Hands-On Machine Learning with Scikit-Learn and PyTorch — Study Repo

Personal study notes and code for Aurélien Géron's book
Hands-On Machine Learning with Scikit-Learn, Keras & PyTorch (2025 edition)

---

📁 Repository Structure

homl-study/
├── data/                          ← Shared datasets (all chapters use ../data/)
│
├── 01_the_machine_learning_landscape/
│   └── 01_the_machine_learning_landscape.ipynb
├── 02_end_to_end_machine_learning_project/
│   └── 02_end_to_end_machine_learning_project.ipynb
├── 03_classification/
│   └── 03_classification.ipynb
├── 04_training_linear_models/
│   └── 04_training_linear_models.ipynb
├── 05_decision_trees/
│   └── 05_decision_trees.ipynb
├── 06_ensemble_learning_and_random_forests/
│   └── 06_ensemble_learning_and_random_forests.ipynb
├── 07_dimensionality_reduction/
│   └── 07_dimensionality_reduction.ipynb
├── 08_unsupervised_learning/
│   └── 08_unsupervised_learning.ipynb
├── 09_artificial_neural_networks/
│   └── 09_artificial_neural_networks.ipynb
├── 10_neural_nets_with_pytorch/
│   └── 10_neural_nets_with_pytorch.ipynb
├── 11_training_deep_neural_networks/
│   └── 11_training_deep_neural_networks.ipynb
├── 12_deep_computer_vision_with_cnns/
│   └── 12_deep_computer_vision_with_cnns.ipynb
├── 13_processing_sequences_using_rnns_and_cnns/
│   └── 13_processing_sequences_using_rnns_and_cnns.ipynb
├── 14_nlp_with_rnns_and_attention/
│   └── 14_nlp_with_rnns_and_attention.ipynb
├── 15_transformers_for_nlp_and_chatbots/
│   └── 15_transformers_for_nlp_and_chatbots.ipynb
├── 16_vision_and_multimodal_transformers/
│   └── 16_vision_and_multimodal_transformers.ipynb
├── 17_speeding_up_transformers/
│   └── 17_speeding_up_transformers.ipynb
├── 18_autoencoders_gans_and_diffusion_models/
│   └── 18_autoencoders_gans_and_diffusion_models.ipynb
└── 19_reinforcement_learning/
    └── 19_reinforcement_learning.ipynb

---

🗂️ Accessing Datasets

All datasets live in data/ at the root.
From any chapter notebook you can always load data with:

import pandas as pd
from pathlib import Path

DATA_PATH = Path("../data")          # works from any chapter subfolder
housing  = pd.read_csv(DATA_PATH / "housing.csv")

---

⚙️ Environment Setup

1 — Clone & enter the repo

git clone <your-repo-url>
cd homl-study

2 — Install uv

curl -LsSf https://astral.sh/uv/install.sh | sh

3 — Create the environment and install the notebook stack

uv sync

This installs the dependencies required by the current scikit-learn and Polars notebooks, including JupyterLab, Plotnine, and the Chapter 2 stack.

4 — Optional: install the deep learning stack for later chapters

uv sync --extra deep-learning

This adds PyTorch, TorchVision, TorchAudio, and Transformers for the later chapters.

5 — Register a Jupyter kernel

uv run python -m ipykernel install --user --name homl-study --display-name "Python (homl-study)"

6 — Launch JupyterLab

uv run jupyter lab

---

📖 Chapter Index

#	Chapter	Topics
01	The Machine Learning Landscape	ML types, challenges, model selection
02	End-to-End ML Project	Pipeline, preprocessing, evaluation
03	Classification	Binary/multiclass, metrics, SVM
04	Training Linear Models	Linear/Logistic regression, regularization
05	Decision Trees	CART, depth, pruning
06	Ensemble Learning and Random Forests	Bagging, boosting, stacking
07	Dimensionality Reduction	PCA, LLE, UMAP
08	Unsupervised Learning	K-Means, DBSCAN, GMM
09	Artificial Neural Networks	Perceptron, MLP, backprop
10	Neural Nets with PyTorch	PyTorch tensors, autograd, nn.Module
11	Training Deep Neural Networks	Optimisers, BN, dropout, schedules
12	Deep Computer Vision with CNNs	Conv layers, ResNet, transfer learning
13	Processing Sequences Using RNNs and CNNs	RNN, LSTM, GRU, WaveNet
14	NLP with RNNs and Attention	Seq2Seq, attention mechanism
15	Transformers for NLP and Chatbots	BERT, GPT, fine-tuning
16	Vision and Multimodal Transformers	ViT, CLIP, DALL-E
17	Speeding Up Transformers	Distillation, quantisation, pruning
18	Autoencoders, GANs, and Diffusion Models	VAE, GAN, DDPM
19	Reinforcement Learning	MDP, Q-Learning, PPO

---

🔗 Official Resources

• Book: Aurélien Géron — Hands-On ML (O'Reilly)
• Official notebooks: github.com/ageron/handson-ml3

---

📝 Notes

• Each chapter notebook starts clean — notes and experiments are added progressively while reading.
• Datasets that are downloaded programmatically are saved to data/ so they are reused across chapters.