strisunshinewentingwang/pytorch

Pull Requests Commits

[PyTorch Edge] Internal Optimized Quantized Matmul (#73244)

salilsdesai committed 2 years ago

c6f991e7

[Profiler] Prefer TSC to wall clock when available (#73855)

Taylor Robie committed 2 years ago

0b1f3bd1

Remove trailing semicolon. (#74031)

xkszltl committed 2 years ago

198d727d

Revert D34803275: [Quant][core][gpu][improvement] Refactored implementation for conv2d_cudnn to use packed parameters

suo committed 2 years ago

22b87678

Revert D34641680: [Quant][core][refactorization] Refactored qconv_unpack.cpp into an implementation file and higher level call registration and definition file

suo committed 2 years ago

dab5659d

[Quant][core][refactorization] Refactored qconv_unpack.cpp into an implementation file and higher level call registration and definition file (#73773)

dzdang committed 2 years ago

a1e9a0cb

[Quant][core][gpu][improvement] Refactored implementation for conv2d_cudnn to use packed parameters (#73510)

dzdang committed 2 years ago

85c790ed

[AutoAccept][Codemod][FBSourceClangFormatLinter] Daily `arc lint --take CLANGFORMAT`

CodemodService FBSourceClangFormatLinterBot committed 2 years ago

ef9023e9

[BC-Breaking] Remove redundant fsdp prefix (#73791)

mrshenli committed 2 years ago

247e4cca

[fx][acc_tracer] fix defaulted placeholder normalization (#73406)

alexbeloi committed 2 years ago

74a3b9c6

[BE] Add optional step to upload coredumps

malfet committed 2 years ago

bbdb7584

[PyTorch Distributed] Update documentation about NCCL environment variables (#74006)

kwen2501 committed 2 years ago

1f04a00c

[ONNX] Remove dangling print in repeat_interleave

shubhambhokare1 committed 2 years ago

56aa1ab0

[JIT] add keep_unique_names arg to canonicalize python bindings (#74074)

davidberard98 committed 2 years ago

b5244b84

Revert D33716039: [pytorch][PR] Add ONEDNN quantization backend

jerryzh168 committed 2 years ago

5a897536

[PyTorch] Move get_nested_tensor_impl to header (#73998)

swolchok committed 2 years ago

44997efd

[PyTorch] Add stub NestedTensor_is_contiguous function (#73997)

swolchok committed 2 years ago

34247626

Add ONEDNN quantization backend (#69820)

Xia-Weiwen committed 2 years ago

989b2485

regenerate flatbuffer header (#73810)

qihqi committed 2 years ago

3e556efc

[Profiler] Specialized AppendOnlyQueue (#73409)

Taylor Robie committed 2 years ago

5a58820f

[jiterator] sigmoid : complex dtypes (#73643)

kshitij12345 committed 2 years ago

66356130

[quant][fx] Fully align convert with the reference model design and simplify the implementation (#73863)

jerryzh168 committed 2 years ago

7ddf212f

Automated submodule update: FBGEMM (#74088)

facebook-github-bot committed 2 years ago

7070fe4d

linalg_solve_triangular should not be a method

lezcano committed 2 years ago

8d5cdb55

Revert D34604068: [PyTorch] [Model Tracer] Use c10::Synchronized<T> for kernel dtype tracer

suo committed 2 years ago

f12703cb

Revert D34604067: [PyTorch] [Model Tracer] Use c10::Synchronized<T> for custom classes tracer

suo committed 2 years ago

689cd224

Revert D34604066: [PyTorch] [Model Tracer] Use c10::Synchronized<T> for build features tracer

suo committed 2 years ago

747c6fdd

Revert D34645509: [PyTorch] Update Synchronized<T>::withLock() to return the type/value from the aceepted callable

suo committed 2 years ago

a554fc68

Revert D34645508: [PyTorch] Use c10::Synchronized<T> in RWSafeLeftRightWrapper

suo committed 2 years ago

b65adf84

fix: nn.Module allowing for expected Mixin MRO

MattiaSarti committed 2 years ago

1ac519e6

Newer Older