The systems studied are: Double-well, Ala2, Chignolin, and BPTI
An environment.yml file is provided in the project root. Run:
conda env create -f environment.yml
conda activate grexThe environment.yml already includes the following dependencies (including pip-installed packages):
- einops
- pytorch
- numpy
- bgflow
- OpenMM
- bgmol
- nflows
- allegro (for Graph Neural Networks)
- nequip (for Graph Neural Networks)
- matplotlib
If you need a CUDA-enabled PyTorch build, create the environment first, then install the matching version from the official PyTorch guide.
