spaCy/examples/training/train_ner.py

70 lines
2.2 KiB
Python
Raw Normal View History

from __future__ import unicode_literals, print_function
import random
from spacy.lang.en import English
2017-05-31 11:42:12 +00:00
from spacy.gold import GoldParse, biluo_tags_from_offsets
2017-01-27 11:27:10 +00:00
2017-05-31 11:42:12 +00:00
def reformat_train_data(tokenizer, examples):
"""Reformat data to match JSON format"""
output = []
for i, (text, entity_offsets) in enumerate(examples):
doc = tokenizer(text)
ner_tags = biluo_tags_from_offsets(tokenizer(text), entity_offsets)
words = [w.text for w in doc]
tags = ['-'] * len(doc)
heads = [0] * len(doc)
deps = [''] * len(doc)
sentence = (range(len(doc)), words, tags, heads, deps, ner_tags)
output.append((text, [(sentence, [])]))
return output
def main(model_dir=None):
train_data = [
(
'Who is Shaka Khan?',
[(len('Who is '), len('Who is Shaka Khan'), 'PERSON')]
),
(
'I like London and Berlin.',
[(len('I like '), len('I like London'), 'LOC'),
(len('I like London and '), len('I like London and Berlin'), 'LOC')]
)
]
nlp = English(pipeline=['tensorizer', 'ner'])
2017-05-31 11:42:12 +00:00
get_data = lambda: reformat_train_data(nlp.tokenizer, train_data)
optimizer = nlp.begin_training(get_data)
for itn in range(100):
random.shuffle(train_data)
losses = {}
for raw_text, entity_offsets in train_data:
doc = nlp.make_doc(raw_text)
gold = GoldParse(doc, entities=entity_offsets)
nlp.update(
[doc], # Batch of Doc objects
[gold], # Batch of GoldParse objects
drop=0.5, # Dropout -- make it harder to memorise data
sgd=optimizer, # Callable to update weights
losses=losses)
print(losses)
print("Save to", model_dir)
nlp.to_disk(model_dir)
print("Load from", model_dir)
nlp = spacy.lang.en.English(pipeline=['tensorizer', 'ner'])
nlp.from_disk(model_dir)
for raw_text, _ in train_data:
2017-05-31 11:42:12 +00:00
doc = nlp(raw_text)
for word in doc:
print(word.text, word.ent_type_, word.ent_iob_)
if __name__ == '__main__':
2017-05-31 11:42:12 +00:00
import plac
plac.call(main)
# Who "" 2
# is "" 2
# Shaka "" PERSON 3
# Khan "" PERSON 1
# ? "" 2