Machine learning training method, device, server and medium

A machine learning and training method technology, applied in the computer field, can solve the problems of low efficiency and high cost of machine learning training, and achieve the effect of reducing labor costs and improving efficiency

Active Publication Date: 2021-03-09
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF16 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] It can be seen that manual labeling of training data not only leads to low efficiency of machine learning training but also high cost.

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
  • Machine learning training method, device, server and medium
  • Machine learning training method, device, server and medium
  • Machine learning training method, device, server and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] figure 1 It is a flow chart of the machine learning training method provided by Embodiment 1 of the present invention. This embodiment is applicable to the situation of machine learning training involved in unmanned driving technology. The method can be executed by a machine learning training device, which can adopt It can be realized by means of software and / or hardware, and can be integrated on a server. Such as figure 1 As shown, the method may include:

[0030] S110. Determine the type and initial state of the identified object, and send a movement instruction to a mobile device carrying the identified object, where the movement instruction includes a movement speed and a movement direction.

[0031] The identified objects in this embodiment include all types of obstacles that may be encountered during vehicle driving, for example, various automobiles, bicycles, motorcycles, pedestrians (in this embodiment, deployed dummy is used instead), trees, Traffic signs, c...

Embodiment 2

[0047] figure 2 It is a flow chart of the machine learning training method provided in Embodiment 2 of the present invention. This embodiment is further optimized on the basis of the above embodiments. Such as figure 2 As shown, the method may include:

[0048] S210. Determine the type and initial state of the identified object, and send a movement instruction to a mobile device carrying the identified object, where the movement instruction is a rotation instruction, including a rotation angular velocity and a rotation direction.

[0049] In this embodiment, the rotation of the recognized object is taken as an example to illustrate the machine learning training method. Specifically, the mobile device includes a controllable mechanical wheel device deployed at the bottom of the identified object, and the controllable mechanical wheel device is fixed on the ground. After receiving the rotation instruction, the mobile device drives the identified object to rotate together. ...

Embodiment 3

[0058] image 3 It is a flow chart of the machine learning training method provided by Embodiment 3 of the present invention, and this embodiment is further optimized on the basis of the foregoing embodiments. Such as image 3 As shown, the method may include:

[0059] S310. Determine the type and initial state of the identified object, and send a movement instruction to the mobile device carrying the identified object, wherein the movement instruction is a traveling instruction, including a traveling trajectory, traveling speed, and traveling direction.

[0060] In this embodiment, the movement of the recognized object is taken as an example to illustrate the machine learning training method. In this embodiment, the mobile device includes a movable carrying device deployed at the bottom of the identified object. After receiving the traveling instruction, the mobile device drives the identified object to travel along the traveling track.

[0061] S320. During the moving pr...

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 embodiment of the present invention discloses a machine learning training method, device, server and medium, wherein the method includes: determining the type and initial state of the recognized object, and sending a movement instruction to the mobile device carrying the recognized object; During the moving process of the identified object, the image data of the identified object is collected by the fixed acquisition device according to the set frequency; according to the initial state, moving speed and moving direction of the identified object, and / or the distance between the acquisition device and the identified object Determine the real-time state of the recognized object corresponding to each frame of image data; use the image data of the recognized object as the input value, and use the type and real-time state of the recognized object as the expected output value to perform machine learning training. The embodiment of the present invention solves the problem of low training efficiency caused by manual labeling of training data in the process of machine learning training, improves the efficiency of machine learning training, and reduces labor costs.

Description

technical field [0001] The embodiments of the present invention relate to the field of computer technology, and in particular to a machine learning training method, device, server and medium. Background technique [0002] With the development of artificial intelligence technology, machine learning has been widely used in the field of intelligent driving. For example, in unmanned driving technology, machine learning can be used to identify obstacles during driving, predict the movement status of obstacles, or provide driving navigation. [0003] However, the training data currently used for machine learning needs to be manually labeled. For example, after collecting the sample data, manually label the status of various obstacles in the sample data based on data processing tools, and then perform machine learning training. Especially when the recognition accuracy of obstacles is high, the marking information of obstacles in each state is required to be more detailed and compr...

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): G06K9/00G06K9/62G06N20/00
CPCG06V20/41G06V20/10G06F18/214
Inventor 胡太群
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) 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