accelerate
https://github.com/huggingface/accelerate
Python
🚀 A simple way to train and use PyTorch models with multi-GPU, TPU, mixed-precision
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported1 Subscribers
Add a CodeTriage badge to accelerate
Help out
- Issues
- DeepSpeedEngineWrapper.backward() does a bit too much
- get_max_memory() returns allocated memory for XPU instead of total device memory
- ValueError: FlatParameter requires uniform dtype but got torch.float16 and torch.float32
- Some adjustment for supporting Deepspeed-Ulysses
- Plan to support FSDP2?
- Dataloader WeightedRandomSampler + Distributed Training
- gather objects in TPU is not supported
- handle weight sharing with init_on_device
- [Feature Request] Allows registering custom trackers to internal tracker type registry
- Unable to specify HYBRID_SHARD for FSDP which requires process group or device_mesh to be passed
- Docs
- Python not yet supported