diff --git a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/debug.json b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/debug.json
index d7335f0e99253986b046024554661180e4680bf2..8380d11ca8d1f99abeb36b63401da334cb6e6fa8 100644
--- a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/debug.json
+++ b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/debug.json
@@ -1,28 +1,28 @@
 {
-    "all_tr_losses": "[]",
-    "all_val_eval_metrics": "[]",
-    "all_val_losses": "[]",
+    "all_tr_losses": "[np.float32(-0.00619347), np.float32(-0.10341911), np.float32(-0.14032517), np.float32(-0.23299958), np.float32(-0.3604002), np.float32(-0.39218843), np.float32(-0.41602528), np.float32(-0.43798658), np.float32(-0.44917554), np.float32(-0.4739833), np.float32(-0.48512065), np.float32(-0.5037208), np.float32(-0.51640576), np.float32(-0.5277429), np.float32(-0.54611784), np.float32(-0.53375), np.float32(-0.5287878), np.float32(-0.5542937), np.float32(-0.564019), np.float32(-0.54377896), np.float32(-0.5562092), np.float32(-0.5558722), np.float32(-0.5597891), np.float32(-0.560288), np.float32(-0.5619807), np.float32(-0.5994977), np.float32(-0.5956461), np.float32(-0.5856571), np.float32(-0.58555466), np.float32(-0.5763671), np.float32(-0.57683206), np.float32(-0.6019021), np.float32(-0.5865283), np.float32(-0.60887104), np.float32(-0.6025605), np.float32(-0.5926334), np.float32(-0.60531735), np.float32(-0.59241414), np.float32(-0.58998334), np.float32(-0.596204), np.float32(-0.59940493), np.float32(-0.6056025), np.float32(-0.6083389), np.float32(-0.6026457), np.float32(-0.6297592), np.float32(-0.6125459), np.float32(-0.6156312), np.float32(-0.63373965), np.float32(-0.6381857), np.float32(-0.5988075), np.float32(-0.6258546), np.float32(-0.627999), np.float32(-0.6357876), np.float32(-0.6227955), np.float32(-0.6353634), np.float32(-0.6266343), np.float32(-0.63454187), np.float32(-0.6360167), np.float32(-0.6414164), np.float32(-0.60830635), np.float32(-0.64541763), np.float32(-0.64118505), np.float32(-0.63678193), np.float32(-0.65113395), np.float32(-0.6356421), np.float32(-0.6580051), np.float32(-0.6221771), np.float32(-0.6404337), np.float32(-0.639511), np.float32(-0.6571746), np.float32(-0.63965994), np.float32(-0.6642626), np.float32(-0.6519752), np.float32(-0.6377228), np.float32(-0.5759572), np.float32(-0.64472395), np.float32(-0.63760936), np.float32(-0.6343812), np.float32(-0.61629355), np.float32(-0.630627), np.float32(-0.6542434), np.float32(-0.6696007), np.float32(-0.6607596), np.float32(-0.63885653), np.float32(-0.6651032), np.float32(-0.63741034), np.float32(-0.63297486), np.float32(-0.6577664), np.float32(-0.62819195), np.float32(-0.6408937), np.float32(-0.6297852), np.float32(-0.64417166), np.float32(-0.64698964), np.float32(-0.67661065), np.float32(-0.6715639), np.float32(-0.607041), np.float32(-0.6678004), np.float32(-0.67538106), np.float32(-0.67728096), np.float32(-0.67555153), np.float32(-0.6557017), np.float32(-0.6743387), np.float32(-0.6760017), np.float32(-0.65949), np.float32(-0.67742866), np.float32(-0.68200195), np.float32(-0.66392124), np.float32(-0.6514964), np.float32(-0.6440743), np.float32(-0.6549479), np.float32(-0.67750925), np.float32(-0.6570667), np.float32(-0.68106955), np.float32(-0.65679663), np.float32(-0.66916406), np.float32(-0.6765846), np.float32(-0.67356825), np.float32(-0.677326), np.float32(-0.6720115), np.float32(-0.67883986), np.float32(-0.68930095), np.float32(-0.6935803), np.float32(-0.6924679), np.float32(-0.68172985), np.float32(-0.6964419), np.float32(-0.6892777), np.float32(-0.6906535), np.float32(-0.6896339), np.float32(-0.7025404), np.float32(-0.69486934), np.float32(-0.6670771), np.float32(-0.6703752), np.float32(-0.662454), np.float32(-0.6648569), np.float32(-0.690099), np.float32(-0.67067564), np.float32(-0.69051474), np.float32(-0.69158095), np.float32(-0.6997817), np.float32(-0.6977838), np.float32(-0.7082537), np.float32(-0.68389326), np.float32(-0.6728365), np.float32(-0.69166607), np.float32(-0.636832), np.float32(-0.65991604), np.float32(-0.68275285), np.float32(-0.69209373), np.float32(-0.69461286), np.float32(-0.67649)]",
+    "all_val_eval_metrics": "[np.float32(0.0), np.float32(0.01889988), np.float32(0.079845764), np.float32(0.13679223), np.float32(0.17857546), np.float32(0.16732009), np.float32(0.19895451), np.float32(0.23460743), np.float32(0.21075362), np.float32(0.19120714), np.float32(0.3070274), np.float32(0.22336139), np.float32(0.19358961), np.float32(0.102281615), np.float32(0.21802288), np.float32(0.31315908), np.float32(0.24491273), np.float32(0.39693174), np.float32(0.39315566), np.float32(0.38618338), np.float32(0.31200612), np.float32(0.2930611), np.float32(0.24451402), np.float32(0.14668918), np.float32(0.21874493), np.float32(0.29099023), np.float32(0.13072735), np.float32(0.16563323), np.float32(0.1763662), np.float32(0.3780376), np.float32(0.2616179), np.float32(0.28131905), np.float32(0.30587274), np.float32(0.41512328), np.float32(0.2894128), np.float32(0.30601633), np.float32(0.2499519), np.float32(0.11581434), np.float32(0.38822114), np.float32(0.25973985), np.float32(0.25034806), np.float32(0.43522474), np.float32(0.06621247), np.float32(0.30407128), np.float32(0.14799121), np.float32(0.27650657), np.float32(0.28254282), np.float32(0.32050848), np.float32(0.22262205), np.float32(0.30573395), np.float32(0.27453232), np.float32(0.28947836), np.float32(0.30398092), np.float32(0.30879176), np.float32(0.31541094), np.float32(0.19862843), np.float32(0.31306612), np.float32(0.29480532), np.float32(0.40565488), np.float32(0.3148551), np.float32(0.33322158), np.float32(0.40734234), np.float32(0.4045507), np.float32(0.2957369), np.float32(0.36678064), np.float32(0.41890493), np.float32(0.3286036), np.float32(0.34367356), np.float32(0.30242655), np.float32(0.41726828), np.float32(0.34221464), np.float32(0.26905036), np.float32(0.34067935), np.float32(0.051818848), np.float32(0.34491572), np.float32(0.27824086), np.float32(0.28361464), np.float32(0.21679063), np.float32(0.29471993), np.float32(0.31009927), np.float32(0.38593453), np.float32(0.36058825), np.float32(0.44014895), np.float32(0.2723917), np.float32(0.2864226), np.float32(0.36489522), np.float32(0.33008677), np.float32(0.39029258), np.float32(0.29772595), np.float32(0.28772676), np.float32(0.36572763), np.float32(0.28292063), np.float32(0.40132305), np.float32(0.41895068), np.float32(0.42759106), np.float32(0.279621), np.float32(0.37066427), np.float32(0.38915932), np.float32(0.3620605), np.float32(0.24416025), np.float32(0.31823605), np.float32(0.3094102), np.float32(0.4071442), np.float32(0.4421763), np.float32(0.42380676), np.float32(0.3315647), np.float32(0.18343663), np.float32(0.4196011), np.float32(0.4404834), np.float32(0.44806555), np.float32(0.3706316), np.float32(0.40587637), np.float32(0.4205264), np.float32(0.41722485), np.float32(0.4371354), np.float32(0.28766468), np.float32(0.34353897), np.float32(0.44338265), np.float32(0.4761868), np.float32(0.32436955), np.float32(0.34003386), np.float32(0.30145895), np.float32(0.4609412), np.float32(0.4317714), np.float32(0.3996956), np.float32(0.41997683), np.float32(0.42910394), np.float32(0.32100934), np.float32(0.36465207), np.float32(0.41524178), np.float32(0.35485768), np.float32(0.25318298), np.float32(0.40483916), np.float32(0.27552342), np.float32(0.31775463), np.float32(0.4412951), np.float32(0.44127825), np.float32(0.41657934), np.float32(0.40827367), np.float32(0.41474578), np.float32(0.40083566), np.float32(0.31862867), np.float32(0.42084605), np.float32(0.4308671), np.float32(0.28507146), np.float32(0.23676999), np.float32(0.4359449), np.float32(0.4233735), np.float32(0.19071323), np.float32(0.33704674)]",
+    "all_val_losses": "[np.float32(-0.07470091), np.float32(-0.1155705), np.float32(-0.19333363), np.float32(-0.27809945), np.float32(-0.37054774), np.float32(-0.3565181), np.float32(-0.3831776), np.float32(-0.401925), np.float32(-0.38338625), np.float32(-0.36536282), np.float32(-0.45441592), np.float32(-0.37744793), np.float32(-0.3529129), np.float32(-0.2656958), np.float32(-0.37089065), np.float32(-0.463354), np.float32(-0.37575164), np.float32(-0.51616085), np.float32(-0.52158797), np.float32(-0.49252146), np.float32(-0.46345484), np.float32(-0.44432026), np.float32(-0.3875055), np.float32(-0.29651347), np.float32(-0.374561), np.float32(-0.44172478), np.float32(-0.3020468), np.float32(-0.3249737), np.float32(-0.33035675), np.float32(-0.48948795), np.float32(-0.41371545), np.float32(-0.43393955), np.float32(-0.46594632), np.float32(-0.5379311), np.float32(-0.4374967), np.float32(-0.46398762), np.float32(-0.40548974), np.float32(-0.26895827), np.float32(-0.5160663), np.float32(-0.4170287), np.float32(-0.3874871), np.float32(-0.56852984), np.float32(-0.20573968), np.float32(-0.4403778), np.float32(-0.28382945), np.float32(-0.4249569), np.float32(-0.4226168), np.float32(-0.4591214), np.float32(-0.36560845), np.float32(-0.43757272), np.float32(-0.4334856), np.float32(-0.44014493), np.float32(-0.45037532), np.float32(-0.45998502), np.float32(-0.4603461), np.float32(-0.34373438), np.float32(-0.46866146), np.float32(-0.44739568), np.float32(-0.52187514), np.float32(-0.46080753), np.float32(-0.478531), np.float32(-0.53193355), np.float32(-0.5221857), np.float32(-0.4397172), np.float32(-0.50099033), np.float32(-0.544881), np.float32(-0.48568553), np.float32(-0.48982775), np.float32(-0.45025373), np.float32(-0.5496275), np.float32(-0.49164483), np.float32(-0.4180044), np.float32(-0.48074684), np.float32(-0.20900002), np.float32(-0.49134067), np.float32(-0.43336174), np.float32(-0.43484515), np.float32(-0.3493756), np.float32(-0.44303352), np.float32(-0.46257916), np.float32(-0.52277595), np.float32(-0.50028944), np.float32(-0.5594064), np.float32(-0.42554745), np.float32(-0.43165272), np.float32(-0.49765787), np.float32(-0.4590839), np.float32(-0.51956576), np.float32(-0.4452071), np.float32(-0.43306908), np.float32(-0.4986575), np.float32(-0.42399925), np.float32(-0.5183483), np.float32(-0.5418989), np.float32(-0.54387313), np.float32(-0.41721138), np.float32(-0.5027693), np.float32(-0.50631696), np.float32(-0.5093173), np.float32(-0.3821223), np.float32(-0.4637021), np.float32(-0.44346252), np.float32(-0.52291477), np.float32(-0.5620501), np.float32(-0.55358887), np.float32(-0.48309964), np.float32(-0.32760236), np.float32(-0.57253706), np.float32(-0.5717862), np.float32(-0.5657669), np.float32(-0.49317795), np.float32(-0.5244601), np.float32(-0.544916), np.float32(-0.5373737), np.float32(-0.56530374), np.float32(-0.44027734), np.float32(-0.48818517), np.float32(-0.566051), np.float32(-0.6162163), np.float32(-0.45978385), np.float32(-0.47115567), np.float32(-0.44392464), np.float32(-0.5962756), np.float32(-0.5525789), np.float32(-0.5332187), np.float32(-0.5693787), np.float32(-0.5575467), np.float32(-0.46964335), np.float32(-0.5128202), np.float32(-0.53805345), np.float32(-0.48748544), np.float32(-0.3792828), np.float32(-0.52583545), np.float32(-0.4357098), np.float32(-0.46287668), np.float32(-0.5607835), np.float32(-0.5687721), np.float32(-0.5376203), np.float32(-0.5400239), np.float32(-0.5449283), np.float32(-0.51622903), np.float32(-0.45223755), np.float32(-0.5456331), np.float32(-0.5487034), np.float32(-0.43057808), np.float32(-0.3716735), np.float32(-0.55640674), np.float32(-0.5403391), np.float32(-0.33054656), np.float32(-0.47454056)]",
     "all_val_losses_tr_mode": "[]",
     "also_val_in_tr_mode": "False",
-    "amp_grad_scaler": "None",
+    "amp_grad_scaler": "<torch.cuda.amp.grad_scaler.GradScaler object at 0x7efb50930ad0>",
     "base_num_features": "30",
     "basic_generator_patch_size": "[205 205 205]",
     "batch_dice": "False",
     "batch_size": "2",
-    "best_MA_tr_loss_for_patience": "None",
-    "best_epoch_based_on_MA_tr_loss": "None",
-    "best_val_eval_criterion_MA": "None",
+    "best_MA_tr_loss_for_patience": "-0.6829764",
+    "best_epoch_based_on_MA_tr_loss": "140",
+    "best_val_eval_criterion_MA": "0.3936047",
     "classes": "[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36]",
     "conv_per_stage": "2",
     "data_aug_params": "{'selected_data_channels': None, 'selected_seg_channels': [0], 'do_elastic': False, 'elastic_deform_alpha': (0.0, 900.0), 'elastic_deform_sigma': (9.0, 13.0), 'p_eldef': 0.2, 'do_scaling': True, 'scale_range': (0.7, 1.4), 'independent_scale_factor_for_each_axis': False, 'p_independent_scale_per_axis': 1, 'p_scale': 0.2, 'do_rotation': True, 'rotation_x': (-0.5235987755982988, 0.5235987755982988), 'rotation_y': (-0.5235987755982988, 0.5235987755982988), 'rotation_z': (-0.5235987755982988, 0.5235987755982988), 'rotation_p_per_axis': 1, 'p_rot': 0.2, 'random_crop': False, 'random_crop_dist_to_border': None, 'do_gamma': True, 'gamma_retain_stats': True, 'gamma_range': (0.7, 1.5), 'p_gamma': 0.3, 'do_mirror': True, 'mirror_axes': (0, 1, 2), 'dummy_2D': False, 'mask_was_used_for_normalization': OrderedDict({0: False}), 'border_mode_data': 'constant', 'all_segmentation_labels': None, 'move_last_seg_chanel_to_data': False, 'cascade_do_cascade_augmentations': False, 'cascade_random_binary_transform_p': 0.4, 'cascade_random_binary_transform_p_per_label': 1, 'cascade_random_binary_transform_size': (1, 8), 'cascade_remove_conn_comp_p': 0.2, 'cascade_remove_conn_comp_max_size_percent_threshold': 0.15, 'cascade_remove_conn_comp_fill_with_other_class_p': 0.0, 'do_additive_brightness': False, 'additive_brightness_p_per_sample': 0.15, 'additive_brightness_p_per_channel': 0.5, 'additive_brightness_mu': 0.0, 'additive_brightness_sigma': 0.1, 'num_threads': 8, 'num_cached_per_thread': 2, 'patch_size_for_spatialtransform': array([128, 128, 128])}",
     "dataset_directory": "/marimo/TopBrain2026/nnUNet_preprocessed/Task001_Brain2026",
     "deep_supervision_scales": "[[1, 1, 1], [np.float64(0.5), np.float64(0.5), np.float64(0.5)], [np.float64(0.25), np.float64(0.25), np.float64(0.25)], [np.float64(0.125), np.float64(0.125), np.float64(0.125)], [np.float64(0.0625), np.float64(0.0625), np.float64(0.0625)], [np.float64(0.03125), np.float64(0.03125), np.float64(0.03125)]]",
     "deterministic": "False",
-    "dl_tr": "<nnunet_mednext.training.dataloading.dataset_loading.DataLoader3D object at 0x7efe777c1fd0>",
-    "dl_val": "<nnunet_mednext.training.dataloading.dataset_loading.DataLoader3D object at 0x7efe778811d0>",
+    "dl_tr": "<nnunet_mednext.training.dataloading.dataset_loading.DataLoader3D object at 0x7efb50b09fd0>",
+    "dl_val": "<nnunet_mednext.training.dataloading.dataset_loading.DataLoader3D object at 0x7efb509411d0>",
     "do_dummy_2D_aug": "False",
     "ds_loss_weights": "[0.51612903 0.25806452 0.12903226 0.06451613 0.03225806 0.        ]",
-    "epoch": "0",
+    "epoch": "150",
     "experiment_name": "nnUNetTrainerV2_MedNeXt_L_kernel3",
     "fold": "0",
     "folder_with_preprocessed_data": "/marimo/TopBrain2026/nnUNet_preprocessed/Task001_Brain2026/nnUNetData_plans_v2.1_trgSp_1x1x1_stage0",
@@ -32,7 +32,7 @@
     "inference_pad_kwargs": "{'constant_values': 0}",
     "init_args": "('/marimo/TopBrain2026/nnUNet_preprocessed/Task001_Brain2026/nnUNetPlansv2.1_trgSp_1x1x1_plans_3D.pkl', 0, '/marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1', '/marimo/TopBrain2026/nnUNet_preprocessed/Task001_Brain2026', False, 0, True, False, True)",
     "initial_lr": "0.001",
-    "log_file": "/marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_11_7_14_01_19.txt",
+    "log_file": "/marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_11_8_06_34_55.txt",
     "lr_scheduler": "None",
     "lr_scheduler_eps": "0.001",
     "lr_scheduler_patience": "30",
@@ -53,7 +53,7 @@
     "online_eval_fp": "[]",
     "online_eval_tp": "[]",
     "only_keep_largest_connected_component": "None",
-    "optimizer": "AdamW (\nParameter Group 0\n    amsgrad: False\n    betas: (0.9, 0.999)\n    capturable: False\n    decoupled_weight_decay: True\n    differentiable: False\n    eps: 0.0001\n    foreach: None\n    fused: None\n    lr: 0.001\n    maximize: False\n    weight_decay: 3e-05\n)",
+    "optimizer": "AdamW (\nParameter Group 0\n    amsgrad: False\n    betas: (0.9, 0.999)\n    capturable: False\n    decoupled_weight_decay: True\n    differentiable: False\n    eps: 0.0001\n    foreach: None\n    fused: None\n    lr: 0.0008639269723028191\n    maximize: False\n    weight_decay: 3e-05\n)",
     "output_folder": "/marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0",
     "output_folder_base": "/marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1",
     "oversample_foreground_percent": "0.33",
@@ -71,7 +71,7 @@
     "save_latest_only": "True",
     "stage": "0",
     "threeD": "True",
-    "tr_gen": "<batchgenerators.dataloading.multi_threaded_augmenter.MultiThreadedAugmenter object at 0x7efe777c3620>",
+    "tr_gen": "<batchgenerators.dataloading.multi_threaded_augmenter.MultiThreadedAugmenter object at 0x7efb50b0b620>",
     "train_loss_MA": "None",
     "train_loss_MA_alpha": "0.93",
     "train_loss_MA_eps": "0.0005",
@@ -82,7 +82,7 @@
     "use_progress_bar": "False",
     "val_eval_criterion_MA": "None",
     "val_eval_criterion_alpha": "0.9",
-    "val_gen": "<batchgenerators.dataloading.multi_threaded_augmenter.MultiThreadedAugmenter object at 0x7efe77881f90>",
+    "val_gen": "<batchgenerators.dataloading.multi_threaded_augmenter.MultiThreadedAugmenter object at 0x7efb50941f90>",
     "was_initialized": "True",
     "weight_decay": "3e-05"
 }
\ No newline at end of file
diff --git a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_best.model b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_best.model
index 12edbd5b065b5ecf7b66a36b3d59b05dbf78187e..2be27dcb710c3854eb67ac0ac0fbb0b0fefc5578 100644
--- a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_best.model
+++ b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_best.model
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71b300c6f03bbb40200c5396976042168c7af5755eab82e1511dc1549c51fbf6
-size 742451083
+oid sha256:cd4ae6c473961f9685d13a6fed5a5906d0df7e2d2956e77a88d48eff7a24ea61
+size 742474955
diff --git a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model
index 80044a0c6e5cb74b6da7b0670aa0fd9ca95f6d08..0cb418523f050b2425b58a2d001b2905698a34f4 100644
--- a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model
+++ b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c83313b4056715d82d4ca00c4b259041c229e7e527ca096d60f58b64c67247b8
-size 742452729
+oid sha256:7aaa3e88cb534cdfa25b193a71a4db5e503fd66e3f8c713455a56bb47be33041
+size 742479673
diff --git a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png
index ee7b5ce88a336085767ab9e140f84a92ed71f356..58b3c5b1e8e456473779589d04517a2784a96b9f 100644
--- a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png
+++ b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c26703f1241db000cad13c59ca4d752fceb74387c195a234be2d7fb028987709
-size 602181
+oid sha256:8af23ca9066b61b525ef239a8d824318039dee4f4a0ba0fecdefa10bdfc3749f
+size 759873
diff --git a/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_11_8_06_34_55.txt b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_11_8_06_34_55.txt
new file mode 100644
index 0000000000000000000000000000000000000000..de0ac0ea42bce82550e5184f08b7f3b92411ed6c
--- /dev/null
+++ b/MedNeXt_Large/fold_0_CT/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_11_8_06_34_55.txt
@@ -0,0 +1,1994 @@
+Starting... 
+2026-11-08 06:34:55.597209: Creating new 5-fold cross-validation split... 
+2026-11-08 06:34:55.598096: Desired fold for training: 0 
+2026-11-08 06:34:55.598166: This split has 20 training and 5 validation cases. 
