.9385, 0.0, 0.9599, nan, nan] 
+2024-08-30 09:07:48.398971: Epoch time: 54.26 s 
+2024-08-30 09:07:49.755149:  
+2024-08-30 09:07:49.756004: Epoch 872 
+2024-08-30 09:07:49.756696: Current learning rate: 0.00157 
+2024-08-30 09:08:43.106714: train_loss -0.8371 
+2024-08-30 09:08:43.107915: val_loss -0.8562 
+2024-08-30 09:08:43.108848: Pseudo dice [0.9368, 0.9445, 0.9531, 0.9557, 0.9561, 0.962, 0.9412, 0.9402, 0.953, 0.9583, 0.9591, 0.9518, 0.9686, 0.9571, 0.9379, 0.8977, 0.8283, 0.0, 0.9554, nan, 0.0] 
+2024-08-30 09:08:43.109610: Epoch time: 53.35 s 
+2024-08-30 09:08:44.639570:  
+2024-08-30 09:08:44.640445: Epoch 873 
+2024-08-30 09:08:44.641129: Current learning rate: 0.00156 
+2024-08-30 09:09:35.810926: train_loss -0.843 
+2024-08-30 09:09:35.813389: val_loss -0.8493 
+2024-08-30 09:09:35.814319: Pseudo dice [0.9132, 0.9213, 0.9236, 0.9295, 0.9249, 0.9122, 0.8885, 0.8997, 0.8974, 0.9023, 0.9263, 0.9364, 0.9549, 0.9696, 0.9662, 0.9468, 0.8956, 0.0, 0.9346, nan, nan] 
+2024-08-30 09:09:35.815161: Epoch time: 51.17 s 
+2024-08-30 09:09:37.314640:  
+2024-08-30 09:09:37.315544: Epoch 874 
+2024-08-30 09:09:37.316256: Current learning rate: 0.00155 
+2024-08-30 09:10:31.548849: train_loss -0.8447 
+2024-08-30 09:10:31.549944: val_loss -0.8468 
+2024-08-30 09:10:31.550888: Pseudo dice [0.9228, 0.9285, 0.9166, 0.931, 0.9581, 0.9519, 0.9376, 0.9329, 0.9187, 0.9142, 0.9283, 0.9442, 0.923, 0.8854, 0.851, 0.8311, 0.7589, 0.0, 0.9417, nan, nan] 
+2024-08-30 09:10:31.551782: Epoch time: 54.24 s 
+2024-08-30 09:10:32.872042:  
+2024-08-30 09:10:32.872933: Epoch 875 
+2024-08-30 09:10:32.873559: Current learning rate: 0.00154 
+2024-08-30 09:11:25.913647: train_loss -0.8416 
+2024-08-30 09:11:25.916418: val_loss -0.8512 
+2024-08-30 09:11:25.917590: Pseudo dice [0.9287, 0.9457, 0.9462, 0.923, 0.8986, 0.9299, 0.9323, 0.9177, 0.9126, 0.9143, 0.9129, 0.9478, 0.9649, 0.9579, 0.9596, 0.947, 0.865, 0.0, 0.9397, nan, nan] 
+2024-08-30 09:11:25.918908: Epoch time: 53.04 s 
+2024-08-30 09:11:27.242563:  
+2024-08-30 09:11:27.243446: Epoch 876 
+2024-08-30 09:11:27.244117: Current learning rate: 0.00153 
+2024-08-30 09:12:16.963555: train_loss -0.8312 
+2024-08-30 09:12:16.965032: val_loss -0.8509 
+2024-08-30 09:12:16.966198: Pseudo dice [0.9027, 0.9104, 0.8725, 0.8879, 0.9106, 0.9152, 0.9027, 0.883, 0.8969, 0.9002, 0.9077, 0.9196, 0.9159, 0.9291, 0.9282, 0.947, 0.8979, nan, 0.9356, nan, nan] 
+2024-08-30 09:12:16.967159: Epoch time: 49.72 s 
+2024-08-30 09:12:18.284918:  
+2024-08-30 09:12:18.285751: Epoch 877 
+2024-08-30 09:12:18.286430: Current learning rate: 0.00152 
+2024-08-30 09:13:09.038544: train_loss -0.8394 
+2024-08-30 09:13:09.040658: val_loss -0.8522 
+2024-08-30 09:13:09.041566: Pseudo dice [0.9443, 0.9627, 0.9609, 0.9649, 0.9532, 0.9516, 0.9521, 0.937, 0.9204, 0.9069, 0.9125, 0.923, 0.9625, 0.9268, 0.939, 0.9383, 0.8919, 0.0, 0.9295, nan, 0.0] 
+2024-08-30 09:13:09.042206: Epoch time: 50.76 s 
+2024-08-30 09:13:11.051424:  
+2024-08-30 09:13:11.052295: Epoch 878 
+2024-08-30 09:13:11.052982: Current learning rate: 0.00151 
+2024-08-30 09:14:01.156067: train_loss -0.8369 
+2024-08-30 09:14:01.157804: val_loss -0.8611 
+2024-08-30 09:14:01.158995: Pseudo dice [0.9301, 0.9615, 0.9623, 0.9619, 0.9621, 0.9645, 0.965, 0.9523, 0.9359, 0.9281, 0.9196, 0.9195, 0.9647, 0.967, 0.9233, 0.8982, 0.9088, nan, 0.9421, nan, nan] 
+2024-08-30 09:14:01.159832: Epoch time: 50.11 s 
+2024-08-30 09:14:02.494009:  
+2024-08-30 09:14:02.494841: Epoch 879 
+2024-08-30 09:14:02.495541: Current learning rate: 0.00149 
+2024-08-30 09:14:53.631948: train_loss -0.836 
+2024-08-30 09:14:53.633824: val_loss -0.8332 
+2024-08-30 09:14:53.634913: Pseudo dice [0.9079, 0.876, 0.8622, 0.9014, 0.9182, 0.9402, 0.9176, 0.8935, 0.8962, 0.9322, 0.9457, 0.9458, 0.9375, 0.8962, 0.8779, 0.8574, 0.8324, 0.0, 0.9115, nan, nan] 
+2024-08-30 09:14:53.636594: Epoch time: 51.14 s 
+2024-08-30 09:14:55.305416:  
+2024-08-30 09:14:55.306445: Epoch 880 
+2024-08-30 09:14:55.307124: Current learning rate: 0.00148 
+2024-08-30 09:15:48.906780: train_loss -0.8415 
+2024-08-30 09:15:48.908741: val_loss -0.8419 
+2024-08-30 09:15:48.909892: Pseudo dice [0.9333, 0.9413, 0.946, 0.9542, 0.9622, 0.9293, 0.8882, 0.8673, 0.8643, 0.867, 0.8732, 0.8859, 0.9404, 0.9425, 0.8989, 0.8451, 0.7916, 0.0, 0.9407, nan, 0.0] 
+2024-08-30 09:15:48.910892: Epoch time: 53.6 s 
+2024-08-30 09:15:50.297559:  
+2024-08-30 09:15:50.298381: Epoch 881 
+2024-08-30 09:15:50.299062: Current learning rate: 0.00147 
+2024-08-30 09:16:42.509722: train_loss -0.8383 
+2024-08-30 09:16:42.511638: val_loss -0.8445 
+2024-08-30 09:16:42.512475: Pseudo dice [0.906, 0.94, 0.9296, 0.9233, 0.9014, 0.8822, 0.889, 0.8779, 0.8718, 0.8774, 0.8619, 0.8644, 0.949, 0.9536, 0.9521, 0.9286, 0.8987, 0.0, 0.9398, nan, 0.0] 
+2024-08-30 09:16:42.513572: Epoch time: 52.22 s 
+2024-08-30 09:16:43.979096:  
+2024-08-30 09:16:43.979967: Epoch 882 
+2024-08-30 09:16:43.980589: Current learning rate: 0.00146 
+2024-08-30 09:17:36.578169: train_loss -0.8375 
+2024-08-30 09:17:36.579384: val_loss -0.8352 
+2024-08-30 09:17:36.580246: Pseudo dice [0.9135, 0.9247, 0.9195, 0.9266, 0.933, 0.9166, 0.9008, 0.9009, 0.9012, 0.8523, 0.8407, 0.8942, 0.8531, 0.7992, 0.819, 0.8354, 0.8685, 0.0, 0.9542, nan, nan] 
+2024-08-30 09:17:36.581056: Epoch time: 52.6 s 
+2024-08-30 09:17:38.009704:  
+2024-08-30 09:17:38.010733: Epoch 883 
+2024-08-30 09:17:38.011462: Current learning rate: 0.00145 
+2024-08-30 09:18:28.543527: train_loss -0.8386 
+2024-08-30 09:18:28.545922: val_loss -0.8393 
+2024-08-30 09:18:28.546789: Pseudo dice [0.9315, 0.9584, 0.9628, 0.9606, 0.9533, 0.9611, 0.9706, 0.9649, 0.9607, 0.966, 0.9659, 0.9589, 0.9582, 0.9026, 0.8605, 0.7682, 0.7065, 0.0, 0.9238, nan, 0.0] 
+2024-08-30 09:18:28.547521: Epoch time: 50.54 s 
+2024-08-30 09:18:29.982323:  
+2024-08-30 09:18:29.983205: Epoch 884 
+2024-08-30 09:18:29.983942: Current learning rate: 0.00144 
+2024-08-30 09:19:22.844262: train_loss -0.8407 
+2024-08-30 09:19:22.845461: val_loss -0.8468 
+2024-08-30 09:19:22.846434: Pseudo dice [0.8979, 0.9238, 0.9197, 0.932, 0.9348, 0.9316, 0.9168, 0.9015, 0.9097, 0.9157, 0.8999, 0.8794, 0.9356, 0.9102, 0.8713, 0.8698, 0.9059, nan, 0.95, nan, 0.0] 
+2024-08-30 09:19:22.847541: Epoch time: 52.86 s 
+2024-08-30 09:19:24.125946:  
+2024-08-30 09:19:24.126772: Epoch 885 
+2024-08-30 09:19:24.127533: Current learning rate: 0.00143 
+2024-08-30 09:20:17.535192: train_loss -0.8435 
+2024-08-30 09:20:17.538976: val_loss -0.8716 
+2024-08-30 09:20:17.539990: Pseudo dice [0.9421, 0.9584, 0.9499, 0.9534, 0.9657, 0.9699, 0.9641, 0.9614, 0.9628, 0.9674, 0.9511, 0.9531, 0.9645, 0.961, 0.9554, 0.9447, 0.8955, nan, 0.9404, nan, nan] 
+2024-08-30 09:20:17.540793: Epoch time: 53.41 s 
+2024-08-30 09:20:19.032748:  
+2024-08-30 09:20:19.033603: Epoch 886 
+2024-08-30 09:20:19.034249: Current learning rate: 0.00142 
+2024-08-30 09:21:10.380071: train_loss -0.8451 
+2024-08-30 09:21:10.391248: val_loss -0.8491 
+2024-08-30 09:21:10.392237: Pseudo dice [0.923, 0.9434, 0.9261, 0.9186, 0.9428, 0.9434, 0.9265, 0.89, 0.8506, 0.8471, 0.899, 0.9226, 0.9481, 0.9505, 0.9462, 0.9108, 0.8979, 0.0, 0.9532, nan, 0.0] 
+2024-08-30 09:21:10.392935: Epoch time: 51.35 s 
+2024-08-30 09:21:11.701781:  
+2024-08-30 09:21:11.703192: Epoch 887 
+2024-08-30 09:21:11.703910: Current learning rate: 0.00141 
+2024-08-30 09:22:04.421584: train_loss -0.8447 
+2024-08-30 09:22:04.423617: val_loss -0.8645 
+2024-08-30 09:22:04.424931: Pseudo dice [0.9246, 0.9471, 0.9482, 0.954, 0.9557, 0.9558, 0.9515, 0.9305, 0.9371, 0.9582, 0.9573, 0.961, 0.9651, 0.9669, 0.9637, 0.9546, 0.9001, 0.0, 0.9515, nan, nan] 
+2024-08-30 09:22:04.425783: Epoch time: 52.72 s 
+2024-08-30 09:22:05.792814:  
+2024-08-30 09:22:05.793650: Epoch 888 
+2024-08-30 09:22:05.794380: Current learning rate: 0.00139 
+2024-08-30 09:22:54.214429: train_loss -0.8439 
+2024-08-30 09:22:54.215655: val_loss -0.8541 
+2024-08-30 09:22:54.216516: Pseudo dice [0.9066, 0.9146, 0.9315, 0.9335, 0.9391, 0.9587, 0.9621, 0.9562, 0.9455, 0.9305, 0.929, 0.9336, 0.9524, 0.9605, 0.9197, 0.9003, 0.8764, 0.0, 0.9465, nan, 0.0] 
