diff --git a/.gitattributes b/.gitattributes
index 1988b35c43333956b21df59f1bb3ec9cdf2fa97b..66ffd6ec06e8f2ba3f2fbb6899536819136845fc 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png filter=lfs diff=lfs merge=lfs -text
+nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png filter=lfs diff=lfs merge=lfs -text
diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/dataset.json b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/dataset.json
new file mode 100644
index 0000000000000000000000000000000000000000..998fada443cfc594f132bb17b98c2cc25192d896
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/dataset.json
@@ -0,0 +1,46 @@
+{
+    "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"
+}
\ No newline at end of file
diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/dataset_fingerprint.json b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/dataset_fingerprint.json
new file mode 100644
index 0000000000000000000000000000000000000000..e278577ae2fadbf012a75ccc57cd8c1b9c4ef448
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/dataset_fingerprint.json
@@ -0,0 +1,268 @@
+{
+    "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
+        }
+    },
+    "median_relative_size_after_cropping": 1.0,
+    "shapes_after_crop": [
+        [
+            243,
+            327,
+            284
+        ],
+        [
+            203,
+            347,
+            289
+        ],
+        [
+            182,
+            313,
+            281
+        ],
+        [
+            224,
+            431,
+            351
+        ],
+        [
+            184,
+            417,
+            332
+        ],
+        [
+            216,
+            369,
+            286
+        ],
+        [
+            208,
+            321,
+            258
+        ],
+        [
+            204,
+            338,
+            270
+        ],
+        [
+            196,
+            386,
+            295
+        ],
+        [
+            295,
+            334,
+            269
+        ],
+        [
+            311,
+            371,
+            266
+        ],
+        [
+            168,
+            433,
+            326
+        ],
+        [
+            211,
+            328,
+            247
+        ],
+        [
+            203,
+            390,
+            316
+        ],
+        [
+            182,
+            365,
+            306
+        ],
+        [
+            527,
+            430,
+            319
+        ],
+        [
+            202,
+            379,
+            286
+        ],
+        [
+            200,
+            387,
+            290
+        ],
+        [
+            205,
+            345,
+            275
+        ],
+        [
+            224,
+            351,
+            287
+        ],
+        [
+            204,
+            396,
+            338
+        ],
+        [
+            223,
+            366,
+            300
+        ],
+        [
+            206,
+            340,
+            271
+        ],
+        [
+            214,
+            316,
+            274
+        ],
+        [
+            212,
+            360,
+            275
+        ]
+    ],
+    "spacings": [
+        [
+            0.625,
+            0.5078129768371582,
+            0.5078129768371582
+        ],
+        [
+            0.699951171875,
+            0.45703125,
+            0.45703125
+        ],
+        [
+            0.75,
+            0.50390625,
+            0.50390625
+        ],
+        [
+            0.70001220703125,
+            0.40625,
+            0.40625
+        ],
+        [
+            0.75,
+            0.3984375,
+            0.3984375
+        ],
+        [
+            0.75,
+            0.484375,
+            0.484375
+        ],
+        [
+            0.75,
+            0.544921875,
+            0.544921875
+        ],
+        [
+            0.75,
+            0.50390625,
+            0.50390625
+        ],
+        [
+            0.75,
+            0.431640625,
+            0.431640625
+        ],
+        [
+            0.5,
+            0.498046875,
+            0.498046875
+        ],
+        [
+            0.5,
+            0.498046875,
+            0.498046875
+        ],
+        [
+            0.7000007629394531,
+            0.40234375,
+            0.40234375
+        ],
+        [
+            0.75,
+            0.552734375,
+            0.552734375
+        ],
+        [
+            0.75,
+            0.431640625,
+            0.431640625
+        ],
+        [
+            0.75,
+            0.431640625,
+            0.431640625
+        ],
+        [
+            0.29998779296875,
+            0.4375,
+            0.4375
+        ],
+        [
+            0.75,
+            0.482421875,
+            0.482421875
+        ],
+        [
+            0.75,
+            0.4609375,
+            0.4609375
+        ],
+        [
+            0.70001220703125,
+            0.4921875,
+            0.4921875
+        ],
+        [
+            0.70001220703125,
+            0.486328125,
+            0.486328125
+        ],
+        [
+            0.70001220703125,
+            0.419921875,
+            0.419921875
+        ],
+        [
+            0.699951171875,
+            0.451171875,
+            0.451171875
+        ],
+        [
+            0.70001220703125,
+            0.48828125,
+            0.48828125
+        ],
+        [
+            0.70001220703125,
+            0.482421875,
+            0.482421875
+        ],
+        [
+            0.70001220703125,
+            0.486328125,
+            0.486328125
+        ]
+    ]
+}
\ No newline at end of file
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
new file mode 100644
index 0000000000000000000000000000000000000000..647bac8a3a2d74c76dfd4cc0b6213d0f0a5ccc35
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_best.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec265a5e4069931beeff4d6c0eaa4d86a377a6715215022586346bc011b1fdaa
+size 819603011
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
new file mode 100644
index 0000000000000000000000000000000000000000..8e850faf265d76d3611126d4ebe3bd3d9ad32162
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_latest.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ebbc2a75943c78e20cff26bcbe296f2bd9d35e82d579fbe87b5cf793c0d2e1e
+size 819681067
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
new file mode 100644
index 0000000000000000000000000000000000000000..529526b84da83e73f16c6dd698a2972fea151509
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json
@@ -0,0 +1,55 @@
+{
+    "_best_ema": "None",
+    "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>",
+    "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.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",
+    "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": "cuda:0",
+    "disable_checkpointing": "False",
+    "enable_deep_supervision": "True",
+    "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",
+    "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>",
+    "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>",
+    "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>",
+    "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)",
+    "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",
+    "plans_manager": "{'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}}}",
+    "preprocessed_dataset_folder": "/marimo/TopBrain2026/nnUNet_preprocessed/Dataset001_CT/nnUNetPlans_3d_fullres",
+    "preprocessed_dataset_folder_base": "/marimo/TopBrain2026/nnUNet_preprocessed/Dataset001_CT",
+    "probabilistic_oversampling": "False",
+    "save_every": "50",
+    "torch_version": "2.11.0+cu130",
+    "was_initialized": "True",
+    "weight_decay": "3e-05"
+}
\ No newline at end of file
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
new file mode 100644
index 0000000000000000000000000000000000000000..e121c164daaa72811e05f0b7edf1e77fcf62ad32
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:908c4d7e293e3bb3d6191ee4890521f342070d3c6d80a0f2a8f664a4464c5fd8
+size 887460
diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_15_10_11_07.txt b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_15_10_11_07.txt
new file mode 100644
index 0000000000000000000000000000000000000000..7f90c5b35dc03b7a81265bb09d780ab2174782ac
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_15_10_11_07.txt
@@ -0,0 +1,1154 @@
+
+#######################################################################
+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 10:11:09.341023: do_dummy_2d_data_aug: False 
+2026-10-15 10:11:09.342103: Creating new 5-fold cross-validation split... 
+2026-10-15 10:11:09.375782: Desired fold for training: 0 
+2026-10-15 10:11:09.375918: 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 10:11:14.838681: Unable to plot network architecture: 
+2026-10-15 10:11:14.839002: No module named 'hiddenlayer' 
+2026-10-15 10:11:14.880241:  
+2026-10-15 10:11:14.881575: Epoch 0 
+2026-10-15 10:11:14.882620: Current learning rate: 0.01 
+2026-10-15 10:16:01.258272: train_loss 0.2745 
+2026-10-15 10:16:01.259218: val_loss 0.0353 
+2026-10-15 10:16:01.259994: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0)] 
+2026-10-15 10:16:01.260497: Epoch time: 286.39 s 
+2026-10-15 10:16:01.260739: Yayy! New best EMA pseudo Dice: 0.0 
+2026-10-15 10:16:04.441941:  
+2026-10-15 10:16:04.442687: Epoch 1 
+2026-10-15 10:16:04.443163: Current learning rate: 0.00999 
+2026-10-15 10:19:51.781059: train_loss 0.0211 
+2026-10-15 10:19:51.781606: val_loss 0.0169 
+2026-10-15 10:19:51.782359: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0)] 
+2026-10-15 10:19:51.782881: Epoch time: 227.34 s 
+2026-10-15 10:19:52.597464:  
+2026-10-15 10:19:52.598228: Epoch 2 
+2026-10-15 10:19:52.598694: Current learning rate: 0.00998 
+2026-10-15 10:23:49.050312: train_loss 0.0124 
+2026-10-15 10:23:49.051181: val_loss 0.0148 
+2026-10-15 10:23:49.051554: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0)] 
+2026-10-15 10:23:49.051950: Epoch time: 236.45 s 
+2026-10-15 10:23:49.866432:  
+2026-10-15 10:23:49.867146: Epoch 3 
+2026-10-15 10:23:49.867508: Current learning rate: 0.00997 
+2026-10-15 10:27:52.116905: train_loss 0.0075 
+2026-10-15 10:27:52.117396: val_loss 0.0123 
+2026-10-15 10:27:52.117897: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0)] 
+2026-10-15 10:27:52.118715: Epoch time: 242.25 s 
+2026-10-15 10:27:52.860156:  
+2026-10-15 10:27:52.860776: Epoch 4 
+2026-10-15 10:27:52.861158: Current learning rate: 0.00996 
+2026-10-15 10:31:38.679676: train_loss 0.0037 
+2026-10-15 10:31:38.680134: val_loss 0.0076 
+2026-10-15 10:31:38.680773: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0004)] 
+2026-10-15 10:31:38.681248: Epoch time: 225.82 s 
+2026-10-15 10:31:38.681506: Yayy! New best EMA pseudo Dice: 0.0 
+2026-10-15 10:31:40.054143:  
+2026-10-15 10:31:40.055121: Epoch 5 
+2026-10-15 10:31:40.055558: Current learning rate: 0.00995 
+2026-10-15 10:35:40.690847: train_loss 0.0006 
+2026-10-15 10:35:40.691999: val_loss 0.0006 
+2026-10-15 10:35:40.692535: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0076), np.float32(0.0002)] 
+2026-10-15 10:35:40.693151: Epoch time: 240.64 s 
+2026-10-15 10:35:40.693803: Yayy! New best EMA pseudo Dice: 0.0 
+2026-10-15 10:35:42.262830:  
+2026-10-15 10:35:42.263587: Epoch 6 
+2026-10-15 10:35:42.269671: Current learning rate: 0.00995 
+2026-10-15 10:39:37.529706: train_loss -0.0029 
+2026-10-15 10:39:37.530501: val_loss -0.0037 
+2026-10-15 10:39:37.531068: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.032), np.float32(0.0)] 
+2026-10-15 10:39:37.531658: Epoch time: 235.27 s 
+2026-10-15 10:39:37.532104: Yayy! New best EMA pseudo Dice: 9.999999747378752e-05 
+2026-10-15 10:39:40.832937:  
+2026-10-15 10:39:40.833808: Epoch 7 
+2026-10-15 10:39:40.834175: Current learning rate: 0.00994 
+2026-10-15 10:43:22.195490: train_loss -0.0069 
+2026-10-15 10:43:22.196366: val_loss -0.0029 
+2026-10-15 10:43:22.196992: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1306), np.float32(0.0)] 
+2026-10-15 10:43:22.197313: Epoch time: 221.36 s 
+2026-10-15 10:43:22.197501: Yayy! New best EMA pseudo Dice: 0.0005000000237487257 
+2026-10-15 10:43:25.476846:  
+2026-10-15 10:43:25.477607: Epoch 8 
+2026-10-15 10:43:25.477941: Current learning rate: 0.00993 
+2026-10-15 10:47:16.019985: train_loss -0.0088 
+2026-10-15 10:47:16.020628: val_loss -0.0076 
+2026-10-15 10:47:16.021594: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0931), np.float32(0.0)] 
+2026-10-15 10:47:16.022079: Epoch time: 230.55 s 
+2026-10-15 10:47:16.022502: Yayy! New best EMA pseudo Dice: 0.000699999975040555 
+2026-10-15 10:47:17.674738:  
+2026-10-15 10:47:17.675623: Epoch 9 
+2026-10-15 10:47:17.676093: Current learning rate: 0.00992 
+2026-10-15 10:51:17.527801: train_loss -0.011 
+2026-10-15 10:51:17.528640: val_loss -0.0094 
+2026-10-15 10:51:17.529274: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1615), np.float32(0.0)] 
+2026-10-15 10:51:17.529747: Epoch time: 239.86 s 
+2026-10-15 10:51:17.530054: Yayy! New best EMA pseudo Dice: 0.0010999999940395355 
+2026-10-15 10:51:18.883509:  
+2026-10-15 10:51:18.884310: Epoch 10 
+2026-10-15 10:51:18.884722: Current learning rate: 0.00991 
+2026-10-15 10:55:33.185403: train_loss -0.0159 
+2026-10-15 10:55:33.186226: val_loss -0.0163 
+2026-10-15 10:55:33.187055: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2079), np.float32(0.0)] 
+2026-10-15 10:55:33.187535: Epoch time: 254.31 s 
+2026-10-15 10:55:33.187915: Yayy! New best EMA pseudo Dice: 0.001500000013038516 
+2026-10-15 10:55:34.597394:  
+2026-10-15 10:55:34.598055: Epoch 11 
+2026-10-15 10:55:34.598585: Current learning rate: 0.0099 
+2026-10-15 10:59:39.948561: train_loss -0.0201 
+2026-10-15 10:59:39.949459: val_loss -0.0199 
+2026-10-15 10:59:39.950181: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2032), np.float32(0.0)] 
+2026-10-15 10:59:39.950550: Epoch time: 245.35 s 
+2026-10-15 10:59:39.950726: Yayy! New best EMA pseudo Dice: 0.0019000000320374966 
+2026-10-15 10:59:41.512831:  
+2026-10-15 10:59:41.513554: Epoch 12 
+2026-10-15 10:59:41.516731: Current learning rate: 0.00989 
+2026-10-15 11:03:53.471608: train_loss -0.0218 
+2026-10-15 11:03:53.472306: val_loss -0.0222 
+2026-10-15 11:03:53.473329: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.251), np.float32(0.0)] 
+2026-10-15 11:03:53.473911: Epoch time: 251.96 s 
+2026-10-15 11:03:53.474272: Yayy! New best EMA pseudo Dice: 0.002400000113993883 
+2026-10-15 11:03:54.866746:  
+2026-10-15 11:03:54.867610: Epoch 13 
+2026-10-15 11:03:54.868031: Current learning rate: 0.00988 
+2026-10-15 11:08:01.309684: train_loss -0.027 
+2026-10-15 11:08:01.310196: val_loss -0.0239 
+2026-10-15 11:08:01.310801: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.271), np.float32(0.0)] 
+2026-10-15 11:08:01.311882: Epoch time: 246.45 s 
+2026-10-15 11:08:01.312250: Yayy! New best EMA pseudo Dice: 0.003000000026077032 
+2026-10-15 11:08:02.807968:  
+2026-10-15 11:08:02.808764: Epoch 14 
+2026-10-15 11:08:02.809280: Current learning rate: 0.00987 
+2026-10-15 11:12:09.532907: train_loss -0.0298 
+2026-10-15 11:12:09.533835: val_loss -0.0278 
+2026-10-15 11:12:09.534362: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0035), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0006), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3234), np.float32(0.0)] 
+2026-10-15 11:12:09.534731: Epoch time: 246.73 s 
+2026-10-15 11:12:09.535079: Yayy! New best EMA pseudo Dice: 0.003599999938160181 
+2026-10-15 11:12:11.110775:  
+2026-10-15 11:12:11.112599: Epoch 15 
+2026-10-15 11:12:11.113054: Current learning rate: 0.00986 
+2026-10-15 11:16:05.799047: train_loss -0.0327 
+2026-10-15 11:16:05.800213: val_loss -0.0317 
+2026-10-15 11:16:05.801149: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(1e-04), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0277), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3193), np.float32(0.0)] 
+2026-10-15 11:16:05.802058: Epoch time: 234.69 s 
+2026-10-15 11:16:05.803288: Yayy! New best EMA pseudo Dice: 0.00419999985024333 
+2026-10-15 11:16:07.327223:  
+2026-10-15 11:16:07.327946: Epoch 16 
+2026-10-15 11:16:07.328338: Current learning rate: 0.00986 
+2026-10-15 11:20:13.101494: train_loss -0.0371 
+2026-10-15 11:20:13.102301: val_loss -0.0371 
+2026-10-15 11:20:13.102952: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1092), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3267), np.float32(0.0)] 
+2026-10-15 11:20:13.103506: Epoch time: 245.78 s 
+2026-10-15 11:20:13.104112: Yayy! New best EMA pseudo Dice: 0.004999999888241291 
+2026-10-15 11:20:14.668957:  
+2026-10-15 11:20:14.669819: Epoch 17 
+2026-10-15 11:20:14.673756: Current learning rate: 0.00985 
+2026-10-15 11:24:13.680163: train_loss -0.0416 
+2026-10-15 11:24:13.680981: val_loss -0.0417 
+2026-10-15 11:24:13.681422: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1132), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3571), np.float32(0.0)] 
+2026-10-15 11:24:13.681835: Epoch time: 239.01 s 
+2026-10-15 11:24:13.682098: Yayy! New best EMA pseudo Dice: 0.005799999926239252 
+2026-10-15 11:24:15.220016:  
+2026-10-15 11:24:15.220840: Epoch 18 
+2026-10-15 11:24:15.221201: Current learning rate: 0.00984 
+2026-10-15 11:28:09.925040: train_loss -0.0503 
+2026-10-15 11:28:09.925706: val_loss -0.0502 
+2026-10-15 11:28:09.926513: Pseudo dice [np.float32(0.0), np.float32(0.0), np.float32(0.0462), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0953), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4076), np.float32(0.0)] 
+2026-10-15 11:28:09.927138: Epoch time: 234.71 s 
+2026-10-15 11:28:09.927672: Yayy! New best EMA pseudo Dice: 0.0066999997943639755 
+2026-10-15 11:28:11.298033:  
+2026-10-15 11:28:11.298724: Epoch 19 
+2026-10-15 11:28:11.299104: Current learning rate: 0.00983 
+2026-10-15 11:32:04.989652: train_loss -0.0548 
+2026-10-15 11:32:04.990674: val_loss -0.0509 
+2026-10-15 11:32:04.991344: Pseudo dice [np.float32(0.0), np.float32(0.0991), np.float32(0.0838), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0524), np.float32(0.0), np.float32(0.0122), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3651), np.float32(0.0)] 
+2026-10-15 11:32:04.991879: Epoch time: 233.69 s 
+2026-10-15 11:32:04.992092: Yayy! New best EMA pseudo Dice: 0.007799999788403511 
+2026-10-15 11:32:08.577351:  
+2026-10-15 11:32:08.578137: Epoch 20 
+2026-10-15 11:32:08.578523: Current learning rate: 0.00982 
+2026-10-15 11:36:03.509032: train_loss -0.0587 
+2026-10-15 11:36:03.509677: val_loss -0.0547 
+2026-10-15 11:36:03.510311: Pseudo dice [np.float32(0.0), np.float32(0.1064), np.float32(0.1744), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0115), np.float32(0.0), np.float32(0.0612), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.364), np.float32(0.0)] 
+2026-10-15 11:36:03.511117: Epoch time: 234.93 s 
+2026-10-15 11:36:03.511485: Yayy! New best EMA pseudo Dice: 0.008999999612569809 
+2026-10-15 11:36:05.169315:  
+2026-10-15 11:36:05.170258: Epoch 21 
+2026-10-15 11:36:05.170926: Current learning rate: 0.00981 
