ERROR: Create an Entity Ruler

[judmesa]

Hi, I would like to ask for your help because now I am trying to create and entity ruler but I am having the same error

-------------Code--------------
import os
import spacy
from pptx import Presentation
from spacy.matcher import Matcher
from spacy.matcher import PhraseMatcher
from spacy.lang.en import English
from spacy.pipeline import EntityRuler
import json

nlp=spacy.load("en_core_web_md")
#Inicializa el matcher con el vocabulario compartido
phrase_matcher = PhraseMatcher(nlp.vocab)
.
.
.
def generate_rules(patterns):
nlp=English()
ruler=EntityRuler(nlp)
ruler.add_patterns(patterns)
nlp.add_pipe('ruler')
nlp.to_disk("Stations_NER")

Error:

ValueError: [E002] Can't find factory for 'ruler' for language English (en). This usually happens when spaCy calls nlp.create_pipe with a custom component name that's not registered on the current language class. If you're using a Transformer, make sure to install 'spacy-transformers'. If you're using a custom component, make sure you've added the decorator @Language.component (for function components) or @Language.factory (for class components).

Available factories: attribute_ruler, tok2vec, merge_noun_chunks, merge_entities, merge_subtokens, token_splitter, doc_cleaner, parser, beam_parser, entity_linker, ner, beam_ner, entity_ruler, lemmatizer, tagger, morphologizer, senter, sentencizer, textcat, spancat, textcat_multilabel, en.lemmatizer

[pmbaumgartner]

Could you try adding the component with the name "entity_ruler" and not "ruler", like this:

nlp.add_pipe("entity_ruler")

There's also a self-contained example of the entity ruler component in the docs here you can use for reference.

If that doesn't work, it looks to me like you've excluded some of the code you might be using. It's not clear to me how generate_rules should be working without seeing how you're calling the function and what might exist prior to that call. Could you include a more complete snippet of your code to help us figure out what the issue might be?

[darrkj]

It looks like you are using both methods to construct an entity ruler. The first is ruler=EntityRuler(nlp) and the second is nlp.add_pipe('ruler'). As stated in the other response the add_pipe method should receive the factory name for the component and not a local variable name, thus entity_ruler instead of ruler.

From that snippet you have I think this would resolve the error and also remove one of the two creation methods.

nlp=English()
ruler=nlp.add_pipe('entity_ruler')
ruler.add_patterns(patterns)

I made this notebook so that I can refer back to some of the nuance around the two methods that can create an entity_ruler if its helpful.

ent_ruler_test.pdf