strisunshinewentingwang/pytorch

Pull Requests Commits

Fix div(scalar, tensor) case for nvFuser (#80341)

IvanYashchuk committed 2 years ago

471c8330

Validate Sparse Compressed tensor inputs (#79385)

pearu committed 2 years ago

cde365a7

Add __all__ for torch.nn.modules, torch.distributed.elastic, torch.nn.utils submodules (#80240)

pytorchmergebot committed 2 years ago

9db3c517

TensorImpl:::size_custom to support NestedTensor.size (#80236)

cpuhrsch committed 2 years ago

2258db5d

[composite compliance] embedding_bag (#79986)

kshitij12345 committed 2 years ago

4300f64a

[ci] skip slow gradcheck jobs (#80311)

suo committed 2 years ago

590d3e57

[Checkpoint Wrapper] Fix assert (#80283)

awgu committed 2 years ago

77942548

[composite compliance] fill (#80277)

kshitij12345 committed 2 years ago

1b18c2e9

[quant] Modify APoT nonuniform quantization workflow (#80075)

asl3 committed 2 years ago

777c12f2

Only sync CUDA if the operation is run on GPU (#80328)

Flamefire committed 2 years ago

71d9592a

Copy Tensor for tests to avoid in-place transform modifying the original tensor (#80331)

Flamefire committed 2 years ago

3b8589ac

Add Div reference (#77936)

rdspring1 committed 2 years ago

1d0d506e

support nested_tensor * scalar (#80284)

Yifan Shen committed 2 years ago

09f79e94

[quant] Modify APoT qparam quantization levels calculation (#80303)

asl3 committed 2 years ago

0eee81aa

Avoid unnecessary copy of rvalue ref shared ptr in SharedPtrWrapper (#79697)

ezgicicek committed 2 years ago

044c9a65

Disable AVX512 CPU dispatch by default (#80253)

malfet committed 2 years ago

14813536

[c10d] Make barrier as a custom op (#79777)

Jiewen Tan committed 2 years ago

238eaf20

[c10d] Make alltoall as a custom op (#79691)

Jiewen Tan committed 2 years ago

dcd17357

cmake: Fix variable typo for USE_SYSTEM_PYBIND11. (#80272)

cdluminate committed 2 years ago

799d7137

functionalization <> LTC integration (take 3) (#80251)

bdhirsh committed 2 years ago

c2d395cf

[ONNX] Move no runtime tests out of runtime test cases (#80078)

titaiwangms committed 2 years ago

33761c80

[composite compliance] quantile, nanquantile (#80024)

kshitij12345 committed 2 years ago

8a5cc940

Ensure torch._refs registrations also get triggered on import torch (#80270)

ezyang committed 2 years ago

4331bc43

Introduce Z3 types and utility functions for constraint generation (#80084)

migeed-z committed 2 years ago

443db9b5

[MPS] Add equal operator (#80195)

malfet committed 2 years ago

f11cce30

[FX] OSS is_fx_tracing (#80255)

RenfeiChen-FB committed 2 years ago

ee0e8361

`permute` for COO sparse tensors (#79707)

nikitaved committed 2 years ago

417677bf

[torchdynamo hash update] update the pinned torchdynamo hash (#80262)

pytorchmergebot committed 2 years ago

3764ffef

AT_DISPATCH: Expose switch-case like macro syntax (#79978)

peterbell10 committed 2 years ago

2c43876f

Fix interpretation torch -> torch._refs in case of nested torch calls under TorchRefsMode (#80135)

IvanYashchuk committed 2 years ago

0d5bc541

Newer Older