Pure integer quantization method for lightweight neural network (LNN)
US11934954B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Sep 22, 2021 |
| Grant date | Mar 19, 2024 |
| Priority date | — |
| Expiry date | Sep 22, 2041 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N3/08
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A pure integer quantization method for a lightweight neural network (LNN) is provided. The method includes the following steps: acquiring a maximum value of each pixel in each of the channels of the feature map of a current layer; dividing a value of each pixel in each of the channels of the feature map by a t-th power of the maximum value, t∈[0,1]; multiplying a weight in each of the channels by the maximum value of each pixel in each of the channels of the corresponding feature map; and convolving the processed feature map with the processed weight to acquire the feature map of a next layer. The algorithm is verified on SkyNet and MobileNet respectively, and lossless INT8 quantization on SkyNet and maximum quantization accuracy so far on MobileNetv2 are achieved.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.