diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_best.pth b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_best.pth
index 647bac8a3a2d74c76dfd4cc0b6213d0f0a5ccc35..88da7c1fc0dc8aeae940f5b8f70b3d5483c72a3f 100644
--- a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_best.pth
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_best.pth
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec265a5e4069931beeff4d6c0eaa4d86a377a6715215022586346bc011b1fdaa
-size 819603011
+oid sha256:ddb8648534e1a8c3ab9dc584d43129ee611d10cb04b51f6f88ef15282e007bac
+size 819886531
diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_latest.pth b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_latest.pth
index 8e850faf265d76d3611126d4ebe3bd3d9ad32162..0be807343cf0e6bb044e854f371f732623b7f70a 100644
--- a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_latest.pth
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_latest.pth
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ebbc2a75943c78e20cff26bcbe296f2bd9d35e82d579fbe87b5cf793c0d2e1e
-size 819681067
+oid sha256:87ca9fb7760da47dc0f9f196b4f1012442b5eb0aa70cf8eddd1c0f3ff5812200
+size 820041515
diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json
index 529526b84da83e73f16c6dd698a2972fea151509..372bcb3fde1d31ec359fb06196eeb75b6867d10d 100644
--- a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json
@@ -1,17 +1,17 @@
 {
-    "_best_ema": "None",
+    "_best_ema": "0.44475228",
     "batch_size": "2",
     "configuration_manager": "{'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [128, 256, 192], 'median_image_size_in_voxels': [214.0, 352.0, 281.0], 'spacing': [0.70001220703125, 0.482421875, 0.482421875], '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}",
     "configuration_name": "3d_fullres",
     "cudnn_version": 91900,
-    "current_epoch": "0",
-    "dataloader_train": "<batchgenerators.dataloading.nondet_multi_threaded_augmenter.NonDetMultiThreadedAugmenter object at 0x7f8a521f9400>",
-    "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader.nnUNetDataLoader object at 0x7f8a521f92b0>",
+    "current_epoch": "150",
+    "dataloader_train": "<batchgenerators.dataloading.nondet_multi_threaded_augmenter.NonDetMultiThreadedAugmenter object at 0x7f02be189400>",
+    "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader.nnUNetDataLoader object at 0x7f02be1892b0>",
     "dataloader_train.generator.transforms": "ComposeTransforms( transforms = [SpatialTransform( patch_size = [128, 256, 192], patch_center_dist_from_border = [0, 0, 0], random_crop = False, p_elastic_deform = 0, elastic_deform_scale = (0, 0.2), elastic_deform_magnitude = (0, 0.2), p_rotation = 0.2, rotation = (-0.5235987755982988, 0.5235987755982988), p_rot_per_axis = 1, p_scaling = 0.2, scaling = (0.7, 1.4), p_synchronize_scaling_across_axes = 1, p_synchronize_def_scale_across_axes = 0, bg_style_seg_sampling = False, seg_tiebreak = 'nearest', mode_seg = 'bilinear', border_mode_seg = 'constant', center_deformation = True, mode_image = 'bilinear', padding_mode_image = 'zeros', padding_value_seg = -1, padding_value_image = 0, align_corners = False, _grid_cache = {} ), RandomTransform(p=0.1, transform=GaussianNoiseTransform( noise_variance = (0, 0.1), p_per_channel = 1, synchronize_channels = True )), RandomTransform(p=0.2, transform=GaussianBlurTransform( blur_sigma = (0.5, 1.0), benchmark = True, synchronize_channels = False, synchronize_axes = False, p_per_channel = 0.5, benchmark_use_fft = {}, benchmark_num_runs = 9 )), RandomTransform(p=0.15, transform=MultiplicativeBrightnessTransform( multiplier_range = BGContrast(contrast_range=(0.75, 1.25)), synchronize_channels = False, p_per_channel = 1 )), RandomTransform(p=0.15, transform=ContrastTransform( contrast_range = BGContrast(contrast_range=(0.75, 1.25)), preserve_range = True, synchronize_channels = False, p_per_channel = 1.0 )), RandomTransform(p=0.25, transform=SimulateLowResolutionTransform( scale = (0.5, 1), synchronize_channels = False, synchronize_axes = True, ignore_axes = None, allowed_channels = None, p_per_channel = 0.5, upmodes = {1: 'linear', 2: 'bilinear', 3: 'trilinear'} )), RandomTransform(p=0.1, transform=GammaTransform( gamma = BGContrast(contrast_range=(0.7, 1.5)), p_invert_image = 1.0, synchronize_channels = False, p_per_channel = 1.0, p_retain_stats = 1.0 )), RandomTransform(p=0.3, transform=GammaTransform( gamma = BGContrast(contrast_range=(0.7, 1.5)), p_invert_image = 0.0, synchronize_channels = False, p_per_channel = 1.0, p_retain_stats = 1.0 )), MirrorTransform( allowed_axes = (0, 1, 2) ), RemoveLabelTransform( segmentation_channels = None, label_value = -1, set_to = 0 ), DownsampleSegForDSTransform( ds_scales = [[np.float64(1.0), np.float64(1.0), np.float64(1.0)], [np.float64(0.5), np.float64(0.5), np.float64(0.5)], [np.float64(0.25), np.float64(0.25), np.float64(0.25)], [np.float64(0.125), np.float64(0.125), np.float64(0.125)], [np.float64(0.0625), np.float64(0.0625), np.float64(0.0625)]] )] )",
     "dataloader_train.num_processes": "8",
     "dataloader_train.transform": "None",
-    "dataloader_val": "<batchgenerators.dataloading.nondet_multi_threaded_augmenter.NonDetMultiThreadedAugmenter object at 0x7f8a521d65d0>",
-    "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader.nnUNetDataLoader object at 0x7f8a521d6490>",
+    "dataloader_val": "<batchgenerators.dataloading.nondet_multi_threaded_augmenter.NonDetMultiThreadedAugmenter object at 0x7f02be16a710>",
+    "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader.nnUNetDataLoader object at 0x7f02be16a5d0>",
     "dataloader_val.generator.transforms": "ComposeTransforms( transforms = [RemoveLabelTransform( segmentation_channels = None, label_value = -1, set_to = 0 ), DownsampleSegForDSTransform( ds_scales = [[np.float64(1.0), np.float64(1.0), np.float64(1.0)], [np.float64(0.5), np.float64(0.5), np.float64(0.5)], [np.float64(0.25), np.float64(0.25), np.float64(0.25)], [np.float64(0.125), np.float64(0.125), np.float64(0.125)], [np.float64(0.0625), np.float64(0.0625), np.float64(0.0625)]] )] )",
     "dataloader_val.num_processes": "4",
     "dataloader_val.transform": "None",
@@ -22,25 +22,25 @@
     "fold": "0",
     "folder_with_segs_from_previous_stage": "None",
     "gpu_name": "NVIDIA RTX PRO 6000 Blackwell Server Edition",
-    "grad_scaler": "<torch.amp.grad_scaler.GradScaler object at 0x7f8c586ade80>",
-    "hostname": "36b518f8-d155-4ff1-999a-434dd4465158-dg7s5",
+    "grad_scaler": "<torch.amp.grad_scaler.GradScaler object at 0x7f04c37f5e80>",
+    "hostname": "1751cc12-bbcf-4769-b784-aa824fd1e9c3-p7sh2",
     "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 0x7f8c586add30>",
+    "label_manager": "<nnunetv2.utilities.label_handling.label_handling.LabelManager object at 0x7f04c37f5d30>",
     "local_rank": "0",
-    "log_file": "/marimo/TopBrain2026/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_15_10_11_07.txt",
-    "logger": "<nnunetv2.training.logging.nnunet_logger.MetaLogger object at 0x7f8c586adfd0>",
+    "log_file": "/marimo/TopBrain2026/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_15_20_21_46.txt",
+    "logger": "<nnunetv2.training.logging.nnunet_logger.MetaLogger object at 0x7f04c37f5fd0>",
     "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 0x7f8c586afa10>",
+    "lr_scheduler": "<nnunetv2.training.lr_scheduler.polylr.PolyLRScheduler object at 0x7f04c37f7a10>",
     "my_init_kwargs": "{'plans': {'dataset_name': 'Dataset001_CT', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [0.70001220703125, 0.482421875, 0.482421875], 'original_median_shape_after_transp': [206, 365, 286], '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': 70, 'patch_size': [384, 320], 'median_image_size_in_voxels': [352.0, 281.0], 'spacing': [0.482421875, 0.482421875], '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, 256, 192], 'median_image_size_in_voxels': [214.0, 352.0, 281.0], 'spacing': [0.70001220703125, 0.482421875, 0.482421875], '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': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 1813.9854736328125, 'mean': 211.27015686035156, 'median': 192.95703125, 'min': -91.01630401611328, 'percentile_00_5': 66.00397491455078, 'percentile_99_5': 618.0081787109375, 'std': 107.36483764648438}}}, 'configuration': '3d_fullres', 'fold': 0, 'dataset_json': {'channel_names': {'0': 'CT'}, 'labels': {'background': 0, 'BA': 1, 'R-P1P2': 2, 'L-P1P2': 3, 'R-ICA-C6-C7': 4, 'R-M1': 5, 'L-ICA-C6-C7': 6, 'L-M1': 7, 'R-Pcom': 8, 'L-Pcom': 9, 'Acom': 10, 'R-A1A2': 11, 'L-A1A2': 12, 'R-A3': 13, 'L-A3': 14, '3rd-A2': 15, '3rd-A3': 16, 'R-M2': 17, 'R-M3': 18, 'L-M2': 19, 'L-M3': 20, 'R-P3P4': 21, 'L-P3P4': 22, 'R-VA': 23, 'L-VA': 24, 'R-SCA': 25, 'L-SCA': 26, 'R-AICA': 27, 'L-AICA': 28, 'R-PICA': 29, 'L-PICA': 30, 'R-AChA': 31, 'L-AChA': 32, 'R-OA': 33, 'L-OA': 34, 'R-ICA-C1-C5': 35, 'L-ICA-C1-C5': 36}, 'numTraining': 25, 'file_ending': '.nii.gz'}, 'device': device(type='cuda')}",
     "network": "ResidualEncoderUNet",
     "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: 0.01\n    maximize: False\n    momentum: 0.99\n    nesterov: True\n    weight_decay: 3e-05\n)",
+    "optimizer": "SGD (\nParameter Group 0\n    dampening: 0\n    differentiable: False\n    foreach: None\n    fused: None\n    initial_lr: 0.01\n    lr: 0.008648416647233837\n    maximize: False\n    momentum: 0.99\n    nesterov: True\n    weight_decay: 3e-05\n)",
     "output_folder": "/marimo/TopBrain2026/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0",
     "output_folder_base": "/marimo/TopBrain2026/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres",
     "oversample_foreground_percent": "0.33",
diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png
index e121c164daaa72811e05f0b7edf1e77fcf62ad32..e82519c442c01dcc61630231e57611599bef873e 100644
--- a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:908c4d7e293e3bb3d6191ee4890521f342070d3c6d80a0f2a8f664a4464c5fd8
-size 887460
+oid sha256:e56beee1d25bec4a88ceaf9bbff0ea2cb786ab33dcc0e6c9da62db3f32f9fe11
+size 1176887
diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_15_20_21_46.txt b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_15_20_21_46.txt
new file mode 100644
index 0000000000000000000000000000000000000000..db2b047470c111461b0ed65bf2ffe523a4f0f474
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_15_20_21_46.txt
@@ -0,0 +1,1085 @@
+
+#######################################################################
+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.
+#######################################################################
+ 
+2026-10-15 20:21:50.009508: do_dummy_2d_data_aug: False 
+2026-10-15 20:21:50.010777: Creating new 5-fold cross-validation split... 
+2026-10-15 20:21:50.044950: Desired fold for training: 0 
+2026-10-15 20:21:50.045110: This split has 20 training and 5 validation cases. 
+
+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, 256, 192], 'median_image_size_in_voxels': [214.0, 352.0, 281.0], 'spacing': [0.70001220703125, 0.482421875, 0.482421875], '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} 
+ 
+These are the global plan.json settings:
+ {'dataset_name': 'Dataset001_CT', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [0.70001220703125, 0.482421875, 0.482421875], 'original_median_shape_after_transp': [206, 365, 286], 'image_reader_writer': 'SimpleITKIO', 'transpose_forward': [0, 1, 2], 'transpose_backward': [0, 1, 2], 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 1813.9854736328125, 'mean': 211.27015686035156, 'median': 192.95703125, 'min': -91.01630401611328, 'percentile_00_5': 66.00397491455078, 'percentile_99_5': 618.0081787109375, 'std': 107.36483764648438}}} 
+ 
+2026-10-15 20:21:55.129702: Unable to plot network architecture: 
+2026-10-15 20:21:55.130102: No module named 'hiddenlayer' 
+2026-10-15 20:21:55.206727:  
+2026-10-15 20:21:55.207088: Epoch 150 
+2026-10-15 20:21:55.208083: Current learning rate: 0.00864 
+2026-10-15 20:26:32.214953: train_loss -0.6176 
+2026-10-15 20:26:32.215713: val_loss -0.4924 
+2026-10-15 20:26:32.216014: Pseudo dice [np.float32(0.8794), np.float32(0.5692), np.float32(0.6652), np.float32(0.6454), np.float32(0.6243), np.float32(0.6509), np.float32(0.6065), np.float32(0.0), np.float32(0.0), np.float32(0.2104), np.float32(0.5795), np.float32(0.4719), np.float32(0.6035), np.float32(0.3894), np.float32(0.0), np.float32(0.0), np.float32(0.5772), np.float32(0.4278), np.float32(0.5749), np.float32(0.4543), np.float32(0.5328), np.float32(0.6131), np.float32(0.7443), np.float32(0.7096), np.float32(0.4637), np.float32(0.3608), np.float32(0.0), np.float32(0.0), np.float32(0.0996), np.float32(0.2445), np.float32(0.0), np.float32(0.0), np.float32(0.3148), np.float32(0.4101), np.float32(0.6913), np.float32(0.6486)] 
+2026-10-15 20:26:32.216563: Epoch time: 277.01 s 
+2026-10-15 20:26:33.089919:  
+2026-10-15 20:26:33.090434: Epoch 151 
+2026-10-15 20:26:33.090868: Current learning rate: 0.00863 
+2026-10-15 20:30:11.877503: train_loss -0.6157 
+2026-10-15 20:30:11.878175: val_loss -0.383 
+2026-10-15 20:30:11.878867: Pseudo dice [np.float32(0.8606), np.float32(0.3992), np.float32(0.4289), np.float32(0.3143), np.float32(0.2503), np.float32(0.3843), np.float32(0.3768), np.float32(0.0), np.float32(0.0), np.float32(0.4304), np.float32(0.3418), np.float32(0.4253), np.float32(0.4735), np.float32(0.4413), np.float32(0.0), np.float32(0.0), np.float32(0.1939), np.float32(0.2779), np.float32(0.3498), np.float32(0.3153), np.float32(0.4211), np.float32(0.3834), np.float32(0.4784), np.float32(0.4179), np.float32(0.1636), np.float32(0.1579), np.float32(0.0), np.float32(0.0), np.float32(0.2519), np.float32(0.4141), np.float32(0.0), np.float32(0.0), np.float32(0.197), np.float32(0.3174), np.float32(0.3196), np.float32(0.3955)] 
+2026-10-15 20:30:11.879346: Epoch time: 218.79 s 
+2026-10-15 20:30:12.795329:  
+2026-10-15 20:30:12.795961: Epoch 152 
+2026-10-15 20:30:12.796344: Current learning rate: 0.00862 
+2026-10-15 20:33:44.884905: train_loss -0.5938 
+2026-10-15 20:33:44.885508: val_loss -0.3777 
+2026-10-15 20:33:44.886019: Pseudo dice [np.float32(0.8648), np.float32(0.4539), np.float32(0.4251), np.float32(0.4034), np.float32(0.3532), np.float32(0.4426), np.float32(0.3967), np.float32(0.0), np.float32(0.0), np.float32(0.3933), np.float32(0.3723), np.float32(0.404), np.float32(0.5449), np.float32(0.5064), np.float32(0.0), np.float32(0.0), np.float32(0.3178), np.float32(0.2762), np.float32(0.3534), np.float32(0.2828), np.float32(0.3785), np.float32(0.313), np.float32(0.3627), np.float32(0.2902), np.float32(0.27), np.float32(0.2169), np.float32(0.0), np.float32(0.0), np.float32(0.0803), np.float32(0.1484), np.float32(0.0), np.float32(0.0), np.float32(0.2353), np.float32(0.3455), np.float32(0.3646), np.float32(0.4107)] 
+2026-10-15 20:33:44.886569: Epoch time: 212.09 s 
+2026-10-15 20:33:45.671234:  
+2026-10-15 20:33:45.672122: Epoch 153 
+2026-10-15 20:33:45.672618: Current learning rate: 0.00861 
+2026-10-15 20:37:28.888096: train_loss -0.5944 
+2026-10-15 20:37:28.888610: val_loss -0.4975 
+2026-10-15 20:37:28.889553: Pseudo dice [np.float32(0.8435), np.float32(0.6321), np.float32(0.6562), np.float32(0.7028), np.float32(0.6316), np.float32(0.7166), np.float32(0.6222), np.float32(0.0), np.float32(0.0), np.float32(0.221), np.float32(0.5767), np.float32(0.5511), np.float32(0.554), np.float32(0.3898), np.float32(0.0), np.float32(0.0), np.float32(0.6552), np.float32(0.4224), np.float32(0.5945), np.float32(0.4915), np.float32(0.4591), np.float32(0.4549), np.float32(0.6486), np.float32(0.7095), np.float32(0.3471), np.float32(0.3229), np.float32(0.0), np.float32(0.0), np.float32(0.3483), np.float32(0.2649), np.float32(0.0), np.float32(0.0), np.float32(0.2724), np.float32(0.3541), np.float32(0.7279), np.float32(0.7091)] 
+2026-10-15 20:37:28.889905: Epoch time: 223.22 s 
+2026-10-15 20:37:29.825186:  
+2026-10-15 20:37:29.825838: Epoch 154 
+2026-10-15 20:37:29.826045: Current learning rate: 0.0086 
+2026-10-15 20:41:17.789315: train_loss -0.6137 
+2026-10-15 20:41:17.790052: val_loss -0.4866 
+2026-10-15 20:41:17.790905: Pseudo dice [np.float32(0.8368), np.float32(0.5231), np.float32(0.6768), np.float32(0.646), np.float32(0.6152), np.float32(0.7073), np.float32(0.6371), np.float32(0.0), np.float32(0.0), np.float32(0.3264), np.float32(0.5762), np.float32(0.521), np.float32(0.5559), np.float32(0.3873), np.float32(0.0), np.float32(0.0), np.float32(0.6197), np.float32(0.5156), np.float32(0.6898), np.float32(0.5758), np.float32(0.5284), np.float32(0.6173), np.float32(0.6269), np.float32(0.6022), np.float32(0.4448), np.float32(0.3493), np.float32(0.0), np.float32(0.0), np.float32(0.1139), np.float32(0.4621), np.float32(0.0), np.float32(0.0), np.float32(0.4195), np.float32(0.6443), np.float32(0.7097), np.float32(0.7382)] 
+2026-10-15 20:41:17.791664: Epoch time: 227.97 s 
+2026-10-15 20:41:18.665806:  
+2026-10-15 20:41:18.666466: Epoch 155 
+2026-10-15 20:41:18.666806: Current learning rate: 0.00859 
+2026-10-15 20:44:59.329580: train_loss -0.6455 
+2026-10-15 20:44:59.330507: val_loss -0.5234 
+2026-10-15 20:44:59.331163: Pseudo dice [np.float32(0.8696), np.float32(0.6245), np.float32(0.6917), np.float32(0.6717), np.float32(0.6311), np.float32(0.7076), np.float32(0.6321), np.float32(0.0), np.float32(0.0), np.float32(0.4896), np.float32(0.5423), np.float32(0.5463), np.float32(0.5297), np.float32(0.472), np.float32(0.0), np.float32(0.0), np.float32(0.6468), np.float32(0.5103), np.float32(0.6296), np.float32(0.5143), np.float32(0.5612), np.float32(0.5672), np.float32(0.7443), np.float32(0.719), np.float32(0.4875), np.float32(0.4813), np.float32(0.0), np.float32(0.0), np.float32(0.2616), np.float32(0.5676), np.float32(0.0), np.float32(0.0), np.float32(0.5918), np.float32(0.5965), np.float32(0.7207), np.float32(0.7392)] 
+2026-10-15 20:44:59.331656: Epoch time: 220.67 s 
+2026-10-15 20:45:00.175071:  
+2026-10-15 20:45:00.175674: Epoch 156 
+2026-10-15 20:45:00.176024: Current learning rate: 0.00858 
+2026-10-15 20:48:40.531237: train_loss -0.6285 
+2026-10-15 20:48:40.531844: val_loss -0.4871 
+2026-10-15 20:48:40.532475: Pseudo dice [np.float32(0.8723), np.float32(0.5467), np.float32(0.6612), np.float32(0.6261), np.float32(0.5783), np.float32(0.6871), np.float32(0.5725), np.float32(0.0), np.float32(0.0), np.float32(0.3492), np.float32(0.5558), np.float32(0.6057), np.float32(0.5009), np.float32(0.4173), np.float32(0.0), np.float32(0.0), np.float32(0.6143), np.float32(0.4444), np.float32(0.5924), np.float32(0.536), np.float32(0.5071), np.float32(0.5539), np.float32(0.7322), np.float32(0.741), np.float32(0.5451), np.float32(0.4165), np.float32(0.0), np.float32(0.0), np.float32(0.1225), np.float32(0.3701), np.float32(0.0), np.float32(0.0), np.float32(0.2379), np.float32(0.5093), np.float32(0.7037), np.float32(0.72)] 
+2026-10-15 20:48:40.532841: Epoch time: 220.36 s 
+2026-10-15 20:48:41.369691:  
+2026-10-15 20:48:41.370312: Epoch 157 
+2026-10-15 20:48:41.370658: Current learning rate: 0.00858 
+2026-10-15 20:52:19.494077: train_loss -0.6369 
+2026-10-15 20:52:19.494560: val_loss -0.4484 
+2026-10-15 20:52:19.495342: Pseudo dice [np.float32(0.8789), np.float32(0.4927), np.float32(0.5511), np.float32(0.5821), np.float32(0.4745), np.float32(0.6163), np.float32(0.4949), np.float32(0.0), np.float32(0.0), np.float32(0.2229), np.float32(0.4449), np.float32(0.516), np.float32(0.4476), np.float32(0.4177), np.float32(0.0), np.float32(0.0), np.float32(0.5421), np.float32(0.4351), np.float32(0.5758), np.float32(0.4926), np.float32(0.4715), np.float32(0.5163), np.float32(0.5883), np.float32(0.5167), np.float32(0.4442), np.float32(0.3883), np.float32(0.0), np.float32(0.0), np.float32(0.163), np.float32(0.5074), np.float32(0.0), np.float32(0.0), np.float32(0.4808), np.float32(0.5583), np.float32(0.5106), np.float32(0.5483)] 
+2026-10-15 20:52:19.495875: Epoch time: 218.13 s 
+2026-10-15 20:52:20.300190:  
+2026-10-15 20:52:20.300879: Epoch 158 
+2026-10-15 20:52:20.301259: Current learning rate: 0.00857 
+2026-10-15 20:55:36.105014: train_loss -0.6558 
+2026-10-15 20:55:36.105702: val_loss -0.5386 
+2026-10-15 20:55:36.106450: Pseudo dice [np.float32(0.8931), np.float32(0.6321), np.float32(0.7081), np.float32(0.7273), np.float32(0.6803), np.float32(0.754), np.float32(0.6718), np.float32(0.0), np.float32(0.0), np.float32(0.4511), np.float32(0.6294), np.float32(0.5847), np.float32(0.5281), np.float32(0.5632), np.float32(0.0), np.float32(0.0), np.float32(0.719), np.float32(0.5592), np.float32(0.6824), np.float32(0.556), np.float32(0.6224), np.float32(0.6246), np.float32(0.735), np.float32(0.7273), np.float32(0.513), np.float32(0.4271), np.float32(0.0), np.float32(0.0), np.float32(0.2251), np.float32(0.4334), np.float32(0.0), np.float32(0.0), np.float32(0.4192), np.float32(0.6123), np.float32(0.7361), np.float32(0.7705)] 
