Apparatus and system of compression/decompression, chip, and electronic device

A compression device and decompression technology, which is applied in the field of artificial neural networks, can solve the problems of very high memory bandwidth requirements and large storage space

Active Publication Date: 2018-07-10
SHANGHAI CAMBRICON INFORMATION TECH CO LTD
View PDF1 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If it is not compressed, not only a large amount of storage space is requ

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
  • Apparatus and system of compression/decompression, chip, and electronic device
  • Apparatus and system of compression/decompression, chip, and electronic device
  • Apparatus and system of compression/decompression, chip, and electronic device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0098]Video coding and decoding technology is a very mature technology. Traditional video coding and decoding technologies use prediction, transformation, and entropy coding technologies. After the rise of deep learning, video coding and decoding using deep neural networks has become a new research hotspot.

[0099] In a broad sense, neural network data refers to the collection of neural network information, including neural network numerical data and neural network structure information.

[0100] The neural network numerical data includes: weight data and bias data of neurons, which are actually numerical data. After careful research and comparison, the applicant found that the numerical data of the neural network has the same local correlation as the pixels of the video image, so using the method of video encoding and decoding to encode and decode the neural network model, and then compressing the neural network model, will be A feasible technical route.

[0101] In additio...

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 an apparatus and system of compression/decompression of neural network data, a chip, and an electronic device. The apparatus of compression of neural network data includes a model conversion module and a data coding module which is connected with the model conversion module. The apparatus of decompression of neural network data includes a data decoding module and a model conversion module which is connected with the data decoding module. The system of compression of neural network data includes the above compression apparatus and decompression apparatus. The apparatus and system of compression/decompression of neural network data can compress/decompress the neural network data, can achieve relatively higher compression ratio, and can greatly reduce the storage spaceand transmission pressure of the neural network model.

Description

technical field [0001] The present disclosure relates to the technical field of artificial neural networks, and in particular to a device and system for compressing / decompressing neural network data, a chip, and an electronic device. Background technique [0002] Artificial Neural Networks (ANNs), referred to as Neural Networks (NNs), is an algorithmic mathematical model that imitates the behavioral characteristics of animal neural networks and performs distributed parallel information processing. This kind of network depends on the complexity of the system, and achieves the purpose of processing information by adjusting the interconnection relationship between a large number of internal nodes. [0003] At present, neural networks have made great progress in many fields such as intelligent control and machine learning. With the rise of deep learning, neural networks have once again become a hot topic in the field of artificial intelligence research. With the extensive comb...

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
IPC IPC(8): H04N19/182H04N19/42H04N19/50H04N19/61H04N19/625H04N19/124H04N19/13H04N19/91G06N3/04G06N3/08
CPCH04N19/124H04N19/13H04N19/182H04N19/42H04N19/50H04N19/61H04N19/625H04N19/91G06N3/08G06N3/045G06N3/063G06N3/048H04N19/172H04N19/60G06N3/04
Inventor 不公告发明人
Owner SHANGHAI CAMBRICON INFORMATION TECH CO LTD
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