allennlp.modules.seq2seq_encoders.seq2seq_encoder#

Seq2SeqEncoder#

Seq2SeqEncoder(self, stateful:bool=False) -> None

A Seq2SeqEncoder is a Module that takes as input a sequence of vectors and returns a modified sequence of vectors. Input shape : (batch_size, sequence_length, input_dim); output shape : (batch_size, sequence_length, output_dim).

We add two methods to the basic Module API: :func:get_input_dim() and :func:get_output_dim(). You might need this if you want to construct a Linear layer using the output of this encoder, or to raise sensible errors for mis-matching input dimensions.

get_input_dim#

Seq2SeqEncoder.get_input_dim(self) -> int

Returns the dimension of the vector input for each element in the sequence input to a Seq2SeqEncoder. This is not the shape of the input tensor, but the last element of that shape.

get_output_dim#

Seq2SeqEncoder.get_output_dim(self) -> int

Returns the dimension of each vector in the sequence output by this Seq2SeqEncoder. This is not the shape of the returned tensor, but the last element of that shape.

is_bidirectional#

Seq2SeqEncoder.is_bidirectional(self) -> bool

Returns True if this encoder is bidirectional. If so, we assume the forward direction of the encoder is the first half of the final dimension, and the backward direction is the second half.