MoisesHer opened a new pull request #18921: URL: https://github.com/apache/incubator-mxnet/pull/18921
## Description ## Implementation of einsum on GPU though [cuTensor](https://developer.nvidia.com/cutensor) library ## Checklist ## ### Essentials ### - [ ] Changes are complete (i.e. I finished coding on this PR) - [x] All changes have test coverage: - Unit tests are added for different einsum scenarios - [x] Code is well-documented: - [x] To the best of my knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change ### Changes ### - [x] Added a new implementation of einsum operator though cuTensor library ## Comments ## TODO: - Include Broadcast cases - Include Identity cases - Include single operator with cuTensor - Infer right part of equation - Doubts about operator signature: same signature for FWD & BWD? - Cache vs. Signature approaches (alignment at descriptors creation) ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected]
