diff --git a/.gitattributes b/.gitattributes
index c45ddb50a869f1574c57997eedf434fb68188c11..5bba2e84b1425e7074becb135247e0c6d812181f 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -37,3 +37,4 @@ MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTra
 nnUNet_Large/fold_0/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png filter=lfs diff=lfs merge=lfs -text
 MedNeXt_Base/fold_1/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_1/progress.png filter=lfs diff=lfs merge=lfs -text
 nnUNet_Large/fold_1/nnUNet_results/Dataset001_CT/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_1/progress.png filter=lfs diff=lfs merge=lfs -text
+nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png filter=lfs diff=lfs merge=lfs -text
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/.DS_Store b/nnUNet_Large/fold_0_MRA/nnUNet_results/.DS_Store
new file mode 100644
index 0000000000000000000000000000000000000000..8894020dc419c0150d330d6dda63f88feb30bb16
Binary files /dev/null and b/nnUNet_Large/fold_0_MRA/nnUNet_results/.DS_Store differ
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/.DS_Store b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/.DS_Store
new file mode 100644
index 0000000000000000000000000000000000000000..740a789da77fbd15ea2522e6dcdb6922c1a315b7
Binary files /dev/null and b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/.DS_Store differ
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/dataset.json b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/dataset.json
new file mode 100644
index 0000000000000000000000000000000000000000..b8995e1f67ee44362dd9d7e5c7246c18f05baf0b
--- /dev/null
+++ b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/dataset.json
@@ -0,0 +1,46 @@
+{
+    "channel_names": {
+        "0": "MRA"
+    },
+    "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_MRA/nnUNet_results/Dataset001_MRA/dataset_fingerprint.json b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/dataset_fingerprint.json
new file mode 100644
index 0000000000000000000000000000000000000000..6b5b7e30e25c19b83c992d94e874cc0d529dea62
--- /dev/null
+++ b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/dataset_fingerprint.json
@@ -0,0 +1,268 @@
+{
+    "foreground_intensity_properties_per_channel": {
+        "0": {
+            "max": 1421.0,
+            "mean": 324.5860900878906,
+            "median": 274.9985656738281,
+            "min": 52.99739456176758,
+            "percentile_00_5": 88.00049591064453,
+            "percentile_99_5": 896.009521484375,
+            "std": 170.6415252685547
+        }
+    },
+    "median_relative_size_after_cropping": 0.983957219251337,
+    "shapes_after_crop": [
+        [
+            189,
+            577,
+            508
+        ],
+        [
+            248,
+            536,
+            480
+        ],
+        [
+            179,
+            566,
+            510
+        ],
+        [
+            200,
+            607,
+            500
+        ],
+        [
+            196,
+            566,
+            470
+        ],
+        [
+            139,
+            507,
+            429
+        ],
+        [
+            200,
+            552,
+            490
+        ],
+        [
+            181,
+            590,
+            485
+        ],
+        [
+            194,
+            542,
+            459
+        ],
+        [
+            153,
+            566,
+            470
+        ],
+        [
+            174,
+            610,
+            469
+        ],
+        [
+            186,
+            582,
+            470
+        ],
+        [
+            174,
+            622,
+            485
+        ],
+        [
+            189,
+            575,
+            485
+        ],
+        [
+            200,
+            541,
+            464
+        ],
+        [
+            189,
+            612,
+            515
+        ],
+        [
+            176,
+            604,
+            490
+        ],
+        [
+            176,
+            610,
+            480
+        ],
+        [
+            174,
+            576,
+            479
+        ],
+        [
+            199,
+            578,
+            495
+        ],
+        [
+            194,
+            577,
+            495
+        ],
+        [
+            149,
+            569,
+            475
+        ],
+        [
+            190,
+            540,
+            450
+        ],
+        [
+            174,
+            549,
+            464
+        ],
+        [
+            181,
+            583,
+            470
+        ]
+    ],
+    "spacings": [
+        [
+            0.5999941825866699,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.6000015735626221,
+            0.296875,
+            0.2968749701976776
+        ],
+        [
+            0.5999999642372131,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.6000019907951355,
+            0.296875,
+            0.2968749701976776
+        ],
+        [
+            0.6000010967254639,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.5000007748603821,
+            0.3515625,
+            0.3515625
+        ],
+        [
+            0.5999962091445923,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.5999972820281982,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.5999994277954102,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.6000003218650818,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.5999979376792908,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.5999984741210938,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.6000016331672668,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.599994421005249,
+            0.296875,
+            0.2968749701976776
+        ],
+        [
+            0.5999991297721863,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.599997878074646,
+            0.2968750298023224,
+            0.296875
+        ],
+        [
+            0.5999959707260132,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.600004255771637,
+            0.296875,
+            0.2968749701976776
+        ],
+        [
+            0.599998414516449,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.5999965071678162,
+            0.296875,
+            0.2968749701976776
+        ],
+        [
+            0.6000003814697266,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.6000002026557922,
+            0.2968749701976776,
+            0.296875
+        ],
+        [
+            0.5999941229820251,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.6000006794929504,
+            0.296875,
+            0.296875
+        ],
+        [
+            0.599998414516449,
+            0.296875,
+            0.296875
+        ]
+    ]
+}
\ No newline at end of file
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/.DS_Store b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/.DS_Store
new file mode 100644
index 0000000000000000000000000000000000000000..dd9c52bdb64be24635a28c9dd3c9af79c6239331
Binary files /dev/null and b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/.DS_Store differ
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_best.pth b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_best.pth
new file mode 100644
index 0000000000000000000000000000000000000000..aaf699fc7975a5a77802ba3192b7ac57e7624f2d
--- /dev/null
+++ b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_best.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2524f8977f6f2189b8836d06f50409f234e292df293ab1310ad94f499159415a
+size 818930499
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_latest.pth b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_latest.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f1647ab4686f24cf17fce29f268a802b71a3ac35
--- /dev/null
+++ b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/checkpoint_latest.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98a74c45f513784cf382a6b6a78f7d28152c49eee5de560101baf04b703b04a7
+size 819382635
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/debug.json
new file mode 100644
index 0000000000000000000000000000000000000000..f5051ac84b31e5aa73b3a8517cc016500e3f051f
--- /dev/null
+++ b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/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': [80, 256, 224], 'median_image_size_in_voxels': [186.0, 577.0, 480.0], 'spacing': [0.5999984741210938, 0.296875, 0.296875], 'normalization_schemes': ['ZScoreNormalization'], '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': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 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': True}",
+    "configuration_name": "3d_fullres",
+    "cudnn_version": 91900,
+    "current_epoch": "0",
+    "dataloader_train": "<batchgenerators.dataloading.nondet_multi_threaded_augmenter.NonDetMultiThreadedAugmenter object at 0x7f7d7daf5550>",
+    "dataloader_train.generator": "<nnunetv2.training.dataloading.data_loader.nnUNetDataLoader object at 0x7f7d7daf5400>",
+    "dataloader_train.generator.transforms": "ComposeTransforms( transforms = [Convert3DTo2DTransform(  ), SpatialTransform( patch_size = [256, 224], patch_center_dist_from_border = [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 = (-3.141592653589793, 3.141592653589793), 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 = {} ), Convert2DTo3DTransform(  ), 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 = (0,), 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(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)]] )] )",
+    "dataloader_train.num_processes": "8",
+    "dataloader_train.transform": "None",
+    "dataloader_val": "<batchgenerators.dataloading.nondet_multi_threaded_augmenter.NonDetMultiThreadedAugmenter object at 0x7f7d7dace710>",
+    "dataloader_val.generator": "<nnunetv2.training.dataloading.data_loader.nnUNetDataLoader object at 0x7f7d7dace5d0>",
+    "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(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)]] )] )",
+    "dataloader_val.num_processes": "4",
+    "dataloader_val.transform": "None",
+    "dataset_json": "{'channel_names': {'0': 'MRA'}, '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 0x7f7d7de39d30>",
+    "hostname": "0ec791b4-7f9c-4cdc-b373-e23e9835edc5-rddl6",
+    "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 0x7f7d7de39be0>",
+    "local_rank": "0",
+    "log_file": "/marimo/TopBrain2026/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_18_08_03_10.txt",
+    "logger": "<nnunetv2.training.logging.nnunet_logger.MetaLogger object at 0x7f7d7de39e80>",
+    "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 0x7f7d7de3b8c0>",
+    "my_init_kwargs": "{'plans': {'dataset_name': 'Dataset001_MRA', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [0.5999984741210938, 0.296875, 0.296875], 'original_median_shape_after_transp': [186, 576, 480], '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': 26, 'patch_size': [640, 512], 'median_image_size_in_voxels': [577.0, 480.0], 'spacing': [0.296875, 0.296875], 'normalization_schemes': ['ZScoreNormalization'], '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': 8, 'features_per_stage': [32, 64, 128, 256, 512, 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], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 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_lowres': {'data_identifier': 'nnUNetResEncUNetLPlans_3d_lowres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [96, 256, 256], 'median_image_size_in_voxels': [147, 442, 368], 'spacing': [0.7600601158948237, 0.387354538949307, 0.387354538949307], 'normalization_schemes': ['ZScoreNormalization'], '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, 320, 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], [3, 3, 3]], 'strides': [[1, 1, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2], [1, 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.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, 'next_stage': '3d_cascade_fullres'}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [80, 256, 224], 'median_image_size_in_voxels': [186.0, 577.0, 480.0], 'spacing': [0.5999984741210938, 0.296875, 0.296875], 'normalization_schemes': ['ZScoreNormalization'], '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': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 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': True}, '3d_cascade_fullres': {'inherits_from': '3d_fullres', 'previous_stage': '3d_lowres'}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 1421.0, 'mean': 324.5860900878906, 'median': 274.9985656738281, 'min': 52.99739456176758, 'percentile_00_5': 88.00049591064453, 'percentile_99_5': 896.009521484375, 'std': 170.6415252685547}}}, 'configuration': '3d_fullres', 'fold': 0, 'dataset_json': {'channel_names': {'0': 'MRA'}, '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_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0",
+    "output_folder_base": "/marimo/TopBrain2026/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres",
+    "oversample_foreground_percent": "0.33",
+    "plans_manager": "{'dataset_name': 'Dataset001_MRA', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [0.5999984741210938, 0.296875, 0.296875], 'original_median_shape_after_transp': [186, 576, 480], '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': 26, 'patch_size': [640, 512], 'median_image_size_in_voxels': [577.0, 480.0], 'spacing': [0.296875, 0.296875], 'normalization_schemes': ['ZScoreNormalization'], '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': 8, 'features_per_stage': [32, 64, 128, 256, 512, 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], [3, 3]], 'strides': [[1, 1], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2], [2, 2]], 'n_blocks_per_stage': [1, 3, 4, 6, 6, 6, 6, 6], 'n_conv_per_stage_decoder': [1, 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_lowres': {'data_identifier': 'nnUNetResEncUNetLPlans_3d_lowres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [96, 256, 256], 'median_image_size_in_voxels': [147, 442, 368], 'spacing': [0.7600601158948237, 0.387354538949307, 0.387354538949307], 'normalization_schemes': ['ZScoreNormalization'], '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, 320, 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], [3, 3, 3]], 'strides': [[1, 1, 1], [2, 2, 2], [2, 2, 2], [2, 2, 2], [2, 2, 2], [1, 2, 2], [1, 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.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, 'next_stage': '3d_cascade_fullres'}, '3d_fullres': {'data_identifier': 'nnUNetPlans_3d_fullres', 'preprocessor_name': 'DefaultPreprocessor', 'batch_size': 2, 'patch_size': [80, 256, 224], 'median_image_size_in_voxels': [186.0, 577.0, 480.0], 'spacing': [0.5999984741210938, 0.296875, 0.296875], 'normalization_schemes': ['ZScoreNormalization'], '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': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 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': True}, '3d_cascade_fullres': {'inherits_from': '3d_fullres', 'previous_stage': '3d_lowres'}}, 'experiment_planner_used': 'nnUNetPlannerResEncL', 'label_manager': 'LabelManager', 'foreground_intensity_properties_per_channel': {'0': {'max': 1421.0, 'mean': 324.5860900878906, 'median': 274.9985656738281, 'min': 52.99739456176758, 'percentile_00_5': 88.00049591064453, 'percentile_99_5': 896.009521484375, 'std': 170.6415252685547}}}",
+    "preprocessed_dataset_folder": "/marimo/TopBrain2026/nnUNet_preprocessed/Dataset001_MRA/nnUNetPlans_3d_fullres",
+    "preprocessed_dataset_folder_base": "/marimo/TopBrain2026/nnUNet_preprocessed/Dataset001_MRA",
+    "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_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png
new file mode 100644
index 0000000000000000000000000000000000000000..58d83b89338f4d7de7f706f83d16facb59e7645f
--- /dev/null
+++ b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/progress.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9c71e5d5613b8c3b9f0f706563ba2cde9e6d8a7e105d942f4eeb1203e404ec6
+size 755864
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_18_08_03_10.txt b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_18_08_03_10.txt
new file mode 100644
index 0000000000000000000000000000000000000000..b5424a931c9cd14be5e130674547a64a6df12b6a
--- /dev/null
+++ b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/nnUNetTrainer__nnUNetResEncUNetLPlans__3d_fullres/fold_0/training_log_2026_10_18_08_03_10.txt
@@ -0,0 +1,2222 @@
+
+#######################################################################
+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.
+#######################################################################
+ 
+2026-10-18 08:03:12.740476: do_dummy_2d_data_aug: True 
+2026-10-18 08:03:12.741107: Creating new 5-fold cross-validation split... 
+2026-10-18 08:03:12.784535: Desired fold for training: 0 
+2026-10-18 08:03:12.784745: 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': [80, 256, 224], 'median_image_size_in_voxels': [186.0, 577.0, 480.0], 'spacing': [0.5999984741210938, 0.296875, 0.296875], 'normalization_schemes': ['ZScoreNormalization'], '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': [[1, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3], [3, 3, 3]], 'strides': [[1, 1, 1], [1, 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': True} 
+ 
+These are the global plan.json settings:
+ {'dataset_name': 'Dataset001_MRA', 'plans_name': 'nnUNetResEncUNetLPlans', 'original_median_spacing_after_transp': [0.5999984741210938, 0.296875, 0.296875], 'original_median_shape_after_transp': [186, 576, 480], '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': 1421.0, 'mean': 324.5860900878906, 'median': 274.9985656738281, 'min': 52.99739456176758, 'percentile_00_5': 88.00049591064453, 'percentile_99_5': 896.009521484375, 'std': 170.6415252685547}}} 
+ 
+2026-10-18 08:03:16.144817: Unable to plot network architecture: 
+2026-10-18 08:03:16.145278: No module named 'hiddenlayer' 
+2026-10-18 08:03:16.189988:  
+2026-10-18 08:03:16.190642: Epoch 0 
+2026-10-18 08:03:16.191339: Current learning rate: 0.01 
+2026-10-18 08:05:16.706444: train_loss 0.2235 
+2026-10-18 08:05:16.707389: val_loss 0.0378 
+2026-10-18 08:05:16.708023: Pseudo dice [np.float32(0.0055), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2475), np.float32(0.0)] 
+2026-10-18 08:05:16.708351: Epoch time: 120.52 s 
+2026-10-18 08:05:16.708686: Yayy! New best EMA pseudo Dice: 0.007000000216066837 
+2026-10-18 08:05:18.426202:  
+2026-10-18 08:05:18.427683: Epoch 1 
+2026-10-18 08:05:18.427966: Current learning rate: 0.00999 
+2026-10-18 08:07:07.485682: train_loss 0.0345 
+2026-10-18 08:07:07.486177: val_loss 0.0275 
+2026-10-18 08:07:07.486668: Pseudo dice [np.float32(0.0435), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2997), np.float32(0.0)] 
+2026-10-18 08:07:07.487502: Epoch time: 109.06 s 
+2026-10-18 08:07:07.488055: Yayy! New best EMA pseudo Dice: 0.007300000172108412 
+2026-10-18 08:07:10.715364:  
+2026-10-18 08:07:10.716054: Epoch 2 
+2026-10-18 08:07:10.716236: Current learning rate: 0.00998 
+2026-10-18 08:08:59.775845: train_loss 0.0237 
+2026-10-18 08:08:59.776572: val_loss 0.0168 
+2026-10-18 08:08:59.776909: Pseudo dice [np.float32(0.2593), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2381), np.float32(0.2838)] 
+2026-10-18 08:08:59.777395: Epoch time: 109.06 s 
+2026-10-18 08:08:59.777607: Yayy! New best EMA pseudo Dice: 0.008700000122189522 
+2026-10-18 08:09:01.933906:  
+2026-10-18 08:09:01.934643: Epoch 3 
+2026-10-18 08:09:01.935002: Current learning rate: 0.00997 
+2026-10-18 08:10:51.005153: train_loss 0.015 
+2026-10-18 08:10:51.005963: val_loss 0.0094 
+2026-10-18 08:10:51.006711: Pseudo dice [np.float32(0.2615), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), 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(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.376), np.float32(0.0)] 
+2026-10-18 08:10:51.007214: Epoch time: 109.07 s 
+2026-10-18 08:10:51.007704: Yayy! New best EMA pseudo Dice: 0.009600000455975533 
+2026-10-18 08:10:52.298186:  
+2026-10-18 08:10:52.298721: Epoch 4 
+2026-10-18 08:10:52.298994: Current learning rate: 0.00996 
+2026-10-18 08:12:41.370266: train_loss 0.0094 
+2026-10-18 08:12:41.370898: val_loss 0.0008 
+2026-10-18 08:12:41.371562: Pseudo dice [np.float32(0.3655), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2942), np.float32(0.3635)] 
+2026-10-18 08:12:41.371938: Epoch time: 109.07 s 
+2026-10-18 08:12:41.372285: Yayy! New best EMA pseudo Dice: 0.011500000022351742 
+2026-10-18 08:12:42.571954:  
+2026-10-18 08:12:42.572616: Epoch 5 
+2026-10-18 08:12:42.572962: Current learning rate: 0.00995 
+2026-10-18 08:14:31.680166: train_loss 0.0025 
+2026-10-18 08:14:31.680644: val_loss -0.0018 
+2026-10-18 08:14:31.681175: Pseudo dice [np.float32(0.3584), np.float32(0.0), 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(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6366)] 
+2026-10-18 08:14:31.682255: Epoch time: 109.11 s 
+2026-10-18 08:14:31.682557: Yayy! New best EMA pseudo Dice: 0.013100000098347664 
+2026-10-18 08:14:33.171835:  
+2026-10-18 08:14:33.172385: Epoch 6 
+2026-10-18 08:14:33.172707: Current learning rate: 0.00995 
+2026-10-18 08:16:22.283517: train_loss -0.0064 
+2026-10-18 08:16:22.284212: val_loss -0.0087 
+2026-10-18 08:16:22.285072: Pseudo dice [np.float32(0.4329), np.float32(0.0292), np.float32(0.0068), np.float32(0.0075), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1665), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4258), np.float32(0.0776)] 
+2026-10-18 08:16:22.285531: Epoch time: 109.11 s 
+2026-10-18 08:16:22.285827: Yayy! New best EMA pseudo Dice: 0.014999999664723873 
+2026-10-18 08:16:23.518965:  
+2026-10-18 08:16:23.519524: Epoch 7 
+2026-10-18 08:16:23.520288: Current learning rate: 0.00994 
+2026-10-18 08:18:12.643409: train_loss -0.0098 
+2026-10-18 08:18:12.643907: val_loss -0.0183 
+2026-10-18 08:18:12.644472: Pseudo dice [np.float32(0.5568), np.float32(0.1155), np.float32(0.0612), np.float32(0.0026), 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(0.0377), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0084), np.float32(0.5963)] 
+2026-10-18 08:18:12.645471: Epoch time: 109.13 s 
+2026-10-18 08:18:12.645756: Yayy! New best EMA pseudo Dice: 0.01730000041425228 
+2026-10-18 08:18:14.265397:  
+2026-10-18 08:18:14.266060: Epoch 8 
+2026-10-18 08:18:14.266426: Current learning rate: 0.00993 
+2026-10-18 08:20:03.438622: train_loss -0.016 
+2026-10-18 08:20:03.440186: val_loss -0.0176 
+2026-10-18 08:20:03.440722: Pseudo dice [np.float32(0.5139), np.float32(0.001), np.float32(0.2196), np.float32(0.1441), 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(0.006), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(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(0.0), np.float32(0.0141), np.float32(0.6052)] 
+2026-10-18 08:20:03.441218: Epoch time: 109.18 s 
+2026-10-18 08:20:03.441540: Yayy! New best EMA pseudo Dice: 0.01979999989271164 
+2026-10-18 08:20:04.761828:  
+2026-10-18 08:20:04.762503: Epoch 9 
+2026-10-18 08:20:04.762896: Current learning rate: 0.00992 
+2026-10-18 08:21:53.906164: train_loss -0.021 
+2026-10-18 08:21:53.906785: val_loss -0.028 
+2026-10-18 08:21:53.907615: Pseudo dice [np.float32(0.5818), np.float32(0.0), np.float32(0.2869), np.float32(0.2713), np.float32(0.0), np.float32(0.003), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0103), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0008), np.float32(0.6211)] 
+2026-10-18 08:21:53.908121: Epoch time: 109.15 s 
+2026-10-18 08:21:53.908602: Yayy! New best EMA pseudo Dice: 0.022700000554323196 
+2026-10-18 08:21:55.206832:  
+2026-10-18 08:21:55.212089: Epoch 10 
+2026-10-18 08:21:55.215296: Current learning rate: 0.00991 
+2026-10-18 08:23:44.355624: train_loss -0.0269 
+2026-10-18 08:23:44.356262: val_loss -0.0313 
+2026-10-18 08:23:44.356745: Pseudo dice [np.float32(0.6499), np.float32(0.0449), np.float32(0.2368), np.float32(0.3251), np.float32(0.0), np.float32(0.0101), np.float32(0.0182), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0032), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(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(0.0147), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4972), np.float32(0.1652)] 
+2026-10-18 08:23:44.357407: Epoch time: 109.15 s 
+2026-10-18 08:23:44.357760: Yayy! New best EMA pseudo Dice: 0.02590000070631504 
+2026-10-18 08:23:45.650115:  
+2026-10-18 08:23:45.654079: Epoch 11 
+2026-10-18 08:23:45.656090: Current learning rate: 0.0099 
+2026-10-18 08:25:34.830459: train_loss -0.0319 
+2026-10-18 08:25:34.831010: val_loss -0.0349 
+2026-10-18 08:25:34.831755: Pseudo dice [np.float32(0.7065), np.float32(0.0054), np.float32(0.2458), np.float32(0.3597), np.float32(0.0), np.float32(0.2184), np.float32(0.1793), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0007), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0008), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4612), np.float32(0.0337)] 
+2026-10-18 08:25:34.832417: Epoch time: 109.18 s 
+2026-10-18 08:25:34.832867: Yayy! New best EMA pseudo Dice: 0.029500000178813934 
+2026-10-18 08:25:36.035686:  
+2026-10-18 08:25:36.036337: Epoch 12 
+2026-10-18 08:25:36.036618: Current learning rate: 0.00989 
+2026-10-18 08:27:25.181589: train_loss -0.0399 
+2026-10-18 08:27:25.182104: val_loss -0.0425 
+2026-10-18 08:27:25.182513: Pseudo dice [np.float32(0.7207), np.float32(0.0717), np.float32(0.3103), np.float32(0.0459), np.float32(0.0), np.float32(0.5224), np.float32(0.1981), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0011), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0127), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4004), np.float32(0.5192)] 
+2026-10-18 08:27:25.183021: Epoch time: 109.15 s 
+2026-10-18 08:27:25.183729: Yayy! New best EMA pseudo Dice: 0.034299999475479126 
+2026-10-18 08:27:28.559644:  
+2026-10-18 08:27:28.560264: Epoch 13 
+2026-10-18 08:27:28.560589: Current learning rate: 0.00988 
+2026-10-18 08:29:17.741308: train_loss -0.0462 
+2026-10-18 08:29:17.742030: val_loss -0.0445 
+2026-10-18 08:29:17.742627: Pseudo dice [np.float32(0.7809), np.float32(0.1384), np.float32(0.3032), np.float32(0.1991), np.float32(0.0), np.float32(0.5738), np.float32(0.1998), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0164), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0079), np.float32(0.0018), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5614), np.float32(0.0547)] 
+2026-10-18 08:29:17.742935: Epoch time: 109.18 s 
+2026-10-18 08:29:17.743101: Yayy! New best EMA pseudo Dice: 0.03880000114440918 
+2026-10-18 08:29:18.960661:  
+2026-10-18 08:29:18.961377: Epoch 14 
+2026-10-18 08:29:18.961705: Current learning rate: 0.00987 
+2026-10-18 08:31:08.142354: train_loss -0.0526 
+2026-10-18 08:31:08.142959: val_loss -0.0494 
+2026-10-18 08:31:08.143496: Pseudo dice [np.float32(0.6872), np.float32(0.1712), np.float32(0.3187), np.float32(0.3504), np.float32(0.0995), np.float32(0.0068), np.float32(0.2394), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0336), np.float32(0.0433), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0706), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0007), np.float32(0.0002), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5176), np.float32(0.0372)] 
+2026-10-18 08:31:08.144121: Epoch time: 109.18 s 
+2026-10-18 08:31:08.144394: Yayy! New best EMA pseudo Dice: 0.041999999433755875 
+2026-10-18 08:31:09.387275:  
+2026-10-18 08:31:09.387906: Epoch 15 
+2026-10-18 08:31:09.388290: Current learning rate: 0.00986 
+2026-10-18 08:32:58.531019: train_loss -0.0606 
+2026-10-18 08:32:58.531552: val_loss -0.0583 
+2026-10-18 08:32:58.532131: Pseudo dice [np.float32(0.7465), np.float32(0.1717), np.float32(0.4626), np.float32(0.001), np.float32(0.006), np.float32(0.5397), np.float32(0.2901), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0118), np.float32(0.0313), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1598), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1202), np.float32(0.0027), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0002), np.float32(0.6583)] 
+2026-10-18 08:32:58.532847: Epoch time: 109.15 s 
+2026-10-18 08:32:58.533081: Yayy! New best EMA pseudo Dice: 0.04670000076293945 
+2026-10-18 08:32:59.778861:  
+2026-10-18 08:32:59.779557: Epoch 16 
+2026-10-18 08:32:59.779895: Current learning rate: 0.00986 
+2026-10-18 08:34:48.983396: train_loss -0.0664 
+2026-10-18 08:34:48.983893: val_loss -0.0616 
+2026-10-18 08:34:48.984792: Pseudo dice [np.float32(0.7429), np.float32(0.1417), np.float32(0.4459), np.float32(0.0305), np.float32(0.1162), np.float32(0.5759), np.float32(0.3179), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0524), np.float32(0.0152), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0017), np.float32(0.0), np.float32(0.1687), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.167), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5175), np.float32(0.0375)] 
+2026-10-18 08:34:48.985484: Epoch time: 109.21 s 
+2026-10-18 08:34:48.985682: Yayy! New best EMA pseudo Dice: 0.05130000039935112 
+2026-10-18 08:34:50.227646:  
+2026-10-18 08:34:50.228043: Epoch 17 
+2026-10-18 08:34:50.228493: Current learning rate: 0.00985 
+2026-10-18 08:36:39.432772: train_loss -0.0734 
+2026-10-18 08:36:39.433554: val_loss -0.068 
+2026-10-18 08:36:39.433968: Pseudo dice [np.float32(0.7584), np.float32(0.0985), np.float32(0.4721), np.float32(0.2707), np.float32(0.2288), np.float32(0.454), np.float32(0.3621), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1702), np.float32(0.0749), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0004), np.float32(0.0), np.float32(0.2451), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1796), np.float32(0.0895), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4605), np.float32(0.0921)] 
+2026-10-18 08:36:39.434815: Epoch time: 109.21 s 
+2026-10-18 08:36:39.435256: Yayy! New best EMA pseudo Dice: 0.05719999969005585 
+2026-10-18 08:36:40.692591:  
+2026-10-18 08:36:40.693148: Epoch 18 
+2026-10-18 08:36:40.693571: Current learning rate: 0.00984 
+2026-10-18 08:38:29.867582: train_loss -0.0827 
+2026-10-18 08:38:29.868238: val_loss -0.075 
+2026-10-18 08:38:29.868789: Pseudo dice [np.float32(0.7924), np.float32(0.1991), np.float32(0.4924), np.float32(0.4733), np.float32(0.1967), np.float32(0.3421), np.float32(0.0586), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2407), np.float32(0.1688), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0965), np.float32(0.0), np.float32(0.253), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.261), np.float32(0.0663), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5373), np.float32(0.0506)] 
+2026-10-18 08:38:29.869430: Epoch time: 109.18 s 
+2026-10-18 08:38:29.869606: Yayy! New best EMA pseudo Dice: 0.06319999694824219 
+2026-10-18 08:38:31.119051:  
+2026-10-18 08:38:31.119684: Epoch 19 
+2026-10-18 08:38:31.119959: Current learning rate: 0.00983 
+2026-10-18 08:40:20.313084: train_loss -0.0863 
+2026-10-18 08:40:20.313718: val_loss -0.0793 
+2026-10-18 08:40:20.314308: Pseudo dice [np.float32(0.7909), np.float32(0.1944), np.float32(0.3653), np.float32(0.5169), np.float32(0.263), np.float32(0.2632), np.float32(0.287), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3269), np.float32(0.0574), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1188), np.float32(0.0), np.float32(0.247), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3393), np.float32(0.0312), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4562), np.float32(0.4736)] 
+2026-10-18 08:40:20.314791: Epoch time: 109.2 s 
+2026-10-18 08:40:20.315115: Yayy! New best EMA pseudo Dice: 0.07000000029802322 
+2026-10-18 08:40:23.831604:  
+2026-10-18 08:40:23.832299: Epoch 20 
+2026-10-18 08:40:23.832441: Current learning rate: 0.00982 
+2026-10-18 08:42:13.036867: train_loss -0.0906 
+2026-10-18 08:42:13.037506: val_loss -0.0822 
+2026-10-18 08:42:13.038220: Pseudo dice [np.float32(0.7598), np.float32(0.2952), np.float32(0.4017), np.float32(0.4024), np.float32(0.3528), np.float32(0.3596), np.float32(0.3277), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3913), np.float32(0.1262), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2011), np.float32(0.0), np.float32(0.0494), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0084), np.float32(0.3295), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3987), np.float32(0.68)] 
+2026-10-18 08:42:13.038818: Epoch time: 109.21 s 
+2026-10-18 08:42:13.039257: Yayy! New best EMA pseudo Dice: 0.0771000012755394 
+2026-10-18 08:42:14.297353:  
+2026-10-18 08:42:14.298124: Epoch 21 
+2026-10-18 08:42:14.298500: Current learning rate: 0.00981 
+2026-10-18 08:44:03.483438: train_loss -0.1069 
+2026-10-18 08:44:03.484142: val_loss -0.0995 
+2026-10-18 08:44:03.484674: Pseudo dice [np.float32(0.8001), np.float32(0.238), np.float32(0.5369), np.float32(0.4655), np.float32(0.3481), np.float32(0.6665), np.float32(0.4371), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.254), np.float32(0.2793), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2181), np.float32(0.0), np.float32(0.2596), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.36), np.float32(0.0179), np.float32(0.0), np.float32(0.006), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4594), np.float32(0.4926)] 
+2026-10-18 08:44:03.485155: Epoch time: 109.19 s 
+2026-10-18 08:44:03.485471: Yayy! New best EMA pseudo Dice: 0.08560000360012054 
+2026-10-18 08:44:04.714446:  
+2026-10-18 08:44:04.714849: Epoch 22 
+2026-10-18 08:44:04.715164: Current learning rate: 0.0098 
+2026-10-18 08:45:53.925480: train_loss -0.1145 
+2026-10-18 08:45:53.926295: val_loss -0.1062 
+2026-10-18 08:45:53.926770: Pseudo dice [np.float32(0.8061), np.float32(0.3548), np.float32(0.3414), np.float32(0.5501), np.float32(0.3862), np.float32(0.6644), np.float32(0.4874), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4076), np.float32(0.2592), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2861), np.float32(0.0), np.float32(0.1277), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0013), np.float32(0.2639), np.float32(0.0), np.float32(0.1568), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4803), np.float32(0.6089)] 
+2026-10-18 08:45:53.927217: Epoch time: 109.21 s 
+2026-10-18 08:45:53.927608: Yayy! New best EMA pseudo Dice: 0.09430000185966492 
+2026-10-18 08:45:55.171663:  
+2026-10-18 08:45:55.172410: Epoch 23 
+2026-10-18 08:45:55.172625: Current learning rate: 0.00979 
+2026-10-18 08:47:44.352115: train_loss -0.1194 
+2026-10-18 08:47:44.352835: val_loss -0.1179 
+2026-10-18 08:47:44.353554: Pseudo dice [np.float32(0.8147), np.float32(0.2974), np.float32(0.4999), np.float32(0.5662), np.float32(0.5085), np.float32(0.6504), np.float32(0.4516), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3784), np.float32(0.3613), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3512), np.float32(0.0), np.float32(0.0793), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0413), np.float32(0.4719), np.float32(0.0023), np.float32(0.1709), 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(0.507), np.float32(0.5061)] 
+2026-10-18 08:47:44.353917: Epoch time: 109.18 s 
+2026-10-18 08:47:44.354110: Yayy! New best EMA pseudo Dice: 0.10329999774694443 
+2026-10-18 08:47:46.078522:  
+2026-10-18 08:47:46.079252: Epoch 24 
+2026-10-18 08:47:46.079550: Current learning rate: 0.00978 
+2026-10-18 08:49:35.283485: train_loss -0.1363 
+2026-10-18 08:49:35.284021: val_loss -0.107 
+2026-10-18 08:49:35.284489: Pseudo dice [np.float32(0.8072), np.float32(0.1557), np.float32(0.3851), np.float32(0.5119), np.float32(0.4609), np.float32(0.6119), np.float32(0.4139), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4243), np.float32(0.5023), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3588), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3741), np.float32(0.0052), np.float32(0.008), np.float32(0.2138), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.48), np.float32(0.4991)] 
+2026-10-18 08:49:35.285507: Epoch time: 109.21 s 
+2026-10-18 08:49:35.285840: Yayy! New best EMA pseudo Dice: 0.11020000278949738 
+2026-10-18 08:49:38.898418:  
+2026-10-18 08:49:38.899105: Epoch 25 
+2026-10-18 08:49:38.899401: Current learning rate: 0.00977 
+2026-10-18 08:51:28.107710: train_loss -0.1543 
+2026-10-18 08:51:28.108286: val_loss -0.1465 
+2026-10-18 08:51:28.109051: Pseudo dice [np.float32(0.8569), np.float32(0.4863), np.float32(0.5221), np.float32(0.5857), np.float32(0.5638), np.float32(0.5738), np.float32(0.5827), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4705), np.float32(0.4424), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3814), np.float32(0.0), np.float32(0.0548), np.float32(0.0), np.float32(0.0), np.float32(1e-04), np.float32(0.4019), np.float32(0.0184), np.float32(0.3244), np.float32(0.2688), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6173), np.float32(0.6622)] 
+2026-10-18 08:51:28.109572: Epoch time: 109.21 s 
+2026-10-18 08:51:28.109744: Yayy! New best EMA pseudo Dice: 0.120899997651577 
+2026-10-18 08:51:31.414575:  
+2026-10-18 08:51:31.415242: Epoch 26 
+2026-10-18 08:51:31.415558: Current learning rate: 0.00977 
+2026-10-18 08:53:20.604390: train_loss -0.1658 
+2026-10-18 08:53:20.604841: val_loss -0.1472 
+2026-10-18 08:53:20.605924: Pseudo dice [np.float32(0.8672), np.float32(0.4124), np.float32(0.6316), np.float32(0.628), np.float32(0.5725), np.float32(0.716), np.float32(0.6127), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.416), np.float32(0.5238), np.float32(0.0003), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2073), np.float32(0.0), np.float32(0.325), np.float32(0.0), np.float32(0.0), np.float32(0.0008), np.float32(0.3611), np.float32(0.1565), np.float32(0.3143), np.float32(0.4488), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5108), np.float32(0.484)] 
+2026-10-18 08:53:20.606380: Epoch time: 109.19 s 
+2026-10-18 08:53:20.606675: Yayy! New best EMA pseudo Dice: 0.1316000074148178 
+2026-10-18 08:53:21.821157:  
+2026-10-18 08:53:21.821847: Epoch 27 
+2026-10-18 08:53:21.822183: Current learning rate: 0.00976 
+2026-10-18 08:55:11.011388: train_loss -0.179 
+2026-10-18 08:55:11.011891: val_loss -0.1656 
+2026-10-18 08:55:11.012939: Pseudo dice [np.float32(0.8142), np.float32(0.4681), np.float32(0.6632), np.float32(0.6675), np.float32(0.5112), np.float32(0.7364), np.float32(0.5625), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5309), np.float32(0.5479), np.float32(0.0432), np.float32(0.0022), np.float32(0.0), np.float32(0.0), np.float32(0.4123), np.float32(0.0), np.float32(0.1384), np.float32(0.0), np.float32(0.0), np.float32(0.0505), np.float32(0.0468), np.float32(0.4491), np.float32(0.4028), np.float32(0.4802), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0005), np.float32(0.0), np.float32(0.0), np.float32(0.6649), np.float32(0.5768)] 
+2026-10-18 08:55:11.013330: Epoch time: 109.19 s 
+2026-10-18 08:55:11.013803: Yayy! New best EMA pseudo Dice: 0.1428000032901764 
+2026-10-18 08:55:12.264536:  
+2026-10-18 08:55:12.265221: Epoch 28 
+2026-10-18 08:55:12.265617: Current learning rate: 0.00975 
+2026-10-18 08:57:01.446785: train_loss -0.1902 
+2026-10-18 08:57:01.447578: val_loss -0.1754 
+2026-10-18 08:57:01.448048: Pseudo dice [np.float32(0.852), np.float32(0.529), np.float32(0.6955), np.float32(0.6926), np.float32(0.3987), np.float32(0.7403), np.float32(0.5142), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4514), np.float32(0.5651), np.float32(0.0008), np.float32(0.0715), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4452), np.float32(0.0), np.float32(0.0), np.float32(0.0013), np.float32(0.0226), np.float32(0.4806), np.float32(0.3626), np.float32(0.4423), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1005), np.float32(0.0), np.float32(0.0002), np.float32(0.0), np.float32(0.0), np.float32(0.6998), np.float32(0.697)] 
+2026-10-18 08:57:01.448462: Epoch time: 109.18 s 
+2026-10-18 08:57:01.448828: Yayy! New best EMA pseudo Dice: 0.15279999375343323 
+2026-10-18 08:57:02.910400:  
+2026-10-18 08:57:02.911054: Epoch 29 
+2026-10-18 08:57:02.911329: Current learning rate: 0.00974 
+2026-10-18 08:58:52.123456: train_loss -0.2004 
+2026-10-18 08:58:52.124028: val_loss -0.1688 
+2026-10-18 08:58:52.124522: Pseudo dice [np.float32(0.8684), np.float32(0.4752), np.float32(0.521), np.float32(0.6411), np.float32(0.4428), np.float32(0.6922), np.float32(0.4261), np.float32(0.0), np.float32(0.0016), np.float32(0.0), np.float32(0.5236), np.float32(0.2957), np.float32(0.0187), np.float32(0.1496), np.float32(0.0), np.float32(0.0), np.float32(0.4768), np.float32(0.0003), np.float32(0.0017), np.float32(0.0), np.float32(0.0008), np.float32(0.0243), np.float32(0.0463), np.float32(0.492), np.float32(0.4858), np.float32(0.5267), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.1734), np.float32(0.0), np.float32(0.0134), np.float32(0.0), np.float32(0.0), np.float32(0.736), np.float32(0.7401)] 
