Different RMSE in Regression Learner App everytime I start a new Session

Everytime I start a new Session in the Regression Learner App I get different RMSEs (Validation) although I am using the same model and the same dataset, features, fold numbers for k-fold. Could this occur due to k-fold partitions getting randomly initialized every time I start a new session?

Answer 1

[Mohammad Sami]

The partitions would be randomly initiated. You can set a random number seed before you begin, to try and get the same results.

https://www.mathworks.com/help/matlab/math/generate-random-numbers-that-are-repeatable.html

[Matthias Scherer]

And how can I use the random number seed for the regression learner app? Since I need the regression learner app to choose the best model for my data, but setting a number seed would be something I do in command window and I don`t know how to use this to affect the partitions of the k-fold cross validation in the regression learner app.