comaniac commented on pull request #9261: URL: https://github.com/apache/tvm/pull/9261#issuecomment-942878845
Right. Probably just need some experiments to see how many kernels should be generated to cover a range of batch sizes, and we may just hardcore the dispatching logic first to make it work. With these restrictions for now, C source codegen is also doable and it also has less overhead. -- 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]
