Deep learning processor and electronic equipment

A deep learning and processor technology, applied in the field of deep learning, can solve problems such as high bandwidth requirements, poor computing performance, and inability to apply to changing application scenarios, reducing bandwidth requirements and improving efficiency.

Inactive Publication Date: 2019-06-11
SUZHOU LANGCHAO INTELLIGENT TECH CO LTD
View PDF3 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The Cambrian diannao series chips of related technologies and Google's TPU are deep learning processing chips that are widely used, but the above chips implement convolution calculations based o

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
  • Deep learning processor and electronic equipment
  • Deep learning processor and electronic equipment
  • Deep learning processor and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments It is a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

[0033] see below figure 1 , figure 1 It is a schematic structural diagram of a deep learning processor provided in the embodiment of the present application. The deep learning processor may specifically include:

[0034] The scheduling module 100 is configured to receive a data processing request, and analyze macro instructions in the target instruc...

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 deep learning processor, and the processor comprises a scheduling module which is used for receiving a data processing request, and analyzing a macro instruction in a targetinstruction set according to the data processing request so as to obtain an analysis result; The convolution calculation module is used for executing a two-dimensional convolution calculation operation corresponding to the analysis result in a three-dimensional convolution pulse array to obtain a data processing result; Wherein the three-dimensional convolution pulse array is obtained by arranginga plurality of operation units according to a multiplication and addition device array construction rule. According to the method, the requirement of convolution calculation for bandwidth can be reduced, the efficiency of convolution calculation is improved, and the processing flow of deep learning is optimized. The invention further discloses the electronic equipment which has the above beneficial effects.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to a deep learning processor and an electronic device. Background technique [0002] Convolutional neural networks play an important role in deep learning and are widely used in image classification, object recognition, object tracking, etc. With the increase in the amount of data in data centers and the increase in the real-time detection requirements of mobile terminals, performance and power consumption have become key issues for deep learning. [0003] The Cambrian diannao series chips of related technologies and Google’s TPU are deep learning processing chips that are widely used, but the above chips implement convolution calculations based on matrix multiplication or two-dimensional systolic arrays, and the above related technologies require more bandwidth and computing performance. Poor, unable to apply to changing application scenarios. [0004] Therefore, ho...

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): G06N3/063
Inventor 张新赵雅倩方兴董刚杨宏斌范宝余刘栩辰
Owner SUZHOU LANGCHAO INTELLIGENT 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