back to course overview
PCA (Principal Component Analysis) in Python - ML From Scratch 11
In this Machine Learning from Scratch Tutorial, we are going to implement a PCA algorithm using only built-in Python modules and numpy. We will also learn about the concept and the math behind this popular ML algorithm.
All algorithms from this course can be found on GitHub together with example tests.
import numpy as np class PCA: def __init__(self, n_components): self.n_components = n_components self.components = None self.mean = None def fit(self, X): # Mean centering self.mean = np.mean(X, axis=0) X = X - self.mean # covariance, function needs samples as columns cov = np.cov(X.T) # eigenvalues, eigenvectors eigenvalues, eigenvectors = np.linalg.eig(cov) # -> eigenvector v = [:,i] column vector, transpose for easier calculations # sort eigenvectors eigenvectors = eigenvectors.T idxs = np.argsort(eigenvalues)[::-1] eigenvalues = eigenvalues[idxs] eigenvectors = eigenvectors[idxs] # store first n eigenvectors self.components = eigenvectors[0:self.n_components] def transform(self, X): # project data X = X - self.mean return np.dot(X, self.components.T)
Join My Newsletter! Get Python and ML tips emailed directly to your inbox. Each month you’ll get a summary of all the content I created, including the newest videos, articles, promotions, tips, and more.
Learn all the necessary basics to get started with TensorFlow 2 and Keras.
Learn all the necessary basics to get started with this deep learning framework.
Implement popular Machine Learning algorithms from scratch using only built-in Python modules and numpy.