Compression and acceleration method of structured network model based on multi-level pruning
A network model and structured technology, applied in biological neural network models, neural learning methods, neural architectures, etc., can solve problems such as difficulty in taking into account parameter reduction network acceleration, limited compression effect, inability to obtain compressed storage and reduced operations, etc. The effect of reducing the number of floating-point operations, improving operating efficiency, and reducing hardware dependencies
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0049] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings.
[0050] see figure 1 , the specific implementation steps of the multi-level pruning-based structured network model compression acceleration method proposed by the present invention are as follows:
[0051] S1: Obtain a pre-training model, train the original network model to be processed on the training data set, and obtain a complete network model;
[0052] S2: Based on the pre-training model, measure the sensitivity of the convolutional layer of the original network model, and obtain the sensitivity-pruning rate curve of each convolutional layer through the control variable method;
[0053] S3: Iterative pruning between sensitivity layers, single-layer pruning is performed on the current network model according to the sensitivity order from low...
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