+2026-11-08 06:34:56.249124: TRAINING KEYS:
+ odict_keys([np.str_('topcow_ct_001'), np.str_('topcow_ct_002'), np.str_('topcow_ct_003'), np.str_('topcow_ct_004'), np.str_('topcow_ct_005'), np.str_('topcow_ct_006'), np.str_('topcow_ct_007'), np.str_('topcow_ct_008'), np.str_('topcow_ct_011'), np.str_('topcow_ct_012'), np.str_('topcow_ct_013'), np.str_('topcow_ct_014'), np.str_('topcow_ct_016'), np.str_('topcow_ct_017'), np.str_('topcow_ct_020'), np.str_('topcow_ct_021'), np.str_('topcow_ct_023'), np.str_('topcow_ct_025'), np.str_('topcow_ct_026'), np.str_('topcow_ct_027')]) 
+2026-11-08 06:34:56.249498: VALIDATION KEYS:
+ odict_keys([np.str_('topcow_ct_010'), np.str_('topcow_ct_015'), np.str_('topcow_ct_018'), np.str_('topcow_ct_022'), np.str_('topcow_ct_024')]) 
+2026-11-08 06:34:57.935640: loading checkpoint /marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_L_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model train= True 
+2026-11-08 06:34:58.541518: lr: 0.000864 
+2026-11-08 06:35:02.047134: Unable to plot network architecture: 
+2026-11-08 06:35:02.050959: No module named 'hiddenlayer' 
+2026-11-08 06:35:02.051960: 
+printing the network instead:
+ 
+2026-11-08 06:35:02.052563: MedNeXt(
+  (stem): Conv3d(1, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+  (enc_block_0): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=32)
+      (norm): GroupNorm(32, 32, eps=1e-05, affine=True)
+      (conv2): Conv3d(32, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(96, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=32)
+      (norm): GroupNorm(32, 32, eps=1e-05, affine=True)
+      (conv2): Conv3d(32, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(96, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=32)
+      (norm): GroupNorm(32, 32, eps=1e-05, affine=True)
+      (conv2): Conv3d(32, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(96, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (down_0): MedNeXtDownBlock(
+    (conv1): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), groups=32)
+    (norm): GroupNorm(32, 32, eps=1e-05, affine=True)
+    (conv2): Conv3d(32, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (act): GELU(approximate='none')
+    (conv3): Conv3d(128, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (res_conv): Conv3d(32, 64, kernel_size=(1, 1, 1), stride=(2, 2, 2))
+  )
+  (enc_block_1): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=64)
+      (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+      (conv2): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=64)
+      (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+      (conv2): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=64)
+      (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+      (conv2): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (3): MedNeXtBlock(
+      (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=64)
+      (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+      (conv2): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (down_1): MedNeXtDownBlock(
+    (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), groups=64)
+    (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+    (conv2): Conv3d(64, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (act): GELU(approximate='none')
+    (conv3): Conv3d(512, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (res_conv): Conv3d(64, 128, kernel_size=(1, 1, 1), stride=(2, 2, 2))
+  )
+  (enc_block_2): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (3): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (4): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (5): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (6): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (7): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (down_2): MedNeXtDownBlock(
+    (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), groups=128)
+    (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+    (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (act): GELU(approximate='none')
+    (conv3): Conv3d(1024, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (res_conv): Conv3d(128, 256, kernel_size=(1, 1, 1), stride=(2, 2, 2))
+  )
+  (enc_block_3): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (3): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (4): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (5): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (6): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (7): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (down_3): MedNeXtDownBlock(
+    (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), groups=256)
+    (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+    (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (act): GELU(approximate='none')
+    (conv3): Conv3d(2048, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (res_conv): Conv3d(256, 512, kernel_size=(1, 1, 1), stride=(2, 2, 2))
+  )
+  (bottleneck): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(512, 512, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=512)
+      (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+      (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(4096, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(512, 512, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=512)
+      (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+      (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(4096, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(512, 512, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=512)
+      (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+      (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(4096, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (3): MedNeXtBlock(
+      (conv1): Conv3d(512, 512, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=512)
+      (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+      (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(4096, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (4): MedNeXtBlock(
+      (conv1): Conv3d(512, 512, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=512)
+      (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+      (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(4096, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (5): MedNeXtBlock(
+      (conv1): Conv3d(512, 512, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=512)
+      (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+      (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(4096, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (6): MedNeXtBlock(
+      (conv1): Conv3d(512, 512, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=512)
+      (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+      (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(4096, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (7): MedNeXtBlock(
+      (conv1): Conv3d(512, 512, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=512)
+      (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+      (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(4096, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (up_3): MedNeXtUpBlock(
+    (conv1): ConvTranspose3d(512, 512, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), groups=512)
+    (norm): GroupNorm(512, 512, eps=1e-05, affine=True)
+    (conv2): Conv3d(512, 4096, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (act): GELU(approximate='none')
+    (conv3): Conv3d(4096, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (res_conv): ConvTranspose3d(512, 256, kernel_size=(1, 1, 1), stride=(2, 2, 2))
+  )
+  (dec_block_3): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (3): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (4): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (5): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (6): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (7): MedNeXtBlock(
+      (conv1): Conv3d(256, 256, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=256)
+      (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+      (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(2048, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (up_2): MedNeXtUpBlock(
+    (conv1): ConvTranspose3d(256, 256, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), groups=256)
+    (norm): GroupNorm(256, 256, eps=1e-05, affine=True)
+    (conv2): Conv3d(256, 2048, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (act): GELU(approximate='none')
+    (conv3): Conv3d(2048, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (res_conv): ConvTranspose3d(256, 128, kernel_size=(1, 1, 1), stride=(2, 2, 2))
+  )
+  (dec_block_2): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (3): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (4): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (5): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (6): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (7): MedNeXtBlock(
+      (conv1): Conv3d(128, 128, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=128)
+      (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+      (conv2): Conv3d(128, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(1024, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (up_1): MedNeXtUpBlock(
+    (conv1): ConvTranspose3d(128, 128, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), groups=128)
+    (norm): GroupNorm(128, 128, eps=1e-05, affine=True)
+    (conv2): Conv3d(128, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (act): GELU(approximate='none')
+    (conv3): Conv3d(512, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (res_conv): ConvTranspose3d(128, 64, kernel_size=(1, 1, 1), stride=(2, 2, 2))
+  )
+  (dec_block_1): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=64)
+      (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+      (conv2): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=64)
+      (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+      (conv2): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=64)
+      (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+      (conv2): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (3): MedNeXtBlock(
+      (conv1): Conv3d(64, 64, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=64)
+      (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+      (conv2): Conv3d(64, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(256, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (up_0): MedNeXtUpBlock(
+    (conv1): ConvTranspose3d(64, 64, kernel_size=(3, 3, 3), stride=(2, 2, 2), padding=(1, 1, 1), groups=64)
+    (norm): GroupNorm(64, 64, eps=1e-05, affine=True)
+    (conv2): Conv3d(64, 192, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (act): GELU(approximate='none')
+    (conv3): Conv3d(192, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    (res_conv): ConvTranspose3d(64, 32, kernel_size=(1, 1, 1), stride=(2, 2, 2))
+  )
+  (dec_block_0): Sequential(
+    (0): MedNeXtBlock(
+      (conv1): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=32)
+      (norm): GroupNorm(32, 32, eps=1e-05, affine=True)
+      (conv2): Conv3d(32, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(96, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (1): MedNeXtBlock(
+      (conv1): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=32)
+      (norm): GroupNorm(32, 32, eps=1e-05, affine=True)
+      (conv2): Conv3d(32, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(96, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+    (2): MedNeXtBlock(
+      (conv1): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1), groups=32)
+      (norm): GroupNorm(32, 32, eps=1e-05, affine=True)
+      (conv2): Conv3d(32, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+      (act): GELU(approximate='none')
+      (conv3): Conv3d(96, 32, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+    )
+  )
+  (out_0): OutBlock(
+    (conv_out): ConvTranspose3d(32, 37, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+  )
+  (out_1): OutBlock(
+    (conv_out): ConvTranspose3d(64, 37, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+  )
+  (out_2): OutBlock(
+    (conv_out): ConvTranspose3d(128, 37, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+  )
+  (out_3): OutBlock(
+    (conv_out): ConvTranspose3d(256, 37, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+  )
+  (out_4): OutBlock(
+    (conv_out): ConvTranspose3d(512, 37, kernel_size=(1, 1, 1), stride=(1, 1, 1))
+  )
+) 
+2026-11-08 06:35:02.081839: 
+ 
+2026-11-08 06:35:02.082298: 
+epoch:  150 
+2026-11-08 06:39:14.092094: train loss : -0.7036 
+2026-11-08 06:39:22.251725: validation loss: -0.4477 
+2026-11-08 06:39:22.252942: Average global foreground Dice: [np.float32(0.8982), np.float32(0.4956), np.float32(0.4466), np.float32(0.518), np.float32(0.3817), np.float32(0.5771), np.float32(0.4682), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4171), np.float32(0.4011), np.float32(0.4782), np.float32(0.4313), np.float32(0.0), np.float32(0.0), np.float32(0.3637), np.float32(0.2902), np.float32(0.4008), np.float32(0.4269), np.float32(0.3605), np.float32(0.3387), np.float32(0.6082), np.float32(0.5889), np.float32(0.226), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4665), np.float32(0.0), np.float32(0.0), np.float32(0.2964), np.float32(0.2464), np.float32(0.5461), np.float32(0.5772)] 
+2026-11-08 06:39:22.253397: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 06:39:22.638951: lr: 0.000863 
+2026-11-08 06:39:22.639203: This epoch took 260.556458 s
+ 
+2026-11-08 06:39:22.639416: 
+epoch:  151 
+2026-11-08 06:43:41.540851: train loss : -0.6953 
+2026-11-08 06:43:49.639580: validation loss: -0.5453 
+2026-11-08 06:43:49.641012: Average global foreground Dice: [np.float32(0.8989), np.float32(0.5992), np.float32(0.6531), np.float32(0.7134), np.float32(0.6556), np.float32(0.7477), np.float32(0.6638), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5268), np.float32(0.5853), np.float32(0.5491), np.float32(0.5161), np.float32(0.0), np.float32(0.0), np.float32(0.6489), np.float32(0.4656), np.float32(0.6326), np.float32(0.5313), np.float32(0.4645), np.float32(0.4819), np.float32(0.8074), np.float32(0.7904), np.float32(0.3829), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6626), np.float32(0.0), np.float32(0.0), np.float32(0.3286), np.float32(0.4183), np.float32(0.7604), np.float32(0.7873)] 
+2026-11-08 06:43:49.641330: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 06:43:49.948251: lr: 0.000862 
+2026-11-08 06:43:49.948601: This epoch took 267.309072 s
+ 
+2026-11-08 06:43:49.948708: 
+epoch:  152 
+2026-11-08 06:47:33.976600: train loss : -0.7160 
+2026-11-08 06:47:42.082564: validation loss: -0.5483 
+2026-11-08 06:47:42.083726: Average global foreground Dice: [np.float32(0.9159), np.float32(0.6239), np.float32(0.7012), np.float32(0.7498), np.float32(0.647), np.float32(0.8025), np.float32(0.6739), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5235), np.float32(0.6349), np.float32(0.5601), np.float32(0.4965), np.float32(0.0), np.float32(0.0), np.float32(0.6718), np.float32(0.5486), np.float32(0.6665), np.float32(0.6132), np.float32(0.5043), np.float32(0.5299), np.float32(0.7368), np.float32(0.7284), np.float32(0.4983), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6677), np.float32(0.0), np.float32(0.0), np.float32(0.4457), np.float32(0.6141), np.float32(0.7731), np.float32(0.7898)] 
+2026-11-08 06:47:42.084113: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 06:47:42.426038: lr: 0.000861 
+2026-11-08 06:47:42.426290: This epoch took 232.477508 s
+ 
+2026-11-08 06:47:42.426373: 
+epoch:  153 
+2026-11-08 06:51:47.199722: train loss : -0.7220 
+2026-11-08 06:51:55.288226: validation loss: -0.5280 
+2026-11-08 06:51:55.289269: Average global foreground Dice: [np.float32(0.9018), np.float32(0.5368), np.float32(0.6221), np.float32(0.6716), np.float32(0.5427), np.float32(0.706), np.float32(0.6188), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5359), np.float32(0.5711), np.float32(0.5175), np.float32(0.4232), np.float32(0.0), np.float32(0.0), np.float32(0.5543), np.float32(0.4283), np.float32(0.568), np.float32(0.498), np.float32(0.4816), np.float32(0.4394), np.float32(0.7687), np.float32(0.7486), np.float32(0.389), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6636), np.float32(0.0), np.float32(0.0), np.float32(0.4547), np.float32(0.5003), np.float32(0.7105), np.float32(0.7343)] 
+2026-11-08 06:51:55.289428: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 06:51:55.601287: lr: 0.00086 
+2026-11-08 06:51:55.601562: This epoch took 253.175131 s
+ 
+2026-11-08 06:51:55.601803: 
+epoch:  154 
+2026-11-08 06:56:01.745496: train loss : -0.6994 
+2026-11-08 06:56:09.868502: validation loss: -0.5048 
+2026-11-08 06:56:09.869454: Average global foreground Dice: [np.float32(0.8895), np.float32(0.5287), np.float32(0.6179), np.float32(0.6512), np.float32(0.5759), np.float32(0.7327), np.float32(0.5358), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4858), np.float32(0.4849), np.float32(0.4633), np.float32(0.3874), np.float32(0.0), np.float32(0.0), np.float32(0.6321), np.float32(0.5035), np.float32(0.5411), np.float32(0.5456), np.float32(0.4534), np.float32(0.5577), np.float32(0.7018), np.float32(0.664), np.float32(0.4947), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5613), np.float32(0.0), np.float32(0.0), np.float32(0.4328), np.float32(0.5401), np.float32(0.7228), np.float32(0.7285)] 
+2026-11-08 06:56:09.869627: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 06:56:10.173309: lr: 0.000859 
+2026-11-08 06:56:10.173613: This epoch took 254.571723 s
+ 
+2026-11-08 06:56:10.173790: 
+epoch:  155 
+2026-11-08 07:00:15.523197: train loss : -0.7082 
+2026-11-08 07:00:23.679559: validation loss: -0.5267 
+2026-11-08 07:00:23.680502: Average global foreground Dice: [np.float32(0.9082), np.float32(0.5804), np.float32(0.6009), np.float32(0.6401), np.float32(0.5856), np.float32(0.7064), np.float32(0.5841), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5206), np.float32(0.5708), np.float32(0.5902), np.float32(0.5364), np.float32(0.0), np.float32(0.0), np.float32(0.5285), np.float32(0.4682), np.float32(0.5436), np.float32(0.5067), np.float32(0.4588), np.float32(0.4609), np.float32(0.7832), np.float32(0.7168), np.float32(0.3289), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6398), np.float32(0.0), np.float32(0.0), np.float32(0.3847), np.float32(0.4774), np.float32(0.6933), np.float32(0.6988)] 
+2026-11-08 07:00:23.680971: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:00:24.039869: lr: 0.000858 
+2026-11-08 07:00:24.040113: This epoch took 253.866249 s
+ 
+2026-11-08 07:00:24.040205: 
+epoch:  156 
+2026-11-08 07:04:23.306114: train loss : -0.7107 
+2026-11-08 07:04:31.408556: validation loss: -0.4551 
+2026-11-08 07:04:31.409545: Average global foreground Dice: [np.float32(0.9077), np.float32(0.4716), np.float32(0.4772), np.float32(0.5116), np.float32(0.4229), np.float32(0.4801), np.float32(0.362), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3709), np.float32(0.4009), np.float32(0.4699), np.float32(0.2904), np.float32(0.0), np.float32(0.0), np.float32(0.4556), np.float32(0.3839), np.float32(0.4295), np.float32(0.492), np.float32(0.426), np.float32(0.3436), np.float32(0.7008), np.float32(0.6726), np.float32(0.3233), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6455), np.float32(0.0), np.float32(0.0), np.float32(0.2936), np.float32(0.3626), np.float32(0.5856), np.float32(0.6129)] 
+2026-11-08 07:04:31.409773: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:04:31.774192: lr: 0.000858 
+2026-11-08 07:04:31.774449: This epoch took 247.733557 s
+ 
+2026-11-08 07:04:31.774537: 
+epoch:  157 
+2026-11-08 07:08:35.290781: train loss : -0.7198 
+2026-11-08 07:08:43.504027: validation loss: -0.5234 
+2026-11-08 07:08:43.505086: Average global foreground Dice: [np.float32(0.8988), np.float32(0.5496), np.float32(0.6422), np.float32(0.6274), np.float32(0.538), np.float32(0.6881), np.float32(0.56), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4545), np.float32(0.5803), np.float32(0.5291), np.float32(0.4905), np.float32(0.0), np.float32(0.0), np.float32(0.6189), np.float32(0.4938), np.float32(0.6174), np.float32(0.5465), np.float32(0.465), np.float32(0.5566), np.float32(0.78), np.float32(0.7314), np.float32(0.4851), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5761), np.float32(0.0), np.float32(0.0), np.float32(0.4387), np.float32(0.549), np.float32(0.6896), np.float32(0.7282)] 
+2026-11-08 07:08:43.505415: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:08:43.811817: lr: 0.000857 
+2026-11-08 07:08:43.812056: This epoch took 252.037416 s
+ 
+2026-11-08 07:08:43.812130: 
+epoch:  158 
+2026-11-08 07:13:06.772872: train loss : -0.7141 
+2026-11-08 07:13:14.866557: validation loss: -0.5589 
+2026-11-08 07:13:14.867498: Average global foreground Dice: [np.float32(0.9218), np.float32(0.5891), np.float32(0.7205), np.float32(0.7503), np.float32(0.6655), np.float32(0.7998), np.float32(0.6971), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5703), np.float32(0.6658), np.float32(0.5128), np.float32(0.0699), np.float32(0.0), np.float32(0.0), np.float32(0.7292), np.float32(0.5387), np.float32(0.7148), np.float32(0.6391), np.float32(0.5089), np.float32(0.5509), np.float32(0.7934), np.float32(0.8027), np.float32(0.5623), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6951), np.float32(0.0), np.float32(0.0), np.float32(0.5063), np.float32(0.6268), np.float32(0.7737), np.float32(0.8068)] 
+2026-11-08 07:13:14.867656: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:13:15.177427: lr: 0.000856 
+2026-11-08 07:13:15.177637: This epoch took 271.365453 s
+ 
+2026-11-08 07:13:15.177710: 
+epoch:  159 
+2026-11-08 07:17:14.812462: train loss : -0.7191 
+2026-11-08 07:17:22.901579: validation loss: -0.5480 
+2026-11-08 07:17:22.902714: Average global foreground Dice: [np.float32(0.9232), np.float32(0.6017), np.float32(0.7079), np.float32(0.7423), np.float32(0.6347), np.float32(0.7808), np.float32(0.6664), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5697), np.float32(0.64), np.float32(0.5912), np.float32(0.4697), np.float32(0.0), np.float32(0.0), np.float32(0.7111), np.float32(0.5619), np.float32(0.703), np.float32(0.6263), np.float32(0.4474), np.float32(0.5082), np.float32(0.7369), np.float32(0.7344), np.float32(0.4881), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.609), np.float32(0.0), np.float32(0.0), np.float32(0.4853), np.float32(0.5683), np.float32(0.7549), np.float32(0.7795)] 
+2026-11-08 07:17:22.903155: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:17:23.214660: lr: 0.000855 
+2026-11-08 07:17:23.214898: This epoch took 248.037135 s
+ 
+2026-11-08 07:17:23.214980: 
+epoch:  160 
+2026-11-08 07:21:33.462729: train loss : -0.7282 
+2026-11-08 07:21:41.590191: validation loss: -0.3759 
+2026-11-08 07:21:41.591160: Average global foreground Dice: [np.float32(0.903), np.float32(0.3859), np.float32(0.3638), np.float32(0.3485), np.float32(0.3565), np.float32(0.3826), np.float32(0.3263), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2465), np.float32(0.3571), np.float32(0.3952), np.float32(0.3287), np.float32(0.0), np.float32(0.0), np.float32(0.3221), np.float32(0.2813), np.float32(0.3072), np.float32(0.337), np.float32(0.2473), np.float32(0.2473), np.float32(0.5578), np.float32(0.4946), np.float32(0.2701), np.float32(0.2135), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3383), np.float32(0.0), np.float32(0.0), np.float32(0.3363), np.float32(0.2701), np.float32(0.394), np.float32(0.4172)] 
+2026-11-08 07:21:41.591542: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:21:41.901997: lr: 0.000854 
+2026-11-08 07:21:41.902439: This epoch took 258.687407 s
+ 
+2026-11-08 07:21:41.902797: 
+epoch:  161 
+2026-11-08 07:25:42.282009: train loss : -0.7387 
+2026-11-08 07:25:50.384531: validation loss: -0.3894 
+2026-11-08 07:25:50.385467: Average global foreground Dice: [np.float32(0.8986), np.float32(0.3757), np.float32(0.3668), np.float32(0.3444), np.float32(0.366), np.float32(0.4047), np.float32(0.3674), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.2419), np.float32(0.3504), np.float32(0.4067), np.float32(0.2755), np.float32(0.0), np.float32(0.0), np.float32(0.3716), np.float32(0.2886), np.float32(0.3618), np.float32(0.3466), np.float32(0.278), np.float32(0.2674), np.float32(0.5807), np.float32(0.5389), np.float32(0.2715), np.float32(0.2397), np.float32(0.0), np.float32(0.0), np.float32(0.0253), np.float32(0.494), np.float32(0.0), np.float32(0.0), np.float32(0.3222), np.float32(0.3454), np.float32(0.4136), np.float32(0.4485)] 
+2026-11-08 07:25:50.385762: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:25:50.838963: lr: 0.000853 
+2026-11-08 07:25:50.839304: This epoch took 248.936274 s
+ 
+2026-11-08 07:25:50.839397: 
+epoch:  162 
+2026-11-08 07:30:09.929976: train loss : -0.7284 
+2026-11-08 07:30:18.075415: validation loss: -0.2589 
+2026-11-08 07:30:18.076794: Average global foreground Dice: [np.float32(0.9029), np.float32(0.1739), np.float32(0.1436), np.float32(0.2018), np.float32(0.092), np.float32(0.1864), np.float32(0.1536), np.float32(0.0), np.float32(0.0), np.float32(0.4075), np.float32(0.1671), np.float32(0.1818), np.float32(0.249), np.float32(0.1876), np.float32(0.0), np.float32(0.0), np.float32(0.0957), np.float32(0.065), np.float32(0.1636), np.float32(0.1439), np.float32(0.1859), np.float32(0.1033), np.float32(0.2561), np.float32(0.1761), np.float32(0.0701), np.float32(0.0459), np.float32(0.0), np.float32(0.0), np.float32(0.0007), np.float32(0.0967), np.float32(0.0), np.float32(0.0), np.float32(0.0937), np.float32(0.1845), np.float32(0.1765), np.float32(0.2048)] 
+2026-11-08 07:30:18.077437: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:30:18.418729: lr: 0.000852 
+2026-11-08 07:30:18.419173: This epoch took 267.579704 s
+ 
+2026-11-08 07:30:18.419488: 
+epoch:  163 
+2026-11-08 07:34:29.963385: train loss : -0.7078 
+2026-11-08 07:34:38.109675: validation loss: -0.5029 
+2026-11-08 07:34:38.110633: Average global foreground Dice: [np.float32(0.9066), np.float32(0.5227), np.float32(0.6088), np.float32(0.6146), np.float32(0.5149), np.float32(0.6612), np.float32(0.554), np.float32(0.0), np.float32(0.0), np.float32(0.3881), np.float32(0.4238), np.float32(0.5285), np.float32(0.5179), np.float32(0.4144), np.float32(0.0), np.float32(0.0), np.float32(0.4956), np.float32(0.3783), np.float32(0.5255), np.float32(0.4625), np.float32(0.4874), np.float32(0.4872), np.float32(0.7384), np.float32(0.7109), np.float32(0.3771), np.float32(0.2049), np.float32(0.0), np.float32(0.0), np.float32(0.1203), np.float32(0.5473), np.float32(0.0), np.float32(0.0), np.float32(0.4414), np.float32(0.3984), np.float32(0.6506), np.float32(0.662)] 
+2026-11-08 07:34:38.110824: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:34:38.427434: lr: 0.000851 
+2026-11-08 07:34:38.427808: This epoch took 260.008081 s
+ 
+2026-11-08 07:34:38.428090: 
+epoch:  164 
+2026-11-08 07:38:55.836800: train loss : -0.7234 
+2026-11-08 07:39:03.925744: validation loss: -0.5163 
+2026-11-08 07:39:03.927048: Average global foreground Dice: [np.float32(0.8858), np.float32(0.5326), np.float32(0.6505), np.float32(0.6502), np.float32(0.539), np.float32(0.7393), np.float32(0.6146), np.float32(0.0), np.float32(0.0), np.float32(0.4883), np.float32(0.4826), np.float32(0.6197), np.float32(0.5079), np.float32(0.297), np.float32(0.0), np.float32(0.0), np.float32(0.6498), np.float32(0.4827), np.float32(0.651), np.float32(0.5634), np.float32(0.4704), np.float32(0.4859), np.float32(0.7104), np.float32(0.6525), np.float32(0.5218), np.float32(0.3283), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5824), np.float32(0.0), np.float32(0.0), np.float32(0.4064), np.float32(0.5336), np.float32(0.6853), np.float32(0.7156)] 
+2026-11-08 07:39:03.927563: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:39:04.238304: lr: 0.00085 
+2026-11-08 07:39:04.238597: This epoch took 265.810377 s
+ 
+2026-11-08 07:39:04.238798: 
+epoch:  165 
+2026-11-08 07:43:16.343533: train loss : -0.7389 
+2026-11-08 07:43:24.430272: validation loss: -0.5916 
+2026-11-08 07:43:24.431398: Average global foreground Dice: [np.float32(0.923), np.float32(0.634), np.float32(0.7283), np.float32(0.7858), np.float32(0.7063), np.float32(0.8434), np.float32(0.735), np.float32(0.0), np.float32(0.0), np.float32(0.3626), np.float32(0.5128), np.float32(0.6882), np.float32(0.6262), np.float32(0.4849), np.float32(0.0), np.float32(0.0), np.float32(0.7625), np.float32(0.5896), np.float32(0.7293), np.float32(0.6492), np.float32(0.476), np.float32(0.631), np.float32(0.7929), np.float32(0.7733), np.float32(0.5457), np.float32(0.2575), np.float32(0.0), np.float32(0.0), np.float32(0.1916), np.float32(0.6833), np.float32(0.0), np.float32(0.0), np.float32(0.629), np.float32(0.6651), np.float32(0.8094), np.float32(0.8258)] 
+2026-11-08 07:43:24.431830: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:43:24.751519: lr: 0.000849 
+2026-11-08 07:43:24.751735: This epoch took 260.512856 s
+ 
+2026-11-08 07:43:24.751807: 
+epoch:  166 
+2026-11-08 07:47:58.403579: train loss : -0.7416 
+2026-11-08 07:48:06.528991: validation loss: -0.5897 
+2026-11-08 07:48:06.530029: Average global foreground Dice: [np.float32(0.8999), np.float32(0.6173), np.float32(0.7256), np.float32(0.7752), np.float32(0.7062), np.float32(0.8273), np.float32(0.711), np.float32(0.0), np.float32(0.0), np.float32(0.3474), np.float32(0.5334), np.float32(0.6653), np.float32(0.5757), np.float32(0.4618), np.float32(0.0), np.float32(0.0), np.float32(0.7533), np.float32(0.5611), np.float32(0.7234), np.float32(0.5656), np.float32(0.5122), np.float32(0.5641), np.float32(0.8556), np.float32(0.816), np.float32(0.5084), np.float32(0.3791), np.float32(0.0), np.float32(0.0), np.float32(0.3597), np.float32(0.7337), np.float32(0.0), np.float32(0.0), np.float32(0.5613), np.float32(0.6616), np.float32(0.7988), np.float32(0.8249)] 
+2026-11-08 07:48:06.530442: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:48:06.843990: lr: 0.000848 
+2026-11-08 07:48:06.844400: This epoch took 282.092531 s
+ 
+2026-11-08 07:48:06.844651: 
+epoch:  167 
+2026-11-08 07:52:08.580078: train loss : -0.7495 
+2026-11-08 07:52:16.666114: validation loss: -0.5982 
+2026-11-08 07:52:16.667119: Average global foreground Dice: [np.float32(0.9244), np.float32(0.6659), np.float32(0.7397), np.float32(0.8042), np.float32(0.7569), np.float32(0.8437), np.float32(0.7429), np.float32(0.0), np.float32(0.0), np.float32(0.3568), np.float32(0.5224), np.float32(0.6015), np.float32(0.6076), np.float32(0.4814), np.float32(0.0), np.float32(0.0), np.float32(0.7497), np.float32(0.6202), np.float32(0.7163), np.float32(0.6002), np.float32(0.5451), np.float32(0.6188), np.float32(0.8625), np.float32(0.8464), np.float32(0.542), np.float32(0.3377), np.float32(0.0), np.float32(0.0), np.float32(0.2965), np.float32(0.6042), np.float32(0.0), np.float32(0.0), np.float32(0.5899), np.float32(0.6235), np.float32(0.8281), np.float32(0.8469)] 
+2026-11-08 07:52:16.667301: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:52:17.002777: lr: 0.000847 
+2026-11-08 07:52:17.003163: This epoch took 250.158329 s
+ 
+2026-11-08 07:52:17.003348: 
+epoch:  168 
+2026-11-08 07:56:19.550771: train loss : -0.7241 
+2026-11-08 07:56:27.659830: validation loss: -0.4503 
+2026-11-08 07:56:27.660972: Average global foreground Dice: [np.float32(0.8965), np.float32(0.5068), np.float32(0.4848), np.float32(0.4323), np.float32(0.4453), np.float32(0.5073), np.float32(0.4794), np.float32(0.0), np.float32(0.0), np.float32(0.3092), np.float32(0.3096), np.float32(0.3653), np.float32(0.4844), np.float32(0.3981), np.float32(0.0), np.float32(0.0), np.float32(0.4598), np.float32(0.4008), np.float32(0.4544), np.float32(0.4452), np.float32(0.3372), np.float32(0.3017), np.float32(0.6609), np.float32(0.6476), np.float32(0.2801), np.float32(0.3063), np.float32(0.0), np.float32(0.0), np.float32(0.0138), np.float32(0.6104), np.float32(0.0), np.float32(0.0), np.float32(0.32), np.float32(0.5134), np.float32(0.5263), np.float32(0.5779)] 
+2026-11-08 07:56:27.661143: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 07:56:27.968183: lr: 0.000847 
+2026-11-08 07:56:27.968457: This epoch took 250.965032 s
+ 
+2026-11-08 07:56:27.968593: 
+epoch:  169 
+2026-11-08 08:00:02.578750: train loss : -0.7471 
+2026-11-08 08:00:10.680595: validation loss: -0.5943 
+2026-11-08 08:00:10.681685: Average global foreground Dice: [np.float32(0.9274), np.float32(0.6228), np.float32(0.7654), np.float32(0.806), np.float32(0.7535), np.float32(0.8403), np.float32(0.7478), np.float32(0.0), np.float32(0.0), np.float32(0.2998), np.float32(0.5061), np.float32(0.6284), np.float32(0.5784), np.float32(0.4854), np.float32(0.0), np.float32(0.0), np.float32(0.7721), np.float32(0.5797), np.float32(0.7391), np.float32(0.6259), np.float32(0.5016), np.float32(0.5671), np.float32(0.833), np.float32(0.8069), np.float32(0.5064), np.float32(0.4325), np.float32(0.0), np.float32(0.0), np.float32(0.1368), np.float32(0.5337), np.float32(0.0), np.float32(0.0), np.float32(0.5427), np.float32(0.6075), np.float32(0.8387), np.float32(0.8419)] 
+2026-11-08 08:00:10.681875: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:00:10.992233: lr: 0.000846 
+2026-11-08 08:00:11.159306: saving checkpoint... 
+2026-11-08 08:00:11.719981: done, saving took 0.73 seconds 
+2026-11-08 08:00:11.728047: This epoch took 223.759244 s
+ 
+2026-11-08 08:00:11.728339: 
+epoch:  170 
+2026-11-08 08:04:36.175252: train loss : -0.7544 
+2026-11-08 08:04:44.248398: validation loss: -0.5725 
+2026-11-08 08:04:44.249378: Average global foreground Dice: [np.float32(0.9196), np.float32(0.574), np.float32(0.7011), np.float32(0.7554), np.float32(0.6781), np.float32(0.7959), np.float32(0.7015), np.float32(0.0), np.float32(0.0), np.float32(0.5574), np.float32(0.5704), np.float32(0.6029), np.float32(0.5972), np.float32(0.4245), np.float32(0.0), np.float32(0.0), np.float32(0.729), np.float32(0.5536), np.float32(0.7104), np.float32(0.6549), np.float32(0.4019), np.float32(0.5464), np.float32(0.8212), np.float32(0.8036), np.float32(0.5389), np.float32(0.3838), np.float32(0.0), np.float32(0.0), np.float32(0.0431), np.float32(0.5706), np.float32(0.0), np.float32(0.0), np.float32(0.5034), np.float32(0.6002), np.float32(0.7861), np.float32(0.8045)] 
+2026-11-08 08:04:44.249676: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:04:44.555610: lr: 0.000845 
+2026-11-08 08:04:44.800095: saving checkpoint... 
+2026-11-08 08:04:45.354386: done, saving took 0.80 seconds 
+2026-11-08 08:04:45.358911: This epoch took 273.630378 s
+ 
+2026-11-08 08:04:45.359143: 
+epoch:  171 
+2026-11-08 08:08:44.269482: train loss : -0.7289 
+2026-11-08 08:08:52.343063: validation loss: -0.5789 
+2026-11-08 08:08:52.344343: Average global foreground Dice: [np.float32(0.9123), np.float32(0.6132), np.float32(0.7262), np.float32(0.7676), np.float32(0.7202), np.float32(0.8177), np.float32(0.7113), np.float32(0.0), np.float32(0.0), np.float32(0.2646), np.float32(0.4414), np.float32(0.6013), np.float32(0.5886), np.float32(0.4796), np.float32(0.0), np.float32(0.0), np.float32(0.7541), np.float32(0.6025), np.float32(0.7264), np.float32(0.6286), np.float32(0.4795), np.float32(0.5286), np.float32(0.8386), np.float32(0.8197), np.float32(0.5022), np.float32(0.4418), np.float32(0.0), np.float32(0.0), np.float32(0.2272), np.float32(0.6887), np.float32(0.0), np.float32(0.0), np.float32(0.6118), np.float32(0.664), np.float32(0.8161), np.float32(0.8353)] 
+2026-11-08 08:08:52.344543: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:08:52.649900: lr: 0.000844 
+2026-11-08 08:08:52.891826: saving checkpoint... 
