diff --git a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/debug.json b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/debug.json index 85db40ddca0a5b527547dc0fc0d9650a6b98be51..12ada0c8864006d8ce4b2c73b9324fd647d6aac5 100644 --- a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/debug.json +++ b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_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(-0.0070340233), np.float32(-0.112925336), np.float32(-0.149424), np.float32(-0.24931183), np.float32(-0.3653992), np.float32(-0.4054248), np.float32(-0.4184123), np.float32(-0.4408323), np.float32(-0.453771), np.float32(-0.46679142), np.float32(-0.4926229), np.float32(-0.497237), np.float32(-0.5037522), np.float32(-0.5236966), np.float32(-0.5303798), np.float32(-0.5430756), np.float32(-0.5220591), np.float32(-0.54732054), np.float32(-0.55173415), np.float32(-0.54539686), np.float32(-0.5562473), np.float32(-0.569614), np.float32(-0.56031716), np.float32(-0.55873436), np.float32(-0.583387), np.float32(-0.57455236), np.float32(-0.56759095), np.float32(-0.57634044), np.float32(-0.57277), np.float32(-0.5671812), np.float32(-0.5757392), np.float32(-0.5868576), np.float32(-0.6032146), np.float32(-0.5818451), np.float32(-0.5953993), np.float32(-0.59622484), np.float32(-0.6023635), np.float32(-0.5787899), np.float32(-0.59631795), np.float32(-0.59777343), np.float32(-0.6084471), np.float32(-0.59590524), np.float32(-0.5952621), np.float32(-0.59705234), np.float32(-0.59937954), np.float32(-0.6108142), np.float32(-0.6067993), np.float32(-0.60814834), np.float32(-0.6224911), np.float32(-0.59962267), np.float32(-0.604713), np.float32(-0.6233558), np.float32(-0.60915476), np.float32(-0.60716516), np.float32(-0.6054515), np.float32(-0.60841954), np.float32(-0.61409914), np.float32(-0.61729646), np.float32(-0.624039), np.float32(-0.6156721), np.float32(-0.5952519), np.float32(-0.6133857), np.float32(-0.62473387), np.float32(-0.6275945), np.float32(-0.613734), np.float32(-0.6179021), np.float32(-0.63152903), np.float32(-0.6230633), np.float32(-0.6216848), np.float32(-0.6282575), np.float32(-0.6334524), np.float32(-0.6400968), np.float32(-0.6301526), np.float32(-0.6454364), np.float32(-0.6435081), np.float32(-0.64598316), np.float32(-0.62908745), np.float32(-0.63671166), np.float32(-0.6493793), np.float32(-0.62353706), np.float32(-0.62096846), np.float32(-0.6384387), np.float32(-0.65811276), np.float32(-0.6526504), np.float32(-0.64388406), np.float32(-0.62566626), np.float32(-0.6478479), np.float32(-0.65355676), np.float32(-0.63560194), np.float32(-0.64542437), np.float32(-0.6440785), np.float32(-0.6549733), np.float32(-0.6478916), np.float32(-0.66092473), np.float32(-0.6603601), np.float32(-0.6557561), np.float32(-0.65770054), np.float32(-0.6407093), np.float32(-0.63614523), np.float32(-0.643041), np.float32(-0.65757555), np.float32(-0.6631133), np.float32(-0.65738547), np.float32(-0.66176903), np.float32(-0.66551936), np.float32(-0.64996916), np.float32(-0.6743488), np.float32(-0.676429), np.float32(-0.66399914), np.float32(-0.64735883), np.float32(-0.6507397), np.float32(-0.6459762), np.float32(-0.6571735), np.float32(-0.66718346), np.float32(-0.6727607), np.float32(-0.6649681), np.float32(-0.66306514), np.float32(-0.65537), np.float32(-0.6784558), np.float32(-0.6786621), np.float32(-0.68335277), np.float32(-0.67020345), np.float32(-0.6617534), np.float32(-0.6638445), np.float32(-0.65342796), np.float32(-0.6810618), np.float32(-0.67128867), np.float32(-0.6593141), np.float32(-0.68332595), np.float32(-0.67636645), np.float32(-0.6799999), np.float32(-0.67537194), np.float32(-0.6700514), np.float32(-0.6828408), np.float32(-0.67794436), np.float32(-0.6719297), np.float32(-0.6800618), np.float32(-0.6841686), np.float32(-0.6898927), np.float32(-0.68613166), np.float32(-0.67369896), np.float32(-0.6735082), np.float32(-0.6717074), np.float32(-0.6758409), np.float32(-0.68632656), np.float32(-0.68161964), np.float32(-0.68469787), np.float32(-0.67211473), np.float32(-0.69289905), np.float32(-0.69227135)]", + "all_val_eval_metrics": "[np.float32(0.0011664525), np.float32(0.029798958), np.float32(0.10751925), np.float32(0.1615676), np.float32(0.18584761), np.float32(0.20465608), np.float32(0.22476462), np.float32(0.19481742), np.float32(0.23174174), np.float32(0.23604053), np.float32(0.20344113), np.float32(0.31062725), np.float32(0.18995923), np.float32(0.3097726), np.float32(0.19800368), np.float32(0.19229077), np.float32(0.19176525), np.float32(0.23728853), np.float32(0.1913711), np.float32(0.1794259), np.float32(0.2786304), np.float32(0.26555455), np.float32(0.09428714), np.float32(0.1596486), np.float32(0.24961919), np.float32(0.22431633), np.float32(0.38822305), np.float32(0.38185933), np.float32(0.24736166), np.float32(0.39289182), np.float32(0.22700326), np.float32(0.32254508), np.float32(0.28615728), np.float32(0.32509568), np.float32(0.31190765), np.float32(0.26043546), np.float32(0.15944834), np.float32(0.32492843), np.float32(0.4217823), np.float32(0.31053406), np.float32(0.23190148), np.float32(0.23828319), np.float32(0.20825638), np.float32(0.27926186), np.float32(0.26981542), np.float32(0.079810485), np.float32(0.2462118), np.float32(0.3971989), np.float32(0.27671918), np.float32(0.23514067), np.float32(0.23783977), np.float32(0.2247137), np.float32(0.28919682), np.float32(0.3561383), np.float32(0.25764725), np.float32(0.15586914), np.float32(0.20107143), np.float32(0.37710765), np.float32(0.21210715), np.float32(0.20294048), np.float32(0.23477657), np.float32(0.27131715), np.float32(0.3276398), np.float32(0.18512993), np.float32(0.11023216), np.float32(0.36166704), np.float32(0.32125852), np.float32(0.26142946), np.float32(0.23939334), np.float32(0.42732307), np.float32(0.40783), np.float32(0.23708016), np.float32(0.3485414), np.float32(0.3965518), np.float32(0.3141595), np.float32(0.38469905), np.float32(0.41976678), np.float32(0.37905142), np.float32(0.39596683), np.float32(0.40836498), np.float32(0.32638636), np.float32(0.35011303), np.float32(0.43009245), np.float32(0.42253843), np.float32(0.42221224), np.float32(0.40806997), np.float32(0.3981628), np.float32(0.28888717), np.float32(0.13885729), np.float32(0.17607224), np.float32(0.28374916), np.float32(0.1514493), np.float32(0.340153), np.float32(0.37989685), np.float32(0.355068), np.float32(0.4429344), np.float32(0.2588037), np.float32(0.3757391), np.float32(0.28273213), np.float32(0.28199303), np.float32(0.3266425), np.float32(0.3496346), np.float32(0.41317594), np.float32(0.36998555), np.float32(0.36282882), np.float32(0.24376665), np.float32(0.101161726), np.float32(0.34726354), np.float32(0.29614216), np.float32(0.2957395), np.float32(0.23718722), np.float32(0.3151909), np.float32(0.3584239), np.float32(0.32977325), np.float32(0.42857748), np.float32(0.36363065), np.float32(0.41707474), np.float32(0.4317904), np.float32(0.43115947), np.float32(0.31943572), np.float32(0.41912553), np.float32(0.3232277), np.float32(0.35386702), np.float32(0.42394257), np.float32(0.35168216), np.float32(0.34109056), np.float32(0.42523924), np.float32(0.2932022), np.float32(0.32453033), np.float32(0.2163529), np.float32(0.30713332), np.float32(0.23269309), np.float32(0.2740538), np.float32(0.18327111), np.float32(0.17305344), np.float32(0.38473746), np.float32(0.4319843), np.float32(0.26609248), np.float32(0.24322264), np.float32(0.41959116), np.float32(0.2844654), np.float32(0.30811104), np.float32(0.39612392), np.float32(0.29253542), np.float32(0.25733575), np.float32(0.4661514), np.float32(0.43742335), np.float32(0.4092793), np.float32(0.32796857), np.float32(0.3271492)]", + "all_val_losses": "[np.float32(-0.08916956), np.float32(-0.12137714), np.float32(-0.19052878), np.float32(-0.27197948), np.float32(-0.35847667), np.float32(-0.3656161), np.float32(-0.3837471), np.float32(-0.37134486), np.float32(-0.38080525), np.float32(-0.3865542), np.float32(-0.34625113), np.float32(-0.4382327), np.float32(-0.34963414), np.float32(-0.44697908), np.float32(-0.35495007), np.float32(-0.34786165), np.float32(-0.328925), np.float32(-0.38663968), np.float32(-0.34285665), np.float32(-0.34025463), np.float32(-0.43056634), np.float32(-0.42888004), np.float32(-0.24380608), np.float32(-0.31894898), np.float32(-0.41352454), np.float32(-0.36627468), np.float32(-0.51443326), np.float32(-0.5033348), np.float32(-0.4029147), np.float32(-0.5247224), np.float32(-0.38410532), np.float32(-0.47579148), np.float32(-0.4414705), np.float32(-0.48525086), np.float32(-0.46963403), np.float32(-0.41693023), np.float32(-0.30781966), np.float32(-0.46632886), np.float32(-0.55609494), np.float32(-0.45017087), np.float32(-0.37657946), np.float32(-0.3769211), np.float32(-0.35871455), np.float32(-0.41582888), np.float32(-0.42275128), np.float32(-0.23435621), np.float32(-0.4102009), np.float32(-0.52136445), np.float32(-0.4140325), np.float32(-0.3954364), np.float32(-0.39677578), np.float32(-0.36896148), np.float32(-0.43835983), np.float32(-0.49288714), np.float32(-0.40663648), np.float32(-0.29846582), np.float32(-0.35555604), np.float32(-0.5088454), np.float32(-0.36394536), np.float32(-0.35268646), np.float32(-0.39231655), np.float32(-0.4206043), np.float32(-0.46953148), np.float32(-0.3373498), np.float32(-0.25576064), np.float32(-0.50856274), np.float32(-0.47127828), np.float32(-0.4074003), np.float32(-0.39541566), np.float32(-0.560495), np.float32(-0.52555054), np.float32(-0.37941378), np.float32(-0.496115), np.float32(-0.51557064), np.float32(-0.44516554), np.float32(-0.49990037), np.float32(-0.53338265), np.float32(-0.5053932), np.float32(-0.5112909), np.float32(-0.5288888), np.float32(-0.4588317), np.float32(-0.4826864), np.float32(-0.5523021), np.float32(-0.54035044), np.float32(-0.5418796), np.float32(-0.5422781), np.float32(-0.5271104), np.float32(-0.44290447), np.float32(-0.2746993), np.float32(-0.31811422), np.float32(-0.43207818), np.float32(-0.2968474), np.float32(-0.484511), np.float32(-0.50244784), np.float32(-0.49928546), np.float32(-0.5860114), np.float32(-0.39656302), np.float32(-0.5046189), np.float32(-0.404435), np.float32(-0.4337685), np.float32(-0.47761184), np.float32(-0.49191648), np.float32(-0.53908813), np.float32(-0.49863213), np.float32(-0.5059263), np.float32(-0.3825668), np.float32(-0.23190159), np.float32(-0.49644583), np.float32(-0.44283733), np.float32(-0.4501389), np.float32(-0.37808523), np.float32(-0.46120661), np.float32(-0.49043617), np.float32(-0.4458477), np.float32(-0.55790114), np.float32(-0.4830499), np.float32(-0.531731), np.float32(-0.55577785), np.float32(-0.55736005), np.float32(-0.4518967), np.float32(-0.5539367), np.float32(-0.46596026), np.float32(-0.49285492), np.float32(-0.54964453), np.float32(-0.4895441), np.float32(-0.4870645), np.float32(-0.56225604), np.float32(-0.44263142), np.float32(-0.47941762), np.float32(-0.35377324), np.float32(-0.44927585), np.float32(-0.38186455), np.float32(-0.4111146), np.float32(-0.31940573), np.float32(-0.30479708), np.float32(-0.5189724), np.float32(-0.5632127), np.float32(-0.40663326), np.float32(-0.38516548), np.float32(-0.5373224), np.float32(-0.4343089), np.float32(-0.4555755), np.float32(-0.5264908), np.float32(-0.42551222), np.float32(-0.40095982), np.float32(-0.5992088), np.float32(-0.55658764), np.float32(-0.54091954), np.float32(-0.47537354), np.float32(-0.475381)]", "all_val_losses_tr_mode": "[]", "also_val_in_tr_mode": "False", - "amp_grad_scaler": "None", + "amp_grad_scaler": "", "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": "-0.6763729", + "best_epoch_based_on_MA_tr_loss": "148", + "best_val_eval_criterion_MA": "0.37075663", "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": "", - "dl_val": "", + "dl_tr": "", + "dl_val": "", "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_B_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_B_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_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_10_14_17_59_39.txt", + "log_file": "/marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_10_15_09_20_22.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_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0", "output_folder_base": "/marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1", "oversample_foreground_percent": "0.33", @@ -71,7 +71,7 @@ "save_latest_only": "True", "stage": "0", "threeD": "True", - "tr_gen": "", + "tr_gen": "", "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": "", + "val_gen": "", "was_initialized": "True", "weight_decay": "3e-05" } \ No newline at end of file diff --git a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_best.model b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_best.model index fb6bc9afa1a5ff3bc53e4bd3727d66ce42e4353f..7b3db630add3c7ea9b97829893d1eaf51d050b49 100644 --- a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_best.model +++ b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_best.model @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:d5e4833231da769ea890195f72985ef0a4e9dfa46f798e79288d0fd34f9a0b0d -size 127055170 +oid sha256:fb5e050c7b2086aaa4a830e11c6b6ff65d85b20b91c6902e54cbf266467c1fc6 +size 127087618 diff --git a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model index c220f229dd92a5971a842a276c55968765f825e8..7b62fe210e0e4b72094f26f357d8346b045b2066 100644 --- a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model +++ b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:d307523c8b0a5b40e46fb6ca84ce1894088fb2a5aaf5a213dc39f40b2fcb0cb5 -size 127068464 +oid sha256:25fe6d097d8cc7bb2c28d3100ecad18f5e6a974076148ab5d1a474d94cdbe95e +size 127095408 diff --git a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png index 9ce527f993c199da555307b5a3a366667ff3dd90..1e38ed1d11cdefaef2eaadfeaa90eadee16d3da0 100644 --- a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png +++ b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/progress.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:d794ee913d4c24393287e1cb5a7f9f5d4242e4003d06c066d4049acd8bdafba7 -size 590870 +oid sha256:8be1cbd14a4ab3bcbf7d82c7d653d5c650dec2db533aca646f558e85a16295ae +size 700059 diff --git a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_10_15_09_15_15.txt b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_10_15_09_15_15.txt new file mode 100644 index 0000000000000000000000000000000000000000..17e3a16537425d7efdd711ce9d105528fcac7337 --- /dev/null +++ b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_10_15_09_15_15.txt @@ -0,0 +1,9 @@ +Starting... +2026-10-15 09:15:15.737655: Creating new 5-fold cross-validation split... +2026-10-15 09:15:15.738538: Desired fold for training: 0 +2026-10-15 09:15:15.738618: This split has 20 training and 5 validation cases. +2026-10-15 09:15:16.412861: 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')]) +2026-10-15 09:15:16.413221: 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')]) +2026-10-15 09:15:17.936467: loading checkpoint /marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model train= True diff --git a/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_10_15_09_20_22.txt b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_10_15_09_20_22.txt new file mode 100644 index 0000000000000000000000000000000000000000..b3b20dab2d45e24102597561ad51b4f55874e13c --- /dev/null +++ b/MedNeXt_Base/fold_0/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/training_log_2026_10_15_09_20_22.txt @@ -0,0 +1,1687 @@ +Starting... +2026-10-15 09:20:22.612488: Using splits from existing split file: /marimo/TopBrain2026/nnUNet_preprocessed/Task001_Brain2026/splits_final.pkl +2026-10-15 09:20:22.612878: The split file contains 5 splits. +2026-10-15 09:20:22.612949: Desired fold for training: 0 +2026-10-15 09:20:22.612996: This split has 20 training and 5 validation cases. +2026-10-15 09:20:22.955559: 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')]) +2026-10-15 09:20:22.955884: 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')]) +2026-10-15 09:20:24.486312: loading checkpoint /marimo/TopBrain2026/nnUNet_results/nnUNet/3d_fullres/Task001_Brain2026/nnUNetTrainerV2_MedNeXt_B_kernel3__nnUNetPlansv2.1_trgSp_1x1x1/fold_0/model_latest.model train= True +2026-10-15 09:20:24.642566: lr: 0.000864 +2026-10-15 09:20:28.340510: Unable to plot network architecture: +2026-10-15 09:20:28.340992: No module named 'hiddenlayer' +2026-10-15 09:20:28.341311: +printing the network instead: + +2026-10-15 09:20:28.341676: 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, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(64, 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, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(64, 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, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(96, 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, 192, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(192, 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, 192, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(192, 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, 256, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(256, 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, 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)) + ) + (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, 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)) + ) + ) + (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, 512, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(512, 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, 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)) + ) + (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, 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)) + ) + ) + (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, 1024, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(1024, 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, 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)) + ) + (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, 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)) + ) + ) + (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, 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)) + (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, 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)) + ) + (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, 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)) + ) + ) + (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, 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)) + (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, 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)) + ) + (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, 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)) + ) + ) + (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, 384, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(384, 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, 192, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(192, 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, 192, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(192, 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, 128, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(128, 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, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(64, 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, 64, kernel_size=(1, 1, 1), stride=(1, 1, 1)) + (act): GELU(approximate='none') + (conv3): Conv3d(64, 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)) + ) +) +2026-10-15 09:20:28.369015: + +2026-10-15 09:20:28.369302: +epoch: 150 +2026-10-15 09:24:55.412928: train loss : -0.6938 +2026-10-15 09:25:00.842565: validation loss: -0.4459 +2026-10-15 09:25:00.843721: Average global foreground Dice: [np.float32(0.8985), np.float32(0.4463), np.float32(0.5015), np.float32(0.472), np.float32(0.3698), np.float32(0.4691), np.float32(0.3738), np.float32(0.0), np.float32(0.0), np.float32(0.5173), np.float32(0.3542), np.float32(0.4159), np.float32(0.567), np.float32(0.4723), np.float32(0.0), np.float32(0.0), np.float32(0.3864), np.float32(0.392), np.float32(0.3813), np.float32(0.3766), np.float32(0.4202), np.float32(0.4285), np.float32(0.671), np.float32(0.5863), np.float32(0.2759), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6336), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4794), np.float32(0.4925)] +2026-10-15 09:25:00.843973: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 09:25:01.241145: lr: 0.000863 +2026-10-15 09:25:01.241518: This epoch took 272.871872 s + +2026-10-15 09:25:01.241812: +epoch: 151 +2026-10-15 09:28:50.805775: train loss : -0.6980 +2026-10-15 09:28:56.190940: validation loss: -0.3751 +2026-10-15 09:28:56.192176: Average global foreground Dice: [np.float32(0.9091), np.float32(0.4007), np.float32(0.4105), np.float32(0.318), np.float32(0.3284), np.float32(0.3751), np.float32(0.4016), np.float32(0.0), np.float32(0.0), np.float32(0.4103), np.float32(0.2691), np.float32(0.2512), np.float32(0.4351), np.float32(0.3583), np.float32(0.0), np.float32(0.0), np.float32(0.4085), np.float32(0.3529), np.float32(0.3909), np.float32(0.3656), np.float32(0.304), np.float32(0.2389), np.float32(0.5334), np.float32(0.4275), np.float32(0.3396), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4408), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3138), np.float32(0.3601)] +2026-10-15 09:28:56.192589: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 09:28:56.561703: lr: 0.000862 +2026-10-15 09:28:56.562049: This epoch took 235.320138 s + +2026-10-15 09:28:56.562461: +epoch: 152 +2026-10-15 09:33:37.384174: train loss : -0.6999 +2026-10-15 09:33:42.759374: validation loss: -0.4903 +2026-10-15 09:33:42.760257: Average global foreground Dice: [np.float32(0.9079), np.float32(0.5497), np.float32(0.6527), np.float32(0.637), np.float32(0.5256), np.float32(0.6471), np.float32(0.5781), np.float32(0.0), np.float32(0.0), np.float32(0.3795), np.float32(0.4749), np.float32(0.4573), np.float32(0.4839), np.float32(0.3296), np.float32(0.0), np.float32(0.0), np.float32(0.6111), np.float32(0.4653), np.float32(0.5835), np.float32(0.5334), np.float32(0.4753), np.float32(0.5151), np.float32(0.6753), np.float32(0.6096), np.float32(0.5022), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5444), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6299), np.float32(0.6128)] +2026-10-15 09:33:42.760419: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 09:33:43.140778: lr: 0.000861 +2026-10-15 09:33:43.141083: This epoch took 286.578470 s + +2026-10-15 09:33:43.141186: +epoch: 153 +2026-10-15 09:38:15.500610: train loss : -0.6925 +2026-10-15 09:38:20.900752: validation loss: -0.5598 +2026-10-15 