wse-research/bert-address-en0.1.1

This is a NER-model trained using the spacy library in Python. It recognizes addresses (postal codes, cities, streets and housenumbers) and uses a separate spacy model to find connected entities. It is based on spacy's English transformer pipeline (roberta-base). You can use it as any other spacy model.

This model was created as part of the Qanary-NER-automl-component's multi result branch. To include it in this component, refer to the corresponding Readme chapter. There are images containing this, and other, models already available to download. A list can be found in the final chapter of the Readme.