+2026-10-15 11:40:17.290807: train_loss -0.0683 
+2026-10-15 11:40:17.292532: val_loss -0.0681 
+2026-10-15 11:40:17.293083: Pseudo dice [np.float32(0.4927), np.float32(0.0173), np.float32(0.1983), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.001), np.float32(0.0), np.float32(0.0347), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4313), np.float32(0.0)] 
+2026-10-15 11:40:17.293415: Epoch time: 252.12 s 
+2026-10-15 11:40:17.293861: Yayy! New best EMA pseudo Dice: 0.011300000362098217 
+2026-10-15 11:40:18.803432:  
+2026-10-15 11:40:18.804014: Epoch 22 
+2026-10-15 11:40:18.804400: Current learning rate: 0.0098 
+2026-10-15 11:44:00.040262: train_loss -0.0779 
+2026-10-15 11:44:00.040771: val_loss -0.0817 
+2026-10-15 11:44:00.041385: Pseudo dice [np.float32(0.6964), np.float32(0.0448), np.float32(0.2104), np.float32(0.0), np.float32(1e-04), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(1e-04), np.float32(0.0), np.float32(0.0), np.float32(0.0004), np.float32(0.0), np.float32(0.0443), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5239), np.float32(0.0064)] 
+2026-10-15 11:44:00.042296: Epoch time: 221.24 s 
+2026-10-15 11:44:00.042699: Yayy! New best EMA pseudo Dice: 0.014399999752640724 
+2026-10-15 11:44:01.344092:  
+2026-10-15 11:44:01.344841: Epoch 23 
+2026-10-15 11:44:01.345269: Current learning rate: 0.00979 
+2026-10-15 11:47:58.058422: train_loss -0.0906 
+2026-10-15 11:47:58.059337: val_loss -0.0712 
+2026-10-15 11:47:58.060171: Pseudo dice [np.float32(0.7674), np.float32(0.0871), np.float32(0.2407), np.float32(0.0576), np.float32(0.0104), np.float32(0.0), np.float32(1e-04), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(1e-04), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0145), np.float32(0.0), np.float32(0.1505), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0252), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5205), np.float32(0.0016)] 
+2026-10-15 11:47:58.060464: Epoch time: 236.72 s 
+2026-10-15 11:47:58.060899: Yayy! New best EMA pseudo Dice: 0.018200000748038292 
+2026-10-15 11:47:59.446013:  
+2026-10-15 11:47:59.446846: Epoch 24 
+2026-10-15 11:47:59.447252: Current learning rate: 0.00978 
+2026-10-15 11:51:30.095692: train_loss -0.1001 
+2026-10-15 11:51:30.096201: val_loss -0.1008 
+2026-10-15 11:51:30.096712: Pseudo dice [np.float32(0.8156), np.float32(0.2212), np.float32(0.2375), np.float32(0.1137), np.float32(1e-04), np.float32(0.298), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1831), np.float32(0.0005), np.float32(0.0), np.float32(0.0012), np.float32(0.0), np.float32(0.0), np.float32(0.2517), np.float32(0.0), np.float32(0.0004), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3092), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5089), np.float32(0.0137)] 
+2026-10-15 11:51:30.097330: Epoch time: 210.65 s 
+2026-10-15 11:51:30.097556: Yayy! New best EMA pseudo Dice: 0.02459999918937683 
+2026-10-15 11:51:31.412577:  
+2026-10-15 11:51:31.413239: Epoch 25 
+2026-10-15 11:51:31.413760: Current learning rate: 0.00977 
+2026-10-15 11:55:29.507211: train_loss -0.1246 
+2026-10-15 11:55:29.508092: val_loss -0.1039 
+2026-10-15 11:55:29.508818: Pseudo dice [np.float32(0.7783), np.float32(0.2999), np.float32(0.3236), np.float32(0.3595), np.float32(0.3031), np.float32(0.2865), np.float32(0.0113), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3432), np.float32(0.2601), np.float32(0.0002), np.float32(1e-04), np.float32(0.0), np.float32(0.0), np.float32(0.3089), np.float32(0.0), np.float32(0.0288), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0525), np.float32(0.3625), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4517), np.float32(0.1443)] 
+2026-10-15 11:55:29.509386: Epoch time: 238.1 s 
+2026-10-15 11:55:29.509832: Yayy! New best EMA pseudo Dice: 0.0340999998152256 
+2026-10-15 11:55:32.715665:  
+2026-10-15 11:55:32.716703: Epoch 26 
+2026-10-15 11:55:32.717012: Current learning rate: 0.00977 
+2026-10-15 11:59:30.838586: train_loss -0.1471 
+2026-10-15 11:59:30.839328: val_loss -0.1169 
+2026-10-15 11:59:30.839773: Pseudo dice [np.float32(0.858), np.float32(0.325), np.float32(0.2772), np.float32(0.3865), np.float32(0.4542), np.float32(0.203), np.float32(0.0637), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3499), np.float32(0.3962), np.float32(0.0052), np.float32(0.0018), np.float32(0.0), np.float32(0.0), np.float32(0.355), np.float32(0.0), np.float32(0.0031), np.float32(0.0008), np.float32(0.0), np.float32(0.0), np.float32(0.3446), np.float32(0.1392), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4254), np.float32(0.3617)] 
+2026-10-15 11:59:30.840139: Epoch time: 238.13 s 
+2026-10-15 11:59:30.840509: Yayy! New best EMA pseudo Dice: 0.04450000077486038 
+2026-10-15 11:59:32.206273:  
+2026-10-15 11:59:32.207080: Epoch 27 
+2026-10-15 11:59:32.207517: Current learning rate: 0.00976 
+2026-10-15 12:03:23.494396: train_loss -0.1703 
+2026-10-15 12:03:23.495362: val_loss -0.0684 
+2026-10-15 12:03:23.495916: Pseudo dice [np.float32(0.8333), np.float32(0.2679), np.float32(0.0779), np.float32(0.0755), np.float32(0.1412), np.float32(0.1238), np.float32(0.0826), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0227), np.float32(0.3485), np.float32(0.1456), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2891), np.float32(0.0), np.float32(0.1608), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3758), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1442), np.float32(0.0508)] 
+2026-10-15 12:03:23.496201: Epoch time: 231.29 s 
+2026-10-15 12:03:23.496512: Yayy! New best EMA pseudo Dice: 0.048700001090765 
+2026-10-15 12:03:24.803399:  
+2026-10-15 12:03:24.804149: Epoch 28 
+2026-10-15 12:03:24.804926: Current learning rate: 0.00975 
+2026-10-15 12:07:37.648367: train_loss -0.1784 
+2026-10-15 12:07:37.649101: val_loss -0.0703 
+2026-10-15 12:07:37.649817: Pseudo dice [np.float32(0.8041), np.float32(0.1788), np.float32(0.1452), np.float32(0.1124), np.float32(0.1704), np.float32(0.1938), np.float32(0.1344), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1397), np.float32(0.2749), np.float32(0.195), np.float32(0.0279), np.float32(0.0), np.float32(0.0), np.float32(0.0858), np.float32(0.0), np.float32(0.3387), np.float32(0.0423), np.float32(0.0), np.float32(0.0), np.float32(0.211), np.float32(0.3896), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0949), np.float32(0.1434)] 
+2026-10-15 12:07:37.650234: Epoch time: 252.85 s 
+2026-10-15 12:07:37.650665: Yayy! New best EMA pseudo Dice: 0.054099999368190765 
+2026-10-15 12:07:39.266299:  
+2026-10-15 12:07:39.266870: Epoch 29 
+2026-10-15 12:07:39.267253: Current learning rate: 0.00974 
+2026-10-15 12:11:10.021181: train_loss -0.1821 
+2026-10-15 12:11:10.022110: val_loss -0.0826 
+2026-10-15 12:11:10.022926: Pseudo dice [np.float32(0.834), np.float32(0.206), np.float32(0.2717), np.float32(0.1475), np.float32(0.1039), np.float32(0.1688), np.float32(0.0546), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0709), np.float32(0.4211), np.float32(0.2752), np.float32(0.1655), np.float32(0.0), np.float32(0.0), np.float32(0.2783), np.float32(0.0), np.float32(0.1691), np.float32(0.1079), np.float32(0.0), np.float32(0.0015), np.float32(0.1619), np.float32(0.4656), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1125), np.float32(0.1171)] 
+2026-10-15 12:11:10.023453: Epoch time: 210.76 s 
+2026-10-15 12:11:10.023817: Yayy! New best EMA pseudo Dice: 0.06019999831914902 
+2026-10-15 12:11:11.323348:  
+2026-10-15 12:11:11.324088: Epoch 30 
+2026-10-15 12:11:11.324512: Current learning rate: 0.00973 
+2026-10-15 12:15:05.778519: train_loss -0.1939 
+2026-10-15 12:15:05.779112: val_loss -0.0829 
+2026-10-15 12:15:05.779643: Pseudo dice [np.float32(0.8719), np.float32(0.2503), np.float32(0.1317), np.float32(0.1472), np.float32(0.2443), np.float32(0.049), np.float32(0.0278), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.199), np.float32(0.1826), np.float32(0.3502), np.float32(0.2595), np.float32(0.0), np.float32(0.0), np.float32(0.2328), np.float32(0.0), np.float32(0.1057), np.float32(0.1828), np.float32(0.0011), np.float32(0.0325), np.float32(0.1434), np.float32(0.5215), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1313), np.float32(0.0693)] 
+2026-10-15 12:15:05.780376: Epoch time: 234.46 s 
+2026-10-15 12:15:05.780764: Yayy! New best EMA pseudo Dice: 0.06560000032186508 
+2026-10-15 12:15:08.730918:  
+2026-10-15 12:15:08.731772: Epoch 31 
+2026-10-15 12:15:08.732052: Current learning rate: 0.00972 
+2026-10-15 12:19:18.638136: train_loss -0.2184 
+2026-10-15 12:19:18.639155: val_loss -0.1399 
+2026-10-15 12:19:18.639818: Pseudo dice [np.float32(0.8466), np.float32(0.3704), np.float32(0.4014), np.float32(0.3087), np.float32(0.2324), np.float32(0.4202), np.float32(0.3248), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3054), np.float32(0.4069), np.float32(0.4707), np.float32(0.0057), np.float32(0.0), np.float32(0.0), np.float32(0.2485), np.float32(0.1361), np.float32(0.1839), np.float32(0.1851), np.float32(0.172), np.float32(0.0792), np.float32(0.2907), np.float32(0.4821), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3498), np.float32(0.4155)] 
+2026-10-15 12:19:18.640097: Epoch time: 249.91 s 
+2026-10-15 12:19:18.640522: Yayy! New best EMA pseudo Dice: 0.07750000059604645 
+2026-10-15 12:19:20.022254:  
+2026-10-15 12:19:20.023008: Epoch 32 
+2026-10-15 12:19:20.023346: Current learning rate: 0.00971 
+2026-10-15 12:23:35.684535: train_loss -0.2261 
+2026-10-15 12:23:35.685210: val_loss -0.1384 
+2026-10-15 12:23:35.685723: Pseudo dice [np.float32(0.8508), np.float32(0.3183), np.float32(0.3124), np.float32(0.3232), np.float32(0.1592), np.float32(0.3851), np.float32(0.2325), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3312), np.float32(0.3848), np.float32(0.408), np.float32(0.1398), np.float32(0.0), np.float32(0.0), np.float32(0.2403), np.float32(0.069), np.float32(0.2205), np.float32(0.2206), np.float32(0.2466), np.float32(0.1375), np.float32(0.3359), np.float32(0.546), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.001), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3453), np.float32(0.3356)] 
+2026-10-15 12:23:35.686637: Epoch time: 255.66 s 
+2026-10-15 12:23:35.686988: Yayy! New best EMA pseudo Dice: 0.08789999783039093 
+2026-10-15 12:23:37.016361:  
+2026-10-15 12:23:37.017082: Epoch 33 
+2026-10-15 12:23:37.017547: Current learning rate: 0.0097 
+2026-10-15 12:27:24.864258: train_loss -0.2297 
+2026-10-15 12:27:24.864988: val_loss -0.1555 
+2026-10-15 12:27:24.865531: Pseudo dice [np.float32(0.84), np.float32(0.4699), np.float32(0.3849), np.float32(0.4438), np.float32(0.4451), np.float32(0.3623), np.float32(0.1922), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3535), np.float32(0.3843), np.float32(0.2803), np.float32(0.2527), np.float32(0.0), np.float32(0.0), np.float32(0.4295), np.float32(0.135), np.float32(0.2539), np.float32(0.2745), np.float32(0.1581), np.float32(0.3088), np.float32(0.4973), np.float32(0.4094), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0028), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4025), np.float32(0.3319)] 
+2026-10-15 12:27:24.866209: Epoch time: 227.85 s 
+2026-10-15 12:27:24.866499: Yayy! New best EMA pseudo Dice: 0.10029999911785126 
+2026-10-15 12:27:26.254089:  
+2026-10-15 12:27:26.254799: Epoch 34 
+2026-10-15 12:27:26.255108: Current learning rate: 0.00969 
+2026-10-15 12:31:32.903912: train_loss -0.2521 
+2026-10-15 12:31:32.904875: val_loss -0.1223 
+2026-10-15 12:31:32.905638: Pseudo dice [np.float32(0.873), np.float32(0.2658), np.float32(0.2761), np.float32(0.2388), np.float32(0.318), np.float32(0.2901), np.float32(0.2128), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2018), np.float32(0.2617), np.float32(0.4139), np.float32(0.1247), np.float32(0.0), np.float32(0.0), np.float32(0.271), np.float32(0.2906), np.float32(0.3116), np.float32(0.2283), np.float32(0.3005), np.float32(0.1295), np.float32(0.5039), np.float32(0.1516), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1972), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2663), np.float32(0.3044)] 
+2026-10-15 12:31:32.906114: Epoch time: 246.65 s 
+2026-10-15 12:31:32.906454: Yayy! New best EMA pseudo Dice: 0.10809999704360962 
+2026-10-15 12:31:35.883035:  
+2026-10-15 12:31:35.883509: Epoch 35 
+2026-10-15 12:31:35.883853: Current learning rate: 0.00968 
+2026-10-15 12:35:33.186227: train_loss -0.257 
+2026-10-15 12:35:33.187224: val_loss -0.0741 
+2026-10-15 12:35:33.187923: Pseudo dice [np.float32(0.8648), np.float32(0.1547), np.float32(0.0768), np.float32(0.1205), np.float32(0.0513), np.float32(0.0767), np.float32(0.0365), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1124), np.float32(0.2277), np.float32(0.3757), np.float32(0.1747), np.float32(0.0), np.float32(0.0), np.float32(0.0542), np.float32(0.1048), np.float32(0.1529), np.float32(0.2755), np.float32(0.1), np.float32(0.1376), np.float32(0.4464), np.float32(0.2545), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0433), np.float32(0.1451), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1124), np.float32(0.0962)] 
+2026-10-15 12:35:33.188427: Epoch time: 237.31 s 
+2026-10-15 12:35:33.188773: Yayy! New best EMA pseudo Dice: 0.10899999737739563 
+2026-10-15 12:35:34.634643:  
+2026-10-15 12:35:34.635380: Epoch 36 
+2026-10-15 12:35:34.635717: Current learning rate: 0.00968 
+2026-10-15 12:39:37.052629: train_loss -0.2602 
+2026-10-15 12:39:37.053118: val_loss -0.1114 
+2026-10-15 12:39:37.053594: Pseudo dice [np.float32(0.8904), np.float32(0.2764), np.float32(0.2264), np.float32(0.1649), np.float32(0.2785), np.float32(0.1701), np.float32(0.2505), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.161), np.float32(0.2361), np.float32(0.3282), np.float32(0.2303), np.float32(0.0), np.float32(0.0), np.float32(0.2702), np.float32(0.2673), np.float32(0.2301), np.float32(0.1727), np.float32(0.2526), np.float32(0.152), np.float32(0.3021), np.float32(0.3855), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0105), np.float32(0.2652), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2198), np.float32(0.2164)] 
+2026-10-15 12:39:37.054320: Epoch time: 242.42 s 
+2026-10-15 12:39:37.054719: Yayy! New best EMA pseudo Dice: 0.11460000276565552 
+2026-10-15 12:39:38.384887:  
+2026-10-15 12:39:38.385702: Epoch 37 
+2026-10-15 12:39:38.385954: Current learning rate: 0.00967 
+2026-10-15 12:43:18.879138: train_loss -0.2837 
+2026-10-15 12:43:18.879973: val_loss -0.2249 
+2026-10-15 12:43:18.880702: Pseudo dice [np.float32(0.8736), np.float32(0.5133), np.float32(0.5804), np.float32(0.56), np.float32(0.5278), np.float32(0.5805), np.float32(0.5648), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3993), np.float32(0.4933), np.float32(0.3615), np.float32(0.3365), np.float32(0.0), np.float32(0.0), np.float32(0.4838), np.float32(0.3603), np.float32(0.3585), np.float32(0.3965), np.float32(0.3345), np.float32(0.4687), np.float32(0.5585), np.float32(0.6662), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2144), np.float32(0.3943), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6209), np.float32(0.5761)] 
+2026-10-15 12:43:18.881135: Epoch time: 220.5 s 
+2026-10-15 12:43:18.881402: Yayy! New best EMA pseudo Dice: 0.13429999351501465 
+2026-10-15 12:43:20.233273:  
+2026-10-15 12:43:20.233890: Epoch 38 
+2026-10-15 12:43:20.234128: Current learning rate: 0.00966 
+2026-10-15 12:47:04.170926: train_loss -0.2841 
+2026-10-15 12:47:04.171515: val_loss -0.0896 
+2026-10-15 12:47:04.172497: Pseudo dice [np.float32(0.8591), np.float32(0.2754), np.float32(0.1691), np.float32(0.1491), np.float32(0.226), np.float32(0.1832), np.float32(0.1973), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1472), np.float32(0.2572), np.float32(0.4401), np.float32(0.0571), np.float32(0.0), np.float32(0.0), np.float32(0.1837), np.float32(0.2535), np.float32(0.1524), np.float32(0.1069), np.float32(0.1971), np.float32(0.202), np.float32(0.3057), np.float32(0.3524), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0563), np.float32(0.2898), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1723), np.float32(0.1411)] 
+2026-10-15 12:47:04.173401: Epoch time: 223.94 s 
+2026-10-15 12:47:04.176063: Yayy! New best EMA pseudo Dice: 0.13580000400543213 
+2026-10-15 12:47:05.576128:  
+2026-10-15 12:47:05.576854: Epoch 39 
+2026-10-15 12:47:05.577054: Current learning rate: 0.00965 
+2026-10-15 12:51:01.069834: train_loss -0.2892 
+2026-10-15 12:51:01.070539: val_loss -0.1778 
+2026-10-15 12:51:01.071593: Pseudo dice [np.float32(0.876), np.float32(0.4319), np.float32(0.4427), np.float32(0.386), np.float32(0.3738), np.float32(0.463), np.float32(0.3323), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3096), np.float32(0.3588), np.float32(0.4327), np.float32(0.2862), np.float32(0.0), np.float32(0.0), np.float32(0.2855), np.float32(0.2498), np.float32(0.2533), np.float32(0.2842), np.float32(0.2752), np.float32(0.3466), np.float32(0.4814), np.float32(0.4695), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.132), np.float32(0.3341), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4606), np.float32(0.423)] 
+2026-10-15 12:51:01.072093: Epoch time: 235.5 s 
+2026-10-15 12:51:01.072396: Yayy! New best EMA pseudo Dice: 0.14640000462532043 
+2026-10-15 12:51:04.056952:  
+2026-10-15 12:51:04.057587: Epoch 40 
+2026-10-15 12:51:04.057907: Current learning rate: 0.00964 
+2026-10-15 12:55:31.969077: train_loss -0.3062 
+2026-10-15 12:55:31.969590: val_loss -0.1423 
+2026-10-15 12:55:31.970170: Pseudo dice [np.float32(0.8679), np.float32(0.3589), np.float32(0.3154), np.float32(0.3211), np.float32(0.3897), np.float32(0.3179), np.float32(0.3555), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2405), np.float32(0.2811), np.float32(0.2563), np.float32(0.3721), np.float32(0.0), np.float32(0.0), np.float32(0.3481), np.float32(0.2585), np.float32(0.3292), np.float32(0.2435), np.float32(0.2682), np.float32(0.2203), np.float32(0.4342), np.float32(0.3295), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1468), np.float32(0.248), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3769), np.float32(0.3842)] 
+2026-10-15 12:55:31.970768: Epoch time: 267.91 s 
+2026-10-15 12:55:31.971172: Yayy! New best EMA pseudo Dice: 0.15299999713897705 
+2026-10-15 12:55:34.117765:  
+2026-10-15 12:55:34.118560: Epoch 41 
+2026-10-15 12:55:34.118897: Current learning rate: 0.00963 
+2026-10-15 12:59:12.433598: train_loss -0.2675 
+2026-10-15 12:59:12.434042: val_loss -0.0769 
+2026-10-15 12:59:12.434497: Pseudo dice [np.float32(0.7807), np.float32(0.095), np.float32(0.1456), np.float32(0.2182), np.float32(0.0436), np.float32(0.1265), np.float32(0.2167), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2505), np.float32(0.1422), np.float32(0.3612), np.float32(0.1181), np.float32(0.0), np.float32(0.0), np.float32(0.058), np.float32(0.1765), np.float32(0.1708), np.float32(0.1795), np.float32(0.0313), np.float32(0.1705), np.float32(0.1631), np.float32(0.1026), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0709), np.float32(0.0819), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2387), np.float32(0.0861)] 
+2026-10-15 12:59:12.434886: Epoch time: 218.32 s 
