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);
引用
Masaki Kitayama (2024). MATLAB-Kernel-PCA (https://github.com/kitayama1234/MATLAB-Kernel-PCA), GitHub. 取得済み .
MATLAB リリースの互換性
プラットフォームの互換性
Windows macOS Linuxカテゴリ
- AI, Data Science, and Statistics > Statistics and Machine Learning Toolbox > Dimensionality Reduction and Feature Extraction >
タグ
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!