+2024-08-30 09:22:54.217762: Epoch time: 48.42 s 
+2024-08-30 09:22:55.815449:  
+2024-08-30 09:22:55.816295: Epoch 889 
+2024-08-30 09:22:55.816990: Current learning rate: 0.00138 
+2024-08-30 09:23:47.761132: train_loss -0.8433 
+2024-08-30 09:23:47.763311: val_loss -0.8429 
+2024-08-30 09:23:47.764520: Pseudo dice [0.9395, 0.9501, 0.9441, 0.955, 0.9434, 0.9245, 0.8983, 0.8933, 0.9066, 0.9214, 0.9116, 0.8908, 0.9325, 0.9154, 0.9253, 0.9247, 0.8588, 0.0, 0.9472, nan, 0.0] 
+2024-08-30 09:23:47.765725: Epoch time: 51.95 s 
+2024-08-30 09:23:49.115229:  
+2024-08-30 09:23:49.116137: Epoch 890 
+2024-08-30 09:23:49.116842: Current learning rate: 0.00137 
+2024-08-30 09:24:41.991803: train_loss -0.8406 
+2024-08-30 09:24:41.993486: val_loss -0.8592 
+2024-08-30 09:24:41.994666: Pseudo dice [0.9254, 0.9517, 0.9499, 0.9592, 0.9642, 0.9661, 0.9649, 0.9717, 0.9627, 0.9479, 0.9179, 0.9195, 0.9294, 0.9121, 0.9151, 0.9349, 0.9356, 0.0, 0.9414, nan, 0.0] 
+2024-08-30 09:24:41.995376: Epoch time: 52.88 s 
+2024-08-30 09:24:44.173131:  
+2024-08-30 09:24:44.174046: Epoch 891 
+2024-08-30 09:24:44.174711: Current learning rate: 0.00136 
+2024-08-30 09:25:37.189787: train_loss -0.8482 
+2024-08-30 09:25:37.191763: val_loss -0.8563 
+2024-08-30 09:25:37.192794: Pseudo dice [0.9182, 0.9365, 0.9428, 0.9554, 0.9561, 0.9469, 0.8903, 0.8663, 0.9116, 0.9309, 0.9478, 0.9225, 0.9344, 0.961, 0.9478, 0.9201, 0.8967, 0.0, 0.9472, nan, 0.0] 
+2024-08-30 09:25:37.193596: Epoch time: 53.02 s 
+2024-08-30 09:25:38.676145:  
+2024-08-30 09:25:38.677552: Epoch 892 
+2024-08-30 09:25:38.678295: Current learning rate: 0.00135 
+2024-08-30 09:26:31.698332: train_loss -0.8475 
+2024-08-30 09:26:31.699583: val_loss -0.8362 
+2024-08-30 09:26:31.700533: Pseudo dice [0.8954, 0.8561, 0.8009, 0.8196, 0.8477, 0.8508, 0.8414, 0.8371, 0.8449, 0.8873, 0.924, 0.9297, 0.9266, 0.8997, 0.9062, 0.9172, 0.9063, 0.0, 0.935, nan, 0.0] 
+2024-08-30 09:26:31.701498: Epoch time: 53.03 s 
+2024-08-30 09:26:33.110759:  
+2024-08-30 09:26:33.111785: Epoch 893 
+2024-08-30 09:26:33.112508: Current learning rate: 0.00134 
+2024-08-30 09:27:26.704813: train_loss -0.8476 
+2024-08-30 09:27:26.706740: val_loss -0.8651 
+2024-08-30 09:27:26.707685: Pseudo dice [0.9512, 0.9608, 0.9606, 0.9644, 0.9625, 0.9649, 0.9582, 0.9636, 0.9649, 0.9654, 0.9666, 0.9673, 0.9705, 0.9697, 0.9489, 0.9124, 0.8684, nan, 0.9334, nan, nan] 
+2024-08-30 09:27:26.708473: Epoch time: 53.6 s 
+2024-08-30 09:27:28.148740:  
+2024-08-30 09:27:28.149570: Epoch 894 
+2024-08-30 09:27:28.150343: Current learning rate: 0.00133 
+2024-08-30 09:28:20.282127: train_loss -0.8424 
+2024-08-30 09:28:20.283353: val_loss -0.8536 
+2024-08-30 09:28:20.284453: Pseudo dice [0.9245, 0.9397, 0.9433, 0.9622, 0.9543, 0.9532, 0.9441, 0.9087, 0.8798, 0.901, 0.9302, 0.9616, 0.9584, 0.9219, 0.8654, 0.8241, 0.8262, 0.0, 0.9555, nan, 0.0] 
+2024-08-30 09:28:20.285292: Epoch time: 52.14 s 
+2024-08-30 09:28:21.591115:  
+2024-08-30 09:28:21.592039: Epoch 895 
+2024-08-30 09:28:21.592731: Current learning rate: 0.00132 
+2024-08-30 09:29:13.610711: train_loss -0.8371 
+2024-08-30 09:29:13.613045: val_loss -0.8447 
+2024-08-30 09:29:13.613831: Pseudo dice [0.9361, 0.9286, 0.913, 0.9208, 0.9298, 0.9371, 0.926, 0.9355, 0.9389, 0.9503, 0.9576, 0.9415, 0.9358, 0.9062, 0.8308, 0.8366, 0.8701, 0.0, 0.9601, nan, 0.0] 
+2024-08-30 09:29:13.614555: Epoch time: 52.02 s 
+2024-08-30 09:29:15.072567:  
+2024-08-30 09:29:15.073473: Epoch 896 
+2024-08-30 09:29:15.074170: Current learning rate: 0.0013 
+2024-08-30 09:30:05.941520: train_loss -0.8454 
+2024-08-30 09:30:05.942658: val_loss -0.8582 
+2024-08-30 09:30:05.943528: Pseudo dice [0.9268, 0.9585, 0.964, 0.9661, 0.9691, 0.9724, 0.9673, 0.9629, 0.9707, 0.9727, 0.9701, 0.9685, 0.9605, 0.942, 0.9377, 0.9236, 0.8866, 0.0, 0.9449, nan, nan] 
+2024-08-30 09:30:05.944242: Epoch time: 50.87 s 
+2024-08-30 09:30:07.385942:  
+2024-08-30 09:30:07.386874: Epoch 897 
+2024-08-30 09:30:07.387588: Current learning rate: 0.00129 
+2024-08-30 09:31:00.393564: train_loss -0.8407 
+2024-08-30 09:31:00.395521: val_loss -0.865 
+2024-08-30 09:31:00.396342: Pseudo dice [0.9355, 0.9507, 0.9489, 0.9609, 0.9657, 0.9578, 0.9545, 0.9639, 0.9379, 0.9462, 0.9622, 0.9518, 0.9627, 0.9566, 0.9622, 0.9657, 0.9541, 0.0, 0.9479, nan, nan] 
+2024-08-30 09:31:00.397204: Epoch time: 53.01 s 
+2024-08-30 09:31:01.742066:  
+2024-08-30 09:31:01.742928: Epoch 898 
+2024-08-30 09:31:01.743632: Current learning rate: 0.00128 
+2024-08-30 09:31:53.428476: train_loss -0.8454 
+2024-08-30 09:31:53.429614: val_loss -0.8472 
+2024-08-30 09:31:53.430410: Pseudo dice [0.9033, 0.9079, 0.896, 0.902, 0.9158, 0.9104, 0.915, 0.9356, 0.9496, 0.961, 0.9447, 0.933, 0.957, 0.9114, 0.8735, 0.8528, 0.8598, 0.0, 0.9625, nan, 0.0] 
+2024-08-30 09:31:53.431237: Epoch time: 51.69 s 
+2024-08-30 09:31:54.702623:  
+2024-08-30 09:31:54.703510: Epoch 899 
+2024-08-30 09:31:54.704171: Current learning rate: 0.00127 
+2024-08-30 09:32:46.171492: train_loss -0.8466 
+2024-08-30 09:32:46.173654: val_loss -0.8601 
+2024-08-30 09:32:46.174922: Pseudo dice [0.9211, 0.9391, 0.9364, 0.9198, 0.9132, 0.9096, 0.9293, 0.9308, 0.9449, 0.9456, 0.9465, 0.9689, 0.9747, 0.9552, 0.9235, 0.9176, 0.9041, 0.0, 0.9545, nan, nan] 
+2024-08-30 09:32:46.175740: Epoch time: 51.47 s 
+2024-08-30 09:32:50.069995:  
+2024-08-30 09:32:50.070864: Epoch 900 
+2024-08-30 09:32:50.071557: Current learning rate: 0.00126 
+2024-08-30 09:33:43.996930: train_loss -0.8401 
+2024-08-30 09:33:43.998721: val_loss -0.8438 
+2024-08-30 09:33:43.999696: Pseudo dice [0.9336, 0.9499, 0.9461, 0.9209, 0.8947, 0.9152, 0.9317, 0.9441, 0.9511, 0.9566, 0.9508, 0.9434, 0.9473, 0.9065, 0.8691, 0.8217, 0.7558, 0.0, 0.9578, nan, nan] 
+2024-08-30 09:33:44.000448: Epoch time: 53.93 s 
+2024-08-30 09:33:45.373267:  
+2024-08-30 09:33:45.374262: Epoch 901 
+2024-08-30 09:33:45.375159: Current learning rate: 0.00125 
+2024-08-30 09:34:37.210720: train_loss -0.8382 
+2024-08-30 09:34:37.212937: val_loss -0.8283 
+2024-08-30 09:34:37.213904: Pseudo dice [0.9264, 0.9371, 0.9033, 0.8907, 0.889, 0.8903, 0.8915, 0.8855, 0.9019, 0.8993, 0.8945, 0.8742, 0.8479, 0.7991, 0.7934, 0.8049, 0.7821, 0.0, 0.9344, nan, 0.0] 
+2024-08-30 09:34:37.215112: Epoch time: 51.84 s 
+2024-08-30 09:34:38.528410:  
+2024-08-30 09:34:38.529321: Epoch 902 
+2024-08-30 09:34:38.530048: Current learning rate: 0.00124 
+2024-08-30 09:35:29.587760: train_loss -0.8428 
+2024-08-30 09:35:29.588944: val_loss -0.8584 
+2024-08-30 09:35:29.589844: Pseudo dice [0.9253, 0.9443, 0.9539, 0.9341, 0.9249, 0.9271, 0.9274, 0.9276, 0.9304, 0.9191, 0.9291, 0.9536, 0.9517, 0.9346, 0.9329, 0.9325, 0.9259, nan, 0.9478, nan, nan] 
+2024-08-30 09:35:29.590628: Epoch time: 51.06 s 
+2024-08-30 09:35:31.090321:  
+2024-08-30 09:35:31.091160: Epoch 903 
+2024-08-30 09:35:31.091836: Current learning rate: 0.00122 
+2024-08-30 09:36:24.885172: train_loss -0.8496 
+2024-08-30 09:36:24.887835: val_loss -0.8514 
+2024-08-30 09:36:24.888869: Pseudo dice [0.9331, 0.9464, 0.956, 0.964, 0.9385, 0.923, 0.9325, 0.9447, 0.9396, 0.9333, 0.9441, 0.9419, 0.9493, 0.954, 0.9463, 0.9194, 0.8672, 0.0, 0.9469, nan, 0.0] 
+2024-08-30 09:36:24.889669: Epoch time: 53.8 s 
+2024-08-30 09:36:26.996505:  
+2024-08-30 09:36:26.997940: Epoch 904 
+2024-08-30 09:36:26.998565: Current learning rate: 0.00121 
+2024-08-30 09:37:17.406534: train_loss -0.8477 
+2024-08-30 09:37:17.407805: val_loss -0.8589 
+2024-08-30 09:37:17.408911: Pseudo dice [0.9487, 0.956, 0.948, 0.9506, 0.9625, 0.968, 0.9658, 0.9577, 0.9534, 0.9472, 0.9472, 0.9547, 0.9637, 0.9661, 0.9411, 0.906, 0.8633, 0.0, 0.9433, nan, nan] 
+2024-08-30 09:37:17.409878: Epoch time: 50.41 s 
+2024-08-30 09:37:18.711897:  
+2024-08-30 09:37:18.712965: Epoch 905 
+2024-08-30 09:37:18.713766: Current learning rate: 0.0012 
+2024-08-30 09:38:09.880386: train_loss -0.8408 
+2024-08-30 09:38:09.882540: val_loss -0.8431 
+2024-08-30 09:38:09.883490: Pseudo dice [0.9321, 0.9468, 0.93, 0.9346, 0.9374, 0.9305, 0.8557, 0.8492, 0.9061, 0.9543, 0.9579, 0.9045, 0.8719, 0.8568, 0.8335, 0.862, 0.8597, 0.0, 0.9552, nan, nan] 
