mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 12:21:27 +01:00
f4d8bc46c7
### Description This PR is to enable TF32 as fp32 internal precision for matmul/linear/conv in `mkldnn backend`. Since we have refined fp32 precision API in https://github.com/pytorch/pytorch/pull/125888, we can easily extend the API to support TF32 for `mkldnn backend`. ``` torch.backends.mkldnn.matmul.fp32_precision = 'tf32' torch.backends.mkldnn.conv.fp32_precision = "tf32" ``` Related kernel update and UTs update are done. And the wrapper `bf32_on_and _off` is updated to `reduced_f32_on_and_off`, and it can run tests 3 times, one is reduced_f32 OFF, the other two are reduced_f32 ON (including `bf32 ON` and `tf32 ON`). Pull Request resolved: https://github.com/pytorch/pytorch/pull/157520 Approved by: https://github.com/mingfeima, https://github.com/jansel |
||
|---|---|---|
| .. | ||
| amp_examples.rst | ||
| autograd.rst | ||
| broadcasting.rst | ||
| cpu_threading_torchscript_inference.rst | ||
| cuda.rst | ||
| custom_operators.rst | ||
| ddp.rst | ||
| extending.func.rst | ||
| extending.rst | ||
| faq.rst | ||
| fsdp.rst | ||
| get_start_xpu.rst | ||
| gradcheck.rst | ||
| hip.rst | ||
| large_scale_deployments.rst | ||
| libtorch_stable_abi.md | ||
| mkldnn.rst | ||
| modules.rst | ||
| mps.rst | ||
| multiprocessing.rst | ||
| numerical_accuracy.rst | ||
| out.rst | ||
| randomness.rst | ||
| serialization.rst | ||
| windows.rst | ||