An intelligent modeling method and system based on dynamic parameter configuration and process supervision

A technology of dynamic parameters and modeling methods, applied in the field of deep learning to build neural networks, can solve problems such as inability to supervise the model training process in real time, and achieve efficient and convenient construction, unified scheduling and allocation

Active Publication Date: 2022-03-15
THE 28TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Tensorboard visualization relies on the log file output when the tensorflow program is running. You need to wait until the program is finished running to obtain the log file. Currently, this process is separated from the entire model training process, and it is impossible to monitor the model training process in real time.

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
  • An intelligent modeling method and system based on dynamic parameter configuration and process supervision
  • An intelligent modeling method and system based on dynamic parameter configuration and process supervision
  • An intelligent modeling method and system based on dynamic parameter configuration and process supervision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] The scope of the present invention will be further described below in conjunction with the accompanying drawings and specific examples.

[0053] Such as figure 1 As shown, an embodiment of the present invention provides an intelligent modeling method based on dynamic parameter configuration and process supervision, first receiving the modeling task basic information, training algorithm, training data set, and training container resources by interactive interface configuration. Then, according to the training algorithm selected by the user, obtain the associated container mirror, the code script storage path, the mount path, the training data set mount path, the model output path, and the command of the algorithm in the container; if the training algorithm has pre-training The model, also obtains the storage path of the pre-training model and the mount path in the container; then call the method of starting the container task in the container cluster management system API, s...

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 proposes an intelligent modeling method and system based on dynamic parameter configuration and process supervision. For developers, it provides a visual interface to dynamically edit and adjust the parameters of the model construction code script, so that the operation of the code script can be dynamically adapted to the developer's subjective wishes, thereby building and training an intelligent model that meets the needs of the developer; at the same time, through the code The script integrates the framework method of tensorboard, starts a monitoring service, and records the changes of some key data information and parameter indicators during the operation process along with the operation of the code script for model building and training, so as to supervise the process of intelligent modeling tasks and facilitate development Personnel make decisions based on changes in important parameters during model training, and readjust script parameters for secondary training and construction of the model. All of the above cooperate and complement each other, so that developers can efficiently and conveniently construct intelligent models.

Description

Technical field [0001] The invention belongs to depth learning to build a neural network, and in particular, the present invention relates to an intelligent modeling method and system based on dynamic parameter configuration and process supervision. Background technique [0002] Docker is an open source container project based on Go language. It was born in the beginning of 2013. The initiator was DotCloud. There are currently a number of related items (including Docker Swordac, Kubernates, etc.), which gradually forms an ecosystem surrounding the Docker container. Docker's idea is to manage the application's package, distribution, deployment, and run life cycle, to achieve a package, the purpose of running everywhere. The application components here can be either a web application, a compilation environment, or a set of database platform services, or even an operating system or cluster. Based on multiple open source technologies on the Linux platform, Docker provides high effici...

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): G06F9/455G06N3/04G06N3/08
CPCG06F9/45558G06N3/04G06N3/08G06F2009/45575
Inventor 徐伟民崔隽吴姗姗后弘毅郝大鑫
Owner THE 28TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP
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