+2026-10-18 08:58:52.125332: Epoch time: 109.22 s 
+2026-10-18 08:58:52.125616: Yayy! New best EMA pseudo Dice: 0.16189999878406525 
+2026-10-18 08:58:53.338327:  
+2026-10-18 08:58:53.339019: Epoch 30 
+2026-10-18 08:58:53.339210: Current learning rate: 0.00973 
+2026-10-18 09:00:42.515765: train_loss -0.2285 
+2026-10-18 09:00:42.516246: val_loss -0.1411 
+2026-10-18 09:00:42.516968: Pseudo dice [np.float32(0.8689), np.float32(0.471), np.float32(0.5318), np.float32(0.5314), np.float32(0.4137), np.float32(0.3964), np.float32(0.2461), np.float32(0.006), np.float32(0.0), np.float32(0.0), np.float32(0.4257), np.float32(0.3919), np.float32(0.0282), np.float32(0.3124), np.float32(0.0), np.float32(0.0), np.float32(0.2908), np.float32(0.0095), np.float32(0.3912), np.float32(0.001), np.float32(0.0026), np.float32(0.1934), np.float32(0.2207), np.float32(0.3989), np.float32(0.3465), np.float32(0.3065), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2286), np.float32(0.0), np.float32(0.0103), np.float32(0.0), np.float32(0.0889), np.float32(0.5241), np.float32(0.3282)] 
+2026-10-18 09:00:42.517856: Epoch time: 109.18 s 
+2026-10-18 09:00:42.518136: Yayy! New best EMA pseudo Dice: 0.1678999960422516 
+2026-10-18 09:00:43.735231:  
+2026-10-18 09:00:43.735953: Epoch 31 
+2026-10-18 09:00:43.736282: Current learning rate: 0.00972 
+2026-10-18 09:02:32.921381: train_loss -0.2251 
+2026-10-18 09:02:32.921939: val_loss -0.2124 
+2026-10-18 09:02:32.922606: Pseudo dice [np.float32(0.8777), np.float32(0.5029), np.float32(0.7048), np.float32(0.6936), np.float32(0.3806), np.float32(0.8712), np.float32(0.5481), np.float32(0.0017), np.float32(0.0072), np.float32(1e-04), np.float32(0.6478), np.float32(0.6283), np.float32(0.3499), np.float32(0.1969), np.float32(0.0), np.float32(0.0), np.float32(0.0337), np.float32(0.0), np.float32(0.4854), np.float32(0.0836), np.float32(0.0), np.float32(0.2225), np.float32(0.4225), np.float32(0.4214), np.float32(0.5342), np.float32(0.5065), np.float32(0.0), np.float32(0.4068), np.float32(0.0), np.float32(0.2876), np.float32(0.0), np.float32(0.0663), np.float32(0.515), np.float32(0.4969), np.float32(0.7281), np.float32(0.7826)] 
+2026-10-18 09:02:32.923473: Epoch time: 109.19 s 
+2026-10-18 09:02:32.923801: Yayy! New best EMA pseudo Dice: 0.18549999594688416 
+2026-10-18 09:02:36.152371:  
+2026-10-18 09:02:36.152846: Epoch 32 
+2026-10-18 09:02:36.153525: Current learning rate: 0.00971 
+2026-10-18 09:04:25.361412: train_loss -0.2497 
+2026-10-18 09:04:25.362223: val_loss -0.2346 
+2026-10-18 09:04:25.362818: Pseudo dice [np.float32(0.7997), np.float32(0.6127), np.float32(0.744), np.float32(0.7961), np.float32(0.6097), np.float32(0.8751), np.float32(0.7398), np.float32(0.1249), np.float32(0.0035), np.float32(0.0), np.float32(0.652), np.float32(0.6797), np.float32(0.2273), np.float32(0.3433), np.float32(0.0), np.float32(0.0), np.float32(0.4343), np.float32(0.0018), np.float32(0.113), np.float32(0.1715), np.float32(0.0025), np.float32(0.2391), np.float32(0.09), np.float32(0.5563), np.float32(0.5495), np.float32(0.5261), np.float32(0.0), np.float32(0.4775), np.float32(0.0), np.float32(0.3953), np.float32(0.0), np.float32(0.2325), np.float32(0.5772), np.float32(0.5692), np.float32(0.6729), np.float32(0.6745)] 
+2026-10-18 09:04:25.363422: Epoch time: 109.21 s 
+2026-10-18 09:04:25.363848: Yayy! New best EMA pseudo Dice: 0.2045000046491623 
+2026-10-18 09:04:26.583450:  
+2026-10-18 09:04:26.584168: Epoch 33 
+2026-10-18 09:04:26.584437: Current learning rate: 0.0097 
+2026-10-18 09:06:15.745859: train_loss -0.2682 
+2026-10-18 09:06:15.746513: val_loss -0.2521 
+2026-10-18 09:06:15.747164: Pseudo dice [np.float32(0.9045), np.float32(0.5947), np.float32(0.7221), np.float32(0.6833), np.float32(0.5583), np.float32(0.8472), np.float32(0.6757), np.float32(0.2504), np.float32(0.0043), np.float32(0.0), np.float32(0.608), np.float32(0.6776), np.float32(0.204), np.float32(0.3422), np.float32(0.0), np.float32(0.0), np.float32(0.5219), np.float32(0.1123), np.float32(0.3145), np.float32(0.1205), np.float32(0.1666), np.float32(0.2134), np.float32(0.0757), np.float32(0.5096), np.float32(0.5552), np.float32(0.6096), np.float32(0.1088), np.float32(0.0), np.float32(0.0627), np.float32(0.3259), np.float32(0.0), np.float32(0.2843), np.float32(0.5691), np.float32(0.5476), np.float32(0.8201), np.float32(0.8505)] 
+2026-10-18 09:06:15.747882: Epoch time: 109.16 s 
+2026-10-18 09:06:15.748205: Yayy! New best EMA pseudo Dice: 0.2224999964237213 
+2026-10-18 09:06:16.979305:  
+2026-10-18 09:06:16.979928: Epoch 34 
+2026-10-18 09:06:16.980226: Current learning rate: 0.00969 
+2026-10-18 09:08:06.119884: train_loss -0.2748 
+2026-10-18 09:08:06.120676: val_loss -0.2551 
+2026-10-18 09:08:06.121205: Pseudo dice [np.float32(0.9077), np.float32(0.6335), np.float32(0.7348), np.float32(0.6946), np.float32(0.5227), np.float32(0.7951), np.float32(0.6195), np.float32(0.3869), np.float32(0.0019), np.float32(0.0002), np.float32(0.6146), np.float32(0.5961), np.float32(0.3218), np.float32(0.3357), np.float32(0.0), np.float32(0.0), np.float32(0.3625), np.float32(0.0008), np.float32(0.0842), np.float32(0.2609), np.float32(0.1687), np.float32(0.258), np.float32(0.3304), np.float32(0.5084), np.float32(0.6705), np.float32(0.597), np.float32(0.0782), np.float32(0.2396), np.float32(0.2088), np.float32(0.5288), np.float32(0.0), np.float32(0.1899), np.float32(0.5162), np.float32(0.5742), np.float32(0.7029), np.float32(0.6879)] 
+2026-10-18 09:08:06.121669: Epoch time: 109.14 s 
+2026-10-18 09:08:06.121968: Yayy! New best EMA pseudo Dice: 0.2395000010728836 
+2026-10-18 09:08:07.388448:  
+2026-10-18 09:08:07.388977: Epoch 35 
+2026-10-18 09:08:07.389132: Current learning rate: 0.00968 
+2026-10-18 09:09:56.540282: train_loss -0.2977 
+2026-10-18 09:09:56.540888: val_loss -0.2244 
+2026-10-18 09:09:56.541735: Pseudo dice [np.float32(0.8748), np.float32(0.542), np.float32(0.6267), np.float32(0.6493), np.float32(0.3914), np.float32(0.7508), np.float32(0.516), np.float32(0.1443), np.float32(0.0015), np.float32(0.0008), np.float32(0.5638), np.float32(0.4887), np.float32(0.327), np.float32(0.2282), np.float32(0.0), np.float32(0.0), np.float32(0.3547), np.float32(0.28), np.float32(0.4774), np.float32(0.0761), np.float32(0.0772), np.float32(0.2048), np.float32(0.4882), np.float32(0.436), np.float32(0.6055), np.float32(0.4988), np.float32(0.1039), np.float32(0.4841), np.float32(0.1155), np.float32(0.3654), np.float32(0.0), np.float32(0.013), np.float32(0.4934), np.float32(0.5159), np.float32(0.6516), np.float32(0.6346)] 
+2026-10-18 09:09:56.542055: Epoch time: 109.15 s 
+2026-10-18 09:09:56.542369: Yayy! New best EMA pseudo Dice: 0.2515999972820282 
+2026-10-18 09:09:57.804991:  
+2026-10-18 09:09:57.805613: Epoch 36 
+2026-10-18 09:09:57.805930: Current learning rate: 0.00968 
+2026-10-18 09:11:46.943698: train_loss -0.2946 
+2026-10-18 09:11:46.944296: val_loss -0.2733 
+2026-10-18 09:11:46.945280: Pseudo dice [np.float32(0.8382), np.float32(0.6463), np.float32(0.7281), np.float32(0.8129), np.float32(0.5926), np.float32(0.8914), np.float32(0.7663), np.float32(0.5645), np.float32(0.0037), np.float32(0.0003), np.float32(0.7022), np.float32(0.716), np.float32(0.0583), np.float32(0.3145), np.float32(0.0), np.float32(0.0), np.float32(0.4236), np.float32(0.2836), np.float32(0.4018), np.float32(0.0), np.float32(0.1043), np.float32(0.3225), np.float32(0.3268), np.float32(0.447), np.float32(0.6636), np.float32(0.5881), np.float32(0.1857), np.float32(0.3274), np.float32(0.3772), np.float32(0.4786), np.float32(0.0), np.float32(0.3473), np.float32(0.6636), np.float32(0.6093), np.float32(0.7026), np.float32(0.7536)] 
+2026-10-18 09:11:46.945927: Epoch time: 109.14 s 
+2026-10-18 09:11:46.946278: Yayy! New best EMA pseudo Dice: 0.26989999413490295 
+2026-10-18 09:11:48.200928:  
+2026-10-18 09:11:48.201535: Epoch 37 
+2026-10-18 09:11:48.201879: Current learning rate: 0.00967 
+2026-10-18 09:13:37.315098: train_loss -0.2918 
+2026-10-18 09:13:37.315835: val_loss -0.242 
+2026-10-18 09:13:37.316592: Pseudo dice [np.float32(0.8727), np.float32(0.6347), np.float32(0.7059), np.float32(0.6847), np.float32(0.5623), np.float32(0.7469), np.float32(0.3588), np.float32(0.3243), np.float32(0.0067), np.float32(0.0), np.float32(0.6146), np.float32(0.654), np.float32(0.277), np.float32(0.0156), np.float32(0.0), np.float32(0.0), np.float32(0.5413), np.float32(0.2741), np.float32(0.0765), np.float32(0.004), np.float32(0.1851), np.float32(0.3407), np.float32(0.2933), np.float32(0.4922), np.float32(0.6398), np.float32(0.5111), np.float32(0.2056), np.float32(0.364), np.float32(0.3075), np.float32(0.5012), np.float32(0.0), np.float32(0.1203), np.float32(0.6337), np.float32(0.5157), np.float32(0.5977), np.float32(0.6508)] 
+2026-10-18 09:13:37.317002: Epoch time: 109.12 s 
+2026-10-18 09:13:37.317257: Yayy! New best EMA pseudo Dice: 0.2809999883174896 
+2026-10-18 09:13:38.636137:  
+2026-10-18 09:13:38.637116: Epoch 38 
+2026-10-18 09:13:38.637472: Current learning rate: 0.00966 
+2026-10-18 09:15:27.785958: train_loss -0.3163 
+2026-10-18 09:15:27.786540: val_loss -0.3007 
+2026-10-18 09:15:27.787037: Pseudo dice [np.float32(0.873), np.float32(0.6933), np.float32(0.7895), np.float32(0.7638), np.float32(0.5815), np.float32(0.8642), np.float32(0.7654), np.float32(0.4396), np.float32(0.0), np.float32(0.0684), np.float32(0.6408), np.float32(0.6558), np.float32(0.3246), np.float32(0.368), np.float32(0.0), np.float32(0.0), np.float32(0.3283), np.float32(0.1614), np.float32(0.5413), np.float32(0.1991), np.float32(0.1654), np.float32(0.3702), np.float32(0.232), np.float32(0.5355), np.float32(0.668), np.float32(0.6269), np.float32(0.1856), np.float32(0.6419), np.float32(0.2557), np.float32(0.5777), np.float32(0.0), np.float32(0.2194), np.float32(0.707), np.float32(0.6011), np.float32(0.8023), np.float32(0.8291)] 
+2026-10-18 09:15:27.787718: Epoch time: 109.15 s 
+2026-10-18 09:15:27.787948: Yayy! New best EMA pseudo Dice: 0.2985999882221222 
+2026-10-18 09:15:29.130767:  
+2026-10-18 09:15:29.131530: Epoch 39 
+2026-10-18 09:15:29.131886: Current learning rate: 0.00965 
+2026-10-18 09:17:18.317898: train_loss -0.3488 
+2026-10-18 09:17:18.318668: val_loss -0.2854 
+2026-10-18 09:17:18.319211: Pseudo dice [np.float32(0.8936), np.float32(0.6631), np.float32(0.6274), np.float32(0.8345), np.float32(0.5609), np.float32(0.8888), np.float32(0.7445), np.float32(0.4805), np.float32(0.0026), np.float32(0.1596), np.float32(0.6513), np.float32(0.6172), np.float32(0.4546), np.float32(0.2188), np.float32(0.0), np.float32(0.0), np.float32(0.5513), np.float32(0.245), np.float32(0.2043), np.float32(0.1416), np.float32(0.3187), np.float32(0.4254), np.float32(0.3814), np.float32(0.519), np.float32(0.6393), np.float32(0.5623), np.float32(0.1434), np.float32(0.5058), np.float32(0.3689), np.float32(0.5482), np.float32(0.0), np.float32(0.2619), np.float32(0.6988), np.float32(0.6664), np.float32(0.822), np.float32(0.8468)] 
+2026-10-18 09:17:18.319731: Epoch time: 109.19 s 
+2026-10-18 09:17:18.320061: Yayy! New best EMA pseudo Dice: 0.3149999976158142 
+2026-10-18 09:17:19.664848:  
+2026-10-18 09:17:19.665513: Epoch 40 
+2026-10-18 09:17:19.665777: Current learning rate: 0.00964 
+2026-10-18 09:19:08.818648: train_loss -0.3394 
+2026-10-18 09:19:08.819147: val_loss -0.2774 
+2026-10-18 09:19:08.819722: Pseudo dice [np.float32(0.8684), np.float32(0.6694), np.float32(0.7424), np.float32(0.7742), np.float32(0.5514), np.float32(0.8924), np.float32(0.7592), np.float32(0.5461), np.float32(0.0012), np.float32(0.3593), np.float32(0.6754), np.float32(0.6841), np.float32(0.1745), np.float32(0.1625), np.float32(0.0), np.float32(0.0), np.float32(0.475), np.float32(0.2979), np.float32(0.2018), np.float32(0.0998), np.float32(0.2395), np.float32(0.4263), np.float32(0.3806), np.float32(0.5117), np.float32(0.7019), np.float32(0.6268), np.float32(0.2555), np.float32(0.5623), np.float32(0.4245), np.float32(0.3529), np.float32(0.0018), np.float32(0.268), np.float32(0.6957), np.float32(0.6534), np.float32(0.8203), np.float32(0.855)] 
+2026-10-18 09:19:08.820656: Epoch time: 109.16 s 
+2026-10-18 09:19:08.820843: Yayy! New best EMA pseudo Dice: 0.32989999651908875 
+2026-10-18 09:19:11.051289:  
+2026-10-18 09:19:11.051858: Epoch 41 
+2026-10-18 09:19:11.052331: Current learning rate: 0.00963 
+2026-10-18 09:21:00.216801: train_loss -0.3297 
+2026-10-18 09:21:00.217512: val_loss -0.257 
+2026-10-18 09:21:00.218058: Pseudo dice [np.float32(0.8205), np.float32(0.5257), np.float32(0.7236), np.float32(0.7559), np.float32(0.6909), np.float32(0.847), np.float32(0.7676), np.float32(0.4272), np.float32(0.0), np.float32(0.3291), np.float32(0.5232), np.float32(0.6923), np.float32(0.3929), np.float32(0.1191), np.float32(0.0), np.float32(0.0), np.float32(0.3786), np.float32(0.0699), np.float32(0.4243), np.float32(0.2458), np.float32(0.223), np.float32(0.2283), np.float32(0.307), np.float32(0.3135), np.float32(0.6241), np.float32(0.5919), np.float32(0.1692), np.float32(0.4147), np.float32(0.3241), np.float32(0.5134), np.float32(0.0072), np.float32(0.3975), np.float32(0.5572), np.float32(0.5613), np.float32(0.7515), np.float32(0.7947)] 
+2026-10-18 09:21:00.218665: Epoch time: 109.17 s 
+2026-10-18 09:21:00.218993: Yayy! New best EMA pseudo Dice: 0.3400000035762787 
+2026-10-18 09:21:01.434217:  
+2026-10-18 09:21:01.434855: Epoch 42 
+2026-10-18 09:21:01.435106: Current learning rate: 0.00962 
+2026-10-18 09:22:50.574046: train_loss -0.3345 
+2026-10-18 09:22:50.574615: val_loss -0.3017 
+2026-10-18 09:22:50.575334: Pseudo dice [np.float32(0.9347), np.float32(0.6038), np.float32(0.7325), np.float32(0.7308), np.float32(0.5217), np.float32(0.8165), np.float32(0.7096), np.float32(0.4008), np.float32(0.0), np.float32(0.3595), np.float32(0.6024), np.float32(0.6675), np.float32(0.4666), np.float32(0.0615), np.float32(0.0), np.float32(0.0), np.float32(0.3892), np.float32(0.0908), np.float32(0.3588), np.float32(0.2908), np.float32(0.4632), np.float32(0.5371), np.float32(0.4621), np.float32(0.5985), np.float32(0.6002), np.float32(0.5412), np.float32(0.2814), np.float32(0.4791), np.float32(0.549), np.float32(0.606), np.float32(0.2424), np.float32(0.4333), np.float32(0.57), np.float32(0.6129), np.float32(0.7348), np.float32(0.7622)] 
+2026-10-18 09:22:50.575611: Epoch time: 109.14 s 
+2026-10-18 09:22:50.575933: Yayy! New best EMA pseudo Dice: 0.3537999987602234 
+2026-10-18 09:22:51.795281:  
+2026-10-18 09:22:51.796128: Epoch 43 
+2026-10-18 09:22:51.796417: Current learning rate: 0.00961 
+2026-10-18 09:24:40.950742: train_loss -0.3594 
+2026-10-18 09:24:40.951429: val_loss -0.2848 
+2026-10-18 09:24:40.951763: Pseudo dice [np.float32(0.8668), np.float32(0.6416), np.float32(0.7295), np.float32(0.6632), np.float32(0.6216), np.float32(0.7133), np.float32(0.6665), np.float32(0.4755), np.float32(0.0015), np.float32(0.4446), np.float32(0.5854), np.float32(0.5511), np.float32(0.526), np.float32(0.2132), np.float32(0.0), np.float32(0.0), np.float32(0.504), np.float32(0.3562), np.float32(0.4435), np.float32(0.2561), np.float32(0.4159), np.float32(0.5849), np.float32(0.5348), np.float32(0.3437), np.float32(0.5827), np.float32(0.5769), np.float32(0.1504), np.float32(0.3966), np.float32(0.4372), np.float32(0.5946), np.float32(0.2408), np.float32(0.2908), np.float32(0.5604), np.float32(0.4381), np.float32(0.7047), np.float32(0.6612)] 
+2026-10-18 09:24:40.952187: Epoch time: 109.16 s 
+2026-10-18 09:24:40.952880: Yayy! New best EMA pseudo Dice: 0.36500000953674316 
+2026-10-18 09:24:42.261529:  
+2026-10-18 09:24:42.262113: Epoch 44 
+2026-10-18 09:24:42.262489: Current learning rate: 0.0096 
+2026-10-18 09:26:31.397297: train_loss -0.3671 
+2026-10-18 09:26:31.397962: val_loss -0.2289 
+2026-10-18 09:26:31.398716: Pseudo dice [np.float32(0.8988), np.float32(0.5569), np.float32(0.5793), np.float32(0.5386), np.float32(0.5744), np.float32(0.7071), np.float32(0.5594), np.float32(0.3467), np.float32(0.0009), np.float32(0.5362), np.float32(0.366), np.float32(0.4702), np.float32(0.446), np.float32(0.0048), np.float32(0.0), np.float32(0.0), np.float32(0.5657), np.float32(0.3096), np.float32(0.2548), np.float32(0.0745), np.float32(0.2939), np.float32(0.5471), np.float32(0.4527), np.float32(0.5124), np.float32(0.5181), np.float32(0.4801), np.float32(0.226), np.float32(0.4848), np.float32(0.4004), np.float32(0.462), np.float32(0.3998), np.float32(0.2624), np.float32(0.4595), np.float32(0.5158), np.float32(0.6154), np.float32(0.6681)] 
+2026-10-18 09:26:31.399280: Epoch time: 109.14 s 
+2026-10-18 09:26:31.399459: Yayy! New best EMA pseudo Dice: 0.37049999833106995 
+2026-10-18 09:26:32.629673:  
+2026-10-18 09:26:32.630418: Epoch 45 
+2026-10-18 09:26:32.630775: Current learning rate: 0.00959 
+2026-10-18 09:28:21.790668: train_loss -0.3765 
+2026-10-18 09:28:21.791488: val_loss -0.3064 
+2026-10-18 09:28:21.791833: Pseudo dice [np.float32(0.8952), np.float32(0.608), np.float32(0.7265), np.float32(0.7594), np.float32(0.588), np.float32(0.8231), np.float32(0.6639), np.float32(0.488), np.float32(0.0032), np.float32(0.4242), np.float32(0.6749), np.float32(0.6136), np.float32(0.1849), np.float32(0.1643), np.float32(0.0), np.float32(0.0), np.float32(0.5105), np.float32(0.2782), np.float32(0.457), np.float32(0.1968), np.float32(0.4201), np.float32(0.5902), np.float32(0.3672), np.float32(0.5818), np.float32(0.6505), np.float32(0.6132), np.float32(0.2341), np.float32(0.662), np.float32(0.4888), np.float32(0.6004), np.float32(0.4167), np.float32(0.2349), np.float32(0.6711), np.float32(0.492), np.float32(0.818), np.float32(0.8396)] 
+2026-10-18 09:28:21.792374: Epoch time: 109.16 s 
+2026-10-18 09:28:21.792780: Yayy! New best EMA pseudo Dice: 0.38269999623298645 
+2026-10-18 09:28:23.242270:  
+2026-10-18 09:28:23.243189: Epoch 46 
+2026-10-18 09:28:23.243499: Current learning rate: 0.00959 
+2026-10-18 09:30:12.398107: train_loss -0.3893 
+2026-10-18 09:30:12.398856: val_loss -0.3389 
+2026-10-18 09:30:12.399373: Pseudo dice [np.float32(0.8965), np.float32(0.6595), np.float32(0.7732), np.float32(0.7426), np.float32(0.602), np.float32(0.7933), np.float32(0.6628), np.float32(0.5019), np.float32(0.0003), np.float32(0.3837), np.float32(0.6659), np.float32(0.6765), np.float32(0.4586), np.float32(0.2298), np.float32(0.0), np.float32(0.0), np.float32(0.6436), np.float32(0.2778), np.float32(0.5531), np.float32(0.2201), np.float32(0.3263), np.float32(0.4764), np.float32(0.4324), np.float32(0.6576), np.float32(0.6425), np.float32(0.6723), np.float32(0.322), np.float32(0.7352), np.float32(0.4815), np.float32(0.5902), np.float32(0.4831), np.float32(0.3121), np.float32(0.7244), np.float32(0.6573), np.float32(0.8075), np.float32(0.8126)] 
+2026-10-18 09:30:12.400015: Epoch time: 109.16 s 
+2026-10-18 09:30:12.400223: Yayy! New best EMA pseudo Dice: 0.3968999981880188 
+2026-10-18 09:30:13.629965:  
+2026-10-18 09:30:13.630746: Epoch 47 
+2026-10-18 09:30:13.631145: Current learning rate: 0.00958 
+2026-10-18 09:32:02.726107: train_loss -0.3973 
+2026-10-18 09:32:02.726968: val_loss -0.3593 
+2026-10-18 09:32:02.727490: Pseudo dice [np.float32(0.9018), np.float32(0.6546), np.float32(0.6556), np.float32(0.767), np.float32(0.5661), np.float32(0.9066), np.float32(0.8164), np.float32(0.6704), np.float32(0.0021), np.float32(0.5146), np.float32(0.7024), np.float32(0.7278), np.float32(0.5224), np.float32(0.2654), np.float32(0.0), np.float32(0.0), np.float32(0.5316), np.float32(0.3173), np.float32(0.6217), np.float32(0.3087), np.float32(0.2631), np.float32(0.533), np.float32(0.517), np.float32(0.7204), np.float32(0.6834), np.float32(0.649), np.float32(0.2473), np.float32(0.6537), np.float32(0.6699), np.float32(0.6965), np.float32(0.3811), np.float32(0.395), np.float32(0.6929), np.float32(0.6197), np.float32(0.8941), np.float32(0.9322)] 
+2026-10-18 09:32:02.727872: Epoch time: 109.1 s 
+2026-10-18 09:32:02.728215: Yayy! New best EMA pseudo Dice: 0.41269999742507935 
+2026-10-18 09:32:03.988443:  
+2026-10-18 09:32:03.989132: Epoch 48 
+2026-10-18 09:32:03.989458: Current learning rate: 0.00957 
+2026-10-18 09:33:53.096440: train_loss -0.4112 
+2026-10-18 09:33:53.096975: val_loss -0.3593 
+2026-10-18 09:33:53.097646: Pseudo dice [np.float32(0.9038), np.float32(0.6771), np.float32(0.8155), np.float32(0.8527), np.float32(0.7047), np.float32(0.8968), np.float32(0.8365), np.float32(0.7104), np.float32(0.0012), np.float32(0.537), np.float32(0.6698), np.float32(0.6767), np.float32(0.1692), np.float32(0.3301), np.float32(0.0), np.float32(0.0), np.float32(0.7119), np.float32(0.31), np.float32(0.7396), np.float32(0.4501), np.float32(0.3177), np.float32(0.5979), np.float32(0.4289), np.float32(0.6047), np.float32(0.7114), np.float32(0.6701), np.float32(0.3395), np.float32(0.5722), np.float32(0.41), np.float32(0.5762), np.float32(0.5538), np.float32(0.3819), np.float32(0.7798), np.float32(0.5916), np.float32(0.8828), np.float32(0.878)] 
+2026-10-18 09:33:53.098401: Epoch time: 109.11 s 
+2026-10-18 09:33:53.098748: Yayy! New best EMA pseudo Dice: 0.4277999997138977 
+2026-10-18 09:33:54.321779:  
+2026-10-18 09:33:54.322421: Epoch 49 
+2026-10-18 09:33:54.322682: Current learning rate: 0.00956 
+2026-10-18 09:35:43.475859: train_loss -0.4058 
+2026-10-18 09:35:43.476478: val_loss -0.3414 
+2026-10-18 09:35:43.477413: Pseudo dice [np.float32(0.9069), np.float32(0.6092), np.float32(0.7649), np.float32(0.7956), np.float32(0.5829), np.float32(0.8729), np.float32(0.6934), np.float32(0.6713), np.float32(0.0), np.float32(0.4358), np.float32(0.6434), np.float32(0.7357), np.float32(0.5427), np.float32(0.2124), np.float32(0.0), np.float32(0.0), np.float32(0.5247), np.float32(0.2425), np.float32(0.6302), np.float32(0.343), np.float32(0.2781), np.float32(0.5115), np.float32(0.5165), np.float32(0.5874), np.float32(0.6404), np.float32(0.6847), np.float32(0.4374), np.float32(0.7067), np.float32(0.6021), np.float32(0.6001), np.float32(0.614), np.float32(0.3173), np.float32(0.7603), np.float32(0.576), np.float32(0.8873), np.float32(0.9079)] 
+2026-10-18 09:35:43.477850: Epoch time: 109.16 s 
+2026-10-18 09:35:46.764632: Yayy! New best EMA pseudo Dice: 0.4401000142097473 
+2026-10-18 09:35:47.981087:  
+2026-10-18 09:35:47.981818: Epoch 50 
+2026-10-18 09:35:47.982070: Current learning rate: 0.00955 
+2026-10-18 09:37:37.038174: train_loss -0.409 
+2026-10-18 09:37:37.039029: val_loss -0.3134 
+2026-10-18 09:37:37.039548: Pseudo dice [np.float32(0.918), np.float32(0.6243), np.float32(0.7466), np.float32(0.748), np.float32(0.543), np.float32(0.8588), np.float32(0.7643), np.float32(0.6815), np.float32(0.0007), np.float32(0.484), np.float32(0.6195), np.float32(0.6956), np.float32(0.423), np.float32(0.3973), np.float32(0.0), np.float32(0.0), np.float32(0.4907), np.float32(0.2202), np.float32(0.6021), np.float32(0.3264), np.float32(0.3073), np.float32(0.6999), np.float32(0.3203), np.float32(0.5946), np.float32(0.6261), np.float32(0.6492), np.float32(0.151), np.float32(0.4039), np.float32(0.4864), np.float32(0.6241), np.float32(0.4479), np.float32(0.2614), np.float32(0.5294), np.float32(0.3641), np.float32(0.7162), np.float32(0.8433)] 
+2026-10-18 09:37:37.040074: Epoch time: 109.06 s 
+2026-10-18 09:37:37.040522: Yayy! New best EMA pseudo Dice: 0.4465999901294708 
+2026-10-18 09:37:38.693141:  
+2026-10-18 09:37:38.693942: Epoch 51 
+2026-10-18 09:37:38.694288: Current learning rate: 0.00954 
+2026-10-18 09:39:27.718748: train_loss -0.4022 
+2026-10-18 09:39:27.719281: val_loss -0.368 
+2026-10-18 09:39:27.719753: Pseudo dice [np.float32(0.8784), np.float32(0.6859), np.float32(0.762), np.float32(0.8219), np.float32(0.6925), np.float32(0.9036), np.float32(0.7652), np.float32(0.661), np.float32(0.0051), np.float32(0.5978), np.float32(0.7204), np.float32(0.7058), np.float32(0.4925), np.float32(0.2672), np.float32(0.0), np.float32(0.0), np.float32(0.5523), np.float32(0.3582), np.float32(0.5102), np.float32(0.2105), np.float32(0.4245), np.float32(0.5287), np.float32(0.5545), np.float32(0.6723), np.float32(0.6927), np.float32(0.6071), np.float32(0.3137), np.float32(0.7146), np.float32(0.5895), np.float32(0.6968), np.float32(0.5978), np.float32(0.3226), np.float32(0.6852), np.float32(0.6314), np.float32(0.8305), np.float32(0.855)] 
+2026-10-18 09:39:27.720041: Epoch time: 109.03 s 
+2026-10-18 09:39:27.720568: Yayy! New best EMA pseudo Dice: 0.45829999446868896 
+2026-10-18 09:39:28.910178:  
+2026-10-18 09:39:28.910600: Epoch 52 
+2026-10-18 09:39:28.910893: Current learning rate: 0.00953 
+2026-10-18 09:41:18.073827: train_loss -0.3876 
+2026-10-18 09:41:18.074766: val_loss -0.3402 
+2026-10-18 09:41:18.075320: Pseudo dice [np.float32(0.9098), np.float32(0.6485), np.float32(0.7346), np.float32(0.7496), np.float32(0.5577), np.float32(0.8499), np.float32(0.7342), np.float32(0.6406), np.float32(0.0006), np.float32(0.2515), np.float32(0.679), np.float32(0.6933), np.float32(0.3897), np.float32(0.3118), np.float32(0.0), np.float32(0.0), np.float32(0.4797), np.float32(0.2773), np.float32(0.617), np.float32(0.4164), np.float32(0.3731), np.float32(0.7197), np.float32(0.5422), np.float32(0.6687), np.float32(0.6426), np.float32(0.5832), np.float32(0.3211), np.float32(0.624), np.float32(0.567), np.float32(0.6402), np.float32(0.6054), np.float32(0.3029), np.float32(0.6802), np.float32(0.5344), np.float32(0.8434), np.float32(0.8685)] 
+2026-10-18 09:41:18.075596: Epoch time: 109.17 s 
+2026-10-18 09:41:18.075893: Yayy! New best EMA pseudo Dice: 0.4666000008583069 
+2026-10-18 09:41:19.255729:  
+2026-10-18 09:41:19.256427: Epoch 53 
+2026-10-18 09:41:19.256778: Current learning rate: 0.00952 
+2026-10-18 09:43:08.314264: train_loss -0.4098 
+2026-10-18 09:43:08.314818: val_loss -0.3643 
+2026-10-18 09:43:08.315415: Pseudo dice [np.float32(0.8717), np.float32(0.6759), np.float32(0.7866), np.float32(0.7577), np.float32(0.579), np.float32(0.8297), np.float32(0.7138), np.float32(0.5742), np.float32(0.015), np.float32(0.2641), np.float32(0.6383), np.float32(0.7105), np.float32(0.5669), np.float32(0.4327), np.float32(0.0), np.float32(0.0), np.float32(0.5204), np.float32(0.2843), np.float32(0.4479), np.float32(0.3439), np.float32(0.1957), np.float32(0.6356), np.float32(0.5987), np.float32(0.6672), np.float32(0.6067), np.float32(0.6678), np.float32(0.3717), np.float32(0.784), np.float32(0.663), np.float32(0.619), np.float32(0.4694), np.float32(0.3507), np.float32(0.5884), np.float32(0.5693), np.float32(0.8879), np.float32(0.9044)] 
+2026-10-18 09:43:08.316250: Epoch time: 109.06 s 
+2026-10-18 09:43:08.316620: Yayy! New best EMA pseudo Dice: 0.47429999709129333 
+2026-10-18 09:43:09.524216:  
+2026-10-18 09:43:09.524823: Epoch 54 
+2026-10-18 09:43:09.525129: Current learning rate: 0.00951 
+2026-10-18 09:44:58.635067: train_loss -0.4118 
+2026-10-18 09:44:58.635697: val_loss -0.3767 
+2026-10-18 09:44:58.636275: Pseudo dice [np.float32(0.9202), np.float32(0.7128), np.float32(0.8537), np.float32(0.8529), np.float32(0.5905), np.float32(0.921), np.float32(0.8109), np.float32(0.472), np.float32(0.0), np.float32(0.6005), np.float32(0.7663), np.float32(0.7997), np.float32(0.588), np.float32(0.3101), np.float32(0.0003), np.float32(0.0), np.float32(0.6373), np.float32(0.3021), np.float32(0.6924), np.float32(0.4398), np.float32(0.2896), np.float32(0.6211), np.float32(0.5348), np.float32(0.7285), np.float32(0.6973), np.float32(0.6445), np.float32(0.4359), np.float32(0.8128), np.float32(0.5745), np.float32(0.6457), np.float32(0.572), np.float32(0.2748), np.float32(0.735), np.float32(0.6461), np.float32(0.8764), np.float32(0.9016)] 
+2026-10-18 09:44:58.636866: Epoch time: 109.11 s 
+2026-10-18 09:44:58.637135: Yayy! New best EMA pseudo Dice: 0.48590001463890076 
+2026-10-18 09:44:59.914429:  
+2026-10-18 09:44:59.915133: Epoch 55 
+2026-10-18 09:44:59.915556: Current learning rate: 0.0095 
+2026-10-18 09:46:48.993209: train_loss -0.4086 
+2026-10-18 09:46:48.993820: val_loss -0.3724 
+2026-10-18 09:46:48.994619: Pseudo dice [np.float32(0.9148), np.float32(0.7311), np.float32(0.8054), np.float32(0.7891), np.float32(0.635), np.float32(0.8931), np.float32(0.7516), np.float32(0.4491), np.float32(0.0012), np.float32(0.4544), np.float32(0.6303), np.float32(0.7289), np.float32(0.6029), np.float32(0.36), np.float32(0.0), np.float32(0.0), np.float32(0.7055), np.float32(0.3127), np.float32(0.735), np.float32(0.316), np.float32(0.2995), np.float32(0.5739), np.float32(0.6318), np.float32(0.6833), np.float32(0.7168), np.float32(0.7077), np.float32(0.4764), np.float32(0.8104), np.float32(0.5985), np.float32(0.7554), np.float32(0.6298), np.float32(0.4045), np.float32(0.7102), np.float32(0.6635), np.float32(0.914), np.float32(0.9495)] 
+2026-10-18 09:46:48.995174: Epoch time: 109.08 s 
+2026-10-18 09:46:48.995481: Yayy! New best EMA pseudo Dice: 0.4966000020503998 
+2026-10-18 09:46:50.219731:  
+2026-10-18 09:46:50.220413: Epoch 56 
+2026-10-18 09:46:50.220721: Current learning rate: 0.00949 
+2026-10-18 09:48:39.324257: train_loss -0.4191 
+2026-10-18 09:48:39.324816: val_loss -0.3506 
+2026-10-18 09:48:39.325418: Pseudo dice [np.float32(0.9015), np.float32(0.6525), np.float32(0.8193), np.float32(0.7866), np.float32(0.5018), np.float32(0.8989), np.float32(0.7541), np.float32(0.3991), np.float32(0.0025), np.float32(0.572), np.float32(0.7434), np.float32(0.7473), np.float32(0.5699), np.float32(0.2707), np.float32(0.0002), np.float32(0.0), np.float32(0.6628), np.float32(0.4196), np.float32(0.6744), np.float32(0.5069), np.float32(0.3189), np.float32(0.5806), np.float32(0.6687), np.float32(0.7564), np.float32(0.6384), np.float32(0.6572), np.float32(0.474), np.float32(0.6469), np.float32(0.6322), np.float32(0.7039), np.float32(0.5321), np.float32(0.3914), np.float32(0.7007), np.float32(0.6096), np.float32(0.86), np.float32(0.9234)] 
+2026-10-18 09:48:39.326070: Epoch time: 109.11 s 
+2026-10-18 09:48:39.326486: Yayy! New best EMA pseudo Dice: 0.5052000284194946 
+2026-10-18 09:48:42.588894:  
+2026-10-18 09:48:42.589487: Epoch 57 
+2026-10-18 09:48:42.589762: Current learning rate: 0.00949 
+2026-10-18 09:50:31.683545: train_loss -0.4432 
+2026-10-18 09:50:31.684168: val_loss -0.3565 
+2026-10-18 09:50:31.684946: Pseudo dice [np.float32(0.8961), np.float32(0.6555), np.float32(0.8264), np.float32(0.8103), np.float32(0.6204), np.float32(0.8943), np.float32(0.8165), np.float32(0.605), np.float32(0.0), np.float32(0.3812), np.float32(0.7469), np.float32(0.7327), np.float32(0.5592), np.float32(0.4192), np.float32(0.0), np.float32(0.0), np.float32(0.6507), np.float32(0.4194), np.float32(0.6547), np.float32(0.426), np.float32(0.2763), np.float32(0.6005), np.float32(0.5676), np.float32(0.7089), np.float32(0.688), np.float32(0.6343), np.float32(0.3641), np.float32(0.7757), np.float32(0.4784), np.float32(0.6366), np.float32(0.4914), np.float32(0.4655), np.float32(0.7268), np.float32(0.6388), np.float32(0.884), np.float32(0.9197)] 
+2026-10-18 09:50:31.685230: Epoch time: 109.1 s 
+2026-10-18 09:50:31.685556: Yayy! New best EMA pseudo Dice: 0.5130000114440918 
+2026-10-18 09:50:32.919274:  
+2026-10-18 09:50:32.919990: Epoch 58 
+2026-10-18 09:50:32.920332: Current learning rate: 0.00948 
+2026-10-18 09:52:22.010911: train_loss -0.4295 
+2026-10-18 09:52:22.011680: val_loss -0.3864 
+2026-10-18 09:52:22.012371: Pseudo dice [np.float32(0.9207), np.float32(0.6949), np.float32(0.8456), np.float32(0.8661), np.float32(0.5654), np.float32(0.9139), np.float32(0.7242), np.float32(0.5047), np.float32(0.0), np.float32(0.5431), np.float32(0.7543), np.float32(0.7585), np.float32(0.5524), np.float32(0.3056), np.float32(0.0004), np.float32(0.0), np.float32(0.6295), np.float32(0.4331), np.float32(0.5384), np.float32(0.2906), np.float32(0.2074), np.float32(0.6605), np.float32(0.5826), np.float32(0.7912), np.float32(0.7353), np.float32(0.6969), np.float32(0.4942), np.float32(0.8367), np.float32(0.5728), np.float32(0.6397), np.float32(0.6751), np.float32(0.404), np.float32(0.7791), np.float32(0.6707), np.float32(0.8929), np.float32(0.935)] 
+2026-10-18 09:52:22.012961: Epoch time: 109.09 s 
+2026-10-18 09:52:22.013320: Yayy! New best EMA pseudo Dice: 0.5212000012397766 
+2026-10-18 09:52:23.245276:  
+2026-10-18 09:52:23.245974: Epoch 59 
+2026-10-18 09:52:23.246380: Current learning rate: 0.00947 
+2026-10-18 09:54:12.311130: train_loss -0.4433 
+2026-10-18 09:54:12.311667: val_loss -0.3376 
+2026-10-18 09:54:12.312634: Pseudo dice [np.float32(0.9129), np.float32(0.6537), np.float32(0.8213), np.float32(0.7775), np.float32(0.6372), np.float32(0.8824), np.float32(0.7625), np.float32(0.401), np.float32(0.0), np.float32(0.4286), np.float32(0.6865), np.float32(0.7034), np.float32(0.4219), np.float32(0.3239), np.float32(0.076), np.float32(0.0), np.float32(0.6378), np.float32(0.3636), np.float32(0.6925), np.float32(0.412), np.float32(0.4065), np.float32(0.6252), np.float32(0.5176), np.float32(0.6708), np.float32(0.6326), np.float32(0.5916), np.float32(0.4149), np.float32(0.8123), np.float32(0.4884), np.float32(0.6065), np.float32(0.6223), np.float32(0.2945), np.float32(0.7116), np.float32(0.6257), np.float32(0.7975), np.float32(0.8545)] 
+2026-10-18 09:54:12.313080: Epoch time: 109.07 s 
+2026-10-18 09:54:12.313474: Yayy! New best EMA pseudo Dice: 0.5253000259399414 
+2026-10-18 09:54:13.966148:  
+2026-10-18 09:54:13.966830: Epoch 60 
+2026-10-18 09:54:13.967046: Current learning rate: 0.00946 
+2026-10-18 09:56:02.981938: train_loss -0.4377 
+2026-10-18 09:56:02.982568: val_loss -0.3754 
+2026-10-18 09:56:02.983336: Pseudo dice [np.float32(0.9146), np.float32(0.6582), np.float32(0.8075), np.float32(0.8026), np.float32(0.6466), np.float32(0.8587), np.float32(0.7589), np.float32(0.5898), np.float32(0.0), np.float32(0.5269), np.float32(0.758), np.float32(0.7645), np.float32(0.4959), np.float32(0.4981), np.float32(0.0157), np.float32(0.0), np.float32(0.6852), np.float32(0.2546), np.float32(0.7605), np.float32(0.5354), np.float32(0.2766), np.float32(0.6127), np.float32(0.5228), np.float32(0.646), np.float32(0.6464), np.float32(0.6734), np.float32(0.312), np.float32(0.7217), np.float32(0.5319), np.float32(0.7483), np.float32(0.585), np.float32(0.2976), np.float32(0.735), np.float32(0.6445), np.float32(0.8183), np.float32(0.8444)] 
+2026-10-18 09:56:02.983950: Epoch time: 109.02 s 
+2026-10-18 09:56:02.984157: Yayy! New best EMA pseudo Dice: 0.531000018119812 
+2026-10-18 09:56:04.200572:  
+2026-10-18 09:56:04.201286: Epoch 61 
+2026-10-18 09:56:04.201484: Current learning rate: 0.00945 
+2026-10-18 09:57:53.215113: train_loss -0.454 
+2026-10-18 09:57:53.216058: val_loss -0.3872 
+2026-10-18 09:57:53.216658: Pseudo dice [np.float32(0.9199), np.float32(0.6789), np.float32(0.8331), np.float32(0.8621), np.float32(0.658), np.float32(0.9177), np.float32(0.7712), np.float32(0.5899), np.float32(0.0), np.float32(0.3712), np.float32(0.7574), np.float32(0.749), np.float32(0.6635), np.float32(0.5731), np.float32(0.145), np.float32(0.0), np.float32(0.7254), np.float32(0.3676), np.float32(0.7283), np.float32(0.3586), np.float32(0.3038), np.float32(0.6102), np.float32(0.4256), np.float32(0.74), np.float32(0.6771), np.float32(0.7055), np.float32(0.3048), np.float32(0.7612), np.float32(0.4425), np.float32(0.7018), np.float32(0.6397), np.float32(0.4522), np.float32(0.756), np.float32(0.727), np.float32(0.8831), np.float32(0.9145)] 
+2026-10-18 09:57:53.217003: Epoch time: 109.02 s 
+2026-10-18 09:57:53.217330: Yayy! New best EMA pseudo Dice: 0.5382000207901001 
+2026-10-18 09:57:54.437756:  
+2026-10-18 09:57:54.438376: Epoch 62 
+2026-10-18 09:57:54.438697: Current learning rate: 0.00944 
+2026-10-18 09:59:43.501647: train_loss -0.4518 
+2026-10-18 09:59:43.502430: val_loss -0.4214 
+2026-10-18 09:59:43.502940: Pseudo dice [np.float32(0.9364), np.float32(0.6717), np.float32(0.842), np.float32(0.8179), np.float32(0.718), np.float32(0.885), np.float32(0.7617), np.float32(0.5388), np.float32(0.0), np.float32(0.5017), np.float32(0.7149), np.float32(0.7631), np.float32(0.5362), np.float32(0.5086), np.float32(0.2428), np.float32(0.0), np.float32(0.5784), np.float32(0.3019), np.float32(0.6708), np.float32(0.3919), np.float32(0.4169), np.float32(0.6248), np.float32(0.717), np.float32(0.7925), np.float32(0.6803), np.float32(0.655), np.float32(0.4335), np.float32(0.595), np.float32(0.4489), np.float32(0.6506), np.float32(0.5867), np.float32(0.4764), np.float32(0.7002), np.float32(0.6364), np.float32(0.8968), np.float32(0.9145)] 
+2026-10-18 09:59:43.503512: Epoch time: 109.07 s 
+2026-10-18 09:59:43.503888: Yayy! New best EMA pseudo Dice: 0.5443999767303467 
+2026-10-18 09:59:45.748132:  
+2026-10-18 09:59:45.749464: Epoch 63 
+2026-10-18 09:59:45.749912: Current learning rate: 0.00943 
+2026-10-18 10:01:34.846892: train_loss -0.4499 
+2026-10-18 10:01:34.847513: val_loss -0.3225 
+2026-10-18 10:01:34.848056: Pseudo dice [np.float32(0.9363), np.float32(0.6236), np.float32(0.6172), np.float32(0.6911), np.float32(0.4148), np.float32(0.7875), np.float32(0.6297), np.float32(0.6677), np.float32(0.0054), np.float32(0.4427), np.float32(0.6287), np.float32(0.653), np.float32(0.5938), np.float32(0.4061), np.float32(0.2306), np.float32(0.0), np.float32(0.4176), np.float32(0.28), np.float32(0.5184), np.float32(0.3028), np.float32(0.3782), np.float32(0.5509), np.float32(0.51), np.float32(0.6105), np.float32(0.5693), np.float32(0.5038), np.float32(0.3889), np.float32(0.4572), np.float32(0.5276), np.float32(0.4678), np.float32(0.5208), np.float32(0.139), np.float32(0.5353), np.float32(0.5228), np.float32(0.7704), np.float32(0.7926)] 
+2026-10-18 10:01:34.848704: Epoch time: 109.1 s 
+2026-10-18 10:01:35.573406:  
+2026-10-18 10:01:35.574185: Epoch 64 
+2026-10-18 10:01:35.574444: Current learning rate: 0.00942 
+2026-10-18 10:03:24.600413: train_loss -0.4642 
+2026-10-18 10:03:24.601233: val_loss -0.3705 
+2026-10-18 10:03:24.601844: Pseudo dice [np.float32(0.9248), np.float32(0.6281), np.float32(0.8041), np.float32(0.796), np.float32(0.6507), np.float32(0.8764), np.float32(0.7803), np.float32(0.4844), np.float32(0.0), np.float32(0.5165), np.float32(0.7389), np.float32(0.7517), np.float32(0.6068), np.float32(0.5043), np.float32(0.3445), np.float32(0.0), np.float32(0.6348), np.float32(0.2459), np.float32(0.7095), np.float32(0.5301), np.float32(0.2769), np.float32(0.5888), np.float32(0.5895), np.float32(0.703), np.float32(0.6433), np.float32(0.6486), np.float32(0.2499), np.float32(0.571), np.float32(0.3707), np.float32(0.6983), np.float32(0.4876), np.float32(0.4342), np.float32(0.7295), np.float32(0.6658), np.float32(0.7969), np.float32(0.7959)] 
+2026-10-18 10:03:24.602171: Epoch time: 109.03 s 
+2026-10-18 10:03:25.554281:  
+2026-10-18 10:03:25.554932: Epoch 65 
+2026-10-18 10:03:25.555238: Current learning rate: 0.00941 
+2026-10-18 10:05:14.705933: train_loss -0.4683 
+2026-10-18 10:05:14.706695: val_loss -0.4061 
+2026-10-18 10:05:14.707186: Pseudo dice [np.float32(0.9307), np.float32(0.6759), np.float32(0.8691), np.float32(0.8802), np.float32(0.8163), np.float32(0.9271), np.float32(0.8419), np.float32(0.6585), np.float32(0.002), np.float32(0.4365), np.float32(0.773), np.float32(0.7547), np.float32(0.5298), np.float32(0.4047), np.float32(0.145), np.float32(0.0), np.float32(0.7003), np.float32(0.4087), np.float32(0.7766), np.float32(0.5087), np.float32(0.3669), np.float32(0.6052), np.float32(0.592), np.float32(0.6898), np.float32(0.6631), np.float32(0.7418), np.float32(0.4461), np.float32(0.7899), np.float32(0.5282), np.float32(0.7191), np.float32(0.553), np.float32(0.5299), np.float32(0.8245), np.float32(0.6754), np.float32(0.8617), np.float32(0.9184)] 
+2026-10-18 10:05:14.707701: Epoch time: 109.15 s 
+2026-10-18 10:05:14.707989: Yayy! New best EMA pseudo Dice: 0.5522000193595886 
+2026-10-18 10:05:15.929843:  
+2026-10-18 10:05:15.930401: Epoch 66 
+2026-10-18 10:05:15.930663: Current learning rate: 0.0094 
+2026-10-18 10:07:05.033640: train_loss -0.4525 
+2026-10-18 10:07:05.034134: val_loss -0.3383 
+2026-10-18 10:07:05.034688: Pseudo dice [np.float32(0.8894), np.float32(0.6198), np.float32(0.7168), np.float32(0.6608), np.float32(0.5377), np.float32(0.6973), np.float32(0.6666), np.float32(0.5362), np.float32(0.0035), np.float32(0.5132), np.float32(0.586), np.float32(0.4306), np.float32(0.5247), np.float32(0.4469), np.float32(0.2108), np.float32(0.0), np.float32(0.6657), np.float32(0.4416), np.float32(0.6005), np.float32(0.4347), np.float32(0.4206), np.float32(0.6796), np.float32(0.6349), np.float32(0.7498), np.float32(0.6545), np.float32(0.5708), np.float32(0.3666), np.float32(0.7393), np.float32(0.5128), np.float32(0.594), np.float32(0.5313), np.float32(0.2514), np.float32(0.632), np.float32(0.5026), np.float32(0.8355), np.float32(0.847)] 
+2026-10-18 10:07:05.035660: Epoch time: 109.11 s 
+2026-10-18 10:07:05.764140:  
+2026-10-18 10:07:05.764807: Epoch 67 
+2026-10-18 10:07:05.765094: Current learning rate: 0.00939 
+2026-10-18 10:08:54.824969: train_loss -0.485 
+2026-10-18 10:08:54.825757: val_loss -0.418 