+2026-10-15 12:59:13.301956:  
+2026-10-15 12:59:13.302572: Epoch 42 
+2026-10-15 12:59:13.303019: Current learning rate: 0.00962 
+2026-10-15 13:03:02.906473: train_loss -0.2765 
+2026-10-15 13:03:02.907197: val_loss -0.2053 
+2026-10-15 13:03:02.907715: Pseudo dice [np.float32(0.8943), np.float32(0.4535), np.float32(0.4096), np.float32(0.4153), np.float32(0.377), np.float32(0.4942), np.float32(0.4103), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2849), np.float32(0.4439), np.float32(0.4902), np.float32(0.4007), np.float32(0.0), np.float32(0.0), np.float32(0.3883), np.float32(0.2698), np.float32(0.372), np.float32(0.2161), np.float32(0.3804), np.float32(0.3543), np.float32(0.5043), np.float32(0.5814), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2503), np.float32(0.5357), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4857), np.float32(0.537)] 
+2026-10-15 13:03:02.908716: Epoch time: 229.61 s 
+2026-10-15 13:03:02.909173: Yayy! New best EMA pseudo Dice: 0.16169999539852142 
+2026-10-15 13:03:04.312842:  
+2026-10-15 13:03:04.313497: Epoch 43 
+2026-10-15 13:03:04.313864: Current learning rate: 0.00961 
+2026-10-15 13:06:55.062329: train_loss -0.2953 
+2026-10-15 13:06:55.063043: val_loss -0.1291 
+2026-10-15 13:06:55.064416: Pseudo dice [np.float32(0.866), np.float32(0.3844), np.float32(0.2231), np.float32(0.1027), np.float32(0.1983), np.float32(0.2322), np.float32(0.2756), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1547), np.float32(0.3349), np.float32(0.2534), np.float32(0.2912), np.float32(0.0), np.float32(0.0), np.float32(0.221), np.float32(0.1821), np.float32(0.1983), np.float32(0.2395), np.float32(0.2066), np.float32(0.514), np.float32(0.4913), np.float32(0.4829), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3396), np.float32(0.3403), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1518), np.float32(0.2401)] 
+2026-10-15 13:06:55.065054: Epoch time: 230.75 s 
+2026-10-15 13:06:55.065393: Yayy! New best EMA pseudo Dice: 0.1647000014781952 
+2026-10-15 13:06:58.450720:  
+2026-10-15 13:06:58.451541: Epoch 44 
+2026-10-15 13:06:58.452018: Current learning rate: 0.0096 
+2026-10-15 13:10:49.251657: train_loss -0.3092 
+2026-10-15 13:10:49.252212: val_loss -0.2493 
+2026-10-15 13:10:49.252797: Pseudo dice [np.float32(0.8642), np.float32(0.4878), np.float32(0.5927), np.float32(0.6271), np.float32(0.525), np.float32(0.6981), np.float32(0.5775), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5402), np.float32(0.5411), np.float32(0.4498), np.float32(0.4159), np.float32(0.0), np.float32(0.0), np.float32(0.5795), np.float32(0.3808), np.float32(0.5749), np.float32(0.3281), np.float32(0.4111), np.float32(0.4855), np.float32(0.5844), np.float32(0.6457), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3735), np.float32(0.4629), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6447), np.float32(0.6487)] 
+2026-10-15 13:10:49.253868: Epoch time: 230.8 s 
+2026-10-15 13:10:49.254219: Yayy! New best EMA pseudo Dice: 0.18279999494552612 
+2026-10-15 13:10:50.681340:  
+2026-10-15 13:10:50.682213: Epoch 45 
+2026-10-15 13:10:50.682456: Current learning rate: 0.00959 
+2026-10-15 13:14:53.964915: train_loss -0.3072 
+2026-10-15 13:14:53.965756: val_loss -0.2739 
+2026-10-15 13:14:53.966215: Pseudo dice [np.float32(0.878), np.float32(0.5283), np.float32(0.664), np.float32(0.6614), np.float32(0.6689), np.float32(0.7087), np.float32(0.6379), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.546), np.float32(0.5654), np.float32(0.5177), np.float32(0.4332), np.float32(0.0), np.float32(0.0), np.float32(0.6779), np.float32(0.4203), np.float32(0.6532), np.float32(0.4523), np.float32(0.4351), np.float32(0.5106), np.float32(0.7015), np.float32(0.6443), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.398), np.float32(0.4906), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7066), np.float32(0.7181)] 
+2026-10-15 13:14:53.966655: Epoch time: 243.29 s 
+2026-10-15 13:14:53.966986: Yayy! New best EMA pseudo Dice: 0.20239999890327454 
+2026-10-15 13:14:55.579574:  
+2026-10-15 13:14:55.580339: Epoch 46 
+2026-10-15 13:14:55.580912: Current learning rate: 0.00959 
+2026-10-15 13:18:46.568477: train_loss -0.3239 
+2026-10-15 13:18:46.569095: val_loss -0.2161 
+2026-10-15 13:18:46.569880: Pseudo dice [np.float32(0.8769), np.float32(0.5056), np.float32(0.5086), np.float32(0.4316), np.float32(0.4971), np.float32(0.5332), np.float32(0.3001), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4059), np.float32(0.4135), np.float32(0.3684), np.float32(0.4226), np.float32(0.0), np.float32(0.0), np.float32(0.4428), np.float32(0.3232), np.float32(0.3412), np.float32(0.2985), np.float32(0.4491), np.float32(0.4234), np.float32(0.5862), np.float32(0.4958), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4067), np.float32(0.4273), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5626), np.float32(0.6125)] 
+2026-10-15 13:18:46.570616: Epoch time: 230.99 s 
+2026-10-15 13:18:46.570951: Yayy! New best EMA pseudo Dice: 0.21170000731945038 
+2026-10-15 13:18:47.919495:  
+2026-10-15 13:18:47.920373: Epoch 47 
+2026-10-15 13:18:47.920809: Current learning rate: 0.00958 
+2026-10-15 13:22:36.899659: train_loss -0.3265 
+2026-10-15 13:22:36.900410: val_loss -0.2137 
+2026-10-15 13:22:36.901187: Pseudo dice [np.float32(0.8883), np.float32(0.4771), np.float32(0.4345), np.float32(0.4671), np.float32(0.4395), np.float32(0.4947), np.float32(0.4344), np.float32(0.0), np.float32(0.0), np.float32(0.3125), np.float32(0.3249), np.float32(0.3756), np.float32(0.466), np.float32(0.3215), np.float32(0.0), np.float32(0.0), np.float32(0.371), np.float32(0.3045), np.float32(0.345), np.float32(0.2246), np.float32(0.3809), np.float32(0.3648), np.float32(0.5595), np.float32(0.521), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3456), np.float32(0.4265), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5096), np.float32(0.5355)] 
+2026-10-15 13:22:36.901755: Epoch time: 228.98 s 
+2026-10-15 13:22:36.902393: Yayy! New best EMA pseudo Dice: 0.219200000166893 
+2026-10-15 13:22:39.389871:  
+2026-10-15 13:22:39.390730: Epoch 48 
+2026-10-15 13:22:39.391134: Current learning rate: 0.00957 
+2026-10-15 13:26:37.002963: train_loss -0.3376 
+2026-10-15 13:26:37.003740: val_loss -0.2724 
+2026-10-15 13:26:37.004588: Pseudo dice [np.float32(0.895), np.float32(0.5753), np.float32(0.6092), np.float32(0.5299), np.float32(0.4985), np.float32(0.5793), np.float32(0.5252), np.float32(0.0), np.float32(0.0), np.float32(0.2808), np.float32(0.4136), np.float32(0.4944), np.float32(0.5061), np.float32(0.4597), np.float32(0.0), np.float32(0.0), np.float32(0.5191), np.float32(0.3898), np.float32(0.4778), np.float32(0.3559), np.float32(0.5226), np.float32(0.5254), np.float32(0.7393), np.float32(0.6908), np.float32(0.0012), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4463), np.float32(0.4868), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.628), np.float32(0.6415)] 
+2026-10-15 13:26:37.005137: Epoch time: 237.62 s 
+2026-10-15 13:26:37.005539: Yayy! New best EMA pseudo Dice: 0.23280000686645508 
+2026-10-15 13:26:38.635043:  
+2026-10-15 13:26:38.635866: Epoch 49 
+2026-10-15 13:26:38.636261: Current learning rate: 0.00956 
+2026-10-15 13:30:31.925087: train_loss -0.3458 
+2026-10-15 13:30:31.925922: val_loss -0.2079 
+2026-10-15 13:30:31.926291: Pseudo dice [np.float32(0.8643), np.float32(0.4605), np.float32(0.5042), np.float32(0.4016), np.float32(0.3775), np.float32(0.4256), np.float32(0.381), np.float32(0.0), np.float32(0.0), np.float32(0.4189), np.float32(0.3422), np.float32(0.24), np.float32(0.4548), np.float32(0.3026), np.float32(0.0), np.float32(0.0), np.float32(0.3398), np.float32(0.2974), np.float32(0.4221), np.float32(0.3765), np.float32(0.4217), np.float32(0.4127), np.float32(0.5806), np.float32(0.4508), np.float32(0.0006), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3885), np.float32(0.4337), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5016), np.float32(0.5725)] 
+2026-10-15 13:30:31.926896: Epoch time: 233.29 s 
+2026-10-15 13:30:32.490288: Yayy! New best EMA pseudo Dice: 0.23829999566078186 
+2026-10-15 13:30:33.864461:  
+2026-10-15 13:30:33.865163: Epoch 50 
+2026-10-15 13:30:33.865528: Current learning rate: 0.00955 
+2026-10-15 13:34:57.708671: train_loss -0.376 
+2026-10-15 13:34:57.709353: val_loss -0.2613 
+2026-10-15 13:34:57.709847: Pseudo dice [np.float32(0.8562), np.float32(0.4718), np.float32(0.542), np.float32(0.4986), np.float32(0.366), np.float32(0.3997), np.float32(0.4299), np.float32(0.0), np.float32(0.0), np.float32(0.3871), np.float32(0.4841), np.float32(0.4476), np.float32(0.4795), np.float32(0.4387), np.float32(0.0), np.float32(0.0), np.float32(0.4628), np.float32(0.3623), np.float32(0.4225), np.float32(0.3635), np.float32(0.4062), np.float32(0.5811), np.float32(0.6013), np.float32(0.4217), np.float32(0.2944), np.float32(0.0827), np.float32(0.0), np.float32(0.0), np.float32(0.2771), np.float32(0.3732), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4992), np.float32(0.4943)] 
+2026-10-15 13:34:57.710359: Epoch time: 263.85 s 
+2026-10-15 13:34:57.710592: Yayy! New best EMA pseudo Dice: 0.24629999697208405 
+2026-10-15 13:34:59.886240:  
+2026-10-15 13:34:59.886927: Epoch 51 
+2026-10-15 13:34:59.887473: Current learning rate: 0.00954 
+2026-10-15 13:38:56.869529: train_loss -0.3829 
+2026-10-15 13:38:56.870686: val_loss -0.3087 
+2026-10-15 13:38:56.871275: Pseudo dice [np.float32(0.8869), np.float32(0.5561), np.float32(0.6084), np.float32(0.549), np.float32(0.4441), np.float32(0.5396), np.float32(0.4941), np.float32(0.0), np.float32(0.0), np.float32(0.468), np.float32(0.4542), np.float32(0.466), np.float32(0.3703), np.float32(0.4586), np.float32(0.0), np.float32(0.0), np.float32(0.4336), np.float32(0.3075), np.float32(0.4903), np.float32(0.4708), np.float32(0.506), np.float32(0.529), np.float32(0.6693), np.float32(0.6421), np.float32(0.3087), np.float32(0.1925), np.float32(0.0), np.float32(0.0), np.float32(0.4399), np.float32(0.4983), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6352), np.float32(0.6734)] 
+2026-10-15 13:38:56.871730: Epoch time: 236.99 s 
+2026-10-15 13:38:56.872301: Yayy! New best EMA pseudo Dice: 0.257999986410141 
+2026-10-15 13:39:00.087840:  
+2026-10-15 13:39:00.088592: Epoch 52 
+2026-10-15 13:39:00.088962: Current learning rate: 0.00953 
+2026-10-15 13:42:55.048645: train_loss -0.4268 
+2026-10-15 13:42:55.049533: val_loss -0.2587 
+2026-10-15 13:42:55.050113: Pseudo dice [np.float32(0.9021), np.float32(0.4582), np.float32(0.4676), np.float32(0.3962), np.float32(0.3715), np.float32(0.4467), np.float32(0.4062), np.float32(0.0), np.float32(0.0), np.float32(0.4102), np.float32(0.3511), np.float32(0.4108), np.float32(0.3967), np.float32(0.379), np.float32(0.0), np.float32(0.0), np.float32(0.4029), np.float32(0.3172), np.float32(0.4201), np.float32(0.2931), np.float32(0.4318), np.float32(0.4333), np.float32(0.5394), np.float32(0.5062), np.float32(0.2774), np.float32(0.2461), np.float32(0.0), np.float32(0.0), np.float32(0.3783), np.float32(0.476), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4207), np.float32(0.4744)] 
+2026-10-15 13:42:55.050703: Epoch time: 234.96 s 
+2026-10-15 13:42:55.050931: Yayy! New best EMA pseudo Dice: 0.262800008058548 
+2026-10-15 13:42:56.456637:  
+2026-10-15 13:42:56.457395: Epoch 53 
+2026-10-15 13:42:56.457778: Current learning rate: 0.00952 
+2026-10-15 13:47:11.686376: train_loss -0.4314 
+2026-10-15 13:47:11.687245: val_loss -0.3716 
+2026-10-15 13:47:11.687732: Pseudo dice [np.float32(0.8667), np.float32(0.6125), np.float32(0.6966), np.float32(0.7582), np.float32(0.697), np.float32(0.7976), np.float32(0.6524), np.float32(0.0), np.float32(0.0), np.float32(0.442), np.float32(0.6083), np.float32(0.6351), np.float32(0.4501), np.float32(0.4385), np.float32(0.0), np.float32(0.0), np.float32(0.6891), np.float32(0.4666), np.float32(0.6724), np.float32(0.5204), np.float32(0.556), np.float32(0.6048), np.float32(0.735), np.float32(0.6976), np.float32(0.4128), np.float32(0.3284), np.float32(0.0929), np.float32(0.0), np.float32(0.4292), np.float32(0.5219), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7944), np.float32(0.8143)] 
+2026-10-15 13:47:11.688083: Epoch time: 255.23 s 
+2026-10-15 13:47:11.688460: Yayy! New best EMA pseudo Dice: 0.2809000015258789 
+2026-10-15 13:47:13.100166:  
+2026-10-15 13:47:13.100814: Epoch 54 
+2026-10-15 13:47:13.101201: Current learning rate: 0.00951 
+2026-10-15 13:51:13.757971: train_loss -0.4372 
+2026-10-15 13:51:13.758843: val_loss -0.4174 
+2026-10-15 13:51:13.759195: Pseudo dice [np.float32(0.8876), np.float32(0.6824), np.float32(0.7664), np.float32(0.8198), np.float32(0.7989), np.float32(0.8678), np.float32(0.7797), np.float32(0.0), np.float32(0.0), np.float32(0.5439), np.float32(0.7004), np.float32(0.7443), np.float32(0.5667), np.float32(0.5024), np.float32(0.0), np.float32(0.0), np.float32(0.747), np.float32(0.4835), np.float32(0.718), np.float32(0.5269), np.float32(0.5983), np.float32(0.661), np.float32(0.7722), np.float32(0.8289), np.float32(0.4136), np.float32(0.2622), np.float32(0.1167), np.float32(0.0), np.float32(0.4358), np.float32(0.5292), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1356), np.float32(0.8345), np.float32(0.8203)] 
+2026-10-15 13:51:13.759423: Epoch time: 240.66 s 
+2026-10-15 13:51:13.759851: Yayy! New best EMA pseudo Dice: 0.30160000920295715 
+2026-10-15 13:51:15.322397:  
+2026-10-15 13:51:15.323105: Epoch 55 
+2026-10-15 13:51:15.323484: Current learning rate: 0.0095 
+2026-10-15 13:55:32.583525: train_loss -0.4214 
+2026-10-15 13:55:32.584568: val_loss -0.1895 
+2026-10-15 13:55:32.585318: Pseudo dice [np.float32(0.8938), np.float32(0.3316), np.float32(0.3254), np.float32(0.2752), np.float32(0.2028), np.float32(0.2609), np.float32(0.2476), np.float32(0.0), np.float32(0.0), np.float32(0.4377), np.float32(0.2464), np.float32(0.3353), np.float32(0.3861), np.float32(0.3941), np.float32(0.0), np.float32(0.0), np.float32(0.1721), np.float32(0.1091), np.float32(0.2523), np.float32(0.1808), np.float32(0.3023), np.float32(0.2362), np.float32(0.4051), np.float32(0.3205), np.float32(0.1309), np.float32(0.141), np.float32(0.0099), np.float32(0.0), np.float32(0.1057), np.float32(0.2676), np.float32(0.0), np.float32(0.0), np.float32(0.2122), np.float32(0.2076), np.float32(0.2543), np.float32(0.2552)] 
+2026-10-15 13:55:32.585832: Epoch time: 257.26 s 
+2026-10-15 13:55:33.365339:  
+2026-10-15 13:55:33.366234: Epoch 56 
+2026-10-15 13:55:33.366668: Current learning rate: 0.00949 
+2026-10-15 13:59:29.569143: train_loss -0.4421 
+2026-10-15 13:59:29.569952: val_loss -0.3566 
+2026-10-15 13:59:29.570668: Pseudo dice [np.float32(0.9005), np.float32(0.5661), np.float32(0.6636), np.float32(0.6795), np.float32(0.6766), np.float32(0.6805), np.float32(0.6344), np.float32(0.0), np.float32(0.0), np.float32(0.4516), np.float32(0.4975), np.float32(0.5195), np.float32(0.4483), np.float32(0.4072), np.float32(0.0), np.float32(0.0), np.float32(0.685), np.float32(0.5091), np.float32(0.6449), np.float32(0.5499), np.float32(0.5161), np.float32(0.6027), np.float32(0.7359), np.float32(0.7135), np.float32(0.4699), np.float32(0.3399), np.float32(0.1873), np.float32(0.0), np.float32(0.506), np.float32(0.4651), np.float32(0.0), np.float32(0.0), np.float32(0.2935), np.float32(0.1658), np.float32(0.7249), np.float32(0.7403)] 
+2026-10-15 13:59:29.571097: Epoch time: 236.21 s 
+2026-10-15 13:59:29.571411: Yayy! New best EMA pseudo Dice: 0.3084000051021576 
+2026-10-15 13:59:30.920393:  
+2026-10-15 13:59:30.921067: Epoch 57 
+2026-10-15 13:59:30.921461: Current learning rate: 0.00949 
+2026-10-15 14:03:15.809427: train_loss -0.4424 
+2026-10-15 14:03:15.809902: val_loss -0.2884 
+2026-10-15 14:03:15.810847: Pseudo dice [np.float32(0.8842), np.float32(0.4996), np.float32(0.4561), np.float32(0.4305), np.float32(0.4683), np.float32(0.4853), np.float32(0.4377), np.float32(0.0), np.float32(0.0), np.float32(0.454), np.float32(0.3559), np.float32(0.4298), np.float32(0.408), np.float32(0.3187), np.float32(0.0), np.float32(0.0), np.float32(0.5061), np.float32(0.3312), np.float32(0.4636), np.float32(0.3861), np.float32(0.3927), np.float32(0.3923), np.float32(0.6246), np.float32(0.5775), np.float32(0.4129), np.float32(0.3052), np.float32(0.0571), np.float32(0.0), np.float32(0.3951), np.float32(0.4087), np.float32(0.0), np.float32(0.0), np.float32(0.3058), np.float32(0.3448), np.float32(0.5003), np.float32(0.5356)] 
+2026-10-15 14:03:15.811454: Epoch time: 224.89 s 
+2026-10-15 14:03:15.811882: Yayy! New best EMA pseudo Dice: 0.3125 
+2026-10-15 14:03:17.208945:  
+2026-10-15 14:03:17.209616: Epoch 58 
+2026-10-15 14:03:17.209955: Current learning rate: 0.00948 
+2026-10-15 14:07:20.180268: train_loss -0.4798 
+2026-10-15 14:07:20.181212: val_loss -0.4046 
+2026-10-15 14:07:20.181736: Pseudo dice [np.float32(0.8574), np.float32(0.5215), np.float32(0.5753), np.float32(0.5338), np.float32(0.5285), np.float32(0.5869), np.float32(0.5494), np.float32(0.0), np.float32(0.0), np.float32(0.3973), np.float32(0.4044), np.float32(0.5518), np.float32(0.486), np.float32(0.4836), np.float32(0.0), np.float32(0.0), np.float32(0.5023), np.float32(0.2617), np.float32(0.5266), np.float32(0.4393), np.float32(0.4633), np.float32(0.4676), np.float32(0.6538), np.float32(0.5583), np.float32(0.3557), np.float32(0.2736), np.float32(0.0624), np.float32(0.0), np.float32(0.359), np.float32(0.3962), np.float32(0.0), np.float32(0.0), np.float32(0.2462), np.float32(0.3067), np.float32(0.586), np.float32(0.5585)] 
+2026-10-15 14:07:20.182198: Epoch time: 242.97 s 
+2026-10-15 14:07:20.182629: Yayy! New best EMA pseudo Dice: 0.31869998574256897 
+2026-10-15 14:07:21.905607:  
+2026-10-15 14:07:21.906335: Epoch 59 
+2026-10-15 14:07:21.906574: Current learning rate: 0.00947 
+2026-10-15 14:11:01.114987: train_loss -0.5382 
+2026-10-15 14:11:01.115603: val_loss -0.4351 
+2026-10-15 14:11:01.116179: Pseudo dice [np.float32(0.8639), np.float32(0.5435), np.float32(0.5646), np.float32(0.6066), np.float32(0.5254), np.float32(0.6854), np.float32(0.5712), np.float32(0.0), np.float32(0.0), np.float32(0.5282), np.float32(0.5335), np.float32(0.5741), np.float32(0.4231), np.float32(0.4687), np.float32(0.0), np.float32(0.0), np.float32(0.5251), np.float32(0.4198), np.float32(0.5767), np.float32(0.4362), np.float32(0.5094), np.float32(0.4437), np.float32(0.7028), np.float32(0.6682), np.float32(0.3366), np.float32(0.2513), np.float32(0.101), np.float32(0.0), np.float32(0.458), np.float32(0.5579), np.float32(0.0), np.float32(0.0), np.float32(0.3601), np.float32(0.3245), np.float32(0.6535), np.float32(0.6495)] 
+2026-10-15 14:11:01.117165: Epoch time: 219.21 s 
+2026-10-15 14:11:01.117477: Yayy! New best EMA pseudo Dice: 0.3280999958515167 
+2026-10-15 14:11:02.745856:  
+2026-10-15 14:11:02.746643: Epoch 60 
+2026-10-15 14:11:02.746975: Current learning rate: 0.00946 
+2026-10-15 14:15:05.076037: train_loss -0.5261 
+2026-10-15 14:15:05.077090: val_loss -0.275 
