The Top 10 Machine Learning Open Source Projects

Awesome Scalability is a curated reading list of patterns, papers, blog posts, and talks for building scalable, reliable, and performant large-scale systems. Maintained by Binh Nguyen, the repository organizes hundreds of links into a structured reference covering everything from load balancing and caching to distributed consensus and incident management.

This is not a code library or framework. You won't install anything from it. What you get is a single-page index of battle-tested resources organized by system design topic, with enough breadth to cover the full stack of scaling problems — from database sharding to organizational scaling. With over 71,000 GitHub stars, it has become one of the most widely bookmarked system design references on GitHub.

Robust Speech Recognition via Large-Scale Weak Supervision

🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

A curated list of awesome Machine Learning frameworks, libraries and software.

Learn how to develop, deploy and iterate on production-grade ML applications.

Open Source Computer Vision Library

A latent text-to-image diffusion model

PyTorch is an open-source machine learning framework built around tensors, automatic differentiation, and GPU acceleration. The core repo at github.com/pytorch/pytorch describes it plainly: you get tensor computation similar to NumPy, plus deep neural network support built on a tape-based autograd system. That combination is why so many teams use it for model research, fine-tuning, distributed training, and production inference.

You can use PyTorch for straightforward feed-forward models, large training jobs spread across many GPUs, custom CUDA operators, and non-Python integrations through LibTorch. You can also stay in eager mode for day-to-day model work and opt into torch.compile when you want the compiler stack to try to speed up execution without forcing a full rewrite.

An Open Source Machine Learning Framework for Everyone

List of Computer Science courses with video lectures.