strisunshinewentingwang/pytorch

Pull Requests Commits

Quantized: Use per-operator headers (#75569)

peterbell10 committed 1 year ago

7134b9bc

[MPS] Fix printTensor() for MPS (#86534)

malfet committed 1 year ago

67434c70

[vision hash update] update the pinned vision hash (#86490)

pytorchmergebot committed 1 year ago

9998f910

[torchdynamo hash update] update the pinned torchdynamo hash (#86489)

pytorchmergebot committed 1 year ago

92ac84c9

QuantizedCPU: Use per-operator headers (#71217)

peterbell10 committed 1 year ago

492d1be5

cuDNN/miopen: Use per-operator headers (#71216)

peterbell10 committed 1 year ago

4bfe2a24

Re-land*4 "SymIntify cat and narrow" (#86468)

ezyang committed 1 year ago

33f0e98a

Revert "Re-enable torchdynamo tests (#86297)"

pytorchmergebot committed 1 year ago

8ea2ed0f

Enable aten-aten decomps (#85921)

eellison committed 1 year ago

d3f7c34c

[FSDP] Add `keep_low_precision_grads` support when CPU offloading (#86495)

awgu committed 1 year ago

af9c6bc8

Revert "Enable aten-aten decomps (#85921)"

pytorchmergebot committed 1 year ago

7ec12a55

[vulkan] Add buffer to buffer copies (#86424)

SS-JIA committed 1 year ago

b0ceb8ea

[vulkan] Clean up convolution code (#86423)

SS-JIA committed 1 year ago

511d81cd

make g2p ~30% faster on mobile by suppressing a log (#85907)

codyohl committed 1 year ago

b645c237

[JIT] Allow freezing modules that contain mutable interfaces (#86039)

davidberard98 committed 1 year ago

bac26155

better error message fix (#86422)

hongxiayang committed 1 year ago

04490e90

[quant][ao_migration] nn.intrinsic.quantized migration to ao (#86172)

zaf committed 1 year ago

3a028731

Caching allocator tracing (#86241)

zdevito committed 1 year ago

91b1bae1

Fix index_select decomp (#86469)

SherlockNoMad committed 1 year ago

8a3a54e0

Skip dynamo for all optim test as they are all flaky otherwise (#86482)

albanD committed 1 year ago

a079dad7

Add multi-grad hooks (#86260)

soulitzer committed 1 year ago

ba3fde6a

Try to fix shutdown test in edge cases (#86464)

albanD committed 1 year ago

97e56c17

Enable aten-aten decomps (#85921)

eellison committed 1 year ago

62e4f51e

[FSDP] Add initial `summon_full_params(with_grads=True)` (#85738)

awgu committed 1 year ago

a95889ba

[optim] fix: empty grad support for SparseAdam (#86459)

kshitij12345 committed 1 year ago

82229d1e

Revert "Disable mac m1 jobs (#86463)"

pytorchmergebot committed 1 year ago

66d480d3

Disable mac m1 jobs (#86463)

huydhn committed 1 year ago

ac632b43

[ao] fixing public v private for fuser_method_mappings.py (#86029)

HDCharles committed 1 year ago

ac74976a

[FSDP] Add `use_orig_params` (#84911)

awgu committed 1 year ago

be682bef

Add reference counter in FileStore (#85601)

KinglittleQ committed 1 year ago

b43ae1c4

Newer Older