+2026-10-18 10:08:54.826269: Pseudo dice [np.float32(0.9307), np.float32(0.7466), np.float32(0.8564), np.float32(0.8567), np.float32(0.7651), np.float32(0.9256), np.float32(0.8586), np.float32(0.5583), np.float32(0.0013), np.float32(0.5181), np.float32(0.7525), np.float32(0.7939), np.float32(0.5489), np.float32(0.5561), np.float32(0.1374), np.float32(0.0), np.float32(0.7672), np.float32(0.4951), np.float32(0.8117), np.float32(0.5294), np.float32(0.4532), np.float32(0.7191), np.float32(0.6407), np.float32(0.7173), np.float32(0.7579), np.float32(0.6692), np.float32(0.4655), np.float32(0.772), np.float32(0.6112), np.float32(0.709), np.float32(0.6387), np.float32(0.4957), np.float32(0.6729), np.float32(0.6955), np.float32(0.8803), np.float32(0.9176)] 
+2026-10-18 10:08:54.826638: Epoch time: 109.06 s 
+2026-10-18 10:08:54.826789: Yayy! New best EMA pseudo Dice: 0.5609999895095825 
+2026-10-18 10:08:56.892412:  
+2026-10-18 10:08:56.895241: Epoch 68 
+2026-10-18 10:08:56.897156: Current learning rate: 0.00939 
+2026-10-18 10:10:45.959337: train_loss -0.4628 
+2026-10-18 10:10:45.959935: val_loss -0.3912 
+2026-10-18 10:10:45.960387: Pseudo dice [np.float32(0.9079), np.float32(0.7169), np.float32(0.8473), np.float32(0.8171), np.float32(0.6088), np.float32(0.9195), np.float32(0.8146), np.float32(0.6482), np.float32(0.0), np.float32(0.3958), np.float32(0.7679), np.float32(0.774), np.float32(0.6315), np.float32(0.5552), np.float32(0.1737), np.float32(0.0), np.float32(0.7412), np.float32(0.4553), np.float32(0.7876), np.float32(0.5787), np.float32(0.4632), np.float32(0.689), np.float32(0.6437), np.float32(0.7453), np.float32(0.6716), np.float32(0.6456), np.float32(0.3671), np.float32(0.5209), np.float32(0.4656), np.float32(0.6801), np.float32(0.7336), np.float32(0.4263), np.float32(0.7941), np.float32(0.6821), np.float32(0.9095), np.float32(0.9514)] 
+2026-10-18 10:10:45.960816: Epoch time: 109.07 s 
+2026-10-18 10:10:45.961281: Yayy! New best EMA pseudo Dice: 0.5674999952316284 
+2026-10-18 10:10:47.183845:  
+2026-10-18 10:10:47.184291: Epoch 69 
+2026-10-18 10:10:47.184524: Current learning rate: 0.00938 
+2026-10-18 10:12:36.244846: train_loss -0.4711 
+2026-10-18 10:12:36.245830: val_loss -0.3706 
+2026-10-18 10:12:36.246742: Pseudo dice [np.float32(0.9194), np.float32(0.7042), np.float32(0.8676), np.float32(0.8205), np.float32(0.7348), np.float32(0.9217), np.float32(0.7884), np.float32(0.6186), np.float32(0.0), np.float32(0.4336), np.float32(0.7731), np.float32(0.7724), np.float32(0.6029), np.float32(0.5485), np.float32(0.2798), np.float32(0.0), np.float32(0.7901), np.float32(0.4795), np.float32(0.7827), np.float32(0.572), np.float32(0.416), np.float32(0.663), np.float32(0.366), np.float32(0.6139), np.float32(0.6788), np.float32(0.6925), np.float32(0.3119), np.float32(0.7363), np.float32(0.4066), np.float32(0.569), np.float32(0.6275), np.float32(0.4985), np.float32(0.7938), np.float32(0.7146), np.float32(0.8174), np.float32(0.886)] 
+2026-10-18 10:12:36.247186: Epoch time: 109.06 s 
+2026-10-18 10:12:36.247619: Yayy! New best EMA pseudo Dice: 0.5723999738693237 
+2026-10-18 10:12:37.482033:  
+2026-10-18 10:12:37.482824: Epoch 70 
+2026-10-18 10:12:37.483046: Current learning rate: 0.00937 
+2026-10-18 10:14:26.549511: train_loss -0.4672 
+2026-10-18 10:14:26.550224: val_loss -0.3872 
+2026-10-18 10:14:26.551060: Pseudo dice [np.float32(0.9202), np.float32(0.6343), np.float32(0.7823), np.float32(0.7002), np.float32(0.5816), np.float32(0.8141), np.float32(0.7247), np.float32(0.5838), np.float32(0.0), np.float32(0.4966), np.float32(0.6505), np.float32(0.7288), np.float32(0.5404), np.float32(0.5698), np.float32(0.3603), np.float32(0.0), np.float32(0.6522), np.float32(0.3586), np.float32(0.708), np.float32(0.4245), np.float32(0.3784), np.float32(0.5197), np.float32(0.5206), np.float32(0.6166), np.float32(0.5713), np.float32(0.6057), np.float32(0.3294), np.float32(0.565), np.float32(0.4328), np.float32(0.7137), np.float32(0.6794), np.float32(0.266), np.float32(0.6001), np.float32(0.554), np.float32(0.7679), np.float32(0.8476)] 
+2026-10-18 10:14:26.551566: Epoch time: 109.07 s 
+2026-10-18 10:14:27.297053:  
+2026-10-18 10:14:27.298674: Epoch 71 
+2026-10-18 10:14:27.299031: Current learning rate: 0.00936 
+2026-10-18 10:16:16.411434: train_loss -0.4537 
+2026-10-18 10:16:16.411924: val_loss -0.3972 
+2026-10-18 10:16:16.412518: Pseudo dice [np.float32(0.9258), np.float32(0.7052), np.float32(0.8644), np.float32(0.8686), np.float32(0.7438), np.float32(0.9193), np.float32(0.8053), np.float32(0.5501), np.float32(0.0), np.float32(0.3703), np.float32(0.7502), np.float32(0.7531), np.float32(0.5788), np.float32(0.3646), np.float32(0.4511), np.float32(0.0), np.float32(0.7595), np.float32(0.4155), np.float32(0.7796), np.float32(0.4809), np.float32(0.3621), np.float32(0.6914), np.float32(0.5693), np.float32(0.7394), np.float32(0.7284), np.float32(0.7199), np.float32(0.3056), np.float32(0.6846), np.float32(0.4881), np.float32(0.6478), np.float32(0.6921), np.float32(0.566), np.float32(0.8216), np.float32(0.6968), np.float32(0.8977), np.float32(0.9187)] 
+2026-10-18 10:16:16.413270: Epoch time: 109.12 s 
+2026-10-18 10:16:16.413657: Yayy! New best EMA pseudo Dice: 0.5770000219345093 
+2026-10-18 10:16:19.760298:  
+2026-10-18 10:16:19.761038: Epoch 72 
+2026-10-18 10:16:19.764355: Current learning rate: 0.00935 
+2026-10-18 10:18:08.867435: train_loss -0.4687 
+2026-10-18 10:18:08.868393: val_loss -0.3978 
+2026-10-18 10:18:08.868904: Pseudo dice [np.float32(0.9346), np.float32(0.6601), np.float32(0.8437), np.float32(0.8072), np.float32(0.6722), np.float32(0.89), np.float32(0.7717), np.float32(0.4307), np.float32(0.0007), np.float32(0.4236), np.float32(0.706), np.float32(0.7731), np.float32(0.5302), np.float32(0.6167), np.float32(0.0749), np.float32(0.0), np.float32(0.6296), np.float32(0.2962), np.float32(0.71), np.float32(0.478), np.float32(0.3503), np.float32(0.6938), np.float32(0.6053), np.float32(0.6822), np.float32(0.6348), np.float32(0.696), np.float32(0.3562), np.float32(0.7167), np.float32(0.4107), np.float32(0.7044), np.float32(0.6469), np.float32(0.257), np.float32(0.706), np.float32(0.5902), np.float32(0.8688), np.float32(0.9147)] 
+2026-10-18 10:18:08.869405: Epoch time: 109.11 s 
+2026-10-18 10:18:08.869689: Yayy! New best EMA pseudo Dice: 0.5777999758720398 
+2026-10-18 10:18:11.320791:  
+2026-10-18 10:18:11.321503: Epoch 73 
+2026-10-18 10:18:11.321916: Current learning rate: 0.00934 
+2026-10-18 10:20:00.342822: train_loss -0.4494 
+2026-10-18 10:20:00.343319: val_loss -0.4177 
+2026-10-18 10:20:00.343771: Pseudo dice [np.float32(0.931), np.float32(0.7452), np.float32(0.8644), np.float32(0.8735), np.float32(0.7409), np.float32(0.9193), np.float32(0.8615), np.float32(0.5755), np.float32(0.0), np.float32(0.497), np.float32(0.7962), np.float32(0.8182), np.float32(0.5694), np.float32(0.379), np.float32(0.2221), np.float32(0.0), np.float32(0.745), np.float32(0.4112), np.float32(0.7758), np.float32(0.3486), np.float32(0.39), np.float32(0.615), np.float32(0.73), np.float32(0.7605), np.float32(0.7253), np.float32(0.762), np.float32(0.5642), np.float32(0.7233), np.float32(0.6231), np.float32(0.6781), np.float32(0.4752), np.float32(0.4603), np.float32(0.7671), np.float32(0.721), np.float32(0.908), np.float32(0.9372)] 
+2026-10-18 10:20:00.344382: Epoch time: 109.02 s 
+2026-10-18 10:20:00.344548: Yayy! New best EMA pseudo Dice: 0.5837000012397766 
+2026-10-18 10:20:01.543284:  
+2026-10-18 10:20:01.543985: Epoch 74 
+2026-10-18 10:20:01.544383: Current learning rate: 0.00933 
+2026-10-18 10:21:50.489974: train_loss -0.4164 
+2026-10-18 10:21:50.490697: val_loss -0.3509 
+2026-10-18 10:21:50.491306: Pseudo dice [np.float32(0.9368), np.float32(0.6415), np.float32(0.7521), np.float32(0.7223), np.float32(0.4962), np.float32(0.8692), np.float32(0.7499), np.float32(0.4572), np.float32(0.0007), np.float32(0.2322), np.float32(0.6987), np.float32(0.7224), np.float32(0.5002), np.float32(0.2937), np.float32(0.0152), np.float32(0.0), np.float32(0.5898), np.float32(0.352), np.float32(0.6608), np.float32(0.3604), np.float32(0.3886), np.float32(0.5493), np.float32(0.6278), np.float32(0.6508), np.float32(0.6439), np.float32(0.5428), np.float32(0.4511), np.float32(0.7468), np.float32(0.5096), np.float32(0.516), np.float32(0.6072), np.float32(0.3973), np.float32(0.7277), np.float32(0.5855), np.float32(0.8374), np.float32(0.8925)] 
+2026-10-18 10:21:50.491808: Epoch time: 108.95 s 
+2026-10-18 10:21:51.237592:  
+2026-10-18 10:21:51.238210: Epoch 75 
+2026-10-18 10:21:51.238536: Current learning rate: 0.00932 
+2026-10-18 10:23:40.296629: train_loss -0.4646 
+2026-10-18 10:23:40.297175: val_loss -0.35 
+2026-10-18 10:23:40.297770: Pseudo dice [np.float32(0.9348), np.float32(0.6564), np.float32(0.7736), np.float32(0.6337), np.float32(0.5719), np.float32(0.7546), np.float32(0.7449), np.float32(0.5701), np.float32(0.0009), np.float32(0.5306), np.float32(0.5733), np.float32(0.6346), np.float32(0.5319), np.float32(0.4546), np.float32(0.3427), np.float32(0.0), np.float32(0.5761), np.float32(0.3106), np.float32(0.713), np.float32(0.4163), np.float32(0.3744), np.float32(0.7168), np.float32(0.5271), np.float32(0.633), np.float32(0.5962), np.float32(0.5709), np.float32(0.3936), np.float32(0.7343), np.float32(0.424), np.float32(0.5731), np.float32(0.609), np.float32(0.2283), np.float32(0.5006), np.float32(0.4783), np.float32(0.7631), np.float32(0.8229)] 
+2026-10-18 10:23:40.298418: Epoch time: 109.06 s 
+2026-10-18 10:23:41.043569:  
+2026-10-18 10:23:41.044202: Epoch 76 
+2026-10-18 10:23:41.044465: Current learning rate: 0.00931 
+2026-10-18 10:25:30.083427: train_loss -0.4663 
+2026-10-18 10:25:30.083912: val_loss -0.4085 
+2026-10-18 10:25:30.084404: Pseudo dice [np.float32(0.899), np.float32(0.6935), np.float32(0.8738), np.float32(0.8265), np.float32(0.71), np.float32(0.9066), np.float32(0.8211), np.float32(0.5474), np.float32(0.0), np.float32(0.2797), np.float32(0.6968), np.float32(0.788), np.float32(0.6055), np.float32(0.5338), np.float32(0.0363), np.float32(0.0), np.float32(0.7332), np.float32(0.4172), np.float32(0.7742), np.float32(0.5521), np.float32(0.452), np.float32(0.7133), np.float32(0.4198), np.float32(0.5707), np.float32(0.7017), np.float32(0.7053), np.float32(0.403), np.float32(0.8655), np.float32(0.3734), np.float32(0.526), np.float32(0.5243), np.float32(0.617), np.float32(0.7692), np.float32(0.6603), np.float32(0.8866), np.float32(0.9448)] 
+2026-10-18 10:25:30.085057: Epoch time: 109.04 s 
+2026-10-18 10:25:30.783549:  
+2026-10-18 10:25:30.784137: Epoch 77 
+2026-10-18 10:25:30.784336: Current learning rate: 0.0093 
+2026-10-18 10:27:19.679186: train_loss -0.4719 
+2026-10-18 10:27:19.680009: val_loss -0.3763 
+2026-10-18 10:27:19.680526: Pseudo dice [np.float32(0.9216), np.float32(0.6438), np.float32(0.7826), np.float32(0.7786), np.float32(0.6868), np.float32(0.8798), np.float32(0.7411), np.float32(0.5995), np.float32(0.001), np.float32(0.6009), np.float32(0.7488), np.float32(0.7979), np.float32(0.6185), np.float32(0.4754), np.float32(0.0191), np.float32(0.0), np.float32(0.7064), np.float32(0.4624), np.float32(0.6917), np.float32(0.6182), np.float32(0.3786), np.float32(0.6728), np.float32(0.6262), np.float32(0.7217), np.float32(0.6041), np.float32(0.6479), np.float32(0.4576), np.float32(0.6945), np.float32(0.4869), np.float32(0.6331), np.float32(0.5802), np.float32(0.4257), np.float32(0.7409), np.float32(0.6277), np.float32(0.8408), np.float32(0.8841)] 
+2026-10-18 10:27:19.681024: Epoch time: 108.9 s 
+2026-10-18 10:27:20.419341:  
+2026-10-18 10:27:20.419953: Epoch 78 
+2026-10-18 10:27:20.420228: Current learning rate: 0.0093 
+2026-10-18 10:29:09.412081: train_loss -0.4845 
+2026-10-18 10:29:09.412565: val_loss -0.4239 
+2026-10-18 10:29:09.413101: Pseudo dice [np.float32(0.9375), np.float32(0.6658), np.float32(0.8369), np.float32(0.8664), np.float32(0.7459), np.float32(0.9057), np.float32(0.7656), np.float32(0.4804), np.float32(0.0003), np.float32(0.4108), np.float32(0.7487), np.float32(0.7539), np.float32(0.6499), np.float32(0.6227), np.float32(0.4396), np.float32(0.0), np.float32(0.808), np.float32(0.3604), np.float32(0.8182), np.float32(0.5417), np.float32(0.4302), np.float32(0.6404), np.float32(0.6901), np.float32(0.7515), np.float32(0.7127), np.float32(0.7208), np.float32(0.5268), np.float32(0.8662), np.float32(0.5236), np.float32(0.6685), np.float32(0.6599), np.float32(0.3972), np.float32(0.7911), np.float32(0.7187), np.float32(0.8941), np.float32(0.9137)] 
+2026-10-18 10:29:09.413866: Epoch time: 108.99 s 
+2026-10-18 10:29:09.414189: Yayy! New best EMA pseudo Dice: 0.588699996471405 
+2026-10-18 10:29:10.822458:  
+2026-10-18 10:29:10.827109: Epoch 79 
+2026-10-18 10:29:10.829163: Current learning rate: 0.00929 
+2026-10-18 10:30:59.783380: train_loss -0.466 
+2026-10-18 10:30:59.783841: val_loss -0.3559 
+2026-10-18 10:30:59.784372: Pseudo dice [np.float32(0.9241), np.float32(0.6701), np.float32(0.7358), np.float32(0.7486), np.float32(0.6287), np.float32(0.8317), np.float32(0.6725), np.float32(0.5296), np.float32(0.0002), np.float32(0.4407), np.float32(0.6408), np.float32(0.6496), np.float32(0.5554), np.float32(0.521), np.float32(0.0186), np.float32(0.0), np.float32(0.7199), np.float32(0.4665), np.float32(0.7032), np.float32(0.3412), np.float32(0.2906), np.float32(0.5592), np.float32(0.6352), np.float32(0.6394), np.float32(0.6839), np.float32(0.5715), np.float32(0.4185), np.float32(0.6858), np.float32(0.4359), np.float32(0.6618), np.float32(0.5219), np.float32(0.3471), np.float32(0.725), np.float32(0.6682), np.float32(0.8175), np.float32(0.8461)] 
+2026-10-18 10:30:59.785225: Epoch time: 108.96 s 
+2026-10-18 10:31:00.540580:  
+2026-10-18 10:31:00.541183: Epoch 80 
+2026-10-18 10:31:00.541438: Current learning rate: 0.00928 
+2026-10-18 10:32:49.533848: train_loss -0.4756 
+2026-10-18 10:32:49.534651: val_loss -0.3759 
+2026-10-18 10:32:49.535209: Pseudo dice [np.float32(0.9369), np.float32(0.6981), np.float32(0.8205), np.float32(0.7715), np.float32(0.4295), np.float32(0.8401), np.float32(0.5219), np.float32(0.7314), np.float32(0.0087), np.float32(0.4095), np.float32(0.6843), np.float32(0.7242), np.float32(0.6027), np.float32(0.4748), np.float32(0.401), np.float32(0.0), np.float32(0.5396), np.float32(0.3881), np.float32(0.6293), np.float32(0.4335), np.float32(0.5197), np.float32(0.7778), np.float32(0.598), np.float32(0.7393), np.float32(0.6812), np.float32(0.6926), np.float32(0.416), np.float32(0.7735), np.float32(0.6176), np.float32(0.6817), np.float32(0.612), np.float32(0.1962), np.float32(0.7087), np.float32(0.5424), np.float32(0.8938), np.float32(0.9178)] 
+2026-10-18 10:32:49.535704: Epoch time: 109.0 s 
+2026-10-18 10:32:50.246344:  
+2026-10-18 10:32:50.247045: Epoch 81 
+2026-10-18 10:32:50.247379: Current learning rate: 0.00927 
+2026-10-18 10:34:39.234519: train_loss -0.457 
+2026-10-18 10:34:39.235036: val_loss -0.3703 
+2026-10-18 10:34:39.235478: Pseudo dice [np.float32(0.9068), np.float32(0.6531), np.float32(0.8418), np.float32(0.8144), np.float32(0.6157), np.float32(0.8947), np.float32(0.7752), np.float32(0.5356), np.float32(0.0), np.float32(0.2494), np.float32(0.7224), np.float32(0.7732), np.float32(0.6447), np.float32(0.4109), np.float32(0.3189), np.float32(0.0), np.float32(0.6487), np.float32(0.5301), np.float32(0.7083), np.float32(0.507), np.float32(0.2608), np.float32(0.6512), np.float32(0.6406), np.float32(0.6386), np.float32(0.6201), np.float32(0.6357), np.float32(0.3275), np.float32(0.5503), np.float32(0.4983), np.float32(0.6306), np.float32(0.5567), np.float32(0.5211), np.float32(0.7041), np.float32(0.6705), np.float32(0.7476), np.float32(0.7806)] 
+2026-10-18 10:34:39.236200: Epoch time: 108.99 s 
+2026-10-18 10:34:39.964676:  
+2026-10-18 10:34:39.965216: Epoch 82 
+2026-10-18 10:34:39.965394: Current learning rate: 0.00926 
+2026-10-18 10:36:28.933428: train_loss -0.4845 
+2026-10-18 10:36:28.934033: val_loss -0.3201 
+2026-10-18 10:36:28.934674: Pseudo dice [np.float32(0.9049), np.float32(0.5613), np.float32(0.7028), np.float32(0.6717), np.float32(0.5656), np.float32(0.7458), np.float32(0.6392), np.float32(0.4577), np.float32(0.0), np.float32(0.5228), np.float32(0.6612), np.float32(0.679), np.float32(0.4573), np.float32(0.226), np.float32(0.1463), np.float32(0.0), np.float32(0.5631), np.float32(0.2814), np.float32(0.5263), np.float32(0.4622), np.float32(0.3026), np.float32(0.5492), np.float32(0.4867), np.float32(0.5581), np.float32(0.5864), np.float32(0.5077), np.float32(0.3612), np.float32(0.6446), np.float32(0.4698), np.float32(0.535), np.float32(0.5132), np.float32(0.3582), np.float32(0.607), np.float32(0.5349), np.float32(0.7153), np.float32(0.7426)] 
+2026-10-18 10:36:28.934928: Epoch time: 108.97 s 
+2026-10-18 10:36:29.898720:  
+2026-10-18 10:36:29.899298: Epoch 83 
+2026-10-18 10:36:29.899600: Current learning rate: 0.00925 
+2026-10-18 10:38:18.960377: train_loss -0.4286 
+2026-10-18 10:38:18.960999: val_loss -0.3488 
+2026-10-18 10:38:18.961643: Pseudo dice [np.float32(0.9126), np.float32(0.6322), np.float32(0.7883), np.float32(0.7515), np.float32(0.5887), np.float32(0.8205), np.float32(0.7006), np.float32(0.4826), np.float32(0.0), np.float32(0.5032), np.float32(0.6312), np.float32(0.7186), np.float32(0.6264), np.float32(0.5344), np.float32(0.0139), np.float32(0.0), np.float32(0.6314), np.float32(0.2592), np.float32(0.5795), np.float32(0.2662), np.float32(0.3604), np.float32(0.6671), np.float32(0.5297), np.float32(0.5483), np.float32(0.6348), np.float32(0.6249), np.float32(0.3142), np.float32(0.6804), np.float32(0.626), np.float32(0.7035), np.float32(0.5681), np.float32(0.4229), np.float32(0.6897), np.float32(0.5768), np.float32(0.7627), np.float32(0.7999)] 
+2026-10-18 10:38:18.961943: Epoch time: 109.06 s 
+2026-10-18 10:38:19.658996:  
+2026-10-18 10:38:19.659541: Epoch 84 
+2026-10-18 10:38:19.659815: Current learning rate: 0.00924 
+2026-10-18 10:40:08.693285: train_loss -0.4492 
+2026-10-18 10:40:08.694015: val_loss -0.3444 
+2026-10-18 10:40:08.694760: Pseudo dice [np.float32(0.9168), np.float32(0.6301), np.float32(0.7631), np.float32(0.794), np.float32(0.5297), np.float32(0.8138), np.float32(0.779), np.float32(0.5829), np.float32(0.0051), np.float32(0.5129), np.float32(0.7414), np.float32(0.7603), np.float32(0.5963), np.float32(0.5234), np.float32(0.1126), np.float32(0.0), np.float32(0.511), np.float32(0.2569), np.float32(0.7339), np.float32(0.4195), np.float32(0.2168), np.float32(0.473), np.float32(0.4634), np.float32(0.5761), np.float32(0.6326), np.float32(0.6545), np.float32(0.3013), np.float32(0.601), np.float32(0.524), np.float32(0.5537), np.float32(0.3319), np.float32(0.4429), np.float32(0.7179), np.float32(0.6689), np.float32(0.717), np.float32(0.8034)] 
+2026-10-18 10:40:08.695295: Epoch time: 109.04 s 
+2026-10-18 10:40:09.413599:  
+2026-10-18 10:40:09.414238: Epoch 85 
+2026-10-18 10:40:09.414589: Current learning rate: 0.00923 
+2026-10-18 10:41:58.486008: train_loss -0.4419 
+2026-10-18 10:41:58.486992: val_loss -0.3568 
+2026-10-18 10:41:58.487490: Pseudo dice [np.float32(0.8948), np.float32(0.7368), np.float32(0.8299), np.float32(0.8411), np.float32(0.7053), np.float32(0.9054), np.float32(0.8427), np.float32(0.6018), np.float32(0.0), np.float32(0.4055), np.float32(0.7357), np.float32(0.779), np.float32(0.695), np.float32(0.5567), np.float32(0.0459), np.float32(0.0), np.float32(0.7011), np.float32(0.3686), np.float32(0.7863), np.float32(0.5214), np.float32(0.4221), np.float32(0.6808), np.float32(0.5111), np.float32(0.4753), np.float32(0.7407), np.float32(0.6549), np.float32(0.3083), np.float32(0.5727), np.float32(0.4474), np.float32(0.5889), np.float32(0.6369), np.float32(0.4127), np.float32(0.7564), np.float32(0.6066), np.float32(0.7987), np.float32(0.8381)] 
+2026-10-18 10:41:58.487756: Epoch time: 109.07 s 
+2026-10-18 10:41:59.192662:  
+2026-10-18 10:41:59.193428: Epoch 86 
+2026-10-18 10:41:59.193874: Current learning rate: 0.00922 
+2026-10-18 10:43:48.327845: train_loss -0.475 
+2026-10-18 10:43:48.328712: val_loss -0.4249 
+2026-10-18 10:43:48.329419: Pseudo dice [np.float32(0.9263), np.float32(0.7002), np.float32(0.8382), np.float32(0.8737), np.float32(0.7189), np.float32(0.9106), np.float32(0.7944), np.float32(0.5038), np.float32(0.0004), np.float32(0.4352), np.float32(0.7608), np.float32(0.7801), np.float32(0.6088), np.float32(0.3437), np.float32(0.2466), np.float32(0.0), np.float32(0.7537), np.float32(0.4076), np.float32(0.7745), np.float32(0.3858), np.float32(0.4466), np.float32(0.6324), np.float32(0.7734), np.float32(0.803), np.float32(0.7209), np.float32(0.7212), np.float32(0.4797), np.float32(0.8065), np.float32(0.6411), np.float32(0.6922), np.float32(0.7148), np.float32(0.3644), np.float32(0.7816), np.float32(0.5868), np.float32(0.9257), np.float32(0.9327)] 
+2026-10-18 10:43:48.329654: Epoch time: 109.14 s 
+2026-10-18 10:43:49.073008:  
+2026-10-18 10:43:49.073464: Epoch 87 
+2026-10-18 10:43:49.073757: Current learning rate: 0.00921 
+2026-10-18 10:45:38.166233: train_loss -0.5033 
+2026-10-18 10:45:38.167060: val_loss -0.3853 
+2026-10-18 10:45:38.167608: Pseudo dice [np.float32(0.9233), np.float32(0.6682), np.float32(0.8337), np.float32(0.836), np.float32(0.6559), np.float32(0.879), np.float32(0.7853), np.float32(0.4408), np.float32(0.0006), np.float32(0.4525), np.float32(0.7328), np.float32(0.7688), np.float32(0.6454), np.float32(0.3577), np.float32(0.1121), np.float32(0.0), np.float32(0.5459), np.float32(0.1132), np.float32(0.6979), np.float32(0.49), np.float32(0.4789), np.float32(0.745), np.float32(0.5631), np.float32(0.6666), np.float32(0.6903), np.float32(0.7387), np.float32(0.2673), np.float32(0.752), np.float32(0.4806), np.float32(0.6876), np.float32(0.448), np.float32(0.4666), np.float32(0.7415), np.float32(0.6611), np.float32(0.8803), np.float32(0.8911)] 
+2026-10-18 10:45:38.168030: Epoch time: 109.1 s 
+2026-10-18 10:45:38.892291:  
+2026-10-18 10:45:38.893807: Epoch 88 
+2026-10-18 10:45:38.894182: Current learning rate: 0.0092 
+2026-10-18 10:47:28.029364: train_loss -0.48 
+2026-10-18 10:47:28.030176: val_loss -0.3663 
+2026-10-18 10:47:28.030708: Pseudo dice [np.float32(0.9261), np.float32(0.6107), np.float32(0.7966), np.float32(0.6811), np.float32(0.6249), np.float32(0.8033), np.float32(0.7824), np.float32(0.4917), np.float32(0.0015), np.float32(0.518), np.float32(0.7038), np.float32(0.7638), np.float32(0.5408), np.float32(0.4173), np.float32(0.4848), np.float32(0.0), np.float32(0.6444), np.float32(0.4408), np.float32(0.7789), np.float32(0.6258), np.float32(0.3248), np.float32(0.713), np.float32(0.6904), np.float32(0.772), np.float32(0.6466), np.float32(0.6631), np.float32(0.4912), np.float32(0.7793), np.float32(0.5629), np.float32(0.7263), np.float32(0.4709), np.float32(0.4682), np.float32(0.6812), np.float32(0.5625), np.float32(0.7793), np.float32(0.8723)] 
+2026-10-18 10:47:28.031219: Epoch time: 109.14 s 
+2026-10-18 10:47:28.776448:  
+2026-10-18 10:47:28.777119: Epoch 89 
+2026-10-18 10:47:28.777439: Current learning rate: 0.0092 
+2026-10-18 10:49:17.934982: train_loss -0.4778 
+2026-10-18 10:49:17.935703: val_loss -0.425 
+2026-10-18 10:49:17.936216: Pseudo dice [np.float32(0.9098), np.float32(0.7421), np.float32(0.8793), np.float32(0.8774), np.float32(0.7567), np.float32(0.9246), np.float32(0.8631), np.float32(0.5832), np.float32(0.0073), np.float32(0.3585), np.float32(0.7336), np.float32(0.7743), np.float32(0.4854), np.float32(0.4312), np.float32(0.0417), np.float32(0.0), np.float32(0.7839), np.float32(0.5), np.float32(0.8427), np.float32(0.6233), np.float32(0.5467), np.float32(0.6306), np.float32(0.7554), np.float32(0.7866), np.float32(0.7193), np.float32(0.7544), np.float32(0.465), np.float32(0.7974), np.float32(0.6288), np.float32(0.7293), np.float32(0.6491), np.float32(0.5038), np.float32(0.7384), np.float32(0.6913), np.float32(0.8869), np.float32(0.9257)] 
+2026-10-18 10:49:17.936779: Epoch time: 109.16 s 
+2026-10-18 10:49:17.937159: Yayy! New best EMA pseudo Dice: 0.590499997138977 
+2026-10-18 10:49:19.527280:  
+2026-10-18 10:49:19.530198: Epoch 90 
+2026-10-18 10:49:19.530945: Current learning rate: 0.00919 
+2026-10-18 10:51:08.691317: train_loss -0.5111 
+2026-10-18 10:51:08.692168: val_loss -0.3891 
+2026-10-18 10:51:08.692784: Pseudo dice [np.float32(0.923), np.float32(0.7322), np.float32(0.8342), np.float32(0.7864), np.float32(0.6159), np.float32(0.8393), np.float32(0.7452), np.float32(0.5511), np.float32(0.0024), np.float32(0.4579), np.float32(0.695), np.float32(0.7362), np.float32(0.6611), np.float32(0.5971), np.float32(0.1706), np.float32(0.0), np.float32(0.6962), np.float32(0.5402), np.float32(0.7201), np.float32(0.579), np.float32(0.4658), np.float32(0.7599), np.float32(0.5144), np.float32(0.6112), np.float32(0.6726), np.float32(0.6725), np.float32(0.295), np.float32(0.5381), np.float32(0.6344), np.float32(0.6264), np.float32(0.6505), np.float32(0.474), np.float32(0.7342), np.float32(0.5931), np.float32(0.7747), np.float32(0.7666)] 
+2026-10-18 10:51:08.693268: Epoch time: 109.17 s 
+2026-10-18 10:51:08.693613: Yayy! New best EMA pseudo Dice: 0.5916000008583069 
+2026-10-18 10:51:10.212659:  
+2026-10-18 10:51:10.214103: Epoch 91 
+2026-10-18 10:51:10.214651: Current learning rate: 0.00918 
+2026-10-18 10:52:59.374079: train_loss -0.4859 
+2026-10-18 10:52:59.375663: val_loss -0.3652 
+2026-10-18 10:52:59.376347: Pseudo dice [np.float32(0.9284), np.float32(0.6133), np.float32(0.7703), np.float32(0.6729), np.float32(0.5274), np.float32(0.7895), np.float32(0.751), np.float32(0.4961), np.float32(0.0), np.float32(0.4919), np.float32(0.6938), np.float32(0.74), np.float32(0.607), np.float32(0.629), np.float32(0.4449), np.float32(0.0), np.float32(0.5396), np.float32(0.4571), np.float32(0.5663), np.float32(0.4226), np.float32(0.4143), np.float32(0.7465), np.float32(0.6373), np.float32(0.6144), np.float32(0.5891), np.float32(0.5703), np.float32(0.2132), np.float32(0.5861), np.float32(0.5696), np.float32(0.7159), np.float32(0.5557), np.float32(0.4626), np.float32(0.5985), np.float32(0.5964), np.float32(0.6656), np.float32(0.676)] 
+2026-10-18 10:52:59.376809: Epoch time: 109.16 s 
+2026-10-18 10:53:00.088995:  
+2026-10-18 10:53:00.089569: Epoch 92 
+2026-10-18 10:53:00.089833: Current learning rate: 0.00917 
+2026-10-18 10:54:49.205735: train_loss -0.4624 
+2026-10-18 10:54:49.206406: val_loss -0.4061 
+2026-10-18 10:54:49.207072: Pseudo dice [np.float32(0.9259), np.float32(0.709), np.float32(0.8672), np.float32(0.8821), np.float32(0.7666), np.float32(0.9219), np.float32(0.8134), np.float32(0.5114), np.float32(0.0), np.float32(0.4043), np.float32(0.7417), np.float32(0.7487), np.float32(0.5535), np.float32(0.2988), np.float32(0.0697), np.float32(0.0), np.float32(0.7795), np.float32(0.4588), np.float32(0.808), np.float32(0.5402), np.float32(0.5252), np.float32(0.7578), np.float32(0.629), np.float32(0.6319), np.float32(0.7244), np.float32(0.7486), np.float32(0.4767), np.float32(0.7838), np.float32(0.4888), np.float32(0.6375), np.float32(0.6695), np.float32(0.455), np.float32(0.8237), np.float32(0.5672), np.float32(0.8845), np.float32(0.9223)] 
+2026-10-18 10:54:49.207563: Epoch time: 109.12 s 
+2026-10-18 10:54:49.207874: Yayy! New best EMA pseudo Dice: 0.5927000045776367 
+2026-10-18 10:54:50.661256:  
+2026-10-18 10:54:50.661923: Epoch 93 
+2026-10-18 10:54:50.662302: Current learning rate: 0.00916 
+2026-10-18 10:56:39.718695: train_loss -0.4858 
+2026-10-18 10:56:39.719425: val_loss -0.4206 
+2026-10-18 10:56:39.720086: Pseudo dice [np.float32(0.9386), np.float32(0.7332), np.float32(0.8627), np.float32(0.8604), np.float32(0.7045), np.float32(0.9095), np.float32(0.8168), np.float32(0.4881), np.float32(0.0), np.float32(0.5076), np.float32(0.7634), np.float32(0.7934), np.float32(0.5909), np.float32(0.3597), np.float32(0.0893), np.float32(0.0), np.float32(0.6997), np.float32(0.4759), np.float32(0.7674), np.float32(0.6073), np.float32(0.4477), np.float32(0.7439), np.float32(0.7379), np.float32(0.8037), np.float32(0.692), np.float32(0.6492), np.float32(0.4728), np.float32(0.7623), np.float32(0.6944), np.float32(0.7123), np.float32(0.6329), np.float32(0.5082), np.float32(0.7587), np.float32(0.6778), np.float32(0.9176), np.float32(0.9312)] 
+2026-10-18 10:56:39.720504: Epoch time: 109.06 s 
+2026-10-18 10:56:39.720951: Yayy! New best EMA pseudo Dice: 0.597599983215332 
+2026-10-18 10:56:40.948484:  
+2026-10-18 10:56:40.949181: Epoch 94 
+2026-10-18 10:56:40.949475: Current learning rate: 0.00915 
+2026-10-18 10:58:30.096716: train_loss -0.4931 
+2026-10-18 10:58:30.097386: val_loss -0.4426 
+2026-10-18 10:58:30.098139: Pseudo dice [np.float32(0.9089), np.float32(0.6993), np.float32(0.8808), np.float32(0.8655), np.float32(0.7142), np.float32(0.9194), np.float32(0.8014), np.float32(0.6169), np.float32(0.0), np.float32(0.4775), np.float32(0.7775), np.float32(0.8213), np.float32(0.658), np.float32(0.4542), np.float32(0.5609), np.float32(0.0), np.float32(0.7353), np.float32(0.3595), np.float32(0.7739), np.float32(0.4846), np.float32(0.3493), np.float32(0.6136), np.float32(0.6399), np.float32(0.7096), np.float32(0.7154), np.float32(0.7159), np.float32(0.4467), np.float32(0.7747), np.float32(0.3298), np.float32(0.5815), np.float32(0.6348), np.float32(0.5366), np.float32(0.7834), np.float32(0.6946), np.float32(0.8422), np.float32(0.9098)] 
+2026-10-18 10:58:30.098745: Epoch time: 109.15 s 
+2026-10-18 10:58:30.099206: Yayy! New best EMA pseudo Dice: 0.6011000275611877 
+2026-10-18 10:58:31.331252:  
+2026-10-18 10:58:31.331826: Epoch 95 
+2026-10-18 10:58:31.332100: Current learning rate: 0.00914 
+2026-10-18 11:00:20.417785: train_loss -0.4852 
+2026-10-18 11:00:20.418433: val_loss -0.4345 
+2026-10-18 11:00:20.419162: Pseudo dice [np.float32(0.9389), np.float32(0.7738), np.float32(0.8666), np.float32(0.8905), np.float32(0.6893), np.float32(0.9227), np.float32(0.8223), np.float32(0.498), np.float32(0.0022), np.float32(0.451), np.float32(0.7681), np.float32(0.7845), np.float32(0.6679), np.float32(0.5261), np.float32(0.0416), np.float32(0.0), np.float32(0.7502), np.float32(0.4414), np.float32(0.7978), np.float32(0.613), np.float32(0.4151), np.float32(0.7701), np.float32(0.8021), np.float32(0.8297), np.float32(0.7921), np.float32(0.697), np.float32(0.4682), np.float32(0.8349), np.float32(0.6211), np.float32(0.7191), np.float32(0.7207), np.float32(0.4345), np.float32(0.8104), np.float32(0.6991), np.float32(0.944), np.float32(0.9524)] 
+2026-10-18 11:00:20.419577: Epoch time: 109.09 s 
+2026-10-18 11:00:20.419942: Yayy! New best EMA pseudo Dice: 0.6069999933242798 
+2026-10-18 11:00:21.641912:  
+2026-10-18 11:00:21.642651: Epoch 96 
+2026-10-18 11:00:21.642988: Current learning rate: 0.00913 
+2026-10-18 11:02:10.752823: train_loss -0.5209 
+2026-10-18 11:02:10.753946: val_loss -0.4624 
+2026-10-18 11:02:10.754372: Pseudo dice [np.float32(0.9608), np.float32(0.7253), np.float32(0.9007), np.float32(0.9079), np.float32(0.7397), np.float32(0.9255), np.float32(0.867), np.float32(0.5224), np.float32(0.0018), np.float32(0.4589), np.float32(0.8048), np.float32(0.8207), np.float32(0.6834), np.float32(0.5137), np.float32(0.5811), np.float32(0.0), np.float32(0.7946), np.float32(0.5213), np.float32(0.8236), np.float32(0.5722), np.float32(0.3558), np.float32(0.6953), np.float32(0.8468), np.float32(0.8701), np.float32(0.7404), np.float32(0.7574), np.float32(0.4676), np.float32(0.8877), np.float32(0.6416), np.float32(0.722), np.float32(0.6422), np.float32(0.5116), np.float32(0.8355), np.float32(0.6634), np.float32(0.9661), np.float32(0.9721)] 
+2026-10-18 11:02:10.754884: Epoch time: 109.11 s 
+2026-10-18 11:02:10.755249: Yayy! New best EMA pseudo Dice: 0.6148999929428101 
+2026-10-18 11:02:11.989228:  
+2026-10-18 11:02:11.989713: Epoch 97 
+2026-10-18 11:02:11.989863: Current learning rate: 0.00912 
+2026-10-18 11:04:01.084850: train_loss -0.5088 
+2026-10-18 11:04:01.085635: val_loss -0.385 
+2026-10-18 11:04:01.086441: Pseudo dice [np.float32(0.9272), np.float32(0.6894), np.float32(0.797), np.float32(0.7587), np.float32(0.5014), np.float32(0.8876), np.float32(0.7056), np.float32(0.5966), np.float32(0.0), np.float32(0.4183), np.float32(0.7495), np.float32(0.7963), np.float32(0.6856), np.float32(0.461), np.float32(0.3836), np.float32(0.0), np.float32(0.7233), np.float32(0.5784), np.float32(0.7137), np.float32(0.6317), np.float32(0.5184), np.float32(0.7099), np.float32(0.6851), np.float32(0.7652), np.float32(0.6378), np.float32(0.6297), np.float32(0.3795), np.float32(0.742), np.float32(0.5256), np.float32(0.7233), np.float32(0.6067), np.float32(0.4188), np.float32(0.7552), np.float32(0.6939), np.float32(0.8471), np.float32(0.9118)] 
+2026-10-18 11:04:01.086850: Epoch time: 109.1 s 
+2026-10-18 11:04:01.087215: Yayy! New best EMA pseudo Dice: 0.616100013256073 
+2026-10-18 11:04:05.804944:  
+2026-10-18 11:04:05.805712: Epoch 98 
+2026-10-18 11:04:05.806249: Current learning rate: 0.00911 
+2026-10-18 11:05:54.926946: train_loss -0.5206 
+2026-10-18 11:05:54.927492: val_loss -0.3895 
+2026-10-18 11:05:54.928350: Pseudo dice [np.float32(0.9252), np.float32(0.6702), np.float32(0.8472), np.float32(0.8392), np.float32(0.7393), np.float32(0.8938), np.float32(0.8386), np.float32(0.5234), np.float32(0.0), np.float32(0.4957), np.float32(0.7768), np.float32(0.8109), np.float32(0.653), np.float32(0.588), np.float32(0.2544), np.float32(0.0), np.float32(0.7043), np.float32(0.3709), np.float32(0.7848), np.float32(0.5905), np.float32(0.4889), np.float32(0.649), np.float32(0.6549), np.float32(0.7257), np.float32(0.7163), np.float32(0.7026), np.float32(0.4741), np.float32(0.7205), np.float32(0.5484), np.float32(0.5487), np.float32(0.5961), np.float32(0.4869), np.float32(0.7576), np.float32(0.6858), np.float32(0.8723), np.float32(0.9003)] 
+2026-10-18 11:05:54.929186: Epoch time: 109.12 s 
+2026-10-18 11:05:54.929505: Yayy! New best EMA pseudo Dice: 0.617900013923645 
+2026-10-18 11:05:56.140217:  
+2026-10-18 11:05:56.140937: Epoch 99 
+2026-10-18 11:05:56.141229: Current learning rate: 0.0091 
+2026-10-18 11:07:45.228977: train_loss -0.5268 
+2026-10-18 11:07:45.229786: val_loss -0.4218 
+2026-10-18 11:07:45.230261: Pseudo dice [np.float32(0.9347), np.float32(0.7145), np.float32(0.8441), np.float32(0.8359), np.float32(0.7225), np.float32(0.8726), np.float32(0.774), np.float32(0.5352), np.float32(0.0), np.float32(0.505), np.float32(0.7463), np.float32(0.8026), np.float32(0.654), np.float32(0.5539), np.float32(0.0543), np.float32(0.0), np.float32(0.776), np.float32(0.526), np.float32(0.7556), np.float32(0.5814), np.float32(0.348), np.float32(0.6984), np.float32(0.6103), np.float32(0.6034), np.float32(0.6949), np.float32(0.73), np.float32(0.4505), np.float32(0.8019), np.float32(0.5637), np.float32(0.5955), np.float32(0.553), np.float32(0.4802), np.float32(0.7868), np.float32(0.6323), np.float32(0.8934), np.float32(0.9009)] 
+2026-10-18 11:07:45.230736: Epoch time: 109.09 s 
+2026-10-18 11:07:46.004167: Yayy! New best EMA pseudo Dice: 0.6187000274658203 
+2026-10-18 11:07:47.766881:  
+2026-10-18 11:07:47.767555: Epoch 100 
+2026-10-18 11:07:47.767867: Current learning rate: 0.0091 
+2026-10-18 11:09:36.936120: train_loss -0.5113 
+2026-10-18 11:09:36.937085: val_loss -0.3721 
+2026-10-18 11:09:36.937650: Pseudo dice [np.float32(0.9242), np.float32(0.705), np.float32(0.8467), np.float32(0.8436), np.float32(0.6844), np.float32(0.9119), np.float32(0.8136), np.float32(0.6055), np.float32(0.0), np.float32(0.4477), np.float32(0.7774), np.float32(0.8289), np.float32(0.6734), np.float32(0.5459), np.float32(0.3703), np.float32(0.0), np.float32(0.7045), np.float32(0.3401), np.float32(0.7489), np.float32(0.489), np.float32(0.4695), np.float32(0.7041), np.float32(0.4298), np.float32(0.4574), np.float32(0.7025), np.float32(0.6381), np.float32(0.2439), np.float32(0.391), np.float32(0.4467), np.float32(0.6123), np.float32(0.5815), np.float32(0.5021), np.float32(0.7556), np.float32(0.6305), np.float32(0.671), np.float32(0.7111)] 
+2026-10-18 11:09:36.938052: Epoch time: 109.17 s 
+2026-10-18 11:09:37.683757:  
+2026-10-18 11:09:37.684454: Epoch 101 
+2026-10-18 11:09:37.684880: Current learning rate: 0.00909 
+2026-10-18 11:11:26.823143: train_loss -0.4642 
+2026-10-18 11:11:26.823596: val_loss -0.4197 
+2026-10-18 11:11:26.823928: Pseudo dice [np.float32(0.9329), np.float32(0.7416), np.float32(0.8347), np.float32(0.8387), np.float32(0.576), np.float32(0.9053), np.float32(0.7936), np.float32(0.5533), np.float32(0.0), np.float32(0.3842), np.float32(0.7768), np.float32(0.7928), np.float32(0.6201), np.float32(0.5315), np.float32(0.2678), np.float32(0.0), np.float32(0.6569), np.float32(0.3698), np.float32(0.704), np.float32(0.4444), np.float32(0.3483), np.float32(0.6319), np.float32(0.6258), np.float32(0.7578), np.float32(0.6872), np.float32(0.6754), np.float32(0.4533), np.float32(0.7682), np.float32(0.6932), np.float32(0.7331), np.float32(0.6682), np.float32(0.4119), np.float32(0.7814), np.float32(0.6806), np.float32(0.8718), np.float32(0.9169)] 
+2026-10-18 11:11:26.824482: Epoch time: 109.14 s 
+2026-10-18 11:11:27.533345:  
+2026-10-18 11:11:27.534102: Epoch 102 
+2026-10-18 11:11:27.534536: Current learning rate: 0.00908 
+2026-10-18 11:13:16.624532: train_loss -0.5015 
+2026-10-18 11:13:16.625256: val_loss -0.3614 
+2026-10-18 11:13:16.626011: Pseudo dice [np.float32(0.9318), np.float32(0.5877), np.float32(0.7858), np.float32(0.7342), np.float32(0.6805), np.float32(0.8488), np.float32(0.8125), np.float32(0.6192), np.float32(0.0), np.float32(0.4651), np.float32(0.7066), np.float32(0.7638), np.float32(0.6819), np.float32(0.5655), np.float32(0.3931), np.float32(0.0), np.float32(0.6985), np.float32(0.5043), np.float32(0.7708), np.float32(0.6276), np.float32(0.3539), np.float32(0.6776), np.float32(0.5723), np.float32(0.5781), np.float32(0.5678), np.float32(0.537), np.float32(0.1353), np.float32(0.3812), np.float32(0.3618), np.float32(0.5638), np.float32(0.6251), np.float32(0.4274), np.float32(0.7145), np.float32(0.6281), np.float32(0.646), np.float32(0.7088)] 
+2026-10-18 11:13:16.626314: Epoch time: 109.09 s 
+2026-10-18 11:13:17.364748:  
+2026-10-18 11:13:17.365415: Epoch 103 
+2026-10-18 11:13:17.365690: Current learning rate: 0.00907 
+2026-10-18 11:15:06.517877: train_loss -0.4957 
+2026-10-18 11:15:06.518512: val_loss -0.3906 
+2026-10-18 11:15:06.519521: Pseudo dice [np.float32(0.9245), np.float32(0.7248), np.float32(0.8687), np.float32(0.8912), np.float32(0.6343), np.float32(0.936), np.float32(0.8217), np.float32(0.6405), np.float32(0.0006), np.float32(0.3927), np.float32(0.7902), np.float32(0.8391), np.float32(0.6465), np.float32(0.5622), np.float32(0.4895), np.float32(0.0), np.float32(0.7504), np.float32(0.5323), np.float32(0.792), np.float32(0.5792), np.float32(0.3313), np.float32(0.6774), np.float32(0.4802), np.float32(0.5123), np.float32(0.7541), np.float32(0.7105), np.float32(0.4816), np.float32(0.7857), np.float32(0.4193), np.float32(0.5137), np.float32(0.5878), np.float32(0.5001), np.float32(0.8046), np.float32(0.7114), np.float32(0.8503), np.float32(0.8883)] 
+2026-10-18 11:15:06.519804: Epoch time: 109.16 s 
+2026-10-18 11:15:07.244534:  
+2026-10-18 11:15:07.245223: Epoch 104 
+2026-10-18 11:15:07.245429: Current learning rate: 0.00906 
+2026-10-18 11:16:56.403344: train_loss -0.4917 
+2026-10-18 11:16:56.404448: val_loss -0.3631 
+2026-10-18 11:16:56.405434: Pseudo dice [np.float32(0.9196), np.float32(0.6624), np.float32(0.7983), np.float32(0.766), np.float32(0.6629), np.float32(0.858), np.float32(0.8035), np.float32(0.5197), np.float32(0.0015), np.float32(0.47), np.float32(0.6171), np.float32(0.71), np.float32(0.6213), np.float32(0.5561), np.float32(0.5353), np.float32(0.0), np.float32(0.5405), np.float32(0.3558), np.float32(0.6119), np.float32(0.3567), np.float32(0.433), np.float32(0.7191), np.float32(0.5623), np.float32(0.6721), np.float32(0.6363), np.float32(0.599), np.float32(0.521), np.float32(0.7086), np.float32(0.5763), np.float32(0.6517), np.float32(0.5082), np.float32(0.4168), np.float32(0.6658), np.float32(0.6293), np.float32(0.8872), np.float32(0.9173)] 
+2026-10-18 11:16:56.405848: Epoch time: 109.16 s 
+2026-10-18 11:16:57.115680:  
+2026-10-18 11:16:57.116283: Epoch 105 
+2026-10-18 11:16:57.116578: Current learning rate: 0.00905 
+2026-10-18 11:18:46.262599: train_loss -0.5087 
+2026-10-18 11:18:46.263075: val_loss -0.396 
+2026-10-18 11:18:46.263608: Pseudo dice [np.float32(0.8805), np.float32(0.7056), np.float32(0.8239), np.float32(0.8568), np.float32(0.6896), np.float32(0.9112), np.float32(0.7945), np.float32(0.5979), np.float32(0.0012), np.float32(0.4787), np.float32(0.7123), np.float32(0.736), np.float32(0.5258), np.float32(0.4141), np.float32(0.4691), np.float32(0.0), np.float32(0.6951), np.float32(0.3208), np.float32(0.7261), np.float32(0.5532), np.float32(0.3667), np.float32(0.6994), np.float32(0.5638), np.float32(0.6346), np.float32(0.6923), np.float32(0.674), np.float32(0.4561), np.float32(0.7087), np.float32(0.5385), np.float32(0.607), np.float32(0.7049), np.float32(0.4148), np.float32(0.7395), np.float32(0.6762), np.float32(0.912), np.float32(0.93)] 
+2026-10-18 11:18:46.264326: Epoch time: 109.15 s 
+2026-10-18 11:18:46.986408:  
+2026-10-18 11:18:46.987204: Epoch 106 
+2026-10-18 11:18:46.987576: Current learning rate: 0.00904 
