Mean and SD obtained from a Trained Model - Classification Learner Toolbox

Hello everyone!

I just want to know the meaning of the matrices mu and sigma that are obtained when we export a trained model from Classification Learner. For example, I got 12 features trained and both matrices of mean and standard deviation came out with 12 elements. If I need to know the mean and SD because I used k-fold, how to proceed?

Thank you all in advance.

Link of refference: k-nearest neighbor classification - MATLAB (mathworks.com)

Answer 1

[UDAYA PEDDIRAJU]

Hi Gustavo,

The matrices "mu" and "sigma" represent the mean and standard deviation of each feature used for normalization during training. For k-fold, calculate these for each fold separately or use overall dataset values.