strisunshinewentingwang/pytorch

Pull Requests Commits

Move unboxing to after dispatch for ops with manual kernel registrations (#36398)

smessmer committed 4 years ago

c856a2cb

Fix view_complex_as_float for empty tensors (#36415)

pbelevich committed 4 years ago

7e8c27ed

Revert D20961711: [pytorch][PR] Returns float tensors for complex inputs to abs

houseroad committed 4 years ago

742c7797

[DistAutograd x JIT] Capture global state, dist autograd current context id, before thread switching triggered by JIT future.wait() (#36395)

xush6528 committed 4 years ago

ae452a81

Revert D20931186: Enable c10 unboxing for ops with TensorList

houseroad committed 4 years ago

0dbb21f8

Updating submodules

svcscm committed 4 years ago

409346ee

Catch exception in distributed engine callbacks. (#36118)

pritamdamania committed 4 years ago

5b331e86

Updating submodules

svcscm committed 4 years ago

d71aeece

Upstream generic device test patch. (#36321)

Ailing Zhang committed 4 years ago

e8923989

Remove experimental c10 ops (#36394)

smessmer committed 4 years ago

4305c7f9

Move fakelowp tests from glow to caffe2 (#36409)

Yinghai Lu committed 4 years ago

6920b135

Revert D20958928: [pytorch][PR] Port masked_select cuda from TH to ATen

Mike Ruberry committed 4 years ago

bd476112

Revert D20523080: [pytorch] reduce memory footprint in fused conv QAT ops

Mike Ruberry committed 4 years ago

86e8c49f

hardswish: add cuda kernels (#36350)

vkuzo committed 4 years ago

eddbee19

[caffe2] Use cpuinfo in perfkernels to simplify build dependency (#36371)

Dmytro Dzhulgakov committed 4 years ago

7576cf8d

Port masked_select cuda from TH to ATen (#35429)

kurtamohler committed 4 years ago

343f2c09

Open source the missing part of FakeFp16 ops (#36353)

Yinghai Lu committed 4 years ago

d27dccfd

Updating submodules

svcscm committed 4 years ago

c029aaa2

Fix the typo in operator name string (#36296)

iseeyuan committed 4 years ago

f999d600

[pytorch] reduce memory footprint in fused conv QAT ops (#35002)

kmatzen committed 4 years ago

82be7c75

Canonicalize includes in c10, and add tests for it (#36299)

zasdfgbnm committed 4 years ago

15c74864

[TensorExpr] add support for Reduction Ops (#35866)

nickgg committed 4 years ago

42457e63

[Shape Inference] Infer shape info for second input of elementwise ops (#36365)

ChunliF committed 4 years ago

5177906d

Enable c10 unboxing for ops with TensorList (#36330)

smessmer committed 4 years ago

4a98ba81

Updating submodules

svcscm committed 4 years ago

e574ff35

Fix the clang-format error caused in register prim ops change. (#36393)

iseeyuan committed 4 years ago

d73ee763

Fixed include file header guard. (#36329)

valmikir committed 4 years ago

247f2df8

Add missing TORCH_API annotation (#36391)

ezyang committed 4 years ago

79973a16

Re-enable test_avg_pool3d_nhwc (#36259)

Di Wu committed 4 years ago

b0c90fad

Add torch.Tensor.as_subclass method. (#34369)

hameerabbasi committed 4 years ago

1875c2e4

Newer Older