Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and device for testing operator precision in neural network and computer readable storage medium

A neural network and operator technology, applied in the field of artificial intelligence, can solve problems such as no testing methods or devices, and achieve the effect of flexible evaluation methods

Pending Publication Date: 2021-06-25
CAMBRICON TECH CO LTD
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is no testing method or device for this aspect in the prior art

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
  • Method and device for testing operator precision in neural network and computer readable storage medium
  • Method and device for testing operator precision in neural network and computer readable storage medium
  • Method and device for testing operator precision in neural network and computer readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] The technical solution disclosed in the present disclosure provides a method, device and computer-readable storage medium for testing the accuracy of an operator in a neural network. Specifically, the present disclosure proposes to use the difference between the benchmark result obtained for the operator to be tested and the test result to evaluate the accuracy level of the operator, so as to reflect the fault tolerance of the operator to a certain extent. In the context of this disclosure, the evaluation of operator accuracy mentioned here may not only include the evaluation of the accuracy brought by the operation of the operator itself on different hardware platforms, but also include the evaluation of the accuracy of the operator due to the conversion of data types. The precision brought by the same or different hardware platforms, or the precision brought by the evaluation operator based on the combination of the hardware platform and data type conversion.

[0024]...

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

The invention discloses a method for testing operator precision in a neural network, a testing device and a storage medium, the testing device can be included in a combined processing device, and the combined processing device can further comprise a universal interconnection interface and other processing devices. And the testing device interacts with other processing devices to jointly complete calculation operation specified by a user. The combined processing device can further comprise a storage device, and the storage device is connected with the equipment and the other processing devices and used for storing data of the equipment and the other processing devices. According to the scheme disclosed by the invention, the operator precision of the neural network can be effectively evaluated, so that the overall network efficiency of the neural network is determined.

Description

technical field [0001] This disclosure relates generally to the field of artificial intelligence. More specifically, it relates to a method, device and storage medium for testing the accuracy of operators in a neural network. Background technique [0002] In recent years, thanks to the increase in the amount of data, the enhancement of computing power, the maturity of learning algorithms, and the enrichment of application scenarios, artificial intelligence technologies represented by machine learning and knowledge graphs have gradually become more popular. Especially in recent years, more and more people have begun to pay attention to deep learning with neural network as the main model. Deep learning can not only be used to solve representation learning problems in machine learning, but also is increasingly used to solve some general artificial intelligence problems, such as reasoning or decision-making, due to its powerful capabilities. The deep learning framework is 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
IPC IPC(8): G06N3/02G06F17/15
CPCG06N3/02G06F17/15
Inventor 不公告发明人
Owner CAMBRICON TECH CO LTD