+2026-10-15 20:55:36.106885: Epoch time: 195.81 s 
+2026-10-15 20:55:36.964439:  
+2026-10-15 20:55:36.965212: Epoch 159 
+2026-10-15 20:55:36.965479: Current learning rate: 0.00856 
+2026-10-15 20:59:21.555902: train_loss -0.6477 
+2026-10-15 20:59:21.556807: val_loss -0.4556 
+2026-10-15 20:59:21.557342: Pseudo dice [np.float32(0.8875), np.float32(0.4772), np.float32(0.572), np.float32(0.5546), np.float32(0.5967), np.float32(0.6197), np.float32(0.6101), np.float32(0.0), np.float32(0.0), np.float32(0.3862), np.float32(0.4502), np.float32(0.5248), np.float32(0.4285), np.float32(0.4405), np.float32(0.0), np.float32(0.0), np.float32(0.6407), np.float32(0.5263), np.float32(0.5726), np.float32(0.4222), np.float32(0.4374), np.float32(0.5056), np.float32(0.5989), np.float32(0.588), np.float32(0.4464), np.float32(0.4418), np.float32(0.0), np.float32(0.0), np.float32(0.2328), np.float32(0.4211), np.float32(0.0), np.float32(0.0), np.float32(0.5125), np.float32(0.5006), np.float32(0.585), np.float32(0.6264)] 
+2026-10-15 20:59:21.557852: Epoch time: 224.59 s 
+2026-10-15 20:59:22.675364:  
+2026-10-15 20:59:22.675877: Epoch 160 
+2026-10-15 20:59:22.676202: Current learning rate: 0.00855 
+2026-10-15 21:03:15.603655: train_loss -0.6505 
+2026-10-15 21:03:15.604338: val_loss -0.4085 
+2026-10-15 21:03:15.605186: Pseudo dice [np.float32(0.8677), np.float32(0.4027), np.float32(0.4711), np.float32(0.4636), np.float32(0.5203), np.float32(0.521), np.float32(0.5054), np.float32(0.0), np.float32(0.0), np.float32(0.4), np.float32(0.3632), np.float32(0.4034), np.float32(0.3903), np.float32(0.2975), np.float32(0.0), np.float32(0.0), np.float32(0.6057), np.float32(0.4574), np.float32(0.5879), np.float32(0.4575), np.float32(0.3035), np.float32(0.444), np.float32(0.5653), np.float32(0.4765), np.float32(0.5249), np.float32(0.4713), np.float32(0.0), np.float32(0.0), np.float32(0.1084), np.float32(0.2093), np.float32(0.0), np.float32(0.0), np.float32(0.4951), np.float32(0.5239), np.float32(0.4944), np.float32(0.5173)] 
+2026-10-15 21:03:15.605721: Epoch time: 232.93 s 
+2026-10-15 21:03:16.505777:  
+2026-10-15 21:03:16.506488: Epoch 161 
+2026-10-15 21:03:16.506839: Current learning rate: 0.00854 
+2026-10-15 21:06:50.913783: train_loss -0.638 
+2026-10-15 21:06:50.914802: val_loss -0.3621 
+2026-10-15 21:06:50.915442: Pseudo dice [np.float32(0.8916), np.float32(0.3596), np.float32(0.3928), np.float32(0.298), np.float32(0.4121), np.float32(0.3544), np.float32(0.3411), np.float32(0.0), np.float32(0.0), np.float32(0.4052), np.float32(0.2784), np.float32(0.3588), np.float32(0.4164), np.float32(0.3488), np.float32(0.0), np.float32(0.0), np.float32(0.4707), np.float32(0.4196), np.float32(0.4325), np.float32(0.355), np.float32(0.3033), np.float32(0.3453), np.float32(0.3476), np.float32(0.1783), np.float32(0.3178), np.float32(0.337), np.float32(0.0), np.float32(0.0), np.float32(0.1164), np.float32(0.1629), np.float32(0.0), np.float32(0.0), np.float32(0.3305), np.float32(0.2929), np.float32(0.332), np.float32(0.3281)] 
+2026-10-15 21:06:50.915826: Epoch time: 214.41 s 
+2026-10-15 21:06:51.790059:  
+2026-10-15 21:06:51.790711: Epoch 162 
+2026-10-15 21:06:51.791170: Current learning rate: 0.00853 
+2026-10-15 21:10:30.244097: train_loss -0.6348 
+2026-10-15 21:10:30.244554: val_loss -0.4156 
+2026-10-15 21:10:30.244982: Pseudo dice [np.float32(0.8754), np.float32(0.4413), np.float32(0.5329), np.float32(0.4469), np.float32(0.4665), np.float32(0.5655), np.float32(0.4415), np.float32(0.0), np.float32(0.0), np.float32(0.4315), np.float32(0.3485), np.float32(0.4746), np.float32(0.3449), np.float32(0.3015), np.float32(0.0), np.float32(0.0), np.float32(0.5478), np.float32(0.3887), np.float32(0.5571), np.float32(0.417), np.float32(0.4652), np.float32(0.5128), np.float32(0.6121), np.float32(0.5154), np.float32(0.3875), np.float32(0.3256), np.float32(0.0), np.float32(0.0), np.float32(0.265), np.float32(0.5251), np.float32(0.0), np.float32(0.0), np.float32(0.4334), np.float32(0.4554), np.float32(0.4806), np.float32(0.5261)] 
+2026-10-15 21:10:30.245356: Epoch time: 218.46 s 
+2026-10-15 21:10:31.107819:  
+2026-10-15 21:10:31.108323: Epoch 163 
+2026-10-15 21:10:31.108676: Current learning rate: 0.00852 
+2026-10-15 21:14:19.230263: train_loss -0.6499 
+2026-10-15 21:14:19.231092: val_loss -0.479 
+2026-10-15 21:14:19.231854: Pseudo dice [np.float32(0.8703), np.float32(0.5276), np.float32(0.617), np.float32(0.6376), np.float32(0.5472), np.float32(0.6935), np.float32(0.6014), np.float32(0.0), np.float32(0.0), np.float32(0.3868), np.float32(0.5369), np.float32(0.5542), np.float32(0.5095), np.float32(0.4801), np.float32(0.0), np.float32(0.0), np.float32(0.4908), np.float32(0.3485), np.float32(0.5731), np.float32(0.393), np.float32(0.4737), np.float32(0.4736), np.float32(0.6571), np.float32(0.5548), np.float32(0.3781), np.float32(0.2548), np.float32(0.0), np.float32(0.0), np.float32(0.2816), np.float32(0.3358), np.float32(0.0), np.float32(0.0), np.float32(0.4884), np.float32(0.4952), np.float32(0.6097), np.float32(0.678)] 
+2026-10-15 21:14:19.232201: Epoch time: 228.12 s 
+2026-10-15 21:14:20.070452:  
+2026-10-15 21:14:20.071299: Epoch 164 
+2026-10-15 21:14:20.071654: Current learning rate: 0.00851 
+2026-10-15 21:17:53.266051: train_loss -0.678 
+2026-10-15 21:17:53.266768: val_loss -0.4861 
+2026-10-15 21:17:53.267158: Pseudo dice [np.float32(0.8892), np.float32(0.5264), np.float32(0.6074), np.float32(0.6229), np.float32(0.5519), np.float32(0.6619), np.float32(0.5734), np.float32(0.0), np.float32(0.0), np.float32(0.5098), np.float32(0.5264), np.float32(0.5505), np.float32(0.5106), np.float32(0.4602), np.float32(0.0), np.float32(0.0), np.float32(0.5855), np.float32(0.5111), np.float32(0.6283), np.float32(0.5435), np.float32(0.5274), np.float32(0.5427), np.float32(0.6838), np.float32(0.6214), np.float32(0.4305), np.float32(0.3822), np.float32(0.0), np.float32(0.0), np.float32(0.1569), np.float32(0.5434), np.float32(0.0), np.float32(0.0), np.float32(0.5345), np.float32(0.5975), np.float32(0.6106), np.float32(0.6403)] 
+2026-10-15 21:17:53.267628: Epoch time: 213.2 s 
+2026-10-15 21:17:54.087203:  
+2026-10-15 21:17:54.088055: Epoch 165 
+2026-10-15 21:17:54.088285: Current learning rate: 0.0085 
+2026-10-15 21:21:31.309762: train_loss -0.6827 
+2026-10-15 21:21:31.310306: val_loss -0.5181 
+2026-10-15 21:21:31.310904: Pseudo dice [np.float32(0.8858), np.float32(0.5724), np.float32(0.6809), np.float32(0.6634), np.float32(0.6079), np.float32(0.73), np.float32(0.6287), np.float32(0.0), np.float32(0.0), np.float32(0.5585), np.float32(0.5494), np.float32(0.5731), np.float32(0.5411), np.float32(0.4554), np.float32(0.0), np.float32(0.0), np.float32(0.6477), np.float32(0.4921), np.float32(0.642), np.float32(0.4817), np.float32(0.5366), np.float32(0.5899), np.float32(0.6384), np.float32(0.5866), np.float32(0.5108), np.float32(0.3973), np.float32(0.0), np.float32(0.0), np.float32(0.3448), np.float32(0.6458), np.float32(0.0), np.float32(0.0), np.float32(0.5386), np.float32(0.6225), np.float32(0.6767), np.float32(0.714)] 
+2026-10-15 21:21:31.311928: Epoch time: 217.23 s 
+2026-10-15 21:21:32.250334:  
+2026-10-15 21:21:32.251030: Epoch 166 
+2026-10-15 21:21:32.251235: Current learning rate: 0.00849 
+2026-10-15 21:25:29.796159: train_loss -0.6703 
+2026-10-15 21:25:29.796793: val_loss -0.4836 
+2026-10-15 21:25:29.797469: Pseudo dice [np.float32(0.8622), np.float32(0.5249), np.float32(0.5911), np.float32(0.5685), np.float32(0.501), np.float32(0.6291), np.float32(0.5157), np.float32(0.0), np.float32(0.0), np.float32(0.5582), np.float32(0.4899), np.float32(0.5481), np.float32(0.4706), np.float32(0.4881), np.float32(0.0), np.float32(0.0), np.float32(0.5274), np.float32(0.4216), np.float32(0.5939), np.float32(0.4995), np.float32(0.4743), np.float32(0.5167), np.float32(0.6303), np.float32(0.5908), np.float32(0.4002), np.float32(0.3523), np.float32(0.0), np.float32(0.0), np.float32(0.236), np.float32(0.3044), np.float32(0.0), np.float32(0.0), np.float32(0.4237), np.float32(0.6075), np.float32(0.5583), np.float32(0.619)] 
+2026-10-15 21:25:29.798212: Epoch time: 237.55 s 
+2026-10-15 21:25:30.623280:  
+2026-10-15 21:25:30.623971: Epoch 167 
+2026-10-15 21:25:30.624380: Current learning rate: 0.00848 
+2026-10-15 21:29:10.859401: train_loss -0.6729 
+2026-10-15 21:29:10.860350: val_loss -0.5599 
+2026-10-15 21:29:10.861074: Pseudo dice [np.float32(0.8871), np.float32(0.629), np.float32(0.7544), np.float32(0.7884), np.float32(0.7411), np.float32(0.8166), np.float32(0.7064), np.float32(0.0), np.float32(0.0), np.float32(0.2846), np.float32(0.6263), np.float32(0.6451), np.float32(0.5446), np.float32(0.5092), np.float32(0.0), np.float32(0.0), np.float32(0.7439), np.float32(0.5798), np.float32(0.732), np.float32(0.5978), np.float32(0.5777), np.float32(0.6611), np.float32(0.8014), np.float32(0.7113), np.float32(0.5534), np.float32(0.5105), np.float32(0.0), np.float32(0.0), np.float32(0.438), np.float32(0.524), np.float32(0.0), np.float32(0.0), np.float32(0.6492), np.float32(0.6619), np.float32(0.8243), np.float32(0.8218)] 
+2026-10-15 21:29:10.861599: Epoch time: 220.24 s 
+2026-10-15 21:29:11.687243:  
+2026-10-15 21:29:11.687885: Epoch 168 
+2026-10-15 21:29:11.688257: Current learning rate: 0.00847 
+2026-10-15 21:33:06.305250: train_loss -0.6655 
+2026-10-15 21:33:06.305735: val_loss -0.2897 
+2026-10-15 21:33:06.306388: Pseudo dice [np.float32(0.8765), np.float32(0.2245), np.float32(0.2128), np.float32(0.1772), np.float32(0.2166), np.float32(0.2587), np.float32(0.2179), np.float32(0.0), np.float32(0.0), np.float32(0.272), np.float32(0.1628), np.float32(0.223), np.float32(0.3839), np.float32(0.3108), np.float32(0.0), np.float32(0.0), np.float32(0.1474), np.float32(0.1634), np.float32(0.186), np.float32(0.1564), np.float32(0.159), np.float32(0.1394), np.float32(0.3723), np.float32(0.2554), np.float32(0.146), np.float32(0.1494), np.float32(0.0), np.float32(0.0), np.float32(0.1127), np.float32(0.2628), np.float32(0.0), np.float32(0.0), np.float32(0.1911), np.float32(0.3002), np.float32(0.2145), np.float32(0.2605)] 
+2026-10-15 21:33:06.306754: Epoch time: 234.62 s 
+2026-10-15 21:33:07.105986:  
+2026-10-15 21:33:07.106624: Epoch 169 
+2026-10-15 21:33:07.106987: Current learning rate: 0.00847 
+2026-10-15 21:36:40.293227: train_loss -0.6858 
+2026-10-15 21:36:40.294323: val_loss -0.4385 
+2026-10-15 21:36:40.295170: Pseudo dice [np.float32(0.8745), np.float32(0.5139), np.float32(0.5572), np.float32(0.5138), np.float32(0.4506), np.float32(0.5814), np.float32(0.5071), np.float32(0.0), np.float32(0.0), np.float32(0.2995), np.float32(0.4412), np.float32(0.483), np.float32(0.4981), np.float32(0.4565), np.float32(0.0), np.float32(0.0), np.float32(0.3755), np.float32(0.3531), np.float32(0.5015), np.float32(0.4199), np.float32(0.5014), np.float32(0.4337), np.float32(0.642), np.float32(0.5728), np.float32(0.3028), np.float32(0.302), np.float32(0.0), np.float32(0.0), np.float32(0.1016), np.float32(0.475), np.float32(0.0), np.float32(0.0), np.float32(0.4227), np.float32(0.5245), np.float32(0.5245), np.float32(0.5808)] 
+2026-10-15 21:36:40.295533: Epoch time: 213.19 s 
+2026-10-15 21:36:41.185253:  
+2026-10-15 21:36:41.185845: Epoch 170 
+2026-10-15 21:36:41.186188: Current learning rate: 0.00846 
+2026-10-15 21:40:19.628311: train_loss -0.6507 
+2026-10-15 21:40:19.629444: val_loss -0.2545 
+2026-10-15 21:40:19.630021: Pseudo dice [np.float32(0.8661), np.float32(0.2107), np.float32(0.2017), np.float32(0.2824), np.float32(0.1484), np.float32(0.2175), np.float32(0.1527), np.float32(0.0), np.float32(0.0), np.float32(0.4881), np.float32(0.1863), np.float32(0.1915), np.float32(0.4309), np.float32(0.2545), np.float32(0.0), np.float32(0.0), np.float32(0.1275), np.float32(0.1098), np.float32(0.116), np.float32(0.1224), np.float32(0.1262), np.float32(0.1342), np.float32(0.1865), np.float32(0.3211), np.float32(0.0951), np.float32(0.0994), np.float32(0.0), np.float32(0.0), np.float32(0.0048), np.float32(0.1563), np.float32(0.0), np.float32(0.0), np.float32(0.1067), np.float32(0.1741), np.float32(0.189), np.float32(0.1737)] 
+2026-10-15 21:40:19.630262: Epoch time: 218.45 s 
+2026-10-15 21:40:20.544052:  
+2026-10-15 21:40:20.544662: Epoch 171 
+2026-10-15 21:40:20.545275: Current learning rate: 0.00845 
+2026-10-15 21:43:50.633165: train_loss -0.6383 
+2026-10-15 21:43:50.633682: val_loss -0.4266 
+2026-10-15 21:43:50.634240: Pseudo dice [np.float32(0.8516), np.float32(0.5127), np.float32(0.503), np.float32(0.4789), np.float32(0.4334), np.float32(0.5515), np.float32(0.4371), np.float32(0.0), np.float32(0.0), np.float32(0.5347), np.float32(0.4433), np.float32(0.4483), np.float32(0.508), np.float32(0.5248), np.float32(0.0), np.float32(0.0), np.float32(0.4002), np.float32(0.2716), np.float32(0.3689), np.float32(0.2833), np.float32(0.3811), np.float32(0.3974), np.float32(0.5667), np.float32(0.5204), np.float32(0.2586), np.float32(0.2551), np.float32(0.0), np.float32(0.0), np.float32(0.0271), np.float32(0.3478), np.float32(0.0), np.float32(0.0), np.float32(0.3761), np.float32(0.5014), np.float32(0.5085), np.float32(0.5422)] 
+2026-10-15 21:43:50.635304: Epoch time: 210.09 s 
+2026-10-15 21:43:51.555779:  
+2026-10-15 21:43:51.556444: Epoch 172 
+2026-10-15 21:43:51.556792: Current learning rate: 0.00844 
+2026-10-15 21:47:28.843901: train_loss -0.6651 
+2026-10-15 21:47:28.844664: val_loss -0.5405 
+2026-10-15 21:47:28.845024: Pseudo dice [np.float32(0.9015), np.float32(0.6421), np.float32(0.7252), np.float32(0.7294), np.float32(0.6635), np.float32(0.7701), np.float32(0.6578), np.float32(0.0), np.float32(0.0), np.float32(0.3691), np.float32(0.5894), np.float32(0.6224), np.float32(0.4654), np.float32(0.3979), np.float32(0.0), np.float32(0.0), np.float32(0.6642), np.float32(0.5495), np.float32(0.6863), np.float32(0.5189), np.float32(0.5341), np.float32(0.6177), np.float32(0.807), np.float32(0.7681), np.float32(0.4959), np.float32(0.4727), np.float32(0.0), np.float32(0.0), np.float32(0.3371), np.float32(0.546), np.float32(0.0), np.float32(0.0), np.float32(0.5289), np.float32(0.667), np.float32(0.7582), np.float32(0.7769)] 
+2026-10-15 21:47:28.845418: Epoch time: 217.29 s 
+2026-10-15 21:47:29.758345:  
+2026-10-15 21:47:29.759177: Epoch 173 
+2026-10-15 21:47:29.759555: Current learning rate: 0.00843 
+2026-10-15 21:50:57.936112: train_loss -0.6591 
+2026-10-15 21:50:57.936774: val_loss -0.5506 
+2026-10-15 21:50:57.937269: Pseudo dice [np.float32(0.8847), np.float32(0.6352), np.float32(0.764), np.float32(0.7627), np.float32(0.7076), np.float32(0.7885), np.float32(0.6804), np.float32(0.0), np.float32(0.0), np.float32(0.4585), np.float32(0.6023), np.float32(0.6324), np.float32(0.5411), np.float32(0.4836), np.float32(0.0), np.float32(0.0), np.float32(0.7271), np.float32(0.5442), np.float32(0.692), np.float32(0.5648), np.float32(0.5586), np.float32(0.6201), np.float32(0.7801), np.float32(0.7491), np.float32(0.5154), np.float32(0.489), np.float32(0.0), np.float32(0.0), np.float32(0.31), np.float32(0.3971), np.float32(0.0), np.float32(0.0), np.float32(0.6736), np.float32(0.7039), np.float32(0.8017), np.float32(0.8063)] 
+2026-10-15 21:50:57.937817: Epoch time: 208.18 s 
+2026-10-15 21:50:58.784833:  
+2026-10-15 21:50:58.785423: Epoch 174 
+2026-10-15 21:50:58.785751: Current learning rate: 0.00842 
+2026-10-15 21:54:52.391681: train_loss -0.6718 
+2026-10-15 21:54:52.392528: val_loss -0.5451 
+2026-10-15 21:54:52.393286: Pseudo dice [np.float32(0.8943), np.float32(0.64), np.float32(0.7142), np.float32(0.7496), np.float32(0.6582), np.float32(0.8023), np.float32(0.6531), np.float32(0.0), np.float32(0.0), np.float32(0.5303), np.float32(0.6529), np.float32(0.6404), np.float32(0.5188), np.float32(0.4526), np.float32(0.0), np.float32(0.0), np.float32(0.6879), np.float32(0.5561), np.float32(0.6527), np.float32(0.5704), np.float32(0.6063), np.float32(0.6253), np.float32(0.8026), np.float32(0.7723), np.float32(0.5661), np.float32(0.4134), np.float32(0.0), np.float32(0.0), np.float32(0.313), np.float32(0.5415), np.float32(0.0), np.float32(0.0), np.float32(0.4911), np.float32(0.5387), np.float32(0.7905), np.float32(0.7986)] 
+2026-10-15 21:54:52.393707: Epoch time: 233.61 s 
+2026-10-15 21:54:53.308384:  
+2026-10-15 21:54:53.309164: Epoch 175 
+2026-10-15 21:54:53.309450: Current learning rate: 0.00841 
+2026-10-15 21:58:30.333216: train_loss -0.6833 
+2026-10-15 21:58:30.334002: val_loss -0.5181 
+2026-10-15 21:58:30.334445: Pseudo dice [np.float32(0.8844), np.float32(0.6121), np.float32(0.653), np.float32(0.6634), np.float32(0.5918), np.float32(0.7124), np.float32(0.571), np.float32(0.0), np.float32(0.0), np.float32(0.5339), np.float32(0.5777), np.float32(0.5683), np.float32(0.5517), np.float32(0.5114), np.float32(0.0), np.float32(0.0), np.float32(0.6219), np.float32(0.5104), np.float32(0.6092), np.float32(0.5329), np.float32(0.5536), np.float32(0.5416), np.float32(0.7186), np.float32(0.7068), np.float32(0.4471), np.float32(0.5002), np.float32(0.0), np.float32(0.0), np.float32(0.1312), np.float32(0.5582), np.float32(0.0), np.float32(0.0), np.float32(0.544), np.float32(0.6676), np.float32(0.7046), np.float32(0.7417)] 
+2026-10-15 21:58:30.335098: Epoch time: 217.03 s 
+2026-10-15 21:58:31.226075:  
+2026-10-15 21:58:31.226692: Epoch 176 
+2026-10-15 21:58:31.227036: Current learning rate: 0.0084 
+2026-10-15 22:02:10.379271: train_loss -0.6819 
+2026-10-15 22:02:10.380152: val_loss -0.5032 
+2026-10-15 22:02:10.380582: Pseudo dice [np.float32(0.8708), np.float32(0.5607), np.float32(0.6684), np.float32(0.6697), np.float32(0.5873), np.float32(0.7289), np.float32(0.6178), np.float32(0.0), np.float32(0.0), np.float32(0.3877), np.float32(0.5404), np.float32(0.5086), np.float32(0.4702), np.float32(0.4533), np.float32(0.0), np.float32(0.0), np.float32(0.6154), np.float32(0.4981), np.float32(0.6589), np.float32(0.5191), np.float32(0.515), np.float32(0.5726), np.float32(0.6862), np.float32(0.6664), np.float32(0.4827), np.float32(0.4508), np.float32(0.0), np.float32(0.0), np.float32(0.3951), np.float32(0.6095), np.float32(0.0), np.float32(0.0), np.float32(0.6454), np.float32(0.6925), np.float32(0.7144), np.float32(0.7518)] 
+2026-10-15 22:02:10.381013: Epoch time: 219.16 s 
+2026-10-15 22:02:11.574675:  
+2026-10-15 22:02:11.575303: Epoch 177 
+2026-10-15 22:02:11.575788: Current learning rate: 0.00839 
+2026-10-15 22:05:48.817606: train_loss -0.6659 
+2026-10-15 22:05:48.818132: val_loss -0.43 
+2026-10-15 22:05:48.819031: Pseudo dice [np.float32(0.8793), np.float32(0.4683), np.float32(0.5295), np.float32(0.5082), np.float32(0.4735), np.float32(0.4894), np.float32(0.4662), np.float32(0.0), np.float32(0.0), np.float32(0.2177), np.float32(0.3695), np.float32(0.3629), np.float32(0.3705), np.float32(0.3358), np.float32(0.0), np.float32(0.0), np.float32(0.4778), np.float32(0.389), np.float32(0.4533), np.float32(0.327), np.float32(0.3518), np.float32(0.4162), np.float32(0.6676), np.float32(0.5955), np.float32(0.3487), np.float32(0.2613), np.float32(0.0), np.float32(0.0), np.float32(0.3357), np.float32(0.3804), np.float32(0.0), np.float32(0.0), np.float32(0.3982), np.float32(0.3143), np.float32(0.5264), np.float32(0.5309)] 
+2026-10-15 22:05:48.819278: Epoch time: 217.25 s 
+2026-10-15 22:05:49.682479:  
+2026-10-15 22:05:49.683112: Epoch 178 
+2026-10-15 22:05:49.683449: Current learning rate: 0.00838 
+2026-10-15 22:09:27.920772: train_loss -0.6661 
+2026-10-15 22:09:27.921253: val_loss -0.5508 
+2026-10-15 22:09:27.921818: Pseudo dice [np.float32(0.9033), np.float32(0.6494), np.float32(0.7137), np.float32(0.7265), np.float32(0.6441), np.float32(0.7602), np.float32(0.6401), np.float32(0.0), np.float32(0.0), np.float32(0.4985), np.float32(0.6262), np.float32(0.6034), np.float32(0.5316), np.float32(0.5106), np.float32(0.0), np.float32(0.0), np.float32(0.6828), np.float32(0.5234), np.float32(0.6692), np.float32(0.4955), np.float32(0.5902), np.float32(0.5882), np.float32(0.6781), np.float32(0.6237), np.float32(0.4857), np.float32(0.4453), np.float32(0.0), np.float32(0.0), np.float32(0.4717), np.float32(0.4845), np.float32(0.0), np.float32(0.0), np.float32(0.4384), np.float32(0.6944), np.float32(0.732), np.float32(0.7566)] 
+2026-10-15 22:09:27.922767: Epoch time: 218.24 s 
+2026-10-15 22:09:28.789633:  
+2026-10-15 22:09:28.790361: Epoch 179 
