adding ner model to existing spacy model give empty response of tag_ and dep_

, ,
1

Hi, I have added ner model to existing model but having some issues. Here is how I added the ner to existing model.

import spacy
nlp = spacy.load("ur_model")
# nlp.remove_pipe("ner")
print(nlp.pipe_names)  # ['tagger', 'parser']

nlp_entity = spacy.load("ur_ner")

# Get the ner pipe from this model and add it to base model
nlp.add_pipe("ner", source=nlp_entity)
print(nlp.pipe_names)  # ['tagger', 'parser', 'ner']

nlp.to_disk("./ur_ner")

Now when I use this model tagger and parser are showing empty string instead of tags.

import spacy
nlp = spacy.load("ur_ner")
for token in doc:
    print(token.pos_, token.ent_type_, token.dep_)
 Other 
 Other 
 Other 
 Location 
 Other 
 Other 
 Other 
 Other
 Other 
 Other 
 Other

But my exising model used give tags on both tag_ and deps_

nlp = spacy.load("ur_model")
doc = nlp("پانچواں حصہ تحریک پاکستان کے نامور افراد کے ذکر پر مشتمل")
for token in doc:
    print(token.pos_, token.ent_type_, token.dep_)
    
ADJ  amod
NOUN  nsubj
PROPN  compound
PROPN  nmod
ADP  case
ADJ  amod
NOUN  nmod
ADP  case
NOUN  obl
ADP  case
ADJ  ROOT
2

Hi Muhammad,

you seem to be asking a question that's primarily about spaCy, not Prodigy. For these kinds of questions, it'd be better to check the spaCy discussion forum. The maintainers of the spaCy keep an eye on that forum and that way the larger community can also find your question.