Model card and evaluation performance.
Hi there,
Since this is the large implementation of this model would it be possible to have an equivalent model card and performance evaluation on the SQuAD 2.0 like the base model does? There isn't much out there in terms of comparing performance so this would be very helpful, thanks!
Charles
I agree that it would be awesome. Let us know if we can help!
Hi 馃憢馃徑 I agree - I'll take the time to create a model card for this one too. I'm planning on making my way through various models we have at deepset. Started with roberta-base-squad2 and tinyroberta-squad2. Will try to get to this as soon as possible :) Will let you know here when I have something!
Thank you very much for this, it's very appreciated. If there is anything I can do to help let me know, happy to contribute.
Hey everyone, we've added evaluation results on Squad 2 along with other datasets!