+2026-10-15 22:09:28.790580: Current learning rate: 0.00837 
+2026-10-15 22:13:05.949159: train_loss -0.6554 
+2026-10-15 22:13:05.950453: val_loss -0.2789 
+2026-10-15 22:13:05.951179: Pseudo dice [np.float32(0.8645), np.float32(0.1939), np.float32(0.1644), np.float32(0.1465), np.float32(0.0968), np.float32(0.1956), np.float32(0.1571), np.float32(0.0), np.float32(0.0), np.float32(0.5052), np.float32(0.1636), np.float32(0.2864), np.float32(0.4009), np.float32(0.2287), np.float32(0.0), np.float32(0.0), np.float32(0.1404), np.float32(0.0583), np.float32(0.172), np.float32(0.1704), np.float32(0.217), np.float32(0.1335), np.float32(0.2771), np.float32(0.122), np.float32(0.1335), np.float32(0.0583), np.float32(0.0), np.float32(0.0), np.float32(0.0842), np.float32(0.1488), np.float32(0.0), np.float32(0.0), np.float32(0.1016), np.float32(0.1828), np.float32(0.1267), np.float32(0.1276)] 
+2026-10-15 22:13:05.951773: Epoch time: 217.16 s 
+2026-10-15 22:13:06.773097:  
+2026-10-15 22:13:06.773894: Epoch 180 
+2026-10-15 22:13:06.774228: Current learning rate: 0.00836 
+2026-10-15 22:16:28.487344: train_loss -0.6101 
+2026-10-15 22:16:28.487888: val_loss -0.2669 
+2026-10-15 22:16:28.488346: Pseudo dice [np.float32(0.8578), np.float32(0.2934), np.float32(0.1822), np.float32(0.188), np.float32(0.1892), np.float32(0.1414), np.float32(0.159), np.float32(0.0), np.float32(0.0), np.float32(0.549), np.float32(0.1588), np.float32(0.1949), np.float32(0.3297), np.float32(0.2531), np.float32(0.0), np.float32(0.0), np.float32(0.1733), np.float32(0.1057), np.float32(0.1495), np.float32(0.116), np.float32(0.1345), np.float32(0.0987), np.float32(0.3843), np.float32(0.2592), np.float32(0.2695), np.float32(0.1189), np.float32(0.0), np.float32(0.0), np.float32(0.016), np.float32(0.1986), np.float32(0.0), np.float32(0.0), np.float32(0.1078), np.float32(0.1407), np.float32(0.147), np.float32(0.1681)] 
+2026-10-15 22:16:28.488908: Epoch time: 201.72 s 
+2026-10-15 22:16:29.395586:  
+2026-10-15 22:16:29.396239: Epoch 181 
+2026-10-15 22:16:29.396439: Current learning rate: 0.00836 
+2026-10-15 22:20:27.176245: train_loss -0.6326 
+2026-10-15 22:20:27.177095: val_loss -0.5155 
+2026-10-15 22:20:27.177952: Pseudo dice [np.float32(0.8947), np.float32(0.6078), np.float32(0.6949), np.float32(0.7044), np.float32(0.6681), np.float32(0.7187), np.float32(0.6494), np.float32(0.0), np.float32(0.0), np.float32(0.5607), np.float32(0.5496), np.float32(0.5612), np.float32(0.5083), np.float32(0.4593), np.float32(0.0), np.float32(0.0), np.float32(0.6631), np.float32(0.524), np.float32(0.6071), np.float32(0.4918), np.float32(0.5378), np.float32(0.5995), np.float32(0.7281), np.float32(0.6751), np.float32(0.51), np.float32(0.3868), np.float32(0.0), np.float32(0.0), np.float32(0.3281), np.float32(0.5876), np.float32(0.0), np.float32(0.0), np.float32(0.408), np.float32(0.5674), np.float32(0.7281), np.float32(0.7428)] 
+2026-10-15 22:20:27.178399: Epoch time: 237.78 s 
+2026-10-15 22:20:28.096251:  
+2026-10-15 22:20:28.096992: Epoch 182 
+2026-10-15 22:20:28.097347: Current learning rate: 0.00835 
+2026-10-15 22:24:07.113622: train_loss -0.668 
+2026-10-15 22:24:07.114152: val_loss -0.37 
+2026-10-15 22:24:07.114776: Pseudo dice [np.float32(0.8854), np.float32(0.3891), np.float32(0.3589), np.float32(0.3184), np.float32(0.2794), np.float32(0.3627), np.float32(0.318), np.float32(0.0), np.float32(0.0), np.float32(0.4597), np.float32(0.2781), np.float32(0.2762), np.float32(0.4358), np.float32(0.3612), np.float32(0.0), np.float32(0.0), np.float32(0.2964), np.float32(0.2819), np.float32(0.3019), np.float32(0.2205), np.float32(0.3304), np.float32(0.2934), np.float32(0.498), np.float32(0.4702), np.float32(0.2074), np.float32(0.2319), np.float32(0.0), np.float32(0.0), np.float32(0.3299), np.float32(0.3683), np.float32(0.0), np.float32(0.0), np.float32(0.2611), np.float32(0.3367), np.float32(0.3528), np.float32(0.4059)] 
+2026-10-15 22:24:07.115583: Epoch time: 219.02 s 
+2026-10-15 22:24:08.017253:  
+2026-10-15 22:24:08.017941: Epoch 183 
+2026-10-15 22:24:08.018310: Current learning rate: 0.00834 
+2026-10-15 22:27:50.436228: train_loss -0.6547 
+2026-10-15 22:27:50.436731: val_loss -0.4209 
+2026-10-15 22:27:50.437657: Pseudo dice [np.float32(0.8766), np.float32(0.4946), np.float32(0.4628), np.float32(0.4065), np.float32(0.4857), np.float32(0.4698), np.float32(0.4795), np.float32(0.0), np.float32(0.0), np.float32(0.4967), np.float32(0.3655), np.float32(0.352), np.float32(0.328), np.float32(0.3338), np.float32(0.0), np.float32(0.0), np.float32(0.4459), np.float32(0.3769), np.float32(0.399), np.float32(0.3374), np.float32(0.4122), np.float32(0.4139), np.float32(0.6197), np.float32(0.5183), np.float32(0.2814), np.float32(0.2893), np.float32(0.0), np.float32(0.0), np.float32(0.0676), np.float32(0.0728), np.float32(0.0), np.float32(0.0), np.float32(0.3121), np.float32(0.4192), np.float32(0.4913), np.float32(0.5063)] 
+2026-10-15 22:27:50.438136: Epoch time: 222.42 s 
+2026-10-15 22:27:51.284384:  
+2026-10-15 22:27:51.285142: Epoch 184 
+2026-10-15 22:27:51.285564: Current learning rate: 0.00833 
+2026-10-15 22:31:42.360451: train_loss -0.6604 
+2026-10-15 22:31:42.361365: val_loss -0.6119 
+2026-10-15 22:31:42.361910: Pseudo dice [np.float32(0.889), np.float32(0.7165), np.float32(0.8254), np.float32(0.8545), np.float32(0.8152), np.float32(0.8754), np.float32(0.7534), np.float32(0.0), np.float32(0.0), np.float32(0.5601), np.float32(0.6831), np.float32(0.7293), np.float32(0.5971), np.float32(0.564), np.float32(0.0), np.float32(0.0), np.float32(0.7838), np.float32(0.5919), np.float32(0.7431), np.float32(0.6187), np.float32(0.6526), np.float32(0.6703), np.float32(0.8714), np.float32(0.8486), np.float32(0.5265), np.float32(0.4374), np.float32(0.0), np.float32(0.0), np.float32(0.4153), np.float32(0.5816), np.float32(0.0), np.float32(0.0), np.float32(0.6053), np.float32(0.7014), np.float32(0.8896), np.float32(0.8754)] 
+2026-10-15 22:31:42.362323: Epoch time: 231.08 s 
+2026-10-15 22:31:43.210842:  
+2026-10-15 22:31:43.211546: Epoch 185 
+2026-10-15 22:31:43.212013: Current learning rate: 0.00832 
+2026-10-15 22:35:40.602397: train_loss -0.6659 
+2026-10-15 22:35:40.603357: val_loss -0.4766 
+2026-10-15 22:35:40.603935: Pseudo dice [np.float32(0.8724), np.float32(0.4973), np.float32(0.5446), np.float32(0.5114), np.float32(0.3947), np.float32(0.5824), np.float32(0.5332), np.float32(0.0), np.float32(0.0), np.float32(0.5133), np.float32(0.4604), np.float32(0.4976), np.float32(0.4455), np.float32(0.4336), np.float32(0.0), np.float32(0.0), np.float32(0.4146), np.float32(0.2749), np.float32(0.541), np.float32(0.4151), np.float32(0.4593), np.float32(0.4361), np.float32(0.7445), np.float32(0.6863), np.float32(0.3191), np.float32(0.26), np.float32(0.0), np.float32(0.0), np.float32(0.4901), np.float32(0.5647), np.float32(0.0), np.float32(0.0), np.float32(0.429), np.float32(0.5524), np.float32(0.6127), np.float32(0.6717)] 
+2026-10-15 22:35:40.604410: Epoch time: 237.39 s 
+2026-10-15 22:35:41.434569:  
+2026-10-15 22:35:41.435364: Epoch 186 
+2026-10-15 22:35:41.435703: Current learning rate: 0.00831 
+2026-10-15 22:39:19.761578: train_loss -0.6827 
+2026-10-15 22:39:19.762361: val_loss -0.5568 
+2026-10-15 22:39:19.762995: Pseudo dice [np.float32(0.8804), np.float32(0.6405), np.float32(0.7393), np.float32(0.7576), np.float32(0.6907), np.float32(0.7995), np.float32(0.7008), np.float32(0.0), np.float32(0.0), np.float32(0.4758), np.float32(0.5954), np.float32(0.6342), np.float32(0.5314), np.float32(0.4925), np.float32(0.0), np.float32(0.0), np.float32(0.7074), np.float32(0.5612), np.float32(0.6861), np.float32(0.5933), np.float32(0.5711), np.float32(0.6502), np.float32(0.8365), np.float32(0.8042), np.float32(0.487), np.float32(0.4613), np.float32(0.0), np.float32(0.0), np.float32(0.4396), np.float32(0.6026), np.float32(0.0), np.float32(0.0), np.float32(0.516), np.float32(0.6554), np.float32(0.8148), np.float32(0.8315)] 
+2026-10-15 22:39:19.763558: Epoch time: 218.33 s 
+2026-10-15 22:39:20.644390:  
+2026-10-15 22:39:20.645117: Epoch 187 
+2026-10-15 22:39:20.645424: Current learning rate: 0.0083 
+2026-10-15 22:43:02.710765: train_loss -0.6726 
+2026-10-15 22:43:02.711798: val_loss -0.2971 
+2026-10-15 22:43:02.712599: Pseudo dice [np.float32(0.8568), np.float32(0.3113), np.float32(0.2441), np.float32(0.1732), np.float32(0.2458), np.float32(0.2617), np.float32(0.2373), np.float32(0.0), np.float32(0.0), np.float32(0.5063), np.float32(0.1795), np.float32(0.2586), np.float32(0.3696), np.float32(0.3523), np.float32(0.0), np.float32(0.0), np.float32(0.2173), np.float32(0.1946), np.float32(0.1983), np.float32(0.1592), np.float32(0.1575), np.float32(0.1429), np.float32(0.2582), np.float32(0.2366), np.float32(0.1299), np.float32(0.1948), np.float32(0.0), np.float32(0.0), np.float32(0.0603), np.float32(0.0892), np.float32(0.0), np.float32(0.0), np.float32(0.22), np.float32(0.1934), np.float32(0.21), np.float32(0.2335)] 
+2026-10-15 22:43:02.713112: Epoch time: 222.07 s 
+2026-10-15 22:43:03.694199:  
+2026-10-15 22:43:03.694865: Epoch 188 
+2026-10-15 22:43:03.695258: Current learning rate: 0.00829 
+2026-10-15 22:46:52.651265: train_loss -0.6779 
+2026-10-15 22:46:52.652383: val_loss -0.4209 
+2026-10-15 22:46:52.652989: Pseudo dice [np.float32(0.8762), np.float32(0.487), np.float32(0.4943), np.float32(0.4269), np.float32(0.3535), np.float32(0.4803), np.float32(0.4112), np.float32(0.0), np.float32(0.0), np.float32(0.526), np.float32(0.3806), np.float32(0.4311), np.float32(0.4706), np.float32(0.4621), np.float32(0.0), np.float32(0.0), np.float32(0.3223), np.float32(0.2536), np.float32(0.3726), np.float32(0.2744), np.float32(0.3857), np.float32(0.3137), np.float32(0.4976), np.float32(0.3595), np.float32(0.1962), np.float32(0.2373), np.float32(0.0), np.float32(0.0), np.float32(0.2336), np.float32(0.335), np.float32(0.0), np.float32(0.0), np.float32(0.2986), np.float32(0.4114), np.float32(0.4525), np.float32(0.5187)] 
+2026-10-15 22:46:52.653543: Epoch time: 228.96 s 
+2026-10-15 22:46:53.553581:  
+2026-10-15 22:46:53.554318: Epoch 189 
+2026-10-15 22:46:53.554633: Current learning rate: 0.00828 
+2026-10-15 22:50:28.878043: train_loss -0.6838 
+2026-10-15 22:50:28.878635: val_loss -0.4731 
+2026-10-15 22:50:28.879437: Pseudo dice [np.float32(0.8691), np.float32(0.5171), np.float32(0.5566), np.float32(0.5346), np.float32(0.456), np.float32(0.582), np.float32(0.5076), np.float32(0.0), np.float32(0.0), np.float32(0.529), np.float32(0.4583), np.float32(0.4347), np.float32(0.5469), np.float32(0.5285), np.float32(0.0), np.float32(0.0), np.float32(0.3595), np.float32(0.2563), np.float32(0.4675), np.float32(0.3821), np.float32(0.4524), np.float32(0.3864), np.float32(0.6129), np.float32(0.5023), np.float32(0.2743), np.float32(0.2916), np.float32(0.0), np.float32(0.0), np.float32(0.3389), np.float32(0.5917), np.float32(0.0), np.float32(0.0), np.float32(0.3772), np.float32(0.52), np.float32(0.5607), np.float32(0.6065)] 
+2026-10-15 22:50:28.879895: Epoch time: 215.33 s 
+2026-10-15 22:50:29.737268:  
+2026-10-15 22:50:29.738003: Epoch 190 
+2026-10-15 22:50:29.738389: Current learning rate: 0.00827 
+2026-10-15 22:54:03.315979: train_loss -0.6716 
+2026-10-15 22:54:03.317125: val_loss -0.3895 
+2026-10-15 22:54:03.317625: Pseudo dice [np.float32(0.8813), np.float32(0.4892), np.float32(0.4753), np.float32(0.3853), np.float32(0.422), np.float32(0.4289), np.float32(0.3692), np.float32(0.0), np.float32(0.0), np.float32(0.5203), np.float32(0.2625), np.float32(0.4405), np.float32(0.4402), np.float32(0.3809), np.float32(0.0), np.float32(0.0), np.float32(0.4442), np.float32(0.3755), np.float32(0.3276), np.float32(0.2228), np.float32(0.3948), np.float32(0.3078), np.float32(0.542), np.float32(0.4151), np.float32(0.287), np.float32(0.2695), np.float32(0.0), np.float32(0.0), np.float32(0.0712), np.float32(0.349), np.float32(0.0), np.float32(0.0), np.float32(0.435), np.float32(0.3677), np.float32(0.4479), np.float32(0.472)] 
+2026-10-15 22:54:03.318068: Epoch time: 213.58 s 
+2026-10-15 22:54:04.214272:  
+2026-10-15 22:54:04.215006: Epoch 191 
+2026-10-15 22:54:04.215221: Current learning rate: 0.00826 
+2026-10-15 22:57:39.944777: train_loss -0.659 
+2026-10-15 22:57:39.945718: val_loss -0.4632 
+2026-10-15 22:57:39.946159: Pseudo dice [np.float32(0.9009), np.float32(0.5484), np.float32(0.54), np.float32(0.5023), np.float32(0.4884), np.float32(0.6077), np.float32(0.4923), np.float32(0.0), np.float32(0.0), np.float32(0.4851), np.float32(0.4509), np.float32(0.4771), np.float32(0.4896), np.float32(0.4624), np.float32(0.0), np.float32(0.0), np.float32(0.3707), np.float32(0.3252), np.float32(0.4389), np.float32(0.3556), np.float32(0.4834), np.float32(0.371), np.float32(0.7158), np.float32(0.5885), np.float32(0.2668), np.float32(0.2757), np.float32(0.0), np.float32(0.0), np.float32(0.1316), np.float32(0.5267), np.float32(0.0), np.float32(0.0), np.float32(0.4354), np.float32(0.4934), np.float32(0.6194), np.float32(0.6619)] 
+2026-10-15 22:57:39.946615: Epoch time: 215.73 s 
+2026-10-15 22:57:40.818842:  
+2026-10-15 22:57:40.819659: Epoch 192 
+2026-10-15 22:57:40.820014: Current learning rate: 0.00825 
+2026-10-15 23:01:23.613080: train_loss -0.6811 
+2026-10-15 23:01:23.614214: val_loss -0.4314 
+2026-10-15 23:01:23.614898: Pseudo dice [np.float32(0.8855), np.float32(0.5003), np.float32(0.485), np.float32(0.4723), np.float32(0.3645), np.float32(0.5247), np.float32(0.4495), np.float32(0.0), np.float32(0.0), np.float32(0.5388), np.float32(0.4273), np.float32(0.4461), np.float32(0.5051), np.float32(0.4093), np.float32(0.0), np.float32(0.0), np.float32(0.3379), np.float32(0.3038), np.float32(0.4164), np.float32(0.3586), np.float32(0.4437), np.float32(0.3848), np.float32(0.5985), np.float32(0.5576), np.float32(0.231), np.float32(0.2297), np.float32(0.0), np.float32(0.0), np.float32(0.3597), np.float32(0.5381), np.float32(0.0), np.float32(0.0), np.float32(0.23), np.float32(0.3847), np.float32(0.5277), np.float32(0.5764)] 
+2026-10-15 23:01:23.615410: Epoch time: 222.8 s 
+2026-10-15 23:01:24.533961:  
+2026-10-15 23:01:24.534685: Epoch 193 
+2026-10-15 23:01:24.535035: Current learning rate: 0.00824 
+2026-10-15 23:05:13.339517: train_loss -0.6794 
+2026-10-15 23:05:13.340316: val_loss -0.3905 
+2026-10-15 23:05:13.340874: Pseudo dice [np.float32(0.8856), np.float32(0.415), np.float32(0.3976), np.float32(0.3633), np.float32(0.3514), np.float32(0.4203), np.float32(0.3806), np.float32(0.0), np.float32(0.0), np.float32(0.6004), np.float32(0.3043), np.float32(0.3555), np.float32(0.3698), np.float32(0.3941), np.float32(0.0), np.float32(0.0), np.float32(0.3646), np.float32(0.2871), np.float32(0.3976), np.float32(0.2856), np.float32(0.3169), np.float32(0.3), np.float32(0.5852), np.float32(0.4968), np.float32(0.2085), np.float32(0.269), np.float32(0.0), np.float32(0.0), np.float32(0.1909), np.float32(0.3663), np.float32(0.0), np.float32(0.0), np.float32(0.3343), np.float32(0.4453), np.float32(0.4109), np.float32(0.4376)] 
+2026-10-15 23:05:13.341321: Epoch time: 228.81 s 
+2026-10-15 23:05:14.280496:  
+2026-10-15 23:05:14.281227: Epoch 194 
+2026-10-15 23:05:14.281552: Current learning rate: 0.00824 
+2026-10-15 23:09:03.902767: train_loss -0.6901 
+2026-10-15 23:09:03.903902: val_loss -0.4878 
+2026-10-15 23:09:03.904454: Pseudo dice [np.float32(0.8843), np.float32(0.5652), np.float32(0.5921), np.float32(0.6127), np.float32(0.486), np.float32(0.6687), np.float32(0.5521), np.float32(0.0), np.float32(0.0), np.float32(0.2742), np.float32(0.5382), np.float32(0.575), np.float32(0.5729), np.float32(0.5568), np.float32(0.0), np.float32(0.0), np.float32(0.5045), np.float32(0.4473), np.float32(0.5162), np.float32(0.4368), np.float32(0.5314), np.float32(0.4653), np.float32(0.642), np.float32(0.6314), np.float32(0.2984), np.float32(0.3542), np.float32(0.0), np.float32(0.0), np.float32(0.3694), np.float32(0.668), np.float32(0.0), np.float32(0.0), np.float32(0.5266), np.float32(0.5767), np.float32(0.6337), np.float32(0.6702)] 
+2026-10-15 23:09:03.904747: Epoch time: 229.62 s 
+2026-10-15 23:09:04.773144:  
+2026-10-15 23:09:04.774083: Epoch 195 
+2026-10-15 23:09:04.774562: Current learning rate: 0.00823 
+2026-10-15 23:12:43.652663: train_loss -0.6836 
+2026-10-15 23:12:43.653520: val_loss -0.4443 
+2026-10-15 23:12:43.654054: Pseudo dice [np.float32(0.8659), np.float32(0.5076), np.float32(0.5256), np.float32(0.5138), np.float32(0.6018), np.float32(0.5646), np.float32(0.5406), np.float32(0.0), np.float32(0.0), np.float32(0.4731), np.float32(0.3961), np.float32(0.4579), np.float32(0.421), np.float32(0.4007), np.float32(0.0), np.float32(0.0), np.float32(0.5744), np.float32(0.4603), np.float32(0.5157), np.float32(0.3747), np.float32(0.3968), np.float32(0.4612), np.float32(0.654), np.float32(0.5812), np.float32(0.4609), np.float32(0.445), np.float32(0.0), np.float32(0.0), np.float32(0.2461), np.float32(0.2743), np.float32(0.0), np.float32(0.0), np.float32(0.3531), np.float32(0.3145), np.float32(0.5728), np.float32(0.6034)] 
+2026-10-15 23:12:43.654524: Epoch time: 218.88 s 
+2026-10-15 23:12:44.554662:  
+2026-10-15 23:12:44.555496: Epoch 196 
+2026-10-15 23:12:44.555892: Current learning rate: 0.00822 
+2026-10-15 23:16:50.951006: train_loss -0.6592 
+2026-10-15 23:16:50.952378: val_loss -0.5081 
+2026-10-15 23:16:50.952869: Pseudo dice [np.float32(0.8873), np.float32(0.5855), np.float32(0.7225), np.float32(0.6816), np.float32(0.6073), np.float32(0.7398), np.float32(0.6283), np.float32(0.0), np.float32(0.0), np.float32(0.4106), np.float32(0.5722), np.float32(0.5917), np.float32(0.4944), np.float32(0.4407), np.float32(0.0), np.float32(0.0), np.float32(0.6477), np.float32(0.491), np.float32(0.5846), np.float32(0.5346), np.float32(0.5448), np.float32(0.5546), np.float32(0.7771), np.float32(0.7214), np.float32(0.5553), np.float32(0.38), np.float32(0.0), np.float32(0.0), np.float32(0.1249), np.float32(0.3418), np.float32(0.0), np.float32(0.0), np.float32(0.4314), np.float32(0.5473), np.float32(0.7273), np.float32(0.7458)] 
+2026-10-15 23:16:50.953222: Epoch time: 246.4 s 
+2026-10-15 23:16:52.122999:  
+2026-10-15 23:16:52.123843: Epoch 197 
+2026-10-15 23:16:52.124188: Current learning rate: 0.00821 
+2026-10-15 23:20:49.208151: train_loss -0.68 
+2026-10-15 23:20:49.209895: val_loss -0.508 
+2026-10-15 23:20:49.210471: Pseudo dice [np.float32(0.8726), np.float32(0.5013), np.float32(0.667), np.float32(0.6869), np.float32(0.6337), np.float32(0.7123), np.float32(0.658), np.float32(0.0), np.float32(0.0), np.float32(0.6262), np.float32(0.5411), np.float32(0.5436), np.float32(0.4963), np.float32(0.4027), np.float32(0.0), np.float32(0.0), np.float32(0.678), np.float32(0.4888), np.float32(0.681), np.float32(0.5172), np.float32(0.4954), np.float32(0.5597), np.float32(0.7397), np.float32(0.7267), np.float32(0.5363), np.float32(0.4291), np.float32(0.0), np.float32(0.0), np.float32(0.2317), np.float32(0.5969), np.float32(0.0), np.float32(0.0), np.float32(0.5853), np.float32(0.6133), np.float32(0.6756), np.float32(0.7215)] 
+2026-10-15 23:20:49.210984: Epoch time: 237.09 s 
+2026-10-15 23:20:50.036488:  
+2026-10-15 23:20:50.037219: Epoch 198 
+2026-10-15 23:20:50.037542: Current learning rate: 0.0082 
+2026-10-15 23:24:32.289107: train_loss -0.6867 
+2026-10-15 23:24:32.289942: val_loss -0.5178 
+2026-10-15 23:24:32.290475: Pseudo dice [np.float32(0.8893), np.float32(0.5861), np.float32(0.6446), np.float32(0.6575), np.float32(0.554), np.float32(0.7033), np.float32(0.5873), np.float32(0.0), np.float32(0.0), np.float32(0.5765), np.float32(0.5469), np.float32(0.5582), np.float32(0.5683), np.float32(0.5447), np.float32(0.0), np.float32(0.0), np.float32(0.5395), np.float32(0.4207), np.float32(0.5836), np.float32(0.5164), np.float32(0.5352), np.float32(0.4986), np.float32(0.673), np.float32(0.6474), np.float32(0.3753), np.float32(0.3625), np.float32(0.0), np.float32(0.0), np.float32(0.4448), np.float32(0.6351), np.float32(0.0), np.float32(0.0), np.float32(0.4078), np.float32(0.5948), np.float32(0.6565), np.float32(0.693)] 
+2026-10-15 23:24:32.290997: Epoch time: 222.25 s 
+2026-10-15 23:24:33.213296:  
+2026-10-15 23:24:33.213922: Epoch 199 
+2026-10-15 23:24:33.214201: Current learning rate: 0.00819 
+2026-10-15 23:28:21.404001: train_loss -0.6841 
+2026-10-15 23:28:21.404979: val_loss -0.4882 