+2026-10-18 11:20:36.103374: train_loss -0.5205 
+2026-10-18 11:20:36.103894: val_loss -0.3993 
+2026-10-18 11:20:36.104885: Pseudo dice [np.float32(0.9127), np.float32(0.7645), np.float32(0.8728), np.float32(0.8884), np.float32(0.8236), np.float32(0.9302), np.float32(0.8482), np.float32(0.6461), np.float32(0.0003), np.float32(0.3605), np.float32(0.7624), np.float32(0.7958), np.float32(0.6875), np.float32(0.4993), np.float32(0.4265), np.float32(0.0), np.float32(0.7912), np.float32(0.5458), np.float32(0.7968), np.float32(0.6317), np.float32(0.4617), np.float32(0.7787), np.float32(0.7094), np.float32(0.781), np.float32(0.7442), np.float32(0.7169), np.float32(0.4363), np.float32(0.8459), np.float32(0.615), np.float32(0.5421), np.float32(0.699), np.float32(0.4404), np.float32(0.7921), np.float32(0.7065), np.float32(0.9154), np.float32(0.9477)] 
+2026-10-18 11:20:36.105447: Epoch time: 109.12 s 
+2026-10-18 11:20:36.105804: Yayy! New best EMA pseudo Dice: 0.6187000274658203 
+2026-10-18 11:20:39.352837:  
+2026-10-18 11:20:39.353493: Epoch 107 
+2026-10-18 11:20:39.353782: Current learning rate: 0.00903 
+2026-10-18 11:22:28.461826: train_loss -0.5184 
+2026-10-18 11:22:28.462468: val_loss -0.3804 
+2026-10-18 11:22:28.463337: Pseudo dice [np.float32(0.8999), np.float32(0.7294), np.float32(0.7852), np.float32(0.835), np.float32(0.5182), np.float32(0.8976), np.float32(0.7968), np.float32(0.5834), np.float32(0.0002), np.float32(0.4567), np.float32(0.7741), np.float32(0.7513), np.float32(0.7054), np.float32(0.5903), np.float32(0.4313), np.float32(0.0), np.float32(0.7237), np.float32(0.4462), np.float32(0.7707), np.float32(0.5733), np.float32(0.3629), np.float32(0.6668), np.float32(0.5386), np.float32(0.5393), np.float32(0.6983), np.float32(0.6998), np.float32(0.3489), np.float32(0.5586), np.float32(0.616), np.float32(0.5592), np.float32(0.4964), np.float32(0.347), np.float32(0.6955), np.float32(0.5801), np.float32(0.8106), np.float32(0.8511)] 
+2026-10-18 11:22:28.463645: Epoch time: 109.11 s 
+2026-10-18 11:22:29.182015:  
+2026-10-18 11:22:29.182752: Epoch 108 
+2026-10-18 11:22:29.183084: Current learning rate: 0.00902 
+2026-10-18 11:24:18.323895: train_loss -0.4965 
+2026-10-18 11:24:18.324566: val_loss -0.4227 
+2026-10-18 11:24:18.325358: Pseudo dice [np.float32(0.9378), np.float32(0.6586), np.float32(0.8328), np.float32(0.8767), np.float32(0.7348), np.float32(0.9125), np.float32(0.8412), np.float32(0.5963), np.float32(1e-04), np.float32(0.3427), np.float32(0.7773), np.float32(0.752), np.float32(0.6966), np.float32(0.5841), np.float32(0.4619), np.float32(0.0), np.float32(0.714), np.float32(0.4059), np.float32(0.6764), np.float32(0.4437), np.float32(0.305), np.float32(0.5966), np.float32(0.714), np.float32(0.807), np.float32(0.6702), np.float32(0.67), np.float32(0.4713), np.float32(0.8295), np.float32(0.537), np.float32(0.5961), np.float32(0.65), np.float32(0.4624), np.float32(0.7049), np.float32(0.7007), np.float32(0.8965), np.float32(0.9308)] 
+2026-10-18 11:24:18.325890: Epoch time: 109.14 s 
+2026-10-18 11:24:19.058058:  
+2026-10-18 11:24:19.058783: Epoch 109 
+2026-10-18 11:24:19.059169: Current learning rate: 0.00901 
+2026-10-18 11:26:08.212390: train_loss -0.5108 
+2026-10-18 11:26:08.213336: val_loss -0.4154 
+2026-10-18 11:26:08.214041: Pseudo dice [np.float32(0.8985), np.float32(0.7409), np.float32(0.8241), np.float32(0.8352), np.float32(0.6481), np.float32(0.8873), np.float32(0.7999), np.float32(0.5833), np.float32(0.0), np.float32(0.4626), np.float32(0.7344), np.float32(0.6943), np.float32(0.6437), np.float32(0.5104), np.float32(0.4996), np.float32(0.0), np.float32(0.6925), np.float32(0.4351), np.float32(0.7071), np.float32(0.4116), np.float32(0.2978), np.float32(0.6227), np.float32(0.6784), np.float32(0.7942), np.float32(0.7163), np.float32(0.7016), np.float32(0.4763), np.float32(0.7158), np.float32(0.612), np.float32(0.7176), np.float32(0.5774), np.float32(0.3981), np.float32(0.7323), np.float32(0.6787), np.float32(0.8824), np.float32(0.9069)] 
+2026-10-18 11:26:08.214547: Epoch time: 109.16 s 
+2026-10-18 11:26:08.214877: Yayy! New best EMA pseudo Dice: 0.6191999912261963 
+2026-10-18 11:26:11.871690:  
+2026-10-18 11:26:11.872395: Epoch 110 
+2026-10-18 11:26:11.872731: Current learning rate: 0.009 
+2026-10-18 11:28:01.010001: train_loss -0.5177 
+2026-10-18 11:28:01.010738: val_loss -0.4221 
+2026-10-18 11:28:01.011511: Pseudo dice [np.float32(0.9324), np.float32(0.7393), np.float32(0.8407), np.float32(0.8328), np.float32(0.6774), np.float32(0.8864), np.float32(0.7879), np.float32(0.5414), np.float32(0.0), np.float32(0.3805), np.float32(0.7616), np.float32(0.7646), np.float32(0.6172), np.float32(0.0496), np.float32(0.5396), np.float32(0.0), np.float32(0.6757), np.float32(0.2424), np.float32(0.7078), np.float32(0.4791), np.float32(0.4734), np.float32(0.6393), np.float32(0.782), np.float32(0.8185), np.float32(0.7372), np.float32(0.7353), np.float32(0.6138), np.float32(0.8599), np.float32(0.643), np.float32(0.7008), np.float32(0.6829), np.float32(0.4716), np.float32(0.7936), np.float32(0.5402), np.float32(0.9235), np.float32(0.9495)] 
+2026-10-18 11:28:01.011951: Epoch time: 109.14 s 
+2026-10-18 11:28:01.012412: Yayy! New best EMA pseudo Dice: 0.6207000017166138 
+2026-10-18 11:28:02.256288:  
+2026-10-18 11:28:02.257045: Epoch 111 
+2026-10-18 11:28:02.257314: Current learning rate: 0.009 
+2026-10-18 11:29:51.334691: train_loss -0.5296 
+2026-10-18 11:29:51.335218: val_loss -0.4211 
+2026-10-18 11:29:51.335724: Pseudo dice [np.float32(0.9318), np.float32(0.6874), np.float32(0.8497), np.float32(0.8058), np.float32(0.5548), np.float32(0.9181), np.float32(0.7864), np.float32(0.5315), np.float32(0.0003), np.float32(0.4305), np.float32(0.7826), np.float32(0.8454), np.float32(0.5705), np.float32(0.6054), np.float32(0.422), np.float32(0.0), np.float32(0.672), np.float32(0.491), np.float32(0.7812), np.float32(0.6483), np.float32(0.2845), np.float32(0.7204), np.float32(0.6208), np.float32(0.7624), np.float32(0.6361), np.float32(0.6925), np.float32(0.4105), np.float32(0.8065), np.float32(0.498), np.float32(0.6529), np.float32(0.4754), np.float32(0.5887), np.float32(0.8), np.float32(0.7124), np.float32(0.9053), np.float32(0.9442)] 
+2026-10-18 11:29:51.336420: Epoch time: 109.08 s 
+2026-10-18 11:29:51.336659: Yayy! New best EMA pseudo Dice: 0.621999979019165 
+2026-10-18 11:29:52.581011:  
+2026-10-18 11:29:52.581557: Epoch 112 
+2026-10-18 11:29:52.582179: Current learning rate: 0.00899 
+2026-10-18 11:31:41.786009: train_loss -0.4536 
+2026-10-18 11:31:41.786729: val_loss -0.3705 
+2026-10-18 11:31:41.787411: Pseudo dice [np.float32(0.9041), np.float32(0.5699), np.float32(0.7365), np.float32(0.783), np.float32(0.7057), np.float32(0.8593), np.float32(0.8303), np.float32(0.4132), np.float32(0.0008), np.float32(0.4452), np.float32(0.7595), np.float32(0.765), np.float32(0.4498), np.float32(0.2305), np.float32(0.5605), np.float32(0.0), np.float32(0.6712), np.float32(0.3492), np.float32(0.7263), np.float32(0.5192), np.float32(0.2922), np.float32(0.6337), np.float32(0.3972), np.float32(0.4402), np.float32(0.5846), np.float32(0.5824), np.float32(0.1704), np.float32(0.6805), np.float32(0.3118), np.float32(0.3929), np.float32(0.5401), np.float32(0.4311), np.float32(0.7906), np.float32(0.6471), np.float32(0.7885), np.float32(0.8341)] 
+2026-10-18 11:31:41.787784: Epoch time: 109.21 s 
+2026-10-18 11:31:42.510514:  
+2026-10-18 11:31:42.511247: Epoch 113 
+2026-10-18 11:31:42.511534: Current learning rate: 0.00898 
+2026-10-18 11:33:31.666664: train_loss -0.4468 
+2026-10-18 11:33:31.667440: val_loss -0.315 
+2026-10-18 11:33:31.667995: Pseudo dice [np.float32(0.9096), np.float32(0.4843), np.float32(0.6977), np.float32(0.6446), np.float32(0.5925), np.float32(0.6993), np.float32(0.7806), np.float32(0.2875), np.float32(0.0), np.float32(0.485), np.float32(0.6134), np.float32(0.6622), np.float32(0.496), np.float32(0.4147), np.float32(0.4209), np.float32(0.0), np.float32(0.6309), np.float32(0.2482), np.float32(0.6892), np.float32(0.4909), np.float32(0.4447), np.float32(0.6402), np.float32(0.5332), np.float32(0.5745), np.float32(0.4768), np.float32(0.4321), np.float32(0.2062), np.float32(0.4472), np.float32(0.2669), np.float32(0.4943), np.float32(0.5142), np.float32(0.3683), np.float32(0.6546), np.float32(0.4348), np.float32(0.6929), np.float32(0.7253)] 
+2026-10-18 11:33:31.668566: Epoch time: 109.16 s 
+2026-10-18 11:33:32.388241:  
+2026-10-18 11:33:32.388896: Epoch 114 
+2026-10-18 11:33:32.389221: Current learning rate: 0.00897 
+2026-10-18 11:35:21.496738: train_loss -0.4857 
+2026-10-18 11:35:21.497480: val_loss -0.3712 
+2026-10-18 11:35:21.498277: Pseudo dice [np.float32(0.9027), np.float32(0.6543), np.float32(0.7999), np.float32(0.8001), np.float32(0.6847), np.float32(0.856), np.float32(0.7395), np.float32(0.4631), np.float32(0.0), np.float32(0.0849), np.float32(0.7849), np.float32(0.7995), np.float32(0.5704), np.float32(0.5469), np.float32(0.5376), np.float32(0.0), np.float32(0.6361), np.float32(0.3584), np.float32(0.6616), np.float32(0.3911), np.float32(0.4506), np.float32(0.6067), np.float32(0.3628), np.float32(0.5596), np.float32(0.6364), np.float32(0.5891), np.float32(0.1471), np.float32(0.5684), np.float32(0.4187), np.float32(0.5007), np.float32(0.6088), np.float32(0.3826), np.float32(0.6989), np.float32(0.5475), np.float32(0.6824), np.float32(0.7487)] 
+2026-10-18 11:35:21.498549: Epoch time: 109.11 s 
+2026-10-18 11:35:22.280676:  
+2026-10-18 11:35:22.281325: Epoch 115 
+2026-10-18 11:35:22.281521: Current learning rate: 0.00896 
+2026-10-18 11:37:11.409908: train_loss -0.4838 
+2026-10-18 11:37:11.410800: val_loss -0.3749 
+2026-10-18 11:37:11.411343: Pseudo dice [np.float32(0.9251), np.float32(0.6336), np.float32(0.7392), np.float32(0.8029), np.float32(0.5957), np.float32(0.8698), np.float32(0.776), np.float32(0.5194), np.float32(0.0026), np.float32(0.0668), np.float32(0.7441), np.float32(0.7379), np.float32(0.5696), np.float32(0.1942), np.float32(0.1902), np.float32(0.0), np.float32(0.6668), np.float32(0.2639), np.float32(0.701), np.float32(0.5243), np.float32(0.4128), np.float32(0.6376), np.float32(0.4395), np.float32(0.6488), np.float32(0.606), np.float32(0.6619), np.float32(0.3561), np.float32(0.699), np.float32(0.4671), np.float32(0.5738), np.float32(0.6879), np.float32(0.3921), np.float32(0.7547), np.float32(0.5349), np.float32(0.7783), np.float32(0.7797)] 
+2026-10-18 11:37:11.411790: Epoch time: 109.13 s 
+2026-10-18 11:37:12.133781:  
+2026-10-18 11:37:12.134378: Epoch 116 
+2026-10-18 11:37:12.134710: Current learning rate: 0.00895 
+2026-10-18 11:39:01.241156: train_loss -0.4796 
+2026-10-18 11:39:01.241661: val_loss -0.4123 
+2026-10-18 11:39:01.242459: Pseudo dice [np.float32(0.9247), np.float32(0.6376), np.float32(0.7548), np.float32(0.8156), np.float32(0.6627), np.float32(0.8699), np.float32(0.7654), np.float32(0.5003), np.float32(0.0), np.float32(0.1956), np.float32(0.6935), np.float32(0.6976), np.float32(0.5433), np.float32(0.3501), np.float32(0.6065), np.float32(0.0), np.float32(0.6958), np.float32(0.3844), np.float32(0.6786), np.float32(0.2708), np.float32(0.3594), np.float32(0.52), np.float32(0.7134), np.float32(0.8022), np.float32(0.6728), np.float32(0.7061), np.float32(0.3493), np.float32(0.7501), np.float32(0.4546), np.float32(0.6871), np.float32(0.4959), np.float32(0.4696), np.float32(0.7831), np.float32(0.6416), np.float32(0.8721), np.float32(0.9076)] 
+2026-10-18 11:39:01.243001: Epoch time: 109.11 s 
+2026-10-18 11:39:01.992602:  
+2026-10-18 11:39:01.993267: Epoch 117 
+2026-10-18 11:39:01.993582: Current learning rate: 0.00894 
+2026-10-18 11:40:51.149275: train_loss -0.5273 
+2026-10-18 11:40:51.150078: val_loss -0.4491 
+2026-10-18 11:40:51.150754: Pseudo dice [np.float32(0.9248), np.float32(0.7365), np.float32(0.8537), np.float32(0.8548), np.float32(0.6588), np.float32(0.9081), np.float32(0.7938), np.float32(0.553), np.float32(0.0), np.float32(0.4105), np.float32(0.7613), np.float32(0.7997), np.float32(0.6922), np.float32(0.5937), np.float32(0.1884), np.float32(0.0), np.float32(0.6978), np.float32(0.4143), np.float32(0.7528), np.float32(0.5348), np.float32(0.193), np.float32(0.6849), np.float32(0.6364), np.float32(0.738), np.float32(0.7387), np.float32(0.6706), np.float32(0.4144), np.float32(0.7962), np.float32(0.5103), np.float32(0.562), np.float32(0.5523), np.float32(0.4745), np.float32(0.6769), np.float32(0.5926), np.float32(0.7959), np.float32(0.8327)] 
+2026-10-18 11:40:51.151235: Epoch time: 109.16 s 
+2026-10-18 11:40:52.152370:  
+2026-10-18 11:40:52.152931: Epoch 118 
+2026-10-18 11:40:52.153185: Current learning rate: 0.00893 
+2026-10-18 11:42:41.326071: train_loss -0.5448 
+2026-10-18 11:42:41.326930: val_loss -0.4997 
+2026-10-18 11:42:41.327622: Pseudo dice [np.float32(0.9394), np.float32(0.7919), np.float32(0.8782), np.float32(0.8796), np.float32(0.642), np.float32(0.9224), np.float32(0.8562), np.float32(0.484), np.float32(0.0), np.float32(0.4545), np.float32(0.765), np.float32(0.7883), np.float32(0.6171), np.float32(0.4742), np.float32(0.0), np.float32(0.0), np.float32(0.6941), np.float32(0.4404), np.float32(0.7281), np.float32(0.2178), np.float32(0.3989), np.float32(0.6826), np.float32(0.7456), np.float32(0.818), np.float32(0.7617), np.float32(0.7599), np.float32(0.2904), np.float32(0.8523), np.float32(0.0504), np.float32(0.7225), np.float32(0.5488), np.float32(0.3901), np.float32(0.738), np.float32(0.6977), np.float32(0.9365), np.float32(0.9585)] 
+2026-10-18 11:42:41.328033: Epoch time: 109.18 s 
+2026-10-18 11:42:42.068076:  
+2026-10-18 11:42:42.068802: Epoch 119 
+2026-10-18 11:42:42.069211: Current learning rate: 0.00892 
+2026-10-18 11:44:31.271051: train_loss -0.5533 
+2026-10-18 11:44:31.271664: val_loss -0.4843 
+2026-10-18 11:44:31.272445: Pseudo dice [np.float32(0.9119), np.float32(0.7321), np.float32(0.8849), np.float32(0.8326), np.float32(0.6207), np.float32(0.9199), np.float32(0.8274), np.float32(0.403), np.float32(0.0), np.float32(0.2952), np.float32(0.7477), np.float32(0.7542), np.float32(0.6076), np.float32(0.3453), np.float32(0.0), np.float32(0.0), np.float32(0.7566), np.float32(0.589), np.float32(0.7651), np.float32(0.1335), np.float32(0.4779), np.float32(0.5559), np.float32(0.5469), np.float32(0.598), np.float32(0.7116), np.float32(0.6829), np.float32(0.3445), np.float32(0.7798), np.float32(0.2688), np.float32(0.6715), np.float32(0.5449), np.float32(0.5559), np.float32(0.7857), np.float32(0.6791), np.float32(0.9209), np.float32(0.9318)] 
+2026-10-18 11:44:31.273638: Epoch time: 109.21 s 
+2026-10-18 11:44:32.010115:  
+2026-10-18 11:44:32.010766: Epoch 120 
+2026-10-18 11:44:32.011075: Current learning rate: 0.00891 
+2026-10-18 11:46:21.212896: train_loss -0.5607 
+2026-10-18 11:46:21.213609: val_loss -0.4928 
+2026-10-18 11:46:21.214392: Pseudo dice [np.float32(0.911), np.float32(0.6641), np.float32(0.8642), np.float32(0.8803), np.float32(0.774), np.float32(0.9057), np.float32(0.8422), np.float32(0.4739), np.float32(0.0), np.float32(0.4582), np.float32(0.7589), np.float32(0.792), np.float32(0.6144), np.float32(0.457), np.float32(0.0), np.float32(0.0), np.float32(0.7647), np.float32(0.3178), np.float32(0.7752), np.float32(0.2723), np.float32(0.291), np.float32(0.7095), np.float32(0.7721), np.float32(0.742), np.float32(0.5946), np.float32(0.7655), np.float32(0.4349), np.float32(0.6457), np.float32(0.0), np.float32(0.6023), np.float32(0.319), np.float32(0.4591), np.float32(0.788), np.float32(0.6169), np.float32(0.8982), np.float32(0.9088)] 
+2026-10-18 11:46:21.214878: Epoch time: 109.2 s 
+2026-10-18 11:46:21.972457:  
+2026-10-18 11:46:21.973140: Epoch 121 
+2026-10-18 11:46:21.973504: Current learning rate: 0.0089 
+2026-10-18 11:48:11.112002: train_loss -0.5549 
+2026-10-18 11:48:11.112697: val_loss -0.4623 
+2026-10-18 11:48:11.113215: Pseudo dice [np.float32(0.9351), np.float32(0.7205), np.float32(0.8392), np.float32(0.839), np.float32(0.7325), np.float32(0.9098), np.float32(0.8527), np.float32(0.3898), np.float32(0.0), np.float32(0.5021), np.float32(0.7246), np.float32(0.7844), np.float32(0.5994), np.float32(0.3895), np.float32(0.0), np.float32(0.0), np.float32(0.7592), np.float32(0.4616), np.float32(0.7608), np.float32(0.4868), np.float32(0.3317), np.float32(0.5759), np.float32(0.5706), np.float32(0.5645), np.float32(0.6876), np.float32(0.6947), np.float32(0.3676), np.float32(0.5646), np.float32(0.0), np.float32(0.4737), np.float32(0.5243), np.float32(0.4882), np.float32(0.7726), np.float32(0.6985), np.float32(0.8048), np.float32(0.8532)] 
+2026-10-18 11:48:11.114075: Epoch time: 109.14 s 
+2026-10-18 11:48:11.839473:  
+2026-10-18 11:48:11.840188: Epoch 122 
+2026-10-18 11:48:11.840724: Current learning rate: 0.00889 
+2026-10-18 11:50:01.004153: train_loss -0.5612 
+2026-10-18 11:50:01.005092: val_loss -0.4559 
+2026-10-18 11:50:01.005962: Pseudo dice [np.float32(0.9069), np.float32(0.6374), np.float32(0.807), np.float32(0.7878), np.float32(0.5299), np.float32(0.8697), np.float32(0.7878), np.float32(0.5355), np.float32(0.0), np.float32(0.4686), np.float32(0.7219), np.float32(0.7643), np.float32(0.6554), np.float32(0.6653), np.float32(0.0), np.float32(0.0), np.float32(0.7223), np.float32(0.2959), np.float32(0.7185), np.float32(0.5721), np.float32(0.0002), np.float32(0.3743), np.float32(0.6331), np.float32(0.7013), np.float32(0.6018), np.float32(0.6148), np.float32(0.3872), np.float32(0.7242), np.float32(0.0), np.float32(0.0629), np.float32(0.071), np.float32(0.3401), np.float32(0.6377), np.float32(0.6233), np.float32(0.9001), np.float32(0.9159)] 
+2026-10-18 11:50:01.006373: Epoch time: 109.17 s 
+2026-10-18 11:50:01.766334:  
+2026-10-18 11:50:01.767006: Epoch 123 
+2026-10-18 11:50:01.767431: Current learning rate: 0.00889 
+2026-10-18 11:51:50.897328: train_loss -0.5674 
+2026-10-18 11:51:50.897925: val_loss -0.4897 
+2026-10-18 11:51:50.898460: Pseudo dice [np.float32(0.8865), np.float32(0.786), np.float32(0.8582), np.float32(0.9036), np.float32(0.6855), np.float32(0.9265), np.float32(0.7946), np.float32(0.2965), np.float32(0.0), np.float32(0.3835), np.float32(0.7847), np.float32(0.7894), np.float32(0.6327), np.float32(0.3463), np.float32(0.0), np.float32(0.0), np.float32(0.777), np.float32(0.3305), np.float32(0.7873), np.float32(0.5556), np.float32(0.5369), np.float32(0.664), np.float32(0.5922), np.float32(0.6967), np.float32(0.7396), np.float32(0.7517), np.float32(0.395), np.float32(0.784), np.float32(0.0), np.float32(0.0308), np.float32(0.74), np.float32(0.2656), np.float32(0.8185), np.float32(0.7432), np.float32(0.9206), np.float32(0.9491)] 
+2026-10-18 11:51:50.899136: Epoch time: 109.13 s 
+2026-10-18 11:51:51.629086:  
+2026-10-18 11:51:51.629813: Epoch 124 
+2026-10-18 11:51:51.630159: Current learning rate: 0.00888 
+2026-10-18 11:53:40.775341: train_loss -0.5667 
+2026-10-18 11:53:40.776286: val_loss -0.4753 
+2026-10-18 11:53:40.777259: Pseudo dice [np.float32(0.9109), np.float32(0.6755), np.float32(0.8512), np.float32(0.8611), np.float32(0.6529), np.float32(0.8977), np.float32(0.8054), np.float32(0.4458), np.float32(0.0), np.float32(0.2996), np.float32(0.7683), np.float32(0.7647), np.float32(0.6018), np.float32(0.4154), np.float32(0.0), np.float32(0.0), np.float32(0.6256), np.float32(0.456), np.float32(0.7851), np.float32(0.3602), np.float32(0.4444), np.float32(0.7103), np.float32(0.4486), np.float32(0.6486), np.float32(0.6998), np.float32(0.7049), np.float32(0.2672), np.float32(0.7999), np.float32(0.0), np.float32(0.4735), np.float32(0.5217), np.float32(0.3142), np.float32(0.7694), np.float32(0.6852), np.float32(0.8755), np.float32(0.8896)] 
+2026-10-18 11:53:40.777613: Epoch time: 109.15 s 
+2026-10-18 11:53:41.515920:  
+2026-10-18 11:53:41.516609: Epoch 125 
+2026-10-18 11:53:41.516884: Current learning rate: 0.00887 
+2026-10-18 11:55:30.637388: train_loss -0.5634 
+2026-10-18 11:55:30.637922: val_loss -0.4636 
+2026-10-18 11:55:30.638658: Pseudo dice [np.float32(0.8852), np.float32(0.6705), np.float32(0.8367), np.float32(0.8849), np.float32(0.6864), np.float32(0.925), np.float32(0.8233), np.float32(0.5154), np.float32(0.0), np.float32(0.1455), np.float32(0.7577), np.float32(0.7752), np.float32(0.6824), np.float32(0.5687), np.float32(0.0), np.float32(0.0), np.float32(0.7479), np.float32(0.5378), np.float32(0.7553), np.float32(0.4117), np.float32(0.2505), np.float32(0.5342), np.float32(0.6008), np.float32(0.6223), np.float32(0.6367), np.float32(0.6856), np.float32(0.4424), np.float32(0.8421), np.float32(0.0), np.float32(0.6751), np.float32(0.5445), np.float32(0.5028), np.float32(0.7898), np.float32(0.7573), np.float32(0.8391), np.float32(0.8903)] 
+2026-10-18 11:55:30.639267: Epoch time: 109.12 s 
+2026-10-18 11:55:31.378358:  
+2026-10-18 11:55:31.379055: Epoch 126 
+2026-10-18 11:55:31.379276: Current learning rate: 0.00886 
+2026-10-18 11:57:20.524007: train_loss -0.5565 
+2026-10-18 11:57:20.524699: val_loss -0.3376 
+2026-10-18 11:57:20.525069: Pseudo dice [np.float32(0.8729), np.float32(0.5892), np.float32(0.7218), np.float32(0.6865), np.float32(0.6027), np.float32(0.7063), np.float32(0.6705), np.float32(0.2328), np.float32(0.0), np.float32(0.5016), np.float32(0.6539), np.float32(0.6487), np.float32(0.5064), np.float32(0.2417), np.float32(0.0), np.float32(0.0), np.float32(0.676), np.float32(0.2432), np.float32(0.7235), np.float32(0.1471), np.float32(0.3037), np.float32(0.3919), np.float32(0.2378), np.float32(0.2257), np.float32(0.5208), np.float32(0.4627), np.float32(0.1649), np.float32(0.165), np.float32(0.0), np.float32(0.0015), np.float32(0.479), np.float32(0.2818), np.float32(0.5761), np.float32(0.4685), np.float32(0.4814), np.float32(0.5249)] 
+2026-10-18 11:57:20.525588: Epoch time: 109.15 s 
+2026-10-18 11:57:21.258341:  
+2026-10-18 11:57:21.259028: Epoch 127 
+2026-10-18 11:57:21.259203: Current learning rate: 0.00885 
+2026-10-18 11:59:10.398475: train_loss -0.5447 
+2026-10-18 11:59:10.399435: val_loss -0.4748 
+2026-10-18 11:59:10.399938: Pseudo dice [np.float32(0.9183), np.float32(0.7465), np.float32(0.8517), np.float32(0.8735), np.float32(0.6627), np.float32(0.9166), np.float32(0.8091), np.float32(0.5021), np.float32(0.0), np.float32(0.2547), np.float32(0.6756), np.float32(0.6062), np.float32(0.6168), np.float32(0.4025), np.float32(0.0), np.float32(0.0), np.float32(0.6855), np.float32(0.3633), np.float32(0.7707), np.float32(0.1004), np.float32(0.4493), np.float32(0.5721), np.float32(0.651), np.float32(0.7171), np.float32(0.6869), np.float32(0.7306), np.float32(0.2925), np.float32(0.6677), np.float32(0.0), np.float32(0.5175), np.float32(0.6622), np.float32(0.4642), np.float32(0.6299), np.float32(0.5736), np.float32(0.9143), np.float32(0.9408)] 
+2026-10-18 11:59:10.400422: Epoch time: 109.14 s 
+2026-10-18 11:59:11.169595:  
+2026-10-18 11:59:11.170304: Epoch 128 
+2026-10-18 11:59:11.170647: Current learning rate: 0.00884 
+2026-10-18 12:01:00.363936: train_loss -0.5609 
+2026-10-18 12:01:00.364700: val_loss -0.5082 
+2026-10-18 12:01:00.365611: Pseudo dice [np.float32(0.9334), np.float32(0.7698), np.float32(0.8434), np.float32(0.8733), np.float32(0.7286), np.float32(0.9145), np.float32(0.8406), np.float32(0.5639), np.float32(0.0), np.float32(0.4737), np.float32(0.7765), np.float32(0.8076), np.float32(0.6515), np.float32(0.5248), np.float32(0.0), np.float32(0.0), np.float32(0.6725), np.float32(0.5666), np.float32(0.7055), np.float32(0.3123), np.float32(0.3738), np.float32(0.3774), np.float32(0.82), np.float32(0.7502), np.float32(0.7669), np.float32(0.6982), np.float32(0.4074), np.float32(0.7847), np.float32(0.0), np.float32(0.5756), np.float32(0.6491), np.float32(0.422), np.float32(0.8026), np.float32(0.7094), np.float32(0.9197), np.float32(0.9487)] 
+2026-10-18 12:01:00.366031: Epoch time: 109.2 s 
+2026-10-18 12:01:01.121880:  
+2026-10-18 12:01:01.122769: Epoch 129 
+2026-10-18 12:01:01.123188: Current learning rate: 0.00883 
+2026-10-18 12:02:50.312372: train_loss -0.52 
+2026-10-18 12:02:50.312977: val_loss -0.4252 
+2026-10-18 12:02:50.313523: Pseudo dice [np.float32(0.8871), np.float32(0.6728), np.float32(0.7962), np.float32(0.708), np.float32(0.6232), np.float32(0.7998), np.float32(0.7642), np.float32(0.4939), np.float32(0.0), np.float32(0.3496), np.float32(0.6511), np.float32(0.6994), np.float32(0.5662), np.float32(0.4484), np.float32(0.0), np.float32(0.0), np.float32(0.7089), np.float32(0.5176), np.float32(0.7189), np.float32(0.5358), np.float32(0.396), np.float32(0.6287), np.float32(0.5187), np.float32(0.624), np.float32(0.6786), np.float32(0.5566), np.float32(0.3114), np.float32(0.7741), np.float32(0.0), np.float32(0.45), np.float32(0.3681), np.float32(0.2208), np.float32(0.6027), np.float32(0.5482), np.float32(0.7768), np.float32(0.832)] 
+2026-10-18 12:02:50.314727: Epoch time: 109.19 s 
+2026-10-18 12:02:51.053848:  
+2026-10-18 12:02:51.054611: Epoch 130 
+2026-10-18 12:02:51.054890: Current learning rate: 0.00882 
+2026-10-18 12:04:40.212880: train_loss -0.5294 
+2026-10-18 12:04:40.213515: val_loss -0.4165 
+2026-10-18 12:04:40.213987: Pseudo dice [np.float32(0.9161), np.float32(0.5745), np.float32(0.7694), np.float32(0.7207), np.float32(0.4259), np.float32(0.7932), np.float32(0.7197), np.float32(0.2374), np.float32(0.0), np.float32(0.314), np.float32(0.5748), np.float32(0.6279), np.float32(0.4831), np.float32(0.2187), np.float32(0.0), np.float32(0.0), np.float32(0.6091), np.float32(0.4643), np.float32(0.6728), np.float32(0.1746), np.float32(0.2773), np.float32(0.6425), np.float32(0.4528), np.float32(0.6687), np.float32(0.5991), np.float32(0.5677), np.float32(0.3136), np.float32(0.6677), np.float32(0.0), np.float32(0.5634), np.float32(0.4731), np.float32(0.5039), np.float32(0.6132), np.float32(0.6164), np.float32(0.7679), np.float32(0.8717)] 
+2026-10-18 12:04:40.214640: Epoch time: 109.16 s 
+2026-10-18 12:04:40.973256:  
+2026-10-18 12:04:40.974064: Epoch 131 
+2026-10-18 12:04:40.974404: Current learning rate: 0.00881 
+2026-10-18 12:06:30.133240: train_loss -0.5573 
+2026-10-18 12:06:30.134024: val_loss -0.4795 
+2026-10-18 12:06:30.134593: Pseudo dice [np.float32(0.937), np.float32(0.7239), np.float32(0.8396), np.float32(0.7732), np.float32(0.3072), np.float32(0.8747), np.float32(0.7568), np.float32(0.5736), np.float32(0.0), np.float32(0.32), np.float32(0.6976), np.float32(0.7808), np.float32(0.5851), np.float32(0.6266), np.float32(0.0), np.float32(0.0), np.float32(0.6491), np.float32(0.2518), np.float32(0.7419), np.float32(0.3403), np.float32(0.3136), np.float32(0.6838), np.float32(0.8056), np.float32(0.7974), np.float32(0.6848), np.float32(0.6705), np.float32(0.3712), np.float32(0.4046), np.float32(0.0), np.float32(0.5925), np.float32(0.5351), np.float32(0.4903), np.float32(0.6562), np.float32(0.6332), np.float32(0.8756), np.float32(0.9288)] 
+2026-10-18 12:06:30.135103: Epoch time: 109.16 s 
+2026-10-18 12:06:30.880903:  
+2026-10-18 12:06:30.881551: Epoch 132 
+2026-10-18 12:06:30.881948: Current learning rate: 0.0088 
+2026-10-18 12:08:20.003516: train_loss -0.5221 
+2026-10-18 12:08:20.004056: val_loss -0.3582 
+2026-10-18 12:08:20.004625: Pseudo dice [np.float32(0.9386), np.float32(0.5309), np.float32(0.6925), np.float32(0.6428), np.float32(0.3717), np.float32(0.725), np.float32(0.6887), np.float32(0.3026), np.float32(0.0), np.float32(0.1759), np.float32(0.5025), np.float32(0.5519), np.float32(0.2513), np.float32(0.2417), np.float32(0.0), np.float32(0.0), np.float32(0.4142), np.float32(0.2858), np.float32(0.5595), np.float32(0.3895), np.float32(0.2446), np.float32(0.7215), np.float32(0.5579), np.float32(0.4643), np.float32(0.5078), np.float32(0.5342), np.float32(0.1984), np.float32(0.493), np.float32(0.0), np.float32(0.4664), np.float32(0.496), np.float32(0.2722), np.float32(0.4454), np.float32(0.5428), np.float32(0.7045), np.float32(0.7266)] 
+2026-10-18 12:08:20.005672: Epoch time: 109.12 s 
+2026-10-18 12:08:20.760530:  
+2026-10-18 12:08:20.761160: Epoch 133 
+2026-10-18 12:08:20.761523: Current learning rate: 0.00879 
+2026-10-18 12:10:09.944912: train_loss -0.5295 
+2026-10-18 12:10:09.945445: val_loss -0.4332 
+2026-10-18 12:10:09.946459: Pseudo dice [np.float32(0.9242), np.float32(0.6773), np.float32(0.8052), np.float32(0.7983), np.float32(0.6979), np.float32(0.8646), np.float32(0.7566), np.float32(0.4744), np.float32(0.0017), np.float32(0.3229), np.float32(0.6851), np.float32(0.7402), np.float32(0.4896), np.float32(0.5312), np.float32(0.0), np.float32(0.0), np.float32(0.7235), np.float32(0.5219), np.float32(0.7126), np.float32(0.2704), np.float32(0.2358), np.float32(0.6483), np.float32(0.5308), np.float32(0.5111), np.float32(0.645), np.float32(0.6209), np.float32(0.2571), np.float32(0.7174), np.float32(0.0), np.float32(0.4571), np.float32(0.4435), np.float32(0.5112), np.float32(0.7112), np.float32(0.6479), np.float32(0.718), np.float32(0.7804)] 
+2026-10-18 12:10:09.946959: Epoch time: 109.19 s 
+2026-10-18 12:10:10.683373:  
+2026-10-18 12:10:10.683962: Epoch 134 
+2026-10-18 12:10:10.684281: Current learning rate: 0.00879 
+2026-10-18 12:11:59.828445: train_loss -0.5773 
+2026-10-18 12:11:59.829320: val_loss -0.4839 
+2026-10-18 12:11:59.829854: Pseudo dice [np.float32(0.9196), np.float32(0.7155), np.float32(0.8252), np.float32(0.8877), np.float32(0.771), np.float32(0.9223), np.float32(0.8298), np.float32(0.6149), np.float32(0.0), np.float32(0.4678), np.float32(0.7968), np.float32(0.8034), np.float32(0.5295), np.float32(0.3417), np.float32(0.0), np.float32(0.0), np.float32(0.713), np.float32(0.3921), np.float32(0.7384), np.float32(0.5248), np.float32(0.3727), np.float32(0.66), np.float32(0.2856), np.float32(0.5589), np.float32(0.6994), np.float32(0.6655), np.float32(0.3323), np.float32(0.7363), np.float32(0.0), np.float32(0.2457), np.float32(0.5903), np.float32(0.2783), np.float32(0.8091), np.float32(0.7106), np.float32(0.8291), np.float32(0.8583)] 
+2026-10-18 12:11:59.830148: Epoch time: 109.15 s 
+2026-10-18 12:12:00.597911:  
+2026-10-18 12:12:00.598625: Epoch 135 
+2026-10-18 12:12:00.598947: Current learning rate: 0.00878 
+2026-10-18 12:13:49.732037: train_loss -0.5822 
+2026-10-18 12:13:49.732598: val_loss -0.5072 
+2026-10-18 12:13:49.733127: Pseudo dice [np.float32(0.8962), np.float32(0.7269), np.float32(0.8612), np.float32(0.849), np.float32(0.6368), np.float32(0.8943), np.float32(0.8038), np.float32(0.5808), np.float32(0.0), np.float32(0.475), np.float32(0.7975), np.float32(0.822), np.float32(0.6643), np.float32(0.4615), np.float32(0.0), np.float32(0.0), np.float32(0.712), np.float32(0.3859), np.float32(0.7569), np.float32(0.4751), np.float32(0.3844), np.float32(0.5786), np.float32(0.7219), np.float32(0.765), np.float32(0.6759), np.float32(0.6897), np.float32(0.3109), np.float32(0.4835), np.float32(0.0), np.float32(0.5356), np.float32(0.2204), np.float32(0.4732), np.float32(0.5678), np.float32(0.6561), np.float32(0.871), np.float32(0.9105)] 
+2026-10-18 12:13:49.733464: Epoch time: 109.14 s 
+2026-10-18 12:13:50.476451:  
+2026-10-18 12:13:50.477120: Epoch 136 
+2026-10-18 12:13:50.477395: Current learning rate: 0.00877 
+2026-10-18 12:15:39.621397: train_loss -0.5709 
+2026-10-18 12:15:39.622258: val_loss -0.4773 
+2026-10-18 12:15:39.622986: Pseudo dice [np.float32(0.9157), np.float32(0.7653), np.float32(0.8516), np.float32(0.7855), np.float32(0.7037), np.float32(0.8254), np.float32(0.7483), np.float32(0.5259), np.float32(0.0), np.float32(0.417), np.float32(0.6888), np.float32(0.7394), np.float32(0.5723), np.float32(0.4909), np.float32(0.0), np.float32(0.0), np.float32(0.725), np.float32(0.5183), np.float32(0.725), np.float32(0.5021), np.float32(0.2815), np.float32(0.5674), np.float32(0.6821), np.float32(0.6809), np.float32(0.7344), np.float32(0.6619), np.float32(0.4422), np.float32(0.7336), np.float32(0.0), np.float32(0.6721), np.float32(0.5398), np.float32(0.2176), np.float32(0.6692), np.float32(0.647), np.float32(0.8083), np.float32(0.8102)] 
+2026-10-18 12:15:39.623384: Epoch time: 109.15 s 
+2026-10-18 12:15:40.368525:  
+2026-10-18 12:15:40.369337: Epoch 137 
+2026-10-18 12:15:40.369661: Current learning rate: 0.00876 
+2026-10-18 12:17:29.535407: train_loss -0.5897 
+2026-10-18 12:17:29.536000: val_loss -0.5349 
+2026-10-18 12:17:29.536889: Pseudo dice [np.float32(0.9423), np.float32(0.7212), np.float32(0.8626), np.float32(0.8926), np.float32(0.6856), np.float32(0.9047), np.float32(0.8594), np.float32(0.572), np.float32(0.0), np.float32(0.3644), np.float32(0.7786), np.float32(0.791), np.float32(0.6576), np.float32(0.4499), np.float32(0.0), np.float32(0.0), np.float32(0.7615), np.float32(0.4748), np.float32(0.8025), np.float32(0.5613), np.float32(0.3003), np.float32(0.6719), np.float32(0.7503), np.float32(0.7426), np.float32(0.7294), np.float32(0.726), np.float32(0.4933), np.float32(0.8011), np.float32(0.0), np.float32(0.6043), np.float32(0.6887), np.float32(0.4651), np.float32(0.8192), np.float32(0.7282), np.float32(0.9284), np.float32(0.9434)] 
+2026-10-18 12:17:29.537312: Epoch time: 109.17 s 
+2026-10-18 12:17:30.291738:  
+2026-10-18 12:17:30.292356: Epoch 138 
+2026-10-18 12:17:30.292774: Current learning rate: 0.00875 
+2026-10-18 12:19:19.420222: train_loss -0.6083 
+2026-10-18 12:19:19.421139: val_loss -0.4696 
+2026-10-18 12:19:19.421665: Pseudo dice [np.float32(0.9031), np.float32(0.6929), np.float32(0.7803), np.float32(0.7939), np.float32(0.5406), np.float32(0.9219), np.float32(0.8531), np.float32(0.4491), np.float32(0.0), np.float32(0.4231), np.float32(0.7301), np.float32(0.7893), np.float32(0.6811), np.float32(0.5394), np.float32(0.0), np.float32(0.0), np.float32(0.5677), np.float32(0.3899), np.float32(0.736), np.float32(0.5464), np.float32(0.4515), np.float32(0.6186), np.float32(0.3979), np.float32(0.5074), np.float32(0.652), np.float32(0.5973), np.float32(0.275), np.float32(0.6935), np.float32(0.0), np.float32(0.3485), np.float32(0.493), np.float32(0.3446), np.float32(0.7536), np.float32(0.7365), np.float32(0.7255), np.float32(0.8279)] 
+2026-10-18 12:19:19.422060: Epoch time: 109.13 s 
+2026-10-18 12:19:20.442700:  
+2026-10-18 12:19:20.443805: Epoch 139 
+2026-10-18 12:19:20.444098: Current learning rate: 0.00874 
+2026-10-18 12:21:09.602507: train_loss -0.5742 
+2026-10-18 12:21:09.603231: val_loss -0.509 
+2026-10-18 12:21:09.603724: Pseudo dice [np.float32(0.9314), np.float32(0.6893), np.float32(0.8535), np.float32(0.844), np.float32(0.6423), np.float32(0.8943), np.float32(0.8172), np.float32(0.3395), np.float32(0.0), np.float32(0.5382), np.float32(0.779), np.float32(0.7936), np.float32(0.6398), np.float32(0.6113), np.float32(0.0), np.float32(0.0), np.float32(0.7489), np.float32(0.0438), np.float32(0.7879), np.float32(0.4738), np.float32(0.4311), np.float32(0.7358), np.float32(0.6986), np.float32(0.7673), np.float32(0.6693), np.float32(0.6796), np.float32(0.375), np.float32(0.8065), np.float32(0.0), np.float32(0.721), np.float32(0.5229), np.float32(0.4868), np.float32(0.7848), np.float32(0.6903), np.float32(0.839), np.float32(0.8749)] 
+2026-10-18 12:21:09.604258: Epoch time: 109.16 s 
+2026-10-18 12:21:10.329367:  
+2026-10-18 12:21:10.329955: Epoch 140 
+2026-10-18 12:21:10.330119: Current learning rate: 0.00873 
+2026-10-18 12:22:59.497810: train_loss -0.5894 
+2026-10-18 12:22:59.498373: val_loss -0.3649 
+2026-10-18 12:22:59.498945: Pseudo dice [np.float32(0.9195), np.float32(0.5776), np.float32(0.7242), np.float32(0.6262), np.float32(0.4342), np.float32(0.6522), np.float32(0.771), np.float32(0.1258), np.float32(0.0), np.float32(0.4054), np.float32(0.5088), np.float32(0.5513), np.float32(0.4102), np.float32(0.2631), np.float32(0.0), np.float32(0.0), np.float32(0.3833), np.float32(0.0963), np.float32(0.6507), np.float32(0.4101), np.float32(0.3548), np.float32(0.6306), np.float32(0.3144), np.float32(0.5691), np.float32(0.5044), np.float32(0.5894), np.float32(0.1535), np.float32(0.5832), np.float32(0.0), np.float32(0.4807), np.float32(0.6413), np.float32(0.3784), np.float32(0.7137), np.float32(0.5156), np.float32(0.6258), np.float32(0.6818)] 
+2026-10-18 12:22:59.499450: Epoch time: 109.17 s 
+2026-10-18 12:23:00.259678:  
+2026-10-18 12:23:00.260328: Epoch 141 
+2026-10-18 12:23:00.260652: Current learning rate: 0.00872 
+2026-10-18 12:24:49.439719: train_loss -0.5792 
+2026-10-18 12:24:49.440702: val_loss -0.5265 
+2026-10-18 12:24:49.441401: Pseudo dice [np.float32(0.951), np.float32(0.6763), np.float32(0.9068), np.float32(0.8666), np.float32(0.7013), np.float32(0.9143), np.float32(0.8387), np.float32(0.2117), np.float32(0.0), np.float32(0.5011), np.float32(0.8218), np.float32(0.825), np.float32(0.6391), np.float32(0.4138), np.float32(0.0), np.float32(0.0), np.float32(0.7573), np.float32(0.4278), np.float32(0.7688), np.float32(0.5486), np.float32(0.2554), np.float32(0.6389), np.float32(0.8168), np.float32(0.8179), np.float32(0.6886), np.float32(0.7388), np.float32(0.4262), np.float32(0.8729), np.float32(0.0), np.float32(0.5526), np.float32(0.5199), np.float32(0.6486), np.float32(0.704), np.float32(0.6889), np.float32(0.9088), np.float32(0.9362)] 
+2026-10-18 12:24:49.441674: Epoch time: 109.18 s 
+2026-10-18 12:24:50.172330:  
+2026-10-18 12:24:50.173027: Epoch 142 
+2026-10-18 12:24:50.173322: Current learning rate: 0.00871 
+2026-10-18 12:26:39.383708: train_loss -0.5662 
+2026-10-18 12:26:39.384236: val_loss -0.4596 
+2026-10-18 12:26:39.385255: Pseudo dice [np.float32(0.9109), np.float32(0.7426), np.float32(0.8305), np.float32(0.7401), np.float32(0.6134), np.float32(0.8376), np.float32(0.809), np.float32(0.4013), np.float32(0.0), np.float32(0.4213), np.float32(0.7348), np.float32(0.7853), np.float32(0.5935), np.float32(0.502), np.float32(0.0), np.float32(0.0), np.float32(0.7088), np.float32(0.3836), np.float32(0.7659), np.float32(0.5767), np.float32(0.4211), np.float32(0.726), np.float32(0.499), np.float32(0.6748), np.float32(0.6369), np.float32(0.6795), np.float32(0.2388), np.float32(0.7351), np.float32(0.0), np.float32(0.6413), np.float32(0.4171), np.float32(0.4229), np.float32(0.527), np.float32(0.5428), np.float32(0.8669), np.float32(0.8915)] 
+2026-10-18 12:26:39.385820: Epoch time: 109.21 s 
+2026-10-18 12:26:40.117192:  
+2026-10-18 12:26:40.117941: Epoch 143 
+2026-10-18 12:26:40.118389: Current learning rate: 0.0087 
+2026-10-18 12:28:29.283650: train_loss -0.563 
+2026-10-18 12:28:29.284227: val_loss -0.483 
+2026-10-18 12:28:29.284831: Pseudo dice [np.float32(0.9208), np.float32(0.699), np.float32(0.8066), np.float32(0.8237), np.float32(0.565), np.float32(0.9049), np.float32(0.7466), np.float32(0.3869), np.float32(0.0), np.float32(0.2288), np.float32(0.7108), np.float32(0.7761), np.float32(0.4873), np.float32(0.2546), np.float32(0.0), np.float32(0.0), np.float32(0.71), np.float32(0.4674), np.float32(0.7287), np.float32(0.6163), np.float32(0.3062), np.float32(0.5605), np.float32(0.6114), np.float32(0.6944), np.float32(0.6956), np.float32(0.6852), np.float32(0.3571), np.float32(0.7906), np.float32(0.0), np.float32(0.616), np.float32(0.6435), np.float32(0.4497), np.float32(0.7652), np.float32(0.6521), np.float32(0.8379), np.float32(0.9139)] 
+2026-10-18 12:28:29.285920: Epoch time: 109.17 s 
+2026-10-18 12:28:30.049464:  
+2026-10-18 12:28:30.050163: Epoch 144 
+2026-10-18 12:28:30.050516: Current learning rate: 0.00869 
+2026-10-18 12:30:19.208150: train_loss -0.5502 
+2026-10-18 12:30:19.208638: val_loss -0.4928 
+2026-10-18 12:30:19.209260: Pseudo dice [np.float32(0.9074), np.float32(0.6993), np.float32(0.8054), np.float32(0.8378), np.float32(0.6488), np.float32(0.8929), np.float32(0.8136), np.float32(0.508), np.float32(0.0), np.float32(0.4834), np.float32(0.761), np.float32(0.7902), np.float32(0.6681), np.float32(0.4388), np.float32(0.0), np.float32(0.0), np.float32(0.7978), np.float32(0.57), np.float32(0.7828), np.float32(0.4035), np.float32(0.4371), np.float32(0.6074), np.float32(0.7064), np.float32(0.6902), np.float32(0.6032), np.float32(0.6631), np.float32(0.4119), np.float32(0.7271), np.float32(0.0), np.float32(0.5625), np.float32(0.6589), np.float32(0.3764), np.float32(0.7514), np.float32(0.686), np.float32(0.8788), np.float32(0.8969)] 
+2026-10-18 12:30:19.210038: Epoch time: 109.16 s 
+2026-10-18 12:30:19.942292:  
+2026-10-18 12:30:19.942953: Epoch 145 
+2026-10-18 12:30:19.943225: Current learning rate: 0.00868 
+2026-10-18 12:32:09.100886: train_loss -0.565 
+2026-10-18 12:32:09.101441: val_loss -0.4043 
+2026-10-18 12:32:09.101982: Pseudo dice [np.float32(0.9178), np.float32(0.6047), np.float32(0.6241), np.float32(0.7271), np.float32(0.5909), np.float32(0.7514), np.float32(0.6063), np.float32(0.3732), np.float32(0.0), np.float32(0.1502), np.float32(0.6554), np.float32(0.6425), np.float32(0.5971), np.float32(0.4188), np.float32(0.0), np.float32(0.0), np.float32(0.582), np.float32(0.4021), np.float32(0.5618), np.float32(0.2339), np.float32(0.4129), np.float32(0.6094), np.float32(0.5359), np.float32(0.5783), np.float32(0.5975), np.float32(0.4948), np.float32(0.2494), np.float32(0.6246), np.float32(0.0), np.float32(0.5662), np.float32(0.4681), np.float32(0.1659), np.float32(0.6929), np.float32(0.5286), np.float32(0.7772), np.float32(0.7791)] 
+2026-10-18 12:32:09.102921: Epoch time: 109.16 s 
+2026-10-18 12:32:09.844165:  