+2026-11-08 08:08:54.694449: done, saving took 2.04 seconds 
+2026-11-08 08:08:54.696640: This epoch took 249.337371 s
+ 
+2026-11-08 08:08:54.696775: 
+epoch:  172 
+2026-11-08 08:12:56.481053: train loss : -0.7140 
+2026-11-08 08:13:04.581964: validation loss: -0.3769 
+2026-11-08 08:13:04.583106: Average global foreground Dice: [np.float32(0.9066), np.float32(0.3833), np.float32(0.323), np.float32(0.4258), np.float32(0.4171), np.float32(0.4507), np.float32(0.2985), np.float32(0.0), np.float32(0.0), np.float32(0.3731), np.float32(0.2974), np.float32(0.4269), np.float32(0.2941), np.float32(0.2814), np.float32(0.0), np.float32(0.0), np.float32(0.2963), np.float32(0.2652), np.float32(0.2707), np.float32(0.3542), np.float32(0.3071), np.float32(0.4059), np.float32(0.3641), np.float32(0.2807), np.float32(0.2472), np.float32(0.1469), np.float32(0.0), np.float32(0.0), np.float32(0.0647), np.float32(0.4654), np.float32(0.0), np.float32(0.0), np.float32(0.2755), np.float32(0.3313), np.float32(0.4343), np.float32(0.2657)] 
+2026-11-08 08:13:04.583470: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:13:04.912192: lr: 0.000843 
+2026-11-08 08:13:04.912477: This epoch took 250.215633 s
+ 
+2026-11-08 08:13:04.912730: 
+epoch:  173 
+2026-11-08 08:16:57.676951: train loss : -0.6889 
+2026-11-08 08:17:05.758542: validation loss: -0.4779 
+2026-11-08 08:17:05.759474: Average global foreground Dice: [np.float32(0.9166), np.float32(0.4902), np.float32(0.4893), np.float32(0.5257), np.float32(0.4222), np.float32(0.5894), np.float32(0.4802), np.float32(0.0), np.float32(0.0), np.float32(0.5628), np.float32(0.4314), np.float32(0.4444), np.float32(0.4891), np.float32(0.4243), np.float32(0.0), np.float32(0.0), np.float32(0.3793), np.float32(0.3152), np.float32(0.4051), np.float32(0.4106), np.float32(0.3914), np.float32(0.3108), np.float32(0.6187), np.float32(0.5676), np.float32(0.2437), np.float32(0.0999), np.float32(0.0), np.float32(0.0), np.float32(0.0955), np.float32(0.5931), np.float32(0.0), np.float32(0.0), np.float32(0.3159), np.float32(0.4526), np.float32(0.561), np.float32(0.5643)] 
+2026-11-08 08:17:05.759734: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:17:06.064640: lr: 0.000842 
+2026-11-08 08:17:06.064800: This epoch took 241.151912 s
+ 
+2026-11-08 08:17:06.064862: 
+epoch:  174 
+2026-11-08 08:21:48.581504: train loss : -0.7197 
+2026-11-08 08:21:56.704403: validation loss: -0.6042 
+2026-11-08 08:21:56.705511: Average global foreground Dice: [np.float32(0.9141), np.float32(0.6389), np.float32(0.756), np.float32(0.7578), np.float32(0.7416), np.float32(0.7982), np.float32(0.7599), np.float32(0.0), np.float32(0.0), np.float32(0.378), np.float32(0.512), np.float32(0.5962), np.float32(0.6045), np.float32(0.4512), np.float32(0.0), np.float32(0.0), np.float32(0.7366), np.float32(0.5347), np.float32(0.7232), np.float32(0.6362), np.float32(0.5256), np.float32(0.5937), np.float32(0.875), np.float32(0.8495), np.float32(0.548), np.float32(0.3122), np.float32(0.0), np.float32(0.0), np.float32(0.2686), np.float32(0.5327), np.float32(0.0), np.float32(0.0), np.float32(0.5483), np.float32(0.634), np.float32(0.8576), np.float32(0.8553)] 
+2026-11-08 08:21:56.705923: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:21:57.018651: lr: 0.000841 
+2026-11-08 08:21:57.018900: This epoch took 290.953975 s
+ 
+2026-11-08 08:21:57.019169: 
+epoch:  175 
+2026-11-08 08:26:21.760727: train loss : -0.7377 
+2026-11-08 08:26:29.881428: validation loss: -0.5631 
+2026-11-08 08:26:29.882489: Average global foreground Dice: [np.float32(0.9006), np.float32(0.5622), np.float32(0.6643), np.float32(0.6748), np.float32(0.5889), np.float32(0.7023), np.float32(0.6146), np.float32(0.0), np.float32(0.0), np.float32(0.4521), np.float32(0.5024), np.float32(0.5611), np.float32(0.6043), np.float32(0.5254), np.float32(0.0), np.float32(0.0), np.float32(0.6126), np.float32(0.5766), np.float32(0.5843), np.float32(0.5779), np.float32(0.5906), np.float32(0.5741), np.float32(0.7233), np.float32(0.6956), np.float32(0.4559), np.float32(0.1858), np.float32(0.0), np.float32(0.0), np.float32(0.2326), np.float32(0.5305), np.float32(0.0), np.float32(0.0), np.float32(0.4354), np.float32(0.4471), np.float32(0.7159), np.float32(0.7187)] 
+2026-11-08 08:26:29.882750: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:26:30.204556: lr: 0.00084 
+2026-11-08 08:26:30.205016: This epoch took 273.185724 s
+ 
+2026-11-08 08:26:30.205172: 
+epoch:  176 
+2026-11-08 08:30:52.508360: train loss : -0.7423 
+2026-11-08 08:31:00.635769: validation loss: -0.4971 
+2026-11-08 08:31:00.636724: Average global foreground Dice: [np.float32(0.9166), np.float32(0.5216), np.float32(0.5438), np.float32(0.5524), np.float32(0.4375), np.float32(0.576), np.float32(0.5188), np.float32(0.0), np.float32(0.0), np.float32(0.3564), np.float32(0.4042), np.float32(0.4593), np.float32(0.5549), np.float32(0.4537), np.float32(0.0), np.float32(0.0), np.float32(0.4362), np.float32(0.4281), np.float32(0.4778), np.float32(0.5195), np.float32(0.5392), np.float32(0.4854), np.float32(0.6497), np.float32(0.6035), np.float32(0.2983), np.float32(0.2518), np.float32(0.0), np.float32(0.0), np.float32(0.2631), np.float32(0.5818), np.float32(0.0), np.float32(0.0), np.float32(0.2712), np.float32(0.3174), np.float32(0.5726), np.float32(0.6051)] 
+2026-11-08 08:31:00.636875: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:31:00.955812: lr: 0.000839 
+2026-11-08 08:31:00.956040: This epoch took 270.750666 s
+ 
+2026-11-08 08:31:00.956115: 
+epoch:  177 
+2026-11-08 08:35:32.401975: train loss : -0.7335 
+2026-11-08 08:35:40.534026: validation loss: -0.4939 
+2026-11-08 08:35:40.534955: Average global foreground Dice: [np.float32(0.9164), np.float32(0.4272), np.float32(0.5672), np.float32(0.5683), np.float32(0.5499), np.float32(0.6066), np.float32(0.6147), np.float32(0.0), np.float32(0.0), np.float32(0.3121), np.float32(0.3737), np.float32(0.4953), np.float32(0.5353), np.float32(0.3873), np.float32(0.0), np.float32(0.0), np.float32(0.5213), np.float32(0.4209), np.float32(0.4757), np.float32(0.4588), np.float32(0.4292), np.float32(0.4441), np.float32(0.7636), np.float32(0.7379), np.float32(0.3831), np.float32(0.2453), np.float32(0.0), np.float32(0.0), np.float32(0.3508), np.float32(0.5531), np.float32(0.0), np.float32(0.0), np.float32(0.471), np.float32(0.3782), np.float32(0.6415), np.float32(0.6477)] 
+2026-11-08 08:35:40.535107: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:35:40.850255: lr: 0.000838 
+2026-11-08 08:35:40.850542: This epoch took 279.894374 s
+ 
+2026-11-08 08:35:40.850811: 
+epoch:  178 
+2026-11-08 08:39:36.852829: train loss : -0.7482 
+2026-11-08 08:39:44.939254: validation loss: -0.5644 
+2026-11-08 08:39:44.940404: Average global foreground Dice: [np.float32(0.9244), np.float32(0.5828), np.float32(0.7137), np.float32(0.7512), np.float32(0.7044), np.float32(0.7847), np.float32(0.7062), np.float32(0.0), np.float32(0.0), np.float32(0.3484), np.float32(0.5231), np.float32(0.5914), np.float32(0.55), np.float32(0.456), np.float32(0.0), np.float32(0.0), np.float32(0.7157), np.float32(0.5881), np.float32(0.7026), np.float32(0.6335), np.float32(0.4765), np.float32(0.5453), np.float32(0.8121), np.float32(0.7914), np.float32(0.5199), np.float32(0.3366), np.float32(0.0), np.float32(0.0), np.float32(0.1252), np.float32(0.5259), np.float32(0.0), np.float32(0.0), np.float32(0.5474), np.float32(0.541), np.float32(0.7863), np.float32(0.8078)] 
+2026-11-08 08:39:44.941575: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:39:45.256793: lr: 0.000837 
+2026-11-08 08:39:45.257026: This epoch took 244.406087 s
+ 
+2026-11-08 08:39:45.257104: 
+epoch:  179 
+2026-11-08 08:43:35.061432: train loss : -0.7606 
+2026-11-08 08:43:43.144073: validation loss: -0.5635 
+2026-11-08 08:43:43.145112: Average global foreground Dice: [np.float32(0.8974), np.float32(0.544), np.float32(0.6883), np.float32(0.6765), np.float32(0.6089), np.float32(0.7435), np.float32(0.6586), np.float32(0.0), np.float32(0.0), np.float32(0.4432), np.float32(0.4994), np.float32(0.574), np.float32(0.5758), np.float32(0.5092), np.float32(0.0), np.float32(0.0), np.float32(0.622), np.float32(0.5371), np.float32(0.6222), np.float32(0.5616), np.float32(0.5094), np.float32(0.5429), np.float32(0.7953), np.float32(0.7694), np.float32(0.436), np.float32(0.354), np.float32(0.0), np.float32(0.0), np.float32(0.1532), np.float32(0.6803), np.float32(0.0), np.float32(0.0), np.float32(0.4816), np.float32(0.525), np.float32(0.7503), np.float32(0.7779)] 
+2026-11-08 08:43:43.145746: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:43:43.461851: lr: 0.000836 
+2026-11-08 08:43:43.507468: saving checkpoint... 
+2026-11-08 08:43:44.049109: done, saving took 0.59 seconds 
+2026-11-08 08:43:44.051839: This epoch took 238.794682 s
+ 
+2026-11-08 08:43:44.052029: 
+epoch:  180 
+2026-11-08 08:47:41.232867: train loss : -0.7544 
+2026-11-08 08:47:49.336790: validation loss: -0.5944 
+2026-11-08 08:47:49.337775: Average global foreground Dice: [np.float32(0.8994), np.float32(0.5925), np.float32(0.718), np.float32(0.7641), np.float32(0.6995), np.float32(0.8114), np.float32(0.7187), np.float32(0.0), np.float32(0.0), np.float32(0.5131), np.float32(0.5659), np.float32(0.6422), np.float32(0.5989), np.float32(0.539), np.float32(0.0), np.float32(0.0), np.float32(0.7321), np.float32(0.6044), np.float32(0.7114), np.float32(0.6348), np.float32(0.5684), np.float32(0.5269), np.float32(0.8202), np.float32(0.7888), np.float32(0.4929), np.float32(0.3875), np.float32(0.0), np.float32(0.0), np.float32(0.3706), np.float32(0.7051), np.float32(0.0), np.float32(0.0), np.float32(0.5914), np.float32(0.6222), np.float32(0.8003), np.float32(0.8153)] 
+2026-11-08 08:47:49.338054: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:47:49.651908: lr: 0.000836 
+2026-11-08 08:47:49.679724: saving checkpoint... 
+2026-11-08 08:47:52.193110: done, saving took 2.54 seconds 
+2026-11-08 08:47:52.195759: This epoch took 248.143632 s
+ 
+2026-11-08 08:47:52.195970: 
+epoch:  181 
+2026-11-08 08:51:51.481182: train loss : -0.7610 
+2026-11-08 08:51:59.604853: validation loss: -0.5934 
+2026-11-08 08:51:59.605946: Average global foreground Dice: [np.float32(0.9043), np.float32(0.6138), np.float32(0.7444), np.float32(0.7535), np.float32(0.6926), np.float32(0.8076), np.float32(0.7185), np.float32(0.0), np.float32(0.0), np.float32(0.3996), np.float32(0.5469), np.float32(0.6303), np.float32(0.601), np.float32(0.5198), np.float32(0.0), np.float32(0.0), np.float32(0.7375), np.float32(0.6079), np.float32(0.7287), np.float32(0.6418), np.float32(0.5276), np.float32(0.5129), np.float32(0.7909), np.float32(0.7527), np.float32(0.4558), np.float32(0.371), np.float32(0.0), np.float32(0.0), np.float32(0.3028), np.float32(0.6528), np.float32(0.0), np.float32(0.0), np.float32(0.585), np.float32(0.6114), np.float32(0.7928), np.float32(0.8139)] 
+2026-11-08 08:51:59.606322: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:51:59.921690: lr: 0.000835 
+2026-11-08 08:51:59.948657: saving checkpoint... 
+2026-11-08 08:52:00.496108: done, saving took 0.57 seconds 
+2026-11-08 08:52:00.498399: This epoch took 248.302325 s
+ 
+2026-11-08 08:52:00.498492: 
+epoch:  182 
+2026-11-08 08:55:54.096785: train loss : -0.7551 
+2026-11-08 08:56:02.177687: validation loss: -0.5899 
+2026-11-08 08:56:02.178762: Average global foreground Dice: [np.float32(0.907), np.float32(0.633), np.float32(0.7331), np.float32(0.7739), np.float32(0.7126), np.float32(0.8206), np.float32(0.7218), np.float32(0.0), np.float32(0.0), np.float32(0.4702), np.float32(0.5574), np.float32(0.5789), np.float32(0.6183), np.float32(0.5055), np.float32(0.0), np.float32(0.0), np.float32(0.7544), np.float32(0.5551), np.float32(0.7312), np.float32(0.6235), np.float32(0.4941), np.float32(0.5235), np.float32(0.8285), np.float32(0.808), np.float32(0.492), np.float32(0.4041), np.float32(0.0), np.float32(0.0), np.float32(0.2368), np.float32(0.5339), np.float32(0.0), np.float32(0.0), np.float32(0.6289), np.float32(0.5744), np.float32(0.8145), np.float32(0.828)] 
+2026-11-08 08:56:02.179011: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 08:56:02.512269: lr: 0.000834 
+2026-11-08 08:56:02.537523: saving checkpoint... 
+2026-11-08 08:56:03.082482: done, saving took 0.57 seconds 
+2026-11-08 08:56:03.085006: This epoch took 242.586462 s
+ 
+2026-11-08 08:56:03.085208: 
+epoch:  183 
+2026-11-08 09:00:13.722949: train loss : -0.7421 
+2026-11-08 09:00:21.834566: validation loss: -0.5234 
+2026-11-08 09:00:21.835525: Average global foreground Dice: [np.float32(0.8968), np.float32(0.5791), np.float32(0.5912), np.float32(0.6237), np.float32(0.565), np.float32(0.6756), np.float32(0.5941), np.float32(0.0), np.float32(0.0), np.float32(0.3333), np.float32(0.4393), np.float32(0.4599), np.float32(0.5522), np.float32(0.504), np.float32(0.0), np.float32(0.0), np.float32(0.5576), np.float32(0.5096), np.float32(0.5479), np.float32(0.532), np.float32(0.3863), np.float32(0.4188), np.float32(0.7694), np.float32(0.7379), np.float32(0.3693), np.float32(0.3157), np.float32(0.0), np.float32(0.0), np.float32(0.2096), np.float32(0.5786), np.float32(0.0), np.float32(0.0), np.float32(0.4884), np.float32(0.4729), np.float32(0.6802), np.float32(0.713)] 
+2026-11-08 09:00:21.835694: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:00:22.153759: lr: 0.000833 
+2026-11-08 09:00:22.153984: This epoch took 259.068685 s
+ 
+2026-11-08 09:00:22.154105: 
+epoch:  184 
+2026-11-08 09:04:20.378325: train loss : -0.7272 
+2026-11-08 09:04:28.470909: validation loss: -0.5975 
+2026-11-08 09:04:28.471864: Average global foreground Dice: [np.float32(0.9014), np.float32(0.62), np.float32(0.7623), np.float32(0.8035), np.float32(0.7379), np.float32(0.8458), np.float32(0.7555), np.float32(0.0), np.float32(0.0), np.float32(0.3275), np.float32(0.5384), np.float32(0.6521), np.float32(0.6102), np.float32(0.4625), np.float32(0.0), np.float32(0.0), np.float32(0.7691), np.float32(0.6098), np.float32(0.7481), np.float32(0.6651), np.float32(0.547), np.float32(0.5205), np.float32(0.8445), np.float32(0.8355), np.float32(0.536), np.float32(0.3424), np.float32(0.0), np.float32(0.0), np.float32(0.03), np.float32(0.5647), np.float32(0.0), np.float32(0.0), np.float32(0.5818), np.float32(0.6662), np.float32(0.8321), np.float32(0.8499)] 
+2026-11-08 09:04:28.472102: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:04:28.789430: lr: 0.000832 
+2026-11-08 09:04:28.831724: saving checkpoint... 
+2026-11-08 09:04:29.385328: done, saving took 0.59 seconds 
+2026-11-08 09:04:29.387842: This epoch took 247.233536 s
+ 
+2026-11-08 09:04:29.388063: 
+epoch:  185 
+2026-11-08 09:09:03.498354: train loss : -0.7527 
+2026-11-08 09:09:11.617552: validation loss: -0.1786 
+2026-11-08 09:09:11.618543: Average global foreground Dice: [np.float32(0.8955), np.float32(0.0537), np.float32(0.0108), np.float32(0.104), np.float32(0.0265), np.float32(0.008), np.float32(0.0449), np.float32(0.0), np.float32(0.0), np.float32(0.521), np.float32(0.0594), np.float32(0.0217), np.float32(0.1775), np.float32(0.0344), np.float32(0.0), np.float32(0.0), np.float32(0.0301), np.float32(0.0341), np.float32(0.0433), np.float32(0.0235), np.float32(0.0246), np.float32(0.0394), np.float32(0.052), np.float32(0.0617), np.float32(0.0213), np.float32(0.0019), np.float32(0.0), np.float32(0.0), np.float32(0.0026), np.float32(0.2128), np.float32(0.0), np.float32(0.0), np.float32(0.0079), np.float32(0.0477), np.float32(0.018), np.float32(0.0334)] 
+2026-11-08 09:09:11.618855: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:09:11.937633: lr: 0.000831 
+2026-11-08 09:09:11.937869: This epoch took 282.549698 s
+ 
+2026-11-08 09:09:11.937954: 
+epoch:  186 
+2026-11-08 09:13:16.211576: train loss : -0.7489 
+2026-11-08 09:13:24.357885: validation loss: -0.4265 
+2026-11-08 09:13:24.359194: Average global foreground Dice: [np.float32(0.8588), np.float32(0.4173), np.float32(0.4194), np.float32(0.4203), np.float32(0.3403), np.float32(0.4888), np.float32(0.4209), np.float32(0.0), np.float32(0.0), np.float32(0.4705), np.float32(0.3395), np.float32(0.4216), np.float32(0.4445), np.float32(0.4211), np.float32(0.0), np.float32(0.0), np.float32(0.2871), np.float32(0.2471), np.float32(0.3298), np.float32(0.3407), np.float32(0.369), np.float32(0.2278), np.float32(0.6287), np.float32(0.5819), np.float32(0.2058), np.float32(0.1657), np.float32(0.0), np.float32(0.0), np.float32(0.1165), np.float32(0.5896), np.float32(0.0), np.float32(0.0), np.float32(0.0694), np.float32(0.2824), np.float32(0.4791), np.float32(0.5442)] 
+2026-11-08 09:13:24.359671: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:13:24.704251: lr: 0.00083 
+2026-11-08 09:13:24.704461: This epoch took 252.766454 s
+ 
+2026-11-08 09:13:24.704763: 
+epoch:  187 
+2026-11-08 09:17:39.271345: train loss : -0.7434 
+2026-11-08 09:17:47.401448: validation loss: -0.6121 
+2026-11-08 09:17:47.402646: Average global foreground Dice: [np.float32(0.9004), np.float32(0.622), np.float32(0.7826), np.float32(0.8073), np.float32(0.7615), np.float32(0.8607), np.float32(0.7629), np.float32(0.0), np.float32(0.0), np.float32(0.2554), np.float32(0.5622), np.float32(0.6428), np.float32(0.6323), np.float32(0.5355), np.float32(0.0), np.float32(0.0), np.float32(0.7795), np.float32(0.608), np.float32(0.733), np.float32(0.6174), np.float32(0.4977), np.float32(0.5935), np.float32(0.8904), np.float32(0.8628), np.float32(0.5185), np.float32(0.4168), np.float32(0.0), np.float32(0.0), np.float32(0.2337), np.float32(0.5311), np.float32(0.0), np.float32(0.0), np.float32(0.593), np.float32(0.6733), np.float32(0.8518), np.float32(0.8585)] 
+2026-11-08 09:17:47.403162: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:17:47.740624: lr: 0.000829 
+2026-11-08 09:17:47.740883: This epoch took 263.035740 s
+ 
+2026-11-08 09:17:47.740984: 
+epoch:  188 
+2026-11-08 09:21:52.302044: train loss : -0.7467 
+2026-11-08 09:22:00.399971: validation loss: -0.5043 
+2026-11-08 09:22:00.400973: Average global foreground Dice: [np.float32(0.9109), np.float32(0.5276), np.float32(0.5307), np.float32(0.5548), np.float32(0.4588), np.float32(0.6294), np.float32(0.5087), np.float32(0.0), np.float32(0.0), np.float32(0.3627), np.float32(0.4271), np.float32(0.4866), np.float32(0.5568), np.float32(0.5111), np.float32(0.0), np.float32(0.0), np.float32(0.4183), np.float32(0.3705), np.float32(0.4431), np.float32(0.4418), np.float32(0.5074), np.float32(0.2845), np.float32(0.6427), np.float32(0.5937), np.float32(0.2343), np.float32(0.229), np.float32(0.0), np.float32(0.0), np.float32(0.3199), np.float32(0.6286), np.float32(0.0), np.float32(0.0), np.float32(0.3966), np.float32(0.5085), np.float32(0.596), np.float32(0.6312)] 
+2026-11-08 09:22:00.401268: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:22:00.718448: lr: 0.000828 
+2026-11-08 09:22:00.718758: This epoch took 252.977695 s
+ 
+2026-11-08 09:22:00.718963: 
+epoch:  189 
+2026-11-08 09:26:20.606042: train loss : -0.7617 
+2026-11-08 09:26:28.739012: validation loss: -0.4431 
+2026-11-08 09:26:28.740182: Average global foreground Dice: [np.float32(0.8959), np.float32(0.4274), np.float32(0.4336), np.float32(0.4364), np.float32(0.4058), np.float32(0.4981), np.float32(0.418), np.float32(0.0), np.float32(0.0), np.float32(0.4221), np.float32(0.3194), np.float32(0.3489), np.float32(0.5021), np.float32(0.4499), np.float32(0.0), np.float32(0.0), np.float32(0.3742), np.float32(0.2961), np.float32(0.3501), np.float32(0.3626), np.float32(0.3588), np.float32(0.2091), np.float32(0.6838), np.float32(0.6257), np.float32(0.2371), np.float32(0.1968), np.float32(0.0), np.float32(0.0), np.float32(0.2258), np.float32(0.5818), np.float32(0.0), np.float32(0.0), np.float32(0.3865), np.float32(0.3989), np.float32(0.4875), np.float32(0.5185)] 
+2026-11-08 09:26:28.740692: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:26:29.050446: lr: 0.000827 
+2026-11-08 09:26:29.050854: This epoch took 268.331820 s
+ 
+2026-11-08 09:26:29.051083: 
+epoch:  190 
+2026-11-08 09:30:32.081775: train loss : -0.7621 
+2026-11-08 09:30:40.206843: validation loss: -0.5713 
+2026-11-08 09:30:40.207823: Average global foreground Dice: [np.float32(0.9166), np.float32(0.5986), np.float32(0.701), np.float32(0.7226), np.float32(0.6174), np.float32(0.7736), np.float32(0.6596), np.float32(0.1322), np.float32(0.0), np.float32(0.5697), np.float32(0.5656), np.float32(0.6509), np.float32(0.6032), np.float32(0.4805), np.float32(0.0), np.float32(0.0), np.float32(0.7001), np.float32(0.5558), np.float32(0.6862), np.float32(0.6279), np.float32(0.4804), np.float32(0.5899), np.float32(0.7195), np.float32(0.7145), np.float32(0.5082), np.float32(0.315), np.float32(0.0), np.float32(0.0), np.float32(0.4566), np.float32(0.642), np.float32(0.0), np.float32(0.0), np.float32(0.4957), np.float32(0.6016), np.float32(0.7502), np.float32(0.7792)] 
+2026-11-08 09:30:40.208092: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:30:40.528581: lr: 0.000826 
+2026-11-08 09:30:40.528855: This epoch took 251.477691 s
+ 
+2026-11-08 09:30:40.528945: 
+epoch:  191 
+2026-11-08 09:34:27.781224: train loss : -0.7586 
+2026-11-08 09:34:35.892637: validation loss: -0.5935 
+2026-11-08 09:34:35.893857: Average global foreground Dice: [np.float32(0.9066), np.float32(0.6009), np.float32(0.7548), np.float32(0.7756), np.float32(0.7094), np.float32(0.8361), np.float32(0.7487), np.float32(0.1581), np.float32(0.0), np.float32(0.3787), np.float32(0.5253), np.float32(0.5992), np.float32(0.5961), np.float32(0.4472), np.float32(0.0), np.float32(0.0), np.float32(0.7486), np.float32(0.5754), np.float32(0.7323), np.float32(0.6389), np.float32(0.4745), np.float32(0.6443), np.float32(0.8267), np.float32(0.8039), np.float32(0.5179), np.float32(0.318), np.float32(0.0), np.float32(0.0), np.float32(0.4856), np.float32(0.5592), np.float32(0.0), np.float32(0.0), np.float32(0.6054), np.float32(0.6168), np.float32(0.8251), np.float32(0.8353)] 
+2026-11-08 09:34:35.894311: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:34:36.224107: lr: 0.000825 
+2026-11-08 09:34:36.224592: This epoch took 235.695583 s
+ 
+2026-11-08 09:34:36.224923: 
+epoch:  192 
+2026-11-08 09:38:49.182223: train loss : -0.7570 
+2026-11-08 09:38:57.297388: validation loss: -0.5227 
+2026-11-08 09:38:57.298475: Average global foreground Dice: [np.float32(0.9007), np.float32(0.5236), np.float32(0.6044), np.float32(0.5954), np.float32(0.5315), np.float32(0.6591), np.float32(0.5816), np.float32(0.0135), np.float32(0.0), np.float32(0.6135), np.float32(0.4772), np.float32(0.5506), np.float32(0.555), np.float32(0.4618), np.float32(0.0), np.float32(0.0), np.float32(0.5358), np.float32(0.4562), np.float32(0.5499), np.float32(0.5192), np.float32(0.4518), np.float32(0.4131), np.float32(0.7393), np.float32(0.7003), np.float32(0.3311), np.float32(0.3233), np.float32(0.0), np.float32(0.0), np.float32(0.309), np.float32(0.5402), np.float32(0.0), np.float32(0.0), np.float32(0.4512), np.float32(0.5468), np.float32(0.6412), np.float32(0.6743)] 
+2026-11-08 09:38:57.298863: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:38:57.622823: lr: 0.000824 
+2026-11-08 09:38:57.623037: This epoch took 261.397790 s
+ 
+2026-11-08 09:38:57.623119: 
+epoch:  193 
+2026-11-08 09:43:03.192372: train loss : -0.7633 
+2026-11-08 09:43:11.276368: validation loss: -0.2684 
+2026-11-08 09:43:11.277313: Average global foreground Dice: [np.float32(0.8842), np.float32(0.1831), np.float32(0.1759), np.float32(0.2389), np.float32(0.2052), np.float32(0.2195), np.float32(0.2083), np.float32(0.0188), np.float32(0.0), np.float32(0.3338), np.float32(0.2089), np.float32(0.1824), np.float32(0.3729), np.float32(0.2044), np.float32(0.0), np.float32(0.0), np.float32(0.2375), np.float32(0.1762), np.float32(0.2098), np.float32(0.2442), np.float32(0.0292), np.float32(0.1225), np.float32(0.1727), np.float32(0.0678), np.float32(0.1003), np.float32(0.0568), np.float32(0.0), np.float32(0.0), np.float32(0.0052), np.float32(0.0308), np.float32(0.0), np.float32(0.0), np.float32(0.1053), np.float32(0.1701), np.float32(0.2107), np.float32(0.1728)] 
+2026-11-08 09:43:11.277471: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:43:11.634162: lr: 0.000824 
+2026-11-08 09:43:11.634406: This epoch took 254.011208 s
+ 
+2026-11-08 09:43:11.634489: 
+epoch:  194 
+2026-11-08 09:47:06.611609: train loss : -0.7475 
+2026-11-08 09:47:14.742356: validation loss: -0.5925 
+2026-11-08 09:47:14.743527: Average global foreground Dice: [np.float32(0.9202), np.float32(0.6139), np.float32(0.7348), np.float32(0.7403), np.float32(0.6666), np.float32(0.7949), np.float32(0.6832), np.float32(0.2904), np.float32(0.0), np.float32(0.6073), np.float32(0.6129), np.float32(0.6546), np.float32(0.6168), np.float32(0.5011), np.float32(0.0), np.float32(0.0), np.float32(0.7228), np.float32(0.5783), np.float32(0.6995), np.float32(0.6164), np.float32(0.4926), np.float32(0.5264), np.float32(0.7996), np.float32(0.7898), np.float32(0.4945), np.float32(0.4218), np.float32(0.0), np.float32(0.0), np.float32(0.4187), np.float32(0.6336), np.float32(0.0), np.float32(0.0), np.float32(0.6038), np.float32(0.6552), np.float32(0.7815), np.float32(0.8033)] 
+2026-11-08 09:47:14.743803: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:47:15.071764: lr: 0.000823 
+2026-11-08 09:47:15.071992: This epoch took 243.437445 s
+ 
+2026-11-08 09:47:15.072073: 
+epoch:  195 
+2026-11-08 09:51:07.270556: train loss : -0.7028 
+2026-11-08 09:51:15.340655: validation loss: -0.4478 
+2026-11-08 09:51:15.341617: Average global foreground Dice: [np.float32(0.8945), np.float32(0.445), np.float32(0.4585), np.float32(0.4574), np.float32(0.3631), np.float32(0.4994), np.float32(0.4395), np.float32(0.0629), np.float32(0.0), np.float32(0.5667), np.float32(0.3703), np.float32(0.4542), np.float32(0.4808), np.float32(0.4119), np.float32(0.0), np.float32(0.0), np.float32(0.2943), np.float32(0.2567), np.float32(0.3396), np.float32(0.3541), np.float32(0.3537), np.float32(0.2708), np.float32(0.6027), np.float32(0.5528), np.float32(0.1591), np.float32(0.2069), np.float32(0.0), np.float32(0.0), np.float32(0.2144), np.float32(0.5991), np.float32(0.0), np.float32(0.0), np.float32(0.266), np.float32(0.3446), np.float32(0.4878), np.float32(0.5216)] 
+2026-11-08 09:51:15.342019: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:51:15.666333: lr: 0.000822 
+2026-11-08 09:51:15.666626: This epoch took 240.594447 s
+ 
+2026-11-08 09:51:15.666750: 
+epoch:  196 
+2026-11-08 09:54:45.905033: train loss : -0.7433 
+2026-11-08 09:54:53.986669: validation loss: -0.5901 
+2026-11-08 09:54:53.987773: Average global foreground Dice: [np.float32(0.9149), np.float32(0.6125), np.float32(0.7103), np.float32(0.7432), np.float32(0.6677), np.float32(0.794), np.float32(0.6632), np.float32(0.209), np.float32(0.0), np.float32(0.6282), np.float32(0.5761), np.float32(0.6307), np.float32(0.6236), np.float32(0.5011), np.float32(0.0), np.float32(0.0), np.float32(0.7155), np.float32(0.5703), np.float32(0.6822), np.float32(0.612), np.float32(0.4698), np.float32(0.5645), np.float32(0.8155), np.float32(0.8021), np.float32(0.5213), np.float32(0.2778), np.float32(0.0), np.float32(0.0), np.float32(0.3906), np.float32(0.649), np.float32(0.0), np.float32(0.0), np.float32(0.5826), np.float32(0.6592), np.float32(0.7773), np.float32(0.7934)] 
+2026-11-08 09:54:53.988035: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:54:54.328856: lr: 0.000821 
+2026-11-08 09:54:54.329136: This epoch took 218.662324 s
+ 
+2026-11-08 09:54:54.329248: 
+epoch:  197 
+2026-11-08 09:58:52.333107: train loss : -0.7541 
+2026-11-08 09:59:00.434044: validation loss: -0.5845 
+2026-11-08 09:59:00.435213: Average global foreground Dice: [np.float32(0.9108), np.float32(0.6365), np.float32(0.6914), np.float32(0.7297), np.float32(0.6386), np.float32(0.7735), np.float32(0.6657), np.float32(0.7072), np.float32(0.0), np.float32(0.5608), np.float32(0.6145), np.float32(0.667), np.float32(0.6119), np.float32(0.4731), np.float32(0.0), np.float32(0.0), np.float32(0.6984), np.float32(0.5301), np.float32(0.6697), np.float32(0.6083), np.float32(0.5073), np.float32(0.5421), np.float32(0.7838), np.float32(0.7641), np.float32(0.4827), np.float32(0.2547), np.float32(0.0), np.float32(0.0), np.float32(0.1771), np.float32(0.557), np.float32(0.0), np.float32(0.0), np.float32(0.587), np.float32(0.6365), np.float32(0.7669), np.float32(0.7896)] 
+2026-11-08 09:59:00.435460: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 09:59:00.761014: lr: 0.00082 
+2026-11-08 09:59:00.761406: This epoch took 246.432102 s
+ 
+2026-11-08 09:59:00.761610: 
+epoch:  198 
+2026-11-08 10:03:18.043310: train loss : -0.7519 
+2026-11-08 10:03:26.159024: validation loss: -0.4906 
+2026-11-08 10:03:26.160143: Average global foreground Dice: [np.float32(0.9083), np.float32(0.5472), np.float32(0.5177), np.float32(0.5563), np.float32(0.4601), np.float32(0.5916), np.float32(0.4972), np.float32(0.529), np.float32(0.0), np.float32(0.3877), np.float32(0.4272), np.float32(0.4938), np.float32(0.5569), np.float32(0.3788), np.float32(0.0), np.float32(0.0), np.float32(0.4027), np.float32(0.3637), np.float32(0.4181), np.float32(0.4611), np.float32(0.404), np.float32(0.3569), np.float32(0.6446), np.float32(0.608), np.float32(0.2424), np.float32(0.1938), np.float32(0.0), np.float32(0.0), np.float32(0.1258), np.float32(0.5089), np.float32(0.0), np.float32(0.0), np.float32(0.4042), np.float32(0.4909), np.float32(0.599), np.float32(0.6016)] 
+2026-11-08 10:03:26.160511: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:03:26.490248: lr: 0.000819 
+2026-11-08 10:03:26.490502: This epoch took 265.728714 s
+ 
+2026-11-08 10:03:26.490586: 
+epoch:  199 
+2026-11-08 10:07:45.128059: train loss : -0.7588 
+2026-11-08 10:07:53.239424: validation loss: -0.5666 
+2026-11-08 10:07:53.240591: Average global foreground Dice: [np.float32(0.9202), np.float32(0.5739), np.float32(0.6906), np.float32(0.7018), np.float32(0.6248), np.float32(0.7573), np.float32(0.6674), np.float32(0.5265), np.float32(0.0), np.float32(0.3963), np.float32(0.536), np.float32(0.6166), np.float32(0.5943), np.float32(0.5218), np.float32(0.0), np.float32(0.0), np.float32(0.6607), np.float32(0.5385), np.float32(0.6462), np.float32(0.5965), np.float32(0.4445), np.float32(0.5634), np.float32(0.7587), np.float32(0.7437), np.float32(0.4429), np.float32(0.2974), np.float32(0.0), np.float32(0.0), np.float32(0.2772), np.float32(0.6856), np.float32(0.0), np.float32(0.0), np.float32(0.5912), np.float32(0.6235), np.float32(0.7242), np.float32(0.7532)] 
+2026-11-08 10:07:53.240812: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:07:53.592353: lr: 0.000818 
+2026-11-08 10:07:53.592578: saving scheduled checkpoint file... 
