How to get prediction scores from exported Classification model

15 views (last 30 days)
I have exported a decision tree model from the classification learner, and now I want to test the model with new data. I have used the
predictions = researchquestion2.predictFcn(T)
How can I now get the accuracy, AUC, and confusion matrix figures in similar way to the classification learner app?
Thank you!

Answers (1)

Shashank Gupta
Shashank Gupta on 25 May 2021
The ability to export figures from the Classification Learner App is not available in the Statistics and Machine Learning Toolbox as of now. But there are workaround, you can simply export the data and call out the function explicitly to generate the same plot.
After you exported the model, you can evaluate the model and get the prediction for test data. Confusion matrix can be found out simply by using confusionchart function. Below is the code for the reference.
% Accuracy.
accuracy = sum(YPred == YTest)/numel(YTest)
% Confusion Matrix.
confusionchart(YPred, YTest);
I hope this helps.


Find more on Statistics and Machine Learning Toolbox in Help Center and File Exchange


Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!