09:38:20.901960: Average global foreground Dice: [np.float32(0.9067), np.float32(0.6161), np.float32(0.7387), np.float32(0.7421), np.float32(0.6502), np.float32(0.7874), np.float32(0.6717), np.float32(0.0), np.float32(0.0), np.float32(0.3824), np.float32(0.5136), np.float32(0.6199), np.float32(0.5545), np.float32(0.4697), np.float32(0.0), np.float32(0.0), np.float32(0.6744), np.float32(0.5835), np.float32(0.6684), np.float32(0.5916), np.float32(0.5288), np.float32(0.5903), np.float32(0.8722), np.float32(0.8484), np.float32(0.4927), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.626), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7926), np.float32(0.8096)] +2026-10-15 09:38:20.902252: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 09:38:21.284382: lr: 0.00086 +2026-10-15 09:38:21.284693: This epoch took 278.143419 s + +2026-10-15 09:38:21.284773: +epoch: 154 +2026-10-15 09:42:25.091900: train loss : -0.6965 +2026-10-15 09:42:30.461827: validation loss: -0.5515 +2026-10-15 09:42:30.462743: Average global foreground Dice: [np.float32(0.9127), np.float32(0.5596), np.float32(0.7203), np.float32(0.7426), np.float32(0.6118), np.float32(0.7694), np.float32(0.609), np.float32(0.0), np.float32(0.0), np.float32(0.3123), np.float32(0.5403), np.float32(0.5711), np.float32(0.5736), np.float32(0.4952), np.float32(0.0), np.float32(0.0), np.float32(0.6701), np.float32(0.5497), np.float32(0.672), np.float32(0.5553), np.float32(0.5391), np.float32(0.521), np.float32(0.8184), np.float32(0.8085), np.float32(0.4441), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6619), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7773), np.float32(0.7976)] +2026-10-15 09:42:30.462928: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 09:42:30.855106: lr: 0.000859 +2026-10-15 09:42:30.855706: This epoch took 249.570879 s + +2026-10-15 09:42:30.855903: +epoch: 155 +2026-10-15 09:46:31.921699: train loss : -0.6921 +2026-10-15 09:46:37.299468: validation loss: -0.5999 +2026-10-15 09:46:37.300570: Average global foreground Dice: [np.float32(0.9149), np.float32(0.6872), np.float32(0.789), np.float32(0.7965), np.float32(0.7591), np.float32(0.8391), np.float32(0.7386), np.float32(0.0), np.float32(0.0), np.float32(0.3802), np.float32(0.5883), np.float32(0.6686), np.float32(0.611), np.float32(0.5151), np.float32(0.0), np.float32(0.0), np.float32(0.7536), np.float32(0.5646), np.float32(0.7121), np.float32(0.5657), np.float32(0.5751), np.float32(0.6469), np.float32(0.8953), np.float32(0.8852), np.float32(0.4187), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4837), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.8614), np.float32(0.8659)] +2026-10-15 09:46:37.300853: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 09:46:37.627811: lr: 0.000858 +2026-10-15 09:46:37.628057: This epoch took 246.772067 s + +2026-10-15 09:46:37.628142: +epoch: 156 +2026-10-15 09:50:42.270709: train loss : -0.6906 +2026-10-15 09:50:47.626533: validation loss: -0.6163 +2026-10-15 09:50:47.627570: Average global foreground Dice: [np.float32(0.9119), np.float32(0.6472), np.float32(0.8252), np.float32(0.8412), np.float32(0.7932), np.float32(0.8883), np.float32(0.8043), np.float32(0.0), np.float32(0.0), np.float32(0.5632), np.float32(0.6366), np.float32(0.7188), np.float32(0.6678), np.float32(0.526), np.float32(0.0), np.float32(0.0), np.float32(0.7937), np.float32(0.6096), np.float32(0.7543), np.float32(0.6694), np.float32(0.5706), np.float32(0.6922), np.float32(0.8783), np.float32(0.8552), np.float32(0.5443), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3027), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.8811), np.float32(0.8821)] +2026-10-15 09:50:47.627733: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 09:50:47.957906: lr: 0.000858 +2026-10-15 09:50:47.958197: This epoch took 250.329997 s + +2026-10-15 09:50:47.958480: +epoch: 157 +2026-10-15 09:55:32.907363: train loss : -0.6951 +2026-10-15 09:55:38.269463: validation loss: -0.5519 +2026-10-15 09:55:38.270355: Average global foreground Dice: [np.float32(0.9213), np.float32(0.5868), np.float32(0.7458), np.float32(0.7578), np.float32(0.6904), np.float32(0.7786), np.float32(0.7061), np.float32(0.0), np.float32(0.0), np.float32(0.4623), np.float32(0.578), np.float32(0.6046), np.float32(0.5916), np.float32(0.4714), np.float32(0.0), np.float32(0.0), np.float32(0.7294), np.float32(0.5803), np.float32(0.7171), np.float32(0.5922), np.float32(0.5305), np.float32(0.6301), np.float32(0.7704), np.float32(0.766), np.float32(0.5019), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4885), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7881), np.float32(0.7991)] +2026-10-15 09:55:38.270556: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 09:55:38.646388: lr: 0.000857 +2026-10-15 09:55:38.646759: This epoch took 290.688177 s + +2026-10-15 09:55:38.647052: +epoch: 158 +2026-10-15 10:00:03.924671: train loss : -0.7034 +2026-10-15 10:00:09.311004: validation loss: -0.5576 +2026-10-15 10:00:09.312212: Average global foreground Dice: [np.float32(0.9206), np.float32(0.5938), np.float32(0.6962), np.float32(0.7295), np.float32(0.6206), np.float32(0.7517), np.float32(0.6513), np.float32(0.0), np.float32(0.0), np.float32(0.4647), np.float32(0.5265), np.float32(0.6062), np.float32(0.5839), np.float32(0.4813), np.float32(0.0), np.float32(0.0), np.float32(0.684), np.float32(0.5773), np.float32(0.6666), np.float32(0.6005), np.float32(0.6014), np.float32(0.6148), np.float32(0.8276), np.float32(0.7997), np.float32(0.4766), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5143), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.753), np.float32(0.7766)] +2026-10-15 10:00:09.312607: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:00:09.786287: lr: 0.000856 +2026-10-15 10:00:09.796962: saving checkpoint... +2026-10-15 10:00:09.927428: done, saving took 0.14 seconds +2026-10-15 10:00:09.929323: This epoch took 271.282138 s + +2026-10-15 10:00:09.929485: +epoch: 159 +2026-10-15 10:04:18.348564: train loss : -0.7042 +2026-10-15 10:04:23.750313: validation loss: -0.5677 +2026-10-15 10:04:23.751590: Average global foreground Dice: [np.float32(0.9018), np.float32(0.6182), np.float32(0.7593), np.float32(0.7734), np.float32(0.694), np.float32(0.804), np.float32(0.7187), np.float32(0.0), np.float32(0.0), np.float32(0.5378), np.float32(0.5581), np.float32(0.6241), np.float32(0.6138), np.float32(0.4068), np.float32(0.0), np.float32(0.0), np.float32(0.7407), np.float32(0.58), np.float32(0.701), np.float32(0.6174), np.float32(0.5373), np.float32(0.6253), np.float32(0.8225), np.float32(0.7851), np.float32(0.4931), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4022), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.8001), np.float32(0.7987)] +2026-10-15 10:04:23.751782: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:04:24.068095: lr: 0.000855 +2026-10-15 10:04:24.077444: saving checkpoint... +2026-10-15 10:04:24.211333: done, saving took 0.14 seconds +2026-10-15 10:04:24.213369: This epoch took 254.283808 s + +2026-10-15 10:04:24.213577: +epoch: 160 +2026-10-15 10:08:30.508311: train loss : -0.6932 +2026-10-15 10:08:35.866472: validation loss: -0.5247 +2026-10-15 10:08:35.867581: Average global foreground Dice: [np.float32(0.8967), np.float32(0.5603), np.float32(0.6215), np.float32(0.6084), np.float32(0.5124), np.float32(0.6232), np.float32(0.5699), np.float32(0.0), np.float32(0.0), np.float32(0.4907), np.float32(0.4855), np.float32(0.5443), np.float32(0.5786), np.float32(0.4694), np.float32(0.0), np.float32(0.0), np.float32(0.459), np.float32(0.3692), np.float32(0.4409), np.float32(0.3975), np.float32(0.3839), np.float32(0.5342), np.float32(0.7778), np.float32(0.7136), np.float32(0.2953), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4951), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6644), np.float32(0.6717)] +2026-10-15 10:08:35.867751: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:08:36.204543: lr: 0.000854 +2026-10-15 10:08:36.205009: This epoch took 251.991284 s + +2026-10-15 10:08:36.205375: +epoch: 161 +2026-10-15 10:12:29.347573: train loss : -0.6971 +2026-10-15 10:12:34.708118: validation loss: -0.5785 +2026-10-15 10:12:34.709145: Average global foreground Dice: [np.float32(0.9168), np.float32(0.6121), np.float32(0.808), np.float32(0.8031), np.float32(0.7657), np.float32(0.8033), np.float32(0.7449), np.float32(0.0), np.float32(0.0), np.float32(0.5672), np.float32(0.5441), np.float32(0.6467), np.float32(0.6039), np.float32(0.499), np.float32(0.0), np.float32(0.0), np.float32(0.7779), np.float32(0.6268), np.float32(0.7566), np.float32(0.6658), np.float32(0.5188), np.float32(0.6794), np.float32(0.8262), np.float32(0.7892), np.float32(0.4683), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5687), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7888), np.float32(0.7879)] +2026-10-15 10:12:34.709332: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:12:35.016625: lr: 0.000853 +2026-10-15 10:12:35.048110: saving checkpoint... +2026-10-15 10:12:35.181099: done, saving took 0.16 seconds +2026-10-15 10:12:35.182688: This epoch took 238.977055 s + +2026-10-15 10:12:35.182796: +epoch: 162 +2026-10-15 10:16:48.334260: train loss : -0.6824 +2026-10-15 10:16:53.704224: validation loss: -0.4149 +2026-10-15 10:16:53.705386: Average global foreground Dice: [np.float32(0.8929), np.float32(0.4177), np.float32(0.4516), np.float32(0.4178), np.float32(0.3045), np.float32(0.412), np.float32(0.3723), np.float32(0.0), np.float32(0.0), np.float32(0.4764), np.float32(0.3098), np.float32(0.2855), np.float32(0.4187), np.float32(0.2728), np.float32(0.0), np.float32(0.0), np.float32(0.3283), np.float32(0.2796), np.float32(0.3028), np.float32(0.3105), np.float32(0.2893), np.float32(0.3519), np.float32(0.6237), np.float32(0.5462), np.float32(0.1933), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4768), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4679), np.float32(0.5056)] +2026-10-15 10:16:53.705606: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:16:54.022901: lr: 0.000852 +2026-10-15 10:16:54.023370: This epoch took 258.840517 s + +2026-10-15 10:16:54.023679: +epoch: 163 +2026-10-15 10:20:59.983810: train loss : -0.6878 +2026-10-15 10:21:05.364127: validation loss: -0.4218 +2026-10-15 10:21:05.365062: Average global foreground Dice: [np.float32(0.9147), np.float32(0.4724), np.float32(0.449), np.float32(0.463), np.float32(0.376), np.float32(0.4244), np.float32(0.3149), np.float32(0.0), np.float32(0.0), np.float32(0.36), np.float32(0.3496), np.float32(0.369), np.float32(0.5499), np.float32(0.1147), np.float32(0.0), np.float32(0.0), np.float32(0.3741), np.float32(0.3177), np.float32(0.3233), np.float32(0.3332), np.float32(0.3456), np.float32(0.3365), np.float32(0.6333), np.float32(0.5915), np.float32(0.2439), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5532), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5224), np.float32(0.4933)] +2026-10-15 10:21:05.365309: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:21:05.692219: lr: 0.000851 +2026-10-15 10:21:05.692480: This epoch took 251.668570 s + +2026-10-15 10:21:05.692571: +epoch: 164 +2026-10-15 10:25:17.490341: train loss : -0.6852 +2026-10-15 10:25:22.909553: validation loss: -0.5862 +2026-10-15 10:25:22.911085: Average global foreground Dice: [np.float32(0.904), np.float32(0.5991), np.float32(0.7851), np.float32(0.7747), np.float32(0.7223), np.float32(0.7601), np.float32(0.7205), np.float32(0.0), np.float32(0.0), np.float32(0.309), np.float32(0.6111), np.float32(0.6481), np.float32(0.5897), np.float32(0.4536), np.float32(0.0), np.float32(0.0), np.float32(0.7501), np.float32(0.5613), np.float32(0.7023), np.float32(0.6734), np.float32(0.5419), np.float32(0.6286), np.float32(0.8736), np.float32(0.8529), np.float32(0.4569), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4039), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.8279), np.float32(0.832)] +2026-10-15 10:25:22.911587: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:25:23.258509: lr: 0.00085 +2026-10-15 10:25:23.259004: This epoch took 257.566374 s + +2026-10-15 10:25:23.259282: +epoch: 165 +2026-10-15 10:29:14.446470: train loss : -0.6993 +2026-10-15 10:29:19.817528: validation loss: -0.4364 +2026-10-15 10:29:19.818496: Average global foreground Dice: [np.float32(0.9101), np.float32(0.5091), np.float32(0.4868), np.float32(0.5575), np.float32(0.4399), np.float32(0.5245), np.float32(0.3958), np.float32(0.0), np.float32(0.0), np.float32(0.2338), np.float32(0.3977), np.float32(0.3627), np.float32(0.4975), np.float32(0.3823), np.float32(0.0), np.float32(0.0), np.float32(0.4441), np.float32(0.3562), np.float32(0.3162), np.float32(0.3403), np.float32(0.3807), np.float32(0.3488), np.float32(0.6754), np.float32(0.6236), np.float32(0.3339), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4822), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6041), np.float32(0.5626)] +2026-10-15 10:29:19.818890: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:29:20.204973: lr: 0.000849 +2026-10-15 10:29:20.205200: This epoch took 236.945716 s + +2026-10-15 10:29:20.205521: +epoch: 166 +2026-10-15 10:33:30.567509: train loss : -0.6949 +2026-10-15 10:33:35.943846: validation loss: -0.5304 +2026-10-15 10:33:35.944789: Average global foreground Dice: [np.float32(0.9104), np.float32(0.5576), np.float32(0.6691), np.float32(0.676), np.float32(0.5283), np.float32(0.7016), np.float32(0.5483), np.float32(0.0), np.float32(0.0), np.float32(0.5064), np.float32(0.4787), np.float32(0.5168), np.float32(0.5245), np.float32(0.4199), np.float32(0.0), np.float32(0.0), np.float32(0.5724), np.float32(0.5115), np.float32(0.5016), np.float32(0.4452), np.float32(0.51), np.float32(0.5731), np.float32(0.8214), np.float32(0.7999), np.float32(0.4504), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5218), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7853), np.float32(0.7956)] +2026-10-15 10:33:35.945307: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:33:36.256202: lr: 0.000848 +2026-10-15 10:33:36.256613: This epoch took 256.051030 s + +2026-10-15 10:33:36.256840: +epoch: 167 +2026-10-15 10:37:40.566662: train loss : -0.6694 +2026-10-15 10:37:45.936547: validation loss: -0.5874 +2026-10-15 10:37:45.937750: Average global foreground Dice: [np.float32(0.9001), np.float32(0.6558), np.float32(0.7917), np.float32(0.8032), np.float32(0.7322), np.float32(0.8308), np.float32(0.7566), np.float32(0.0), np.float32(0.0), np.float32(0.544), np.float32(0.6108), np.float32(0.6665), np.float32(0.5951), np.float32(0.5024), np.float32(0.0), np.float32(0.0), np.float32(0.7553), np.float32(0.6008), np.float32(0.7013), np.float32(0.6495), np.float32(0.5598), np.float32(0.6497), np.float32(0.8601), np.float32(0.8211), np.float32(0.455), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0593), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.8401), np.float32(0.8487)] +2026-10-15 10:37:45.938055: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:37:46.246491: lr: 0.000847 +2026-10-15 10:37:46.246808: This epoch took 249.989825 s + +2026-10-15 10:37:46.246986: +epoch: 168 +2026-10-15 10:42:24.765892: train loss : -0.6923 +2026-10-15 10:42:30.141917: validation loss: -0.5565 +2026-10-15 10:42:30.142849: Average global foreground Dice: [np.float32(0.9033), np.float32(0.5868), np.float32(0.752), np.float32(0.7545), np.float32(0.6784), np.float32(0.7983), np.float32(0.6901), np.float32(0.0), np.float32(0.0), np.float32(0.5956), np.float32(0.5341), np.float32(0.6072), np.float32(0.5719), np.float32(0.4424), np.float32(0.0), np.float32(0.0), np.float32(0.7551), np.float32(0.5977), np.float32(0.7018), np.float32(0.6553), np.float32(0.5689), np.float32(0.6133), np.float32(0.8463), np.float32(0.8274), np.float32(0.5159), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.445), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.8087), np.float32(0.8196)] +2026-10-15 10:42:30.143080: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:42:30.513539: lr: 0.000847 +2026-10-15 10:42:30.513941: This epoch took 284.266891 s + +2026-10-15 10:42:30.514135: +epoch: 169 +2026-10-15 10:46:28.008495: train loss : -0.6753 +2026-10-15 10:46:33.386147: validation loss: -0.4892 +2026-10-15 10:46:33.387092: Average global foreground Dice: [np.float32(0.9159), np.float32(0.5295), np.float32(0.6519), np.float32(0.6536), np.float32(0.5049), np.float32(0.6884), np.float32(0.5472), np.float32(0.0), np.float32(0.0), np.float32(0.4679), np.float32(0.4816), np.float32(0.5498), np.float32(0.6001), np.float32(0.4207), np.float32(0.0), np.float32(0.0), np.float32(0.5972), np.float32(0.4218), np.float32(0.5344), np.float32(0.5239), np.float32(0.5108), np.float32(0.542), np.float32(0.6529), np.float32(0.5575), np.float32(0.4199), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5109), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6249), np.float32(0.6466)] +2026-10-15 10:46:33.387284: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:46:33.719828: lr: 0.000846 +2026-10-15 10:46:33.720313: This epoch took 243.206008 s + +2026-10-15 10:46:33.720500: +epoch: 170 +2026-10-15 10:50:46.044479: train loss : -0.6924 +2026-10-15 10:50:51.428372: validation loss: -0.5335 +2026-10-15 10:50:51.429280: Average global foreground Dice: [np.float32(0.9052), np.float32(0.5858), np.float32(0.6558), np.float32(0.6482), np.float32(0.5394), np.float32(0.6969), np.float32(0.5629), np.float32(0.0), np.float32(0.0), np.float32(0.4871), np.float32(0.4614), np.float32(0.4998), np.float32(0.5086), np.float32(0.4213), np.float32(0.0), np.float32(0.0), np.float32(0.5672), np.float32(0.5063), np.float32(0.5286), np.float32(0.5079), np.float32(0.6317), np.float32(0.6155), np.float32(0.8201), np.float32(0.7935), np.float32(0.3634), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6741), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7134), np.float32(0.7367)] +2026-10-15 10:50:51.429433: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:50:51.740501: lr: 0.000845 +2026-10-15 10:50:51.740730: This epoch took 258.020134 s + +2026-10-15 10:50:51.740807: +epoch: 171 +2026-10-15 10:55:18.487100: train loss : -0.6950 +2026-10-15 10:55:23.853805: validation loss: -0.4773 +2026-10-15 10:55:23.854764: Average global foreground Dice: [np.float32(0.9079), np.float32(0.5253), np.float32(0.5599), np.float32(0.574), np.float32(0.4772), np.float32(0.6037), np.float32(0.4955), np.float32(0.0), np.float32(0.0), np.float32(0.546), np.float32(0.4282), np.float32(0.4927), np.float32(0.5631), np.float32(0.4693), np.float32(0.0), np.float32(0.0), np.float32(0.4593), np.float32(0.3866), np.float32(0.4321), np.float32(0.4489), np.float32(0.408), np.float32(0.3086), np.float32(0.6956), np.float32(0.6486), np.float32(0.2364), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5582), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5809), np.float32(0.622)] +2026-10-15 10:55:23.854948: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:55:24.205813: lr: 0.000844 +2026-10-15 10:55:24.206065: This epoch took 272.465206 s + +2026-10-15 10:55:24.206143: +epoch: 172 +2026-10-15 10:59:37.599926: train loss : -0.6916 +2026-10-15 10:59:42.964988: validation loss: -0.5807 +2026-10-15 10:59:42.966004: Average global foreground Dice: [np.float32(0.921), np.float32(0.6477), np.float32(0.7735), np.float32(0.7688), np.float32(0.7305), np.float32(0.8227), np.float32(0.7349), np.float32(0.0), np.float32(0.0), np.float32(0.3258), np.float32(0.5155), np.float32(0.6558), np.float32(0.5668), np.float32(0.5324), np.float32(0.0), np.float32(0.0), np.float32(0.7553), np.float32(0.5846), np.float32(0.7178), np.float32(0.6515), np.float32(0.4899), np.float32(0.588), np.float32(0.8666), np.float32(0.8621), np.float32(0.4395), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6019), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.8305), np.float32(0.8499)] +2026-10-15 10:59:42.966227: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 10:59:43.286660: lr: 0.000843 +2026-10-15 10:59:43.296908: saving checkpoint... +2026-10-15 10:59:43.424747: done, saving took 0.14 seconds +2026-10-15 10:59:43.426509: This epoch took 259.220302 s + +2026-10-15 10:59:43.426708: +epoch: 173 +2026-10-15 11:03:46.538672: train loss : -0.6998 +2026-10-15 11:03:51.903830: validation loss: -0.5504 +2026-10-15 11:03:51.904803: Average global foreground Dice: [np.float32(0.9013), np.float32(0.5941), np.float32(0.7168), np.float32(0.7033), np.float32(0.6446), np.float32(0.7364), np.float32(0.6483), np.float32(0.0), np.float32(0.0), np.float32(0.5655), np.float32(0.5217), np.float32(0.6114), np.float32(0.5395), np.float32(0.452), np.float32(0.0), np.float32(0.0), np.float32(0.6965), np.float32(0.5278), np.float32(0.653), np.float32(0.6325), np.float32(0.4462), np.float32(0.5778), np.float32(0.7366), np.float32(0.7185), np.float32(0.4266), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4232), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6951), np.float32(0.6806)] +2026-10-15 11:03:51.905070: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:03:52.223768: lr: 0.000842 +2026-10-15 11:03:52.232692: saving checkpoint... +2026-10-15 11:03:52.366912: done, saving took 0.14 seconds +2026-10-15 11:03:52.368937: This epoch took 248.942135 s + +2026-10-15 11:03:52.369145: +epoch: 174 +2026-10-15 11:08:32.706704: train loss : -0.7040 +2026-10-15 11:08:38.073480: validation loss: -0.5433 +2026-10-15 11:08:38.074423: Average global foreground Dice: [np.float32(0.9066), np.float32(0.6252), np.float32(0.6804), np.float32(0.6994), np.float32(0.623), np.float32(0.7173), np.float32(0.6074), np.float32(0.0), np.float32(0.0), np.float32(0.4288), np.float32(0.5314), np.float32(0.6081), np.float32(0.5948), np.float32(0.5134), np.float32(0.0), np.float32(0.0), np.float32(0.6543), np.float32(0.5215), np.float32(0.5901), np.float32(0.5378), np.float32(0.4932), np.float32(0.4547), np.float32(0.7937), np.float32(0.771), np.float32(0.3765), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5942), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7187), np.float32(0.7389)] +2026-10-15 11:08:38.074773: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:08:38.431971: lr: 0.000841 +2026-10-15 11:08:38.462953: saving checkpoint... +2026-10-15 11:08:38.628817: done, saving took 0.20 seconds +2026-10-15 11:08:38.631384: This epoch took 286.262135 s + +2026-10-15 11:08:38.631577: +epoch: 175 +2026-10-15 11:12:34.923455: train loss : -0.7004 +2026-10-15 11:12:40.306276: validation loss: -0.5210 +2026-10-15 11:12:40.307228: Average global foreground Dice: [np.float32(0.9171), np.float32(0.6086), np.float32(0.6434), np.float32(0.6494), np.float32(0.5679), np.float32(0.6738), np.float32(0.5833), np.float32(0.0), np.float32(0.0), np.float32(0.4426), np.float32(0.4946), np.float32(0.5537), np.float32(0.5723), np.float32(0.5167), np.float32(0.0), np.float32(0.0), np.float32(0.5839), np.float32(0.4432), np.float32(0.5484), np.float32(0.5359), np.float32(0.5521), np.float32(0.4323), np.float32(0.7688), np.float32(0.7301), np.float32(0.3549), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4218), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6869), np.float32(0.6954)] +2026-10-15 11:12:40.307422: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:12:40.627537: lr: 0.00084 +2026-10-15 11:12:40.627981: This epoch took 241.996248 s + +2026-10-15 11:12:40.628149: +epoch: 176 +2026-10-15 11:17:13.692590: train loss : -0.7064 +2026-10-15 11:17:19.082099: validation loss: -0.4039 +2026-10-15 11:17:19.083049: Average global foreground Dice: [np.float32(0.9148), np.float32(0.4474), np.float32(0.4317), np.float32(0.4052), np.float32(0.3695), np.float32(0.4402), np.float32(0.3809), np.float32(0.0), np.float32(0.0), np.float32(0.4336), np.float32(0.2695), np.float32(0.3243), np.float32(0.3617), np.float32(0.2961), np.float32(0.0), np.float32(0.0), np.float32(0.31), np.float32(0.2494), np.float32(0.337), np.float32(0.3149), np.float32(0.3234), np.float32(0.2671), np.float32(0.5834), np.float32(0.5411), np.float32(0.2171), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4272), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4575), np.float32(0.4882)] +2026-10-15 11:17:19.083642: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:17:19.402128: lr: 0.000839 +2026-10-15 11:17:19.402599: This epoch took 278.774364 s + +2026-10-15 11:17:19.402921: +epoch: 177 +2026-10-15 11:21:36.833133: train loss : -0.7082 +2026-10-15 11:21:42.215650: validation loss: -0.4504 +2026-10-15 11:21:42.216846: Average global foreground Dice: [np.float32(0.9002), np.float32(0.477), np.float32(0.4967), np.float32(0.4562), np.float32(0.3234), np.float32(0.4976), np.float32(0.4309), np.float32(0.0), np.float32(0.0), np.float32(0.504), np.float32(0.3635), np.float32(0.4058), np.float32(0.4703), np.float32(0.3172), np.float32(0.0), np.float32(0.0), np.float32(0.3141), np.float32(0.2377), np.float32(0.3607), np.float32(0.3639), np.float32(0.497), np.float32(0.4258), np.float32(0.6522), np.float32(0.6249), np.float32(0.1989), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5756), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5016), np.float32(0.5366)] +2026-10-15 11:21:42.217253: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:21:42.535421: lr: 0.000838 +2026-10-15 11:21:42.535737: This epoch took 263.132699 s + +2026-10-15 11:21:42.535854: +epoch: 178 +2026-10-15 11:25:55.002807: train loss : -0.6978 +2026-10-15 11:26:00.377385: validation loss: -0.4101 +2026-10-15 11:26:00.378558: Average global foreground Dice: [np.float32(0.9092), np.float32(0.4617), np.float32(0.497), np.float32(0.5603), np.float32(0.4282), np.float32(0.3555), np.float32(0.2141), np.float32(0.0), np.float32(0.0), np.float32(0.5066), np.float32(0.4358), np.float32(0.3732), np.float32(0.4659), np.float32(0.2811), np.float32(0.0), np.float32(0.0), np.float32(0.4627), np.float32(0.3365), np.float32(0.3086), np.float32(0.3466), np.float32(0.2849), np.float32(0.4659), np.float32(0.5025), np.float32(0.5514), np.float32(0.3665), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3595), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4939), np.float32(0.464)] +2026-10-15 11:26:00.378826: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:26:00.714897: lr: 0.000837 +2026-10-15 11:26:00.715534: This epoch took 258.179570 s + +2026-10-15 11:26:00.715793: +epoch: 179 +2026-10-15 11:30:41.356635: train loss : -0.6898 +2026-10-15 11:30:46.735731: validation loss: -0.4674 +2026-10-15 11:30:46.736943: Average global foreground Dice: [np.float32(0.9017), np.float32(0.5067), np.float32(0.5825), np.float32(0.5458), np.float32(0.3901), np.float32(0.5231), np.float32(0.4371), np.float32(0.0), np.float32(0.0), np.float32(0.3423), np.float32(0.4129), np.float32(0.4482), np.float32(0.53), np.float32(0.4658), np.float32(0.0), np.float32(0.0), np.float32(0.4189), np.float32(0.3236), np.float32(0.3931), np.float32(0.4285), np.float32(0.4823), np.float32(0.5124), np.float32(0.6795), np.float32(0.6434), np.float32(0.3408), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4684), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5678), np.float32(0.5868)] +2026-10-15 11:30:46.737202: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:30:47.056301: lr: 0.000836 +2026-10-15 11:30:47.056725: This epoch took 286.340585 s + +2026-10-15 11:30:47.057064: +epoch: 180 +2026-10-15 11:34:38.479797: train loss : -0.7034 +2026-10-15 11:34:43.836121: validation loss: -0.5286 +2026-10-15 11:34:43.837158: Average global foreground Dice: [np.float32(0.9071), np.float32(0.5925), np.float32(0.6717), np.float32(0.7001), np.float32(0.5815), np.float32(0.7263), np.float32(0.6203), np.float32(0.0), np.float32(0.0), np.float32(0.4263), np.float32(0.5085), np.float32(0.599), np.float32(0.5305), np.float32(0.4724), np.float32(0.0), np.float32(0.0), np.float32(0.6009), np.float32(0.4664), np.float32(0.5562), np.float32(0.51), np.float32(0.5374), np.float32(0.5438), np.float32(0.7628), np.float32(0.7232), np.float32(0.4032), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6156), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.722), np.float32(0.7324)] +2026-10-15 11:34:43.837319: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:34:44.157102: lr: 0.000836 +2026-10-15 11:34:44.157378: This epoch took 237.100041 s + +2026-10-15 11:34:44.157471: +epoch: 181 +2026-10-15 11:38:38.797832: train loss : -0.7118 +2026-10-15 11:38:44.158185: validation loss: -0.5721 +2026-10-15 11:38:44.159348: Average global foreground Dice: [np.float32(0.9078), np.float32(0.6452), np.float32(0.7672), np.float32(0.7722), np.float32(0.6944), np.float32(0.8149), np.float32(0.7164), np.float32(0.0), np.float32(0.0), np.float32(0.3233), np.float32(0.5333), np.float32(0.6398), np.float32(0.5845), np.float32(0.5141), np.float32(0.0), np.float32(0.0), np.float32(0.7329), np.float32(0.5936), np.float32(0.6924), np.float32(0.6171), np.float32(0.5853), np.float32(0.5707), np.float32(0.8259), np.float32(0.8113), np.float32(0.4565), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.646), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.8091), np.float32(0.83)] +2026-10-15 11:38:44.159738: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:38:44.483052: lr: 0.000835 +2026-10-15 11:38:44.483332: This epoch took 240.325791 s + +2026-10-15 11:38:44.483422: +epoch: 182 +2026-10-15 11:42:56.702842: train loss : -0.7056 +2026-10-15 11:43:02.051866: validation loss: -0.5318 +2026-10-15 11:43:02.053028: Average global foreground Dice: [np.float32(0.9084), np.float32(0.5585), np.float32(0.6714), np.float32(0.684), np.float32(0.6004), np.float32(0.7232), np.float32(0.6252), np.float32(0.0), np.float32(0.0), np.float32(0.5238), np.float32(0.4536), np.float32(0.5158), np.float32(0.5561), np.float32(0.4889), np.float32(0.0), np.float32(0.0), np.float32(0.6041), np.float32(0.4926), np.float32(0.569), np.float32(0.5265), np.float32(0.4949), np.float32(0.5353), np.float32(0.8222), np.float32(0.7815), np.float32(0.4134), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.61), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7095), np.float32(0.7367)] +2026-10-15 11:43:02.053375: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:43:02.370156: lr: 0.000834 +2026-10-15 11:43:02.370443: This epoch took 257.886958 s + +2026-10-15 11:43:02.370522: +epoch: 183 +2026-10-15 11:47:38.436914: train loss : -0.7063 +2026-10-15 11:47:43.809004: validation loss: -0.4966 +2026-10-15 11:47:43.809922: Average global foreground Dice: [np.float32(0.9124), np.float32(0.5708), np.float32(0.6132), np.float32(0.5667), np.float32(0.4927), np.float32(0.6155), np.float32(0.518), np.float32(0.0101), np.float32(0.0), np.float32(0.5141), np.float32(0.4395), np.float32(0.4485), np.float32(0.552), np.float32(0.5051), np.float32(0.0), np.float32(0.0), np.float32(0.4806), np.float32(0.3769), np.float32(0.4639), np.float32(0.4), np.float32(0.5228), np.float32(0.4506), np.float32(0.7953), np.float32(0.7663), np.float32(0.3246), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3281), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6394), np.float32(0.6711)] +2026-10-15 11:47:43.810220: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:47:44.118323: lr: 0.000833 +2026-10-15 11:47:44.118732: This epoch took 281.748156 s + +2026-10-15 11:47:44.118939: +epoch: 184 +2026-10-15 11:52:11.045222: train loss : -0.7093 +2026-10-15 11:52:16.409976: validation loss: -0.5250 +2026-10-15 11:52:16.411002: Average global foreground Dice: [np.float32(0.9199), np.float32(0.5628), np.float32(0.6348), np.float32(0.6222), np.float32(0.5107), np.float32(0.6482), np.float32(0.5324), np.float32(0.0), np.float32(0.0), np.float32(0.5411), np.float32(0.4963), np.float32(0.5496), np.float32(0.5536), np.float32(0.4858), np.float32(0.0), np.float32(0.0), np.float32(0.538), np.float32(0.4394), np.float32(0.5361), np.float32(0.4541), np.float32(0.5793), np.float32(0.5271), np.float32(0.7927), np.float32(0.745), np.float32(0.3459), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6679), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6728), np.float32(0.6988)] +2026-10-15 11:52:16.411410: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:52:16.786626: lr: 0.000832 +2026-10-15 11:52:16.786857: This epoch took 272.667708 s + +2026-10-15 11:52:16.786963: +epoch: 185 +2026-10-15 11:56:23.661555: train loss : -0.7136 +2026-10-15 11:56:29.051832: validation loss: -0.4789 +2026-10-15 11:56:29.052762: Average global foreground Dice: [np.float32(0.9109), np.float32(0.4801), np.float32(0.5261), np.float32(0.4963), np.float32(0.3772), np.float32(0.5516), np.float32(0.4587), np.float32(0.5354), np.float32(0.0), np.float32(0.4253), np.float32(0.3815), np.float32(0.5007), np.float32(0.5434), np.float32(0.4776), np.float32(0.0), np.float32(0.0), np.float32(0.3634), np.float32(0.3083), np.float32(0.3931), np.float32(0.3754), np.float32(0.4674), np.float32(0.4324), np.float32(0.7289), np.float32(0.6401), np.float32(0.2402), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6086), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5552), np.float32(0.5982)] +2026-10-15 11:56:29.052936: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 11:56:29.364721: lr: 0.000831 +2026-10-15 11:56:29.365036: This epoch took 252.577994 s + +2026-10-15 11:56:29.365137: +epoch: 186 +2026-10-15 12:00:08.723451: train loss : -0.7159 +2026-10-15 12:00:14.087222: validation loss: -0.5127 +2026-10-15 12:00:14.088697: Average global foreground Dice: [np.float32(0.9059), np.float32(0.5466), np.float32(0.5945), np.float32(0.5655), np.float32(0.473), np.float32(0.6083), np.float32(0.5301), np.float32(0.2074), np.float32(0.0), np.float32(0.5262), np.float32(0.455), np.float32(0.5311), np.float32(0.5697), np.float32(0.5063), np.float32(0.0), np.float32(0.0), np.float32(0.4333), np.float32(0.3685), np.float32(0.4443), np.float32(0.4399), np.float32(0.4497), np.float32(0.3806), np.float32(0.7772), np.float32(0.6976), np.float32(0.2562), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4221), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.638), np.float32(0.6565)] +2026-10-15 12:00:14.088973: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:00:14.429458: lr: 0.00083 +2026-10-15 12:00:14.430158: This epoch took 225.064943 s + +2026-10-15 12:00:14.430532: +epoch: 187 +2026-10-15 12:04:24.939393: train loss : -0.7043 +2026-10-15 12:04:30.304870: validation loss: -0.5179 +2026-10-15 12:04:30.305996: Average global foreground Dice: [np.float32(0.8924), np.float32(0.5239), np.float32(0.575), np.float32(0.603), np.float32(0.51), np.float32(0.5615), np.float32(0.5774), np.float32(0.5833), np.float32(0.0), np.float32(0.4753), np.float32(0.4809), np.float32(0.5083), np.float32(0.5867), np.float32(0.5319), np.float32(0.0), np.float32(0.0), np.float32(0.4267), np.float32(0.3069), np.float32(0.4625), np.float32(0.4576), np.float32(0.455), np.float32(0.4652), np.float32(0.7746), np.float32(0.7508), np.float32(0.2604), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4095), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6461), np.float32(0.6718)] +2026-10-15 12:04:30.306394: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:04:30.639826: lr: 0.000829 +2026-10-15 12:04:30.640306: This epoch took 256.209574 s + +2026-10-15 12:04:30.640624: +epoch: 188 +2026-10-15 12:08:32.914756: train loss : -0.7090 +2026-10-15 12:08:38.280727: validation loss: -0.5083 +2026-10-15 12:08:38.281827: Average global foreground Dice: [np.float32(0.8949), np.float32(0.5621), np.float32(0.6575), np.float32(0.6768), np.float32(0.552), np.float32(0.679), np.float32(0.5616), np.float32(0.5333), np.float32(0.0), np.float32(0.4542), np.float32(0.5142), np.float32(0.528), np.float32(0.6387), np.float32(0.5892), np.float32(0.0), np.float32(0.0), np.float32(0.5445), np.float32(0.3821), np.float32(0.5268), np.float32(0.5143), np.float32(0.5616), np.float32(0.5037), np.float32(0.7621), np.float32(0.7347), np.float32(0.4419), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3846), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7068), np.float32(0.7104)] +2026-10-15 12:08:38.282210: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:08:38.597366: lr: 0.000828 +2026-10-15 12:08:38.597609: This epoch took 247.956760 s + +2026-10-15 12:08:38.597692: +epoch: 189 +2026-10-15 12:13:20.057681: train loss : -0.6979 +2026-10-15 12:13:25.421581: validation loss: -0.5310 +2026-10-15 12:13:25.422532: Average global foreground Dice: [np.float32(0.8873), np.float32(0.5812), np.float32(0.6769), np.float32(0.6912), np.float32(0.561), np.float32(0.7315), np.float32(0.6056), np.float32(0.4621), np.float32(0.0), np.float32(0.4757), np.float32(0.5236), np.float32(0.6127), np.float32(0.5719), np.float32(0.5565), np.float32(0.0), np.float32(0.0), np.float32(0.5702), np.float32(0.4773), np.float32(0.5334), np.float32(0.4931), np.float32(0.5735), np.float32(0.4516), np.float32(0.7326), np.float32(0.6784), np.float32(0.3431), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6792), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6732), np.float32(0.6926)] +2026-10-15 12:13:25.422718: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:13:25.736189: lr: 0.000827 +2026-10-15 12:13:25.736441: This epoch took 287.138677 s + +2026-10-15 12:13:25.736533: +epoch: 190 +2026-10-15 12:17:45.171271: train loss : -0.7071 +2026-10-15 12:17:50.549040: validation loss: -0.4240 +2026-10-15 12:17:50.550028: Average global foreground Dice: [np.float32(0.9113), np.float32(0.4804), np.float32(0.4851), np.float32(0.4738), np.float32(0.3506), np.float32(0.5103), np.float32(0.4215), np.float32(0.4716), np.float32(0.0), np.float32(0.5865), np.float32(0.4029), np.float32(0.4525), np.float32(0.5108), np.float32(0.3339), np.float32(0.0), np.float32(0.0), np.float32(0.335), np.float32(0.2637), np.float32(0.3398), np.float32(0.3138), np.float32(0.4122), np.float32(0.3296), np.float32(0.5198), np.float32(0.5106), np.float32(0.1884), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5045), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4947), np.float32(0.539)] +2026-10-15 12:17:50.550295: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:17:50.881126: lr: 0.000826 +2026-10-15 12:17:50.881412: This epoch took 265.144819 s + +2026-10-15 12:17:50.881708: +epoch: 191 +2026-10-15 12:22:04.834384: train loss : -0.7110 +2026-10-15 12:22:10.257146: validation loss: -0.5596 +2026-10-15 12:22:10.258050: Average global foreground Dice: [np.float32(0.9163), np.float32(0.6375), np.float32(0.7165), np.float32(0.7464), np.float32(0.6429), np.float32(0.7649), np.float32(0.6538), np.float32(0.3552), np.float32(0.0), np.float32(0.6113), np.float32(0.5525), np.float32(0.6298), np.float32(0.5694), np.float32(0.5105), np.float32(0.0), np.float32(0.0), np.float32(0.6525), np.float32(0.5072), np.float32(0.6411), np.float32(0.5413), np.float32(0.5112), np.float32(0.5661), np.float32(0.8202), np.float32(0.7973), np.float32(0.416), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6183), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.7501), np.float32(0.7785)] +2026-10-15 12:22:10.258211: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:22:10.584505: lr: 0.000825 +2026-10-15 12:22:10.584764: This epoch took 259.702775 s + +2026-10-15 12:22:10.585090: +epoch: 192 +2026-10-15 12:26:20.938271: train loss : -0.7108 +2026-10-15 12:26:26.315439: validation loss: -0.5134 +2026-10-15 12:26:26.316339: Average global foreground Dice: [np.float32(0.9031), np.float32(0.5856), np.float32(0.6207), np.float32(0.6098), np.float32(0.4862), np.float32(0.659), np.float32(0.5258), np.float32(0.4535), np.float32(0.0), np.float32(0.6299), np.float32(0.5231), np.float32(0.5915), np.float32(0.5561), np.float32(0.5332), np.float32(0.0), np.float32(0.0), np.float32(0.459), np.float32(0.3772), np.float32(0.4461), np.float32(0.4248), np.float32(0.4949), np.float32(0.3178), np.float32(0.6941), np.float32(0.6584), np.float32(0.2777), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.571), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6349), np.float32(0.6657)] +2026-10-15 12:26:26.316494: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:26:26.648364: lr: 0.000824 +2026-10-15 12:26:26.648839: This epoch took 256.063625 s + +2026-10-15 12:26:26.649161: +epoch: 193 +2026-10-15 12:31:07.663731: train loss : -0.6984 +2026-10-15 12:31:13.023915: validation loss: -0.5810 +2026-10-15 12:31:13.025068: Average global foreground Dice: [np.float32(0.901), np.float32(0.5935), np.float32(0.7565), np.float32(0.7896), np.float32(0.7048), np.float32(0.8063), np.float32(0.717), np.float32(0.4503), np.float32(0.0), np.float32(0.4434), np.float32(0.5818), np.float32(0.6832), np.float32(0.6091), np.float32(0.4894), np.float32(0.0), np.float32(0.0), np.float32(0.7547), np.float32(0.5807), np.float32(0.7289), np.float32(0.6651), np.float32(0.595), np.float32(0.586), np.float32(0.8544), np.float32(0.8244), np.float32(0.5036), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.416), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.4212), np.float32(0.8302), np.float32(0.8325)] +2026-10-15 12:31:13.025351: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:31:13.420168: lr: 0.000824 +2026-10-15 12:31:13.420650: This epoch took 286.771132 s + +2026-10-15 12:31:13.420910: +epoch: 194 +2026-10-15 12:35:25.262506: train loss : -0.7164 +2026-10-15 12:35:30.651463: validation loss: -0.6046 +2026-10-15 12:35:30.652761: Average global foreground Dice: [np.float32(0.9187), np.float32(0.6469), np.float32(0.7782), np.float32(0.7759), np.float32(0.6997), np.float32(0.8046), np.float32(0.6968), np.float32(0.4449), np.float32(0.0), np.float32(0.5656), np.float32(0.6426), np.float32(0.6932), np.float32(0.6468), np.float32(0.5941), np.float32(0.0), np.float32(0.0), np.float32(0.7499), np.float32(0.6335), np.float32(0.7024), np.float32(0.6332), np.float32(0.6196), np.float32(0.6261), np.float32(0.8571), np.float32(0.8414), np.float32(0.4735), np.float32(0.4171), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6804), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.3558), np.float32(0.8134), np.float32(0.8209)] +2026-10-15 12:35:30.653003: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:35:30.981265: lr: 0.000823 +2026-10-15 12:35:30.990071: saving checkpoint... +2026-10-15 12:35:31.124012: done, saving took 0.14 seconds +2026-10-15 12:35:31.125479: This epoch took 257.704462 s + +2026-10-15 12:35:31.125649: +epoch: 195 +2026-10-15 12:40:15.985698: train loss : -0.7238 +2026-10-15 12:40:21.359234: validation loss: -0.5842 +2026-10-15 12:40:21.360180: Average global foreground Dice: [np.float32(0.9037), np.float32(0.6427), np.float32(0.7157), np.float32(0.7453), np.float32(0.6558), np.float32(0.7631), np.float32(0.662), np.float32(0.4819), np.float32(0.0), np.float32(0.5732), np.float32(0.6167), np.float32(0.6846), np.float32(0.6314), np.float32(0.5782), np.float32(0.0), np.float32(0.0), np.float32(0.7031), np.float32(0.5849), np.float32(0.66), np.float32(0.6024), np.float32(0.5208), np.float32(0.5684), np.float32(0.7968), np.float32(0.7624), np.float32(0.3992), np.float32(0.3441), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.706), np.float32(0.0), np.float32(0.0), np.float32(0.3321), np.float32(0.3547), np.float32(0.7496), np.float32(0.746)] +2026-10-15 12:40:21.360529: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:40:21.680756: lr: 0.000822 +2026-10-15 12:40:21.720781: saving checkpoint... +2026-10-15 12:40:21.858829: done, saving took 0.18 seconds +2026-10-15 12:40:21.860703: This epoch took 290.734968 s + +2026-10-15 12:40:21.860956: +epoch: 196 +2026-10-15 12:44:34.208299: train loss : -0.7276 +2026-10-15 12:44:39.603502: validation loss: -0.4756 +2026-10-15 12:44:39.604368: Average global foreground Dice: [np.float32(0.8944), np.float32(0.5087), np.float32(0.5679), np.float32(0.5254), np.float32(0.4485), np.float32(0.5618), np.float32(0.5172), np.float32(0.1726), np.float32(0.0), np.float32(0.5784), np.float32(0.3927), np.float32(0.4865), np.float32(0.5014), np.float32(0.4599), np.float32(0.0), np.float32(0.0), np.float32(0.4645), np.float32(0.413), np.float32(0.4646), np.float32(0.4399), np.float32(0.3485), np.float32(0.3855), np.float32(0.6272), np.float32(0.5422), np.float32(0.2622), np.float32(0.2173), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5362), np.float32(0.0), np.float32(0.0), np.float32(0.2728), np.float32(0.2551), np.float32(0.5238), np.float32(0.5621)] +2026-10-15 12:44:39.604531: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:44:39.927708: lr: 0.000821 +2026-10-15 12:44:39.927963: This epoch took 258.066920 s + +2026-10-15 12:44:39.928050: +epoch: 197 +2026-10-15 12:49:26.927648: train loss : -0.7308 +2026-10-15 12:49:32.299302: validation loss: -0.5561 +2026-10-15 12:49:32.300236: Average global foreground Dice: [np.float32(0.9104), np.float32(0.6018), np.float32(0.7108), np.float32(0.714), np.float32(0.6255), np.float32(0.7452), np.float32(0.6538), np.float32(0.5971), np.float32(0.0), np.float32(0.4214), np.float32(0.5304), np.float32(0.6266), np.float32(0.6141), np.float32(0.5466), np.float32(0.0), np.float32(0.0), np.float32(0.6474), np.float32(0.5117), np.float32(0.622), np.float32(0.5396), np.float32(0.5356), np.float32(0.5181), np.float32(0.7919), np.float32(0.7583), np.float32(0.4168), np.float32(0.3186), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6468), np.float32(0.0), np.float32(0.0), np.float32(0.2433), np.float32(0.3786), np.float32(0.7287), np.float32(0.7599)] +2026-10-15 