+2026-10-15 23:28:21.405590: Pseudo dice [np.float32(0.8724), np.float32(0.5493), np.float32(0.5671), np.float32(0.5574), np.float32(0.4605), np.float32(0.6062), np.float32(0.5113), np.float32(0.0), np.float32(0.0), np.float32(0.5075), np.float32(0.4844), np.float32(0.5342), np.float32(0.5073), np.float32(0.4794), np.float32(0.0), np.float32(0.0), np.float32(0.4483), np.float32(0.3466), np.float32(0.5461), np.float32(0.4552), np.float32(0.5308), np.float32(0.5081), np.float32(0.6605), np.float32(0.5631), np.float32(0.2755), np.float32(0.3002), np.float32(0.0), np.float32(0.0), np.float32(0.3381), np.float32(0.5099), np.float32(0.0), np.float32(0.0), np.float32(0.376), np.float32(0.4832), np.float32(0.5673), np.float32(0.6094)] 
+2026-10-15 23:28:21.406096: Epoch time: 228.19 s 
+2026-10-15 23:28:23.890305:  
+2026-10-15 23:28:23.907226: Epoch 200 
+2026-10-15 23:28:23.907869: Current learning rate: 0.00818 
+2026-10-15 23:32:10.857079: train_loss -0.6894 
+2026-10-15 23:32:10.857942: val_loss -0.4416 
+2026-10-15 23:32:10.858656: Pseudo dice [np.float32(0.8802), np.float32(0.4877), np.float32(0.4989), np.float32(0.5268), np.float32(0.4779), np.float32(0.5664), np.float32(0.5027), np.float32(0.0), np.float32(0.0), np.float32(0.3896), np.float32(0.4217), np.float32(0.467), np.float32(0.4869), np.float32(0.4103), np.float32(0.0), np.float32(0.0), np.float32(0.4583), np.float32(0.3211), np.float32(0.4651), np.float32(0.3608), np.float32(0.4704), np.float32(0.4171), np.float32(0.5872), np.float32(0.493), np.float32(0.2648), np.float32(0.2956), np.float32(0.0), np.float32(0.0), np.float32(0.2396), np.float32(0.3761), np.float32(0.0), np.float32(0.0), np.float32(0.4204), np.float32(0.5163), np.float32(0.5477), np.float32(0.5728)] 
+2026-10-15 23:32:10.859181: Epoch time: 226.97 s 
+2026-10-15 23:32:11.722595:  
+2026-10-15 23:32:11.723258: Epoch 201 
+2026-10-15 23:32:11.723621: Current learning rate: 0.00817 
+2026-10-15 23:35:55.561038: train_loss -0.696 
+2026-10-15 23:35:55.562248: val_loss -0.495 
+2026-10-15 23:35:55.562919: Pseudo dice [np.float32(0.88), np.float32(0.5652), np.float32(0.5828), np.float32(0.61), np.float32(0.498), np.float32(0.6448), np.float32(0.5304), np.float32(0.0), np.float32(0.0), np.float32(0.5729), np.float32(0.5102), np.float32(0.5585), np.float32(0.5476), np.float32(0.5042), np.float32(0.0), np.float32(0.0), np.float32(0.4344), np.float32(0.3469), np.float32(0.5397), np.float32(0.4561), np.float32(0.499), np.float32(0.4687), np.float32(0.7073), np.float32(0.6422), np.float32(0.3366), np.float32(0.3045), np.float32(0.0), np.float32(0.0), np.float32(0.2979), np.float32(0.5686), np.float32(0.0), np.float32(0.0), np.float32(0.4895), np.float32(0.5619), np.float32(0.6331), np.float32(0.6693)] 
+2026-10-15 23:35:55.563324: Epoch time: 223.84 s 
+2026-10-15 23:35:56.464192:  
+2026-10-15 23:35:56.464961: Epoch 202 
+2026-10-15 23:35:56.465318: Current learning rate: 0.00816 
+2026-10-15 23:39:37.605042: train_loss -0.7016 
+2026-10-15 23:39:37.605585: val_loss -0.4467 
+2026-10-15 23:39:37.606057: Pseudo dice [np.float32(0.8815), np.float32(0.5282), np.float32(0.5117), np.float32(0.4982), np.float32(0.4601), np.float32(0.5656), np.float32(0.4782), np.float32(0.0), np.float32(0.0), np.float32(0.4982), np.float32(0.3948), np.float32(0.4483), np.float32(0.5101), np.float32(0.4973), np.float32(0.0), np.float32(0.0), np.float32(0.4012), np.float32(0.3411), np.float32(0.4265), np.float32(0.3851), np.float32(0.4765), np.float32(0.392), np.float32(0.7513), np.float32(0.6668), np.float32(0.238), np.float32(0.2773), np.float32(0.0), np.float32(0.0), np.float32(0.3188), np.float32(0.6798), np.float32(0.0), np.float32(0.0), np.float32(0.4456), np.float32(0.4791), np.float32(0.5682), np.float32(0.6223)] 
+2026-10-15 23:39:37.606750: Epoch time: 221.14 s 
+2026-10-15 23:39:38.520495:  
+2026-10-15 23:39:38.521343: Epoch 203 
+2026-10-15 23:39:38.521706: Current learning rate: 0.00815 
+2026-10-15 23:43:27.763731: train_loss -0.6902 
+2026-10-15 23:43:27.764643: val_loss -0.4754 
+2026-10-15 23:43:27.765501: Pseudo dice [np.float32(0.8723), np.float32(0.5123), np.float32(0.5743), np.float32(0.6255), np.float32(0.5418), np.float32(0.6589), np.float32(0.5604), np.float32(0.0), np.float32(0.0), np.float32(0.4915), np.float32(0.5031), np.float32(0.5414), np.float32(0.4601), np.float32(0.4311), np.float32(0.0), np.float32(0.0), np.float32(0.5218), np.float32(0.4379), np.float32(0.5697), np.float32(0.4442), np.float32(0.436), np.float32(0.4094), np.float32(0.5605), np.float32(0.4069), np.float32(0.29), np.float32(0.2415), np.float32(0.0), np.float32(0.0), np.float32(0.2162), np.float32(0.5415), np.float32(0.0), np.float32(0.0), np.float32(0.4758), np.float32(0.5472), np.float32(0.5754), np.float32(0.6025)] 
+2026-10-15 23:43:27.765998: Epoch time: 229.25 s 
+2026-10-15 23:43:28.660586:  
+2026-10-15 23:43:28.661161: Epoch 204 
+2026-10-15 23:43:28.661506: Current learning rate: 0.00814 
+2026-10-15 23:46:33.046749: train_loss -0.6804 
+2026-10-15 23:46:33.047757: val_loss -0.4728 
+2026-10-15 23:46:33.048359: Pseudo dice [np.float32(0.8644), np.float32(0.5127), np.float32(0.567), np.float32(0.5868), np.float32(0.4488), np.float32(0.6319), np.float32(0.4885), np.float32(0.0), np.float32(0.0), np.float32(0.5137), np.float32(0.5066), np.float32(0.55), np.float32(0.4668), np.float32(0.4888), np.float32(0.0), np.float32(0.0), np.float32(0.4549), np.float32(0.4211), np.float32(0.4635), np.float32(0.4285), np.float32(0.5186), np.float32(0.4472), np.float32(0.5881), np.float32(0.5658), np.float32(0.268), np.float32(0.2758), np.float32(0.0), np.float32(0.0), np.float32(0.2105), np.float32(0.4887), np.float32(0.0), np.float32(0.0), np.float32(0.5105), np.float32(0.583), np.float32(0.576), np.float32(0.6238)] 
+2026-10-15 23:46:33.048843: Epoch time: 184.39 s 
+2026-10-15 23:46:33.877187:  
+2026-10-15 23:46:33.877941: Epoch 205 
+2026-10-15 23:46:33.878415: Current learning rate: 0.00813 
+2026-10-15 23:50:00.051935: train_loss -0.6849 
+2026-10-15 23:50:00.052860: val_loss -0.5734 
+2026-10-15 23:50:00.053291: Pseudo dice [np.float32(0.9048), np.float32(0.6457), np.float32(0.7411), np.float32(0.7902), np.float32(0.6856), np.float32(0.8105), np.float32(0.6897), np.float32(0.0), np.float32(0.0), np.float32(0.5321), np.float32(0.6376), np.float32(0.625), np.float32(0.5928), np.float32(0.526), np.float32(0.0), np.float32(0.0), np.float32(0.6923), np.float32(0.5559), np.float32(0.6606), np.float32(0.4922), np.float32(0.5894), np.float32(0.6016), np.float32(0.8084), np.float32(0.7861), np.float32(0.4477), np.float32(0.5243), np.float32(0.0), np.float32(0.0), np.float32(0.3968), np.float32(0.5939), np.float32(0.0), np.float32(0.0), np.float32(0.5766), np.float32(0.4949), np.float32(0.7953), np.float32(0.8129)] 
+2026-10-15 23:50:00.053807: Epoch time: 206.18 s 
+2026-10-15 23:50:00.942886:  
+2026-10-15 23:50:00.943901: Epoch 206 
+2026-10-15 23:50:00.944414: Current learning rate: 0.00813 
+2026-10-15 23:53:43.096306: train_loss -0.661 
+2026-10-15 23:53:43.097134: val_loss -0.5508 
+2026-10-15 23:53:43.097821: Pseudo dice [np.float32(0.8544), np.float32(0.6316), np.float32(0.7261), np.float32(0.7694), np.float32(0.6757), np.float32(0.8078), np.float32(0.6831), np.float32(0.0), np.float32(0.0), np.float32(0.5105), np.float32(0.6636), np.float32(0.6801), np.float32(0.5685), np.float32(0.5607), np.float32(0.0), np.float32(0.0), np.float32(0.7148), np.float32(0.5675), np.float32(0.7237), np.float32(0.607), np.float32(0.5937), np.float32(0.5935), np.float32(0.7068), np.float32(0.7097), np.float32(0.5059), np.float32(0.5157), np.float32(0.0), np.float32(0.0), np.float32(0.1427), np.float32(0.5569), np.float32(0.0), np.float32(0.0), np.float32(0.6278), np.float32(0.6975), np.float32(0.7793), np.float32(0.795)] 
+2026-10-15 23:53:43.098159: Epoch time: 222.16 s 
+2026-10-15 23:53:43.969354:  
+2026-10-15 23:53:43.970457: Epoch 207 
+2026-10-15 23:53:43.970930: Current learning rate: 0.00812 
+2026-10-15 23:57:17.510639: train_loss -0.638 
+2026-10-15 23:57:17.511454: val_loss -0.5416 
+2026-10-15 23:57:17.512243: Pseudo dice [np.float32(0.8832), np.float32(0.6282), np.float32(0.7047), np.float32(0.7068), np.float32(0.672), np.float32(0.795), np.float32(0.6788), np.float32(0.0), np.float32(0.0), np.float32(0.5418), np.float32(0.6466), np.float32(0.6591), np.float32(0.5499), np.float32(0.4808), np.float32(0.0), np.float32(0.0), np.float32(0.6838), np.float32(0.5193), np.float32(0.6416), np.float32(0.5369), np.float32(0.5076), np.float32(0.5406), np.float32(0.7488), np.float32(0.7573), np.float32(0.5102), np.float32(0.4745), np.float32(0.0), np.float32(0.0), np.float32(0.34), np.float32(0.4964), np.float32(0.0), np.float32(0.0), np.float32(0.5669), np.float32(0.6618), np.float32(0.729), np.float32(0.7525)] 
+2026-10-15 23:57:17.512791: Epoch time: 213.54 s 
+2026-10-15 23:57:18.333204:  
+2026-10-15 23:57:18.334059: Epoch 208 
+2026-10-15 23:57:18.334418: Current learning rate: 0.00811 
+2026-10-16 00:01:07.136338: train_loss -0.6576 
+2026-10-16 00:01:07.137321: val_loss -0.5218 
+2026-10-16 00:01:07.137866: Pseudo dice [np.float32(0.8722), np.float32(0.6007), np.float32(0.6615), np.float32(0.6575), np.float32(0.6115), np.float32(0.7177), np.float32(0.6242), np.float32(0.0), np.float32(0.0), np.float32(0.4531), np.float32(0.5517), np.float32(0.6147), np.float32(0.5529), np.float32(0.4914), np.float32(0.0), np.float32(0.0), np.float32(0.5901), np.float32(0.4715), np.float32(0.6324), np.float32(0.5366), np.float32(0.5481), np.float32(0.5071), np.float32(0.632), np.float32(0.6394), np.float32(0.4354), np.float32(0.467), np.float32(0.0), np.float32(0.0), np.float32(0.3134), np.float32(0.6435), np.float32(0.0), np.float32(0.0), np.float32(0.5803), np.float32(0.5687), np.float32(0.657), np.float32(0.7076)] 
+2026-10-16 00:01:07.138380: Epoch time: 228.81 s 
+2026-10-16 00:01:07.921203:  
+2026-10-16 00:01:07.921964: Epoch 209 
+2026-10-16 00:01:07.922404: Current learning rate: 0.0081 
+2026-10-16 00:05:00.541996: train_loss -0.6624 
+2026-10-16 00:05:00.542925: val_loss -0.5322 
+2026-10-16 00:05:00.543590: Pseudo dice [np.float32(0.8818), np.float32(0.6115), np.float32(0.7178), np.float32(0.7472), np.float32(0.7015), np.float32(0.7671), np.float32(0.7037), np.float32(0.0), np.float32(0.0), np.float32(0.3681), np.float32(0.6079), np.float32(0.6409), np.float32(0.5454), np.float32(0.5122), np.float32(0.0), np.float32(0.0), np.float32(0.6956), np.float32(0.5101), np.float32(0.6915), np.float32(0.4973), np.float32(0.5157), np.float32(0.5758), np.float32(0.7539), np.float32(0.7436), np.float32(0.4936), np.float32(0.4618), np.float32(0.0), np.float32(0.0), np.float32(0.3085), np.float32(0.6268), np.float32(0.0), np.float32(0.0), np.float32(0.5271), np.float32(0.62), np.float32(0.756), np.float32(0.7752)] 
+2026-10-16 00:05:00.544107: Epoch time: 232.62 s 
+2026-10-16 00:05:01.410558:  
+2026-10-16 00:05:01.411002: Epoch 210 
+2026-10-16 00:05:01.411354: Current learning rate: 0.00809 
+2026-10-16 00:09:02.243983: train_loss -0.6898 
+2026-10-16 00:09:02.244859: val_loss -0.4577 
+2026-10-16 00:09:02.245564: Pseudo dice [np.float32(0.8746), np.float32(0.4999), np.float32(0.545), np.float32(0.554), np.float32(0.4555), np.float32(0.5987), np.float32(0.5075), np.float32(0.0), np.float32(0.0), np.float32(0.4698), np.float32(0.4455), np.float32(0.4494), np.float32(0.4974), np.float32(0.4677), np.float32(0.0), np.float32(0.0), np.float32(0.493), np.float32(0.4429), np.float32(0.5396), np.float32(0.4521), np.float32(0.4722), np.float32(0.4418), np.float32(0.5443), np.float32(0.4584), np.float32(0.2946), np.float32(0.3137), np.float32(0.0), np.float32(0.0), np.float32(0.2069), np.float32(0.3783), np.float32(0.0), np.float32(0.0), np.float32(0.4683), np.float32(0.558), np.float32(0.537), np.float32(0.595)] 
+2026-10-16 00:09:02.246010: Epoch time: 240.84 s 
+2026-10-16 00:09:03.129668:  
+2026-10-16 00:09:03.130475: Epoch 211 
+2026-10-16 00:09:03.130919: Current learning rate: 0.00808 
+2026-10-16 00:12:36.909064: train_loss -0.6849 
+2026-10-16 00:12:36.909644: val_loss -0.4787 
+2026-10-16 00:12:36.910270: Pseudo dice [np.float32(0.8955), np.float32(0.5326), np.float32(0.5827), np.float32(0.6019), np.float32(0.4993), np.float32(0.6389), np.float32(0.5253), np.float32(0.0), np.float32(0.0), np.float32(0.432), np.float32(0.4916), np.float32(0.5511), np.float32(0.4267), np.float32(0.3939), np.float32(0.0), np.float32(0.0), np.float32(0.5104), np.float32(0.3832), np.float32(0.5721), np.float32(0.482), np.float32(0.4875), np.float32(0.5184), np.float32(0.7041), np.float32(0.6373), np.float32(0.4252), np.float32(0.3387), np.float32(0.0), np.float32(0.0), np.float32(0.3622), np.float32(0.5325), np.float32(0.0), np.float32(0.0), np.float32(0.4581), np.float32(0.4789), np.float32(0.6128), np.float32(0.6592)] 
+2026-10-16 00:12:36.911000: Epoch time: 213.78 s 
+2026-10-16 00:12:37.826095:  
+2026-10-16 00:12:37.827184: Epoch 212 
+2026-10-16 00:12:37.827678: Current learning rate: 0.00807 
+2026-10-16 00:16:10.120943: train_loss -0.692 
+2026-10-16 00:16:10.121860: val_loss -0.4991 
+2026-10-16 00:16:10.123446: Pseudo dice [np.float32(0.8875), np.float32(0.5229), np.float32(0.6502), np.float32(0.6554), np.float32(0.5848), np.float32(0.7279), np.float32(0.6154), np.float32(0.0), np.float32(0.0), np.float32(0.4249), np.float32(0.5796), np.float32(0.6074), np.float32(0.5194), np.float32(0.4829), np.float32(0.0), np.float32(0.0), np.float32(0.6012), np.float32(0.4416), np.float32(0.6386), np.float32(0.4939), np.float32(0.4875), np.float32(0.5545), np.float32(0.6616), np.float32(0.6715), np.float32(0.4438), np.float32(0.3659), np.float32(0.0), np.float32(0.0), np.float32(0.3357), np.float32(0.5418), np.float32(0.0), np.float32(0.0), np.float32(0.5013), np.float32(0.5537), np.float32(0.6595), np.float32(0.7054)] 
+2026-10-16 00:16:10.124279: Epoch time: 212.3 s 
+2026-10-16 00:16:11.011327:  
+2026-10-16 00:16:11.011969: Epoch 213 
+2026-10-16 00:16:11.012368: Current learning rate: 0.00806 
+2026-10-16 00:19:46.743674: train_loss -0.6889 
+2026-10-16 00:19:46.744756: val_loss -0.5111 
+2026-10-16 00:19:46.745633: Pseudo dice [np.float32(0.8893), np.float32(0.5724), np.float32(0.6336), np.float32(0.6716), np.float32(0.5618), np.float32(0.7289), np.float32(0.5959), np.float32(0.0), np.float32(0.0), np.float32(0.5014), np.float32(0.563), np.float32(0.5457), np.float32(0.4787), np.float32(0.4464), np.float32(0.0), np.float32(0.0), np.float32(0.6573), np.float32(0.4749), np.float32(0.6043), np.float32(0.5116), np.float32(0.486), np.float32(0.5229), np.float32(0.7095), np.float32(0.678), np.float32(0.492), np.float32(0.405), np.float32(0.0), np.float32(0.0), np.float32(0.5023), np.float32(0.6422), np.float32(0.0), np.float32(0.0), np.float32(0.5765), np.float32(0.6462), np.float32(0.7058), np.float32(0.7416)] 
+2026-10-16 00:19:46.746086: Epoch time: 215.73 s 
+2026-10-16 00:19:47.635370:  
+2026-10-16 00:19:47.635931: Epoch 214 
+2026-10-16 00:19:47.636268: Current learning rate: 0.00805 
+2026-10-16 00:23:36.132726: train_loss -0.6761 
+2026-10-16 00:23:36.133739: val_loss -0.3806 
+2026-10-16 00:23:36.134330: Pseudo dice [np.float32(0.8658), np.float32(0.3741), np.float32(0.4484), np.float32(0.395), np.float32(0.4035), np.float32(0.4614), np.float32(0.4432), np.float32(0.0), np.float32(0.0), np.float32(0.4728), np.float32(0.3528), np.float32(0.3991), np.float32(0.3749), np.float32(0.3382), np.float32(0.0), np.float32(0.0), np.float32(0.5128), np.float32(0.3905), np.float32(0.4722), np.float32(0.3944), np.float32(0.3248), np.float32(0.3935), np.float32(0.5636), np.float32(0.5436), np.float32(0.4172), np.float32(0.2508), np.float32(0.0), np.float32(0.0), np.float32(0.2345), np.float32(0.3164), np.float32(0.0), np.float32(0.0), np.float32(0.3856), np.float32(0.4221), np.float32(0.4305), np.float32(0.4706)] 
+2026-10-16 00:23:36.134866: Epoch time: 228.5 s 
+2026-10-16 00:23:37.015018:  
+2026-10-16 00:23:37.015861: Epoch 215 
+2026-10-16 00:23:37.016230: Current learning rate: 0.00804 
+2026-10-16 00:27:11.808915: train_loss -0.6679 
+2026-10-16 00:27:11.809543: val_loss -0.4305 
+2026-10-16 00:27:11.810572: Pseudo dice [np.float32(0.8805), np.float32(0.4558), np.float32(0.4516), np.float32(0.4598), np.float32(0.3632), np.float32(0.5206), np.float32(0.4142), np.float32(0.0), np.float32(0.0), np.float32(0.3146), np.float32(0.3808), np.float32(0.3932), np.float32(0.4791), np.float32(0.4869), np.float32(0.0), np.float32(0.0), np.float32(0.3937), np.float32(0.2784), np.float32(0.367), np.float32(0.3174), np.float32(0.3947), np.float32(0.3045), np.float32(0.6285), np.float32(0.561), np.float32(0.2923), np.float32(0.2336), np.float32(0.0), np.float32(0.0), np.float32(0.4198), np.float32(0.6309), np.float32(0.0), np.float32(0.0), np.float32(0.4152), np.float32(0.4313), np.float32(0.5124), np.float32(0.5695)] 
+2026-10-16 00:27:11.811444: Epoch time: 214.8 s 
+2026-10-16 00:27:12.678755:  
+2026-10-16 00:27:12.679531: Epoch 216 
+2026-10-16 00:27:12.680038: Current learning rate: 0.00803 
+2026-10-16 00:30:53.495461: train_loss -0.629 
+2026-10-16 00:30:53.496011: val_loss -0.537 
+2026-10-16 00:30:53.496671: Pseudo dice [np.float32(0.8684), np.float32(0.6124), np.float32(0.7078), np.float32(0.7), np.float32(0.6699), np.float32(0.7187), np.float32(0.6966), np.float32(0.0), np.float32(0.0), np.float32(0.4996), np.float32(0.6288), np.float32(0.627), np.float32(0.5401), np.float32(0.5409), np.float32(0.0), np.float32(0.0), np.float32(0.723), np.float32(0.5417), np.float32(0.6991), np.float32(0.4784), np.float32(0.5471), np.float32(0.5707), np.float32(0.7804), np.float32(0.7469), np.float32(0.5035), np.float32(0.4764), np.float32(0.0), np.float32(0.0), np.float32(0.4503), np.float32(0.4048), np.float32(0.0), np.float32(0.0), np.float32(0.6621), np.float32(0.6045), np.float32(0.7344), np.float32(0.7475)] 
+2026-10-16 00:30:53.497943: Epoch time: 220.82 s 
+2026-10-16 00:30:54.366798:  
+2026-10-16 00:30:54.367539: Epoch 217 
+2026-10-16 00:30:54.367934: Current learning rate: 0.00802 
+2026-10-16 00:34:35.201900: train_loss -0.6539 
+2026-10-16 00:34:35.202734: val_loss -0.4113 
+2026-10-16 00:34:35.203331: Pseudo dice [np.float32(0.868), np.float32(0.412), np.float32(0.4847), np.float32(0.41), np.float32(0.3565), np.float32(0.4388), np.float32(0.4038), np.float32(0.0), np.float32(0.0), np.float32(0.4276), np.float32(0.3986), np.float32(0.441), np.float32(0.4359), np.float32(0.4481), np.float32(0.0), np.float32(0.0), np.float32(0.4225), np.float32(0.3489), np.float32(0.4154), np.float32(0.3377), np.float32(0.4336), np.float32(0.5211), np.float32(0.5937), np.float32(0.5786), np.float32(0.3411), np.float32(0.2931), np.float32(0.0), np.float32(0.0), np.float32(0.2134), np.float32(0.4029), np.float32(0.0), np.float32(0.0), np.float32(0.337), np.float32(0.4398), np.float32(0.4363), np.float32(0.5105)] 
+2026-10-16 00:34:35.203621: Epoch time: 220.84 s 
+2026-10-16 00:34:36.072248:  
+2026-10-16 00:34:36.073097: Epoch 218 
+2026-10-16 00:34:36.073589: Current learning rate: 0.00801 
+2026-10-16 00:38:09.700147: train_loss -0.6798 
+2026-10-16 00:38:09.700959: val_loss -0.5305 
+2026-10-16 00:38:09.701707: Pseudo dice [np.float32(0.9101), np.float32(0.6034), np.float32(0.7094), np.float32(0.743), np.float32(0.6642), np.float32(0.743), np.float32(0.6483), np.float32(0.0), np.float32(0.0), np.float32(0.4921), np.float32(0.6476), np.float32(0.6009), np.float32(0.51), np.float32(0.4352), np.float32(0.0), np.float32(0.0), np.float32(0.6976), np.float32(0.5148), np.float32(0.6646), np.float32(0.5545), np.float32(0.5633), np.float32(0.5977), np.float32(0.7816), np.float32(0.7897), np.float32(0.5296), np.float32(0.5177), np.float32(0.0), np.float32(0.0), np.float32(0.4234), np.float32(0.5764), np.float32(0.0), np.float32(0.0), np.float32(0.6528), np.float32(0.6148), np.float32(0.7567), np.float32(0.7801)] 
+2026-10-16 00:38:09.702240: Epoch time: 213.63 s 
+2026-10-16 00:38:10.562404:  
+2026-10-16 00:38:10.563130: Epoch 219 
+2026-10-16 00:38:10.563543: Current learning rate: 0.00801 
+2026-10-16 00:42:06.669288: train_loss -0.6778 
+2026-10-16 00:42:06.669886: val_loss -0.4783 
+2026-10-16 00:42:06.670391: Pseudo dice [np.float32(0.8638), np.float32(0.5484), np.float32(0.553), np.float32(0.5904), np.float32(0.5251), np.float32(0.6289), np.float32(0.5354), np.float32(0.0), np.float32(0.0), np.float32(0.5526), np.float32(0.4665), np.float32(0.523), np.float32(0.4896), np.float32(0.4872), np.float32(0.0), np.float32(0.0), np.float32(0.4739), np.float32(0.4118), np.float32(0.4767), np.float32(0.3554), np.float32(0.4946), np.float32(0.4277), np.float32(0.6663), np.float32(0.6151), np.float32(0.3259), np.float32(0.3001), np.float32(0.0), np.float32(0.0), np.float32(0.1839), np.float32(0.4332), np.float32(0.0), np.float32(0.0), np.float32(0.4641), np.float32(0.5451), np.float32(0.6123), np.float32(0.6357)] 
