Get started with JAX by Aleksa Gordić
A collection of tutorials and resources to help developers learn JAX, Flax, and Haiku for machine learning.
What is Get started with JAX by Aleksa Gordić?
Get Started with JAX is a learning resource repository designed to help developers master the JAX ecosystem, including Flax and Haiku. It provides a series of video tutorials and Jupyter notebooks that cover everything from basic JAX operations to building neural networks from scratch and training on multiple TPU cores. The project aims to lower the entry barrier for using JAX as an alternative to established frameworks like PyTorch and TensorFlow.
Machine learning practitioners, researchers, and students who want to learn JAX, Flax, and Haiku for developing and training neural networks. It's especially useful for those transitioning from PyTorch or TensorFlow.
It offers a curated, hands-on learning path with practical examples and Colab-ready notebooks, saving learners time by focusing on high-quality, personally vetted content rather than overwhelming them with every available resource.
Overview
The purpose of this repo is to make it easy to get started with JAX, Flax, and Haiku. It contains my "Machine Learning with JAX" series of tutorials (YouTube videos and Jupyter Notebooks) as well as the content I found useful while learning about the JAX ecosystem.
Use Cases
Best For
- Learning JAX fundamentals like jit, grad, and vmap
- Transitioning from PyTorch or TensorFlow to the JAX ecosystem
- Building neural networks from scratch using pure JAX
- Training machine learning models on multiple TPU cores
- Getting started with Flax for neural network development
- Following structured video tutorials with accompanying code
Not Ideal For
- Experienced JAX users seeking comprehensive documentation or the latest advanced features beyond tutorials
- Learners preferring text-based or offline resources, as content is video-heavy and requires internet for YouTube and Google Colab
- Teams needing production deployment guides or code integration, since the focus is purely on educational examples
- Developers looking for immediate, complete coverage of Haiku, as that tutorial is marked 'coming soon' in the README
Pros & Cons
Pros
Offers personally vetted, high-quality content to streamline learning without overwhelming users, as emphasized in the README's philosophy of avoiding exhaustive resources.
Interactive Jupyter notebooks can be opened directly in Google Colab, eliminating local setup, which is especially convenient for Windows users as noted in the README.
Covers from JAX basics like jit and grad to advanced multi-device training on TPUs, with step-by-step video guides and accompanying notebooks.
Includes a tutorial on building and training a neural network from scratch using only JAX, providing deep understanding without relying on higher-level libraries.
Cons
The Haiku tutorial is listed as 'coming soon,' leaving a gap in ecosystem coverage and making it less useful for those specifically interested in Haiku.
Relies on YouTube for videos and Google Colab for notebooks, which may not suit offline learning or environments with restricted internet access.
As a curated resource, it doesn't cover all aspects of JAX; users might need to supplement with other materials like the awesome-jax repo for exhaustive knowledge.
Open Source Alternative To
Get started with JAX by Aleksa Gordić is an open-source alternative to the following products:
TensorFlow is an open-source machine learning framework developed by Google for building and deploying ML models across various platforms.
PyTorch is an open-source machine learning framework that provides tensor computation with strong GPU acceleration and deep neural networks built on a tape-based autograd system.
Frequently Asked Questions
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.