strisunshinewentingwang/pytorch

Pull Requests Commits

[TensorExpr] Nuke TE shape inference. (#65554)

Mikhail Zolotukhin committed 2 years ago

24b9b304

[Pytorch Edge] Improve bundled inputs name error handling (#65856)

JacobSzwejbka committed 2 years ago

18e46881

Revert D31386275: Migrate THCState to ATen

suo committed 2 years ago

2d155282

[PyTorch] Split c10 Type.cpp into two files to allow targets to include one of them (#66445)

larryliu0820 committed 2 years ago

d8532e35

[deploy] fix oss build (#66347)

suo committed 2 years ago

07ec250f

Update test to handle NaN compare after NLL change

wconstab committed 2 years ago

6338decd

Move softmax ops to codegen, following merge conflict

wconstab committed 2 years ago

e1711558

Update alias_info handling in EagerFallback

wconstab committed 2 years ago

c6894f9f

Fix batch_isend_irecv tests for err case (#63112)

zasdfgbnm committed 2 years ago

9a85167d

[FX] Fix issue where GraphModule.delete_all_unused_submodules deletes submodules from called leaf modules (#66430)

James Reed committed 2 years ago

3eb94436

Migrate THCState to ATen (#65948)

peterbell10 committed 2 years ago

a6774d6e

Call `PyArray_Check` only if NumPy is available (#66433)

malfet committed 2 years ago

e7b5712c

Quantization docs: add pages for Numeric Suite (Eager and FX) (#66380)

vkuzo committed 2 years ago

565cf47a

Improve quantization API docs (#66379)

vkuzo committed 2 years ago

8b125869

Merge branch 'master' into wconstab/lazy_tensor_staging

wconstab committed 2 years ago

263ef2d5

Fix type checking errors in torch/quantization/fx/qconfig_utils.py (#66428)

EshikaShah committed 2 years ago

88ed93c2

Back out "Revert D31495086: open source engine_layer_visualize.py" (#66431)

jerryzh168 committed 2 years ago

25965619

Do not enforce unused vars rule for torch_deploy (#66447)

malfet committed 2 years ago

ae5a9a45

Chunk: Converter (#66028)

FusionPower committed 2 years ago

7baf4f6b

[NNC] Normalize loops in SplitWithTail (#66242)

anijain2305 committed 2 years ago

cc24e4e5

[RFC] Reduce logging noise from AdagradOptimizer (#66443)

atuljangra committed 2 years ago

49f16053

[torchelastic] Fix failing tests (#66440)

aivanou committed 2 years ago

c03f8517

[TensorExpr] Adding missing python binding for operators (#66336)

anijain2305 committed 2 years ago

1d14fbda

[LTC] Code-gen native_layer_norm_backward

Jiewen Tan committed 2 years ago

7feb787a

[LTC] Supported optional operands in the code-gen and generate native_layer_norm

Jiewen Tan committed 2 years ago

1c7ca78a

[LTC] Add support for code-gen operators that have tuple output

Jiewen Tan committed 2 years ago

635b9667

[LTC] Move at::dtype and at::shape from leaf IR nodes to the base Node

Jiewen Tan committed 2 years ago

726c3654

[LTC] Code-gen layer_norm

Jiewen Tan committed 2 years ago

d9ef926a

Wextra fix for Integration.cpp (#66321)

r-barnes committed 2 years ago

08fab7ae

[PyTorch][JIT] Return a reference from caching specializations of getTypePtr (#66342)

swolchok committed 2 years ago

8c468ce0

Newer Older