word2vec简单使用

word2vec

from gensim.models import word2vec
import logging

logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)
#加载text8语料库
sentences = word2vec.Text8Corpus('text8')
#建模
model = word2vec.Word2Vec(sentences, min_count=1)
#模型保存
model.save('text8.model')
#模型加载
model1 = word2vec.Word2Vec.load("text8.model")

#测试
print (model.similarity('dogs','cats'))
print (model.most_similar('dog')
print (model1.most_similar(positive=['woman', 'king'], negative=['man'], topn=1))
print (model1.doesnt_match("breakfast cereal dinner lunch";.split()))