Classification model evaluation index determination method and device, equipment and medium

Abstract

The invention discloses a classification model evaluation index determination method and device, equipment and a medium, and relates to the technical field of deep learning. The classification model evaluation index determination method comprises the steps of obtaining a current sample set of a classification model; taking the current sample set and a fixed number of historical sample sets continuously acquired with the current sample set as a current sliding window; and determining a current evaluation index value of the classification model according to the current sliding window. By means of the classification model evaluation index determination method, the accuracy and credibility of the model evaluation result can be improved, and then the stability of model training is improved.

Description

technical field [0001] The embodiments of the present application relate to computer technology, in particular to the field of deep learning technology, and in particular to a method, device, device and medium for determining an evaluation index of a classification model. Background technique [0002] Classification task is the most typical and widely used task in machine learning tasks. And binary classification, that is, tasks with only two categories of positive samples and negative samples, is the most common scenario in classification tasks. Based on this, how to evaluate the results of the binary classification task is particularly important. Commonly used evaluation indicators for binary classification tasks include Precision, Recall, F1-score, AUC, etc. Among them, AUC (Area Under Curve, area under the curve) is the most commonly used and most important evaluation index. [0003] Specifically, AUC is defined as the area enclosed by the ROC curve and the coordinate...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, after using dynamically collected samples to train the binary classification model, when evaluating the training results, either only use the predicted value results of the newly collected samples to calculate the AUC value of the classification model, or use the predicted values ​​of all samples As a result, the AUC value of the classification model is recalculated, which will lead to inaccurate and low credibility of the final calculated model evaluation results.

Images