A Reconfigurable CNN High Concurrency Convolution Accelerator
An accelerator and convolution technology, applied in neural architecture, biological neural network models, etc., can solve problems such as complex configuration, low efficiency, and high power consumption, and achieve the effects of reducing occupied resources, simplifying control parts, and improving parallelism
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0031] The present invention will be described in detail below in conjunction with the accompanying drawings and specific implementation cases.
[0032] Such as figure 1 , the reconfigurable convolution accelerator is mainly composed of a main controller, a feature map address generation unit, a weight address generation unit, a reconfigurable computing unit, a result address generation unit and a storage exchange unit. The main controller is responsible for receiving operation configuration information, including feature map size, number of feature map channels, convolution kernel size, convolution kernel channel number, output result size, output result channel number, convolution stride and convolution mode, receiving start signal to start each sub-module, internally calculate the index value of each convolution operation cycle according to the configuration information, according to the index value and convolution mode, the feature map address generation unit and the weigh...
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