-
Notifications
You must be signed in to change notification settings - Fork 3.6k
Description
π Feature
Motivation
Native PyTorch FSDP is a powerful alternative to regular DDP.
Pitch
Add the strategy to Lite. Build it using the existing implementation in Trainer.
If you enjoy Lightning, check out our other projects! β‘
-
Metrics: Machine learning metrics for distributed, scalable PyTorch applications.
-
Lite: enables pure PyTorch users to scale their existing code on any kind of device while retaining full control over their own loops and optimization logic.
-
Flash: The fastest way to get a Lightning baseline! A collection of tasks for fast prototyping, baselining, fine-tuning, and solving problems with deep learning.
-
Bolts: Pretrained SOTA Deep Learning models, callbacks, and more for research and production with PyTorch Lightning and PyTorch.
-
Lightning Transformers: Flexible interface for high-performance research using SOTA Transformers leveraging PyTorch Lightning, Transformers, and Hydra.