NaN values in feature files before training neural network

[gj0706]

After implementing extract.py in feature_extraction folder, I tried to run agg.py in neural_network folder but got ValueError("Input contains NaN").

Do you have any instruction on how to handle the NaN values? I tried preprocess.py in preprocessing folder but got an error :
FileNotFoundError: [Errno 2] File b'../features/raw/processed/included_fids.csv' does not exist: b'../features/raw/processed/included_fids.csv'

Do you have any idea why this happened? Thank you!

[mauk95]

I also get NaN values in feature files generated by extract.py in raw/ folder. The downloaded feature files don't contain NaNs. Don't get this either.