+2026-11-08 10:07:53.622159: saving checkpoint... 
+2026-11-08 10:07:54.176855: done, saving took 0.58 seconds 
+2026-11-08 10:07:54.178859: done 
+2026-11-08 10:07:54.179022: This epoch took 267.688382 s
+ 
+2026-11-08 10:07:54.179101: 
+epoch:  200 
+2026-11-08 10:12:20.173901: train loss : -0.7476 
+2026-11-08 10:12:28.276603: validation loss: -0.5607 
+2026-11-08 10:12:28.277847: Average global foreground Dice: [np.float32(0.9237), np.float32(0.5706), np.float32(0.6956), np.float32(0.7133), np.float32(0.6489), np.float32(0.7828), np.float32(0.6715), np.float32(0.1118), np.float32(0.0), np.float32(0.489), np.float32(0.569), np.float32(0.6599), np.float32(0.532), np.float32(0.3271), np.float32(0.0), np.float32(0.0), np.float32(0.7036), np.float32(0.5651), np.float32(0.6965), np.float32(0.5983), np.float32(0.3784), np.float32(0.4782), np.float32(0.7374), np.float32(0.7347), np.float32(0.4758), np.float32(0.4094), np.float32(0.0), np.float32(0.0), np.float32(0.1264), np.float32(0.6238), np.float32(0.0), np.float32(0.0), np.float32(0.398), np.float32(0.5544), np.float32(0.757), np.float32(0.7901)] 
+2026-11-08 10:12:28.278299: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:12:28.604707: lr: 0.000817 
+2026-11-08 10:12:28.605012: This epoch took 274.425852 s
+ 
+2026-11-08 10:12:28.605120: 
+epoch:  201 
+2026-11-08 10:16:34.826148: train loss : -0.7613 
+2026-11-08 10:16:42.947858: validation loss: -0.5070 
+2026-11-08 10:16:42.948992: Average global foreground Dice: [np.float32(0.8985), np.float32(0.5193), np.float32(0.5696), np.float32(0.5867), np.float32(0.5085), np.float32(0.652), np.float32(0.5674), np.float32(0.2825), np.float32(0.0), np.float32(0.3349), np.float32(0.4772), np.float32(0.5901), np.float32(0.5431), np.float32(0.4483), np.float32(0.0), np.float32(0.0), np.float32(0.5223), np.float32(0.457), np.float32(0.525), np.float32(0.4628), np.float32(0.3676), np.float32(0.376), np.float32(0.6495), np.float32(0.6137), np.float32(0.3085), np.float32(0.2309), np.float32(0.0), np.float32(0.0), np.float32(0.1184), np.float32(0.5042), np.float32(0.0), np.float32(0.0), np.float32(0.4511), np.float32(0.5242), np.float32(0.6158), np.float32(0.6588)] 
+2026-11-08 10:16:42.949216: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:16:43.273912: lr: 0.000816 
+2026-11-08 10:16:43.274128: This epoch took 254.668950 s
+ 
+2026-11-08 10:16:43.274259: 
+epoch:  202 
+2026-11-08 10:20:30.234514: train loss : -0.7743 
+2026-11-08 10:20:38.321459: validation loss: -0.5231 
+2026-11-08 10:20:38.322538: Average global foreground Dice: [np.float32(0.9263), np.float32(0.5611), np.float32(0.5995), np.float32(0.599), np.float32(0.5118), np.float32(0.667), np.float32(0.5797), np.float32(0.5397), np.float32(0.0), np.float32(0.4198), np.float32(0.4717), np.float32(0.5848), np.float32(0.5691), np.float32(0.4846), np.float32(0.0), np.float32(0.0), np.float32(0.4838), np.float32(0.3905), np.float32(0.5108), np.float32(0.484), np.float32(0.4163), np.float32(0.4001), np.float32(0.6844), np.float32(0.654), np.float32(0.3203), np.float32(0.2548), np.float32(0.0), np.float32(0.0), np.float32(0.2581), np.float32(0.6346), np.float32(0.0), np.float32(0.0), np.float32(0.4672), np.float32(0.4961), np.float32(0.624), np.float32(0.6607)] 
+2026-11-08 10:20:38.323270: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:20:38.653855: lr: 0.000815 
+2026-11-08 10:20:38.654083: This epoch took 235.379624 s
+ 
+2026-11-08 10:20:38.654167: 
+epoch:  203 
+2026-11-08 10:25:13.701142: train loss : -0.7695 
+2026-11-08 10:25:21.789190: validation loss: -0.4871 
+2026-11-08 10:25:21.790168: Average global foreground Dice: [np.float32(0.9211), np.float32(0.5424), np.float32(0.5086), np.float32(0.5249), np.float32(0.4143), np.float32(0.5907), np.float32(0.4886), np.float32(0.6003), np.float32(0.0), np.float32(0.3736), np.float32(0.4454), np.float32(0.4796), np.float32(0.5378), np.float32(0.4595), np.float32(0.0), np.float32(0.0), np.float32(0.3708), np.float32(0.3096), np.float32(0.413), np.float32(0.3983), np.float32(0.473), np.float32(0.3606), np.float32(0.6817), np.float32(0.6446), np.float32(0.2092), np.float32(0.2064), np.float32(0.0), np.float32(0.0), np.float32(0.2846), np.float32(0.6381), np.float32(0.0), np.float32(0.0), np.float32(0.3931), np.float32(0.4659), np.float32(0.5861), np.float32(0.6091)] 
+2026-11-08 10:25:21.791559: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:25:22.118867: lr: 0.000814 
+2026-11-08 10:25:22.119136: This epoch took 283.464821 s
+ 
+2026-11-08 10:25:22.119346: 
+epoch:  204 
+2026-11-08 10:29:42.224822: train loss : -0.7594 
+2026-11-08 10:29:50.356427: validation loss: -0.5172 
+2026-11-08 10:29:50.357475: Average global foreground Dice: [np.float32(0.9062), np.float32(0.5918), np.float32(0.5506), np.float32(0.5638), np.float32(0.4189), np.float32(0.636), np.float32(0.5152), np.float32(0.8258), np.float32(0.0), np.float32(0.4253), np.float32(0.4826), np.float32(0.5237), np.float32(0.5353), np.float32(0.5172), np.float32(0.0), np.float32(0.0), np.float32(0.3645), np.float32(0.303), np.float32(0.4382), np.float32(0.4269), np.float32(0.4705), np.float32(0.3308), np.float32(0.6549), np.float32(0.6277), np.float32(0.2131), np.float32(0.1986), np.float32(0.0), np.float32(0.0), np.float32(0.3143), np.float32(0.698), np.float32(0.0), np.float32(0.0), np.float32(0.4564), np.float32(0.5206), np.float32(0.5989), np.float32(0.6365)] 
+2026-11-08 10:29:50.357918: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:29:50.680254: lr: 0.000813 
+2026-11-08 10:29:50.680700: This epoch took 268.561270 s
+ 
+2026-11-08 10:29:50.680944: 
+epoch:  205 
+2026-11-08 10:34:11.837477: train loss : -0.7669 
+2026-11-08 10:34:19.970954: validation loss: -0.4934 
+2026-11-08 10:34:19.971907: Average global foreground Dice: [np.float32(0.8999), np.float32(0.5255), np.float32(0.5096), np.float32(0.5141), np.float32(0.4461), np.float32(0.5912), np.float32(0.4786), np.float32(0.5808), np.float32(0.0), np.float32(0.4002), np.float32(0.4405), np.float32(0.4903), np.float32(0.5527), np.float32(0.4699), np.float32(0.0), np.float32(0.0), np.float32(0.3914), np.float32(0.3191), np.float32(0.4209), np.float32(0.4147), np.float32(0.413), np.float32(0.3257), np.float32(0.6649), np.float32(0.6027), np.float32(0.1809), np.float32(0.2763), np.float32(0.0), np.float32(0.0), np.float32(0.2524), np.float32(0.6293), np.float32(0.0), np.float32(0.0), np.float32(0.4217), np.float32(0.479), np.float32(0.5731), np.float32(0.6091)] 
+2026-11-08 10:34:19.972186: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:34:20.305225: lr: 0.000813 
+2026-11-08 10:34:20.305501: This epoch took 269.624452 s
+ 
+2026-11-08 10:34:20.305592: 
+epoch:  206 
+2026-11-08 10:38:29.469471: train loss : -0.7621 
+2026-11-08 10:38:37.587368: validation loss: -0.6090 
+2026-11-08 10:38:37.588389: Average global foreground Dice: [np.float32(0.9118), np.float32(0.6606), np.float32(0.7483), np.float32(0.7867), np.float32(0.7143), np.float32(0.8314), np.float32(0.7312), np.float32(0.7442), np.float32(0.0), np.float32(0.4246), np.float32(0.5892), np.float32(0.6757), np.float32(0.6295), np.float32(0.516), np.float32(0.0), np.float32(0.0), np.float32(0.7613), np.float32(0.6222), np.float32(0.7285), np.float32(0.6427), np.float32(0.5243), np.float32(0.6329), np.float32(0.852), np.float32(0.8316), np.float32(0.4983), np.float32(0.2793), np.float32(0.0), np.float32(0.0), np.float32(0.3058), np.float32(0.613), np.float32(0.0), np.float32(0.0), np.float32(0.5861), np.float32(0.6407), np.float32(0.8118), np.float32(0.8317)] 
+2026-11-08 10:38:37.588778: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:38:37.910482: lr: 0.000812 
+2026-11-08 10:38:37.910913: This epoch took 257.605221 s
+ 
+2026-11-08 10:38:37.911225: 
+epoch:  207 
+2026-11-08 10:42:26.385466: train loss : -0.7635 
+2026-11-08 10:42:34.490051: validation loss: -0.5841 
+2026-11-08 10:42:34.491192: Average global foreground Dice: [np.float32(0.9059), np.float32(0.6397), np.float32(0.7282), np.float32(0.7633), np.float32(0.6748), np.float32(0.8096), np.float32(0.7196), np.float32(0.4257), np.float32(0.0), np.float32(0.5814), np.float32(0.5365), np.float32(0.6281), np.float32(0.4855), np.float32(0.4437), np.float32(0.0), np.float32(0.0), np.float32(0.7282), np.float32(0.5552), np.float32(0.7273), np.float32(0.5909), np.float32(0.5613), np.float32(0.5983), np.float32(0.8531), np.float32(0.8312), np.float32(0.5126), np.float32(0.1953), np.float32(0.0), np.float32(0.0), np.float32(0.2568), np.float32(0.5096), np.float32(0.0), np.float32(0.0), np.float32(0.5791), np.float32(0.599), np.float32(0.8115), np.float32(0.8318)] 
+2026-11-08 10:42:34.491564: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:42:34.808905: lr: 0.000811 
+2026-11-08 10:42:34.809159: This epoch took 236.897704 s
+ 
+2026-11-08 10:42:34.809254: 
+epoch:  208 
+2026-11-08 10:46:17.529366: train loss : -0.7498 
+2026-11-08 10:46:25.647563: validation loss: -0.6128 
+2026-11-08 10:46:25.648438: Average global foreground Dice: [np.float32(0.9128), np.float32(0.6828), np.float32(0.7518), np.float32(0.7758), np.float32(0.6465), np.float32(0.8117), np.float32(0.7042), np.float32(0.7232), np.float32(0.0), np.float32(0.5923), np.float32(0.6202), np.float32(0.6689), np.float32(0.6211), np.float32(0.5116), np.float32(0.0), np.float32(0.0), np.float32(0.7245), np.float32(0.5716), np.float32(0.7028), np.float32(0.6486), np.float32(0.5881), np.float32(0.621), np.float32(0.8172), np.float32(0.8111), np.float32(0.4522), np.float32(0.3717), np.float32(0.0), np.float32(0.0), np.float32(0.3786), np.float32(0.7444), np.float32(0.0), np.float32(0.0), np.float32(0.4644), np.float32(0.581), np.float32(0.7998), np.float32(0.8153)] 
+2026-11-08 10:46:25.648782: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:46:25.964938: lr: 0.00081 
+2026-11-08 10:46:25.965181: This epoch took 231.155856 s
+ 
+2026-11-08 10:46:25.965372: 
+epoch:  209 
+2026-11-08 10:50:26.951096: train loss : -0.7584 
+2026-11-08 10:50:35.064054: validation loss: -0.5180 
+2026-11-08 10:50:35.065185: Average global foreground Dice: [np.float32(0.9045), np.float32(0.5368), np.float32(0.5813), np.float32(0.6502), np.float32(0.6457), np.float32(0.708), np.float32(0.6816), np.float32(0.3519), np.float32(0.0), np.float32(0.5248), np.float32(0.5426), np.float32(0.5749), np.float32(0.5919), np.float32(0.4737), np.float32(0.0), np.float32(0.0), np.float32(0.6931), np.float32(0.5759), np.float32(0.6471), np.float32(0.5788), np.float32(0.3916), np.float32(0.398), np.float32(0.676), np.float32(0.6048), np.float32(0.4302), np.float32(0.3335), np.float32(0.0), np.float32(0.0), np.float32(0.1551), np.float32(0.4738), np.float32(0.0), np.float32(0.0), np.float32(0.557), np.float32(0.3595), np.float32(0.6293), np.float32(0.6574)] 
+2026-11-08 10:50:35.065798: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:50:35.384300: lr: 0.000809 
+2026-11-08 10:50:35.384513: This epoch took 249.419054 s
+ 
+2026-11-08 10:50:35.384591: 
+epoch:  210 
+2026-11-08 10:54:00.027876: train loss : -0.7610 
+2026-11-08 10:54:08.104856: validation loss: -0.5046 
+2026-11-08 10:54:08.105874: Average global foreground Dice: [np.float32(0.9215), np.float32(0.5337), np.float32(0.5552), np.float32(0.5638), np.float32(0.4928), np.float32(0.6239), np.float32(0.5415), np.float32(0.2182), np.float32(0.0), np.float32(0.5697), np.float32(0.4393), np.float32(0.4987), np.float32(0.5506), np.float32(0.407), np.float32(0.0), np.float32(0.0), np.float32(0.5191), np.float32(0.4415), np.float32(0.5258), np.float32(0.5193), np.float32(0.4863), np.float32(0.4519), np.float32(0.6723), np.float32(0.6375), np.float32(0.3601), np.float32(0.2432), np.float32(0.0), np.float32(0.0), np.float32(0.3279), np.float32(0.5834), np.float32(0.0), np.float32(0.0), np.float32(0.4337), np.float32(0.4795), np.float32(0.6132), np.float32(0.6379)] 
+2026-11-08 10:54:08.106217: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:54:08.421975: lr: 0.000808 
+2026-11-08 10:54:08.422253: This epoch took 213.037609 s
+ 
+2026-11-08 10:54:08.422442: 
+epoch:  211 
+2026-11-08 10:57:42.300522: train loss : -0.7683 
+2026-11-08 10:57:50.391903: validation loss: -0.5471 
+2026-11-08 10:57:50.392833: Average global foreground Dice: [np.float32(0.9204), np.float32(0.6006), np.float32(0.6474), np.float32(0.6565), np.float32(0.5694), np.float32(0.7175), np.float32(0.5973), np.float32(0.4934), np.float32(0.0), np.float32(0.5778), np.float32(0.4871), np.float32(0.5702), np.float32(0.5138), np.float32(0.4689), np.float32(0.0), np.float32(0.0), np.float32(0.6386), np.float32(0.5203), np.float32(0.6336), np.float32(0.6125), np.float32(0.4842), np.float32(0.4897), np.float32(0.7339), np.float32(0.7126), np.float32(0.4162), np.float32(0.3169), np.float32(0.0), np.float32(0.0), np.float32(0.3713), np.float32(0.6875), np.float32(0.0), np.float32(0.0), np.float32(0.5439), np.float32(0.5779), np.float32(0.7108), np.float32(0.7477)] 
+2026-11-08 10:57:50.393106: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 10:57:50.704569: lr: 0.000807 
+2026-11-08 10:57:50.731664: saving checkpoint... 
+2026-11-08 10:57:51.240191: done, saving took 0.54 seconds 
+2026-11-08 10:57:51.241986: This epoch took 222.819468 s
+ 
+2026-11-08 10:57:51.242173: 
+epoch:  212 
+2026-11-08 11:01:02.169520: train loss : -0.7699 
+2026-11-08 11:01:10.247419: validation loss: -0.5190 
+2026-11-08 11:01:10.248291: Average global foreground Dice: [np.float32(0.917), np.float32(0.5647), np.float32(0.6518), np.float32(0.6471), np.float32(0.6264), np.float32(0.6974), np.float32(0.6339), np.float32(0.4671), np.float32(0.0), np.float32(0.4564), np.float32(0.4374), np.float32(0.563), np.float32(0.5201), np.float32(0.4867), np.float32(0.0), np.float32(0.0), np.float32(0.6294), np.float32(0.5362), np.float32(0.6219), np.float32(0.546), np.float32(0.4775), np.float32(0.4715), np.float32(0.7156), np.float32(0.6867), np.float32(0.4221), np.float32(0.3513), np.float32(0.0), np.float32(0.0), np.float32(0.104), np.float32(0.2675), np.float32(0.0), np.float32(0.0), np.float32(0.527), np.float32(0.5057), np.float32(0.6795), np.float32(0.7219)] 
+2026-11-08 11:01:10.248494: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:01:10.557623: lr: 0.000806 
+2026-11-08 11:01:10.557839: This epoch took 199.315504 s
+ 
+2026-11-08 11:01:10.557917: 
+epoch:  213 
+2026-11-08 11:04:28.895521: train loss : -0.7597 
+2026-11-08 11:04:36.979832: validation loss: -0.5701 
+2026-11-08 11:04:36.980885: Average global foreground Dice: [np.float32(0.9096), np.float32(0.6301), np.float32(0.7143), np.float32(0.7347), np.float32(0.6695), np.float32(0.7946), np.float32(0.6841), np.float32(0.6344), np.float32(0.0), np.float32(0.1522), np.float32(0.5539), np.float32(0.6873), np.float32(0.6292), np.float32(0.5618), np.float32(0.0), np.float32(0.0), np.float32(0.7159), np.float32(0.5963), np.float32(0.6969), np.float32(0.628), np.float32(0.4364), np.float32(0.4693), np.float32(0.7784), np.float32(0.7566), np.float32(0.4604), np.float32(0.4142), np.float32(0.0), np.float32(0.0), np.float32(0.2614), np.float32(0.6381), np.float32(0.0), np.float32(0.0), np.float32(0.5791), np.float32(0.6094), np.float32(0.7519), np.float32(0.8001)] 
+2026-11-08 11:04:36.981369: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:04:37.301978: lr: 0.000805 
+2026-11-08 11:04:37.327092: saving checkpoint... 
+2026-11-08 11:04:37.835827: done, saving took 0.53 seconds 
+2026-11-08 11:04:37.838094: This epoch took 207.280120 s
+ 
+2026-11-08 11:04:37.838299: 
+epoch:  214 
+2026-11-08 11:08:14.175224: train loss : -0.7593 
+2026-11-08 11:08:22.256639: validation loss: -0.5872 
+2026-11-08 11:08:22.257536: Average global foreground Dice: [np.float32(0.9078), np.float32(0.6408), np.float32(0.7406), np.float32(0.7888), np.float32(0.726), np.float32(0.8308), np.float32(0.7404), np.float32(0.8083), np.float32(0.0), np.float32(0.4734), np.float32(0.5956), np.float32(0.6645), np.float32(0.6113), np.float32(0.496), np.float32(0.0), np.float32(0.0), np.float32(0.766), np.float32(0.5878), np.float32(0.7351), np.float32(0.628), np.float32(0.519), np.float32(0.6193), np.float32(0.8493), np.float32(0.8264), np.float32(0.5256), np.float32(0.3697), np.float32(0.0), np.float32(0.0), np.float32(0.1863), np.float32(0.3757), np.float32(0.0), np.float32(0.0), np.float32(0.4262), np.float32(0.5649), np.float32(0.8225), np.float32(0.8426)] 
+2026-11-08 11:08:22.257802: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:08:22.582100: lr: 0.000804 
+2026-11-08 11:08:22.610527: saving checkpoint... 
+2026-11-08 11:08:23.125458: done, saving took 0.54 seconds 
+2026-11-08 11:08:23.128910: This epoch took 225.290540 s
+ 
+2026-11-08 11:08:23.129063: 
+epoch:  215 
+2026-11-08 11:11:50.505450: train loss : -0.7729 
+2026-11-08 11:11:58.595568: validation loss: -0.5713 
+2026-11-08 11:11:58.596480: Average global foreground Dice: [np.float32(0.9205), np.float32(0.6088), np.float32(0.6741), np.float32(0.736), np.float32(0.6588), np.float32(0.7772), np.float32(0.6883), np.float32(0.773), np.float32(0.0), np.float32(0.3308), np.float32(0.5082), np.float32(0.6189), np.float32(0.5845), np.float32(0.4669), np.float32(0.0), np.float32(0.0), np.float32(0.699), np.float32(0.5604), np.float32(0.6732), np.float32(0.6261), np.float32(0.551), np.float32(0.5972), np.float32(0.7839), np.float32(0.7765), np.float32(0.5141), np.float32(0.3187), np.float32(0.0), np.float32(0.0), np.float32(0.3289), np.float32(0.5038), np.float32(0.0), np.float32(0.0), np.float32(0.5951), np.float32(0.6239), np.float32(0.7666), np.float32(0.7865)] 
+2026-11-08 11:11:58.596753: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:11:58.925667: lr: 0.000803 
+2026-11-08 11:11:58.952099: saving checkpoint... 
+2026-11-08 11:12:01.590555: done, saving took 2.66 seconds 
+2026-11-08 11:12:01.592549: This epoch took 218.463405 s
+ 
+2026-11-08 11:12:01.592938: 
+epoch:  216 
+2026-11-08 11:15:34.159244: train loss : -0.7685 
+2026-11-08 11:15:42.241069: validation loss: -0.5434 
+2026-11-08 11:15:42.241916: Average global foreground Dice: [np.float32(0.925), np.float32(0.5894), np.float32(0.6691), np.float32(0.699), np.float32(0.6325), np.float32(0.742), np.float32(0.6505), np.float32(0.7594), np.float32(0.0), np.float32(0.4112), np.float32(0.4911), np.float32(0.6165), np.float32(0.5693), np.float32(0.4983), np.float32(0.0), np.float32(0.0), np.float32(0.6824), np.float32(0.5427), np.float32(0.6562), np.float32(0.5949), np.float32(0.4896), np.float32(0.5603), np.float32(0.7585), np.float32(0.724), np.float32(0.4781), np.float32(0.2799), np.float32(0.0), np.float32(0.0), np.float32(0.3265), np.float32(0.5043), np.float32(0.0), np.float32(0.0), np.float32(0.5568), np.float32(0.6224), np.float32(0.7313), np.float32(0.7591)] 
+2026-11-08 11:15:42.242296: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:15:42.550064: lr: 0.000802 
+2026-11-08 11:15:42.576432: saving checkpoint... 
+2026-11-08 11:15:43.112593: done, saving took 0.56 seconds 
+2026-11-08 11:15:43.114950: This epoch took 221.521719 s
+ 
+2026-11-08 11:15:43.115285: 
+epoch:  217 
+2026-11-08 11:19:06.995856: train loss : -0.7639 
+2026-11-08 11:19:15.120458: validation loss: -0.5755 
+2026-11-08 11:19:15.121614: Average global foreground Dice: [np.float32(0.9265), np.float32(0.6042), np.float32(0.7115), np.float32(0.7154), np.float32(0.6451), np.float32(0.7616), np.float32(0.6761), np.float32(0.6667), np.float32(0.0), np.float32(0.5834), np.float32(0.4688), np.float32(0.5957), np.float32(0.5653), np.float32(0.4851), np.float32(0.0), np.float32(0.0), np.float32(0.7015), np.float32(0.5626), np.float32(0.6887), np.float32(0.6228), np.float32(0.5401), np.float32(0.562), np.float32(0.8039), np.float32(0.7691), np.float32(0.4687), np.float32(0.4139), np.float32(0.0), np.float32(0.0), np.float32(0.2694), np.float32(0.6164), np.float32(0.0), np.float32(0.0), np.float32(0.5865), np.float32(0.6303), np.float32(0.7602), np.float32(0.7782)] 
+2026-11-08 11:19:15.121998: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:19:15.434830: lr: 0.000801 
+2026-11-08 11:19:15.463590: saving checkpoint... 