+2026-10-16 00:42:06.671071: Epoch time: 236.11 s 
+2026-10-16 00:42:07.775448:  
+2026-10-16 00:42:07.776013: Epoch 220 
+2026-10-16 00:42:07.776460: Current learning rate: 0.008 
+2026-10-16 00:45:31.910388: train_loss -0.6867 
+2026-10-16 00:45:31.911210: val_loss -0.5355 
+2026-10-16 00:45:31.911917: Pseudo dice [np.float32(0.8997), np.float32(0.6226), np.float32(0.6565), np.float32(0.6652), np.float32(0.6232), np.float32(0.7063), np.float32(0.6206), np.float32(0.0), np.float32(0.0), np.float32(0.5406), np.float32(0.5561), np.float32(0.5573), np.float32(0.5317), np.float32(0.4943), np.float32(0.0), np.float32(0.0), np.float32(0.6445), np.float32(0.5116), np.float32(0.6287), np.float32(0.5356), np.float32(0.5626), np.float32(0.5562), np.float32(0.7368), np.float32(0.7016), np.float32(0.4534), np.float32(0.46), np.float32(0.0), np.float32(0.0), np.float32(0.4365), np.float32(0.5661), np.float32(0.0), np.float32(0.0), np.float32(0.6028), np.float32(0.6965), np.float32(0.6983), np.float32(0.7329)] 
+2026-10-16 00:45:31.912409: Epoch time: 204.14 s 
+2026-10-16 00:45:32.755526:  
+2026-10-16 00:45:32.756204: Epoch 221 
+2026-10-16 00:45:32.756607: Current learning rate: 0.00799 
+2026-10-16 00:49:34.794486: train_loss -0.6884 
+2026-10-16 00:49:34.795145: val_loss -0.5425 
+2026-10-16 00:49:34.795884: Pseudo dice [np.float32(0.8932), np.float32(0.5877), np.float32(0.7299), np.float32(0.7531), np.float32(0.7007), np.float32(0.7955), np.float32(0.6843), np.float32(0.0), np.float32(0.0), np.float32(0.4753), np.float32(0.6459), np.float32(0.6251), np.float32(0.5118), np.float32(0.4894), np.float32(0.0), np.float32(0.0), np.float32(0.7102), np.float32(0.5464), np.float32(0.699), np.float32(0.5244), np.float32(0.5756), np.float32(0.6363), np.float32(0.7631), np.float32(0.7266), np.float32(0.4725), np.float32(0.4526), np.float32(0.0), np.float32(0.0), np.float32(0.3728), np.float32(0.6475), np.float32(0.0), np.float32(0.0), np.float32(0.5231), np.float32(0.6278), np.float32(0.7738), np.float32(0.8036)] 
+2026-10-16 00:49:34.796287: Epoch time: 242.04 s 
+2026-10-16 00:49:35.640354:  
+2026-10-16 00:49:35.640922: Epoch 222 
+2026-10-16 00:49:35.641452: Current learning rate: 0.00798 
+2026-10-16 00:53:41.174312: train_loss -0.6793 
+2026-10-16 00:53:41.175315: val_loss -0.4813 
+2026-10-16 00:53:41.175797: Pseudo dice [np.float32(0.8964), np.float32(0.574), np.float32(0.5643), np.float32(0.5448), np.float32(0.4991), np.float32(0.6117), np.float32(0.5091), np.float32(0.0), np.float32(0.0), np.float32(0.5435), np.float32(0.46), np.float32(0.4939), np.float32(0.539), np.float32(0.4979), np.float32(0.0), np.float32(0.0), np.float32(0.4496), np.float32(0.3946), np.float32(0.4759), np.float32(0.4126), np.float32(0.4928), np.float32(0.4137), np.float32(0.6264), np.float32(0.5918), np.float32(0.31), np.float32(0.3724), np.float32(0.0), np.float32(0.0), np.float32(0.1273), np.float32(0.4529), np.float32(0.0), np.float32(0.0), np.float32(0.4477), np.float32(0.5516), np.float32(0.5911), np.float32(0.63)] 
+2026-10-16 00:53:41.176322: Epoch time: 245.54 s 
+2026-10-16 00:53:42.039235:  
+2026-10-16 00:53:42.039913: Epoch 223 
+2026-10-16 00:53:42.040387: Current learning rate: 0.00797 
+2026-10-16 00:57:14.393570: train_loss -0.6952 
+2026-10-16 00:57:14.394216: val_loss -0.4821 
+2026-10-16 00:57:14.395231: Pseudo dice [np.float32(0.8715), np.float32(0.5345), np.float32(0.6308), np.float32(0.596), np.float32(0.5721), np.float32(0.5767), np.float32(0.5709), np.float32(0.0), np.float32(0.0), np.float32(0.4057), np.float32(0.4684), np.float32(0.5002), np.float32(0.5443), np.float32(0.4332), np.float32(0.0), np.float32(0.0), np.float32(0.5932), np.float32(0.5034), np.float32(0.5681), np.float32(0.4227), np.float32(0.5222), np.float32(0.572), np.float32(0.7042), np.float32(0.6412), np.float32(0.4384), np.float32(0.3264), np.float32(0.0), np.float32(0.0), np.float32(0.3727), np.float32(0.5446), np.float32(0.0), np.float32(0.0), np.float32(0.424), np.float32(0.5698), np.float32(0.6297), np.float32(0.6635)] 
+2026-10-16 00:57:14.395745: Epoch time: 212.36 s 
+2026-10-16 00:57:15.306243:  
+2026-10-16 00:57:15.306905: Epoch 224 
+2026-10-16 00:57:15.307290: Current learning rate: 0.00796 
+2026-10-16 01:01:12.799540: train_loss -0.6889 
+2026-10-16 01:01:12.800509: val_loss -0.5351 
+2026-10-16 01:01:12.801096: Pseudo dice [np.float32(0.8803), np.float32(0.602), np.float32(0.7229), np.float32(0.7178), np.float32(0.6616), np.float32(0.764), np.float32(0.666), np.float32(0.0), np.float32(0.0), np.float32(0.5633), np.float32(0.6022), np.float32(0.5434), np.float32(0.5123), np.float32(0.4487), np.float32(0.0), np.float32(0.0), np.float32(0.7274), np.float32(0.5789), np.float32(0.6979), np.float32(0.5474), np.float32(0.549), np.float32(0.5459), np.float32(0.8158), np.float32(0.7734), np.float32(0.4853), np.float32(0.4283), np.float32(0.0), np.float32(0.0), np.float32(0.4434), np.float32(0.6565), np.float32(0.0), np.float32(0.0), np.float32(0.5423), np.float32(0.6188), np.float32(0.7952), np.float32(0.8224)] 
+2026-10-16 01:01:12.801608: Epoch time: 237.5 s 
+2026-10-16 01:01:13.642627:  
+2026-10-16 01:01:13.643373: Epoch 225 
+2026-10-16 01:01:13.643871: Current learning rate: 0.00795 
+2026-10-16 01:04:51.111587: train_loss -0.6989 
+2026-10-16 01:04:51.112471: val_loss -0.5552 
+2026-10-16 01:04:51.113034: Pseudo dice [np.float32(0.8944), np.float32(0.6284), np.float32(0.7178), np.float32(0.7442), np.float32(0.6738), np.float32(0.7858), np.float32(0.6879), np.float32(0.0), np.float32(0.0), np.float32(0.448), np.float32(0.6453), np.float32(0.6422), np.float32(0.5915), np.float32(0.5525), np.float32(0.0), np.float32(0.0), np.float32(0.6843), np.float32(0.533), np.float32(0.6803), np.float32(0.5434), np.float32(0.5891), np.float32(0.5922), np.float32(0.8344), np.float32(0.8024), np.float32(0.4967), np.float32(0.4291), np.float32(0.0), np.float32(0.0), np.float32(0.4732), np.float32(0.7185), np.float32(0.0), np.float32(0.0), np.float32(0.5831), np.float32(0.5965), np.float32(0.7858), np.float32(0.8095)] 
+2026-10-16 01:04:51.113301: Epoch time: 217.47 s 
+2026-10-16 01:04:51.987706:  
+2026-10-16 01:04:51.988549: Epoch 226 
+2026-10-16 01:04:51.988931: Current learning rate: 0.00794 
+2026-10-16 01:08:25.764913: train_loss -0.6984 
+2026-10-16 01:08:25.765537: val_loss -0.5524 
+2026-10-16 01:08:25.765954: Pseudo dice [np.float32(0.8856), np.float32(0.6197), np.float32(0.7363), np.float32(0.717), np.float32(0.6835), np.float32(0.7788), np.float32(0.6854), np.float32(0.0), np.float32(0.0), np.float32(0.47), np.float32(0.5982), np.float32(0.5743), np.float32(0.5524), np.float32(0.4958), np.float32(0.0), np.float32(0.0), np.float32(0.706), np.float32(0.541), np.float32(0.7073), np.float32(0.5336), np.float32(0.5719), np.float32(0.6103), np.float32(0.7764), np.float32(0.7544), np.float32(0.5078), np.float32(0.4413), np.float32(0.0), np.float32(0.0), np.float32(0.4071), np.float32(0.6213), np.float32(0.0), np.float32(0.0), np.float32(0.653), np.float32(0.7039), np.float32(0.758), np.float32(0.7883)] 
+2026-10-16 01:08:25.766562: Epoch time: 213.78 s 
+2026-10-16 01:08:26.654470:  
+2026-10-16 01:08:26.655378: Epoch 227 
+2026-10-16 01:08:26.655828: Current learning rate: 0.00793 
+2026-10-16 01:11:52.225635: train_loss -0.6926 
+2026-10-16 01:11:52.226330: val_loss -0.5997 
+2026-10-16 01:11:52.226985: Pseudo dice [np.float32(0.8733), np.float32(0.6374), np.float32(0.8153), np.float32(0.8313), np.float32(0.8115), np.float32(0.8531), np.float32(0.792), np.float32(0.0), np.float32(0.0), np.float32(0.5688), np.float32(0.7032), np.float32(0.6959), np.float32(0.5826), np.float32(0.5025), np.float32(0.0), np.float32(0.0), np.float32(0.7944), np.float32(0.5738), np.float32(0.7406), np.float32(0.5625), np.float32(0.561), np.float32(0.6764), np.float32(0.8626), np.float32(0.8324), np.float32(0.6219), np.float32(0.507), np.float32(0.0), np.float32(0.0), np.float32(0.4949), np.float32(0.5991), np.float32(0.0), np.float32(0.0), np.float32(0.5614), np.float32(0.6144), np.float32(0.889), np.float32(0.8844)] 
+2026-10-16 01:11:52.227531: Epoch time: 205.57 s 
+2026-10-16 01:11:52.227779: Yayy! New best EMA pseudo Dice: 0.4535999894142151 
+2026-10-16 01:11:53.819153:  
+2026-10-16 01:11:53.820007: Epoch 228 
+2026-10-16 01:11:53.820298: Current learning rate: 0.00792 
+2026-10-16 01:15:16.867968: train_loss -0.6395 
+2026-10-16 01:15:16.868839: val_loss -0.5334 
+2026-10-16 01:15:16.869549: Pseudo dice [np.float32(0.8972), np.float32(0.63), np.float32(0.7225), np.float32(0.7244), np.float32(0.6911), np.float32(0.7604), np.float32(0.6399), np.float32(0.0), np.float32(0.0), np.float32(0.4459), np.float32(0.5744), np.float32(0.6165), np.float32(0.5219), np.float32(0.4459), np.float32(0.0), np.float32(0.0), np.float32(0.7065), np.float32(0.5136), np.float32(0.6818), np.float32(0.496), np.float32(0.5787), np.float32(0.6053), np.float32(0.7252), np.float32(0.6846), np.float32(0.5303), np.float32(0.5227), np.float32(0.0), np.float32(0.0), np.float32(0.287), np.float32(0.4924), np.float32(0.0), np.float32(0.0), np.float32(0.6185), np.float32(0.6053), np.float32(0.7391), np.float32(0.766)] 
+2026-10-16 01:15:16.870079: Epoch time: 203.05 s 
+2026-10-16 01:15:16.870437: Yayy! New best EMA pseudo Dice: 0.4560999870300293 
+2026-10-16 01:15:18.780061:  
+2026-10-16 01:15:18.784651: Epoch 229 
+2026-10-16 01:15:18.786649: Current learning rate: 0.00791 
+2026-10-16 01:19:01.655020: train_loss -0.6707 
+2026-10-16 01:19:01.656260: val_loss -0.5221 
+2026-10-16 01:19:01.656654: Pseudo dice [np.float32(0.8854), np.float32(0.5715), np.float32(0.6792), np.float32(0.7129), np.float32(0.6606), np.float32(0.7508), np.float32(0.6413), np.float32(0.0), np.float32(0.0), np.float32(0.413), np.float32(0.5932), np.float32(0.4907), np.float32(0.5405), np.float32(0.4948), np.float32(0.0), np.float32(0.0), np.float32(0.7063), np.float32(0.5274), np.float32(0.6877), np.float32(0.5572), np.float32(0.5655), np.float32(0.5988), np.float32(0.7649), np.float32(0.725), np.float32(0.4656), np.float32(0.5116), np.float32(0.0), np.float32(0.0), np.float32(0.1111), np.float32(0.4542), np.float32(0.0), np.float32(0.0), np.float32(0.5617), np.float32(0.6639), np.float32(0.7694), np.float32(0.8008)] 
+2026-10-16 01:19:01.656917: Epoch time: 222.88 s 
+2026-10-16 01:19:01.657359: Yayy! New best EMA pseudo Dice: 0.45739999413490295 
+2026-10-16 01:19:05.121473:  
+2026-10-16 01:19:05.122206: Epoch 230 
+2026-10-16 01:19:05.122537: Current learning rate: 0.0079 
+2026-10-16 01:22:53.952848: train_loss -0.682 
+2026-10-16 01:22:53.953565: val_loss -0.5627 
+2026-10-16 01:22:53.954316: Pseudo dice [np.float32(0.887), np.float32(0.6426), np.float32(0.7773), np.float32(0.7989), np.float32(0.7552), np.float32(0.8293), np.float32(0.7254), np.float32(0.0), np.float32(0.0), np.float32(0.6173), np.float32(0.6641), np.float32(0.6675), np.float32(0.5563), np.float32(0.4892), np.float32(0.0), np.float32(0.0), np.float32(0.7633), np.float32(0.5327), np.float32(0.717), np.float32(0.5381), np.float32(0.6121), np.float32(0.6573), np.float32(0.8387), np.float32(0.8249), np.float32(0.5073), np.float32(0.4891), np.float32(0.0), np.float32(0.0), np.float32(0.3719), np.float32(0.489), np.float32(0.0), np.float32(0.0), np.float32(0.6887), np.float32(0.6973), np.float32(0.8176), np.float32(0.844)] 
+2026-10-16 01:22:53.954731: Epoch time: 228.83 s 
+2026-10-16 01:22:53.954999: Yayy! New best EMA pseudo Dice: 0.46389999985694885 
+2026-10-16 01:22:55.462703:  
+2026-10-16 01:22:55.463372: Epoch 231 
+2026-10-16 01:22:55.463737: Current learning rate: 0.00789 
+2026-10-16 01:26:44.537766: train_loss -0.6558 
+2026-10-16 01:26:44.538367: val_loss -0.5257 
+2026-10-16 01:26:44.539177: Pseudo dice [np.float32(0.8951), np.float32(0.5767), np.float32(0.6653), np.float32(0.6868), np.float32(0.6977), np.float32(0.711), np.float32(0.6385), np.float32(0.0), np.float32(0.0), np.float32(0.5352), np.float32(0.577), np.float32(0.6197), np.float32(0.5569), np.float32(0.5454), np.float32(0.0), np.float32(0.0), np.float32(0.6616), np.float32(0.4961), np.float32(0.5997), np.float32(0.5105), np.float32(0.5765), np.float32(0.6129), np.float32(0.7382), np.float32(0.7337), np.float32(0.4976), np.float32(0.2635), np.float32(0.0), np.float32(0.0), np.float32(0.3245), np.float32(0.5004), np.float32(0.0), np.float32(0.0), np.float32(0.3742), np.float32(0.4348), np.float32(0.7256), np.float32(0.7617)] 
+2026-10-16 01:26:44.539711: Epoch time: 229.08 s 
+2026-10-16 01:26:45.354918:  
+2026-10-16 01:26:45.355781: Epoch 232 
+2026-10-16 01:26:45.356233: Current learning rate: 0.00789 
+2026-10-16 01:30:33.579337: train_loss -0.6465 
+2026-10-16 01:30:33.580058: val_loss -0.4896 
+2026-10-16 01:30:33.580410: Pseudo dice [np.float32(0.901), np.float32(0.5992), np.float32(0.5885), np.float32(0.554), np.float32(0.5101), np.float32(0.6475), np.float32(0.5309), np.float32(0.0), np.float32(0.0), np.float32(0.3904), np.float32(0.5021), np.float32(0.5659), np.float32(0.5078), np.float32(0.5433), np.float32(0.0), np.float32(0.0), np.float32(0.4706), np.float32(0.3853), np.float32(0.4783), np.float32(0.3679), np.float32(0.5023), np.float32(0.4512), np.float32(0.6863), np.float32(0.6594), np.float32(0.2657), np.float32(0.3347), np.float32(0.0), np.float32(0.0), np.float32(0.3248), np.float32(0.5386), np.float32(0.0), np.float32(0.0), np.float32(0.417), np.float32(0.4854), np.float32(0.6371), np.float32(0.6611)] 
+2026-10-16 01:30:33.580930: Epoch time: 228.23 s 
+2026-10-16 01:30:34.483139:  
+2026-10-16 01:30:34.483864: Epoch 233 
+2026-10-16 01:30:34.484241: Current learning rate: 0.00788 
+2026-10-16 01:34:15.222069: train_loss -0.6242 
+2026-10-16 01:34:15.222877: val_loss -0.5109 
+2026-10-16 01:34:15.223571: Pseudo dice [np.float32(0.8825), np.float32(0.5688), np.float32(0.6708), np.float32(0.6809), np.float32(0.6537), np.float32(0.7211), np.float32(0.6532), np.float32(0.0), np.float32(0.0), np.float32(0.4148), np.float32(0.5424), np.float32(0.5832), np.float32(0.4578), np.float32(0.3377), np.float32(0.0), np.float32(0.0), np.float32(0.6512), np.float32(0.4803), np.float32(0.6479), np.float32(0.5141), np.float32(0.4992), np.float32(0.5861), np.float32(0.7406), np.float32(0.7558), np.float32(0.5357), np.float32(0.4205), np.float32(0.0), np.float32(0.0), np.float32(0.4005), np.float32(0.5127), np.float32(0.0), np.float32(0.0), np.float32(0.5349), np.float32(0.6293), np.float32(0.7198), np.float32(0.7373)] 
+2026-10-16 01:34:15.223976: Epoch time: 220.74 s 
+2026-10-16 01:34:16.073627:  
+2026-10-16 01:34:16.074244: Epoch 234 
+2026-10-16 01:34:16.074999: Current learning rate: 0.00787 
+2026-10-16 01:37:45.267690: train_loss -0.6474 
+2026-10-16 01:37:45.268352: val_loss -0.5797 
+2026-10-16 01:37:45.269248: Pseudo dice [np.float32(0.8916), np.float32(0.6407), np.float32(0.7314), np.float32(0.7864), np.float32(0.7797), np.float32(0.8276), np.float32(0.7386), np.float32(0.0), np.float32(0.0), np.float32(0.515), np.float32(0.6239), np.float32(0.6278), np.float32(0.5419), np.float32(0.4524), np.float32(0.0), np.float32(0.0), np.float32(0.7655), np.float32(0.5447), np.float32(0.7023), np.float32(0.5075), np.float32(0.5559), np.float32(0.6444), np.float32(0.8774), np.float32(0.8601), np.float32(0.5896), np.float32(0.5079), np.float32(0.0), np.float32(0.0), np.float32(0.4338), np.float32(0.3849), np.float32(0.0), np.float32(0.0), np.float32(0.6114), np.float32(0.635), np.float32(0.8548), np.float32(0.8639)] 
+2026-10-16 01:37:45.269776: Epoch time: 209.2 s 
+2026-10-16 01:37:46.073608:  
+2026-10-16 01:37:46.074380: Epoch 235 
+2026-10-16 01:37:46.074720: Current learning rate: 0.00786 
+2026-10-16 01:41:36.677012: train_loss -0.635 
+2026-10-16 01:41:36.677973: val_loss -0.55 
+2026-10-16 01:41:36.678651: Pseudo dice [np.float32(0.8588), np.float32(0.6017), np.float32(0.7337), np.float32(0.749), np.float32(0.6496), np.float32(0.7823), np.float32(0.6676), np.float32(0.0), np.float32(0.0), np.float32(0.3623), np.float32(0.6433), np.float32(0.6491), np.float32(0.5504), np.float32(0.4637), np.float32(0.0), np.float32(0.0), np.float32(0.6612), np.float32(0.5163), np.float32(0.6269), np.float32(0.5251), np.float32(0.5744), np.float32(0.6189), np.float32(0.7902), np.float32(0.7973), np.float32(0.4728), np.float32(0.406), np.float32(0.0), np.float32(0.0), np.float32(0.0969), np.float32(0.5114), np.float32(0.0), np.float32(0.0), np.float32(0.5383), np.float32(0.6215), np.float32(0.7943), np.float32(0.8103)] 
+2026-10-16 01:41:36.679033: Epoch time: 230.61 s 
+2026-10-16 01:41:36.679391: Yayy! New best EMA pseudo Dice: 0.4643000066280365 
+2026-10-16 01:41:38.421825:  
+2026-10-16 01:41:38.422700: Epoch 236 
+2026-10-16 01:41:38.422976: Current learning rate: 0.00785 
+2026-10-16 01:45:38.711785: train_loss -0.6381 
+2026-10-16 01:45:38.712320: val_loss -0.4989 
+2026-10-16 01:45:38.713416: Pseudo dice [np.float32(0.8772), np.float32(0.5582), np.float32(0.6375), np.float32(0.6447), np.float32(0.6021), np.float32(0.6821), np.float32(0.5922), np.float32(0.0), np.float32(0.0), np.float32(0.4663), np.float32(0.5553), np.float32(0.5715), np.float32(0.5467), np.float32(0.4478), np.float32(0.0), np.float32(0.0), np.float32(0.6219), np.float32(0.4839), np.float32(0.655), np.float32(0.5559), np.float32(0.5357), np.float32(0.5628), np.float32(0.7269), np.float32(0.6754), np.float32(0.4432), np.float32(0.3023), np.float32(0.0), np.float32(0.0), np.float32(0.1927), np.float32(0.3967), np.float32(0.0), np.float32(0.0), np.float32(0.5266), np.float32(0.6155), np.float32(0.6692), np.float32(0.6821)] 
+2026-10-16 01:45:38.713933: Epoch time: 240.29 s 
+2026-10-16 01:45:39.966224:  
+2026-10-16 01:45:39.967064: Epoch 237 
+2026-10-16 01:45:39.967505: Current learning rate: 0.00784 
+2026-10-16 01:49:23.307363: train_loss -0.655 
+2026-10-16 01:49:23.308104: val_loss -0.2992 
+2026-10-16 01:49:23.308620: Pseudo dice [np.float32(0.833), np.float32(0.1732), np.float32(0.214), np.float32(0.1723), np.float32(0.0139), np.float32(0.23), np.float32(0.2191), np.float32(0.0), np.float32(0.0), np.float32(0.5069), np.float32(0.2745), np.float32(0.2013), np.float32(0.2966), np.float32(0.2028), np.float32(0.0), np.float32(0.0), np.float32(0.0254), np.float32(0.015), np.float32(0.2439), np.float32(0.289), np.float32(0.2112), np.float32(0.1418), np.float32(0.3455), np.float32(0.2501), np.float32(0.0439), np.float32(0.0405), np.float32(0.0), np.float32(0.0), np.float32(0.0577), np.float32(0.2253), np.float32(0.0), np.float32(0.0), np.float32(0.0458), np.float32(0.1482), np.float32(0.1934), np.float32(0.2183)] 
+2026-10-16 01:49:23.309468: Epoch time: 223.34 s 
+2026-10-16 01:49:24.108404:  
+2026-10-16 01:49:24.109122: Epoch 238 
+2026-10-16 01:49:24.109509: Current learning rate: 0.00783 
+2026-10-16 01:53:09.851008: train_loss -0.6345 
+2026-10-16 01:53:09.851667: val_loss -0.5288 
+2026-10-16 01:53:09.852158: Pseudo dice [np.float32(0.8781), np.float32(0.6194), np.float32(0.6999), np.float32(0.7053), np.float32(0.6572), np.float32(0.7567), np.float32(0.6513), np.float32(0.0), np.float32(0.0), np.float32(0.4883), np.float32(0.5855), np.float32(0.586), np.float32(0.558), np.float32(0.4672), np.float32(0.0), np.float32(0.0), np.float32(0.7091), np.float32(0.5397), np.float32(0.6986), np.float32(0.5392), np.float32(0.51), np.float32(0.6203), np.float32(0.776), np.float32(0.7271), np.float32(0.5777), np.float32(0.457), np.float32(0.0), np.float32(0.0), np.float32(0.4301), np.float32(0.5897), np.float32(0.0), np.float32(0.0), np.float32(0.4623), np.float32(0.6397), np.float32(0.7274), np.float32(0.7616)] 
+2026-10-16 01:53:09.853051: Epoch time: 225.74 s 
+2026-10-16 01:53:10.705709:  
+2026-10-16 01:53:10.706488: Epoch 239 
+2026-10-16 01:53:10.706905: Current learning rate: 0.00782 
+2026-10-16 01:56:45.327053: train_loss -0.6669 
+2026-10-16 01:56:45.327743: val_loss -0.5092 
+2026-10-16 01:56:45.328382: Pseudo dice [np.float32(0.8912), np.float32(0.5725), np.float32(0.639), np.float32(0.6326), np.float32(0.5733), np.float32(0.6721), np.float32(0.5908), np.float32(0.0), np.float32(0.0), np.float32(0.3589), np.float32(0.5573), np.float32(0.5581), np.float32(0.5425), np.float32(0.4771), np.float32(0.0), np.float32(0.0), np.float32(0.6454), np.float32(0.5231), np.float32(0.6462), np.float32(0.5776), np.float32(0.5288), np.float32(0.5165), np.float32(0.6648), np.float32(0.6491), np.float32(0.4992), np.float32(0.3648), np.float32(0.0), np.float32(0.0), np.float32(0.2796), np.float32(0.4923), np.float32(0.0), np.float32(0.0), np.float32(0.5473), np.float32(0.5757), np.float32(0.6638), np.float32(0.6803)] 
