machine-learning

Why we included this project

Built around Jupyter notebooks, this repository documents one practitioner's path through data science and machine learning, and it reads like a course you could follow at your own pace. Each topic pairs the math with working Python code, using the scientific stack for from-scratch implementations and libraries like scikit-learn, PyTorch, and XGBoost where those make more sense. The coverage goes well past deep learning into time series, recommender systems, reinforcement learning, model deployment, and even A/B testing and operations research. Notebooks are available both interactively and as rendered HTML, so you can run them or just read through the notes. It is a learning resource, not a library you install, so expect to work through the material rather than import it.