strisunshinewentingwang/pytorch

Pull Requests Commits

[rpc][jit] support rpc_sync in TorchScript (#43043)

wanchaol committed 4 years ago

3e5df5f2

[rpc] rename some functions (#43042)

wanchaol committed 4 years ago

8bec7cfa

MinMax based observers: respect device affinity for state_dict (#44537)

vkuzo committed 4 years ago

70dfeb44

Simplify target handling in nn gradcheck. (#44507)

gchanan committed 4 years ago

192c4111

[Cmake] Drop quotation marks around `$ENV{MAX_JOBS}` (#44557)

malfet committed 4 years ago

8a574c71

[caffe2] adds Cancel to OperatorBase and NetBase (#44145)

dahsh committed 4 years ago

2b8f0b20

Fix SmoothL1Loss when target.requires_grad is True. (#44486)

gchanan committed 4 years ago

5579b53a

[NNC] Add loop slicing transforms (#43854)

Cheng Chang committed 4 years ago

b7ef4eec

Update fallback kernel for Autograd keys. (#44349)

Ailing Zhang committed 4 years ago

39bb455e

[quant][graphmode][fx][fix] Support dictionary output (#44508)

jerryzh168 committed 4 years ago

11fb51d0

[pytorch] Remove mobile nonvariadic run_method (#44235)

ann-ss committed 4 years ago

442957d8

[pytorch] Replace mobile run_method with get_method and operator() (#44202)

ann-ss committed 4 years ago

a61318a5

add typing annotations for a few torch.utils.* modules (#43806)

guilhermeleobas committed 4 years ago

cdf5e2ae

Update interpolate to use new upsample overloads (#43025)

dreiss committed 4 years ago

7d78a6fc

Add nondeterministic check to new upsample overloads

dreiss committed 4 years ago

df6ea625

Fix L1Loss when target.requires_grad is True. (#44471)

gchanan committed 4 years ago

3de2c0b4

[dper3] Export PackSegments and UnpackSegments to Pytorch

Brandon Lin committed 4 years ago

ea558206

[PyTorch Mobile] Move some string ops to register_prim_ops.cpp and make them selective (#44500)

iseeyuan committed 4 years ago

b73b44f9

In common_distributed, fix TEST_SKIPS multiprocessing manager (#44525)

rohan-varma committed 4 years ago

567c51cc

Fix MSELoss when target.requires_grad is True. (#44437)

gchanan committed 4 years ago

d07d25a8

Update submodule gloo to have latest commits to enable it can work on Windows (#44529)

gunandrose4u committed 4 years ago

9a3b83cb

torch.view_as_complex fails with segfault for a zero dimensional tensor (#44175)

guol-fnst committed 4 years ago

b6b1c01a

Use TP Tensor.metadata to carry device info (#44396)

mrshenli committed 4 years ago

a9754fb8

Add missing rpc.shutdown() (#44417)

mrshenli committed 4 years ago

f44de7cd

C++ APIs Transformer NN Module Top Layer (#44333)

lixinyu committed 4 years ago

77cc7d1e

Clarify track_running_stats docs; Make SyncBatchNorm track_running_stats behavior consistent (#44445)

ssnl committed 4 years ago

09892de8

[NNC] Don't attempt to refactor conditional scalars (#44223)

nickgg committed 4 years ago

30fccc53

[quant] conv_transpose1d_prepack / conv_transpose1d_unpack (#40360)

z-a-f committed 4 years ago

c967e772

[JIT] Remove profiling nodes in autodiff forward graph (#44420)

Elias Ellison committed 4 years ago

8b898666

[JIT] Add a python hook for a function to interpret JIT graphs. (#44493)

Mikhail Zolotukhin committed 4 years ago

c6febc64

Newer Older