Depth convolution neural network acceleration and compression method based on parameter quantification

A deep convolution and neural network technology, applied in the field of image processing, can solve the problem that the acceleration or compression effect of the network model needs to be studied, and the acceleration and compression of the convolutional neural network are not considered at the same time.

Active Publication Date: 2015-12-23
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF3 Cites 51 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these algorithms do not consider the acceleration and compression of convolutional neural networks at the same time, and are only tested on a few layers of the network, and the acceleration or compression effect of the entire network model remains to be studied

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Depth convolution neural network acceleration and compression method based on parameter quantification
  • Depth convolution neural network acceleration and compression method based on parameter quantification
  • Depth convolution neural network acceleration and compression method based on parameter quantification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0013] The method for accelerating and compressing a deep convolutional neural network based on parameter quantization provided by an embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0014] figure 1 The flow chart of the method for accelerating and compressing the deep convolutional neural network based on parameter quantization provided by the embodiment of the present invention.

[0015] refer to figure 1 , in step S101, quantize the parameters of the deep convolutional neural network to obtain a plurality of sub-codebooks and index values ​​corresponding to the plurality of sub-codebooks.

[0016] In step S102, the feature map of the output of the deep convolutional neural network is obtained according to the plurality of sub-codebooks and index values ​​respectively corresponding to the plurality of sub-codebooks.

[0017] Here, the deep convolutional neural network includes multiple convolutional layers...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a depth convolution neural network acceleration and compression method based on parameter quantification, and the method comprises the steps: carrying out the quantification of parameters of a depth convolution neural network, and obtaining a plurality of sub-codebooks and indexes corresponding to the sub-codebooks; and obtaining an output characteristic graph of the depth convolution neural network according to the plurality of sub-codebooks and indexes corresponding to the sub-codebooks. The method can achieve the acceleration and compression of the depth convolution neural network.

Description

technical field [0001] The invention relates to image processing technology, in particular to an acceleration and compression method of a deep convolutional neural network based on parameter quantization. Background technique [0002] In recent years, deep convolutional neural networks have made great breakthroughs in many fields such as computer vision, speech processing, and machine learning, and have significantly improved the performance of machine algorithms in multiple tasks such as image classification, object detection, and speech recognition. And it has been widely used in Internet, video surveillance and other industries. [0003] The training process of the deep convolutional neural network is based on a large-scale data set containing artificially labeled information to learn and adjust the network parameters. Generally speaking, a large-capacity, high-complexity deep convolutional network can learn data more comprehensively, thereby achieving better performance...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/02
Inventor 程健吴家祥冷聪卢汉清
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap