wepe/MachineLearning ? reverse-engineered prompt

Build me a Python project that collects the classic machine learning algorithms in one place, like k nearest neighbors, logistic regression, naive Bayes, decision trees, PCA, KMeans, SVM, ridge regression, and a few simple deep learning examples too.

I want it to feel like a learning notebook, not a polished product, with clear code that is easy to read, run, and compare. Use NumPy for the core implementations and Matplotlib for plots and visualizations where it makes sense, like showing clustering results or reduced dimensional data. If it helps, include small demos on simple datasets such as MNIST or toy data so each algorithm has something practical to run.

Please organize the code into separate folders by topic, add short explanations in the README, and make sure each example can be run on its own. Keep it simple and educational, and look up current docs online if you need to.