12:49:32.300435: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:49:32.626871: lr: 0.00082 +2026-10-15 12:49:32.636111: saving checkpoint... +2026-10-15 12:49:32.765180: done, saving took 0.14 seconds +2026-10-15 12:49:32.766982: This epoch took 292.838861 s + +2026-10-15 12:49:32.767177: +epoch: 198 +2026-10-15 12:53:33.538070: train loss : -0.7354 +2026-10-15 12:53:38.920575: validation loss: -0.5370 +2026-10-15 12:53:38.921468: Average global foreground Dice: [np.float32(0.9012), np.float32(0.5658), np.float32(0.721), np.float32(0.6936), np.float32(0.6572), np.float32(0.7028), np.float32(0.6614), np.float32(0.5529), np.float32(0.0), np.float32(0.5893), np.float32(0.48), np.float32(0.5645), np.float32(0.5804), np.float32(0.4025), np.float32(0.0), np.float32(0.0), np.float32(0.6913), np.float32(0.4859), np.float32(0.6459), np.float32(0.5669), np.float32(0.499), np.float32(0.5499), np.float32(0.8192), np.float32(0.804), np.float32(0.4902), np.float32(0.289), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5699), np.float32(0.0), np.float32(0.0), np.float32(0.4927), np.float32(0.5195), np.float32(0.7412), np.float32(0.7731)] +2026-10-15 12:53:38.921630: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:53:39.229420: lr: 0.000819 +2026-10-15 12:53:39.242973: saving checkpoint... +2026-10-15 12:53:39.369278: done, saving took 0.14 seconds +2026-10-15 12:53:39.371322: This epoch took 246.604056 s + +2026-10-15 12:53:39.371684: +epoch: 199 +2026-10-15 12:57:46.509769: train loss : -0.7432 +2026-10-15 12:57:51.879611: validation loss: -0.4663 +2026-10-15 12:57:51.880580: Average global foreground Dice: [np.float32(0.8997), np.float32(0.4924), np.float32(0.5851), np.float32(0.5626), np.float32(0.4303), np.float32(0.6005), np.float32(0.5163), np.float32(0.5377), np.float32(0.0), np.float32(0.6256), np.float32(0.4222), np.float32(0.4386), np.float32(0.5194), np.float32(0.4005), np.float32(0.0), np.float32(0.0), np.float32(0.415), np.float32(0.297), np.float32(0.4261), np.float32(0.38), np.float32(0.4428), np.float32(0.4451), np.float32(0.5682), np.float32(0.5239), np.float32(0.2875), np.float32(0.2066), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5612), np.float32(0.0), np.float32(0.0), np.float32(0.2827), np.float32(0.3361), np.float32(0.5473), np.float32(0.5794)] +2026-10-15 12:57:51.880801: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 12:57:52.194075: lr: 0.000818 +2026-10-15 12:57:52.194308: saving scheduled checkpoint file... +2026-10-15 12:57:52.203587: saving checkpoint... +2026-10-15 12:57:52.332202: done, saving took 0.14 seconds +2026-10-15 12:57:52.333725: done +2026-10-15 12:57:52.333920: This epoch took 252.962109 s + +2026-10-15 12:57:52.333988: +epoch: 200 +2026-10-15 13:02:11.160837: train loss : -0.7392 +2026-10-15 13:02:16.528792: validation loss: -0.5740 +2026-10-15 13:02:16.529800: Average global foreground Dice: [np.float32(0.8954), np.float32(0.5998), np.float32(0.7051), np.float32(0.7539), np.float32(0.6181), np.float32(0.7877), np.float32(0.6468), np.float32(0.5893), np.float32(0.0), np.float32(0.5911), np.float32(0.6061), np.float32(0.6372), np.float32(0.5857), np.float32(0.5505), np.float32(0.0), np.float32(0.0), np.float32(0.6634), np.float32(0.4837), np.float32(0.641), np.float32(0.5671), np.float32(0.5201), np.float32(0.5638), np.float32(0.7539), np.float32(0.7251), np.float32(0.4815), np.float32(0.2366), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6082), np.float32(0.0), np.float32(0.0), np.float32(0.3798), np.float32(0.4755), np.float32(0.7558), np.float32(0.7714)] +2026-10-15 13:02:16.530297: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:02:16.868081: lr: 0.000817 +2026-10-15 13:02:16.921740: saving checkpoint... +2026-10-15 13:02:17.062000: done, saving took 0.19 seconds +2026-10-15 13:02:17.063643: This epoch took 264.729601 s + +2026-10-15 13:02:17.063896: +epoch: 201 +2026-10-15 13:06:19.400918: train loss : -0.7455 +2026-10-15 13:06:24.788914: validation loss: -0.5869 +2026-10-15 13:06:24.789775: Average global foreground Dice: [np.float32(0.9089), np.float32(0.6218), np.float32(0.7731), np.float32(0.7895), np.float32(0.682), np.float32(0.8029), np.float32(0.7109), np.float32(0.4267), np.float32(0.0), np.float32(0.5087), np.float32(0.6055), np.float32(0.6109), np.float32(0.6057), np.float32(0.541), np.float32(0.0), np.float32(0.0), np.float32(0.684), np.float32(0.5328), np.float32(0.6666), np.float32(0.5888), np.float32(0.5709), np.float32(0.5742), np.float32(0.8262), np.float32(0.8204), np.float32(0.5267), np.float32(0.3574), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5558), np.float32(0.0), np.float32(0.0), np.float32(0.404), np.float32(0.3759), np.float32(0.8142), np.float32(0.8233)] +2026-10-15 13:06:24.789928: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:06:25.122713: lr: 0.000816 +2026-10-15 13:06:25.180897: saving checkpoint... +2026-10-15 13:06:25.321846: done, saving took 0.20 seconds +2026-10-15 13:06:25.323467: This epoch took 248.259470 s + +2026-10-15 13:06:25.323638: +epoch: 202 +2026-10-15 13:10:39.404091: train loss : -0.7466 +2026-10-15 13:10:44.771746: validation loss: -0.5721 +2026-10-15 13:10:44.773045: Average global foreground Dice: [np.float32(0.9149), np.float32(0.6565), np.float32(0.7247), np.float32(0.7668), np.float32(0.6548), np.float32(0.7919), np.float32(0.6587), np.float32(0.5652), np.float32(0.0), np.float32(0.2952), np.float32(0.5868), np.float32(0.6386), np.float32(0.6053), np.float32(0.5358), np.float32(0.0), np.float32(0.0), np.float32(0.6778), np.float32(0.4764), np.float32(0.6308), np.float32(0.5176), np.float32(0.5125), np.float32(0.509), np.float32(0.8411), np.float32(0.8181), np.float32(0.4393), np.float32(0.4178), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6461), np.float32(0.0), np.float32(0.0), np.float32(0.5022), np.float32(0.5357), np.float32(0.8044), np.float32(0.8169)] +2026-10-15 13:10:44.773285: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:10:45.096061: lr: 0.000815 +2026-10-15 13:10:45.154367: saving checkpoint... +2026-10-15 13:10:45.298647: done, saving took 0.20 seconds +2026-10-15 13:10:45.300508: This epoch took 259.976678 s + +2026-10-15 13:10:45.300710: +epoch: 203 +2026-10-15 13:15:02.191937: train loss : -0.7453 +2026-10-15 13:15:07.584644: validation loss: -0.5778 +2026-10-15 13:15:07.585792: Average global foreground Dice: [np.float32(0.9066), np.float32(0.6298), np.float32(0.7223), np.float32(0.7085), np.float32(0.6603), np.float32(0.738), np.float32(0.6803), np.float32(0.6779), np.float32(0.0), np.float32(0.412), np.float32(0.5088), np.float32(0.5984), np.float32(0.5778), np.float32(0.5777), np.float32(0.0), np.float32(0.0), np.float32(0.6275), np.float32(0.5307), np.float32(0.5851), np.float32(0.4979), np.float32(0.5355), np.float32(0.5551), np.float32(0.8622), np.float32(0.8318), np.float32(0.389), np.float32(0.2616), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6311), np.float32(0.0), np.float32(0.0), np.float32(0.5506), np.float32(0.5679), np.float32(0.7614), np.float32(0.7821)] +2026-10-15 13:15:07.586183: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:15:07.924852: lr: 0.000814 +2026-10-15 13:15:07.981874: saving checkpoint... +2026-10-15 13:15:08.117962: done, saving took 0.19 seconds +2026-10-15 13:15:08.124208: This epoch took 262.823412 s + +2026-10-15 13:15:08.124391: +epoch: 204 +2026-10-15 13:19:32.747334: train loss : -0.7433 +2026-10-15 13:19:38.117491: validation loss: -0.5556 +2026-10-15 13:19:38.118569: Average global foreground Dice: [np.float32(0.9119), np.float32(0.5802), np.float32(0.7326), np.float32(0.7354), np.float32(0.6382), np.float32(0.7488), np.float32(0.6421), np.float32(0.6161), np.float32(0.0), np.float32(0.5699), np.float32(0.494), np.float32(0.6229), np.float32(0.5053), np.float32(0.4335), np.float32(0.0), np.float32(0.0), np.float32(0.6602), np.float32(0.5328), np.float32(0.6114), np.float32(0.6061), np.float32(0.5167), np.float32(0.5629), np.float32(0.8022), np.float32(0.7821), np.float32(0.4701), np.float32(0.3498), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.583), np.float32(0.0), np.float32(0.0), np.float32(0.3073), np.float32(0.4382), np.float32(0.7751), np.float32(0.8013)] +2026-10-15 13:19:38.119195: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:19:38.492155: lr: 0.000813 +2026-10-15 13:19:38.555932: saving checkpoint... +2026-10-15 13:19:38.700285: done, saving took 0.21 seconds +2026-10-15 13:19:38.702338: This epoch took 270.577865 s + +2026-10-15 13:19:38.702435: +epoch: 205 +2026-10-15 13:23:58.285418: train loss : -0.7439 +2026-10-15 13:24:03.657444: validation loss: -0.5297 +2026-10-15 13:24:03.658505: Average global foreground Dice: [np.float32(0.9051), np.float32(0.5538), np.float32(0.6123), np.float32(0.6408), np.float32(0.5529), np.float32(0.6374), np.float32(0.5537), np.float32(0.5539), np.float32(0.0), np.float32(0.5644), np.float32(0.4213), np.float32(0.4948), np.float32(0.5453), np.float32(0.4444), np.float32(0.0), np.float32(0.0), np.float32(0.5139), np.float32(0.378), np.float32(0.4968), np.float32(0.4986), np.float32(0.5204), np.float32(0.4636), np.float32(0.7914), np.float32(0.7455), np.float32(0.3946), np.float32(0.2878), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.6527), np.float32(0.0), np.float32(0.0), np.float32(0.3771), np.float32(0.3389), np.float32(0.7001), np.float32(0.7264)] +2026-10-15 13:24:03.659178: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:24:03.971680: lr: 0.000813 +2026-10-15 13:24:03.972090: This epoch took 265.269599 s + +2026-10-15 13:24:03.972340: +epoch: 206 +2026-10-15 13:28:06.611153: train loss : -0.7458 +2026-10-15 13:28:11.983717: validation loss: -0.5496 +2026-10-15 13:28:11.984798: Average global foreground Dice: [np.float32(0.9097), np.float32(0.5929), np.float32(0.658), np.float32(0.6711), np.float32(0.5628), np.float32(0.718), np.float32(0.6086), np.float32(0.5471), np.float32(0.0), np.float32(0.4621), np.float32(0.5191), np.float32(0.5157), np.float32(0.5237), np.float32(0.4949), np.float32(0.0), np.float32(0.0), np.float32(0.6126), np.float32(0.4526), np.float32(0.5816), np.float32(0.5282), np.float32(0.4752), np.float32(0.4322), np.float32(0.7802), np.float32(0.7666), np.float32(0.4196), np.float32(0.2969), np.float32(0.0), np.float32(0.0), np.float32(0.0044), np.float32(0.6766), np.float32(0.0), np.float32(0.0), np.float32(0.4079), np.float32(0.471), np.float32(0.7285), np.float32(0.7557)] +2026-10-15 13:28:11.985096: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:28:12.327836: lr: 0.000812 +2026-10-15 13:28:12.328310: This epoch took 248.355874 s + +2026-10-15 13:28:12.328622: +epoch: 207 +2026-10-15 13:32:24.543721: train loss : -0.7487 +2026-10-15 13:32:29.904728: validation loss: -0.5236 +2026-10-15 13:32:29.905702: Average global foreground Dice: [np.float32(0.9003), np.float32(0.5581), np.float32(0.6188), np.float32(0.656), np.float32(0.5286), np.float32(0.6828), np.float32(0.5649), np.float32(0.5427), np.float32(0.0), np.float32(0.5696), np.float32(0.5182), np.float32(0.574), np.float32(0.5521), np.float32(0.5593), np.float32(0.0), np.float32(0.0), np.float32(0.4518), np.float32(0.3292), np.float32(0.4416), np.float32(0.4078), np.float32(0.5178), np.float32(0.3575), np.float32(0.6958), np.float32(0.674), np.float32(0.3208), np.float32(0.309), np.float32(0.0), np.float32(0.0), np.float32(0.0601), np.float32(0.6183), np.float32(0.0), np.float32(0.0), np.float32(0.3637), np.float32(0.2257), np.float32(0.6538), np.float32(0.6801)] +2026-10-15 13:32:29.905882: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:32:30.312026: lr: 0.000811 +2026-10-15 13:32:30.312265: This epoch took 257.983356 s + +2026-10-15 13:32:30.312452: +epoch: 208 +2026-10-15 13:36:30.922675: train loss : -0.7605 +2026-10-15 13:36:36.295078: validation loss: -0.5820 +2026-10-15 13:36:36.296024: Average global foreground Dice: [np.float32(0.9019), np.float32(0.6506), np.float32(0.7546), np.float32(0.7804), np.float32(0.7027), np.float32(0.8127), np.float32(0.7177), np.float32(0.5214), np.float32(0.0), np.float32(0.5214), np.float32(0.5626), np.float32(0.6568), np.float32(0.5898), np.float32(0.5215), np.float32(0.0), np.float32(0.0), np.float32(0.7329), np.float32(0.5775), np.float32(0.6999), np.float32(0.6438), np.float32(0.5103), np.float32(0.5095), np.float32(0.7663), np.float32(0.7315), np.float32(0.4508), np.float32(0.3854), np.float32(0.0), np.float32(0.0), np.float32(0.116), np.float32(0.6598), np.float32(0.0), np.float32(0.0), np.float32(0.4757), np.float32(0.443), np.float32(0.7888), np.float32(0.8093)] +2026-10-15 13:36:36.296196: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:36:36.642406: lr: 0.00081 +2026-10-15 13:36:36.704773: saving checkpoint... +2026-10-15 13:36:36.838538: done, saving took 0.20 seconds +2026-10-15 13:36:36.840713: This epoch took 246.528052 s + +2026-10-15 13:36:36.840929: +epoch: 209 +2026-10-15 13:40:46.132646: train loss : -0.7479 +2026-10-15 13:40:51.512957: validation loss: -0.5574 +2026-10-15 13:40:51.513934: Average global foreground Dice: [np.float32(0.9119), np.float32(0.6458), np.float32(0.6536), np.float32(0.6585), np.float32(0.5892), np.float32(0.6897), np.float32(0.6031), np.float32(0.6051), np.float32(0.0), np.float32(0.4791), np.float32(0.5016), np.float32(0.6019), np.float32(0.643), np.float32(0.6001), np.float32(0.0), np.float32(0.0), np.float32(0.5522), np.float32(0.4582), np.float32(0.5341), np.float32(0.4688), np.float32(0.4632), np.float32(0.3888), np.float32(0.8353), np.float32(0.8132), np.float32(0.2868), np.float32(0.3906), np.float32(0.0), np.float32(0.0), np.float32(0.0033), np.float32(0.4105), np.float32(0.0), np.float32(0.0), np.float32(0.3064), np.float32(0.4729), np.float32(0.7232), np.float32(0.7456)] +2026-10-15 13:40:51.514218: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:40:51.833796: lr: 0.000809 +2026-10-15 13:40:51.893245: saving checkpoint... +2026-10-15 13:40:52.043891: done, saving took 0.21 seconds +2026-10-15 13:40:52.046320: This epoch took 255.205289 s + +2026-10-15 13:40:52.046838: +epoch: 210 +2026-10-15 13:44:37.238445: train loss : -0.7638 +2026-10-15 13:44:42.610000: validation loss: -0.5810 +2026-10-15 13:44:42.611017: Average global foreground Dice: [np.float32(0.9064), np.float32(0.6047), np.float32(0.7635), np.float32(0.7419), np.float32(0.6472), np.float32(0.7588), np.float32(0.6963), np.float32(0.422), np.float32(0.0), np.float32(0.4011), np.float32(0.5288), np.float32(0.6169), np.float32(0.5889), np.float32(0.5399), np.float32(0.0), np.float32(0.0), np.float32(0.6322), np.float32(0.4252), np.float32(0.6197), np.float32(0.4825), np.float32(0.5908), np.float32(0.5739), np.float32(0.8476), np.float32(0.8295), np.float32(0.4693), np.float32(0.4105), np.float32(0.0), np.float32(0.0), np.float32(0.2011), np.float32(0.6419), np.float32(0.0), np.float32(0.0), np.float32(0.2586), np.float32(0.3942), np.float32(0.7897), np.float32(0.8056)] +2026-10-15 13:44:42.611335: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:44:42.973515: lr: 0.000808 +2026-10-15 13:44:43.043059: saving checkpoint... +2026-10-15 13:44:43.182278: done, saving took 0.21 seconds +2026-10-15 13:44:43.184049: This epoch took 231.137008 s + +2026-10-15 13:44:43.184271: +epoch: 211 +2026-10-15 13:49:01.644411: train loss : -0.7536 +2026-10-15 13:49:07.020956: validation loss: -0.5865 +2026-10-15 13:49:07.022110: Average global foreground Dice: [np.float32(0.9118), np.float32(0.6603), np.float32(0.7784), np.float32(0.7772), np.float32(0.7221), np.float32(0.8094), np.float32(0.7276), np.float32(0.5719), np.float32(0.0), np.float32(0.347), np.float32(0.5937), np.float32(0.6645), np.float32(0.5995), np.float32(0.5643), np.float32(0.0), np.float32(0.0), np.float32(0.7176), np.float32(0.569), np.float32(0.6818), np.float32(0.6026), np.float32(0.5695), np.float32(0.541), np.float32(0.8913), np.float32(0.8725), np.float32(0.4549), np.float32(0.4396), np.float32(0.0), np.float32(0.0), np.float32(0.1569), np.float32(0.5766), np.float32(0.0), np.float32(0.0), np.float32(0.3067), np.float32(0.4644), np.float32(0.829), np.float32(0.8447)] +2026-10-15 13:49:07.022558: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:49:07.341209: lr: 0.000807 +2026-10-15 13:49:07.386028: saving checkpoint... +2026-10-15 13:49:07.520376: done, saving took 0.18 seconds +2026-10-15 13:49:07.522188: This epoch took 264.337819 s + +2026-10-15 13:49:07.522312: +epoch: 212 +2026-10-15 13:53:17.145618: train loss : -0.7551 +2026-10-15 13:53:22.527586: validation loss: -0.5107 +2026-10-15 13:53:22.528606: Average global foreground Dice: [np.float32(0.9103), np.float32(0.526), np.float32(0.5741), np.float32(0.5811), np.float32(0.4514), np.float32(0.637), np.float32(0.5638), np.float32(0.4527), np.float32(0.0), np.float32(0.5172), np.float32(0.4647), np.float32(0.5037), np.float32(0.5495), np.float32(0.437), np.float32(0.0), np.float32(0.0), np.float32(0.427), np.float32(0.3152), np.float32(0.4616), np.float32(0.4413), np.float32(0.473), np.float32(0.373), np.float32(0.7243), np.float32(0.6678), np.float32(0.2524), np.float32(0.2618), np.float32(0.0), np.float32(0.0), np.float32(0.0724), np.float32(0.4978), np.float32(0.0), np.float32(0.0), np.float32(0.3462), np.float32(0.3338), np.float32(0.6252), np.float32(0.6376)] +2026-10-15 13:53:22.528984: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:53:22.845361: lr: 0.000806 +2026-10-15 13:53:22.845823: This epoch took 255.323450 s + +2026-10-15 13:53:22.846104: +epoch: 213 +2026-10-15 13:57:33.639026: train loss : -0.7612 +2026-10-15 13:57:39.010386: validation loss: -0.5151 +2026-10-15 13:57:39.011412: Average global foreground Dice: [np.float32(0.9164), np.float32(0.4984), np.float32(0.5588), np.float32(0.5915), np.float32(0.4799), np.float32(0.6369), np.float32(0.5624), np.float32(0.4856), np.float32(0.0), np.float32(0.2761), np.float32(0.4362), np.float32(0.4845), np.float32(0.5202), np.float32(0.3922), np.float32(0.0), np.float32(0.0), np.float32(0.4968), np.float32(0.4039), np.float32(0.5119), np.float32(0.4693), np.float32(0.4412), np.float32(0.3923), np.float32(0.8086), np.float32(0.769), np.float32(0.3269), np.float32(0.2416), np.float32(0.0), np.float32(0.0), np.float32(0.345), np.float32(0.5697), np.float32(0.0), np.float32(0.0), np.float32(0.3635), np.float32(0.4396), np.float32(0.6591), np.float32(0.6616)] +2026-10-15 13:57:39.011822: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 13:57:39.385564: lr: 0.000805 +2026-10-15 13:57:39.385818: This epoch took 256.539465 s + +2026-10-15 13:57:39.385912: +epoch: 214 +2026-10-15 14:01:54.872615: train loss : -0.7559 +2026-10-15 14:02:00.237483: validation loss: -0.5350 +2026-10-15 14:02:00.238376: Average global foreground Dice: [np.float32(0.9021), np.float32(0.5844), np.float32(0.6336), np.float32(0.6305), np.float32(0.5834), np.float32(0.672), np.float32(0.6048), np.float32(0.5934), np.float32(0.0), np.float32(0.4495), np.float32(0.478), np.float32(0.5622), np.float32(0.5431), np.float32(0.4389), np.float32(0.0), np.float32(0.0), np.float32(0.6058), np.float32(0.4537), np.float32(0.6009), np.float32(0.5817), np.float32(0.4851), np.float32(0.4223), np.float32(0.7771), np.float32(0.7383), np.float32(0.3643), np.float32(0.286), np.float32(0.0), np.float32(0.0), np.float32(0.1255), np.float32(0.4849), np.float32(0.0), np.float32(0.0), np.float32(0.3507), np.float32(0.4743), np.float32(0.6719), np.float32(0.6886)] +2026-10-15 14:02:00.238571: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:02:00.604558: lr: 0.000804 +2026-10-15 14:02:00.605452: This epoch took 261.219486 s + +2026-10-15 14:02:00.605804: +epoch: 215 +2026-10-15 14:06:03.136288: train loss : -0.7481 +2026-10-15 14:06:08.522089: validation loss: -0.5010 +2026-10-15 14:06:08.523182: Average global foreground Dice: [np.float32(0.9039), np.float32(0.4911), np.float32(0.5363), np.float32(0.5034), np.float32(0.4116), np.float32(0.5455), np.float32(0.5006), np.float32(0.4526), np.float32(0.0), np.float32(0.527), np.float32(0.4156), np.float32(0.4957), np.float32(0.5631), np.float32(0.4778), np.float32(0.0), np.float32(0.0), np.float32(0.3674), np.float32(0.2869), np.float32(0.3922), np.float32(0.387), np.float32(0.4521), np.float32(0.4122), np.float32(0.7619), np.float32(0.7663), np.float32(0.1954), np.float32(0.2077), np.float32(0.0), np.float32(0.0), np.float32(0.3556), np.float32(0.6037), np.float32(0.0), np.float32(0.0), np.float32(0.1987), np.float32(0.3341), np.float32(0.5928), np.float32(0.6555)] +2026-10-15 14:06:08.523569: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:06:08.929045: lr: 0.000803 +2026-10-15 14:06:08.929378: This epoch took 248.323408 s + +2026-10-15 14:06:08.929616: +epoch: 216 +2026-10-15 14:10:21.787811: train loss : -0.7580 +2026-10-15 14:10:27.185673: validation loss: -0.4624 +2026-10-15 14:10:27.186535: Average global foreground Dice: [np.float32(0.924), np.float32(0.4884), np.float32(0.5014), np.float32(0.4953), np.float32(0.4033), np.float32(0.5382), np.float32(0.4626), np.float32(0.5528), np.float32(0.0), np.float32(0.4152), np.float32(0.3798), np.float32(0.4379), np.float32(0.5474), np.float32(0.4465), np.float32(0.0), np.float32(0.0), np.float32(0.3569), np.float32(0.2728), np.float32(0.3672), np.float32(0.3793), np.float32(0.4422), np.float32(0.3767), np.float32(0.6742), np.float32(0.6253), np.float32(0.1594), np.float32(0.1823), np.float32(0.0), np.float32(0.0), np.float32(0.149), np.float32(0.422), np.float32(0.0), np.float32(0.0), np.float32(0.3222), np.float32(0.3323), np.float32(0.5528), np.float32(0.582)] +2026-10-15 14:10:27.186688: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:10:27.558791: lr: 0.000802 +2026-10-15 14:10:27.559243: This epoch took 258.629495 s + +2026-10-15 14:10:27.559572: +epoch: 217 +2026-10-15 14:14:13.324489: train loss : -0.7545 +2026-10-15 14:14:18.698500: validation loss: -0.5319 +2026-10-15 14:14:18.699667: Average global foreground Dice: [np.float32(0.902), np.float32(0.5525), np.float32(0.5586), np.float32(0.5919), np.float32(0.511), np.float32(0.6005), np.float32(0.5492), np.float32(0.6053), np.float32(0.0), np.float32(0.5267), np.float32(0.4718), np.float32(0.5197), np.float32(0.6195), np.float32(0.5607), np.float32(0.0), np.float32(0.0), np.float32(0.4557), np.float32(0.3913), np.float32(0.4444), np.float32(0.4009), np.float32(0.5017), np.float32(0.4467), np.float32(0.7401), np.float32(0.6941), np.float32(0.1695), np.float32(0.2852), np.float32(0.0), np.float32(0.0), np.float32(0.0017), np.float32(0.4925), np.float32(0.0), np.float32(0.0), np.float32(0.3391), np.float32(0.3942), np.float32(0.6202), np.float32(0.6443)] +2026-10-15 