- Run
git clone [email protected]:Deceptrax123/Drug-Interaction-Using-GNNs.git
- Create a Virtual environment and activate it
- Run
pip install -r requirements.txt
- Create a new branch for every new model such as GAT, GCN, SAGE etc. For instance, run
git checkout -b GCN
to create and run a Graph Convolutional Model - Commit and Push all changes to that specific branch only. I will review and merge if no further changes are needed.
You may download all the Graph Tensors with their folds over here
- Before contributing, run the following command:
export PYTHONPATH="/path/to/your/project"
- Follow the same directoy structure as follws:
├── Dataset
│ ├── Molecule_dataset.py
├── Metrics
│ ├── metrics.py
├── GAT
│ ├── train.py
│ ├── model.py
├── GCN
│ ├── train.py
│ ├── model.py
├── SAGE
│ ├── ....
│ ├── ....
├── requirements.txt
├── contributing.md
└── .gitignore