JohnMachado11/Neural-Networks-from-Scratch-in-Python ? reverse-engineered prompt

Build me a Python learning project that teaches neural networks from scratch using only Python and NumPy.

I want it to walk through the whole journey step by step, starting with a single neuron, then layers, activation functions, loss, optimization, derivatives, gradients, backpropagation, and training. It should include examples for classification and regression, some regularization like L1, L2, and dropout, and a simple model object that can be trained, evaluated, saved, loaded, and used for prediction later.

Please organize it like a hands on tutorial with small self contained examples that build on each other, so someone can run the code and actually see how everything works. If it helps, include a real dataset example near the end and make sure the project is easy to follow for someone learning deep learning for the first time. Look up current docs online if you need to.