Model optimization method, equipment and storage medium

Abstract

The embodiment of the invention provides a model optimization method, equipment and a storage medium. In the embodiment of the invention, a plurality of sample data can be marked in advance, so that a plurality of sample sets are obtained, and on the basis, target sample sets meeting preset requirements can be selected from the plurality of sample sets in batches; and a to-be-promoted model is trained based on the selected target sample set. Therefore, in the embodiment, the sample data and the marking information can be synthesized, and the target sample sets can be selected in batches and added into the training set. Therefore, a large batch of target sample sets can be efficiently mined, so that the value of massive backflow data is fully played; a target sample set carrying essential knowledge can be more accurately and comprehensively mined from the backflow data, so that the structure of the training set can be optimized, the quality of the training set can be improved, and the model performance can be continuously improved; in addition, the mode of mining the target sample sets in batches can greatly reduce the number of sample selection times of the to-be-improved model, so that the model optimization efficiency can be effectively improved.

Description

technical field [0001] The present application relates to the technical field of machine learning, and in particular to a model optimization method, device and storage medium. Background technique [0002] The traditional active learning model is usually: A=(C, Q, S, L, U). Where C is a group or a classifier, and L is the marked sample used for training. Q is a query function, which is used to query information with a large amount of information from the unlabeled sample pool U, and S is a supervisor, who can label the samples queried by Q. The model starts learning with a small number of initially labeled samples L, selects the most useful samples through a certain query function Q, and asks the supervisor for labels, and then uses the acquired new knowledge to train the classifier and perform the next round of queries. [0003] However, the model trained in this way has reached the performance bottleneck and cannot meet the increasing performance requirements of the mode...

AI Technical Summary

Problems solved by technology

[0003] However, the model trained in this way has reached the performance bottleneck and cannot meet the increasing model performance requirements

Images