An optimization method for deep learning of edge computing device
An edge computing and deep learning technology, applied in the optimization field of deep learning, can solve the problems of impracticality, large engineering volume and cost, and high cost, and achieve the effect of reducing system energy consumption, optimizing system energy efficiency, and improving energy consumption ratio.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment
[0040] The deep learning optimization method in the embodiment of the present invention is mainly aimed at edge computing devices based on general embedded systems. It is mainly based on CPU and GPU as the computing core, but the present invention is not limited to this. The method is suitable for all calculations. Platform deployment of deep learning applications has good results.
[0041] A deep learning optimization method for edge computing devices based on general embedded systems, starting from two aspects of the system layer and application layer, at the system layer through DVFS for adaptive dynamic frequency modulation of computing chips such as CPU and GPU, without affecting On the premise of computing performance, try to reduce the system energy consumption and increase the energy consumption ratio; at the application layer, the amount of calculation of the deep neural network model is reduced by means of model lightweight, layer fusion, and branch reduction, so that th...
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