pytorch
https://github.com/pytorch/pytorch
Python
Tensors and Dynamic neural networks in Python with strong GPU acceleration
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 supported926 Subscribers
View all SubscribersAdd a CodeTriage badge to pytorch
Help out
- Issues
- DPP training incompatibility with checkpoint and detach
- Cannot call CUDAGeneratorImpl::current_seed during CUDA graph capture
- Make FSDP easier to debug when erroring in backward pass
- deepcopy of LazyLinear fails
- [FSDP] `test_summon_single_param()` is misleading
- [MPS] Bug on training CNN+LSTM
- torch.nn.functional.log_softmax parameter '_stacklevel' undocumented
- Support for CSR Tensor with NN layers
- make_fx + aot_autograd segfaults
- FSDP init can crash with shared parameters
- Docs
- Python not yet supported