spacy-annotator
SpaCy annotator for Named Entity Recognition (NER) using ipywidgets. The annotator allows users to quickly assign (custom) labels to one or more entities in the text.
Features:
- The annotator supports pandas dataframe (see
pandas_annotations.py): it adds annotations in a separate 'annotation' column of the dataframe; - If a spaCy model is passed into the annotator, the model is used to identify entities in text.
This trick of pre-labelling the example using the current best model available allows for accelerated labelling - also known as of noisy pre-labelling; - The annotations adhere to spaCy format and are ready to serve as input to a spaCy NER model.
No additional code required!
Note: not using pandas dataframe? No problem. You can always label entities from text stored in a simple python list.
Blog post: medium/enrico.alemani/spacy-annotator
Example
For code, see spacy_annotator demo notebook.
Contributing
- Fork the repo on GitHub;
- Clone the project to your own machine;
- Commit changes to your own branch; and
- Push your work back up to your own fork;
- Submit a Pull request so that I can review your changes.
Dependencies
Spacy-annotator works with SpaCy 3.X, and ipywidgets 7.X.
References
spacy-annotator is based on spaCy and pigeon (see also PigeonXT).
Many thanks to them for making their awesome libraries publicly available. Another interesting project is Doccano.
Note: spaCy is a great library and, most importantly, free to use. So please also consider using the https://prodi.gy/ annotator to keep supporting the spaCy deveopment.