strisunshinewentingwang/pytorch

Pull Requests Commits

[fx2trt] save and load TRTModule for OSS (#65958)

842974287 committed 2 years ago

60fe854f

Revert "Revert D31227448: [pytorch][PR] fixing sorting in stride indices" (#66176)

jjsjann123 committed 2 years ago

321345d7

[fx2trt] More controls over output dtypes (#65959)

842974287 committed 2 years ago

74477ba2

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

CodemodService FBSourceClangFormatLinterBot committed 2 years ago

227f91e7

[quant][embedding qat] Add basic EmbeddingBag QAT fakeQuant workflow (#65443)

b-koopman committed 2 years ago

a58ff186

[PyTorch Edge] Leave out field for debug_handle if not being built with eager symbolication support (#66131)

dhruvbird committed 2 years ago

64caee13

Periodic jobs should not have CIFLOW_DEFAULT label (#66300)

malfet committed 2 years ago

ebe530a9

TBB: Use static partitioner to match OpenMP scheduling (#65327)

peterbell10 committed 2 years ago

bd9eee4e

Parallel: Deduplicate parallel functions in different backends (#65326)

malfet committed 2 years ago

d5033410

[BE] Cleanup python_function.cpp (#66296)

malfet committed 2 years ago

e1817d89

docker: Ensure libgnutls30 for all docker builds (#66258)

seemethere committed 2 years ago

ca363d1e

Fix https://github.com/pytorch/pytorch/issues/61982 (#66015)

rohan-varma committed 2 years ago

38f5144e

Rename cuda/Resize.cu to cuda/Resize.cpp (#65943)

peterbell10 committed 2 years ago

20f2e55d

Upgrade to ubuntu:trusty-20190515 (#63468)

ashishsolankiii committed 2 years ago

86de09e4

[Static Runtime] Group graph nodes into input aliases & output aliases (#65517)

d1jang committed 2 years ago

416f5930

[Bootcamp][Pytorch Core] Add testing for complex non-vanilla SGD

mikaylagawarecki committed 2 years ago

0e2d1b22

Support Registering a Variable Length List of Builtin Modules for torch::deploy Builtin Libraries (#66021)

shunting314 committed 2 years ago

5e7d8ec8

[Static Runtime] Cleanup LLVMCodeGen memory after code gen completes (#66218)

navahgar committed 2 years ago

40dd2711

[nnc] Added a cache to use singleton instances of PytorchLLVMJIT for every triple,cpu,attrs combination (#66217)

navahgar committed 2 years ago

7e5ef5e5

[nnc] Use given kernel function name while emitting code (#66216)

navahgar committed 2 years ago

c30dc527

add gather to ShardedTensor (#65671)

dracifer committed 2 years ago

3cc40253

OpInfo for 2d fft functions (#66128)

peterbell10 committed 2 years ago

f445ed19

C++ API and docs for hfftn (#66127)

peterbell10 committed 2 years ago

2213c463

slow_conv3d: Use at::sum for grad_bias accumulation (#65758)

peterbell10 committed 2 years ago

e6a4f746

Add inplace_variant for resize_ OpInfo (#66135)

IvanYashchuk committed 2 years ago

2e4e5b02

Chunk: acc_ops (#66010)

FusionPower committed 2 years ago

361b34eb

Update `torch.fx.passes.split_module` docstring (#65542)

Patrick Spencer committed 2 years ago

9fb6ba24

[Static Runtime] Support aten::to.prim_dtype overload (#64928)

Mike Iovine committed 2 years ago

d5f64afc

[pytorch][PR] Add hash and int128 utils for Lazy Tensor Core" (#66181)

wconstab committed 2 years ago

a8c0b362

[Part 1] upstreaming fairscale fsdp to PyTorch -- sharding, core data flow and hooks (#63881)

zhaojuanmao committed 2 years ago

61fca037

Newer Older