datactive / bigbang

Scientific analysis of collaborative communities
http://datactive.github.io/bigbang/
MIT License
149 stars 52 forks source link

Adding entity recognition and example notebook #554

Closed effyli closed 2 years ago

effyli commented 2 years ago

Adding a named entity recognition method with huggingface. Adding an example notebook on how to use the recognizer with emails and visualize the recognized entities.

sbenthall commented 2 years ago

Thanks for this PR!

The failing test seems to be unrelated -- it's due to deprecation in some dependencies: https://github.com/scverse/anndata/commit/7b1077d5e1a63fef0d47d716d19ea14ad2ad10de

Once I've gotten this test fixed I'll merge this PR.

sbenthall commented 2 years ago

Does this code have any additional dependencies not yet included in requirements.txt?

sbenthall commented 2 years ago
import transformers[torch]
import contractions
import spacy
Christovis commented 2 years ago

As this is still in an early experimental stage, I would opt for not merging it for now and wait until this work has more matured.