vp_suite.model_blocks.predrnn

class ActionConditionalSpatioTemporalLSTMCell(in_channel, num_hidden, height, width, filter_size, stride, layer_norm)

Bases: vp_suite.base.base_model_block.VPModelBlock

Action-Conditional Spatio-Temporal LSTM Cell (v2) as introduced in “PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learning” by Wang et al. (https://arxiv.org/pdf/2103.09504.pdf) and implemented in (https://github.com/thuml/predrnn-pytorch).

CODE_REFERENCE = 'https://github.com/thuml/predrnn-pytorch'

The code location of the reference implementation.

MATCHES_REFERENCE: str = 'Yes'

A comment indicating whether the implementation in this package matches the reference.

NAME: str = 'Spatio-Temporal LSTM Cell (Action-Conditional)'

The clear-text name for this model block.

PAPER_REFERENCE = 'https://arxiv.org/abs/2103.09504'

The publication where this model was introduced first.

__init__(in_channel, num_hidden, height, width, filter_size, stride, layer_norm)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x_t, h_t, c_t, m_t, a_t)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool

class SpatioTemporalLSTMCell(in_channel, num_hidden, height, width, filter_size, stride, layer_norm)

Bases: vp_suite.base.base_model_block.VPModelBlock

Spatio-Temporal LSTM Cell (v2) as introduced in “PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learning” by Wang et al. (https://arxiv.org/pdf/2103.09504.pdf) and implemented in (https://github.com/thuml/predrnn-pytorch).

CODE_REFERENCE = 'https://github.com/thuml/predrnn-pytorch'

The code location of the reference implementation.

MATCHES_REFERENCE: str = 'Yes'

A comment indicating whether the implementation in this package matches the reference.

NAME: str = 'Spatio-Temporal LSTM Cell'

The clear-text name for this model block.

PAPER_REFERENCE = 'https://arxiv.org/abs/2103.09504'

The publication where this model was introduced first.

__init__(in_channel, num_hidden, height, width, filter_size, stride, layer_norm)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x_t, h_t, c_t, m_t)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool