comaniac commented on pull request #7220: URL: https://github.com/apache/tvm/pull/7220#issuecomment-755762632
> The PR looks reasonable. I was wondering from time to time, when compiler cache could be potentially useful? I think it's useful to reduce the compile time when building a model? Like we only extract 25 tasks from a model with 50 conv2ds, meaning that we only compile 25 of them instead of 50. ---------------------------------------------------------------- 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]
