strisunshinewentingwang/pytorch

Pull Requests Commits

Replace `GatherRangesToDense` operator in Dper from c2 to pt.

bugra committed 3 years ago

68a50a71

Refactor request_callback_no_python.cpp processRpc function (#47816)

H-Huang committed 3 years ago

55d5b273

Add basic ldexp operator for numpy compatibility (#45370)

Randall Hunt committed 3 years ago

562d4c3b

implement torch.addr using TensorIterator based kernels (#47664)

RockingJavaBean committed 3 years ago

ec256ab2

[TensorExpr] Add even more operator tests. (#48292)

Mikhail Zolotukhin committed 3 years ago

eb49dabe

[TensorExpr] Add more operator tests. (#48282)

Mikhail Zolotukhin committed 3 years ago

efd41db3

remove having no deadline for the test (#48226)

Hector Yuen committed 3 years ago

56129bde

[pytorch][codegen] add autograd data model (#48249)

ljk53 committed 3 years ago

de284b6d

[Pytorch] Weaker memory ordering for c10::intrusive_ptr (#48221)

swolchok committed 3 years ago

fa412758

[JIT] Resolve `torch.device` in recursive compilation of classes (#47734)

Meghan Lele committed 3 years ago

d6b37495

Add TorchBind-based Python and TorchScript binding for ProcessGroup (#47907)

gmagogsfm committed 3 years ago

28580d3c

fix a bug in leakyReLU (#48265)

Zhi-Zheng Wu committed 3 years ago

7828a220

[FX] Add Node.all_input_nodes (#48270)

James Reed committed 3 years ago

998c4cac

third_party: Update pybind to point to fork (#48117)

seemethere committed 3 years ago

aa8aa30a

enable cuda11.1 and cudnn 8.0.5 in CI (#48242)

mszhanyi committed 3 years ago

84d4e9c4

[Gradient Compression] Add a comment on _orthogonalize. (#48253)

Yi Wang committed 3 years ago

1a6666c9

Revert D25056091: migrate export_caffe2_op_to_c10.h macros to the new dispatcher registration API

bdhirsh committed 3 years ago

3c936ecd

migrate export_caffe2_op_to_c10.h macros to the new dispatcher registration API (#48097)

bdhirsh committed 3 years ago

0ea4982c

add kl_based_partition (#48197)

scottxu0730 committed 3 years ago

4b56aef0

Add MessageTypeFlags enum for RPC Messages (#48143)

H-Huang committed 3 years ago

c0723a0a

Dont skip NCCL backend when testing all_reduce_cuda (#48231)

rohan-varma committed 3 years ago

feb6487a

Refactor CuFFTConfig to not use tensor objects (#46909)

peterbell10 committed 3 years ago

685cd968

[Caffe2] Optimize MishOp on CPU (#48212)

xiaomengy committed 3 years ago

2039ff3f

[pytorch][codegen] move is_abstract property to NativeFunction model (#48252)

ljk53 committed 3 years ago

f98ab184

Fix collect_env.py with older version of PyTorch (#48076)

skyline75489 committed 3 years ago

9b19880c

Added linalg.tensorinv (#45969)

IvanYashchuk committed 3 years ago

343b3e5c

Add blas compare example (#47058)

Taylor Robie committed 3 years ago

678fe9f0

Implement in-place method torch.cumsum_ and torch.cumprod_ (#47651)

Kiyosora committed 3 years ago

008f840e

[PyTorch] Declare the instantiation of PackedConvWeightsQnnp<2>::prepack (#48256)

iseeyuan committed 3 years ago

fe6bb2d2

docker: Make CUDA_VERSION configurable (#48199)

seemethere committed 3 years ago

1dd4f433

Newer Older