+2026-10-15 14:15:05.077595: Pseudo dice [np.float32(0.8799), np.float32(0.287), np.float32(0.2516), np.float32(0.2209), np.float32(0.2755), np.float32(0.2733), np.float32(0.2363), np.float32(0.0), np.float32(0.0), np.float32(0.5557), np.float32(0.2164), np.float32(0.201), np.float32(0.3908), np.float32(0.3258), np.float32(0.0), np.float32(0.0), np.float32(0.2537), np.float32(0.2531), np.float32(0.2472), np.float32(0.2168), np.float32(0.1884), np.float32(0.1854), np.float32(0.4658), np.float32(0.333), np.float32(0.1279), np.float32(0.1521), np.float32(0.0617), np.float32(0.0), np.float32(0.1782), np.float32(0.2547), np.float32(0.0), np.float32(0.0), np.float32(0.2831), np.float32(0.314), np.float32(0.2584), np.float32(0.3347)] 
+2026-10-15 14:15:05.078074: Epoch time: 242.33 s 
+2026-10-15 14:15:05.950129:  
+2026-10-15 14:15:05.951380: Epoch 61 
+2026-10-15 14:15:05.951890: Current learning rate: 0.00945 
+2026-10-15 14:19:01.468551: train_loss -0.5548 
+2026-10-15 14:19:01.469128: val_loss -0.385 
+2026-10-15 14:19:01.469863: Pseudo dice [np.float32(0.8991), np.float32(0.4908), np.float32(0.4729), np.float32(0.451), np.float32(0.3751), np.float32(0.5104), np.float32(0.397), np.float32(0.0), np.float32(0.0), np.float32(0.4408), np.float32(0.3687), np.float32(0.4113), np.float32(0.4964), np.float32(0.4588), np.float32(0.0), np.float32(0.0), np.float32(0.3275), np.float32(0.2984), np.float32(0.3808), np.float32(0.3178), np.float32(0.4434), np.float32(0.3552), np.float32(0.5964), np.float32(0.5559), np.float32(0.2132), np.float32(0.2457), np.float32(0.0205), np.float32(0.0), np.float32(0.3916), np.float32(0.5047), np.float32(0.0), np.float32(0.0), np.float32(0.369), np.float32(0.2918), np.float32(0.4997), np.float32(0.5314)] 
+2026-10-15 14:19:01.470787: Epoch time: 235.52 s 
+2026-10-15 14:19:02.283033:  
+2026-10-15 14:19:02.283756: Epoch 62 
+2026-10-15 14:19:02.284240: Current learning rate: 0.00944 
+2026-10-15 14:23:09.500122: train_loss -0.5558 
+2026-10-15 14:23:09.501100: val_loss -0.3108 
+2026-10-15 14:23:09.501777: Pseudo dice [np.float32(0.9004), np.float32(0.2932), np.float32(0.3131), np.float32(0.2601), np.float32(0.1455), np.float32(0.312), np.float32(0.247), np.float32(0.0), np.float32(0.0), np.float32(0.5261), np.float32(0.316), np.float32(0.3045), np.float32(0.3571), np.float32(0.4495), np.float32(0.0), np.float32(0.0), np.float32(0.1511), np.float32(0.0832), np.float32(0.2849), np.float32(0.3139), np.float32(0.3067), np.float32(0.2632), np.float32(0.5574), np.float32(0.4747), np.float32(0.121), np.float32(0.1376), np.float32(0.0172), np.float32(0.0), np.float32(0.2376), np.float32(0.355), np.float32(0.0), np.float32(0.0), np.float32(0.22), np.float32(0.3034), np.float32(0.2958), np.float32(0.3748)] 
+2026-10-15 14:23:09.502212: Epoch time: 247.22 s 
+2026-10-15 14:23:10.313658:  
+2026-10-15 14:23:10.314368: Epoch 63 
+2026-10-15 14:23:10.314747: Current learning rate: 0.00943 
+2026-10-15 14:26:52.817147: train_loss -0.5151 
+2026-10-15 14:26:52.817655: val_loss -0.3428 
+2026-10-15 14:26:52.818188: Pseudo dice [np.float32(0.8662), np.float32(0.4521), np.float32(0.3748), np.float32(0.3707), np.float32(0.4451), np.float32(0.4032), np.float32(0.4293), np.float32(0.0), np.float32(0.0), np.float32(0.3664), np.float32(0.2744), np.float32(0.3085), np.float32(0.3139), np.float32(0.3179), np.float32(0.0), np.float32(0.0), np.float32(0.4092), np.float32(0.3843), np.float32(0.4287), np.float32(0.3795), np.float32(0.3645), np.float32(0.3034), np.float32(0.5448), np.float32(0.5348), np.float32(0.3136), np.float32(0.2227), np.float32(0.0393), np.float32(0.0), np.float32(0.3301), np.float32(0.3413), np.float32(0.0), np.float32(0.0), np.float32(0.3351), np.float32(0.3396), np.float32(0.4217), np.float32(0.4063)] 
+2026-10-15 14:26:52.819226: Epoch time: 222.51 s 
+2026-10-15 14:26:54.000695:  
+2026-10-15 14:26:54.001582: Epoch 64 
+2026-10-15 14:26:54.001969: Current learning rate: 0.00942 
+2026-10-15 14:30:43.306380: train_loss -0.5312 
+2026-10-15 14:30:43.307238: val_loss -0.3595 
+2026-10-15 14:30:43.307942: Pseudo dice [np.float32(0.868), np.float32(0.4385), np.float32(0.4374), np.float32(0.4041), np.float32(0.3595), np.float32(0.4736), np.float32(0.4078), np.float32(0.0), np.float32(0.0), np.float32(0.4817), np.float32(0.3391), np.float32(0.3816), np.float32(0.3998), np.float32(0.3557), np.float32(0.0), np.float32(0.0), np.float32(0.3299), np.float32(0.2866), np.float32(0.3634), np.float32(0.3221), np.float32(0.3822), np.float32(0.3212), np.float32(0.5635), np.float32(0.4532), np.float32(0.2043), np.float32(0.1699), np.float32(0.0152), np.float32(0.0), np.float32(0.3966), np.float32(0.3943), np.float32(0.0), np.float32(0.0), np.float32(0.3127), np.float32(0.3159), np.float32(0.4688), np.float32(0.4916)] 
+2026-10-15 14:30:43.308309: Epoch time: 229.31 s 
+2026-10-15 14:30:44.121683:  
+2026-10-15 14:30:44.122543: Epoch 65 
+2026-10-15 14:30:44.122871: Current learning rate: 0.00941 
+2026-10-15 14:34:56.158214: train_loss -0.5316 
+2026-10-15 14:34:56.159132: val_loss -0.4626 
+2026-10-15 14:34:56.159702: Pseudo dice [np.float32(0.8885), np.float32(0.6098), np.float32(0.7269), np.float32(0.6714), np.float32(0.6252), np.float32(0.7176), np.float32(0.5974), np.float32(0.0), np.float32(0.0), np.float32(0.3559), np.float32(0.5554), np.float32(0.5452), np.float32(0.5157), np.float32(0.4622), np.float32(0.0), np.float32(0.0), np.float32(0.6589), np.float32(0.4751), np.float32(0.594), np.float32(0.4689), np.float32(0.5675), np.float32(0.6434), np.float32(0.6841), np.float32(0.6274), np.float32(0.457), np.float32(0.4168), np.float32(0.202), np.float32(0.0), np.float32(0.4452), np.float32(0.437), np.float32(0.0), np.float32(0.0), np.float32(0.3859), np.float32(0.276), np.float32(0.6734), np.float32(0.6965)] 
+2026-10-15 14:34:56.160198: Epoch time: 252.04 s 
+2026-10-15 14:34:56.993925:  
+2026-10-15 14:34:56.994744: Epoch 66 
+2026-10-15 14:34:56.995158: Current learning rate: 0.0094 
+2026-10-15 14:38:52.268592: train_loss -0.5458 
+2026-10-15 14:38:52.269370: val_loss -0.4059 
+2026-10-15 14:38:52.270039: Pseudo dice [np.float32(0.8881), np.float32(0.4984), np.float32(0.5063), np.float32(0.4827), np.float32(0.4101), np.float32(0.5112), np.float32(0.4594), np.float32(0.0), np.float32(0.0), np.float32(0.5412), np.float32(0.4278), np.float32(0.4869), np.float32(0.5135), np.float32(0.4699), np.float32(0.0), np.float32(0.0), np.float32(0.4197), np.float32(0.3625), np.float32(0.4604), np.float32(0.4594), np.float32(0.4899), np.float32(0.4587), np.float32(0.5705), np.float32(0.544), np.float32(0.2515), np.float32(0.2543), np.float32(0.0923), np.float32(0.0), np.float32(0.449), np.float32(0.4715), np.float32(0.0), np.float32(0.0), np.float32(0.3213), np.float32(0.382), np.float32(0.5461), np.float32(0.582)] 
+2026-10-15 14:38:52.270446: Epoch time: 235.28 s 
+2026-10-15 14:38:52.270753: Yayy! New best EMA pseudo Dice: 0.3294999897480011 
+2026-10-15 14:38:53.645700:  
+2026-10-15 14:38:53.646255: Epoch 67 
+2026-10-15 14:38:53.646611: Current learning rate: 0.00939 
+2026-10-15 14:42:41.834153: train_loss -0.5637 
+2026-10-15 14:42:41.835506: val_loss -0.4597 
+2026-10-15 14:42:41.836027: Pseudo dice [np.float32(0.8877), np.float32(0.6382), np.float32(0.6692), np.float32(0.6669), np.float32(0.619), np.float32(0.719), np.float32(0.6082), np.float32(0.0), np.float32(0.0), np.float32(0.3438), np.float32(0.5711), np.float32(0.5796), np.float32(0.4997), np.float32(0.4416), np.float32(0.0), np.float32(0.0), np.float32(0.6542), np.float32(0.5119), np.float32(0.6378), np.float32(0.4947), np.float32(0.528), np.float32(0.5684), np.float32(0.6975), np.float32(0.6999), np.float32(0.4227), np.float32(0.5025), np.float32(0.2323), np.float32(0.0), np.float32(0.3858), np.float32(0.4134), np.float32(0.0), np.float32(0.0), np.float32(0.4092), np.float32(0.398), np.float32(0.7103), np.float32(0.7624)] 
+2026-10-15 14:42:41.836589: Epoch time: 228.19 s 
+2026-10-15 14:42:41.837106: Yayy! New best EMA pseudo Dice: 0.3418000042438507 
+2026-10-15 14:42:43.478784:  
+2026-10-15 14:42:43.479601: Epoch 68 
+2026-10-15 14:42:43.479964: Current learning rate: 0.00939 
+2026-10-15 14:46:46.209984: train_loss -0.5694 
+2026-10-15 14:46:46.210590: val_loss -0.2417 
+2026-10-15 14:46:46.211170: Pseudo dice [np.float32(0.8696), np.float32(0.2076), np.float32(0.1805), np.float32(0.1159), np.float32(0.1714), np.float32(0.1652), np.float32(0.1736), np.float32(0.0), np.float32(0.0), np.float32(0.5298), np.float32(0.1491), np.float32(0.1982), np.float32(0.3437), np.float32(0.2932), np.float32(0.0), np.float32(0.0), np.float32(0.1179), np.float32(0.1432), np.float32(0.2166), np.float32(0.1631), np.float32(0.1563), np.float32(0.1628), np.float32(0.4497), np.float32(0.3202), np.float32(0.1185), np.float32(0.1576), np.float32(0.0712), np.float32(0.0), np.float32(0.0932), np.float32(0.2673), np.float32(0.0), np.float32(0.0), np.float32(0.1093), np.float32(0.1195), np.float32(0.1445), np.float32(0.2078)] 
+2026-10-15 14:46:46.212043: Epoch time: 242.73 s 
+2026-10-15 14:46:47.060985:  
+2026-10-15 14:46:47.061670: Epoch 69 
+2026-10-15 14:46:47.062041: Current learning rate: 0.00938 
+2026-10-15 14:50:51.176293: train_loss -0.5184 
+2026-10-15 14:50:51.176807: val_loss -0.4104 
+2026-10-15 14:50:51.177373: Pseudo dice [np.float32(0.8777), np.float32(0.5199), np.float32(0.5252), np.float32(0.4841), np.float32(0.447), np.float32(0.5895), np.float32(0.5107), np.float32(0.0), np.float32(0.0), np.float32(0.4833), np.float32(0.4195), np.float32(0.4583), np.float32(0.5679), np.float32(0.5697), np.float32(0.0), np.float32(0.0), np.float32(0.3681), np.float32(0.2631), np.float32(0.4112), np.float32(0.3527), np.float32(0.4347), np.float32(0.4308), np.float32(0.631), np.float32(0.5718), np.float32(0.2082), np.float32(0.2539), np.float32(0.0659), np.float32(0.0), np.float32(0.3902), np.float32(0.4291), np.float32(0.0), np.float32(0.0), np.float32(0.4041), np.float32(0.3651), np.float32(0.537), np.float32(0.5867)] 
+2026-10-15 14:50:51.177661: Epoch time: 244.12 s 
+2026-10-15 14:50:51.997091:  
+2026-10-15 14:50:51.997740: Epoch 70 
+2026-10-15 14:50:51.998037: Current learning rate: 0.00937 
+2026-10-15 14:54:40.845196: train_loss -0.5222 
+2026-10-15 14:54:40.846083: val_loss -0.3529 
+2026-10-15 14:54:40.846938: Pseudo dice [np.float32(0.8938), np.float32(0.4044), np.float32(0.4573), np.float32(0.342), np.float32(0.431), np.float32(0.4565), np.float32(0.389), np.float32(0.0), np.float32(0.0), np.float32(0.3182), np.float32(0.3391), np.float32(0.3767), np.float32(0.4805), np.float32(0.4274), np.float32(0.0), np.float32(0.0), np.float32(0.5144), np.float32(0.3994), np.float32(0.4032), np.float32(0.3399), np.float32(0.388), np.float32(0.3655), np.float32(0.5665), np.float32(0.5363), np.float32(0.3053), np.float32(0.2008), np.float32(0.0619), np.float32(0.0), np.float32(0.2392), np.float32(0.2649), np.float32(0.0), np.float32(0.0), np.float32(0.2963), np.float32(0.297), np.float32(0.4346), np.float32(0.4658)] 
+2026-10-15 14:54:40.847410: Epoch time: 228.85 s 
+2026-10-15 14:54:41.700794:  
+2026-10-15 14:54:41.701574: Epoch 71 
+2026-10-15 14:54:41.702204: Current learning rate: 0.00936 
+2026-10-15 14:58:28.119424: train_loss -0.5437 
+2026-10-15 14:58:28.120296: val_loss -0.3937 
+2026-10-15 14:58:28.120785: Pseudo dice [np.float32(0.8814), np.float32(0.4292), np.float32(0.4489), np.float32(0.392), np.float32(0.3294), np.float32(0.4339), np.float32(0.375), np.float32(0.0), np.float32(0.0), np.float32(0.574), np.float32(0.3617), np.float32(0.4151), np.float32(0.4951), np.float32(0.4312), np.float32(0.0), np.float32(0.0), np.float32(0.4341), np.float32(0.4004), np.float32(0.3213), np.float32(0.2668), np.float32(0.4587), np.float32(0.4429), np.float32(0.6183), np.float32(0.6412), np.float32(0.299), np.float32(0.207), np.float32(0.1169), np.float32(0.0), np.float32(0.4264), np.float32(0.4219), np.float32(0.0), np.float32(0.0), np.float32(0.2544), np.float32(0.3143), np.float32(0.4871), np.float32(0.5545)] 
+2026-10-15 14:58:28.121132: Epoch time: 226.42 s 
+2026-10-15 14:58:28.915858:  
+2026-10-15 14:58:28.916474: Epoch 72 
+2026-10-15 14:58:28.916813: Current learning rate: 0.00935 
+2026-10-15 15:02:24.422826: train_loss -0.581 
+2026-10-15 15:02:24.423929: val_loss -0.3739 
+2026-10-15 15:02:24.424611: Pseudo dice [np.float32(0.9002), np.float32(0.4527), np.float32(0.4581), np.float32(0.3896), np.float32(0.3314), np.float32(0.4612), np.float32(0.3919), np.float32(0.0), np.float32(0.0), np.float32(0.4181), np.float32(0.3779), np.float32(0.3693), np.float32(0.4528), np.float32(0.4763), np.float32(0.0), np.float32(0.0), np.float32(0.3881), np.float32(0.3084), np.float32(0.3467), np.float32(0.3644), np.float32(0.4694), np.float32(0.4118), np.float32(0.4639), np.float32(0.4262), np.float32(0.2077), np.float32(0.2118), np.float32(0.0553), np.float32(0.0), np.float32(0.2941), np.float32(0.417), np.float32(0.0), np.float32(0.0), np.float32(0.3786), np.float32(0.2516), np.float32(0.3789), np.float32(0.4293)] 
+2026-10-15 15:02:24.425122: Epoch time: 235.51 s 
+2026-10-15 15:02:25.286784:  
+2026-10-15 15:02:25.287574: Epoch 73 
+2026-10-15 15:02:25.287938: Current learning rate: 0.00934 
+2026-10-15 15:06:26.987514: train_loss -0.5809 
+2026-10-15 15:06:26.988281: val_loss -0.3907 
+2026-10-15 15:06:26.989014: Pseudo dice [np.float32(0.8981), np.float32(0.4802), np.float32(0.4955), np.float32(0.4414), np.float32(0.4202), np.float32(0.5118), np.float32(0.4597), np.float32(0.0), np.float32(0.0), np.float32(0.5379), np.float32(0.3452), np.float32(0.4572), np.float32(0.4564), np.float32(0.435), np.float32(0.0), np.float32(0.0), np.float32(0.4067), np.float32(0.3612), np.float32(0.42), np.float32(0.4215), np.float32(0.433), np.float32(0.3875), np.float32(0.6109), np.float32(0.5509), np.float32(0.3006), np.float32(0.2826), np.float32(0.1203), np.float32(0.0), np.float32(0.4186), np.float32(0.4969), np.float32(0.0), np.float32(0.0), np.float32(0.4169), np.float32(0.485), np.float32(0.4913), np.float32(0.541)] 
+2026-10-15 15:06:26.989427: Epoch time: 241.7 s 
+2026-10-15 15:06:27.775288:  
+2026-10-15 15:06:27.775995: Epoch 74 
+2026-10-15 15:06:27.776395: Current learning rate: 0.00933 
+2026-10-15 15:10:39.029348: train_loss -0.5815 
+2026-10-15 15:10:39.030047: val_loss -0.2476 
+2026-10-15 15:10:39.030689: Pseudo dice [np.float32(0.8627), np.float32(0.2352), np.float32(0.1938), np.float32(0.1282), np.float32(0.2271), np.float32(0.2213), np.float32(0.1775), np.float32(0.0), np.float32(0.0), np.float32(0.5018), np.float32(0.1554), np.float32(0.2), np.float32(0.355), np.float32(0.3414), np.float32(0.0), np.float32(0.0), np.float32(0.2582), np.float32(0.2428), np.float32(0.2111), np.float32(0.1532), np.float32(0.1984), np.float32(0.1724), np.float32(0.2397), np.float32(0.2227), np.float32(0.146), np.float32(0.1688), np.float32(0.0), np.float32(0.0), np.float32(0.124), np.float32(0.3085), np.float32(0.0), np.float32(0.0), np.float32(0.2168), np.float32(0.128), np.float32(0.1418), np.float32(0.2068)] 
+2026-10-15 15:10:39.031789: Epoch time: 251.26 s 
+2026-10-15 15:10:39.862349:  
+2026-10-15 15:10:39.863297: Epoch 75 
+2026-10-15 15:10:39.863771: Current learning rate: 0.00932 
+2026-10-15 15:14:45.769601: train_loss -0.5519 
+2026-10-15 15:14:45.770295: val_loss -0.401 
+2026-10-15 15:14:45.770777: Pseudo dice [np.float32(0.8909), np.float32(0.5219), np.float32(0.5699), np.float32(0.4897), np.float32(0.3153), np.float32(0.4946), np.float32(0.4745), np.float32(0.0), np.float32(0.0), np.float32(0.5369), np.float32(0.4338), np.float32(0.4165), np.float32(0.5292), np.float32(0.4466), np.float32(0.0), np.float32(0.0), np.float32(0.2582), np.float32(0.3325), np.float32(0.3688), np.float32(0.3844), np.float32(0.5034), np.float32(0.5805), np.float32(0.6103), np.float32(0.6117), np.float32(0.3022), np.float32(0.2379), np.float32(0.0556), np.float32(0.0), np.float32(0.4571), np.float32(0.4745), np.float32(0.0), np.float32(0.0), np.float32(0.1522), np.float32(0.3227), np.float32(0.5159), np.float32(0.6047)] 
+2026-10-15 15:14:45.771353: Epoch time: 245.91 s 
+2026-10-15 15:14:46.617442:  
+2026-10-15 15:14:46.618113: Epoch 76 
+2026-10-15 15:14:46.618583: Current learning rate: 0.00931 
+2026-10-15 15:18:52.690766: train_loss -0.5603 
+2026-10-15 15:18:52.691726: val_loss -0.4106 
+2026-10-15 15:18:52.692294: Pseudo dice [np.float32(0.8719), np.float32(0.4904), np.float32(0.4954), np.float32(0.4888), np.float32(0.4323), np.float32(0.5431), np.float32(0.4642), np.float32(0.0), np.float32(0.0), np.float32(0.5502), np.float32(0.4197), np.float32(0.4335), np.float32(0.4854), np.float32(0.424), np.float32(0.0), np.float32(0.0), np.float32(0.3926), np.float32(0.3677), np.float32(0.4605), np.float32(0.3355), np.float32(0.5024), np.float32(0.4813), np.float32(0.6656), np.float32(0.608), np.float32(0.2811), np.float32(0.2274), np.float32(0.0663), np.float32(0.0), np.float32(0.361), np.float32(0.4382), np.float32(0.0), np.float32(0.0), np.float32(0.3412), np.float32(0.2668), np.float32(0.5335), np.float32(0.5581)] 
+2026-10-15 15:18:52.692728: Epoch time: 246.08 s 
+2026-10-15 15:18:53.511070:  
+2026-10-15 15:18:53.511787: Epoch 77 
+2026-10-15 15:18:53.512189: Current learning rate: 0.0093 
+2026-10-15 15:22:53.736098: train_loss -0.5738 
+2026-10-15 15:22:53.737132: val_loss -0.4379 
+2026-10-15 15:22:53.737925: Pseudo dice [np.float32(0.8742), np.float32(0.5441), np.float32(0.5973), np.float32(0.5679), np.float32(0.4912), np.float32(0.6251), np.float32(0.5267), np.float32(0.0), np.float32(0.0), np.float32(0.5727), np.float32(0.5074), np.float32(0.5556), np.float32(0.4801), np.float32(0.496), np.float32(0.0), np.float32(0.0), np.float32(0.5463), np.float32(0.4468), np.float32(0.5418), np.float32(0.4719), np.float32(0.5217), np.float32(0.5304), np.float32(0.6174), np.float32(0.6098), np.float32(0.4656), np.float32(0.4094), np.float32(0.1897), np.float32(0.0), np.float32(0.411), np.float32(0.4978), np.float32(0.0), np.float32(0.0), np.float32(0.3265), np.float32(0.4522), np.float32(0.5812), np.float32(0.6251)] 
+2026-10-15 15:22:53.738473: Epoch time: 240.23 s 
+2026-10-15 15:22:54.563632:  
+2026-10-15 15:22:54.564328: Epoch 78 
+2026-10-15 15:22:54.564831: Current learning rate: 0.0093 
+2026-10-15 15:27:05.942054: train_loss -0.5879 
+2026-10-15 15:27:05.942933: val_loss -0.4607 
+2026-10-15 15:27:05.943475: Pseudo dice [np.float32(0.893), np.float32(0.5405), np.float32(0.6386), np.float32(0.6397), np.float32(0.5672), np.float32(0.6636), np.float32(0.5825), np.float32(0.0), np.float32(0.0), np.float32(0.4425), np.float32(0.5237), np.float32(0.5537), np.float32(0.4866), np.float32(0.3868), np.float32(0.0), np.float32(0.0), np.float32(0.6528), np.float32(0.5011), np.float32(0.5971), np.float32(0.448), np.float32(0.5119), np.float32(0.5732), np.float32(0.6954), np.float32(0.6854), np.float32(0.5552), np.float32(0.474), np.float32(0.3089), np.float32(0.0), np.float32(0.5089), np.float32(0.526), np.float32(0.0), np.float32(0.0), np.float32(0.4562), np.float32(0.5778), np.float32(0.644), np.float32(0.6886)] 
+2026-10-15 15:27:05.944061: Epoch time: 251.38 s 
+2026-10-15 15:27:05.944364: Yayy! New best EMA pseudo Dice: 0.34630000591278076 
+2026-10-15 15:27:07.934858:  
+2026-10-15 15:27:07.935571: Epoch 79 
+2026-10-15 15:27:07.936005: Current learning rate: 0.00929 
+2026-10-15 15:31:02.621294: train_loss -0.6042 
+2026-10-15 15:31:02.621865: val_loss -0.4237 
