As part of a work for the "Kernel Methods" course from the MVA master at ENS-Paris-Saclay. The challenge was to implement Kernel Methods to solve a image classification problem, adhering to the crucial ”Do It Yourself” rule, meaning no use of ML libraries. This repository contains an implementation in Python and an analysis report for this challenge.
To reproduce our experiments and results just run the 'experimentNotebook.ipynb'.