Model training method and system, nonvolatile storage medium and computer terminal

Abstract

The invention discloses a model training method and system, a nonvolatile storage medium and a computer terminal. The method comprises the steps that each GPU in a plurality of GPUs is provided with a first feature extraction network and a first full connection layer, and the network structures of the first feature extraction networks in the plurality of GPUs are the same; the first GPU in the plurality of GPUs is used for initializing a first feature extraction network of the first GPU and extracting a first sample feature in the target training data set by adopting the initialized first feature extraction network; inputting the first sample feature into a first full connection layer of the first GPU for processing; determining a prediction error of the first GPU based on the processing result; determining a target prediction error of a target neural network model based on the prediction error of the first GPU and the received prediction errors of other GPUs; and updating parameters of the target neural network model based on the target prediction error.

Description

Technical field [0001] The application relates to the field of machine learning, in particular to a model training method and system, a nonvolatile storage medium and a computer terminal. technical background [0002] The purpose of unsupervised learning or self-supervised learning is to learn models or features with strong expressive ability through unlabeled data. Usually, it is necessary to define a pretext task to guide the training of the model. Agents include, but are not limited to, change prediction, image completion, spatial or temporal sequence prediction, clustering, data generation, etc. [0003] Among them, in the field of unsupervised learning, the method of instance classification regards each data sample in the data set as a class, and it can adopt the same training network as supervised classification, which can make full use of all negative examples in the data set, so it is a more potential scheme. [0004] However, in order to realize the case classification m...

AI Technical Summary

Problems solved by technology

[0006] The embodiment of the present application provides a model training method and system, a non-volatile storage medium, and a computer terminal, to at least solve the problem that the classification model needs to support the same number of classification results through the training process when the sample size in the data set is relatively large. , resulting in the technical problem of large consumption of GPU computing resources

Images