strisunshinewentingwang/pytorch

Pull Requests Commits

Revert D24667127: [c10d] switch ProcessGroupNCCL:Options to be managed by intrusive_ptr

wanchaol committed 4 years ago

2204374f

Convert from higher order functions to classes in tools.codegen.gen (#47008)

ezyang committed 4 years ago

0c64f9f5

Fix classmethod override argument passing. (#47114)

hameerabbasi committed 4 years ago

d478605d

[quant][graphmode][fx] Support standalone_module_class (#47705)

jerryzh168 committed 4 years ago

1239d067

Allow for string literal return during symbolic tracing (#47618)

Ansley Ussery committed 4 years ago

4cb73f5a

Stop including TypeDefault.h from MPSCNNTests.mm (#46998)

ezyang committed 4 years ago

48ed577f

[fbcode][pytorch mobile] Create model reader utilities.

xcheng16 committed 4 years ago

88ec72e1

Revert D24859919: [pytorch][PR] Grammatically updated the tech docs

ejguan committed 4 years ago

5647f0ca

[c10d] switch ProcessGroupNCCL:Options to be managed by intrusive_ptr (#47075)

wanchaol committed 4 years ago

ae5c2feb

[c10d] switch Store to be managed by intrusive_ptr (#47074)

wanchaol committed 4 years ago

0cfe3451

[c10d] switch ProcessGroup::Work to be managed by intrusive_ptr (#44046)

wanchaol committed 4 years ago

0650a616

Unbreak backward compatibility tests (#47726)

Natalia Gimelshein committed 4 years ago

cbf439ca

[vulkan] Apply new changes to vulkan api v1 (#47721)

IvanKobzarev committed 4 years ago

bfec376e

Use local env for building CUDA extensions on Windows (#47150)

peterjc123 committed 4 years ago

d73a8db2

Fix output type of torch.max for Tensor subclasses. (#47110)

hameerabbasi committed 4 years ago

7908bf27

Fixed einsum compatibility/performance issues (#46398)

heitorschueroff committed 4 years ago

a5c65b86

[vulkan] tentative fix for conv2d_pw, and fix checks for addmm (#47723)

SS-JIA committed 4 years ago

51a661c0

Support default args in symbolic tracing (#47615)

Ansley Ussery committed 4 years ago

e914a1b9

Add max_src_column_width to autograd profiler (#46257)

Will Feng (DPER) committed 4 years ago

a5e9fa1b

Disable test_distributed_for for multigpu test env (#47703)

mrshenli committed 4 years ago

1b954749

[ONNX] Improve stability of gemm export (#46570)

BowenBao committed 4 years ago

4de40dad

Vulkan MobileNetv2 unit test. (#47616)

Ashkan Aliabadi committed 4 years ago

69532c42

Fix kthvalue error for scalar input (#47600)

heitorschueroff committed 4 years ago

bf6a156f

[numpy] torch.{all, any} : Extend Dtype Support (#44790)

kshitij12345 committed 4 years ago

6575e674

Back out "[pytorch][PR] The dimension being reduced should not be coalesced by TensorIterator" (#47642)

Natalia Gimelshein committed 4 years ago

c9d37675

add unittest for operator benchmark (#47678)

mengwa41 committed 4 years ago

f692af20

Grammatically updated the tech docs (#47345)

ashishpandey2600 committed 4 years ago

a843d48e

fix assert_allclose doesnt check shape (#47580)

walterddr committed 4 years ago

febc76a5

[pytorch] fix debug symbol flag for android clang (#46331)

ljk53 committed 4 years ago

8e3af9fa

[complex] torch.sqrt: fix edge values (#47424)

kshitij12345 committed 4 years ago

baa2f777

Newer Older