Add custom NER model from prodigy to spacy pipeline - spaCy V3

Hi, I ran into the same problem as shown here .
The task is the same as the former question: to add a prodigy trained NER to the spacy pipeline.
My code:

The error:
Screenshot 2022-10-05 at 23.47.04

Do you know how to make it work? Looking forward to your reply.

hi @ruiyeNLP!

Thanks for your question.

Can you explain what is wrong with your /model-last/? Does it not have the transformer layer instead of a tok2vec?

For example, what the output of this:

print(trained_model.pipe_names)

If it's just the ['transformer', 'ner'] then you should be good (it's tough to see if that's the case from the image you provided).

If it's not, can you provide what was your recipe to train /model-last/ (or the config.cfg you provided)?

Moving forward, could you refrain from posting images on the Forum? You can simply copy/paste and render code in Markdown. This makes your code (e.g., certain errors) searchable for others. We greatly appreciate it :slight_smile: