d3rlpy.models.encoders.PixelEncoderFactory

class d3rlpy.models.encoders.PixelEncoderFactory(filters=None, feature_size=512, activation='relu', use_batch_norm=False)[source]

Pixel encoder factory class.

This is the default encoder factory for image observation.

Parameters

  • filters (list) – list of tuples consisting with (filter_size, kernel_size, stride). If None, Nature DQN-based architecture is used.

  • feature_size (int) – the last linear layer size.

  • activation (str) – activation function name.

  • use_batch_norm (bool) – flag to insert batch normalization layers.

Methods

create(observation_shape)[source]

Returns PyTorch’s state enocder module.

Parameters

observation_shape (Sequence[int]) – observation shape.

Returns

an enocder object.

Return type

d3rlpy.models.torch.encoders.PixelEncoder

create_with_action(observation_shape, action_size, discrete_action=False)[source]

Returns PyTorch’s state-action enocder module.

Parameters

  • observation_shape (Sequence[int]) – observation shape.

  • action_size (int) – action size. If None, the encoder does not take action as input.

  • discrete_action (bool) – flag if action-space is discrete.

Returns

an enocder object.

Return type

d3rlpy.models.torch.encoders.PixelEncoderWithAction

get_params(deep=False)[source]

Returns encoder parameters.

Parameters

deep (bool) – flag to deeply copy the parameters.

Returns

encoder parameters.

Return type

Dict[str, Any]

get_type()

Returns encoder type.

Returns

encoder type.

Return type

str

Attributes

TYPE: ClassVar[str] = 'pixel'