Data fixed-point acceleration method, device, electronic equipment and storage medium

A fixed-point and data-based technology, applied in the direction of machine actuators, electrical digital data processing, program control design, etc., can solve the problems of slow simulation, time-consuming, long iteration cycle, etc.

Active Publication Date: 2021-07-06
SHENZHEN INTELLIFUSION TECHNOLOGIES CO LTD
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Usually, this simulation is very slow, and it takes a lot of time to test the fixed-point algorithm model, which makes the iteration cycle of the fixed-point algorithm of the deep learning neural network algorithm longer

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Data fixed-point acceleration method, device, electronic equipment and storage medium
  • Data fixed-point acceleration method, device, electronic equipment and storage medium
  • Data fixed-point acceleration method, device, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0097] As an optional implementation, the method also includes:

[0098] According to the output data, it is judged whether the test of the fixed-point algorithm model is over;

[0099] If the test of the fixed-point algorithm model is not over, perform fixed-point adjustment on the first model parameters.

[0100] In this optional implementation, after obtaining the output data, it can be judged whether the output data meets the preset requirements of the fixed-point algorithm model, if the output data meets the preset requirements of the fixed-point algorithm model If the requirement is set, it can be determined that the test of the fixed-point algorithm model is over, otherwise, if the output data does not meet the preset requirements of the fixed-point algorithm model, it indicates that the first model parameter of the fixed-point algorithm model is inaccurate , it is also necessary to adjust the first model parameter of the fixed-point algorithm model, therefore, the tes...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

A fixed-point acceleration method for data, the method comprising: acquiring test data and first model parameters; inputting the test data and the first model parameters into a fixed-point algorithm model that needs parameter fixed-point adjustment, Obtain the first eigenvalue; Convert the data format of the first eigenvalue from a fixed-point format to a floating-point format to obtain a second eigenvalue represented by the floating-point format; use a graphics processor GPU acceleration mechanism to Perform fixed-point operations on the second eigenvalue to obtain output data. The invention also provides a fixed-point accelerating device, electronic equipment and computer storage medium. The invention can accelerate the calculation of the fixed-point algorithm model, and accelerate the iteration period of the fixed-point algorithm.

Description

technical field [0001] The present invention relates to the technical field of intelligent terminals, in particular to a data fixed-point acceleration method, device, electronic equipment and storage medium. Background technique [0002] Deep learning neural network algorithms require high computing power and a large amount of storage space, which limits the application of deep learning neural network algorithms on embedded terminal devices. By fixed-pointing the deep learning neural network algorithm, the dependence on storage space can be reduced, so that the deep learning neural network algorithm can be expanded on the terminal device. [0003] Due to the high complexity of the deep learning neural network algorithm, usually only a relatively low quantization bit width can be used to ensure the operating efficiency on the terminal device, which will cause a relatively large loss of precision. In order to reduce the loss of precision, in the process of fixed-pointing the ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06F9/30
CPCG06F9/30007G06F9/30025
Inventor 韦国恒
Owner SHENZHEN INTELLIFUSION TECHNOLOGIES CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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