+2026-11-08 11:19:15.983425: done, saving took 0.55 seconds 
+2026-11-08 11:19:15.985522: This epoch took 212.870136 s
+ 
+2026-11-08 11:19:15.985699: 
+epoch:  218 
+2026-11-08 11:22:32.732940: train loss : -0.7773 
+2026-11-08 11:22:40.791333: validation loss: -0.6033 
+2026-11-08 11:22:40.792125: Average global foreground Dice: [np.float32(0.9225), np.float32(0.5927), np.float32(0.7406), np.float32(0.7824), np.float32(0.7005), np.float32(0.8325), np.float32(0.7529), np.float32(0.656), np.float32(0.0), np.float32(0.5915), np.float32(0.5652), np.float32(0.6392), np.float32(0.5938), np.float32(0.5252), np.float32(0.0), np.float32(0.0), np.float32(0.7472), np.float32(0.5792), np.float32(0.7333), np.float32(0.6544), np.float32(0.5399), np.float32(0.6423), np.float32(0.8112), np.float32(0.8047), np.float32(0.5071), np.float32(0.351), np.float32(0.0), np.float32(0.0), np.float32(0.4125), np.float32(0.5922), np.float32(0.0), np.float32(0.0), np.float32(0.6023), np.float32(0.684), np.float32(0.7974), np.float32(0.8175)] 
+2026-11-08 11:22:40.792272: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:22:41.127702: lr: 0.000801 
+2026-11-08 11:22:41.284601: saving checkpoint... 
+2026-11-08 11:22:41.965465: done, saving took 0.84 seconds 
+2026-11-08 11:22:41.967946: This epoch took 205.982101 s
+ 
+2026-11-08 11:22:41.968081: 
+epoch:  219 
+2026-11-08 11:26:14.611464: train loss : -0.7735 
+2026-11-08 11:26:22.695432: validation loss: -0.5693 
+2026-11-08 11:26:22.696580: Average global foreground Dice: [np.float32(0.9124), np.float32(0.5502), np.float32(0.7097), np.float32(0.7389), np.float32(0.6656), np.float32(0.7828), np.float32(0.6936), np.float32(0.634), np.float32(0.0), np.float32(0.4341), np.float32(0.5522), np.float32(0.5733), np.float32(0.5871), np.float32(0.4918), np.float32(0.0), np.float32(0.0), np.float32(0.7244), np.float32(0.5386), np.float32(0.7121), np.float32(0.6135), np.float32(0.543), np.float32(0.6477), np.float32(0.786), np.float32(0.7539), np.float32(0.4718), np.float32(0.3238), np.float32(0.0), np.float32(0.0), np.float32(0.3837), np.float32(0.5136), np.float32(0.0), np.float32(0.0), np.float32(0.6366), np.float32(0.6593), np.float32(0.7633), np.float32(0.7904)] 
+2026-11-08 11:26:22.696924: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:26:23.005907: lr: 0.0008 
+2026-11-08 11:26:23.032453: saving checkpoint... 
+2026-11-08 11:26:23.550040: done, saving took 0.54 seconds 
+2026-11-08 11:26:23.552400: This epoch took 221.584253 s
+ 
+2026-11-08 11:26:23.552599: 
+epoch:  220 
+2026-11-08 11:29:48.975880: train loss : -0.7722 
+2026-11-08 11:29:57.066762: validation loss: -0.5645 
+2026-11-08 11:29:57.067813: Average global foreground Dice: [np.float32(0.9168), np.float32(0.5765), np.float32(0.6804), np.float32(0.7225), np.float32(0.6073), np.float32(0.7746), np.float32(0.6763), np.float32(0.4646), np.float32(0.0), np.float32(0.5853), np.float32(0.5798), np.float32(0.6122), np.float32(0.5381), np.float32(0.4834), np.float32(0.0), np.float32(0.0), np.float32(0.6949), np.float32(0.5299), np.float32(0.6647), np.float32(0.5661), np.float32(0.4922), np.float32(0.5636), np.float32(0.7686), np.float32(0.778), np.float32(0.4817), np.float32(0.3795), np.float32(0.0), np.float32(0.0), np.float32(0.2357), np.float32(0.6798), np.float32(0.0), np.float32(0.0), np.float32(0.4341), np.float32(0.4787), np.float32(0.7621), np.float32(0.79)] 
+2026-11-08 11:29:57.067997: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:29:57.377409: lr: 0.000799 
+2026-11-08 11:29:57.406549: saving checkpoint... 
+2026-11-08 11:29:57.918766: done, saving took 0.54 seconds 
+2026-11-08 11:29:57.921375: This epoch took 214.368680 s
+ 
+2026-11-08 11:29:57.921495: 
+epoch:  221 
+2026-11-08 11:33:11.123705: train loss : -0.7376 
+2026-11-08 11:33:19.204370: validation loss: -0.5125 
+2026-11-08 11:33:19.205292: Average global foreground Dice: [np.float32(0.9169), np.float32(0.5005), np.float32(0.5946), np.float32(0.5734), np.float32(0.4668), np.float32(0.6347), np.float32(0.5501), np.float32(0.618), np.float32(0.0), np.float32(0.5398), np.float32(0.432), np.float32(0.5182), np.float32(0.5541), np.float32(0.4789), np.float32(0.0), np.float32(0.0), np.float32(0.4706), np.float32(0.4232), np.float32(0.4999), np.float32(0.4753), np.float32(0.5188), np.float32(0.5209), np.float32(0.6785), np.float32(0.7019), np.float32(0.3259), np.float32(0.2605), np.float32(0.0), np.float32(0.0), np.float32(0.3422), np.float32(0.6954), np.float32(0.0), np.float32(0.0), np.float32(0.4641), np.float32(0.5185), np.float32(0.6516), np.float32(0.7011)] 
+2026-11-08 11:33:19.205533: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:33:19.520031: lr: 0.000798 
+2026-11-08 11:33:19.520283: This epoch took 201.598719 s
+ 
+2026-11-08 11:33:19.520364: 
+epoch:  222 
+2026-11-08 11:37:02.802021: train loss : -0.7516 
+2026-11-08 11:37:10.927247: validation loss: -0.4445 
+2026-11-08 11:37:10.928123: Average global foreground Dice: [np.float32(0.882), np.float32(0.468), np.float32(0.4594), np.float32(0.457), np.float32(0.4126), np.float32(0.5295), np.float32(0.4207), np.float32(0.4861), np.float32(0.0), np.float32(0.5594), np.float32(0.3633), np.float32(0.4575), np.float32(0.5485), np.float32(0.5561), np.float32(0.0), np.float32(0.0), np.float32(0.3437), np.float32(0.3002), np.float32(0.3626), np.float32(0.3577), np.float32(0.3366), np.float32(0.2326), np.float32(0.6165), np.float32(0.5483), np.float32(0.1623), np.float32(0.2208), np.float32(0.0), np.float32(0.0), np.float32(0.0005), np.float32(0.351), np.float32(0.0), np.float32(0.0), np.float32(0.287), np.float32(0.3002), np.float32(0.507), np.float32(0.5585)] 
+2026-11-08 11:37:10.928342: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:37:11.235618: lr: 0.000797 
+2026-11-08 11:37:11.236010: This epoch took 231.715590 s
+ 
+2026-11-08 11:37:11.236197: 
+epoch:  223 
+2026-11-08 11:40:36.987815: train loss : -0.7520 
+2026-11-08 11:40:45.063305: validation loss: -0.6122 
+2026-11-08 11:40:45.064257: Average global foreground Dice: [np.float32(0.9238), np.float32(0.6508), np.float32(0.7788), np.float32(0.7913), np.float32(0.7546), np.float32(0.8304), np.float32(0.7397), np.float32(0.623), np.float32(0.0), np.float32(0.569), np.float32(0.556), np.float32(0.6875), np.float32(0.5535), np.float32(0.5455), np.float32(0.0), np.float32(0.0), np.float32(0.7711), np.float32(0.6371), np.float32(0.7493), np.float32(0.6535), np.float32(0.5305), np.float32(0.6169), np.float32(0.8436), np.float32(0.8233), np.float32(0.4693), np.float32(0.3435), np.float32(0.0), np.float32(0.0), np.float32(0.2616), np.float32(0.5649), np.float32(0.0), np.float32(0.0), np.float32(0.6643), np.float32(0.6843), np.float32(0.839), np.float32(0.858)] 
+2026-11-08 11:40:45.064584: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:40:45.378205: lr: 0.000796 
+2026-11-08 11:40:45.378642: This epoch took 214.142361 s
+ 
+2026-11-08 11:40:45.378933: 
+epoch:  224 
+2026-11-08 11:44:09.868077: train loss : -0.7715 
+2026-11-08 11:44:17.936680: validation loss: -0.6108 
+2026-11-08 11:44:17.937627: Average global foreground Dice: [np.float32(0.9204), np.float32(0.661), np.float32(0.7639), np.float32(0.7699), np.float32(0.7083), np.float32(0.8235), np.float32(0.7105), np.float32(0.6255), np.float32(0.0), np.float32(0.4798), np.float32(0.6201), np.float32(0.6987), np.float32(0.6239), np.float32(0.5527), np.float32(0.0), np.float32(0.0), np.float32(0.7547), np.float32(0.6004), np.float32(0.7318), np.float32(0.6572), np.float32(0.4832), np.float32(0.5107), np.float32(0.8494), np.float32(0.8265), np.float32(0.4494), np.float32(0.3925), np.float32(0.0), np.float32(0.0), np.float32(0.3045), np.float32(0.6689), np.float32(0.0), np.float32(0.0), np.float32(0.6801), np.float32(0.685), np.float32(0.8146), np.float32(0.8341)] 
+2026-11-08 11:44:17.937807: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:44:18.261758: lr: 0.000795 
+2026-11-08 11:44:18.261967: This epoch took 212.882825 s
+ 
+2026-11-08 11:44:18.262036: 
+epoch:  225 
+2026-11-08 11:48:02.931180: train loss : -0.7534 
+2026-11-08 11:48:11.061587: validation loss: -0.6339 
+2026-11-08 11:48:11.062541: Average global foreground Dice: [np.float32(0.9129), np.float32(0.6764), np.float32(0.7915), np.float32(0.8393), np.float32(0.8087), np.float32(0.8634), np.float32(0.7759), np.float32(0.3985), np.float32(0.0), np.float32(0.2758), np.float32(0.5928), np.float32(0.7182), np.float32(0.5614), np.float32(0.5198), np.float32(0.0), np.float32(0.0), np.float32(0.7969), np.float32(0.612), np.float32(0.7619), np.float32(0.6591), np.float32(0.522), np.float32(0.6891), np.float32(0.893), np.float32(0.8835), np.float32(0.4634), np.float32(0.5048), np.float32(0.0), np.float32(0.0), np.float32(0.2238), np.float32(0.612), np.float32(0.0), np.float32(0.0), np.float32(0.6327), np.float32(0.6456), np.float32(0.8693), np.float32(0.8851)] 
+2026-11-08 11:48:11.062901: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:48:11.379620: lr: 0.000794 
+2026-11-08 11:48:11.406233: saving checkpoint... 
+2026-11-08 11:48:13.961208: done, saving took 2.58 seconds 
+2026-11-08 11:48:13.963591: This epoch took 235.701497 s
+ 
+2026-11-08 11:48:13.963706: 
+epoch:  226 
+2026-11-08 11:51:25.999016: train loss : -0.7674 
+2026-11-08 11:51:34.081890: validation loss: -0.6565 
+2026-11-08 11:51:34.083235: Average global foreground Dice: [np.float32(0.9249), np.float32(0.7073), np.float32(0.8016), np.float32(0.8173), np.float32(0.8051), np.float32(0.8712), np.float32(0.8103), np.float32(0.7191), np.float32(0.0), np.float32(0.5545), np.float32(0.6042), np.float32(0.7264), np.float32(0.6227), np.float32(0.591), np.float32(0.0), np.float32(0.0), np.float32(0.7922), np.float32(0.6209), np.float32(0.7653), np.float32(0.6364), np.float32(0.5534), np.float32(0.5857), np.float32(0.9143), np.float32(0.9152), np.float32(0.4695), np.float32(0.4427), np.float32(0.0), np.float32(0.0), np.float32(0.2609), np.float32(0.6761), np.float32(0.0), np.float32(0.0), np.float32(0.6724), np.float32(0.5947), np.float32(0.8762), np.float32(0.8906)] 
+2026-11-08 11:51:34.083601: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:51:34.393393: lr: 0.000793 
+2026-11-08 11:51:34.420950: saving checkpoint... 
+2026-11-08 11:51:34.931233: done, saving took 0.54 seconds 
+2026-11-08 11:51:34.933280: This epoch took 200.969513 s
+ 
+2026-11-08 11:51:34.933376: 
+epoch:  227 
+2026-11-08 11:55:14.891375: train loss : -0.7712 
+2026-11-08 11:55:22.993559: validation loss: -0.5319 
+2026-11-08 11:55:22.994541: Average global foreground Dice: [np.float32(0.9051), np.float32(0.5721), np.float32(0.5868), np.float32(0.5931), np.float32(0.5088), np.float32(0.6561), np.float32(0.5366), np.float32(0.5188), np.float32(0.0), np.float32(0.4509), np.float32(0.5103), np.float32(0.5906), np.float32(0.5941), np.float32(0.5654), np.float32(0.0), np.float32(0.0), np.float32(0.5155), np.float32(0.4882), np.float32(0.5121), np.float32(0.4851), np.float32(0.3159), np.float32(0.375), np.float32(0.7719), np.float32(0.7294), np.float32(0.2684), np.float32(0.3078), np.float32(0.0), np.float32(0.0), np.float32(0.069), np.float32(0.7198), np.float32(0.0), np.float32(0.0), np.float32(0.5175), np.float32(0.5113), np.float32(0.6556), np.float32(0.6839)] 
+2026-11-08 11:55:22.994741: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:55:23.319765: lr: 0.000792 
+2026-11-08 11:55:23.320085: This epoch took 228.386653 s
+ 
+2026-11-08 11:55:23.320304: 
+epoch:  228 
+2026-11-08 11:59:00.697432: train loss : -0.7700 
+2026-11-08 11:59:08.771459: validation loss: -0.4947 
+2026-11-08 11:59:08.772565: Average global foreground Dice: [np.float32(0.9071), np.float32(0.5136), np.float32(0.5263), np.float32(0.524), np.float32(0.4557), np.float32(0.5868), np.float32(0.4983), np.float32(0.651), np.float32(0.0), np.float32(0.3405), np.float32(0.4192), np.float32(0.5313), np.float32(0.532), np.float32(0.4625), np.float32(0.0), np.float32(0.0), np.float32(0.4235), np.float32(0.3915), np.float32(0.4376), np.float32(0.4584), np.float32(0.4265), np.float32(0.4131), np.float32(0.6761), np.float32(0.6235), np.float32(0.246), np.float32(0.2373), np.float32(0.0), np.float32(0.0), np.float32(0.0821), np.float32(0.6233), np.float32(0.0), np.float32(0.0), np.float32(0.4174), np.float32(0.502), np.float32(0.5775), np.float32(0.6225)] 
+2026-11-08 11:59:08.772990: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 11:59:09.092102: lr: 0.000791 
+2026-11-08 11:59:09.092544: This epoch took 225.772147 s
+ 
+2026-11-08 11:59:09.092719: 
+epoch:  229 
+2026-11-08 12:02:50.397024: train loss : -0.7740 
+2026-11-08 12:02:58.472943: validation loss: -0.5408 
+2026-11-08 12:02:58.474016: Average global foreground Dice: [np.float32(0.904), np.float32(0.5497), np.float32(0.6468), np.float32(0.6404), np.float32(0.5916), np.float32(0.6919), np.float32(0.6165), np.float32(0.7573), np.float32(0.0), np.float32(0.4908), np.float32(0.4518), np.float32(0.5785), np.float32(0.5239), np.float32(0.4451), np.float32(0.0), np.float32(0.0), np.float32(0.6264), np.float32(0.5141), np.float32(0.5996), np.float32(0.5489), np.float32(0.3989), np.float32(0.5394), np.float32(0.83), np.float32(0.7754), np.float32(0.4272), np.float32(0.3015), np.float32(0.0), np.float32(0.0), np.float32(0.1263), np.float32(0.6005), np.float32(0.0), np.float32(0.0), np.float32(0.4887), np.float32(0.4891), np.float32(0.6824), np.float32(0.7113)] 
+2026-11-08 12:02:58.474343: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:02:58.797402: lr: 0.00079 
+2026-11-08 12:02:58.797624: This epoch took 229.704732 s
+ 
+2026-11-08 12:02:58.797704: 
+epoch:  230 
+2026-11-08 12:06:22.457360: train loss : -0.7742 
+2026-11-08 12:06:30.534508: validation loss: -0.4665 
+2026-11-08 12:06:30.535359: Average global foreground Dice: [np.float32(0.9021), np.float32(0.4888), np.float32(0.4806), np.float32(0.4939), np.float32(0.3653), np.float32(0.5616), np.float32(0.4655), np.float32(0.6104), np.float32(0.0), np.float32(0.5484), np.float32(0.4129), np.float32(0.4391), np.float32(0.5247), np.float32(0.385), np.float32(0.0), np.float32(0.0), np.float32(0.3428), np.float32(0.2994), np.float32(0.3965), np.float32(0.4085), np.float32(0.3932), np.float32(0.3042), np.float32(0.6088), np.float32(0.5586), np.float32(0.1675), np.float32(0.2065), np.float32(0.0), np.float32(0.0), np.float32(0.1158), np.float32(0.6178), np.float32(0.0), np.float32(0.0), np.float32(0.3654), np.float32(0.4529), np.float32(0.5353), np.float32(0.5667)] 
+2026-11-08 12:06:30.535607: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:06:30.851966: lr: 0.000789 
+2026-11-08 12:06:30.852273: This epoch took 212.054508 s
+ 
+2026-11-08 12:06:30.852482: 
+epoch:  231 
+2026-11-08 12:09:59.158743: train loss : -0.7735 
+2026-11-08 12:10:07.251741: validation loss: -0.5161 
+2026-11-08 12:10:07.252646: Average global foreground Dice: [np.float32(0.9075), np.float32(0.5469), np.float32(0.5678), np.float32(0.5587), np.float32(0.4759), np.float32(0.6314), np.float32(0.5388), np.float32(0.6288), np.float32(0.0), np.float32(0.5016), np.float32(0.4425), np.float32(0.547), np.float32(0.5473), np.float32(0.4652), np.float32(0.0), np.float32(0.0), np.float32(0.4284), np.float32(0.3753), np.float32(0.4753), np.float32(0.4742), np.float32(0.4498), np.float32(0.4203), np.float32(0.7075), np.float32(0.6543), np.float32(0.2716), np.float32(0.3154), np.float32(0.0), np.float32(0.0), np.float32(0.2936), np.float32(0.6363), np.float32(0.0), np.float32(0.0), np.float32(0.4813), np.float32(0.5211), np.float32(0.5946), np.float32(0.6477)] 
+2026-11-08 12:10:07.252807: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:10:07.572986: lr: 0.000789 
+2026-11-08 12:10:07.573208: This epoch took 216.720568 s
+ 
+2026-11-08 12:10:07.573391: 
+epoch:  232 
+2026-11-08 12:13:47.067764: train loss : -0.7800 
+2026-11-08 12:13:55.140361: validation loss: -0.5357 
+2026-11-08 12:13:55.141249: Average global foreground Dice: [np.float32(0.8837), np.float32(0.5628), np.float32(0.6677), np.float32(0.6841), np.float32(0.6131), np.float32(0.7337), np.float32(0.6588), np.float32(0.5854), np.float32(0.0), np.float32(0.4496), np.float32(0.477), np.float32(0.5679), np.float32(0.5354), np.float32(0.4266), np.float32(0.0), np.float32(0.0), np.float32(0.5886), np.float32(0.5102), np.float32(0.5761), np.float32(0.5461), np.float32(0.3773), np.float32(0.4712), np.float32(0.7437), np.float32(0.659), np.float32(0.4308), np.float32(0.3465), np.float32(0.0), np.float32(0.0), np.float32(0.3858), np.float32(0.5352), np.float32(0.0), np.float32(0.0), np.float32(0.5147), np.float32(0.4796), np.float32(0.7264), np.float32(0.7527)] 
+2026-11-08 12:13:55.141671: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:13:55.455301: lr: 0.000788 
+2026-11-08 12:13:55.455738: This epoch took 227.882075 s
+ 
+2026-11-08 12:13:55.456032: 
+epoch:  233 
+2026-11-08 12:16:50.151986: train loss : -0.7851 
+2026-11-08 12:16:58.247047: validation loss: -0.5590 
+2026-11-08 12:16:58.248097: Average global foreground Dice: [np.float32(0.9122), np.float32(0.6143), np.float32(0.6424), np.float32(0.6675), np.float32(0.602), np.float32(0.7178), np.float32(0.6379), np.float32(0.596), np.float32(0.0), np.float32(0.4401), np.float32(0.5306), np.float32(0.5931), np.float32(0.5967), np.float32(0.5225), np.float32(0.0), np.float32(0.0), np.float32(0.5927), np.float32(0.4972), np.float32(0.5821), np.float32(0.5648), np.float32(0.4465), np.float32(0.3823), np.float32(0.7446), np.float32(0.7093), np.float32(0.3448), np.float32(0.3569), np.float32(0.0), np.float32(0.0), np.float32(0.2134), np.float32(0.6807), np.float32(0.0), np.float32(0.0), np.float32(0.5414), np.float32(0.5717), np.float32(0.7099), np.float32(0.7322)] 
+2026-11-08 12:16:58.248358: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:16:58.707694: lr: 0.000787 
+2026-11-08 12:16:58.707940: This epoch took 183.251747 s
+ 
+2026-11-08 12:16:58.708117: 
+epoch:  234 
+2026-11-08 12:20:14.878362: train loss : -0.7814 
+2026-11-08 12:20:22.960231: validation loss: -0.5078 
+2026-11-08 12:20:22.961139: Average global foreground Dice: [np.float32(0.9062), np.float32(0.5138), np.float32(0.5741), np.float32(0.5566), np.float32(0.4999), np.float32(0.6446), np.float32(0.5556), np.float32(0.6232), np.float32(0.0), np.float32(0.4523), np.float32(0.4034), np.float32(0.5019), np.float32(0.5538), np.float32(0.5203), np.float32(0.0), np.float32(0.0), np.float32(0.521), np.float32(0.4511), np.float32(0.522), np.float32(0.4866), np.float32(0.4417), np.float32(0.384), np.float32(0.7326), np.float32(0.6986), np.float32(0.3289), np.float32(0.324), np.float32(0.0), np.float32(0.0), np.float32(0.0981), np.float32(0.6922), np.float32(0.0), np.float32(0.0), np.float32(0.5139), np.float32(0.4693), np.float32(0.6353), np.float32(0.6831)] 
+2026-11-08 12:20:22.961492: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:20:23.279488: lr: 0.000786 
+2026-11-08 12:20:23.279826: This epoch took 204.571592 s
+ 
+2026-11-08 12:20:23.279919: 
+epoch:  235 
+2026-11-08 12:23:37.271367: train loss : -0.7830 
+2026-11-08 12:23:45.347934: validation loss: -0.6006 
+2026-11-08 12:23:45.348802: Average global foreground Dice: [np.float32(0.9215), np.float32(0.6431), np.float32(0.7274), np.float32(0.7348), np.float32(0.6564), np.float32(0.7965), np.float32(0.6778), np.float32(0.5653), np.float32(0.0), np.float32(0.4609), np.float32(0.5985), np.float32(0.6604), np.float32(0.6326), np.float32(0.6086), np.float32(0.0), np.float32(0.0), np.float32(0.7179), np.float32(0.6206), np.float32(0.7032), np.float32(0.6358), np.float32(0.5235), np.float32(0.5072), np.float32(0.806), np.float32(0.7831), np.float32(0.4251), np.float32(0.5036), np.float32(0.0), np.float32(0.0), np.float32(0.2757), np.float32(0.7211), np.float32(0.0), np.float32(0.0), np.float32(0.6674), np.float32(0.6187), np.float32(0.7788), np.float32(0.8041)] 
+2026-11-08 12:23:45.348958: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:23:45.666002: lr: 0.000785 
+2026-11-08 12:23:45.666350: This epoch took 202.386373 s
+ 
+2026-11-08 12:23:45.666465: 
+epoch:  236 
+2026-11-08 12:27:00.116868: train loss : -0.7758 
+2026-11-08 12:27:08.183725: validation loss: -0.5550 
+2026-11-08 12:27:08.184736: Average global foreground Dice: [np.float32(0.9082), np.float32(0.5634), np.float32(0.7259), np.float32(0.7207), np.float32(0.6329), np.float32(0.7705), np.float32(0.6708), np.float32(0.4975), np.float32(0.0), np.float32(0.4789), np.float32(0.4841), np.float32(0.568), np.float32(0.5172), np.float32(0.4206), np.float32(0.0), np.float32(0.0), np.float32(0.6975), np.float32(0.5657), np.float32(0.6945), np.float32(0.6304), np.float32(0.455), np.float32(0.5784), np.float32(0.7712), np.float32(0.7503), np.float32(0.5222), np.float32(0.4706), np.float32(0.0), np.float32(0.0), np.float32(0.2188), np.float32(0.6439), np.float32(0.0), np.float32(0.0), np.float32(0.5501), np.float32(0.584), np.float32(0.7616), np.float32(0.7846)] 
+2026-11-08 12:27:08.185091: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:27:08.512816: lr: 0.000784 
+2026-11-08 12:27:08.513021: This epoch took 202.846491 s
+ 
+2026-11-08 12:27:08.513084: 
+epoch:  237 
+2026-11-08 12:30:25.934021: train loss : -0.7879 
+2026-11-08 12:30:34.003915: validation loss: -0.5911 
+2026-11-08 12:30:34.004852: Average global foreground Dice: [np.float32(0.9104), np.float32(0.6302), np.float32(0.7318), np.float32(0.7557), np.float32(0.6608), np.float32(0.7982), np.float32(0.6917), np.float32(0.5791), np.float32(0.0), np.float32(0.3622), np.float32(0.577), np.float32(0.6417), np.float32(0.5757), np.float32(0.4695), np.float32(0.0), np.float32(0.0), np.float32(0.7246), np.float32(0.5854), np.float32(0.707), np.float32(0.6529), np.float32(0.5648), np.float32(0.6045), np.float32(0.7802), np.float32(0.7526), np.float32(0.5206), np.float32(0.4911), np.float32(0.0), np.float32(0.0), np.float32(0.3421), np.float32(0.742), np.float32(0.0), np.float32(0.0), np.float32(0.4846), np.float32(0.642), np.float32(0.7819), np.float32(0.8008)] 
+2026-11-08 12:30:34.005085: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:30:34.309919: lr: 0.000783 
+2026-11-08 12:30:34.310285: This epoch took 205.797145 s
+ 
+2026-11-08 12:30:34.310557: 
+epoch:  238 
+2026-11-08 12:33:56.101488: train loss : -0.7867 
+2026-11-08 12:34:04.157409: validation loss: -0.5926 
+2026-11-08 12:34:04.158322: Average global foreground Dice: [np.float32(0.9069), np.float32(0.6129), np.float32(0.7465), np.float32(0.752), np.float32(0.6811), np.float32(0.7968), np.float32(0.7121), np.float32(0.5544), np.float32(0.0), np.float32(0.2885), np.float32(0.5169), np.float32(0.6548), np.float32(0.6042), np.float32(0.5005), np.float32(0.0), np.float32(0.0), np.float32(0.7309), np.float32(0.5662), np.float32(0.7148), np.float32(0.6441), np.float32(0.5151), np.float32(0.5816), np.float32(0.8217), np.float32(0.792), np.float32(0.48), np.float32(0.4903), np.float32(0.0), np.float32(0.0), np.float32(0.2921), np.float32(0.718), np.float32(0.0), np.float32(0.0), np.float32(0.6299), np.float32(0.6434), np.float32(0.7941), np.float32(0.8142)] 
+2026-11-08 12:34:04.158592: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:34:04.477865: lr: 0.000782 
+2026-11-08 12:34:04.478101: This epoch took 210.167324 s
+ 
+2026-11-08 12:34:04.478183: 
+epoch:  239 
+2026-11-08 12:37:07.479262: train loss : -0.7902 
+2026-11-08 12:37:15.548097: validation loss: -0.5940 
+2026-11-08 12:37:15.549081: Average global foreground Dice: [np.float32(0.9131), np.float32(0.6296), np.float32(0.739), np.float32(0.7326), np.float32(0.6507), np.float32(0.8007), np.float32(0.7083), np.float32(0.6447), np.float32(0.0), np.float32(0.3591), np.float32(0.5305), np.float32(0.6231), np.float32(0.608), np.float32(0.5641), np.float32(0.0), np.float32(0.0), np.float32(0.7267), np.float32(0.5738), np.float32(0.6977), np.float32(0.6139), np.float32(0.5715), np.float32(0.6203), np.float32(0.8057), np.float32(0.7862), np.float32(0.4599), np.float32(0.4761), np.float32(0.0), np.float32(0.0), np.float32(0.2363), np.float32(0.7133), np.float32(0.0), np.float32(0.0), np.float32(0.6044), np.float32(0.5876), np.float32(0.7753), np.float32(0.8064)] 
+2026-11-08 12:37:15.549269: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:37:15.861133: lr: 0.000781 
+2026-11-08 12:37:15.861499: This epoch took 191.383237 s
+ 
+2026-11-08 12:37:15.861737: 
+epoch:  240 
+2026-11-08 12:40:22.595057: train loss : -0.7839 
+2026-11-08 12:40:30.685205: validation loss: -0.5361 
+2026-11-08 12:40:30.686049: Average global foreground Dice: [np.float32(0.9031), np.float32(0.556), np.float32(0.6081), np.float32(0.6084), np.float32(0.5573), np.float32(0.6704), np.float32(0.5672), np.float32(0.5449), np.float32(0.0), np.float32(0.3869), np.float32(0.4825), np.float32(0.5789), np.float32(0.6446), np.float32(0.5245), np.float32(0.0), np.float32(0.0), np.float32(0.5241), np.float32(0.4728), np.float32(0.5434), np.float32(0.5314), np.float32(0.4703), np.float32(0.4203), np.float32(0.7334), np.float32(0.6955), np.float32(0.3029), np.float32(0.3221), np.float32(0.0), np.float32(0.0), np.float32(0.1516), np.float32(0.6263), np.float32(0.0), np.float32(0.0), np.float32(0.5361), np.float32(0.5768), np.float32(0.6707), np.float32(0.7086)] 
+2026-11-08 12:40:30.686288: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:40:30.997385: lr: 0.00078 
+2026-11-08 12:40:30.997598: This epoch took 195.135674 s
+ 
+2026-11-08 12:40:30.997675: 
+epoch:  241 
+2026-11-08 12:44:04.527006: train loss : -0.7683 
+2026-11-08 12:44:12.610352: validation loss: -0.5935 
+2026-11-08 12:44:12.611375: Average global foreground Dice: [np.float32(0.9043), np.float32(0.6077), np.float32(0.746), np.float32(0.7737), np.float32(0.6611), np.float32(0.809), np.float32(0.7204), np.float32(0.6205), np.float32(0.0), np.float32(0.5498), np.float32(0.5415), np.float32(0.6352), np.float32(0.6004), np.float32(0.5148), np.float32(0.0), np.float32(0.0), np.float32(0.6737), np.float32(0.6034), np.float32(0.688), np.float32(0.6075), np.float32(0.4749), np.float32(0.5947), np.float32(0.7886), np.float32(0.7731), np.float32(0.4596), np.float32(0.3947), np.float32(0.0), np.float32(0.0), np.float32(0.3169), np.float32(0.6557), np.float32(0.0), np.float32(0.0), np.float32(0.5684), np.float32(0.5776), np.float32(0.7849), np.float32(0.8129)] 
+2026-11-08 12:44:12.611556: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:44:12.922024: lr: 0.000779 
+2026-11-08 12:44:12.922418: This epoch took 221.924694 s
+ 
+2026-11-08 12:44:12.922721: 
+epoch:  242 
+2026-11-08 12:47:28.064813: train loss : -0.7751 
+2026-11-08 12:47:36.146805: validation loss: -0.5751 
+2026-11-08 12:47:36.147652: Average global foreground Dice: [np.float32(0.9138), np.float32(0.6223), np.float32(0.7358), np.float32(0.7561), np.float32(0.6558), np.float32(0.7978), np.float32(0.6982), np.float32(0.4989), np.float32(0.1465), np.float32(0.2947), np.float32(0.5796), np.float32(0.6534), np.float32(0.6025), np.float32(0.4861), np.float32(0.0), np.float32(0.0), np.float32(0.7012), np.float32(0.5751), np.float32(0.6975), np.float32(0.6376), np.float32(0.5277), np.float32(0.6342), np.float32(0.7279), np.float32(0.7095), np.float32(0.4899), np.float32(0.3739), np.float32(0.0), np.float32(0.0), np.float32(0.2247), np.float32(0.6826), np.float32(0.0), np.float32(0.0), np.float32(0.484), np.float32(0.5403), np.float32(0.7762), np.float32(0.7942)] 
+2026-11-08 12:47:36.147812: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:47:36.460096: lr: 0.000778 
+2026-11-08 12:47:36.460518: This epoch took 203.537562 s
+ 
+2026-11-08 12:47:36.460898: 
+epoch:  243 
+2026-11-08 12:50:57.650346: train loss : -0.7797 
+2026-11-08 12:51:05.720583: validation loss: -0.5992 
+2026-11-08 12:51:05.721486: Average global foreground Dice: [np.float32(0.9113), np.float32(0.666), np.float32(0.7458), np.float32(0.7833), np.float32(0.6911), np.float32(0.8204), np.float32(0.7133), np.float32(0.7258), np.float32(0.2011), np.float32(0.4709), np.float32(0.6403), np.float32(0.6933), np.float32(0.5871), np.float32(0.5128), np.float32(0.0), np.float32(0.0), np.float32(0.7277), np.float32(0.5869), np.float32(0.7241), np.float32(0.6304), np.float32(0.5026), np.float32(0.5649), np.float32(0.7947), np.float32(0.7706), np.float32(0.4869), np.float32(0.4451), np.float32(0.0), np.float32(0.0), np.float32(0.2247), np.float32(0.7378), np.float32(0.0), np.float32(0.0), np.float32(0.5), np.float32(0.6434), np.float32(0.8051), np.float32(0.8225)] 
+2026-11-08 12:51:05.721696: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:51:06.030323: lr: 0.000777 
+2026-11-08 12:51:06.030585: This epoch took 209.569463 s
+ 
+2026-11-08 12:51:06.030678: 
+epoch:  244 
+2026-11-08 12:54:22.801482: train loss : -0.7896 
+2026-11-08 12:54:30.888410: validation loss: -0.5897 
+2026-11-08 12:54:30.889287: Average global foreground Dice: [np.float32(0.9078), np.float32(0.6346), np.float32(0.7295), np.float32(0.7125), np.float32(0.6567), np.float32(0.7727), np.float32(0.6867), np.float32(0.5088), np.float32(0.0773), np.float32(0.4229), np.float32(0.5618), np.float32(0.6646), np.float32(0.6024), np.float32(0.5525), np.float32(0.0), np.float32(0.0), np.float32(0.6734), np.float32(0.5857), np.float32(0.6689), np.float32(0.6203), np.float32(0.4862), np.float32(0.5396), np.float32(0.8075), np.float32(0.7772), np.float32(0.4244), np.float32(0.4438), np.float32(0.0), np.float32(0.0), np.float32(0.1957), np.float32(0.6998), np.float32(0.0), np.float32(0.0), np.float32(0.5509), np.float32(0.5966), np.float32(0.7733), np.float32(0.8009)] 
+2026-11-08 12:54:30.889536: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:54:31.205660: lr: 0.000777 
+2026-11-08 12:54:31.206023: This epoch took 205.175277 s
+ 
+2026-11-08 12:54:31.206315: 
+epoch:  245 
+2026-11-08 12:57:49.106233: train loss : -0.7849 
+2026-11-08 12:57:57.198600: validation loss: -0.5875 
+2026-11-08 12:57:57.199579: Average global foreground Dice: [np.float32(0.9175), np.float32(0.6338), np.float32(0.7477), np.float32(0.7798), np.float32(0.6848), np.float32(0.8129), np.float32(0.711), np.float32(0.6004), np.float32(0.1241), np.float32(0.5051), np.float32(0.5488), np.float32(0.6696), np.float32(0.591), np.float32(0.5096), np.float32(0.0), np.float32(0.0), np.float32(0.7036), np.float32(0.5802), np.float32(0.7059), np.float32(0.6264), np.float32(0.5011), np.float32(0.6461), np.float32(0.8034), np.float32(0.7684), np.float32(0.5086), np.float32(0.4086), np.float32(0.0), np.float32(0.0), np.float32(0.3222), np.float32(0.6344), np.float32(0.0), np.float32(0.0), np.float32(0.5964), np.float32(0.5754), np.float32(0.8019), np.float32(0.8161)] 
+2026-11-08 12:57:57.199887: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 12:57:57.526768: lr: 0.000776 
+2026-11-08 12:57:57.553019: saving checkpoint... 
