strisunshinewentingwang/pytorch

Pull Requests Commits

replace LegacyTracedModule with torchscript used in add_graph (#21339)

lanpa committed 5 years ago

54413cf9

Change PyTorch tests to use non-default CUDA stream (#21474)

huba committed 5 years ago

b144ba66

Set "scalar_check: false" for TH methods that can't return scalars. (#21496)

gchanan committed 5 years ago

5cc3a3e2

ONNX Export Slice and Flip ops for opset 10

lara-hdr committed 5 years ago

cc85c3db

Make torchvision install chatty. (#21476)

ezyang committed 5 years ago

3eced796

updating gemmlowp tp 3fb5c

harouwu committed 5 years ago

1503c734

Make test_cuda.py work on Python 2. (#21466)

ezyang committed 5 years ago

8c9a88bd

Fix batch norm multiplier init (#13774)

Kaixhin committed 5 years ago

c6046587

Implement at::match(Dimname, Dimname) and at::unify(Dimname, Dimname) (#21281)

zou3519 committed 5 years ago

c6048476

Added at::Dimname (#21280)

zou3519 committed 5 years ago

4727685e

Revert "Revert D15632268: [pytorch][PR] Continuation of Port max_unpool1d, max_unpool2d and max_unpool3d to ATen" (#21427)

ezyang committed 5 years ago

e27c2f14

Super resolution export to Caffe2 is broken, skip it. (#21479)

ezyang committed 5 years ago

d6af6588

add cancelAsyncCallback method to OperatorBase (#21492)

Peng Gong committed 5 years ago

78a37659

Adding gradient to Boolean Mask operator (#21423)

David Zhang committed 5 years ago

696b2c89

Updating submodules

svcscm committed 5 years ago

d3d195e0

Defer constructing error strings for definitions under If's until they're needed. (#21429)

Owen Anderson committed 5 years ago

772fd79d

Fix unused variable warning

asuhan committed 5 years ago

abc0d3e5

Add warning for Turing GPUs and CUDA <= 9000 (#21468)

ptrblck committed 5 years ago

bad67015

Turn XLA back on for default set (but filtered out using should_run_job) (#21470)

ezyang committed 5 years ago

63d4bbb0

add more info back to BenchResult (#21502)

hl475 committed 5 years ago

f4339139

Revert D15629687: Deprecate torch::jit::RegisterOperators

ezyang committed 5 years ago

d51bd219

Move jit testing utils to their own file (#21491)

davidriazati committed 5 years ago

37ab35c8

Fix typo (#21426)

smessmer committed 5 years ago

e87f77de

Deprecate torch::jit::RegisterOperators (#21368)

smessmer committed 5 years ago

d714abf5

ReshapeOp supports empty tensor (#21230)

David Zhang committed 5 years ago

cb2ec07f

support ceil mode by padding changes (#21310)

Aapo Kyrola committed 5 years ago

b161832f

Remove unneeded headers (#21393)

dskhudia committed 5 years ago

80a083ef

Add autograd for to_sparse. (#20458)

nairbv committed 5 years ago

8a9ea55b

Bilinear Upsampling increased throughput (#19306)

ThisIsIsaac committed 5 years ago

87d10d49

Fix numerically instability of `SigmoidTransform` (#19802)

zuoxingdong committed 5 years ago

c5d5d45f

Newer Older