+2024-08-30 09:38:09.884215: Epoch time: 51.17 s 
+2024-08-30 09:38:11.401536:  
+2024-08-30 09:38:11.402362: Epoch 906 
+2024-08-30 09:38:11.403127: Current learning rate: 0.00119 
+2024-08-30 09:39:07.168600: train_loss -0.8398 
+2024-08-30 09:39:07.169694: val_loss -0.8607 
+2024-08-30 09:39:07.170516: Pseudo dice [0.946, 0.9336, 0.9198, 0.9418, 0.9606, 0.9684, 0.9698, 0.9646, 0.9532, 0.9202, 0.8896, 0.9159, 0.9412, 0.9115, 0.9136, 0.9261, 0.9213, nan, 0.9496, nan, nan] 
+2024-08-30 09:39:07.171190: Epoch time: 55.77 s 
+2024-08-30 09:39:08.707077:  
+2024-08-30 09:39:08.708077: Epoch 907 
+2024-08-30 09:39:08.708838: Current learning rate: 0.00118 
+2024-08-30 09:40:01.850905: train_loss -0.8436 
+2024-08-30 09:40:01.852901: val_loss -0.8652 
+2024-08-30 09:40:01.853704: Pseudo dice [0.9414, 0.9518, 0.9357, 0.9309, 0.9494, 0.9599, 0.9594, 0.9684, 0.972, 0.9718, 0.9647, 0.9489, 0.9495, 0.9579, 0.9734, 0.9602, 0.9444, nan, 0.9474, nan, nan] 
+2024-08-30 09:40:01.854403: Epoch time: 53.15 s 
+2024-08-30 09:40:01.855169: Yayy! New best EMA pseudo Dice: 0.8798 
+2024-08-30 09:40:05.744240:  
+2024-08-30 09:40:05.745144: Epoch 908 
+2024-08-30 09:40:05.745888: Current learning rate: 0.00117 
+2024-08-30 09:40:56.679379: train_loss -0.8408 
+2024-08-30 09:40:56.681270: val_loss -0.8655 
+2024-08-30 09:40:56.682125: Pseudo dice [0.9359, 0.9569, 0.952, 0.9585, 0.9448, 0.9418, 0.9537, 0.9588, 0.9417, 0.9218, 0.9166, 0.9346, 0.9431, 0.9404, 0.9281, 0.9238, 0.913, 0.0, 0.9597, nan, nan] 
+2024-08-30 09:40:56.682843: Epoch time: 50.94 s 
+2024-08-30 09:40:56.683533: Yayy! New best EMA pseudo Dice: 0.8809 
+2024-08-30 09:41:00.972686:  
+2024-08-30 09:41:00.973612: Epoch 909 
+2024-08-30 09:41:00.974352: Current learning rate: 0.00116 
+2024-08-30 09:41:52.229133: train_loss -0.8412 
+2024-08-30 09:41:52.231776: val_loss -0.8727 
+2024-08-30 09:41:52.232930: Pseudo dice [0.918, 0.9451, 0.9422, 0.9508, 0.9556, 0.964, 0.9644, 0.9671, 0.9645, 0.9655, 0.9627, 0.9657, 0.973, 0.97, 0.9637, 0.9673, 0.9297, nan, 0.9402, nan, nan] 
+2024-08-30 09:41:52.234144: Epoch time: 51.26 s 
+2024-08-30 09:41:52.235236: Yayy! New best EMA pseudo Dice: 0.8884 
+2024-08-30 09:41:56.418429:  
+2024-08-30 09:41:56.419302: Epoch 910 
+2024-08-30 09:41:56.420158: Current learning rate: 0.00115 
+2024-08-30 09:42:49.296465: train_loss -0.8396 
+2024-08-30 09:42:49.297770: val_loss -0.8364 
+2024-08-30 09:42:49.298769: Pseudo dice [0.9266, 0.943, 0.9451, 0.9304, 0.9353, 0.9334, 0.9443, 0.9358, 0.9083, 0.9158, 0.8974, 0.8976, 0.8958, 0.8884, 0.8805, 0.8676, 0.8506, 0.0, 0.9446, nan, nan] 
+2024-08-30 09:42:49.299514: Epoch time: 52.88 s 
+2024-08-30 09:42:50.844342:  
+2024-08-30 09:42:50.845278: Epoch 911 
+2024-08-30 09:42:50.845987: Current learning rate: 0.00113 
+2024-08-30 09:43:43.097344: train_loss -0.8408 
+2024-08-30 09:43:43.099696: val_loss -0.8599 
+2024-08-30 09:43:43.100677: Pseudo dice [0.9409, 0.9516, 0.9496, 0.9518, 0.9491, 0.9517, 0.9611, 0.9717, 0.9486, 0.9486, 0.9627, 0.9627, 0.9714, 0.9641, 0.9298, 0.8623, 0.8033, 0.0, 0.9354, nan, nan] 
+2024-08-30 09:43:43.101291: Epoch time: 52.26 s 
+2024-08-30 09:43:44.533529:  
+2024-08-30 09:43:44.534385: Epoch 912 
+2024-08-30 09:43:44.535038: Current learning rate: 0.00112 
+2024-08-30 09:44:34.996822: train_loss -0.8517 
+2024-08-30 09:44:34.998015: val_loss -0.8617 
+2024-08-30 09:44:34.998905: Pseudo dice [0.943, 0.9519, 0.9501, 0.9613, 0.9655, 0.9678, 0.9643, 0.9611, 0.9606, 0.9492, 0.942, 0.9565, 0.9475, 0.9256, 0.9272, 0.9134, 0.8621, 0.0, 0.9564, nan, nan] 
+2024-08-30 09:44:34.999667: Epoch time: 50.47 s 
+2024-08-30 09:44:36.408237:  
+2024-08-30 09:44:36.409230: Epoch 913 
+2024-08-30 09:44:36.409955: Current learning rate: 0.00111 
+2024-08-30 09:45:28.878945: train_loss -0.8486 
+2024-08-30 09:45:28.881683: val_loss -0.8403 
+2024-08-30 09:45:28.882527: Pseudo dice [0.9299, 0.9509, 0.9427, 0.9301, 0.9125, 0.8868, 0.8903, 0.8965, 0.8822, 0.8931, 0.9089, 0.9366, 0.9198, 0.9021, 0.9141, 0.9048, 0.8939, 0.0, 0.9509, nan, 0.0] 
+2024-08-30 09:45:28.883183: Epoch time: 52.47 s 
+2024-08-30 09:45:30.225083:  
+2024-08-30 09:45:30.225914: Epoch 914 
+2024-08-30 09:45:30.226596: Current learning rate: 0.0011 
+2024-08-30 09:46:21.714068: train_loss -0.8507 
+2024-08-30 09:46:21.715301: val_loss -0.8477 
+2024-08-30 09:46:21.716247: Pseudo dice [0.93, 0.9427, 0.9337, 0.9243, 0.9162, 0.9224, 0.924, 0.8984, 0.8946, 0.9114, 0.8878, 0.8954, 0.9646, 0.9548, 0.9262, 0.9193, 0.9156, 0.0, 0.9416, nan, 0.0] 
+2024-08-30 09:46:21.717107: Epoch time: 51.49 s 
+2024-08-30 09:46:23.155884:  
+2024-08-30 09:46:23.156791: Epoch 915 
+2024-08-30 09:46:23.157521: Current learning rate: 0.00109 
+2024-08-30 09:47:17.947239: train_loss -0.8474 
+2024-08-30 09:47:17.949703: val_loss -0.8591 
+2024-08-30 09:47:17.950529: Pseudo dice [0.9407, 0.9571, 0.9483, 0.947, 0.9568, 0.9628, 0.9556, 0.9411, 0.9132, 0.9306, 0.9651, 0.969, 0.9726, 0.9683, 0.9356, 0.8999, 0.8389, 0.0, 0.9438, nan, nan] 
+2024-08-30 09:47:17.951321: Epoch time: 54.79 s 
+2024-08-30 09:47:20.035912:  
+2024-08-30 09:47:20.064198: Epoch 916 
+2024-08-30 09:47:20.108490: Current learning rate: 0.00108 
+2024-08-30 09:48:12.034337: train_loss -0.8468 
+2024-08-30 09:48:12.035556: val_loss -0.8684 
+2024-08-30 09:48:12.036769: Pseudo dice [0.9457, 0.9523, 0.9467, 0.9505, 0.9529, 0.9644, 0.9497, 0.9333, 0.934, 0.9413, 0.9452, 0.9597, 0.9665, 0.9701, 0.9652, 0.9567, 0.9102, 0.0, 0.9573, nan, 0.0] 
+2024-08-30 09:48:12.037589: Epoch time: 52.0 s 
+2024-08-30 09:48:13.400879:  
+2024-08-30 09:48:13.401742: Epoch 917 
+2024-08-30 09:48:13.402465: Current learning rate: 0.00106 
+2024-08-30 09:49:05.424017: train_loss -0.842 
+2024-08-30 09:49:05.426039: val_loss -0.8653 
+2024-08-30 09:49:05.426991: Pseudo dice [0.923, 0.9549, 0.9529, 0.9535, 0.9501, 0.9481, 0.9586, 0.9605, 0.9647, 0.9662, 0.9659, 0.9611, 0.9658, 0.9694, 0.959, 0.9398, 0.9007, 0.0, 0.9568, nan, nan] 
+2024-08-30 09:49:05.427838: Epoch time: 52.03 s 
+2024-08-30 09:49:06.859180:  
+2024-08-30 09:49:06.860131: Epoch 918 
+2024-08-30 09:49:06.860854: Current learning rate: 0.00105 
+2024-08-30 09:49:59.689979: train_loss -0.8431 
+2024-08-30 09:49:59.692385: val_loss -0.8596 
+2024-08-30 09:49:59.693380: Pseudo dice [0.9444, 0.9573, 0.9579, 0.9557, 0.9212, 0.9125, 0.9291, 0.945, 0.961, 0.9641, 0.955, 0.9588, 0.9684, 0.9568, 0.9043, 0.889, 0.8703, 0.0, 0.9561, nan, nan] 
+2024-08-30 09:49:59.694192: Epoch time: 52.83 s 
+2024-08-30 09:50:01.139813:  
+2024-08-30 09:50:01.140732: Epoch 919 
+2024-08-30 09:50:01.141433: Current learning rate: 0.00104 
+2024-08-30 09:50:52.872537: train_loss -0.8485 
+2024-08-30 09:50:52.874662: val_loss -0.8641 
+2024-08-30 09:50:52.875826: Pseudo dice [0.9312, 0.9414, 0.9545, 0.9463, 0.9217, 0.9152, 0.9335, 0.9404, 0.9481, 0.9544, 0.9634, 0.9661, 0.9726, 0.9641, 0.9549, 0.9482, 0.9081, 0.0, 0.9539, nan, nan] 
+2024-08-30 09:50:52.876757: Epoch time: 51.74 s 
+2024-08-30 09:50:54.344548:  
+2024-08-30 09:50:54.345468: Epoch 920 
+2024-08-30 09:50:54.346212: Current learning rate: 0.00103 
+2024-08-30 09:51:46.919124: train_loss -0.8516 
+2024-08-30 09:51:47.050050: val_loss -0.8514 
+2024-08-30 09:51:47.141948: Pseudo dice [0.9304, 0.9619, 0.9617, 0.9574, 0.953, 0.9654, 0.9728, 0.9746, 0.9669, 0.9523, 0.9483, 0.9447, 0.9217, 0.8759, 0.8593, 0.8695, 0.857, 0.0, 0.946, nan, nan] 
+2024-08-30 09:51:47.169266: Epoch time: 52.58 s 
+2024-08-30 09:51:48.569705:  
+2024-08-30 09:51:48.589292: Epoch 921 
+2024-08-30 09:51:48.589994: Current learning rate: 0.00102 
+2024-08-30 09:52:41.788446: train_loss -0.8457 
+2024-08-30 09:52:41.790469: val_loss -0.8683 
+2024-08-30 09:52:41.791498: Pseudo dice [0.9368, 0.955, 0.9573, 0.9629, 0.9651, 0.9514, 0.9302, 0.9205, 0.9281, 0.9265, 0.9274, 0.9183, 0.9399, 0.946, 0.9521, 0.9707, 0.9617, nan, 0.9598, nan, nan] 
+2024-08-30 09:52:41.792266: Epoch time: 53.22 s 
+2024-08-30 09:52:43.170865:  
+2024-08-30 09:52:43.171814: Epoch 922 
+2024-08-30 09:52:43.172486: Current learning rate: 0.00101 
