Spared_EGN

EGN state-of-the-art model adapted for SPARED datasets.

Environment set up

Create environment:

cd Spared_EGN
conda env create -f environment.yml -n egn_spared
conda activate egn_spared
pip install -r requirements.txt

Running the complete EGN framework (exemplar building + gene expression prediction)

python run_full_egn.py --dataset [SPARED_dataset_name] --lr [learning_rate]

Running each EGN phase separately

• Building the exemplars:

python exemplars.py --dataset [SPARED_dataset_name]

• Training and testing the gene expression prediction model:

python main.py --dataset [SPARED_dataset_name] --lr [learning_rate]