Model training method and device and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and provides a model training method and device and a storage medium, and the model training method comprises the steps: obtaining a to-be-trained routing model; obtaining multiple pieces of sample transaction data and multiple pieces of attribute data; obtaining at least two target conditions; inputting the plurality of pieces of sample transaction data and the plurality of pieces of attribute data into a to-be-trained routing model to obtain at least two target statistical results; determining the number of statistical results meeting corresponding target conditions in the at least two target statistical results, and if the number meets a model output condition, determining the to-be-trained routing model as an alternative routing model; and if the number does not meet a model output condition, adjusting parameters of the to-be-trained routing model, determining the routing model after parameter adjustment as the to-be-trained routing model, and outputting one or more alternative routing models until at least two statistical results meet corresponding target conditions. By implementing the application, the payment channel selection accuracy and rationality can be improved.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, and in particular to a model training method, device and storage medium. Background technique [0002] In the payment system of the e-commerce platform, a very important module is the payment routing. In the payment and settlement link, from multiple payment channels that can be selected, the only payment channel is found according to the transaction dimension. But it is very difficult to choose a suitable payment channel in the e-commerce platform. At present, when training the model used for payment channel selection, the model training is carried out with reference to the single attribute data of the payment channel, such as referring to the minimum cost as the training goal until the model converges. This training method is relatively single due to the single reference factor , resulting in low accuracy of the trained model when selecting payment channels. Content...

AI Technical Summary

Problems solved by technology

But it is very difficult to choose a suitable payment channel in the e-commerce platform

At present, when training the model used for payment channel selection, the model training is carried out with reference to the single attribute data of the payment channel, such as referring to the minimum cost as the training goal until the model converges. This training method is relatively single due to the single reference factor , resulting in low accuracy of the trained model when selecting payment channels

Images