pygaggle
pygaggle copied to clipboard
Add duoBERT
I was looking into adding duo support (k-way ranking potentially) for MSMARCO and TREC-CAR.
For RelevanceExample, I was wondering if we can have documents as Union(List[Text], List[List[Text]]) instead of just List[Text]. I could also do something like this in the evaluate method in RerankerEvaluator. Another way is to add an is_duo argument to the various reranks/class.
Thoughts? @daemon @rodrigonogueira4
Yeah, that sounds good to me! For now, I would worry only about duo and forget about the general case as we don't even know how to make it computationally viable
@wiltan-uw is on this!
I'm wondering what the relationships between this issue and this repo is: https://github.com/castorini/duobert
Are we planning on deprecating the duobert
repo, i.e., but into archive real-only mode?
Yeah, I think archiving duobert repo is a good idea. It is in tensorflow, and it requires some manual steps to do reranking