Hello! I'm Josh Starmer and welcome to the github repository for my book, The StatQuest Illustrated Guide to Neural Networks and AI. Here are the coding tutorials associated with each chapter. Enjoy!
Shameless Self Promotion!!!! If you don't already have The StatQuest Illustrated Guide to Neural Networks and AI, you should get it! The book starts with the basics, showing you how a simple neural network fits a shape to data and then builds from there, one step at a time, until you have mastered the concepts behind today’s revolution in AI. Furthermore, putting the theory into practice is easy because each major concept is paired with a PyTorch tutorial that teaches you how to code each neural network from scratch.
Here's what people are saying about The StatQuest Illustrated Guide to Neural Networks and AI
This book is a game changer. With simple yet thorough diagrams, useful and relevant code examples, and his usual fun analogies, Starmer one more time manages to demystify the complexity behind neural networks. Topics such as transformers, LSTMs, and attention mechanisms will not be a mystery anymore after reading this gem. - Luis Serrano, Founder and CEO of Serrano Academy and author of Grokking Machine Learning
Fellow fans of StatQuest should rejoice for this accessible introduction to some of the most important concepts in AI. It reads like a playful comic that gently and visually guides you along the math that powers today's AI revolution - Jay Alammar, author of The Illustrated Transformer and, with Maarten Grootendorst, Hands-On Large Language Models
The physcial copies of the book are available on Amazon. If your more of a digital kind of person, you can also purchase a PDF.
All of the tutorials listed below can be run immediately in the cloud for free with the Google Colab and Lightning Studio links. However, if you would prefer to run the tutorials on your own computer, there are instructions for setting up your environment with docker file below.
| Chapter | Google Colab | Lightning Studio |
|---|---|---|
| 01: Fundamental Concepts in Neural Networks and AI!!! |  |
 |
| 02: Optimizing Weights and Biases with Backpropagation!!! | |
|
| 03: Networks with Multiple Inputs and Outputs!!! | |
|
| 04: Simplifying Outputs with ArgMax and SoftMax!!! | |
|
| 05: Speeding Up Training with Cross Entropy!!! | (see chapter 4) | |
| 06: Image Classification with Convolutional Neural Networks!!! | |
|
| 07: Stock Prediction with Recurrent Neural Networks (RNNs)!!! | (see chapter 8) | |
| 08: Better Stock Prediction with Long Short-Term Memory (LSTM)!!! | |
|
| 09: Converting Words to Numbers with Word Embedding!!! | |
|
| 10: Language Translation with Seq2Seq and Encoder-Decoder Models!!! | |
|
| 11: Better Language Translation with Attention!!! | |
|
| 12: Even Better Language Translation with Transformers!!! | |
|
| 13: Generating Lots of Text with Decoder-Only Transformers!!! | |
|
| 14: Classification and Clustering with Encoder-Only Transformers!!! | |
|
Run the following commands to install Docker, Buildx, and Compose on Ubuntu/Linux (including WSL2 on Windows 11) and launch Jupyter with course dependencies:
# 1
chmod +x install_docker.sh
# 2
./install_docker.sh
# 3
docker build -t pytorch-jupyter .
# 4
docker run -p 8888:8888 --name jupyter-env pytorch-jupyter
# 5
# go to this url to see the notebooks in this repo in your Jupyter setup!: http://localhost:8888/