+2026-11-08 12:57:58.075036: done, saving took 0.55 seconds 
+2026-11-08 12:57:58.077046: This epoch took 206.870506 s
+ 
+2026-11-08 12:57:58.077173: 
+epoch:  246 
+2026-11-08 13:01:26.491482: train loss : -0.7545 
+2026-11-08 13:01:34.595106: validation loss: -0.5757 
+2026-11-08 13:01:34.596056: Average global foreground Dice: [np.float32(0.9186), np.float32(0.6038), np.float32(0.7276), np.float32(0.7537), np.float32(0.6524), np.float32(0.8062), np.float32(0.6956), np.float32(0.575), np.float32(0.0496), np.float32(0.4949), np.float32(0.4969), np.float32(0.5932), np.float32(0.5822), np.float32(0.4739), np.float32(0.0), np.float32(0.0), np.float32(0.7229), np.float32(0.5763), np.float32(0.7173), np.float32(0.6328), np.float32(0.5031), np.float32(0.6041), np.float32(0.8115), np.float32(0.7983), np.float32(0.5111), np.float32(0.2383), np.float32(0.0), np.float32(0.0), np.float32(0.3219), np.float32(0.6343), np.float32(0.0), np.float32(0.0), np.float32(0.6344), np.float32(0.6347), np.float32(0.7848), np.float32(0.7994)] 
+2026-11-08 13:01:34.596779: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:01:34.969292: lr: 0.000775 
+2026-11-08 13:01:34.993970: saving checkpoint... 
+2026-11-08 13:01:35.575736: done, saving took 0.61 seconds 
+2026-11-08 13:01:35.578579: This epoch took 217.501291 s
+ 
+2026-11-08 13:01:35.578759: 
+epoch:  247 
+2026-11-08 13:05:44.140059: train loss : -0.7593 
+2026-11-08 13:05:52.241399: validation loss: -0.4326 
+2026-11-08 13:05:52.242406: Average global foreground Dice: [np.float32(0.8889), np.float32(0.4283), np.float32(0.4819), np.float32(0.468), np.float32(0.3949), np.float32(0.5308), np.float32(0.4494), np.float32(0.3096), np.float32(0.0216), np.float32(0.4469), np.float32(0.3635), np.float32(0.4978), np.float32(0.5559), np.float32(0.3646), np.float32(0.0), np.float32(0.0), np.float32(0.4298), np.float32(0.3773), np.float32(0.4514), np.float32(0.4413), np.float32(0.3672), np.float32(0.3303), np.float32(0.6342), np.float32(0.5688), np.float32(0.2364), np.float32(0.2636), np.float32(0.0), np.float32(0.0), np.float32(0.1134), np.float32(0.6161), np.float32(0.0), np.float32(0.0), np.float32(0.3866), np.float32(0.4707), np.float32(0.5029), np.float32(0.5239)] 
+2026-11-08 13:05:52.242588: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:05:52.557301: lr: 0.000774 
+2026-11-08 13:05:52.557549: This epoch took 256.978700 s
+ 
+2026-11-08 13:05:52.557628: 
+epoch:  248 
+2026-11-08 13:09:58.271067: train loss : -0.7808 
+2026-11-08 13:10:06.375757: validation loss: -0.5911 
+2026-11-08 13:10:06.376825: Average global foreground Dice: [np.float32(0.9224), np.float32(0.6249), np.float32(0.7723), np.float32(0.8011), np.float32(0.7312), np.float32(0.8325), np.float32(0.7394), np.float32(0.4247), np.float32(0.2503), np.float32(0.3082), np.float32(0.5758), np.float32(0.677), np.float32(0.6262), np.float32(0.5101), np.float32(0.0), np.float32(0.0), np.float32(0.7593), np.float32(0.6099), np.float32(0.7491), np.float32(0.6388), np.float32(0.4504), np.float32(0.59), np.float32(0.8754), np.float32(0.8618), np.float32(0.4819), np.float32(0.5163), np.float32(0.0), np.float32(0.0), np.float32(0.3409), np.float32(0.6985), np.float32(0.0), np.float32(0.0), np.float32(0.6763), np.float32(0.6868), np.float32(0.8173), np.float32(0.8376)] 
+2026-11-08 13:10:06.377007: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:10:06.694623: lr: 0.000773 
+2026-11-08 13:10:06.694952: This epoch took 254.137263 s
+ 
+2026-11-08 13:10:06.695160: 
+epoch:  249 
+2026-11-08 13:14:13.408612: train loss : -0.7669 
+2026-11-08 13:14:21.497257: validation loss: -0.3358 
+2026-11-08 13:14:21.498289: Average global foreground Dice: [np.float32(0.9016), np.float32(0.3066), np.float32(0.2731), np.float32(0.2345), np.float32(0.2905), np.float32(0.3283), np.float32(0.357), np.float32(0.3052), np.float32(0.0592), np.float32(0.3945), np.float32(0.2071), np.float32(0.3692), np.float32(0.4727), np.float32(0.4227), np.float32(0.0), np.float32(0.0312), np.float32(0.2746), np.float32(0.291), np.float32(0.3119), np.float32(0.271), np.float32(0.1478), np.float32(0.0686), np.float32(0.4668), np.float32(0.3935), np.float32(0.1566), np.float32(0.2389), np.float32(0.0), np.float32(0.0), np.float32(0.0485), np.float32(0.1971), np.float32(0.0), np.float32(0.0), np.float32(0.2582), np.float32(0.3045), np.float32(0.2451), np.float32(0.34)] 
+2026-11-08 13:14:21.498478: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:14:21.819882: lr: 0.000772 
+2026-11-08 13:14:21.820257: saving scheduled checkpoint file... 
+2026-11-08 13:14:21.845670: saving checkpoint... 
+2026-11-08 13:14:24.187883: done, saving took 2.37 seconds 
+2026-11-08 13:14:24.190691: done 
+2026-11-08 13:14:24.190899: This epoch took 257.495649 s
+ 
+2026-11-08 13:14:24.191043: 
+epoch:  250 
+2026-11-08 13:18:36.061359: train loss : -0.7580 
+2026-11-08 13:18:44.165776: validation loss: -0.3593 
+2026-11-08 13:18:44.166679: Average global foreground Dice: [np.float32(0.8869), np.float32(0.3207), np.float32(0.3499), np.float32(0.3497), np.float32(0.3867), np.float32(0.3551), np.float32(0.3658), np.float32(0.2994), np.float32(0.1848), np.float32(0.2457), np.float32(0.1971), np.float32(0.3116), np.float32(0.4523), np.float32(0.2929), np.float32(0.0), np.float32(0.0), np.float32(0.4004), np.float32(0.3067), np.float32(0.3846), np.float32(0.3682), np.float32(0.1796), np.float32(0.3013), np.float32(0.4355), np.float32(0.3602), np.float32(0.2307), np.float32(0.232), np.float32(0.0), np.float32(0.0), np.float32(0.042), np.float32(0.4128), np.float32(0.0), np.float32(0.0), np.float32(0.3577), np.float32(0.361), np.float32(0.3454), np.float32(0.3559)] 
+2026-11-08 13:18:44.166830: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:18:44.490849: lr: 0.000771 
+2026-11-08 13:18:44.491191: This epoch took 260.300030 s
+ 
+2026-11-08 13:18:44.491443: 
+epoch:  251 
+2026-11-08 13:22:57.490518: train loss : -0.7777 
+2026-11-08 13:23:05.610837: validation loss: -0.5707 
+2026-11-08 13:23:05.611749: Average global foreground Dice: [np.float32(0.9243), np.float32(0.5379), np.float32(0.7037), np.float32(0.7409), np.float32(0.642), np.float32(0.7861), np.float32(0.6958), np.float32(0.2222), np.float32(0.1464), np.float32(0.3026), np.float32(0.5455), np.float32(0.6362), np.float32(0.6182), np.float32(0.4519), np.float32(0.0), np.float32(0.0), np.float32(0.7274), np.float32(0.5491), np.float32(0.7109), np.float32(0.6287), np.float32(0.5038), np.float32(0.6756), np.float32(0.802), np.float32(0.7815), np.float32(0.5077), np.float32(0.3024), np.float32(0.0), np.float32(0.0), np.float32(0.4042), np.float32(0.6725), np.float32(0.0), np.float32(0.0), np.float32(0.5671), np.float32(0.6429), np.float32(0.776), np.float32(0.7941)] 
+2026-11-08 13:23:05.611906: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:23:05.928988: lr: 0.00077 
+2026-11-08 13:23:05.929466: This epoch took 261.437918 s
+ 
+2026-11-08 13:23:05.929807: 
+epoch:  252 
+2026-11-08 13:27:06.867136: train loss : -0.7849 
+2026-11-08 13:27:14.986605: validation loss: -0.5998 
+2026-11-08 13:27:14.987720: Average global foreground Dice: [np.float32(0.9197), np.float32(0.6128), np.float32(0.7271), np.float32(0.7638), np.float32(0.6806), np.float32(0.8155), np.float32(0.7018), np.float32(0.5369), np.float32(0.205), np.float32(0.4005), np.float32(0.5904), np.float32(0.6696), np.float32(0.6181), np.float32(0.5254), np.float32(0.0), np.float32(0.0), np.float32(0.742), np.float32(0.5765), np.float32(0.7052), np.float32(0.6329), np.float32(0.5335), np.float32(0.6355), np.float32(0.7939), np.float32(0.7846), np.float32(0.4851), np.float32(0.4232), np.float32(0.0), np.float32(0.0), np.float32(0.2747), np.float32(0.6967), np.float32(0.0), np.float32(0.0), np.float32(0.6229), np.float32(0.6649), np.float32(0.7951), np.float32(0.8166)] 
+2026-11-08 13:27:14.988191: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:27:15.308959: lr: 0.000769 
+2026-11-08 13:27:15.309219: This epoch took 249.379176 s
+ 
+2026-11-08 13:27:15.309329: 
+epoch:  253 
+2026-11-08 13:31:19.651224: train loss : -0.7889 
+2026-11-08 13:31:27.766133: validation loss: -0.5823 
+2026-11-08 13:31:27.767325: Average global foreground Dice: [np.float32(0.9048), np.float32(0.5671), np.float32(0.7329), np.float32(0.7538), np.float32(0.6716), np.float32(0.7856), np.float32(0.7206), np.float32(0.4244), np.float32(0.3981), np.float32(0.2335), np.float32(0.5061), np.float32(0.6752), np.float32(0.617), np.float32(0.4495), np.float32(0.0), np.float32(0.0), np.float32(0.749), np.float32(0.5859), np.float32(0.7363), np.float32(0.638), np.float32(0.5223), np.float32(0.621), np.float32(0.8081), np.float32(0.7892), np.float32(0.536), np.float32(0.4585), np.float32(0.0), np.float32(0.0), np.float32(0.4012), np.float32(0.6813), np.float32(0.0), np.float32(0.0), np.float32(0.561), np.float32(0.6671), np.float32(0.7937), np.float32(0.8108)] 
+2026-11-08 13:31:27.767583: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:31:28.096960: lr: 0.000768 
+2026-11-08 13:31:28.097212: This epoch took 252.787809 s
+ 
+2026-11-08 13:31:28.097407: 
+epoch:  254 
+2026-11-08 13:35:31.297916: train loss : -0.7868 
+2026-11-08 13:35:39.390765: validation loss: -0.4969 
+2026-11-08 13:35:39.391799: Average global foreground Dice: [np.float32(0.9222), np.float32(0.5189), np.float32(0.5166), np.float32(0.5296), np.float32(0.4597), np.float32(0.5818), np.float32(0.4983), np.float32(0.6657), np.float32(0.2696), np.float32(0.3931), np.float32(0.4185), np.float32(0.482), np.float32(0.5527), np.float32(0.4804), np.float32(0.0), np.float32(0.0), np.float32(0.415), np.float32(0.3577), np.float32(0.437), np.float32(0.4444), np.float32(0.4161), np.float32(0.3483), np.float32(0.6257), np.float32(0.5786), np.float32(0.2345), np.float32(0.3016), np.float32(0.0), np.float32(0.0), np.float32(0.235), np.float32(0.6716), np.float32(0.0), np.float32(0.0), np.float32(0.4711), np.float32(0.541), np.float32(0.5622), np.float32(0.6044)] 
+2026-11-08 13:35:39.391955: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:35:39.712743: lr: 0.000767 
+2026-11-08 13:35:39.712973: This epoch took 251.615487 s
+ 
+2026-11-08 13:35:39.713082: 
+epoch:  255 
+2026-11-08 13:39:19.154073: train loss : -0.7809 
+2026-11-08 13:39:27.248322: validation loss: -0.5289 
+2026-11-08 13:39:27.249301: Average global foreground Dice: [np.float32(0.9214), np.float32(0.5223), np.float32(0.5607), np.float32(0.5878), np.float32(0.4959), np.float32(0.66), np.float32(0.549), np.float32(0.658), np.float32(0.218), np.float32(0.4526), np.float32(0.4957), np.float32(0.5561), np.float32(0.5901), np.float32(0.555), np.float32(0.0), np.float32(0.0), np.float32(0.4642), np.float32(0.4174), np.float32(0.4797), np.float32(0.4784), np.float32(0.2601), np.float32(0.3742), np.float32(0.6792), np.float32(0.6436), np.float32(0.2431), np.float32(0.2925), np.float32(0.0), np.float32(0.0), np.float32(0.1603), np.float32(0.6913), np.float32(0.0), np.float32(0.0), np.float32(0.4818), np.float32(0.5771), np.float32(0.6212), np.float32(0.6687)] 
+2026-11-08 13:39:27.249738: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:39:27.579552: lr: 0.000766 
+2026-11-08 13:39:27.579795: This epoch took 227.866652 s
+ 
+2026-11-08 13:39:27.579895: 
+epoch:  256 
+2026-11-08 13:43:28.333590: train loss : -0.7845 
+2026-11-08 13:43:36.456283: validation loss: -0.4970 
+2026-11-08 13:43:36.457211: Average global foreground Dice: [np.float32(0.9227), np.float32(0.5252), np.float32(0.5352), np.float32(0.5445), np.float32(0.4532), np.float32(0.6042), np.float32(0.507), np.float32(0.6361), np.float32(0.0107), np.float32(0.3945), np.float32(0.414), np.float32(0.4795), np.float32(0.5857), np.float32(0.5256), np.float32(0.0), np.float32(0.0), np.float32(0.3949), np.float32(0.3398), np.float32(0.4095), np.float32(0.4191), np.float32(0.43), np.float32(0.3137), np.float32(0.6836), np.float32(0.6465), np.float32(0.1676), np.float32(0.2502), np.float32(0.0), np.float32(0.0), np.float32(0.2921), np.float32(0.6493), np.float32(0.0), np.float32(0.0), np.float32(0.446), np.float32(0.5114), np.float32(0.588), np.float32(0.6149)] 
+2026-11-08 13:43:36.457632: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:43:36.775343: lr: 0.000765 
+2026-11-08 13:43:36.775579: This epoch took 249.195604 s
+ 
+2026-11-08 13:43:36.775657: 
+epoch:  257 
+2026-11-08 13:47:17.326981: train loss : -0.7838 
+2026-11-08 13:47:25.448118: validation loss: -0.5108 
+2026-11-08 13:47:25.449238: Average global foreground Dice: [np.float32(0.9012), np.float32(0.5296), np.float32(0.5358), np.float32(0.5591), np.float32(0.4501), np.float32(0.6351), np.float32(0.5268), np.float32(0.5548), np.float32(0.1209), np.float32(0.3597), np.float32(0.4622), np.float32(0.5586), np.float32(0.6052), np.float32(0.5369), np.float32(0.0), np.float32(0.0), np.float32(0.3938), np.float32(0.3623), np.float32(0.4198), np.float32(0.3977), np.float32(0.4577), np.float32(0.4024), np.float32(0.6553), np.float32(0.6262), np.float32(0.1962), np.float32(0.275), np.float32(0.0), np.float32(0.0), np.float32(0.2337), np.float32(0.5871), np.float32(0.0), np.float32(0.0), np.float32(0.4543), np.float32(0.493), np.float32(0.5961), np.float32(0.6389)] 
+2026-11-08 13:47:25.449450: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:47:25.775007: lr: 0.000764 
+2026-11-08 13:47:25.775462: This epoch took 228.999739 s
+ 
+2026-11-08 13:47:25.775827: 
+epoch:  258 
+2026-11-08 13:51:51.266478: train loss : -0.7849 
+2026-11-08 13:51:59.388815: validation loss: -0.4842 
+2026-11-08 13:51:59.389990: Average global foreground Dice: [np.float32(0.9133), np.float32(0.4813), np.float32(0.5327), np.float32(0.5356), np.float32(0.3937), np.float32(0.5972), np.float32(0.5151), np.float32(0.4001), np.float32(0.2408), np.float32(0.4516), np.float32(0.4419), np.float32(0.4877), np.float32(0.5603), np.float32(0.4549), np.float32(0.0), np.float32(0.0), np.float32(0.3475), np.float32(0.3052), np.float32(0.4096), np.float32(0.4011), np.float32(0.4684), np.float32(0.3956), np.float32(0.6089), np.float32(0.5968), np.float32(0.1829), np.float32(0.2394), np.float32(0.0), np.float32(0.0), np.float32(0.3355), np.float32(0.6233), np.float32(0.0), np.float32(0.0), np.float32(0.4148), np.float32(0.4571), np.float32(0.5618), np.float32(0.5994)] 
+2026-11-08 13:51:59.390340: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:51:59.726866: lr: 0.000764 
+2026-11-08 13:51:59.727164: This epoch took 273.951062 s
+ 
+2026-11-08 13:51:59.727292: 
+epoch:  259 
+2026-11-08 13:55:30.556866: train loss : -0.7917 
+2026-11-08 13:55:38.654211: validation loss: -0.5307 
+2026-11-08 13:55:38.655509: Average global foreground Dice: [np.float32(0.9122), np.float32(0.5577), np.float32(0.5579), np.float32(0.5769), np.float32(0.4846), np.float32(0.6395), np.float32(0.5297), np.float32(0.5898), np.float32(0.1754), np.float32(0.35), np.float32(0.4561), np.float32(0.548), np.float32(0.6145), np.float32(0.5888), np.float32(0.0), np.float32(0.0), np.float32(0.4274), np.float32(0.3903), np.float32(0.4506), np.float32(0.4349), np.float32(0.4448), np.float32(0.3526), np.float32(0.7424), np.float32(0.6992), np.float32(0.2259), np.float32(0.3176), np.float32(0.0), np.float32(0.0), np.float32(0.3062), np.float32(0.6872), np.float32(0.0), np.float32(0.0), np.float32(0.4549), np.float32(0.5367), np.float32(0.6303), np.float32(0.6642)] 
+2026-11-08 13:55:38.655945: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:55:38.977923: lr: 0.000763 
+2026-11-08 13:55:38.978218: This epoch took 219.250842 s
+ 
+2026-11-08 13:55:38.978417: 
+epoch:  260 
+2026-11-08 13:59:44.961431: train loss : -0.7962 
+2026-11-08 13:59:53.053486: validation loss: -0.5269 
+2026-11-08 13:59:53.054441: Average global foreground Dice: [np.float32(0.9114), np.float32(0.5523), np.float32(0.5741), np.float32(0.5879), np.float32(0.4676), np.float32(0.6492), np.float32(0.5399), np.float32(0.788), np.float32(0.1904), np.float32(0.4176), np.float32(0.4886), np.float32(0.5592), np.float32(0.606), np.float32(0.5613), np.float32(0.0), np.float32(0.0), np.float32(0.4371), np.float32(0.3694), np.float32(0.4701), np.float32(0.4456), np.float32(0.4293), np.float32(0.3534), np.float32(0.7229), np.float32(0.6939), np.float32(0.218), np.float32(0.3176), np.float32(0.0), np.float32(0.0), np.float32(0.3264), np.float32(0.6718), np.float32(0.0), np.float32(0.0), np.float32(0.4799), np.float32(0.4671), np.float32(0.6268), np.float32(0.6623)] 
+2026-11-08 13:59:53.054793: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 13:59:53.374535: lr: 0.000762 
+2026-11-08 13:59:53.374840: This epoch took 254.396348 s
+ 
+2026-11-08 13:59:53.375070: 
+epoch:  261 
+2026-11-08 14:04:05.395996: train loss : -0.7909 
+2026-11-08 14:04:13.512120: validation loss: -0.4849 
+2026-11-08 14:04:13.513073: Average global foreground Dice: [np.float32(0.9004), np.float32(0.4969), np.float32(0.5144), np.float32(0.5148), np.float32(0.4048), np.float32(0.5868), np.float32(0.4958), np.float32(0.651), np.float32(0.2113), np.float32(0.4309), np.float32(0.4305), np.float32(0.485), np.float32(0.5693), np.float32(0.5119), np.float32(0.0), np.float32(0.0), np.float32(0.3545), np.float32(0.3085), np.float32(0.395), np.float32(0.3856), np.float32(0.4185), np.float32(0.3616), np.float32(0.6938), np.float32(0.6549), np.float32(0.1972), np.float32(0.2154), np.float32(0.0), np.float32(0.0), np.float32(0.22), np.float32(0.6134), np.float32(0.0), np.float32(0.0), np.float32(0.4073), np.float32(0.3606), np.float32(0.5641), np.float32(0.6031)] 
+2026-11-08 14:04:13.513258: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:04:13.838464: lr: 0.000761 
+2026-11-08 14:04:13.838738: This epoch took 260.463481 s
+ 
+2026-11-08 14:04:13.838839: 
+epoch:  262 
+2026-11-08 14:08:07.668242: train loss : -0.7944 
+2026-11-08 14:08:15.769372: validation loss: -0.4605 
+2026-11-08 14:08:15.770534: Average global foreground Dice: [np.float32(0.8914), np.float32(0.4841), np.float32(0.4748), np.float32(0.4867), np.float32(0.3968), np.float32(0.5468), np.float32(0.4499), np.float32(0.5898), np.float32(0.3645), np.float32(0.538), np.float32(0.3993), np.float32(0.4976), np.float32(0.537), np.float32(0.4778), np.float32(0.0), np.float32(0.0), np.float32(0.3403), np.float32(0.2809), np.float32(0.3786), np.float32(0.3623), np.float32(0.3597), np.float32(0.2878), np.float32(0.6421), np.float32(0.5785), np.float32(0.1821), np.float32(0.2365), np.float32(0.0), np.float32(0.0), np.float32(0.3681), np.float32(0.6078), np.float32(0.0), np.float32(0.0), np.float32(0.409), np.float32(0.4607), np.float32(0.5215), np.float32(0.5678)] 
+2026-11-08 14:08:15.770961: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:08:16.092356: lr: 0.00076 
+2026-11-08 14:08:16.092661: This epoch took 242.253752 s
+ 
+2026-11-08 14:08:16.093124: 
+epoch:  263 
+2026-11-08 14:12:29.437338: train loss : -0.7811 
+2026-11-08 14:12:37.533603: validation loss: -0.4352 
+2026-11-08 14:12:37.534591: Average global foreground Dice: [np.float32(0.9068), np.float32(0.4286), np.float32(0.4308), np.float32(0.4136), np.float32(0.3894), np.float32(0.4541), np.float32(0.4337), np.float32(0.5406), np.float32(0.4672), np.float32(0.3093), np.float32(0.3421), np.float32(0.4614), np.float32(0.494), np.float32(0.4298), np.float32(0.0), np.float32(0.0), np.float32(0.3281), np.float32(0.3242), np.float32(0.3733), np.float32(0.3447), np.float32(0.316), np.float32(0.149), np.float32(0.6125), np.float32(0.5954), np.float32(0.1561), np.float32(0.1626), np.float32(0.0), np.float32(0.0), np.float32(0.1656), np.float32(0.6445), np.float32(0.0), np.float32(0.0), np.float32(0.4038), np.float32(0.422), np.float32(0.4829), np.float32(0.5499)] 
+2026-11-08 14:12:37.534816: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:12:37.862526: lr: 0.000759 
+2026-11-08 14:12:37.862782: This epoch took 261.769435 s
+ 
+2026-11-08 14:12:37.862870: 
+epoch:  264 
+2026-11-08 14:17:01.450676: train loss : -0.7633 
+2026-11-08 14:17:09.579385: validation loss: -0.5654 
+2026-11-08 14:17:09.580464: Average global foreground Dice: [np.float32(0.9041), np.float32(0.616), np.float32(0.6469), np.float32(0.6221), np.float32(0.5477), np.float32(0.6635), np.float32(0.5862), np.float32(0.5093), np.float32(0.3027), np.float32(0.3522), np.float32(0.4881), np.float32(0.5202), np.float32(0.6351), np.float32(0.5674), np.float32(0.0), np.float32(0.0), np.float32(0.501), np.float32(0.4071), np.float32(0.4749), np.float32(0.4453), np.float32(0.5022), np.float32(0.5474), np.float32(0.7814), np.float32(0.735), np.float32(0.2791), np.float32(0.3167), np.float32(0.0), np.float32(0.0), np.float32(0.4189), np.float32(0.6433), np.float32(0.0), np.float32(0.0), np.float32(0.3983), np.float32(0.4059), np.float32(0.6861), np.float32(0.7036)] 
+2026-11-08 14:17:09.580830: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:17:09.899476: lr: 0.000758 
+2026-11-08 14:17:09.899861: This epoch took 272.036933 s
+ 
+2026-11-08 14:17:09.900147: 
+epoch:  265 
+2026-11-08 14:21:11.307887: train loss : -0.7911 
+2026-11-08 14:21:19.456057: validation loss: -0.6104 
+2026-11-08 14:21:19.457161: Average global foreground Dice: [np.float32(0.9091), np.float32(0.6315), np.float32(0.7276), np.float32(0.7092), np.float32(0.64), np.float32(0.7604), np.float32(0.6822), np.float32(0.4644), np.float32(0.3083), np.float32(0.3728), np.float32(0.5421), np.float32(0.5926), np.float32(0.6603), np.float32(0.5958), np.float32(0.0), np.float32(0.0), np.float32(0.6103), np.float32(0.5214), np.float32(0.5858), np.float32(0.5467), np.float32(0.545), np.float32(0.6947), np.float32(0.8581), np.float32(0.8308), np.float32(0.3909), np.float32(0.3499), np.float32(0.0), np.float32(0.0), np.float32(0.3881), np.float32(0.6445), np.float32(0.0), np.float32(0.0), np.float32(0.4923), np.float32(0.5396), np.float32(0.7702), np.float32(0.7975)] 
+2026-11-08 14:21:19.457478: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:21:19.808300: lr: 0.000757 
+2026-11-08 14:21:19.808627: This epoch took 249.908307 s
