You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2021/12/21 20:51:27 UTC

[GitHub] [tvm] masahi commented on pull request #9782: [AMP] Disallow converting layer norm to fp16

masahi commented on pull request #9782:
URL: https://github.com/apache/tvm/pull/9782#issuecomment-999085773


   > Meanwhile, do you think ToMixedPrecision should be applied later than the expression mutation passes (e.g., after SimplifyExpr, SimplifyInference, etc, and before FuseOps), so that we could make sure all ops (types) in the IR are basically fixed?
   
   I thought about making `SimplifyInference` a prereq pass for `ToMixedPrecision`, but I think putting `ToMixedPrecision` later in the pipeline makes it a bit inconvenient to use this pass (because it is fundamentally an optional feature). So it is more natural to invoke `ToMixedPrecision` somewhere in a user script. Always running `SimplifyInference` before `ToMixedPrecision` may also break someone's workflow that depends on a particular pattern be present in a mod.


-- 
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: commits-unsubscribe@tvm.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org