strisunshinewentingwang/pytorch

Pull Requests Commits

is_complex doc fix (#35680)

anjali411 committed 4 years ago

2c6d1e57

[jit] Minor: in unpickler, string tweak in readBytes() (#35550)

jjlilley committed 4 years ago

3ba88589

[jit] In RPC Server, handle TorchScript continuations asynchronously (#34109)

jjlilley committed 4 years ago

8d64a384

[ROCm] Remove installation of ca-certificates and apt-transport-https in test.sh (#35676)

iotamudelta committed 4 years ago

e5746eec

[quant][graphmode] Quantization support for at::sort (#35571)

jerryzh168 committed 4 years ago

9650f465

[Onnxifi] Add initializers to the C2 net passed into Glow (#35764)

Yinghai Lu committed 4 years ago

8de01aac

[easy] ThroughputBenchmark: print out aten's parallel settings before execution (#35632)

Alexander Sidorov committed 4 years ago

7d5350c2

bfloat16: vectorized clamp, clamp_min and clmap_max (#35082)

XiaobingSuper committed 4 years ago

07dbf0db

Updating submodules

svcscm committed 4 years ago

8e49afa9

Fix a bug in subgraph rewriters. (#35704)

kimishpatel committed 4 years ago

063275fd

[rref] Handle exceptions returned via remote() calls (#35331)

jjlilley committed 4 years ago

f182b437

hswish and hardsigmoid: improve docs (#35431)

vkuzo committed 4 years ago

b4c43427

Move libtorch to py3 and cleanup other CircleCI config (#35700)

orionr committed 4 years ago

3d6b5bac

Update collect_env.py to detect relevant conda-installed numpy and cudatoolkit (#35646)

rgommers committed 4 years ago

8ff05031

[caffe2] explicitly pass use_offsets=false when calling fbgemm embedding kernels (#35711)

jspark1105 committed 4 years ago

ada64721

[caffe2] Switch to using `public_include_directories

mzlee committed 4 years ago

81c24127

Disables complex floor, ceil, trunc (to be compatible with NumPy) (#35592)

Mike Ruberry committed 4 years ago

d2343bea

Revert D20749588: [pytorch][PR] Use `std::abs` instead of `abs` in lbfgs.cpp

zou3519 committed 4 years ago

539d3ff3

[JIT] Improve the error message when registering a custom class twice (#35568)

rohan-varma committed 4 years ago

59268d4c

Recording of TorchScript functions (#34710)

Ilia Cherniavskii committed 4 years ago

800d5617

clang-format (#35752)

suo committed 4 years ago

8fef8d19

[quant][graph] Add quant fusion for dynamic quantization (#35586)

supriyar committed 4 years ago

a090de38

[quant][graph] Add pass to insert quant dequant for dynamic quantization (#35448)

supriyar committed 4 years ago

1f7ee7b6

Update clang_format.yml

suo committed 4 years ago

Verified b0f84298

[Shape Inference] Try to infer input of elementwise ops (#35701)

ChunliF committed 4 years ago

35087b8d

Disable c10::import(ns) (#35398)

ezyang committed 4 years ago

4f4ed5c1

Do not link BLAS into torch_cuda/torch_hip (#35724)

ezyang committed 4 years ago

726baf69

Updating submodules

svcscm committed 4 years ago

cd760fbd

[quant][graphmode][refactor] getGeneralTensorInputs(Node*) -> getPassThroughInputs(Value*) (#35558)

jerryzh168 committed 4 years ago

9018538a

[quant][graphmode][fix] docs for InsertObservers (#35557)

jerryzh168 committed 4 years ago

8add1843

Newer Older