agrabows opened a new pull request #20932: URL: https://github.com/apache/incubator-mxnet/pull/20932
## Description ## Elementwise binary broadcast operators such as add, subtract, multiply, divide in MXNet NumPy module support mixed inputs of float and integer type. Goal of this change is to add support of those operators for bfloat with fp32 and fp64 mixed input cases. Additionally oneDNN will be used when processing fp32 and bfloat cases. ## Checklist ## ### Essentials ### - [x] PR's title starts with a category (e.g. [BUGFIX], [MODEL], [TUTORIAL], [FEATURE], [DOC], etc) - [x] Changes are complete (i.e. I finished coding on this PR) - [x] All changes have test coverage ### Changes ### - [x] Add support for bfloat16 and float32 through oneDNN primitive and without oneDNN. - [x] Add support for bfloat16 and float64. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
