strisunshinewentingwang/pytorch

Pull Requests Commits

Set test owners for module: ci (#66796)

janeyx99 committed 2 years ago

f4a7273b

[sharded_tensor] support gloo/mpi backend in tests (#65855)

wanchaol committed 2 years ago

8532061b

fx quant: enable linear-bn1d fusion for PTQ (#66484)

vkuzo committed 2 years ago

d549c8de

[fx2trt]Add support for negative dim in softmax (#66760)

842974287 committed 2 years ago

9d287d0b

[quant][embedding qat] Enable quint4 in EmbeddingBag QAT workflow (#66348)

b-koopman committed 2 years ago

aa7da7b0

Fix `nn.functional.max_poolNd` dispatch (for arg: `return_indices`) (#62544)

krshrimali committed 2 years ago

909694fd

Deduplicate codegenOutputQuery to query maximum CUDA compute capabilities (#55901)

Flamefire committed 2 years ago

e4a9ee8d

Adding StreamWrapper to ensure file object will be closed (#66715)

NivekT committed 2 years ago

811f5a2b

Add relative and absolute tolerances for matrix_rank, pinv (#63102)

IvanYashchuk committed 2 years ago

0d203a16

[PyTorch] Allow override for macro `HAS_DEMANGLE` (#66540)

larryliu0820 committed 2 years ago

53aac4b6

Revert D31577488: Migrate THCState to ATen

ngimel committed 2 years ago

3b4cb9dd

Revert D31547709: Remove THCGeneral.cpp

ngimel committed 2 years ago

719d43a2

Implement Python Array API `asarray` function. (#60627)

ysiraichi committed 2 years ago

8854817f

Make aotCompile support multiple input sizes (#66727)

priyaramani committed 2 years ago

9e3a2bab

Refactor: move method to func compilation work to compileMethod, add option to specify method name (#66726)

priyaramani committed 2 years ago

962c6476

Remove THCGeneral.cpp (#66391)

peterbell10 committed 2 years ago

aa0c3187

add frozen_numpy as a builtin library to torch::deploy (#66297)

shunting314 committed 2 years ago

8c5928bd

[TS] Return early if device doesn't match (#66694)

842974287 committed 2 years ago

42f13846

Suppress deprecated copy in vec256_qint.h (#66646)

r-barnes committed 2 years ago

32ac001e

Migrate THCState to ATen (#66480)

peterbell10 committed 2 years ago

65adf1df

Revert D30652629: use irange for loops

Xue Li committed 2 years ago

2f099c75

sort_out_cuda: Use custom kernels to fill index tensors (#66668)

peterbell10 committed 2 years ago

1e2b2ee5

Clean up test running scripts (#65508)

driazati committed 2 years ago

9ba39d20

[Vulkan] cat operator for channel dimension (#66669)

beback4u committed 2 years ago

2c761caa

Promote integral inputs to floating for `torch.logsumexp` (#63393)

Kiyosora committed 2 years ago

06cfdfae

[Static Runtime] Determine function for `ProcessedNode::run()` statically (#66692)

d1jang committed 2 years ago

67e003f0

Revert D31656999: Add meta support to tensor range factories

malfet committed 2 years ago

d1b61219

Add `warn_only` kwarg to `use_deterministic_algorithms` (#66233)

kurtamohler committed 2 years ago

a2564895

use irange for loops (#66234)

r-barnes committed 2 years ago

687c2267

EmbeddingBackward exclusive_scan thrust->cub (#66566)

zasdfgbnm committed 2 years ago

b5b7d6a3

Newer Older