-
Notifications
You must be signed in to change notification settings - Fork 22.1k
Pull requests: pytorch/pytorch
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[PyTorch] Remove unnecessary include of c10/util/Exception.h in irange.h
fb-exported
topic: not user facing
topic category
#136202
opened Sep 17, 2024 by
swolchok
Loading…
[MPS] Fix 5D+ reductions over negative dimentions
ciflow/mps
Run MPS tests (subset of trunk)
release notes: mps
Release notes category
topic: bug fixes
topic category
#136198
opened Sep 17, 2024 by
malfet
Loading…
Increase number of shards for linux aarch64 jobs
ciflow/linux-aarch64
linux aarch64 CI workflow
topic: not user facing
topic category
#136195
opened Sep 17, 2024 by
atalman
Loading…
Make avoid_device_init() aware of hpu device
open source
topic: not user facing
topic category
#136194
opened Sep 17, 2024 by
australopitek
Loading…
[inductor] Pass release notes category
device_type argument to do_bench
module: inductor
open source
release notes: releng
#136189
opened Sep 17, 2024 by
galexite
Loading…
[QAT] Make Fused modules torchscriptable
fb-exported
release notes: AO frontend
#136188
opened Sep 17, 2024 by
Ninja91
Loading…
[Dynamo][autograd.Function] Use fake tensor prop to infer fwd output
ciflow/inductor
module: dynamo
topic: not user facing
topic category
#136184
opened Sep 17, 2024 by
yanboliang
Loading…
fix a *single* missing quote in named_tensor.rst
open source
#136183
opened Sep 17, 2024 by
yoinked-h
Loading…
[AOTI][Tooling][8/n] Add option to pinpoint kernel names in debug printer
ciflow/inductor
fb-exported
module: inductor
#136182
opened Sep 17, 2024 by
YUNQIUGUO
Loading…
[cuDNN][conv][A100] Bump tolerances for Related to torch.cuda, and CUDA support in general
module: cudnn
Related to torch.backends.cudnn, and CuDNN support
module: functorch
Pertaining to torch.func or pytorch/functorch
open source
topic: not user facing
topic category
vmap_autograd_grad conv2d on A100
module: cuda
#136178
opened Sep 16, 2024 by
eqy
Loading…
add basic_modules_ListOfLinears_inductor_gpu_force_shape_pad
ciflow/inductor
module: dynamo
topic: not user facing
topic category
#136175
opened Sep 16, 2024 by
laithsakka
Loading…
[PyTorch CUDA Allocator] Allow reuse of non-split blocks with better rounding
fb-exported
release notes: cuda
release notes category
#136174
opened Sep 16, 2024 by
banitag1
Loading…
Make AOTAutogradCache support remote FXGraphCache
ciflow/inductor
fb-exported
module: inductor
release notes: AO frontend
#136173
opened Sep 16, 2024 by
jamesjwu
Loading…
[does not work] add partitioner
ciflow/inductor
module: dynamo
#136172
opened Sep 16, 2024 by
anijain2305
Loading…
[prototype] Invoke subgraph higher order op
ciflow/inductor
module: dynamo
#136171
opened Sep 16, 2024 by
anijain2305
Loading…
Update PyTorch/XLA pin.
ciflow/inductor
open source
topic: not user facing
topic category
#136170
opened Sep 16, 2024 by
ysiraichi
Loading…
[inductor] Add a OperatorBench to benchmark custom operations
ciflow/inductor
module: dynamo
release notes: inductor
#136169
opened Sep 16, 2024 by
FindHao
Loading…
Remove unused video_image targets
fb-exported
topic: not user facing
topic category
#136167
opened Sep 16, 2024 by
scotts
Loading…
fix a bug with empty split size in torch.split
fb-exported
#136166
opened Sep 16, 2024 by
bearzx
Loading…
Fix fast_expand recursion error
ciflow/inductor
open source
release notes: fx
release notes category
#136163
opened Sep 16, 2024 by
isuruf
Loading…
Fix ROCm skip decorator for test_ddp_tp and multiprocess UTs
ciflow/rocm
module: rocm
AMD GPU support for Pytorch
oncall: distributed
Add this issue/PR to distributed oncall triage queue
open source
topic: not user facing
topic category
#136161
opened Sep 16, 2024 by
pragupta
Loading…
[c10d] Make test compatible for new pytest
release notes: distributed (c10d)
release notes category
#136158
opened Sep 16, 2024 by
fduwjj
Loading…
Previous Next
ProTip!
Exclude everything labeled
bug with -label:bug.