FrozenGene commented on pull request #10650: URL: https://github.com/apache/tvm/pull/10650#issuecomment-1070537214
@DzAvril Thanks for detail experiment and explaining. As autotvm tuning conv2d, it then will call `traverse_inline` to fuse injective ops in your figure. So the miss information is conv2d's schedule is good but fuse it get worse performance? -- 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]