+2024-08-30 09:53:36.126686: train_loss -0.8445 
+2024-08-30 09:53:36.127753: val_loss -0.8666 
+2024-08-30 09:53:36.128754: Pseudo dice [0.9288, 0.9428, 0.9474, 0.9483, 0.9434, 0.9349, 0.9364, 0.9442, 0.9444, 0.939, 0.939, 0.9428, 0.955, 0.967, 0.9688, 0.9616, 0.9327, 0.0, 0.951, nan, nan] 
+2024-08-30 09:53:36.129586: Epoch time: 52.96 s 
+2024-08-30 09:53:36.130816: Yayy! New best EMA pseudo Dice: 0.8885 
+2024-08-30 09:53:39.801655:  
+2024-08-30 09:53:39.802550: Epoch 923 
+2024-08-30 09:53:39.803259: Current learning rate: 0.001 
+2024-08-30 09:54:33.398716: train_loss -0.8484 
+2024-08-30 09:54:33.401361: val_loss -0.8515 
+2024-08-30 09:54:33.402522: Pseudo dice [0.8997, 0.9143, 0.9175, 0.9186, 0.9427, 0.9346, 0.9287, 0.9462, 0.9505, 0.9428, 0.9277, 0.9285, 0.8983, 0.8973, 0.9088, 0.9193, 0.9334, nan, 0.9575, nan, 0.0] 
+2024-08-30 09:54:33.403441: Epoch time: 53.6 s 
+2024-08-30 09:54:34.808954:  
+2024-08-30 09:54:34.809764: Epoch 924 
+2024-08-30 09:54:34.810424: Current learning rate: 0.00098 
+2024-08-30 09:55:28.088682: train_loss -0.8511 
+2024-08-30 09:55:28.090126: val_loss -0.8449 
+2024-08-30 09:55:28.091139: Pseudo dice [0.9267, 0.9525, 0.954, 0.9556, 0.967, 0.9433, 0.923, 0.9156, 0.909, 0.916, 0.9098, 0.9217, 0.9491, 0.9631, 0.9363, 0.8794, 0.8242, 0.0, 0.9476, nan, nan] 
+2024-08-30 09:55:28.091930: Epoch time: 53.28 s 
+2024-08-30 09:55:29.546050:  
+2024-08-30 09:55:29.546935: Epoch 925 
+2024-08-30 09:55:29.547622: Current learning rate: 0.00097 
+2024-08-30 09:56:20.578186: train_loss -0.8467 
+2024-08-30 09:56:20.580002: val_loss -0.8509 
+2024-08-30 09:56:20.580908: Pseudo dice [0.897, 0.9127, 0.9111, 0.9206, 0.9238, 0.9235, 0.9344, 0.9569, 0.9414, 0.9223, 0.9145, 0.9139, 0.9645, 0.9502, 0.9306, 0.9167, 0.875, 0.0, 0.9509, nan, 0.0] 
+2024-08-30 09:56:20.581635: Epoch time: 51.04 s 
+2024-08-30 09:56:22.278639:  
+2024-08-30 09:56:22.279523: Epoch 926 
+2024-08-30 09:56:22.280306: Current learning rate: 0.00096 
+2024-08-30 09:57:17.439591: train_loss -0.8471 
+2024-08-30 09:57:17.441475: val_loss -0.858 
+2024-08-30 09:57:17.442336: Pseudo dice [0.9342, 0.9531, 0.955, 0.9623, 0.9609, 0.9706, 0.9524, 0.9177, 0.9257, 0.9353, 0.9205, 0.9225, 0.9538, 0.947, 0.9409, 0.9246, 0.9254, 0.0, 0.9634, nan, 0.0] 
+2024-08-30 09:57:17.443338: Epoch time: 55.16 s 
+2024-08-30 09:57:18.847193:  
+2024-08-30 09:57:18.848045: Epoch 927 
+2024-08-30 09:57:18.848760: Current learning rate: 0.00095 
+2024-08-30 09:58:12.705588: train_loss -0.835 
+2024-08-30 09:58:12.707983: val_loss -0.8488 
+2024-08-30 09:58:12.709411: Pseudo dice [0.9369, 0.9562, 0.9525, 0.9439, 0.9239, 0.9238, 0.933, 0.9301, 0.9272, 0.936, 0.9399, 0.9489, 0.9658, 0.937, 0.8944, 0.8545, 0.8314, 0.0, 0.937, nan, nan] 
+2024-08-30 09:58:12.710254: Epoch time: 53.86 s 
+2024-08-30 09:58:14.085509:  
+2024-08-30 09:58:14.086400: Epoch 928 
+2024-08-30 09:58:14.087110: Current learning rate: 0.00094 
+2024-08-30 09:59:08.547267: train_loss -0.8446 
+2024-08-30 09:59:08.548403: val_loss -0.8622 
+2024-08-30 09:59:08.549146: Pseudo dice [0.9348, 0.9537, 0.9508, 0.9588, 0.9615, 0.9586, 0.957, 0.9564, 0.9631, 0.9582, 0.9505, 0.9516, 0.9662, 0.941, 0.9272, 0.9393, 0.9324, 0.0, 0.9551, nan, nan] 
+2024-08-30 09:59:08.549765: Epoch time: 54.46 s 
+2024-08-30 09:59:10.562996:  
+2024-08-30 09:59:10.563841: Epoch 929 
+2024-08-30 09:59:10.564563: Current learning rate: 0.00092 
+2024-08-30 10:00:07.417935: train_loss -0.8513 
+2024-08-30 10:00:07.420214: val_loss -0.847 
+2024-08-30 10:00:07.421000: Pseudo dice [0.9407, 0.9197, 0.9079, 0.9191, 0.9232, 0.9127, 0.9074, 0.9006, 0.8912, 0.9228, 0.9414, 0.9477, 0.9045, 0.8925, 0.8886, 0.859, 0.8429, 0.0, 0.952, nan, 0.0] 
+2024-08-30 10:00:07.421667: Epoch time: 56.86 s 
+2024-08-30 10:00:08.787585:  
+2024-08-30 10:00:08.788558: Epoch 930 
+2024-08-30 10:00:08.789189: Current learning rate: 0.00091 
+2024-08-30 10:01:01.196343: train_loss -0.8495 
+2024-08-30 10:01:01.197545: val_loss -0.8592 
+2024-08-30 10:01:01.198351: Pseudo dice [0.9444, 0.9518, 0.9453, 0.9306, 0.9506, 0.9457, 0.9509, 0.9376, 0.9328, 0.9368, 0.9286, 0.9335, 0.9487, 0.9432, 0.9172, 0.9221, 0.905, 0.0, 0.9579, nan, nan] 
+2024-08-30 10:01:01.199045: Epoch time: 52.41 s 
+2024-08-30 10:01:02.768086:  
+2024-08-30 10:01:02.768924: Epoch 931 
+2024-08-30 10:01:02.769610: Current learning rate: 0.0009 
+2024-08-30 10:01:55.521782: train_loss -0.8537 
+2024-08-30 10:01:55.524310: val_loss -0.8603 
+2024-08-30 10:01:55.525145: Pseudo dice [0.9412, 0.9493, 0.9577, 0.9637, 0.9387, 0.9095, 0.9091, 0.9166, 0.9083, 0.9041, 0.9196, 0.9138, 0.9356, 0.9538, 0.944, 0.9366, 0.9278, nan, 0.9592, nan, nan] 
+2024-08-30 10:01:55.526204: Epoch time: 52.76 s 
+2024-08-30 10:01:56.917265:  
+2024-08-30 10:01:56.918360: Epoch 932 
+2024-08-30 10:01:56.919020: Current learning rate: 0.00089 
+2024-08-30 10:02:50.055418: train_loss -0.8415 
+2024-08-30 10:02:50.056636: val_loss -0.8556 
+2024-08-30 10:02:50.057668: Pseudo dice [0.9323, 0.9378, 0.9381, 0.9547, 0.957, 0.9645, 0.963, 0.9492, 0.9467, 0.9584, 0.9704, 0.9722, 0.9571, 0.9301, 0.8894, 0.8112, 0.7112, 0.0, 0.9598, nan, nan] 
+2024-08-30 10:02:50.058403: Epoch time: 53.14 s 
+2024-08-30 10:02:51.538301:  
+2024-08-30 10:02:51.539323: Epoch 933 
+2024-08-30 10:02:51.540074: Current learning rate: 0.00088 
+2024-08-30 10:03:42.849202: train_loss -0.8494 
+2024-08-30 10:03:42.851120: val_loss -0.8589 
+2024-08-30 10:03:42.851995: Pseudo dice [0.9421, 0.9578, 0.952, 0.9497, 0.9338, 0.9144, 0.9034, 0.9141, 0.9011, 0.8926, 0.8806, 0.8981, 0.9358, 0.9459, 0.95, 0.9464, 0.9092, 0.0, 0.9503, nan, 0.0] 
+2024-08-30 10:03:42.852779: Epoch time: 51.31 s 
+2024-08-30 10:03:44.188414:  
+2024-08-30 10:03:44.189330: Epoch 934 
+2024-08-30 10:03:44.190004: Current learning rate: 0.00087 
+2024-08-30 10:04:36.412654: train_loss -0.8525 
+2024-08-30 10:04:36.414294: val_loss -0.8557 
+2024-08-30 10:04:36.415700: Pseudo dice [0.898, 0.9104, 0.9188, 0.9263, 0.9027, 0.9003, 0.9326, 0.9562, 0.9608, 0.9633, 0.9462, 0.9462, 0.9568, 0.9351, 0.9318, 0.925, 0.8876, 0.0, 0.9538, nan, nan] 
+2024-08-30 10:04:36.416319: Epoch time: 52.23 s 
+2024-08-30 10:04:38.010796:  
+2024-08-30 10:04:38.011621: Epoch 935 
+2024-08-30 10:04:38.012287: Current learning rate: 0.00085 
+2024-08-30 10:05:31.319292: train_loss -0.845 
+2024-08-30 10:05:31.321297: val_loss -0.8617 
+2024-08-30 10:05:31.322178: Pseudo dice [0.9201, 0.9441, 0.9469, 0.9533, 0.9553, 0.9651, 0.9658, 0.9542, 0.9534, 0.953, 0.9597, 0.9705, 0.9708, 0.9728, 0.9694, 0.9424, 0.8878, 0.0, 0.9479, nan, nan] 
+2024-08-30 10:05:31.322860: Epoch time: 53.31 s 
+2024-08-30 10:05:32.736392:  
+2024-08-30 10:05:32.737269: Epoch 936 
+2024-08-30 10:05:32.737977: Current learning rate: 0.00084 
+2024-08-30 10:06:25.349432: train_loss -0.8485 
+2024-08-30 10:06:25.350676: val_loss -0.8688 
+2024-08-30 10:06:25.351754: Pseudo dice [0.9399, 0.9506, 0.9487, 0.9533, 0.9548, 0.9631, 0.9571, 0.9414, 0.9548, 0.9739, 0.9745, 0.9736, 0.9703, 0.9209, 0.9098, 0.9307, 0.9211, 0.0, 0.9639, nan, nan] 
+2024-08-30 10:06:25.352515: Epoch time: 52.62 s 
+2024-08-30 10:06:27.018242:  
+2024-08-30 10:06:27.019112: Epoch 937 
+2024-08-30 10:06:27.019696: Current learning rate: 0.00083 
+2024-08-30 10:07:17.556547: train_loss -0.8495 
+2024-08-30 10:07:17.560977: val_loss -0.8565 
+2024-08-30 10:07:17.561835: Pseudo dice [0.9357, 0.9529, 0.9492, 0.9554, 0.9551, 0.9555, 0.9568, 0.9557, 0.954, 0.9616, 0.9713, 0.9707, 0.9661, 0.9526, 0.9377, 0.9244, 0.875, 0.0, 0.9574, nan, nan] 
+2024-08-30 10:07:17.562643: Epoch time: 50.54 s 
+2024-08-30 10:07:19.018479:  
+2024-08-30 10:07:19.019373: Epoch 938 
+2024-08-30 10:07:19.020112: Current learning rate: 0.00082 
+2024-08-30 10:08:10.795385: train_loss -0.8497 
+2024-08-30 10:08:10.796620: val_loss -0.8659 
+2024-08-30 10:08:10.797538: Pseudo dice [0.947, 0.9601, 0.9543, 0.9229, 0.8806, 0.8887, 0.902, 0.9312, 0.9388, 0.9307, 0.9311, 0.9564, 0.975, 0.9585, 0.9433, 0.9293, 0.9085, 0.0, 0.962, nan, nan] 
+2024-08-30 10:08:10.798293: Epoch time: 51.78 s 
+2024-08-30 10:08:12.109109:  
+2024-08-30 10:08:12.110064: Epoch 939 
