Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020134532 - DEEP MODEL TRAINING METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM

Publication Number WO/2020/134532
Publication Date 02.07.2020
International Application No. PCT/CN2019/114493
International Filing Date 30.10.2019
IPC
G06N 3/08 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
CPC
G06K 9/62
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
G06N 3/08
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
Applicants
  • 北京市商汤科技开发有限公司 BEIJING SENSETIME TECHNOLOGY DEVELOPMENT CO., LTD. [CN]/[CN]
Inventors
  • 李嘉辉 LI, Jiahui
Agents
  • 北京派特恩知识产权代理有限公司 CHINA PAT INTELLECTUAL PROPERTY OFFICE
Priority Data
201811646430.529.12.2018CN
Publication Language Chinese (ZH)
Filing Language Chinese (ZH)
Designated States
Title
(EN) DEEP MODEL TRAINING METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM
(FR) PROCÉDÉ ET APPAREIL D'APPRENTISSAGE DE MODÈLE PROFOND, DISPOSITIF ÉLECTRONIQUE, ET SUPPORT DE STOCKAGE
(ZH) 深度模型训练方法及装置、电子设备及存储介质
Abstract
(EN)
A deep model training method and apparatus, an electronic device, and a storage medium, the deep model training method comprising: acquiring n+1th annotation information outputted by a model to be trained, the model to be trained having undergone n rounds of training (S110); on the basis of training data and the n+1th annotation information, generating an n+1th training sample (S120); and carrying out an n+1th round of training on the model to be trained by using the n+1th training sample (S130).
(FR)
L'invention porte sur un procédé et un appareil d'apprentissage de modèle profond, un dispositif électronique et un support de stockage, le procédé d'apprentissage de modèle profond comprenant : l'acquisition de n+1èmes informations d'annotation fournies par un modèle à entraîner, le modèle à entraîner ayant subi n cycles d'apprentissage (S110) ; sur la base des données d'apprentissage et des n+1èmes informations d'annotation, la génération d'un n+1ème échantillon d'apprentissage (S120) ; et la réalisation d'un n+1ème cycle d'apprentissage sur le modèle à entraîner en utilisant le n+1ème échantillon d'apprentissage (S130).
(ZH)
一种深度模型训练方法及装置、电子设备及存储介质,所述深度模型训练方法包括:获取待训练模型输出的第n+1标注信息,所述待训练模型已经过n轮训练(S110);基于所述训练数据及所述第n+1轮标注信息生成第n+1训练样本(S120);将所述第n+1训练样本对所述待训练模型进行第n+1轮训练(S130)。
Latest bibliographic data on file with the International Bureau