strisunshinewentingwang/pytorch

Pull Requests Commits

[cuDNN V8 API] Enable cuDNN v8 API by default (#75466)

eqy committed 2 years ago

14ab3ff4

Make PartialTensot a torch.Tensor subclass

pritamdamania87 committed 2 years ago

068d35a6

Revert "[flaky test reporting] print stack trace for flaky reruns (#77664)"

pytorchmergebot committed 2 years ago

b87e5f38

Fix MPS interaction with autograd engine

albanD committed 2 years ago

090eddf1

Bitwise ops improvements (#77621)

zasdfgbnm committed 2 years ago

f2745580

fix complex abs/angle output format (#77585)

ngimel committed 2 years ago

059c439e

[GH1] Add builds as required status checks (#77596)

janeyx99 committed 2 years ago

54bc94bb

change sparse_mask to support Half

george-qi committed 2 years ago

8f20ea19

FSDP parameter sync

rohan-varma committed 2 years ago

6f954d7b

rohan-varma committed 2 years ago

8ae0b275

[BE] remove unused RDS pipeline from print_test_stats.py (#77654)

janeyx99 committed 2 years ago

a580e07d

[flaky test reporting] print stack trace for flaky reruns (#77664)

janeyx99 committed 2 years ago

646a20c8

Revert "Use weakref.proxy when saving module to internal dictionaries to not increase refcount (#76435)"

pytorchmergebot committed 2 years ago

d8b80eda

Fix typo in torch.package code and docs (#77604)

sadra-barikbin committed 2 years ago

71d61bb7

add channels last support for PixelShuffle and PixelUnshuffle

mingfeima committed 2 years ago

c0034947

fix grad(torch.tensor()) using lift() operator

bdhirsh committed 2 years ago

cfc87cad

Remove operators that support BFloat16 in the fp32 cast policy list of AutocastCPU (#77623)

CaoE committed 2 years ago

541a3789

Add Sparse Compressed tensor support to torch.clone

pearu committed 2 years ago

dc882ed3

updating tolerance override (#77402)

jjsjann123 committed 2 years ago

98a20ebb

A few forward AD formulas

lezcano committed 2 years ago

369d9f41

Fix mypy lint failure

ezyang committed 2 years ago

4d3930fe

[complex32, jiterator] angle (#76692)

khushi-411 committed 2 years ago

387cccb1

[chalf] enable testing for multiple ops (#77404)

kshitij12345 committed 2 years ago

5a16f849

Revert "Add Sparse Compressed tensor support to torch.clone"

pytorchmergebot committed 2 years ago

0d1329c4

[composite compliance] backward: index_select, trace, repeat_interleave (#77578)

kshitij12345 committed 2 years ago

47809fe6

fix torch.jit.tracing for at::lift (#77588)

bdhirsh committed 2 years ago

f9f4896a

`nn.functional.glu`: forward over reserse AD support (#77309)

nikitaved committed 2 years ago

3ccf35e0

Reland "[JIT] during freezing, cast optional bias to half if weight is half"

davidberard98 committed 2 years ago

d0dc7cb7

quant docs: best practices for quantization accuracy debugging

vkuzo committed 2 years ago

9cc92d53

Reduce structured kernels' `set_output` boilerplate with new overloads.

ysiraichi committed 2 years ago

9d442507

Newer Older