+2026-10-18 12:32:09.844841: Epoch 146 
+2026-10-18 12:32:09.845231: Current learning rate: 0.00868 
+2026-10-18 12:33:59.010014: train_loss -0.5649 
+2026-10-18 12:33:59.010791: val_loss -0.4852 
+2026-10-18 12:33:59.011457: Pseudo dice [np.float32(0.9056), np.float32(0.6751), np.float32(0.8276), np.float32(0.8171), np.float32(0.5772), np.float32(0.8846), np.float32(0.8335), np.float32(0.4187), np.float32(0.0), np.float32(0.488), np.float32(0.7467), np.float32(0.7131), np.float32(0.4663), np.float32(0.3033), np.float32(0.0), np.float32(0.0), np.float32(0.7489), np.float32(0.4146), np.float32(0.7588), np.float32(0.2976), np.float32(0.2456), np.float32(0.5972), np.float32(0.7265), np.float32(0.6943), np.float32(0.616), np.float32(0.6273), np.float32(0.365), np.float32(0.7292), np.float32(0.0), np.float32(0.5932), np.float32(0.5774), np.float32(0.5548), np.float32(0.7162), np.float32(0.6361), np.float32(0.8715), np.float32(0.9136)] 
+2026-10-18 12:33:59.011921: Epoch time: 109.17 s 
+2026-10-18 12:33:59.747102:  
+2026-10-18 12:33:59.747800: Epoch 147 
+2026-10-18 12:33:59.748036: Current learning rate: 0.00867 
+2026-10-18 12:35:48.869816: train_loss -0.5646 
+2026-10-18 12:35:48.870451: val_loss -0.4892 
+2026-10-18 12:35:48.871071: Pseudo dice [np.float32(0.9174), np.float32(0.7125), np.float32(0.8176), np.float32(0.8729), np.float32(0.6981), np.float32(0.9061), np.float32(0.8092), np.float32(0.3903), np.float32(0.0), np.float32(0.538), np.float32(0.7668), np.float32(0.7779), np.float32(0.6181), np.float32(0.6159), np.float32(0.0), np.float32(0.0), np.float32(0.7354), np.float32(0.454), np.float32(0.7773), np.float32(0.514), np.float32(0.1939), np.float32(0.4813), np.float32(0.693), np.float32(0.7671), np.float32(0.6728), np.float32(0.7069), np.float32(0.3888), np.float32(0.7135), np.float32(0.0), np.float32(0.2587), np.float32(0.694), np.float32(0.3408), np.float32(0.7477), np.float32(0.6153), np.float32(0.8572), np.float32(0.8966)] 
+2026-10-18 12:35:48.871501: Epoch time: 109.12 s 
+2026-10-18 12:35:49.600926:  
+2026-10-18 12:35:49.601631: Epoch 148 
+2026-10-18 12:35:49.602126: Current learning rate: 0.00866 
+2026-10-18 12:37:38.732774: train_loss -0.5875 
+2026-10-18 12:37:38.733500: val_loss -0.5366 
+2026-10-18 12:37:38.734224: Pseudo dice [np.float32(0.921), np.float32(0.7481), np.float32(0.885), np.float32(0.8458), np.float32(0.7064), np.float32(0.9074), np.float32(0.8397), np.float32(0.4223), np.float32(0.0), np.float32(0.4623), np.float32(0.7698), np.float32(0.8083), np.float32(0.6112), np.float32(0.5981), np.float32(0.0), np.float32(0.0), np.float32(0.7386), np.float32(0.3707), np.float32(0.8007), np.float32(0.4053), np.float32(0.409), np.float32(0.6953), np.float32(0.7178), np.float32(0.8038), np.float32(0.7285), np.float32(0.7041), np.float32(0.2427), np.float32(0.7944), np.float32(0.0), np.float32(0.6016), np.float32(0.0), np.float32(0.2999), np.float32(0.7421), np.float32(0.5819), np.float32(0.887), np.float32(0.9422)] 
+2026-10-18 12:37:38.734766: Epoch time: 109.13 s 
+2026-10-18 12:37:39.491128:  
+2026-10-18 12:37:39.491832: Epoch 149 
+2026-10-18 12:37:39.492161: Current learning rate: 0.00865 
+2026-10-18 12:39:28.683300: train_loss -0.5973 
+2026-10-18 12:39:28.684092: val_loss -0.5448 
+2026-10-18 12:39:28.684631: Pseudo dice [np.float32(0.9309), np.float32(0.7704), np.float32(0.8431), np.float32(0.8681), np.float32(0.74), np.float32(0.9107), np.float32(0.8294), np.float32(0.3838), np.float32(0.0), np.float32(0.4691), np.float32(0.792), np.float32(0.8109), np.float32(0.6806), np.float32(0.5569), np.float32(0.0), np.float32(0.0), np.float32(0.7641), np.float32(0.6188), np.float32(0.8113), np.float32(0.6112), np.float32(0.3005), np.float32(0.7239), np.float32(0.8369), np.float32(0.8234), np.float32(0.7382), np.float32(0.7363), np.float32(0.4004), np.float32(0.8786), np.float32(0.0), np.float32(0.7871), np.float32(0.5508), np.float32(0.4115), np.float32(0.764), np.float32(0.7263), np.float32(0.9524), np.float32(0.9635)] 
+2026-10-18 12:39:28.685008: Epoch time: 109.19 s 
+2026-10-18 12:39:30.744430:  
+2026-10-18 12:39:30.745221: Epoch 150 
+2026-10-18 12:39:30.745567: Current learning rate: 0.00864 
+2026-10-18 12:41:19.932524: train_loss -0.6091 
+2026-10-18 12:41:19.933247: val_loss -0.553 
+2026-10-18 12:41:19.933916: Pseudo dice [np.float32(0.9345), np.float32(0.7404), np.float32(0.8736), np.float32(0.8817), np.float32(0.7589), np.float32(0.9253), np.float32(0.8561), np.float32(0.4383), np.float32(0.0), np.float32(0.3622), np.float32(0.7834), np.float32(0.804), np.float32(0.7059), np.float32(0.6385), np.float32(0.0), np.float32(0.0), np.float32(0.8033), np.float32(0.4776), np.float32(0.7776), np.float32(0.4914), np.float32(0.2974), np.float32(0.7193), np.float32(0.7546), np.float32(0.8171), np.float32(0.7124), np.float32(0.7423), np.float32(0.4985), np.float32(0.8667), np.float32(0.0), np.float32(0.7687), np.float32(0.6096), np.float32(0.3647), np.float32(0.8068), np.float32(0.6982), np.float32(0.9446), np.float32(0.9495)] 
+2026-10-18 12:41:19.934530: Epoch time: 109.19 s 
+2026-10-18 12:41:20.682427:  
+2026-10-18 12:41:20.683126: Epoch 151 
+2026-10-18 12:41:20.683477: Current learning rate: 0.00863 
+2026-10-18 12:43:09.881922: train_loss -0.5953 
+2026-10-18 12:43:09.882489: val_loss -0.5343 
+2026-10-18 12:43:09.884134: Pseudo dice [np.float32(0.9107), np.float32(0.7612), np.float32(0.8678), np.float32(0.8737), np.float32(0.5793), np.float32(0.893), np.float32(0.8274), np.float32(0.4933), np.float32(0.0), np.float32(0.1938), np.float32(0.7662), np.float32(0.8001), np.float32(0.6543), np.float32(0.5033), np.float32(0.0), np.float32(0.0), np.float32(0.7574), np.float32(0.5845), np.float32(0.7963), np.float32(0.5327), np.float32(0.2982), np.float32(0.6893), np.float32(0.679), np.float32(0.7411), np.float32(0.7819), np.float32(0.6818), np.float32(0.3651), np.float32(0.8416), np.float32(0.0), np.float32(0.7516), np.float32(0.6203), np.float32(0.3117), np.float32(0.8104), np.float32(0.6278), np.float32(0.875), np.float32(0.9156)] 
+2026-10-18 12:43:09.884747: Epoch time: 109.2 s 
+2026-10-18 12:43:10.625574:  
+2026-10-18 12:43:10.626283: Epoch 152 
+2026-10-18 12:43:10.626554: Current learning rate: 0.00862 
+2026-10-18 12:44:59.811836: train_loss -0.5685 
+2026-10-18 12:44:59.812660: val_loss -0.5099 
+2026-10-18 12:44:59.813022: Pseudo dice [np.float32(0.9167), np.float32(0.7838), np.float32(0.8835), np.float32(0.8519), np.float32(0.5368), np.float32(0.9124), np.float32(0.8239), np.float32(0.3657), np.float32(0.0), np.float32(0.0), np.float32(0.7467), np.float32(0.7971), np.float32(0.5423), np.float32(0.4503), np.float32(0.0), np.float32(0.0), np.float32(0.7643), np.float32(0.4597), np.float32(0.7944), np.float32(0.4867), np.float32(0.3186), np.float32(0.6107), np.float32(0.7638), np.float32(0.7915), np.float32(0.7521), np.float32(0.7622), np.float32(0.4285), np.float32(0.8252), np.float32(0.0), np.float32(0.727), np.float32(0.503), np.float32(0.5754), np.float32(0.7246), np.float32(0.6593), np.float32(0.9317), np.float32(0.9549)] 
+2026-10-18 12:44:59.813461: Epoch time: 109.19 s 
+2026-10-18 12:45:00.563922:  
+2026-10-18 12:45:00.564558: Epoch 153 
+2026-10-18 12:45:00.564956: Current learning rate: 0.00861 
+2026-10-18 12:46:49.750210: train_loss -0.5462 
+2026-10-18 12:46:49.750952: val_loss -0.4365 
+2026-10-18 12:46:49.751754: Pseudo dice [np.float32(0.9174), np.float32(0.6358), np.float32(0.7656), np.float32(0.7365), np.float32(0.5503), np.float32(0.7831), np.float32(0.7325), np.float32(0.4915), np.float32(0.0), np.float32(0.504), np.float32(0.6614), np.float32(0.7198), np.float32(0.5889), np.float32(0.4358), np.float32(0.0), np.float32(0.0), np.float32(0.5173), np.float32(0.2837), np.float32(0.5872), np.float32(0.3475), np.float32(0.2854), np.float32(0.5825), np.float32(0.5799), np.float32(0.5903), np.float32(0.6127), np.float32(0.5855), np.float32(0.206), np.float32(0.5344), np.float32(0.0), np.float32(0.6522), np.float32(0.5202), np.float32(0.2665), np.float32(0.7233), np.float32(0.6424), np.float32(0.7822), np.float32(0.8107)] 
+2026-10-18 12:46:49.752023: Epoch time: 109.19 s 
+2026-10-18 12:46:50.496872:  
+2026-10-18 12:46:50.497349: Epoch 154 
+2026-10-18 12:46:50.497653: Current learning rate: 0.0086 
+2026-10-18 12:48:39.671944: train_loss -0.5729 
+2026-10-18 12:48:39.672734: val_loss -0.5203 
+2026-10-18 12:48:39.673685: Pseudo dice [np.float32(0.9415), np.float32(0.7451), np.float32(0.8513), np.float32(0.8565), np.float32(0.7411), np.float32(0.8883), np.float32(0.7607), np.float32(0.6464), np.float32(0.0), np.float32(0.5037), np.float32(0.7807), np.float32(0.8389), np.float32(0.689), np.float32(0.6483), np.float32(0.0), np.float32(0.0), np.float32(0.7813), np.float32(0.533), np.float32(0.7749), np.float32(0.4855), np.float32(0.4168), np.float32(0.6279), np.float32(0.6612), np.float32(0.6518), np.float32(0.6648), np.float32(0.6763), np.float32(0.603), np.float32(0.7361), np.float32(0.0), np.float32(0.553), np.float32(0.5568), np.float32(0.3046), np.float32(0.7124), np.float32(0.5838), np.float32(0.8792), np.float32(0.8886)] 
+2026-10-18 12:48:39.674117: Epoch time: 109.18 s 
+2026-10-18 12:48:40.425345:  
+2026-10-18 12:48:40.425997: Epoch 155 
+2026-10-18 12:48:40.426201: Current learning rate: 0.00859 
+2026-10-18 12:50:29.599536: train_loss -0.5997 
+2026-10-18 12:50:29.600381: val_loss -0.512 
+2026-10-18 12:50:29.600889: Pseudo dice [np.float32(0.9255), np.float32(0.6628), np.float32(0.8859), np.float32(0.8719), np.float32(0.7062), np.float32(0.9352), np.float32(0.8285), np.float32(0.5466), np.float32(0.0), np.float32(0.3513), np.float32(0.7834), np.float32(0.8297), np.float32(0.6462), np.float32(0.5147), np.float32(0.0), np.float32(0.0), np.float32(0.7882), np.float32(0.4998), np.float32(0.7908), np.float32(0.4447), np.float32(0.1875), np.float32(0.5759), np.float32(0.8276), np.float32(0.7813), np.float32(0.6189), np.float32(0.6972), np.float32(0.386), np.float32(0.8036), np.float32(0.0), np.float32(0.6128), np.float32(0.4794), np.float32(0.606), np.float32(0.4831), np.float32(0.6426), np.float32(0.9242), np.float32(0.9383)] 
+2026-10-18 12:50:29.601340: Epoch time: 109.18 s 
+2026-10-18 12:50:30.382509:  
+2026-10-18 12:50:30.383115: Epoch 156 
+2026-10-18 12:50:30.383273: Current learning rate: 0.00858 
+2026-10-18 12:52:19.541280: train_loss -0.6048 
+2026-10-18 12:52:19.541898: val_loss -0.514 
+2026-10-18 12:52:19.542529: Pseudo dice [np.float32(0.9388), np.float32(0.7235), np.float32(0.858), np.float32(0.8747), np.float32(0.7514), np.float32(0.901), np.float32(0.7808), np.float32(0.4595), np.float32(0.0), np.float32(0.2399), np.float32(0.7305), np.float32(0.7485), np.float32(0.6428), np.float32(0.4025), np.float32(0.0), np.float32(0.0), np.float32(0.7387), np.float32(0.4779), np.float32(0.7478), np.float32(0.5754), np.float32(0.269), np.float32(0.6762), np.float32(0.7869), np.float32(0.7571), np.float32(0.6871), np.float32(0.6761), np.float32(0.4037), np.float32(0.6859), np.float32(0.0), np.float32(0.6793), np.float32(0.658), np.float32(0.3428), np.float32(0.8208), np.float32(0.6562), np.float32(0.876), np.float32(0.8867)] 
+2026-10-18 12:52:19.543128: Epoch time: 109.16 s 
+2026-10-18 12:52:20.290262:  
+2026-10-18 12:52:20.290984: Epoch 157 
+2026-10-18 12:52:20.291287: Current learning rate: 0.00858 
+2026-10-18 12:54:09.470895: train_loss -0.5898 
+2026-10-18 12:54:09.471913: val_loss -0.5052 
+2026-10-18 12:54:09.472717: Pseudo dice [np.float32(0.8933), np.float32(0.7028), np.float32(0.859), np.float32(0.815), np.float32(0.6314), np.float32(0.8724), np.float32(0.7943), np.float32(0.472), np.float32(0.0), np.float32(0.4038), np.float32(0.7949), np.float32(0.8215), np.float32(0.5441), np.float32(0.4711), np.float32(0.0), np.float32(0.0), np.float32(0.7152), np.float32(0.6149), np.float32(0.8105), np.float32(0.5721), np.float32(0.3564), np.float32(0.7115), np.float32(0.6648), np.float32(0.7541), np.float32(0.7239), np.float32(0.7134), np.float32(0.34), np.float32(0.7088), np.float32(0.0), np.float32(0.6244), np.float32(0.6754), np.float32(0.4987), np.float32(0.7316), np.float32(0.6704), np.float32(0.8447), np.float32(0.8689)] 
+2026-10-18 12:54:09.473101: Epoch time: 109.18 s 
+2026-10-18 12:54:10.498983:  
+2026-10-18 12:54:10.499805: Epoch 158 
+2026-10-18 12:54:10.500204: Current learning rate: 0.00857 
+2026-10-18 12:55:59.720594: train_loss -0.5924 
+2026-10-18 12:55:59.721150: val_loss -0.5049 
+2026-10-18 12:55:59.721761: Pseudo dice [np.float32(0.9331), np.float32(0.6723), np.float32(0.7923), np.float32(0.8722), np.float32(0.7736), np.float32(0.9051), np.float32(0.8353), np.float32(0.7089), np.float32(0.0), np.float32(0.4437), np.float32(0.7617), np.float32(0.7855), np.float32(0.6674), np.float32(0.3559), np.float32(0.0), np.float32(0.0), np.float32(0.7702), np.float32(0.4397), np.float32(0.7378), np.float32(0.4111), np.float32(0.2742), np.float32(0.6328), np.float32(0.5891), np.float32(0.6674), np.float32(0.6249), np.float32(0.719), np.float32(0.263), np.float32(0.8343), np.float32(0.0), np.float32(0.4968), np.float32(0.7295), np.float32(0.4377), np.float32(0.7895), np.float32(0.5581), np.float32(0.8753), np.float32(0.9199)] 
+2026-10-18 12:55:59.722620: Epoch time: 109.22 s 
+2026-10-18 12:56:00.467770:  
+2026-10-18 12:56:00.468379: Epoch 159 
+2026-10-18 12:56:00.468725: Current learning rate: 0.00856 
+2026-10-18 12:57:49.661831: train_loss -0.6268 
+2026-10-18 12:57:49.662673: val_loss -0.523 
+2026-10-18 12:57:49.663271: Pseudo dice [np.float32(0.9251), np.float32(0.7232), np.float32(0.8719), np.float32(0.86), np.float32(0.7262), np.float32(0.9249), np.float32(0.86), np.float32(0.555), np.float32(0.0), np.float32(0.4715), np.float32(0.784), np.float32(0.8165), np.float32(0.6744), np.float32(0.4784), np.float32(0.0), np.float32(0.0), np.float32(0.7288), np.float32(0.4165), np.float32(0.767), np.float32(0.5896), np.float32(0.2522), np.float32(0.665), np.float32(0.6824), np.float32(0.7283), np.float32(0.6951), np.float32(0.7454), np.float32(0.3509), np.float32(0.8093), np.float32(0.0), np.float32(0.662), np.float32(0.5339), np.float32(0.4086), np.float32(0.7478), np.float32(0.7239), np.float32(0.8695), np.float32(0.9009)] 
+2026-10-18 12:57:49.663716: Epoch time: 109.2 s 
+2026-10-18 12:57:50.410758:  
+2026-10-18 12:57:50.411519: Epoch 160 
+2026-10-18 12:57:50.411872: Current learning rate: 0.00855 
+2026-10-18 12:59:39.594185: train_loss -0.5944 
+2026-10-18 12:59:39.594990: val_loss -0.5232 
+2026-10-18 12:59:39.595812: Pseudo dice [np.float32(0.9147), np.float32(0.7264), np.float32(0.8451), np.float32(0.8805), np.float32(0.6886), np.float32(0.9318), np.float32(0.822), np.float32(0.5025), np.float32(0.0), np.float32(0.4426), np.float32(0.7504), np.float32(0.7775), np.float32(0.6751), np.float32(0.5012), np.float32(0.0), np.float32(0.0), np.float32(0.664), np.float32(0.4853), np.float32(0.7489), np.float32(0.6937), np.float32(0.2662), np.float32(0.5619), np.float32(0.687), np.float32(0.757), np.float32(0.7257), np.float32(0.6941), np.float32(0.2204), np.float32(0.3095), np.float32(0.0), np.float32(0.7531), np.float32(0.5307), np.float32(0.5311), np.float32(0.7988), np.float32(0.6707), np.float32(0.9067), np.float32(0.9302)] 
+2026-10-18 12:59:39.596298: Epoch time: 109.19 s 
+2026-10-18 12:59:40.346646:  
+2026-10-18 12:59:40.347402: Epoch 161 
+2026-10-18 12:59:40.347785: Current learning rate: 0.00854 
+2026-10-18 13:01:29.465783: train_loss -0.6073 
+2026-10-18 13:01:29.466269: val_loss -0.5304 
+2026-10-18 13:01:29.466784: Pseudo dice [np.float32(0.9066), np.float32(0.7688), np.float32(0.8589), np.float32(0.8457), np.float32(0.6807), np.float32(0.8941), np.float32(0.8375), np.float32(0.4194), np.float32(0.0), np.float32(0.4138), np.float32(0.7277), np.float32(0.8095), np.float32(0.5735), np.float32(0.453), np.float32(0.0), np.float32(0.0), np.float32(0.6909), np.float32(0.19), np.float32(0.7862), np.float32(0.4654), np.float32(0.5525), np.float32(0.5623), np.float32(0.6666), np.float32(0.7276), np.float32(0.7008), np.float32(0.7064), np.float32(0.3211), np.float32(0.6802), np.float32(0.0), np.float32(0.6694), np.float32(0.721), np.float32(0.4224), np.float32(0.8113), np.float32(0.5943), np.float32(0.8744), np.float32(0.94)] 
+2026-10-18 13:01:29.467476: Epoch time: 109.12 s 
+2026-10-18 13:01:30.222297:  
+2026-10-18 13:01:30.222883: Epoch 162 
+2026-10-18 13:01:30.223198: Current learning rate: 0.00853 
+2026-10-18 13:03:19.388451: train_loss -0.6216 
+2026-10-18 13:03:19.389143: val_loss -0.5349 
+2026-10-18 13:03:19.389616: Pseudo dice [np.float32(0.9328), np.float32(0.7174), np.float32(0.8822), np.float32(0.8721), np.float32(0.7288), np.float32(0.9199), np.float32(0.8463), np.float32(0.4673), np.float32(0.0), np.float32(0.42), np.float32(0.7955), np.float32(0.8605), np.float32(0.6616), np.float32(0.6198), np.float32(0.0), np.float32(0.0), np.float32(0.7287), np.float32(0.5331), np.float32(0.7851), np.float32(0.5604), np.float32(0.2999), np.float32(0.7287), np.float32(0.7388), np.float32(0.7246), np.float32(0.725), np.float32(0.7356), np.float32(0.3501), np.float32(0.767), np.float32(0.0), np.float32(0.7398), np.float32(0.7003), np.float32(0.3896), np.float32(0.8436), np.float32(0.7555), np.float32(0.8543), np.float32(0.91)] 
+2026-10-18 13:03:19.390067: Epoch time: 109.17 s 
+2026-10-18 13:03:20.147891:  
+2026-10-18 13:03:20.148525: Epoch 163 
+2026-10-18 13:03:20.148816: Current learning rate: 0.00852 
+2026-10-18 13:05:09.288440: train_loss -0.5809 
+2026-10-18 13:05:09.289171: val_loss -0.4726 
+2026-10-18 13:05:09.289650: Pseudo dice [np.float32(0.9079), np.float32(0.6766), np.float32(0.8008), np.float32(0.7634), np.float32(0.6298), np.float32(0.8454), np.float32(0.7009), np.float32(0.4397), np.float32(0.0), np.float32(0.4402), np.float32(0.6412), np.float32(0.7237), np.float32(0.3257), np.float32(0.4068), np.float32(0.0), np.float32(0.0), np.float32(0.6011), np.float32(0.3634), np.float32(0.7122), np.float32(0.4758), np.float32(0.4442), np.float32(0.6509), np.float32(0.6194), np.float32(0.7531), np.float32(0.6657), np.float32(0.6333), np.float32(0.1288), np.float32(0.806), np.float32(0.0), np.float32(0.7498), np.float32(0.5793), np.float32(0.4272), np.float32(0.667), np.float32(0.6038), np.float32(0.8026), np.float32(0.8821)] 
+2026-10-18 13:05:09.290138: Epoch time: 109.14 s 
+2026-10-18 13:05:10.039061:  
+2026-10-18 13:05:10.039795: Epoch 164 
+2026-10-18 13:05:10.040220: Current learning rate: 0.00851 
+2026-10-18 13:06:59.148291: train_loss -0.5773 
+2026-10-18 13:06:59.149218: val_loss -0.515 
+2026-10-18 13:06:59.149787: Pseudo dice [np.float32(0.9147), np.float32(0.7447), np.float32(0.7997), np.float32(0.8305), np.float32(0.7846), np.float32(0.8931), np.float32(0.7945), np.float32(0.4823), np.float32(0.0), np.float32(0.4742), np.float32(0.7456), np.float32(0.7783), np.float32(0.6544), np.float32(0.6553), np.float32(0.0), np.float32(0.0), np.float32(0.7408), np.float32(0.451), np.float32(0.7525), np.float32(0.5291), np.float32(0.1895), np.float32(0.5189), np.float32(0.7933), np.float32(0.8213), np.float32(0.7215), np.float32(0.7041), np.float32(0.4161), np.float32(0.8116), np.float32(0.0), np.float32(0.4749), np.float32(0.6159), np.float32(0.3624), np.float32(0.769), np.float32(0.6964), np.float32(0.863), np.float32(0.9109)] 
+2026-10-18 13:06:59.150315: Epoch time: 109.11 s 
+2026-10-18 13:06:59.887302:  
+2026-10-18 13:06:59.887978: Epoch 165 
+2026-10-18 13:06:59.888204: Current learning rate: 0.0085 
+2026-10-18 13:08:49.031813: train_loss -0.6021 
+2026-10-18 13:08:49.032421: val_loss -0.462 
+2026-10-18 13:08:49.033178: Pseudo dice [np.float32(0.9236), np.float32(0.7428), np.float32(0.8009), np.float32(0.7842), np.float32(0.6877), np.float32(0.8899), np.float32(0.8019), np.float32(0.5044), np.float32(0.0), np.float32(0.4713), np.float32(0.7098), np.float32(0.7681), np.float32(0.5449), np.float32(0.381), np.float32(0.0), np.float32(0.0), np.float32(0.7314), np.float32(0.4488), np.float32(0.7317), np.float32(0.4283), np.float32(0.4011), np.float32(0.6652), np.float32(0.4461), np.float32(0.4221), np.float32(0.6484), np.float32(0.7044), np.float32(0.3173), np.float32(0.6382), np.float32(0.0), np.float32(0.3979), np.float32(0.7182), np.float32(0.4408), np.float32(0.7719), np.float32(0.6586), np.float32(0.6618), np.float32(0.7324)] 
+2026-10-18 13:08:49.033848: Epoch time: 109.15 s 
+2026-10-18 13:08:49.770471:  
+2026-10-18 13:08:49.771116: Epoch 166 
+2026-10-18 13:08:49.771457: Current learning rate: 0.00849 
+2026-10-18 13:10:38.917195: train_loss -0.6028 
+2026-10-18 13:10:38.917998: val_loss -0.5339 
+2026-10-18 13:10:38.919203: Pseudo dice [np.float32(0.9458), np.float32(0.7648), np.float32(0.856), np.float32(0.8986), np.float32(0.7024), np.float32(0.9255), np.float32(0.8511), np.float32(0.4005), np.float32(0.0), np.float32(0.0), np.float32(0.7392), np.float32(0.7946), np.float32(0.6561), np.float32(0.4564), np.float32(0.0), np.float32(0.0), np.float32(0.753), np.float32(0.5315), np.float32(0.8179), np.float32(0.6941), np.float32(0.4284), np.float32(0.6856), np.float32(0.7918), np.float32(0.7835), np.float32(0.7579), np.float32(0.8003), np.float32(0.4334), np.float32(0.8799), np.float32(0.0), np.float32(0.7216), np.float32(0.709), np.float32(0.5879), np.float32(0.8376), np.float32(0.7153), np.float32(0.9374), np.float32(0.9532)] 
+2026-10-18 13:10:38.920105: Epoch time: 109.15 s 
+2026-10-18 13:10:39.672014:  
+2026-10-18 13:10:39.672803: Epoch 167 
+2026-10-18 13:10:39.673125: Current learning rate: 0.00848 
+2026-10-18 13:12:28.831147: train_loss -0.5541 
+2026-10-18 13:12:28.831637: val_loss -0.5057 
+2026-10-18 13:12:28.832637: Pseudo dice [np.float32(0.9168), np.float32(0.738), np.float32(0.8452), np.float32(0.835), np.float32(0.4989), np.float32(0.9061), np.float32(0.8142), np.float32(0.5072), np.float32(0.0), np.float32(0.1854), np.float32(0.7512), np.float32(0.7918), np.float32(0.6213), np.float32(0.5499), np.float32(0.0), np.float32(0.0), np.float32(0.7393), np.float32(0.5234), np.float32(0.7497), np.float32(0.5457), np.float32(0.3202), np.float32(0.651), np.float32(0.6828), np.float32(0.6839), np.float32(0.7039), np.float32(0.5718), np.float32(0.2426), np.float32(0.5763), np.float32(0.0), np.float32(0.6824), np.float32(0.0954), np.float32(0.3752), np.float32(0.7727), np.float32(0.6685), np.float32(0.866), np.float32(0.8998)] 
+2026-10-18 13:12:28.833032: Epoch time: 109.16 s 
+2026-10-18 13:12:29.599030:  
+2026-10-18 13:12:29.599759: Epoch 168 
+2026-10-18 13:12:29.600171: Current learning rate: 0.00847 
+2026-10-18 13:14:18.762390: train_loss -0.5672 
+2026-10-18 13:14:18.763009: val_loss -0.4598 
+2026-10-18 13:14:18.763911: Pseudo dice [np.float32(0.9126), np.float32(0.6968), np.float32(0.8347), np.float32(0.7651), np.float32(0.5931), np.float32(0.8364), np.float32(0.7466), np.float32(0.6082), np.float32(0.0), np.float32(0.2378), np.float32(0.6008), np.float32(0.6597), np.float32(0.5764), np.float32(0.3607), np.float32(0.0), np.float32(0.0), np.float32(0.5485), np.float32(0.3742), np.float32(0.643), np.float32(0.441), np.float32(0.5032), np.float32(0.7427), np.float32(0.8143), np.float32(0.8438), np.float32(0.7001), np.float32(0.6698), np.float32(0.5706), np.float32(0.8164), np.float32(0.0), np.float32(0.7217), np.float32(0.592), np.float32(0.4232), np.float32(0.6932), np.float32(0.5516), np.float32(0.8749), np.float32(0.8977)] 
+2026-10-18 13:14:18.764344: Epoch time: 109.17 s 
+2026-10-18 13:14:19.524195:  
+2026-10-18 13:14:19.524882: Epoch 169 
+2026-10-18 13:14:19.525299: Current learning rate: 0.00847 
+2026-10-18 13:16:08.659772: train_loss -0.5959 
+2026-10-18 13:16:08.660595: val_loss -0.5231 
+2026-10-18 13:16:08.661106: Pseudo dice [np.float32(0.9156), np.float32(0.7172), np.float32(0.8645), np.float32(0.867), np.float32(0.698), np.float32(0.921), np.float32(0.8419), np.float32(0.4823), np.float32(0.0), np.float32(0.3436), np.float32(0.7521), np.float32(0.8099), np.float32(0.6295), np.float32(0.4987), np.float32(0.0), np.float32(0.0), np.float32(0.7273), np.float32(0.5646), np.float32(0.7957), np.float32(0.6931), np.float32(0.3556), np.float32(0.7447), np.float32(0.8114), np.float32(0.7994), np.float32(0.6709), np.float32(0.7203), np.float32(0.3959), np.float32(0.7287), np.float32(0.0), np.float32(0.6625), np.float32(0.6845), np.float32(0.3468), np.float32(0.7839), np.float32(0.7219), np.float32(0.9061), np.float32(0.939)] 
+2026-10-18 13:16:08.661360: Epoch time: 109.14 s 
+2026-10-18 13:16:09.422317:  
+2026-10-18 13:16:09.423032: Epoch 170 
+2026-10-18 13:16:09.423351: Current learning rate: 0.00846 
+2026-10-18 13:17:58.587013: train_loss -0.6078 
+2026-10-18 13:17:58.587514: val_loss -0.4605 
+2026-10-18 13:17:58.587863: Pseudo dice [np.float32(0.9427), np.float32(0.7037), np.float32(0.824), np.float32(0.7851), np.float32(0.707), np.float32(0.8362), np.float32(0.7915), np.float32(0.4446), np.float32(0.0), np.float32(0.3316), np.float32(0.6871), np.float32(0.7764), np.float32(0.515), np.float32(0.4942), np.float32(0.0), np.float32(0.0), np.float32(0.7045), np.float32(0.4766), np.float32(0.7127), np.float32(0.4932), np.float32(0.3091), np.float32(0.6781), np.float32(0.5733), np.float32(0.6346), np.float32(0.6646), np.float32(0.5228), np.float32(0.3103), np.float32(0.3328), np.float32(0.0), np.float32(0.3974), np.float32(0.69), np.float32(0.4249), np.float32(0.7442), np.float32(0.6192), np.float32(0.8361), np.float32(0.8528)] 
+2026-10-18 13:17:58.588883: Epoch time: 109.17 s 
+2026-10-18 13:17:59.336099:  
+2026-10-18 13:17:59.336733: Epoch 171 
+2026-10-18 13:17:59.337071: Current learning rate: 0.00845 
+2026-10-18 13:19:48.467817: train_loss -0.6195 
+2026-10-18 13:19:48.469130: val_loss -0.4791 
+2026-10-18 13:19:48.469713: Pseudo dice [np.float32(0.9185), np.float32(0.5967), np.float32(0.8127), np.float32(0.806), np.float32(0.6593), np.float32(0.8625), np.float32(0.7876), np.float32(0.4157), np.float32(0.0), np.float32(0.4251), np.float32(0.6971), np.float32(0.7431), np.float32(0.4941), np.float32(0.3193), np.float32(0.0), np.float32(0.0), np.float32(0.5122), np.float32(0.134), np.float32(0.6489), np.float32(0.3944), np.float32(0.4785), np.float32(0.7556), np.float32(0.7073), np.float32(0.7588), np.float32(0.6763), np.float32(0.6506), np.float32(0.478), np.float32(0.8058), np.float32(0.0), np.float32(0.644), np.float32(0.6407), np.float32(0.455), np.float32(0.7349), np.float32(0.6895), np.float32(0.8612), np.float32(0.9142)] 
+2026-10-18 13:19:48.470069: Epoch time: 109.13 s 
+2026-10-18 13:19:49.227639:  
+2026-10-18 13:19:49.228307: Epoch 172 
+2026-10-18 13:19:49.228513: Current learning rate: 0.00844 
+2026-10-18 13:21:38.345941: train_loss -0.6149 
+2026-10-18 13:21:38.346718: val_loss -0.4874 
+2026-10-18 13:21:38.347393: Pseudo dice [np.float32(0.9142), np.float32(0.6509), np.float32(0.8208), np.float32(0.7408), np.float32(0.607), np.float32(0.8328), np.float32(0.7533), np.float32(0.4756), np.float32(0.0), np.float32(0.3604), np.float32(0.6511), np.float32(0.7707), np.float32(0.629), np.float32(0.4437), np.float32(0.0), np.float32(0.0), np.float32(0.7193), np.float32(0.4134), np.float32(0.729), np.float32(0.7057), np.float32(0.1667), np.float32(0.7005), np.float32(0.7795), np.float32(0.7652), np.float32(0.6395), np.float32(0.6547), np.float32(0.3787), np.float32(0.4927), np.float32(0.0), np.float32(0.6588), np.float32(0.5917), np.float32(0.3487), np.float32(0.6361), np.float32(0.5434), np.float32(0.8653), np.float32(0.8998)] 
+2026-10-18 13:21:38.347861: Epoch time: 109.12 s 
+2026-10-18 13:21:39.097543:  
+2026-10-18 13:21:39.098196: Epoch 173 
+2026-10-18 13:21:39.098599: Current learning rate: 0.00843 
+2026-10-18 13:23:28.212904: train_loss -0.6026 
+2026-10-18 13:23:28.213682: val_loss -0.5162 
+2026-10-18 13:23:28.214237: Pseudo dice [np.float32(0.8774), np.float32(0.7325), np.float32(0.874), np.float32(0.8591), np.float32(0.6672), np.float32(0.9102), np.float32(0.8263), np.float32(0.3093), np.float32(0.0), np.float32(0.3585), np.float32(0.7502), np.float32(0.8012), np.float32(0.411), np.float32(0.0114), np.float32(0.0), np.float32(0.0), np.float32(0.665), np.float32(0.4427), np.float32(0.7197), np.float32(0.6208), np.float32(0.2474), np.float32(0.6998), np.float32(0.6519), np.float32(0.7077), np.float32(0.7022), np.float32(0.7036), np.float32(0.5527), np.float32(0.8179), np.float32(0.0), np.float32(0.6503), np.float32(0.6428), np.float32(0.4955), np.float32(0.8114), np.float32(0.7215), np.float32(0.8748), np.float32(0.9)] 
+2026-10-18 13:23:28.214740: Epoch time: 109.12 s 
+2026-10-18 13:23:28.967490:  
+2026-10-18 13:23:28.968287: Epoch 174 
+2026-10-18 13:23:28.968633: Current learning rate: 0.00842 
+2026-10-18 13:25:18.161010: train_loss -0.6173 
+2026-10-18 13:25:18.161850: val_loss -0.526 
+2026-10-18 13:25:18.162648: Pseudo dice [np.float32(0.9099), np.float32(0.7637), np.float32(0.8349), np.float32(0.85), np.float32(0.7385), np.float32(0.9059), np.float32(0.834), np.float32(0.3469), np.float32(0.0), np.float32(0.4986), np.float32(0.7673), np.float32(0.8266), np.float32(0.6646), np.float32(0.4324), np.float32(0.0), np.float32(0.0), np.float32(0.7653), np.float32(0.5294), np.float32(0.7918), np.float32(0.5576), np.float32(0.4091), np.float32(0.7062), np.float32(0.7251), np.float32(0.7133), np.float32(0.7643), np.float32(0.6428), np.float32(0.404), np.float32(0.5777), np.float32(0.0), np.float32(0.4816), np.float32(0.612), np.float32(0.3038), np.float32(0.7614), np.float32(0.6184), np.float32(0.8649), np.float32(0.9364)] 
+2026-10-18 13:25:18.162978: Epoch time: 109.2 s 
+2026-10-18 13:25:18.916760:  
+2026-10-18 13:25:18.917679: Epoch 175 
+2026-10-18 13:25:18.918016: Current learning rate: 0.00841 
+2026-10-18 13:27:08.147226: train_loss -0.6351 
+2026-10-18 13:27:08.147992: val_loss -0.5199 
+2026-10-18 13:27:08.148577: Pseudo dice [np.float32(0.9057), np.float32(0.7576), np.float32(0.869), np.float32(0.8731), np.float32(0.7893), np.float32(0.9103), np.float32(0.854), np.float32(0.6649), np.float32(0.0), np.float32(0.3871), np.float32(0.769), np.float32(0.8051), np.float32(0.6452), np.float32(0.4764), np.float32(0.0), np.float32(0.0), np.float32(0.6983), np.float32(0.2992), np.float32(0.7382), np.float32(0.3946), np.float32(0.2885), np.float32(0.6829), np.float32(0.4994), np.float32(0.5204), np.float32(0.7548), np.float32(0.7339), np.float32(0.4037), np.float32(0.7829), np.float32(0.0), np.float32(0.4767), np.float32(0.6956), np.float32(0.5397), np.float32(0.8097), np.float32(0.7437), np.float32(0.8545), np.float32(0.91)] 
+2026-10-18 13:27:08.149185: Epoch time: 109.23 s 
+2026-10-18 13:27:08.900077:  
+2026-10-18 13:27:08.900896: Epoch 176 
+2026-10-18 13:27:08.901242: Current learning rate: 0.0084 
+2026-10-18 13:28:58.002112: train_loss -0.6199 
+2026-10-18 13:28:58.002637: val_loss -0.4312 
+2026-10-18 13:28:58.003131: Pseudo dice [np.float32(0.9027), np.float32(0.6122), np.float32(0.8233), np.float32(0.7982), np.float32(0.4467), np.float32(0.828), np.float32(0.7708), np.float32(0.3054), np.float32(0.0), np.float32(0.3686), np.float32(0.6502), np.float32(0.7211), np.float32(0.5301), np.float32(0.3573), np.float32(0.0), np.float32(0.0), np.float32(0.497), np.float32(0.0918), np.float32(0.6167), np.float32(0.4416), np.float32(0.3561), np.float32(0.7334), np.float32(0.6093), np.float32(0.6549), np.float32(0.5986), np.float32(0.6385), np.float32(0.4067), np.float32(0.8447), np.float32(0.0), np.float32(0.6175), np.float32(0.599), np.float32(0.5271), np.float32(0.7772), np.float32(0.6662), np.float32(0.7875), np.float32(0.8194)] 
+2026-10-18 13:28:58.003863: Epoch time: 109.1 s 
+2026-10-18 13:28:59.015343:  
+2026-10-18 13:28:59.015983: Epoch 177 
+2026-10-18 13:28:59.016387: Current learning rate: 0.00839 
+2026-10-18 13:30:48.199565: train_loss -0.6064 
+2026-10-18 13:30:48.200190: val_loss -0.495 
+2026-10-18 13:30:48.200804: Pseudo dice [np.float32(0.9058), np.float32(0.683), np.float32(0.7894), np.float32(0.8266), np.float32(0.6983), np.float32(0.8976), np.float32(0.8348), np.float32(0.5717), np.float32(0.0), np.float32(0.3602), np.float32(0.7852), np.float32(0.8165), np.float32(0.5829), np.float32(0.5504), np.float32(0.0), np.float32(0.0), np.float32(0.7516), np.float32(0.6165), np.float32(0.7596), np.float32(0.5707), np.float32(0.4267), np.float32(0.6932), np.float32(0.4628), np.float32(0.6526), np.float32(0.6595), np.float32(0.6743), np.float32(0.2859), np.float32(0.7506), np.float32(0.0), np.float32(0.4626), np.float32(0.6483), np.float32(0.3971), np.float32(0.7646), np.float32(0.6907), np.float32(0.7892), np.float32(0.8703)] 
+2026-10-18 13:30:48.201567: Epoch time: 109.19 s 
+2026-10-18 13:30:48.968111:  
+2026-10-18 13:30:48.968702: Epoch 178 
+2026-10-18 13:30:48.969077: Current learning rate: 0.00838 
+2026-10-18 13:32:38.128401: train_loss -0.6258 
+2026-10-18 13:32:38.129237: val_loss -0.4566 
+2026-10-18 13:32:38.129780: Pseudo dice [np.float32(0.9007), np.float32(0.637), np.float32(0.8231), np.float32(0.8549), np.float32(0.6824), np.float32(0.8912), np.float32(0.8308), np.float32(0.4969), np.float32(0.0), np.float32(0.4682), np.float32(0.7271), np.float32(0.8038), np.float32(0.5084), np.float32(0.2184), np.float32(0.0), np.float32(0.0), np.float32(0.6438), np.float32(0.4108), np.float32(0.7504), np.float32(0.595), np.float32(0.3811), np.float32(0.6932), np.float32(0.3299), np.float32(0.4498), np.float32(0.6009), np.float32(0.7066), np.float32(0.3656), np.float32(0.7158), np.float32(0.0), np.float32(0.3806), np.float32(0.7175), np.float32(0.4224), np.float32(0.8266), np.float32(0.7426), np.float32(0.7983), np.float32(0.8445)] 
+2026-10-18 13:32:38.130341: Epoch time: 109.16 s 
+2026-10-18 13:32:38.878754:  
+2026-10-18 13:32:38.879463: Epoch 179 
+2026-10-18 13:32:38.879780: Current learning rate: 0.00837 
+2026-10-18 13:34:28.065977: train_loss -0.6199 
+2026-10-18 13:34:28.066410: val_loss -0.4447 
+2026-10-18 13:34:28.067006: Pseudo dice [np.float32(0.913), np.float32(0.6778), np.float32(0.8113), np.float32(0.7499), np.float32(0.5089), np.float32(0.8594), np.float32(0.7115), np.float32(0.4217), np.float32(0.0), np.float32(0.267), np.float32(0.6937), np.float32(0.7371), np.float32(0.4663), np.float32(0.2194), np.float32(0.0), np.float32(0.0), np.float32(0.658), np.float32(0.4101), np.float32(0.723), np.float32(0.5707), np.float32(0.294), np.float32(0.701), np.float32(0.5517), np.float32(0.619), np.float32(0.6975), np.float32(0.7168), np.float32(0.1901), np.float32(0.7444), np.float32(0.0), np.float32(0.624), np.float32(0.4569), np.float32(0.5378), np.float32(0.7391), np.float32(0.7034), np.float32(0.8224), np.float32(0.8805)] 
+2026-10-18 13:34:28.067703: Epoch time: 109.19 s 
+2026-10-18 13:34:28.808213:  
+2026-10-18 13:34:28.808710: Epoch 180 
+2026-10-18 13:34:28.809093: Current learning rate: 0.00836 
+2026-10-18 13:36:17.904222: train_loss -0.5831 
+2026-10-18 13:36:17.905064: val_loss -0.3773 
+2026-10-18 13:36:17.905669: Pseudo dice [np.float32(0.9255), np.float32(0.6094), np.float32(0.6644), np.float32(0.5618), np.float32(0.4381), np.float32(0.6308), np.float32(0.6047), np.float32(0.2558), np.float32(0.0), np.float32(0.2646), np.float32(0.4837), np.float32(0.553), np.float32(0.3104), np.float32(0.1898), np.float32(0.0), np.float32(0.0), np.float32(0.3829), np.float32(0.0496), np.float32(0.515), np.float32(0.3632), np.float32(0.1795), np.float32(0.5272), np.float32(0.629), np.float32(0.7271), np.float32(0.6044), np.float32(0.4765), np.float32(0.3993), np.float32(0.677), np.float32(0.0), np.float32(0.4731), np.float32(0.645), np.float32(0.1624), np.float32(0.5345), np.float32(0.3746), np.float32(0.8081), np.float32(0.846)] 
+2026-10-18 13:36:17.906174: Epoch time: 109.1 s 
+2026-10-18 13:36:18.687418:  
+2026-10-18 13:36:18.688051: Epoch 181 
+2026-10-18 13:36:18.688422: Current learning rate: 0.00836 
+2026-10-18 13:38:07.825862: train_loss -0.608 
+2026-10-18 13:38:07.826463: val_loss -0.5027 
+2026-10-18 13:38:07.826970: Pseudo dice [np.float32(0.9284), np.float32(0.7201), np.float32(0.8575), np.float32(0.8261), np.float32(0.6446), np.float32(0.8864), np.float32(0.8136), np.float32(0.361), np.float32(0.0), np.float32(0.4161), np.float32(0.741), np.float32(0.7528), np.float32(0.5787), np.float32(0.533), np.float32(0.0), np.float32(0.0), np.float32(0.6731), np.float32(0.2779), np.float32(0.7049), np.float32(0.4692), np.float32(0.3761), np.float32(0.7374), np.float32(0.7214), np.float32(0.7798), np.float32(0.6888), np.float32(0.697), np.float32(0.4029), np.float32(0.8043), np.float32(0.0), np.float32(0.5588), np.float32(0.693), np.float32(0.5015), np.float32(0.782), np.float32(0.6908), np.float32(0.8984), np.float32(0.9123)] 
+2026-10-18 13:38:07.827975: Epoch time: 109.14 s 
+2026-10-18 13:38:08.574547:  
+2026-10-18 13:38:08.574949: Epoch 182 
+2026-10-18 13:38:08.575332: Current learning rate: 0.00835 
+2026-10-18 13:39:57.704091: train_loss -0.6077 
+2026-10-18 13:39:57.704871: val_loss -0.4703 
+2026-10-18 13:39:57.705293: Pseudo dice [np.float32(0.9273), np.float32(0.688), np.float32(0.8351), np.float32(0.802), np.float32(0.5361), np.float32(0.8892), np.float32(0.8009), np.float32(0.4954), np.float32(0.0), np.float32(0.2477), np.float32(0.7078), np.float32(0.7617), np.float32(0.6062), np.float32(0.4689), np.float32(0.0), np.float32(0.0), np.float32(0.5619), np.float32(0.1791), np.float32(0.7131), np.float32(0.5164), np.float32(0.2516), np.float32(0.6075), np.float32(0.7548), np.float32(0.7764), np.float32(0.6832), np.float32(0.6763), np.float32(0.3593), np.float32(0.8374), np.float32(0.0), np.float32(0.6328), np.float32(0.5577), np.float32(0.4269), np.float32(0.7034), np.float32(0.6308), np.float32(0.8563), np.float32(0.918)] 
+2026-10-18 13:39:57.705537: Epoch time: 109.13 s 
+2026-10-18 13:39:58.448824:  
+2026-10-18 13:39:58.449510: Epoch 183 
+2026-10-18 13:39:58.449850: Current learning rate: 0.00834 
+2026-10-18 13:41:47.590708: train_loss -0.614 
+2026-10-18 13:41:47.591565: val_loss -0.5199 
+2026-10-18 13:41:47.591951: Pseudo dice [np.float32(0.9169), np.float32(0.7389), np.float32(0.8754), np.float32(0.8857), np.float32(0.6651), np.float32(0.9239), np.float32(0.8623), np.float32(0.5671), np.float32(0.0), np.float32(0.3972), np.float32(0.7573), np.float32(0.8294), np.float32(0.7135), np.float32(0.6032), np.float32(0.0), np.float32(0.0), np.float32(0.7456), np.float32(0.5023), np.float32(0.781), np.float32(0.6305), np.float32(0.4167), np.float32(0.6946), np.float32(0.8336), np.float32(0.8779), np.float32(0.7205), np.float32(0.6685), np.float32(0.3276), np.float32(0.717), np.float32(0.0), np.float32(0.6539), np.float32(0.4895), np.float32(0.5336), np.float32(0.7866), np.float32(0.7549), np.float32(0.9385), np.float32(0.9518)] 
+2026-10-18 13:41:47.592395: Epoch time: 109.14 s 
+2026-10-18 13:41:48.377354:  
+2026-10-18 13:41:48.378073: Epoch 184 
+2026-10-18 13:41:48.378433: Current learning rate: 0.00833 
+2026-10-18 13:43:37.526509: train_loss -0.6239 
+2026-10-18 13:43:37.527058: val_loss -0.5372 
+2026-10-18 13:43:37.527504: Pseudo dice [np.float32(0.9336), np.float32(0.74), np.float32(0.8787), np.float32(0.8708), np.float32(0.7399), np.float32(0.925), np.float32(0.8366), np.float32(0.4507), np.float32(0.0), np.float32(0.355), np.float32(0.7852), np.float32(0.8053), np.float32(0.5762), np.float32(0.521), np.float32(0.0), np.float32(0.0), np.float32(0.7786), np.float32(0.5996), np.float32(0.7696), np.float32(0.6382), np.float32(0.3506), np.float32(0.7448), np.float32(0.6566), np.float32(0.7981), np.float32(0.7161), np.float32(0.7004), np.float32(0.5313), np.float32(0.6901), np.float32(0.0), np.float32(0.5725), np.float32(0.7249), np.float32(0.3543), np.float32(0.82), np.float32(0.7564), np.float32(0.9232), np.float32(0.9342)] 
+2026-10-18 13:43:37.528052: Epoch time: 109.15 s 
+2026-10-18 13:43:38.250104:  
+2026-10-18 13:43:38.250697: Epoch 185 
+2026-10-18 13:43:38.251037: Current learning rate: 0.00832 
+2026-10-18 13:45:27.347481: train_loss -0.639 
+2026-10-18 13:45:27.348053: val_loss -0.544 
+2026-10-18 13:45:27.348589: Pseudo dice [np.float32(0.9258), np.float32(0.7653), np.float32(0.9027), np.float32(0.9145), np.float32(0.664), np.float32(0.9297), np.float32(0.8593), np.float32(0.5698), np.float32(0.0), np.float32(0.3114), np.float32(0.7232), np.float32(0.7948), np.float32(0.6998), np.float32(0.5462), np.float32(0.0), np.float32(0.0), np.float32(0.7994), np.float32(0.4201), np.float32(0.8008), np.float32(0.6289), np.float32(0.3184), np.float32(0.7472), np.float32(0.7062), np.float32(0.7709), np.float32(0.7293), np.float32(0.7264), np.float32(0.3216), np.float32(0.7353), np.float32(0.0), np.float32(0.2184), np.float32(0.6983), np.float32(0.3522), np.float32(0.8537), np.float32(0.6835), np.float32(0.922), np.float32(0.9489)] 
+2026-10-18 13:45:27.349076: Epoch time: 109.1 s 
+2026-10-18 13:45:28.086723:  
+2026-10-18 13:45:28.087417: Epoch 186 
+2026-10-18 13:45:28.087752: Current learning rate: 0.00831 
+2026-10-18 13:47:17.144109: train_loss -0.6371 
+2026-10-18 13:47:17.144815: val_loss -0.5332 
