strisunshinewentingwang/pytorch

Pull Requests Commits

[ONNX] Add test for Batchnorm (#45633)

KsenijaS committed 3 years ago

7f8b02f5

Mark parts of spectral tests as slow (#46509)

malfet committed 3 years ago

172ed51a

Refactor scalar list APIs to use overloads (#45673)

Iurii Zdebskyi committed 3 years ago

e7564b07

Add myself as owner of C++ APIs related folder (#46487)

lixinyu committed 3 years ago

f06ea4bc

Enable TP_USE_CUDA and TP_ENABLE_CUDA_IPC (#46523)

mrshenli committed 3 years ago

eadc59df

detect inplace modifications of views earlier (fix #21875) (#46204)

bdhirsh committed 3 years ago

00c779a9

[RFC] Switch PyTorch Selective Build (Custom Build) to use the SelectiveBuilder abstraction (#45722)

dhruvbird committed 3 years ago

0c5cd8c2

Change codecov comment style to be less verbose (#46499)

malfet committed 3 years ago

bcd68dfa

Delete libtorch test jobs (#46508)

malfet committed 3 years ago

351670f0

Disable profiling when `getGraphExecutorOptimize` is unset (#46479)

Krovatkin committed 3 years ago

c3466dab

Expose script_if_tracing as public API (#46494)

gmagogsfm committed 3 years ago

6a2f40dc

torch.fft: Two dimensional FFT functions (#45164)

peterbell10 committed 3 years ago

da95eec6

[Metal] Add the Python binding for optimize_for_mobile (#46456)

xta0 committed 3 years ago

495070b3

[c10] add operator= of intrusive_ptr to weak_intrusive_ptr (#44045)

wanchaol committed 3 years ago

e8ff0f6c

[Metal] Enable optimize_for_mobile on Linux (#46384)

xta0 committed 3 years ago

cc471c6d

[TensorExpr] Re-enable test for torch.cat, add a test for torch.cat being a single node in a fusion group. (#46447)

Mikhail Zolotukhin committed 3 years ago

5233ff9f

[TensorExpr] Properly handle input types promotion and special case of empty inputs for aten::cat. (#46500)

Mikhail Zolotukhin committed 3 years ago

d6de9d57

[TensorExpr] Fix shape inference logic for aten::cat. (#46482)

Mikhail Zolotukhin committed 3 years ago

0f668d95

Remove catchAllKernel_. (#46354)

Ailing Zhang committed 3 years ago

58f14d3a

[GPU] Introduce USE_PYTORCH_METAL (#46383)

xta0 committed 3 years ago

04e5fcc0

Add flatten loops transformation (#46365)

navahgar committed 3 years ago

fa108bd2

[GPU] Add metal to DispatchKeySet (#46455)

xta0 committed 3 years ago

5da4a086

[WIP] Move catchAll to Math (#45939)

Ailing Zhang committed 3 years ago

8c629ecc

[Gradient Compression] Rename the first arg of pybinding of _register_comm_hook: ddp_model -> reducer. (#46498)

Yi Wang committed 3 years ago

d1ca7ef3

caffe2: use at::mt19937 instead of std::mt19937 (10x speedup) (#43987)

d4l3k committed 3 years ago

0c9787c7

[hotfix] remove test.pypi.org (#46492)

Rong Rong committed 3 years ago

e6e83bf8

Run __setstate__ when cloning modules (#45858)

bzinodev committed 3 years ago

11cc7f14

Removing hack so that NCCL is not removed in base Docker commands (#46495)

janeyx99 committed 3 years ago

478fa180

type check for torch.quantization.stubs (#46475)

Rong Rong committed 3 years ago

89108ba6

Move NCCL installation for xenial-cuda10.1 to Docker image instead of for every job (#46476)

janeyx99 committed 3 years ago

997e672a

Newer Older