rushter/MLAlgorithms ? reverse-engineered prompt

I want to build a Python project to help people learn the internals of machine learning algorithms. The main idea is to have a collection of minimal and clean "from-scratch" implementations that are easy to read, not like a big, optimized library. The code should primarily use numpy and scipy.

It should cover a good range of algorithms. Let's include some classics like linear regression, k-means, and Support Vector Machines (SVMs). Let's also add some ensemble methods like Random Forests and gradient boosting. For deep learning, I'd like to see simple versions of a multi-layer perceptron (MLP), a CNN, and maybe an LSTM.

For each algorithm you implement in the main library folder, please also create a corresponding script in an examples directory that shows a simple use case. Make sure the whole thing is structured as an installable Python package.