+2026-10-16 01:56:45.329427: Epoch time: 214.62 s 
+2026-10-16 01:56:46.207350:  
+2026-10-16 01:56:46.208226: Epoch 240 
+2026-10-16 01:56:46.208725: Current learning rate: 0.00781 
+2026-10-16 02:00:28.049899: train_loss -0.6886 
+2026-10-16 02:00:28.050808: val_loss -0.4869 
+2026-10-16 02:00:28.051660: Pseudo dice [np.float32(0.8899), np.float32(0.5483), np.float32(0.6233), np.float32(0.6369), np.float32(0.5732), np.float32(0.6909), np.float32(0.5876), np.float32(0.0), np.float32(0.0), np.float32(0.3998), np.float32(0.5097), np.float32(0.4932), np.float32(0.4858), np.float32(0.4354), np.float32(0.0), np.float32(0.0), np.float32(0.6431), np.float32(0.4867), np.float32(0.6295), np.float32(0.5268), np.float32(0.465), np.float32(0.5415), np.float32(0.5954), np.float32(0.5447), np.float32(0.4767), np.float32(0.441), np.float32(0.0), np.float32(0.0), np.float32(0.2709), np.float32(0.3423), np.float32(0.0), np.float32(0.0), np.float32(0.5121), np.float32(0.5778), np.float32(0.6222), np.float32(0.6608)] 
+2026-10-16 02:00:28.052042: Epoch time: 221.84 s 
+2026-10-16 02:00:28.907981:  
+2026-10-16 02:00:28.908731: Epoch 241 
+2026-10-16 02:00:28.909225: Current learning rate: 0.0078 
+2026-10-16 02:04:15.915865: train_loss -0.6457 
+2026-10-16 02:04:15.916606: val_loss -0.4419 
+2026-10-16 02:04:15.917589: Pseudo dice [np.float32(0.8849), np.float32(0.4711), np.float32(0.4844), np.float32(0.4524), np.float32(0.3711), np.float32(0.536), np.float32(0.4489), np.float32(0.0), np.float32(0.0), np.float32(0.2598), np.float32(0.4078), np.float32(0.4406), np.float32(0.5303), np.float32(0.4226), np.float32(0.0), np.float32(0.0), np.float32(0.3299), np.float32(0.2821), np.float32(0.4067), np.float32(0.3532), np.float32(0.3546), np.float32(0.3627), np.float32(0.5598), np.float32(0.4917), np.float32(0.2198), np.float32(0.2015), np.float32(0.0), np.float32(0.0), np.float32(0.2424), np.float32(0.5522), np.float32(0.0), np.float32(0.0), np.float32(0.2566), np.float32(0.4762), np.float32(0.5429), np.float32(0.5404)] 
+2026-10-16 02:04:15.918047: Epoch time: 227.01 s 
+2026-10-16 02:04:16.788677:  
+2026-10-16 02:04:16.789325: Epoch 242 
+2026-10-16 02:04:16.789687: Current learning rate: 0.00779 
+2026-10-16 02:07:51.881486: train_loss -0.605 
+2026-10-16 02:07:51.882003: val_loss -0.4371 
+2026-10-16 02:07:51.883056: Pseudo dice [np.float32(0.887), np.float32(0.4985), np.float32(0.5468), np.float32(0.4948), np.float32(0.4432), np.float32(0.4521), np.float32(0.3089), np.float32(0.0), np.float32(0.0), np.float32(0.2487), np.float32(0.3872), np.float32(0.4765), np.float32(0.4962), np.float32(0.3856), np.float32(0.0), np.float32(0.0), np.float32(0.4609), np.float32(0.4319), np.float32(0.3681), np.float32(0.3255), np.float32(0.428), np.float32(0.4618), np.float32(0.7461), np.float32(0.7228), np.float32(0.413), np.float32(0.3153), np.float32(0.0), np.float32(0.0), np.float32(0.1714), np.float32(0.4754), np.float32(0.0), np.float32(0.0), np.float32(0.3937), np.float32(0.3267), np.float32(0.5793), np.float32(0.5955)] 
+2026-10-16 02:07:51.883513: Epoch time: 215.09 s 
+2026-10-16 02:07:52.681958:  
+2026-10-16 02:07:52.682482: Epoch 243 
+2026-10-16 02:07:52.682907: Current learning rate: 0.00778 
+2026-10-16 02:11:38.008731: train_loss -0.6444 
+2026-10-16 02:11:38.009671: val_loss -0.3684 
+2026-10-16 02:11:38.010112: Pseudo dice [np.float32(0.8834), np.float32(0.3968), np.float32(0.356), np.float32(0.3065), np.float32(0.3615), np.float32(0.3743), np.float32(0.338), np.float32(0.0), np.float32(0.0), np.float32(0.3318), np.float32(0.2639), np.float32(0.3227), np.float32(0.4472), np.float32(0.3247), np.float32(0.0), np.float32(0.0), np.float32(0.2711), np.float32(0.2031), np.float32(0.3395), np.float32(0.284), np.float32(0.357), np.float32(0.3289), np.float32(0.5907), np.float32(0.5044), np.float32(0.2569), np.float32(0.2148), np.float32(0.0), np.float32(0.0), np.float32(0.265), np.float32(0.2998), np.float32(0.0), np.float32(0.0), np.float32(0.2507), np.float32(0.3257), np.float32(0.3751), np.float32(0.4426)] 
+2026-10-16 02:11:38.010618: Epoch time: 225.33 s 
+2026-10-16 02:11:38.839176:  
+2026-10-16 02:11:38.839815: Epoch 244 
+2026-10-16 02:11:38.840164: Current learning rate: 0.00777 
+2026-10-16 02:15:14.809575: train_loss -0.6772 
+2026-10-16 02:15:14.810429: val_loss -0.4594 
+2026-10-16 02:15:14.811006: Pseudo dice [np.float32(0.8859), np.float32(0.5054), np.float32(0.511), np.float32(0.505), np.float32(0.4263), np.float32(0.5246), np.float32(0.4347), np.float32(0.0), np.float32(0.0), np.float32(0.3038), np.float32(0.4217), np.float32(0.4792), np.float32(0.462), np.float32(0.4469), np.float32(0.0), np.float32(0.0), np.float32(0.3854), np.float32(0.3786), np.float32(0.3613), np.float32(0.3443), np.float32(0.4715), np.float32(0.4033), np.float32(0.6696), np.float32(0.6333), np.float32(0.3302), np.float32(0.2405), np.float32(0.0), np.float32(0.0), np.float32(0.3409), np.float32(0.5538), np.float32(0.0), np.float32(0.0), np.float32(0.4144), np.float32(0.4821), np.float32(0.5621), np.float32(0.5935)] 
+2026-10-16 02:15:14.811454: Epoch time: 215.97 s 
+2026-10-16 02:15:15.702956:  
+2026-10-16 02:15:15.703770: Epoch 245 
+2026-10-16 02:15:15.704150: Current learning rate: 0.00777 
+2026-10-16 02:18:42.108104: train_loss -0.6897 
+2026-10-16 02:18:42.108813: val_loss -0.4994 
+2026-10-16 02:18:42.109474: Pseudo dice [np.float32(0.8912), np.float32(0.5345), np.float32(0.6233), np.float32(0.6167), np.float32(0.5598), np.float32(0.6428), np.float32(0.5521), np.float32(0.0), np.float32(0.0), np.float32(0.5046), np.float32(0.4925), np.float32(0.5271), np.float32(0.4816), np.float32(0.4032), np.float32(0.0), np.float32(0.0), np.float32(0.5654), np.float32(0.4477), np.float32(0.5642), np.float32(0.469), np.float32(0.4829), np.float32(0.5513), np.float32(0.7131), np.float32(0.6777), np.float32(0.4179), np.float32(0.3764), np.float32(0.0), np.float32(0.0), np.float32(0.315), np.float32(0.4181), np.float32(0.0), np.float32(0.0), np.float32(0.3902), np.float32(0.5743), np.float32(0.6705), np.float32(0.7005)] 
+2026-10-16 02:18:42.109912: Epoch time: 206.41 s 
+2026-10-16 02:18:42.935025:  
+2026-10-16 02:18:42.935662: Epoch 246 
+2026-10-16 02:18:42.936143: Current learning rate: 0.00776 
+2026-10-16 02:22:33.147081: train_loss -0.6557 
+2026-10-16 02:22:33.147829: val_loss -0.3839 
+2026-10-16 02:22:33.148527: Pseudo dice [np.float32(0.8795), np.float32(0.4757), np.float32(0.4119), np.float32(0.3472), np.float32(0.4157), np.float32(0.4166), np.float32(0.3672), np.float32(0.0), np.float32(0.0), np.float32(0.4969), np.float32(0.2495), np.float32(0.3476), np.float32(0.333), np.float32(0.3586), np.float32(0.0), np.float32(0.0), np.float32(0.4191), np.float32(0.3395), np.float32(0.3056), np.float32(0.2819), np.float32(0.3409), np.float32(0.3555), np.float32(0.581), np.float32(0.4937), np.float32(0.3004), np.float32(0.2915), np.float32(0.0), np.float32(0.0), np.float32(0.0656), np.float32(0.3186), np.float32(0.0), np.float32(0.0), np.float32(0.3202), np.float32(0.3689), np.float32(0.3782), np.float32(0.4017)] 
+2026-10-16 02:22:33.149181: Epoch time: 230.21 s 
+2026-10-16 02:22:33.971944:  
+2026-10-16 02:22:33.972729: Epoch 247 
+2026-10-16 02:22:33.973147: Current learning rate: 0.00775 
+2026-10-16 02:26:10.327205: train_loss -0.6481 
+2026-10-16 02:26:10.327857: val_loss -0.4059 
+2026-10-16 02:26:10.328328: Pseudo dice [np.float32(0.8887), np.float32(0.441), np.float32(0.4328), np.float32(0.3737), np.float32(0.4298), np.float32(0.423), np.float32(0.4018), np.float32(0.0), np.float32(0.0), np.float32(0.4569), np.float32(0.273), np.float32(0.3689), np.float32(0.3853), np.float32(0.4038), np.float32(0.0), np.float32(0.0), np.float32(0.3833), np.float32(0.3033), np.float32(0.3909), np.float32(0.3371), np.float32(0.3675), np.float32(0.3414), np.float32(0.6116), np.float32(0.5505), np.float32(0.2378), np.float32(0.3313), np.float32(0.0), np.float32(0.0), np.float32(0.2021), np.float32(0.4033), np.float32(0.0), np.float32(0.0), np.float32(0.4167), np.float32(0.4566), np.float32(0.4495), np.float32(0.4601)] 
+2026-10-16 02:26:10.329298: Epoch time: 216.36 s 
+2026-10-16 02:26:11.160292:  
+2026-10-16 02:26:11.161025: Epoch 248 
+2026-10-16 02:26:11.161301: Current learning rate: 0.00774 
+2026-10-16 02:30:11.249267: train_loss -0.6796 
+2026-10-16 02:30:11.249861: val_loss -0.4506 
+2026-10-16 02:30:11.250325: Pseudo dice [np.float32(0.8783), np.float32(0.5095), np.float32(0.515), np.float32(0.497), np.float32(0.4397), np.float32(0.5481), np.float32(0.418), np.float32(0.0), np.float32(0.0), np.float32(0.3207), np.float32(0.4258), np.float32(0.4144), np.float32(0.5479), np.float32(0.3957), np.float32(0.0), np.float32(0.0), np.float32(0.4171), np.float32(0.3133), np.float32(0.4101), np.float32(0.368), np.float32(0.4396), np.float32(0.404), np.float32(0.6575), np.float32(0.5861), np.float32(0.2818), np.float32(0.2042), np.float32(0.0), np.float32(0.0), np.float32(0.4951), np.float32(0.5481), np.float32(0.0), np.float32(0.0), np.float32(0.3812), np.float32(0.5164), np.float32(0.556), np.float32(0.564)] 
+2026-10-16 02:30:11.250867: Epoch time: 240.09 s 
+2026-10-16 02:30:12.125170:  
+2026-10-16 02:30:12.125819: Epoch 249 
+2026-10-16 02:30:12.126296: Current learning rate: 0.00773 
+2026-10-16 02:33:41.897454: train_loss -0.6814 
+2026-10-16 02:33:41.898246: val_loss -0.3637 
+2026-10-16 02:33:41.898879: Pseudo dice [np.float32(0.8946), np.float32(0.3819), np.float32(0.3387), np.float32(0.325), np.float32(0.3993), np.float32(0.3859), np.float32(0.3606), np.float32(0.0), np.float32(0.0), np.float32(0.3755), np.float32(0.2483), np.float32(0.3025), np.float32(0.4061), np.float32(0.3827), np.float32(0.0), np.float32(0.0), np.float32(0.3683), np.float32(0.2587), np.float32(0.3327), np.float32(0.2959), np.float32(0.3217), np.float32(0.3299), np.float32(0.5771), np.float32(0.5086), np.float32(0.2484), np.float32(0.297), np.float32(0.0), np.float32(0.0), np.float32(0.0037), np.float32(0.1745), np.float32(0.0), np.float32(0.0), np.float32(0.4039), np.float32(0.4561), np.float32(0.4041), np.float32(0.4422)] 
+2026-10-16 02:33:41.899235: Epoch time: 209.77 s 
+2026-10-16 02:33:44.348000:  
+2026-10-16 02:33:44.348750: Epoch 250 
+2026-10-16 02:33:44.349140: Current learning rate: 0.00772 
+2026-10-16 02:37:19.278697: train_loss -0.6959 
+2026-10-16 02:37:19.280077: val_loss -0.4477 
+2026-10-16 02:37:19.280591: Pseudo dice [np.float32(0.8939), np.float32(0.4983), np.float32(0.489), np.float32(0.4872), np.float32(0.4492), np.float32(0.541), np.float32(0.4797), np.float32(0.0), np.float32(0.0), np.float32(0.4551), np.float32(0.4404), np.float32(0.4671), np.float32(0.5707), np.float32(0.5364), np.float32(0.0), np.float32(0.0), np.float32(0.3652), np.float32(0.3105), np.float32(0.4064), np.float32(0.3324), np.float32(0.442), np.float32(0.3411), np.float32(0.6705), np.float32(0.5934), np.float32(0.2244), np.float32(0.2257), np.float32(0.0), np.float32(0.0), np.float32(0.2498), np.float32(0.6077), np.float32(0.0), np.float32(0.0), np.float32(0.3998), np.float32(0.5387), np.float32(0.5537), np.float32(0.59)] 
+2026-10-16 02:37:19.281019: Epoch time: 214.93 s 
+2026-10-16 02:37:20.068406:  
+2026-10-16 02:37:20.069083: Epoch 251 
+2026-10-16 02:37:20.069354: Current learning rate: 0.00771 
+2026-10-16 02:40:53.139470: train_loss -0.6999 
+2026-10-16 02:40:53.140163: val_loss -0.4341 
+2026-10-16 02:40:53.140786: Pseudo dice [np.float32(0.8913), np.float32(0.4715), np.float32(0.4667), np.float32(0.4883), np.float32(0.3774), np.float32(0.5412), np.float32(0.443), np.float32(0.0), np.float32(0.0), np.float32(0.5025), np.float32(0.4011), np.float32(0.4157), np.float32(0.5233), np.float32(0.4764), np.float32(0.0), np.float32(0.0), np.float32(0.2823), np.float32(0.252), np.float32(0.3575), np.float32(0.3002), np.float32(0.4469), np.float32(0.3155), np.float32(0.676), np.float32(0.5971), np.float32(0.2071), np.float32(0.2408), np.float32(0.0), np.float32(0.0), np.float32(0.0124), np.float32(0.3328), np.float32(0.0), np.float32(0.0), np.float32(0.3215), np.float32(0.4009), np.float32(0.5517), np.float32(0.5945)] 
+2026-10-16 02:40:53.141600: Epoch time: 213.07 s 
+2026-10-16 02:40:54.015777:  
+2026-10-16 02:40:54.016657: Epoch 252 
+2026-10-16 02:40:54.017025: Current learning rate: 0.0077 
+2026-10-16 02:44:31.390295: train_loss -0.6825 
+2026-10-16 02:44:31.391266: val_loss -0.4032 
+2026-10-16 02:44:31.391799: Pseudo dice [np.float32(0.8918), np.float32(0.4604), np.float32(0.4337), np.float32(0.3933), np.float32(0.4107), np.float32(0.4414), np.float32(0.3996), np.float32(0.0), np.float32(0.0), np.float32(0.2912), np.float32(0.318), np.float32(0.3124), np.float32(0.506), np.float32(0.4624), np.float32(0.0), np.float32(0.0), np.float32(0.3591), np.float32(0.2809), np.float32(0.408), np.float32(0.3205), np.float32(0.3421), np.float32(0.3013), np.float32(0.5559), np.float32(0.5293), np.float32(0.2172), np.float32(0.2746), np.float32(0.0), np.float32(0.0), np.float32(0.0481), np.float32(0.3047), np.float32(0.0), np.float32(0.0), np.float32(0.3779), np.float32(0.503), np.float32(0.4509), np.float32(0.5033)] 
+2026-10-16 02:44:31.392227: Epoch time: 217.38 s 
+2026-10-16 02:44:32.239164:  
+2026-10-16 02:44:32.239910: Epoch 253 
+2026-10-16 02:44:32.240232: Current learning rate: 0.00769 
+2026-10-16 02:48:07.285092: train_loss -0.693 
+2026-10-16 02:48:07.285846: val_loss -0.418 
+2026-10-16 02:48:07.286393: Pseudo dice [np.float32(0.8965), np.float32(0.47), np.float32(0.4404), np.float32(0.4273), np.float32(0.4752), np.float32(0.4636), np.float32(0.4454), np.float32(0.0), np.float32(0.0), np.float32(0.3923), np.float32(0.2999), np.float32(0.3819), np.float32(0.4438), np.float32(0.3916), np.float32(0.0), np.float32(0.0), np.float32(0.3979), np.float32(0.3053), np.float32(0.3908), np.float32(0.3187), np.float32(0.3501), np.float32(0.3315), np.float32(0.6467), np.float32(0.5247), np.float32(0.2196), np.float32(0.3087), np.float32(0.0), np.float32(0.0), np.float32(0.1792), np.float32(0.2471), np.float32(0.0), np.float32(0.0), np.float32(0.4441), np.float32(0.5228), np.float32(0.4946), np.float32(0.5442)] 
+2026-10-16 02:48:07.287131: Epoch time: 215.05 s 
+2026-10-16 02:48:08.142728:  
+2026-10-16 02:48:08.143615: Epoch 254 
+2026-10-16 02:48:08.144054: Current learning rate: 0.00768 
+2026-10-16 02:51:43.947535: train_loss -0.7021 
+2026-10-16 02:51:43.948079: val_loss -0.5181 
+2026-10-16 02:51:43.948599: Pseudo dice [np.float32(0.9117), np.float32(0.597), np.float32(0.63), np.float32(0.6418), np.float32(0.5335), np.float32(0.675), np.float32(0.564), np.float32(0.0), np.float32(0.0), np.float32(0.5039), np.float32(0.5478), np.float32(0.538), np.float32(0.5703), np.float32(0.5454), np.float32(0.0), np.float32(0.0), np.float32(0.4912), np.float32(0.3882), np.float32(0.5514), np.float32(0.5059), np.float32(0.5215), np.float32(0.4724), np.float32(0.7243), np.float32(0.6865), np.float32(0.3511), np.float32(0.4227), np.float32(0.0), np.float32(0.0), np.float32(0.348), np.float32(0.5792), np.float32(0.0), np.float32(0.0), np.float32(0.4633), np.float32(0.6772), np.float32(0.6677), np.float32(0.7199)] 
+2026-10-16 02:51:43.949422: Epoch time: 215.81 s 
+2026-10-16 02:51:44.840056:  
+2026-10-16 02:51:44.840879: Epoch 255 
+2026-10-16 02:51:44.841260: Current learning rate: 0.00767 
+2026-10-16 02:55:34.121502: train_loss -0.6963 
+2026-10-16 02:55:34.122513: val_loss -0.4221 
+2026-10-16 02:55:34.123100: Pseudo dice [np.float32(0.8501), np.float32(0.495), np.float32(0.4713), np.float32(0.4393), np.float32(0.4357), np.float32(0.4581), np.float32(0.425), np.float32(0.0), np.float32(0.0), np.float32(0.4648), np.float32(0.3365), np.float32(0.3114), np.float32(0.4506), np.float32(0.4398), np.float32(0.0), np.float32(0.0), np.float32(0.4491), np.float32(0.3802), np.float32(0.3996), np.float32(0.3414), np.float32(0.441), np.float32(0.425), np.float32(0.6009), np.float32(0.5543), np.float32(0.3311), np.float32(0.2987), np.float32(0.0), np.float32(0.0), np.float32(0.0996), np.float32(0.4264), np.float32(0.0), np.float32(0.0), np.float32(0.4365), np.float32(0.5234), np.float32(0.5016), np.float32(0.5341)] 
+2026-10-16 02:55:34.123369: Epoch time: 229.28 s 
+2026-10-16 02:55:34.908551:  
+2026-10-16 02:55:34.909427: Epoch 256 
+2026-10-16 02:55:34.909868: Current learning rate: 0.00766 
+2026-10-16 02:59:43.603414: train_loss -0.6927 
+2026-10-16 02:59:43.603965: val_loss -0.3974 
+2026-10-16 02:59:43.604578: Pseudo dice [np.float32(0.8829), np.float32(0.4582), np.float32(0.4312), np.float32(0.4077), np.float32(0.3952), np.float32(0.4924), np.float32(0.4139), np.float32(0.0), np.float32(0.0), np.float32(0.3204), np.float32(0.3546), np.float32(0.3691), np.float32(0.52), np.float32(0.5237), np.float32(0.0), np.float32(0.0), np.float32(0.2882), np.float32(0.248), np.float32(0.3899), np.float32(0.3698), np.float32(0.3833), np.float32(0.3555), np.float32(0.5419), np.float32(0.4707), np.float32(0.1903), np.float32(0.2347), np.float32(0.0), np.float32(0.0), np.float32(0.1401), np.float32(0.4824), np.float32(0.0), np.float32(0.0), np.float32(0.3578), np.float32(0.4169), np.float32(0.4247), np.float32(0.4869)] 
+2026-10-16 02:59:43.605285: Epoch time: 248.7 s 
+2026-10-16 02:59:44.421497:  
+2026-10-16 02:59:44.422212: Epoch 257 
+2026-10-16 02:59:44.422618: Current learning rate: 0.00765 
+2026-10-16 03:03:26.195150: train_loss -0.6761 
+2026-10-16 03:03:26.196090: val_loss -0.2884 
+2026-10-16 03:03:26.196559: Pseudo dice [np.float32(0.8797), np.float32(0.2432), np.float32(0.2534), np.float32(0.1684), np.float32(0.1488), np.float32(0.2403), np.float32(0.2116), np.float32(0.0), np.float32(0.0), np.float32(0.4269), np.float32(0.1986), np.float32(0.2247), np.float32(0.3786), np.float32(0.3779), np.float32(0.0), np.float32(0.0), np.float32(0.1296), np.float32(0.1557), np.float32(0.1647), np.float32(0.1634), np.float32(0.1843), np.float32(0.1454), np.float32(0.352), np.float32(0.2851), np.float32(0.0827), np.float32(0.157), np.float32(0.0), np.float32(0.0), np.float32(0.0365), np.float32(0.3465), np.float32(0.0), np.float32(0.0), np.float32(0.1778), np.float32(0.2302), np.float32(0.1877), np.float32(0.2364)] 
+2026-10-16 03:03:26.197071: Epoch time: 221.78 s 
+2026-10-16 03:03:27.050995:  
+2026-10-16 03:03:27.051661: Epoch 258 
+2026-10-16 03:03:27.052048: Current learning rate: 0.00764 
+2026-10-16 03:07:34.061167: train_loss -0.6646 
+2026-10-16 03:07:34.062173: val_loss -0.3458 
+2026-10-16 03:07:34.062696: Pseudo dice [np.float32(0.8816), np.float32(0.3311), np.float32(0.2914), np.float32(0.2495), np.float32(0.3316), np.float32(0.3011), np.float32(0.2847), np.float32(0.0), np.float32(0.0), np.float32(0.3668), np.float32(0.2205), np.float32(0.2136), np.float32(0.4553), np.float32(0.3976), np.float32(0.0), np.float32(0.0), np.float32(0.2744), np.float32(0.2515), np.float32(0.2595), np.float32(0.2234), np.float32(0.2705), np.float32(0.2592), np.float32(0.4818), np.float32(0.3741), np.float32(0.1725), np.float32(0.2364), np.float32(0.0), np.float32(0.0), np.float32(0.0958), np.float32(0.1212), np.float32(0.0), np.float32(0.0), np.float32(0.3171), np.float32(0.3341), np.float32(0.3059), np.float32(0.3477)] 
+2026-10-16 03:07:34.063158: Epoch time: 247.01 s 
+2026-10-16 03:07:34.909931:  
+2026-10-16 03:07:34.910727: Epoch 259 
+2026-10-16 03:07:34.911010: Current learning rate: 0.00764 
+2026-10-16 03:11:24.334383: train_loss -0.6811 
+2026-10-16 03:11:24.335294: val_loss -0.3899 
+2026-10-16 03:11:24.335898: Pseudo dice [np.float32(0.8923), np.float32(0.4398), np.float32(0.3798), np.float32(0.3781), np.float32(0.4309), np.float32(0.4333), np.float32(0.4114), np.float32(0.0), np.float32(0.0), np.float32(0.3206), np.float32(0.2796), np.float32(0.3468), np.float32(0.4688), np.float32(0.3835), np.float32(0.0), np.float32(0.0), np.float32(0.3623), np.float32(0.291), np.float32(0.3671), np.float32(0.2908), np.float32(0.3619), np.float32(0.2728), np.float32(0.61), np.float32(0.5118), np.float32(0.1856), np.float32(0.2178), np.float32(0.0), np.float32(0.0), np.float32(0.1971), np.float32(0.3282), np.float32(0.0), np.float32(0.0), np.float32(0.3745), np.float32(0.4385), np.float32(0.4234), np.float32(0.4821)] 
+2026-10-16 03:11:24.336482: Epoch time: 229.43 s 
+2026-10-16 03:11:25.130899:  
+2026-10-16 03:11:25.131639: Epoch 260 
+2026-10-16 03:11:25.132013: Current learning rate: 0.00763 
+2026-10-16 03:15:01.728825: train_loss -0.6943 
+2026-10-16 03:15:01.729385: val_loss -0.495 
