strisunshinewentingwang/pytorch

Pull Requests Commits

add --gpu-max-threads-per-block=256 to hipMAGMA build (#54161)

jeffdaily committed 3 years ago

ab8e9188

Automated submodule update: FBGEMM (#54118)

facebook-github-bot committed 3 years ago

80a4a500

Skips failing pinv and pinverse test (#54392)

Mike Ruberry committed 3 years ago

fc58b3f1

Rewrite functional.tensordot to be TorchScript-able (#53672)

gmagogsfm committed 3 years ago

f48a9712

Merge channel hierarchies. (#54333)

beauby committed 3 years ago

cffa70d3

[StaticRuntime] Copy version of reshape/flatten (#54353)

Hao Lu committed 3 years ago

8294bff2

Tag distributed team for review for /torch/nn/parallel (#54221)

rohan-varma committed 3 years ago

08e43125

[nnc] Remove cached argv from LLVMCodeGen to fix race condition (#54286)

bertmaher committed 3 years ago

98baad57

Support non-tensor inputs and outputs for checkpointed functions. (#52422)

pritamdamania committed 3 years ago

4fa47e5e

Support for Half / bfloat16 / index_select and better testing (#53898)

lezcano committed 3 years ago

9d9986fd

[package] Make exporters write to buffer in fbcode (#54303)

Meghan Lele committed 3 years ago

d58c00a5

Fix `torch.linalg.qr` example (#54342)

malfet committed 3 years ago

41b1ea21

update distributed doc table for alltoall nccl (#54277)

wanchaol committed 3 years ago

270d675f

Remove legacy constructor calls from _torch_ folder. (#53889)

ysiraichi committed 3 years ago

27048c1d

Allow linking against vcomp on Windows (#54132)

skyline75489 committed 3 years ago

6a4d2c61

port div to structured (#53680)

bdhirsh committed 3 years ago

6f7a5a47

port sub to structured, fix sub.Scalar bug (#53679)

bdhirsh committed 3 years ago

fa482aa4

port at::pow to structured (#53669)

bdhirsh committed 3 years ago

779cae9e

Add codeowners for onnx export (#54287)

BowenBao committed 3 years ago

454dd7ba

Backup .circleci/config.yml before regenerating (#54345)

samestep committed 3 years ago

679f07a0

[caffe2] expose whether FBGEMM is available to the Python code (#54274)

simpkins committed 3 years ago

da18313d

[PyPer] Port c2 add to pt (#54229)

Hao Lu committed 3 years ago

f1cbd102

.github: Add workflow to build libtorch (#53292)

samestep committed 3 years ago

fa07d0c8

[FX][EZ] Fix type correctness on GraphModule.graph (#54305)

James Reed committed 3 years ago

05a03a6c

port at::mul to structured (#52692)

bdhirsh committed 3 years ago

bc4f5211

`torch.prod` backward for complex types. (#48125)

nikitaved committed 3 years ago

61b07458

Refactor Unary Ops tests (#49712)

Iurii Zdebskyi committed 3 years ago

cc7a28d7

Fix inaccurate dispatch tables (#54127)

ezyang committed 3 years ago

645a3e9a

Add Tensor::is_cpu, genericize TensorIterator (#54079)

ezyang committed 3 years ago

49f13361

Refactor tensor_new.cpp to use TensorOptions instead of DispatchKey (#54034)

ezyang committed 3 years ago

e0aebe24

Newer Older