+2026-10-18 13:47:17.145365: Pseudo dice [np.float32(0.9239), np.float32(0.7244), np.float32(0.8921), np.float32(0.8784), np.float32(0.721), np.float32(0.8845), np.float32(0.8367), np.float32(0.3826), np.float32(0.0), np.float32(0.3545), np.float32(0.7507), np.float32(0.7745), np.float32(0.5883), np.float32(0.5523), np.float32(0.0), np.float32(0.0), np.float32(0.7801), np.float32(0.3949), np.float32(0.7788), np.float32(0.4881), np.float32(0.3166), np.float32(0.7167), np.float32(0.6914), np.float32(0.7008), np.float32(0.7749), np.float32(0.774), np.float32(0.5563), np.float32(0.8535), np.float32(0.0), np.float32(0.5875), np.float32(0.6799), np.float32(0.3542), np.float32(0.8064), np.float32(0.7221), np.float32(0.9413), np.float32(0.9487)] 
+2026-10-18 13:47:17.145756: Epoch time: 109.06 s 
+2026-10-18 13:47:17.868013:  
+2026-10-18 13:47:17.868634: Epoch 187 
+2026-10-18 13:47:17.868929: Current learning rate: 0.0083 
+2026-10-18 13:49:06.983087: train_loss -0.6573 
+2026-10-18 13:49:06.983948: val_loss -0.5177 
+2026-10-18 13:49:06.984621: Pseudo dice [np.float32(0.8998), np.float32(0.7097), np.float32(0.8547), np.float32(0.847), np.float32(0.6933), np.float32(0.9033), np.float32(0.7949), np.float32(0.3485), np.float32(0.0), np.float32(0.3728), np.float32(0.7819), np.float32(0.818), np.float32(0.6104), np.float32(0.4656), np.float32(0.0), np.float32(0.0), np.float32(0.7577), np.float32(0.4769), np.float32(0.6878), np.float32(0.4641), np.float32(0.389), np.float32(0.7669), np.float32(0.7432), np.float32(0.8393), np.float32(0.6678), np.float32(0.7028), np.float32(0.333), np.float32(0.682), np.float32(0.0), np.float32(0.6505), np.float32(0.735), np.float32(0.3512), np.float32(0.7584), np.float32(0.6976), np.float32(0.9065), np.float32(0.9293)] 
+2026-10-18 13:49:06.985051: Epoch time: 109.12 s 
+2026-10-18 13:49:07.726857:  
+2026-10-18 13:49:07.727479: Epoch 188 
+2026-10-18 13:49:07.727687: Current learning rate: 0.00829 
+2026-10-18 13:50:56.842829: train_loss -0.6401 
+2026-10-18 13:50:56.843617: val_loss -0.5199 
+2026-10-18 13:50:56.844360: Pseudo dice [np.float32(0.9209), np.float32(0.7294), np.float32(0.865), np.float32(0.8948), np.float32(0.7318), np.float32(0.9291), np.float32(0.84), np.float32(0.5718), np.float32(0.0), np.float32(0.1996), np.float32(0.7772), np.float32(0.8353), np.float32(0.6399), np.float32(0.502), np.float32(0.0), np.float32(0.0), np.float32(0.7862), np.float32(0.5182), np.float32(0.7783), np.float32(0.4746), np.float32(0.3913), np.float32(0.6861), np.float32(0.6477), np.float32(0.6861), np.float32(0.7147), np.float32(0.6718), np.float32(0.3408), np.float32(0.5638), np.float32(0.0), np.float32(0.6107), np.float32(0.507), np.float32(0.4662), np.float32(0.8136), np.float32(0.739), np.float32(0.9129), np.float32(0.9291)] 
+2026-10-18 13:50:56.844827: Epoch time: 109.12 s 
+2026-10-18 13:50:57.595323:  
+2026-10-18 13:50:57.596045: Epoch 189 
+2026-10-18 13:50:57.596222: Current learning rate: 0.00828 
+2026-10-18 13:52:46.718417: train_loss -0.5867 
+2026-10-18 13:52:46.719229: val_loss -0.5178 
+2026-10-18 13:52:46.719826: Pseudo dice [np.float32(0.915), np.float32(0.7465), np.float32(0.8584), np.float32(0.8374), np.float32(0.651), np.float32(0.9084), np.float32(0.7557), np.float32(0.1814), np.float32(0.0), np.float32(0.4692), np.float32(0.7636), np.float32(0.7705), np.float32(0.6294), np.float32(0.5299), np.float32(0.0), np.float32(0.0), np.float32(0.6929), np.float32(0.4529), np.float32(0.7077), np.float32(0.4831), np.float32(0.2478), np.float32(0.7094), np.float32(0.6938), np.float32(0.7236), np.float32(0.7629), np.float32(0.6753), np.float32(0.4714), np.float32(0.8066), np.float32(0.0), np.float32(0.6317), np.float32(0.5297), np.float32(0.5408), np.float32(0.7994), np.float32(0.6962), np.float32(0.918), np.float32(0.9351)] 
+2026-10-18 13:52:46.720249: Epoch time: 109.13 s 
+2026-10-18 13:52:47.467837:  
+2026-10-18 13:52:47.468488: Epoch 190 
+2026-10-18 13:52:47.468691: Current learning rate: 0.00827 
+2026-10-18 13:54:36.632434: train_loss -0.6115 
+2026-10-18 13:54:36.633032: val_loss -0.4991 
+2026-10-18 13:54:36.634098: Pseudo dice [np.float32(0.9008), np.float32(0.6741), np.float32(0.8374), np.float32(0.7872), np.float32(0.5337), np.float32(0.8866), np.float32(0.8356), np.float32(0.4918), np.float32(0.0), np.float32(0.4693), np.float32(0.7416), np.float32(0.8143), np.float32(0.5174), np.float32(0.5902), np.float32(0.0), np.float32(0.0), np.float32(0.5146), np.float32(0.2433), np.float32(0.7424), np.float32(0.5678), np.float32(0.3866), np.float32(0.6188), np.float32(0.6679), np.float32(0.6923), np.float32(0.6681), np.float32(0.6391), np.float32(0.4031), np.float32(0.687), np.float32(0.0), np.float32(0.4317), np.float32(0.6801), np.float32(0.3843), np.float32(0.7796), np.float32(0.6284), np.float32(0.8723), np.float32(0.918)] 
+2026-10-18 13:54:36.634718: Epoch time: 109.17 s 
+2026-10-18 13:54:37.417824:  
+2026-10-18 13:54:37.418756: Epoch 191 
+2026-10-18 13:54:37.419095: Current learning rate: 0.00826 
+2026-10-18 13:56:26.542006: train_loss -0.6096 
+2026-10-18 13:56:26.542621: val_loss -0.4937 
+2026-10-18 13:56:26.543389: Pseudo dice [np.float32(0.9374), np.float32(0.7254), np.float32(0.8417), np.float32(0.8097), np.float32(0.6079), np.float32(0.8864), np.float32(0.7425), np.float32(0.5348), np.float32(0.0), np.float32(0.3673), np.float32(0.715), np.float32(0.7429), np.float32(0.6631), np.float32(0.5781), np.float32(0.0), np.float32(0.0), np.float32(0.713), np.float32(0.5103), np.float32(0.7519), np.float32(0.5015), np.float32(0.4585), np.float32(0.6893), np.float32(0.7438), np.float32(0.726), np.float32(0.6672), np.float32(0.6891), np.float32(0.401), np.float32(0.737), np.float32(0.0), np.float32(0.4649), np.float32(0.6493), np.float32(0.453), np.float32(0.7109), np.float32(0.6509), np.float32(0.8836), np.float32(0.9064)] 
+2026-10-18 13:56:26.543934: Epoch time: 109.13 s 
+2026-10-18 13:56:27.316463:  
+2026-10-18 13:56:27.317176: Epoch 192 
+2026-10-18 13:56:27.317496: Current learning rate: 0.00825 
+2026-10-18 13:58:16.412584: train_loss -0.5942 
+2026-10-18 13:58:16.413301: val_loss -0.5153 
+2026-10-18 13:58:16.413922: Pseudo dice [np.float32(0.9373), np.float32(0.7162), np.float32(0.8518), np.float32(0.8553), np.float32(0.6708), np.float32(0.9039), np.float32(0.7694), np.float32(0.3794), np.float32(0.0), np.float32(0.4184), np.float32(0.7877), np.float32(0.7837), np.float32(0.6749), np.float32(0.552), np.float32(0.0), np.float32(0.0), np.float32(0.762), np.float32(0.4461), np.float32(0.754), np.float32(0.5353), np.float32(0.313), np.float32(0.7606), np.float32(0.6235), np.float32(0.6933), np.float32(0.7096), np.float32(0.695), np.float32(0.5515), np.float32(0.7699), np.float32(0.0), np.float32(0.6026), np.float32(0.6781), np.float32(0.3594), np.float32(0.8309), np.float32(0.7416), np.float32(0.8906), np.float32(0.9248)] 
+2026-10-18 13:58:16.414321: Epoch time: 109.1 s 
+2026-10-18 13:58:17.162615:  
+2026-10-18 13:58:17.163305: Epoch 193 
+2026-10-18 13:58:17.163610: Current learning rate: 0.00824 
+2026-10-18 14:00:06.275337: train_loss -0.6152 
+2026-10-18 14:00:06.276031: val_loss -0.5139 
+2026-10-18 14:00:06.276451: Pseudo dice [np.float32(0.9356), np.float32(0.6182), np.float32(0.8592), np.float32(0.8683), np.float32(0.6531), np.float32(0.9129), np.float32(0.8373), np.float32(0.4185), np.float32(0.0), np.float32(0.5016), np.float32(0.7947), np.float32(0.8096), np.float32(0.6423), np.float32(0.4372), np.float32(0.0), np.float32(0.0), np.float32(0.6849), np.float32(0.3638), np.float32(0.7465), np.float32(0.5665), np.float32(0.2302), np.float32(0.6567), np.float32(0.7869), np.float32(0.8275), np.float32(0.6776), np.float32(0.6777), np.float32(0.3968), np.float32(0.7973), np.float32(0.0), np.float32(0.4753), np.float32(0.4938), np.float32(0.1025), np.float32(0.8534), np.float32(0.7602), np.float32(0.892), np.float32(0.9307)] 
+2026-10-18 14:00:06.277118: Epoch time: 109.11 s 
+2026-10-18 14:00:07.036597:  
+2026-10-18 14:00:07.037129: Epoch 194 
+2026-10-18 14:00:07.037657: Current learning rate: 0.00824 
+2026-10-18 14:01:56.185618: train_loss -0.5923 
+2026-10-18 14:01:56.186464: val_loss -0.4705 
+2026-10-18 14:01:56.186944: Pseudo dice [np.float32(0.9303), np.float32(0.6905), np.float32(0.7448), np.float32(0.7103), np.float32(0.6204), np.float32(0.7933), np.float32(0.6983), np.float32(0.4948), np.float32(0.0), np.float32(0.4258), np.float32(0.6543), np.float32(0.6786), np.float32(0.6136), np.float32(0.5543), np.float32(0.0), np.float32(0.0), np.float32(0.6023), np.float32(0.1633), np.float32(0.6975), np.float32(0.5728), np.float32(0.3482), np.float32(0.6667), np.float32(0.7028), np.float32(0.7484), np.float32(0.6163), np.float32(0.5934), np.float32(0.2965), np.float32(0.6207), np.float32(0.0), np.float32(0.6843), np.float32(0.6334), np.float32(0.2252), np.float32(0.6986), np.float32(0.5527), np.float32(0.8271), np.float32(0.8801)] 
+2026-10-18 14:01:56.187451: Epoch time: 109.15 s 
+2026-10-18 14:01:56.946699:  
+2026-10-18 14:01:56.947340: Epoch 195 
+2026-10-18 14:01:56.947645: Current learning rate: 0.00823 
+2026-10-18 14:03:46.051679: train_loss -0.6145 
+2026-10-18 14:03:46.052362: val_loss -0.5121 
+2026-10-18 14:03:46.052651: Pseudo dice [np.float32(0.9312), np.float32(0.7067), np.float32(0.8331), np.float32(0.8418), np.float32(0.6891), np.float32(0.8779), np.float32(0.7478), np.float32(0.4731), np.float32(0.0), np.float32(0.4515), np.float32(0.7428), np.float32(0.7687), np.float32(0.6761), np.float32(0.5468), np.float32(0.0), np.float32(0.0), np.float32(0.6643), np.float32(0.4663), np.float32(0.7119), np.float32(0.5234), np.float32(0.3203), np.float32(0.7381), np.float32(0.7042), np.float32(0.7006), np.float32(0.7064), np.float32(0.5566), np.float32(0.4002), np.float32(0.6831), np.float32(0.0), np.float32(0.6933), np.float32(0.7701), np.float32(0.2796), np.float32(0.7482), np.float32(0.6898), np.float32(0.8449), np.float32(0.9)] 
+2026-10-18 14:03:46.052848: Epoch time: 109.11 s 
+2026-10-18 14:03:47.047330:  
+2026-10-18 14:03:47.047917: Epoch 196 
+2026-10-18 14:03:47.048082: Current learning rate: 0.00822 
+2026-10-18 14:05:36.189212: train_loss -0.6216 
+2026-10-18 14:05:36.189768: val_loss -0.4981 
+2026-10-18 14:05:36.191151: Pseudo dice [np.float32(0.9074), np.float32(0.7431), np.float32(0.8459), np.float32(0.8617), np.float32(0.6758), np.float32(0.9124), np.float32(0.8385), np.float32(0.4282), np.float32(0.0), np.float32(0.2905), np.float32(0.7649), np.float32(0.7701), np.float32(0.5984), np.float32(0.4749), np.float32(0.0), np.float32(0.0), np.float32(0.7142), np.float32(0.4285), np.float32(0.8), np.float32(0.6634), np.float32(0.4349), np.float32(0.7324), np.float32(0.5916), np.float32(0.7451), np.float32(0.717), np.float32(0.6791), np.float32(0.3686), np.float32(0.7233), np.float32(0.0), np.float32(0.3915), np.float32(0.6927), np.float32(0.2613), np.float32(0.7776), np.float32(0.713), np.float32(0.8638), np.float32(0.9205)] 
+2026-10-18 14:05:36.191693: Epoch time: 109.14 s 
+2026-10-18 14:05:36.944294:  
+2026-10-18 14:05:36.944986: Epoch 197 
+2026-10-18 14:05:36.945347: Current learning rate: 0.00821 
+2026-10-18 14:07:26.053664: train_loss -0.6281 
+2026-10-18 14:07:26.054617: val_loss -0.4706 
+2026-10-18 14:07:26.055125: Pseudo dice [np.float32(0.9372), np.float32(0.7134), np.float32(0.8268), np.float32(0.8254), np.float32(0.7211), np.float32(0.8768), np.float32(0.778), np.float32(0.5866), np.float32(0.0), np.float32(0.1019), np.float32(0.7728), np.float32(0.7789), np.float32(0.7125), np.float32(0.7156), np.float32(0.0), np.float32(0.0), np.float32(0.8139), np.float32(0.4727), np.float32(0.7313), np.float32(0.5235), np.float32(0.3034), np.float32(0.6221), np.float32(0.5285), np.float32(0.6177), np.float32(0.6078), np.float32(0.7069), np.float32(0.2949), np.float32(0.6973), np.float32(0.0), np.float32(0.3957), np.float32(0.6663), np.float32(0.5307), np.float32(0.7744), np.float32(0.7518), np.float32(0.779), np.float32(0.7989)] 
+2026-10-18 14:07:26.055427: Epoch time: 109.11 s 
+2026-10-18 14:07:26.813582:  
+2026-10-18 14:07:26.814271: Epoch 198 
+2026-10-18 14:07:26.814614: Current learning rate: 0.0082 
+2026-10-18 14:09:15.903484: train_loss -0.641 
+2026-10-18 14:09:15.904327: val_loss -0.4473 
+2026-10-18 14:09:15.904893: Pseudo dice [np.float32(0.93), np.float32(0.6989), np.float32(0.7244), np.float32(0.6473), np.float32(0.5599), np.float32(0.6769), np.float32(0.6977), np.float32(0.2201), np.float32(0.0), np.float32(0.468), np.float32(0.5792), np.float32(0.5789), np.float32(0.5154), np.float32(0.2637), np.float32(0.0), np.float32(0.0), np.float32(0.7335), np.float32(0.6027), np.float32(0.7099), np.float32(0.5364), np.float32(0.5012), np.float32(0.7055), np.float32(0.757), np.float32(0.7314), np.float32(0.6636), np.float32(0.6029), np.float32(0.4001), np.float32(0.6835), np.float32(0.0), np.float32(0.6321), np.float32(0.7054), np.float32(0.5076), np.float32(0.6962), np.float32(0.5288), np.float32(0.7503), np.float32(0.7857)] 
+2026-10-18 14:09:15.905458: Epoch time: 109.09 s 
+2026-10-18 14:09:16.655023:  
+2026-10-18 14:09:16.655697: Epoch 199 
+2026-10-18 14:09:16.655907: Current learning rate: 0.00819 
+2026-10-18 14:11:05.836722: train_loss -0.6263 
+2026-10-18 14:11:05.838191: val_loss -0.4264 
+2026-10-18 14:11:05.839071: Pseudo dice [np.float32(0.9354), np.float32(0.5754), np.float32(0.7714), np.float32(0.6862), np.float32(0.4367), np.float32(0.813), np.float32(0.7697), np.float32(0.3007), np.float32(0.0), np.float32(0.5874), np.float32(0.686), np.float32(0.7221), np.float32(0.5681), np.float32(0.3394), np.float32(0.0), np.float32(0.0), np.float32(0.2985), np.float32(0.1224), np.float32(0.6528), np.float32(0.476), np.float32(0.2253), np.float32(0.6408), np.float32(0.4535), np.float32(0.6012), np.float32(0.5673), np.float32(0.5914), np.float32(0.358), np.float32(0.4768), np.float32(0.0), np.float32(0.4529), np.float32(0.3201), np.float32(0.4499), np.float32(0.6292), np.float32(0.5927), np.float32(0.6923), np.float32(0.8086)] 
+2026-10-18 14:11:05.840311: Epoch time: 109.18 s 
+2026-10-18 14:11:07.134056:  
+2026-10-18 14:11:07.134726: Epoch 200 
+2026-10-18 14:11:07.135018: Current learning rate: 0.00818 
+2026-10-18 14:12:56.258893: train_loss -0.585 
+2026-10-18 14:12:56.259699: val_loss -0.5099 
+2026-10-18 14:12:56.260417: Pseudo dice [np.float32(0.9271), np.float32(0.7345), np.float32(0.862), np.float32(0.8843), np.float32(0.7981), np.float32(0.9293), np.float32(0.8602), np.float32(0.6127), np.float32(0.0), np.float32(0.3661), np.float32(0.7465), np.float32(0.7976), np.float32(0.5393), np.float32(0.5505), np.float32(0.0), np.float32(0.0), np.float32(0.7533), np.float32(0.5003), np.float32(0.7266), np.float32(0.4512), np.float32(0.2942), np.float32(0.6328), np.float32(0.4525), np.float32(0.5545), np.float32(0.7491), np.float32(0.7079), np.float32(0.4343), np.float32(0.7552), np.float32(0.0), np.float32(0.3998), np.float32(0.5054), np.float32(0.4912), np.float32(0.7356), np.float32(0.5846), np.float32(0.8242), np.float32(0.8684)] 
+2026-10-18 14:12:56.260898: Epoch time: 109.13 s 
+2026-10-18 14:12:57.015805:  
+2026-10-18 14:12:57.016525: Epoch 201 
+2026-10-18 14:12:57.016726: Current learning rate: 0.00817 
+2026-10-18 14:14:46.124918: train_loss -0.6256 
+2026-10-18 14:14:46.125661: val_loss -0.5464 
+2026-10-18 14:14:46.126325: Pseudo dice [np.float32(0.9416), np.float32(0.7388), np.float32(0.8712), np.float32(0.8899), np.float32(0.7894), np.float32(0.9261), np.float32(0.8695), np.float32(0.5091), np.float32(0.0), np.float32(0.5396), np.float32(0.8017), np.float32(0.8436), np.float32(0.5699), np.float32(0.3206), np.float32(0.0), np.float32(0.0), np.float32(0.7558), np.float32(0.4911), np.float32(0.8025), np.float32(0.6524), np.float32(0.4062), np.float32(0.7), np.float32(0.7126), np.float32(0.7615), np.float32(0.7203), np.float32(0.7322), np.float32(0.359), np.float32(0.771), np.float32(0.0), np.float32(0.6616), np.float32(0.6835), np.float32(0.208), np.float32(0.8143), np.float32(0.7168), np.float32(0.9252), np.float32(0.9479)] 
+2026-10-18 14:14:46.126848: Epoch time: 109.11 s 
+2026-10-18 14:14:46.925252:  
+2026-10-18 14:14:46.925691: Epoch 202 
+2026-10-18 14:14:46.926077: Current learning rate: 0.00816 
+2026-10-18 14:16:36.107917: train_loss -0.6216 
+2026-10-18 14:16:36.108413: val_loss -0.4607 
+2026-10-18 14:16:36.109019: Pseudo dice [np.float32(0.8363), np.float32(0.6849), np.float32(0.8228), np.float32(0.8468), np.float32(0.6721), np.float32(0.8823), np.float32(0.77), np.float32(0.4406), np.float32(0.0014), np.float32(0.387), np.float32(0.7848), np.float32(0.7924), np.float32(0.5824), np.float32(0.5594), np.float32(0.0), np.float32(0.0), np.float32(0.6909), np.float32(0.4537), np.float32(0.7657), np.float32(0.4493), np.float32(0.3066), np.float32(0.506), np.float32(0.4707), np.float32(0.7008), np.float32(0.6907), np.float32(0.75), np.float32(0.2754), np.float32(0.7804), np.float32(0.0), np.float32(0.4518), np.float32(0.5891), np.float32(0.4337), np.float32(0.7596), np.float32(0.7061), np.float32(0.8675), np.float32(0.899)] 
+2026-10-18 14:16:36.109583: Epoch time: 109.18 s 
+2026-10-18 14:16:36.868933:  
+2026-10-18 14:16:36.869624: Epoch 203 
+2026-10-18 14:16:36.869998: Current learning rate: 0.00815 
+2026-10-18 14:18:26.001935: train_loss -0.5343 
+2026-10-18 14:18:26.003195: val_loss -0.4942 
+2026-10-18 14:18:26.003791: Pseudo dice [np.float32(0.8821), np.float32(0.6317), np.float32(0.8297), np.float32(0.8453), np.float32(0.6284), np.float32(0.8903), np.float32(0.804), np.float32(0.2), np.float32(0.0), np.float32(0.4769), np.float32(0.7471), np.float32(0.7711), np.float32(0.6083), np.float32(0.454), np.float32(0.0), np.float32(0.0), np.float32(0.4249), np.float32(0.0616), np.float32(0.7189), np.float32(0.2402), np.float32(0.2261), np.float32(0.6632), np.float32(0.7859), np.float32(0.7841), np.float32(0.6385), np.float32(0.6722), np.float32(0.5767), np.float32(0.7187), np.float32(0.0), np.float32(0.5671), np.float32(0.6259), np.float32(0.4843), np.float32(0.7774), np.float32(0.5553), np.float32(0.8966), np.float32(0.9365)] 
+2026-10-18 14:18:26.004474: Epoch time: 109.14 s 
+2026-10-18 14:18:26.743597:  
+2026-10-18 14:18:26.744272: Epoch 204 
+2026-10-18 14:18:26.744469: Current learning rate: 0.00814 
+2026-10-18 14:20:15.868668: train_loss -0.5845 
+2026-10-18 14:20:15.869221: val_loss -0.5011 
+2026-10-18 14:20:15.869785: Pseudo dice [np.float32(0.9093), np.float32(0.7033), np.float32(0.8431), np.float32(0.838), np.float32(0.7137), np.float32(0.8757), np.float32(0.828), np.float32(0.3311), np.float32(0.0), np.float32(0.1657), np.float32(0.7582), np.float32(0.7855), np.float32(0.2457), np.float32(0.5761), np.float32(0.0), np.float32(0.0), np.float32(0.6948), np.float32(0.3274), np.float32(0.6979), np.float32(0.3468), np.float32(0.4278), np.float32(0.5898), np.float32(0.7437), np.float32(0.6867), np.float32(0.6939), np.float32(0.6757), np.float32(0.3991), np.float32(0.7176), np.float32(0.0), np.float32(0.5455), np.float32(0.6938), np.float32(0.4511), np.float32(0.7196), np.float32(0.5839), np.float32(0.8045), np.float32(0.8336)] 
+2026-10-18 14:20:15.870660: Epoch time: 109.13 s 
+2026-10-18 14:20:16.620578:  
+2026-10-18 14:20:16.621187: Epoch 205 
+2026-10-18 14:20:16.621394: Current learning rate: 0.00813 
+2026-10-18 14:22:05.779367: train_loss -0.5942 
+2026-10-18 14:22:05.780020: val_loss -0.5445 
+2026-10-18 14:22:05.780860: Pseudo dice [np.float32(0.9247), np.float32(0.7319), np.float32(0.8633), np.float32(0.8816), np.float32(0.7047), np.float32(0.8792), np.float32(0.7959), np.float32(0.5837), np.float32(0.0), np.float32(0.4035), np.float32(0.7498), np.float32(0.811), np.float32(0.5339), np.float32(0.5453), np.float32(0.0), np.float32(0.0), np.float32(0.7289), np.float32(0.5083), np.float32(0.7605), np.float32(0.4896), np.float32(0.3878), np.float32(0.7068), np.float32(0.7162), np.float32(0.7466), np.float32(0.7152), np.float32(0.7276), np.float32(0.5652), np.float32(0.8104), np.float32(0.0), np.float32(0.7228), np.float32(0.6246), np.float32(0.5762), np.float32(0.7806), np.float32(0.7264), np.float32(0.9147), np.float32(0.9269)] 
+2026-10-18 14:22:05.781329: Epoch time: 109.16 s 
+2026-10-18 14:22:06.500939:  
+2026-10-18 14:22:06.501609: Epoch 206 
+2026-10-18 14:22:06.501983: Current learning rate: 0.00813 
+2026-10-18 14:23:55.626191: train_loss -0.6245 
+2026-10-18 14:23:55.627019: val_loss -0.5549 
+2026-10-18 14:23:55.627543: Pseudo dice [np.float32(0.9307), np.float32(0.7891), np.float32(0.8979), np.float32(0.899), np.float32(0.6726), np.float32(0.9296), np.float32(0.8519), np.float32(0.431), np.float32(0.0), np.float32(0.4688), np.float32(0.8106), np.float32(0.8534), np.float32(0.7337), np.float32(0.5487), np.float32(0.0), np.float32(0.0), np.float32(0.7859), np.float32(0.5706), np.float32(0.8049), np.float32(0.5317), np.float32(0.3353), np.float32(0.6375), np.float32(0.692), np.float32(0.7736), np.float32(0.769), np.float32(0.6778), np.float32(0.5118), np.float32(0.7908), np.float32(0.0), np.float32(0.6967), np.float32(0.5913), np.float32(0.5043), np.float32(0.8131), np.float32(0.7033), np.float32(0.9192), np.float32(0.938)] 
+2026-10-18 14:23:55.628000: Epoch time: 109.13 s 
+2026-10-18 14:23:56.351741:  
+2026-10-18 14:23:56.352453: Epoch 207 
+2026-10-18 14:23:56.352833: Current learning rate: 0.00812 
+2026-10-18 14:25:45.493031: train_loss -0.6343 
+2026-10-18 14:25:45.493726: val_loss -0.5015 
+2026-10-18 14:25:45.494403: Pseudo dice [np.float32(0.931), np.float32(0.6383), np.float32(0.8397), np.float32(0.8511), np.float32(0.6931), np.float32(0.8965), np.float32(0.7978), np.float32(0.4086), np.float32(0.0), np.float32(0.5333), np.float32(0.7707), np.float32(0.8379), np.float32(0.683), np.float32(0.6235), np.float32(0.0), np.float32(0.0), np.float32(0.7691), np.float32(0.5396), np.float32(0.7361), np.float32(0.5041), np.float32(0.3082), np.float32(0.6621), np.float32(0.5141), np.float32(0.7055), np.float32(0.7628), np.float32(0.7188), np.float32(0.4442), np.float32(0.7837), np.float32(0.0), np.float32(0.4471), np.float32(0.4586), np.float32(0.5119), np.float32(0.7811), np.float32(0.6816), np.float32(0.836), np.float32(0.8761)] 
+2026-10-18 14:25:45.494703: Epoch time: 109.14 s 
+2026-10-18 14:25:46.219844:  
+2026-10-18 14:25:46.220552: Epoch 208 
+2026-10-18 14:25:46.220856: Current learning rate: 0.00811 
+2026-10-18 14:27:35.392730: train_loss -0.6175 
+2026-10-18 14:27:35.393512: val_loss -0.5058 
+2026-10-18 14:27:35.394267: Pseudo dice [np.float32(0.9315), np.float32(0.7328), np.float32(0.8682), np.float32(0.8709), np.float32(0.6032), np.float32(0.928), np.float32(0.8485), np.float32(0.4576), np.float32(0.0), np.float32(0.4465), np.float32(0.7039), np.float32(0.7785), np.float32(0.6962), np.float32(0.5367), np.float32(0.0), np.float32(0.0), np.float32(0.689), np.float32(0.4359), np.float32(0.8084), np.float32(0.5374), np.float32(0.3086), np.float32(0.6912), np.float32(0.5484), np.float32(0.6076), np.float32(0.6928), np.float32(0.6364), np.float32(0.3285), np.float32(0.7997), np.float32(0.0), np.float32(0.5561), np.float32(0.6579), np.float32(0.4886), np.float32(0.8213), np.float32(0.7122), np.float32(0.734), np.float32(0.8369)] 
+2026-10-18 14:27:35.394803: Epoch time: 109.17 s 
+2026-10-18 14:27:36.111558:  
+2026-10-18 14:27:36.112243: Epoch 209 
+2026-10-18 14:27:36.112537: Current learning rate: 0.0081 
+2026-10-18 14:29:25.279727: train_loss -0.6162 
+2026-10-18 14:29:25.280436: val_loss -0.5204 
+2026-10-18 14:29:25.281141: Pseudo dice [np.float32(0.927), np.float32(0.7325), np.float32(0.8685), np.float32(0.9015), np.float32(0.805), np.float32(0.9122), np.float32(0.8438), np.float32(0.6182), np.float32(0.0), np.float32(0.4866), np.float32(0.7995), np.float32(0.8261), np.float32(0.5425), np.float32(0.5707), np.float32(0.0), np.float32(0.0), np.float32(0.6542), np.float32(0.1866), np.float32(0.7437), np.float32(0.4914), np.float32(0.2473), np.float32(0.6573), np.float32(0.7203), np.float32(0.7885), np.float32(0.6878), np.float32(0.6799), np.float32(0.3313), np.float32(0.7856), np.float32(0.0), np.float32(0.6044), np.float32(0.6796), np.float32(0.2513), np.float32(0.8544), np.float32(0.7061), np.float32(0.9135), np.float32(0.9455)] 
+2026-10-18 14:29:25.281474: Epoch time: 109.17 s 
+2026-10-18 14:29:26.011257:  
+2026-10-18 14:29:26.011829: Epoch 210 
+2026-10-18 14:29:26.012155: Current learning rate: 0.00809 
+2026-10-18 14:31:15.082058: train_loss -0.6284 
+2026-10-18 14:31:15.082945: val_loss -0.536 
+2026-10-18 14:31:15.083534: Pseudo dice [np.float32(0.9311), np.float32(0.7915), np.float32(0.8879), np.float32(0.8346), np.float32(0.7509), np.float32(0.9206), np.float32(0.8304), np.float32(0.6304), np.float32(0.0), np.float32(0.4408), np.float32(0.8048), np.float32(0.8304), np.float32(0.7012), np.float32(0.6289), np.float32(0.0), np.float32(0.0), np.float32(0.7518), np.float32(0.3896), np.float32(0.808), np.float32(0.6611), np.float32(0.4128), np.float32(0.7448), np.float32(0.687), np.float32(0.7152), np.float32(0.7263), np.float32(0.7117), np.float32(0.3545), np.float32(0.8051), np.float32(0.0), np.float32(0.6732), np.float32(0.568), np.float32(0.2996), np.float32(0.808), np.float32(0.7386), np.float32(0.879), np.float32(0.9193)] 
+2026-10-18 14:31:15.083898: Epoch time: 109.07 s 
+2026-10-18 14:31:15.796383:  
+2026-10-18 14:31:15.797126: Epoch 211 
+2026-10-18 14:31:15.797490: Current learning rate: 0.00808 
+2026-10-18 14:33:04.933097: train_loss -0.6393 
+2026-10-18 14:33:04.934026: val_loss -0.5063 
+2026-10-18 14:33:04.934608: Pseudo dice [np.float32(0.9422), np.float32(0.7068), np.float32(0.8246), np.float32(0.7608), np.float32(0.6651), np.float32(0.7712), np.float32(0.711), np.float32(0.455), np.float32(0.0), np.float32(0.4267), np.float32(0.7104), np.float32(0.7349), np.float32(0.6218), np.float32(0.2825), np.float32(0.0), np.float32(0.0), np.float32(0.7048), np.float32(0.4585), np.float32(0.6594), np.float32(0.5325), np.float32(0.4515), np.float32(0.6723), np.float32(0.8385), np.float32(0.8649), np.float32(0.7327), np.float32(0.6633), np.float32(0.4813), np.float32(0.7593), np.float32(0.0), np.float32(0.6513), np.float32(0.7381), np.float32(0.234), np.float32(0.7803), np.float32(0.6407), np.float32(0.8894), np.float32(0.8891)] 
+2026-10-18 14:33:04.935024: Epoch time: 109.14 s 
+2026-10-18 14:33:05.692095:  
+2026-10-18 14:33:05.692703: Epoch 212 
+2026-10-18 14:33:05.692985: Current learning rate: 0.00807 
+2026-10-18 14:34:54.829040: train_loss -0.6351 
+2026-10-18 14:34:54.829797: val_loss -0.5037 
+2026-10-18 14:34:54.830594: Pseudo dice [np.float32(0.923), np.float32(0.68), np.float32(0.8421), np.float32(0.8879), np.float32(0.6678), np.float32(0.9009), np.float32(0.753), np.float32(0.3776), np.float32(0.0), np.float32(0.5292), np.float32(0.7836), np.float32(0.8103), np.float32(0.6583), np.float32(0.6028), np.float32(0.0), np.float32(0.0), np.float32(0.6926), np.float32(0.4504), np.float32(0.7463), np.float32(0.5487), np.float32(0.3966), np.float32(0.7332), np.float32(0.5963), np.float32(0.6979), np.float32(0.7643), np.float32(0.6708), np.float32(0.3318), np.float32(0.7428), np.float32(0.0), np.float32(0.4343), np.float32(0.5641), np.float32(0.2797), np.float32(0.8105), np.float32(0.7652), np.float32(0.8447), np.float32(0.8708)] 
+2026-10-18 14:34:54.831106: Epoch time: 109.14 s 
+2026-10-18 14:34:55.825655:  
+2026-10-18 14:34:55.826335: Epoch 213 
+2026-10-18 14:34:55.826780: Current learning rate: 0.00806 
+2026-10-18 14:36:45.027648: train_loss -0.6191 
+2026-10-18 14:36:45.028217: val_loss -0.512 
+2026-10-18 14:36:45.029436: Pseudo dice [np.float32(0.9204), np.float32(0.7542), np.float32(0.8634), np.float32(0.8754), np.float32(0.7021), np.float32(0.9166), np.float32(0.8091), np.float32(0.6109), np.float32(0.0), np.float32(0.418), np.float32(0.7735), np.float32(0.7967), np.float32(0.5268), np.float32(0.5837), np.float32(0.0), np.float32(0.0), np.float32(0.7686), np.float32(0.53), np.float32(0.7062), np.float32(0.6362), np.float32(0.3695), np.float32(0.7246), np.float32(0.702), np.float32(0.692), np.float32(0.7602), np.float32(0.6776), np.float32(0.4221), np.float32(0.8115), np.float32(0.0), np.float32(0.4332), np.float32(0.6843), np.float32(0.5125), np.float32(0.7872), np.float32(0.7226), np.float32(0.8303), np.float32(0.8715)] 
+2026-10-18 14:36:45.029848: Epoch time: 109.2 s 
+2026-10-18 14:36:45.760435:  
+2026-10-18 14:36:45.761315: Epoch 214 
+2026-10-18 14:36:45.761787: Current learning rate: 0.00805 
+2026-10-18 14:38:34.914954: train_loss -0.5935 
+2026-10-18 14:38:34.915886: val_loss -0.4891 
+2026-10-18 14:38:34.917038: Pseudo dice [np.float32(0.8806), np.float32(0.6774), np.float32(0.8628), np.float32(0.8635), np.float32(0.6176), np.float32(0.8976), np.float32(0.8327), np.float32(0.2449), np.float32(0.0), np.float32(0.4125), np.float32(0.7579), np.float32(0.7908), np.float32(0.6535), np.float32(0.6272), np.float32(0.0), np.float32(0.0), np.float32(0.6774), np.float32(0.3822), np.float32(0.7557), np.float32(0.5544), np.float32(0.2492), np.float32(0.7578), np.float32(0.6389), np.float32(0.8156), np.float32(0.6217), np.float32(0.6958), np.float32(0.4029), np.float32(0.6846), np.float32(0.0), np.float32(0.6479), np.float32(0.4294), np.float32(0.4111), np.float32(0.7625), np.float32(0.7112), np.float32(0.8818), np.float32(0.9252)] 
+2026-10-18 14:38:34.917796: Epoch time: 109.16 s 
+2026-10-18 14:38:35.663323:  
+2026-10-18 14:38:35.663976: Epoch 215 
+2026-10-18 14:38:35.664360: Current learning rate: 0.00804 
+2026-10-18 14:40:24.875757: train_loss -0.606 
+2026-10-18 14:40:24.876461: val_loss -0.4947 
+2026-10-18 14:40:24.876977: Pseudo dice [np.float32(0.9345), np.float32(0.6453), np.float32(0.8089), np.float32(0.8387), np.float32(0.6505), np.float32(0.8737), np.float32(0.7618), np.float32(0.4219), np.float32(0.0), np.float32(0.3515), np.float32(0.7648), np.float32(0.7687), np.float32(0.6115), np.float32(0.5161), np.float32(0.0), np.float32(0.0), np.float32(0.6988), np.float32(0.3157), np.float32(0.6776), np.float32(0.5243), np.float32(0.2791), np.float32(0.6803), np.float32(0.801), np.float32(0.8142), np.float32(0.6584), np.float32(0.7114), np.float32(0.4059), np.float32(0.861), np.float32(0.0), np.float32(0.6548), np.float32(0.5694), np.float32(0.4782), np.float32(0.8219), np.float32(0.7129), np.float32(0.8722), np.float32(0.9004)] 
+2026-10-18 14:40:24.877626: Epoch time: 109.21 s 
+2026-10-18 14:40:25.596479:  
+2026-10-18 14:40:25.597256: Epoch 216 
+2026-10-18 14:40:25.597595: Current learning rate: 0.00803 
+2026-10-18 14:42:14.762887: train_loss -0.6358 
+2026-10-18 14:42:14.763592: val_loss -0.5216 
+2026-10-18 14:42:14.764339: Pseudo dice [np.float32(0.9395), np.float32(0.7029), np.float32(0.8708), np.float32(0.856), np.float32(0.6653), np.float32(0.8984), np.float32(0.8279), np.float32(0.3841), np.float32(0.0), np.float32(0.5212), np.float32(0.7942), np.float32(0.8267), np.float32(0.6433), np.float32(0.453), np.float32(0.0), np.float32(0.0), np.float32(0.7607), np.float32(0.3984), np.float32(0.7994), np.float32(0.3541), np.float32(0.38), np.float32(0.6711), np.float32(0.7946), np.float32(0.8526), np.float32(0.6585), np.float32(0.7419), np.float32(0.4213), np.float32(0.8309), np.float32(0.0), np.float32(0.6516), np.float32(0.6461), np.float32(0.1961), np.float32(0.7858), np.float32(0.6826), np.float32(0.928), np.float32(0.9493)] 
+2026-10-18 14:42:14.764932: Epoch time: 109.17 s 
+2026-10-18 14:42:15.502759:  
+2026-10-18 14:42:15.503512: Epoch 217 
+2026-10-18 14:42:15.503807: Current learning rate: 0.00802 
+2026-10-18 14:44:04.692869: train_loss -0.5918 
+2026-10-18 14:44:04.693465: val_loss -0.5084 
+2026-10-18 14:44:04.694450: Pseudo dice [np.float32(0.9272), np.float32(0.7169), np.float32(0.8722), np.float32(0.8447), np.float32(0.592), np.float32(0.9015), np.float32(0.7996), np.float32(0.4242), np.float32(0.0), np.float32(0.3811), np.float32(0.7266), np.float32(0.7075), np.float32(0.6713), np.float32(0.6744), np.float32(0.0), np.float32(0.0), np.float32(0.7433), np.float32(0.4262), np.float32(0.7003), np.float32(0.5513), np.float32(0.407), np.float32(0.731), np.float32(0.6723), np.float32(0.6976), np.float32(0.7414), np.float32(0.7565), np.float32(0.2921), np.float32(0.8162), np.float32(0.0), np.float32(0.5166), np.float32(0.6692), np.float32(0.3986), np.float32(0.7252), np.float32(0.7388), np.float32(0.8412), np.float32(0.8572)] 
+2026-10-18 14:44:04.695110: Epoch time: 109.19 s 
+2026-10-18 14:44:05.450507:  
+2026-10-18 14:44:05.451156: Epoch 218 
+2026-10-18 14:44:05.451502: Current learning rate: 0.00801 
+2026-10-18 14:45:54.640360: train_loss -0.608 
+2026-10-18 14:45:54.640834: val_loss -0.4719 
+2026-10-18 14:45:54.641377: Pseudo dice [np.float32(0.932), np.float32(0.6954), np.float32(0.832), np.float32(0.7808), np.float32(0.4801), np.float32(0.8763), np.float32(0.8319), np.float32(0.2159), np.float32(0.0), np.float32(0.3371), np.float32(0.6996), np.float32(0.7543), np.float32(0.5454), np.float32(0.5599), np.float32(0.0), np.float32(0.0), np.float32(0.5557), np.float32(0.1909), np.float32(0.7248), np.float32(0.5784), np.float32(0.411), np.float32(0.7047), np.float32(0.5019), np.float32(0.5935), np.float32(0.6831), np.float32(0.7075), np.float32(0.5327), np.float32(0.8314), np.float32(0.0), np.float32(0.5894), np.float32(0.6765), np.float32(0.4172), np.float32(0.7128), np.float32(0.6671), np.float32(0.781), np.float32(0.874)] 
+2026-10-18 14:45:54.642175: Epoch time: 109.19 s 
+2026-10-18 14:45:55.365224:  
+2026-10-18 14:45:55.365684: Epoch 219 
+2026-10-18 14:45:55.365808: Current learning rate: 0.00801 
+2026-10-18 14:47:44.533114: train_loss -0.6241 
+2026-10-18 14:47:44.534107: val_loss -0.5219 
+2026-10-18 14:47:44.534697: Pseudo dice [np.float32(0.9383), np.float32(0.7325), np.float32(0.8535), np.float32(0.8419), np.float32(0.7591), np.float32(0.8877), np.float32(0.8295), np.float32(0.4713), np.float32(0.0), np.float32(0.4823), np.float32(0.7669), np.float32(0.7879), np.float32(0.5199), np.float32(0.5169), np.float32(0.0), np.float32(0.0), np.float32(0.6521), np.float32(0.3602), np.float32(0.742), np.float32(0.5038), np.float32(0.4853), np.float32(0.682), np.float32(0.7414), np.float32(0.7614), np.float32(0.7041), np.float32(0.7325), np.float32(0.3419), np.float32(0.6861), np.float32(0.0), np.float32(0.666), np.float32(0.6047), np.float32(0.394), np.float32(0.805), np.float32(0.671), np.float32(0.8911), np.float32(0.915)] 
+2026-10-18 14:47:44.535098: Epoch time: 109.17 s 
+2026-10-18 14:47:45.272470:  
+2026-10-18 14:47:45.273243: Epoch 220 
+2026-10-18 14:47:45.273450: Current learning rate: 0.008 
+2026-10-18 14:49:34.419091: train_loss -0.6193 
+2026-10-18 14:49:34.419739: val_loss -0.5541 
+2026-10-18 14:49:34.420242: Pseudo dice [np.float32(0.9381), np.float32(0.7741), np.float32(0.8861), np.float32(0.8297), np.float32(0.6674), np.float32(0.8774), np.float32(0.8455), np.float32(0.3159), np.float32(0.0), np.float32(0.3906), np.float32(0.8057), np.float32(0.8206), np.float32(0.6994), np.float32(0.6484), np.float32(0.0), np.float32(0.0), np.float32(0.7668), np.float32(0.5406), np.float32(0.7548), np.float32(0.5569), np.float32(0.4708), np.float32(0.6767), np.float32(0.6375), np.float32(0.7592), np.float32(0.7726), np.float32(0.7482), np.float32(0.42), np.float32(0.8802), np.float32(0.0), np.float32(0.5394), np.float32(0.6099), np.float32(0.4686), np.float32(0.8156), np.float32(0.6315), np.float32(0.8937), np.float32(0.9208)] 
+2026-10-18 14:49:34.420893: Epoch time: 109.15 s 
+2026-10-18 14:49:35.263284:  
+2026-10-18 14:49:35.264442: Epoch 221 
+2026-10-18 14:49:35.264666: Current learning rate: 0.00799 
+2026-10-18 14:51:24.427629: train_loss -0.6164 
+2026-10-18 14:51:24.428380: val_loss -0.5119 
+2026-10-18 14:51:24.428898: Pseudo dice [np.float32(0.9336), np.float32(0.7411), np.float32(0.8252), np.float32(0.8684), np.float32(0.7172), np.float32(0.916), np.float32(0.8548), np.float32(0.6367), np.float32(0.0), np.float32(0.3753), np.float32(0.7344), np.float32(0.7845), np.float32(0.6354), np.float32(0.6138), np.float32(0.0), np.float32(0.0), np.float32(0.7696), np.float32(0.4271), np.float32(0.7336), np.float32(0.5299), np.float32(0.2272), np.float32(0.5864), np.float32(0.7033), np.float32(0.6861), np.float32(0.7143), np.float32(0.6941), np.float32(0.3569), np.float32(0.6979), np.float32(0.0), np.float32(0.641), np.float32(0.72), np.float32(0.3983), np.float32(0.7771), np.float32(0.7024), np.float32(0.9082), np.float32(0.9238)] 
+2026-10-18 14:51:24.429696: Epoch time: 109.17 s 
+2026-10-18 14:51:25.167755:  
+2026-10-18 14:51:25.168628: Epoch 222 
+2026-10-18 14:51:25.169014: Current learning rate: 0.00798 
+2026-10-18 14:53:14.294781: train_loss -0.6456 
+2026-10-18 14:53:14.295292: val_loss -0.5496 
+2026-10-18 14:53:14.295797: Pseudo dice [np.float32(0.9399), np.float32(0.765), np.float32(0.8824), np.float32(0.8198), np.float32(0.673), np.float32(0.9291), np.float32(0.84), np.float32(0.5172), np.float32(0.0), np.float32(0.5065), np.float32(0.773), np.float32(0.8388), np.float32(0.6538), np.float32(0.6301), np.float32(0.0), np.float32(0.0), np.float32(0.7722), np.float32(0.5632), np.float32(0.7978), np.float32(0.5349), np.float32(0.3486), np.float32(0.7618), np.float32(0.7267), np.float32(0.7885), np.float32(0.7027), np.float32(0.7097), np.float32(0.4782), np.float32(0.8001), np.float32(0.0), np.float32(0.6438), np.float32(0.7404), np.float32(0.5381), np.float32(0.7255), np.float32(0.6728), np.float32(0.9), np.float32(0.9381)] 
+2026-10-18 14:53:14.296439: Epoch time: 109.13 s 
+2026-10-18 14:53:15.041410:  
+2026-10-18 14:53:15.042254: Epoch 223 
+2026-10-18 14:53:15.042457: Current learning rate: 0.00797 
+2026-10-18 14:55:04.132410: train_loss -0.6385 
+2026-10-18 14:55:04.132909: val_loss -0.5395 
+2026-10-18 14:55:04.134087: Pseudo dice [np.float32(0.9364), np.float32(0.7611), np.float32(0.8773), np.float32(0.8796), np.float32(0.7216), np.float32(0.938), np.float32(0.8357), np.float32(0.5904), np.float32(0.0), np.float32(0.4423), np.float32(0.8072), np.float32(0.8341), np.float32(0.6981), np.float32(0.6159), np.float32(0.0), np.float32(0.0), np.float32(0.7828), np.float32(0.4391), np.float32(0.8053), np.float32(0.5838), np.float32(0.3069), np.float32(0.7278), np.float32(0.7263), np.float32(0.7103), np.float32(0.6986), np.float32(0.7479), np.float32(0.336), np.float32(0.7943), np.float32(0.0), np.float32(0.3077), np.float32(0.7276), np.float32(0.5322), np.float32(0.7812), np.float32(0.7043), np.float32(0.9099), np.float32(0.9271)] 
+2026-10-18 14:55:04.134502: Epoch time: 109.09 s 
+2026-10-18 14:55:04.883930:  
+2026-10-18 14:55:04.884639: Epoch 224 
+2026-10-18 14:55:04.885014: Current learning rate: 0.00796 
+2026-10-18 14:56:53.940628: train_loss -0.6211 
+2026-10-18 14:56:53.941405: val_loss -0.5146 
+2026-10-18 14:56:53.941852: Pseudo dice [np.float32(0.9429), np.float32(0.7256), np.float32(0.7942), np.float32(0.8524), np.float32(0.5865), np.float32(0.8938), np.float32(0.8268), np.float32(0.4782), np.float32(0.0), np.float32(0.2907), np.float32(0.7456), np.float32(0.7774), np.float32(0.6783), np.float32(0.6966), np.float32(0.0), np.float32(0.0), np.float32(0.5755), np.float32(0.2452), np.float32(0.7495), np.float32(0.5436), np.float32(0.2161), np.float32(0.5874), np.float32(0.674), np.float32(0.7821), np.float32(0.6716), np.float32(0.708), np.float32(0.3609), np.float32(0.7198), np.float32(0.0), np.float32(0.4597), np.float32(0.7417), np.float32(0.3365), np.float32(0.8064), np.float32(0.6556), np.float32(0.8636), np.float32(0.9086)] 
+2026-10-18 14:56:53.942322: Epoch time: 109.06 s 
+2026-10-18 14:56:54.689366:  
+2026-10-18 14:56:54.689907: Epoch 225 
+2026-10-18 14:56:54.690310: Current learning rate: 0.00795 
+2026-10-18 14:58:43.843962: train_loss -0.6139 
+2026-10-18 14:58:43.844873: val_loss -0.5392 
+2026-10-18 14:58:43.845460: Pseudo dice [np.float32(0.9385), np.float32(0.752), np.float32(0.8909), np.float32(0.8482), np.float32(0.7107), np.float32(0.9314), np.float32(0.8634), np.float32(0.5671), np.float32(0.0), np.float32(0.3853), np.float32(0.7754), np.float32(0.8608), np.float32(0.7046), np.float32(0.5702), np.float32(0.0), np.float32(0.0), np.float32(0.7377), np.float32(0.5042), np.float32(0.8398), np.float32(0.5861), np.float32(0.496), np.float32(0.7373), np.float32(0.6599), np.float32(0.5987), np.float32(0.694), np.float32(0.7435), np.float32(0.3508), np.float32(0.7389), np.float32(0.0), np.float32(0.4864), np.float32(0.7049), np.float32(0.4249), np.float32(0.8028), np.float32(0.7066), np.float32(0.8226), np.float32(0.8686)] 
+2026-10-18 14:58:43.845722: Epoch time: 109.16 s 
+2026-10-18 14:58:44.551124:  
+2026-10-18 14:58:44.551849: Epoch 226 
+2026-10-18 14:58:44.552208: Current learning rate: 0.00794 
+2026-10-18 15:00:33.643116: train_loss -0.6351 
+2026-10-18 15:00:33.643834: val_loss -0.438 
+2026-10-18 15:00:33.644723: Pseudo dice [np.float32(0.9136), np.float32(0.6816), np.float32(0.7798), np.float32(0.7455), np.float32(0.6171), np.float32(0.8499), np.float32(0.7771), np.float32(0.5708), np.float32(0.0), np.float32(0.3637), np.float32(0.6616), np.float32(0.7401), np.float32(0.6277), np.float32(0.5224), np.float32(0.0), np.float32(0.0), np.float32(0.4846), np.float32(0.0603), np.float32(0.6238), np.float32(0.5278), np.float32(0.1362), np.float32(0.491), np.float32(0.4875), np.float32(0.5163), np.float32(0.6869), np.float32(0.6285), np.float32(0.314), np.float32(0.6419), np.float32(0.0), np.float32(0.2668), np.float32(0.6316), np.float32(0.3124), np.float32(0.7286), np.float32(0.6952), np.float32(0.7216), np.float32(0.6963)] 
