Further train NER model from existing Model

[mitro02 (AL)]

Hi all,

since the commands have changed a bit, I am confused how to load an existing NER Model in Prodi.gy‘s ner train workflow.

Training a NER-Model from scratch (with spaCy‘s Blank:de model) is working fine. I‘d like to load this NER-Model in the next step to train it further with another annotated dataset (Same Entity).

How should I use the ner train Command to load an existing NER model?

[ines (Ines Montani)]

Hi! I just replied on your other thread here:

Commands for training NER-Model in prodigy

In that case, you can just use the --base-model argument, e.g. --base-model /path/to/your/model. In general, we'd recommend retraining a model on all annotations from scratch, rather than updating a previously trained artifact, since this will give you better and more reliable results. But the base model setting can still be useful if you want to update one of the trained pipelines provided by spaCy etc.