Convolutional neural network channel self-selection compression and acceleration method based on knowledge migration
A convolutional neural network and neural network technology, applied in the field of convolutional neural network channel self-selection compression and acceleration, can solve the problem of capacity reduction and achieve high compression ratio and acceleration ratio
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0028] The present invention will be further described in detail below with reference to the drawings and embodiments, but the implementation of the present invention is not limited to this.
[0029] Such as figure 1 As shown, the method for self-selected compression and acceleration of convolutional neural network channels based on knowledge transfer provided by this embodiment includes the following steps:
[0030] S1. Extract the migration guidance knowledge from the trained complex convolutional neural network CN1 with good enough performance.
[0031] The migration guidance knowledge can be extracted from different locations in the network. In this embodiment, the ResNet56 network is used as the complex convolutional neural network CN1, such as figure 2 As shown, the residual module in the figure includes two convolutional layers of 3*3 kernel size, and in the first residual module of each stage, the step size of the first convolutional layer is 2, use For dimensionality reduct...
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.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap