This class describes a recurrent (RNN-like) sequence operation. More...

#include <recurrent.hpp>

Public Member Functions
	RecurrentSequence (OpInputMap &args, ngraph::op::RecurrentSequenceDirection direction)
	Constructs a RecurrentSequence class object. More...

OutputVector	run_sequence (const RecurrentCellFunction &kernel)
	Carry out all steps of recurrent sequence with provided cell kernel. More...

Detailed Description

This class describes a recurrent (RNN-like) sequence operation.

. This class takes care of orchestration of computations carried

out on data sequence. Use have to only provide kernel function which would be executed on current time-step input data and the sequence direction mode.

. This class assumes the RNN-like sequence operation. This

means that the operator should have inputs and outputs the same as RNN operator. Especially the cell/kernel should have input related to hidden cell state.

Constructor & Destructor Documentation

◆ RecurrentSequence()

ngraph::onnx_import::recurrent::RecurrentSequence::RecurrentSequence	(	OpInputMap &	args,
		ngraph::op::RecurrentSequenceDirection	direction
	)

Constructs a RecurrentSequence class object.

Parameters

[in]	args	The map with recurrent sequence operator inputs.
[in]	attrs	The structure containing operator attributes.
[in]	direction	The sequence direction mode {FORWARD, REVERSE, BIDIRECTIONAL}.

Member Function Documentation

◆ run_sequence()

OutputVector ngraph::onnx_import::recurrent::RecurrentSequence::run_sequence ( const RecurrentCellFunction & kernel )

Carry out all steps of recurrent sequence with provided cell kernel.

Parameters

[in] kernel The cell kernel function.

Returns: The node vector containing results from all sequence steps.

The documentation for this class was generated from the following file:

frontend/onnx_import/include/onnx_import/utils/recurrent.hpp