Sample acquisition method and device applied to model training, equipment and storage medium

Abstract

The embodiment of the invention discloses a sample acquisition method and a device applied to model training. The method comprises the following steps: randomly sampling a data source to obtain a training sample set; training the model according to the training sample set to obtain a prediction error rate of the training sample set; re-determining the sampling rate of the data source according tothe prediction error rate, and sampling the data source according to the re-determined sampling rate to obtain an updated training sample set; iteratively executing the steps of training the model according to the updated training sample set, re-determining the sampling rate of the data source according to the obtained prediction error rate, and obtaining the updated training sample set accordingto the re-determined sampling rate; and obtaining the updated training sample set as a target sample set. According to the method of the embodiment of the invention, the target sample set close to thefeature distribution of the prediction sample set can be selected from the data source, and the target sample set is used for training the model.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular, to a sample acquisition method, device, device, and computer-readable storage medium applied to model training. Background technique [0002] With the rapid development of Internet information technology, there are more and more Internet applications. In order to broaden the audience of Internet applications and make Internet applications bring higher profits to enterprises, it is necessary to promote Internet applications to more users. [0003] In the early stage of Internet application promotion, such as game applications, in the absence of initial players, the feedback data of the first batch of players obtained through game promotion is required as a sample for promotion modeling, or the player data of similar game applications is used as a sample. samples for generalization modeling. Since promotional modeling of game applications often requires a large amo...

AI Technical Summary

Problems solved by technology

Since the promotion modeling of game applications often requires a large amount of player data, the effect of promotion modeling through the former method is better, but the sample acquisition cost for promotion modeling is relatively high

Although generalization modeling in the latter way can easily obtain a large number of samples from similar types of game applications, since the feature distribution of samples in these game applications is not exactly the same as that in current game applications, the generalization model It is easy to produce deviations in actual forecasting, resulting in poor forecasting results

Images