+2026-10-18 15:00:33.645358: Epoch time: 109.09 s 
+2026-10-18 15:00:34.363965:  
+2026-10-18 15:00:34.364718: Epoch 227 
+2026-10-18 15:00:34.365137: Current learning rate: 0.00793 
+2026-10-18 15:02:23.511601: train_loss -0.5991 
+2026-10-18 15:02:23.512251: val_loss -0.4605 
+2026-10-18 15:02:23.512973: Pseudo dice [np.float32(0.9315), np.float32(0.7142), np.float32(0.8276), np.float32(0.6972), np.float32(0.4191), np.float32(0.7965), np.float32(0.7473), np.float32(0.4553), np.float32(0.0), np.float32(0.2106), np.float32(0.646), np.float32(0.6748), np.float32(0.677), np.float32(0.39), np.float32(0.0), np.float32(0.0), np.float32(0.7045), np.float32(0.3612), np.float32(0.6889), np.float32(0.4693), np.float32(0.3895), np.float32(0.7536), np.float32(0.593), np.float32(0.6994), np.float32(0.6699), np.float32(0.621), np.float32(0.2807), np.float32(0.5661), np.float32(0.0), np.float32(0.7485), np.float32(0.6885), np.float32(0.4184), np.float32(0.6351), np.float32(0.5255), np.float32(0.7565), np.float32(0.8009)] 
+2026-10-18 15:02:23.513360: Epoch time: 109.15 s 
+2026-10-18 15:02:24.229319:  
+2026-10-18 15:02:24.230052: Epoch 228 
+2026-10-18 15:02:24.230474: Current learning rate: 0.00792 
+2026-10-18 15:04:13.365730: train_loss -0.5997 
+2026-10-18 15:04:13.366327: val_loss -0.396 
+2026-10-18 15:04:13.366872: Pseudo dice [np.float32(0.9298), np.float32(0.5904), np.float32(0.7133), np.float32(0.5568), np.float32(0.5021), np.float32(0.6877), np.float32(0.6954), np.float32(0.1791), np.float32(0.0), np.float32(0.375), np.float32(0.5497), np.float32(0.6343), np.float32(0.4578), np.float32(0.4746), np.float32(0.0), np.float32(0.0), np.float32(0.5124), np.float32(0.2477), np.float32(0.5583), np.float32(0.3414), np.float32(0.2945), np.float32(0.6174), np.float32(0.4327), np.float32(0.4816), np.float32(0.5655), np.float32(0.4845), np.float32(0.1765), np.float32(0.3581), np.float32(0.0), np.float32(0.292), np.float32(0.5475), np.float32(0.354), np.float32(0.5576), np.float32(0.4698), np.float32(0.633), np.float32(0.7024)] 
+2026-10-18 15:04:13.367557: Epoch time: 109.14 s 
+2026-10-18 15:04:14.083436:  
+2026-10-18 15:04:14.084079: Epoch 229 
+2026-10-18 15:04:14.084520: Current learning rate: 0.00791 
+2026-10-18 15:06:03.196051: train_loss -0.6329 
+2026-10-18 15:06:03.196966: val_loss -0.4736 
+2026-10-18 15:06:03.197431: Pseudo dice [np.float32(0.9403), np.float32(0.6588), np.float32(0.7416), np.float32(0.6367), np.float32(0.6045), np.float32(0.6853), np.float32(0.692), np.float32(0.6024), np.float32(0.0), np.float32(0.4178), np.float32(0.5968), np.float32(0.6259), np.float32(0.6633), np.float32(0.5398), np.float32(0.0), np.float32(0.0), np.float32(0.7239), np.float32(0.4307), np.float32(0.7503), np.float32(0.5873), np.float32(0.3509), np.float32(0.7406), np.float32(0.7948), np.float32(0.7991), np.float32(0.6371), np.float32(0.6054), np.float32(0.3408), np.float32(0.6616), np.float32(0.0), np.float32(0.6118), np.float32(0.5859), np.float32(0.1531), np.float32(0.6135), np.float32(0.5137), np.float32(0.828), np.float32(0.8344)] 
+2026-10-18 15:06:03.197844: Epoch time: 109.11 s 
+2026-10-18 15:06:03.911154:  
+2026-10-18 15:06:03.911858: Epoch 230 
+2026-10-18 15:06:03.912089: Current learning rate: 0.0079 
+2026-10-18 15:07:53.026953: train_loss -0.6315 
+2026-10-18 15:07:53.027697: val_loss -0.5248 
+2026-10-18 15:07:53.028362: Pseudo dice [np.float32(0.9449), np.float32(0.733), np.float32(0.8343), np.float32(0.7924), np.float32(0.7076), np.float32(0.8658), np.float32(0.8098), np.float32(0.6615), np.float32(0.0), np.float32(0.4144), np.float32(0.6954), np.float32(0.7721), np.float32(0.4786), np.float32(0.4337), np.float32(0.0), np.float32(0.0), np.float32(0.7169), np.float32(0.4533), np.float32(0.7654), np.float32(0.6383), np.float32(0.352), np.float32(0.6624), np.float32(0.7866), np.float32(0.8006), np.float32(0.6778), np.float32(0.6474), np.float32(0.3671), np.float32(0.7079), np.float32(0.0), np.float32(0.6199), np.float32(0.7137), np.float32(0.425), np.float32(0.6807), np.float32(0.6133), np.float32(0.8648), np.float32(0.8933)] 
+2026-10-18 15:07:53.028741: Epoch time: 109.12 s 
+2026-10-18 15:07:53.764774:  
+2026-10-18 15:07:53.765558: Epoch 231 
+2026-10-18 15:07:53.765972: Current learning rate: 0.00789 
+2026-10-18 15:09:42.916759: train_loss -0.6109 
+2026-10-18 15:09:42.917293: val_loss -0.4934 
+2026-10-18 15:09:42.917863: Pseudo dice [np.float32(0.91), np.float32(0.7057), np.float32(0.8461), np.float32(0.8668), np.float32(0.743), np.float32(0.8892), np.float32(0.7702), np.float32(0.3649), np.float32(0.0), np.float32(0.5515), np.float32(0.7879), np.float32(0.7941), np.float32(0.6525), np.float32(0.4777), np.float32(0.0), np.float32(0.0), np.float32(0.8005), np.float32(0.4557), np.float32(0.7889), np.float32(0.4664), np.float32(0.3237), np.float32(0.6517), np.float32(0.3626), np.float32(0.4626), np.float32(0.6343), np.float32(0.6864), np.float32(0.303), np.float32(0.6112), np.float32(0.0), np.float32(0.4017), np.float32(0.7142), np.float32(0.4441), np.float32(0.7059), np.float32(0.5973), np.float32(0.8116), np.float32(0.8701)] 
+2026-10-18 15:09:42.918575: Epoch time: 109.15 s 
+2026-10-18 15:09:43.637097:  
+2026-10-18 15:09:43.637766: Epoch 232 
+2026-10-18 15:09:43.638095: Current learning rate: 0.00789 
+2026-10-18 15:11:32.760794: train_loss -0.6136 
+2026-10-18 15:11:32.761519: val_loss -0.5422 
+2026-10-18 15:11:32.762050: Pseudo dice [np.float32(0.9295), np.float32(0.7544), np.float32(0.8423), np.float32(0.8775), np.float32(0.7358), np.float32(0.9248), np.float32(0.8594), np.float32(0.4922), np.float32(0.0), np.float32(0.5044), np.float32(0.8157), np.float32(0.8076), np.float32(0.6013), np.float32(0.5439), np.float32(0.0), np.float32(0.0), np.float32(0.8029), np.float32(0.5111), np.float32(0.81), np.float32(0.4263), np.float32(0.4163), np.float32(0.708), np.float32(0.645), np.float32(0.7819), np.float32(0.6363), np.float32(0.6539), np.float32(0.46), np.float32(0.8425), np.float32(0.0), np.float32(0.6513), np.float32(0.6994), np.float32(0.3985), np.float32(0.8075), np.float32(0.6839), np.float32(0.892), np.float32(0.9405)] 
+2026-10-18 15:11:32.762675: Epoch time: 109.13 s 
+2026-10-18 15:11:33.505810:  
+2026-10-18 15:11:33.506492: Epoch 233 
+2026-10-18 15:11:33.506795: Current learning rate: 0.00788 
+2026-10-18 15:13:22.612271: train_loss -0.5588 
+2026-10-18 15:13:22.612887: val_loss -0.4891 
+2026-10-18 15:13:22.613445: Pseudo dice [np.float32(0.9323), np.float32(0.7357), np.float32(0.8357), np.float32(0.8644), np.float32(0.6663), np.float32(0.9033), np.float32(0.7905), np.float32(0.5246), np.float32(0.0014), np.float32(0.405), np.float32(0.7516), np.float32(0.7741), np.float32(0.4713), np.float32(0.6248), np.float32(0.0), np.float32(0.0), np.float32(0.6326), np.float32(0.1953), np.float32(0.6539), np.float32(0.3678), np.float32(0.0785), np.float32(0.664), np.float32(0.5946), np.float32(0.6994), np.float32(0.6585), np.float32(0.7294), np.float32(0.381), np.float32(0.5773), np.float32(0.0), np.float32(0.6844), np.float32(0.6277), np.float32(0.4842), np.float32(0.7813), np.float32(0.7014), np.float32(0.8479), np.float32(0.8933)] 
+2026-10-18 15:13:22.613857: Epoch time: 109.11 s 
+2026-10-18 15:13:23.593569:  
+2026-10-18 15:13:23.594274: Epoch 234 
+2026-10-18 15:13:23.594557: Current learning rate: 0.00787 
+2026-10-18 15:15:12.794915: train_loss -0.6147 
+2026-10-18 15:15:12.795483: val_loss -0.4832 
+2026-10-18 15:15:12.796491: Pseudo dice [np.float32(0.9319), np.float32(0.7539), np.float32(0.8387), np.float32(0.8215), np.float32(0.7227), np.float32(0.8832), np.float32(0.8403), np.float32(0.4696), np.float32(0.0), np.float32(0.3757), np.float32(0.7203), np.float32(0.7707), np.float32(0.6123), np.float32(0.5234), np.float32(0.0), np.float32(0.0), np.float32(0.6878), np.float32(0.3522), np.float32(0.7125), np.float32(0.5107), np.float32(0.4018), np.float32(0.7366), np.float32(0.5411), np.float32(0.4299), np.float32(0.7066), np.float32(0.6721), np.float32(0.1674), np.float32(0.4577), np.float32(0.0), np.float32(0.3936), np.float32(0.6531), np.float32(0.4551), np.float32(0.7707), np.float32(0.6752), np.float32(0.8393), np.float32(0.8756)] 
+2026-10-18 15:15:12.796867: Epoch time: 109.2 s 
+2026-10-18 15:15:13.535495:  
+2026-10-18 15:15:13.536180: Epoch 235 
+2026-10-18 15:15:13.536551: Current learning rate: 0.00786 
+2026-10-18 15:17:02.658559: train_loss -0.5956 
+2026-10-18 15:17:02.659090: val_loss -0.4456 
+2026-10-18 15:17:02.660163: Pseudo dice [np.float32(0.9284), np.float32(0.6976), np.float32(0.7622), np.float32(0.6588), np.float32(0.5366), np.float32(0.7333), np.float32(0.7144), np.float32(0.4066), np.float32(0.0), np.float32(0.4046), np.float32(0.6241), np.float32(0.6604), np.float32(0.6347), np.float32(0.4809), np.float32(0.0), np.float32(0.0), np.float32(0.6609), np.float32(0.4223), np.float32(0.6374), np.float32(0.5679), np.float32(0.4344), np.float32(0.7183), np.float32(0.6218), np.float32(0.5836), np.float32(0.5646), np.float32(0.5887), np.float32(0.3869), np.float32(0.5498), np.float32(0.0), np.float32(0.5191), np.float32(0.6533), np.float32(0.2682), np.float32(0.62), np.float32(0.4422), np.float32(0.712), np.float32(0.7613)] 
+2026-10-18 15:17:02.660635: Epoch time: 109.13 s 
+2026-10-18 15:17:03.381157:  
+2026-10-18 15:17:03.381902: Epoch 236 
+2026-10-18 15:17:03.382200: Current learning rate: 0.00785 
+2026-10-18 15:18:52.568776: train_loss -0.6189 
+2026-10-18 15:18:52.569316: val_loss -0.5041 
+2026-10-18 15:18:52.569855: Pseudo dice [np.float32(0.9078), np.float32(0.762), np.float32(0.796), np.float32(0.8211), np.float32(0.65), np.float32(0.8891), np.float32(0.7676), np.float32(0.5611), np.float32(0.0), np.float32(0.3918), np.float32(0.7252), np.float32(0.7773), np.float32(0.6337), np.float32(0.5219), np.float32(0.0), np.float32(0.0), np.float32(0.7494), np.float32(0.4771), np.float32(0.7878), np.float32(0.5652), np.float32(0.5091), np.float32(0.774), np.float32(0.5621), np.float32(0.6881), np.float32(0.7196), np.float32(0.6664), np.float32(0.4183), np.float32(0.7383), np.float32(0.0), np.float32(0.5748), np.float32(0.548), np.float32(0.2137), np.float32(0.7289), np.float32(0.6635), np.float32(0.8206), np.float32(0.8603)] 
+2026-10-18 15:18:52.570898: Epoch time: 109.19 s 
+2026-10-18 15:18:53.308478:  
+2026-10-18 15:18:53.309262: Epoch 237 
+2026-10-18 15:18:53.309498: Current learning rate: 0.00784 
+2026-10-18 15:20:42.471598: train_loss -0.6359 
+2026-10-18 15:20:42.472482: val_loss -0.5275 
+2026-10-18 15:20:42.473057: Pseudo dice [np.float32(0.9255), np.float32(0.6882), np.float32(0.8329), np.float32(0.847), np.float32(0.7149), np.float32(0.8969), np.float32(0.8356), np.float32(0.6165), np.float32(0.0), np.float32(0.4586), np.float32(0.7752), np.float32(0.7973), np.float32(0.6611), np.float32(0.4607), np.float32(0.0), np.float32(0.0), np.float32(0.7721), np.float32(0.5217), np.float32(0.7811), np.float32(0.467), np.float32(0.1868), np.float32(0.5777), np.float32(0.7136), np.float32(0.7979), np.float32(0.6173), np.float32(0.6979), np.float32(0.3509), np.float32(0.7552), np.float32(0.0), np.float32(0.733), np.float32(0.7736), np.float32(0.3345), np.float32(0.768), np.float32(0.6423), np.float32(0.8843), np.float32(0.9009)] 
+2026-10-18 15:20:42.473537: Epoch time: 109.17 s 
+2026-10-18 15:20:43.199433:  
+2026-10-18 15:20:43.200168: Epoch 238 
+2026-10-18 15:20:43.200584: Current learning rate: 0.00783 
+2026-10-18 15:22:32.379778: train_loss -0.6375 
+2026-10-18 15:22:32.380519: val_loss -0.5362 
+2026-10-18 15:22:32.381017: Pseudo dice [np.float32(0.922), np.float32(0.7417), np.float32(0.8911), np.float32(0.8975), np.float32(0.7575), np.float32(0.922), np.float32(0.8653), np.float32(0.3635), np.float32(0.0), np.float32(0.5539), np.float32(0.8151), np.float32(0.8227), np.float32(0.6736), np.float32(0.5919), np.float32(0.0), np.float32(0.0), np.float32(0.7999), np.float32(0.5618), np.float32(0.8351), np.float32(0.6173), np.float32(0.4362), np.float32(0.7472), np.float32(0.5142), np.float32(0.6343), np.float32(0.7251), np.float32(0.7757), np.float32(0.4705), np.float32(0.6297), np.float32(0.0), np.float32(0.6197), np.float32(0.7567), np.float32(0.3809), np.float32(0.7933), np.float32(0.6358), np.float32(0.8394), np.float32(0.8847)] 
+2026-10-18 15:22:32.381594: Epoch time: 109.18 s 
+2026-10-18 15:22:33.092867:  
+2026-10-18 15:22:33.093603: Epoch 239 
+2026-10-18 15:22:33.094015: Current learning rate: 0.00782 
+2026-10-18 15:24:22.270418: train_loss -0.6211 
+2026-10-18 15:24:22.271057: val_loss -0.5347 
+2026-10-18 15:24:22.271943: Pseudo dice [np.float32(0.916), np.float32(0.7065), np.float32(0.8564), np.float32(0.8625), np.float32(0.7775), np.float32(0.9134), np.float32(0.8359), np.float32(0.7451), np.float32(0.0), np.float32(0.458), np.float32(0.7542), np.float32(0.8158), np.float32(0.5882), np.float32(0.5736), np.float32(0.0), np.float32(0.0), np.float32(0.7182), np.float32(0.5151), np.float32(0.805), np.float32(0.5434), np.float32(0.3889), np.float32(0.6312), np.float32(0.6661), np.float32(0.746), np.float32(0.658), np.float32(0.7835), np.float32(0.4949), np.float32(0.7445), np.float32(0.0), np.float32(0.4745), np.float32(0.6875), np.float32(0.4039), np.float32(0.7835), np.float32(0.6843), np.float32(0.8511), np.float32(0.9054)] 
+2026-10-18 15:24:22.272506: Epoch time: 109.18 s 
+2026-10-18 15:24:23.009622:  
+2026-10-18 15:24:23.010242: Epoch 240 
+2026-10-18 15:24:23.010623: Current learning rate: 0.00781 
+2026-10-18 15:26:12.147019: train_loss -0.6456 
+2026-10-18 15:26:12.147748: val_loss -0.5401 
+2026-10-18 15:26:12.148509: Pseudo dice [np.float32(0.9239), np.float32(0.7498), np.float32(0.8509), np.float32(0.8887), np.float32(0.6946), np.float32(0.8884), np.float32(0.8286), np.float32(0.411), np.float32(0.0), np.float32(0.3076), np.float32(0.7811), np.float32(0.8295), np.float32(0.6262), np.float32(0.4862), np.float32(0.0), np.float32(0.0), np.float32(0.742), np.float32(0.4011), np.float32(0.7754), np.float32(0.5922), np.float32(0.2839), np.float32(0.5627), np.float32(0.8274), np.float32(0.8573), np.float32(0.7001), np.float32(0.7184), np.float32(0.5041), np.float32(0.7828), np.float32(0.0), np.float32(0.6613), np.float32(0.6348), np.float32(0.3989), np.float32(0.7636), np.float32(0.6841), np.float32(0.902), np.float32(0.9233)] 
+2026-10-18 15:26:12.149153: Epoch time: 109.14 s 
+2026-10-18 15:26:12.873729:  
+2026-10-18 15:26:12.874406: Epoch 241 
+2026-10-18 15:26:12.874731: Current learning rate: 0.0078 
+2026-10-18 15:28:02.065376: train_loss -0.6208 
+2026-10-18 15:28:02.066180: val_loss -0.5116 
+2026-10-18 15:28:02.066674: Pseudo dice [np.float32(0.9144), np.float32(0.7374), np.float32(0.8814), np.float32(0.907), np.float32(0.7459), np.float32(0.9362), np.float32(0.8557), np.float32(0.5205), np.float32(0.0), np.float32(0.452), np.float32(0.7893), np.float32(0.8352), np.float32(0.6518), np.float32(0.5688), np.float32(0.0), np.float32(0.0), np.float32(0.7923), np.float32(0.5045), np.float32(0.8103), np.float32(0.5759), np.float32(0.3711), np.float32(0.7091), np.float32(0.5978), np.float32(0.5586), np.float32(0.7121), np.float32(0.7547), np.float32(0.4947), np.float32(0.8101), np.float32(0.0), np.float32(0.568), np.float32(0.5622), np.float32(0.4398), np.float32(0.796), np.float32(0.7143), np.float32(0.8549), np.float32(0.8978)] 
+2026-10-18 15:28:02.067075: Epoch time: 109.19 s 
+2026-10-18 15:28:02.811088:  
+2026-10-18 15:28:02.812151: Epoch 242 
+2026-10-18 15:28:02.812708: Current learning rate: 0.00779 
+2026-10-18 15:29:51.990199: train_loss -0.6386 
+2026-10-18 15:29:51.990797: val_loss -0.5136 
+2026-10-18 15:29:51.991314: Pseudo dice [np.float32(0.9351), np.float32(0.7137), np.float32(0.8267), np.float32(0.8112), np.float32(0.7332), np.float32(0.8927), np.float32(0.8341), np.float32(0.5015), np.float32(0.0), np.float32(0.3598), np.float32(0.722), np.float32(0.7556), np.float32(0.6119), np.float32(0.2338), np.float32(0.0), np.float32(0.0), np.float32(0.8057), np.float32(0.5108), np.float32(0.7938), np.float32(0.5062), np.float32(0.2419), np.float32(0.6956), np.float32(0.8023), np.float32(0.8424), np.float32(0.703), np.float32(0.6813), np.float32(0.4477), np.float32(0.7811), np.float32(0.0), np.float32(0.6287), np.float32(0.58), np.float32(0.529), np.float32(0.8009), np.float32(0.6428), np.float32(0.8792), np.float32(0.9284)] 
+2026-10-18 15:29:51.992035: Epoch time: 109.18 s 
+2026-10-18 15:29:52.745380:  
+2026-10-18 15:29:52.746192: Epoch 243 
+2026-10-18 15:29:52.746571: Current learning rate: 0.00778 
+2026-10-18 15:31:41.930526: train_loss -0.6254 
+2026-10-18 15:31:41.931175: val_loss -0.3424 
+2026-10-18 15:31:41.931772: Pseudo dice [np.float32(0.941), np.float32(0.5178), np.float32(0.5768), np.float32(0.3977), np.float32(0.3313), np.float32(0.4864), np.float32(0.6159), np.float32(0.2489), np.float32(0.0), np.float32(0.4104), np.float32(0.4204), np.float32(0.518), np.float32(0.5691), np.float32(0.4994), np.float32(0.0), np.float32(0.0), np.float32(0.4576), np.float32(0.391), np.float32(0.5453), np.float32(0.4445), np.float32(0.2132), np.float32(0.6771), np.float32(0.7533), np.float32(0.6906), np.float32(0.5001), np.float32(0.3462), np.float32(0.3042), np.float32(0.3922), np.float32(0.0), np.float32(0.4468), np.float32(0.2164), np.float32(0.3104), np.float32(0.301), np.float32(0.2645), np.float32(0.5888), np.float32(0.5462)] 
+2026-10-18 15:31:41.932309: Epoch time: 109.19 s 
+2026-10-18 15:31:42.656456:  
+2026-10-18 15:31:42.657173: Epoch 244 
+2026-10-18 15:31:42.657341: Current learning rate: 0.00777 
+2026-10-18 15:33:31.817403: train_loss -0.5806 
+2026-10-18 15:33:31.817869: val_loss -0.5057 
+2026-10-18 15:33:31.818886: Pseudo dice [np.float32(0.9232), np.float32(0.7658), np.float32(0.837), np.float32(0.8231), np.float32(0.5887), np.float32(0.8886), np.float32(0.8266), np.float32(0.3965), np.float32(0.0), np.float32(0.4502), np.float32(0.7413), np.float32(0.8041), np.float32(0.7196), np.float32(0.567), np.float32(0.0), np.float32(0.0), np.float32(0.7587), np.float32(0.5112), np.float32(0.7803), np.float32(0.4977), np.float32(0.4404), np.float32(0.5898), np.float32(0.6684), np.float32(0.7651), np.float32(0.7044), np.float32(0.5766), np.float32(0.3339), np.float32(0.7454), np.float32(0.0), np.float32(0.4909), np.float32(0.5991), np.float32(0.2554), np.float32(0.772), np.float32(0.642), np.float32(0.8273), np.float32(0.897)] 
+2026-10-18 15:33:31.819223: Epoch time: 109.16 s 
+2026-10-18 15:33:32.573013:  
+2026-10-18 15:33:32.573656: Epoch 245 
+2026-10-18 15:33:32.573983: Current learning rate: 0.00777 
+2026-10-18 15:35:21.764197: train_loss -0.6208 
+2026-10-18 15:35:21.764731: val_loss -0.5149 
+2026-10-18 15:35:21.765501: Pseudo dice [np.float32(0.9145), np.float32(0.7295), np.float32(0.8544), np.float32(0.848), np.float32(0.6889), np.float32(0.8711), np.float32(0.8396), np.float32(0.3028), np.float32(0.0), np.float32(0.4925), np.float32(0.7178), np.float32(0.7731), np.float32(0.6302), np.float32(0.4009), np.float32(0.0), np.float32(0.0), np.float32(0.7571), np.float32(0.4023), np.float32(0.764), np.float32(0.3157), np.float32(0.3689), np.float32(0.6839), np.float32(0.7294), np.float32(0.7445), np.float32(0.7352), np.float32(0.7305), np.float32(0.3859), np.float32(0.7304), np.float32(0.0), np.float32(0.5592), np.float32(0.7773), np.float32(0.5457), np.float32(0.74), np.float32(0.5617), np.float32(0.8765), np.float32(0.8985)] 
+2026-10-18 15:35:21.766156: Epoch time: 109.19 s 
+2026-10-18 15:35:22.554964:  
+2026-10-18 15:35:22.555795: Epoch 246 
+2026-10-18 15:35:22.556083: Current learning rate: 0.00776 
+2026-10-18 15:37:11.691893: train_loss -0.6233 
+2026-10-18 15:37:11.692509: val_loss -0.5503 
+2026-10-18 15:37:11.693058: Pseudo dice [np.float32(0.945), np.float32(0.7283), np.float32(0.8624), np.float32(0.8969), np.float32(0.7696), np.float32(0.9309), np.float32(0.8585), np.float32(0.5293), np.float32(0.0), np.float32(0.4926), np.float32(0.7774), np.float32(0.826), np.float32(0.5859), np.float32(0.5645), np.float32(0.0), np.float32(0.0), np.float32(0.7541), np.float32(0.3847), np.float32(0.8086), np.float32(0.6221), np.float32(0.3213), np.float32(0.6815), np.float32(0.8428), np.float32(0.8115), np.float32(0.7397), np.float32(0.7174), np.float32(0.5762), np.float32(0.8309), np.float32(0.0), np.float32(0.4403), np.float32(0.6808), np.float32(0.5084), np.float32(0.838), np.float32(0.7631), np.float32(0.9432), np.float32(0.9643)] 
+2026-10-18 15:37:11.693808: Epoch time: 109.14 s 
+2026-10-18 15:37:12.412985:  
+2026-10-18 15:37:12.413765: Epoch 247 
+2026-10-18 15:37:12.414118: Current learning rate: 0.00775 
+2026-10-18 15:39:01.585340: train_loss -0.6108 
+2026-10-18 15:39:01.586279: val_loss -0.5329 
+2026-10-18 15:39:01.586828: Pseudo dice [np.float32(0.9399), np.float32(0.7391), np.float32(0.8578), np.float32(0.8638), np.float32(0.5789), np.float32(0.8986), np.float32(0.8295), np.float32(0.4552), np.float32(0.0), np.float32(0.5142), np.float32(0.7803), np.float32(0.7783), np.float32(0.6582), np.float32(0.6102), np.float32(0.0), np.float32(0.0), np.float32(0.7234), np.float32(0.4807), np.float32(0.7981), np.float32(0.6306), np.float32(0.3596), np.float32(0.8259), np.float32(0.7915), np.float32(0.8336), np.float32(0.7085), np.float32(0.7535), np.float32(0.517), np.float32(0.8272), np.float32(0.0), np.float32(0.7275), np.float32(0.6643), np.float32(0.4869), np.float32(0.8188), np.float32(0.7259), np.float32(0.9102), np.float32(0.9421)] 
+2026-10-18 15:39:01.587151: Epoch time: 109.17 s 
+2026-10-18 15:39:02.351097:  
+2026-10-18 15:39:02.351782: Epoch 248 
+2026-10-18 15:39:02.352116: Current learning rate: 0.00774 
+2026-10-18 15:40:51.517434: train_loss -0.6346 
+2026-10-18 15:40:51.518047: val_loss -0.3832 
+2026-10-18 15:40:51.518708: Pseudo dice [np.float32(0.937), np.float32(0.5583), np.float32(0.678), np.float32(0.578), np.float32(0.4798), np.float32(0.6819), np.float32(0.6707), np.float32(0.3296), np.float32(0.0), np.float32(0.4673), np.float32(0.5935), np.float32(0.6556), np.float32(0.559), np.float32(0.6083), np.float32(0.0), np.float32(0.0), np.float32(0.657), np.float32(0.2277), np.float32(0.6334), np.float32(0.4655), np.float32(0.2781), np.float32(0.4993), np.float32(0.3778), np.float32(0.3423), np.float32(0.5019), np.float32(0.5179), np.float32(0.091), np.float32(0.1713), np.float32(0.0), np.float32(0.4049), np.float32(0.6643), np.float32(0.4369), np.float32(0.5825), np.float32(0.4285), np.float32(0.5463), np.float32(0.6359)] 
+2026-10-18 15:40:51.519249: Epoch time: 109.17 s 
+2026-10-18 15:40:52.228475:  
+2026-10-18 15:40:52.229243: Epoch 249 
+2026-10-18 15:40:52.229453: Current learning rate: 0.00773 
+2026-10-18 15:42:41.393008: train_loss -0.6125 
+2026-10-18 15:42:41.394394: val_loss -0.4244 
+2026-10-18 15:42:41.395236: Pseudo dice [np.float32(0.9013), np.float32(0.6251), np.float32(0.7278), np.float32(0.6758), np.float32(0.5336), np.float32(0.7947), np.float32(0.7592), np.float32(0.4888), np.float32(0.0), np.float32(0.3961), np.float32(0.6867), np.float32(0.7265), np.float32(0.6331), np.float32(0.6701), np.float32(0.0), np.float32(0.0), np.float32(0.5297), np.float32(0.1655), np.float32(0.6029), np.float32(0.5441), np.float32(0.3844), np.float32(0.6508), np.float32(0.5687), np.float32(0.6353), np.float32(0.5833), np.float32(0.6071), np.float32(0.4656), np.float32(0.691), np.float32(0.0), np.float32(0.5324), np.float32(0.6006), np.float32(0.3419), np.float32(0.7272), np.float32(0.651), np.float32(0.7308), np.float32(0.7353)] 
+2026-10-18 15:42:41.395904: Epoch time: 109.17 s 
+2026-10-18 15:42:42.657726:  
+2026-10-18 15:42:42.658408: Epoch 250 
+2026-10-18 15:42:42.658631: Current learning rate: 0.00772 
+2026-10-18 15:44:31.842059: train_loss -0.6212 
+2026-10-18 15:44:31.842805: val_loss -0.5238 
+2026-10-18 15:44:31.843364: Pseudo dice [np.float32(0.9295), np.float32(0.7346), np.float32(0.8399), np.float32(0.8194), np.float32(0.6882), np.float32(0.8985), np.float32(0.8261), np.float32(0.5131), np.float32(0.0), np.float32(0.3509), np.float32(0.7098), np.float32(0.7872), np.float32(0.6479), np.float32(0.6325), np.float32(0.0), np.float32(0.0), np.float32(0.6102), np.float32(0.3907), np.float32(0.6728), np.float32(0.4136), np.float32(0.2917), np.float32(0.745), np.float32(0.789), np.float32(0.8268), np.float32(0.7801), np.float32(0.7204), np.float32(0.5248), np.float32(0.7773), np.float32(0.0), np.float32(0.7221), np.float32(0.6479), np.float32(0.4715), np.float32(0.7388), np.float32(0.6705), np.float32(0.8981), np.float32(0.9333)] 
+2026-10-18 15:44:31.844016: Epoch time: 109.19 s 
+2026-10-18 15:44:32.568227:  
+2026-10-18 15:44:32.568969: Epoch 251 
+2026-10-18 15:44:32.569297: Current learning rate: 0.00771 
+2026-10-18 15:46:21.699198: train_loss -0.6487 
+2026-10-18 15:46:21.699892: val_loss -0.5353 
+2026-10-18 15:46:21.700654: Pseudo dice [np.float32(0.9401), np.float32(0.741), np.float32(0.882), np.float32(0.8985), np.float32(0.6276), np.float32(0.9219), np.float32(0.8572), np.float32(0.5582), np.float32(0.0), np.float32(0.2911), np.float32(0.7425), np.float32(0.7802), np.float32(0.6266), np.float32(0.5112), np.float32(0.0), np.float32(0.0), np.float32(0.7392), np.float32(0.5233), np.float32(0.7947), np.float32(0.5985), np.float32(0.3874), np.float32(0.7939), np.float32(0.7171), np.float32(0.7737), np.float32(0.7329), np.float32(0.7512), np.float32(0.4226), np.float32(0.726), np.float32(0.0), np.float32(0.7088), np.float32(0.6064), np.float32(0.4683), np.float32(0.7724), np.float32(0.64), np.float32(0.9424), np.float32(0.9508)] 
+2026-10-18 15:46:21.700955: Epoch time: 109.13 s 
+2026-10-18 15:46:22.425026:  
+2026-10-18 15:46:22.425623: Epoch 252 
+2026-10-18 15:46:22.425828: Current learning rate: 0.0077 
+2026-10-18 15:48:11.580235: train_loss -0.6411 
+2026-10-18 15:48:11.581128: val_loss -0.5294 
+2026-10-18 15:48:11.581714: Pseudo dice [np.float32(0.9255), np.float32(0.7062), np.float32(0.8817), np.float32(0.8488), np.float32(0.6096), np.float32(0.8963), np.float32(0.8186), np.float32(0.3972), np.float32(0.0), np.float32(0.4516), np.float32(0.7257), np.float32(0.7871), np.float32(0.5563), np.float32(0.3424), np.float32(0.0), np.float32(0.0), np.float32(0.7721), np.float32(0.5833), np.float32(0.8041), np.float32(0.475), np.float32(0.1994), np.float32(0.7647), np.float32(0.845), np.float32(0.8456), np.float32(0.7045), np.float32(0.7227), np.float32(0.5118), np.float32(0.8266), np.float32(0.0), np.float32(0.7305), np.float32(0.5952), np.float32(0.559), np.float32(0.8007), np.float32(0.6972), np.float32(0.9285), np.float32(0.947)] 
+2026-10-18 15:48:11.582315: Epoch time: 109.16 s 
+2026-10-18 15:48:12.310168:  
+2026-10-18 15:48:12.310841: Epoch 253 
+2026-10-18 15:48:12.311169: Current learning rate: 0.00769 
+2026-10-18 15:50:01.489137: train_loss -0.6288 
+2026-10-18 15:50:01.489701: val_loss -0.512 
+2026-10-18 15:50:01.490623: Pseudo dice [np.float32(0.9452), np.float32(0.7373), np.float32(0.8353), np.float32(0.8529), np.float32(0.7368), np.float32(0.9104), np.float32(0.8496), np.float32(0.5676), np.float32(0.0), np.float32(0.3538), np.float32(0.7268), np.float32(0.7587), np.float32(0.4834), np.float32(0.2892), np.float32(0.0), np.float32(0.0), np.float32(0.7739), np.float32(0.5063), np.float32(0.8077), np.float32(0.5703), np.float32(0.4424), np.float32(0.7509), np.float32(0.6038), np.float32(0.5773), np.float32(0.7523), np.float32(0.7761), np.float32(0.3568), np.float32(0.7742), np.float32(0.0), np.float32(0.6713), np.float32(0.7694), np.float32(0.4286), np.float32(0.7828), np.float32(0.68), np.float32(0.8382), np.float32(0.87)] 
+2026-10-18 15:50:01.490931: Epoch time: 109.18 s 
+2026-10-18 15:50:02.204230:  
+2026-10-18 15:50:02.204864: Epoch 254 
+2026-10-18 15:50:02.205219: Current learning rate: 0.00768 
+2026-10-18 15:51:51.373124: train_loss -0.6632 
+2026-10-18 15:51:51.373770: val_loss -0.5376 
+2026-10-18 15:51:51.374252: Pseudo dice [np.float32(0.9254), np.float32(0.7012), np.float32(0.8657), np.float32(0.889), np.float32(0.7043), np.float32(0.9362), np.float32(0.8586), np.float32(0.6091), np.float32(0.0), np.float32(0.432), np.float32(0.7987), np.float32(0.8089), np.float32(0.6822), np.float32(0.6488), np.float32(0.0), np.float32(0.0), np.float32(0.7504), np.float32(0.5146), np.float32(0.7604), np.float32(0.5956), np.float32(0.2211), np.float32(0.7654), np.float32(0.7335), np.float32(0.7363), np.float32(0.7304), np.float32(0.7237), np.float32(0.2589), np.float32(0.7684), np.float32(0.0), np.float32(0.6081), np.float32(0.7271), np.float32(0.6117), np.float32(0.8247), np.float32(0.7161), np.float32(0.9219), np.float32(0.9364)] 
+2026-10-18 15:51:51.374875: Epoch time: 109.17 s 
+2026-10-18 15:51:52.350727:  
+2026-10-18 15:51:52.351431: Epoch 255 
+2026-10-18 15:51:52.351793: Current learning rate: 0.00767 
+2026-10-18 15:53:41.556334: train_loss -0.6473 
+2026-10-18 15:53:41.556993: val_loss -0.5002 
+2026-10-18 15:53:41.557765: Pseudo dice [np.float32(0.9159), np.float32(0.7182), np.float32(0.8588), np.float32(0.8483), np.float32(0.5661), np.float32(0.919), np.float32(0.8348), np.float32(0.3674), np.float32(0.0), np.float32(0.4184), np.float32(0.7822), np.float32(0.8438), np.float32(0.6924), np.float32(0.622), np.float32(0.0), np.float32(0.0), np.float32(0.6423), np.float32(0.2672), np.float32(0.7574), np.float32(0.536), np.float32(0.2592), np.float32(0.6995), np.float32(0.5044), np.float32(0.6968), np.float32(0.7335), np.float32(0.7104), np.float32(0.346), np.float32(0.7398), np.float32(0.0), np.float32(0.5276), np.float32(0.6486), np.float32(0.4497), np.float32(0.777), np.float32(0.6398), np.float32(0.8769), np.float32(0.9248)] 
+2026-10-18 15:53:41.558156: Epoch time: 109.21 s 
+2026-10-18 15:53:42.275718:  
+2026-10-18 15:53:42.276393: Epoch 256 
+2026-10-18 15:53:42.276579: Current learning rate: 0.00766 
+2026-10-18 15:55:31.440289: train_loss -0.6394 
+2026-10-18 15:55:31.441166: val_loss -0.5456 
+2026-10-18 15:55:31.441764: Pseudo dice [np.float32(0.8857), np.float32(0.7556), np.float32(0.8762), np.float32(0.8776), np.float32(0.7154), np.float32(0.9316), np.float32(0.8667), np.float32(0.5321), np.float32(0.0), np.float32(0.4476), np.float32(0.7683), np.float32(0.7963), np.float32(0.7542), np.float32(0.6737), np.float32(0.0), np.float32(0.0), np.float32(0.7799), np.float32(0.477), np.float32(0.8294), np.float32(0.6385), np.float32(0.33), np.float32(0.7936), np.float32(0.7032), np.float32(0.752), np.float32(0.7217), np.float32(0.7434), np.float32(0.4031), np.float32(0.458), np.float32(0.0), np.float32(0.5197), np.float32(0.7503), np.float32(0.3453), np.float32(0.7317), np.float32(0.6749), np.float32(0.88), np.float32(0.8905)] 
+2026-10-18 15:55:31.442274: Epoch time: 109.17 s 
+2026-10-18 15:55:32.170667:  
+2026-10-18 15:55:32.171321: Epoch 257 
+2026-10-18 15:55:32.171739: Current learning rate: 0.00765 
+2026-10-18 15:57:21.342987: train_loss -0.6326 
+2026-10-18 15:57:21.343597: val_loss -0.547 
+2026-10-18 15:57:21.344354: Pseudo dice [np.float32(0.9309), np.float32(0.7536), np.float32(0.8807), np.float32(0.8879), np.float32(0.6374), np.float32(0.9396), np.float32(0.8819), np.float32(0.3819), np.float32(0.0), np.float32(0.4831), np.float32(0.7719), np.float32(0.8437), np.float32(0.6807), np.float32(0.6553), np.float32(0.0), np.float32(0.0), np.float32(0.7545), np.float32(0.5896), np.float32(0.8174), np.float32(0.6135), np.float32(0.3472), np.float32(0.7169), np.float32(0.8084), np.float32(0.8358), np.float32(0.7409), np.float32(0.7555), np.float32(0.4209), np.float32(0.7673), np.float32(0.0), np.float32(0.7157), np.float32(0.6397), np.float32(0.4502), np.float32(0.8118), np.float32(0.747), np.float32(0.9385), np.float32(0.9566)] 
+2026-10-18 15:57:21.344832: Epoch time: 109.17 s 
+2026-10-18 15:57:22.093407:  
+2026-10-18 15:57:22.094096: Epoch 258 
+2026-10-18 15:57:22.094316: Current learning rate: 0.00764 
+2026-10-18 15:59:11.275206: train_loss -0.6348 
+2026-10-18 15:59:11.275709: val_loss -0.5279 
+2026-10-18 15:59:11.276369: Pseudo dice [np.float32(0.9405), np.float32(0.7138), np.float32(0.8838), np.float32(0.853), np.float32(0.6992), np.float32(0.9102), np.float32(0.839), np.float32(0.5344), np.float32(0.0), np.float32(0.4874), np.float32(0.7615), np.float32(0.7587), np.float32(0.654), np.float32(0.5276), np.float32(0.0), np.float32(0.0), np.float32(0.7997), np.float32(0.5651), np.float32(0.7981), np.float32(0.5783), np.float32(0.3637), np.float32(0.5942), np.float32(0.6136), np.float32(0.6955), np.float32(0.7032), np.float32(0.7377), np.float32(0.4012), np.float32(0.6947), np.float32(0.0), np.float32(0.6328), np.float32(0.4731), np.float32(0.5326), np.float32(0.7813), np.float32(0.6547), np.float32(0.8909), np.float32(0.9268)] 
+2026-10-18 15:59:11.276984: Epoch time: 109.18 s 
+2026-10-18 15:59:12.009069:  
+2026-10-18 15:59:12.009877: Epoch 259 
+2026-10-18 15:59:12.010091: Current learning rate: 0.00764 
+2026-10-18 16:01:01.196882: train_loss -0.6426 
+2026-10-18 16:01:01.198273: val_loss -0.5316 
+2026-10-18 16:01:01.199288: Pseudo dice [np.float32(0.9359), np.float32(0.713), np.float32(0.8684), np.float32(0.8805), np.float32(0.7225), np.float32(0.9294), np.float32(0.8709), np.float32(0.3042), np.float32(0.0), np.float32(0.4625), np.float32(0.7828), np.float32(0.7847), np.float32(0.4723), np.float32(0.4656), np.float32(0.0), np.float32(0.0), np.float32(0.7603), np.float32(0.5693), np.float32(0.842), np.float32(0.4438), np.float32(0.3071), np.float32(0.6782), np.float32(0.623), np.float32(0.7179), np.float32(0.7397), np.float32(0.7326), np.float32(0.3314), np.float32(0.6661), np.float32(0.0), np.float32(0.7146), np.float32(0.687), np.float32(0.5503), np.float32(0.8202), np.float32(0.6618), np.float32(0.9099), np.float32(0.9186)] 
+2026-10-18 16:01:01.199857: Epoch time: 109.19 s 
+2026-10-18 16:01:01.930454:  
+2026-10-18 16:01:01.931142: Epoch 260 
+2026-10-18 16:01:01.931556: Current learning rate: 0.00763 
+2026-10-18 16:02:51.132288: train_loss -0.661 
+2026-10-18 16:02:51.133117: val_loss -0.565 
+2026-10-18 16:02:51.133898: Pseudo dice [np.float32(0.941), np.float32(0.7697), np.float32(0.9013), np.float32(0.9235), np.float32(0.7498), np.float32(0.931), np.float32(0.8479), np.float32(0.3482), np.float32(0.0), np.float32(0.5061), np.float32(0.8218), np.float32(0.8448), np.float32(0.649), np.float32(0.4795), np.float32(0.0), np.float32(0.0), np.float32(0.8059), np.float32(0.5695), np.float32(0.8283), np.float32(0.6486), np.float32(0.2865), np.float32(0.7452), np.float32(0.7927), np.float32(0.7885), np.float32(0.7612), np.float32(0.7713), np.float32(0.5868), np.float32(0.7951), np.float32(0.0), np.float32(0.6933), np.float32(0.7319), np.float32(0.4292), np.float32(0.8233), np.float32(0.6889), np.float32(0.9417), np.float32(0.944)] 
+2026-10-18 16:02:51.134284: Epoch time: 109.2 s 
+2026-10-18 16:02:51.875808:  
+2026-10-18 16:02:51.876486: Epoch 261 
+2026-10-18 16:02:51.876793: Current learning rate: 0.00762 
+2026-10-18 16:04:41.045460: train_loss -0.6624 
+2026-10-18 16:04:41.046167: val_loss -0.5611 
+2026-10-18 16:04:41.047013: Pseudo dice [np.float32(0.9482), np.float32(0.7825), np.float32(0.8877), np.float32(0.8786), np.float32(0.8124), np.float32(0.9343), np.float32(0.8519), np.float32(0.3545), np.float32(0.0), np.float32(0.36), np.float32(0.7603), np.float32(0.8221), np.float32(0.6813), np.float32(0.5444), np.float32(0.0), np.float32(0.0), np.float32(0.8156), np.float32(0.5299), np.float32(0.8178), np.float32(0.4475), np.float32(0.1799), np.float32(0.6619), np.float32(0.7488), np.float32(0.7992), np.float32(0.7497), np.float32(0.738), np.float32(0.4302), np.float32(0.8398), np.float32(0.0), np.float32(0.7793), np.float32(0.6275), np.float32(0.609), np.float32(0.7587), np.float32(0.7207), np.float32(0.917), np.float32(0.954)] 
+2026-10-18 16:04:41.047539: Epoch time: 109.17 s 
+2026-10-18 16:04:41.779811:  
+2026-10-18 16:04:41.780322: Epoch 262 
+2026-10-18 16:04:41.780861: Current learning rate: 0.00761 
+2026-10-18 16:06:30.930362: train_loss -0.6394 
+2026-10-18 16:06:30.931419: val_loss -0.5337 
+2026-10-18 16:06:30.932328: Pseudo dice [np.float32(0.908), np.float32(0.7109), np.float32(0.8909), np.float32(0.8771), np.float32(0.7292), np.float32(0.9426), np.float32(0.8591), np.float32(0.6025), np.float32(0.0), np.float32(0.4058), np.float32(0.8054), np.float32(0.8195), np.float32(0.6753), np.float32(0.5851), np.float32(0.0), np.float32(0.0), np.float32(0.7962), np.float32(0.4205), np.float32(0.8085), np.float32(0.6202), np.float32(0.3478), np.float32(0.778), np.float32(0.6708), np.float32(0.7423), np.float32(0.7124), np.float32(0.7488), np.float32(0.4208), np.float32(0.8195), np.float32(0.0), np.float32(0.565), np.float32(0.3938), np.float32(0.5365), np.float32(0.7933), np.float32(0.7116), np.float32(0.8978), np.float32(0.9332)] 
+2026-10-18 16:06:30.932750: Epoch time: 109.15 s 
+2026-10-18 16:06:31.720070:  
+2026-10-18 16:06:31.720687: Epoch 263 
+2026-10-18 16:06:31.721046: Current learning rate: 0.0076 
+2026-10-18 16:08:20.873177: train_loss -0.6384 
+2026-10-18 16:08:20.873841: val_loss -0.5337 
+2026-10-18 16:08:20.874294: Pseudo dice [np.float32(0.9074), np.float32(0.7694), np.float32(0.8884), np.float32(0.8884), np.float32(0.7158), np.float32(0.934), np.float32(0.8718), np.float32(0.478), np.float32(0.0), np.float32(0.4913), np.float32(0.7753), np.float32(0.7918), np.float32(0.6352), np.float32(0.6053), np.float32(0.0), np.float32(0.0), np.float32(0.747), np.float32(0.439), np.float32(0.8301), np.float32(0.5228), np.float32(0.3197), np.float32(0.7326), np.float32(0.6485), np.float32(0.7069), np.float32(0.7527), np.float32(0.7599), np.float32(0.3526), np.float32(0.8672), np.float32(0.0), np.float32(0.7029), np.float32(0.5856), np.float32(0.5656), np.float32(0.8245), np.float32(0.7029), np.float32(0.9166), np.float32(0.937)] 
+2026-10-18 16:08:20.875043: Epoch time: 109.16 s 
+2026-10-18 16:08:21.611820:  
+2026-10-18 16:08:21.612348: Epoch 264 
+2026-10-18 16:08:21.612632: Current learning rate: 0.00759 
+2026-10-18 16:10:10.757823: train_loss -0.6401 
+2026-10-18 16:10:10.758496: val_loss -0.5261 
+2026-10-18 16:10:10.759068: Pseudo dice [np.float32(0.9406), np.float32(0.7541), np.float32(0.8514), np.float32(0.8616), np.float32(0.6858), np.float32(0.8997), np.float32(0.8294), np.float32(0.3973), np.float32(0.0029), np.float32(0.4626), np.float32(0.7312), np.float32(0.7771), np.float32(0.6431), np.float32(0.4622), np.float32(0.0), np.float32(0.0), np.float32(0.7293), np.float32(0.482), np.float32(0.7586), np.float32(0.5814), np.float32(0.3777), np.float32(0.736), np.float32(0.7166), np.float32(0.7932), np.float32(0.7277), np.float32(0.7321), np.float32(0.6174), np.float32(0.8402), np.float32(0.0), np.float32(0.592), np.float32(0.694), np.float32(0.4695), np.float32(0.8332), np.float32(0.7148), np.float32(0.899), np.float32(0.9285)] 
+2026-10-18 16:10:10.759494: Epoch time: 109.15 s 
+2026-10-18 16:10:11.490724:  
+2026-10-18 16:10:11.491510: Epoch 265 
+2026-10-18 16:10:11.491740: Current learning rate: 0.00758 
+2026-10-18 16:12:00.620869: train_loss -0.6154 
+2026-10-18 16:12:00.621527: val_loss -0.4078 
+2026-10-18 16:12:00.622112: Pseudo dice [np.float32(0.9535), np.float32(0.6027), np.float32(0.7208), np.float32(0.6196), np.float32(0.4202), np.float32(0.7088), np.float32(0.655), np.float32(0.3891), np.float32(0.0), np.float32(0.4825), np.float32(0.6151), np.float32(0.6482), np.float32(0.6138), np.float32(0.5627), np.float32(0.0), np.float32(0.0), np.float32(0.5734), np.float32(0.3997), np.float32(0.6619), np.float32(0.5158), np.float32(0.2082), np.float32(0.7646), np.float32(0.5685), np.float32(0.6388), np.float32(0.5518), np.float32(0.531), np.float32(0.2284), np.float32(0.5237), np.float32(0.0), np.float32(0.5828), np.float32(0.5609), np.float32(0.4075), np.float32(0.5827), np.float32(0.4753), np.float32(0.6139), np.float32(0.6499)] 
+2026-10-18 16:12:00.622876: Epoch time: 109.13 s 
+2026-10-18 16:12:01.361079:  
+2026-10-18 16:12:01.361688: Epoch 266 
+2026-10-18 16:12:01.361973: Current learning rate: 0.00757 
+2026-10-18 16:13:50.527930: train_loss -0.6635 
+2026-10-18 16:13:50.528776: val_loss -0.4862 
+2026-10-18 16:13:50.529516: Pseudo dice [np.float32(0.9213), np.float32(0.6455), np.float32(0.7809), np.float32(0.7016), np.float32(0.6765), np.float32(0.7678), np.float32(0.7493), np.float32(0.398), np.float32(0.0), np.float32(0.5279), np.float32(0.6422), np.float32(0.6798), np.float32(0.534), np.float32(0.2859), np.float32(0.0), np.float32(0.0), np.float32(0.7006), np.float32(0.5753), np.float32(0.7304), np.float32(0.4995), np.float32(0.3156), np.float32(0.7901), np.float32(0.786), np.float32(0.8443), np.float32(0.6219), np.float32(0.6062), np.float32(0.2918), np.float32(0.5781), np.float32(0.0), np.float32(0.5773), np.float32(0.464), np.float32(0.2636), np.float32(0.6106), np.float32(0.6212), np.float32(0.8022), np.float32(0.845)] 
+2026-10-18 16:13:50.529964: Epoch time: 109.17 s 
+2026-10-18 16:13:51.265856:  
