| 2 | |
| 3 | |
| 4 | class PCA: |
| 5 | |
| 6 | def __init__(self, n_components): |
| 7 | self.n_components = n_components |
| 8 | self.components = None |
| 9 | self.mean = None |
| 10 | |
| 11 | def fit(self, X): |
| 12 | # mean centering |
| 13 | self.mean = np.mean(X, axis=0) |
| 14 | X = X - self.mean |
| 15 | |
| 16 | # covariance, functions needs samples as columns |
| 17 | cov = np.cov(X.T) |
| 18 | |
| 19 | # eigenvectors, eigenvalues |
| 20 | eigenvectors, eigenvalues = np.linalg.eig(cov) |
| 21 | |
| 22 | # eigenvectors v = [:, i] column vector, transpose this for easier calculations |
| 23 | eigenvectors = eigenvectors.T |
| 24 | |
| 25 | # sort eigenvectors |
| 26 | idxs = np.argsort(eigenvalues)[::-1] |
| 27 | eigenvalues = eigenvalues[idxs] |
| 28 | eigenvectors = eigenvectors[idxs] |
| 29 | |
| 30 | self.components = eigenvectors[:self.n_components] |
| 31 | |
| 32 | def transform(self, X): |
| 33 | # projects data |
| 34 | X = X - self.mean |
| 35 | return np.dot(X, self.components.T) |
| 36 | |
| 37 | |
| 38 | # Testing |