strisunshinewentingwang/pytorch

Pull Requests Commits

Update codeowners for MPS

albanD committed 2 years ago

Verified a69b05c7

Bessel functions (#78451)

0x00b1 committed 2 years ago

4a5381ab

Revert "Always convert truthy booleans to 1"

pytorchmergebot committed 2 years ago

78824a7d

Fix embedding jvp support by making embedding_renorm ignore forward mode AD (#78560)

samdow committed 2 years ago

ce7c7bb2

[primTorch] Use the same error message as in ATen for canonicalize_dim (#78541)

IvanYashchuk committed 2 years ago

0be4672a

Reference implementations for rot90, roll, atleast_1d,2d,3d (#78080)

IvanYashchuk committed 2 years ago

48256f3c

[primTorch] Adds broadcast_shapes reference (#78612)

jjsjann123 committed 2 years ago

fea909b4

MPS: Fix issues with view tensors and linspace. (#78690)

kulinseth committed 2 years ago

4858c563

Always convert truthy booleans to 1

peterbell10 committed 2 years ago

3c3c6cd9

Fix docs for torch.real (#78644)

zasdfgbnm committed 2 years ago

388d4431

remove prims::square (#78627)

zasdfgbnm committed 2 years ago

b651148f

Generalize sizes and strides policy on _make_wrapper_subclass

suo committed 2 years ago

876c3593

Revert "[complex32, jiterator] cos, sinh, cosh, tanh (#78458)"

pytorchmergebot committed 2 years ago

64a01f12

Norm decomposition (#78582)

yushangdi committed 2 years ago

02273f05

[ONNX] Update CI test script to run parallel by default (#78200)

BowenBao committed 2 years ago

cfc96895

remove math kernels that have derivative formulas in core

bdhirsh committed 2 years ago

bf629642

[DataPipe] Lazily generate exception message for performance

NivekT committed 2 years ago

575c4202

move generated_srcs_list.bzl into caffe2/build.bzl

Michael Andreas Dagitses committed 2 years ago

7dc5b5bf

Docs: build with Sphinx 5 (#70309)

adamjstewart committed 2 years ago

d90652db

[quant] Factor out common operator configs from native.py (#78407)

jerryzh168 committed 2 years ago

22fd2f2e

[MPS] Do not pass linker command to a compiler (#78630)

malfet committed 2 years ago

634954c5

Don't include libiomp with conda install on MacOS (#78632)

atalman committed 2 years ago

ca7f9488

Modernize FakeTensorMode, throw on non-fake inputs

eellison committed 2 years ago

6671b504

Update distributed/CONTRIBUTING.md to remove ProcessGroupAgent references and add test instructions

H-Huang committed 2 years ago

24b7142d

[quant][core][better-engineering] Rename files in quantized/cpu directory to conform with non-quantized countertpart filenames

dzdang committed 2 years ago

5874a311

enable with semantics

samdow committed 2 years ago

aa06d052

[PyTorchEdge] Extend Flatbuffer to get mobile_info for NMLML workflows

pavithranrao committed 2 years ago

9b81e817

[complex32, jiterator] cos, sinh, cosh, tanh (#78458)

khushi-411 committed 2 years ago

5fbec86f

Revert "[chalf] where(cpu and cuda), pow(cuda) (#77640)"

pytorchmergebot committed 2 years ago

4bb8db85

Move THPStorage definitions out of `torch/csrc/generic` (#78032)

kurtamohler committed 2 years ago

272193d0

Older