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.

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

13,073 Open 28,374 Closed
13,073 Open 28,374 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

DataLoader's pin_memory is default to CUDA if parameter pin_memory_device is not set module: dataloader Related to torch.utils.data.DataLoader and Sampler module: PrivateUse1 private use triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124908 opened Apr 25, 2024 by wizzniu
2.3.0 not backward compatible with torchdata
#124907 opened Apr 25, 2024 by npuichigo
Many cases in distributed/elastic/multiprocessing/redirects_test.py fails when use pytest oncall: distributed Add this issue/PR to distributed oncall triage queue
#124906 opened Apr 25, 2024 by cdzhan
Dataloader crashes with FileNotFoundError randomly when num_workers>0 on Ubuntu 22.04 module: crash Problem manifests as a hard crash, as opposed to a RuntimeError module: dataloader Related to torch.utils.data.DataLoader and Sampler triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124903 opened Apr 25, 2024 by TomHeaven
Add support for IPC features for PrivateUse devices module: PrivateUse1 private use oncall: distributed Add this issue/PR to distributed oncall triage queue
#124902 opened Apr 25, 2024 by mengpenghui
RuntimeError: invalid dtype for bias when use compile + autocast oncall: pt2
#124901 opened Apr 25, 2024 by yiliu30
[dynamo] Crash when context manager object crosses a graph break bug module: dynamo oncall: pt2 topic: not user facing topic category triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124900 opened Apr 25, 2024 by williamwen42
@williamwen42
Certain .pyi files are not encoded as UTF-8 in Windows module: binaries Anything related to official binaries that we release to users module: regression It used to work, and now it doesn't module: typing Related to mypy type annotations triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124897 opened Apr 25, 2024 by marovira 2.3.1
@malfet
7
How to catch NCCL collective timeout in Python needs reproduction Someone else needs to try reproducing the issue given the instructions. No action needed from user oncall: distributed Add this issue/PR to distributed oncall triage queue
#124887 opened Apr 24, 2024 by gkroiz
2
Windows build step did not fail despite error module: ci Related to continuous integration module: windows Windows support for PyTorch triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124886 opened Apr 24, 2024 by clee2000
1
ONNX dynamic sized model export with torch.onnx.dynamo_export fails when torch.nn.functional.interpolate is used module: onnx Related to torch.onnx triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124884 opened Apr 24, 2024 by pinkhamr-fb
Nested wrapper subclasses with torch.compile is broken module: pt2-dispatcher PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op, module: __torch_dispatch__ oncall: pt2
#124878 opened Apr 24, 2024 by bdhirsh
SDPA memory efficient kernel returns NaNs when the query and key are different lengths module: multi-headed-attention oncall: transformer/mha Issues related to Transformers and MultiheadAttention triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124877 opened Apr 24, 2024 by Craigacp
pytorch Windows MKL cmake file don't support static link mkl. intel This tag is for PR from Intel module: mkl Related to our MKL support module: windows Windows support for PyTorch triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124869 opened Apr 24, 2024 by xuhancn
@xuhancn
tensor.dtype.to_complex() crashes kernel after ~100 calls in ipython kernel module: complex Related to complex number support in PyTorch module: crash Problem manifests as a hard crash, as opposed to a RuntimeError 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
#124868 opened Apr 24, 2024 by RichieHakim 2.3.1
@malfet
8
support side effects in HOPs? module: pt2-dispatcher PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op, oncall: pt2
#124866 opened Apr 24, 2024 by zou3519
2
Deprecate unsupported types in operator registration high priority module: custom-operators module: pt2-dispatcher PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op, oncall: pt2 triage review
#124863 opened Apr 24, 2024 by zou3519
1
torch._dynamo.assume_constant_result does not work outside nn.Module module: dynamo oncall: pt2
#124858 opened Apr 24, 2024 by IvanKobzarev
2
Use lintrunner-adapters for our adapters when possible module: lint Issues related to our Python/C++ lint rules (run by Travis) triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124857 opened Apr 24, 2024 by ezyang
2
ShapeEnv canonicalization is still over-aggressive module: dynamic shapes oncall: pt2
#124855 opened Apr 24, 2024 by ezyang
[inductor] unexpected cuda:0 device usage when compiling and runing a model on cuda:1 inductor_pattern_match oncall: pt2
#124854 opened Apr 24, 2024 by vfdev-5
aten::nonzero calls taking a huge amount of time when using MPS backend vs CPU high priority module: crash Problem manifests as a hard crash, as opposed to a RuntimeError module: mps Related to Apple Metal Performance Shaders framework module: performance Issues related to performance, either of kernel code or framework glue triage review
#124850 opened Apr 24, 2024 by theo-costain-arondite 2.3.1
@malfet
2
[inductor][cpu]LoweringException: AttributeError: 'NoneType' object has no attribute 'get_origin_node' in 2024-04-20 nightly release oncall: pt2
#124844 opened Apr 24, 2024 by zxd1997066
Forwardpropagation of high-dimensional tensors through the nn.Linear module becomes multiple times slower since pytorch 2.1.0 module: cuda Related to torch.cuda, and CUDA support in general triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#124838 opened Apr 24, 2024 by StepinSilence
2
[inductor][cpu]RuntimeError: no channels last format strides exist in 1 dimensions in 2024-04-21 nightly release oncall: pt2
#124837 opened Apr 24, 2024 by zxd1997066
ProTip! Updated in the last three days: updated:>2024-04-21.