LSTM#

class pytorch_forecasting.models.nn.rnn.LSTM(input_size: int, hidden_size: int, num_layers: int = 1, bias: bool = True, batch_first: bool = False, dropout: float = 0.0, bidirectional: bool = False, proj_size: int = 0, device=None, dtype=None)[source]#
class pytorch_forecasting.models.nn.rnn.LSTM(*args, **kwargs)

Bases: RNN, LSTM

LSTM that can handle zero-length sequences

Methods

handle_no_encoding(hidden_state, ...)

Mask the hidden_state where there is no encoding.

init_hidden_state(x)

Initialise a hidden_state.

repeat_interleave(hidden_state, n_samples)

Duplicate the hidden_state n_samples times.
handle_no_encoding(hidden_state: Tuple[Tensor, Tensor] | Tensor, no_encoding: BoolTensor, initial_hidden_state: Tuple[Tensor, Tensor] | Tensor) → Tuple[Tensor, Tensor] | Tensor[source]#

Mask the hidden_state where there is no encoding.

Parameters:

  • hidden_state (HiddenState) – hidden state where some entries need replacement

  • no_encoding (torch.BoolTensor) – positions that need replacement

  • initial_hidden_state (HiddenState) – hidden state to use for replacement

Returns:

hidden state with propagated initial hidden state where appropriate

Return type:

HiddenState

init_hidden_state(x: Tensor) → Tuple[Tensor, Tensor] | Tensor[source]#

Initialise a hidden_state.

Parameters:

x (torch.Tensor) – network input

Returns:

default (zero-like) hidden state

Return type:

HiddenState

repeat_interleave(hidden_state: Tuple[Tensor, Tensor] | Tensor, n_samples: int) → Tuple[Tensor, Tensor] | Tensor[source]#

Duplicate the hidden_state n_samples times.

Parameters:

  • hidden_state (HiddenState) – hidden state to repeat

  • n_samples (int) – number of repetitions

Returns:

repeated hidden state

Return type:

HiddenState