What does it mean to find the best configuration in the 5-fold cross-validation under the 3d_fullres configuration?

[angolin22]

When I use this command nnUNetv2_find_best_configuration DATASET_NAME_OR_ID -c 3d_fullres, because I don't know much about his theoretical knowledge, I don't know what he means to decide the best configuration. May I ask what is his theoretical knowledge?

[FabianIsensee]

This command uses the 5-fold cross validation to determine which if the configurations among the ones given with -c is the best. It also checks for possible ensembles and will configure the postprocessing. I recommend you take a look at the paper & documentation if you need more information

[shiblyg]

Hi While I am running this getting this error: Traceback (most recent call last): File "/home/msadi002/envs/nnUnet2/lib/python3.10/site-packages/nnunetv2/evaluation/evaluate_predictions.py", line 26, in key_to_label_or_region return int(key) ValueError: invalid literal for int() with base 10: '(3,)'

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "/home/msadi002/envs/nnUnet2/bin/nnUNetv2_find_best_configuration", line 8, in sys.exit(find_best_configuration_entry_point()) File "/home/msadi002/envs/nnUnet2/lib/python3.10/site-packages/nnunetv2/evaluation/find_best_configuration.py", line 295, in find_best_configuration_entry_point find_best_configuration(dataset_name, model_dict, allow_ensembling=not args.disable_ensembling, File "/home/msadi002/envs/nnUnet2/lib/python3.10/site-packages/nnunetv2/evaluation/find_best_configuration.py", line 103, in find_best_configuration 'result': load_summary_json(join(merged_output_folder, 'summary.json'))['foreground_mean']['Dice'] File "/home/msadi002/envs/nnUnet2/lib/python3.10/site-packages/nnunetv2/evaluation/evaluate_predictions.py", line 54, in load_summary_json results['mean'] = {key_to_label_or_region(k): results['mean'][k] for k in results['mean'].keys()} File "/home/msadi002/envs/nnUnet2/lib/python3.10/site-packages/nnunetv2/evaluation/evaluate_predictions.py", line 54, in results['mean'] = {key_to_label_or_region(k): results['mean'][k] for k in results['mean'].keys()} File "/home/msadi002/envs/nnUnet2/lib/python3.10/site-packages/nnunetv2/evaluation/evaluate_predictions.py", line 31, in key_to_label_or_region return tuple([int(i) for i in splitted]) File "/home/msadi002/envs/nnUnet2/lib/python3.10/site-packages/nnunetv2/evaluation/evaluate_predictions.py", line 31, in return tuple([int(i) for i in splitted]) ValueError: invalid literal for int() with base 10: '' .....It takes 28 s to finish the task.....

[dojoh]

Hey,

which command are you running precisely so I can try to reproduce the problem?

Cheers Ole Johannsen