Machine Learning

Complete interactive reference covering 38 topics — from linear algebra fundamentals to diffusion models. Every topic has core math, visual intuition, and PyTorch code.

30 topics from tokenization to agents — transformer internals, training paradigms (RLHF, DPO, LoRA), inference optimization, RAG, and evaluation benchmarks.

25 topics on deploying and operating ML systems at scale — model serving, monitoring, pipelines, feature stores, and governance. From experiment tracking to incident response.

25 topics on forecasting and temporal data — from stationarity and ARIMA to LSTMs, Transformers, and production backtesting. Interactive visualizations for every model.

4 practical mini-cases that turn the topic library into workflows: load a dataset, build a baseline, choose metrics, avoid common leakage, and copy a notebook-ready starter.

9 hands-on activities — linear regression, K-Means, classification boundaries, neural networks, feature scaling, timeseries forecasting, and more. Build intuition by experimenting.