@Mousius In this case, @YuchenJin 's reply clearly articulated that there is a close co-design of these factors, and changing to adopt dynamic alone would imply a one-step jump to relax -- which is not incremental. The data structure change would come with a set of supporting infra and co-design of things including shape inference, and other things.
Of course, if we do not maintain both behavior and allow an incremental transition. Then it is equivalent to a relay=>relax bridge that would allow part of the pipeline to go through in a more incremental fashion. This approach is consistent with what is being proposed in the RFC (with clear data structure boundaries). I would suggest us not to trivialize the argument here, as it is easier to say why not do X than really go and do things. The learnings from @YuchenJin by concrete code certainly is very convincing, as well as their learnings of why not do things otherwise. This is a case where there are subjective disagreements happens. In such cases, I would encourage again for us to put community into consideration, and the fact that things are not disrupting other existing flows -- Reply to this email directly or view it on GitHub: https://github.com/apache/tvm-rfcs/pull/89#issuecomment-1266970803 You are receiving this because you are subscribed to this thread. Message ID: <apache/tvm-rfcs/pull/89/c1266970...@github.com>
