strisunshinewentingwang/pytorch

Pull Requests Commits

In dynamo tracing for index() use None as the default indicator for end and not -1 (#119151)

tarun292 committed 223 days ago

bc521f2c

Decompose torch.ops.higher_order.auto_functionalized in Inductor (#118673)

zou3519 committed 223 days ago

cf474a09

[export] Implement logging for scuba. (#119585)

zhxchen17 committed 223 days ago

8069b296

Refactor ExportedProgram to expose the functions for pre and postprocessing (#119513)

qihqi committed 223 days ago

757201c2

add get_function to TorchInGraphFunctionVariable (#119314)

laithsakka committed 223 days ago

72d9a381

[sparse] Add in out_dtype support (i8i8->bf16, i32) for cusparselt (#119296)

jcaip committed 223 days ago

1c1dc0e4

Enable x86 CPU vectorization on windows (#118980)

xuhancn committed 223 days ago

5f69d95b

[AOTI][refactor] Move ThreadLocalCachedOutputTensor into a separate header (#119392)

desertfire committed 223 days ago

52a3de6c

Revert "Reify view_func() closures as ViewFuncs (#118404)"

pytorchmergebot committed 223 days ago

24bdd03d

Fix some tests in test_c10d_functional_native.py (#119102)

yifuwang committed 224 days ago

79df8976

Revert "[c10d] PGNCCL refactor part 2: Simplify ProcessGroupNCCL into single-device style (#119421)"

pytorchmergebot committed 224 days ago

0342b227

Remove unused header inclusion (#119667)

cyyever committed 224 days ago

8a3c2410

Add CUDAEvent recording for constant folding to show up. (#119216)

muchulee8 committed 224 days ago

dcb08a70

[executorch hash update] update the pinned executorch hash (#119648)

pytorchupdatebot committed 224 days ago

bc4d0277

add a couple more cases to pointwise_cat perf tests (#119521)

Chillee committed 224 days ago

76fac695

Implement conditional statements in kernel analysis (#119664)

oulgen committed 224 days ago

647564db

[inductor] Update the compile options for CppPythonBindingsCodeCache (#119415)

desertfire committed 224 days ago

663dd5d0

[BE] Properly mark destructor overrides (#119656)

malfet committed 224 days ago

069581b3

[dynamo][eval][perf] Remove unnecessary dict copies. (#119305)

ttsugriy committed 224 days ago

a4cc6b85

Check alignment of ReinterpretView args of custom Triton kernels (#119649)

aakhundov committed 224 days ago

e5f46a1d

[torch][cuda][perf] Avoid unnecessary dicts. (#118011)

ttsugriy committed 224 days ago

b8e44232

[random] Replace for loop with list comprehension. (#119143)

ttsugriy committed 224 days ago

95a8d5b1

[inductor] Use list comprehension to initialize unused_views. (#119618)

ttsugriy committed 224 days ago

4394e0dc

Optimize multi_tensor_apply (#119153)

yifuwang committed 224 days ago

24be7daf

Add lowerings to special functions (#119187)

pearu committed 224 days ago

2c91e13a

[MPS] Enable `bfloat16` support on MacOS 14 (#119641)

malfet committed 224 days ago

4ee8aac4

[BE][EZ] Use `dyspatch_sync_with_rethrow` in searchsorted (#119646)

malfet committed 225 days ago

68e009dd

fix torch.set_float32_matmul_precision doc (#119620)

lancerts committed 225 days ago

6cd82253

Remove unneeded linking of torch_shm_manager in CMake (#119540)

cyyever committed 225 days ago

88183923

Don't skip register-spilling configs in custom Triton kernel auto-tuning (#119634)

aakhundov committed 225 days ago

0bed0501

Newer Older