A classification pipeline for three specific handwritten digits (0, 1, and 2) from the MNIST dataset using classic statistical techniques—specifically, Maximum Likelihood Estimation (MLE), Principal Component Analysis (PCA), Fisher’s Discriminant Analysis (FDA), and Discriminant Analysis (LDA/QDA). The project’s main goal was to see how well these methods, often taught in theory, actually perform on real image data, and whether reducing the dimensionality (with PCA or FDA) would affect classification accuracy in a meaningful way.