search

elsevierlabs-os / nerds

BSD 3-Clause "New" or "Revised" License
27 stars 11 forks source link

About applying NER on prediction for single sentences #7

Open marcusau opened 4 years ago

marcusau commented 4 years ago

Hi,

Thanks for a nice piece of work. Highly appreciated.

You mention in "nerds/nerds/models/bert.py" :

           NOTE: this is an experimental NER that did not perform very well, and
        is only here for reference purposes. It has been superseded by the
        TransformerNER model, which offers the same functionality (and improved
        performance) not only with BERT as the underlying language model (as this 
        one does), but allows other BERT-like language model backends as well.

May I ask where i can find the TransformerNER model?

Thanks a lot

sujitpal commented 4 years ago

@marcusau both the BertNER (bert.py) and TransformerNER (transformer.py) are in the same models directory, but currently only in my fork of the project (https://github.com/sujitpal/nerds). I am working with the core developers of this project to have these models folded in here.