Jan-Christoph Klie

You could try to add the `config` parameter to https://github.com/inception-project/inception-external-recommender/blob/master/ariadne/contrib/adapters.py#L169

I fixed some things and pushed a new example, can you try again?

You need to find out what the labels and ids of the adapter are and change the part where we map numerical ids to string labels.

I think it is https://github.com/inception-project/inception-external-recommender/blob/master/ariadne/contrib/adapters.py#L159 I assumed that label ids start from 0

We already use adapter for your transformer needs. I am currently working on v2 of this project, so I do not want to implement this and then archive the project.

Adapters are a way to efficiently finetune transformers. You do not fine tune the whole thing, but insert a small layer and finetune that. https://adapterhub.ml/ https://github.com/inception-project/inception-external-recommender/blob/master/ariadne/contrib/adapters.py

Thank you for the PR! The basics looks good to me. This code still has the issue that you do not use the tokenization of INCEpTION, which requires you then...

Spacy sadly right now cannot give reasonable confidence scores

Yes, that is ok. We right now dont build external recommender though and would need Jenkins dependency magic to always build them when INCEpTION changes. That would maybe slow our...

Hi, from the code you provided, it seems like you assign the term by yourself, `named_entity[2]` has to be the term text because otherwise your assert would fail. cas_named_entity =...