strisunshinewentingwang/pytorch

Pull Requests Commits

Add api for changing function executor settings, hook up execution with decomposition registry (#74186)

Elias Ellison committed 2 years ago

9c4a6378

Introduce function-local settings for executor, expose in c++ (#74012)

Elias Ellison committed 2 years ago

0ecf1add

[JIT] Make aot autograd decompositions usable in JIT, add script for serializing the decompositions (#73938)

Elias Ellison committed 2 years ago

aacdf291

Remove bailout logic (#73876)

Elias Ellison committed 2 years ago

a6ed6891

Clean up profiling mode and profiling executor strategy (#73875)

Elias Ellison committed 2 years ago

6694fdac

fix docs error in Autograd Mechanics

bhbbbbb committed 2 years ago

ab578764

Back out D34696255 "[pyper] to + lengths_to_offsets" (#74906)

ajyu committed 2 years ago

91a72e90

Add additional CUDA error handling macros (#74865)

r-barnes committed 2 years ago

1249d490

[FX] Fix type of argument min_acc_module_size (#74891)

cgufb committed 2 years ago

bdf468b9

[Profiler] Limit calls to `recordThreadInfo` (#74888)

Taylor Robie committed 2 years ago

d92fd2dd

Fix docstring for torch.roll

joker-eph committed 2 years ago

f17ad06c

Combine android and ios merge rule

linbinyu committed 2 years ago

81d994d5

[GHF] Fix force merge handling

malfet committed 2 years ago

ddf8ffa1

[fx2trt] Enable enum type with lower_precision (#74841)

brad-mengchi committed 2 years ago

00f2962e

Back out "free up dispatch key space (in C++)" (#74859)

bdhirsh committed 2 years ago

9872a06d

Back out "DispatchKeySet perf improvements" (#74858)

bdhirsh committed 2 years ago

a9216cde

[JIT][easy] comment about where nvfuser is called in profiling_graph_executor_impl.cpp

davidberard98 committed 2 years ago

7cd78ea3

[ci] add timeout to rocm test

suo committed 2 years ago

ffd96089

[ci] add an easier-to-understand error for workflow consolidation FC breaks

suo committed 2 years ago

87590b62

[ci] unpin workflows from master

suo committed 2 years ago

ca75b578

[fix] fix op name in dispatch

kshitij12345 committed 2 years ago

fde1d060

Add lazy tensor python bindings (#74508)

wconstab committed 2 years ago

ff206ed0

[FSDP] exclude from typing (#74833)

rohan-varma committed 2 years ago

0928da10

[vulkan] Remove unnecessary include in vulkan_api_test (#74699)

SS-JIA committed 2 years ago

3cbf308a

Cleanup C10::Scalar stringification (#73462)

Taylor Robie committed 2 years ago

0e2d5f14

add model test for Android

linbinyu committed 2 years ago

91ef3c82

[GHF] Adding PyTorch Compilers Devs to merge_rules

osalpekar committed 2 years ago

51e50a2d

Add C++ implementation of histogramdd

peterbell10 committed 2 years ago

c7f9da57

Add typing for torch.return_type

peterbell10 committed 2 years ago

71003c74

Add strided layout support for to_dense

cpuhrsch committed 2 years ago

e55b73d6

Newer Older