Skip to content

Issues: pytorch/pytorch

[v2.3.1] Release Tracker
#125425 opened May 2, 2024 by atalman
Open 4
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.

Sign up for GitHub

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

12,999 Open 28,809 Closed
12,999 Open 28,809 Closed
Author
Filter by author
Label
Filter by label
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Milestones
Filter by milestone
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

all_gather with gloo backend does not work in inference mode
#126032 opened May 12, 2024 by youkaichao
Inductor error: AttributeError: 'View' object has no attribute 'freeze_layout' high priority module: inductor oncall: pt2 triage review
#126029 opened May 12, 2024 by ezyang
Graph break: torch._dynamo.exc.Unsupported: non-function or method super: <slot wrapper '__setattr__' of 'object' objects> high priority module: dynamo oncall: pt2 triage review
#126028 opened May 12, 2024 by ezyang
Graph break: torch._dynamo.exc.Unsupported: non-function or method super: <slot wrapper '__setitem__' of 'collections.OrderedDict' objects> high priority module: dynamo oncall: pt2 triage review
#126027 opened May 12, 2024 by ezyang
Graph break: Unsupported: torch.* op returned non-Tensor bool call_function <built-in function _is_any_autocast_enabled> good first issue high priority module: dynamo oncall: pt2 triage review
#126026 opened May 12, 2024 by ezyang
torch.compile does not support strided NestedTensor oncall: pt2
#126025 opened May 12, 2024 by ezyang
1
Can not use torch.compile with dynamic=Ture when using multi-threads oncall: pt2
#126024 opened May 12, 2024 by CaoE
1
"Enable " ON)
#126023 opened May 12, 2024 by dylankr1
replace this with a more robust fix
#126022 opened May 12, 2024 by dylankr1
Incorrect output from torch.histc when min and max are equal.
#126021 opened May 12, 2024 by Manas-Tiwari
CUDA is not working!
#126020 opened May 12, 2024 by Smellycod
2
Remove out parameter from functions in PyTorch to simplify the functions and the documentation
#126016 opened May 12, 2024 by hyperkai
Issue torch.cuda.mem_get_info runtime error: hip error: invalid argument
#126015 opened May 12, 2024 by zhangnju
The unexpected behavior of can_cast()
#126014 opened May 12, 2024 by hyperkai
can_cast() with from argument doesn't work but can_cast() without from argument works
#126012 opened May 11, 2024 by hyperkai
1
1
what(): new_refcount != 1 INTERNAL ASSERT FAILED oncall: pt2
#126010 opened May 11, 2024 by ezyang
predict_fn error: backend='torch_tensorrt' raised: TypeError: pybind11::init(): factory function returned nullptr
#126009 opened May 11, 2024 by geraldstanje
Torch distribution blocking ports and failing to connect
#126008 opened May 11, 2024 by nightsSeeker
The arguments tesnor1 and tensor2 of result_type() don't work but tensor and other work
#126006 opened May 11, 2024 by hyperkai
CUDNN not detected in official nightly devel image
#126005 opened May 11, 2024 by bhack
3
Inform when weights loading promotes types
#126000 opened May 11, 2024 by isentropic
1
Got different outputs with and without torch.compile module: correctness (silent) issue that returns an incorrect result silently needs reproduction Someone else needs to try reproducing the issue given the instructions. No action needed from user oncall: pt2
#125999 opened May 11, 2024 by kunlunl
1
StateDictOptions claims default strict is False, but it's actually True
#125992 opened May 11, 2024 by ad8e
1
DISABLED test_transformer_training_is_seq_parallel_True (__main__.DistTensorParallelExampleTest) module: rocm AMD GPU support for Pytorch skipped Denotes a (flaky) test currently skipped in CI.
#125991 opened May 11, 2024 by huydhn
1
torchrun c10d backend doesn't seem to work with python 3.12, giving segmentation fault because of calling obmalloc without holding GIL oncall: distributed Add this issue/PR to distributed oncall triage queue
#125990 opened May 11, 2024 by TanyaAdams1
1
ProTip! no:milestone will show everything without a milestone.