Shuffle

Intel® oneAPI Deep Neural Network Developer Guide and Reference

Download PDF

ID 768875

Date 2/28/2024

Version

Public

A newer version of this document is available. Customers should click here to go to the newest version.

Visible to Intel only — GUID: GUID-1648AD19-BCE3-4CC9-B317-8DF0C6EB68EE

View Details

Shuffle

API Reference

General

The shuffle primitive shuffles data along the shuffle axis (here designated as ) with group parameter . If the shuffle axis is thought of as a matrix in row-major order, then the shuffle operation transposes the shuffle axis to a matrix in row-major order.

Forward

The formal definition is as follows (variable names follow the standard Naming Conventions):

where

dimension is called a shuffle axis,
is a group_size,
is the outermost indices (to the left from shuffle axis),
is the innermost indices (to the right from shuffle axis), and
and relate to each other as define by the system:

Here, and .

Difference Between Forward Training and Forward Inference

There is no difference between the dnnl_forward_training and dnnl_forward_inference propagation kinds.

Backward

The backward propagation computes , based on .

Essentially, backward propagation is the same as forward propagation with replaced by .

Execution Arguments

When executed, the inputs and outputs should be mapped to an execution argument index as specified by the following table.

Primitive input/output	Execution argument index
	DNNL_ARG_SRC
	DNNL_ARG_DST
	DNNL_ARG_DIFF_SRC
	DNNL_ARG_DIFF_DST

Data Types

The shuffle primitive supports the following combinations of data types:

Propagation	Source / Destination
forward / backward	f32, bf16, f16
forward	s32, s8, u8

WARNING:

There might be hardware and/or implementation specific restrictions. Check the Implementation Limitations section below.

Data Layouts

The shuffle primitive works with arbitrary data tensors. There is no special meaning associated with any logical dimensions. However, the shuffle axis is typically referred to as channels (hence in formulas we use ).

Shuffle operation typically appear in CNN topologies. Hence, in the library the shuffle primitive is optimized for the corresponding memory formats:

Spatial	Logical tensor	Shuffle Axis	Implementations optimized for memory formats
2D	NCHW	1 (C)	dnnl_nchw ( dnnl_abcd ), dnnl_nhwc ( dnnl_acdb ), optimized^
3D	NCDHW	1 (C)	dnnl_ncdhw ( dnnl_abcde ), dnnl_ndhwc ( dnnl_acdeb ), optimized^

Here optimized^ means the format that comes out of any preceding compute-intensive primitive.

Post-Ops and Attributes

The shuffle primitive does not support any post-ops or attributes.

Implementation Limitations

Refer to Data Types for limitations related to data types support.
GPU
- Only tensors of 6 or fewer dimensions are supported.

Performance Tips

N/A

Example

Shuffle Primitive Example

This C++ API example demonstrates how to create and execute a Shuffle primitive.

Key optimizations included in this example:

Shuffle along axis 1 (channels).

Select Your Language

Using Intel.com Search

Quick Links

Recent Searches

Advanced Search

Only search in

Intel® oneAPI Deep Neural Network Developer Guide and Reference

Shuffle

General

Execution Arguments

Data Types

Data Layouts

Implementation Limitations

Performance Tips

Example