ChaiBapchya opened a new pull request #16104: Faster Transpose 2D URL: https://github.com/apache/incubator-mxnet/pull/16104 ## Description ## Faster 2D transpose ## Checklist ## ### Essentials ### Please feel free to remove inapplicable items for your PR. - [x] Changes are complete (i.e. I finished coding on this PR) - [x] All changes have test coverage: - [x] Code is well-documented: - [x] To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change ### Changes ### - [ ] Bypass expression template way of execution - [ ] Implement cache-optimal transpose ## Performance ## | Shape | Branch | MKL | | P50/Median | P90 Time | P99 Time | Avg Time | |----------------|---------------|-----|-------|------------|------------|------------|------------| | (1024, 1024) | new transpose | off | int64 | 2.97046 | 3.02903 | 5.23159 | 3.12827 | | | master | off | int64 | 7.87614 | 7.92953 | 10.18385 | 8.04834 | | (10000, 10000) | new transpose | off | int64 | 804.39574 | 836.96015 | 858.47099 | 812.50145 | | | master | off | int64 | 2973.75703 | 3045.19813 | 3070.58677 | 2980.31202 | | (1024, 1024) | new transpose | off | int32 | 2.97993 | 3.04809 | 5.3323 | 3.14604 | | | master | off | int32 | 7.417 | 7.47346 | 9.63628 | 7.58372 | | (10000, 10000) | new transpose | off | int32 | 806.47784 | 845.70507 | 868.55227 | 815.48621 | | | master | off | int32 | 2941.61317 | 3007.3768 | 3053.5961 | 2966.09583 |
---------------------------------------------------------------- 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: us...@infra.apache.org With regards, Apache Git Services