+2026-10-15 15:31:02.622600: Pseudo dice [np.float32(0.8781), np.float32(0.5302), np.float32(0.5584), np.float32(0.5267), np.float32(0.4559), np.float32(0.5915), np.float32(0.4786), np.float32(0.0), np.float32(0.0), np.float32(0.5166), np.float32(0.4488), np.float32(0.4621), np.float32(0.5349), np.float32(0.4806), np.float32(0.0), np.float32(0.0), np.float32(0.3723), np.float32(0.3289), np.float32(0.4452), np.float32(0.3789), np.float32(0.4968), np.float32(0.4832), np.float32(0.6999), np.float32(0.5903), np.float32(0.3703), np.float32(0.3423), np.float32(0.1553), np.float32(0.0), np.float32(0.4788), np.float32(0.442), np.float32(0.0), np.float32(0.0), np.float32(0.4093), np.float32(0.4327), np.float32(0.6144), np.float32(0.6523)] 
+2026-10-15 15:31:02.623519: Epoch time: 234.69 s 
+2026-10-15 15:31:02.623873: Yayy! New best EMA pseudo Dice: 0.35100001096725464 
+2026-10-15 15:31:04.756045:  
+2026-10-15 15:31:04.763643: Epoch 80 
+2026-10-15 15:31:04.770716: Current learning rate: 0.00928 
+2026-10-15 15:34:56.173238: train_loss -0.5735 
+2026-10-15 15:34:56.174118: val_loss -0.2447 
+2026-10-15 15:34:56.174752: Pseudo dice [np.float32(0.8478), np.float32(0.1532), np.float32(0.1586), np.float32(0.1036), np.float32(0.1436), np.float32(0.1528), np.float32(0.1723), np.float32(0.0), np.float32(0.0), np.float32(0.4905), np.float32(0.1292), np.float32(0.1733), np.float32(0.2846), np.float32(0.3054), np.float32(0.0), np.float32(0.0), np.float32(0.1157), np.float32(0.135), np.float32(0.1661), np.float32(0.0983), np.float32(0.1424), np.float32(0.145), np.float32(0.2857), np.float32(0.3389), np.float32(0.1186), np.float32(0.1465), np.float32(0.0133), np.float32(0.0), np.float32(0.0753), np.float32(0.2041), np.float32(0.0), np.float32(0.0), np.float32(0.1129), np.float32(0.1299), np.float32(0.1348), np.float32(0.1379)] 
+2026-10-15 15:34:56.175282: Epoch time: 231.42 s 
+2026-10-15 15:34:57.020772:  
+2026-10-15 15:34:57.021520: Epoch 81 
+2026-10-15 15:34:57.021933: Current learning rate: 0.00927 
+2026-10-15 15:38:50.853407: train_loss -0.5838 
+2026-10-15 15:38:50.854231: val_loss -0.2954 
+2026-10-15 15:38:50.854743: Pseudo dice [np.float32(0.8863), np.float32(0.324), np.float32(0.3068), np.float32(0.218), np.float32(0.1949), np.float32(0.3186), np.float32(0.2176), np.float32(0.0), np.float32(0.0), np.float32(0.5074), np.float32(0.193), np.float32(0.3342), np.float32(0.3922), np.float32(0.3843), np.float32(0.0), np.float32(0.0), np.float32(0.1857), np.float32(0.1765), np.float32(0.1858), np.float32(0.2183), np.float32(0.2988), np.float32(0.2422), np.float32(0.3625), np.float32(0.3668), np.float32(0.1005), np.float32(0.1713), np.float32(0.0194), np.float32(0.0), np.float32(0.2212), np.float32(0.2857), np.float32(0.0), np.float32(0.0), np.float32(0.1567), np.float32(0.2345), np.float32(0.2536), np.float32(0.331)] 
+2026-10-15 15:38:50.855035: Epoch time: 233.84 s 
+2026-10-15 15:38:51.990869:  
+2026-10-15 15:38:51.991498: Epoch 82 
+2026-10-15 15:38:51.991987: Current learning rate: 0.00926 
+2026-10-15 15:43:02.754085: train_loss -0.6028 
+2026-10-15 15:43:02.754673: val_loss -0.3221 
+2026-10-15 15:43:02.755513: Pseudo dice [np.float32(0.8832), np.float32(0.2874), np.float32(0.3665), np.float32(0.2703), np.float32(0.1999), np.float32(0.294), np.float32(0.2949), np.float32(0.0), np.float32(0.0), np.float32(0.4463), np.float32(0.2947), np.float32(0.3324), np.float32(0.4494), np.float32(0.4318), np.float32(0.0), np.float32(0.0), np.float32(0.1635), np.float32(0.1549), np.float32(0.2422), np.float32(0.2707), np.float32(0.3412), np.float32(0.2965), np.float32(0.3699), np.float32(0.3469), np.float32(0.1181), np.float32(0.1809), np.float32(0.0067), np.float32(0.0), np.float32(0.2266), np.float32(0.3508), np.float32(0.0), np.float32(0.0), np.float32(0.2151), np.float32(0.2955), np.float32(0.3234), np.float32(0.385)] 
+2026-10-15 15:43:02.756399: Epoch time: 250.77 s 
+2026-10-15 15:43:03.566153:  
+2026-10-15 15:43:03.566936: Epoch 83 
+2026-10-15 15:43:03.567312: Current learning rate: 0.00925 
+2026-10-15 15:46:42.359665: train_loss -0.6058 
+2026-10-15 15:46:42.360320: val_loss -0.437 
+2026-10-15 15:46:42.360832: Pseudo dice [np.float32(0.8913), np.float32(0.5216), np.float32(0.5935), np.float32(0.5414), np.float32(0.4691), np.float32(0.5154), np.float32(0.5552), np.float32(0.0), np.float32(0.0), np.float32(0.4476), np.float32(0.5324), np.float32(0.5456), np.float32(0.5331), np.float32(0.4985), np.float32(0.0), np.float32(0.0), np.float32(0.4954), np.float32(0.4092), np.float32(0.5798), np.float32(0.5379), np.float32(0.5492), np.float32(0.5412), np.float32(0.6037), np.float32(0.6088), np.float32(0.4048), np.float32(0.4633), np.float32(0.2026), np.float32(0.0), np.float32(0.3), np.float32(0.4736), np.float32(0.0), np.float32(0.0), np.float32(0.438), np.float32(0.4651), np.float32(0.545), np.float32(0.586)] 
+2026-10-15 15:46:42.361867: Epoch time: 218.8 s 
+2026-10-15 15:46:43.181442:  
+2026-10-15 15:46:43.182091: Epoch 84 
+2026-10-15 15:46:43.182493: Current learning rate: 0.00924 
+2026-10-15 15:50:35.578101: train_loss -0.6036 
+2026-10-15 15:50:35.578655: val_loss -0.4292 
+2026-10-15 15:50:35.579741: Pseudo dice [np.float32(0.87), np.float32(0.5268), np.float32(0.6117), np.float32(0.5298), np.float32(0.4729), np.float32(0.5443), np.float32(0.4638), np.float32(0.0), np.float32(0.0), np.float32(0.5486), np.float32(0.4382), np.float32(0.5203), np.float32(0.4937), np.float32(0.5023), np.float32(0.0), np.float32(0.0), np.float32(0.4743), np.float32(0.4157), np.float32(0.398), np.float32(0.3997), np.float32(0.5193), np.float32(0.5486), np.float32(0.6078), np.float32(0.6567), np.float32(0.4241), np.float32(0.4682), np.float32(0.2039), np.float32(0.0), np.float32(0.2767), np.float32(0.4458), np.float32(0.0), np.float32(0.0), np.float32(0.314), np.float32(0.1598), np.float32(0.5884), np.float32(0.5874)] 
+2026-10-15 15:50:35.580319: Epoch time: 232.4 s 
+2026-10-15 15:50:36.362583:  
+2026-10-15 15:50:36.363401: Epoch 85 
+2026-10-15 15:50:36.363791: Current learning rate: 0.00923 
+2026-10-15 15:54:20.431179: train_loss -0.5756 
+2026-10-15 15:54:20.431898: val_loss -0.4183 
+2026-10-15 15:54:20.432380: Pseudo dice [np.float32(0.8856), np.float32(0.4978), np.float32(0.4964), np.float32(0.4575), np.float32(0.3776), np.float32(0.5132), np.float32(0.4545), np.float32(0.0), np.float32(0.0), np.float32(0.443), np.float32(0.3967), np.float32(0.4374), np.float32(0.5375), np.float32(0.466), np.float32(0.0), np.float32(0.0), np.float32(0.3637), np.float32(0.3757), np.float32(0.3774), np.float32(0.3092), np.float32(0.4482), np.float32(0.3817), np.float32(0.7033), np.float32(0.6637), np.float32(0.1806), np.float32(0.2767), np.float32(0.1206), np.float32(0.0), np.float32(0.3428), np.float32(0.4628), np.float32(0.0), np.float32(0.0), np.float32(0.3055), np.float32(0.3889), np.float32(0.5452), np.float32(0.5716)] 
+2026-10-15 15:54:20.432989: Epoch time: 224.07 s 
+2026-10-15 15:54:21.287447:  
+2026-10-15 15:54:21.288132: Epoch 86 
+2026-10-15 15:54:21.288591: Current learning rate: 0.00922 
+2026-10-15 15:57:50.151712: train_loss -0.5888 
+2026-10-15 15:57:50.152857: val_loss -0.4514 
+2026-10-15 15:57:50.153527: Pseudo dice [np.float32(0.8924), np.float32(0.5677), np.float32(0.5664), np.float32(0.5352), np.float32(0.4544), np.float32(0.5475), np.float32(0.4985), np.float32(0.0), np.float32(0.0), np.float32(0.5168), np.float32(0.4733), np.float32(0.5116), np.float32(0.5099), np.float32(0.5254), np.float32(0.0), np.float32(0.0), np.float32(0.4104), np.float32(0.298), np.float32(0.4746), np.float32(0.4012), np.float32(0.4255), np.float32(0.4637), np.float32(0.7229), np.float32(0.7162), np.float32(0.2796), np.float32(0.2954), np.float32(0.0697), np.float32(0.0), np.float32(0.4208), np.float32(0.5333), np.float32(0.0), np.float32(0.0), np.float32(0.3677), np.float32(0.3864), np.float32(0.5661), np.float32(0.6204)] 
+2026-10-15 15:57:50.153833: Epoch time: 208.87 s 
+2026-10-15 15:57:51.014896:  
+2026-10-15 15:57:51.015623: Epoch 87 
+2026-10-15 15:57:51.016077: Current learning rate: 0.00921 
+2026-10-15 16:01:33.435545: train_loss -0.6032 
+2026-10-15 16:01:33.436137: val_loss -0.4245 
+2026-10-15 16:01:33.436785: Pseudo dice [np.float32(0.8853), np.float32(0.5267), np.float32(0.5205), np.float32(0.482), np.float32(0.4381), np.float32(0.5171), np.float32(0.4515), np.float32(0.0), np.float32(0.0), np.float32(0.549), np.float32(0.403), np.float32(0.4649), np.float32(0.5241), np.float32(0.461), np.float32(0.0), np.float32(0.0), np.float32(0.3695), np.float32(0.2906), np.float32(0.4374), np.float32(0.3443), np.float32(0.4767), np.float32(0.3889), np.float32(0.7114), np.float32(0.6613), np.float32(0.195), np.float32(0.3118), np.float32(0.0714), np.float32(0.0), np.float32(0.4069), np.float32(0.4419), np.float32(0.0), np.float32(0.0), np.float32(0.3367), np.float32(0.44), np.float32(0.61), np.float32(0.6479)] 
+2026-10-15 16:01:33.437995: Epoch time: 222.42 s 
+2026-10-15 16:01:34.217922:  
+2026-10-15 16:01:34.218628: Epoch 88 
+2026-10-15 16:01:34.219012: Current learning rate: 0.0092 
+2026-10-15 16:05:28.706721: train_loss -0.6163 
+2026-10-15 16:05:28.707185: val_loss -0.4053 
+2026-10-15 16:05:28.707679: Pseudo dice [np.float32(0.8833), np.float32(0.4782), np.float32(0.4725), np.float32(0.4163), np.float32(0.3889), np.float32(0.4814), np.float32(0.4228), np.float32(0.0), np.float32(0.0), np.float32(0.5549), np.float32(0.363), np.float32(0.4238), np.float32(0.534), np.float32(0.5066), np.float32(0.0), np.float32(0.0), np.float32(0.2783), np.float32(0.279), np.float32(0.3647), np.float32(0.3068), np.float32(0.4259), np.float32(0.3567), np.float32(0.6288), np.float32(0.576), np.float32(0.1522), np.float32(0.2483), np.float32(0.0304), np.float32(0.0), np.float32(0.3608), np.float32(0.5652), np.float32(0.0), np.float32(0.0), np.float32(0.3949), np.float32(0.4428), np.float32(0.5178), np.float32(0.5694)] 
+2026-10-15 16:05:28.708766: Epoch time: 234.49 s 
+2026-10-15 16:05:29.486773:  
+2026-10-15 16:05:29.487509: Epoch 89 
+2026-10-15 16:05:29.487964: Current learning rate: 0.0092 
+2026-10-15 16:09:04.473880: train_loss -0.5971 
+2026-10-15 16:09:04.474700: val_loss -0.393 
+2026-10-15 16:09:04.475308: Pseudo dice [np.float32(0.8964), np.float32(0.4711), np.float32(0.4779), np.float32(0.4738), np.float32(0.4095), np.float32(0.5227), np.float32(0.4353), np.float32(0.0), np.float32(0.0), np.float32(0.5703), np.float32(0.3801), np.float32(0.383), np.float32(0.4675), np.float32(0.4014), np.float32(0.0), np.float32(0.0), np.float32(0.3758), np.float32(0.2993), np.float32(0.3909), np.float32(0.3472), np.float32(0.3962), np.float32(0.3965), np.float32(0.443), np.float32(0.5199), np.float32(0.2359), np.float32(0.27), np.float32(0.0741), np.float32(0.0), np.float32(0.377), np.float32(0.4131), np.float32(0.0), np.float32(0.0), np.float32(0.4385), np.float32(0.5005), np.float32(0.5269), np.float32(0.5245)] 
+2026-10-15 16:09:04.475655: Epoch time: 214.99 s 
+2026-10-15 16:09:05.271988:  
+2026-10-15 16:09:05.272842: Epoch 90 
+2026-10-15 16:09:05.273213: Current learning rate: 0.00919 
+2026-10-15 16:13:24.580649: train_loss -0.5448 
+2026-10-15 16:13:24.581333: val_loss -0.3262 
+2026-10-15 16:13:24.582059: Pseudo dice [np.float32(0.8575), np.float32(0.3552), np.float32(0.3424), np.float32(0.3068), np.float32(0.3954), np.float32(0.363), np.float32(0.3302), np.float32(0.0), np.float32(0.0), np.float32(0.4603), np.float32(0.2211), np.float32(0.3413), np.float32(0.4382), np.float32(0.4201), np.float32(0.0), np.float32(0.0), np.float32(0.2971), np.float32(0.2447), np.float32(0.3163), np.float32(0.173), np.float32(0.3059), np.float32(0.2026), np.float32(0.3394), np.float32(0.4079), np.float32(0.1598), np.float32(0.2384), np.float32(0.0754), np.float32(0.0), np.float32(0.2641), np.float32(0.4332), np.float32(0.0), np.float32(0.0), np.float32(0.3632), np.float32(0.3934), np.float32(0.3804), np.float32(0.44)] 
+2026-10-15 16:13:24.582628: Epoch time: 259.31 s 
+2026-10-15 16:13:25.414824:  
+2026-10-15 16:13:25.415614: Epoch 91 
+2026-10-15 16:13:25.416065: Current learning rate: 0.00918 
+2026-10-15 16:17:18.433225: train_loss -0.5879 
+2026-10-15 16:17:18.433965: val_loss -0.4593 
+2026-10-15 16:17:18.434717: Pseudo dice [np.float32(0.8653), np.float32(0.5716), np.float32(0.5717), np.float32(0.6184), np.float32(0.576), np.float32(0.6636), np.float32(0.5856), np.float32(0.0), np.float32(0.0), np.float32(0.4154), np.float32(0.5141), np.float32(0.5624), np.float32(0.4168), np.float32(0.4826), np.float32(0.0), np.float32(0.0), np.float32(0.543), np.float32(0.4018), np.float32(0.6209), np.float32(0.4744), np.float32(0.4892), np.float32(0.4554), np.float32(0.5997), np.float32(0.5143), np.float32(0.4321), np.float32(0.3753), np.float32(0.167), np.float32(0.0), np.float32(0.421), np.float32(0.5118), np.float32(0.0), np.float32(0.0), np.float32(0.4501), np.float32(0.5716), np.float32(0.6779), np.float32(0.7171)] 
+2026-10-15 16:17:18.435322: Epoch time: 233.02 s 
+2026-10-15 16:17:19.199385:  
+2026-10-15 16:17:19.200192: Epoch 92 
+2026-10-15 16:17:19.200585: Current learning rate: 0.00917 
+2026-10-15 16:21:09.953701: train_loss -0.5648 
+2026-10-15 16:21:09.954626: val_loss -0.4477 
+2026-10-15 16:21:09.955245: Pseudo dice [np.float32(0.8923), np.float32(0.5304), np.float32(0.6303), np.float32(0.6444), np.float32(0.6175), np.float32(0.6294), np.float32(0.6039), np.float32(0.0), np.float32(0.0), np.float32(0.5089), np.float32(0.546), np.float32(0.4809), np.float32(0.4738), np.float32(0.4762), np.float32(0.0), np.float32(0.0), np.float32(0.666), np.float32(0.47), np.float32(0.588), np.float32(0.4517), np.float32(0.4663), np.float32(0.5576), np.float32(0.6894), np.float32(0.7057), np.float32(0.4828), np.float32(0.3474), np.float32(0.2608), np.float32(0.0), np.float32(0.3529), np.float32(0.4802), np.float32(0.0), np.float32(0.0), np.float32(0.5119), np.float32(0.5342), np.float32(0.6584), np.float32(0.6477)] 
+2026-10-15 16:21:09.955707: Epoch time: 230.76 s 
+2026-10-15 16:21:09.956007: Yayy! New best EMA pseudo Dice: 0.3538999855518341 
+2026-10-15 16:21:11.937860:  
+2026-10-15 16:21:11.953630: Epoch 93 
+2026-10-15 16:21:11.954113: Current learning rate: 0.00916 
+2026-10-15 16:24:44.278024: train_loss -0.583 
+2026-10-15 16:24:44.278735: val_loss -0.3582 
+2026-10-15 16:24:44.279620: Pseudo dice [np.float32(0.8792), np.float32(0.437), np.float32(0.4438), np.float32(0.302), np.float32(0.2249), np.float32(0.3359), np.float32(0.2647), np.float32(0.0), np.float32(0.0), np.float32(0.5653), np.float32(0.3243), np.float32(0.3151), np.float32(0.371), np.float32(0.3112), np.float32(0.0), np.float32(0.0), np.float32(0.3548), np.float32(0.456), np.float32(0.3916), np.float32(0.5045), np.float32(0.4636), np.float32(0.507), np.float32(0.4167), np.float32(0.4981), np.float32(0.4757), np.float32(0.2373), np.float32(0.1755), np.float32(0.0), np.float32(0.2886), np.float32(0.4385), np.float32(0.0), np.float32(0.0), np.float32(0.2164), np.float32(0.2646), np.float32(0.3342), np.float32(0.4055)] 
+2026-10-15 16:24:44.280165: Epoch time: 212.34 s 
+2026-10-15 16:24:45.140523:  
+2026-10-15 16:24:45.141148: Epoch 94 
+2026-10-15 16:24:45.141351: Current learning rate: 0.00915 
+2026-10-15 16:28:29.884757: train_loss -0.5713 
+2026-10-15 16:28:29.885893: val_loss -0.4947 
+2026-10-15 16:28:29.886667: Pseudo dice [np.float32(0.8657), np.float32(0.6306), np.float32(0.6751), np.float32(0.6745), np.float32(0.6505), np.float32(0.7356), np.float32(0.608), np.float32(0.0), np.float32(0.0), np.float32(0.4226), np.float32(0.5735), np.float32(0.5809), np.float32(0.541), np.float32(0.5271), np.float32(0.0), np.float32(0.0), np.float32(0.6462), np.float32(0.4665), np.float32(0.6174), np.float32(0.5143), np.float32(0.5729), np.float32(0.5221), np.float32(0.7067), np.float32(0.6577), np.float32(0.5806), np.float32(0.5041), np.float32(0.1159), np.float32(0.0), np.float32(0.4735), np.float32(0.513), np.float32(0.0), np.float32(0.0), np.float32(0.547), np.float32(0.6551), np.float32(0.6914), np.float32(0.6912)] 
+2026-10-15 16:28:29.886992: Epoch time: 224.75 s 
+2026-10-15 16:28:29.887346: Yayy! New best EMA pseudo Dice: 0.3617999851703644 
+2026-10-15 16:28:31.270505:  
+2026-10-15 16:28:31.271574: Epoch 95 
+2026-10-15 16:28:31.271943: Current learning rate: 0.00914 
+2026-10-15 16:32:32.667934: train_loss -0.5812 
+2026-10-15 16:32:32.668483: val_loss -0.3746 
+2026-10-15 16:32:32.669534: Pseudo dice [np.float32(0.8654), np.float32(0.4392), np.float32(0.4102), np.float32(0.338), np.float32(0.3343), np.float32(0.3947), np.float32(0.336), np.float32(0.0), np.float32(0.0), np.float32(0.5939), np.float32(0.2398), np.float32(0.3793), np.float32(0.3966), np.float32(0.369), np.float32(0.0), np.float32(0.0), np.float32(0.4034), np.float32(0.3726), np.float32(0.3133), np.float32(0.3519), np.float32(0.456), np.float32(0.4557), np.float32(0.5491), np.float32(0.4731), np.float32(0.3101), np.float32(0.2809), np.float32(0.2326), np.float32(0.0), np.float32(0.4427), np.float32(0.4521), np.float32(0.0), np.float32(0.0), np.float32(0.2535), np.float32(0.2384), np.float32(0.414), np.float32(0.4193)] 
+2026-10-15 16:32:32.669978: Epoch time: 241.4 s 
+2026-10-15 16:32:33.495105:  
+2026-10-15 16:32:33.495792: Epoch 96 
+2026-10-15 16:32:33.496208: Current learning rate: 0.00913 
+2026-10-15 16:36:18.960636: train_loss -0.5626 
+2026-10-15 16:36:18.961318: val_loss -0.2862 
+2026-10-15 16:36:18.961882: Pseudo dice [np.float32(0.8781), np.float32(0.274), np.float32(0.2621), np.float32(0.2861), np.float32(0.2021), np.float32(0.3048), np.float32(0.2405), np.float32(0.0), np.float32(0.0), np.float32(0.4624), np.float32(0.2903), np.float32(0.3107), np.float32(0.3258), np.float32(0.2923), np.float32(0.0), np.float32(0.0), np.float32(0.192), np.float32(0.1181), np.float32(0.2136), np.float32(0.1947), np.float32(0.2795), np.float32(0.2251), np.float32(0.2529), np.float32(0.3025), np.float32(0.105), np.float32(0.0806), np.float32(0.0696), np.float32(0.0), np.float32(0.101), np.float32(0.2084), np.float32(0.0), np.float32(0.0), np.float32(0.1222), np.float32(0.1237), np.float32(0.2435), np.float32(0.2508)] 
+2026-10-15 16:36:18.962330: Epoch time: 225.47 s 
+2026-10-15 16:36:19.797507:  
+2026-10-15 16:36:19.798306: Epoch 97 
+2026-10-15 16:36:19.798825: Current learning rate: 0.00912 
+2026-10-15 16:40:25.921436: train_loss -0.5567 
+2026-10-15 16:40:25.922399: val_loss -0.4875 
+2026-10-15 16:40:25.922965: Pseudo dice [np.float32(0.895), np.float32(0.6063), np.float32(0.7064), np.float32(0.705), np.float32(0.654), np.float32(0.7331), np.float32(0.6688), np.float32(0.0), np.float32(0.0), np.float32(0.4491), np.float32(0.6044), np.float32(0.5642), np.float32(0.5295), np.float32(0.4731), np.float32(0.0), np.float32(0.0), np.float32(0.6055), np.float32(0.5111), np.float32(0.539), np.float32(0.4053), np.float32(0.537), np.float32(0.5811), np.float32(0.7832), np.float32(0.7805), np.float32(0.5517), np.float32(0.3886), np.float32(0.2176), np.float32(0.0), np.float32(0.6076), np.float32(0.5404), np.float32(0.0), np.float32(0.0), np.float32(0.4623), np.float32(0.5271), np.float32(0.7811), np.float32(0.8012)] 
+2026-10-15 16:40:25.923482: Epoch time: 246.13 s 
+2026-10-15 16:40:26.702204:  
