Model quality evaluation method and device

Abstract

The invention discloses a model quality evaluation method and device. The method includes: obtaining a prediction model and a plurality of test data subsets, wherein each test data subset comprises atleast one piece of test data marked with a real value; sending a prediction request carrying the prediction model and the test data subset to a corresponding prediction terminal in the terminal cluster; when the prediction terminal in the terminal cluster completes prediction, obtaining a prediction value of the test data from the prediction terminal, and obtaining a prediction value of the testdata in each test data subset; classifying the test data according to the predicted value and the true value of the test data to obtain the category to which the test data in the plurality of test data subsets belongs; and calculating a model quality parameter of the prediction model according to the category to which the test data in the plurality of test data subsets belongs. According to the method, prediction is carried out in a clustering mode, and the calculation time of the prediction value is shortened, so that the model quality evaluation efficiency can be improved.

Description

technical field [0001] The present application relates to the field of computers, in particular to a method and device for evaluating model quality. Background technique [0002] Supervised learning refers to using a set of training data to learn the mapping relationship between the data input and output, and then applying this mapping relationship to unknown data to achieve the purpose of classifying or regressing the position data. In the field of prediction, the mapping relationship between the training data and the real value of the training data that the machine can learn through supervised learning can be called a prediction model. The quality of a predictive model can affect the accuracy of the model's output prediction results. Therefore, before a predictive model is put into use, it is necessary to pre-test the quality of the model, that is, how the model performs on new test data to determine the model's performance. Whether it can be put into use, it is clear tha...

AI Technical Summary

Problems solved by technology

The quality of a predictive model can affect the accuracy of the model's output prediction results. Therefore, before a predictive model is put into use, it is necessary to pre-test the quality of the model, that is, how the model performs on new test data to determine the model's performance. Whether it can be put into use, it is clear that the existing technology has the following problems: the method of predictive model quality assessment is inefficient

Images