Aryan06406/From-Scratch ? reverse-engineered prompt

Build me a Python project that implements math and machine learning algorithms from scratch, with a focus on linear algebra, probability, statistics, and a few basic ML methods.

I want the code organized as a clean package, with some small visualization scripts that can be run from the project root and save their output images into a figures folder automatically. Make sure the examples for vector and matrix style operations are easy to run, and that the internal imports work correctly when I use python module commands.

Please keep the implementations self contained, without relying on ready made ML libraries for the core logic. If you need to check current Python docs or best practices for package imports, go ahead and look them up online. Also make the setup simple enough that someone can install the requirements and run the examples without much trouble.