For the textcat.teach command, can I do on a custom model (like the model I trained through ner.train) ?
Like this, can I replace the
blank:en with my model name? Thanks.
Like this, can I replace the
Yes, excatly. The spacy_model argument can either be the name of an installed spaCy model or its path (basically, anything you can load with spacy.load()) or blank: plus the language code, if you want to start with a blank model.
In case you haven't seen it already, the recipe docs also include a more detailed overview of the built-in recipes, plus descriptions of their arguments and examples: Built-in Recipes · Prodigy · An annotation tool for AI, Machine Learning & NLP