garage.np.baselines

Baselines (value functions) which use NumPy as a numerical backend.

class Baseline[source]

Bases: abc.ABC

Inheritance diagram of garage.np.baselines.Baseline

Base class for all baselines.

abstract fit(paths)[source]

Fit regressor based on paths.

Parameters

paths (dict[numpy.ndarray]) – Sample paths.

abstract predict(paths)[source]

Predict value based on paths.

Parameters

paths (dict[numpy.ndarray]) – Sample paths.

Returns

Predicted value.

Return type

numpy.ndarray

class LinearFeatureBaseline(env_spec, reg_coeff=1e-05, name='LinearFeatureBaseline')[source]

Bases: garage.np.baselines.baseline.Baseline

Inheritance diagram of garage.np.baselines.LinearFeatureBaseline

A linear value function (baseline) based on features.

Parameters

  • env_spec (garage.envs.env_spec.EnvSpec) – Environment specification.

  • reg_coeff (float) – Regularization coefficient.

  • name (str) – Name of baseline.

get_param_values()[source]

Get parameter values.

Returns

A list of values of each parameter.

Return type

List[np.ndarray]

set_param_values(flattened_params)[source]

Set param values.

Parameters

flattened_params (np.ndarray) – A numpy array of parameter values.

fit(paths)[source]

Fit regressor based on paths.

Parameters

paths (list[dict]) – Sample paths.

predict(paths)[source]

Predict value based on paths.

Parameters

paths (list[dict]) – Sample paths.

Returns

Predicted value.

Return type

numpy.ndarray

class LinearMultiFeatureBaseline(env_spec, features=None, reg_coeff=1e-05, name='LinearMultiFeatureBaseline')[source]

Bases: garage.np.baselines.LinearFeatureBaseline

Inheritance diagram of garage.np.baselines.LinearMultiFeatureBaseline

A linear value function (baseline) based on features.

Parameters

  • env_spec (garage.envs.env_spec.EnvSpec) – Environment specification.

  • reg_coeff (float) – Regularization coefficient.

  • features (list[str]) – Name of features.

  • name (str) – Name of baseline.

get_param_values()

Get parameter values.

Returns

A list of values of each parameter.

Return type

List[np.ndarray]

set_param_values(flattened_params)

Set param values.

Parameters

flattened_params (np.ndarray) – A numpy array of parameter values.

fit(paths)

Fit regressor based on paths.

Parameters

paths (list[dict]) – Sample paths.

predict(paths)

Predict value based on paths.

Parameters

paths (list[dict]) – Sample paths.

Returns

Predicted value.

Return type

numpy.ndarray

class ZeroBaseline(env_spec)[source]

Bases: garage.np.baselines.baseline.Baseline

Inheritance diagram of garage.np.baselines.ZeroBaseline

Base class for all baselines.

get_param_values(**kwargs)[source]
set_param_values(val, **kwargs)[source]
fit(paths)[source]

Fit regressor based on paths.

Parameters

paths (dict[numpy.ndarray]) – Sample paths.

predict(path)[source]

Predict value based on paths.

Parameters

paths (dict[numpy.ndarray]) – Sample paths.

Returns

Predicted value.

Return type

numpy.ndarray

predict_n(paths)[source]