Principal component analysis (PCA) is a statistical procedure that uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components
Model = Eigenface(X_train,Y_train,num_classes,image_dimensions);
Model.train({{true if m>n}});
Model.show_and_plot_eig_vals()
%Plot the eigenvalues
Model.Give_Basis_Dim(integer=number of eigenfaces to use)
Model.give_test_data(X_test,Y_test)
Model.test_and_give_accuracy()
%You can also plot eignefaces using
Model.plot_eigenfaces()Below is the set of 12 eigenfaces generated
References :