handson-unsupervised-learning

Why we included this project

This repository holds the code that accompanies Ankur Patel's O'Reilly book on unsupervised learning, organized as runnable Jupyter notebooks rather than a static reading list. If you are past the basics of supervised models and want to work with unlabeled data, the examples cover clustering, dimensionality reduction, anomaly detection, and autoencoders, all built on scikit-learn and TensorFlow. The tasks are grounded in real problems: catching credit card fraud, segmenting loan borrowers, and building a movie recommender with restricted Boltzmann machines. There is also a look at generative models such as deep belief networks and GANs for synthetic image generation. Because the notebooks are arranged by chapter with setup instructions for local and Colab environments, it works well as a study companion for teams or individuals learning these techniques.