strisunshinewentingwang/pytorch

Pull Requests Commits

Revert "Make allreduce compatible with fx ProxyTensor (#84126)"

pytorchmergebot committed 2 years ago

b8fe0edc

[xla hash update] update the pinned xla hash (#84164)

pytorchmergebot committed 2 years ago

c032b097

Add nvprims.var_mean (#83508)

IvanYashchuk committed 2 years ago

7e7694b6

[ONNX] Fix type annotations and enable type checking for all apis (#84091)

justinchuby committed 2 years ago

6446da17

_to_copy decomp (#84108)

ngimel committed 2 years ago

e33897cb

Enforce explicit ProcessGroup passed into DefaultState (#84105)

rohan-varma committed 2 years ago

adc9a1e2

[Nested Tensor] detach (#84078)

drisspg committed 2 years ago

092fe71f

[AOT Autograd] Redirect named_parameters to original mod (#84157)

anijain2305 committed 2 years ago

43620b7e

Revert "Don't introduce new overload for SymInt (#83628)"

pytorchmergebot committed 2 years ago

c7edcd69

Revert "[xla hash update] update the pinned xla hash (#84043)"

pytorchmergebot committed 2 years ago

38e5e4a8

[Prim] Implement group_norm_backward (#84037)

malfet committed 2 years ago

bed85cce

[DataPipe] Convert MapDataPipe.shuffle to IterDataPipe (#83202)

ejguan committed 2 years ago

a423c966

[Profiler] Add `disabled` and `global` methods to ProfilerConfig. (#83891)

Taylor Robie committed 2 years ago

69e9f905

[Profiler][Trivial] Cleanup ExperimentalConfig (#83890)

Taylor Robie committed 2 years ago

f4dc7b3a

[fx][pass] Fix type of exception (#84094)

angelayi committed 2 years ago

eb2fa2e0

[Nested Tensor] do not use at::cuda::getDefaultCUDAStream() (#84134)

jeffdaily committed 2 years ago

aa4be48b

Enable cache action for windows and other minor workflows (#84093)

huydhn committed 2 years ago

82efb0e1

Read via FileAdapter when loading files in torch if not flatbuffer (#84028)

Ian Graves committed 2 years ago

3fae89d4

Add support to traverse all python collection objects (#84079)

ejguan committed 2 years ago

e0f0c8e7

[ONNX] Export node and value with scope name (#82040)

BowenBao committed 2 years ago

6a366628

make job pass even if monitoring script fails (#84068)

clee2000 committed 2 years ago

b5c2b0b2

Update Dynamo pin (#83829)

anijain2305 committed 2 years ago

6a586039

[Profiler][Trivial] Add null handling to `AppendOnlyList::copy` memcpy path. (#83963)

Taylor Robie committed 2 years ago

61b9d8fc

[Profiler][Minor] Extend Python bindings (#83622)

Taylor Robie committed 2 years ago

014a333d

[ONNX] Clean up patch functions (#83136)

justinchuby committed 2 years ago

681c3870

Make allreduce compatible with fx ProxyTensor (#84126)

mrshenli committed 2 years ago

ec5b83f7

Update proxy_tensor.py to support List input/output (#83302)

mrshenli committed 2 years ago

f93446ad

Expose ProcessGroup::Work.wait() API to TorchScript (#83303)

mrshenli committed 2 years ago

527a1601

Add type hints to torch.save, torch.load (#83937)

adamjstewart committed 2 years ago

c6348a71

mac circleci workflows (#82780)

clee2000 committed 2 years ago

582c0833

Newer Older