HUBioDataLab / DrugGEN

Official implementation of DrugGEN
GNU General Public License v3.0
50 stars 16 forks source link

[Refactor] seperate the load_data function into sub-fuctions #13

Closed osbm closed 1 year ago

osbm commented 1 year ago

This would allow us to separate preprocessing steps, and increase the readability and documentation. As a next step forward we can directly add these into the dataset, and cache the results.

I generated the Z values from directly pytorch functions. And It seems that the drugs and molecules are passing through the same steps.

If this PR looks ok, i can remove the commented lines