Classification Learner App: creating a table from the parallel coordinate plots

2 views (last 30 days)
Hi,
I used the Classification Learner App to build a Quadratic SVM model for the prediction of a medical outcome based on patient characteristics.
I explored the results including the confusion matrix, ROC curve and parallel coordinate plots.
Is it possible to export the data from the parallel coordinate plots in the form of a table?
In other words, I want to know what the model predicted for each patient (row of data) compared to standard reference.
Thanks a lot!!

Answers (1)

Shashank Gupta
Shashank Gupta on 4 Feb 2021
Everything which is done on classificationLearner app can be regenerated on command line. You just need to export the trained model to MATLAB workspace and find the appropriate data. Check out the trainedModel (exported model) properties. All the data which is used to construct parallel coordinates plot will be mentioned there. you can use link to plot there once you get the data. Generally the things you are looking for are accessible in trainedModel.ClassificationSVM.
I hope this helps
Cheers.

Community Treasure Hunt

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

Start Hunting!