strisunshinewentingwang/pytorch

Pull Requests Commits

Fix cuda/cpu check on NoneType (#88854)

Michael Gschwind committed 2 years ago

ee91c328

Error on ZeroTensor serialization (#88803)

kshitij12345 committed 2 years ago

d15a6b0c

[ONNX] Add test case for onnx::Max scalar type (#88751)

titaiwangms committed 2 years ago

b843f4db

Use `atomicAdd` for `bfloat16` in Ampere and above (#84981)

eqy committed 2 years ago

396c3b1d

[ONNX] Add onnx::Max into standard Op for scalar type alignment (#88750)

titaiwangms committed 2 years ago

a6d72f44

Revert "[dynamo] fixes dict changed during runtime error (#87526)"

pytorchmergebot committed 2 years ago

0de8f047

Update lr_scheduler.pyi to match lr_scheduler.py (#88818)

janeyx99 committed 2 years ago

310335de

Fix FakeTensorProp on Module with Parameters or Buffers (#88700)

wschin committed 2 years ago

86b7aa26

[FSDP][state_dict][1/N] Moving state_dict logic to pre_state_dict_hook (#87900)

fegin committed 2 years ago

c4fc5d37

Disable check for dropout in MultiheadAttention fast_path (#88831)

ecly committed 2 years ago

9d09968b

[vision hash update] update the pinned vision hash (#88853)

pytorchmergebot committed 2 years ago

30823787

Ref for aten.full; symint changes in prim (#88762)

SherlockNoMad committed 2 years ago

495e7b1c

Assert we have triton before scheduling on triton (#88849)

voznesenskym committed 2 years ago

3fbf748f

Add meta support for scalar_tensor and argmax (#88590)

anjali411 committed 2 years ago

fc9e36dd

handle zero dims in reductions (#88280)

Krovatkin committed 2 years ago

c961e45e

[primTorch] Implement group norm reference (#87054)

rdspring1 committed 2 years ago

534ae6ae

[ao] qconfig_mapping.py fixing public v private (#87518)

HDCharles committed 2 years ago

072834d5

[pytorch] Enable memory map file support for Android, Apple, and CXX (#88545)

Rittzz committed 2 years ago

f9221bf5

Revert "Add nondeterministic error for `scatter` (#88244)"

pytorchmergebot committed 2 years ago

84414431

[MPS] Fix `test_embedding_dense_backward` (#88847)

malfet committed 2 years ago

62ef15e3

[Dynamo] Add complete support for Tensor.is_contiguous (#88407)

yanboliang committed 2 years ago

b30222e0

Fix cupti search path in CMake (#88657)

dzhulgakov committed 2 years ago

ae01615d

Symbolic shape: sym_floor , sym_sqrt, sym_int (#88760)

SherlockNoMad committed 2 years ago

d9ad08ce

[Inductor] Fix lowmem_dropout() missing 1 required positional argument: 'p' (#88716)

yanboliang committed 2 years ago

cc04cf50

[ONNX] Create common ExportTestCase base class (#88145)

BowenBao committed 2 years ago

500fd655

[ONNX] Improve diagnostic message formatting (#87830)

BowenBao committed 2 years ago

20ae19aa

[ao] qconfig_mapping_utils.py fixing public v private (#87517)

HDCharles committed 2 years ago

a6610faa

Have kernel names include fused ops (#88624)

mlazos committed 2 years ago

c1553880

[ao] fuser_method_mappings.py fixing public v private (#87516)

HDCharles committed 2 years ago

ad2eba80

[xnnpack][lite-int][on-device] rebuild serialized modules at runtime (#88780)

mcr229 committed 2 years ago

37b468ac

Newer Older