Skip to content

Issues: pytorch/pytorch

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort

Issues list

Output mismatch between torch.compile and eager mode bug high priority module: inductor oncall: cpu inductor CPU Inductor issues for Intel team to triage oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#138908 opened Oct 25, 2024 by vivekvpandya
Solve issue 138464 bug module: edge cases Adversarial inputs unlikely to occur in practice module: NaNs and Infs Problems related to NaN and Inf handling in floating point module: special Functions with no exact solutions, analogous to those in scipy.special open source triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#138653 opened Oct 22, 2024 by vladimirrotariu Loading…
compilation of rrelu_with_noise with bfloat16 input does not capture noise mutation bug module: pt2-dispatcher PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op, oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#136784 opened Sep 26, 2024 by IvanKobzarev
xpu: runtime error in safetensors: Error while trying to find names to remove to save state dict bug module: xpu Intel XPU related issues triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#135550 opened Sep 9, 2024 by dvrogozh
torch.multiprocessing.start_processes is blocking with large input arguments bug module: multiprocessing Related to torch.multiprocessing triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#133010 opened Aug 8, 2024 by d4l3k
Generated some code to try to torch compile it, and it fails for the same code when it's torch compiled bug module: fakeTensor oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#128626 opened Jun 13, 2024 by Skylion007
[BUG] Circular dependency torch.nn.parameter -> torch.nn -> torch.autograd -> torch._C._autograd_init() -> torch.nn.parameter bug module: autograd Related to torch.autograd, and the autograd engine in general module: python frontend For issues relating to PyTorch's Python frontend needs design triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#128609 opened Jun 13, 2024 by XuehaiPan
2nd compile of deepcopy(model) fails on multiple ubuntu-pc (fatal error: Python.h: file not found) bug module: dynamo oncall: pt2 topic: build triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#128121 opened Jun 6, 2024 by alex77g2
[inline-inbuilt-nn-modules] tensordict functional calls with nn.Module silently gives the wrong (non-functional) result bug module: dynamo oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#127173 opened May 25, 2024 by vmoens
Distributed checkpoint state_dict load may report nccl error and hide the real root-cause exception bug has workaround module: distributed_checkpoint oncall: distributed checkpointing Oncall label should be attached to any issues related to distributed checkpointing. oncall: distributed Add this issue/PR to distributed oncall triage queue triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#122529 opened Mar 22, 2024 by bigning
[Inductor][Triton] test_mixed_mm2_cuda fails on Hopper bug module: inductor needs reproduction Someone else needs to try reproducing the issue given the instructions. No action needed from user oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module upstream triton Upstream Triton Issue
#122227 opened Mar 19, 2024 by Aidyn-A
torch.logsumexp throws RuntimeError instead of reducing all dimensions when dim=None actionable bug module: named tensor Named tensor support module: python frontend For issues relating to PyTorch's Python frontend triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#121725 opened Mar 12, 2024 by TheoBoyer
torch.func.grad hangs on nested differentiation bug high priority module: functorch Pertaining to torch.func or pytorch/functorch triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#121711 opened Mar 12, 2024 by rtqichen
copy_() produces wrong results for boolean tensors on mps bug module: mps Related to Apple Metal Performance Shaders framework triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#121439 opened Mar 7, 2024 by BoyuanFeng
2
2
Misleading log_abs_det_jacobian shape for StackTransform? bug module: distributions Related to torch.distributions triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#121220 opened Mar 5, 2024 by LouisRouillard
Restoring SequentialLR has undocumented side-effects on Optimizer actionable bug module: LrScheduler module: optimizer Related to torch.optim triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#119168 opened Feb 5, 2024 by ceisenach
Dynamo fails on scalar bit shifts when dynamic=True bug module: dynamic shapes oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#119152 opened Feb 4, 2024 by ar0ck
Inconsistent Behavior of torch.dsplit with torch.compile bug internal ramp-up task Tasks that are suitable for new folks w/ high-touch guidance from senior PyTorch folks low priority We're unlikely to get around to doing this in the near future module: dynamo oncall: pt2 topic: fuzzer triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#118741 opened Jan 31, 2024 by Gwihwan-Go
Getting "RuntimeError: Trying to backward through the graph a second time" when calling backward on compiled unbind / split ops bug module: aotdispatch umbrella label for AOTAutograd issues module: pt2-dispatcher PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op, oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#118739 opened Jan 31, 2024 by markurcin
Different outputs of div and threshold with a specific input bug module: decompositions Topics related to decomposition (excluding PrimTorch) oncall: pt2 topic: fuzzer triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#118642 opened Jan 30, 2024 by zoux1a
[FSDP] unexpected reshard in backward because of unused gradable input in frozon modules bug module: fsdp triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#117510 opened Jan 16, 2024 by weifengpy
Dynamo fails to track dataclass actionable bug dynamo-dataclasses dynamo-must-fix These bugs affect TorchDynamo reliability. module: dynamo oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#116264 opened Dec 21, 2023 by guilhermeleobas
[Dynamo] bytecode transformed by Dynamo is not serializable by marshal bug low priority We're unlikely to get around to doing this in the near future module: dynamo oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#116013 opened Dec 18, 2023 by youkaichao
Inductor autotuning with None Tensors leads to cuda IMA with user-defined triton kernel bug module: inductor module: user triton related to ability to directly torch.compile triton kernels oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#115344 opened Dec 7, 2023 by RobertCsordas
PT2 QAT flow fails to get reasonable accuracy with mobilenet_v3_large bug low priority We're unlikely to get around to doing this in the near future module: pt2 accuracy oncall: pt2 oncall: quantization Quantization support in PyTorch triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#114859 opened Nov 30, 2023 by leslie-fang-intel
ProTip! What’s not been updated in a month: updated:<2024-09-30.