malt.agents.player.SequentialModelBasedPlayer

class malt.agents.player.SequentialModelBasedPlayer(*args, **kwargs)

Bases: malt.agents.player.ModelBasedPlayer

Model based player with step size equal one.

Examples

>>> import malt
>>> player = SequentialModelBasedPlayer(
...    model = malt.models.supervised_model.SupervisedModel(
...        representation=malt.models.representation.DGLRepresentation(
...            out_features=128
...        ),
...        regressor=malt.models.regressor.NeuralNetworkRegressor(
...            in_features=128,
...        ),
...    ),
...    policy=malt.policy.UtilityFunction(),
...    trainer=malt.trainer.get_default_trainer(),
...    merchant=malt.agents.merchant.DatasetMerchant(
...        malt.data.collections.linear_alkanes(10),
...    ),
...    assayer=malt.agents.assayer.DatasetAssayer(
...        malt.data.collections.linear_alkanes(10),
...    )
... )

>>> while True:
...     if player.step() is None:
...         break

__init__(*args, **kwargs)

Methods

__init__(*args, **kwargs)
assay(dataset)
merchandize(dataset)
prioritize()
step()
train()