Issue in Replicating Model Training for the IDC Dataset

[taoo0316]

After adding the training data downloaded from https://www.kaggle.com/datasets/kasikrit/idc-dataset?resource=download in the folder data/IDC_regular_ps50_idx5/12749, I ran the following line in the terminal:

python run.py -c configs/betavae_IDC.yaml

This gave me the following error message:

Commenting the line out gave me the error message:

FileNotFoundError: [Errno 2] No such file or directory: './data/IDC_regular_ps50_idx5/label.csv'

[wangqianwen0418]

Hi @taoo0316 , thank you for raising this issue. I noticed a few things that might help:

1. The downloaded dataset should be placed directly under the data folder. After unzipping, the folder name will be IDC_regular_ps50_idx5.
2. You can generate the label.csv file by running the script located at server/data/IDC_label.py.

[taoo0316]

Hi @wangqianwen0418, thanks so much for the prompt reply!

1. Got it. I now placed the data under the correct folder.
2. After running :

   python server/data/IDC_label.py

I got the following error message:

FileNotFoundError: [Errno 2] No such file or directory: '../../front/public/assets/results_IDC_all.csv'

Seems like there are some additional files needed? Or maybe I am missing something again.

Thank you:)

[wangqianwen0418]

I'm not sure which version you're using, but is it possible that the generate_IDC_label() function is commented out in the IDC_label.py file?

Thanks again for raising the issue. I will try to put together some docs for reproducing the training process with the example datasets when I have some free time

# %%
if __name__=='__main__':
    generate_IDC_label()
    add_info2results()
# %%

[taoo0316]

Hi @wangqianwen0418, thank you for the reply!

No, I don't think the function is commented out in the version I used.

Sure, some docs for reproducing the training process would be really helpful:) Thank you.