strisunshinewentingwang/pytorch

Pull Requests Commits

Enable Python bindings for UntypedStorage (#68945)

cbalioglu committed 2 years ago

80b19c4c

Additional unit test for sharded linear. (#70476)

pritamdamania committed 2 years ago

f5b19ba6

Add GitHub merge rules (#71514)

malfet committed 2 years ago

a5d5b112

[PyTorch] Fix a bunch of structured kernel refcounting (#71140)

swolchok committed 2 years ago

c59942ac

[flatbuffer] Fix forward flatbuffer type handling with dynamic type. (#71500)

zhxchen17 committed 2 years ago

b98e955b

[Pytorch] Speed up LayerNorm 4-5% (#71423)

swolchok committed 2 years ago

565f78f5

[PyTorch][Static Runtime] Fix extra refcount bumps in layer_norm (#71237)

swolchok committed 2 years ago

958f9cf5

Fix trivial typo at the doc of `torch.lobpcg` (#71464)

nonconvexopt committed 2 years ago

811af259

Update min python version to 3.7 in setup.py and mypy configs (#71494)

malfet committed 2 years ago

dc5cda0c

[acc_ops] Remove usage of kwarg expansion via **locals() for jit scripting support (#71425)

jfix71 committed 2 years ago

06bc6748

[distributed] Make rref_proxy._invoke_rpc trully async when needed. (#70206)

Rodrigo Kumpera committed 2 years ago

ef4bc3fa

[nnc] Update block and thread extents in cuda_codegen to use int64_t (#71428)

navahgar committed 2 years ago

70c9146c

[fx2trt] Issue warnings instead of error if there's possible const folding opportunities (#71031)

842974287 committed 2 years ago

2dbbb1a9

Add trymerge workflow (#71488)

malfet committed 2 years ago

61713acb

Consolidate the overloads of TensorImpl::shallow_copy_and_detach (#68953)

cbalioglu committed 2 years ago

f45e217c

test //c10/... without Google libraries in OSS (#70853)

Michael Dagitses committed 2 years ago

805b7575

port //c10/macros to common build structure (#70852)

Michael Dagitses committed 2 years ago

78e1f9db

use c10/macros/cmake_macros.h in fbcode build (#70851)

Michael Dagitses committed 2 years ago

661d10aa

[fx] add documentation to AccOpProperties (#71450)

alexbeloi committed 2 years ago

bdeec0c7

add rocm GHA workflow (#68552)

jeffdaily committed 2 years ago

7ce6db48

[jit][edge] Create convinience wrapper for dynamic type construcytors. (#71457)

zhxchen17 committed 2 years ago

15e7d181

Allow disabling nvfuser without CUDA (#71358)

jansel committed 2 years ago

ac26f823

Support sparse.sum on empty sparse tensor (#71091)

pearu committed 2 years ago

214f4bf2

[DDP Checkpointing] non-reentrant checkpoint tests (#69060)

rohan-varma committed 2 years ago

3b589c34

Remove simd qualifier for pragma omp loop in upsample_nearest_op.h (#71462)

r-barnes committed 2 years ago

75aaa9f9

[fix] composite compliance: quantile and nanquantile (#70894)

kshitij12345 committed 2 years ago

908fd3d7

Back out "[pytorch][PR] Performance and memory improvements to batched torch.linalg.solve" (#71421)

Mike Ruberry committed 2 years ago

a0ada2d2

Lazy load `pandas` when importing pytorch (#71316)

malfet committed 2 years ago

8a924399

Move sccache compilation log to its own group (#71444)

janeyx99 committed 2 years ago

671a0b53

Adding wheels with py3.10 (#71419)

atalman committed 2 years ago

7ed2a43d

Newer Older