Training NER在Google Colab上使用spacy

!pip install -U spacy
import spacy
from spacy.util import minibatch, compounding

train_data = [
    ("Apple is looking to buy U.K. startup for $1 billion", {"entities": [(0, 5, "ORG")]}),
    ("Microsoft acquires GitHub for $7.5 billion", {"entities": [(0, 9, "ORG")]}),
    # 添加更多的训练数据
]

nlp = spacy.blank("en")

ner = nlp.create_pipe("ner")
nlp.add_pipe(ner, last=True)

ner.add_label("ORG")

n_iter = 10
for _ in range(n_iter):
    losses = {}
    random.shuffle(train_data)
    batches = minibatch(train_data, size=compounding(4.0, 32.0, 1.001))
    for batch in batches:
        texts, annotations = zip(*batch)
        nlp.update(texts, annotations, losses=losses)
    print("Losses:", losses)

nlp.to_disk("trained_ner_model")