High precision neural network engineering method based on look-up table computation
A neural network and high-precision technology, applied in the field of neural network engineering, can solve problems such as loss of network precision, achieve fast computing speed, high computing precision, and reduce precision loss
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0025] The present invention will be further described below in conjunction with the accompanying drawings of the specification, but the present invention is not limited to the following embodiments.
[0026] Such as Figure 1-2 Shown is a specific embodiment of a high-precision neural network engineering method based on table look-up calculation. This embodiment is a high-precision neural network engineering method based on table look-up calculation. The method includes the following steps:
[0027] S1: Based on the original floating-point network model, select part of the sample data as the neural network input to the test set, calculate the neural network from the input layer to the hidden layer and then to the output layer forward, and use the accuracy obtained by the output layer as the benchmark Accuracy, the accuracy loss threshold A is set according to actual needs, and the sparsity rate B is set as the descending search step. The accuracy loss threshold A is preferably 0.0...
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