strisunshinewentingwang/pytorch

Pull Requests Commits

Refactored proxytensor to clean up separate branches (#84325)

Chillee committed 2 years ago

a27a4a02

remove data-dependent shapes from some distributions (#84322)

Chillee committed 2 years ago

8843f5b9

Started storing faketensor/symbolic shape metadata on FX nodes in make_fx (#84114)

Chillee committed 2 years ago

6a3ecda5

[BE] Remove unused `export.h` include (#84305)

malfet committed 2 years ago

79e3a39f

ci: Replace setup-miniconda with test-infra version (#84236)

seemethere committed 2 years ago

abaf8112

[torchdynamo hash update] update the pinned torchdynamo hash (#84317)

pytorchmergebot committed 2 years ago

b343febe

Use register_meta for everything in meta_registrations (#84297)

ezyang committed 2 years ago

8cd296f6

add nvidia-smi to run_torchbench (#83857)

davidberard98 committed 2 years ago

71d99662

Use reentrant mode when invoking prims, delete global prim_fake_mode (#84090)

eellison committed 2 years ago

9c452abc

[Static Runtime] Schema checks for index_put (#84152)

Mike Iovine committed 2 years ago

db7784e7

[Modes] remove inner constructor kwarg (#83925)

samdow committed 2 years ago

7532d5b1

[PyTorch][caffe2] Add CAFFE2_{DECLARE,DEFINE}_KNOWN_TYPE (#83707)

swolchok committed 2 years ago

e23d159b

no ios arm builds on circleci (#84299)

clee2000 committed 2 years ago

af741e82

Upgrade default cuda version of torchbench (#84248)

xuzhao9 committed 2 years ago

e014bd8e

Don't start land checks if the PR hasn't been approved yet (#84239)

huydhn committed 2 years ago

7acdb2d5

[Quant] Remove warnings from using torch.tensor(value) (#84277)

jcaip committed 2 years ago

eabe34cc

Reland symint_numel (#84281)

Krovatkin committed 2 years ago

eda217ab

[ROCm] enable nvfuser (#82498)

jeffdaily committed 2 years ago

d09486ab

Fix several typos (#83823)

OccupyMars2025 committed 2 years ago

f9609d82

[Profiler] Unify global and thread local profiler lookup. (#83894)

Taylor Robie committed 2 years ago

c06a5586

[Profiler][Trivial] Create orchestration folder and move observer management there. (#83893)

Taylor Robie committed 2 years ago

48a596ad

[Profiler] Encapsulate callback handle management. (#83892)

Taylor Robie committed 2 years ago

c26b53f6

Removing multigpu 10.2 . Using 11.6 cuda for multigpu tests instead (#84286)

atalman committed 2 years ago

ddd841b3

Revert "Test distributed backends in parallel (#84034)"

pytorchmergebot committed 2 years ago

772721a4

modify split_by_tags to retain output order (#84136)

ihoffman1028 committed 2 years ago

20018aa7

Add nvfuser support for squeeze (#84117)

IvanYashchuk committed 2 years ago

90161c23

[Nested Tensor]Clean up offsets (#84145)

drisspg committed 2 years ago

174c3c68

Test distributed backends in parallel (#84034)

huydhn committed 2 years ago

3ae5be74

[ONNX] refactor test_pytorch_onnx_onnxruntime_cuda.py (#84218)

titaiwangms committed 2 years ago

641c3952

[Easy][FSDP] Update `StateDictType` doc (#84200)

awgu committed 2 years ago

b8ee8101

Newer Older