strisunshinewentingwang/pytorch

Pull Requests Commits

[DataPipe] Enforcing single valid iterator for IterDataPipes without multiple outputs

NivekT committed 2 years ago

7c52f204

[DataPipe] refactor to separate _IterDataPipeMeta

NivekT committed 2 years ago

e0451d80

Add support for TxT mask layout for masked_softmax in BetterTransformer (#77607)

zrphercule2 committed 2 years ago

4e2f5507

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

pytorchmergebot committed 2 years ago

b64845eb

Python function to extract information on mobile::Module from flatbuffer (#77624)

qihqi committed 2 years ago

3822a472

Implement sym_sizes to create proper IR for sym ints representing tensor sizes (#76836)

Krovatkin committed 2 years ago

fc4c3c9b

Add dispatch mode testing for meta tensors and other stuff

ezyang committed 2 years ago

c1cdb121

Make SymInt constructor explicit

suo committed 2 years ago

7f1e331b

[skip ci] fix comment spacing in SymIntArrayRef.h

suo committed 2 years ago

c673696b

[CUBLAS][TF32] Skip `test_cublas_allow_tf32_get_set` if `TORCH_ALLOW_TF32_CUBLAS_OVERRIDE` is set (#77298)

eqy committed 2 years ago

76b952bb

[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

Newer Older