+ 
+2026-11-08 14:21:19.808790: 
+epoch:  266 
+2026-11-08 14:25:12.691303: train loss : -0.7912 
+2026-11-08 14:25:20.802991: validation loss: -0.6231 
+2026-11-08 14:25:20.804240: Average global foreground Dice: [np.float32(0.9173), np.float32(0.648), np.float32(0.7658), np.float32(0.7533), np.float32(0.6757), np.float32(0.8098), np.float32(0.7417), np.float32(0.6898), np.float32(0.3093), np.float32(0.2172), np.float32(0.5815), np.float32(0.6471), np.float32(0.6455), np.float32(0.5009), np.float32(0.0), np.float32(0.0), np.float32(0.6203), np.float32(0.4892), np.float32(0.6219), np.float32(0.574), np.float32(0.5863), np.float32(0.6744), np.float32(0.8827), np.float32(0.8709), np.float32(0.4811), np.float32(0.3734), np.float32(0.0), np.float32(0.0), np.float32(0.4047), np.float32(0.6734), np.float32(0.0), np.float32(0.0), np.float32(0.5257), np.float32(0.5753), np.float32(0.8102), np.float32(0.8298)] 
+2026-11-08 14:25:20.804602: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:25:21.151231: lr: 0.000756 
+2026-11-08 14:25:21.151462: This epoch took 241.342563 s
+ 
+2026-11-08 14:25:21.151535: 
+epoch:  267 
+2026-11-08 14:29:32.651043: train loss : -0.7942 
+2026-11-08 14:29:40.757148: validation loss: -0.5710 
+2026-11-08 14:29:40.758262: Average global foreground Dice: [np.float32(0.9056), np.float32(0.5718), np.float32(0.6542), np.float32(0.6636), np.float32(0.549), np.float32(0.7234), np.float32(0.6349), np.float32(0.618), np.float32(0.2651), np.float32(0.4983), np.float32(0.5545), np.float32(0.6169), np.float32(0.6362), np.float32(0.5397), np.float32(0.0), np.float32(0.0), np.float32(0.4881), np.float32(0.3905), np.float32(0.5186), np.float32(0.4792), np.float32(0.5311), np.float32(0.5413), np.float32(0.8106), np.float32(0.782), np.float32(0.3419), np.float32(0.2978), np.float32(0.0), np.float32(0.0), np.float32(0.3535), np.float32(0.6665), np.float32(0.0), np.float32(0.0), np.float32(0.4953), np.float32(0.5087), np.float32(0.7008), np.float32(0.7313)] 
+2026-11-08 14:29:40.758632: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:29:41.090489: lr: 0.000755 
+2026-11-08 14:29:41.090725: This epoch took 259.939120 s
+ 
+2026-11-08 14:29:41.090809: 
+epoch:  268 
+2026-11-08 14:34:17.809864: train loss : -0.7816 
+2026-11-08 14:34:25.942850: validation loss: -0.4767 
+2026-11-08 14:34:25.943913: Average global foreground Dice: [np.float32(0.8983), np.float32(0.5005), np.float32(0.582), np.float32(0.4723), np.float32(0.4599), np.float32(0.5059), np.float32(0.4465), np.float32(0.377), np.float32(0.2555), np.float32(0.3407), np.float32(0.3539), np.float32(0.3674), np.float32(0.5371), np.float32(0.3973), np.float32(0.0), np.float32(0.0), np.float32(0.4263), np.float32(0.4475), np.float32(0.3828), np.float32(0.406), np.float32(0.548), np.float32(0.5779), np.float32(0.6577), np.float32(0.612), np.float32(0.2592), np.float32(0.3328), np.float32(0.0), np.float32(0.0), np.float32(0.242), np.float32(0.4765), np.float32(0.0), np.float32(0.0), np.float32(0.3714), np.float32(0.3264), np.float32(0.5331), np.float32(0.5703)] 
+2026-11-08 14:34:25.944328: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:34:26.275054: lr: 0.000754 
+2026-11-08 14:34:26.275482: This epoch took 285.184619 s
+ 
+2026-11-08 14:34:26.275772: 
+epoch:  269 
+2026-11-08 14:38:41.226821: train loss : -0.7799 
+2026-11-08 14:38:49.366748: validation loss: -0.4685 
+2026-11-08 14:38:49.367793: Average global foreground Dice: [np.float32(0.9046), np.float32(0.4745), np.float32(0.491), np.float32(0.4281), np.float32(0.4525), np.float32(0.4001), np.float32(0.4754), np.float32(0.5152), np.float32(0.051), np.float32(0.4187), np.float32(0.2616), np.float32(0.357), np.float32(0.489), np.float32(0.3702), np.float32(0.0), np.float32(0.0), np.float32(0.4875), np.float32(0.4313), np.float32(0.4973), np.float32(0.4659), np.float32(0.6048), np.float32(0.5576), np.float32(0.7027), np.float32(0.6944), np.float32(0.3116), np.float32(0.2903), np.float32(0.0), np.float32(0.0), np.float32(0.2343), np.float32(0.6186), np.float32(0.0), np.float32(0.0), np.float32(0.4191), np.float32(0.4642), np.float32(0.5242), np.float32(0.5731)] 
+2026-11-08 14:38:49.367957: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:38:49.694772: lr: 0.000753 
+2026-11-08 14:38:49.695217: This epoch took 263.419275 s
+ 
+2026-11-08 14:38:49.695550: 
+epoch:  270 
+2026-11-08 14:42:39.620707: train loss : -0.7784 
+2026-11-08 14:42:47.714977: validation loss: -0.5955 
+2026-11-08 14:42:47.716111: Average global foreground Dice: [np.float32(0.9106), np.float32(0.6136), np.float32(0.7569), np.float32(0.7757), np.float32(0.7229), np.float32(0.8135), np.float32(0.7407), np.float32(0.604), np.float32(0.1463), np.float32(0.5701), np.float32(0.6134), np.float32(0.6082), np.float32(0.6083), np.float32(0.4375), np.float32(0.0), np.float32(0.0), np.float32(0.754), np.float32(0.6007), np.float32(0.7265), np.float32(0.6316), np.float32(0.6145), np.float32(0.694), np.float32(0.8531), np.float32(0.84), np.float32(0.5489), np.float32(0.3754), np.float32(0.0), np.float32(0.0), np.float32(0.5099), np.float32(0.6498), np.float32(0.0), np.float32(0.0), np.float32(0.6193), np.float32(0.646), np.float32(0.8174), np.float32(0.8308)] 
+2026-11-08 14:42:47.716350: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:42:48.052349: lr: 0.000752 
+2026-11-08 14:42:48.052807: This epoch took 238.356998 s
+ 
+2026-11-08 14:42:48.053022: 
+epoch:  271 
+2026-11-08 14:46:53.439960: train loss : -0.7879 
+2026-11-08 14:47:01.523563: validation loss: -0.5609 
+2026-11-08 14:47:01.525028: Average global foreground Dice: [np.float32(0.917), np.float32(0.6137), np.float32(0.7138), np.float32(0.7224), np.float32(0.6432), np.float32(0.7714), np.float32(0.6726), np.float32(0.5416), np.float32(0.2694), np.float32(0.5398), np.float32(0.5378), np.float32(0.6181), np.float32(0.5608), np.float32(0.4437), np.float32(0.0), np.float32(0.0), np.float32(0.7026), np.float32(0.5572), np.float32(0.6976), np.float32(0.6097), np.float32(0.468), np.float32(0.5662), np.float32(0.7816), np.float32(0.769), np.float32(0.471), np.float32(0.4255), np.float32(0.0), np.float32(0.0), np.float32(0.1883), np.float32(0.6228), np.float32(0.0), np.float32(0.0), np.float32(0.6005), np.float32(0.5957), np.float32(0.7596), np.float32(0.7869)] 
+2026-11-08 14:47:01.525286: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:47:01.850748: lr: 0.000751 
+2026-11-08 14:47:01.851313: This epoch took 253.798191 s
+ 
+2026-11-08 14:47:01.851510: 
+epoch:  272 
+2026-11-08 14:51:08.965406: train loss : -0.7911 
+2026-11-08 14:51:17.109140: validation loss: -0.5683 
+2026-11-08 14:51:17.110149: Average global foreground Dice: [np.float32(0.9198), np.float32(0.57), np.float32(0.7253), np.float32(0.7352), np.float32(0.6638), np.float32(0.7946), np.float32(0.684), np.float32(0.6105), np.float32(0.2452), np.float32(0.375), np.float32(0.5545), np.float32(0.6159), np.float32(0.5512), np.float32(0.4627), np.float32(0.0), np.float32(0.0), np.float32(0.7198), np.float32(0.5729), np.float32(0.7153), np.float32(0.6067), np.float32(0.55), np.float32(0.6438), np.float32(0.7981), np.float32(0.7943), np.float32(0.4898), np.float32(0.4123), np.float32(0.0), np.float32(0.0), np.float32(0.296), np.float32(0.6282), np.float32(0.0), np.float32(0.0), np.float32(0.6126), np.float32(0.6701), np.float32(0.7721), np.float32(0.8035)] 
+2026-11-08 14:51:17.110325: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:51:17.448854: lr: 0.000751 
+2026-11-08 14:51:17.449245: This epoch took 255.597659 s
+ 
+2026-11-08 14:51:17.449482: 
+epoch:  273 
+2026-11-08 14:56:02.383300: train loss : -0.7883 
+2026-11-08 14:56:10.528120: validation loss: -0.5609 
+2026-11-08 14:56:10.529138: Average global foreground Dice: [np.float32(0.9017), np.float32(0.63), np.float32(0.7246), np.float32(0.7587), np.float32(0.679), np.float32(0.8095), np.float32(0.69), np.float32(0.4721), np.float32(0.2043), np.float32(0.5062), np.float32(0.6112), np.float32(0.623), np.float32(0.5912), np.float32(0.4775), np.float32(0.0), np.float32(0.0), np.float32(0.698), np.float32(0.5914), np.float32(0.6601), np.float32(0.6122), np.float32(0.5053), np.float32(0.5548), np.float32(0.7414), np.float32(0.656), np.float32(0.5162), np.float32(0.3483), np.float32(0.0), np.float32(0.0), np.float32(0.0909), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6279), np.float32(0.6452), np.float32(0.7777), np.float32(0.8075)] 
+2026-11-08 14:56:10.529543: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 14:56:10.857676: lr: 0.00075 
+2026-11-08 14:56:10.858187: This epoch took 293.408585 s
+ 
+2026-11-08 14:56:10.858546: 
+epoch:  274 
+2026-11-08 15:00:22.773308: train loss : -0.7862 
+2026-11-08 15:00:30.890301: validation loss: -0.5960 
+2026-11-08 15:00:30.891232: Average global foreground Dice: [np.float32(0.9233), np.float32(0.5906), np.float32(0.7473), np.float32(0.7933), np.float32(0.7205), np.float32(0.8416), np.float32(0.7316), np.float32(0.4375), np.float32(0.1942), np.float32(0.45), np.float32(0.6039), np.float32(0.6208), np.float32(0.576), np.float32(0.5265), np.float32(0.0), np.float32(0.0), np.float32(0.7475), np.float32(0.5974), np.float32(0.7206), np.float32(0.6232), np.float32(0.5764), np.float32(0.6366), np.float32(0.8342), np.float32(0.8354), np.float32(0.5539), np.float32(0.2738), np.float32(0.0), np.float32(0.0), np.float32(0.4265), np.float32(0.6253), np.float32(0.0), np.float32(0.0), np.float32(0.6959), np.float32(0.6123), np.float32(0.8153), np.float32(0.8376)] 
+2026-11-08 15:00:30.891384: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:00:31.211897: lr: 0.000749 
+2026-11-08 15:00:31.212144: This epoch took 260.353352 s
+ 
+2026-11-08 15:00:31.212225: 
+epoch:  275 
+2026-11-08 15:04:33.539698: train loss : -0.7765 
+2026-11-08 15:04:41.627893: validation loss: -0.6475 
+2026-11-08 15:04:41.628876: Average global foreground Dice: [np.float32(0.9135), np.float32(0.7219), np.float32(0.8068), np.float32(0.8149), np.float32(0.786), np.float32(0.8894), np.float32(0.782), np.float32(0.3896), np.float32(0.17), np.float32(0.3928), np.float32(0.6805), np.float32(0.6824), np.float32(0.6905), np.float32(0.6488), np.float32(0.0), np.float32(0.0), np.float32(0.785), np.float32(0.6051), np.float32(0.738), np.float32(0.6132), np.float32(0.5575), np.float32(0.6771), np.float32(0.9096), np.float32(0.8983), np.float32(0.5458), np.float32(0.4865), np.float32(0.0), np.float32(0.0), np.float32(0.3629), np.float32(0.6789), np.float32(0.0), np.float32(0.0), np.float32(0.3911), np.float32(0.5533), np.float32(0.8917), np.float32(0.8941)] 
+2026-11-08 15:04:41.629086: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:04:41.962672: lr: 0.000748 
+2026-11-08 15:04:41.962933: This epoch took 250.750654 s
+ 
+2026-11-08 15:04:41.963011: 
+epoch:  276 
+2026-11-08 15:08:41.193229: train loss : -0.7807 
+2026-11-08 15:08:49.311415: validation loss: -0.5694 
+2026-11-08 15:08:49.312520: Average global foreground Dice: [np.float32(0.9027), np.float32(0.6358), np.float32(0.716), np.float32(0.7226), np.float32(0.5989), np.float32(0.77), np.float32(0.6692), np.float32(0.5035), np.float32(0.2029), np.float32(0.4389), np.float32(0.627), np.float32(0.623), np.float32(0.6087), np.float32(0.5089), np.float32(0.0), np.float32(0.0), np.float32(0.7144), np.float32(0.5854), np.float32(0.6961), np.float32(0.6184), np.float32(0.482), np.float32(0.6149), np.float32(0.8132), np.float32(0.7963), np.float32(0.5371), np.float32(0.394), np.float32(0.0), np.float32(0.0), np.float32(0.3437), np.float32(0.6585), np.float32(0.0), np.float32(0.0), np.float32(0.5335), np.float32(0.5882), np.float32(0.7692), np.float32(0.7769)] 
+2026-11-08 15:08:49.312690: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:08:49.634329: lr: 0.000747 
+2026-11-08 15:08:49.634622: This epoch took 247.671555 s
+ 
+2026-11-08 15:08:49.634701: 
+epoch:  277 
+2026-11-08 15:13:19.119179: train loss : -0.7887 
+2026-11-08 15:13:27.240751: validation loss: -0.6169 
+2026-11-08 15:13:27.241919: Average global foreground Dice: [np.float32(0.9273), np.float32(0.6589), np.float32(0.7707), np.float32(0.7969), np.float32(0.7083), np.float32(0.8408), np.float32(0.725), np.float32(0.4781), np.float32(0.041), np.float32(0.3263), np.float32(0.602), np.float32(0.6459), np.float32(0.6646), np.float32(0.5909), np.float32(0.0), np.float32(0.0), np.float32(0.7563), np.float32(0.6307), np.float32(0.7315), np.float32(0.6384), np.float32(0.5013), np.float32(0.656), np.float32(0.8527), np.float32(0.8428), np.float32(0.5345), np.float32(0.3967), np.float32(0.0), np.float32(0.0), np.float32(0.3061), np.float32(0.677), np.float32(0.0), np.float32(0.0), np.float32(0.6463), np.float32(0.6673), np.float32(0.8271), np.float32(0.8443)] 
+2026-11-08 15:13:27.242135: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:13:27.565371: lr: 0.000746 
+2026-11-08 15:13:27.565610: This epoch took 277.930852 s
+ 
+2026-11-08 15:13:27.565698: 
+epoch:  278 
+2026-11-08 15:17:09.038267: train loss : -0.7846 
+2026-11-08 15:17:17.142741: validation loss: -0.5726 
+2026-11-08 15:17:17.143851: Average global foreground Dice: [np.float32(0.9057), np.float32(0.5649), np.float32(0.7243), np.float32(0.752), np.float32(0.6368), np.float32(0.7903), np.float32(0.6842), np.float32(0.2314), np.float32(0.0326), np.float32(0.4656), np.float32(0.5918), np.float32(0.6132), np.float32(0.5897), np.float32(0.4863), np.float32(0.0), np.float32(0.0), np.float32(0.7063), np.float32(0.5549), np.float32(0.6989), np.float32(0.5997), np.float32(0.4863), np.float32(0.5864), np.float32(0.8166), np.float32(0.7984), np.float32(0.5415), np.float32(0.3107), np.float32(0.0), np.float32(0.0), np.float32(0.2663), np.float32(0.5567), np.float32(0.0), np.float32(0.0), np.float32(0.577), np.float32(0.614), np.float32(0.7753), np.float32(0.798)] 
+2026-11-08 15:17:17.144035: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:17:17.467546: lr: 0.000745 
+2026-11-08 15:17:17.467764: This epoch took 229.902001 s
+ 
+2026-11-08 15:17:17.467840: 
+epoch:  279 
+2026-11-08 15:21:28.040777: train loss : -0.7643 
+2026-11-08 15:21:36.141430: validation loss: -0.5676 
+2026-11-08 15:21:36.142636: Average global foreground Dice: [np.float32(0.918), np.float32(0.5933), np.float32(0.6748), np.float32(0.7041), np.float32(0.627), np.float32(0.7628), np.float32(0.671), np.float32(0.3026), np.float32(0.223), np.float32(0.419), np.float32(0.5765), np.float32(0.6335), np.float32(0.5719), np.float32(0.5393), np.float32(0.0), np.float32(0.0), np.float32(0.6573), np.float32(0.5145), np.float32(0.6638), np.float32(0.5635), np.float32(0.5226), np.float32(0.5491), np.float32(0.7444), np.float32(0.7384), np.float32(0.4611), np.float32(0.3643), np.float32(0.0), np.float32(0.0), np.float32(0.0152), np.float32(0.6772), np.float32(0.0), np.float32(0.0), np.float32(0.4479), np.float32(0.574), np.float32(0.7452), np.float32(0.7675)] 
+2026-11-08 15:21:36.142846: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:21:36.473418: lr: 0.000744 
+2026-11-08 15:21:36.473715: This epoch took 259.005820 s
+ 
+2026-11-08 15:21:36.473815: 
+epoch:  280 
+2026-11-08 15:25:51.162235: train loss : -0.7884 
+2026-11-08 15:25:59.286121: validation loss: -0.6035 
+2026-11-08 15:25:59.287142: Average global foreground Dice: [np.float32(0.9225), np.float32(0.6289), np.float32(0.7318), np.float32(0.7799), np.float32(0.6976), np.float32(0.8085), np.float32(0.7056), np.float32(0.5293), np.float32(0.1482), np.float32(0.4089), np.float32(0.5939), np.float32(0.6663), np.float32(0.6365), np.float32(0.5835), np.float32(0.0), np.float32(0.0), np.float32(0.7321), np.float32(0.5906), np.float32(0.7144), np.float32(0.6059), np.float32(0.5318), np.float32(0.6119), np.float32(0.8267), np.float32(0.8186), np.float32(0.4992), np.float32(0.4173), np.float32(0.0), np.float32(0.0), np.float32(0.322), np.float32(0.6883), np.float32(0.0), np.float32(0.0), np.float32(0.6416), np.float32(0.6679), np.float32(0.8017), np.float32(0.8207)] 
+2026-11-08 15:25:59.287358: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:25:59.618544: lr: 0.000743 
+2026-11-08 15:25:59.618796: This epoch took 263.144876 s
+ 
+2026-11-08 15:25:59.618872: 
+epoch:  281 
+2026-11-08 15:30:31.369978: train loss : -0.7855 
+2026-11-08 15:30:39.475697: validation loss: -0.6207 
+2026-11-08 15:30:39.476748: Average global foreground Dice: [np.float32(0.921), np.float32(0.6427), np.float32(0.7514), np.float32(0.8026), np.float32(0.7222), np.float32(0.8354), np.float32(0.7394), np.float32(0.393), np.float32(0.3174), np.float32(0.3641), np.float32(0.6392), np.float32(0.6554), np.float32(0.6295), np.float32(0.5017), np.float32(0.0), np.float32(0.0), np.float32(0.7542), np.float32(0.6153), np.float32(0.7307), np.float32(0.6468), np.float32(0.5668), np.float32(0.6629), np.float32(0.8827), np.float32(0.8838), np.float32(0.5443), np.float32(0.467), np.float32(0.0), np.float32(0.0), np.float32(0.4916), np.float32(0.7481), np.float32(0.0), np.float32(0.0), np.float32(0.5934), np.float32(0.6635), np.float32(0.8323), np.float32(0.8437)] 
+2026-11-08 15:30:39.476955: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:30:39.825793: lr: 0.000742 
+2026-11-08 15:30:39.852062: saving checkpoint... 
+2026-11-08 15:30:40.411095: done, saving took 0.58 seconds 
+2026-11-08 15:30:40.413860: This epoch took 280.794932 s
+ 
+2026-11-08 15:30:40.414142: 
+epoch:  282 
+2026-11-08 15:34:39.506574: train loss : -0.7925 
+2026-11-08 15:34:47.585213: validation loss: -0.6027 
+2026-11-08 15:34:47.586151: Average global foreground Dice: [np.float32(0.9259), np.float32(0.6048), np.float32(0.7567), np.float32(0.7904), np.float32(0.7223), np.float32(0.8189), np.float32(0.7441), np.float32(0.5047), np.float32(0.2579), np.float32(0.3896), np.float32(0.6057), np.float32(0.6434), np.float32(0.6196), np.float32(0.5198), np.float32(0.0), np.float32(0.0), np.float32(0.7554), np.float32(0.5928), np.float32(0.7383), np.float32(0.6124), np.float32(0.5504), np.float32(0.6557), np.float32(0.8582), np.float32(0.8657), np.float32(0.53), np.float32(0.4089), np.float32(0.0), np.float32(0.0), np.float32(0.3528), np.float32(0.6969), np.float32(0.0), np.float32(0.0), np.float32(0.556), np.float32(0.4979), np.float32(0.8285), np.float32(0.8349)] 
+2026-11-08 15:34:47.586387: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:34:47.918545: lr: 0.000741 
+2026-11-08 15:34:47.945404: saving checkpoint... 
+2026-11-08 15:34:48.476834: done, saving took 0.56 seconds 
+2026-11-08 15:34:48.479202: This epoch took 248.064862 s
+ 
+2026-11-08 15:34:48.479557: 
+epoch:  283 
+2026-11-08 15:38:44.285887: train loss : -0.7967 
+2026-11-08 15:38:52.384576: validation loss: -0.6226 
+2026-11-08 15:38:52.385650: Average global foreground Dice: [np.float32(0.9262), np.float32(0.6642), np.float32(0.7613), np.float32(0.818), np.float32(0.7394), np.float32(0.8544), np.float32(0.7525), np.float32(0.5735), np.float32(0.2952), np.float32(0.3912), np.float32(0.6413), np.float32(0.6714), np.float32(0.6565), np.float32(0.6424), np.float32(0.0), np.float32(0.0), np.float32(0.7546), np.float32(0.6257), np.float32(0.7473), np.float32(0.614), np.float32(0.5223), np.float32(0.657), np.float32(0.8675), np.float32(0.8608), np.float32(0.4959), np.float32(0.4601), np.float32(0.0), np.float32(0.0), np.float32(0.3363), np.float32(0.7024), np.float32(0.0), np.float32(0.0), np.float32(0.6783), np.float32(0.6222), np.float32(0.8464), np.float32(0.8613)] 
+2026-11-08 15:38:52.385807: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:38:52.708092: lr: 0.00074 
+2026-11-08 15:38:52.735344: saving checkpoint... 
+2026-11-08 15:38:53.282121: done, saving took 0.57 seconds 
+2026-11-08 15:38:53.284688: This epoch took 244.804921 s
+ 
+2026-11-08 15:38:53.284835: 
+epoch:  284 
+2026-11-08 15:42:57.014421: train loss : -0.7881 
+2026-11-08 15:43:05.116403: validation loss: -0.5957 
+2026-11-08 15:43:05.117624: Average global foreground Dice: [np.float32(0.9273), np.float32(0.6307), np.float32(0.7239), np.float32(0.7681), np.float32(0.6933), np.float32(0.8127), np.float32(0.7072), np.float32(0.5381), np.float32(0.1428), np.float32(0.4628), np.float32(0.6061), np.float32(0.6509), np.float32(0.6081), np.float32(0.5586), np.float32(0.0), np.float32(0.0), np.float32(0.7395), np.float32(0.5816), np.float32(0.7221), np.float32(0.6405), np.float32(0.5245), np.float32(0.5481), np.float32(0.81), np.float32(0.803), np.float32(0.505), np.float32(0.429), np.float32(0.0), np.float32(0.0), np.float32(0.1728), np.float32(0.5263), np.float32(0.0), np.float32(0.0), np.float32(0.5686), np.float32(0.6143), np.float32(0.7945), np.float32(0.81)] 
+2026-11-08 15:43:05.118165: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:43:05.474146: lr: 0.000739 
+2026-11-08 15:43:05.503356: saving checkpoint... 
+2026-11-08 15:43:06.044648: done, saving took 0.57 seconds 
+2026-11-08 15:43:06.047219: This epoch took 252.762106 s
+ 
+2026-11-08 15:43:06.047438: 
+epoch:  285 
+2026-11-08 15:46:51.450926: train loss : -0.7952 
+2026-11-08 15:46:59.540971: validation loss: -0.5897 
+2026-11-08 15:46:59.542140: Average global foreground Dice: [np.float32(0.9234), np.float32(0.6285), np.float32(0.7135), np.float32(0.7626), np.float32(0.6941), np.float32(0.8179), np.float32(0.7112), np.float32(0.7515), np.float32(0.2146), np.float32(0.5294), np.float32(0.6246), np.float32(0.6594), np.float32(0.6278), np.float32(0.6194), np.float32(0.0), np.float32(0.0), np.float32(0.7411), np.float32(0.5866), np.float32(0.7268), np.float32(0.6191), np.float32(0.5062), np.float32(0.5755), np.float32(0.7915), np.float32(0.7739), np.float32(0.4968), np.float32(0.2691), np.float32(0.0), np.float32(0.0), np.float32(0.2787), np.float32(0.6768), np.float32(0.0), np.float32(0.0), np.float32(0.6849), np.float32(0.6387), np.float32(0.7972), np.float32(0.8213)] 
+2026-11-08 15:46:59.542413: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:46:59.883537: lr: 0.000738 
+2026-11-08 15:46:59.909444: saving checkpoint... 
