mirror of
				https://github.com/explosion/spaCy.git
				synced 2025-10-31 16:07:41 +03:00 
			
		
		
		
	
		
			
				
	
	
		
			23 lines
		
	
	
		
			772 B
		
	
	
	
		
			Python
		
	
	
	
	
	
			
		
		
	
	
			23 lines
		
	
	
		
			772 B
		
	
	
	
		
			Python
		
	
	
	
	
	
| # Load NER
 | |
| from __future__ import unicode_literals
 | |
| import spacy
 | |
| import pathlib
 | |
| from spacy.pipeline import EntityRecognizer
 | |
| from spacy.vocab import Vocab
 | |
| 
 | |
| def load_model(model_dir):
 | |
|     model_dir = pathlib.Path(model_dir)
 | |
|     nlp = spacy.load('en', parser=False, entity=False, add_vectors=False)
 | |
|     with (model_dir / 'vocab' / 'strings.json').open('r', encoding='utf8') as file_:
 | |
|         nlp.vocab.strings.load(file_)
 | |
|     nlp.vocab.load_lexemes(model_dir / 'vocab' / 'lexemes.bin')
 | |
|     ner = EntityRecognizer.load(model_dir, nlp.vocab, require=True)
 | |
|     return (nlp, ner)
 | |
| 
 | |
| (nlp, ner) = load_model('ner')
 | |
| doc = nlp.make_doc('Who is Shaka Khan?')
 | |
| nlp.tagger(doc)
 | |
| ner(doc)
 | |
| for word in doc:
 | |
|     print(word.text, word.orth, word.lower, word.tag_, word.ent_type_, word.ent_iob)
 |