strisunshinewentingwang/pytorch

Pull Requests Commits

Add docstring for is_inference_mode_enabled (#59047)

soulitzer committed 3 years ago

9e60c7de

BFloat16 support for torch.sort (#58196)

zasdfgbnm committed 3 years ago

1bd22e28

fix unique for discontiguous inputs (#59003)

ngimel committed 3 years ago

f4a890d7

CUDA support in the CSR layout: constructors (#59010)

aocsa committed 3 years ago

b435a27f

[fx2trt] Quantized uru on gpu (#58965)

842974287 committed 3 years ago

7c17e1dd

fix nn.MHA scriptability (#58727)

Basil Hosmer committed 3 years ago

58d1b363

[PyTorch] Rename TI::add_borrowed_{in,out}put to TI::add_{in,out}put (#58608)

swolchok committed 3 years ago

ac67cda2

[PyTorch] Add temporary guardrail to borrowing_ op variants on TensorIterator (#58607)

swolchok committed 3 years ago

7db36c07

[PyTorch] Add TI::add_owned_{in,out}put for clarity & use them (#58606)

swolchok committed 3 years ago

bed0eb54

Document factory_kwargs in nn.Quantize + remove Attributes section (#59025)

jbschlosser committed 3 years ago

4f390eb6

Add UninitializedBuffer to nn docs (#59021)

jbschlosser committed 3 years ago

a749e8ed

[PyTorch] Replace RecordFunction shouldRun callback with atomic bools (#56504)

swolchok committed 3 years ago

de22657e

[nnc] Use BufHandle in loopnest.cache_accesses python API (#59006)

huiguoo committed 3 years ago

ac07c645

concurrency fixes (#58961)

Krovatkin committed 3 years ago

b93e7a76

Revert D28549240: [typing] Pyre fixes for batch_distributed_inference

Rong Rong (AI Infra) committed 3 years ago

97c1179c

Back out "Back out "[ONNX] Process const folding progressively when converts to ONNX (#54569)"" (#58923)

Meghan Lele committed 3 years ago

0d5527de

ShardedTensor framework for ChunkedShardingSpec (#58517)

pritamdamania committed 3 years ago

b420ded6

[typing] Pyre fixes for batch_distributed_inference

Shivansh Dhar committed 3 years ago

671c224b

[remove xla from codegen] rename aten_xla_type.h -> DispatchKeyNativeFunctions.h (#58568)

bdhirsh committed 3 years ago

be47060a

remove xla-specific stuff from codegen (minus CPU fallback) (#58064)

bdhirsh committed 3 years ago

86ce2950

Define the SYCL device version __assert_fail when the NDEBUG defined. (#58906)

chengjunlu committed 3 years ago

511979df

[structured] remainder (#58732)

kshitij12345 committed 3 years ago

2e2a7572

Add 11.3 binaries without conda (#58877)

janeyx99 committed 3 years ago

29487ac7

Revert D28643215: Adds an aten::_ops namespace with unambiguous function names

Alice Ou committed 3 years ago

24508337

Back out "Revert D28664514: [pytorch][PR] various TensorIterator speed improvements"

ngimel committed 3 years ago

12418a4f

Make torch.Tensor(torch.tensor(1.0)) work (#58885)

ezyang committed 3 years ago

17fb651a

[nnc] Concat input shapes must be known to fuse (#58974)

bertmaher committed 3 years ago

e2436274

Revert D28664514: [pytorch][PR] various TensorIterator speed improvements

ngimel committed 3 years ago

8398ebaa

[caffe2] Add support for int32 lengths in BatchSparseToDense (#58062)

qxy11 committed 3 years ago

c06d2afa

Use docker base for clang-lint in CI (#58964)

zhouzhuojie committed 3 years ago

444e195b

Newer Older