MATLAB-Kernel-PCA

KernelPca.m is a MATLAB class file that enables you to do the following three things with a very short code.
1.fitting a kernel pca model with training-data with the three kernel functions (gaussian, polynomial, linear) (demo.m)
2.projection of new data with the fitted pca model (demo.m)
3.confirming the contribution ratio (demo2.m)

See the github page for more detail.
https://github.com/kitayama1234/MATLAB-Kernel-PCA

[Example usage]
% There are a training dataset 'X' and testing dataset 'Xtest'

% train pca model with 'X'
kpca = KernelPca(X, 'gaussian', 'gamma', 2.5, 'AutoScale', true);

% project 'X' using the fitted model
projected_X = project(kpca, X, 2);

% project 'Xtest' using the fitted model
projected_Xtest = project(kpca, Xtest, 2);