molecular-active-learning

Active learning, generator, and ML lives here.

Generator

The generator should be called at the start of every MD pipline to provide a ligand.

python run_generator.py -i -s /tmp/ -n 1 -o <unique place for pipeline to find output>

Aggregator

python learner_agg.py -i ~/Model-generation/output/ -o ./ -s ~/Model-generation/input/john_smiles_kinasei.smi 

After some learning stage

Learner

For testing one can run:

python run_learner.py -f -o tmp/ --data_path tmp/fake_data/ --smiles_file sample_input.csv