Home > Chapter 9: On-policy Prediction with Approximation

rlai.gpi.state_action_value.function_approximation.models.StateActionFunctionApproximationModel

Base class for models that approximate state-action value functions.

rlai.gpi.state_action_value.function_approximation.models.feature_extraction.StateActionFeatureExtractor

Feature extractor.

rlai.models.FunctionApproximationModel

Base class for models that approximate functions.

rlai.models.feature_extraction.FeatureExtractor

Base feature extractor for all others. This class does not define any extraction function, since the signature of
    such a function depends on what, conceptually, we're extracting features from. The definition of this signature is
    deferred to inheriting classes that are closer to their conceptual extraction targets.

rlai.state_value.function_approximation.models.StateFunctionApproximationModel

Base class for models that approximate state-action value functions.

rlai.state_value.function_approximation.models.sklearn.SKLearnSGD

State-action value modeler based on the SKLearnSGD algorithm.

rlai.gpi.state_action_value.function_approximation.models.sklearn.SKLearnSGD

State-action value modeler based on the SKLearnSGD algorithm.

rlai.models.sklearn.SKLearnSGD

Wrapper for the sklearn.linear_model.SGDRegressor implemented by scikit-learn.