strisunshinewentingwang/pytorch

Pull Requests Commits

[pytorch]Migrate _th_ger to Aten and kill resize_scalar in codegen (#33792)

lixinyu committed 4 years ago

57c1b80e

[JIT] Register rpc.rpc_async(..) as a JIT operator (#33329)

xush6528 committed 4 years ago

7d01888a

[jit] Fix iOS build (#34180)

davidriazati committed 4 years ago

9b39ad7f

[pytorch][mobile] support for custom mobile build with dynamic dispatch (#34055)

ljk53 committed 4 years ago

3c042a6a

[quant][graphmode] Skip quantizing input and output in matched module (#32814)

jerryzh168 committed 4 years ago

e5bbd23c

[pytorch_ci] Python target determinator (#33577)

yns88 committed 4 years ago

7cee787a

NNPI op mapping correct SpatialBN NNPI op name (#34176)

amylittleyang committed 4 years ago

7c205787

[Redo][ATen] Remove AT_ASSERTM from Blob::free_() (#34168)

Hao Lu committed 4 years ago

a19db54b

Expose `CUDACachingAllocator` `raw_alloc` and `raw_delete` to python (#33860)

Emilio Castillo committed 4 years ago

31cc3111

[c10] Fix typo in __assert_fail noreturn modifier guard (#34157)

malfet committed 4 years ago

4edff32f

[jit] Add type tags to lists/dicts in pickle (#33255)

davidriazati committed 4 years ago

99e211e6

Apply clang-format to RPC files (#34139)

mrshenli committed 4 years ago

7da24b36

Use double quotes in C++ to stay consistent with Python RPC docs (#34095)

mrshenli committed 4 years ago

3af0dffe

Improve ProcessGroup RpcBackendOptions Constructor API (#34081)

mrshenli committed 4 years ago

f1085a8e

[Aten] Suppress valgrind leaks in libcuda (#34169)

malfet committed 4 years ago

9d1c971b

Make DEBUG == REL_WITH_DEB_INFO on CUDA build (#34153)

zasdfgbnm committed 4 years ago

1beb309e

.circleci: Re-do run nightly pipelines on tag (#34148)

seemethere committed 4 years ago

cb3905e8

Remove deprecated codepath for old-style autograd.Function (#30696) (#33956)

lixinyu committed 4 years ago

7cda964e

[JIT] Add modulelist indexing for integer literal (#29236)

Elias Ellison committed 4 years ago

04378eb6

Turn on strict dtype checking for test_torch.py (#33825)

ezyang committed 4 years ago

ba1bd417

Revert D20171428: [profiler] fix chrome tracing for profiler run with cuda

rohan-varma committed 4 years ago

c5799766

Completely kill CUDA_tensor_apply3 (#34026)

zasdfgbnm committed 4 years ago

f299c2d6

Migrate lerp from CUDA_tensor_apply3 to TensorIterator (#34025)

zasdfgbnm committed 4 years ago

1affaf8d

Migrate bce loss from CUDA_tensor_apply3 to TensorIterator (#34023)

zasdfgbnm committed 4 years ago

27f56632

[gloo] dont hold locks in calls to buffer in ProcessGroupGloo:RecvWork::wait() and (#33926)

rohan-varma committed 4 years ago

92083f31

[profiler] fix chrome tracing for profiler run with cuda (#33987)

rohan-varma committed 4 years ago

c93b1d42

Remove use of `.data` from optimizers (#33640)

Enealor committed 4 years ago

6a97777f

[fix] flake8 lint error (#34146)

jerryzh168 committed 4 years ago

f26bbb5f

Fix HistogramObserver to not do detach on input (#34114)

Dmytro Dzhulgakov committed 4 years ago

a8fc3d8c

[caffe2] fix ambiguous call to 'fmaxType' THCHalfAutoNumerics.cuh (#33569)

Igor Sugak committed 4 years ago

9650253d

Newer Older