The algorithm failed on one or more cases.

Hello, can you please send me the logs of my submission? Are these the warnings only as suggested by organizers on other posts on the same topic? Can I get my Dice score for segmentation if this is only a warning? Or do I have to fix it? Also, I am writing all three file: outputs = [result, trimesh.primitives.Box(), #optional visualization task, leave Box() as placeholder if you do not participate trimesh.primitives.Box()] #optional volumetric meshing task, leave Box() as placeholder if you do not participate

It runs successfully on the grand_challenge Try_out algorithm on the available images.

Thank You Cheers Abbas

Hi

Your algorithm gave the following output:

2023-07-20T19:07:25.573000+00:00 nnUNet_raw is not defined and nnU-Net can only be used on data for which preprocessed files are already present on your system. nnU-Net cannot be used for experiment planning and preprocessing like this. If this is not intended, please read documentation/setting_up_paths.md for information on how to set this up properly. 2023-07-20T19:07:25.573000+00:00 nnUNet_preprocessed is not defined and nnU-Net can not be used for preprocessing or training. If this is not intended, please read documentation/setting_up_paths.md for information on how to set this up. 2023-07-20T19:07:25.573000+00:00 nnUNet_results is not defined and nnU-Net cannot be used for training or inference. If this is not intended behavior, please read documentation/setting_up_paths.md for information on how to set this up. 2023-07-20T19:07:25.573000+00:00 this 2023-07-20T19:07:25.573000+00:00 input FILES ['bc5eae0e-1bc8-468a-98a2-00547e42ed2.mha'] 2023-07-20T19:07:25.573000+00:00 hash 4123413483339228137 2023-07-20T19:07:25.573000+00:00 path /input/images/ct/bc5eae0e-1bc8-468a-98a2-00547... 2023-07-20T19:07:25.573000+00:00 Name: 0, dtype: object 2023-07-20T19:07:25.573000+00:00 0 2023-07-20T19:07:25.573000+00:00 Image Name bc5eae0e-1bc8-468a-98a2-00547e42ed2.mha 2023-07-20T19:07:25.573000+00:00 Working on device: cuda 2023-07-20T19:07:25.573000+00:00 There are 1 cases in the source folder 2023-07-20T19:07:25.573000+00:00 I am process 0 out of 1 (max process ID is 0, we start counting with 0!) 2023-07-20T19:07:25.574000+00:00 There are 1 cases that I would like to predict 2023-07-20T19:08:11.585000+00:00 nnUNet_raw is not defined and nnU-Net can only be used on data for which preprocessed files are already present on your system. nnU-Net cannot be used for experiment planning and preprocessing like this. If this is not intended, please read documentation/setting_up_paths.md for information on how to set this up properly. 2023-07-20T19:08:11.585000+00:00 nnUNet_preprocessed is not defined and nnU-Net can not be used for preprocessing or training. If this is not intended, please read documentation/setting_up_paths.md for information on how to set this up.

The following error was raised: RuntimeError: Background workers died. Look for the error message further up! If there is none then your RAM was full and the worker was killed by the OS. Use fewer workers

Thanks for listing the log file. I tried to solve the error but could not do anything on my behalf. Even the grand challenge website runs with the available images, and if I test it on any image from the training set throws the same error again.

''' RuntimeError: Background workers died. Look for the error message further up! If there is none then your RAM was full and the worker was killed by the OS. Use fewer workers or get more RAM in that case!''''

Any further help will be appreciated. Thanks Cheers Abbas

Hi,

I am not sure if I do understand your question correctly.

You mean it works if you select the option "try out algorithm" on grand-challenge, but it does not work if you test it locally on the training set?

Antonio

Hi, yes, it is: You mean it works if you select the option "try out algorithm" on grand-challenge, but it does not work if you test it locally on the training set?

But I also found that it fails even if I select an image from the training data set like K20.nrrd I think it's the memory issue; we cannot use a large model or, for a large model, a large input image. I was successful with he same image using rather a small model, compared to my previous approach.

Cheers Abbas