14:14:18.700034: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:14:19.015210: lr: 0.000801 +2026-10-15 14:14:19.015547: This epoch took 231.455760 s + +2026-10-15 14:14:19.015733: +epoch: 218 +2026-10-15 14:18:43.304679: train loss : -0.7621 +2026-10-15 14:18:48.674906: validation loss: -0.5787 +2026-10-15 14:18:48.675939: Average global foreground Dice: [np.float32(0.9188), np.float32(0.6019), np.float32(0.7106), np.float32(0.7295), np.float32(0.6497), np.float32(0.7448), np.float32(0.6697), np.float32(0.5819), np.float32(0.0), np.float32(0.4648), np.float32(0.5097), np.float32(0.6169), np.float32(0.6078), np.float32(0.5126), np.float32(0.0), np.float32(0.0), np.float32(0.6374), np.float32(0.5282), np.float32(0.6101), np.float32(0.5753), np.float32(0.4681), np.float32(0.5529), np.float32(0.8328), np.float32(0.8124), np.float32(0.3898), np.float32(0.2923), np.float32(0.0), np.float32(0.0), np.float32(0.2746), np.float32(0.5748), np.float32(0.0), np.float32(0.0), np.float32(0.4755), np.float32(0.5132), np.float32(0.7631), np.float32(0.7754)] +2026-10-15 14:18:48.676194: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:18:49.033491: lr: 0.000801 +2026-10-15 14:18:49.033802: This epoch took 270.017985 s + +2026-10-15 14:18:49.034000: +epoch: 219 +2026-10-15 14:23:09.793128: train loss : -0.7648 +2026-10-15 14:23:15.162612: validation loss: -0.4780 +2026-10-15 14:23:15.163693: Average global foreground Dice: [np.float32(0.9117), np.float32(0.4879), np.float32(0.5258), np.float32(0.5219), np.float32(0.3985), np.float32(0.5333), np.float32(0.4759), np.float32(0.5688), np.float32(0.0), np.float32(0.6025), np.float32(0.3892), np.float32(0.4565), np.float32(0.5383), np.float32(0.3968), np.float32(0.0), np.float32(0.0), np.float32(0.3646), np.float32(0.3126), np.float32(0.3953), np.float32(0.4139), np.float32(0.4471), np.float32(0.3541), np.float32(0.6722), np.float32(0.6239), np.float32(0.2128), np.float32(0.2258), np.float32(0.0), np.float32(0.0), np.float32(0.3833), np.float32(0.5293), np.float32(0.0), np.float32(0.0), np.float32(0.383), np.float32(0.418), np.float32(0.5619), np.float32(0.5783)] +2026-10-15 14:23:15.163920: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:23:15.503847: lr: 0.0008 +2026-10-15 14:23:15.504139: This epoch took 266.470064 s + +2026-10-15 14:23:15.504317: +epoch: 220 +2026-10-15 14:26:55.779856: train loss : -0.7407 +2026-10-15 14:27:01.158585: validation loss: -0.5865 +2026-10-15 14:27:01.159553: Average global foreground Dice: [np.float32(0.8878), np.float32(0.6389), np.float32(0.7856), np.float32(0.7828), np.float32(0.7372), np.float32(0.8119), np.float32(0.7599), np.float32(0.6448), np.float32(0.0), np.float32(0.5143), np.float32(0.531), np.float32(0.6085), np.float32(0.5151), np.float32(0.436), np.float32(0.0), np.float32(0.0), np.float32(0.7228), np.float32(0.5047), np.float32(0.6675), np.float32(0.594), np.float32(0.5624), np.float32(0.6757), np.float32(0.8293), np.float32(0.8012), np.float32(0.4855), np.float32(0.378), np.float32(0.0), np.float32(0.0), np.float32(0.2834), np.float32(0.5606), np.float32(0.0), np.float32(0.0), np.float32(0.4572), np.float32(0.3301), np.float32(0.8533), np.float32(0.8633)] +2026-10-15 14:27:01.159793: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:27:01.489527: lr: 0.000799 +2026-10-15 14:27:01.489933: This epoch took 225.985532 s + +2026-10-15 14:27:01.490184: +epoch: 221 +2026-10-15 14:31:08.710114: train loss : -0.7611 +2026-10-15 14:31:14.080835: validation loss: -0.6102 +2026-10-15 14:31:14.082033: Average global foreground Dice: [np.float32(0.9189), np.float32(0.6567), np.float32(0.7679), np.float32(0.7693), np.float32(0.7134), np.float32(0.8146), np.float32(0.7236), np.float32(0.5899), np.float32(0.0), np.float32(0.518), np.float32(0.5971), np.float32(0.6216), np.float32(0.6205), np.float32(0.564), np.float32(0.0), np.float32(0.0), np.float32(0.7427), np.float32(0.6144), np.float32(0.7063), np.float32(0.6367), np.float32(0.5748), np.float32(0.5649), np.float32(0.871), np.float32(0.8516), np.float32(0.4728), np.float32(0.4441), np.float32(0.0), np.float32(0.0), np.float32(0.3523), np.float32(0.6592), np.float32(0.0), np.float32(0.0), np.float32(0.4921), np.float32(0.5144), np.float32(0.7962), np.float32(0.8091)] +2026-10-15 14:31:14.082443: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:31:14.448412: lr: 0.000798 +2026-10-15 14:31:14.448832: This epoch took 252.958517 s + +2026-10-15 14:31:14.449121: +epoch: 222 +2026-10-15 14:35:21.038770: train loss : -0.7636 +2026-10-15 14:35:26.417894: validation loss: -0.5564 +2026-10-15 14:35:26.419525: Average global foreground Dice: [np.float32(0.9149), np.float32(0.5891), np.float32(0.7015), np.float32(0.7094), np.float32(0.6263), np.float32(0.7416), np.float32(0.6366), np.float32(0.6596), np.float32(0.0), np.float32(0.5003), np.float32(0.5448), np.float32(0.5759), np.float32(0.6103), np.float32(0.4909), np.float32(0.0), np.float32(0.0), np.float32(0.689), np.float32(0.5265), np.float32(0.6572), np.float32(0.5988), np.float32(0.5829), np.float32(0.5283), np.float32(0.7308), np.float32(0.7366), np.float32(0.4969), np.float32(0.3538), np.float32(0.0), np.float32(0.0), np.float32(0.4455), np.float32(0.636), np.float32(0.0), np.float32(0.0), np.float32(0.5361), np.float32(0.4911), np.float32(0.7236), np.float32(0.7489)] +2026-10-15 14:35:26.419953: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:35:26.797944: lr: 0.000797 +2026-10-15 14:35:26.798231: This epoch took 252.348899 s + +2026-10-15 14:35:26.798332: +epoch: 223 +2026-10-15 14:39:39.424775: train loss : -0.7683 +2026-10-15 14:39:44.808348: validation loss: -0.5450 +2026-10-15 14:39:44.809304: Average global foreground Dice: [np.float32(0.9157), np.float32(0.6009), np.float32(0.6884), np.float32(0.6606), np.float32(0.5737), np.float32(0.708), np.float32(0.6193), np.float32(0.5367), np.float32(0.0), np.float32(0.5228), np.float32(0.4846), np.float32(0.5798), np.float32(0.5855), np.float32(0.4097), np.float32(0.0), np.float32(0.0), np.float32(0.5936), np.float32(0.4986), np.float32(0.5983), np.float32(0.5218), np.float32(0.5719), np.float32(0.5657), np.float32(0.803), np.float32(0.7912), np.float32(0.4413), np.float32(0.3346), np.float32(0.0), np.float32(0.0), np.float32(0.185), np.float32(0.5641), np.float32(0.0), np.float32(0.0), np.float32(0.4809), np.float32(0.5005), np.float32(0.7092), np.float32(0.7136)] +2026-10-15 14:39:44.809526: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:39:45.143478: lr: 0.000796 +2026-10-15 14:39:45.143715: This epoch took 258.345317 s + +2026-10-15 14:39:45.143793: +epoch: 224 +2026-10-15 14:43:54.253963: train loss : -0.7617 +2026-10-15 14:43:59.614998: validation loss: -0.5777 +2026-10-15 14:43:59.616102: Average global foreground Dice: [np.float32(0.9194), np.float32(0.5768), np.float32(0.7615), np.float32(0.7545), np.float32(0.6883), np.float32(0.8094), np.float32(0.7182), np.float32(0.6076), np.float32(0.0), np.float32(0.5119), np.float32(0.5356), np.float32(0.6278), np.float32(0.5041), np.float32(0.4568), np.float32(0.0), np.float32(0.0), np.float32(0.7541), np.float32(0.6039), np.float32(0.7197), np.float32(0.6726), np.float32(0.5847), np.float32(0.6449), np.float32(0.8532), np.float32(0.8454), np.float32(0.436), np.float32(0.3257), np.float32(0.0), np.float32(0.0), np.float32(0.16), np.float32(0.5193), np.float32(0.0), np.float32(0.0), np.float32(0.5365), np.float32(0.6195), np.float32(0.7794), np.float32(0.8036)] +2026-10-15 14:43:59.616512: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:43:59.968350: lr: 0.000795 +2026-10-15 14:44:00.013224: saving checkpoint... +2026-10-15 14:44:00.150325: done, saving took 0.18 seconds +2026-10-15 14:44:00.152192: This epoch took 255.008344 s + +2026-10-15 14:44:00.152411: +epoch: 225 +2026-10-15 14:48:07.265486: train loss : -0.7659 +2026-10-15 14:48:12.634393: validation loss: -0.5883 +2026-10-15 14:48:12.635556: Average global foreground Dice: [np.float32(0.9173), np.float32(0.6296), np.float32(0.7588), np.float32(0.7767), np.float32(0.6884), np.float32(0.8057), np.float32(0.7107), np.float32(0.4826), np.float32(0.0), np.float32(0.5274), np.float32(0.5859), np.float32(0.6403), np.float32(0.5759), np.float32(0.4669), np.float32(0.0), np.float32(0.0), np.float32(0.7335), np.float32(0.5712), np.float32(0.7269), np.float32(0.674), np.float32(0.6105), np.float32(0.6565), np.float32(0.8058), np.float32(0.7872), np.float32(0.4915), np.float32(0.3869), np.float32(0.0), np.float32(0.0), np.float32(0.3576), np.float32(0.6017), np.float32(0.0), np.float32(0.0), np.float32(0.3941), np.float32(0.4521), np.float32(0.7967), np.float32(0.8132)] +2026-10-15 14:48:12.635777: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:48:12.960630: lr: 0.000794 +2026-10-15 14:48:12.984798: saving checkpoint... +2026-10-15 14:48:13.115059: done, saving took 0.15 seconds +2026-10-15 14:48:13.117033: This epoch took 252.964538 s + +2026-10-15 14:48:13.117235: +epoch: 226 +2026-10-15 14:52:18.889969: train loss : -0.7680 +2026-10-15 14:52:24.264253: validation loss: -0.5995 +2026-10-15 14:52:24.265363: Average global foreground Dice: [np.float32(0.9124), np.float32(0.6344), np.float32(0.7303), np.float32(0.7702), np.float32(0.6878), np.float32(0.7979), np.float32(0.6976), np.float32(0.7879), np.float32(0.0), np.float32(0.47), np.float32(0.5448), np.float32(0.6143), np.float32(0.5517), np.float32(0.5139), np.float32(0.0), np.float32(0.0), np.float32(0.7256), np.float32(0.5549), np.float32(0.6989), np.float32(0.6271), np.float32(0.5879), np.float32(0.5846), np.float32(0.845), np.float32(0.834), np.float32(0.4712), np.float32(0.3878), np.float32(0.0), np.float32(0.0), np.float32(0.495), np.float32(0.6083), np.float32(0.0), np.float32(0.0), np.float32(0.5402), np.float32(0.4723), np.float32(0.7945), np.float32(0.8032)] +2026-10-15 14:52:24.265832: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:52:24.766582: lr: 0.000793 +2026-10-15 14:52:24.776047: saving checkpoint... +2026-10-15 14:52:24.904227: done, saving took 0.14 seconds +2026-10-15 14:52:24.906182: This epoch took 251.788870 s + +2026-10-15 14:52:24.906387: +epoch: 227 +2026-10-15 14:56:37.136294: train loss : -0.7717 +2026-10-15 14:56:42.500559: validation loss: -0.6201 +2026-10-15 14:56:42.501720: Average global foreground Dice: [np.float32(0.9239), np.float32(0.6758), np.float32(0.7651), np.float32(0.7956), np.float32(0.7125), np.float32(0.8255), np.float32(0.7139), np.float32(0.6218), np.float32(0.0), np.float32(0.5558), np.float32(0.5736), np.float32(0.626), np.float32(0.5911), np.float32(0.5553), np.float32(0.0), np.float32(0.0), np.float32(0.7509), np.float32(0.5982), np.float32(0.7243), np.float32(0.681), np.float32(0.6031), np.float32(0.6401), np.float32(0.8497), np.float32(0.8331), np.float32(0.4797), np.float32(0.4598), np.float32(0.0), np.float32(0.0), np.float32(0.1858), np.float32(0.502), np.float32(0.0), np.float32(0.0), np.float32(0.6037), np.float32(0.6077), np.float32(0.8178), np.float32(0.8399)] +2026-10-15 14:56:42.501929: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 14:56:42.825253: lr: 0.000792 +2026-10-15 14:56:42.834404: saving checkpoint... +2026-10-15 14:56:42.972509: done, saving took 0.15 seconds +2026-10-15 14:56:42.974141: This epoch took 258.067657 s + +2026-10-15 14:56:42.974317: +epoch: 228 +2026-10-15 15:00:51.164204: train loss : -0.7609 +2026-10-15 15:00:56.538038: validation loss: -0.5665 +2026-10-15 15:00:56.538955: Average global foreground Dice: [np.float32(0.9187), np.float32(0.6066), np.float32(0.7152), np.float32(0.6907), np.float32(0.6378), np.float32(0.6968), np.float32(0.6439), np.float32(0.451), np.float32(0.0), np.float32(0.4286), np.float32(0.5251), np.float32(0.5771), np.float32(0.5549), np.float32(0.5232), np.float32(0.0), np.float32(0.0), np.float32(0.6777), np.float32(0.5577), np.float32(0.6578), np.float32(0.6345), np.float32(0.5334), np.float32(0.6124), np.float32(0.8143), np.float32(0.7923), np.float32(0.4841), np.float32(0.3337), np.float32(0.0), np.float32(0.0), np.float32(0.4128), np.float32(0.6626), np.float32(0.0), np.float32(0.0), np.float32(0.4551), np.float32(0.392), np.float32(0.7404), np.float32(0.755)] +2026-10-15 15:00:56.539123: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:00:56.866993: lr: 0.000791 +2026-10-15 15:00:56.877402: saving checkpoint... +2026-10-15 15:00:57.006303: done, saving took 0.14 seconds +2026-10-15 15:00:57.008314: This epoch took 254.033918 s + +2026-10-15 15:00:57.008501: +epoch: 229 +2026-10-15 15:05:18.773022: train loss : -0.7624 +2026-10-15 15:05:24.134137: validation loss: -0.5637 +2026-10-15 15:05:24.135519: Average global foreground Dice: [np.float32(0.9011), np.float32(0.5839), np.float32(0.7087), np.float32(0.6935), np.float32(0.6426), np.float32(0.6613), np.float32(0.6193), np.float32(0.736), np.float32(0.0), np.float32(0.5543), np.float32(0.4902), np.float32(0.6205), np.float32(0.573), np.float32(0.4648), np.float32(0.0), np.float32(0.0), np.float32(0.6347), np.float32(0.5164), np.float32(0.6093), np.float32(0.6), np.float32(0.5629), np.float32(0.5876), np.float32(0.7799), np.float32(0.784), np.float32(0.4596), np.float32(0.3756), np.float32(0.0), np.float32(0.0), np.float32(0.3686), np.float32(0.5629), np.float32(0.0), np.float32(0.0), np.float32(0.3508), np.float32(0.3578), np.float32(0.7466), np.float32(0.7651)] +2026-10-15 15:05:24.135816: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:05:24.443675: lr: 0.00079 +2026-10-15 15:05:24.451386: saving checkpoint... +2026-10-15 15:05:24.582766: done, saving took 0.14 seconds +2026-10-15 15:05:24.584700: This epoch took 267.576121 s + +2026-10-15 15:05:24.584926: +epoch: 230 +2026-10-15 15:09:36.664367: train loss : -0.7557 +2026-10-15 15:09:42.045780: validation loss: -0.5082 +2026-10-15 15:09:42.046758: Average global foreground Dice: [np.float32(0.8964), np.float32(0.6045), np.float32(0.7094), np.float32(0.6766), np.float32(0.5853), np.float32(0.6851), np.float32(0.6512), np.float32(0.7412), np.float32(0.0), np.float32(0.3442), np.float32(0.4638), np.float32(0.577), np.float32(0.5668), np.float32(0.4447), np.float32(0.0), np.float32(0.0), np.float32(0.5393), np.float32(0.4733), np.float32(0.6031), np.float32(0.4903), np.float32(0.4554), np.float32(0.3879), np.float32(0.6205), np.float32(0.645), np.float32(0.387), np.float32(0.3335), np.float32(0.0), np.float32(0.0), np.float32(0.077), np.float32(0.5307), np.float32(0.0), np.float32(0.0), np.float32(0.3717), np.float32(0.4757), np.float32(0.5769), np.float32(0.6876)] +2026-10-15 15:09:42.047098: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:09:42.374478: lr: 0.000789 +2026-10-15 15:09:42.374884: This epoch took 257.789856 s + +2026-10-15 15:09:42.375074: +epoch: 231 +2026-10-15 15:13:41.479865: train loss : -0.7534 +2026-10-15 15:13:46.862889: validation loss: -0.3422 +2026-10-15 15:13:46.863825: Average global foreground Dice: [np.float32(0.9071), np.float32(0.3102), np.float32(0.2271), np.float32(0.3338), np.float32(0.2364), np.float32(0.3217), np.float32(0.2178), np.float32(0.3301), np.float32(0.0), np.float32(0.5286), np.float32(0.2246), np.float32(0.3043), np.float32(0.3789), np.float32(0.1998), np.float32(0.0), np.float32(0.0), np.float32(0.2708), np.float32(0.2368), np.float32(0.2175), np.float32(0.1911), np.float32(0.2319), np.float32(0.2445), np.float32(0.3869), np.float32(0.4756), np.float32(0.1549), np.float32(0.093), np.float32(0.0), np.float32(0.0), np.float32(0.0977), np.float32(0.3507), np.float32(0.0), np.float32(0.0), np.float32(0.2281), np.float32(0.36), np.float32(0.3265), np.float32(0.401)] +2026-10-15 15:13:46.864198: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:13:47.188364: lr: 0.000789 +2026-10-15 15:13:47.188621: This epoch took 244.813460 s + +2026-10-15 15:13:47.188701: +epoch: 232 +2026-10-15 15:18:09.133425: train loss : -0.7557 +2026-10-15 15:18:14.528033: validation loss: -0.4816 +2026-10-15 15:18:14.529227: Average global foreground Dice: [np.float32(0.9066), np.float32(0.4746), np.float32(0.5413), np.float32(0.5723), np.float32(0.5112), np.float32(0.5864), np.float32(0.5154), np.float32(0.4595), np.float32(0.0), np.float32(0.4855), np.float32(0.423), np.float32(0.4944), np.float32(0.4854), np.float32(0.4077), np.float32(0.0), np.float32(0.0), np.float32(0.6152), np.float32(0.5291), np.float32(0.5898), np.float32(0.5459), np.float32(0.4533), np.float32(0.4263), np.float32(0.5681), np.float32(0.6102), np.float32(0.3917), np.float32(0.3385), np.float32(0.0), np.float32(0.0), np.float32(0.1742), np.float32(0.5695), np.float32(0.0), np.float32(0.0), np.float32(0.5118), np.float32(0.4571), np.float32(0.5449), np.float32(0.5911)] +2026-10-15 15:18:14.529411: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:18:14.855673: lr: 0.000788 +2026-10-15 15:18:14.855935: This epoch took 267.667182 s + +2026-10-15 15:18:14.856026: +epoch: 233 +2026-10-15 15:23:02.410480: train loss : -0.7663 +2026-10-15 15:23:07.783528: validation loss: -0.4876 +2026-10-15 15:23:07.784814: Average global foreground Dice: [np.float32(0.9073), np.float32(0.5086), np.float32(0.5678), np.float32(0.5841), np.float32(0.5474), np.float32(0.6237), np.float32(0.5771), np.float32(0.4909), np.float32(0.0), np.float32(0.5324), np.float32(0.4416), np.float32(0.5359), np.float32(0.5296), np.float32(0.3707), np.float32(0.0), np.float32(0.0), np.float32(0.5545), np.float32(0.4318), np.float32(0.5531), np.float32(0.5216), np.float32(0.4566), np.float32(0.4099), np.float32(0.6487), np.float32(0.5982), np.float32(0.3418), np.float32(0.2354), np.float32(0.0), np.float32(0.0), np.float32(0.3358), np.float32(0.5862), np.float32(0.0), np.float32(0.0), np.float32(0.3903), np.float32(0.3444), np.float32(0.5772), np.float32(0.5815)] +2026-10-15 15:23:07.785515: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:23:08.100740: lr: 0.000787 +2026-10-15 15:23:08.100988: This epoch took 293.244886 s + +2026-10-15 15:23:08.101218: +epoch: 234 +2026-10-15 15:26:54.724004: train loss : -0.7682 +2026-10-15 15:27:00.097998: validation loss: -0.5371 +2026-10-15 15:27:00.098994: Average global foreground Dice: [np.float32(0.9093), np.float32(0.5789), np.float32(0.6587), np.float32(0.678), np.float32(0.5871), np.float32(0.7137), np.float32(0.638), np.float32(0.5136), np.float32(0.0), np.float32(0.5074), np.float32(0.4988), np.float32(0.6243), np.float32(0.5333), np.float32(0.4339), np.float32(0.0), np.float32(0.0), np.float32(0.6195), np.float32(0.482), np.float32(0.6332), np.float32(0.5773), np.float32(0.5191), np.float32(0.4448), np.float32(0.7489), np.float32(0.7037), np.float32(0.4191), np.float32(0.3248), np.float32(0.0), np.float32(0.0), np.float32(0.3629), np.float32(0.6143), np.float32(0.0), np.float32(0.0), np.float32(0.3428), np.float32(0.3998), np.float32(0.6622), np.float32(0.6847)] +2026-10-15 15:27:00.100001: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:27:00.460853: lr: 0.000786 +2026-10-15 15:27:00.461090: This epoch took 232.359772 s + +2026-10-15 15:27:00.461175: +epoch: 235 +2026-10-15 15:31:21.922833: train loss : -0.7618 +2026-10-15 15:31:27.309381: validation loss: -0.5183 +2026-10-15 15:31:27.310457: Average global foreground Dice: [np.float32(0.9058), np.float32(0.5607), np.float32(0.6464), np.float32(0.6448), np.float32(0.5919), np.float32(0.6543), np.float32(0.6303), np.float32(0.4617), np.float32(0.0), np.float32(0.4204), np.float32(0.4012), np.float32(0.5159), np.float32(0.4965), np.float32(0.4375), np.float32(0.0), np.float32(0.0), np.float32(0.6451), np.float32(0.5011), np.float32(0.6357), np.float32(0.5558), np.float32(0.4888), np.float32(0.4732), np.float32(0.7083), np.float32(0.7373), np.float32(0.468), np.float32(0.3893), np.float32(0.0), np.float32(0.0), np.float32(0.2953), np.float32(0.5732), np.float32(0.0), np.float32(0.0), np.float32(0.422), np.float32(0.4212), np.float32(0.6818), np.float32(0.7383)] +2026-10-15 15:31:27.310906: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:31:27.651129: lr: 0.000785 +2026-10-15 15:31:27.651582: This epoch took 267.190219 s + +2026-10-15 15:31:27.651872: +epoch: 236 +2026-10-15 15:35:32.282083: train loss : -0.7688 +2026-10-15 15:35:37.653586: validation loss: -0.5896 +2026-10-15 15:35:37.654767: Average global foreground Dice: [np.float32(0.9053), np.float32(0.6071), np.float32(0.7462), np.float32(0.7646), np.float32(0.6792), np.float32(0.7896), np.float32(0.7035), np.float32(0.3741), np.float32(0.0), np.float32(0.3573), np.float32(0.5714), np.float32(0.6404), np.float32(0.5629), np.float32(0.4834), np.float32(0.0), np.float32(0.0), np.float32(0.7351), np.float32(0.564), np.float32(0.7327), np.float32(0.676), np.float32(0.5252), np.float32(0.6319), np.float32(0.8471), np.float32(0.8204), np.float32(0.4895), np.float32(0.3719), np.float32(0.0), np.float32(0.0), np.float32(0.1579), np.float32(0.6329), np.float32(0.0), np.float32(0.0), np.float32(0.3489), np.float32(0.444), np.float32(0.8102), np.float32(0.8308)] +2026-10-15 15:35:37.655338: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:35:37.979471: lr: 0.000784 +2026-10-15 15:35:37.979807: This epoch took 250.327717 s + +2026-10-15 15:35:37.979939: +epoch: 237 +2026-10-15 15:39:34.703265: train loss : -0.7704 +2026-10-15 15:39:40.076277: validation loss: -0.5801 +2026-10-15 15:39:40.077178: Average global foreground Dice: [np.float32(0.9079), np.float32(0.6189), np.float32(0.7423), np.float32(0.7559), np.float32(0.6681), np.float32(0.7844), np.float32(0.6914), np.float32(0.6624), np.float32(0.0), np.float32(0.5611), np.float32(0.5443), np.float32(0.6328), np.float32(0.5288), np.float32(0.4388), np.float32(0.0), np.float32(0.0), np.float32(0.7369), np.float32(0.6016), np.float32(0.7149), np.float32(0.6728), np.float32(0.5552), np.float32(0.5479), np.float32(0.8231), np.float32(0.7918), np.float32(0.4994), np.float32(0.3532), np.float32(0.0), np.float32(0.0), np.float32(0.3867), np.float32(0.5857), np.float32(0.0), np.float32(0.0), np.float32(0.5286), np.float32(0.5027), np.float32(0.7672), np.float32(0.7948)] +2026-10-15 15:39:40.077454: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:39:40.388060: lr: 0.000783 +2026-10-15 15:39:40.388498: This epoch took 242.408481 s + +2026-10-15 15:39:40.388790: +epoch: 238 +2026-10-15 15:43:43.473937: train loss : -0.7718 +2026-10-15 15:43:48.861720: validation loss: -0.5711 +2026-10-15 15:43:48.862799: Average global foreground Dice: [np.float32(0.9005), np.float32(0.6039), np.float32(0.7363), np.float32(0.7439), np.float32(0.6355), np.float32(0.7763), np.float32(0.6769), np.float32(0.7974), np.float32(0.0), np.float32(0.4437), np.float32(0.5662), np.float32(0.6362), np.float32(0.5603), np.float32(0.4868), np.float32(0.0), np.float32(0.0), np.float32(0.6255), np.float32(0.4689), np.float32(0.623), np.float32(0.5712), np.float32(0.389), np.float32(0.4747), np.float32(0.7393), np.float32(0.7255), np.float32(0.4155), np.float32(0.3992), np.float32(0.0), np.float32(0.0), np.float32(0.4404), np.float32(0.6132), np.float32(0.0), np.float32(0.0), np.float32(0.4857), np.float32(0.4752), np.float32(0.73), np.float32(0.7218)] +2026-10-15 