A kind of convolutional neural network cnn hardware accelerator and acceleration method
A convolutional neural network and hardware accelerator technology, applied in the field of deep learning hardware acceleration, can solve the problems of reduced accelerator efficiency, poor scalability of systolic arrays, and high idle rate of computing units, so as to improve the reuse rate, reduce the amount of data moving, and improve the The effect of computing performance
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0043] The present invention will be further described below in conjunction with the accompanying drawings and specific preferred embodiments, but the protection scope of the present invention is not limited thereby.
[0044] Such as figure 1As shown, the convolutional neural network CNN hardware accelerator in this embodiment includes an input buffer 1 for caching input feature picture data and a plurality of computing units 2 (PE) that share the same input feature picture data for CNN convolution operations, each The computing unit 2 includes a convolution kernel buffer 21, an output buffer 22, and a multiply-add unit 23 composed of multiple MAC components; the CNN hardware accelerator is connected to an external storage component, and the external storage component provides the CNN hardware accelerator with computing data information and result write-back space . The convolution kernel buffer 21 receives the convolution kernel data returned from the external storage unit, ...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com