+2026-10-15 16:40:26.702940: Epoch 98 
+2026-10-15 16:40:26.703332: Current learning rate: 0.00911 
+2026-10-15 16:44:23.394696: train_loss -0.5919 
+2026-10-15 16:44:23.395506: val_loss -0.5136 
+2026-10-15 16:44:23.395938: Pseudo dice [np.float32(0.8878), np.float32(0.6345), np.float32(0.7101), np.float32(0.7215), np.float32(0.6924), np.float32(0.7597), np.float32(0.6675), np.float32(0.0), np.float32(0.0), np.float32(0.4393), np.float32(0.6146), np.float32(0.6164), np.float32(0.4998), np.float32(0.5159), np.float32(0.0), np.float32(0.0), np.float32(0.6866), np.float32(0.5429), np.float32(0.6906), np.float32(0.5225), np.float32(0.6026), np.float32(0.5759), np.float32(0.7498), np.float32(0.7521), np.float32(0.5114), np.float32(0.4858), np.float32(0.2733), np.float32(0.0), np.float32(0.5472), np.float32(0.5471), np.float32(0.0), np.float32(0.0), np.float32(0.4769), np.float32(0.5876), np.float32(0.7496), np.float32(0.7757)] 
+2026-10-15 16:44:23.396398: Epoch time: 236.69 s 
+2026-10-15 16:44:23.396806: Yayy! New best EMA pseudo Dice: 0.3695000112056732 
+2026-10-15 16:44:24.794840:  
+2026-10-15 16:44:24.795569: Epoch 99 
+2026-10-15 16:44:24.795911: Current learning rate: 0.0091 
+2026-10-15 16:48:03.607729: train_loss -0.5618 
+2026-10-15 16:48:03.608325: val_loss -0.4846 
+2026-10-15 16:48:03.608989: Pseudo dice [np.float32(0.8749), np.float32(0.6133), np.float32(0.6315), np.float32(0.7041), np.float32(0.6367), np.float32(0.7092), np.float32(0.6378), np.float32(0.0), np.float32(0.0), np.float32(0.5398), np.float32(0.5915), np.float32(0.5014), np.float32(0.4907), np.float32(0.4802), np.float32(0.0), np.float32(0.0), np.float32(0.6865), np.float32(0.5122), np.float32(0.6543), np.float32(0.5068), np.float32(0.5228), np.float32(0.5847), np.float32(0.6749), np.float32(0.6871), np.float32(0.5363), np.float32(0.4326), np.float32(0.2483), np.float32(0.0), np.float32(0.5084), np.float32(0.5288), np.float32(0.0), np.float32(0.0), np.float32(0.5273), np.float32(0.6358), np.float32(0.7115), np.float32(0.7194)] 
+2026-10-15 16:48:03.609701: Epoch time: 218.82 s 
+2026-10-15 16:48:04.158031: Yayy! New best EMA pseudo Dice: 0.3799999952316284 
+2026-10-15 16:48:06.959981:  
+2026-10-15 16:48:06.962806: Epoch 100 
+2026-10-15 16:48:06.963629: Current learning rate: 0.0091 
+2026-10-15 16:51:51.791735: train_loss -0.5713 
+2026-10-15 16:51:51.792702: val_loss -0.3328 
+2026-10-15 16:51:51.793243: Pseudo dice [np.float32(0.8579), np.float32(0.3116), np.float32(0.3881), np.float32(0.4518), np.float32(0.2907), np.float32(0.3346), np.float32(0.4524), np.float32(0.0), np.float32(0.0), np.float32(0.4994), np.float32(0.3293), np.float32(0.2916), np.float32(0.3025), np.float32(0.2194), np.float32(0.0), np.float32(0.0), np.float32(0.1855), np.float32(0.1291), np.float32(0.3941), np.float32(0.3159), np.float32(0.3349), np.float32(0.4472), np.float32(0.495), np.float32(0.5721), np.float32(0.4152), np.float32(0.2365), np.float32(0.1148), np.float32(0.0), np.float32(0.3152), np.float32(0.3186), np.float32(0.0), np.float32(0.0), np.float32(0.3422), np.float32(0.3673), np.float32(0.3572), np.float32(0.24)] 
+2026-10-15 16:51:51.793637: Epoch time: 224.83 s 
+2026-10-15 16:51:52.682123:  
+2026-10-15 16:51:52.682611: Epoch 101 
+2026-10-15 16:51:52.682894: Current learning rate: 0.00909 
+2026-10-15 16:55:46.374220: train_loss -0.5341 
+2026-10-15 16:55:46.374828: val_loss -0.4228 
+2026-10-15 16:55:46.375328: Pseudo dice [np.float32(0.8867), np.float32(0.5408), np.float32(0.5429), np.float32(0.4944), np.float32(0.4114), np.float32(0.5142), np.float32(0.4308), np.float32(0.0), np.float32(0.0), np.float32(0.4775), np.float32(0.4526), np.float32(0.4644), np.float32(0.5534), np.float32(0.5012), np.float32(0.0), np.float32(0.0), np.float32(0.5065), np.float32(0.4484), np.float32(0.4676), np.float32(0.5004), np.float32(0.5045), np.float32(0.5261), np.float32(0.541), np.float32(0.577), np.float32(0.4248), np.float32(0.3147), np.float32(0.1529), np.float32(0.0), np.float32(0.3526), np.float32(0.5054), np.float32(0.0), np.float32(0.0), np.float32(0.3127), np.float32(0.3543), np.float32(0.4975), np.float32(0.51)] 
+2026-10-15 16:55:46.376288: Epoch time: 233.69 s 
+2026-10-15 16:55:47.214436:  
+2026-10-15 16:55:47.215188: Epoch 102 
+2026-10-15 16:55:47.215537: Current learning rate: 0.00908 
+2026-10-15 16:59:17.060770: train_loss -0.5676 
+2026-10-15 16:59:17.061659: val_loss -0.5272 
+2026-10-15 16:59:17.062181: Pseudo dice [np.float32(0.7921), np.float32(0.6615), np.float32(0.7739), np.float32(0.7978), np.float32(0.7838), np.float32(0.815), np.float32(0.6895), np.float32(0.0), np.float32(0.0), np.float32(0.5099), np.float32(0.6285), np.float32(0.655), np.float32(0.5682), np.float32(0.529), np.float32(0.0), np.float32(0.0), np.float32(0.7429), np.float32(0.5917), np.float32(0.6596), np.float32(0.5812), np.float32(0.595), np.float32(0.6655), np.float32(0.7871), np.float32(0.7942), np.float32(0.5359), np.float32(0.5248), np.float32(0.1627), np.float32(0.0), np.float32(0.5164), np.float32(0.4599), np.float32(0.0), np.float32(0.0), np.float32(0.5765), np.float32(0.5645), np.float32(0.7486), np.float32(0.7869)] 
+2026-10-15 16:59:17.062486: Epoch time: 209.85 s 
+2026-10-15 16:59:17.062837: Yayy! New best EMA pseudo Dice: 0.38600000739097595 
+2026-10-15 16:59:18.753769:  
+2026-10-15 16:59:18.754522: Epoch 103 
+2026-10-15 16:59:18.754982: Current learning rate: 0.00907 
+2026-10-15 17:03:36.685548: train_loss -0.5355 
+2026-10-15 17:03:36.686089: val_loss -0.5199 
+2026-10-15 17:03:36.686791: Pseudo dice [np.float32(0.9037), np.float32(0.6443), np.float32(0.7114), np.float32(0.7376), np.float32(0.7486), np.float32(0.7517), np.float32(0.7028), np.float32(0.0), np.float32(0.0), np.float32(0.4406), np.float32(0.597), np.float32(0.6527), np.float32(0.5505), np.float32(0.4976), np.float32(0.0), np.float32(0.0), np.float32(0.6794), np.float32(0.5135), np.float32(0.7013), np.float32(0.5305), np.float32(0.5289), np.float32(0.5708), np.float32(0.7054), np.float32(0.6191), np.float32(0.485), np.float32(0.4591), np.float32(0.1753), np.float32(0.0), np.float32(0.4474), np.float32(0.3787), np.float32(0.0), np.float32(0.0), np.float32(0.5948), np.float32(0.516), np.float32(0.7691), np.float32(0.7591)] 
+2026-10-15 17:03:36.687901: Epoch time: 257.93 s 
+2026-10-15 17:03:36.688249: Yayy! New best EMA pseudo Dice: 0.39570000767707825 
+2026-10-15 17:03:38.048004:  
+2026-10-15 17:03:38.048787: Epoch 104 
+2026-10-15 17:03:38.049192: Current learning rate: 0.00906 
+2026-10-15 17:07:37.104631: train_loss -0.5797 
+2026-10-15 17:07:37.105488: val_loss -0.4479 
+2026-10-15 17:07:37.106212: Pseudo dice [np.float32(0.8848), np.float32(0.5761), np.float32(0.6281), np.float32(0.5373), np.float32(0.5191), np.float32(0.5897), np.float32(0.5116), np.float32(0.0), np.float32(0.0), np.float32(0.591), np.float32(0.4496), np.float32(0.5105), np.float32(0.5164), np.float32(0.4829), np.float32(0.0), np.float32(0.0), np.float32(0.4134), np.float32(0.3556), np.float32(0.4533), np.float32(0.4309), np.float32(0.4745), np.float32(0.5152), np.float32(0.6665), np.float32(0.6592), np.float32(0.3509), np.float32(0.3472), np.float32(0.0955), np.float32(0.0), np.float32(0.276), np.float32(0.5581), np.float32(0.0), np.float32(0.0), np.float32(0.3827), np.float32(0.4885), np.float32(0.6393), np.float32(0.6289)] 
+2026-10-15 17:07:37.106489: Epoch time: 239.06 s 
+2026-10-15 17:07:37.106850: Yayy! New best EMA pseudo Dice: 0.39649999141693115 
+2026-10-15 17:07:42.200958:  
+2026-10-15 17:07:42.201730: Epoch 105 
+2026-10-15 17:07:42.202033: Current learning rate: 0.00905 
+2026-10-15 17:11:37.815364: train_loss -0.5894 
+2026-10-15 17:11:37.815811: val_loss -0.5216 
+2026-10-15 17:11:37.816785: Pseudo dice [np.float32(0.8844), np.float32(0.6452), np.float32(0.7761), np.float32(0.7856), np.float32(0.7665), np.float32(0.8205), np.float32(0.7125), np.float32(0.0), np.float32(0.0), np.float32(0.3965), np.float32(0.6452), np.float32(0.6373), np.float32(0.515), np.float32(0.4878), np.float32(0.0), np.float32(0.0), np.float32(0.7353), np.float32(0.5357), np.float32(0.7054), np.float32(0.5253), np.float32(0.5186), np.float32(0.625), np.float32(0.7882), np.float32(0.8186), np.float32(0.5421), np.float32(0.4304), np.float32(0.2976), np.float32(0.0), np.float32(0.516), np.float32(0.4826), np.float32(0.0), np.float32(0.0), np.float32(0.4983), np.float32(0.5028), np.float32(0.8523), np.float32(0.8355)] 
+2026-10-15 17:11:37.817167: Epoch time: 235.62 s 
+2026-10-15 17:11:37.817592: Yayy! New best EMA pseudo Dice: 0.4075999855995178 
+2026-10-15 17:11:39.765839:  
+2026-10-15 17:11:39.766552: Epoch 106 
+2026-10-15 17:11:39.767136: Current learning rate: 0.00904 
+2026-10-15 17:15:46.154648: train_loss -0.5309 
+2026-10-15 17:15:46.155411: val_loss -0.534 
+2026-10-15 17:15:46.155958: Pseudo dice [np.float32(0.8874), np.float32(0.6743), np.float32(0.789), np.float32(0.8132), np.float32(0.7789), np.float32(0.8437), np.float32(0.746), np.float32(0.0), np.float32(0.0), np.float32(0.4998), np.float32(0.6903), np.float32(0.6914), np.float32(0.5507), np.float32(0.4978), np.float32(0.0), np.float32(0.0), np.float32(0.7455), np.float32(0.5415), np.float32(0.7127), np.float32(0.5228), np.float32(0.5142), np.float32(0.661), np.float32(0.8068), np.float32(0.782), np.float32(0.3814), np.float32(0.4875), np.float32(0.201), np.float32(0.0), np.float32(0.4396), np.float32(0.4574), np.float32(0.0), np.float32(0.0), np.float32(0.4668), np.float32(0.5367), np.float32(0.8437), np.float32(0.7906)] 
+2026-10-15 17:15:46.156620: Epoch time: 246.39 s 
+2026-10-15 17:15:46.156915: Yayy! New best EMA pseudo Dice: 0.41780000925064087 
+2026-10-15 17:15:47.507029:  
+2026-10-15 17:15:47.507933: Epoch 107 
+2026-10-15 17:15:47.508274: Current learning rate: 0.00903 
+2026-10-15 17:19:49.799307: train_loss -0.5813 
+2026-10-15 17:19:49.800379: val_loss -0.5154 
+2026-10-15 17:19:49.801006: Pseudo dice [np.float32(0.8824), np.float32(0.6466), np.float32(0.7119), np.float32(0.7147), np.float32(0.7048), np.float32(0.7732), np.float32(0.6572), np.float32(0.0), np.float32(0.0), np.float32(0.5329), np.float32(0.629), np.float32(0.6301), np.float32(0.5408), np.float32(0.4951), np.float32(0.0), np.float32(0.0), np.float32(0.6804), np.float32(0.4945), np.float32(0.6652), np.float32(0.519), np.float32(0.5176), np.float32(0.5966), np.float32(0.7718), np.float32(0.7626), np.float32(0.4822), np.float32(0.5445), np.float32(0.2185), np.float32(0.0), np.float32(0.4259), np.float32(0.5934), np.float32(0.0), np.float32(0.0), np.float32(0.567), np.float32(0.6095), np.float32(0.7961), np.float32(0.8108)] 
+2026-10-15 17:19:49.801465: Epoch time: 242.29 s 
+2026-10-15 17:19:49.801839: Yayy! New best EMA pseudo Dice: 0.4259999990463257 
+2026-10-15 17:19:51.400240:  
+2026-10-15 17:19:51.400887: Epoch 108 
+2026-10-15 17:19:51.401188: Current learning rate: 0.00902 
+2026-10-15 17:23:47.099365: train_loss -0.6064 
+2026-10-15 17:23:47.099997: val_loss -0.4726 
+2026-10-15 17:23:47.100798: Pseudo dice [np.float32(0.8791), np.float32(0.5544), np.float32(0.6508), np.float32(0.6426), np.float32(0.5412), np.float32(0.6968), np.float32(0.5714), np.float32(0.0), np.float32(0.0), np.float32(0.4405), np.float32(0.5664), np.float32(0.5919), np.float32(0.4998), np.float32(0.461), np.float32(0.0), np.float32(0.0), np.float32(0.629), np.float32(0.5317), np.float32(0.5865), np.float32(0.4998), np.float32(0.5377), np.float32(0.5772), np.float32(0.6442), np.float32(0.5602), np.float32(0.4072), np.float32(0.2778), np.float32(0.2526), np.float32(0.0), np.float32(0.4296), np.float32(0.44), np.float32(0.0), np.float32(0.0), np.float32(0.3861), np.float32(0.5265), np.float32(0.6862), np.float32(0.6936)] 
+2026-10-15 17:23:47.101293: Epoch time: 235.7 s 
+2026-10-15 17:23:47.101848: Yayy! New best EMA pseudo Dice: 0.42719998955726624 
+2026-10-15 17:23:48.635748:  
+2026-10-15 17:23:48.636481: Epoch 109 
+2026-10-15 17:23:48.636836: Current learning rate: 0.00901 
+2026-10-15 17:27:48.461591: train_loss -0.609 
+2026-10-15 17:27:48.462332: val_loss -0.4199 
+2026-10-15 17:27:48.462964: Pseudo dice [np.float32(0.8912), np.float32(0.4678), np.float32(0.5622), np.float32(0.4857), np.float32(0.4515), np.float32(0.5968), np.float32(0.519), np.float32(0.0), np.float32(0.0), np.float32(0.4475), np.float32(0.4947), np.float32(0.5311), np.float32(0.5119), np.float32(0.4648), np.float32(0.0), np.float32(0.0), np.float32(0.4164), np.float32(0.364), np.float32(0.4691), np.float32(0.425), np.float32(0.4547), np.float32(0.4983), np.float32(0.5855), np.float32(0.5306), np.float32(0.4034), np.float32(0.4059), np.float32(0.1842), np.float32(0.0), np.float32(0.2513), np.float32(0.4388), np.float32(0.0), np.float32(0.0), np.float32(0.3682), np.float32(0.4217), np.float32(0.5474), np.float32(0.5946)] 
+2026-10-15 17:27:48.463682: Epoch time: 239.83 s 
+2026-10-15 17:27:49.324916:  
+2026-10-15 17:27:49.325510: Epoch 110 
+2026-10-15 17:27:49.325873: Current learning rate: 0.009 
+2026-10-15 17:31:58.243887: train_loss -0.6129 
+2026-10-15 17:31:58.244578: val_loss -0.5238 
+2026-10-15 17:31:58.245509: Pseudo dice [np.float32(0.908), np.float32(0.6114), np.float32(0.7712), np.float32(0.7544), np.float32(0.7044), np.float32(0.8207), np.float32(0.706), np.float32(0.0), np.float32(0.0), np.float32(0.4556), np.float32(0.6286), np.float32(0.6594), np.float32(0.5851), np.float32(0.48), np.float32(0.0), np.float32(0.0), np.float32(0.6953), np.float32(0.553), np.float32(0.6428), np.float32(0.5159), np.float32(0.5436), np.float32(0.5886), np.float32(0.8037), np.float32(0.7945), np.float32(0.5607), np.float32(0.4562), np.float32(0.2297), np.float32(0.0), np.float32(0.5021), np.float32(0.5167), np.float32(0.0), np.float32(0.0), np.float32(0.4751), np.float32(0.552), np.float32(0.8325), np.float32(0.8359)] 
+2026-10-15 17:31:58.245876: Epoch time: 248.92 s 
+2026-10-15 17:31:58.246324: Yayy! New best EMA pseudo Dice: 0.4309999942779541 
+2026-10-15 17:32:00.780128:  
+2026-10-15 17:32:00.781054: Epoch 111 
+2026-10-15 17:32:00.781357: Current learning rate: 0.009 
+2026-10-15 17:35:54.385458: train_loss -0.595 
+2026-10-15 17:35:54.386272: val_loss -0.3295 
+2026-10-15 17:35:54.386954: Pseudo dice [np.float32(0.8898), np.float32(0.3173), np.float32(0.3798), np.float32(0.2731), np.float32(0.2319), np.float32(0.3708), np.float32(0.2852), np.float32(0.0), np.float32(0.0), np.float32(0.433), np.float32(0.3131), np.float32(0.3842), np.float32(0.4967), np.float32(0.4734), np.float32(0.0), np.float32(0.0), np.float32(0.2404), np.float32(0.2194), np.float32(0.2329), np.float32(0.2357), np.float32(0.2831), np.float32(0.2536), np.float32(0.4052), np.float32(0.4847), np.float32(0.1972), np.float32(0.1504), np.float32(0.1292), np.float32(0.0), np.float32(0.1893), np.float32(0.336), np.float32(0.0), np.float32(0.0), np.float32(0.176), np.float32(0.2723), np.float32(0.3066), np.float32(0.3753)] 
+2026-10-15 17:35:54.387390: Epoch time: 233.61 s 
+2026-10-15 17:35:55.229891:  
+2026-10-15 17:35:55.230733: Epoch 112 
+2026-10-15 17:35:55.231056: Current learning rate: 0.00899 
+2026-10-15 17:39:43.848546: train_loss -0.6141 
+2026-10-15 17:39:43.849332: val_loss -0.5163 
+2026-10-15 17:39:43.850078: Pseudo dice [np.float32(0.9003), np.float32(0.6224), np.float32(0.7307), np.float32(0.7278), np.float32(0.6836), np.float32(0.7826), np.float32(0.6743), np.float32(0.0), np.float32(0.0), np.float32(0.5057), np.float32(0.6124), np.float32(0.6429), np.float32(0.5587), np.float32(0.5349), np.float32(0.0), np.float32(0.0), np.float32(0.6976), np.float32(0.5644), np.float32(0.6882), np.float32(0.5582), np.float32(0.5943), np.float32(0.6258), np.float32(0.7558), np.float32(0.7233), np.float32(0.5085), np.float32(0.4682), np.float32(0.3458), np.float32(0.0), np.float32(0.5106), np.float32(0.5492), np.float32(0.0), np.float32(0.0), np.float32(0.6209), np.float32(0.6846), np.float32(0.7775), np.float32(0.805)] 
+2026-10-15 17:39:43.850497: Epoch time: 228.62 s 
+2026-10-15 17:39:44.684980:  
+2026-10-15 17:39:44.685714: Epoch 113 
+2026-10-15 17:39:44.686084: Current learning rate: 0.00898 
+2026-10-15 17:43:50.093529: train_loss -0.611 
+2026-10-15 17:43:50.094200: val_loss -0.4449 
+2026-10-15 17:43:50.094949: Pseudo dice [np.float32(0.8939), np.float32(0.5444), np.float32(0.6225), np.float32(0.5623), np.float32(0.5033), np.float32(0.5828), np.float32(0.5245), np.float32(0.0), np.float32(0.0), np.float32(0.4994), np.float32(0.495), np.float32(0.4694), np.float32(0.4616), np.float32(0.4634), np.float32(0.0), np.float32(0.0), np.float32(0.4922), np.float32(0.3731), np.float32(0.4724), np.float32(0.4084), np.float32(0.5144), np.float32(0.5211), np.float32(0.7695), np.float32(0.6913), np.float32(0.4662), np.float32(0.4609), np.float32(0.2296), np.float32(0.0), np.float32(0.3783), np.float32(0.5167), np.float32(0.0), np.float32(0.0), np.float32(0.5053), np.float32(0.49), np.float32(0.6646), np.float32(0.6995)] 
+2026-10-15 17:43:50.095343: Epoch time: 245.41 s 
+2026-10-15 17:43:50.929329:  
+2026-10-15 17:43:50.930240: Epoch 114 
+2026-10-15 17:43:50.930733: Current learning rate: 0.00897 
+2026-10-15 17:47:44.775323: train_loss -0.6143 
+2026-10-15 17:47:44.775886: val_loss -0.4781 
+2026-10-15 17:47:44.776746: Pseudo dice [np.float32(0.8962), np.float32(0.5567), np.float32(0.6623), np.float32(0.6777), np.float32(0.5839), np.float32(0.708), np.float32(0.6079), np.float32(0.0), np.float32(0.0), np.float32(0.5269), np.float32(0.5715), np.float32(0.5693), np.float32(0.5111), np.float32(0.4111), np.float32(0.0), np.float32(0.0), np.float32(0.6579), np.float32(0.5386), np.float32(0.644), np.float32(0.5613), np.float32(0.5454), np.float32(0.6287), np.float32(0.6877), np.float32(0.6442), np.float32(0.5446), np.float32(0.3657), np.float32(0.3055), np.float32(0.0), np.float32(0.5258), np.float32(0.5088), np.float32(0.0), np.float32(0.0), np.float32(0.5039), np.float32(0.5807), np.float32(0.7), np.float32(0.7328)] 
+2026-10-15 17:47:44.777482: Epoch time: 233.85 s 
+2026-10-15 17:47:45.623521:  
+2026-10-15 17:47:45.624484: Epoch 115 
+2026-10-15 17:47:45.624851: Current learning rate: 0.00896 
+2026-10-15 17:51:36.716381: train_loss -0.6199 
+2026-10-15 17:51:36.717300: val_loss -0.5143 
+2026-10-15 17:51:36.717771: Pseudo dice [np.float32(0.8848), np.float32(0.6139), np.float32(0.7144), np.float32(0.6979), np.float32(0.6531), np.float32(0.7541), np.float32(0.6421), np.float32(0.0), np.float32(0.0), np.float32(0.496), np.float32(0.6104), np.float32(0.6478), np.float32(0.5973), np.float32(0.5574), np.float32(0.0), np.float32(0.0), np.float32(0.6573), np.float32(0.5015), np.float32(0.6302), np.float32(0.5265), np.float32(0.5802), np.float32(0.5989), np.float32(0.7776), np.float32(0.7485), np.float32(0.4366), np.float32(0.4415), np.float32(0.1302), np.float32(0.0), np.float32(0.4064), np.float32(0.5314), np.float32(0.0), np.float32(0.0), np.float32(0.5514), np.float32(0.6244), np.float32(0.7469), np.float32(0.7472)] 
+2026-10-15 17:51:36.718284: Epoch time: 231.09 s 
+2026-10-15 17:51:36.718821: Yayy! New best EMA pseudo Dice: 0.4343000054359436 
+2026-10-15 17:51:38.083071:  
+2026-10-15 17:51:38.083833: Epoch 116 
+2026-10-15 17:51:38.084203: Current learning rate: 0.00895 
