Contributing to Lightning AI OSS

https://docs.google.com/presentation/d/18oSdKdsrQuIefmHP0Mfvccjziskwfk6R9E0ZqEUQ0YE/edit?usp=sharing

Transcript

1. Contributing to Lightning AI OSS Aniket Maurya & Jiří Borovec
2. Agenda 2 • Lightning AI open-source ecosystem • Finding GitHub Issues • Fork and Clone the Repository • Create a Pull Request
3. Lightning AI OSS Ecosystem
5. PyTorch Lightning v1.9.3 5 Scaling Foundation Models with Lightning You do the science, we do the engineering https://lightning.ai
6. Fabric v1.9.3 6 Scaling Foundation Models with Lightning Raw PyTorch PyTorch Lightning Fabric Full control More boilerplate Fully managed Less boilerplate
7. TorchMetrics v0.11.3 7 • 90+ PyTorch metrics implementations • Reduces Boilerplate • Distributed-training compatible • Rigorously tested • Automatic accumulation over batches • Automatic synchronization between multiple devices
8. Apps v1.9.3 8 • Lightning Apps are composed of Flow and Works • API for running long tasks like training, deployment and data download • Orchestrator for managing the long running tasks • Prebuilt components for training, serving, and scaling
9. Let’s create our Pull Request
10. Never hesitate to ask for help 💜