strisunshinewentingwang/pytorch

Pull Requests Commits

Clean up op BC check list (#103363)

houseroad committed 1 year ago

18f203a5

[dynamo] graph break on nn.Parameter construction (#103262)

davidberard98 committed 1 year ago

df83fe5b

Revert "Update torchbench pin - torchrec_dlrm moved to canary (#103383)"

pytorchmergebot committed 1 year ago

08f90b34

Revert "Log functional_collectives apis to distributed logger (#103288)"

pytorchmergebot committed 1 year ago

caecb552

Always create ShapeEnv, always apply unspec logic (#103302)

ezyang committed 1 year ago

c3fdfca5

[xla hash update] update the pinned xla hash (#103416)

pytorchupdatebot committed 1 year ago

f4228e70

Log functional_collectives apis to distributed logger (#103288)

wconstab committed 1 year ago

37359c36

[Quant][XNNPACK] Delegate add_relu fusion (#103266)

mcr229 committed 1 year ago

f37be778

Up to 48% speed up using Sklansky method for innermost prefix scan algorithm (#103314)

mfkasim1 committed 1 year ago

8a744c31

Handle nonzero via its meta registration (#103379)

ezyang committed 1 year ago

0863e550

Update torchbench pin - torchrec_dlrm moved to canary (#103383)

anijain2305 committed 1 year ago

114f99bb

Remove not dynamic_shapes case from wrap_listlike (#103301)

ezyang committed 1 year ago

03101a22

add multi swa support for custom device (#103297)

heidongxianhua committed 1 year ago

900226f2

[AOTAutograd] compare with stride hints (#103342)

shunting314 committed 1 year ago

daf75c07

[BE] Deprecate `has_XYZ` attributes (#103279)

malfet committed 1 year ago

4cfa06f7

[Profiler][Easy] Add log msg to assertEqual for flaky test_memory_timeline_no_id (#103326)

aaronenyeshi committed 1 year ago

0496d70a

Simplify has_unpack_var_sequence (#103324)

ezyang committed 1 year ago

919c567c

Inductor cpp wrapper: support ConvTranspose and fix Convolution ir (#103308)

chunyuan-w committed 1 year ago

d61cd03b

Remove config.dynamic_shapes test for tracing size calls (#103325)

ezyang committed 1 year ago

d67b676c

Make torch.compile(dynamic=True) not assume static by default (#99469)

ezyang committed 1 year ago

cf8af57c

[Docker] Update cc/c++ to point t clang/clang++ (#103350)

malfet committed 1 year ago

f474497c

[cpp-extensions] Add clang to the list of supported Linux compilers (#103349)

malfet committed 1 year ago

347463fd

[LibTorch] Fix `append_whole_archive` macro (#103348)

malfet committed 1 year ago

00e16179

[Inductor/cpp] Fix reduction on pre clang-10 (#103347)

malfet committed 1 year ago

5c252f2c

Turn off automatic_dynamic_shapes in prep for dynamic-by-default (#103320)

ezyang committed 1 year ago

414ec6ce

Add distributed category to TORCH_LOGS (#103351)

wconstab committed 1 year ago

a8549357

fix soundness bug with unsupported constraints (#102897)

avikchaudhuri committed 1 year ago

59ee6cd8

Rely on repeat meta reporting dynamic shapes (#103294)

ezyang committed 1 year ago

1b398297

Remove getitem dynamic shapes special case (#103296)

ezyang committed 1 year ago

1d40b394

Delete is_dynamic_shapes test (#103291)

ezyang committed 1 year ago

5987c520

Newer Older