15:43:48.863530: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:43:49.181345: lr: 0.000782 +2026-10-15 15:43:49.181608: This epoch took 248.792612 s + +2026-10-15 15:43:49.181701: +epoch: 239 +2026-10-15 15:48:08.337841: train loss : -0.7605 +2026-10-15 15:48:13.705572: validation loss: -0.5862 +2026-10-15 15:48:13.706566: Average global foreground Dice: [np.float32(0.9091), np.float32(0.6036), np.float32(0.7832), np.float32(0.8037), np.float32(0.7195), np.float32(0.8179), np.float32(0.7607), np.float32(0.3784), np.float32(0.0), np.float32(0.2463), np.float32(0.5726), np.float32(0.6459), np.float32(0.6409), np.float32(0.4972), np.float32(0.0), np.float32(0.0), np.float32(0.7643), np.float32(0.5802), np.float32(0.7366), np.float32(0.6791), np.float32(0.5196), np.float32(0.6326), np.float32(0.822), np.float32(0.808), np.float32(0.5059), np.float32(0.3978), np.float32(0.0), np.float32(0.0), np.float32(0.0224), np.float32(0.3513), np.float32(0.0), np.float32(0.0), np.float32(0.4616), np.float32(0.5639), np.float32(0.8377), np.float32(0.8391)] +2026-10-15 15:48:13.706731: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:48:14.033838: lr: 0.000781 +2026-10-15 15:48:14.034115: This epoch took 264.852343 s + +2026-10-15 15:48:14.034216: +epoch: 240 +2026-10-15 15:52:25.351550: train loss : -0.7606 +2026-10-15 15:52:30.719786: validation loss: -0.5798 +2026-10-15 15:52:30.720683: Average global foreground Dice: [np.float32(0.9206), np.float32(0.5794), np.float32(0.786), np.float32(0.7978), np.float32(0.7469), np.float32(0.8406), np.float32(0.7737), np.float32(0.5815), np.float32(0.0), np.float32(0.4388), np.float32(0.469), np.float32(0.536), np.float32(0.5634), np.float32(0.2938), np.float32(0.0), np.float32(0.0), np.float32(0.7734), np.float32(0.6011), np.float32(0.7508), np.float32(0.6773), np.float32(0.449), np.float32(0.6399), np.float32(0.8335), np.float32(0.8322), np.float32(0.4895), np.float32(0.4584), np.float32(0.0), np.float32(0.0), np.float32(0.2315), np.float32(0.4007), np.float32(0.0), np.float32(0.0), np.float32(0.5873), np.float32(0.6185), np.float32(0.8485), np.float32(0.8621)] +2026-10-15 15:52:30.720840: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:52:31.044146: lr: 0.00078 +2026-10-15 15:52:31.044384: This epoch took 257.010100 s + +2026-10-15 15:52:31.044477: +epoch: 241 +2026-10-15 15:56:30.600777: train loss : -0.7719 +2026-10-15 15:56:35.960133: validation loss: -0.5543 +2026-10-15 15:56:35.961195: Average global foreground Dice: [np.float32(0.9211), np.float32(0.5762), np.float32(0.6848), np.float32(0.688), np.float32(0.5718), np.float32(0.73), np.float32(0.6255), np.float32(0.6473), np.float32(0.0), np.float32(0.6167), np.float32(0.5287), np.float32(0.5874), np.float32(0.5172), np.float32(0.405), np.float32(0.0), np.float32(0.0), np.float32(0.6802), np.float32(0.5211), np.float32(0.6632), np.float32(0.653), np.float32(0.5519), np.float32(0.5548), np.float32(0.7749), np.float32(0.7605), np.float32(0.483), np.float32(0.3623), np.float32(0.0), np.float32(0.0), np.float32(0.4183), np.float32(0.5578), np.float32(0.0), np.float32(0.0), np.float32(0.439), np.float32(0.5644), np.float32(0.7204), np.float32(0.7516)] +2026-10-15 15:56:35.961375: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 15:56:36.282116: lr: 0.000779 +2026-10-15 15:56:36.282448: This epoch took 245.237915 s + +2026-10-15 15:56:36.282706: +epoch: 242 +2026-10-15 16:00:44.136266: train loss : -0.7717 +2026-10-15 16:00:49.537597: validation loss: -0.6031 +2026-10-15 16:00:49.538785: Average global foreground Dice: [np.float32(0.9084), np.float32(0.6466), np.float32(0.7564), np.float32(0.7944), np.float32(0.6912), np.float32(0.8204), np.float32(0.7219), np.float32(0.6945), np.float32(0.0), np.float32(0.5722), np.float32(0.6469), np.float32(0.6499), np.float32(0.6066), np.float32(0.5223), np.float32(0.0), np.float32(0.0), np.float32(0.7586), np.float32(0.5968), np.float32(0.7265), np.float32(0.6788), np.float32(0.6191), np.float32(0.5842), np.float32(0.838), np.float32(0.8055), np.float32(0.5312), np.float32(0.4135), np.float32(0.0), np.float32(0.0), np.float32(0.3381), np.float32(0.4769), np.float32(0.0), np.float32(0.0), np.float32(0.5749), np.float32(0.5314), np.float32(0.8185), np.float32(0.8281)] +2026-10-15 16:00:49.539268: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:00:49.872133: lr: 0.000778 +2026-10-15 16:00:49.872540: This epoch took 253.589745 s + +2026-10-15 16:00:49.872819: +epoch: 243 +2026-10-15 16:05:42.665909: train loss : -0.7714 +2026-10-15 16:05:48.030516: validation loss: -0.6243 +2026-10-15 16:05:48.031964: Average global foreground Dice: [np.float32(0.9209), np.float32(0.6927), np.float32(0.7595), np.float32(0.7864), np.float32(0.6911), np.float32(0.8249), np.float32(0.7104), np.float32(0.6706), np.float32(0.0), np.float32(0.5367), np.float32(0.6056), np.float32(0.622), np.float32(0.6171), np.float32(0.5286), np.float32(0.0), np.float32(0.0), np.float32(0.7389), np.float32(0.5842), np.float32(0.7093), np.float32(0.6596), np.float32(0.6128), np.float32(0.5687), np.float32(0.8618), np.float32(0.8333), np.float32(0.445), np.float32(0.4532), np.float32(0.0), np.float32(0.0), np.float32(0.2947), np.float32(0.5776), np.float32(0.0), np.float32(0.0), np.float32(0.5513), np.float32(0.5813), np.float32(0.83), np.float32(0.8452)] +2026-10-15 16:05:48.032226: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:05:48.363753: lr: 0.000777 +2026-10-15 16:05:48.371969: saving checkpoint... +2026-10-15 16:05:48.507338: done, saving took 0.14 seconds +2026-10-15 16:05:48.509203: This epoch took 298.636198 s + +2026-10-15 16:05:48.509437: +epoch: 244 +2026-10-15 16:09:52.194821: train loss : -0.7626 +2026-10-15 16:09:57.558500: validation loss: -0.4885 +2026-10-15 16:09:57.559492: Average global foreground Dice: [np.float32(0.8917), np.float32(0.4838), np.float32(0.5049), np.float32(0.5152), np.float32(0.3991), np.float32(0.5549), np.float32(0.4751), np.float32(0.6166), np.float32(0.0), np.float32(0.3767), np.float32(0.4116), np.float32(0.4151), np.float32(0.5516), np.float32(0.5013), np.float32(0.0), np.float32(0.0), np.float32(0.366), np.float32(0.2883), np.float32(0.37), np.float32(0.3619), np.float32(0.4867), np.float32(0.4311), np.float32(0.7935), np.float32(0.7658), np.float32(0.1715), np.float32(0.2191), np.float32(0.0), np.float32(0.0), np.float32(0.1532), np.float32(0.4736), np.float32(0.0), np.float32(0.0), np.float32(0.4262), np.float32(0.4225), np.float32(0.5888), np.float32(0.6271)] +2026-10-15 16:09:57.559807: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:09:57.891454: lr: 0.000777 +2026-10-15 16:09:57.891737: This epoch took 249.382096 s + +2026-10-15 16:09:57.891817: +epoch: 245 +2026-10-15 16:14:26.541878: train loss : -0.7473 +2026-10-15 16:14:31.906451: validation loss: -0.5822 +2026-10-15 16:14:31.908007: Average global foreground Dice: [np.float32(0.9075), np.float32(0.6088), np.float32(0.7337), np.float32(0.6816), np.float32(0.6239), np.float32(0.7077), np.float32(0.6663), np.float32(0.6742), np.float32(0.0), np.float32(0.373), np.float32(0.5582), np.float32(0.5803), np.float32(0.5616), np.float32(0.5196), np.float32(0.0), np.float32(0.0), np.float32(0.5674), np.float32(0.4478), np.float32(0.5876), np.float32(0.5355), np.float32(0.5523), np.float32(0.5163), np.float32(0.8721), np.float32(0.8802), np.float32(0.4305), np.float32(0.4394), np.float32(0.0), np.float32(0.0), np.float32(0.2609), np.float32(0.4739), np.float32(0.0), np.float32(0.0), np.float32(0.4321), np.float32(0.5177), np.float32(0.8217), np.float32(0.8386)] +2026-10-15 16:14:31.908420: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:14:32.233788: lr: 0.000776 +2026-10-15 16:14:32.234468: This epoch took 274.342590 s + +2026-10-15 16:14:32.234731: +epoch: 246 +2026-10-15 16:18:51.665642: train loss : -0.7524 +2026-10-15 16:18:57.025818: validation loss: -0.4997 +2026-10-15 16:18:57.027062: Average global foreground Dice: [np.float32(0.9002), np.float32(0.5078), np.float32(0.5484), np.float32(0.5162), np.float32(0.4651), np.float32(0.5512), np.float32(0.4912), np.float32(0.6442), np.float32(0.0), np.float32(0.5716), np.float32(0.4032), np.float32(0.4575), np.float32(0.5306), np.float32(0.5049), np.float32(0.0), np.float32(0.0), np.float32(0.4005), np.float32(0.3304), np.float32(0.3969), np.float32(0.3631), np.float32(0.4434), np.float32(0.3977), np.float32(0.8213), np.float32(0.8189), np.float32(0.2928), np.float32(0.3095), np.float32(0.0), np.float32(0.0), np.float32(0.2687), np.float32(0.5465), np.float32(0.0), np.float32(0.0), np.float32(0.3965), np.float32(0.4491), np.float32(0.6687), np.float32(0.7228)] +2026-10-15 16:18:57.027444: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:18:57.363581: lr: 0.000775 +2026-10-15 16:18:57.363876: This epoch took 265.129050 s + +2026-10-15 16:18:57.364160: +epoch: 247 +2026-10-15 16:23:03.095651: train loss : -0.7641 +2026-10-15 16:23:08.486829: validation loss: -0.5562 +2026-10-15 16:23:08.487842: Average global foreground Dice: [np.float32(0.8932), np.float32(0.6173), np.float32(0.668), np.float32(0.6572), np.float32(0.5771), np.float32(0.7121), np.float32(0.6064), np.float32(0.7309), np.float32(0.0), np.float32(0.4705), np.float32(0.4964), np.float32(0.5531), np.float32(0.5677), np.float32(0.5319), np.float32(0.0), np.float32(0.0), np.float32(0.5707), np.float32(0.4649), np.float32(0.5622), np.float32(0.5078), np.float32(0.5955), np.float32(0.5511), np.float32(0.8111), np.float32(0.7774), np.float32(0.3485), np.float32(0.3774), np.float32(0.0), np.float32(0.0), np.float32(0.1746), np.float32(0.4975), np.float32(0.0), np.float32(0.0), np.float32(0.4786), np.float32(0.5159), np.float32(0.7418), np.float32(0.7736)] +2026-10-15 16:23:08.488114: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:23:08.815042: lr: 0.000774 +2026-10-15 16:23:08.815504: This epoch took 251.451225 s + +2026-10-15 16:23:08.815695: +epoch: 248 +2026-10-15 16:27:25.483648: train loss : -0.7678 +2026-10-15 16:27:30.845421: validation loss: -0.5688 +2026-10-15 16:27:30.846638: Average global foreground Dice: [np.float32(0.9155), np.float32(0.6366), np.float32(0.7055), np.float32(0.689), np.float32(0.6459), np.float32(0.7308), np.float32(0.6382), np.float32(0.7626), np.float32(0.0), np.float32(0.4994), np.float32(0.5359), np.float32(0.5561), np.float32(0.5885), np.float32(0.5234), np.float32(0.0), np.float32(0.0), np.float32(0.6633), np.float32(0.5208), np.float32(0.6177), np.float32(0.5676), np.float32(0.5531), np.float32(0.506), np.float32(0.8035), np.float32(0.7638), np.float32(0.3758), np.float32(0.4533), np.float32(0.0), np.float32(0.0), np.float32(0.3466), np.float32(0.4984), np.float32(0.0), np.float32(0.0), np.float32(0.2859), np.float32(0.1792), np.float32(0.7481), np.float32(0.7693)] +2026-10-15 16:27:30.847419: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:27:31.183757: lr: 0.000773 +2026-10-15 16:27:31.184144: This epoch took 262.368360 s + +2026-10-15 16:27:31.184289: +epoch: 249 +2026-10-15 16:32:15.629494: train loss : -0.7619 +2026-10-15 16:32:20.998749: validation loss: -0.6120 +2026-10-15 16:32:20.999948: Average global foreground Dice: [np.float32(0.9167), np.float32(0.6754), np.float32(0.762), np.float32(0.7611), np.float32(0.7077), np.float32(0.8159), np.float32(0.7201), np.float32(0.7096), np.float32(0.0), np.float32(0.4934), np.float32(0.6054), np.float32(0.6601), np.float32(0.5971), np.float32(0.5443), np.float32(0.0), np.float32(0.0), np.float32(0.7403), np.float32(0.6069), np.float32(0.7149), np.float32(0.643), np.float32(0.5401), np.float32(0.5884), np.float32(0.8151), np.float32(0.7989), np.float32(0.4791), np.float32(0.447), np.float32(0.0), np.float32(0.0), np.float32(0.2519), np.float32(0.516), np.float32(0.0), np.float32(0.0), np.float32(0.5641), np.float32(0.6276), np.float32(0.8086), np.float32(0.8224)] +2026-10-15 16:32:21.000381: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:32:21.325332: lr: 0.000772 +2026-10-15 16:32:21.325553: saving scheduled checkpoint file... +2026-10-15 16:32:21.333054: saving checkpoint... +2026-10-15 16:32:21.462954: done, saving took 0.14 seconds +2026-10-15 16:32:21.464952: done +2026-10-15 16:32:21.465198: This epoch took 290.280850 s + +2026-10-15 16:32:21.465286: +epoch: 250 +2026-10-15 16:36:29.357851: train loss : -0.7690 +2026-10-15 16:36:34.720399: validation loss: -0.5857 +2026-10-15 16:36:34.721477: Average global foreground Dice: [np.float32(0.9022), np.float32(0.6161), np.float32(0.7386), np.float32(0.7526), np.float32(0.6761), np.float32(0.7909), np.float32(0.716), np.float32(0.6445), np.float32(0.0), np.float32(0.509), np.float32(0.5478), np.float32(0.6393), np.float32(0.5661), np.float32(0.446), np.float32(0.0), np.float32(0.0), np.float32(0.7297), np.float32(0.6185), np.float32(0.7212), np.float32(0.6835), np.float32(0.57), np.float32(0.6084), np.float32(0.8774), np.float32(0.8506), np.float32(0.5063), np.float32(0.3674), np.float32(0.0), np.float32(0.0), np.float32(0.3543), np.float32(0.5242), np.float32(0.0), np.float32(0.0), np.float32(0.427), np.float32(0.576), np.float32(0.8153), np.float32(0.8301)] +2026-10-15 16:36:34.721637: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:36:35.046641: lr: 0.000771 +2026-10-15 16:36:35.047106: This epoch took 253.581753 s + +2026-10-15 16:36:35.047319: +epoch: 251 +2026-10-15 16:40:51.580335: train loss : -0.7762 +2026-10-15 16:40:56.946488: validation loss: -0.5713 +2026-10-15 16:40:56.947816: Average global foreground Dice: [np.float32(0.9187), np.float32(0.6222), np.float32(0.749), np.float32(0.7688), np.float32(0.69), np.float32(0.7805), np.float32(0.7139), np.float32(0.6415), np.float32(0.0), np.float32(0.4731), np.float32(0.4985), np.float32(0.6082), np.float32(0.5519), np.float32(0.4539), np.float32(0.0), np.float32(0.0), np.float32(0.7292), np.float32(0.5194), np.float32(0.7087), np.float32(0.6271), np.float32(0.5309), np.float32(0.6125), np.float32(0.8278), np.float32(0.8036), np.float32(0.503), np.float32(0.3645), np.float32(0.0), np.float32(0.0), np.float32(0.3742), np.float32(0.3398), np.float32(0.0), np.float32(0.0), np.float32(0.4902), np.float32(0.6047), np.float32(0.7974), np.float32(0.8204)] +2026-10-15 16:40:56.948294: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:40:57.284004: lr: 0.00077 +2026-10-15 16:40:57.284280: This epoch took 262.236883 s + +2026-10-15 16:40:57.284369: +epoch: 252 +2026-10-15 16:44:49.767235: train loss : -0.7775 +2026-10-15 16:44:55.116945: validation loss: -0.5651 +2026-10-15 16:44:55.118108: Average global foreground Dice: [np.float32(0.913), np.float32(0.5933), np.float32(0.709), np.float32(0.6835), np.float32(0.6384), np.float32(0.7405), np.float32(0.6997), np.float32(0.3972), np.float32(0.0), np.float32(0.5109), np.float32(0.4961), np.float32(0.6078), np.float32(0.5292), np.float32(0.4349), np.float32(0.0), np.float32(0.0), np.float32(0.6588), np.float32(0.5352), np.float32(0.6864), np.float32(0.56), np.float32(0.6224), np.float32(0.433), np.float32(0.8154), np.float32(0.7889), np.float32(0.4592), np.float32(0.3817), np.float32(0.0), np.float32(0.0), np.float32(0.2776), np.float32(0.5852), np.float32(0.0), np.float32(0.0), np.float32(0.4614), np.float32(0.5836), np.float32(0.7478), np.float32(0.7887)] +2026-10-15 16:44:55.118626: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:44:55.507375: lr: 0.000769 +2026-10-15 16:44:55.507683: This epoch took 238.223254 s + +2026-10-15 16:44:55.507919: +epoch: 253 +2026-10-15 16:49:16.718421: train loss : -0.7775 +2026-10-15 16:49:22.075392: validation loss: -0.5330 +2026-10-15 16:49:22.076551: Average global foreground Dice: [np.float32(0.911), np.float32(0.6021), np.float32(0.6606), np.float32(0.6111), np.float32(0.5327), np.float32(0.6732), np.float32(0.5769), np.float32(0.564), np.float32(0.0), np.float32(0.521), np.float32(0.4508), np.float32(0.503), np.float32(0.5665), np.float32(0.5403), np.float32(0.0), np.float32(0.0), np.float32(0.5224), np.float32(0.4064), np.float32(0.5137), np.float32(0.4755), np.float32(0.5321), np.float32(0.4132), np.float32(0.7663), np.float32(0.7291), np.float32(0.3093), np.float32(0.3732), np.float32(0.0), np.float32(0.0), np.float32(0.2182), np.float32(0.4081), np.float32(0.0), np.float32(0.0), np.float32(0.4846), np.float32(0.4873), np.float32(0.6669), np.float32(0.6962)] +2026-10-15 16:49:22.077201: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:49:22.401753: lr: 0.000768 +2026-10-15 16:49:22.402215: This epoch took 266.893995 s + +2026-10-15 16:49:22.402349: +epoch: 254 +2026-10-15 16:53:51.232528: train loss : -0.7752 +2026-10-15 16:53:56.600036: validation loss: -0.5247 +2026-10-15 16:53:56.601133: Average global foreground Dice: [np.float32(0.915), np.float32(0.579), np.float32(0.6268), np.float32(0.5759), np.float32(0.4819), np.float32(0.6137), np.float32(0.553), np.float32(0.6278), np.float32(0.0), np.float32(0.3629), np.float32(0.4381), np.float32(0.503), np.float32(0.5966), np.float32(0.5621), np.float32(0.0), np.float32(0.0), np.float32(0.4881), np.float32(0.4366), np.float32(0.5067), np.float32(0.493), np.float32(0.4273), np.float32(0.4498), np.float32(0.768), np.float32(0.7293), np.float32(0.2528), np.float32(0.3492), np.float32(0.0), np.float32(0.0), np.float32(0.0), np.float32(0.5308), np.float32(0.0), np.float32(0.0), np.float32(0.4654), np.float32(0.4334), np.float32(0.6294), np.float32(0.6593)] +2026-10-15 16:53:56.601502: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:53:56.931118: lr: 0.000767 +2026-10-15 16:53:56.931582: This epoch took 274.529157 s + +2026-10-15 16:53:56.931831: +epoch: 255 +2026-10-15 16:58:00.052852: train loss : -0.7740 +2026-10-15 16:58:05.410902: validation loss: -0.5409 +2026-10-15 16:58:05.411943: Average global foreground Dice: [np.float32(0.9117), np.float32(0.6268), np.float32(0.6887), np.float32(0.6924), np.float32(0.6285), np.float32(0.6976), np.float32(0.6439), np.float32(0.6871), np.float32(0.0), np.float32(0.427), np.float32(0.4874), np.float32(0.5028), np.float32(0.5518), np.float32(0.4471), np.float32(0.0), np.float32(0.0), np.float32(0.663), np.float32(0.5196), np.float32(0.6393), np.float32(0.5968), np.float32(0.5101), np.float32(0.5175), np.float32(0.7642), np.float32(0.7188), np.float32(0.4124), np.float32(0.3336), np.float32(0.0), np.float32(0.0), np.float32(0.1274), np.float32(0.5793), np.float32(0.0), np.float32(0.0), np.float32(0.4582), np.float32(0.4784), np.float32(0.7108), np.float32(0.7297)] +2026-10-15 16:58:05.412113: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 16:58:05.746866: lr: 0.000766 +2026-10-15 16:58:05.747368: This epoch took 248.815243 s + +2026-10-15 16:58:05.747707: +epoch: 256 +2026-10-15 17:02:44.291313: train loss : -0.7576 +2026-10-15 17:02:49.650130: validation loss: -0.5664 +2026-10-15 17:02:49.651297: Average global foreground Dice: [np.float32(0.9218), np.float32(0.5647), np.float32(0.7452), np.float32(0.7552), np.float32(0.6784), np.float32(0.7955), np.float32(0.6983), np.float32(0.599), np.float32(0.0), np.float32(0.5811), np.float32(0.5398), np.float32(0.5707), np.float32(0.5082), np.float32(0.3789), np.float32(0.0), np.float32(0.0), np.float32(0.7096), np.float32(0.5665), np.float32(0.7127), np.float32(0.6641), np.float32(0.5595), np.float32(0.5934), np.float32(0.82), np.float32(0.7931), np.float32(0.5188), np.float32(0.3838), np.float32(0.0), np.float32(0.0), np.float32(0.0747), np.float32(0.4148), np.float32(0.0), np.float32(0.0), np.float32(0.5327), np.float32(0.5728), np.float32(0.7934), np.float32(0.8162)] +2026-10-15 17:02:49.651712: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:02:49.983600: lr: 0.000765 +2026-10-15 17:02:49.984083: This epoch took 284.236075 s + +2026-10-15 17:02:49.984307: +epoch: 257 +2026-10-15 17:06:54.872167: train loss : -0.7774 +2026-10-15 17:07:00.264598: validation loss: -0.6058 +2026-10-15 17:07:00.265885: Average global foreground Dice: [np.float32(0.9153), np.float32(0.6213), np.float32(0.7441), np.float32(0.7647), np.float32(0.6846), np.float32(0.7921), np.float32(0.6985), np.float32(0.6789), np.float32(0.0), np.float32(0.506), np.float32(0.5694), np.float32(0.6289), np.float32(0.5611), np.float32(0.5109), np.float32(0.0), np.float32(0.0), np.float32(0.7173), np.float32(0.5973), np.float32(0.702), np.float32(0.6457), np.float32(0.6036), np.float32(0.5485), np.float32(0.8785), np.float32(0.8656), np.float32(0.4821), np.float32(0.478), np.float32(0.0), np.float32(0.0), np.float32(0.5667), np.float32(0.6573), np.float32(0.0), np.float32(0.0), np.float32(0.5718), np.float32(0.6089), np.float32(0.7896), np.float32(0.8112)] +2026-10-15 17:07:00.266104: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:07:00.597180: lr: 0.000764 +2026-10-15 17:07:00.597684: This epoch took 250.613268 s + +2026-10-15 17:07:00.597998: +epoch: 258 +2026-10-15 17:11:33.265773: train loss : -0.7755 +2026-10-15 17:11:38.629906: validation loss: -0.5865 +2026-10-15 17:11:38.631008: Average global foreground Dice: [np.float32(0.9114), np.float32(0.6314), np.float32(0.7291), np.float32(0.7421), np.float32(0.6573), np.float32(0.7489), np.float32(0.679), np.float32(0.6371), np.float32(0.0), np.float32(0.3989), np.float32(0.5637), np.float32(0.6061), np.float32(0.5791), np.float32(0.5433), np.float32(0.0), np.float32(0.0), np.float32(0.6963), np.float32(0.5979), np.float32(0.6881), np.float32(0.6283), np.float32(0.606), np.float32(0.5944), np.float32(0.8193), np.float32(0.7953), np.float32(0.4773), np.float32(0.4167), np.float32(0.0), np.float32(0.0), np.float32(0.4495), np.float32(0.5833), np.float32(0.0), np.float32(0.0), np.float32(0.5939), np.float32(0.6289), np.float32(0.7713), np.float32(0.7971)] +2026-10-15 17:11:38.631502: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:11:38.959782: lr: 0.000764 +2026-10-15 17:11:38.967833: saving checkpoint... +2026-10-15 17:11:39.111007: done, saving took 0.15 seconds +2026-10-15 17:11:39.112948: This epoch took 278.514669 s + +2026-10-15 17:11:39.113144: +epoch: 259 +2026-10-15 17:15:59.948465: train loss : -0.7703 +2026-10-15 17:16:05.350879: validation loss: -0.5937 +2026-10-15 17:16:05.351915: Average global foreground Dice: [np.float32(0.8997), np.float32(0.6286), np.float32(0.7988), np.float32(0.801), np.float32(0.7329), np.float32(0.8371), np.float32(0.7539), np.float32(0.6409), np.float32(0.0), np.float32(0.4087), np.float32(0.5953), np.float32(0.6142), np.float32(0.5359), np.float32(0.4312), np.float32(0.0), np.float32(0.0), np.float32(0.7449), np.float32(0.6004), np.float32(0.7311), np.float32(0.6485), np.float32(0.5805), np.float32(0.6669), np.float32(0.8853), np.float32(0.8615), np.float32(0.5286), np.float32(0.3775), np.float32(0.0), np.float32(0.0), np.float32(0.2008), np.float32(0.5723), np.float32(0.0), np.float32(0.0), np.float32(0.5954), np.float32(0.5013), np.float32(0.8537), np.float32(0.8618)] +2026-10-15 17:16:05.352211: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:16:05.684869: lr: 0.000763 +2026-10-15 17:16:05.692814: saving checkpoint... +2026-10-15 17:16:05.825255: done, saving took 0.14 seconds +2026-10-15 17:16:05.827112: This epoch took 266.713882 s + +2026-10-15 17:16:05.827253: +epoch: 260 +2026-10-15 17:20:14.137539: train loss : -0.7717 +2026-10-15 17:20:19.493150: validation loss: -0.5874 +2026-10-15 17:20:19.494464: Average global foreground Dice: [np.float32(0.9079), np.float32(0.6183), np.float32(0.7741), np.float32(0.7684), np.float32(0.6739), np.float32(0.8194), np.float32(0.7326), np.float32(0.5809), np.float32(0.0), np.float32(0.3938), np.float32(0.5696), np.float32(0.6268), np.float32(0.5574), np.float32(0.4738), np.float32(0.0), np.float32(0.0), np.float32(0.7363), np.float32(0.5863), np.float32(0.7248), np.float32(0.6393), np.float32(0.5357), np.float32(0.5525), np.float32(0.7869), np.float32(0.7408), np.float32(0.4518), np.float32(0.3949), np.float32(0.0), np.float32(0.0), np.float32(0.2437), np.float32(0.6734), np.float32(0.0), np.float32(0.0), np.float32(0.5602), np.float32(0.5714), np.float32(0.7799), np.float32(0.7962)] +2026-10-15 17:20:19.494848: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:20:19.828498: lr: 0.000762 +2026-10-15 17:20:19.836812: saving checkpoint... +2026-10-15 17:20:19.973956: done, saving took 0.14 seconds +2026-10-15 17:20:19.976055: This epoch took 254.148729 s + +2026-10-15 17:20:19.976295: +epoch: 261 +2026-10-15 17:24:20.124258: train loss : -0.7815 +2026-10-15 17:24:25.510389: validation loss: -0.5801 +2026-10-15 17:24:25.511379: Average global foreground Dice: [np.float32(0.9168), np.float32(0.5902), np.float32(0.7537), np.float32(0.7632), np.float32(0.6567), np.float32(0.7885), np.float32(0.6997), np.float32(0.534), np.float32(0.0), np.float32(0.6187), np.float32(0.5849), np.float32(0.6334), np.float32(0.5171), np.float32(0.4762), np.float32(0.0), np.float32(0.0), np.float32(0.7059), np.float32(0.5529), np.float32(0.7051), np.float32(0.6656), np.float32(0.5507), np.float32(0.6393), np.float32(0.823), np.float32(0.8078), np.float32(0.55), np.float32(0.3525), np.float32(0.0), np.float32(0.0), np.float32(0.3546), np.float32(0.5373), np.float32(0.0), np.float32(0.0), np.float32(0.524), np.float32(0.5388), np.float32(0.7869), np.float32(0.8073)] +2026-10-15 17:24:25.511599: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:24:25.848279: lr: 0.000761 +2026-10-15 17:24:25.856428: saving checkpoint... +2026-10-15 17:24:25.993619: done, saving took 0.14 seconds +2026-10-15 17:24:25.995643: This epoch took 246.019156 s + +2026-10-15 17:24:25.995845: +epoch: 262 +2026-10-15 17:28:39.059524: train loss : -0.7751 +2026-10-15 17:28:44.411296: validation loss: -0.5962 +2026-10-15 17:28:44.412531: Average global foreground Dice: [np.float32(0.9122), np.float32(0.639), np.float32(0.7668), np.float32(0.8133), np.float32(0.7381), np.float32(0.84), np.float32(0.7513), np.float32(0.6326), np.float32(0.0), np.float32(0.548), np.float32(0.5669), np.float32(0.6244), np.float32(0.556), np.float32(0.4659), np.float32(0.0), np.float32(0.0), np.float32(0.759), np.float32(0.6289), np.float32(0.7255), np.float32(0.647), np.float32(0.5878), np.float32(0.64), np.float32(0.8443), np.float32(0.8152), np.float32(0.5358), np.float32(0.3659), np.float32(0.0), np.float32(0.0), np.float32(0.408), np.float32(0.4779), np.float32(0.0), np.float32(0.0), np.float32(0.4834), np.float32(0.5205), np.float32(0.8346), np.float32(0.8414)] +2026-10-15 17:28:44.412956: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:28:44.744965: lr: 0.00076 +2026-10-15 17:28:44.752536: saving checkpoint... +2026-10-15 17:28:44.887006: done, saving took 0.14 seconds +2026-10-15 17:28:44.888709: This epoch took 258.892786 s + +2026-10-15 17:28:44.888961: +epoch: 263 +2026-10-15 17:32:53.138273: train loss : -0.7736 +2026-10-15 17:32:58.524019: validation loss: -0.5808 +2026-10-15 17:32:58.525149: Average global foreground Dice: [np.float32(0.9124), np.float32(0.6424), np.float32(0.7427), np.float32(0.7565), np.float32(0.6688), np.float32(0.7994), np.float32(0.674), np.float32(0.6086), np.float32(0.0), np.float32(0.5499), np.float32(0.5181), np.float32(0.589), np.float32(0.577), np.float32(0.5164), np.float32(0.0), np.float32(0.0), np.float32(0.7246), np.float32(0.595), np.float32(0.6967), np.float32(0.6425), np.float32(0.5463), np.float32(0.6235), np.float32(0.8209), np.float32(0.7961), np.float32(0.4839), np.float32(0.4482), np.float32(0.0), np.float32(0.0), np.float32(0.2678), np.float32(0.6243), np.float32(0.0), np.float32(0.0), np.float32(0.5292), np.float32(0.5744), np.float32(0.7907), np.float32(0.8084)] +2026-10-15 17:32:58.525562: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:32:58.858568: lr: 0.000759 +2026-10-15 17:32:58.867532: saving checkpoint... +2026-10-15 17:32:58.995741: done, saving took 0.14 seconds +2026-10-15 17:32:58.997668: This epoch took 254.108608 s + +2026-10-15 17:32:58.997916: +epoch: 264 +2026-10-15 17:37:18.277098: train loss : -0.7444 +2026-10-15 17:37:23.651245: validation loss: -0.6193 +2026-10-15 17:37:23.652310: Average global foreground Dice: [np.float32(0.9241), np.float32(0.6347), np.float32(0.7798), np.float32(0.8132), np.float32(0.7467), np.float32(0.8376), np.float32(0.7772), np.float32(0.5032), np.float32(0.0), np.float32(0.6386), np.float32(0.5906), np.float32(0.6373), np.float32(0.5348), np.float32(0.5557), np.float32(0.0), np.float32(0.0), np.float32(0.7492), np.float32(0.6031), np.float32(0.7441), np.float32(0.6329), np.float32(0.5974), np.float32(0.6459), np.float32(0.8684), np.float32(0.8687), np.float32(0.456), np.float32(0.3485), np.float32(0.0), np.float32(0.0), np.float32(0.4402), np.float32(0.5611), np.float32(0.0), np.float32(0.0), np.float32(0.5739), np.float32(0.6266), np.float32(0.8424), np.float32(0.8462)] +2026-10-15 17:37:23.652721: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:37:23.986015: lr: 0.000758 +2026-10-15 17:37:23.994977: saving checkpoint... +2026-10-15 17:37:24.133656: done, saving took 0.15 seconds +2026-10-15 17:37:24.135780: This epoch took 265.137779 s + +2026-10-15 17:37:24.135963: +epoch: 265 +2026-10-15 17:41:26.780006: train loss : -0.7669 +2026-10-15 17:41:32.136883: validation loss: -0.6042 +2026-10-15 17:41:32.138307: Average global foreground Dice: [np.float32(0.9147), np.float32(0.6567), np.float32(0.7689), np.float32(0.7757), np.float32(0.7063), np.float32(0.8137), np.float32(0.7207), np.float32(0.5925), np.float32(0.0), np.float32(0.4658), np.float32(0.5732), np.float32(0.6202), np.float32(0.557), np.float32(0.5317), np.float32(0.0), np.float32(0.0), np.float32(0.7457), np.float32(0.6029), np.float32(0.7054), np.float32(0.6343), np.float32(0.5734), np.float32(0.6602), np.float32(0.8298), np.float32(0.8154), np.float32(0.4755), np.float32(0.4452), np.float32(0.0), np.float32(0.0), np.float32(0.4763), np.float32(0.5857), np.float32(0.0), np.float32(0.0), np.float32(0.5503), np.float32(0.6332), np.float32(0.8068), np.float32(0.8313)] +2026-10-15 17:41:32.138683: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:41:32.466403: lr: 0.000757 +2026-10-15 17:41:32.474559: saving checkpoint... +2026-10-15 17:41:32.617072: done, saving took 0.15 seconds +2026-10-15 17:41:32.618776: This epoch took 248.482720 s + +2026-10-15 17:41:32.619101: +epoch: 266 +2026-10-15 17:45:36.957657: train loss : -0.7784 +2026-10-15 17:45:42.329924: validation loss: -0.5836 +2026-10-15 17:45:42.331129: Average global foreground Dice: [np.float32(0.9164), np.float32(0.6099), np.float32(0.7386), np.float32(0.759), np.float32(0.6602), np.float32(0.7932), np.float32(0.6902), np.float32(0.7651), np.float32(0.0), np.float32(0.5418), np.float32(0.5791), np.float32(0.5783), np.float32(0.5482), np.float32(0.5307), np.float32(0.0), np.float32(0.0), np.float32(0.7371), np.float32(0.5631), np.float32(0.7055), np.float32(0.6474), np.float32(0.5751), np.float32(0.6327), np.float32(0.7459), np.float32(0.7477), np.float32(0.4939), np.float32(0.4129), np.float32(0.0), np.float32(0.0), np.float32(0.3748), np.float32(0.6064), np.float32(0.0), np.float32(0.0), np.float32(0.4949), np.float32(0.5779), np.float32(0.7859), np.float32(0.804)] +2026-10-15 17:45:42.331564: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:45:42.656005: lr: 0.000756 +2026-10-15 17:45:42.664573: saving checkpoint... +2026-10-15 17:45:42.916321: done, saving took 0.26 seconds +2026-10-15 17:45:42.918903: This epoch took 250.299707 s + +2026-10-15 17:45:42.919481: +epoch: 267 +2026-10-15 17:50:10.898634: train loss : -0.7629 +2026-10-15 17:50:16.265665: validation loss: -0.4322 +2026-10-15 17:50:16.266885: Average global foreground Dice: [np.float32(0.9216), np.float32(0.4394), np.float32(0.458), np.float32(0.4112), np.float32(0.3502), np.float32(0.4959), np.float32(0.4309), np.float32(0.5284), np.float32(0.0), np.float32(0.6086), np.float32(0.3233), np.float32(0.3717), np.float32(0.4711), np.float32(0.4531), np.float32(0.0), np.float32(0.0), np.float32(0.3519), np.float32(0.2738), np.float32(0.3811), np.float32(0.3992), np.float32(0.3614), np.float32(0.2681), np.float32(0.5563), np.float32(0.5055), np.float32(0.2395), np.float32(0.1634), np.float32(0.0), np.float32(0.0), np.float32(0.272), np.float32(0.5011), np.float32(0.0), np.float32(0.0), np.float32(0.3383), np.float32(0.4097), np.float32(0.4565), np.float32(0.5346)] +2026-10-15 17:50:16.267225: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:50:16.598978: lr: 0.000755 +2026-10-15 17:50:16.599255: This epoch took 273.679630 s + +2026-10-15 17:50:16.599352: +epoch: 268 +2026-10-15 17:54:56.306395: train loss : -0.7669 +2026-10-15 17:55:01.677275: validation loss: -0.5466 +2026-10-15 17:55:01.678199: Average global foreground Dice: [np.float32(0.9128), np.float32(0.5964), np.float32(0.6367), np.float32(0.6854), np.float32(0.5239), np.float32(0.72), np.float32(0.622), np.float32(0.5516), np.float32(0.0), np.float32(0.4878), np.float32(0.535), np.float32(0.5759), np.float32(0.5482), np.float32(0.5149), np.float32(0.0), np.float32(0.0), np.float32(0.5386), np.float32(0.4208), np.float32(0.5553), np.float32(0.4952), np.float32(0.5119), np.float32(0.4855), np.float32(0.7319), np.float32(0.7122), np.float32(0.4018), np.float32(0.2846), np.float32(0.0), np.float32(0.0), np.float32(0.4145), np.float32(0.6071), np.float32(0.0), np.float32(0.0), np.float32(0.4651), np.float32(0.5388), np.float32(0.6703), np.float32(0.7063)] +2026-10-15 17:55:01.678371: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:55:02.002604: lr: 0.000754 +2026-10-15 17:55:02.002972: This epoch took 285.403555 s + +2026-10-15 17:55:02.003103: +epoch: 269 +2026-10-15 17:59:24.824975: train loss : -0.7681 +2026-10-15 17:59:30.205454: validation loss: -0.5427 +2026-10-15 17:59:30.206453: Average global foreground Dice: [np.float32(0.9046), np.float32(0.5645), np.float32(0.6349), np.float32(0.6613), np.float32(0.5758), np.float32(0.7152), np.float32(0.6276), np.float32(0.7369), np.float32(0.0), np.float32(0.6268), np.float32(0.4799), np.float32(0.5649), np.float32(0.5288), np.float32(0.4791), np.float32(0.0), np.float32(0.0), np.float32(0.5794), np.float32(0.4387), np.float32(0.588), np.float32(0.5312), np.float32(0.4719), np.float32(0.4541), np.float32(0.7236), np.float32(0.6696), np.float32(0.3306), np.float32(0.3326), np.float32(0.0), np.float32(0.0), np.float32(0.3833), np.float32(0.6062), np.float32(0.0), np.float32(0.0), np.float32(0.4982), np.float32(0.5326), np.float32(0.6607), np.float32(0.6846)] +2026-10-15 17:59:30.206859: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 17:59:30.540622: lr: 0.000753 +2026-10-15 17:59:30.541035: This epoch took 268.537860 s + +2026-10-15 17:59:30.541412: +epoch: 270 +2026-10-15 18:03:49.537733: train loss : -0.7670 +2026-10-15 18:03:54.912319: validation loss: -0.5644 +2026-10-15 18:03:54.913252: Average global foreground Dice: [np.float32(0.919), np.float32(0.5826), np.float32(0.7435), np.float32(0.753), np.float32(0.6683), np.float32(0.7856), np.float32(0.6911), np.float32(0.5123), np.float32(0.0), np.float32(0.5189), np.float32(0.5613), np.float32(0.6041), np.float32(0.571), np.float32(0.4616), np.float32(0.0), np.float32(0.0), np.float32(0.7265), np.float32(0.5582), np.float32(0.7182), np.float32(0.6341), np.float32(0.4429), np.float32(0.5904), np.float32(0.7924), np.float32(0.7728), np.float32(0.5077), np.float32(0.4176), np.float32(0.0), np.float32(0.0), np.float32(0.0014), np.float32(0.5963), np.float32(0.0), np.float32(0.0), np.float32(0.5627), np.float32(0.624), np.float32(0.7813), np.float32(0.7977)] +2026-10-15 18:03:54.913522: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:03:55.260744: lr: 0.000752 +2026-10-15 18:03:55.260973: This epoch took 264.719419 s + +2026-10-15 18:03:55.261045: +epoch: 271 +2026-10-15 18:08:26.101713: train loss : -0.7792 +2026-10-15 18:08:31.486917: validation loss: -0.5703 +2026-10-15 18:08:31.487935: Average global foreground Dice: [np.float32(0.8936), np.float32(0.5995), np.float32(0.7623), np.float32(0.7415), np.float32(0.6346), np.float32(0.7845), np.float32(0.6321), np.float32(0.5964), np.float32(0.0), np.float32(0.3739), np.float32(0.4956), np.float32(0.5803), np.float32(0.5214), np.float32(0.4335), np.float32(0.0), np.float32(0.0), np.float32(0.6549), np.float32(0.4901), np.float32(0.6386), np.float32(0.5878), np.float32(0.5324), np.float32(0.6587), np.float32(0.8571), np.float32(0.8285), np.float32(0.5267), np.float32(0.3986), np.float32(0.0), np.float32(0.0), np.float32(0.1065), np.float32(0.6359), np.float32(0.0), np.float32(0.0), np.float32(0.4942), np.float32(0.4988), np.float32(0.7971), np.float32(0.8215)] +2026-10-15 18:08:31.488322: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:08:31.803569: lr: 0.000751 +2026-10-15 18:08:31.803819: This epoch took 276.542665 s + +2026-10-15 18:08:31.803908: +epoch: 272 +2026-10-15 18:12:32.089731: train loss : -0.7704 +2026-10-15 18:12:37.456154: validation loss: -0.5513 +2026-10-15 18:12:37.457321: Average global foreground Dice: [np.float32(0.9118), np.float32(0.5974), np.float32(0.6863), np.float32(0.684), np.float32(0.6102), np.float32(0.7249), np.float32(0.6294), np.float32(0.6995), np.float32(0.0), np.float32(0.4386), np.float32(0.458), np.float32(0.5091), np.float32(0.5613), np.float32(0.5041), np.float32(0.0), np.float32(0.0), np.float32(0.6055), np.float32(0.5146), np.float32(0.6079), np.float32(0.5789), np.float32(0.5387), np.float32(0.6032), np.float32(0.7671), np.float32(0.7463), np.float32(0.4665), np.float32(0.403), np.float32(0.0), np.float32(0.0), np.float32(0.1706), np.float32(0.4763), np.float32(0.0), np.float32(0.0), np.float32(0.5203), np.float32(0.4896), np.float32(0.7168), np.float32(0.7596)] +2026-10-15 18:12:37.457936: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:12:37.786175: lr: 0.000751 +2026-10-15 18:12:37.786451: This epoch took 245.982480 s + +2026-10-15 18:12:37.786524: +epoch: 273 +2026-10-15 18:17:00.776521: train loss : -0.7764 +2026-10-15 18:17:06.139514: validation loss: -0.5642 +2026-10-15 18:17:06.141045: Average global foreground Dice: [np.float32(0.9086), np.float32(0.6064), np.float32(0.7225), np.float32(0.7121), np.float32(0.6352), np.float32(0.7603), np.float32(0.6606), np.float32(0.5249), np.float32(0.0), np.float32(0.5629), np.float32(0.4866), np.float32(0.5765), np.float32(0.5347), np.float32(0.469), np.float32(0.0), np.float32(0.0), np.float32(0.6643), np.float32(0.5462), np.float32(0.6501), np.float32(0.6084), np.float32(0.5776), np.float32(0.5318), np.float32(0.7933), np.float32(0.7618), np.float32(0.4814), np.float32(0.3752), np.float32(0.0), np.float32(0.0), np.float32(0.3281), np.float32(0.5639), np.float32(0.0), np.float32(0.0), np.float32(0.5392), np.float32(0.5362), np.float32(0.7312), np.float32(0.759)] +2026-10-15 18:17:06.141448: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:17:06.532485: lr: 0.00075 +2026-10-15 18:17:06.532790: This epoch took 268.746206 s + +2026-10-15 18:17:06.532878: +epoch: 274 +2026-10-15 18:21:12.575582: train loss : -0.7686 +2026-10-15 18:21:17.974641: validation loss: -0.5103 +2026-10-15 18:21:17.975695: Average global foreground Dice: [np.float32(0.8948), np.float32(0.5222), np.float32(0.6335), np.float32(0.5659), np.float32(0.4961), np.float32(0.6399), np.float32(0.5587), np.float32(0.6875), np.float32(0.0), np.float32(0.2836), np.float32(0.371), np.float32(0.5092), np.float32(0.5354), np.float32(0.4703), np.float32(0.0), np.float32(0.0), np.float32(0.5251), np.float32(0.4285), np.float32(0.5163), np.float32(0.4964), np.float32(0.4761), np.float32(0.4604), np.float32(0.7177), np.float32(0.6879), np.float32(0.351), np.float32(0.2925), np.float32(0.0), np.float32(0.0), np.float32(0.2931), np.float32(0.5778), np.float32(0.0), np.float32(0.0), np.float32(0.4656), np.float32(0.4316), np.float32(0.614), np.float32(0.6723)] +2026-10-15 18:21:17.975964: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:21:18.359631: lr: 0.000749 +2026-10-15 18:21:18.360228: This epoch took 251.827282 s + +2026-10-15 18:21:18.360502: +epoch: 275 +2026-10-15 18:25:18.826858: train loss : -0.7744 +2026-10-15 18:25:24.184031: validation loss: -0.3688 +2026-10-15 18:25:24.185248: Average global foreground Dice: [np.float32(0.9003), np.float32(0.3848), np.float32(0.3585), np.float32(0.3582), np.float32(0.2508), np.float32(0.39), np.float32(0.3294), np.float32(0.6715), np.float32(0.0), np.float32(0.5263), np.float32(0.3185), np.float32(0.3533), np.float32(0.4252), np.float32(0.3049), np.float32(0.0), np.float32(0.0), np.float32(0.2451), np.float32(0.2235), np.float32(0.3225), np.float32(0.3849), np.float32(0.335), np.float32(0.3093), np.float32(0.5239), np.float32(0.425), np.float32(0.1349), np.float32(0.1521), np.float32(0.0), np.float32(0.0), np.float32(0.2535), np.float32(0.4436), np.float32(0.0), np.float32(0.0), np.float32(0.21), np.float32(0.2707), np.float32(0.3889), np.float32(0.3944)] +2026-10-15 18:25:24.185653: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:25:24.563935: lr: 0.000748 +2026-10-15 18:25:24.564512: This epoch took 246.203923 s + +2026-10-15 18:25:24.564862: +epoch: 276 +2026-10-15 18:29:59.807990: train loss : -0.7733 +2026-10-15 18:30:05.169063: validation loss: -0.5446 +2026-10-15 18:30:05.170263: Average global foreground Dice: [np.float32(0.902), np.float32(0.6038), np.float32(0.6628), np.float32(0.712), np.float32(0.6064), np.float32(0.7381), np.float32(0.6267), np.float32(0.6493), np.float32(0.0), np.float32(0.3441), np.float32(0.4928), np.float32(0.552), np.float32(0.5603), np.float32(0.4755), np.float32(0.0), np.float32(0.0), np.float32(0.6476), np.float32(0.4912), np.float32(0.6418), np.float32(0.5851), np.float32(0.5197), np.float32(0.5796), np.float32(0.6885), np.float32(0.6411), np.float32(0.4335), np.float32(0.3363), np.float32(0.0), np.float32(0.0), np.float32(0.3393), np.float32(0.5327), np.float32(0.0), np.float32(0.0), np.float32(0.5172), np.float32(0.5697), np.float32(0.6893), np.float32(0.7193)] +2026-10-15 18:30:05.170475: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:30:05.545745: lr: 0.000747 +2026-10-15 18:30:05.546026: This epoch took 280.981057 s + +2026-10-15 18:30:05.546134: +epoch: 277 +2026-10-15 18:34:18.349887: train loss : -0.7736 +2026-10-15 18:34:23.738203: validation loss: -0.6621 +2026-10-15 18:34:23.739196: Average global foreground Dice: [np.float32(0.9061), np.float32(0.6858), np.float32(0.8339), np.float32(0.8571), np.float32(0.8122), np.float32(0.8863), np.float32(0.8208), np.float32(0.7586), np.float32(0.0), np.float32(0.522), np.float32(0.6566), np.float32(0.6785), np.float32(0.664), np.float32(0.5838), np.float32(0.0), np.float32(0.0), np.float32(0.7673), np.float32(0.5963), np.float32(0.738), np.float32(0.6669), np.float32(0.6159), np.float32(0.6828), np.float32(0.8865), np.float32(0.8569), np.float32(0.4926), np.float32(0.4908), np.float32(0.0), np.float32(0.0), np.float32(0.473), np.float32(0.5861), np.float32(0.0), np.float32(0.0), np.float32(0.5706), np.float32(0.5985), np.float32(0.8976), np.float32(0.8968)] +2026-10-15 18:34:23.739398: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:34:24.150709: lr: 0.000746 +2026-10-15 18:34:24.151139: This epoch took 258.604905 s + +2026-10-15 18:34:24.151344: +epoch: 278 +2026-10-15 18:39:11.151805: train loss : -0.7668 +2026-10-15 18:39:16.529616: validation loss: -0.5960 +2026-10-15 18:39:16.530942: Average global foreground Dice: [np.float32(0.9074), np.float32(0.6205), np.float32(0.777), np.float32(0.7778), np.float32(0.6957), np.float32(0.8047), np.float32(0.7358), np.float32(0.4227), np.float32(0.0), np.float32(0.3587), np.float32(0.5894), np.float32(0.5991), np.float32(0.5881), np.float32(0.5056), np.float32(0.0), np.float32(0.0), np.float32(0.7453), np.float32(0.5898), np.float32(0.7209), np.float32(0.6714), np.float32(0.5849), np.float32(0.6064), np.float32(0.8621), np.float32(0.8502), np.float32(0.4921), np.float32(0.4173), np.float32(0.0), np.float32(0.0), np.float32(0.3704), np.float32(0.5505), np.float32(0.0), np.float32(0.0), np.float32(0.5425), np.float32(0.547), np.float32(0.8168), np.float32(0.8261)] +2026-10-15 18:39:16.531424: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:39:16.898824: lr: 0.000745 +2026-10-15 18:39:16.899079: This epoch took 292.747648 s + +2026-10-15 18:39:16.899165: +epoch: 279 +2026-10-15 18:43:48.842218: train loss : -0.7603 +2026-10-15 18:43:54.231299: validation loss: -0.4654 +2026-10-15 18:43:54.232480: Average global foreground Dice: [np.float32(0.9071), np.float32(0.5084), np.float32(0.5258), np.float32(0.5837), np.float32(0.4134), np.float32(0.564), np.float32(0.4632), np.float32(0.7991), np.float32(0.0), np.float32(0.4055), np.float32(0.4458), np.float32(0.4461), np.float32(0.5429), np.float32(0.4435), np.float32(0.0), np.float32(0.0), np.float32(0.4138), np.float32(0.3643), np.float32(0.401), np.float32(0.3797), np.float32(0.4913), np.float32(0.3596), np.float32(0.4531), np.float32(0.505), np.float32(0.3459), np.float32(0.2507), np.float32(0.0), np.float32(0.0), np.float32(0.0095), np.float32(0.4568), np.float32(0.0), np.float32(0.0), np.float32(0.4085), np.float32(0.4195), np.float32(0.4812), np.float32(0.5522)] +2026-10-15 18:43:54.233021: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:43:54.575073: lr: 0.000744 +2026-10-15 18:43:54.575315: This epoch took 