benchmark

benchmark copied to clipboard

TorchBench CI has detected a performance signal. Base PyTorch version: 2.4.0.dev20240606+cu121 Base PyTorch commit: ffaea656b5d8ff6518669494cc8f664b94f8e8b1 Affected PyTorch version: 2.6.0.dev20240916+cu124 Affected PyTorch commit: c64ae601ba9eb3ad2cd3402a14f6ac83c0ab7eba Affected Tests: - test_train[Super_SloMo-cuda-eager]: +14.62303% - test_eval[Super_SloMo-cuda-eager]:...

xuzhao9

TorchBench CI has detected a performance signal. Base PyTorch version: 2.4.0.dev20240606+cu121 Base PyTorch commit: ffaea656b5d8ff6518669494cc8f664b94f8e8b1 Affected PyTorch version: 2.6.0.dev20240915+cu124 Affected PyTorch commit: c64ae601ba9eb3ad2cd3402a14f6ac83c0ab7eba Affected Tests: - test_train[Super_SloMo-cuda-eager]: +14.62265% - test_eval[Super_SloMo-cuda-eager]:...

xuzhao9

TorchBench CI has detected a performance signal. Base PyTorch version: 2.4.0.dev20240606+cu121 Base PyTorch commit: ffaea656b5d8ff6518669494cc8f664b94f8e8b1 Affected PyTorch version: 2.6.0.dev20240914+cu124 Affected PyTorch commit: 081c4a966de77dd5187c1cb97d287cae6f3fbc02 Affected Tests: - test_train[Super_SloMo-cuda-eager]: +14.62073% - test_eval[Super_SloMo-cuda-eager]:...

xuzhao9

TorchBench CI has detected a performance signal. Base PyTorch version: 2.4.0.dev20240606+cu121 Base PyTorch commit: ffaea656b5d8ff6518669494cc8f664b94f8e8b1 Affected PyTorch version: 2.6.0.dev20240913+cu124 Affected PyTorch commit: 0cdc6a8dcd7e294b01d8914385bbe45e79c1770d Affected Tests: - test_train[Super_SloMo-cuda-eager]: +14.63326% - test_eval[Super_SloMo-cuda-eager]:...

xuzhao9

[Inductor][Optimus] Fix group fusion stride layout

5

Summary: context: https://fb.workplace.com/groups/1075192433118967/permalink/1401282167176657/ moving the changes to the group gemm op has compilation errors, see details in D55606636 Differential Revision: D61888433

mengluy0125

TorchBench CI has detected a performance signal. Base PyTorch version: 2.4.0.dev20240606+cu121 Base PyTorch commit: ffaea656b5d8ff6518669494cc8f664b94f8e8b1 Affected PyTorch version: 2.5.0.dev20240828+cu124 Affected PyTorch commit: 55236d0cb7d183f2b64982d8718e9012ae8a40d8 Affected Tests: - test_train[Super_SloMo-cuda-eager]: +14.63055% - test_eval[Super_SloMo-cuda-eager]:...

xuzhao9

Fix group fusion stride layout

2

Summary: X-link: https://github.com/pytorch/pytorch/pull/122839 context: https://fb.workplace.com/groups/1075192433118967/permalink/1401282167176657/ moving the changes to the group gemm op has compilation errors, see details in D55606636 Differential Revision: D55449814

mengluy0125

adding vector add to CI

2

Summary: nothing Differential Revision: D61819148

adamomainz

TorchBench CI has detected a performance signal. Base PyTorch version: 2.4.0.dev20240606+cu121 Base PyTorch commit: ffaea656b5d8ff6518669494cc8f664b94f8e8b1 Affected PyTorch version: 2.5.0.dev20240823+cu124 Affected PyTorch commit: 255cd75a974389985a7a542be1455210a9c369e2 Affected Tests: - test_train[Super_SloMo-cuda-eager]: +14.61805% - test_eval[Super_SloMo-cuda-eager]:...

xuzhao9

[torchao] Dashboard numbers are missing even when the workflow is succeeded

14

TorchAO workflow has fixed a bug (https://github.com/pytorch/benchmark/commit/bf0e5a965b9476b58324017dbf025e859894abb9) and should give reasonable numbers right now, the workflow also succeeded: https://github.com/pytorch/benchmark/actions/workflows/torchao.yml However, we do not see numbers are updated on the TorchAO...

xuzhao9