strisunshinewentingwang/pytorch

Pull Requests Commits

[c10d] add ncclGetLastError to NCCL pg (#83724)

wanchaol committed 2 years ago

976f8bee

Don't detach when making views; force caller to detach (#84893)

ezyang committed 2 years ago

ccade941

Revert "[CUBLAS][CUDA GRAPHS] Explicitly set the workspace for cuBLAS handles (#83461)"

pytorchmergebot committed 2 years ago

2711b9fa

Fix inheritance in TestDataLoaderUtil (#85018)

ZainRizvi committed 2 years ago

a1a95d40

[CUBLAS][CUDA GRAPHS] Explicitly set the workspace for cuBLAS handles (#83461)

eqy committed 2 years ago

713d8b85

Enable manual test config label selection on GHA macos (#84895)

huydhn committed 2 years ago

7b64c885

Revert "[numpy] add `torch.concatenate`, alias of torch.cat (#82946)"

pytorchmergebot committed 2 years ago

fa7bf3e2

Shard distributed tests on non CUDA focal (#84891)

huydhn committed 2 years ago

23b7a5fc

Ensure as_strided_tensorimpl is never called with MPS (#85020)

ezyang committed 2 years ago

3e57c955

[CUDA graphs] Fixes errors in RNG seed (#84967)

Aidyn-A committed 2 years ago

5271494e

[numpy] add `torch.concatenate`, alias of torch.cat (#82946)

khushi-411 committed 2 years ago

270e5e51

Revert "Create Cache for Fusion Reuse in NVFuser in Python Frontend for Primtorch (#83267)"

pytorchmergebot committed 2 years ago

94b67f4c

[MPS] Fix mps to cpu casting from a smaller dtype to a bigger dtype (#84928)

DenisVieriu97 committed 2 years ago

4247cc98

Test tracing consecutive comms on the same input tensor (#84980)

mrshenli committed 2 years ago

1a81ab3b

Remove eager mode support form CommTensor (#84978)

mrshenli committed 2 years ago

0f300592

[ROCM] test_batchnorm_cudnn_nhwc (#84603)

micmelesse committed 2 years ago

b6d6a78c

Revert "Python Dispatcher integration with C++ dispatcher (#84826)"

pytorchmergebot committed 2 years ago

706b9903

[2/N] [Dispatchable Collectives] Extract ProcessGroup::Work into a separate class and update references (#83680)

H-Huang committed 2 years ago

74ead619

Upgrade to CUDNN version for cuda 11.7 (#84964)

atalman committed 2 years ago

54c46e4f

Skip validate_view_consistency for nvFuser tests (#84858)

IvanYashchuk committed 2 years ago

6750946b

Python Dispatcher integration with C++ dispatcher (#84826)

voznesenskym committed 2 years ago

35f6a691

[quant][docs] Add example for the error message for fixed qparam ops (#84666)

jerryzh168 committed 2 years ago

44c30c5d

Remove enable_recursive_torch_dispatch (#84945)

ezyang committed 2 years ago

55ca297d

Removes unnecessary namespace of functions used only in einsum (#84955)

janeyx99 committed 2 years ago

922560b8

[vision hash update] update the pinned vision hash (#84975)

pytorchmergebot committed 2 years ago

d26e9cd9

[torchdynamo hash update] update the pinned torchdynamo hash (#84912)

pytorchmergebot committed 2 years ago

b28d82cb

Fix `expectedFailureMeta` to avoid skipping tests (#84875)

kurtamohler committed 2 years ago

d00cabae

Avoid nested CommTensor wrapping (#84963)

mrshenli committed 2 years ago

8cbbd3a2

Revert "Don't detach when making views; force caller to detach (#84893)"

pytorchmergebot committed 2 years ago

8ca057eb

[ROCm] re-enable tensorexpr and test_openmp (#81367)

arindamroy-eng committed 2 years ago

a185dc2e

Newer Older