Sample generation method, classification model training method, recognition method and corresponding devices

A classification model and generation device technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problems of high labor and time consumption, low efficiency of classification model training, small data processing, and improved training efficiency. , the effect of saving manpower and time

Pending Publication Date: 2020-07-31
度小满科技(北京)有限公司
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Therefore, the training method of the existing classification model needs to consume a lot of manpower and time, resulting in low training efficiency of the classification model

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
  • Sample generation method, classification model training method, recognition method and corresponding devices
  • Sample generation method, classification model training method, recognition method and corresponding devices
  • Sample generation method, classification model training method, recognition method and corresponding devices

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050]When training the classification model, in the face of the problem that manual training of the classification model requires a lot of manpower and time, when performing data feature screening, some people have proposed a method of relying on a single feature of the model itself for data feature screening, for example, The VAR (Variance, variance threshold) method specifically includes: counting the variance of the samples, and removing the samples whose variance exceeds the VAR, so as to obtain samples for training the classification model. Since many samples do not necessarily conform to the normal distribution, the VAR method is likely to affect the quality of the samples, resulting in inaccurate classification models trained. For another example, the SVM (Support Vector Machine, feature vector machine) method is specifically: based on the spatial distance, the space is divided into two intervals, and the segmentation position is used as the benchmark to screen samples ...

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 sample generation method, a classification model training method, a recognition method and corresponding devices. The method comprises the steps: during the generation of samples of a classification model, carrying out the preliminarily screening of the data features in a preset training set based on feature indexes and feature information values to obtain a plurality ofbasic features; reducing the data processing amount during classification model training; then, based on the feature contribution degree of each basic feature, screening out the basic features meetinga preset feature contribution degree condition from the plurality of basic features to serve as target features in the target sample so as to further reduce the data processing amount of classification model training, and generating a classification model based on the target features and the target data corresponding to the target features in the preset training set and the preset verification set. Therefore, automatic screening of the data is realized, and the data processing amount is small, so the manpower and time are greatly saved, and the training efficiency of the classification modelis improved.

Description

technical field [0001] The present invention relates to the technical field of machine learning, and more specifically, relates to a sample generation method, a classification model training method, a recognition method and a corresponding device. Background technique [0002] When training the classification model in the prior art, classification model algorithm engineers are required to manually segment the acquired original data set to obtain a training set and a verification set; then manually filter model features and adjust model parameters on the training set. [0003] Therefore, the training method of the existing classification model needs to consume a lot of manpower and time, resulting in low training efficiency of the classification model. Contents of the invention [0004] In view of this, the present invention discloses a sample generation method, a classification model training method, a recognition method and a corresponding device, so as to obtain a plural...

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 Applications(China)
IPC IPC(8): G06K9/62
CPCG06F18/24G06F18/214
Inventor 郭灿徐庶
Owner 度小满科技(北京)有限公司
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