Is changing the 'KernelScale' in the fitcsvm equivalent to changing the sigma value?
10 views (last 30 days)
Show older comments
Hello,
I am using this command:
cl3 = fitcsvm(X,Y,'KernelFunction','rbf', 'Standardize',true,'BoxConstraint',10,'ClassNames',[-1,1]);
to run the Gaussian Radial Bias kernel. I understand that the default value for sigma = 1. I was told that in order to change the sigma value, use this command:
cl3 = fitcsvm(X, Y, 'KernelFunction', 'rbf', 'KernelScale', 0.2, 'Standardize', true, 'BoxConstraint', 100, 'ClassNames', [-1,1]);
In this case, the sigma value can now be changed by modifying the 'KernelScale'. Is this a correct statement?
3 Comments
Aniket Borkar
on 16 Dec 2016
But where can we set the C and sigma value? Box constraint =100 and kernel scale =0.2. How does these settings can vary C and sigma?
Akshar Agarwal
on 23 Jun 2017
the C value is the box constraint. It is called box constraint because it sets bounds on the lagrange multipliers in the actual function.
Answers (0)
See Also
Categories
Find more on Statistics and Machine Learning Toolbox in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!