schrodinger.active_learning.mq_backend module

class schrodinger.active_learning.mq_backend.Backend(args)

Bases: object

__init__(args)

get_csv_header(fname)

unpackModel(zipped_model)

modelPrep()

This step ensures that only selected keys from TorchGraphConv models are selected from the production folder. This is to avoid loading keys from models that are mentioned in the report but were not saved to slim model dir (which would throw an error that the directory doesn’t exist).

loadHotModels(key, hp)

run()

recv()

validateOutputs()

sendPredictions()

handleRequest(msg)

handleSTOP(msg)

handlePONG(msg)

handleCSV(msg)

predict(columns)

sortWriteResults(ascending=True)

Sort the results from ligand_ml by prediction value. Write the sorted result to output CSV without header.

schrodinger.active_learning.mq_backend.parse_args(argv=None)

schrodinger.active_learning.mq_backend.main()