+2024-08-30 10:08:12.111285: Current learning rate: 0.00081 
+2024-08-30 10:09:05.459191: train_loss -0.8513 
+2024-08-30 10:09:05.461498: val_loss -0.8631 
+2024-08-30 10:09:05.462358: Pseudo dice [0.9396, 0.9531, 0.9499, 0.9481, 0.9592, 0.9571, 0.9458, 0.9453, 0.9477, 0.9393, 0.9074, 0.9308, 0.9717, 0.9697, 0.9644, 0.9672, 0.9457, 0.0, 0.9591, nan, nan] 
+2024-08-30 10:09:05.463133: Epoch time: 53.35 s 
+2024-08-30 10:09:06.766939:  
+2024-08-30 10:09:06.767855: Epoch 940 
+2024-08-30 10:09:06.768591: Current learning rate: 0.00079 
+2024-08-30 10:10:00.599563: train_loss -0.8483 
+2024-08-30 10:10:00.600801: val_loss -0.8512 
+2024-08-30 10:10:00.601717: Pseudo dice [0.9324, 0.9517, 0.9488, 0.9508, 0.9469, 0.9442, 0.9477, 0.9416, 0.9403, 0.9358, 0.9401, 0.9398, 0.9471, 0.9298, 0.9133, 0.9053, 0.8635, 0.0, 0.9542, nan, nan] 
+2024-08-30 10:10:00.602605: Epoch time: 53.84 s 
+2024-08-30 10:10:01.960508:  
+2024-08-30 10:10:01.961484: Epoch 941 
+2024-08-30 10:10:01.963042: Current learning rate: 0.00078 
+2024-08-30 10:10:55.619714: train_loss -0.8418 
+2024-08-30 10:10:55.621739: val_loss -0.8536 
+2024-08-30 10:10:55.623024: Pseudo dice [0.9313, 0.9371, 0.9095, 0.9275, 0.9525, 0.9471, 0.9326, 0.9257, 0.9455, 0.9546, 0.9468, 0.9524, 0.9538, 0.9208, 0.8967, 0.9093, 0.9048, 0.0, 0.9623, nan, nan] 
+2024-08-30 10:10:55.623796: Epoch time: 53.66 s 
+2024-08-30 10:10:57.169628:  
+2024-08-30 10:10:57.170497: Epoch 942 
+2024-08-30 10:10:57.171117: Current learning rate: 0.00077 
+2024-08-30 10:11:50.687285: train_loss -0.8446 
+2024-08-30 10:11:50.689352: val_loss -0.8614 
+2024-08-30 10:11:50.690510: Pseudo dice [0.9353, 0.9151, 0.8989, 0.9208, 0.9384, 0.9407, 0.9502, 0.9572, 0.9437, 0.9392, 0.9362, 0.9302, 0.9763, 0.9756, 0.9776, 0.9761, 0.9484, nan, 0.9563, nan, 0.0] 
+2024-08-30 10:11:50.691401: Epoch time: 53.52 s 
+2024-08-30 10:11:52.028788:  
+2024-08-30 10:11:52.029699: Epoch 943 
+2024-08-30 10:11:52.030868: Current learning rate: 0.00076 
+2024-08-30 10:12:44.592405: train_loss -0.8482 
+2024-08-30 10:12:44.594781: val_loss -0.8415 
+2024-08-30 10:12:44.595926: Pseudo dice [0.936, 0.9144, 0.8553, 0.8094, 0.8132, 0.8317, 0.8481, 0.867, 0.8431, 0.8296, 0.8696, 0.9108, 0.9678, 0.9529, 0.9487, 0.9601, 0.9371, 0.0, 0.9445, nan, 0.0] 
+2024-08-30 10:12:44.597034: Epoch time: 52.57 s 
+2024-08-30 10:12:45.926375:  
+2024-08-30 10:12:45.927188: Epoch 944 
+2024-08-30 10:12:45.927808: Current learning rate: 0.00075 
+2024-08-30 10:13:38.911172: train_loss -0.8484 
+2024-08-30 10:13:38.912382: val_loss -0.8621 
+2024-08-30 10:13:38.913141: Pseudo dice [0.942, 0.9591, 0.9562, 0.9496, 0.9505, 0.9673, 0.9708, 0.9698, 0.9483, 0.9332, 0.9397, 0.9332, 0.9444, 0.9698, 0.9592, 0.9467, 0.9292, 0.0, 0.9606, nan, nan] 
+2024-08-30 10:13:38.913756: Epoch time: 52.99 s 
+2024-08-30 10:13:40.294965:  
+2024-08-30 10:13:40.295834: Epoch 945 
+2024-08-30 10:13:40.296935: Current learning rate: 0.00074 
+2024-08-30 10:14:31.573598: train_loss -0.8465 
+2024-08-30 10:14:31.575586: val_loss -0.8651 
+2024-08-30 10:14:31.576441: Pseudo dice [0.9389, 0.9581, 0.9415, 0.9293, 0.9294, 0.9374, 0.9554, 0.9534, 0.9519, 0.9529, 0.9394, 0.937, 0.9609, 0.9627, 0.9668, 0.9657, 0.9276, 0.0, 0.9481, nan, nan] 
+2024-08-30 10:14:31.577214: Epoch time: 51.28 s 
+2024-08-30 10:14:32.872183:  
+2024-08-30 10:14:32.873069: Epoch 946 
+2024-08-30 10:14:32.873809: Current learning rate: 0.00072 
+2024-08-30 10:15:23.738815: train_loss -0.8518 
+2024-08-30 10:15:23.740160: val_loss -0.8489 
+2024-08-30 10:15:23.741539: Pseudo dice [0.9433, 0.9538, 0.9368, 0.9441, 0.9668, 0.9477, 0.9269, 0.9266, 0.9073, 0.8988, 0.9023, 0.8886, 0.9207, 0.9466, 0.9653, 0.9378, 0.8964, 0.0, 0.9341, nan, nan] 
+2024-08-30 10:15:23.742865: Epoch time: 50.87 s 
+2024-08-30 10:15:25.223688:  
+2024-08-30 10:15:25.225006: Epoch 947 
+2024-08-30 10:15:25.225703: Current learning rate: 0.00071 
+2024-08-30 10:16:17.235266: train_loss -0.8482 
+2024-08-30 10:16:17.237830: val_loss -0.8527 
+2024-08-30 10:16:17.238741: Pseudo dice [0.9489, 0.9583, 0.9581, 0.9618, 0.9507, 0.9305, 0.9245, 0.9277, 0.9346, 0.9362, 0.9359, 0.9237, 0.95, 0.9412, 0.9226, 0.9141, 0.8771, 0.0, 0.9424, nan, 0.0] 
+2024-08-30 10:16:17.239523: Epoch time: 52.01 s 
+2024-08-30 10:16:18.688260:  
+2024-08-30 10:16:18.689226: Epoch 948 
+2024-08-30 10:16:18.689977: Current learning rate: 0.0007 
+2024-08-30 10:17:10.591592: train_loss -0.8459 
+2024-08-30 10:17:10.592764: val_loss -0.8723 
+2024-08-30 10:17:10.593572: Pseudo dice [0.9477, 0.9599, 0.9539, 0.961, 0.9535, 0.9472, 0.9611, 0.9546, 0.9446, 0.9533, 0.9679, 0.9669, 0.972, 0.9611, 0.9489, 0.9493, 0.9316, nan, 0.9558, nan, nan] 
+2024-08-30 10:17:10.595065: Epoch time: 51.91 s 
+2024-08-30 10:17:12.050292:  
+2024-08-30 10:17:12.051152: Epoch 949 
+2024-08-30 10:17:12.051842: Current learning rate: 0.00069 
+2024-08-30 10:18:04.503434: train_loss -0.8511 
+2024-08-30 10:18:04.505441: val_loss -0.8623 
+2024-08-30 10:18:04.506627: Pseudo dice [0.9348, 0.9528, 0.9361, 0.9381, 0.94, 0.9376, 0.953, 0.9522, 0.9423, 0.9401, 0.9352, 0.9533, 0.9705, 0.9692, 0.9558, 0.9346, 0.8951, 0.0, 0.9522, nan, nan] 
+2024-08-30 10:18:04.507421: Epoch time: 52.46 s 
+2024-08-30 10:18:08.332660:  
+2024-08-30 10:18:08.333610: Epoch 950 
+2024-08-30 10:18:08.334377: Current learning rate: 0.00067 
+2024-08-30 10:18:58.365385: train_loss -0.8504 
+2024-08-30 10:18:58.366978: val_loss -0.8706 
+2024-08-30 10:18:58.367857: Pseudo dice [0.9368, 0.9528, 0.955, 0.9555, 0.957, 0.9679, 0.9657, 0.9605, 0.9534, 0.9609, 0.972, 0.971, 0.9625, 0.9421, 0.9312, 0.943, 0.9552, nan, 0.9642, nan, nan] 
+2024-08-30 10:18:58.368901: Epoch time: 50.04 s 
+2024-08-30 10:18:58.370017: Yayy! New best EMA pseudo Dice: 0.8934 
+2024-08-30 10:19:02.179308:  
+2024-08-30 10:19:02.180123: Epoch 951 
+2024-08-30 10:19:02.180810: Current learning rate: 0.00066 
+2024-08-30 10:19:54.884330: train_loss -0.8495 
+2024-08-30 10:19:54.886884: val_loss -0.8577 
+2024-08-30 10:19:54.888390: Pseudo dice [0.9374, 0.9279, 0.9047, 0.9259, 0.9199, 0.9153, 0.914, 0.9093, 0.8888, 0.896, 0.9074, 0.9266, 0.9699, 0.9726, 0.9751, 0.9716, 0.9423, 0.0, 0.9379, nan, 0.0] 
+2024-08-30 10:19:54.889423: Epoch time: 52.71 s 
+2024-08-30 10:19:56.320306:  
+2024-08-30 10:19:56.321118: Epoch 952 
+2024-08-30 10:19:56.321761: Current learning rate: 0.00065 
+2024-08-30 10:20:50.129279: train_loss -0.8535 
+2024-08-30 10:20:50.132541: val_loss -0.8661 
+2024-08-30 10:20:50.133902: Pseudo dice [0.9408, 0.9576, 0.9585, 0.9653, 0.9425, 0.9266, 0.93, 0.9352, 0.9346, 0.9343, 0.9364, 0.9454, 0.9632, 0.9652, 0.9092, 0.865, 0.8257, 0.0, 0.9382, nan, nan] 
+2024-08-30 10:20:50.135259: Epoch time: 53.81 s 
+2024-08-30 10:20:51.537604:  
+2024-08-30 10:20:51.538487: Epoch 953 
+2024-08-30 10:20:51.539279: Current learning rate: 0.00064 
+2024-08-30 10:21:45.772618: train_loss -0.8468 
+2024-08-30 10:21:45.774719: val_loss -0.8624 
+2024-08-30 10:21:45.775752: Pseudo dice [0.9431, 0.9498, 0.9357, 0.9299, 0.9372, 0.9245, 0.928, 0.9424, 0.9433, 0.946, 0.9426, 0.9343, 0.9528, 0.9536, 0.9578, 0.9525, 0.9319, 0.0, 0.9445, nan, 0.0] 
+2024-08-30 10:21:45.776482: Epoch time: 54.24 s 
+2024-08-30 10:21:47.152239:  
+2024-08-30 10:21:47.153165: Epoch 954 
+2024-08-30 10:21:47.153802: Current learning rate: 0.00063 
+2024-08-30 10:22:37.036758: train_loss -0.8478 
+2024-08-30 10:22:37.038927: val_loss -0.8647 
+2024-08-30 10:22:37.040422: Pseudo dice [0.9471, 0.9597, 0.9566, 0.961, 0.9667, 0.9687, 0.9667, 0.9666, 0.972, 0.9698, 0.9631, 0.9627, 0.9761, 0.9688, 0.9384, 0.8997, 0.8372, 0.0, 0.9544, nan, nan] 
+2024-08-30 10:22:37.042530: Epoch time: 49.89 s 
+2024-08-30 10:22:39.197179:  
+2024-08-30 10:22:39.198064: Epoch 955 
+2024-08-30 10:22:39.198757: Current learning rate: 0.00061 
+2024-08-30 10:23:32.983045: train_loss -0.8514 
+2024-08-30 10:23:32.984947: val_loss -0.8497 
+2024-08-30 10:23:32.985809: Pseudo dice [0.9417, 0.9458, 0.9277, 0.9272, 0.9095, 0.8993, 0.8906, 0.8902, 0.8748, 0.8802, 0.8889, 0.9122, 0.9483, 0.9236, 0.9233, 0.9228, 0.8949, 0.0, 0.9582, nan, nan] 
+2024-08-30 10:23:32.986533: Epoch time: 53.79 s 
