An In-depth Guide to PCA with NumPy

Through eigenvalue analysis

Photo by Vincent Guth on Unsplash

Principal Component Analysis (hereafter, PCA) is one of the most popular dimensionality reduction techniques used in machine learning.

It is considered a linear dimensionality reduction technique as it finds a linear combination of input features in a lower dimensional form.

More precisely,

PCA is a linear dimensionality reduction technique that transforms the p number of input variables into a smaller k (k << p) number of uncorrelated variables called principal components by taking advantage of the existing correlations between the input variables in the dataset[ref¹]

ref¹: 3 Easy Steps to Perform Dimensionality Reduction Using Principal Component Analysis (PCA)

We’ve performed PCA so many times before in my previous articles. There, we always used the Scikit-learn popular PCA() function to perform PCA.

Today, we will apply PCA to a dataset without using the PCA() function, but using some specific functions in NumPy. The entire process seems to be tedious, but it is a great way to get an in-depth…