strisunshinewentingwang/pytorch

Pull Requests Commits

[FSDP()][12/N] Easy cleanup (#87925)

awgu committed 2 years ago

8039317c

[FSDP()][10/N][11/N] Introduce composable (ctor only) (#87924)

awgu committed 2 years ago

c1e28731

[FSDP()][9/N] Refactor ctor (continued) (#87923)

awgu committed 2 years ago

78170701

[Static Runtime] Fix ReplaceWithMaybeCopy test in OSS (#88099)

Mike Iovine committed 2 years ago

23fe6c8c

Fix monitoring script for macos (#88159)

huydhn committed 2 years ago

7c6fe21a

Cleaned up the nvFuser Python Frontend Batch Norm printing (#88057)

kevinstephano committed 2 years ago

323c646c

[BE] Use default constructor in `LoggerVoidify` (#88054)

malfet committed 2 years ago

a6acbad5

call contiguous on BMM inputs for NT on CUDA (#88108)

drisspg committed 2 years ago

560786ac

Remove "prims_nvfuser" backend for TorchDynamo (#88083)

IvanYashchuk committed 2 years ago

0eea05b1

[torch::deploy] add gpu unit tests to CI (#88107)

PaliC committed 2 years ago

a8aaee77

Add support for neg to NestedTensor (#88131)

cpuhrsch committed 2 years ago

6a75a0d1

Add labeler with cpu, mkldnn, amp, NNC and quantization paths to start (#87690)

yanbing-j committed 2 years ago

708c050a

[xnnpack][lite-int][4/n] introduce serialization to delegate (#87908)

mcr229 committed 2 years ago

3aa7a528

[xnnpack][lite-int][3/n] flatbuffer serializer class (#87907)

mcr229 committed 2 years ago

8287c1d9

[xnnpack]lite-int][2/n] flatbuffer xnn_value schema (#87906)

mcr229 committed 2 years ago

7bf819b1

[xnnpack][lite-int][1/n] flatbuffer buck rules (#87826)

mcr229 committed 2 years ago

905d532d

[xnnpack][lite-int][graph-build] torchscript -> xnnpack graph (#87824)

mcr229 committed 2 years ago

aa1f9a1b

Also skip large models for normal --accuracy runs (#88086)

ezyang committed 2 years ago

d596b048

Change Nested Tensor logging copy (#88104)

drisspg committed 2 years ago

afd00673

Revert "[dynamo] Error when user nests FX with dynamo (#87797)"

pytorchmergebot committed 2 years ago

c0761a83

Fix `PyTorchStreamWriter` exception handling (#88128)

malfet committed 2 years ago

caaf37a1

[IOS] Update Cocoapods for 1.13 release (#88075)

jmdetloff committed 2 years ago

ea8a5b09

Store `autocast_gpu_dtype` in `custom_fwd` and `custom_bwd` for BFloat16 autocast (#88029)

crcrpar committed 2 years ago

bc03aa60

Remove stale comment (#88135)

ezyang committed 2 years ago

f2b247f0

Fix links to tutorial in torch masked docs (#88129)

cpuhrsch committed 2 years ago

139afc50

fix for auto labeler (#88100)

clee2000 committed 2 years ago

9fed04ba

Fix random "C1041: cannot open program database" errors when compiling on Windows (#88084)

Blackhex committed 2 years ago

ba26bc0f

Do not use unsafe restriding for subclasses (#87610)

bdhirsh committed 2 years ago

73379aca

Create separate files for NT Unary, Binary and Matmul ops (#88091)

cpuhrsch committed 2 years ago

6fe41e76

Changing from sample_inputs to reference_inputs in test_compare_cpu (#86462)

srossross committed 2 years ago

1a9edc81

Newer Older