strisunshinewentingwang/pytorch

Pull Requests Commits

[LT] Support narrow

Jiewen Tan committed 2 years ago

b0081e76

Adds OpInfos for l1_loss and smooth_l1_loss

Mike Ruberry committed 2 years ago

9d17157f

[Distributed] Add a guard for non CPU/CUDA devices

Jiewen Tan committed 2 years ago

1c0a01e7

[quant][improvement][gpu] Fixed errors in test_qlinear_cudnn

dzdang committed 2 years ago

7d8b3662

Adds margin_ranking_loss opinfo

Mike Ruberry committed 2 years ago

2da43ec0

[Quant][fx] Fix get_default_qconfig_dict for fused modules

andrewor14 committed 2 years ago

5dcbcc6d

Adds OpInfos for max_unpool{1, 2, 3}d

Mike Ruberry committed 2 years ago

41c59987

Adds multi_margin_loss opinfo

Mike Ruberry committed 2 years ago

a7dc893d

Adds multilabel_soft_margin_loss opinfo

Mike Ruberry committed 2 years ago

aeb6cf35

ci: Change with-ssh to be on by default

seemethere committed 2 years ago

18b9d6b2

Optimize PReLU (float32) and enable PReLU BFloat16 support in CPU path (#63634)

yanbing-j committed 2 years ago

dc2e6303

Revert "Optimize PReLU (float32) and enable PReLU BFloat16 support in CPU path"

pytorchmergebot committed 2 years ago

e8ed0420

Optimize PReLU (float32) and enable PReLU BFloat16 support in CPU path

yanbing-j committed 2 years ago

263c4c2a

[torch][elastic] Make final agent barrier to shutdown properly

aivanou committed 2 years ago

e54c1f6c

[2] remove caffe2 math.h from maskrcnn ops

linbinyu committed 2 years ago

bc72add5

[lint] Use a problem matcher for GitHub annotations

suo committed 2 years ago

85235c6f

[lint] add test ownership lint to lintrunner

suo committed 2 years ago

3e0e1375

set fetch-depth: 1

clee2000 committed 2 years ago

991c89b2

Revert "Add support to `Tensor[]?` for structured kernel codegen."

malfet committed 2 years ago

3c238c6c

[NVFuser] fix incorrect vector size in chunk move

davidberard98 committed 2 years ago

15892481

[lint] improve retries in stale job

suo committed 2 years ago

9555f3b3

[jiterator] logical_and: complex

khushi-411 committed 2 years ago

fa2c7e2e

Move //xplat/caffe2:caffe2_serialize to shared build structure

kit1980 committed 2 years ago

a5b4839f

Remove spurious warning when using disabled torch function

albanD committed 2 years ago

3467f3fa

Add support to `Tensor[]?` for structured kernel codegen.

ysiraichi committed 2 years ago

ac2de5a0

Utilities for handling Sparse Compressed tensors

pearu committed 2 years ago

0ab03af5

Fix data race on owns_pyobj_ accesses with non-GIL protected threads

ezyang committed 2 years ago

337e3932

Set correct module for APIs in torch module

anjali411 committed 2 years ago

533f0cb2

Preserve Python dispatch keys upon copy_tensor_metadata_except_version_counter

ezyang committed 2 years ago

27728708

Improves the OpInfo out= tests

Mike Ruberry committed 2 years ago

b0976999

Newer Older