strisunshinewentingwang/pytorch

Pull Requests Commits

[c10d] Fix async error in batch_isend_irecv (#82450)

aashaka committed 2 years ago

24a084ed

properly compute batch_element_count (#82927)

ngimel committed 2 years ago

88e43ca4

Merge torchdim into functorch build (#82454)

zdevito committed 2 years ago

d7cebab7

Add a common wrapper for make_fx to handle args and kwargs (#82965)

IvanYashchuk committed 2 years ago

ea391465

add the operations needed for electra model (#82856)

migeed-z committed 2 years ago

0f4f79fb

[TorchTidy] Adding support for storing scalar values in profiling (#81843)

Gamrix committed 2 years ago

1cd1f48d

[composite compliance] eig_backward (#82957)

kshitij12345 committed 2 years ago

10e7a254

[composite compliance] quantile and nanquantile (#81767)

kshitij12345 committed 2 years ago

a3d37f11

Update nan_to_num prims (#82908)

eellison committed 2 years ago

4f61aa79

[functorch] fix unsqueeze_ batching rule (#82899)

zou3519 committed 2 years ago

c3f3b86f

[functorch] Add vmap in-place testing (#82898)

zou3519 committed 2 years ago

62e095ab

[functorch] refactor get_fallback_and_vmap_exhaustive (#82897)

zou3519 committed 2 years ago

7c993d7f

[functorch] eliminate erroneous warning on `import functorch` (#82905)

zou3519 committed 2 years ago

d5eea7f1

[Profiler] Make KinetoEvent a view of Result (Part 4 (final), stragglers) (#81322)

Taylor Robie committed 2 years ago

73ddd412

[Profiler] Make KinetoEvent a view of Result (Part 3, forwarded from `result_`) (#81321)

Taylor Robie committed 2 years ago

c8db7102

[Profiler] Make KinetoEvent a view of Result (Part 2, python and stacks) (#81320)

Taylor Robie committed 2 years ago

7a726a4f

[Profiler] Make KinetoEvent a view of Result (Part 1: trivial fields) (#81319)

Taylor Robie committed 2 years ago

63873ab7

[quant] Remove unneeded lines from APoT linear (#82909)

asl3 committed 2 years ago

b91ff5e3

[torchdynamo hash update] update the pinned torchdynamo hash (#82959)

pytorchmergebot committed 2 years ago

780a303a

Properly compare floats in gemm-block-sparse-microkernel-tester (#82947)

kit1980 committed 2 years ago

4d3f7df3

[Refactoring] making the code more Pythonic (#82929)

se4ml committed 2 years ago

51d12548

[MPS] And native bitwise_[and|or|xor] (#82307)

malfet committed 2 years ago

0377615b

Fix possible out-of-range error of 'as_nd' util function (#82393)

Xia-Weiwen committed 2 years ago

18db7f36

Remove flatbuffer types/headers from flatbuffer_loader.h (#82893)

dbort committed 2 years ago

1d56ea5e

Revert "Add DLPack support for XPU backend by mapping to kDLOneAPI in DLPack (#82867)"

pytorchmergebot committed 2 years ago

0c7ca2d9

Revert "[ONNX] Type utility for converting among JIT, torch and ONNX data types (#81953)"

pytorchmergebot committed 2 years ago

b170a52a

Fix FSDP device_id when CPU offloading (#82892)

rohan-varma committed 2 years ago

ab3c0399

[torchdynamo hash update] update the pinned torchdynamo hash (#82943)

pytorchmergebot committed 2 years ago

73889130

Add EnableTorchFunction (#82647)

ezyang committed 2 years ago

a61c9649

Improved legalize_graph pass in FX (#82874)

Chillee committed 2 years ago

51bbf632

Newer Older