277.676087 s + +2026-10-15 18:43:54.575395: +epoch: 280 +2026-10-15 18:48:14.734300: train loss : -0.7753 +2026-10-15 18:48:20.101560: validation loss: -0.4686 +2026-10-15 18:48:20.102602: Average global foreground Dice: [np.float32(0.9165), np.float32(0.5311), np.float32(0.5553), np.float32(0.5703), np.float32(0.4437), np.float32(0.554), np.float32(0.4639), np.float32(0.5866), np.float32(0.0), np.float32(0.4168), np.float32(0.4315), np.float32(0.4584), np.float32(0.5348), np.float32(0.3536), np.float32(0.0), np.float32(0.0), np.float32(0.407), np.float32(0.3058), np.float32(0.3976), np.float32(0.4348), np.float32(0.4823), np.float32(0.4128), np.float32(0.6291), np.float32(0.5935), np.float32(0.2857), np.float32(0.2287), np.float32(0.0), np.float32(0.0), np.float32(0.2274), np.float32(0.5294), np.float32(0.0), np.float32(0.0), np.float32(0.3651), np.float32(0.3672), np.float32(0.5738), np.float32(0.573)] +2026-10-15 18:48:20.102766: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:48:20.502387: lr: 0.000743 +2026-10-15 18:48:20.502651: This epoch took 265.927200 s + +2026-10-15 18:48:20.502725: +epoch: 281 +2026-10-15 18:52:36.600544: train loss : -0.7812 +2026-10-15 18:52:42.001233: validation loss: -0.5634 +2026-10-15 18:52:42.002563: Average global foreground Dice: [np.float32(0.9025), np.float32(0.5839), np.float32(0.7082), np.float32(0.6695), np.float32(0.6022), np.float32(0.7222), np.float32(0.6376), np.float32(0.6345), np.float32(0.0), np.float32(0.6184), np.float32(0.4923), np.float32(0.536), np.float32(0.5674), np.float32(0.4958), np.float32(0.0), np.float32(0.0), np.float32(0.6476), np.float32(0.505), np.float32(0.6248), np.float32(0.5955), np.float32(0.5775), np.float32(0.6225), np.float32(0.8388), np.float32(0.8193), np.float32(0.4473), np.float32(0.3592), np.float32(0.0), np.float32(0.0), np.float32(0.2759), np.float32(0.6296), np.float32(0.0), np.float32(0.0), np.float32(0.5146), np.float32(0.5369), np.float32(0.7369), np.float32(0.7747)] +2026-10-15 18:52:42.003054: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:52:42.323859: lr: 0.000742 +2026-10-15 18:52:42.324162: This epoch took 261.821381 s + +2026-10-15 18:52:42.324415: +epoch: 282 +2026-10-15 18:57:11.630559: train loss : -0.7821 +2026-10-15 18:57:17.018906: validation loss: -0.5977 +2026-10-15 18:57:17.019897: Average global foreground Dice: [np.float32(0.9057), np.float32(0.6255), np.float32(0.7417), np.float32(0.7623), np.float32(0.6625), np.float32(0.7937), np.float32(0.6941), np.float32(0.4879), np.float32(0.0), np.float32(0.4054), np.float32(0.5443), np.float32(0.6034), np.float32(0.5672), np.float32(0.4698), np.float32(0.0), np.float32(0.0), np.float32(0.6961), np.float32(0.5589), np.float32(0.6681), np.float32(0.6239), np.float32(0.5741), np.float32(0.6082), np.float32(0.8469), np.float32(0.8329), np.float32(0.4841), np.float32(0.3962), np.float32(0.0), np.float32(0.0), np.float32(0.4289), np.float32(0.6323), np.float32(0.0), np.float32(0.0), np.float32(0.5469), np.float32(0.5442), np.float32(0.7912), np.float32(0.8108)] +2026-10-15 18:57:17.020082: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 18:57:17.391092: lr: 0.000741 +2026-10-15 18:57:17.391362: This epoch took 275.066803 s + +2026-10-15 18:57:17.391462: +epoch: 283 +2026-10-15 19:02:16.838090: train loss : -0.7779 +2026-10-15 19:02:22.208829: validation loss: -0.5957 +2026-10-15 19:02:22.209858: Average global foreground Dice: [np.float32(0.9201), np.float32(0.6528), np.float32(0.7496), np.float32(0.754), np.float32(0.6673), np.float32(0.7785), np.float32(0.6796), np.float32(0.5622), np.float32(0.0), np.float32(0.5558), np.float32(0.5605), np.float32(0.641), np.float32(0.589), np.float32(0.5659), np.float32(0.0), np.float32(0.0), np.float32(0.7127), np.float32(0.5513), np.float32(0.6743), np.float32(0.6304), np.float32(0.5399), np.float32(0.5695), np.float32(0.8527), np.float32(0.841), np.float32(0.4532), np.float32(0.4134), np.float32(0.0), np.float32(0.0), np.float32(0.3205), np.float32(0.5484), np.float32(0.0), np.float32(0.0), np.float32(0.5143), np.float32(0.5576), np.float32(0.7882), np.float32(0.8079)] +2026-10-15 19:02:22.210226: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:02:22.563591: lr: 0.00074 +2026-10-15 19:02:22.564072: This epoch took 305.172547 s + +2026-10-15 19:02:22.564482: +epoch: 284 +2026-10-15 19:06:42.901522: train loss : -0.7664 +2026-10-15 19:06:48.273846: validation loss: -0.5544 +2026-10-15 19:06:48.274950: Average global foreground Dice: [np.float32(0.9192), np.float32(0.6314), np.float32(0.7112), np.float32(0.7141), np.float32(0.6016), np.float32(0.7384), np.float32(0.6389), np.float32(0.5318), np.float32(0.0), np.float32(0.4523), np.float32(0.5085), np.float32(0.5477), np.float32(0.516), np.float32(0.4574), np.float32(0.0), np.float32(0.0), np.float32(0.6381), np.float32(0.4847), np.float32(0.607), np.float32(0.5804), np.float32(0.5006), np.float32(0.5477), np.float32(0.8228), np.float32(0.7995), np.float32(0.4421), np.float32(0.3735), np.float32(0.0), np.float32(0.0), np.float32(0.3537), np.float32(0.5008), np.float32(0.0), np.float32(0.0), np.float32(0.3669), np.float32(0.4816), np.float32(0.7511), np.float32(0.7664)] +2026-10-15 19:06:48.275169: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:06:48.600302: lr: 0.000739 +2026-10-15 19:06:48.600595: This epoch took 266.035803 s + +2026-10-15 19:06:48.600844: +epoch: 285 +2026-10-15 19:10:53.548387: train loss : -0.7868 +2026-10-15 19:10:58.932984: validation loss: -0.5517 +2026-10-15 19:10:58.933953: Average global foreground Dice: [np.float32(0.9022), np.float32(0.6001), np.float32(0.7118), np.float32(0.7351), np.float32(0.6327), np.float32(0.7592), np.float32(0.6566), np.float32(0.7071), np.float32(0.0), np.float32(0.4719), np.float32(0.4891), np.float32(0.5975), np.float32(0.5661), np.float32(0.3931), np.float32(0.0), np.float32(0.0), np.float32(0.6514), np.float32(0.4909), np.float32(0.6236), np.float32(0.5581), np.float32(0.5118), np.float32(0.5552), np.float32(0.8166), np.float32(0.7732), np.float32(0.4541), np.float32(0.405), np.float32(0.0), np.float32(0.0), np.float32(0.0007), np.float32(0.5178), np.float32(0.0), np.float32(0.0), np.float32(0.4963), np.float32(0.4714), np.float32(0.7468), np.float32(0.7546)] +2026-10-15 19:10:58.934235: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:10:59.257584: lr: 0.000738 +2026-10-15 19:10:59.257839: This epoch took 250.656776 s + +2026-10-15 19:10:59.257922: +epoch: 286 +2026-10-15 19:15:39.371295: train loss : -0.7810 +2026-10-15 19:15:44.763381: validation loss: -0.5326 +2026-10-15 19:15:44.764407: Average global foreground Dice: [np.float32(0.9068), np.float32(0.5555), np.float32(0.6856), np.float32(0.6822), np.float32(0.6002), np.float32(0.7242), np.float32(0.6285), np.float32(0.5508), np.float32(0.0), np.float32(0.4217), np.float32(0.4422), np.float32(0.5489), np.float32(0.5292), np.float32(0.4256), np.float32(0.0), np.float32(0.0), np.float32(0.6335), np.float32(0.4609), np.float32(0.6024), np.float32(0.5616), np.float32(0.4676), np.float32(0.5298), np.float32(0.8023), np.float32(0.7552), np.float32(0.4113), np.float32(0.3477), np.float32(0.0), np.float32(0.0), np.float32(0.3827), np.float32(0.6449), np.float32(0.0), np.float32(0.0), np.float32(0.3929), np.float32(0.4109), np.float32(0.708), np.float32(0.7286)] +2026-10-15 19:15:44.764599: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:15:45.092743: lr: 0.000738 +2026-10-15 19:15:45.093379: This epoch took 285.835401 s + +2026-10-15 19:15:45.093832: +epoch: 287 +2026-10-15 19:20:17.670391: train loss : -0.7819 +2026-10-15 19:20:23.041054: validation loss: -0.5385 +2026-10-15 19:20:23.042506: Average global foreground Dice: [np.float32(0.9108), np.float32(0.546), np.float32(0.6459), np.float32(0.686), np.float32(0.558), np.float32(0.7153), np.float32(0.6155), np.float32(0.571), np.float32(0.0), np.float32(0.4693), np.float32(0.4808), np.float32(0.547), np.float32(0.5043), np.float32(0.4094), np.float32(0.0), np.float32(0.0), np.float32(0.6165), np.float32(0.479), np.float32(0.5806), np.float32(0.5656), np.float32(0.4813), np.float32(0.5251), np.float32(0.7558), np.float32(0.7199), np.float32(0.4679), np.float32(0.3033), np.float32(0.0), np.float32(0.0), np.float32(0.3839), np.float32(0.6358), np.float32(0.0), np.float32(0.0), np.float32(0.5197), np.float32(0.5123), np.float32(0.6877), np.float32(0.7101)] +2026-10-15 19:20:23.042868: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:20:23.371145: lr: 0.000737 +2026-10-15 19:20:23.371538: This epoch took 278.277553 s + +2026-10-15 19:20:23.371888: +epoch: 288 +2026-10-15 19:24:27.279508: train loss : -0.7846 +2026-10-15 19:24:32.653891: validation loss: -0.4781 +2026-10-15 19:24:32.655088: Average global foreground Dice: [np.float32(0.896), np.float32(0.4735), np.float32(0.5199), np.float32(0.5954), np.float32(0.4769), np.float32(0.5808), np.float32(0.4498), np.float32(0.6306), np.float32(0.0391), np.float32(0.4049), np.float32(0.4069), np.float32(0.498), np.float32(0.4851), np.float32(0.4362), np.float32(0.0), np.float32(0.0), np.float32(0.4609), np.float32(0.3164), np.float32(0.4115), np.float32(0.4174), np.float32(0.4489), np.float32(0.4345), np.float32(0.7808), np.float32(0.7092), np.float32(0.3595), np.float32(0.2567), np.float32(0.0), np.float32(0.0), np.float32(0.3194), np.float32(0.621), np.float32(0.0), np.float32(0.0), np.float32(0.3827), np.float32(0.3951), np.float32(0.6437), np.float32(0.5971)] +2026-10-15 19:24:32.655550: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:24:32.985098: lr: 0.000736 +2026-10-15 19:24:32.985400: This epoch took 249.613366 s + +2026-10-15 19:24:32.985605: +epoch: 289 +2026-10-15 19:28:55.655028: train loss : -0.7753 +2026-10-15 19:29:01.018364: validation loss: -0.4541 +2026-10-15 19:29:01.019434: Average global foreground Dice: [np.float32(0.9002), np.float32(0.4752), np.float32(0.5378), np.float32(0.5547), np.float32(0.4015), np.float32(0.5662), np.float32(0.4496), np.float32(0.3394), np.float32(0.0), np.float32(0.4682), np.float32(0.3972), np.float32(0.4028), np.float32(0.5333), np.float32(0.4515), np.float32(0.0), np.float32(0.0), np.float32(0.4534), np.float32(0.3664), np.float32(0.389), np.float32(0.3716), np.float32(0.4025), np.float32(0.4273), np.float32(0.6772), np.float32(0.6467), np.float32(0.3273), np.float32(0.3103), np.float32(0.0), np.float32(0.0), np.float32(0.2303), np.float32(0.5196), np.float32(0.0), np.float32(0.0), np.float32(0.3291), np.float32(0.2644), np.float32(0.5565), np.float32(0.6243)] +2026-10-15 19:29:01.019896: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:29:01.355187: lr: 0.000735 +2026-10-15 19:29:01.355623: This epoch took 268.369915 s + +2026-10-15 19:29:01.355940: +epoch: 290 +2026-10-15 19:33:02.704212: train loss : -0.7904 +2026-10-15 19:33:08.072243: validation loss: -0.5305 +2026-10-15 19:33:08.073387: Average global foreground Dice: [np.float32(0.913), np.float32(0.611), np.float32(0.6781), np.float32(0.674), np.float32(0.5426), np.float32(0.6781), np.float32(0.5495), np.float32(0.629), np.float32(0.055), np.float32(0.4936), np.float32(0.4869), np.float32(0.5152), np.float32(0.5377), np.float32(0.4662), np.float32(0.0), np.float32(0.0), np.float32(0.6071), np.float32(0.4727), np.float32(0.5477), np.float32(0.5722), np.float32(0.5352), np.float32(0.5656), np.float32(0.763), np.float32(0.7034), np.float32(0.488), np.float32(0.4007), np.float32(0.0), np.float32(0.0), np.float32(0.2259), np.float32(0.573), np.float32(0.0), np.float32(0.0), np.float32(0.4572), np.float32(0.4219), np.float32(0.6989), np.float32(0.7129)] +2026-10-15 19:33:08.073617: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:33:08.393892: lr: 0.000734 +2026-10-15 19:33:08.394377: This epoch took 247.038236 s + +2026-10-15 19:33:08.394783: +epoch: 291 +2026-10-15 19:37:38.217616: train loss : -0.7772 +2026-10-15 19:37:43.599790: validation loss: -0.6400 +2026-10-15 19:37:43.600873: Average global foreground Dice: [np.float32(0.9163), np.float32(0.687), np.float32(0.8123), np.float32(0.8137), np.float32(0.7577), np.float32(0.8288), np.float32(0.7582), np.float32(0.5583), np.float32(0.2191), np.float32(0.5395), np.float32(0.5996), np.float32(0.6311), np.float32(0.6344), np.float32(0.5712), np.float32(0.0), np.float32(0.0), np.float32(0.7572), np.float32(0.5869), np.float32(0.7377), np.float32(0.6617), np.float32(0.5136), np.float32(0.6285), np.float32(0.906), np.float32(0.8818), np.float32(0.5092), np.float32(0.5498), np.float32(0.0), np.float32(0.0), np.float32(0.3844), np.float32(0.6531), np.float32(0.0), np.float32(0.0), np.float32(0.6286), np.float32(0.6505), np.float32(0.8679), np.float32(0.8727)] +2026-10-15 19:37:43.601040: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:37:43.932189: lr: 0.000733 +2026-10-15 19:37:43.932667: This epoch took 275.537589 s + +2026-10-15 19:37:43.933005: +epoch: 292 +2026-10-15 19:42:00.855613: train loss : -0.7801 +2026-10-15 19:42:06.222048: validation loss: -0.4563 +2026-10-15 19:42:06.223046: Average global foreground Dice: [np.float32(0.9008), np.float32(0.5142), np.float32(0.5096), np.float32(0.5173), np.float32(0.3991), np.float32(0.5405), np.float32(0.456), np.float32(0.526), np.float32(0.2002), np.float32(0.3337), np.float32(0.3737), np.float32(0.427), np.float32(0.5584), np.float32(0.4294), np.float32(0.0), np.float32(0.0), np.float32(0.4053), np.float32(0.3149), np.float32(0.4202), np.float32(0.4192), np.float32(0.4097), np.float32(0.3555), np.float32(0.633), np.float32(0.5229), np.float32(0.2184), np.float32(0.2494), np.float32(0.0), np.float32(0.0), np.float32(0.495), np.float32(0.538), np.float32(0.0), np.float32(0.0), np.float32(0.292), np.float32(0.3918), np.float32(0.5335), np.float32(0.5211)] +2026-10-15 19:42:06.223720: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:42:06.546150: lr: 0.000732 +2026-10-15 19:42:06.546506: This epoch took 262.613228 s + +2026-10-15 19:42:06.546694: +epoch: 293 +2026-10-15 19:46:20.985313: train loss : -0.7795 +2026-10-15 19:46:26.362196: validation loss: -0.5757 +2026-10-15 19:46:26.363296: Average global foreground Dice: [np.float32(0.9207), np.float32(0.6338), np.float32(0.7538), np.float32(0.7928), np.float32(0.7133), np.float32(0.8247), np.float32(0.7225), np.float32(0.559), np.float32(0.1374), np.float32(0.3046), np.float32(0.5321), np.float32(0.5994), np.float32(0.5894), np.float32(0.5214), np.float32(0.0), np.float32(0.0), np.float32(0.7462), np.float32(0.6054), np.float32(0.7317), np.float32(0.674), np.float32(0.4981), np.float32(0.5723), np.float32(0.8001), np.float32(0.7948), np.float32(0.5295), np.float32(0.4921), np.float32(0.0), np.float32(0.0), np.float32(0.2373), np.float32(0.5843), np.float32(0.0), np.float32(0.0), np.float32(0.6207), np.float32(0.6414), np.float32(0.7899), np.float32(0.8078)] +2026-10-15 19:46:26.363734: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:46:26.874995: lr: 0.000731 +2026-10-15 19:46:26.875247: This epoch took 260.328475 s + +2026-10-15 19:46:26.875329: +epoch: 294 +2026-10-15 19:51:17.223822: train loss : -0.7792 +2026-10-15 19:51:22.587017: validation loss: -0.5705 +2026-10-15 19:51:22.588036: Average global foreground Dice: [np.float32(0.9138), np.float32(0.6542), np.float32(0.74), np.float32(0.7654), np.float32(0.6825), np.float32(0.8071), np.float32(0.7018), np.float32(0.6779), np.float32(0.1359), np.float32(0.5006), np.float32(0.5461), np.float32(0.5992), np.float32(0.5822), np.float32(0.4842), np.float32(0.0), np.float32(0.0), np.float32(0.7227), np.float32(0.5882), np.float32(0.7077), np.float32(0.6441), np.float32(0.494), np.float32(0.6075), np.float32(0.731), np.float32(0.6995), np.float32(0.4746), np.float32(0.4578), np.float32(0.0), np.float32(0.0), np.float32(0.423), np.float32(0.5798), np.float32(0.0), np.float32(0.0), np.float32(0.5774), np.float32(0.597), np.float32(0.7623), np.float32(0.7887)] +2026-10-15 19:51:22.588345: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:51:22.908532: lr: 0.00073 +2026-10-15 19:51:22.908768: This epoch took 296.033309 s + +2026-10-15 19:51:22.908847: +epoch: 295 +2026-10-15 19:55:45.044394: train loss : -0.7825 +2026-10-15 19:55:50.416735: validation loss: -0.5561 +2026-10-15 19:55:50.417685: Average global foreground Dice: [np.float32(0.9071), np.float32(0.5899), np.float32(0.7169), np.float32(0.7371), np.float32(0.6477), np.float32(0.7704), np.float32(0.663), np.float32(0.704), np.float32(0.0082), np.float32(0.468), np.float32(0.5291), np.float32(0.5828), np.float32(0.5428), np.float32(0.4094), np.float32(0.0), np.float32(0.0), np.float32(0.7108), np.float32(0.5624), np.float32(0.6757), np.float32(0.63), np.float32(0.535), np.float32(0.6199), np.float32(0.7962), np.float32(0.76), np.float32(0.4959), np.float32(0.4554), np.float32(0.0), np.float32(0.0), np.float32(0.4187), np.float32(0.5863), np.float32(0.0), np.float32(0.0), np.float32(0.5735), np.float32(0.6007), np.float32(0.7684), np.float32(0.7922)] +2026-10-15 19:55:50.417922: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 19:55:50.757492: lr: 0.000729 +2026-10-15 19:55:50.757948: This epoch took 267.848885 s + +2026-10-15 19:55:50.758266: +epoch: 296 +2026-10-15 20:00:31.158456: train loss : -0.7814 +2026-10-15 20:00:36.525593: validation loss: -0.5493 +2026-10-15 20:00:36.526625: Average global foreground Dice: [np.float32(0.913), np.float32(0.5485), np.float32(0.7066), np.float32(0.732), np.float32(0.6419), np.float32(0.7737), np.float32(0.6643), np.float32(0.682), np.float32(0.0024), np.float32(0.6057), np.float32(0.5039), np.float32(0.565), np.float32(0.5305), np.float32(0.378), np.float32(0.0), np.float32(0.0), np.float32(0.7172), np.float32(0.5564), np.float32(0.6936), np.float32(0.6001), np.float32(0.5655), np.float32(0.6586), np.float32(0.7999), np.float32(0.7794), np.float32(0.5333), np.float32(0.4276), np.float32(0.0), np.float32(0.0), np.float32(0.3367), np.float32(0.4427), np.float32(0.0), np.float32(0.0), np.float32(0.4836), np.float32(0.4416), np.float32(0.76), np.float32(0.782)] +2026-10-15 20:00:36.526842: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 20:00:36.854281: lr: 0.000728 +2026-10-15 20:00:36.854540: This epoch took 286.095975 s + +2026-10-15 20:00:36.854628: +epoch: 297 +2026-10-15 20:05:09.606462: train loss : -0.7808 +2026-10-15 20:05:14.976234: validation loss: -0.5787 +2026-10-15 20:05:14.977193: Average global foreground Dice: [np.float32(0.9173), np.float32(0.5687), np.float32(0.7442), np.float32(0.7591), np.float32(0.6325), np.float32(0.7791), np.float32(0.682), np.float32(0.5705), np.float32(0.0), np.float32(0.4669), np.float32(0.5631), np.float32(0.6194), np.float32(0.5256), np.float32(0.3502), np.float32(0.0), np.float32(0.0), np.float32(0.7096), np.float32(0.534), np.float32(0.6898), np.float32(0.6424), np.float32(0.5641), np.float32(0.6458), np.float32(0.8189), np.float32(0.8066), np.float32(0.5648), np.float32(0.4682), np.float32(0.0), np.float32(0.0), np.float32(0.4476), np.float32(0.5362), np.float32(0.0), np.float32(0.0), np.float32(0.5649), np.float32(0.5985), np.float32(0.771), np.float32(0.7876)] +2026-10-15 20:05:14.977386: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 20:05:15.308590: lr: 0.000727 +2026-10-15 20:05:15.308857: This epoch took 278.454175 s + +2026-10-15 20:05:15.308986: +epoch: 298 +2026-10-15 20:09:16.315218: train loss : -0.7863 +2026-10-15 20:09:21.670430: validation loss: -0.5607 +2026-10-15 20:09:21.671797: Average global foreground Dice: [np.float32(0.9109), np.float32(0.6047), np.float32(0.7332), np.float32(0.746), np.float32(0.6435), np.float32(0.7828), np.float32(0.6751), np.float32(0.5857), np.float32(0.0184), np.float32(0.4384), np.float32(0.5571), np.float32(0.6147), np.float32(0.5532), np.float32(0.4321), np.float32(0.0), np.float32(0.0), np.float32(0.6905), np.float32(0.5282), np.float32(0.6768), np.float32(0.6424), np.float32(0.5403), np.float32(0.6348), np.float32(0.8005), np.float32(0.7521), np.float32(0.4959), np.float32(0.4428), np.float32(0.0), np.float32(0.0), np.float32(0.4286), np.float32(0.4381), np.float32(0.0), np.float32(0.0), np.float32(0.5557), np.float32(0.6102), np.float32(0.7364), np.float32(0.7529)] +2026-10-15 20:09:21.672182: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 20:09:21.994660: lr: 0.000726 +2026-10-15 20:09:21.994994: This epoch took 246.685776 s + +2026-10-15 20:09:21.995323: +epoch: 299 +2026-10-15 20:13:48.896641: train loss : -0.7888 +2026-10-15 20:13:54.253377: validation loss: -0.5604 +2026-10-15 20:13:54.254313: Average global foreground Dice: [np.float32(0.9112), np.float32(0.6023), np.float32(0.7333), np.float32(0.745), np.float32(0.6441), np.float32(0.7856), np.float32(0.6785), np.float32(0.561), np.float32(0.0508), np.float32(0.4081), np.float32(0.494), np.float32(0.5637), np.float32(0.5514), np.float32(0.4666), np.float32(0.0), np.float32(0.0), np.float32(0.6976), np.float32(0.5487), np.float32(0.6725), np.float32(0.636), np.float32(0.494), np.float32(0.5926), np.float32(0.8489), np.float32(0.8162), np.float32(0.5063), np.float32(0.3851), np.float32(0.0), np.float32(0.0), np.float32(0.3102), np.float32(0.5143), np.float32(0.0), np.float32(0.0), np.float32(0.5673), np.float32(0.5461), np.float32(0.7774), np.float32(0.8039)] +2026-10-15 20:13:54.254614: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 20:13:54.578294: lr: 0.000725 +2026-10-15 20:13:54.578514: saving scheduled checkpoint file... +2026-10-15 20:13:54.587257: saving checkpoint... +2026-10-15 20:13:54.718272: done, saving took 0.14 seconds +2026-10-15 20:13:54.719815: done +2026-10-15 20:13:54.719997: This epoch took 272.724572 s + +2026-10-15 20:13:54.720211: +epoch: 300 +2026-10-15 20:17:56.229600: train loss : -0.7916 +2026-10-15 20:18:01.598840: validation loss: -0.5612 +2026-10-15 20:18:01.600060: Average global foreground Dice: [np.float32(0.9167), np.float32(0.6404), np.float32(0.7113), np.float32(0.7028), np.float32(0.5979), np.float32(0.7151), np.float32(0.6312), np.float32(0.4695), np.float32(0.0429), np.float32(0.4724), np.float32(0.5538), np.float32(0.5891), np.float32(0.6038), np.float32(0.549), np.float32(0.0), np.float32(0.0), np.float32(0.6641), np.float32(0.5637), np.float32(0.6214), np.float32(0.6426), np.float32(0.5745), np.float32(0.586), np.float32(0.6806), np.float32(0.6561), np.float32(0.4196), np.float32(0.4493), np.float32(0.0), np.float32(0.0), np.float32(0.3269), np.float32(0.325), np.float32(0.0), np.float32(0.0), np.float32(0.4953), np.float32(0.4505), np.float32(0.6909), np.float32(0.7238)] +2026-10-15 20:18:01.600713: (interpret this as an estimate for the Dice of the different classes. This is not exact.) +2026-10-15 20:18:01.938170: lr: 0.000724 +2026-10-15 20:18:01.938412: This epoch took 247.218100 s + +2026-10-15 20:18:01.938499: +epoch: 301