+2026-10-15 17:55:25.636365: train_loss -0.6092 
+2026-10-15 17:55:25.637467: val_loss -0.537 
+2026-10-15 17:55:25.637972: Pseudo dice [np.float32(0.8955), np.float32(0.6544), np.float32(0.7578), np.float32(0.7664), np.float32(0.7351), np.float32(0.8179), np.float32(0.7091), np.float32(0.0), np.float32(0.0), np.float32(0.5393), np.float32(0.6551), np.float32(0.6741), np.float32(0.5433), np.float32(0.5589), np.float32(0.0), np.float32(0.0), np.float32(0.7284), np.float32(0.5811), np.float32(0.7082), np.float32(0.5977), np.float32(0.5994), np.float32(0.6572), np.float32(0.8034), np.float32(0.7966), np.float32(0.5194), np.float32(0.4366), np.float32(0.1926), np.float32(0.0), np.float32(0.4386), np.float32(0.5319), np.float32(0.0), np.float32(0.0), np.float32(0.5202), np.float32(0.5807), np.float32(0.8236), np.float32(0.8166)] 
+2026-10-15 17:55:25.638506: Epoch time: 227.56 s 
+2026-10-15 17:55:25.638887: Yayy! New best EMA pseudo Dice: 0.4426000118255615 
+2026-10-15 17:55:26.973623:  
+2026-10-15 17:55:26.974334: Epoch 117 
+2026-10-15 17:55:26.974660: Current learning rate: 0.00894 
+2026-10-15 17:59:25.211046: train_loss -0.6133 
+2026-10-15 17:59:25.211841: val_loss -0.4889 
+2026-10-15 17:59:25.212434: Pseudo dice [np.float32(0.882), np.float32(0.6142), np.float32(0.6458), np.float32(0.6828), np.float32(0.641), np.float32(0.6899), np.float32(0.5916), np.float32(0.0), np.float32(0.0), np.float32(0.438), np.float32(0.5454), np.float32(0.621), np.float32(0.5798), np.float32(0.4785), np.float32(0.0), np.float32(0.0), np.float32(0.6629), np.float32(0.4708), np.float32(0.6152), np.float32(0.5005), np.float32(0.5623), np.float32(0.5369), np.float32(0.7266), np.float32(0.681), np.float32(0.4708), np.float32(0.3544), np.float32(0.2213), np.float32(0.0), np.float32(0.4393), np.float32(0.5671), np.float32(0.0), np.float32(0.0), np.float32(0.5352), np.float32(0.5151), np.float32(0.7225), np.float32(0.7126)] 
+2026-10-15 17:59:25.212908: Epoch time: 238.24 s 
+2026-10-15 17:59:25.213458: Yayy! New best EMA pseudo Dice: 0.4447999894618988 
+2026-10-15 17:59:27.129834:  
+2026-10-15 17:59:27.130517: Epoch 118 
+2026-10-15 17:59:27.131119: Current learning rate: 0.00893 
+2026-10-15 18:03:18.045047: train_loss -0.5832 
+2026-10-15 18:03:18.045899: val_loss -0.4021 
+2026-10-15 18:03:18.046442: Pseudo dice [np.float32(0.8793), np.float32(0.4565), np.float32(0.4901), np.float32(0.4254), np.float32(0.3619), np.float32(0.5032), np.float32(0.4306), np.float32(0.0), np.float32(0.0), np.float32(0.5097), np.float32(0.386), np.float32(0.3848), np.float32(0.4159), np.float32(0.4307), np.float32(0.0), np.float32(0.0), np.float32(0.3865), np.float32(0.2935), np.float32(0.4229), np.float32(0.3696), np.float32(0.4155), np.float32(0.3948), np.float32(0.7241), np.float32(0.6762), np.float32(0.3068), np.float32(0.3204), np.float32(0.1711), np.float32(0.0), np.float32(0.4256), np.float32(0.4753), np.float32(0.0), np.float32(0.0), np.float32(0.341), np.float32(0.5138), np.float32(0.4865), np.float32(0.5735)] 
+2026-10-15 18:03:18.046889: Epoch time: 230.92 s 
+2026-10-15 18:03:18.884027:  
+2026-10-15 18:03:18.884904: Epoch 119 
+2026-10-15 18:03:18.885354: Current learning rate: 0.00892 
+2026-10-15 18:07:06.623154: train_loss -0.6167 
+2026-10-15 18:07:06.624205: val_loss -0.3832 
+2026-10-15 18:07:06.624864: Pseudo dice [np.float32(0.8777), np.float32(0.4725), np.float32(0.4763), np.float32(0.4498), np.float32(0.4431), np.float32(0.4298), np.float32(0.4179), np.float32(0.0), np.float32(0.0), np.float32(0.5534), np.float32(0.3863), np.float32(0.3236), np.float32(0.3736), np.float32(0.3434), np.float32(0.0), np.float32(0.0), np.float32(0.513), np.float32(0.3767), np.float32(0.4876), np.float32(0.375), np.float32(0.4038), np.float32(0.4519), np.float32(0.604), np.float32(0.5605), np.float32(0.4977), np.float32(0.3118), np.float32(0.2251), np.float32(0.0), np.float32(0.3536), np.float32(0.4093), np.float32(0.0), np.float32(0.0), np.float32(0.3979), np.float32(0.4516), np.float32(0.5169), np.float32(0.5164)] 
+2026-10-15 18:07:06.625227: Epoch time: 227.74 s 
+2026-10-15 18:07:07.459400:  
+2026-10-15 18:07:07.460292: Epoch 120 
+2026-10-15 18:07:07.460729: Current learning rate: 0.00891 
+2026-10-15 18:10:55.865626: train_loss -0.6297 
+2026-10-15 18:10:55.866487: val_loss -0.4329 
+2026-10-15 18:10:55.867128: Pseudo dice [np.float32(0.8823), np.float32(0.5406), np.float32(0.5736), np.float32(0.5347), np.float32(0.5143), np.float32(0.5932), np.float32(0.4623), np.float32(0.0), np.float32(0.0), np.float32(0.5198), np.float32(0.383), np.float32(0.4603), np.float32(0.4399), np.float32(0.4039), np.float32(0.0), np.float32(0.0), np.float32(0.5433), np.float32(0.458), np.float32(0.5181), np.float32(0.4406), np.float32(0.5128), np.float32(0.4558), np.float32(0.6909), np.float32(0.6114), np.float32(0.4329), np.float32(0.4312), np.float32(0.1608), np.float32(0.0), np.float32(0.3623), np.float32(0.615), np.float32(0.0), np.float32(0.0), np.float32(0.3134), np.float32(0.4514), np.float32(0.5956), np.float32(0.5942)] 
+2026-10-15 18:10:55.867669: Epoch time: 228.41 s 
+2026-10-15 18:10:56.942252:  
+2026-10-15 18:10:56.943057: Epoch 121 
+2026-10-15 18:10:56.943505: Current learning rate: 0.0089 
+2026-10-15 18:14:54.933514: train_loss -0.6295 
+2026-10-15 18:14:54.934281: val_loss -0.3648 
+2026-10-15 18:14:54.934996: Pseudo dice [np.float32(0.8729), np.float32(0.4261), np.float32(0.3909), np.float32(0.338), np.float32(0.3586), np.float32(0.3963), np.float32(0.337), np.float32(0.0), np.float32(0.0), np.float32(0.5054), np.float32(0.2903), np.float32(0.3404), np.float32(0.4151), np.float32(0.4387), np.float32(0.0), np.float32(0.0), np.float32(0.3067), np.float32(0.2407), np.float32(0.3384), np.float32(0.2519), np.float32(0.3969), np.float32(0.3148), np.float32(0.5869), np.float32(0.4797), np.float32(0.2677), np.float32(0.2614), np.float32(0.0155), np.float32(0.0), np.float32(0.3194), np.float32(0.5091), np.float32(0.0), np.float32(0.0), np.float32(0.3416), np.float32(0.3835), np.float32(0.4072), np.float32(0.436)] 
+2026-10-15 18:14:54.935271: Epoch time: 237.99 s 
+2026-10-15 18:14:55.814889:  
+2026-10-15 18:14:55.815545: Epoch 122 
+2026-10-15 18:14:55.815887: Current learning rate: 0.00889 
+2026-10-15 18:19:04.302614: train_loss -0.638 
+2026-10-15 18:19:04.303378: val_loss -0.4642 
+2026-10-15 18:19:04.304233: Pseudo dice [np.float32(0.8533), np.float32(0.4749), np.float32(0.6334), np.float32(0.6632), np.float32(0.6229), np.float32(0.6515), np.float32(0.6251), np.float32(0.0), np.float32(0.0), np.float32(0.5173), np.float32(0.5176), np.float32(0.5404), np.float32(0.5255), np.float32(0.4495), np.float32(0.0), np.float32(0.0), np.float32(0.68), np.float32(0.5305), np.float32(0.6599), np.float32(0.5516), np.float32(0.4659), np.float32(0.536), np.float32(0.5977), np.float32(0.6086), np.float32(0.5299), np.float32(0.4055), np.float32(0.3081), np.float32(0.0), np.float32(0.2397), np.float32(0.5343), np.float32(0.0), np.float32(0.0), np.float32(0.5045), np.float32(0.5813), np.float32(0.6171), np.float32(0.6422)] 
+2026-10-15 18:19:04.305019: Epoch time: 248.49 s 
+2026-10-15 18:19:05.137392:  
+2026-10-15 18:19:05.138147: Epoch 123 
+2026-10-15 18:19:05.138637: Current learning rate: 0.00889 
+2026-10-15 18:22:51.095520: train_loss -0.6455 
+2026-10-15 18:22:51.096227: val_loss -0.499 
+2026-10-15 18:22:51.096977: Pseudo dice [np.float32(0.8903), np.float32(0.551), np.float32(0.6692), np.float32(0.672), np.float32(0.6122), np.float32(0.7074), np.float32(0.5875), np.float32(0.0), np.float32(0.0), np.float32(0.474), np.float32(0.5404), np.float32(0.5324), np.float32(0.5031), np.float32(0.4365), np.float32(0.0), np.float32(0.0), np.float32(0.6041), np.float32(0.5032), np.float32(0.6135), np.float32(0.5542), np.float32(0.517), np.float32(0.5515), np.float32(0.7523), np.float32(0.7375), np.float32(0.4896), np.float32(0.4642), np.float32(0.337), np.float32(0.0), np.float32(0.5048), np.float32(0.6048), np.float32(0.0), np.float32(0.0), np.float32(0.4081), np.float32(0.4874), np.float32(0.7231), np.float32(0.7538)] 
+2026-10-15 18:22:51.097312: Epoch time: 225.96 s 
+2026-10-15 18:22:51.963982:  
+2026-10-15 18:22:51.965043: Epoch 124 
+2026-10-15 18:22:51.965330: Current learning rate: 0.00888 
+2026-10-15 18:27:04.102104: train_loss -0.6518 
+2026-10-15 18:27:04.103000: val_loss -0.493 
+2026-10-15 18:27:04.103673: Pseudo dice [np.float32(0.8921), np.float32(0.5589), np.float32(0.641), np.float32(0.6682), np.float32(0.6014), np.float32(0.7226), np.float32(0.5862), np.float32(0.0), np.float32(0.0), np.float32(0.5997), np.float32(0.5632), np.float32(0.5631), np.float32(0.521), np.float32(0.472), np.float32(0.0), np.float32(0.0), np.float32(0.7011), np.float32(0.5158), np.float32(0.6474), np.float32(0.5511), np.float32(0.5273), np.float32(0.5658), np.float32(0.6735), np.float32(0.6661), np.float32(0.5617), np.float32(0.3393), np.float32(0.3205), np.float32(0.0), np.float32(0.0977), np.float32(0.5619), np.float32(0.0), np.float32(0.0), np.float32(0.528), np.float32(0.5695), np.float32(0.6767), np.float32(0.7059)] 
+2026-10-15 18:27:04.104357: Epoch time: 252.14 s 
+2026-10-15 18:27:04.912299:  
+2026-10-15 18:27:04.913122: Epoch 125 
+2026-10-15 18:27:04.913474: Current learning rate: 0.00887 
+2026-10-15 18:30:50.144145: train_loss -0.661 
+2026-10-15 18:30:50.145052: val_loss -0.4984 
+2026-10-15 18:30:50.145881: Pseudo dice [np.float32(0.8948), np.float32(0.6105), np.float32(0.7062), np.float32(0.6656), np.float32(0.6604), np.float32(0.7421), np.float32(0.6464), np.float32(0.0), np.float32(0.0), np.float32(0.4586), np.float32(0.5584), np.float32(0.51), np.float32(0.5339), np.float32(0.5054), np.float32(0.0), np.float32(0.0), np.float32(0.6804), np.float32(0.5228), np.float32(0.683), np.float32(0.5203), np.float32(0.5433), np.float32(0.6307), np.float32(0.7118), np.float32(0.6722), np.float32(0.5235), np.float32(0.4281), np.float32(0.3529), np.float32(0.0), np.float32(0.2688), np.float32(0.4934), np.float32(0.0), np.float32(0.0), np.float32(0.3203), np.float32(0.3412), np.float32(0.752), np.float32(0.7866)] 
+2026-10-15 18:30:50.146247: Epoch time: 225.23 s 
+2026-10-15 18:30:51.038546:  
+2026-10-15 18:30:51.039330: Epoch 126 
+2026-10-15 18:30:51.039735: Current learning rate: 0.00886 
+2026-10-15 18:34:23.239242: train_loss -0.6577 
+2026-10-15 18:34:23.239871: val_loss -0.4328 
+2026-10-15 18:34:23.240822: Pseudo dice [np.float32(0.8609), np.float32(0.5083), np.float32(0.5344), np.float32(0.515), np.float32(0.4658), np.float32(0.5784), np.float32(0.4387), np.float32(0.0), np.float32(0.0), np.float32(0.4457), np.float32(0.43), np.float32(0.4782), np.float32(0.4975), np.float32(0.5252), np.float32(0.0), np.float32(0.0), np.float32(0.4415), np.float32(0.401), np.float32(0.4518), np.float32(0.4374), np.float32(0.4517), np.float32(0.4516), np.float32(0.59), np.float32(0.548), np.float32(0.3058), np.float32(0.2945), np.float32(0.1597), np.float32(0.0), np.float32(0.3713), np.float32(0.4909), np.float32(0.0), np.float32(0.0), np.float32(0.2763), np.float32(0.3872), np.float32(0.5388), np.float32(0.5645)] 
+2026-10-15 18:34:23.241258: Epoch time: 212.2 s 
+2026-10-15 18:34:24.025030:  
+2026-10-15 18:34:24.025861: Epoch 127 
+2026-10-15 18:34:24.026142: Current learning rate: 0.00885 
+2026-10-15 18:38:39.447910: train_loss -0.6237 
+2026-10-15 18:38:39.448524: val_loss -0.4994 
+2026-10-15 18:38:39.449551: Pseudo dice [np.float32(0.8868), np.float32(0.5791), np.float32(0.7002), np.float32(0.6539), np.float32(0.6221), np.float32(0.7194), np.float32(0.6093), np.float32(0.0), np.float32(0.0), np.float32(0.4829), np.float32(0.5836), np.float32(0.5976), np.float32(0.5137), np.float32(0.4891), np.float32(0.0), np.float32(0.0), np.float32(0.664), np.float32(0.517), np.float32(0.6354), np.float32(0.5306), np.float32(0.6201), np.float32(0.5865), np.float32(0.7451), np.float32(0.7351), np.float32(0.4721), np.float32(0.4367), np.float32(0.2176), np.float32(0.0), np.float32(0.1753), np.float32(0.6151), np.float32(0.0), np.float32(0.0), np.float32(0.5879), np.float32(0.5919), np.float32(0.7015), np.float32(0.7473)] 
+2026-10-15 18:38:39.450040: Epoch time: 255.42 s 
+2026-10-15 18:38:40.291254:  
+2026-10-15 18:38:40.292017: Epoch 128 
+2026-10-15 18:38:40.292377: Current learning rate: 0.00884 
+2026-10-15 18:42:19.244274: train_loss -0.6392 
+2026-10-15 18:42:19.244829: val_loss -0.4888 
+2026-10-15 18:42:19.245497: Pseudo dice [np.float32(0.8935), np.float32(0.5811), np.float32(0.5841), np.float32(0.57), np.float32(0.5221), np.float32(0.6012), np.float32(0.5528), np.float32(0.0), np.float32(0.0), np.float32(0.5155), np.float32(0.4759), np.float32(0.464), np.float32(0.5638), np.float32(0.557), np.float32(0.0), np.float32(0.0), np.float32(0.4571), np.float32(0.3878), np.float32(0.5126), np.float32(0.4823), np.float32(0.5504), np.float32(0.5132), np.float32(0.7119), np.float32(0.7159), np.float32(0.2695), np.float32(0.3497), np.float32(0.0229), np.float32(0.0), np.float32(0.2904), np.float32(0.6148), np.float32(0.0), np.float32(0.0), np.float32(0.3558), np.float32(0.4509), np.float32(0.6099), np.float32(0.6566)] 
+2026-10-15 18:42:19.246361: Epoch time: 218.96 s 
+2026-10-15 18:42:20.098379:  
+2026-10-15 18:42:20.098915: Epoch 129 
+2026-10-15 18:42:20.099227: Current learning rate: 0.00883 
+2026-10-15 18:46:18.866969: train_loss -0.6559 
+2026-10-15 18:46:18.867732: val_loss -0.4831 
+2026-10-15 18:46:18.868164: Pseudo dice [np.float32(0.8978), np.float32(0.577), np.float32(0.6244), np.float32(0.5767), np.float32(0.4987), np.float32(0.6667), np.float32(0.5388), np.float32(0.0), np.float32(0.0), np.float32(0.5206), np.float32(0.5049), np.float32(0.5068), np.float32(0.481), np.float32(0.4494), np.float32(0.0), np.float32(0.0), np.float32(0.5187), np.float32(0.3762), np.float32(0.5468), np.float32(0.5035), np.float32(0.5332), np.float32(0.5252), np.float32(0.6538), np.float32(0.6272), np.float32(0.399), np.float32(0.3848), np.float32(0.2114), np.float32(0.0), np.float32(0.3317), np.float32(0.5582), np.float32(0.0), np.float32(0.0), np.float32(0.4971), np.float32(0.5668), np.float32(0.6046), np.float32(0.6416)] 
+2026-10-15 18:46:18.868549: Epoch time: 238.77 s 
+2026-10-15 18:46:19.780149:  
+2026-10-15 18:46:19.780940: Epoch 130 
+2026-10-15 18:46:19.781270: Current learning rate: 0.00882 
+2026-10-15 18:49:48.987891: train_loss -0.6632 
+2026-10-15 18:49:48.988751: val_loss -0.4655 
+2026-10-15 18:49:48.989369: Pseudo dice [np.float32(0.8694), np.float32(0.5057), np.float32(0.6286), np.float32(0.5682), np.float32(0.5185), np.float32(0.6526), np.float32(0.5765), np.float32(0.0), np.float32(0.0), np.float32(0.4411), np.float32(0.5059), np.float32(0.5473), np.float32(0.5467), np.float32(0.4202), np.float32(0.0), np.float32(0.0), np.float32(0.564), np.float32(0.4784), np.float32(0.5841), np.float32(0.5109), np.float32(0.5351), np.float32(0.5618), np.float32(0.6207), np.float32(0.5555), np.float32(0.4197), np.float32(0.3256), np.float32(0.0796), np.float32(0.0), np.float32(0.2704), np.float32(0.4033), np.float32(0.0), np.float32(0.0), np.float32(0.3918), np.float32(0.5323), np.float32(0.5861), np.float32(0.6377)] 
+2026-10-15 18:49:48.989699: Epoch time: 209.21 s 
+2026-10-15 18:49:49.760859:  
+2026-10-15 18:49:49.761539: Epoch 131 
+2026-10-15 18:49:49.761865: Current learning rate: 0.00881 
+2026-10-15 18:53:21.046484: train_loss -0.6595 
+2026-10-15 18:53:21.047006: val_loss -0.5426 
+2026-10-15 18:53:21.047479: Pseudo dice [np.float32(0.897), np.float32(0.6413), np.float32(0.7342), np.float32(0.7682), np.float32(0.7341), np.float32(0.8085), np.float32(0.7142), np.float32(0.0), np.float32(0.0), np.float32(0.51), np.float32(0.6216), np.float32(0.6148), np.float32(0.5353), np.float32(0.4949), np.float32(0.0), np.float32(0.0), np.float32(0.7105), np.float32(0.538), np.float32(0.7165), np.float32(0.6228), np.float32(0.5698), np.float32(0.6185), np.float32(0.8097), np.float32(0.8017), np.float32(0.5342), np.float32(0.5005), np.float32(0.3524), np.float32(0.0), np.float32(0.3675), np.float32(0.3819), np.float32(0.0), np.float32(0.0), np.float32(0.5188), np.float32(0.6487), np.float32(0.7952), np.float32(0.811)] 
+2026-10-15 18:53:21.048169: Epoch time: 211.29 s 
+2026-10-15 18:53:21.826292:  
+2026-10-15 18:53:21.826883: Epoch 132 
+2026-10-15 18:53:21.827242: Current learning rate: 0.0088 
+2026-10-15 18:57:02.070890: train_loss -0.6644 
+2026-10-15 18:57:02.072111: val_loss -0.4956 
+2026-10-15 18:57:02.072710: Pseudo dice [np.float32(0.8685), np.float32(0.5582), np.float32(0.6532), np.float32(0.689), np.float32(0.6087), np.float32(0.7197), np.float32(0.6326), np.float32(0.0), np.float32(0.0), np.float32(0.4446), np.float32(0.5507), np.float32(0.5103), np.float32(0.47), np.float32(0.4374), np.float32(0.0), np.float32(0.0), np.float32(0.6799), np.float32(0.4786), np.float32(0.6684), np.float32(0.539), np.float32(0.5665), np.float32(0.5476), np.float32(0.7474), np.float32(0.7125), np.float32(0.5043), np.float32(0.4874), np.float32(0.0417), np.float32(0.0), np.float32(0.2826), np.float32(0.3229), np.float32(0.0), np.float32(0.0), np.float32(0.5613), np.float32(0.6322), np.float32(0.7289), np.float32(0.7642)] 
+2026-10-15 18:57:02.073233: Epoch time: 220.25 s 
+2026-10-15 18:57:02.856236:  
+2026-10-15 18:57:02.856958: Epoch 133 
+2026-10-15 18:57:02.857333: Current learning rate: 0.00879 
+2026-10-15 19:00:31.406790: train_loss -0.6655 
+2026-10-15 19:00:31.407526: val_loss -0.5063 
+2026-10-15 19:00:31.408055: Pseudo dice [np.float32(0.8887), np.float32(0.6061), np.float32(0.7), np.float32(0.7489), np.float32(0.6678), np.float32(0.7715), np.float32(0.6482), np.float32(0.0), np.float32(0.0), np.float32(0.4845), np.float32(0.6043), np.float32(0.593), np.float32(0.5594), np.float32(0.4649), np.float32(0.0), np.float32(0.0), np.float32(0.7567), np.float32(0.5514), np.float32(0.6451), np.float32(0.5099), np.float32(0.6021), np.float32(0.5966), np.float32(0.7429), np.float32(0.7271), np.float32(0.4883), np.float32(0.4741), np.float32(0.2756), np.float32(0.0), np.float32(0.291), np.float32(0.4949), np.float32(0.0), np.float32(0.0), np.float32(0.5927), np.float32(0.5015), np.float32(0.7393), np.float32(0.77)] 
+2026-10-15 19:00:31.408490: Epoch time: 208.55 s 
+2026-10-15 19:00:32.315152:  
+2026-10-15 19:00:32.315959: Epoch 134 
+2026-10-15 19:00:32.316274: Current learning rate: 0.00879 
+2026-10-15 19:03:46.589282: train_loss -0.6676 
+2026-10-15 19:03:46.590150: val_loss -0.4641 
+2026-10-15 19:03:46.590960: Pseudo dice [np.float32(0.8943), np.float32(0.5594), np.float32(0.5853), np.float32(0.5677), np.float32(0.517), np.float32(0.6082), np.float32(0.5381), np.float32(0.0), np.float32(0.0), np.float32(0.4688), np.float32(0.3884), np.float32(0.4524), np.float32(0.4299), np.float32(0.3649), np.float32(0.0), np.float32(0.0), np.float32(0.5737), np.float32(0.4218), np.float32(0.5196), np.float32(0.5351), np.float32(0.5227), np.float32(0.5018), np.float32(0.7323), np.float32(0.7118), np.float32(0.5472), np.float32(0.4031), np.float32(0.1725), np.float32(0.0), np.float32(0.2636), np.float32(0.4342), np.float32(0.0), np.float32(0.0), np.float32(0.4453), np.float32(0.5616), np.float32(0.6965), np.float32(0.721)] 
+2026-10-15 19:03:46.591530: Epoch time: 194.28 s 