+2024-08-30 10:23:34.411091:  
+2024-08-30 10:23:34.412478: Epoch 956 
+2024-08-30 10:23:34.413252: Current learning rate: 0.0006 
+2024-08-30 10:24:25.120140: train_loss -0.8509 
+2024-08-30 10:24:25.122234: val_loss -0.8607 
+2024-08-30 10:24:25.123515: Pseudo dice [0.9355, 0.9569, 0.9581, 0.956, 0.9607, 0.9507, 0.9372, 0.9236, 0.9172, 0.9216, 0.92, 0.9142, 0.9196, 0.9479, 0.9709, 0.9716, 0.9498, nan, 0.9579, nan, nan] 
+2024-08-30 10:24:25.124424: Epoch time: 50.71 s 
+2024-08-30 10:24:26.560558:  
+2024-08-30 10:24:26.561507: Epoch 957 
+2024-08-30 10:24:26.562325: Current learning rate: 0.00059 
+2024-08-30 10:25:18.697950: train_loss -0.8523 
+2024-08-30 10:25:18.699968: val_loss -0.8638 
+2024-08-30 10:25:18.701277: Pseudo dice [0.9488, 0.9585, 0.9532, 0.9477, 0.9519, 0.9542, 0.9468, 0.9408, 0.9509, 0.9569, 0.952, 0.9572, 0.9749, 0.9708, 0.9532, 0.9316, 0.8887, 0.0, 0.9573, nan, nan] 
+2024-08-30 10:25:18.702055: Epoch time: 52.14 s 
+2024-08-30 10:25:20.114854:  
+2024-08-30 10:25:20.115787: Epoch 958 
+2024-08-30 10:25:20.116835: Current learning rate: 0.00058 
+2024-08-30 10:26:16.232098: train_loss -0.8528 
+2024-08-30 10:26:16.234014: val_loss -0.8627 
+2024-08-30 10:26:16.235000: Pseudo dice [0.9405, 0.951, 0.9565, 0.9631, 0.9693, 0.9697, 0.9625, 0.9524, 0.9215, 0.9027, 0.9065, 0.9178, 0.9741, 0.958, 0.9463, 0.9399, 0.9012, 0.0, 0.961, nan, 0.0] 
+2024-08-30 10:26:16.235769: Epoch time: 56.12 s 
+2024-08-30 10:26:17.819944:  
+2024-08-30 10:26:17.820786: Epoch 959 
+2024-08-30 10:26:17.821509: Current learning rate: 0.00056 
+2024-08-30 10:27:11.785599: train_loss -0.8528 
+2024-08-30 10:27:11.787754: val_loss -0.8548 
+2024-08-30 10:27:11.788694: Pseudo dice [0.9354, 0.9556, 0.9546, 0.9597, 0.9667, 0.9674, 0.9668, 0.9666, 0.9609, 0.9453, 0.9371, 0.9278, 0.9338, 0.9075, 0.9006, 0.9045, 0.9078, 0.0, 0.9586, nan, 0.0] 
+2024-08-30 10:27:11.789490: Epoch time: 53.97 s 
+2024-08-30 10:27:13.205766:  
+2024-08-30 10:27:13.206791: Epoch 960 
+2024-08-30 10:27:13.207520: Current learning rate: 0.00055 
+2024-08-30 10:28:03.886426: train_loss -0.8516 
+2024-08-30 10:28:03.888169: val_loss -0.8626 
+2024-08-30 10:28:03.889139: Pseudo dice [0.9472, 0.9578, 0.9634, 0.9636, 0.963, 0.9673, 0.9643, 0.9611, 0.9578, 0.9668, 0.9577, 0.9583, 0.9769, 0.9593, 0.9384, 0.9226, 0.8688, 0.0, 0.9635, nan, 0.0] 
+2024-08-30 10:28:03.889918: Epoch time: 50.68 s 
+2024-08-30 10:28:05.554668:  
+2024-08-30 10:28:05.555649: Epoch 961 
+2024-08-30 10:28:05.556334: Current learning rate: 0.00054 
+2024-08-30 10:28:56.406420: train_loss -0.8509 
+2024-08-30 10:28:56.408435: val_loss -0.865 
+2024-08-30 10:28:56.409498: Pseudo dice [0.9393, 0.9537, 0.9568, 0.9603, 0.9462, 0.9466, 0.9639, 0.9547, 0.9497, 0.9518, 0.9599, 0.9745, 0.9762, 0.9508, 0.9305, 0.9266, 0.9018, 0.0, 0.9513, nan, nan] 
+2024-08-30 10:28:56.410274: Epoch time: 50.85 s 
+2024-08-30 10:28:57.787627:  
+2024-08-30 10:28:57.788509: Epoch 962 
+2024-08-30 10:28:57.789255: Current learning rate: 0.00053 
+2024-08-30 10:29:50.047581: train_loss -0.8475 
+2024-08-30 10:29:50.069606: val_loss -0.8617 
+2024-08-30 10:29:50.070552: Pseudo dice [0.9431, 0.9537, 0.9611, 0.9603, 0.9618, 0.9615, 0.9668, 0.9719, 0.9749, 0.9735, 0.951, 0.9393, 0.9506, 0.9258, 0.8976, 0.886, 0.8648, 0.0, 0.9555, nan, nan] 
+2024-08-30 10:29:50.071630: Epoch time: 52.26 s 
+2024-08-30 10:29:51.671604:  
+2024-08-30 10:29:51.672522: Epoch 963 
+2024-08-30 10:29:51.673182: Current learning rate: 0.00051 
+2024-08-30 10:30:47.847667: train_loss -0.8439 
+2024-08-30 10:30:47.849686: val_loss -0.8609 
+2024-08-30 10:30:47.850615: Pseudo dice [0.9255, 0.9272, 0.9185, 0.9198, 0.9233, 0.9346, 0.9349, 0.9366, 0.9476, 0.9637, 0.9615, 0.9644, 0.963, 0.9656, 0.9522, 0.9402, 0.8983, 0.0, 0.9443, nan, 0.0] 
+2024-08-30 10:30:47.851458: Epoch time: 56.18 s 
+2024-08-30 10:30:49.205772:  
+2024-08-30 10:30:49.206768: Epoch 964 
+2024-08-30 10:30:49.207551: Current learning rate: 0.0005 
+2024-08-30 10:31:39.172107: train_loss -0.8554 
+2024-08-30 10:31:39.174134: val_loss -0.8646 
+2024-08-30 10:31:39.175307: Pseudo dice [0.9501, 0.9612, 0.9612, 0.9636, 0.9571, 0.9479, 0.9443, 0.9442, 0.9453, 0.941, 0.9372, 0.9385, 0.9616, 0.9487, 0.951, 0.9487, 0.9177, 0.0, 0.9416, nan, 0.0] 
+2024-08-30 10:31:39.177831: Epoch time: 49.97 s 
+2024-08-30 10:31:40.590867:  
+2024-08-30 10:31:40.591808: Epoch 965 
+2024-08-30 10:31:40.592421: Current learning rate: 0.00049 
+2024-08-30 10:32:33.608516: train_loss -0.8513 
+2024-08-30 10:32:33.610365: val_loss -0.8555 
+2024-08-30 10:32:33.611141: Pseudo dice [0.9339, 0.9336, 0.9374, 0.9543, 0.9635, 0.9545, 0.9334, 0.9245, 0.9256, 0.9326, 0.9357, 0.9326, 0.9589, 0.9568, 0.9721, 0.9653, 0.9163, 0.0, 0.9506, nan, 0.0] 
+2024-08-30 10:32:33.611813: Epoch time: 53.02 s 
+2024-08-30 10:32:34.942311:  
+2024-08-30 10:32:34.943221: Epoch 966 
+2024-08-30 10:32:34.944027: Current learning rate: 0.00048 
+2024-08-30 10:33:28.128527: train_loss -0.8563 
+2024-08-30 10:33:28.130756: val_loss -0.8634 
+2024-08-30 10:33:28.131817: Pseudo dice [0.9453, 0.9552, 0.9565, 0.9553, 0.965, 0.9665, 0.9683, 0.9665, 0.9685, 0.9705, 0.9719, 0.9666, 0.9477, 0.931, 0.9283, 0.9128, 0.8518, 0.0, 0.9563, nan, nan] 
+2024-08-30 10:33:28.132555: Epoch time: 53.19 s 
+2024-08-30 10:33:29.519173:  
+2024-08-30 10:33:29.519992: Epoch 967 
+2024-08-30 10:33:29.520659: Current learning rate: 0.00046 
+2024-08-30 10:34:18.766661: train_loss -0.8448 
+2024-08-30 10:34:18.769221: val_loss -0.8632 
+2024-08-30 10:34:18.770272: Pseudo dice [0.9457, 0.9606, 0.9631, 0.9632, 0.9679, 0.9677, 0.9682, 0.965, 0.9589, 0.9661, 0.9746, 0.974, 0.9591, 0.9485, 0.9088, 0.9065, 0.8714, 0.0, 0.9564, nan, nan] 
+2024-08-30 10:34:18.772259: Epoch time: 49.25 s 
+2024-08-30 10:34:20.964338:  
+2024-08-30 10:34:20.965245: Epoch 968 
+2024-08-30 10:34:20.965929: Current learning rate: 0.00045 
+2024-08-30 10:35:11.805357: train_loss -0.8609 
+2024-08-30 10:35:11.807266: val_loss -0.8651 
+2024-08-30 10:35:11.808123: Pseudo dice [0.9342, 0.9529, 0.9568, 0.9437, 0.9374, 0.9392, 0.9332, 0.9365, 0.9466, 0.9474, 0.9442, 0.9262, 0.9469, 0.962, 0.9653, 0.9653, 0.9407, 0.0, 0.9573, nan, 0.0] 
+2024-08-30 10:35:11.808921: Epoch time: 50.84 s 
+2024-08-30 10:35:13.364706:  
+2024-08-30 10:35:13.365916: Epoch 969 
+2024-08-30 10:35:13.366628: Current learning rate: 0.00044 
+2024-08-30 10:36:03.890323: train_loss -0.8518 
+2024-08-30 10:36:03.892499: val_loss -0.8605 
+2024-08-30 10:36:03.893434: Pseudo dice [0.9404, 0.9564, 0.9572, 0.9579, 0.9614, 0.9661, 0.9651, 0.9691, 0.9508, 0.9418, 0.9412, 0.9487, 0.9676, 0.9618, 0.9378, 0.9173, 0.896, 0.0, 0.9498, nan, nan] 
+2024-08-30 10:36:03.894784: Epoch time: 50.53 s 
+2024-08-30 10:36:05.254843:  
+2024-08-30 10:36:05.255668: Epoch 970 
+2024-08-30 10:36:05.256326: Current learning rate: 0.00043 
+2024-08-30 10:36:56.767352: train_loss -0.8553 
+2024-08-30 10:36:56.769070: val_loss -0.8669 
+2024-08-30 10:36:56.770013: Pseudo dice [0.9409, 0.9599, 0.9579, 0.9625, 0.9647, 0.9703, 0.9704, 0.9722, 0.9749, 0.9766, 0.9753, 0.9784, 0.973, 0.9625, 0.958, 0.9415, 0.9008, 0.0, 0.9396, nan, nan] 
+2024-08-30 10:36:56.770851: Epoch time: 51.52 s 
+2024-08-30 10:36:58.303058:  
+2024-08-30 10:36:58.303992: Epoch 971 
+2024-08-30 10:36:58.304650: Current learning rate: 0.00041 
+2024-08-30 10:37:51.052966: train_loss -0.8511 
+2024-08-30 10:37:51.057210: val_loss -0.8695 
+2024-08-30 10:37:51.058389: Pseudo dice [0.9424, 0.9567, 0.9528, 0.9567, 0.9566, 0.9596, 0.9671, 0.9708, 0.9727, 0.974, 0.9613, 0.9613, 0.9678, 0.9598, 0.9492, 0.9396, 0.9303, 0.0, 0.9603, nan, nan] 
+2024-08-30 10:37:51.059462: Epoch time: 52.75 s 
+2024-08-30 10:37:52.660736:  
+2024-08-30 10:37:52.661870: Epoch 972 
+2024-08-30 10:37:52.662707: Current learning rate: 0.0004 
+2024-08-30 10:38:46.115569: train_loss -0.8473 
+2024-08-30 10:38:46.117456: val_loss -0.8688 
+2024-08-30 10:38:46.118410: Pseudo dice [0.9467, 0.9542, 0.9552, 0.963, 0.9655, 0.9685, 0.9669, 0.9641, 0.9567, 0.9394, 0.9496, 0.9705, 0.9757, 0.9614, 0.9687, 0.9639, 0.9063, 0.0, 0.9488, nan, nan] 
+2024-08-30 10:38:46.119312: Epoch time: 53.46 s 
