emloop_tensorflow.models.rnn_blocks

Classes

  • RNNBlock: Base block for convolution-like blocks.
class emloop_tensorflow.models.rnn_blocks.RNNBlock(sequence_length=None, **kwargs)[source]

Bases: emloop_tensorflow.models.blocks.BaseBlock

Base block for convolution-like blocks.

Inheritance diagram of RNNBlock

__init__(sequence_length=None, **kwargs)[source]

Try to parse and create new RNNBlock.

_handle_parsed_args(bidirectional, cell_type, num_units)[source]

Handle (most likely save) the arguments matched with the regexp in the code.

Parameters:args – parsed arguments
Return type:None
apply(x)[source]

Apply the RNN block to the given input.

Parameters:x (Tensor) – 3-dim time-major sequences of features [max_time, batch, features]
Return type:Tensor
Returns:output 3-dim tensor [max_time, batch, transformed_features]