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

14,070 Open 31,897 Closed
14,070 Open 31,897 Closed
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
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

When I use register_full_backward_hook and register_forward_hook in nn.MultiheadAttention, The MultiheadAttention executed wrong conditional branch and get wrong backward grads.
#139191 opened Oct 29, 2024 by peter-ni-noob
[MPS] Typo in error message for supported autocast type module: error checking Bugs related to incorrect/lacking error checking 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
#139190 opened Oct 29, 2024 by hvaara
1
2
PCH build fail with sccache-v0.8.2 module: ci Related to continuous integration module: third_party triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#139188 opened Oct 29, 2024 by malfet
1
Operators being traced as method calls in torch.fx
#139185 opened Oct 29, 2024 by acairncross
compile fails when split is called within a device cm module: dynamo module: graph breaks module: __torch_function__ oncall: pt2
#139183 opened Oct 29, 2024 by vmoens
[ROCm] [Upstream Triton] num_stages=0 deprecation with stream pipeliner v2 module: rocm AMD GPU support for Pytorch
#139182 opened Oct 29, 2024 by jataylo
RuntimeError: Storage size calculation overflowed with sizes=[1, 4605674770382112385, 4128, 128]
#139181 opened Oct 29, 2024 by DaozeZhang
[Upstream Triton] attrs_description KeyError: 'cls'
#139179 opened Oct 29, 2024 by jataylo
1
@alexbaden
2
the mpi backend not support to transfer float16/bfloat16 tensor and would post "IndexError: map::at" when transferring float16/bfloat16 tensor
#139178 opened Oct 29, 2024 by Panda-Hero
[v2.6] Inductor issue tracker for Triton release/3.2
#139175 opened Oct 29, 2024 by jataylo
2
pkg_resources module is deprecated by setuptools
#139170 opened Oct 29, 2024 by Harry-Chen
1
1
Deepwise convolution with torch.compile has inconsistent numerical precision with eager mode
#139169 opened Oct 29, 2024 by Thinksky5124
[Dynamo] TypeError: list object is not callable oncall: pt2
#139167 opened Oct 29, 2024 by mieshkiwrk
SubgraphMatcher may fail to match when the matching pattern having call_module IR
#139163 opened Oct 29, 2024 by Maizesty
[ERROR]: fullgraph compilation failed: 'FakeRootModule' object has no attribute 'self___attr_list_0'
#139160 opened Oct 29, 2024 by hyukn
1
Dynamo capture of tensor.data assignment doesn't identical to eager call of tensor.data assignment
#139152 opened Oct 29, 2024 by jerrychenhf
1
DISABLED TCPStoreTest.testMultiTenantStoresUV (__main__.TCPStoreTest) module: flaky-tests Problem is a flaky test in CI module: rocm AMD GPU support for Pytorch skipped Denotes a (flaky) test currently skipped in CI. triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#139150 opened Oct 29, 2024 by pytorch-bot bot
1
DISABLED test_like_channels_last_cpu (__main__.CpuTritonTests) module: flaky-tests Problem is a flaky test in CI module: inductor oncall: pt2 skipped Denotes a (flaky) test currently skipped in CI.
#139149 opened Oct 29, 2024 by pytorch-bot bot
1
Some files in sccache are owned by hostmaster+pytorch module: ci Related to continuous integration module: infra Relates to CI infrastructure triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#139143 opened Oct 29, 2024 by malfet
@jeanschmidt
3
torch.nn.InstanceNorm2d throws "mixed dtype" error with track_running_stats set to True good first issue module: cpu CPU specific problem (e.g., perf, algorithm) module: norms and normalization triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#139140 opened Oct 29, 2024 by jiren-the-gray
@CaoE
1
DISABLED test_manual_with_data_parallel_dp_type_DDP_ScheduleClass0_use_new_runtime_False (__main__.ComposabilityTest) module: flaky-tests Problem is a flaky test in CI oncall: distributed Add this issue/PR to distributed oncall triage queue skipped Denotes a (flaky) test currently skipped in CI.
#139139 opened Oct 29, 2024 by pytorch-bot bot
1
[FlexAttention] Update the way we generate kernel options module: flex attention module: higher order operators torch.cond and similar module: inductor module: pt2-dispatcher PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op, oncall: pt2
#139131 opened Oct 28, 2024 by drisspg
[pgnccl] unstable restarting of PGs module: c10d Issues/PRs related to collective communications and process groups oncall: distributed Add this issue/PR to distributed oncall triage queue
#139129 opened Oct 28, 2024 by shuqiangzhang
@shuqiangzhang
torchao.float8 + torch.compile does not work on HuggingFace's Mixtral model oncall: pt2
#139125 opened Oct 28, 2024 by vkuzo
shift operators not supporting integral types module: unsigned int Related to the new uint16, uint32, uint64 types triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#139124 opened Oct 28, 2024 by lshamis
@malfet
1
ProTip! Adding no:label will show everything without a label.