Efficient data layouts for convolutional neural networks
US10489680B2 · kind B2 · utility
30Cited by
20References
20Claims
0Family size
Assignee
Inventors
Key dates
| Filing date | Oct 3, 2017 |
| Grant date | Nov 26, 2019 |
| Priority date | — |
| Expiry date | May 17, 2038 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06V10/449
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Systems and methods for efficient implementation of a convolutional layer of a convolutional neural network are disclosed. In one aspect, weight values of kernels in a kernel stack of a convolutional layer can be reordered into a tile layout with tiles of runnels. Pixel values of input activation maps of the convolutional layer can be reordered into an interleaved layout comprising a plurality of clusters of input activation map pixels. The output activation maps can be determined using the clusters of the input activation map pixels and kernels tile by tile.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.