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

11,336 Open 23,110 Closed
11,336 Open 23,110 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

gradcheck produces false positives with sparse inputs when masked=False. module: autograd Related to torch.autograd, and the autograd engine in general module: sparse Related to torch.sparse triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103518 opened Jun 13, 2023 by nikitaved
How to add testing capabilities for third party devices
#103506 opened Jun 13, 2023 by Bin1024
[functorch] [FakeTensorMode, meta tensor] + aot_autograd Bug.
#103505 opened Jun 13, 2023 by ConnollyLeon
CUDA error when calling torch._softmax_backward_data when the reduced dimension was large. module: cuda Related to torch.cuda, and CUDA support in general module: edge cases Adversarial inputs unlikely to occur in practice triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103501 opened Jun 13, 2023 by garfield1997
1
CUBLAS_WORKSPACE_CONFIG can not be parsed module: cublas Problem related to cublas support triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103499 opened Jun 13, 2023 by dylanprins
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 6047) of binary: /home/win10-ubuntu/anaconda3/envs/vicuna-7b/bin/python oncall: distributed Add this issue/PR to distributed oncall triage queue
#103498 opened Jun 13, 2023 by Hzzhang-nlp
1
DISABLED test_mem_get_info (__main__.TestCuda) module: flaky-tests Problem is a flaky test in CI module: unknown We do not know who is responsible for this feature, bug, or test case. skipped Denotes a (flaky) test currently skipped in CI.
#103495 opened Jun 13, 2023 by pytorch-bot bot
3
DISABLED test_streaming_backwards_sync (__main__.TestCuda) module: flaky-tests Problem is a flaky test in CI module: unknown We do not know who is responsible for this feature, bug, or test case. skipped Denotes a (flaky) test currently skipped in CI.
#103494 opened Jun 13, 2023 by pytorch-bot bot
3
is_broadcasted in triton.py is not correct module: inductor triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103491 opened Jun 13, 2023 by lezcano
@peterbell10
1
Expose Tensor.retains_grad property in Python API actionable module: autograd Related to torch.autograd, and the autograd engine in general 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
#103485 opened Jun 13, 2023 by ringohoffman
2
No backward implementation for torch._native_multi_head_attention triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103484 opened Jun 13, 2023 by lms-mt
1
torch._dynamo.exc.Unsupported: Tensor.backward with aten_graph=True oncall: pt2
#103483 opened Jun 13, 2023 by gengrill
Document CI retry rules module: devx Related to PyTorch contribution experience (HUD, pytorchbot) triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103482 opened Jun 13, 2023 by huydhn
[inductor] certain reductions cause triton compile error bug module: inductor oncall: pt2 topic: not user facing topic category upstream triton Upstream Triton Issue
#103481 opened Jun 13, 2023 by williamwen42
@williamwen42
1
[Inductor] Inference Fuse Gelu good first issue module: inductor oncall: pt2
#103480 opened Jun 13, 2023 by eellison
[Inductor] Optimize More Cases of Int32 -> Int64 module: inductor triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103475 opened Jun 13, 2023 by eellison
2
Error encountered when tracing model with Dynamo/Functorch for export with trilinear interpolation
#103473 opened Jun 13, 2023 by gs-olive
DISABLED test_aliased_param_return_cpu (__main__.CpuTests) skipped Denotes a (flaky) test currently skipped in CI.
#103466 opened Jun 12, 2023 by eellison
1
torch.round behaves inconsistently when running a traced model
#103465 opened Jun 12, 2023 by mirzadeh
Memory efficient SDP yields wrong gradients module: multi-headed-attention triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103462 opened Jun 12, 2023 by lengstrom
5
[inductor] UnboundLocalError: local variable 'buf0' referenced before assignment oncall: pt2
#103461 opened Jun 12, 2023 by anijain2305
@jansel
[Dynamo] InternalTorchDynamoError: __init__() got an unexpected keyword argument '_saved_tensors' module: dynamo oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103460 opened Jun 12, 2023 by yanboliang
@yanboliang @lyoka
1
torch.nn.functional.gumbel_softmax reporting NaNs in Pytorch 2.0 module: nn Related to torch.nn
#103459 opened Jun 12, 2023 by bryant0918
2
Asynchronous CUDA AveragedModel module: optimizer Related to torch.optim triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103449 opened Jun 12, 2023 by milesial
Deprecation warning on lr_scheduler.step(num_steps) module: optimizer Related to torch.optim triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#103444 opened Jun 12, 2023 by MicaelCarvalho
1
ProTip! Adding no:label will show everything without a label.