+2026-10-16 03:15:01.730091: Pseudo dice [np.float32(0.8987), np.float32(0.5424), np.float32(0.6201), np.float32(0.6246), np.float32(0.5271), np.float32(0.6355), np.float32(0.5721), np.float32(0.0), np.float32(0.0), np.float32(0.4492), np.float32(0.5192), np.float32(0.5293), np.float32(0.473), np.float32(0.3706), np.float32(0.0), np.float32(0.0), np.float32(0.5592), np.float32(0.4264), np.float32(0.6086), np.float32(0.5262), np.float32(0.4945), np.float32(0.5499), np.float32(0.6911), np.float32(0.6683), np.float32(0.3884), np.float32(0.3922), np.float32(0.0), np.float32(0.0), np.float32(0.3731), np.float32(0.5916), np.float32(0.0), np.float32(0.0), np.float32(0.4846), np.float32(0.5658), np.float32(0.6187), np.float32(0.6504)] 
+2026-10-16 03:15:01.730882: Epoch time: 216.6 s 
+2026-10-16 03:15:02.870543:  
+2026-10-16 03:15:02.871472: Epoch 261 
+2026-10-16 03:15:02.871927: Current learning rate: 0.00762 
+2026-10-16 03:18:33.466105: train_loss -0.6932 
+2026-10-16 03:18:33.466791: val_loss -0.5627 
+2026-10-16 03:18:33.467901: Pseudo dice [np.float32(0.899), np.float32(0.6223), np.float32(0.7399), np.float32(0.7589), np.float32(0.6779), np.float32(0.8011), np.float32(0.6822), np.float32(0.0), np.float32(0.0), np.float32(0.5232), np.float32(0.6613), np.float32(0.6392), np.float32(0.5908), np.float32(0.5305), np.float32(0.0), np.float32(0.0), np.float32(0.7005), np.float32(0.5559), np.float32(0.6982), np.float32(0.6172), np.float32(0.559), np.float32(0.6394), np.float32(0.8488), np.float32(0.8071), np.float32(0.4566), np.float32(0.4616), np.float32(0.0), np.float32(0.0), np.float32(0.3073), np.float32(0.6716), np.float32(0.0), np.float32(0.0), np.float32(0.6645), np.float32(0.728), np.float32(0.8099), np.float32(0.8342)] 
+2026-10-16 03:18:33.468359: Epoch time: 210.6 s 
+2026-10-16 03:18:34.328486:  
+2026-10-16 03:18:34.329301: Epoch 262 
+2026-10-16 03:18:34.329672: Current learning rate: 0.00761 
+2026-10-16 03:22:34.053967: train_loss -0.6902 
+2026-10-16 03:22:34.054445: val_loss -0.5263 
+2026-10-16 03:22:34.055035: Pseudo dice [np.float32(0.8913), np.float32(0.5731), np.float32(0.6814), np.float32(0.6962), np.float32(0.6107), np.float32(0.7267), np.float32(0.6275), np.float32(0.0), np.float32(0.0), np.float32(0.5562), np.float32(0.5846), np.float32(0.6145), np.float32(0.5105), np.float32(0.4899), np.float32(0.0), np.float32(0.0), np.float32(0.6762), np.float32(0.5468), np.float32(0.6786), np.float32(0.6117), np.float32(0.5613), np.float32(0.5974), np.float32(0.7905), np.float32(0.7545), np.float32(0.4528), np.float32(0.4848), np.float32(0.0), np.float32(0.0), np.float32(0.3129), np.float32(0.5714), np.float32(0.0), np.float32(0.0), np.float32(0.6133), np.float32(0.6846), np.float32(0.7144), np.float32(0.7581)] 
+2026-10-16 03:22:34.055780: Epoch time: 239.73 s 
+2026-10-16 03:22:34.838389:  
+2026-10-16 03:22:34.839165: Epoch 263 
+2026-10-16 03:22:34.839536: Current learning rate: 0.0076 
+2026-10-16 03:26:28.748930: train_loss -0.6668 
+2026-10-16 03:26:28.749846: val_loss -0.5424 
+2026-10-16 03:26:28.750382: Pseudo dice [np.float32(0.904), np.float32(0.6655), np.float32(0.739), np.float32(0.6412), np.float32(0.6931), np.float32(0.6271), np.float32(0.6834), np.float32(0.0), np.float32(0.0), np.float32(0.5627), np.float32(0.6276), np.float32(0.6011), np.float32(0.5553), np.float32(0.5139), np.float32(0.0), np.float32(0.0), np.float32(0.6465), np.float32(0.4726), np.float32(0.6771), np.float32(0.5653), np.float32(0.5483), np.float32(0.5688), np.float32(0.8112), np.float32(0.7743), np.float32(0.466), np.float32(0.4204), np.float32(0.0), np.float32(0.0), np.float32(0.4365), np.float32(0.616), np.float32(0.0), np.float32(0.0), np.float32(0.4219), np.float32(0.3717), np.float32(0.7763), np.float32(0.805)] 
+2026-10-16 03:26:28.750741: Epoch time: 233.91 s 
+2026-10-16 03:26:29.585689:  
+2026-10-16 03:26:29.586534: Epoch 264 
+2026-10-16 03:26:29.586896: Current learning rate: 0.00759 
+2026-10-16 03:30:21.798477: train_loss -0.6468 
+2026-10-16 03:30:21.798997: val_loss -0.4125 
+2026-10-16 03:30:21.799536: Pseudo dice [np.float32(0.8957), np.float32(0.4005), np.float32(0.4573), np.float32(0.4174), np.float32(0.339), np.float32(0.468), np.float32(0.4023), np.float32(0.0), np.float32(0.0), np.float32(0.5784), np.float32(0.367), np.float32(0.3423), np.float32(0.5019), np.float32(0.42), np.float32(0.0), np.float32(0.0), np.float32(0.3096), np.float32(0.2777), np.float32(0.3674), np.float32(0.3192), np.float32(0.3906), np.float32(0.3596), np.float32(0.64), np.float32(0.5233), np.float32(0.2849), np.float32(0.2231), np.float32(0.0), np.float32(0.0), np.float32(0.0737), np.float32(0.4195), np.float32(0.0), np.float32(0.0), np.float32(0.2695), np.float32(0.3919), np.float32(0.4607), np.float32(0.4502)] 
+2026-10-16 03:30:21.799903: Epoch time: 232.21 s 
+2026-10-16 03:30:22.665271:  
+2026-10-16 03:30:22.665888: Epoch 265 
+2026-10-16 03:30:22.666295: Current learning rate: 0.00758 
+2026-10-16 03:34:17.301404: train_loss -0.6556 
+2026-10-16 03:34:17.302385: val_loss -0.4913 
+2026-10-16 03:34:17.302944: Pseudo dice [np.float32(0.8737), np.float32(0.5679), np.float32(0.5616), np.float32(0.5882), np.float32(0.4953), np.float32(0.6248), np.float32(0.5064), np.float32(0.0), np.float32(0.0), np.float32(0.5718), np.float32(0.4805), np.float32(0.5145), np.float32(0.5794), np.float32(0.5262), np.float32(0.0), np.float32(0.0), np.float32(0.4568), np.float32(0.3911), np.float32(0.5075), np.float32(0.3881), np.float32(0.5468), np.float32(0.4653), np.float32(0.6338), np.float32(0.5743), np.float32(0.2951), np.float32(0.3192), np.float32(0.0), np.float32(0.0), np.float32(0.1426), np.float32(0.4618), np.float32(0.0), np.float32(0.0), np.float32(0.3832), np.float32(0.4542), np.float32(0.5568), np.float32(0.5923)] 
+2026-10-16 03:34:17.303265: Epoch time: 234.64 s 
+2026-10-16 03:34:18.154904:  
+2026-10-16 03:34:18.155678: Epoch 266 
+2026-10-16 03:34:18.156137: Current learning rate: 0.00757 
+2026-10-16 03:38:06.862078: train_loss -0.6178 
+2026-10-16 03:38:06.862950: val_loss -0.4088 
+2026-10-16 03:38:06.863718: Pseudo dice [np.float32(0.8873), np.float32(0.4602), np.float32(0.4563), np.float32(0.4335), np.float32(0.3528), np.float32(0.4744), np.float32(0.3642), np.float32(0.0), np.float32(0.0), np.float32(0.4759), np.float32(0.3877), np.float32(0.4224), np.float32(0.568), np.float32(0.5022), np.float32(0.0), np.float32(0.0), np.float32(0.3134), np.float32(0.2126), np.float32(0.3682), np.float32(0.2878), np.float32(0.4257), np.float32(0.3201), np.float32(0.5023), np.float32(0.4158), np.float32(0.1985), np.float32(0.1731), np.float32(0.0), np.float32(0.0), np.float32(0.2389), np.float32(0.5208), np.float32(0.0), np.float32(0.0), np.float32(0.3693), np.float32(0.3129), np.float32(0.4627), np.float32(0.4351)] 
+2026-10-16 03:38:06.864100: Epoch time: 228.71 s 
+2026-10-16 03:38:07.701181:  
+2026-10-16 03:38:07.701772: Epoch 267 
+2026-10-16 03:38:07.702125: Current learning rate: 0.00756 
+2026-10-16 03:42:05.007257: train_loss -0.6441 
+2026-10-16 03:42:05.007918: val_loss -0.4577 
+2026-10-16 03:42:05.008906: Pseudo dice [np.float32(0.8949), np.float32(0.5458), np.float32(0.5388), np.float32(0.499), np.float32(0.4556), np.float32(0.5467), np.float32(0.4723), np.float32(0.0), np.float32(0.0), np.float32(0.3989), np.float32(0.4391), np.float32(0.5136), np.float32(0.4572), np.float32(0.3859), np.float32(0.0), np.float32(0.0), np.float32(0.3917), np.float32(0.3384), np.float32(0.4392), np.float32(0.3588), np.float32(0.4734), np.float32(0.4032), np.float32(0.7893), np.float32(0.7174), np.float32(0.3598), np.float32(0.2928), np.float32(0.0), np.float32(0.0), np.float32(0.4295), np.float32(0.2666), np.float32(0.0), np.float32(0.0), np.float32(0.3484), np.float32(0.3373), np.float32(0.6022), np.float32(0.6428)] 
+2026-10-16 03:42:05.009306: Epoch time: 237.31 s 
+2026-10-16 03:42:05.916913:  
+2026-10-16 03:42:05.917594: Epoch 268 
+2026-10-16 03:42:05.918120: Current learning rate: 0.00755 
+2026-10-16 03:45:30.308331: train_loss -0.6384 
+2026-10-16 03:45:30.309045: val_loss -0.1926 
+2026-10-16 03:45:30.309497: Pseudo dice [np.float32(0.8479), np.float32(0.1678), np.float32(0.079), np.float32(0.0547), np.float32(0.0389), np.float32(0.0748), np.float32(0.0781), np.float32(0.0), np.float32(0.0), np.float32(0.5119), np.float32(0.1079), np.float32(0.1436), np.float32(0.275), np.float32(0.1348), np.float32(0.0), np.float32(0.0), np.float32(0.04), np.float32(0.0372), np.float32(0.091), np.float32(0.1005), np.float32(0.0611), np.float32(0.0678), np.float32(0.1317), np.float32(0.0478), np.float32(0.112), np.float32(0.0249), np.float32(0.0), np.float32(0.0), np.float32(0.0108), np.float32(0.0571), np.float32(0.0), np.float32(0.0), np.float32(0.0137), np.float32(0.0952), np.float32(0.0112), np.float32(0.0314)] 
+2026-10-16 03:45:30.309984: Epoch time: 204.39 s 
+2026-10-16 03:45:31.237691:  
+2026-10-16 03:45:31.238310: Epoch 269 
+2026-10-16 03:45:31.238761: Current learning rate: 0.00754 
+2026-10-16 03:49:16.370985: train_loss -0.5624 
+2026-10-16 03:49:16.371993: val_loss -0.5224 
+2026-10-16 03:49:16.372661: Pseudo dice [np.float32(0.8682), np.float32(0.6042), np.float32(0.6692), np.float32(0.7257), np.float32(0.6895), np.float32(0.7341), np.float32(0.6426), np.float32(0.0), np.float32(0.0), np.float32(0.4613), np.float32(0.6122), np.float32(0.5799), np.float32(0.4886), np.float32(0.402), np.float32(0.0), np.float32(0.0), np.float32(0.6991), np.float32(0.531), np.float32(0.6548), np.float32(0.51), np.float32(0.4969), np.float32(0.5056), np.float32(0.8175), np.float32(0.7778), np.float32(0.4077), np.float32(0.4723), np.float32(0.0), np.float32(0.0), np.float32(0.092), np.float32(0.5432), np.float32(0.0), np.float32(0.0), np.float32(0.5489), np.float32(0.6179), np.float32(0.7528), np.float32(0.7749)] 
+2026-10-16 03:49:16.373131: Epoch time: 225.14 s 
+2026-10-16 03:49:17.223391:  
+2026-10-16 03:49:17.223947: Epoch 270 
+2026-10-16 03:49:17.224211: Current learning rate: 0.00753 
+2026-10-16 03:53:03.070762: train_loss -0.6341 
+2026-10-16 03:53:03.071932: val_loss -0.4448 
+2026-10-16 03:53:03.072472: Pseudo dice [np.float32(0.8646), np.float32(0.4943), np.float32(0.5527), np.float32(0.4726), np.float32(0.4136), np.float32(0.5288), np.float32(0.4584), np.float32(0.0), np.float32(0.0), np.float32(0.5828), np.float32(0.4584), np.float32(0.4464), np.float32(0.5321), np.float32(0.5062), np.float32(0.0), np.float32(0.0), np.float32(0.3197), np.float32(0.2319), np.float32(0.4354), np.float32(0.348), np.float32(0.4677), np.float32(0.306), np.float32(0.7101), np.float32(0.6374), np.float32(0.2148), np.float32(0.3278), np.float32(0.0), np.float32(0.0), np.float32(0.1159), np.float32(0.3498), np.float32(0.0), np.float32(0.0), np.float32(0.3121), np.float32(0.4777), np.float32(0.5678), np.float32(0.542)] 
+2026-10-16 03:53:03.072959: Epoch time: 225.85 s 
+2026-10-16 03:53:03.962097:  
+2026-10-16 03:53:03.962895: Epoch 271 
+2026-10-16 03:53:03.963307: Current learning rate: 0.00752 
+2026-10-16 03:56:39.535185: train_loss -0.6563 
+2026-10-16 03:56:39.535716: val_loss -0.3143 
+2026-10-16 03:56:39.536224: Pseudo dice [np.float32(0.8646), np.float32(0.2825), np.float32(0.2812), np.float32(0.2583), np.float32(0.3087), np.float32(0.2909), np.float32(0.2623), np.float32(0.0), np.float32(0.0), np.float32(0.3709), np.float32(0.1701), np.float32(0.2876), np.float32(0.3674), np.float32(0.2778), np.float32(0.0), np.float32(0.0), np.float32(0.2459), np.float32(0.181), np.float32(0.2585), np.float32(0.2356), np.float32(0.2071), np.float32(0.2105), np.float32(0.4199), np.float32(0.2874), np.float32(0.1535), np.float32(0.1539), np.float32(0.0), np.float32(0.0), np.float32(0.1758), np.float32(0.2204), np.float32(0.0), np.float32(0.0), np.float32(0.2441), np.float32(0.2013), np.float32(0.2761), np.float32(0.2651)] 
+2026-10-16 03:56:39.537236: Epoch time: 215.58 s 
+2026-10-16 03:56:40.520262:  
+2026-10-16 03:56:40.520907: Epoch 272 
+2026-10-16 03:56:40.521284: Current learning rate: 0.00751 
+2026-10-16 04:00:39.623475: train_loss -0.6313 
+2026-10-16 04:00:39.624130: val_loss -0.4341 
+2026-10-16 04:00:39.625031: Pseudo dice [np.float32(0.8823), np.float32(0.4766), np.float32(0.5105), np.float32(0.4851), np.float32(0.4245), np.float32(0.5573), np.float32(0.4506), np.float32(0.0), np.float32(0.0), np.float32(0.5656), np.float32(0.438), np.float32(0.4852), np.float32(0.5228), np.float32(0.4934), np.float32(0.0), np.float32(0.0), np.float32(0.3979), np.float32(0.3898), np.float32(0.4773), np.float32(0.4289), np.float32(0.3714), np.float32(0.371), np.float32(0.501), np.float32(0.4604), np.float32(0.3231), np.float32(0.2864), np.float32(0.0), np.float32(0.0), np.float32(0.1392), np.float32(0.396), np.float32(0.0), np.float32(0.0), np.float32(0.3958), np.float32(0.5137), np.float32(0.4716), np.float32(0.5375)] 
+2026-10-16 04:00:39.625569: Epoch time: 239.11 s 
+2026-10-16 04:00:40.501366:  
+2026-10-16 04:00:40.502180: Epoch 273 
+2026-10-16 04:00:40.502525: Current learning rate: 0.00751 
+2026-10-16 04:04:18.200418: train_loss -0.6821 
+2026-10-16 04:04:18.201442: val_loss -0.5331 
+2026-10-16 04:04:18.202191: Pseudo dice [np.float32(0.9113), np.float32(0.6289), np.float32(0.7099), np.float32(0.7327), np.float32(0.6877), np.float32(0.7658), np.float32(0.669), np.float32(0.0), np.float32(0.0), np.float32(0.4324), np.float32(0.5602), np.float32(0.6172), np.float32(0.5718), np.float32(0.5294), np.float32(0.0), np.float32(0.0), np.float32(0.6893), np.float32(0.5586), np.float32(0.7101), np.float32(0.5791), np.float32(0.554), np.float32(0.5562), np.float32(0.7615), np.float32(0.7224), np.float32(0.5115), np.float32(0.4456), np.float32(0.0), np.float32(0.0), np.float32(0.0891), np.float32(0.5142), np.float32(0.0), np.float32(0.0), np.float32(0.626), np.float32(0.6351), np.float32(0.7301), np.float32(0.7592)] 
+2026-10-16 04:04:18.202688: Epoch time: 217.7 s 
+2026-10-16 04:04:19.076306:  
+2026-10-16 04:04:19.077013: Epoch 274 
+2026-10-16 04:04:19.077567: Current learning rate: 0.0075 
+2026-10-16 04:08:01.037872: train_loss -0.6891 
+2026-10-16 04:08:01.038589: val_loss -0.3698 
+2026-10-16 04:08:01.039363: Pseudo dice [np.float32(0.8691), np.float32(0.3865), np.float32(0.3619), np.float32(0.447), np.float32(0.3999), np.float32(0.3714), np.float32(0.3268), np.float32(0.0), np.float32(0.0), np.float32(0.324), np.float32(0.3592), np.float32(0.4082), np.float32(0.5076), np.float32(0.3915), np.float32(0.0), np.float32(0.0), np.float32(0.3247), np.float32(0.2874), np.float32(0.3034), np.float32(0.2603), np.float32(0.2799), np.float32(0.2683), np.float32(0.4494), np.float32(0.3775), np.float32(0.305), np.float32(0.216), np.float32(0.0), np.float32(0.0), np.float32(0.1549), np.float32(0.4619), np.float32(0.0), np.float32(0.0), np.float32(0.3262), np.float32(0.4031), np.float32(0.4253), np.float32(0.398)] 
+2026-10-16 04:08:01.039740: Epoch time: 221.96 s 
+2026-10-16 04:08:01.879265:  
+2026-10-16 04:08:01.880001: Epoch 275 
+2026-10-16 04:08:01.880368: Current learning rate: 0.00749 
+2026-10-16 04:12:02.900250: train_loss -0.6857 
+2026-10-16 04:12:02.901351: val_loss -0.4885 
+2026-10-16 04:12:02.901978: Pseudo dice [np.float32(0.8816), np.float32(0.543), np.float32(0.5775), np.float32(0.5971), np.float32(0.593), np.float32(0.6521), np.float32(0.5781), np.float32(0.0), np.float32(0.0), np.float32(0.4936), np.float32(0.4987), np.float32(0.5394), np.float32(0.4236), np.float32(0.4118), np.float32(0.0), np.float32(0.0), np.float32(0.6052), np.float32(0.515), np.float32(0.5584), np.float32(0.5082), np.float32(0.4488), np.float32(0.4401), np.float32(0.7349), np.float32(0.706), np.float32(0.4863), np.float32(0.4369), np.float32(0.0), np.float32(0.0), np.float32(0.3587), np.float32(0.4322), np.float32(0.0), np.float32(0.0), np.float32(0.4364), np.float32(0.545), np.float32(0.6526), np.float32(0.6905)] 
+2026-10-16 04:12:02.902567: Epoch time: 241.02 s 
+2026-10-16 04:12:03.809109:  
+2026-10-16 04:12:03.809722: Epoch 276 
+2026-10-16 04:12:03.810188: Current learning rate: 0.00748 
+2026-10-16 04:15:45.182959: train_loss -0.665 
+2026-10-16 04:15:45.183936: val_loss -0.302 
+2026-10-16 04:15:45.184473: Pseudo dice [np.float32(0.866), np.float32(0.3411), np.float32(0.2682), np.float32(0.2391), np.float32(0.3094), np.float32(0.1398), np.float32(0.1382), np.float32(0.0), np.float32(0.0), np.float32(0.3423), np.float32(0.1758), np.float32(0.1856), np.float32(0.4167), np.float32(0.34), np.float32(0.0), np.float32(0.0), np.float32(0.2849), np.float32(0.233), np.float32(0.215), np.float32(0.1956), np.float32(0.244), np.float32(0.2019), np.float32(0.4135), np.float32(0.2656), np.float32(0.163), np.float32(0.1592), np.float32(0.0), np.float32(0.0), np.float32(0.115), np.float32(0.2399), np.float32(0.0), np.float32(0.0), np.float32(0.1969), np.float32(0.2558), np.float32(0.2738), np.float32(0.2429)] 
+2026-10-16 04:15:45.184876: Epoch time: 221.38 s 
+2026-10-16 04:15:46.022060:  
+2026-10-16 04:15:46.022787: Epoch 277 
+2026-10-16 04:15:46.023173: Current learning rate: 0.00747 
+2026-10-16 04:19:47.719415: train_loss -0.6217 
+2026-10-16 04:19:47.720444: val_loss -0.5295 
+2026-10-16 04:19:47.720980: Pseudo dice [np.float32(0.8893), np.float32(0.6135), np.float32(0.7034), np.float32(0.7365), np.float32(0.6958), np.float32(0.7774), np.float32(0.6273), np.float32(0.0), np.float32(0.0), np.float32(0.4541), np.float32(0.6486), np.float32(0.6623), np.float32(0.5503), np.float32(0.4749), np.float32(0.0), np.float32(0.0), np.float32(0.7279), np.float32(0.5269), np.float32(0.6872), np.float32(0.5791), np.float32(0.5072), np.float32(0.605), np.float32(0.7604), np.float32(0.7653), np.float32(0.5171), np.float32(0.411), np.float32(0.0), np.float32(0.0), np.float32(0.0222), np.float32(0.2839), np.float32(0.0), np.float32(0.0), np.float32(0.5298), np.float32(0.6416), np.float32(0.7689), np.float32(0.7876)] 
+2026-10-16 04:19:47.721479: Epoch time: 241.7 s 
+2026-10-16 04:19:48.618801:  
+2026-10-16 04:19:48.619378: Epoch 278 
+2026-10-16 04:19:48.619648: Current learning rate: 0.00746 
+2026-10-16 04:23:51.342124: train_loss -0.6683 
+2026-10-16 04:23:51.342653: val_loss -0.4481 
+2026-10-16 04:23:51.343419: Pseudo dice [np.float32(0.8863), np.float32(0.4744), np.float32(0.5301), np.float32(0.503), np.float32(0.4037), np.float32(0.5591), np.float32(0.4535), np.float32(0.0), np.float32(0.0), np.float32(0.4247), np.float32(0.4196), np.float32(0.4745), np.float32(0.5622), np.float32(0.4788), np.float32(0.0), np.float32(0.0), np.float32(0.3685), np.float32(0.3696), np.float32(0.4031), np.float32(0.3901), np.float32(0.3966), np.float32(0.3753), np.float32(0.596), np.float32(0.5406), np.float32(0.2627), np.float32(0.242), np.float32(0.0), np.float32(0.0), np.float32(0.3095), np.float32(0.5569), np.float32(0.0), np.float32(0.0), np.float32(0.4463), np.float32(0.5634), np.float32(0.5063), np.float32(0.5593)] 
+2026-10-16 04:23:51.343941: Epoch time: 242.73 s 
+2026-10-16 04:23:52.271286:  
+2026-10-16 04:23:52.271942: Epoch 279 
+2026-10-16 04:23:52.272341: Current learning rate: 0.00745 
+2026-10-16 04:27:56.417547: train_loss -0.6382 
+2026-10-16 04:27:56.418574: val_loss -0.4318 
+2026-10-16 04:27:56.419414: Pseudo dice [np.float32(0.8621), np.float32(0.5309), np.float32(0.5202), np.float32(0.5144), np.float32(0.4699), np.float32(0.4863), np.float32(0.4914), np.float32(0.0), np.float32(0.0), np.float32(0.2572), np.float32(0.3962), np.float32(0.28), np.float32(0.5475), np.float32(0.4583), np.float32(0.0), np.float32(0.0), np.float32(0.4458), np.float32(0.3794), np.float32(0.4348), np.float32(0.3589), np.float32(0.4731), np.float32(0.4233), np.float32(0.6744), np.float32(0.5821), np.float32(0.3043), np.float32(0.2842), np.float32(0.0), np.float32(0.0), np.float32(0.2599), np.float32(0.4855), np.float32(0.0), np.float32(0.0), np.float32(0.4555), np.float32(0.5315), np.float32(0.5355), np.float32(0.5372)] 
+2026-10-16 04:27:56.419835: Epoch time: 244.15 s 
+2026-10-16 04:27:57.200180:  
+2026-10-16 04:27:57.201241: Epoch 280 
+2026-10-16 04:27:57.201623: Current learning rate: 0.00744 
+2026-10-16 04:31:47.017343: train_loss -0.6366 
+2026-10-16 04:31:47.018084: val_loss -0.3356 
+2026-10-16 04:31:47.019139: Pseudo dice [np.float32(0.8637), np.float32(0.3055), np.float32(0.2997), np.float32(0.2836), np.float32(0.3002), np.float32(0.3518), np.float32(0.3249), np.float32(0.0), np.float32(0.0), np.float32(0.2987), np.float32(0.225), np.float32(0.2873), np.float32(0.4226), np.float32(0.3482), np.float32(0.0), np.float32(0.0), np.float32(0.26), np.float32(0.2251), np.float32(0.3181), np.float32(0.3082), np.float32(0.2408), np.float32(0.2044), np.float32(0.3704), np.float32(0.2621), np.float32(0.1295), np.float32(0.2259), np.float32(0.0), np.float32(0.0), np.float32(0.1078), np.float32(0.213), np.float32(0.0), np.float32(0.0), np.float32(0.2953), np.float32(0.3222), np.float32(0.3067), np.float32(0.3232)] 
+2026-10-16 04:31:47.019655: Epoch time: 229.82 s 
+2026-10-16 04:31:47.891433:  
+2026-10-16 04:31:47.892096: Epoch 281 
+2026-10-16 04:31:47.892366: Current learning rate: 0.00743 
+2026-10-16 04:35:49.120739: train_loss -0.6684 
+2026-10-16 04:35:49.122362: val_loss -0.5175 
+2026-10-16 04:35:49.123084: Pseudo dice [np.float32(0.8795), np.float32(0.5865), np.float32(0.6826), np.float32(0.6926), np.float32(0.6154), np.float32(0.7214), np.float32(0.625), np.float32(0.0), np.float32(0.0), np.float32(0.4849), np.float32(0.5471), np.float32(0.5997), np.float32(0.492), np.float32(0.4515), np.float32(0.0), np.float32(0.0), np.float32(0.6561), np.float32(0.5015), np.float32(0.6256), np.float32(0.5692), np.float32(0.5203), np.float32(0.5804), np.float32(0.7385), np.float32(0.6518), np.float32(0.4269), np.float32(0.3787), np.float32(0.0), np.float32(0.0), np.float32(0.3682), np.float32(0.4764), np.float32(0.0), np.float32(0.0), np.float32(0.5604), np.float32(0.6407), np.float32(0.7058), np.float32(0.7313)] 
+2026-10-16 04:35:49.123726: Epoch time: 241.23 s 
+2026-10-16 04:35:50.262958:  
+2026-10-16 04:35:50.263844: Epoch 282 
+2026-10-16 04:35:50.264276: Current learning rate: 0.00742 
+2026-10-16 04:39:28.680897: train_loss -0.6807 
+2026-10-16 04:39:28.681675: val_loss -0.4222 
+2026-10-16 04:39:28.682667: Pseudo dice [np.float32(0.8749), np.float32(0.4643), np.float32(0.4464), np.float32(0.4092), np.float32(0.3856), np.float32(0.4675), np.float32(0.4093), np.float32(0.0), np.float32(0.0), np.float32(0.1521), np.float32(0.3535), np.float32(0.3557), np.float32(0.5144), np.float32(0.5069), np.float32(0.0), np.float32(0.0), np.float32(0.3495), np.float32(0.2723), np.float32(0.3684), np.float32(0.3376), np.float32(0.4245), np.float32(0.3543), np.float32(0.5733), np.float32(0.5088), np.float32(0.2061), np.float32(0.1902), np.float32(0.0), np.float32(0.0), np.float32(0.1308), np.float32(0.3735), np.float32(0.0), np.float32(0.0), np.float32(0.4012), np.float32(0.4628), np.float32(0.456), np.float32(0.4488)] 
+2026-10-16 04:39:28.683099: Epoch time: 218.42 s 
+2026-10-16 04:39:29.530853:  
+2026-10-16 04:39:29.531756: Epoch 283 
+2026-10-16 04:39:29.532106: Current learning rate: 0.00741 
+2026-10-16 04:43:12.048331: train_loss -0.6793 
+2026-10-16 04:43:12.049207: val_loss -0.4529 
+2026-10-16 04:43:12.049772: Pseudo dice [np.float32(0.8798), np.float32(0.4778), np.float32(0.5078), np.float32(0.5085), np.float32(0.3369), np.float32(0.5565), np.float32(0.4384), np.float32(0.0), np.float32(0.0), np.float32(0.436), np.float32(0.4693), np.float32(0.4519), np.float32(0.5266), np.float32(0.4441), np.float32(0.0), np.float32(0.0), np.float32(0.2644), np.float32(0.203), np.float32(0.3769), np.float32(0.3239), np.float32(0.4032), np.float32(0.3676), np.float32(0.6062), np.float32(0.5307), np.float32(0.2231), np.float32(0.19), np.float32(0.0), np.float32(0.0), np.float32(0.1616), np.float32(0.4704), np.float32(0.0), np.float32(0.0), np.float32(0.3247), np.float32(0.388), np.float32(0.5234), np.float32(0.5262)] 
+2026-10-16 04:43:12.050186: Epoch time: 222.52 s 
+2026-10-16 04:43:12.943411:  
+2026-10-16 04:43:12.944031: Epoch 284 
+2026-10-16 04:43:12.944515: Current learning rate: 0.0074 
+2026-10-16 04:46:55.051238: train_loss -0.6948 
+2026-10-16 04:46:55.052532: val_loss -0.4397 
+2026-10-16 04:46:55.053184: Pseudo dice [np.float32(0.8891), np.float32(0.4888), np.float32(0.4719), np.float32(0.4573), np.float32(0.3999), np.float32(0.5041), np.float32(0.4169), np.float32(0.0), np.float32(0.0), np.float32(0.5828), np.float32(0.3939), np.float32(0.4138), np.float32(0.4593), np.float32(0.4538), np.float32(0.0), np.float32(0.0), np.float32(0.3198), np.float32(0.2733), np.float32(0.3514), np.float32(0.2919), np.float32(0.4224), np.float32(0.3431), np.float32(0.6215), np.float32(0.5473), np.float32(0.1923), np.float32(0.2385), np.float32(0.0), np.float32(0.0), np.float32(0.1852), np.float32(0.4773), np.float32(0.0), np.float32(0.0), np.float32(0.3612), np.float32(0.4822), np.float32(0.5017), np.float32(0.5434)] 
+2026-10-16 04:46:55.053874: Epoch time: 222.11 s 
+2026-10-16 04:46:55.915521:  
+2026-10-16 04:46:55.916352: Epoch 285 
+2026-10-16 04:46:55.916766: Current learning rate: 0.00739 
+2026-10-16 04:50:48.824573: train_loss -0.6861 
+2026-10-16 04:50:48.825134: val_loss -0.3718 
+2026-10-16 04:50:48.826136: Pseudo dice [np.float32(0.8373), np.float32(0.3866), np.float32(0.4341), np.float32(0.3724), np.float32(0.4), np.float32(0.4071), np.float32(0.4222), np.float32(0.0), np.float32(0.0), np.float32(0.3946), np.float32(0.2695), np.float32(0.3273), np.float32(0.4459), np.float32(0.4063), np.float32(0.0), np.float32(0.0), np.float32(0.3733), np.float32(0.272), np.float32(0.3835), np.float32(0.2566), np.float32(0.289), np.float32(0.3026), np.float32(0.4741), np.float32(0.4558), np.float32(0.2742), np.float32(0.3142), np.float32(0.0), np.float32(0.0), np.float32(0.1587), np.float32(0.3691), np.float32(0.0), np.float32(0.0), np.float32(0.2071), np.float32(0.2619), np.float32(0.3908), np.float32(0.432)] 
+2026-10-16 04:50:48.826702: Epoch time: 232.91 s 
+2026-10-16 04:50:49.705106:  
+2026-10-16 04:50:49.705790: Epoch 286 
+2026-10-16 04:50:49.706246: Current learning rate: 0.00738 
+2026-10-16 04:54:34.734038: train_loss -0.6506 
+2026-10-16 04:54:34.734869: val_loss -0.5122 
+2026-10-16 04:54:34.735656: Pseudo dice [np.float32(0.8819), np.float32(0.5819), np.float32(0.6661), np.float32(0.6807), np.float32(0.5986), np.float32(0.7045), np.float32(0.5976), np.float32(0.0), np.float32(0.0), np.float32(0.5929), np.float32(0.539), np.float32(0.4928), np.float32(0.5052), np.float32(0.4544), np.float32(0.0), np.float32(0.0), np.float32(0.636), np.float32(0.5226), np.float32(0.6207), np.float32(0.5301), np.float32(0.5501), np.float32(0.5637), np.float32(0.5993), np.float32(0.5052), np.float32(0.4721), np.float32(0.4019), np.float32(0.0), np.float32(0.0), np.float32(0.3352), np.float32(0.5713), np.float32(0.0), np.float32(0.0), np.float32(0.5217), np.float32(0.5448), np.float32(0.6674), np.float32(0.6949)] 
+2026-10-16 04:54:34.735958: Epoch time: 225.03 s 
+2026-10-16 04:54:35.694296:  
+2026-10-16 04:54:35.694900: Epoch 287 
+2026-10-16 04:54:35.695238: Current learning rate: 0.00738 
+2026-10-16 04:58:45.117676: train_loss -0.6781 
+2026-10-16 04:58:45.118259: val_loss -0.5183 
+2026-10-16 04:58:45.118599: Pseudo dice [np.float32(0.8978), np.float32(0.591), np.float32(0.6679), np.float32(0.6644), np.float32(0.6026), np.float32(0.7243), np.float32(0.5976), np.float32(0.0), np.float32(0.0), np.float32(0.3168), np.float32(0.5352), np.float32(0.5692), np.float32(0.5704), np.float32(0.4765), np.float32(0.0), np.float32(0.0), np.float32(0.5613), np.float32(0.4684), np.float32(0.594), np.float32(0.5115), np.float32(0.5243), np.float32(0.5683), np.float32(0.7428), np.float32(0.7178), np.float32(0.4378), np.float32(0.4409), np.float32(0.0), np.float32(0.0), np.float32(0.2592), np.float32(0.615), np.float32(0.0), np.float32(0.0), np.float32(0.4659), np.float32(0.6192), np.float32(0.7158), np.float32(0.7415)] 
+2026-10-16 04:58:45.119015: Epoch time: 249.43 s 
+2026-10-16 04:58:45.959106:  
+2026-10-16 04:58:45.959593: Epoch 288 
+2026-10-16 04:58:45.959863: Current learning rate: 0.00737 
+2026-10-16 05:02:28.640097: train_loss -0.6848 
+2026-10-16 05:02:28.640655: val_loss -0.4646 
+2026-10-16 05:02:28.641778: Pseudo dice [np.float32(0.9029), np.float32(0.491), np.float32(0.5051), np.float32(0.5347), np.float32(0.4148), np.float32(0.5985), np.float32(0.5179), np.float32(0.0), np.float32(0.0), np.float32(0.386), np.float32(0.4615), np.float32(0.4567), np.float32(0.493), np.float32(0.3899), np.float32(0.0), np.float32(0.0), np.float32(0.4019), np.float32(0.2867), np.float32(0.4801), np.float32(0.445), np.float32(0.4586), np.float32(0.4472), np.float32(0.6409), np.float32(0.5542), np.float32(0.2636), np.float32(0.2044), np.float32(0.0), np.float32(0.0), np.float32(0.5015), np.float32(0.5597), np.float32(0.0), np.float32(0.0), np.float32(0.4225), np.float32(0.5103), np.float32(0.5584), np.float32(0.5884)] 
+2026-10-16 05:02:28.642415: Epoch time: 222.68 s 
+2026-10-16 05:02:29.553504:  
+2026-10-16 05:02:29.554214: Epoch 289 
+2026-10-16 05:02:29.554638: Current learning rate: 0.00736 
+2026-10-16 05:06:05.040474: train_loss -0.6924 
+2026-10-16 05:06:05.041401: val_loss -0.3843 
+2026-10-16 05:06:05.042089: Pseudo dice [np.float32(0.873), np.float32(0.4004), np.float32(0.3899), np.float32(0.3547), np.float32(0.2986), np.float32(0.4126), np.float32(0.3281), np.float32(0.0), np.float32(0.0), np.float32(0.5534), np.float32(0.3108), np.float32(0.3176), np.float32(0.4823), np.float32(0.4114), np.float32(0.0), np.float32(0.0), np.float32(0.2138), np.float32(0.2009), np.float32(0.2791), np.float32(0.2691), np.float32(0.3857), np.float32(0.2797), np.float32(0.4811), np.float32(0.3956), np.float32(0.2052), np.float32(0.1945), np.float32(0.0), np.float32(0.0), np.float32(0.2236), np.float32(0.3652), np.float32(0.0), np.float32(0.0), np.float32(0.3173), np.float32(0.3656), np.float32(0.4062), np.float32(0.4521)] 
+2026-10-16 05:06:05.042522: Epoch time: 215.49 s 
+2026-10-16 05:06:05.922202:  
+2026-10-16 05:06:05.923043: Epoch 290 
+2026-10-16 05:06:05.923387: Current learning rate: 0.00735 
+2026-10-16 05:09:56.474397: train_loss -0.6849 
+2026-10-16 05:09:56.475257: val_loss -0.3969 
+2026-10-16 05:09:56.475822: Pseudo dice [np.float32(0.8768), np.float32(0.4183), np.float32(0.4324), np.float32(0.4323), np.float32(0.3435), np.float32(0.484), np.float32(0.36), np.float32(0.0), np.float32(0.0), np.float32(0.317), np.float32(0.3437), np.float32(0.3711), np.float32(0.437), np.float32(0.3561), np.float32(0.0), np.float32(0.0), np.float32(0.3434), np.float32(0.2769), np.float32(0.3588), np.float32(0.336), np.float32(0.3959), np.float32(0.336), np.float32(0.4283), np.float32(0.381), np.float32(0.232), np.float32(0.1995), np.float32(0.0), np.float32(0.0), np.float32(0.1359), np.float32(0.3663), np.float32(0.0), np.float32(0.0), np.float32(0.3767), np.float32(0.4476), np.float32(0.4462), np.float32(0.4241)] 
+2026-10-16 05:09:56.476296: Epoch time: 230.55 s 
+2026-10-16 05:09:57.320269:  
+2026-10-16 05:09:57.320986: Epoch 291 
+2026-10-16 05:09:57.321609: Current learning rate: 0.00734 
+2026-10-16 05:13:38.968712: train_loss -0.6567 
+2026-10-16 05:13:38.969556: val_loss -0.6007 
+2026-10-16 05:13:38.970223: Pseudo dice [np.float32(0.8944), np.float32(0.7019), np.float32(0.825), np.float32(0.8315), np.float32(0.8082), np.float32(0.8783), np.float32(0.8209), np.float32(0.0), np.float32(0.0), np.float32(0.5715), np.float32(0.7411), np.float32(0.6946), np.float32(0.5976), np.float32(0.5278), np.float32(0.0), np.float32(0.0), np.float32(0.7865), np.float32(0.6016), np.float32(0.7764), np.float32(0.6302), np.float32(0.5642), np.float32(0.687), np.float32(0.8443), np.float32(0.8273), np.float32(0.4931), np.float32(0.4536), np.float32(0.0), np.float32(0.0), np.float32(0.1708), np.float32(0.6279), np.float32(0.0), np.float32(0.0), np.float32(0.6933), np.float32(0.7204), np.float32(0.7412), np.float32(0.7671)] 
+2026-10-16 05:13:38.970878: Epoch time: 221.65 s 
+2026-10-16 05:13:39.905561:  
+2026-10-16 05:13:39.906335: Epoch 292 
+2026-10-16 05:13:39.906806: Current learning rate: 0.00733 
+2026-10-16 05:17:58.548686: train_loss -0.6437 
+2026-10-16 05:17:58.549642: val_loss -0.5029 
+2026-10-16 05:17:58.550219: Pseudo dice [np.float32(0.8742), np.float32(0.5517), np.float32(0.6981), np.float32(0.7048), np.float32(0.6465), np.float32(0.7204), np.float32(0.6317), np.float32(0.0), np.float32(0.0), np.float32(0.2958), np.float32(0.5633), np.float32(0.5758), np.float32(0.467), np.float32(0.3444), np.float32(0.0), np.float32(0.0), np.float32(0.6744), np.float32(0.4716), np.float32(0.6839), np.float32(0.5766), np.float32(0.5323), np.float32(0.5558), np.float32(0.6735), np.float32(0.6005), np.float32(0.5176), np.float32(0.4144), np.float32(0.0), np.float32(0.0), np.float32(0.3285), np.float32(0.6043), np.float32(0.0), np.float32(0.0), np.float32(0.5962), np.float32(0.6147), np.float32(0.7123), np.float32(0.7264)] 
+2026-10-16 05:17:58.550744: Epoch time: 258.65 s 
+2026-10-16 05:17:59.382968:  
+2026-10-16 05:17:59.383726: Epoch 293 
+2026-10-16 05:17:59.383992: Current learning rate: 0.00732 
+2026-10-16 05:21:49.484037: train_loss -0.6609 
+2026-10-16 05:21:49.484771: val_loss -0.5427 
+2026-10-16 05:21:49.485389: Pseudo dice [np.float32(0.8828), np.float32(0.5833), np.float32(0.7555), np.float32(0.7351), np.float32(0.7138), np.float32(0.7672), np.float32(0.6645), np.float32(0.0), np.float32(0.0), np.float32(0.5423), np.float32(0.5763), np.float32(0.6001), np.float32(0.4585), np.float32(0.4551), np.float32(0.0), np.float32(0.0), np.float32(0.721), np.float32(0.5909), np.float32(0.6772), np.float32(0.5164), np.float32(0.5274), np.float32(0.6237), np.float32(0.7987), np.float32(0.7938), np.float32(0.5229), np.float32(0.4777), np.float32(0.0), np.float32(0.0), np.float32(0.2491), np.float32(0.6922), np.float32(0.0), np.float32(0.0), np.float32(0.6818), np.float32(0.613), np.float32(0.7884), np.float32(0.7979)] 
+2026-10-16 05:21:49.485729: Epoch time: 230.1 s 
+2026-10-16 05:21:50.402868:  
+2026-10-16 05:21:50.403584: Epoch 294 
+2026-10-16 05:21:50.403963: Current learning rate: 0.00731 
+2026-10-16 05:25:48.874496: train_loss -0.6813 
+2026-10-16 05:25:48.875023: val_loss -0.5362 
+2026-10-16 05:25:48.875902: Pseudo dice [np.float32(0.8317), np.float32(0.5533), np.float32(0.7578), np.float32(0.7437), np.float32(0.6631), np.float32(0.7743), np.float32(0.6809), np.float32(0.0), np.float32(0.0), np.float32(0.5544), np.float32(0.6487), np.float32(0.6499), np.float32(0.4574), np.float32(0.4389), np.float32(0.0), np.float32(0.0), np.float32(0.7157), np.float32(0.558), np.float32(0.6737), np.float32(0.5547), np.float32(0.5484), np.float32(0.6655), np.float32(0.7544), np.float32(0.7173), np.float32(0.556), np.float32(0.5086), np.float32(0.0), np.float32(0.0), np.float32(0.0671), np.float32(0.4462), np.float32(0.0), np.float32(0.0), np.float32(0.5429), np.float32(0.6507), np.float32(0.7887), np.float32(0.8031)] 
+2026-10-16 05:25:48.876295: Epoch time: 238.47 s 
+2026-10-16 05:25:49.803588:  
+2026-10-16 05:25:49.804306: Epoch 295 
+2026-10-16 05:25:49.804721: Current learning rate: 0.0073 
+2026-10-16 05:29:36.192986: train_loss -0.6972 
+2026-10-16 05:29:36.193620: val_loss -0.548 
+2026-10-16 05:29:36.194498: Pseudo dice [np.float32(0.8905), np.float32(0.606), np.float32(0.7166), np.float32(0.7175), np.float32(0.661), np.float32(0.7755), np.float32(0.656), np.float32(0.0), np.float32(0.0), np.float32(0.509), np.float32(0.6002), np.float32(0.5984), np.float32(0.5416), np.float32(0.4655), np.float32(0.0), np.float32(0.0), np.float32(0.6855), np.float32(0.5527), np.float32(0.697), np.float32(0.5851), np.float32(0.5456), np.float32(0.641), np.float32(0.8256), np.float32(0.7986), np.float32(0.4942), np.float32(0.524), np.float32(0.0), np.float32(0.0), np.float32(0.3401), np.float32(0.6264), np.float32(0.0), np.float32(0.0), np.float32(0.668), np.float32(0.7196), np.float32(0.7796), np.float32(0.8096)] 
+2026-10-16 05:29:36.195163: Epoch time: 226.39 s 
+2026-10-16 05:29:37.151873:  
+2026-10-16 05:29:37.152612: Epoch 296 
+2026-10-16 05:29:37.153049: Current learning rate: 0.00729 
+2026-10-16 05:33:37.063326: train_loss -0.7031 
+2026-10-16 05:33:37.064169: val_loss -0.5593 
+2026-10-16 05:33:37.064762: Pseudo dice [np.float32(0.8928), np.float32(0.6351), np.float32(0.7252), np.float32(0.7206), np.float32(0.6655), np.float32(0.7823), np.float32(0.6379), np.float32(0.0), np.float32(0.0), np.float32(0.4328), np.float32(0.6236), np.float32(0.6243), np.float32(0.5465), np.float32(0.4322), np.float32(0.0), np.float32(0.0), np.float32(0.6714), np.float32(0.5427), np.float32(0.604), np.float32(0.5323), np.float32(0.5933), np.float32(0.6311), np.float32(0.8759), np.float32(0.8369), np.float32(0.502), np.float32(0.5151), np.float32(0.0), np.float32(0.0), np.float32(0.3481), np.float32(0.6636), np.float32(0.0), np.float32(0.0), np.float32(0.5594), np.float32(0.6875), np.float32(0.8011), np.float32(0.8155)] 
+2026-10-16 05:33:37.064993: Epoch time: 239.91 s 
+2026-10-16 05:33:37.970223:  
+2026-10-16 05:33:37.970871: Epoch 297 
+2026-10-16 05:33:37.971232: Current learning rate: 0.00728 
+2026-10-16 05:37:34.761542: train_loss -0.6831 
+2026-10-16 05:37:34.762186: val_loss -0.4602 
+2026-10-16 05:37:34.763143: Pseudo dice [np.float32(0.8909), np.float32(0.475), np.float32(0.617), np.float32(0.594), np.float32(0.5414), np.float32(0.6076), np.float32(0.5529), np.float32(0.0), np.float32(0.0), np.float32(0.5826), np.float32(0.4693), np.float32(0.472), np.float32(0.3698), np.float32(0.2579), np.float32(0.0), np.float32(0.0), np.float32(0.6311), np.float32(0.4109), np.float32(0.6169), np.float32(0.555), np.float32(0.4327), np.float32(0.5266), np.float32(0.6386), np.float32(0.6391), np.float32(0.5183), np.float32(0.4204), np.float32(0.0), np.float32(0.0), np.float32(0.1898), np.float32(0.261), np.float32(0.0), np.float32(0.0), np.float32(0.5118), np.float32(0.4975), np.float32(0.6429), np.float32(0.6383)] 
+2026-10-16 05:37:34.763646: Epoch time: 236.79 s 
+2026-10-16 05:37:35.665444:  
+2026-10-16 05:37:35.666168: Epoch 298 
+2026-10-16 05:37:35.666457: Current learning rate: 0.00727 
+2026-10-16 05:41:39.708518: train_loss -0.6657 
+2026-10-16 05:41:39.709491: val_loss -0.449 
+2026-10-16 05:41:39.710152: Pseudo dice [np.float32(0.8929), np.float32(0.5075), np.float32(0.5264), np.float32(0.488), np.float32(0.4238), np.float32(0.5389), np.float32(0.4883), np.float32(0.0), np.float32(0.0), np.float32(0.4935), np.float32(0.4053), np.float32(0.4666), np.float32(0.4449), np.float32(0.356), np.float32(0.0), np.float32(0.0), np.float32(0.4088), np.float32(0.3654), np.float32(0.4565), np.float32(0.3924), np.float32(0.4649), np.float32(0.4216), np.float32(0.7009), np.float32(0.6679), np.float32(0.3145), np.float32(0.2535), np.float32(0.0), np.float32(0.0), np.float32(0.1892), np.float32(0.4965), np.float32(0.0), np.float32(0.0), np.float32(0.4491), np.float32(0.4862), np.float32(0.5782), np.float32(0.5999)] 
+2026-10-16 05:41:39.710712: Epoch time: 244.05 s 
+2026-10-16 05:41:40.575069:  
+2026-10-16 05:41:40.576195: Epoch 299 
+2026-10-16 05:41:40.576596: Current learning rate: 0.00726 
+2026-10-16 05:45:22.964923: train_loss -0.6589 
+2026-10-16 05:45:22.965589: val_loss -0.3258 
+2026-10-16 05:45:22.966172: Pseudo dice [np.float32(0.8818), np.float32(0.3131), np.float32(0.2984), np.float32(0.2507), np.float32(0.2576), np.float32(0.287), np.float32(0.2983), np.float32(0.0), np.float32(0.0), np.float32(0.4071), np.float32(0.255), np.float32(0.2618), np.float32(0.4749), np.float32(0.4307), np.float32(0.0), np.float32(0.0), np.float32(0.2792), np.float32(0.2275), np.float32(0.3188), np.float32(0.239), np.float32(0.3054), np.float32(0.2323), np.float32(0.4525), np.float32(0.3304), np.float32(0.1649), np.float32(0.1759), np.float32(0.0), np.float32(0.0), np.float32(0.1801), np.float32(0.3112), np.float32(0.0), np.float32(0.0), np.float32(0.242), np.float32(0.2729), np.float32(0.2834), np.float32(0.3428)] 
+2026-10-16 05:45:22.967124: Epoch time: 222.39 s 
+2026-10-16 05:45:27.179054:  
+2026-10-16 05:45:27.179804: Epoch 300 
+2026-10-16 05:45:27.180244: Current learning rate: 0.00725 
+2026-10-16 05:49:06.145512: train_loss -0.6802 
+2026-10-16 05:49:06.146255: val_loss -0.525 
+2026-10-16 05:49:06.147004: Pseudo dice [np.float32(0.8972), np.float32(0.6067), np.float32(0.6861), np.float32(0.6912), np.float32(0.6254), np.float32(0.745), np.float32(0.6231), np.float32(0.0), np.float32(0.0), np.float32(0.5401), np.float32(0.5988), np.float32(0.5982), np.float32(0.5698), np.float32(0.5084), np.float32(0.0), np.float32(0.0), np.float32(0.6509), np.float32(0.5006), np.float32(0.6214), np.float32(0.505), np.float32(0.5655), np.float32(0.5605), np.float32(0.7665), np.float32(0.7273), np.float32(0.4431), np.float32(0.3869), np.float32(0.0), np.float32(0.0), np.float32(0.2633), np.float32(0.6023), np.float32(0.0), np.float32(0.0), np.float32(0.5396), np.float32(0.5763), np.float32(0.7311), np.float32(0.7514)] 
+2026-10-16 05:49:06.147522: Epoch time: 218.97 s 
+2026-10-16 05:49:06.961521:  
+2026-10-16 05:49:06.962363: Epoch 301 
+2026-10-16 05:49:06.962747: Current learning rate: 0.00724