Prediction of fragment free energy maps from protein structure with ConvNet
This repo must be kept clean and organized! Please do not change original functions! If you do change them, keep it for your own use, and don't push it here. You can only push required tasks, while making sure that your code satisfy the following:
- Code must be both memory and speed efficient
- Code must be easily redable, clean
- Every function needs unittest, which validates its correctness!
- All arithmethic operations performed on grid(usually large(100x100x100)) must be done with torch.cuda()!
python
3.6 >PyTorch
cuda-10
gcc
7 >TorchProteinLibrary
PyVista