-
data
- test: contain 860 training data and 172 test data
- strict_test: contain 700 training data and 32 test data
-
pretrain
: codes to pretrain CMPNN and save pretrained model -
src
: codes of CMPNN
- test using trained model
python test.py --dataset test
#or test on strict_test set
python test.py --dataset strict_test
- train the model
#train model and test
python main.py
#or train model and test on strict_test set
python main_strict_test.py
- running environment
conda env create -f environment.yaml
#or
pip install -r requirements.txt