0
点赞
收藏
分享

微信扫一扫

word2vec生成词向量

先预处理,进行分词等

import jieba
from gensim.models import Word2Vec
stopword = [line.strip() for line in open('password.txt', 'r',encoding='utf-8').readlines()]
def seg_sentence(sentence):
    """"进行分词"""
    sentence_seged = jieba.cut(sentence.strip())  #分词
    stopwords = [' ']
    # # stopwords = stopwordslist('E:\\pythonimg\\stopword.txt')  # 这里加载停用词的路径  这里可以再加自定义的停用词
    outstr = ''   # 必须字符,不能列表
    for word in sentence_seged:
        if word not in stopwords:
            if word != '\t':
                outstr += word
                outstr += " "
    # return outstr
    return outstr.split(' ')  # 以空格分割 列表

 生成词向量

def vec_produce(sentence,word,size):
    """生成词向量"""
    sentenceseg = seg_sentence(sentence) # 已分词可向量化的句子
    model = Word2Vec(sentences=[sentenceseg], vector_size=size, window=5, min_count=1, workers=4)
    word_vectors = model.wv
    wordvec = word_vectors[word]
    return wordvec
举报

相关推荐

0 条评论