Adaptive iterative convolution neural network model compression method

A convolutional neural network and self-adaptive iterative technology, applied in the field of convolutional neural network model compression, can solve problems such as inability to model compression algorithms, achieve high accuracy and reduce precision loss

Active Publication Date: 2018-12-14
SOUTH CHINA UNIV OF TECH
View PDF5 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These model compression algorithms can only compress the model to a fixed number of bits, and cannot be used as a general model compression algorithm

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
  • Adaptive iterative convolution neural network model compression method
  • Adaptive iterative convolution neural network model compression method
  • Adaptive iterative convolution neural network model compression method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] The present invention is mainly about the self-adaptive iterative convolutional neural network model compression method, so the implementation of the present invention has certain requirements on the hardware, the implementation example set forth below is on the Ubuntu14.04 platform, the graphics card is NVIDIA TiTan X, 12GB video memory, for The convolutional neural network can be trained normally, so it is recommended that the video memory of the graphics card be at least 6GB. In order to make the features and advantages of the method proposed by the present invention more comprehensible, the following will be described in detail in conjunction with the accompanying drawings and specific implementation examples.

[0045] The adaptive iterative convolutional neural network model compression method of the present invention is as follows: figure 1 As shown, it mainly includes the following steps:

[0046] Step 1: Perform data preprocessing on the training data;

[0047...

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

An adaptive iterative convolution neural network model compression method includes: preprocessing the training data, training the convolution neural network with the training data, selecting the optimal model as the model to be compressed, using the adaptive iterative convolution neural network model compression method to compress the model, evaluating the compressed model, and selecting the optimal model as the compressed model. The invention has the following advantages: self-adaptive adjustment of the quantization ratio and less parameters; adaptive iterative compression which can improve the accuracy of the model after compression. The invention supports common convolution neural network model compression and can compress to a specific number of bits according to requirements, so the method of the invention can efficiently compress the convolution neural network model and apply the model to a mobile device.

Description

technical field [0001] The invention belongs to the research in the field of convolutional neural network model compression, in particular to an adaptive iterative convolutional neural network model compression method. Background technique [0002] Since the AlexNet network won the first place in the ImageNet competition in 2012, convolutional neural networks have been applied to various fields of computer vision. Many researchers are studying how to use convolutional neural networks to solve computer vision problems such as image classification, object detection, image semantic segmentation, and image-to-text conversion. At present, researchers have achieved important results in many fields. However, most of the above convolutional neural network models run on desktops or servers, usually requiring GPU acceleration, and the amount of calculation and model size are large. Therefore, it cannot be applied to devices with weaker performance such as FPGAs, smart phones and emb...

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): G06N3/08
CPCG06N3/082
Inventor 余志文马帅
Owner SOUTH CHINA UNIV OF TECH
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