+2026-10-15 19:03:47.453235:  
+2026-10-15 19:03:47.454087: Epoch 135 
+2026-10-15 19:03:47.454421: Current learning rate: 0.00878 
+2026-10-15 19:07:32.453366: train_loss -0.6349 
+2026-10-15 19:07:32.454096: val_loss -0.4869 
+2026-10-15 19:07:32.454963: Pseudo dice [np.float32(0.8851), np.float32(0.5476), np.float32(0.5813), np.float32(0.6337), np.float32(0.5765), np.float32(0.6892), np.float32(0.5796), np.float32(0.0), np.float32(0.0), np.float32(0.4058), np.float32(0.5223), np.float32(0.5421), np.float32(0.4566), np.float32(0.4685), np.float32(0.0), np.float32(0.0), np.float32(0.6013), np.float32(0.4643), np.float32(0.5937), np.float32(0.4839), np.float32(0.4865), np.float32(0.4581), np.float32(0.6356), np.float32(0.6603), np.float32(0.5304), np.float32(0.3681), np.float32(0.001), np.float32(0.0), np.float32(0.386), np.float32(0.4581), np.float32(0.0), np.float32(0.0), np.float32(0.49), np.float32(0.5543), np.float32(0.6786), np.float32(0.6757)] 
+2026-10-15 19:07:32.455440: Epoch time: 225.0 s 
+2026-10-15 19:07:33.306865:  
+2026-10-15 19:07:33.307525: Epoch 136 
+2026-10-15 19:07:33.307827: Current learning rate: 0.00877 
+2026-10-15 19:11:04.981499: train_loss -0.5639 
+2026-10-15 19:11:04.982548: val_loss -0.325 
+2026-10-15 19:11:04.983052: Pseudo dice [np.float32(0.8751), np.float32(0.349), np.float32(0.2694), np.float32(0.2744), np.float32(0.2441), np.float32(0.3165), np.float32(0.2983), np.float32(0.0), np.float32(0.0), np.float32(0.1906), np.float32(0.335), np.float32(0.2109), np.float32(0.3556), np.float32(0.2227), np.float32(0.0), np.float32(0.0), np.float32(0.2452), np.float32(0.2044), np.float32(0.2723), np.float32(0.161), np.float32(0.261), np.float32(0.1893), np.float32(0.4485), np.float32(0.4475), np.float32(0.1256), np.float32(0.1759), np.float32(0.0), np.float32(0.0), np.float32(0.1311), np.float32(0.4107), np.float32(0.0), np.float32(0.0), np.float32(0.1234), np.float32(0.1793), np.float32(0.2984), np.float32(0.3484)] 
+2026-10-15 19:11:04.983559: Epoch time: 211.68 s 
+2026-10-15 19:11:05.854059:  
+2026-10-15 19:11:05.854793: Epoch 137 
+2026-10-15 19:11:05.855127: Current learning rate: 0.00876 
+2026-10-15 19:14:41.689621: train_loss -0.5851 
+2026-10-15 19:14:41.690454: val_loss -0.4494 
+2026-10-15 19:14:41.691148: Pseudo dice [np.float32(0.8956), np.float32(0.5275), np.float32(0.4763), np.float32(0.4778), np.float32(0.3642), np.float32(0.4955), np.float32(0.3855), np.float32(0.0), np.float32(0.0), np.float32(0.4311), np.float32(0.4122), np.float32(0.3925), np.float32(0.4312), np.float32(0.4152), np.float32(0.0), np.float32(0.0), np.float32(0.2871), np.float32(0.1971), np.float32(0.3963), np.float32(0.3531), np.float32(0.477), np.float32(0.4235), np.float32(0.6324), np.float32(0.5739), np.float32(0.2608), np.float32(0.2345), np.float32(0.0), np.float32(0.0), np.float32(0.1332), np.float32(0.4907), np.float32(0.0), np.float32(0.0), np.float32(0.3593), np.float32(0.4713), np.float32(0.5147), np.float32(0.5532)] 
+2026-10-15 19:14:41.691470: Epoch time: 215.84 s 
+2026-10-15 19:14:42.572940:  
+2026-10-15 19:14:42.573728: Epoch 138 
+2026-10-15 19:14:42.574162: Current learning rate: 0.00875 
+2026-10-15 19:18:19.940136: train_loss -0.6423 
+2026-10-15 19:18:19.940791: val_loss -0.4387 
+2026-10-15 19:18:19.941409: Pseudo dice [np.float32(0.8886), np.float32(0.5131), np.float32(0.4992), np.float32(0.5119), np.float32(0.4441), np.float32(0.4988), np.float32(0.4552), np.float32(0.0), np.float32(0.0), np.float32(0.5503), np.float32(0.4108), np.float32(0.4705), np.float32(0.5415), np.float32(0.4127), np.float32(0.0), np.float32(0.0), np.float32(0.3977), np.float32(0.3182), np.float32(0.4021), np.float32(0.3518), np.float32(0.4443), np.float32(0.4391), np.float32(0.5888), np.float32(0.4737), np.float32(0.3045), np.float32(0.2417), np.float32(0.0), np.float32(0.0), np.float32(0.1019), np.float32(0.3244), np.float32(0.0), np.float32(0.0), np.float32(0.3223), np.float32(0.4619), np.float32(0.4828), np.float32(0.4851)] 
+2026-10-15 19:18:19.942071: Epoch time: 217.37 s 
+2026-10-15 19:18:21.090913:  
+2026-10-15 19:18:21.091520: Epoch 139 
+2026-10-15 19:18:21.091982: Current learning rate: 0.00874 
+2026-10-15 19:21:43.007152: train_loss -0.6524 
+2026-10-15 19:21:43.007874: val_loss -0.4845 
+2026-10-15 19:21:43.008359: Pseudo dice [np.float32(0.8905), np.float32(0.5552), np.float32(0.6286), np.float32(0.6342), np.float32(0.5504), np.float32(0.6459), np.float32(0.5477), np.float32(0.0), np.float32(0.0), np.float32(0.3593), np.float32(0.4964), np.float32(0.5571), np.float32(0.5586), np.float32(0.4347), np.float32(0.0), np.float32(0.0), np.float32(0.5429), np.float32(0.45), np.float32(0.5392), np.float32(0.445), np.float32(0.5426), np.float32(0.5676), np.float32(0.6087), np.float32(0.5534), np.float32(0.3736), np.float32(0.4079), np.float32(0.0), np.float32(0.0), np.float32(0.3875), np.float32(0.4943), np.float32(0.0), np.float32(0.0), np.float32(0.5114), np.float32(0.5492), np.float32(0.6412), np.float32(0.6654)] 
+2026-10-15 19:21:43.008766: Epoch time: 201.92 s 
+2026-10-15 19:21:43.799437:  
+2026-10-15 19:21:43.800113: Epoch 140 
+2026-10-15 19:21:43.800522: Current learning rate: 0.00873 
+2026-10-15 19:25:01.673495: train_loss -0.6573 
+2026-10-15 19:25:01.674228: val_loss -0.4418 
+2026-10-15 19:25:01.674819: Pseudo dice [np.float32(0.8865), np.float32(0.5216), np.float32(0.5279), np.float32(0.4655), np.float32(0.4427), np.float32(0.5614), np.float32(0.4653), np.float32(0.0), np.float32(0.0), np.float32(0.4646), np.float32(0.3764), np.float32(0.4479), np.float32(0.439), np.float32(0.4216), np.float32(0.0), np.float32(0.0), np.float32(0.3886), np.float32(0.2747), np.float32(0.4485), np.float32(0.3687), np.float32(0.495), np.float32(0.4209), np.float32(0.6907), np.float32(0.671), np.float32(0.3072), np.float32(0.2863), np.float32(0.0), np.float32(0.0), np.float32(0.0393), np.float32(0.2408), np.float32(0.0), np.float32(0.0), np.float32(0.4401), np.float32(0.5082), np.float32(0.5513), np.float32(0.5836)] 
+2026-10-15 19:25:01.675203: Epoch time: 197.88 s 
+2026-10-15 19:25:02.486371:  
+2026-10-15 19:25:02.487003: Epoch 141 
+2026-10-15 19:25:02.487346: Current learning rate: 0.00872 
+2026-10-15 19:28:26.324012: train_loss -0.6573 
+2026-10-15 19:28:26.324748: val_loss -0.5365 
+2026-10-15 19:28:26.325150: Pseudo dice [np.float32(0.8669), np.float32(0.6211), np.float32(0.7121), np.float32(0.7321), np.float32(0.7019), np.float32(0.7454), np.float32(0.6981), np.float32(0.0), np.float32(0.0), np.float32(0.3516), np.float32(0.6348), np.float32(0.6231), np.float32(0.5429), np.float32(0.4649), np.float32(0.0), np.float32(0.0), np.float32(0.7275), np.float32(0.5748), np.float32(0.6914), np.float32(0.5473), np.float32(0.6063), np.float32(0.6581), np.float32(0.7748), np.float32(0.7377), np.float32(0.4866), np.float32(0.4997), np.float32(0.0), np.float32(0.0), np.float32(0.2378), np.float32(0.4104), np.float32(0.0), np.float32(0.0), np.float32(0.5903), np.float32(0.7049), np.float32(0.7726), np.float32(0.7562)] 
+2026-10-15 19:28:26.325413: Epoch time: 203.84 s 
+2026-10-15 19:28:27.197766:  
+2026-10-15 19:28:27.198538: Epoch 142 
+2026-10-15 19:28:27.198789: Current learning rate: 0.00871 
+2026-10-15 19:31:50.269768: train_loss -0.6711 
+2026-10-15 19:31:50.270339: val_loss -0.521 
+2026-10-15 19:31:50.270841: Pseudo dice [np.float32(0.8956), np.float32(0.5835), np.float32(0.7049), np.float32(0.6916), np.float32(0.6154), np.float32(0.7425), np.float32(0.6368), np.float32(0.0), np.float32(0.0), np.float32(0.3437), np.float32(0.594), np.float32(0.6144), np.float32(0.52), np.float32(0.4404), np.float32(0.0), np.float32(0.0), np.float32(0.6721), np.float32(0.4756), np.float32(0.679), np.float32(0.5344), np.float32(0.5336), np.float32(0.6077), np.float32(0.7295), np.float32(0.6863), np.float32(0.5042), np.float32(0.4954), np.float32(0.0), np.float32(0.0), np.float32(0.2831), np.float32(0.575), np.float32(0.0), np.float32(0.0), np.float32(0.55), np.float32(0.6532), np.float32(0.7107), np.float32(0.73)] 
+2026-10-15 19:31:50.271318: Epoch time: 203.07 s 
+2026-10-15 19:31:51.119110:  
+2026-10-15 19:31:51.119666: Epoch 143 
+2026-10-15 19:31:51.119965: Current learning rate: 0.0087 
+2026-10-15 19:35:17.903918: train_loss -0.6609 
+2026-10-15 19:35:17.904568: val_loss -0.4375 
+2026-10-15 19:35:17.905460: Pseudo dice [np.float32(0.8809), np.float32(0.4864), np.float32(0.4981), np.float32(0.4957), np.float32(0.4427), np.float32(0.5757), np.float32(0.4654), np.float32(0.0), np.float32(0.0), np.float32(0.2702), np.float32(0.4093), np.float32(0.4443), np.float32(0.5171), np.float32(0.4337), np.float32(0.0), np.float32(0.0), np.float32(0.4077), np.float32(0.299), np.float32(0.4005), np.float32(0.3369), np.float32(0.4442), np.float32(0.4366), np.float32(0.5219), np.float32(0.367), np.float32(0.2423), np.float32(0.2853), np.float32(0.0), np.float32(0.0), np.float32(0.2853), np.float32(0.4821), np.float32(0.0), np.float32(0.0), np.float32(0.281), np.float32(0.5019), np.float32(0.5179), np.float32(0.5705)] 
+2026-10-15 19:35:17.905888: Epoch time: 206.79 s 
+2026-10-15 19:35:18.743248:  
+2026-10-15 19:35:18.743945: Epoch 144 
+2026-10-15 19:35:18.744210: Current learning rate: 0.00869 
+2026-10-15 19:38:30.523603: train_loss -0.666 
+2026-10-15 19:38:30.524578: val_loss -0.4521 
+2026-10-15 19:38:30.525391: Pseudo dice [np.float32(0.8906), np.float32(0.4989), np.float32(0.4855), np.float32(0.4728), np.float32(0.4), np.float32(0.5429), np.float32(0.4719), np.float32(0.0), np.float32(0.0), np.float32(0.5373), np.float32(0.4279), np.float32(0.4122), np.float32(0.4865), np.float32(0.4914), np.float32(0.0), np.float32(0.0), np.float32(0.4057), np.float32(0.3542), np.float32(0.4148), np.float32(0.3676), np.float32(0.4593), np.float32(0.4025), np.float32(0.6586), np.float32(0.586), np.float32(0.2132), np.float32(0.2331), np.float32(0.0), np.float32(0.0), np.float32(0.334), np.float32(0.549), np.float32(0.0), np.float32(0.0), np.float32(0.3784), np.float32(0.5146), np.float32(0.5212), np.float32(0.5568)] 
+2026-10-15 19:38:30.525758: Epoch time: 191.78 s 
+2026-10-15 19:38:31.366421:  
+2026-10-15 19:38:31.367098: Epoch 145 
+2026-10-15 19:38:31.367646: Current learning rate: 0.00868 
+2026-10-15 19:41:54.192417: train_loss -0.6759 
+2026-10-15 19:41:54.193025: val_loss -0.5535 
+2026-10-15 19:41:54.193460: Pseudo dice [np.float32(0.8779), np.float32(0.6222), np.float32(0.7486), np.float32(0.7288), np.float32(0.6816), np.float32(0.8013), np.float32(0.6941), np.float32(0.0), np.float32(0.0), np.float32(0.2747), np.float32(0.63), np.float32(0.6548), np.float32(0.5616), np.float32(0.5192), np.float32(0.0), np.float32(0.0), np.float32(0.7314), np.float32(0.582), np.float32(0.7244), np.float32(0.5731), np.float32(0.6119), np.float32(0.6714), np.float32(0.7485), np.float32(0.7248), np.float32(0.4973), np.float32(0.4539), np.float32(0.0), np.float32(0.0), np.float32(0.4405), np.float32(0.5829), np.float32(0.0), np.float32(0.0), np.float32(0.5935), np.float32(0.6827), np.float32(0.806), np.float32(0.8222)] 
+2026-10-15 19:41:54.194148: Epoch time: 202.83 s 
+2026-10-15 19:41:55.005482:  
+2026-10-15 19:41:55.006164: Epoch 146 
+2026-10-15 19:41:55.006548: Current learning rate: 0.00868 
+2026-10-15 19:45:38.816473: train_loss -0.6817 
+2026-10-15 19:45:38.817343: val_loss -0.5436 
+2026-10-15 19:45:38.817924: Pseudo dice [np.float32(0.8853), np.float32(0.6021), np.float32(0.7098), np.float32(0.7274), np.float32(0.6642), np.float32(0.7677), np.float32(0.6437), np.float32(0.0), np.float32(0.0), np.float32(0.5091), np.float32(0.6119), np.float32(0.6366), np.float32(0.5762), np.float32(0.5221), np.float32(0.0), np.float32(0.0), np.float32(0.6872), np.float32(0.5266), np.float32(0.6584), np.float32(0.5721), np.float32(0.6121), np.float32(0.6386), np.float32(0.729), np.float32(0.7379), np.float32(0.5206), np.float32(0.4081), np.float32(0.0), np.float32(0.0), np.float32(0.3699), np.float32(0.6118), np.float32(0.0), np.float32(0.0), np.float32(0.5828), np.float32(0.6862), np.float32(0.777), np.float32(0.8037)] 
+2026-10-15 19:45:38.818403: Epoch time: 223.81 s 
+2026-10-15 19:45:39.667011:  
+2026-10-15 19:45:39.667716: Epoch 147 
+2026-10-15 19:45:39.668111: Current learning rate: 0.00867 
+2026-10-15 19:49:03.211034: train_loss -0.6749 
+2026-10-15 19:49:03.211984: val_loss -0.5594 
+2026-10-15 19:49:03.212336: Pseudo dice [np.float32(0.8909), np.float32(0.627), np.float32(0.7677), np.float32(0.7767), np.float32(0.7273), np.float32(0.7997), np.float32(0.6819), np.float32(0.0), np.float32(0.0), np.float32(0.3631), np.float32(0.657), np.float32(0.6708), np.float32(0.5583), np.float32(0.5276), np.float32(0.0), np.float32(0.0), np.float32(0.753), np.float32(0.5936), np.float32(0.7154), np.float32(0.5866), np.float32(0.6138), np.float32(0.6717), np.float32(0.8001), np.float32(0.8094), np.float32(0.5594), np.float32(0.4465), np.float32(0.0), np.float32(0.0), np.float32(0.2516), np.float32(0.559), np.float32(0.0), np.float32(0.0), np.float32(0.5757), np.float32(0.6369), np.float32(0.8097), np.float32(0.8161)] 
+2026-10-15 19:49:03.212723: Epoch time: 203.55 s 
+2026-10-15 19:49:03.972798:  
+2026-10-15 19:49:03.973482: Epoch 148 
+2026-10-15 19:49:03.973852: Current learning rate: 0.00866 
+2026-10-15 19:52:23.035747: train_loss -0.6658 
+2026-10-15 19:52:23.036515: val_loss -0.4039 
+2026-10-15 19:52:23.037410: Pseudo dice [np.float32(0.8899), np.float32(0.4209), np.float32(0.4456), np.float32(0.4545), np.float32(0.3509), np.float32(0.4893), np.float32(0.4134), np.float32(0.0), np.float32(0.0), np.float32(0.5021), np.float32(0.4595), np.float32(0.447), np.float32(0.4368), np.float32(0.3926), np.float32(0.0), np.float32(0.0), np.float32(0.2557), np.float32(0.2376), np.float32(0.3904), np.float32(0.3379), np.float32(0.3809), np.float32(0.354), np.float32(0.5241), np.float32(0.4704), np.float32(0.332), np.float32(0.219), np.float32(0.0), np.float32(0.0), np.float32(0.2582), np.float32(0.4042), np.float32(0.0), np.float32(0.0), np.float32(0.2345), np.float32(0.4122), np.float32(0.4426), np.float32(0.4937)] 
+2026-10-15 19:52:23.037752: Epoch time: 199.07 s 
+2026-10-15 19:52:23.942538:  
+2026-10-15 19:52:23.943327: Epoch 149 
+2026-10-15 19:52:23.943717: Current learning rate: 0.00865 
+2026-10-15 19:55:52.791834: train_loss -0.6324 
+2026-10-15 19:55:52.792630: val_loss -0.2427 
+2026-10-15 19:55:52.793167: Pseudo dice [np.float32(0.8988), np.float32(0.0996), np.float32(0.1328), np.float32(0.1481), np.float32(0.0711), np.float32(0.1276), np.float32(0.0979), np.float32(0.0), np.float32(0.0), np.float32(0.2912), np.float32(0.1224), np.float32(0.212), np.float32(0.2903), np.float32(0.1347), np.float32(0.0), np.float32(0.0), np.float32(0.0761), np.float32(0.1059), np.float32(0.0978), np.float32(0.0769), np.float32(0.1181), np.float32(0.1177), np.float32(0.2461), np.float32(0.1521), np.float32(0.0763), np.float32(0.0363), np.float32(0.0), np.float32(0.0), np.float32(0.0076), np.float32(0.1009), np.float32(0.0), np.float32(0.0), np.float32(0.0478), np.float32(0.0691), np.float32(0.1158), np.float32(0.1274)] 
+2026-10-15 19:55:52.793565: Epoch time: 208.85 s 
+2026-10-15 19:55:54.899853:  
+2026-10-15 19:55:54.900419: Epoch 150 
+2026-10-15 19:55:54.900902: Current learning rate: 0.00864 
+2026-10-15 19:59:26.946233: train_loss -0.6599 
+2026-10-15 19:59:26.947259: val_loss -0.4913 
+2026-10-15 19:59:26.947922: Pseudo dice [np.float32(0.892), np.float32(0.5607), np.float32(0.6219), np.float32(0.6013), np.float32(0.5162), np.float32(0.6285), np.float32(0.5409), np.float32(0.0), np.float32(0.0), np.float32(0.4963), np.float32(0.5385), np.float32(0.543), np.float32(0.5901), np.float32(0.5464), np.float32(0.0), np.float32(0.0), np.float32(0.5047), np.float32(0.4888), np.float32(0.5263), np.float32(0.4701), np.float32(0.531), np.float32(0.5299), np.float32(0.6004), np.float32(0.5416), np.float32(0.4114), np.float32(0.3922), np.float32(0.0), np.float32(0.0), np.float32(0.3987), np.float32(0.5666), np.float32(0.0), np.float32(0.0), np.float32(0.3006), np.float32(0.3992), np.float32(0.5995), np.float32(0.6138)] 
+2026-10-15 19:59:26.948405: Epoch time: 212.05 s 
+2026-10-15 19:59:27.802187:  
+2026-10-15 19:59:27.802841: Epoch 151 
+2026-10-15 19:59:27.803151: Current learning rate: 0.00863 
diff --git a/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/plans.json b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/plans.json
new file mode 100644
index 0000000000000000000000000000000000000000..5e91dc1c79ce056bedda5be7fd435ccfb89f8dd4
--- /dev/null
+++ b/nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/plans.json
@@ -0,0 +1,356 @@
+{
+    "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": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 7,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        512,
+                        512,
+                        512
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv2d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm2d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": true
+        },
+        "3d_fullres": {
+            "data_identifier": "nnUNetPlans_3d_fullres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                128,
+                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": null
+            },
+            "resampling_fn_seg_kwargs": {
+                "is_seg": true,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "resampling_fn_probabilities": "resample_data_or_seg_to_shape",
+            "resampling_fn_probabilities_kwargs": {
+                "is_seg": false,
+                "order": 1,
+                "order_z": 0,
+                "force_separate_z": null
+            },
+            "architecture": {
+                "network_class_name": "dynamic_network_architectures.architectures.unet.ResidualEncoderUNet",
+                "arch_kwargs": {
+                    "n_stages": 6,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        320,
+                        320
+                    ],
+                    "conv_op": "torch.nn.modules.conv.Conv3d",
+                    "kernel_sizes": [
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        1,
+                        1,
+                        1,
+                        1
+                    ],
+                    "conv_bias": true,
+                    "norm_op": "torch.nn.modules.instancenorm.InstanceNorm3d",
+                    "norm_op_kwargs": {
+                        "eps": 1e-05,
+                        "affine": true
+                    },
+                    "dropout_op": null,
+                    "dropout_op_kwargs": null,
+                    "nonlin": "torch.nn.LeakyReLU",
+                    "nonlin_kwargs": {
+                        "inplace": true
+                    }
+                },
+                "_kw_requires_import": [
+                    "conv_op",
+                    "norm_op",
+                    "dropout_op",
+                    "nonlin"
+                ]
+            },
+            "batch_dice": false
+        }
+    },
+    "experiment_planner_used": "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
+        }
+    }
+}
\ No newline at end of file