About "Processed input"

In the section where we can run the model on a larger dataset, the input comes from a “processed” folder. What processing is done on this? Is it just mean subtracted and divided by std?

1 Like

Yes, but reading the code in util.py, I believe they’ve also grouped the patches into “x” and “y” using the h5py module, and they’ve appended the dimension sizes to the filenames. I’m not sure please correct me if I’m wrong. I wish we could see the processing code for this, it would be really helpful as processing is an integral part of machine learning.