DeepLearning

Why we included this project

People who work through the Deep Learning textbook tend to hit the same wall: the math is dense and the book ships no code, so it's hard to see how the theory translates into running programs. This project closes that gap by restating the book's core ideas with step-by-step derivations and implementing each one from scratch in plain NumPy, deliberately leaving out TensorFlow and PyTorch. The few sklearn calls in the PDF exist only to verify that the hand-built implementations are correct. Beyond the book itself, the author adds full derivations and code for ensemble methods such as random forests, AdaBoost, GBDT, and XGBoost, plus Bayesian linear regression and Gaussian process regression. Treat it as a thorough study companion and reference, not as a library for production use.