New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Minor Code refinement for performance #6

Open

dxyzx0 wants to merge 4 commits into lifanchen-simm:master from dxyzx0:master

dxyzx0 commented Oct 22, 2020

In model.py

These two for-loops in the model.py lowers the speed of training and testing, since it doesn't utilize broadcasting of matrix operation in torch.
In main.py and mol_featurizer.py
I trained my model on a larger dataset with protein length up to 2000 and 70w interactions. We should use joblib to load and dump large files.

Further Refinement:
I like this model very much. I use this model for production. Hence, there's my refinement of the code https://github.com/DingXiangYuanZhiXing/transformerCPI_DTA

dxyzx0 added 4 commits

October 22, 2020 17:29


          Remove some redundant for-loop

a5d84d6

Remove some redundant for-loop with pytorch build-in functionality.


          Replace pickle with joblib

ea8db59

joblib is more robust for large dataset


          Replace pickle with joblib

594af46

joblib is more robust for large dataset


          turn atom_num protein_num into torch.Tensor

1fba6b2

comeonzyx commented Feb 29, 2024

Hello,i would like to ask that do you have "dataset/celegans_uniprot.csv"?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet