garage.tf.baselines package¶
Baseline estimators for TensorFlow-based algorithms.
-
class
ContinuousMLPBaseline
(env_spec, subsample_factor=1.0, num_seq_inputs=1, regressor_args=None, name='ContinuousMLPBaseline')[source]¶ Bases:
garage.np.baselines.base.Baseline
A value function using a MLP network.
-
class
GaussianCNNBaseline
(env_spec, subsample_factor=1.0, regressor_args=None, name='GaussianCNNBaseline')[source]¶ Bases:
garage.np.baselines.base.Baseline
GaussianCNNBaseline With Model.
It fits the input data to a gaussian distribution estimated by a CNN.
Parameters: - env_spec (garage.envs.env_spec.EnvSpec) – Environment specification.
- subsample_factor (float) – The factor to subsample the data. By default it is 1.0, which means using all the data.
- regressor_args (dict) – Arguments for regressor.
-
class
GaussianMLPBaseline
(env_spec, subsample_factor=1.0, num_seq_inputs=1, regressor_args=None, name='GaussianMLPBaseline')[source]¶ Bases:
garage.np.baselines.base.Baseline
A value function using Gaussian MLP network.