Bus error (core dumped) nnUNet

[kabbas570]

Hello, any hint for this error? It happens during inference.

2023-08-14T18:51:18.968000+00:00 0%| | 0/18 [00:00<?, ?it/s] 2023-08-14T18:51:18.968000+00:00 6%|▌ | 1/18 [00:00<00:07, 2.38it/s] 2023-08-14T18:51:18.968000+00:00 11%|█ | 2/18 [00:01<00:10, 1.56it/s] 2023-08-14T18:51:18.968000+00:00 17%|█▋ | 3/18 [00:02<00:10, 1.41it/s] 2023-08-14T18:51:18.968000+00:00 22%|██▏ | 4/18 [00:02<00:10, 1.34it/s] 2023-08-14T18:51:18.968000+00:00 28%|██▊ | 5/18 [00:03<00:09, 1.31it/s] 2023-08-14T18:51:18.968000+00:00 33%|███▎ | 6/18 [00:04<00:09, 1.29it/s] 2023-08-14T18:51:18.968000+00:00 39%|███▉ | 7/18 [00:05<00:08, 1.28it/s] 2023-08-14T18:51:18.968000+00:00 44%|████▍ | 8/18 [00:05<00:07, 1.27it/s] 2023-08-14T18:51:18.968000+00:00 50%|█████ | 9/18 [00:06<00:07, 1.27it/s] 2023-08-14T18:51:18.968000+00:00 56%|█████▌ | 10/18 [00:07<00:06, 1.26it/s] 2023-08-14T18:51:18.968000+00:00 61%|██████ | 11/18 [00:08<00:05, 1.26it/s] 2023-08-14T18:51:18.968000+00:00 67%|██████▋ | 12/18 [00:09<00:04, 1.26it/s] 2023-08-14T18:51:18.968000+00:00 72%|███████▏ | 13/18 [00:09<00:03, 1.26it/s] 2023-08-14T18:51:18.968000+00:00 78%|███████▊ | 14/18 [00:10<00:03, 1.26it/s] 2023-08-14T18:51:18.968000+00:00 83%|████████▎ | 15/18 [00:11<00:02, 1.25it/s] 2023-08-14T18:51:18.968000+00:00 89%|████████▉ | 16/18 [00:12<00:01, 1.25it/s] 2023-08-14T18:51:18.968000+00:00 94%|█████████▍| 17/18 [00:13<00:00, 1.25it/s] 2023-08-14T18:51:18.968000+00:00 100%|██████████| 18/18 [00:13<00:00, 1.25it/s] 2023-08-14T18:51:18.968000+00:00 100%|██████████| 18/18 [00:13<00:00, 1.29it/s]

2023-08-14T18:51:42.973000+00:00 Bus error (core dumped)

[kabbas570]

I am using the following args..

    #Model 1 prediction
    predictor = nnUNetPredictor(
        tile_step_size=0.5,
        use_gaussian=True,
        use_mirroring=True,
        perform_everything_on_gpu=True,
        device=torch.device('cuda', 0),
        verbose=True,
        verbose_preprocessing=True,
        allow_tqdm=True
    )

    predictor.initialize_from_trained_model_folder(
        str(self.nnunet_results / 'Dataset050_SEGA/nnUNetTrainerNoDeepSupervision__nnUNetPlans__3d_lowres'),
       use_folds=(0,1,2,3,4),
       #use_folds=(0,),
        checkpoint_name='checkpoint_final.pth',
    )

    print('Model_1 input files', os.listdir(input_folder))

    predictor.predict_from_files(input_folder,
                                model1_output,
                                save_probabilities=False, overwrite=True,
                                num_processes_preprocessing=1, num_processes_segmentation_export=1,
                                folder_with_segs_from_prev_stage=None, num_parts=1, part_id=0)  

[Kobalt93]

Hi kabbas570, Have you checked the RAM of your system? Is it getting full?

[Kobalt93]

Please free to reopen the issue if it still exists