garage.tf.embeddings.encoder

Encoders in TensorFlow.

class Encoder(name)

Bases: garage.np.embeddings.Encoder, garage.tf.models.Module

Base class for encoders in TensorFlow.

get_latent(self, input_value)

Get a sample of embedding for the given input.

Parameters

input_value (numpy.ndarray) – Tensor to encode.

Returns

An embedding sampled from embedding distribution. dict: Embedding distribution information.

Return type

numpy.ndarray

Note

It returns an embedding and a dict, with keys - mean (numpy.ndarray): Mean of the distribution. - log_std (numpy.ndarray): Log standard deviation of the

distribution.

get_latents(self, input_values)

Get samples of embedding for the given inputs.

Parameters

input_values (numpy.ndarray) – Tensors to encode.

Returns

Embeddings sampled from embedding distribution. dict: Embedding distribution information.

Return type

numpy.ndarray

Note

It returns an embedding and a dict, with keys - mean (list[numpy.ndarray]): Means of the distribution. - log_std (list[numpy.ndarray]): Log standard deviations of the

distribution.

clone(self, name)

Return a clone of the encoder.

It copies the configuration of the primitive and also the parameters.

Parameters

name (str) – Name of the newly created encoder. It has to be different from source encoder if cloned under the same computational graph.

Returns

Newly cloned encoder.

Return type

garage.tf.embeddings.encoder.Encoder

property spec(self)

garage.InOutSpec: Input and output space.

property input_dim(self)

int: Dimension of the encoder input.

property output_dim(self)

int: Dimension of the encoder output (embedding).

reset(self, do_resets=None)

Reset the encoder.

This is effective only to recurrent encoder. do_resets is effective only to vectoried encoder.

For a vectorized encoder, do_resets is an array of boolean indicating which internal states to be reset. The length of do_resets should be equal to the length of inputs.

Parameters

do_resets (numpy.ndarray) – Bool array indicating which states to be reset.

property name(self)

str: Name of this module.

property state_info_specs(self)

State info specification.

Returns

keys and shapes for the information related to the

module’s state when taking an action.

Return type

List[str]

property state_info_keys(self)

State info keys.

Returns

keys for the information related to the module’s state

when taking an input.

Return type

List[str]

terminate(self)

Clean up operation.

get_trainable_vars(self)

Get trainable variables.

Returns

A list of trainable variables in the current

variable scope.

Return type

List[tf.Variable]

get_global_vars(self)

Get global variables.

Returns

A list of global variables in the current

variable scope.

Return type

List[tf.Variable]

get_regularizable_vars(self)

Get all network weight variables in the current scope.

Returns

A list of network weight variables in the

current variable scope.

Return type

List[tf.Variable]

get_params(self)

Get the trainable variables.

Returns

A list of trainable variables in the current

variable scope.

Return type

List[tf.Variable]

get_param_shapes(self)

Get parameter shapes.

Returns

A list of variable shapes.

Return type

List[tuple]

get_param_values(self)

Get param values.

Returns

Values of the parameters evaluated in

the current session

Return type

np.ndarray

set_param_values(self, param_values)

Set param values.

Parameters

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

flat_to_params(self, flattened_params)

Unflatten tensors according to their respective shapes.

Parameters

flattened_params (np.ndarray) – A numpy array of flattened params.

Returns

A list of parameters reshaped to the

shapes specified.

Return type

List[np.ndarray]

class StochasticEncoder(name)

Bases: garage.np.embeddings.StochasticEncoder, garage.tf.models.StochasticModule

Base class for stochastic encoders in TensorFlow.

build(self, embedding_input, name=None)

Build encoder.

After buil, self.distribution is a Gaussian distribution conitioned on embedding_input.

Parameters

• embedding_input (tf.Tensor) – Embedding input.

• name (str) – Name of the model, which is also the name scope.

property distribution(self)

object: Embedding distribution.

property spec(self)

garage.InOutSpec: Input and output space.

property input_dim(self)

int: Dimension of the encoder input.

property output_dim(self)

int: Dimension of the encoder output (embedding).

reset(self, do_resets=None)

Reset the encoder.

This is effective only to recurrent encoder. do_resets is effective only to vectoried encoder.

For a vectorized encoder, do_resets is an array of boolean indicating which internal states to be reset. The length of do_resets should be equal to the length of inputs.

Parameters

do_resets (numpy.ndarray) – Bool array indicating which states to be reset.

property name(self)

str: Name of this module.

property state_info_specs(self)

State info specification.

Returns

keys and shapes for the information related to the

module’s state when taking an action.

Return type

List[str]

property state_info_keys(self)

State info keys.

Returns

keys for the information related to the module’s state

when taking an input.

Return type

List[str]

terminate(self)

Clean up operation.

get_trainable_vars(self)

Get trainable variables.

Returns

A list of trainable variables in the current

variable scope.

Return type

List[tf.Variable]

get_global_vars(self)

Get global variables.

Returns

A list of global variables in the current

variable scope.

Return type

List[tf.Variable]

get_regularizable_vars(self)

Get all network weight variables in the current scope.

Returns

A list of network weight variables in the

current variable scope.

Return type

List[tf.Variable]

get_params(self)

Get the trainable variables.

Returns

A list of trainable variables in the current

variable scope.

Return type

List[tf.Variable]

get_param_shapes(self)

Get parameter shapes.

Returns

A list of variable shapes.

Return type

List[tuple]

get_param_values(self)

Get param values.

Returns

Values of the parameters evaluated in

the current session

Return type

np.ndarray

set_param_values(self, param_values)

Set param values.

Parameters

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

flat_to_params(self, flattened_params)

Unflatten tensors according to their respective shapes.

Parameters

flattened_params (np.ndarray) – A numpy array of flattened params.

Returns

A list of parameters reshaped to the

shapes specified.

Return type

List[np.ndarray]