+2026-10-18 16:13:51.267125: Epoch 267 
+2026-10-18 16:13:51.267811: Current learning rate: 0.00756 
+2026-10-18 16:15:40.412541: train_loss -0.6512 
+2026-10-18 16:15:40.413126: val_loss -0.5406 
+2026-10-18 16:15:40.413674: Pseudo dice [np.float32(0.9357), np.float32(0.7384), np.float32(0.8844), np.float32(0.8686), np.float32(0.7267), np.float32(0.907), np.float32(0.8504), np.float32(0.5252), np.float32(0.0), np.float32(0.2878), np.float32(0.7757), np.float32(0.8224), np.float32(0.6414), np.float32(0.4961), np.float32(0.0), np.float32(0.0), np.float32(0.7576), np.float32(0.4499), np.float32(0.7437), np.float32(0.5259), np.float32(0.2083), np.float32(0.7558), np.float32(0.6775), np.float32(0.7932), np.float32(0.7111), np.float32(0.7169), np.float32(0.446), np.float32(0.7633), np.float32(0.0), np.float32(0.6835), np.float32(0.6714), np.float32(0.5638), np.float32(0.7879), np.float32(0.7224), np.float32(0.8892), np.float32(0.9152)] 
+2026-10-18 16:15:40.414414: Epoch time: 109.15 s 
+2026-10-18 16:15:41.161563:  
+2026-10-18 16:15:41.162143: Epoch 268 
+2026-10-18 16:15:41.162483: Current learning rate: 0.00755 
+2026-10-18 16:17:30.317883: train_loss -0.6767 
+2026-10-18 16:17:30.318574: val_loss -0.5451 
+2026-10-18 16:17:30.319087: Pseudo dice [np.float32(0.9492), np.float32(0.7362), np.float32(0.8596), np.float32(0.8979), np.float32(0.74), np.float32(0.9081), np.float32(0.8593), np.float32(0.6261), np.float32(0.0), np.float32(0.2138), np.float32(0.754), np.float32(0.8089), np.float32(0.514), np.float32(0.5563), np.float32(0.0), np.float32(0.0), np.float32(0.7699), np.float32(0.4995), np.float32(0.7742), np.float32(0.5024), np.float32(0.2288), np.float32(0.6607), np.float32(0.8632), np.float32(0.8538), np.float32(0.6887), np.float32(0.7262), np.float32(0.3945), np.float32(0.6636), np.float32(0.0), np.float32(0.6584), np.float32(0.6376), np.float32(0.5179), np.float32(0.8342), np.float32(0.6924), np.float32(0.8848), np.float32(0.9075)] 
+2026-10-18 16:17:30.319544: Epoch time: 109.16 s 
+2026-10-18 16:17:31.053357:  
+2026-10-18 16:17:31.054109: Epoch 269 
+2026-10-18 16:17:31.054441: Current learning rate: 0.00754 
+2026-10-18 16:19:20.227813: train_loss -0.6654 
+2026-10-18 16:19:20.228399: val_loss -0.5442 
+2026-10-18 16:19:20.229173: Pseudo dice [np.float32(0.9168), np.float32(0.7085), np.float32(0.8939), np.float32(0.8796), np.float32(0.7499), np.float32(0.9011), np.float32(0.8545), np.float32(0.4848), np.float32(0.0), np.float32(0.4681), np.float32(0.7899), np.float32(0.8183), np.float32(0.5979), np.float32(0.7316), np.float32(0.0), np.float32(0.0), np.float32(0.8167), np.float32(0.6301), np.float32(0.8284), np.float32(0.6782), np.float32(0.3843), np.float32(0.8238), np.float32(0.7766), np.float32(0.766), np.float32(0.6963), np.float32(0.7368), np.float32(0.5094), np.float32(0.817), np.float32(0.0), np.float32(0.6458), np.float32(0.6995), np.float32(0.5769), np.float32(0.8208), np.float32(0.6963), np.float32(0.8737), np.float32(0.9151)] 
+2026-10-18 16:19:20.229689: Epoch time: 109.18 s 
+2026-10-18 16:19:20.953060:  
+2026-10-18 16:19:20.953782: Epoch 270 
+2026-10-18 16:19:20.954063: Current learning rate: 0.00753 
+2026-10-18 16:21:10.121292: train_loss -0.6737 
+2026-10-18 16:21:10.122073: val_loss -0.5324 
+2026-10-18 16:21:10.122792: Pseudo dice [np.float32(0.9374), np.float32(0.709), np.float32(0.8639), np.float32(0.879), np.float32(0.7292), np.float32(0.8999), np.float32(0.8489), np.float32(0.2104), np.float32(0.0), np.float32(0.1635), np.float32(0.8083), np.float32(0.7779), np.float32(0.6483), np.float32(0.5366), np.float32(0.0), np.float32(0.0), np.float32(0.7548), np.float32(0.513), np.float32(0.8241), np.float32(0.5623), np.float32(0.3208), np.float32(0.7827), np.float32(0.7644), np.float32(0.7246), np.float32(0.7367), np.float32(0.7163), np.float32(0.6502), np.float32(0.7936), np.float32(0.0), np.float32(0.5511), np.float32(0.6773), np.float32(0.4636), np.float32(0.7567), np.float32(0.5937), np.float32(0.8799), np.float32(0.9098)] 
+2026-10-18 16:21:10.123231: Epoch time: 109.17 s 
+2026-10-18 16:21:10.845430:  
+2026-10-18 16:21:10.846038: Epoch 271 
+2026-10-18 16:21:10.846248: Current learning rate: 0.00752 
+2026-10-18 16:22:59.999567: train_loss -0.6391 
+2026-10-18 16:23:00.000458: val_loss -0.4535 
+2026-10-18 16:23:00.001049: Pseudo dice [np.float32(0.9253), np.float32(0.6271), np.float32(0.7992), np.float32(0.7785), np.float32(0.6954), np.float32(0.8581), np.float32(0.8274), np.float32(0.3955), np.float32(0.0), np.float32(0.3623), np.float32(0.6546), np.float32(0.699), np.float32(0.5264), np.float32(0.3878), np.float32(0.0), np.float32(nan), np.float32(0.587), np.float32(0.3126), np.float32(0.7812), np.float32(0.524), np.float32(0.3628), np.float32(0.6499), np.float32(0.4271), np.float32(0.538), np.float32(0.638), np.float32(0.6321), np.float32(0.3476), np.float32(0.5022), np.float32(0.0), np.float32(0.4813), np.float32(0.6339), np.float32(0.4639), np.float32(0.7188), np.float32(0.6293), np.float32(0.6371), np.float32(0.7384)] 
+2026-10-18 16:23:00.001456: Epoch time: 109.16 s 
+2026-10-18 16:23:00.740531:  
+2026-10-18 16:23:00.741131: Epoch 272 
+2026-10-18 16:23:00.741378: Current learning rate: 0.00751 
+2026-10-18 16:24:49.845906: train_loss -0.6198 
+2026-10-18 16:24:49.846735: val_loss -0.5337 
+2026-10-18 16:24:49.847327: Pseudo dice [np.float32(0.9418), np.float32(0.7663), np.float32(0.8725), np.float32(0.8521), np.float32(0.6518), np.float32(0.8822), np.float32(0.8621), np.float32(0.5358), np.float32(0.0), np.float32(0.4878), np.float32(0.7764), np.float32(0.7556), np.float32(0.6846), np.float32(0.4837), np.float32(0.0), np.float32(0.0), np.float32(0.7488), np.float32(0.5136), np.float32(0.7893), np.float32(0.5725), np.float32(0.3871), np.float32(0.75), np.float32(0.8344), np.float32(0.771), np.float32(0.7364), np.float32(0.7096), np.float32(0.5532), np.float32(0.8263), np.float32(0.0), np.float32(0.6964), np.float32(0.6944), np.float32(0.3744), np.float32(0.7765), np.float32(0.684), np.float32(0.8382), np.float32(0.8586)] 
+2026-10-18 16:24:49.847873: Epoch time: 109.11 s 
+2026-10-18 16:24:50.570018:  
+2026-10-18 16:24:50.570654: Epoch 273 
+2026-10-18 16:24:50.570849: Current learning rate: 0.00751 
+2026-10-18 16:26:39.761804: train_loss -0.6233 
+2026-10-18 16:26:39.762393: val_loss -0.4143 
+2026-10-18 16:26:39.763075: Pseudo dice [np.float32(0.9262), np.float32(0.5965), np.float32(0.745), np.float32(0.6965), np.float32(0.5773), np.float32(0.6852), np.float32(0.7327), np.float32(0.092), np.float32(0.0), np.float32(0.2989), np.float32(0.6457), np.float32(0.6058), np.float32(0.4935), np.float32(0.3193), np.float32(0.0), np.float32(0.0), np.float32(0.6835), np.float32(0.5228), np.float32(0.6543), np.float32(0.6599), np.float32(0.3693), np.float32(0.697), np.float32(0.4678), np.float32(0.4603), np.float32(0.5641), np.float32(0.5918), np.float32(0.1991), np.float32(0.477), np.float32(0.0), np.float32(0.4466), np.float32(0.7169), np.float32(0.3165), np.float32(0.6399), np.float32(0.4974), np.float32(0.6502), np.float32(0.6571)] 
+2026-10-18 16:26:39.763998: Epoch time: 109.19 s 
+2026-10-18 16:26:40.498965:  
+2026-10-18 16:26:40.499511: Epoch 274 
+2026-10-18 16:26:40.499917: Current learning rate: 0.0075 
+2026-10-18 16:28:29.655997: train_loss -0.6419 
+2026-10-18 16:28:29.656960: val_loss -0.5459 
+2026-10-18 16:28:29.657513: Pseudo dice [np.float32(0.9027), np.float32(0.7881), np.float32(0.886), np.float32(0.8673), np.float32(0.7547), np.float32(0.9149), np.float32(0.8668), np.float32(0.4782), np.float32(0.0), np.float32(0.4308), np.float32(0.7678), np.float32(0.7834), np.float32(0.7236), np.float32(0.531), np.float32(0.0), np.float32(0.0), np.float32(0.7937), np.float32(0.5442), np.float32(0.8005), np.float32(0.5644), np.float32(0.5015), np.float32(0.7399), np.float32(0.6311), np.float32(0.6183), np.float32(0.7832), np.float32(0.754), np.float32(0.3506), np.float32(0.7656), np.float32(0.0), np.float32(0.5822), np.float32(0.76), np.float32(0.4739), np.float32(0.8023), np.float32(0.7657), np.float32(0.8727), np.float32(0.8927)] 
+2026-10-18 16:28:29.657960: Epoch time: 109.16 s 
+2026-10-18 16:28:30.390127:  
+2026-10-18 16:28:30.390935: Epoch 275 
+2026-10-18 16:28:30.391272: Current learning rate: 0.00749 
+2026-10-18 16:30:19.574098: train_loss -0.6421 
+2026-10-18 16:30:19.575012: val_loss -0.4951 
+2026-10-18 16:30:19.575572: Pseudo dice [np.float32(0.9212), np.float32(0.6234), np.float32(0.8442), np.float32(0.831), np.float32(0.6763), np.float32(0.8558), np.float32(0.814), np.float32(0.2819), np.float32(0.0), np.float32(0.3595), np.float32(0.7105), np.float32(0.768), np.float32(0.639), np.float32(0.5509), np.float32(0.0), np.float32(0.0), np.float32(0.7806), np.float32(0.5115), np.float32(0.7629), np.float32(0.606), np.float32(0.2922), np.float32(0.6911), np.float32(0.6084), np.float32(0.6272), np.float32(0.6566), np.float32(0.7122), np.float32(0.4923), np.float32(0.8253), np.float32(0.0), np.float32(0.4596), np.float32(0.6226), np.float32(0.538), np.float32(0.7192), np.float32(0.6726), np.float32(0.8346), np.float32(0.8701)] 
+2026-10-18 16:30:19.576140: Epoch time: 109.19 s 
+2026-10-18 16:30:20.582138:  
+2026-10-18 16:30:20.582890: Epoch 276 
+2026-10-18 16:30:20.583215: Current learning rate: 0.00748 
+2026-10-18 16:32:09.805217: train_loss -0.6566 
+2026-10-18 16:32:09.806134: val_loss -0.5636 
+2026-10-18 16:32:09.806767: Pseudo dice [np.float32(0.915), np.float32(0.7471), np.float32(0.911), np.float32(0.9184), np.float32(0.7953), np.float32(0.9332), np.float32(0.8555), np.float32(0.3856), np.float32(0.0), np.float32(0.4186), np.float32(0.7577), np.float32(0.7984), np.float32(0.6848), np.float32(0.3544), np.float32(0.0), np.float32(0.0), np.float32(0.8024), np.float32(0.5712), np.float32(0.8113), np.float32(0.5508), np.float32(0.4783), np.float32(0.7918), np.float32(0.7576), np.float32(0.808), np.float32(0.7846), np.float32(0.8046), np.float32(0.4536), np.float32(0.8102), np.float32(0.0), np.float32(0.634), np.float32(0.7475), np.float32(0.5854), np.float32(0.7801), np.float32(0.7545), np.float32(0.9462), np.float32(0.961)] 
+2026-10-18 16:32:09.807084: Epoch time: 109.23 s 
+2026-10-18 16:32:10.539301:  
+2026-10-18 16:32:10.539909: Epoch 277 
+2026-10-18 16:32:10.540094: Current learning rate: 0.00747 
+2026-10-18 16:33:59.726334: train_loss -0.6648 
+2026-10-18 16:33:59.726989: val_loss -0.528 
+2026-10-18 16:33:59.727477: Pseudo dice [np.float32(0.9175), np.float32(0.6971), np.float32(0.8919), np.float32(0.8826), np.float32(0.8117), np.float32(0.9275), np.float32(0.8618), np.float32(0.4197), np.float32(0.0), np.float32(0.337), np.float32(0.785), np.float32(0.8387), np.float32(0.6737), np.float32(0.5479), np.float32(0.0), np.float32(0.0), np.float32(0.7401), np.float32(0.4262), np.float32(0.7973), np.float32(0.5654), np.float32(0.2273), np.float32(0.6895), np.float32(0.4414), np.float32(0.6077), np.float32(0.7287), np.float32(0.7286), np.float32(0.3795), np.float32(0.6629), np.float32(0.0), np.float32(0.5236), np.float32(0.602), np.float32(0.5767), np.float32(0.8117), np.float32(0.657), np.float32(0.8787), np.float32(0.9173)] 
+2026-10-18 16:33:59.727727: Epoch time: 109.19 s 
+2026-10-18 16:34:00.464962:  
+2026-10-18 16:34:00.465623: Epoch 278 
+2026-10-18 16:34:00.465930: Current learning rate: 0.00746 
+2026-10-18 16:35:49.648232: train_loss -0.621 
+2026-10-18 16:35:49.648865: val_loss -0.523 
+2026-10-18 16:35:49.649424: Pseudo dice [np.float32(0.9303), np.float32(0.7188), np.float32(0.8515), np.float32(0.8603), np.float32(0.7351), np.float32(0.9054), np.float32(0.8743), np.float32(0.3194), np.float32(0.0), np.float32(0.203), np.float32(0.76), np.float32(0.847), np.float32(0.5936), np.float32(0.6343), np.float32(0.0), np.float32(0.0), np.float32(0.7657), np.float32(0.4343), np.float32(0.8023), np.float32(0.5563), np.float32(0.4933), np.float32(0.7916), np.float32(0.7879), np.float32(0.8232), np.float32(0.7061), np.float32(0.7726), np.float32(0.3856), np.float32(0.7987), np.float32(0.0), np.float32(0.6763), np.float32(0.6072), np.float32(0.471), np.float32(0.8086), np.float32(0.6941), np.float32(0.8964), np.float32(0.9301)] 
+2026-10-18 16:35:49.650072: Epoch time: 109.19 s 
+2026-10-18 16:35:50.388582:  
+2026-10-18 16:35:50.389284: Epoch 279 
+2026-10-18 16:35:50.389576: Current learning rate: 0.00745 
+2026-10-18 16:37:39.564537: train_loss -0.6255 
+2026-10-18 16:37:39.565651: val_loss -0.4928 
+2026-10-18 16:37:39.566366: Pseudo dice [np.float32(0.9126), np.float32(0.6961), np.float32(0.855), np.float32(0.7787), np.float32(0.5949), np.float32(0.8883), np.float32(0.759), np.float32(0.276), np.float32(0.0), np.float32(0.3964), np.float32(0.7319), np.float32(0.8254), np.float32(0.4149), np.float32(0.4656), np.float32(0.0), np.float32(0.0), np.float32(0.6661), np.float32(0.4109), np.float32(0.7257), np.float32(0.5416), np.float32(0.3812), np.float32(0.6781), np.float32(0.6943), np.float32(0.8121), np.float32(0.7103), np.float32(0.7445), np.float32(0.3481), np.float32(0.8778), np.float32(0.0), np.float32(0.7507), np.float32(0.2356), np.float32(0.0464), np.float32(0.4671), np.float32(0.5653), np.float32(0.9392), np.float32(0.9487)] 
+2026-10-18 16:37:39.567024: Epoch time: 109.18 s 
+2026-10-18 16:37:40.293588:  
+2026-10-18 16:37:40.294269: Epoch 280 
+2026-10-18 16:37:40.294433: Current learning rate: 0.00744 
+2026-10-18 16:39:29.423846: train_loss -0.6246 
+2026-10-18 16:39:29.424600: val_loss -0.5249 
+2026-10-18 16:39:29.425406: Pseudo dice [np.float32(0.914), np.float32(0.6897), np.float32(0.8777), np.float32(0.8702), np.float32(0.7143), np.float32(0.9025), np.float32(0.8483), np.float32(0.5166), np.float32(0.0), np.float32(0.3855), np.float32(0.7716), np.float32(0.7871), np.float32(0.5126), np.float32(0.4553), np.float32(0.0), np.float32(0.0), np.float32(0.7835), np.float32(0.5684), np.float32(0.7941), np.float32(0.4539), np.float32(0.3355), np.float32(0.7465), np.float32(0.6449), np.float32(0.7191), np.float32(0.7193), np.float32(0.7356), np.float32(0.4859), np.float32(0.6247), np.float32(0.0), np.float32(0.5505), np.float32(0.7018), np.float32(0.5404), np.float32(0.818), np.float32(0.6848), np.float32(0.8841), np.float32(0.9196)] 
+2026-10-18 16:39:29.425853: Epoch time: 109.13 s 
+2026-10-18 16:39:30.147320:  
+2026-10-18 16:39:30.148027: Epoch 281 
+2026-10-18 16:39:30.148330: Current learning rate: 0.00743 
+2026-10-18 16:41:19.260381: train_loss -0.6428 
+2026-10-18 16:41:19.261146: val_loss -0.5061 
+2026-10-18 16:41:19.261712: Pseudo dice [np.float32(0.9277), np.float32(0.7225), np.float32(0.8241), np.float32(0.8574), np.float32(0.7538), np.float32(0.8779), np.float32(0.8409), np.float32(0.3904), np.float32(0.0), np.float32(0.4094), np.float32(0.7639), np.float32(0.8098), np.float32(0.5778), np.float32(0.581), np.float32(0.0), np.float32(0.0), np.float32(0.6568), np.float32(0.3981), np.float32(0.7831), np.float32(0.5476), np.float32(0.2019), np.float32(0.7754), np.float32(0.7069), np.float32(0.7303), np.float32(0.723), np.float32(0.7036), np.float32(0.3019), np.float32(0.5642), np.float32(0.0), np.float32(0.3601), np.float32(0.5609), np.float32(0.4082), np.float32(0.7497), np.float32(0.7207), np.float32(0.8558), np.float32(0.8967)] 
+2026-10-18 16:41:19.261995: Epoch time: 109.12 s 
+2026-10-18 16:41:20.001996:  
+2026-10-18 16:41:20.002713: Epoch 282 
+2026-10-18 16:41:20.003134: Current learning rate: 0.00742 
+2026-10-18 16:43:09.208260: train_loss -0.6451 
+2026-10-18 16:43:09.210174: val_loss -0.5098 
+2026-10-18 16:43:09.211098: Pseudo dice [np.float32(0.9285), np.float32(0.729), np.float32(0.8574), np.float32(0.8207), np.float32(0.7444), np.float32(0.8735), np.float32(0.8334), np.float32(0.5341), np.float32(0.0), np.float32(0.4464), np.float32(0.7632), np.float32(0.8059), np.float32(0.658), np.float32(0.6165), np.float32(0.0), np.float32(0.0), np.float32(0.6908), np.float32(0.2422), np.float32(0.7634), np.float32(0.376), np.float32(0.4591), np.float32(0.7685), np.float32(0.7458), np.float32(0.6975), np.float32(0.6674), np.float32(0.6953), np.float32(0.4274), np.float32(0.7471), np.float32(0.0), np.float32(0.4378), np.float32(0.6877), np.float32(0.5016), np.float32(0.6689), np.float32(0.5799), np.float32(0.8287), np.float32(0.8615)] 
+2026-10-18 16:43:09.214671: Epoch time: 109.21 s 
+2026-10-18 16:43:09.952235:  
+2026-10-18 16:43:09.952864: Epoch 283 
+2026-10-18 16:43:09.953263: Current learning rate: 0.00741 
+2026-10-18 16:44:59.100173: train_loss -0.6567 
+2026-10-18 16:44:59.100949: val_loss -0.5517 
+2026-10-18 16:44:59.101867: Pseudo dice [np.float32(0.925), np.float32(0.7245), np.float32(0.895), np.float32(0.9062), np.float32(0.6829), np.float32(0.9312), np.float32(0.8756), np.float32(0.4251), np.float32(0.0), np.float32(0.3528), np.float32(0.7992), np.float32(0.8059), np.float32(0.6789), np.float32(0.4714), np.float32(0.0), np.float32(0.0), np.float32(0.7837), np.float32(0.4978), np.float32(0.838), np.float32(0.6631), np.float32(0.3261), np.float32(0.7487), np.float32(0.7669), np.float32(0.6146), np.float32(0.7386), np.float32(0.7583), np.float32(0.405), np.float32(0.7333), np.float32(0.0), np.float32(0.4442), np.float32(0.484), np.float32(0.2118), np.float32(0.8138), np.float32(0.6828), np.float32(0.9183), np.float32(0.9314)] 
+2026-10-18 16:44:59.102340: Epoch time: 109.15 s 
+2026-10-18 16:44:59.827710:  
+2026-10-18 16:44:59.828301: Epoch 284 
+2026-10-18 16:44:59.828686: Current learning rate: 0.0074 
+2026-10-18 16:46:48.971285: train_loss -0.6488 
+2026-10-18 16:46:48.971978: val_loss -0.5636 
+2026-10-18 16:46:48.972538: Pseudo dice [np.float32(0.9303), np.float32(0.767), np.float32(0.8827), np.float32(0.8585), np.float32(0.7875), np.float32(0.9241), np.float32(0.8559), np.float32(0.5658), np.float32(0.0), np.float32(0.3987), np.float32(0.775), np.float32(0.8473), np.float32(0.7232), np.float32(0.6963), np.float32(0.0), np.float32(0.0), np.float32(0.7971), np.float32(0.4305), np.float32(0.7625), np.float32(0.474), np.float32(0.3771), np.float32(0.7409), np.float32(0.8131), np.float32(0.8329), np.float32(0.6726), np.float32(0.7087), np.float32(0.3135), np.float32(0.7602), np.float32(0.0), np.float32(0.6218), np.float32(0.7715), np.float32(0.4155), np.float32(0.8151), np.float32(0.7449), np.float32(0.9179), np.float32(0.9515)] 
+2026-10-18 16:46:48.973138: Epoch time: 109.15 s 
+2026-10-18 16:46:49.735370:  
+2026-10-18 16:46:49.736026: Epoch 285 
+2026-10-18 16:46:49.736341: Current learning rate: 0.00739 
+2026-10-18 16:48:38.899959: train_loss -0.6379 
+2026-10-18 16:48:38.900844: val_loss -0.4734 
+2026-10-18 16:48:38.901379: Pseudo dice [np.float32(0.9317), np.float32(0.7105), np.float32(0.8103), np.float32(0.8107), np.float32(0.6967), np.float32(0.8575), np.float32(0.7765), np.float32(0.5126), np.float32(0.0), np.float32(0.4536), np.float32(0.6885), np.float32(0.736), np.float32(0.6637), np.float32(0.4167), np.float32(0.0), np.float32(0.0), np.float32(0.6745), np.float32(0.3424), np.float32(0.7055), np.float32(0.3547), np.float32(0.2131), np.float32(0.5864), np.float32(0.4809), np.float32(0.58), np.float32(0.7153), np.float32(0.6646), np.float32(0.2394), np.float32(0.7608), np.float32(0.0), np.float32(0.5855), np.float32(0.7049), np.float32(0.3388), np.float32(0.7126), np.float32(0.6577), np.float32(0.8091), np.float32(0.8426)] 
+2026-10-18 16:48:38.901782: Epoch time: 109.17 s 
+2026-10-18 16:48:39.631299:  
+2026-10-18 16:48:39.631990: Epoch 286 
+2026-10-18 16:48:39.632308: Current learning rate: 0.00738 
+2026-10-18 16:50:28.775002: train_loss -0.6131 
+2026-10-18 16:50:28.775851: val_loss -0.514 
+2026-10-18 16:50:28.776578: Pseudo dice [np.float32(0.9354), np.float32(0.6932), np.float32(0.8319), np.float32(0.8242), np.float32(0.7624), np.float32(0.8554), np.float32(0.8366), np.float32(0.4476), np.float32(0.0), np.float32(0.4521), np.float32(0.7207), np.float32(0.77), np.float32(0.6879), np.float32(0.5443), np.float32(0.0), np.float32(0.0), np.float32(0.7538), np.float32(0.5329), np.float32(0.7794), np.float32(0.5621), np.float32(0.4845), np.float32(0.6846), np.float32(0.7054), np.float32(0.7888), np.float32(0.7225), np.float32(0.6916), np.float32(0.4591), np.float32(0.8267), np.float32(0.0), np.float32(0.653), np.float32(0.6225), np.float32(0.3749), np.float32(0.7178), np.float32(0.6783), np.float32(0.8188), np.float32(0.8698)] 
+2026-10-18 16:50:28.776973: Epoch time: 109.15 s 
+2026-10-18 16:50:29.545936:  
+2026-10-18 16:50:29.546592: Epoch 287 
+2026-10-18 16:50:29.546864: Current learning rate: 0.00738 
+2026-10-18 16:52:18.672009: train_loss -0.6418 
+2026-10-18 16:52:18.672720: val_loss -0.547 
+2026-10-18 16:52:18.673277: Pseudo dice [np.float32(0.9134), np.float32(0.7455), np.float32(0.872), np.float32(0.8849), np.float32(0.7517), np.float32(0.9106), np.float32(0.8407), np.float32(0.4891), np.float32(0.0), np.float32(0.4968), np.float32(0.7951), np.float32(0.8082), np.float32(0.6553), np.float32(0.584), np.float32(0.0), np.float32(0.0), np.float32(0.7829), np.float32(0.4459), np.float32(0.7731), np.float32(0.4969), np.float32(0.3727), np.float32(0.7633), np.float32(0.691), np.float32(0.7375), np.float32(0.7463), np.float32(0.7087), np.float32(0.4909), np.float32(0.7703), np.float32(0.0), np.float32(0.6522), np.float32(0.6905), np.float32(0.5471), np.float32(0.7977), np.float32(0.704), np.float32(0.9186), np.float32(0.9263)] 
+2026-10-18 16:52:18.673670: Epoch time: 109.13 s 
+2026-10-18 16:52:19.397337:  
+2026-10-18 16:52:19.397950: Epoch 288 
+2026-10-18 16:52:19.398319: Current learning rate: 0.00737 
+2026-10-18 16:54:08.539594: train_loss -0.6489 
+2026-10-18 16:54:08.540454: val_loss -0.5482 
+2026-10-18 16:54:08.540947: Pseudo dice [np.float32(0.9421), np.float32(0.744), np.float32(0.8686), np.float32(0.8684), np.float32(0.749), np.float32(0.9095), np.float32(0.8513), np.float32(0.6341), np.float32(0.0), np.float32(0.5164), np.float32(0.7678), np.float32(0.8081), np.float32(0.711), np.float32(0.6061), np.float32(0.0), np.float32(0.0), np.float32(0.7601), np.float32(0.4837), np.float32(0.7812), np.float32(0.5563), np.float32(0.4985), np.float32(0.7668), np.float32(0.8403), np.float32(0.7587), np.float32(0.6786), np.float32(0.7443), np.float32(0.4831), np.float32(0.8601), np.float32(0.0), np.float32(0.7324), np.float32(0.7226), np.float32(0.5087), np.float32(0.8299), np.float32(0.7372), np.float32(0.9018), np.float32(0.925)] 
+2026-10-18 16:54:08.541517: Epoch time: 109.14 s 
+2026-10-18 16:54:09.314213:  
+2026-10-18 16:54:09.314926: Epoch 289 
+2026-10-18 16:54:09.315228: Current learning rate: 0.00736 
+2026-10-18 16:55:58.435438: train_loss -0.6634 
+2026-10-18 16:55:58.436224: val_loss -0.5495 
+2026-10-18 16:55:58.437140: Pseudo dice [np.float32(0.948), np.float32(0.7912), np.float32(0.8423), np.float32(0.8499), np.float32(0.75), np.float32(0.8937), np.float32(0.8559), np.float32(0.502), np.float32(0.0), np.float32(0.5675), np.float32(0.785), np.float32(0.8117), np.float32(0.6781), np.float32(0.6275), np.float32(0.0), np.float32(0.0), np.float32(0.7811), np.float32(0.5275), np.float32(0.7986), np.float32(0.4653), np.float32(0.3709), np.float32(0.6817), np.float32(0.7502), np.float32(0.7709), np.float32(0.7484), np.float32(0.6772), np.float32(0.5669), np.float32(0.6153), np.float32(0.0), np.float32(0.6972), np.float32(0.735), np.float32(0.4627), np.float32(0.7847), np.float32(0.6601), np.float32(0.8939), np.float32(0.9207)] 
+2026-10-18 16:55:58.437579: Epoch time: 109.12 s 
+2026-10-18 16:55:59.172929:  
+2026-10-18 16:55:59.173581: Epoch 290 
+2026-10-18 16:55:59.173762: Current learning rate: 0.00735 
+2026-10-18 16:57:48.305557: train_loss -0.6549 
+2026-10-18 16:57:48.306391: val_loss -0.5578 
+2026-10-18 16:57:48.307043: Pseudo dice [np.float32(0.9126), np.float32(0.7765), np.float32(0.859), np.float32(0.8638), np.float32(0.7626), np.float32(0.8999), np.float32(0.8118), np.float32(0.67), np.float32(0.0), np.float32(0.4434), np.float32(0.757), np.float32(0.8159), np.float32(0.729), np.float32(0.6544), np.float32(0.0), np.float32(0.0), np.float32(0.7646), np.float32(0.3636), np.float32(0.7744), np.float32(0.6124), np.float32(0.4773), np.float32(0.6948), np.float32(0.7873), np.float32(0.825), np.float32(0.7419), np.float32(0.7637), np.float32(0.3738), np.float32(0.8366), np.float32(0.0), np.float32(0.7816), np.float32(0.2711), np.float32(0.4954), np.float32(0.8063), np.float32(0.7451), np.float32(0.9379), np.float32(0.9571)] 
+2026-10-18 16:57:48.307520: Epoch time: 109.13 s 
+2026-10-18 16:57:49.079312:  
+2026-10-18 16:57:49.080012: Epoch 291 
+2026-10-18 16:57:49.080342: Current learning rate: 0.00734 
+2026-10-18 16:59:38.210161: train_loss -0.6459 
+2026-10-18 16:59:38.210913: val_loss -0.5339 
+2026-10-18 16:59:38.211486: Pseudo dice [np.float32(0.9397), np.float32(0.7468), np.float32(0.8613), np.float32(0.8561), np.float32(0.661), np.float32(0.8991), np.float32(0.8545), np.float32(0.4402), np.float32(0.0), np.float32(0.3435), np.float32(0.7678), np.float32(0.8083), np.float32(0.7477), np.float32(0.6171), np.float32(0.0), np.float32(0.0), np.float32(0.6165), np.float32(0.1882), np.float32(0.7232), np.float32(0.5249), np.float32(0.3491), np.float32(0.6884), np.float32(0.8146), np.float32(0.8454), np.float32(0.7484), np.float32(0.7418), np.float32(0.5352), np.float32(0.8448), np.float32(0.0), np.float32(0.7708), np.float32(0.5976), np.float32(0.3574), np.float32(0.7679), np.float32(0.7546), np.float32(0.9185), np.float32(0.9404)] 
+2026-10-18 16:59:38.212182: Epoch time: 109.13 s 
+2026-10-18 16:59:38.953015:  
+2026-10-18 16:59:38.953581: Epoch 292 
+2026-10-18 16:59:38.953968: Current learning rate: 0.00733 
+2026-10-18 17:01:28.087778: train_loss -0.6275 
+2026-10-18 17:01:28.088664: val_loss -0.5294 
+2026-10-18 17:01:28.089161: Pseudo dice [np.float32(0.9285), np.float32(0.7107), np.float32(0.8852), np.float32(0.8472), np.float32(0.5942), np.float32(0.8899), np.float32(0.7754), np.float32(0.359), np.float32(0.0), np.float32(0.4863), np.float32(0.7478), np.float32(0.7812), np.float32(0.603), np.float32(0.1764), np.float32(0.0), np.float32(0.0), np.float32(0.7437), np.float32(0.5052), np.float32(0.7128), np.float32(0.5935), np.float32(0.4624), np.float32(0.761), np.float32(0.7708), np.float32(0.7338), np.float32(0.7462), np.float32(0.7689), np.float32(0.4538), np.float32(0.8204), np.float32(0.0), np.float32(0.7373), np.float32(0.682), np.float32(0.5378), np.float32(0.7664), np.float32(0.6917), np.float32(0.8761), np.float32(0.9107)] 
+2026-10-18 17:01:28.089454: Epoch time: 109.14 s 
+2026-10-18 17:01:28.824863:  
+2026-10-18 17:01:28.825546: Epoch 293 
+2026-10-18 17:01:28.825815: Current learning rate: 0.00732 
+2026-10-18 17:03:17.959074: train_loss -0.6575 
+2026-10-18 17:03:17.960028: val_loss -0.5283 
+2026-10-18 17:03:17.960545: Pseudo dice [np.float32(0.9358), np.float32(0.7571), np.float32(0.8347), np.float32(0.8639), np.float32(0.7821), np.float32(0.9159), np.float32(0.8363), np.float32(0.4879), np.float32(0.0), np.float32(0.3512), np.float32(0.7746), np.float32(0.8142), np.float32(0.6593), np.float32(0.3126), np.float32(0.0), np.float32(0.0), np.float32(0.7522), np.float32(0.3977), np.float32(0.7794), np.float32(0.5129), np.float32(0.3739), np.float32(0.6956), np.float32(0.8667), np.float32(0.8488), np.float32(0.7285), np.float32(0.686), np.float32(0.3782), np.float32(0.7899), np.float32(0.0), np.float32(0.4343), np.float32(0.6794), np.float32(0.2511), np.float32(0.7643), np.float32(0.7058), np.float32(0.9241), np.float32(0.9454)] 
+2026-10-18 17:03:17.961017: Epoch time: 109.14 s 
+2026-10-18 17:03:18.969223:  
+2026-10-18 17:03:18.969796: Epoch 294 
+2026-10-18 17:03:18.970453: Current learning rate: 0.00731 
+2026-10-18 17:05:08.184722: train_loss -0.6609 
+2026-10-18 17:05:08.185400: val_loss -0.5402 
+2026-10-18 17:05:08.186114: Pseudo dice [np.float32(0.9258), np.float32(0.736), np.float32(0.8691), np.float32(0.8802), np.float32(0.7017), np.float32(0.9195), np.float32(0.8442), np.float32(0.401), np.float32(0.0), np.float32(0.3364), np.float32(0.7705), np.float32(0.7954), np.float32(0.6322), np.float32(0.4294), np.float32(0.0), np.float32(0.0), np.float32(0.7934), np.float32(0.5391), np.float32(0.798), np.float32(0.6333), np.float32(0.437), np.float32(0.6989), np.float32(0.7243), np.float32(0.7524), np.float32(0.7112), np.float32(0.7929), np.float32(0.3951), np.float32(0.8018), np.float32(0.0), np.float32(0.5557), np.float32(0.678), np.float32(0.1093), np.float32(0.809), np.float32(0.7134), np.float32(0.9195), np.float32(0.9388)] 
+2026-10-18 17:05:08.186745: Epoch time: 109.22 s 
+2026-10-18 17:05:08.928057:  
+2026-10-18 17:05:08.928505: Epoch 295 
+2026-10-18 17:05:08.928717: Current learning rate: 0.0073 
+2026-10-18 17:06:58.047680: train_loss -0.6669 
+2026-10-18 17:06:58.048172: val_loss -0.5683 
+2026-10-18 17:06:58.048724: Pseudo dice [np.float32(0.9363), np.float32(0.7988), np.float32(0.8773), np.float32(0.8977), np.float32(0.7598), np.float32(0.9245), np.float32(0.8341), np.float32(0.4814), np.float32(0.0), np.float32(0.406), np.float32(0.788), np.float32(0.836), np.float32(0.6519), np.float32(0.3704), np.float32(0.0), np.float32(0.0), np.float32(0.8249), np.float32(0.5645), np.float32(0.8108), np.float32(0.6692), np.float32(0.4274), np.float32(0.7945), np.float32(0.8228), np.float32(0.8171), np.float32(0.7328), np.float32(0.7706), np.float32(0.4864), np.float32(0.7509), np.float32(0.0), np.float32(0.6992), np.float32(0.7725), np.float32(0.2004), np.float32(0.8465), np.float32(0.7007), np.float32(0.9422), np.float32(0.9634)] 
+2026-10-18 17:06:58.049774: Epoch time: 109.12 s 
+2026-10-18 17:06:58.806658:  
+2026-10-18 17:06:58.807363: Epoch 296 
+2026-10-18 17:06:58.807711: Current learning rate: 0.00729 
+2026-10-18 17:08:47.951080: train_loss -0.6481 
+2026-10-18 17:08:47.951944: val_loss -0.5013 
+2026-10-18 17:08:47.952627: Pseudo dice [np.float32(0.9495), np.float32(0.7176), np.float32(0.8021), np.float32(0.7546), np.float32(0.6177), np.float32(0.8177), np.float32(0.7369), np.float32(0.6783), np.float32(0.0), np.float32(0.3289), np.float32(0.7106), np.float32(0.717), np.float32(0.6304), np.float32(0.6371), np.float32(0.0), np.float32(0.0), np.float32(0.7236), np.float32(0.4535), np.float32(0.6971), np.float32(0.5214), np.float32(0.4358), np.float32(0.7558), np.float32(0.8168), np.float32(0.8474), np.float32(0.6953), np.float32(0.7356), np.float32(0.5269), np.float32(0.7464), np.float32(0.0), np.float32(0.684), np.float32(0.7415), np.float32(0.3561), np.float32(0.7597), np.float32(0.5275), np.float32(0.817), np.float32(0.8299)] 
+2026-10-18 17:08:47.953023: Epoch time: 109.15 s 
+2026-10-18 17:08:48.704368:  
+2026-10-18 17:08:48.704875: Epoch 297 
+2026-10-18 17:08:48.705146: Current learning rate: 0.00728 
+2026-10-18 17:10:37.887274: train_loss -0.6156 
+2026-10-18 17:10:37.888021: val_loss -0.4829 
+2026-10-18 17:10:37.888767: Pseudo dice [np.float32(0.9324), np.float32(0.6745), np.float32(0.7895), np.float32(0.8393), np.float32(0.6341), np.float32(0.8925), np.float32(0.812), np.float32(0.5646), np.float32(0.0), np.float32(0.3581), np.float32(0.748), np.float32(0.7742), np.float32(0.6698), np.float32(0.5131), np.float32(0.0), np.float32(0.0), np.float32(0.7102), np.float32(0.3635), np.float32(0.7199), np.float32(0.3913), np.float32(0.4151), np.float32(0.7311), np.float32(0.4203), np.float32(0.4456), np.float32(0.6517), np.float32(0.6603), np.float32(0.3227), np.float32(0.6013), np.float32(0.0), np.float32(0.5196), np.float32(0.6674), np.float32(0.5161), np.float32(0.7994), np.float32(0.7457), np.float32(0.7886), np.float32(0.8241)] 
+2026-10-18 17:10:37.889077: Epoch time: 109.18 s 
+2026-10-18 17:10:38.631091:  
+2026-10-18 17:10:38.631668: Epoch 298 
+2026-10-18 17:10:38.631876: Current learning rate: 0.00727 
+2026-10-18 17:12:27.767972: train_loss -0.619 
+2026-10-18 17:12:27.768804: val_loss -0.5331 
+2026-10-18 17:12:27.769371: Pseudo dice [np.float32(0.9345), np.float32(0.7496), np.float32(0.8103), np.float32(0.8673), np.float32(0.7488), np.float32(0.9201), np.float32(0.8394), np.float32(0.6703), np.float32(0.0), np.float32(0.2855), np.float32(0.7743), np.float32(0.808), np.float32(0.6944), np.float32(0.6311), np.float32(0.0), np.float32(0.0), np.float32(0.7864), np.float32(0.5515), np.float32(0.7008), np.float32(0.5931), np.float32(0.4272), np.float32(0.6718), np.float32(0.582), np.float32(0.818), np.float32(0.6514), np.float32(0.6545), np.float32(0.4539), np.float32(0.844), np.float32(0.0), np.float32(0.65), np.float32(0.715), np.float32(0.4569), np.float32(0.7948), np.float32(0.7062), np.float32(0.9132), np.float32(0.9497)] 
+2026-10-18 17:12:27.769660: Epoch time: 109.14 s 
+2026-10-18 17:12:28.511600:  
+2026-10-18 17:12:28.512361: Epoch 299 
+2026-10-18 17:12:28.512568: Current learning rate: 0.00726 
+2026-10-18 17:14:17.639216: train_loss -0.6326 
+2026-10-18 17:14:17.639812: val_loss -0.5452 
+2026-10-18 17:14:17.640315: Pseudo dice [np.float32(0.9342), np.float32(0.6819), np.float32(0.8933), np.float32(0.8516), np.float32(0.6899), np.float32(0.9221), np.float32(0.8477), np.float32(0.4196), np.float32(0.0), np.float32(0.5286), np.float32(0.8128), np.float32(0.848), np.float32(0.7731), np.float32(0.7095), np.float32(0.0), np.float32(0.0), np.float32(0.7768), np.float32(0.5435), np.float32(0.8356), np.float32(0.5881), np.float32(0.2774), np.float32(0.5964), np.float32(0.7102), np.float32(0.7432), np.float32(0.6782), np.float32(0.7544), np.float32(0.5197), np.float32(0.8303), np.float32(0.0), np.float32(0.6439), np.float32(0.4476), np.float32(0.5451), np.float32(0.8511), np.float32(0.716), np.float32(0.9033), np.float32(0.9398)] 
+2026-10-18 17:14:17.640910: Epoch time: 109.13 s 
+2026-10-18 17:14:19.157745:  
+2026-10-18 17:14:19.158458: Epoch 300 
+2026-10-18 17:14:19.158824: Current learning rate: 0.00725 
+2026-10-18 17:16:08.290313: train_loss -0.6583 
+2026-10-18 17:16:08.291119: val_loss -0.5627 
+2026-10-18 17:16:08.291784: Pseudo dice [np.float32(0.9415), np.float32(0.7648), np.float32(0.8842), np.float32(0.8404), np.float32(0.7219), np.float32(0.8981), np.float32(0.823), np.float32(0.5611), np.float32(0.0), np.float32(0.465), np.float32(0.7701), np.float32(0.8131), np.float32(0.665), np.float32(0.6918), np.float32(0.0), np.float32(0.0), np.float32(0.763), np.float32(0.5533), np.float32(0.7721), np.float32(0.5985), np.float32(0.5203), np.float32(0.8114), np.float32(0.8456), np.float32(0.8585), np.float32(0.7495), np.float32(0.7409), np.float32(0.5079), np.float32(0.8157), np.float32(0.0), np.float32(0.5461), np.float32(0.7309), np.float32(0.504), np.float32(0.7676), np.float32(0.6992), np.float32(0.9362), np.float32(0.958)] 
+2026-10-18 17:16:08.292038: Epoch time: 109.13 s 
+2026-10-18 17:16:09.024482:  
+2026-10-18 17:16:09.025210: Epoch 301 
+2026-10-18 17:16:09.025540: Current learning rate: 0.00724 
+2026-10-18 17:17:58.120729: train_loss -0.6638 
+2026-10-18 17:17:58.121272: val_loss -0.5442 
+2026-10-18 17:17:58.122129: Pseudo dice [np.float32(0.9329), np.float32(0.7345), np.float32(0.8669), np.float32(0.8886), np.float32(0.7832), np.float32(0.8906), np.float32(0.8656), np.float32(0.6111), np.float32(0.0), np.float32(0.3776), np.float32(0.7743), np.float32(0.7975), np.float32(0.6747), np.float32(0.6303), np.float32(0.0), np.float32(0.0), np.float32(0.8062), np.float32(0.515), np.float32(0.8303), np.float32(0.6669), np.float32(0.3952), np.float32(0.7558), np.float32(0.7433), np.float32(0.7483), np.float32(0.7268), np.float32(0.7509), np.float32(0.4213), np.float32(0.8181), np.float32(0.0), np.float32(0.6278), np.float32(0.6566), np.float32(0.5399), np.float32(0.8249), np.float32(0.6976), np.float32(0.9412), np.float32(0.9478)] 
+2026-10-18 17:17:58.122574: Epoch time: 109.1 s 
+2026-10-18 17:17:58.864273:  
+2026-10-18 17:17:58.864940: Epoch 302 
+2026-10-18 17:17:58.865216: Current learning rate: 0.00724 
diff --git a/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/plans.json b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/plans.json
new file mode 100644
index 0000000000000000000000000000000000000000..7046385975bd04870859998f65e0cd4444bca72d
--- /dev/null
+++ b/nnUNet_Large/fold_0_MRA/nnUNet_results/Dataset001_MRA/plans.json
@@ -0,0 +1,545 @@
+{
+    "dataset_name": "Dataset001_MRA",
+    "plans_name": "nnUNetResEncUNetLPlans",
+    "original_median_spacing_after_transp": [
+        0.5999984741210938,
+        0.296875,
+        0.296875
+    ],
+    "original_median_shape_after_transp": [
+        186,
+        576,
+        480
+    ],
+    "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": 26,
+            "patch_size": [
+                640,
+                512
+            ],
+            "median_image_size_in_voxels": [
+                577.0,
+                480.0
+            ],
+            "spacing": [
+                0.296875,
+                0.296875
+            ],
+            "normalization_schemes": [
+                "ZScoreNormalization"
+            ],
+            "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": 8,
+                    "features_per_stage": [
+                        32,
+                        64,
+                        128,
+                        256,
+                        512,
+                        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
+                        ],
+                        [
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2
+                        ]
+                    ],
+                    "n_blocks_per_stage": [
+                        1,
+                        3,
+                        4,
+                        6,
+                        6,
+                        6,
+                        6,
+                        6
+                    ],
+                    "n_conv_per_stage_decoder": [
+                        1,
+                        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_lowres": {
+            "data_identifier": "nnUNetResEncUNetLPlans_3d_lowres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                96,
+                256,
+                256
+            ],
+            "median_image_size_in_voxels": [
+                147,
+                442,
+                368
+            ],
+            "spacing": [
+                0.7600601158948237,
+                0.387354538949307,
+                0.387354538949307
+            ],
+            "normalization_schemes": [
+                "ZScoreNormalization"
+            ],
+            "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,
+                        320,
+                        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
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            2,
+                            2,
+                            2
+                        ],
+                        [
+                            1,
+                            2,
+                            2
+                        ],
+                        [
+                            1,
+                            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.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,
+            "next_stage": "3d_cascade_fullres"
+        },
+        "3d_fullres": {
+            "data_identifier": "nnUNetPlans_3d_fullres",
+            "preprocessor_name": "DefaultPreprocessor",
+            "batch_size": 2,
+            "patch_size": [
+                80,
+                256,
+                224
+            ],
+            "median_image_size_in_voxels": [
+                186.0,
+                577.0,
+                480.0
+            ],
+            "spacing": [
+                0.5999984741210938,
+                0.296875,
+                0.296875
+            ],
+            "normalization_schemes": [
+                "ZScoreNormalization"
+            ],
+            "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": [
+                        [
+                            1,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ],
+                        [
+                            3,
+                            3,
+                            3
+                        ]
+                    ],
+                    "strides": [
+                        [
+                            1,
+                            1,
+                            1
+                        ],
+                        [
+                            1,
+                            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": true
+        },
+        "3d_cascade_fullres": {
+            "inherits_from": "3d_fullres",
+            "previous_stage": "3d_lowres"
+        }
+    },
+    "experiment_planner_used": "nnUNetPlannerResEncL",
+    "label_manager": "LabelManager",
+    "foreground_intensity_properties_per_channel": {
+        "0": {
+            "max": 1421.0,
+            "mean": 324.5860900878906,
+            "median": 274.9985656738281,
+            "min": 52.99739456176758,
+            "percentile_00_5": 88.00049591064453,
+            "percentile_99_5": 896.009521484375,
+            "std": 170.6415252685547
+        }
+    }
+}
\ No newline at end of file