Method, system and device for searching structural parameters of EfficentNet and medium

A technology of structural parameters and search methods, applied in the field of neural networks, can solve the problems that the accuracy does not meet the requirements, consume computing resources, and inefficient parameter search, and achieve the effect of reducing requirements and consumption.

Pending Publication Date: 2020-06-09
SUZHOU LANGCHAO INTELLIGENT TECH CO LTD
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0016] The method of EfficientNet search parameters α, β, γ has two main disadvantages: one is to use grid search to sample, and the other is to train a new network to converge every time the parameters are changed. These two characteristics make parameter search compare inefficient
The advantage of this is that it can guarantee that the selected network has the highest accuracy, but the disadvantage is that each network has to be trained until the accuracy does not increase, which consumes a lot of computing resources.
In many cases, after training for several epochs, it can be seen that the accuracy will not meet the requirements

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, system and device for searching structural parameters of EfficentNet and medium
  • Method, system and device for searching structural parameters of EfficentNet and medium
  • Method, system and device for searching structural parameters of EfficentNet and medium

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0058] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the following describes the embodiments of the present invention in detail in conjunction with specific embodiments and with reference to the accompanying drawings.

[0059] It should be noted that all the expressions using "first" and "second" in the embodiments of the present invention are used to distinguish two entities with the same name but not the same or parameters that are not the same, as shown in "first" and "second" Only for the convenience of presentation, it should not be construed as a limitation to the embodiments of the present invention, and subsequent embodiments will not describe this one by one.

[0060] According to one aspect of the present invention, an embodiment of the present invention proposes a search method for EfficientNet's structural parameters, such as figure 1 As shown, it may include the steps: S1, randomly initializing structural parameters

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 searching structural parameters of EfficentNet. The method comprises the following steps of: randomly initializing the structural parameters in a search space in apreset range; generating an EfficientNet by utilizing the randomly initialized structure parameters; predicting the final precision of the EfficientNet subjected to the first preset number of eph training; in response to the fact that the predicted final precision is smaller than the threshold value, predicting the final precision of the EfficentNet subjected to the second preset number of epochtrainings again, wherein the second preset number is larger than the first preset number; and in response to the fact that the predicted or re-predicted final precision is smaller than the threshold,stopping training, retaining and recording the corresponding structure parameters, and returning to the step of randomly initializing the structure parameters. The invention further discloses a system, computer equipment and a readable storage medium. Random search is used for replacing an original grid search method, and the final precision is predicted in advance, so that the neural network determined by combining all structural parameters does not need to be trained to be converged, and the consumption of computing resources is greatly reduced.

Description

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

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
Owner SUZHOU LANGCHAO INTELLIGENT TECH 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