+2026-11-08 15:47:00.444378: done, saving took 0.56 seconds 
+2026-11-08 15:47:00.447519: This epoch took 234.399984 s
+ 
+2026-11-08 15:47:00.447701: 
+epoch:  286 
+2026-11-08 15:51:04.720417: train loss : -0.7933 
+2026-11-08 15:51:12.831381: validation loss: -0.5855 
+2026-11-08 15:51:12.832288: Average global foreground Dice: [np.float32(0.922), np.float32(0.6073), np.float32(0.7415), np.float32(0.7525), np.float32(0.6835), np.float32(0.8048), np.float32(0.7078), np.float32(0.6908), np.float32(0.0888), np.float32(0.5556), np.float32(0.6126), np.float32(0.6488), np.float32(0.5653), np.float32(0.5034), np.float32(0.0), np.float32(0.0), np.float32(0.7269), np.float32(0.5852), np.float32(0.7148), np.float32(0.6379), np.float32(0.5152), np.float32(0.5928), np.float32(0.776), np.float32(0.7638), np.float32(0.5005), np.float32(0.3485), np.float32(0.0), np.float32(0.0), np.float32(0.3829), np.float32(0.6564), np.float32(0.0), np.float32(0.0), np.float32(0.6231), np.float32(0.6747), np.float32(0.7828), np.float32(0.8039)] 
+2026-11-08 15:51:12.832527: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:51:13.154855: lr: 0.000738 
+2026-11-08 15:51:13.184091: saving checkpoint... 
+2026-11-08 15:51:16.061039: done, saving took 2.91 seconds 
+2026-11-08 15:51:16.063359: This epoch took 255.615565 s
+ 
+2026-11-08 15:51:16.063505: 
+epoch:  287 
+2026-11-08 15:55:06.375158: train loss : -0.7994 
+2026-11-08 15:55:14.502646: validation loss: -0.5973 
+2026-11-08 15:55:14.503601: Average global foreground Dice: [np.float32(0.9195), np.float32(0.6154), np.float32(0.75), np.float32(0.7595), np.float32(0.7117), np.float32(0.8139), np.float32(0.7185), np.float32(0.7463), np.float32(0.2368), np.float32(0.3524), np.float32(0.6114), np.float32(0.6391), np.float32(0.5693), np.float32(0.5235), np.float32(0.0), np.float32(0.0), np.float32(0.7366), np.float32(0.5919), np.float32(0.7317), np.float32(0.629), np.float32(0.5469), np.float32(0.617), np.float32(0.8335), np.float32(0.8214), np.float32(0.4861), np.float32(0.4097), np.float32(0.0), np.float32(0.0), np.float32(0.2641), np.float32(0.6925), np.float32(0.0), np.float32(0.0), np.float32(0.6489), np.float32(0.6101), np.float32(0.8075), np.float32(0.8231)] 
+2026-11-08 15:55:14.503962: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:55:14.840047: lr: 0.000737 
+2026-11-08 15:55:14.867899: saving checkpoint... 
+2026-11-08 15:55:15.424334: done, saving took 0.58 seconds 
+2026-11-08 15:55:15.426736: This epoch took 239.363146 s
+ 
+2026-11-08 15:55:15.426924: 
+epoch:  288 
+2026-11-08 15:59:32.722124: train loss : -0.7830 
+2026-11-08 15:59:40.835428: validation loss: -0.5539 
+2026-11-08 15:59:40.836445: Average global foreground Dice: [np.float32(0.9244), np.float32(0.5836), np.float32(0.7059), np.float32(0.7167), np.float32(0.6376), np.float32(0.7602), np.float32(0.6555), np.float32(0.6835), np.float32(0.2397), np.float32(0.4964), np.float32(0.5936), np.float32(0.6329), np.float32(0.5443), np.float32(0.4709), np.float32(0.0), np.float32(0.0), np.float32(0.7033), np.float32(0.571), np.float32(0.6899), np.float32(0.5975), np.float32(0.4279), np.float32(0.6177), np.float32(0.7663), np.float32(0.7642), np.float32(0.4976), np.float32(0.4067), np.float32(0.0), np.float32(0.0), np.float32(0.0591), np.float32(0.6945), np.float32(0.0), np.float32(0.0), np.float32(0.6102), np.float32(0.6601), np.float32(0.7411), np.float32(0.7725)] 
+2026-11-08 15:59:40.836697: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 15:59:41.154034: lr: 0.000736 
+2026-11-08 15:59:41.154481: This epoch took 265.727469 s
+ 
+2026-11-08 15:59:41.154636: 
+epoch:  289 
+2026-11-08 16:03:40.388528: train loss : -0.8040 
+2026-11-08 16:03:48.496423: validation loss: -0.5939 
+2026-11-08 16:03:48.497283: Average global foreground Dice: [np.float32(0.9293), np.float32(0.6441), np.float32(0.7606), np.float32(0.7822), np.float32(0.7004), np.float32(0.8296), np.float32(0.724), np.float32(0.7251), np.float32(0.113), np.float32(0.4665), np.float32(0.6594), np.float32(0.6529), np.float32(0.5591), np.float32(0.4812), np.float32(0.0), np.float32(0.0), np.float32(0.7191), np.float32(0.5744), np.float32(0.725), np.float32(0.5921), np.float32(0.4877), np.float32(0.6176), np.float32(0.81), np.float32(0.821), np.float32(0.5366), np.float32(0.4077), np.float32(0.0), np.float32(0.0), np.float32(0.184), np.float32(0.4993), np.float32(0.0), np.float32(0.0), np.float32(0.6103), np.float32(0.6865), np.float32(0.8133), np.float32(0.8327)] 
+2026-11-08 16:03:48.497600: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:03:48.822936: lr: 0.000735 
+2026-11-08 16:03:48.849248: saving checkpoint... 
+2026-11-08 16:03:49.377992: done, saving took 0.55 seconds 
+2026-11-08 16:03:49.380188: This epoch took 248.225482 s
+ 
+2026-11-08 16:03:49.380383: 
+epoch:  290 
+2026-11-08 16:08:01.273930: train loss : -0.7990 
+2026-11-08 16:08:09.401804: validation loss: -0.5921 
+2026-11-08 16:08:09.402787: Average global foreground Dice: [np.float32(0.929), np.float32(0.625), np.float32(0.7181), np.float32(0.7723), np.float32(0.6801), np.float32(0.804), np.float32(0.7023), np.float32(0.6214), np.float32(0.2331), np.float32(0.5503), np.float32(0.6219), np.float32(0.6215), np.float32(0.5722), np.float32(0.4978), np.float32(0.0), np.float32(0.0), np.float32(0.7229), np.float32(0.5442), np.float32(0.7211), np.float32(0.6435), np.float32(0.5786), np.float32(0.6409), np.float32(0.8321), np.float32(0.8398), np.float32(0.5527), np.float32(0.4289), np.float32(0.0), np.float32(0.0), np.float32(0.3099), np.float32(0.6873), np.float32(0.0), np.float32(0.0), np.float32(0.6468), np.float32(0.6599), np.float32(0.7951), np.float32(0.8103)] 
+2026-11-08 16:08:09.403087: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:08:09.721346: lr: 0.000734 
+2026-11-08 16:08:09.747658: saving checkpoint... 
+2026-11-08 16:08:10.265526: done, saving took 0.54 seconds 
+2026-11-08 16:08:10.268252: This epoch took 260.887782 s
+ 
+2026-11-08 16:08:10.268518: 
+epoch:  291 
+2026-11-08 16:12:22.075618: train loss : -0.8021 
+2026-11-08 16:12:30.197031: validation loss: -0.5881 
+2026-11-08 16:12:30.197954: Average global foreground Dice: [np.float32(0.9191), np.float32(0.6138), np.float32(0.7449), np.float32(0.7866), np.float32(0.7082), np.float32(0.8287), np.float32(0.7283), np.float32(0.6965), np.float32(0.2033), np.float32(0.5796), np.float32(0.6261), np.float32(0.6404), np.float32(0.5284), np.float32(0.4677), np.float32(0.0), np.float32(0.0), np.float32(0.7566), np.float32(0.5983), np.float32(0.7361), np.float32(0.6476), np.float32(0.555), np.float32(0.6669), np.float32(0.8138), np.float32(0.8088), np.float32(0.5202), np.float32(0.4397), np.float32(0.0), np.float32(0.0), np.float32(0.2624), np.float32(0.7243), np.float32(0.0), np.float32(0.0), np.float32(0.584), np.float32(0.5979), np.float32(0.813), np.float32(0.8267)] 
+2026-11-08 16:12:30.198108: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:12:30.527073: lr: 0.000733 
+2026-11-08 16:12:30.553462: saving checkpoint... 
+2026-11-08 16:12:31.082022: done, saving took 0.55 seconds 
+2026-11-08 16:12:31.084265: This epoch took 260.815561 s
+ 
+2026-11-08 16:12:31.084538: 
+epoch:  292 
+2026-11-08 16:16:19.727737: train loss : -0.7982 
+2026-11-08 16:16:27.827792: validation loss: -0.6025 
+2026-11-08 16:16:27.828865: Average global foreground Dice: [np.float32(0.9246), np.float32(0.6354), np.float32(0.747), np.float32(0.8), np.float32(0.7196), np.float32(0.8472), np.float32(0.7491), np.float32(0.5195), np.float32(0.1131), np.float32(0.3582), np.float32(0.6371), np.float32(0.6635), np.float32(0.5632), np.float32(0.5079), np.float32(0.0), np.float32(0.0), np.float32(0.7519), np.float32(0.6332), np.float32(0.7322), np.float32(0.6578), np.float32(0.5534), np.float32(0.6005), np.float32(0.8312), np.float32(0.819), np.float32(0.561), np.float32(0.4361), np.float32(0.0), np.float32(0.0), np.float32(0.1491), np.float32(0.7262), np.float32(0.0), np.float32(0.0), np.float32(0.6276), np.float32(0.6257), np.float32(0.8279), np.float32(0.8465)] 
+2026-11-08 16:16:27.829127: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:16:28.163743: lr: 0.000732 
+2026-11-08 16:16:28.189485: saving checkpoint... 
+2026-11-08 16:16:28.740977: done, saving took 0.58 seconds 
+2026-11-08 16:16:28.742917: This epoch took 237.657943 s
+ 
+2026-11-08 16:16:28.743196: 
+epoch:  293 
+2026-11-08 16:20:28.708537: train loss : -0.7993 
+2026-11-08 16:20:36.842228: validation loss: -0.5828 
+2026-11-08 16:20:36.843267: Average global foreground Dice: [np.float32(0.9265), np.float32(0.621), np.float32(0.7551), np.float32(0.7433), np.float32(0.6707), np.float32(0.8074), np.float32(0.7054), np.float32(0.4271), np.float32(0.1551), np.float32(0.482), np.float32(0.6327), np.float32(0.6302), np.float32(0.5768), np.float32(0.5064), np.float32(0.0), np.float32(0.0), np.float32(0.7375), np.float32(0.5985), np.float32(0.7332), np.float32(0.625), np.float32(0.5417), np.float32(0.6117), np.float32(0.8247), np.float32(0.8242), np.float32(0.5012), np.float32(0.4077), np.float32(0.0), np.float32(0.0), np.float32(0.4842), np.float32(0.6556), np.float32(0.0), np.float32(0.0), np.float32(0.6301), np.float32(0.6199), np.float32(0.7825), np.float32(0.8162)] 
+2026-11-08 16:20:36.843559: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:20:37.175746: lr: 0.000731 
+2026-11-08 16:20:37.202152: saving checkpoint... 
+2026-11-08 16:20:39.796758: done, saving took 2.62 seconds 
+2026-11-08 16:20:39.798856: This epoch took 251.055395 s
+ 
+2026-11-08 16:20:39.799070: 
+epoch:  294 
+2026-11-08 16:24:29.723256: train loss : -0.8014 
+2026-11-08 16:24:37.812057: validation loss: -0.5871 
+2026-11-08 16:24:37.812978: Average global foreground Dice: [np.float32(0.9209), np.float32(0.5798), np.float32(0.7552), np.float32(0.7668), np.float32(0.6848), np.float32(0.8211), np.float32(0.728), np.float32(0.5978), np.float32(0.099), np.float32(0.5494), np.float32(0.6275), np.float32(0.6401), np.float32(0.5173), np.float32(0.5289), np.float32(0.0), np.float32(0.0), np.float32(0.7255), np.float32(0.584), np.float32(0.7381), np.float32(0.5817), np.float32(0.4986), np.float32(0.6809), np.float32(0.8106), np.float32(0.8235), np.float32(0.5058), np.float32(0.4099), np.float32(0.0648), np.float32(0.0), np.float32(0.4684), np.float32(0.7017), np.float32(0.0), np.float32(0.0), np.float32(0.6515), np.float32(0.5916), np.float32(0.8034), np.float32(0.8246)] 
+2026-11-08 16:24:37.813141: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:24:38.153470: lr: 0.00073 
+2026-11-08 16:24:38.182081: saving checkpoint... 
+2026-11-08 16:24:38.723321: done, saving took 0.57 seconds 
+2026-11-08 16:24:38.727222: This epoch took 238.927876 s
+ 
+2026-11-08 16:24:38.727483: 
+epoch:  295 
+2026-11-08 16:28:44.084841: train loss : -0.7984 
+2026-11-08 16:28:52.195561: validation loss: -0.5728 
+2026-11-08 16:28:52.196661: Average global foreground Dice: [np.float32(0.9242), np.float32(0.6237), np.float32(0.7278), np.float32(0.7435), np.float32(0.669), np.float32(0.7856), np.float32(0.6897), np.float32(0.6536), np.float32(0.2071), np.float32(0.4789), np.float32(0.5831), np.float32(0.6286), np.float32(0.5679), np.float32(0.5107), np.float32(0.0), np.float32(0.0), np.float32(0.73), np.float32(0.5775), np.float32(0.7182), np.float32(0.5939), np.float32(0.5201), np.float32(0.6318), np.float32(0.7561), np.float32(0.7499), np.float32(0.5102), np.float32(0.4161), np.float32(0.0029), np.float32(0.0), np.float32(0.4272), np.float32(0.6924), np.float32(0.0), np.float32(0.0), np.float32(0.6166), np.float32(0.6677), np.float32(0.7729), np.float32(0.8013)] 
+2026-11-08 16:28:52.197024: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:28:52.532369: lr: 0.000729 
+2026-11-08 16:28:52.559703: saving checkpoint... 
+2026-11-08 16:28:53.113825: done, saving took 0.58 seconds 
+2026-11-08 16:28:53.116664: This epoch took 254.389009 s
+ 
+2026-11-08 16:28:53.116856: 
+epoch:  296 
+2026-11-08 16:32:30.098675: train loss : -0.8010 
+2026-11-08 16:32:38.175727: validation loss: -0.5908 
+2026-11-08 16:32:38.176883: Average global foreground Dice: [np.float32(0.9195), np.float32(0.6342), np.float32(0.7295), np.float32(0.7531), np.float32(0.6789), np.float32(0.8055), np.float32(0.6973), np.float32(0.6277), np.float32(0.3068), np.float32(0.4213), np.float32(0.6431), np.float32(0.651), np.float32(0.5437), np.float32(0.4813), np.float32(0.0), np.float32(0.0), np.float32(0.7297), np.float32(0.5723), np.float32(0.7177), np.float32(0.6172), np.float32(0.5481), np.float32(0.628), np.float32(0.8339), np.float32(0.8289), np.float32(0.5116), np.float32(0.3541), np.float32(0.0946), np.float32(0.0), np.float32(0.3333), np.float32(0.6893), np.float32(0.0), np.float32(0.0), np.float32(0.6049), np.float32(0.6742), np.float32(0.7932), np.float32(0.8143)] 
+2026-11-08 16:32:38.177339: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:32:38.498454: lr: 0.000728 
+2026-11-08 16:32:38.526233: saving checkpoint... 
+2026-11-08 16:32:39.051346: done, saving took 0.55 seconds 
+2026-11-08 16:32:39.054164: This epoch took 225.937225 s
+ 
+2026-11-08 16:32:39.054555: 
+epoch:  297 
+2026-11-08 16:36:14.085147: train loss : -0.7830 
+2026-11-08 16:36:22.192111: validation loss: -0.4813 
+2026-11-08 16:36:22.193256: Average global foreground Dice: [np.float32(0.8797), np.float32(0.5501), np.float32(0.5728), np.float32(0.5235), np.float32(0.4075), np.float32(0.5972), np.float32(0.5295), np.float32(0.6483), np.float32(0.0), np.float32(0.4192), np.float32(0.4504), np.float32(0.5088), np.float32(0.5165), np.float32(0.5158), np.float32(0.0), np.float32(0.0), np.float32(0.4616), np.float32(0.4252), np.float32(0.5188), np.float32(0.4736), np.float32(0.4707), np.float32(0.3689), np.float32(0.6522), np.float32(0.6341), np.float32(0.2611), np.float32(0.2165), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5269), np.float32(0.0), np.float32(0.0), np.float32(0.3182), np.float32(0.365), np.float32(0.5746), np.float32(0.6047)] 
+2026-11-08 16:36:22.193490: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:36:22.523340: lr: 0.000727 
+2026-11-08 16:36:22.523805: This epoch took 223.469141 s
+ 
+2026-11-08 16:36:22.524136: 
+epoch:  298 
+2026-11-08 16:40:24.250469: train loss : -0.7824 
+2026-11-08 16:40:32.341709: validation loss: -0.5279 
+2026-11-08 16:40:32.342717: Average global foreground Dice: [np.float32(0.8982), np.float32(0.5785), np.float32(0.5957), np.float32(0.5913), np.float32(0.4688), np.float32(0.6655), np.float32(0.5266), np.float32(0.5924), np.float32(0.31), np.float32(0.4826), np.float32(0.5134), np.float32(0.586), np.float32(0.588), np.float32(0.5661), np.float32(0.0), np.float32(0.0), np.float32(0.4412), np.float32(0.3956), np.float32(0.4849), np.float32(0.4366), np.float32(0.5056), np.float32(0.3218), np.float32(0.765), np.float32(0.7499), np.float32(0.2561), np.float32(0.2863), np.float32(0.0062), np.float32(0.0), np.float32(0.352), np.float32(0.6871), np.float32(0.0), np.float32(0.0), np.float32(0.4788), np.float32(0.5079), np.float32(0.6386), np.float32(0.6789)] 
+2026-11-08 16:40:32.343065: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:40:32.666407: lr: 0.000726 
+2026-11-08 16:40:32.666664: This epoch took 250.142300 s
+ 
+2026-11-08 16:40:32.666746: 
+epoch:  299 
+2026-11-08 16:44:34.462885: train loss : -0.7948 
+2026-11-08 16:44:42.591600: validation loss: -0.5110 
+2026-11-08 16:44:42.592659: Average global foreground Dice: [np.float32(0.9127), np.float32(0.5528), np.float32(0.6058), np.float32(0.6059), np.float32(0.4874), np.float32(0.6589), np.float32(0.5569), np.float32(0.4507), np.float32(0.3396), np.float32(0.4942), np.float32(0.4867), np.float32(0.5513), np.float32(0.5421), np.float32(0.4642), np.float32(0.0), np.float32(0.0), np.float32(0.4957), np.float32(0.4392), np.float32(0.5423), np.float32(0.5066), np.float32(0.4342), np.float32(0.4589), np.float32(0.7243), np.float32(0.6966), np.float32(0.3666), np.float32(0.2747), np.float32(0.0476), np.float32(0.0), np.float32(0.2348), np.float32(0.6053), np.float32(0.0), np.float32(0.0), np.float32(0.4449), np.float32(0.493), np.float32(0.6245), np.float32(0.6659)] 
+2026-11-08 16:44:42.593056: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:44:42.921793: lr: 0.000725 
+2026-11-08 16:44:42.922080: saving scheduled checkpoint file... 
+2026-11-08 16:44:42.947988: saving checkpoint... 
+2026-11-08 16:44:45.435568: done, saving took 2.51 seconds 
+2026-11-08 16:44:45.438762: done 
+2026-11-08 16:44:45.438970: This epoch took 252.772170 s
+ 
+2026-11-08 16:44:45.439049: 
+epoch:  300 
+2026-11-08 16:48:33.134221: train loss : -0.7947 
+2026-11-08 16:48:41.248115: validation loss: -0.4992 
+2026-11-08 16:48:41.249664: Average global foreground Dice: [np.float32(0.9185), np.float32(0.5627), np.float32(0.5472), np.float32(0.5342), np.float32(0.4109), np.float32(0.601), np.float32(0.4762), np.float32(0.6682), np.float32(0.1753), np.float32(0.5362), np.float32(0.4604), np.float32(0.5397), np.float32(0.5173), np.float32(0.4515), np.float32(0.0), np.float32(0.0), np.float32(0.3758), np.float32(0.3195), np.float32(0.4288), np.float32(0.4162), np.float32(0.4284), np.float32(0.2863), np.float32(0.6923), np.float32(0.6609), np.float32(0.2073), np.float32(0.2458), np.float32(0.0), np.float32(0.0), np.float32(0.4305), np.float32(0.6514), np.float32(0.0), np.float32(0.0), np.float32(0.366), np.float32(0.4588), np.float32(0.5757), np.float32(0.6101)] 
+2026-11-08 16:48:41.250292: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:48:41.574721: lr: 0.000724 
+2026-11-08 16:48:41.574999: This epoch took 236.135870 s
+ 
+2026-11-08 16:48:41.575236: 
+epoch:  301 
+2026-11-08 16:52:42.429124: train loss : -0.8028 
+2026-11-08 16:52:50.562224: validation loss: -0.6148 
+2026-11-08 16:52:50.563259: Average global foreground Dice: [np.float32(0.9278), np.float32(0.6635), np.float32(0.7822), np.float32(0.797), np.float32(0.7359), np.float32(0.8523), np.float32(0.7528), np.float32(0.5173), np.float32(0.2044), np.float32(0.5074), np.float32(0.6911), np.float32(0.6946), np.float32(0.6487), np.float32(0.5783), np.float32(0.0), np.float32(0.0), np.float32(0.7634), np.float32(0.6332), np.float32(0.7467), np.float32(0.6563), np.float32(0.5562), np.float32(0.6195), np.float32(0.857), np.float32(0.8474), np.float32(0.5132), np.float32(0.3985), np.float32(0.0672), np.float32(0.0), np.float32(0.3073), np.float32(0.646), np.float32(0.0), np.float32(0.0), np.float32(0.64), np.float32(0.6467), np.float32(0.8413), np.float32(0.8569)] 
+2026-11-08 16:52:50.563772: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:52:50.895409: lr: 0.000724 
+2026-11-08 16:52:50.895647: This epoch took 249.320285 s
+ 
+2026-11-08 16:52:50.895725: 
+epoch:  302 
+2026-11-08 16:56:33.699323: train loss : -0.7970 
+2026-11-08 16:56:41.787141: validation loss: -0.5317 
+2026-11-08 16:56:41.788015: Average global foreground Dice: [np.float32(0.9095), np.float32(0.5667), np.float32(0.7175), np.float32(0.6809), np.float32(0.6405), np.float32(0.7491), np.float32(0.6736), np.float32(0.5033), np.float32(0.2664), np.float32(0.3154), np.float32(0.5478), np.float32(0.5598), np.float32(0.4496), np.float32(0.4479), np.float32(0.0), np.float32(0.025), np.float32(0.7), np.float32(0.5718), np.float32(0.6898), np.float32(0.6008), np.float32(0.4539), np.float32(0.5944), np.float32(0.7833), np.float32(0.7798), np.float32(0.4981), np.float32(0.4202), np.float32(0.0), np.float32(0.0), np.float32(0.3251), np.float32(0.6827), np.float32(0.0), np.float32(0.0), np.float32(0.6218), np.float32(0.6203), np.float32(0.7337), np.float32(0.7757)] 
+2026-11-08 16:56:41.788149: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 16:56:42.106502: lr: 0.000723 
+2026-11-08 16:56:42.106753: This epoch took 231.210973 s
+ 
+2026-11-08 16:56:42.106989: 
+epoch:  303 
+2026-11-08 17:00:54.391885: train loss : -0.8047 
+2026-11-08 17:01:02.496348: validation loss: -0.6072 
+2026-11-08 17:01:02.497462: Average global foreground Dice: [np.float32(0.9282), np.float32(0.6369), np.float32(0.7807), np.float32(0.7725), np.float32(0.6995), np.float32(0.8284), np.float32(0.7399), np.float32(0.4432), np.float32(0.3091), np.float32(0.5119), np.float32(0.6742), np.float32(0.6727), np.float32(0.6219), np.float32(0.4819), np.float32(0.0), np.float32(0.0), np.float32(0.7518), np.float32(0.6), np.float32(0.7352), np.float32(0.6132), np.float32(0.5511), np.float32(0.6732), np.float32(0.8533), np.float32(0.8583), np.float32(0.5749), np.float32(0.3657), np.float32(0.2447), np.float32(0.0), np.float32(0.4462), np.float32(0.5998), np.float32(0.0), np.float32(0.0), np.float32(0.6166), np.float32(0.6447), np.float32(0.8226), np.float32(0.84)] 
+2026-11-08 17:01:02.497665: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 17:01:02.834051: lr: 0.000722 
+2026-11-08 17:01:02.834285: This epoch took 260.727197 s
+ 
+2026-11-08 17:01:02.834369: 
+epoch:  304 
+2026-11-08 17:05:08.137711: train loss : -0.8081 
+2026-11-08 17:05:16.288526: validation loss: -0.6027 
+2026-11-08 17:05:16.289644: Average global foreground Dice: [np.float32(0.9262), np.float32(0.6134), np.float32(0.7767), np.float32(0.7682), np.float32(0.7001), np.float32(0.8331), np.float32(0.7334), np.float32(0.55), np.float32(0.3053), np.float32(0.4785), np.float32(0.6307), np.float32(0.6559), np.float32(0.5873), np.float32(0.5387), np.float32(0.0), np.float32(0.0), np.float32(0.7568), np.float32(0.6309), np.float32(0.7362), np.float32(0.626), np.float32(0.5043), np.float32(0.664), np.float32(0.8593), np.float32(0.852), np.float32(0.5033), np.float32(0.4434), np.float32(0.183), np.float32(0.0), np.float32(0.3315), np.float32(0.6704), np.float32(0.0), np.float32(0.0), np.float32(0.6464), np.float32(0.6159), np.float32(0.8164), np.float32(0.8413)] 
+2026-11-08 17:05:16.290099: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 17:05:16.641097: lr: 0.000721 
+2026-11-08 17:05:16.641351: This epoch took 253.806924 s
+ 
+2026-11-08 17:05:16.641465: 
+epoch:  305 
+2026-11-08 17:09:26.641321: train loss : -0.7653 
+2026-11-08 17:09:34.727888: validation loss: -0.5621 
+2026-11-08 17:09:34.728862: Average global foreground Dice: [np.float32(0.9229), np.float32(0.5864), np.float32(0.7131), np.float32(0.6951), np.float32(0.6202), np.float32(0.7613), np.float32(0.6577), np.float32(0.4655), np.float32(0.3969), np.float32(0.4364), np.float32(0.5587), np.float32(0.6339), np.float32(0.5935), np.float32(0.5275), np.float32(0.0), np.float32(0.0), np.float32(0.6607), np.float32(0.5983), np.float32(0.6594), np.float32(0.5701), np.float32(0.511), np.float32(0.5545), np.float32(0.7876), np.float32(0.7809), np.float32(0.4029), np.float32(0.3528), np.float32(0.2501), np.float32(0.0), np.float32(0.4414), np.float32(0.6407), np.float32(0.0), np.float32(0.0), np.float32(0.5583), np.float32(0.5072), np.float32(0.7248), np.float32(0.7561)] 
+2026-11-08 17:09:34.729027: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 17:09:35.057897: lr: 0.00072 
+2026-11-08 17:09:35.058167: This epoch took 258.416488 s
+ 
+2026-11-08 17:09:35.058267: 
+epoch:  306 
+2026-11-08 17:13:22.625051: train loss : -0.7926 
+2026-11-08 17:13:30.720788: validation loss: -0.5222 
+2026-11-08 17:13:30.721766: Average global foreground Dice: [np.float32(0.9281), np.float32(0.5816), np.float32(0.6085), np.float32(0.5734), np.float32(0.5151), np.float32(0.6352), np.float32(0.5462), np.float32(0.5588), np.float32(0.4662), np.float32(0.4789), np.float32(0.4676), np.float32(0.5786), np.float32(0.5637), np.float32(0.5735), np.float32(0.0), np.float32(0.0), np.float32(0.4711), np.float32(0.4412), np.float32(0.5009), np.float32(0.4898), np.float32(0.4466), np.float32(0.4504), np.float32(0.7004), np.float32(0.6583), np.float32(0.2254), np.float32(0.2891), np.float32(0.0), np.float32(0.0), np.float32(0.3485), np.float32(0.64), np.float32(0.0), np.float32(0.0), np.float32(0.4763), np.float32(0.3923), np.float32(0.6204), np.float32(0.6607)] 
+2026-11-08 17:13:30.721911: (interpret this as an estimate for the Dice of the different classes. This is not exact.) 
+2026-11-08 17:13:31.043635: lr: 0.000719 
+2026-11-08 17:13:31.044122: This epoch took 235.985790 s
+ 
+2026-11-08 17:13:31.044371: 
+epoch:  307