Stanford CS229 Machine Learning

Notebooks covering the Stanford CS229 ML theory and algorithms course.

Source PDF: cs229.pdf

This folder is the bridge from math foundations to core ML algorithms. Use it when you want to understand how the math turns into modeling choices, error analysis, and learning behavior.

Notebooks

#	Notebook	Topics
01	Linear Regression	Normal equations, least squares, feature scaling
02	Gradient Descent	Batch, stochastic, mini-batch GD, convergence
03	Locally Weighted Regression	Non-parametric regression, bandwidth selection
04	Logistic Regression	Binary classification, sigmoid, cross-entropy
05	Generative Models	Gaussian discriminant analysis, Naive Bayes
06	Support Vector Machines	Margin maximization, kernel trick, SMO
07	Regularization	L1/L2 penalties, bias-variance tradeoff
08	Learning Theory	PAC learning, VC dimension, generalization bounds
09	Decision Trees	CART, pruning, information gain, Gini impurity
10	Neural Networks Basics	Perceptron, feedforward nets, activation functions
11	Neural Networks Advanced	Backprop, dropout, batch norm, architectures
12	ML Strategy	Error analysis, dataset splits, debugging ML
13	Clustering	K-Means, hierarchical, DBSCAN, evaluation
14	Dimensionality Reduction	PCA, t-SNE, autoencoders
15	Reinforcement Learning	MDPs, Q-learning, policy gradient
X01	Anomaly Detection	Gaussian-based, isolation forest
X02	Recommender Systems	Collaborative filtering, matrix factorization
-	Practice Problems	Review exercises across all topics

Prerequisites

foundational/ notebooks 01-04 (linear algebra, calculus, probability, gradient descent)
Python 3.8+, NumPy, Matplotlib, scikit-learn

Suggested Order

Follow the numbered sequence (01-15). X01-X02 are supplementary.

How To Use This Folder Well

Work through the numbered sequence instead of skipping directly to fashionable topics.
Pay special attention to optimization, regularization, learning theory, and ML strategy because those ideas transfer across the entire repo.
Use this folder to connect equations to algorithm behavior, not just to memorize model names.

mml-book/ - deeper math foundations
islp-book/ - statistical learning perspective on similar topics

What Comes Next

Continue to ../islp-book/README.md if you want a more statistical interpretation of similar methods.
Continue to ../advanced/README.md if you want deeper theory after this layer.
Return to ../../02-data-science/README.md or ../../06-neural-networks/README.md to apply the ideas more concretely.