+2024-08-30 10:38:47.479060:  
+2024-08-30 10:38:47.480008: Epoch 973 
+2024-08-30 10:38:47.480727: Current learning rate: 0.00039 
+2024-08-30 10:39:39.407035: train_loss -0.8547 
+2024-08-30 10:39:39.410274: val_loss -0.8459 
+2024-08-30 10:39:39.411667: Pseudo dice [0.9431, 0.9664, 0.9625, 0.9673, 0.9484, 0.9083, 0.8803, 0.8826, 0.9039, 0.897, 0.8781, 0.8948, 0.9488, 0.933, 0.9362, 0.9335, 0.8887, 0.0, 0.9493, nan, 0.0] 
+2024-08-30 10:39:39.412500: Epoch time: 51.93 s 
+2024-08-30 10:39:40.981354:  
+2024-08-30 10:39:40.982215: Epoch 974 
+2024-08-30 10:39:40.982883: Current learning rate: 0.00037 
+2024-08-30 10:40:33.646824: train_loss -0.8498 
+2024-08-30 10:40:33.648756: val_loss -0.8655 
+2024-08-30 10:40:33.649667: Pseudo dice [0.948, 0.9587, 0.9602, 0.9651, 0.9637, 0.957, 0.9566, 0.9663, 0.9671, 0.9683, 0.9659, 0.9546, 0.9351, 0.9272, 0.9344, 0.9261, 0.9274, nan, 0.9631, nan, nan] 
+2024-08-30 10:40:33.650858: Epoch time: 52.67 s 
+2024-08-30 10:40:34.929865:  
+2024-08-30 10:40:34.930692: Epoch 975 
+2024-08-30 10:40:34.931337: Current learning rate: 0.00036 
+2024-08-30 10:41:27.566087: train_loss -0.8482 
+2024-08-30 10:41:27.568153: val_loss -0.8608 
+2024-08-30 10:41:27.569205: Pseudo dice [0.9477, 0.9617, 0.9573, 0.9613, 0.9665, 0.9704, 0.9684, 0.967, 0.9444, 0.9346, 0.9434, 0.9492, 0.9617, 0.967, 0.9451, 0.9407, 0.9227, 0.0, 0.9552, nan, nan] 
+2024-08-30 10:41:27.570002: Epoch time: 52.64 s 
+2024-08-30 10:41:28.975760:  
+2024-08-30 10:41:28.976546: Epoch 976 
+2024-08-30 10:41:28.977172: Current learning rate: 0.00035 
+2024-08-30 10:42:21.734678: train_loss -0.8545 
+2024-08-30 10:42:21.736466: val_loss -0.8763 
+2024-08-30 10:42:21.737272: Pseudo dice [0.9477, 0.9601, 0.9572, 0.9584, 0.9648, 0.969, 0.9733, 0.9764, 0.9763, 0.9755, 0.9749, 0.9739, 0.9763, 0.9746, 0.976, 0.9676, 0.9392, 0.0, 0.9537, nan, nan] 
+2024-08-30 10:42:21.738056: Epoch time: 52.76 s 
+2024-08-30 10:42:23.365764:  
+2024-08-30 10:42:23.366827: Epoch 977 
+2024-08-30 10:42:23.367524: Current learning rate: 0.00034 
+2024-08-30 10:43:16.225462: train_loss -0.8596 
+2024-08-30 10:43:16.227605: val_loss -0.8631 
+2024-08-30 10:43:16.228473: Pseudo dice [0.9426, 0.9546, 0.9606, 0.9654, 0.9691, 0.9702, 0.9716, 0.9671, 0.9671, 0.9696, 0.9652, 0.9478, 0.9705, 0.9558, 0.9122, 0.8836, 0.8467, 0.0, 0.9624, nan, 0.0] 
+2024-08-30 10:43:16.229274: Epoch time: 52.86 s 
+2024-08-30 10:43:17.593902:  
+2024-08-30 10:43:17.594803: Epoch 978 
+2024-08-30 10:43:17.595508: Current learning rate: 0.00032 
+2024-08-30 10:44:08.411565: train_loss -0.8548 
+2024-08-30 10:44:08.414885: val_loss -0.8666 
+2024-08-30 10:44:08.415771: Pseudo dice [0.9365, 0.9421, 0.9448, 0.9496, 0.9406, 0.9489, 0.9591, 0.9543, 0.9571, 0.9643, 0.9624, 0.9663, 0.96, 0.9627, 0.9621, 0.9482, 0.9437, nan, 0.9587, nan, nan] 
+2024-08-30 10:44:08.416511: Epoch time: 50.82 s 
+2024-08-30 10:44:08.417217: Yayy! New best EMA pseudo Dice: 0.8949 
+2024-08-30 10:44:12.315161:  
+2024-08-30 10:44:12.315899: Epoch 979 
+2024-08-30 10:44:12.316523: Current learning rate: 0.00031 
+2024-08-30 10:45:03.216096: train_loss -0.8622 
+2024-08-30 10:45:03.218657: val_loss -0.8621 
+2024-08-30 10:45:03.219755: Pseudo dice [0.9129, 0.9241, 0.9367, 0.933, 0.9333, 0.93, 0.9326, 0.9444, 0.9535, 0.9536, 0.9569, 0.9657, 0.9696, 0.9659, 0.9547, 0.9481, 0.8986, 0.0, 0.9508, nan, nan] 
+2024-08-30 10:45:03.220763: Epoch time: 50.9 s 
+2024-08-30 10:45:04.848525:  
+2024-08-30 10:45:04.849304: Epoch 980 
+2024-08-30 10:45:04.850012: Current learning rate: 0.0003 
+2024-08-30 10:45:55.391572: train_loss -0.8482 
+2024-08-30 10:45:55.393471: val_loss -0.8648 
+2024-08-30 10:45:55.394352: Pseudo dice [0.9452, 0.9577, 0.9545, 0.9547, 0.9622, 0.9648, 0.9702, 0.9624, 0.9351, 0.9312, 0.9371, 0.9542, 0.973, 0.9738, 0.9742, 0.9427, 0.9357, 0.0, 0.9481, nan, nan] 
+2024-08-30 10:45:55.395075: Epoch time: 50.55 s 
+2024-08-30 10:45:55.395663: Yayy! New best EMA pseudo Dice: 0.8956 
+2024-08-30 10:45:59.639190:  
+2024-08-30 10:45:59.640233: Epoch 981 
+2024-08-30 10:45:59.640931: Current learning rate: 0.00028 
+2024-08-30 10:46:51.220230: train_loss -0.8533 
+2024-08-30 10:46:51.221961: val_loss -0.8645 
+2024-08-30 10:46:51.222793: Pseudo dice [0.9084, 0.9289, 0.9261, 0.9328, 0.937, 0.9523, 0.9694, 0.9728, 0.9708, 0.9676, 0.9626, 0.9544, 0.965, 0.9726, 0.9763, 0.9739, 0.9374, 0.0, 0.9653, nan, 0.0] 
+2024-08-30 10:46:51.223505: Epoch time: 51.58 s 
+2024-08-30 10:46:52.604110:  
+2024-08-30 10:46:52.604980: Epoch 982 
+2024-08-30 10:46:52.605722: Current learning rate: 0.00027 
+2024-08-30 10:47:45.156091: train_loss -0.8554 
+2024-08-30 10:47:45.157852: val_loss -0.8633 
+2024-08-30 10:47:45.158680: Pseudo dice [0.9449, 0.9627, 0.9621, 0.9571, 0.9606, 0.9661, 0.9689, 0.9756, 0.9716, 0.967, 0.9673, 0.9673, 0.9699, 0.9623, 0.9641, 0.9725, 0.9054, 0.0, 0.9439, nan, nan] 
+2024-08-30 10:47:45.159342: Epoch time: 52.55 s 
+2024-08-30 10:47:46.722095:  
+2024-08-30 10:47:46.723049: Epoch 983 
+2024-08-30 10:47:46.723816: Current learning rate: 0.00026 
+2024-08-30 10:48:40.935488: train_loss -0.8473 
+2024-08-30 10:48:40.937503: val_loss -0.8671 
+2024-08-30 10:48:40.938333: Pseudo dice [0.9467, 0.9546, 0.9525, 0.9532, 0.957, 0.961, 0.9544, 0.9614, 0.9701, 0.9724, 0.9739, 0.9706, 0.9625, 0.9511, 0.9437, 0.9373, 0.9353, 0.0, 0.9564, nan, nan] 
+2024-08-30 10:48:40.939109: Epoch time: 54.22 s 
+2024-08-30 10:48:42.372318:  
+2024-08-30 10:48:42.373050: Epoch 984 
+2024-08-30 10:48:42.373538: Current learning rate: 0.00024 
+2024-08-30 10:49:31.493529: train_loss -0.8557 
+2024-08-30 10:49:31.495258: val_loss -0.8646 
+2024-08-30 10:49:31.496255: Pseudo dice [0.9378, 0.9574, 0.9616, 0.9586, 0.966, 0.9679, 0.9707, 0.9704, 0.9639, 0.9673, 0.9545, 0.9563, 0.9653, 0.953, 0.9537, 0.9435, 0.9318, nan, 0.9561, nan, nan] 
+2024-08-30 10:49:31.496979: Epoch time: 49.12 s 
+2024-08-30 10:49:31.497591: Yayy! New best EMA pseudo Dice: 0.9012 
+2024-08-30 10:49:36.175643:  
+2024-08-30 10:49:36.176605: Epoch 985 
+2024-08-30 10:49:36.177341: Current learning rate: 0.00023 
+2024-08-30 10:50:23.749144: train_loss -0.8586 
+2024-08-30 10:50:23.751307: val_loss -0.8555 
+2024-08-30 10:50:23.752171: Pseudo dice [0.8816, 0.9021, 0.9041, 0.9093, 0.9088, 0.9426, 0.9738, 0.9772, 0.9757, 0.9724, 0.9509, 0.9501, 0.9708, 0.9572, 0.9328, 0.9296, 0.8655, 0.0, 0.9413, nan, nan] 
+2024-08-30 10:50:23.752908: Epoch time: 47.58 s 
+2024-08-30 10:50:25.481323:  
+2024-08-30 10:50:25.482186: Epoch 986 
+2024-08-30 10:50:25.482784: Current learning rate: 0.00021 
+2024-08-30 10:51:18.112314: train_loss -0.8506 
+2024-08-30 10:51:18.114195: val_loss -0.8693 
+2024-08-30 10:51:18.115175: Pseudo dice [0.9465, 0.9583, 0.96, 0.9615, 0.9657, 0.9728, 0.9767, 0.9735, 0.9725, 0.9713, 0.9715, 0.9567, 0.9247, 0.9496, 0.9447, 0.9297, 0.9349, nan, 0.9532, nan, 0.0] 
+2024-08-30 10:51:18.116014: Epoch time: 52.63 s 
+2024-08-30 10:51:19.735790:  
+2024-08-30 10:51:19.736687: Epoch 987 
+2024-08-30 10:51:19.737489: Current learning rate: 0.0002 
+2024-08-30 10:52:10.007353: train_loss -0.8558 
+2024-08-30 10:52:10.010361: val_loss -0.8589 
+2024-08-30 10:52:10.011275: Pseudo dice [0.9473, 0.962, 0.9418, 0.9268, 0.9331, 0.9368, 0.9345, 0.9339, 0.9359, 0.9444, 0.9562, 0.9696, 0.9589, 0.9211, 0.9083, 0.8948, 0.8692, 0.0, 0.9597, nan, nan] 
+2024-08-30 10:52:10.011992: Epoch time: 50.27 s 
+2024-08-30 10:52:11.381682:  
+2024-08-30 10:52:11.383344: Epoch 988 
+2024-08-30 10:52:11.384220: Current learning rate: 0.00019 
+2024-08-30 10:53:01.295525: train_loss -0.8603 
+2024-08-30 10:53:01.297323: val_loss -0.8651 
+2024-08-30 10:53:01.298688: Pseudo dice [0.9434, 0.9505, 0.9562, 0.9635, 0.9697, 0.9582, 0.9502, 0.9632, 0.9739, 0.975, 0.9757, 0.9708, 0.9752, 0.9574, 0.9289, 0.9098, 0.8922, 0.0, 0.94, nan, nan] 
+2024-08-30 10:53:01.299486: Epoch time: 49.92 s 
+2024-08-30 10:53:02.663652:  
+2024-08-30 10:53:02.664519: Epoch 989 
+2024-08-30 10:53:02.665169: Current learning rate: 0.00017 
+2024-08-30 10:53:57.241668: train_loss -0.8637 
+2024-08-30 10:53:57.243889: val_loss -0.8555 
+2024-08-30 10:53:57.245089: Pseudo dice [0.9463, 0.9575, 0.9632, 0.9629, 0.9615, 0.9643, 0.9715, 0.9745, 0.9738, 0.9749, 0.9727, 0.9755, 0.9715, 0.94, 0.885, 0.8641, 0.8483, 0.0, 0.9524, nan, nan] 
+2024-08-30 10:53:57.246208: Epoch time: 54.58 s 
+2024-08-30 10:53:58.841140:  
+2024-08-30 10:53:58.841934: Epoch 990 
+2024-08-30 10:53:58.845400: Current learning rate: 0.00016 
+2024-08-30 10:54:48.863036: train_loss -0.8539 
+2024-08-30 10:54:48.865016: val_loss -0.8523 
+2024-08-30 10:54:48.866725: Pseudo dice [0.9465, 0.9529, 0.9595, 0.9615, 0.9678, 0.969, 0.9484, 0.9364, 0.9373, 0.9403, 0.9349, 0.9211, 0.9456, 0.9531, 0.9398, 0.9109, 0.8844, 0.0, 0.96, nan, 0.0] 
+2024-08-30 10:54:48.867565: Epoch time: 50.02 s 
+2024-08-30 10:54:50.234685:  
+2024-08-30 10:54:50.235577: Epoch 991 
+2024-08-30 10:54:50.236278: Current learning rate: 0.00014 
+2024-08-30 10:55:42.558640: train_loss -0.8595 
+2024-08-30 10:55:42.563770: val_loss -0.873 
+2024-08-30 10:55:42.565078: Pseudo dice [0.9482, 0.9565, 0.945, 0.9437, 0.9534, 0.9617, 0.972, 0.9719, 0.9712, 0.9729, 0.9746, 0.9741, 0.9722, 0.9761, 0.9763, 0.9711, 0.9205, 0.0, 0.95, nan, nan] 
+2024-08-30 10:55:42.568004: Epoch time: 52.33 s 
+2024-08-30 10:55:43.959184:  
+2024-08-30 10:55:43.960096: Epoch 992 
+2024-08-30 10:55:43.960778: Current learning rate: 0.00013 
+2024-08-30 10:56:34.464707: train_loss -0.8517 
+2024-08-30 10:56:34.466847: val_loss -0.865 
+2024-08-30 10:56:34.468580: Pseudo dice [0.934, 0.9282, 0.9167, 0.9142, 0.9235, 0.9438, 0.9612, 0.9664, 0.9682, 0.9635, 0.9569, 0.9621, 0.9717, 0.9743, 0.9725, 0.9734, 0.9392, 0.0, 0.9582, nan, nan] 
+2024-08-30 10:56:34.469472: Epoch time: 50.51 s 
+2024-08-30 10:56:35.844016:  
+2024-08-30 10:56:35.844839: Epoch 993 
+2024-08-30 10:56:35.845395: Current learning rate: 0.00011 
+2024-08-30 10:57:27.177797: train_loss -0.8557 
+2024-08-30 10:57:27.179776: val_loss -0.8512 
+2024-08-30 10:57:27.180766: Pseudo dice [0.9189, 0.9003, 0.8902, 0.8881, 0.8969, 0.9066, 0.9233, 0.9348, 0.9333, 0.9261, 0.9133, 0.9102, 0.9065, 0.9142, 0.9102, 0.8891, 0.9107, nan, 0.9603, nan, nan] 
+2024-08-30 10:57:27.181525: Epoch time: 51.34 s 
+2024-08-30 10:57:28.516779:  
+2024-08-30 10:57:28.517621: Epoch 994 
+2024-08-30 10:57:28.518277: Current learning rate: 0.0001 
+2024-08-30 10:58:20.734412: train_loss -0.8588 
+2024-08-30 10:58:20.736292: val_loss -0.8618 
+2024-08-30 10:58:20.737302: Pseudo dice [0.9448, 0.9628, 0.9664, 0.9663, 0.9707, 0.9715, 0.9521, 0.933, 0.9258, 0.9169, 0.9189, 0.9543, 0.9757, 0.9759, 0.9676, 0.9625, 0.9047, 0.0, 0.9469, nan, nan] 
+2024-08-30 10:58:20.738145: Epoch time: 52.22 s 
+2024-08-30 10:58:22.078560:  
+2024-08-30 10:58:22.079548: Epoch 995 
+2024-08-30 10:58:22.080193: Current learning rate: 8e-05 
+2024-08-30 10:59:15.489184: train_loss -0.8588 
+2024-08-30 10:59:15.491495: val_loss -0.8563 
+2024-08-30 10:59:15.492667: Pseudo dice [0.9419, 0.9519, 0.9545, 0.9647, 0.9683, 0.9687, 0.9691, 0.967, 0.9676, 0.9707, 0.9644, 0.9673, 0.9713, 0.9637, 0.9456, 0.9309, 0.889, 0.0, 0.9452, nan, nan] 
+2024-08-30 10:59:15.496259: Epoch time: 53.41 s 
+2024-08-30 10:59:16.895959:  
+2024-08-30 10:59:16.896968: Epoch 996 
+2024-08-30 10:59:16.897671: Current learning rate: 7e-05 
+2024-08-30 11:00:08.924720: train_loss -0.8567 
+2024-08-30 11:00:08.926616: val_loss -0.8535 
+2024-08-30 11:00:08.927627: Pseudo dice [0.9473, 0.9595, 0.9598, 0.9639, 0.96, 0.963, 0.968, 0.9631, 0.9458, 0.9417, 0.943, 0.9533, 0.8951, 0.88, 0.8663, 0.8353, 0.7941, 0.0, 0.9559, nan, 0.0] 
+2024-08-30 11:00:08.928442: Epoch time: 52.03 s 
+2024-08-30 11:00:10.462140:  
+2024-08-30 11:00:10.463281: Epoch 997 
+2024-08-30 11:00:10.464050: Current learning rate: 5e-05 
+2024-08-30 11:01:02.995422: train_loss -0.8582 
+2024-08-30 11:01:02.997510: val_loss -0.8605 
+2024-08-30 11:01:02.998402: Pseudo dice [0.9504, 0.9636, 0.9663, 0.9672, 0.9691, 0.9594, 0.9465, 0.9246, 0.8972, 0.898, 0.9106, 0.9234, 0.9464, 0.9293, 0.914, 0.9057, 0.906, 0.0, 0.9568, nan, nan] 
+2024-08-30 11:01:02.999275: Epoch time: 52.54 s 
+2024-08-30 11:01:04.470241:  
+2024-08-30 11:01:04.471272: Epoch 998 
+2024-08-30 11:01:04.471981: Current learning rate: 4e-05 
+2024-08-30 11:01:56.748690: train_loss -0.8576 
+2024-08-30 11:01:56.750474: val_loss -0.8595 
+2024-08-30 11:01:56.751365: Pseudo dice [0.942, 0.9497, 0.9356, 0.9318, 0.944, 0.9526, 0.9614, 0.9519, 0.9392, 0.9468, 0.9599, 0.9453, 0.9154, 0.9521, 0.956, 0.9309, 0.8729, 0.0, 0.954, nan, 0.0] 
+2024-08-30 11:01:56.752067: Epoch time: 52.28 s 
+2024-08-30 11:01:58.278047:  
+2024-08-30 11:01:58.278900: Epoch 999 
+2024-08-30 11:01:58.279588: Current learning rate: 2e-05 
+2024-08-30 11:02:48.987234: train_loss -0.8578 
+2024-08-30 11:02:48.990110: val_loss -0.8588 
+2024-08-30 11:02:48.991422: Pseudo dice [0.9047, 0.8957, 0.8757, 0.8963, 0.9056, 0.9127, 0.9236, 0.9078, 0.9045, 0.9242, 0.9359, 0.9477, 0.9653, 0.9733, 0.9759, 0.9586, 0.9448, 0.0, 0.9615, nan, nan] 
+2024-08-30 11:02:48.992284: Epoch time: 50.71 s 
+2024-08-30 11:02:53.416574: Training done. 

nnUNetTrainer__nnUNetResEncUNetMPlans__3d_fullres/plans.json

@@ -0,0 +1,356 @@
+{
+    "dataset_name": "Dataset601_VertebralBodies",
+    "plans_name": "nnUNetResEncUNetMPlans",
+    "original_median_spacing_after_transp": [
+        1.5,
+        1.5,
+        1.5
+    ],
+    "original_median_shape_after_transp": [
+        239,
+        252,
+        252
+    ],
+    "image_reader_writer": "SimpleITKIO",
+    "transpose_forward": [
+        0,
+        1,
+        2
+    ],
+    "transpose_backward": [
+        0,
+        1,
+        2
+    ],
+    "configurations": {
+        "2d": {
+            "data_identifier": "nnUNetPlans_2d",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 48,
+            "patch_size": [
+                256,
+                256
+            ],
+            "median_image_size_in_voxels": [
+                240.0,
+                240.0
+            ],
+            "spacing": [
+                1.5,
+                1.5
+            ],
+            "normalization_schemes": [
+                "CTNormalization"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 7,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        512,
+                        512,
+                        512
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv2d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm2d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": true
+        },
+        "3d_fullres": {
+            "data_identifier": "nnUNetPlans_3d_fullres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                128,
+                128,
+                128
+            ],
+            "median_image_size_in_voxels": [
+                235.0,
+                240.0,
+                240.0
+            ],
+            "spacing": [
+                1.5,
+                1.5,
+                1.5
+            ],
+            "normalization_schemes": [
+                "CTNormalization"
+            ],
+            "use_mask_for_norm": [
+                false
+            ],
+            "resampling_fn_data": "resample_data_or_seg_to_shape",
+            "resampling_fn_seg": "resample_data_or_seg_to_shape",
+            "resampling_fn_data_kwargs": {
+                "is_seg": false,
+                "order": 3,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 6,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        320,
+                        320
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv3d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm3d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": false
+        }
+    },
+    "experiment_planner_used": "nnUNetPlannerResEncM",
+    "label_manager": "LabelManager",
+    "foreground_intensity_properties_per_channel": {
+        "0": {
+            "max": 10076.0,
+            "mean": 251.18613451837237,
+            "median": 216.0,
+            "min": -2048.0,
+            "percentile_00_5": -158.0,
+            "percentile_99_5": 1100.0,
+            "std": 205.1041738973246
+        }
+    }
+}