训练词向量

def word_vector_gener():
    """
    几种不同的方法来生成词向量
    :return:
    """
    from gensim.models import Word2Vec
    from gensim.test.utils import common_texts
    # 1.word2vec
    # 获取原始数据
    DATA_PATH = './word2vec_data.txt'
    word2evctor = open('./word2vector.txt', 'w', encoding='utf8')
    word_list = []
    finall = []
    # jieba分词
    with open(DATA_PATH, 'r', encoding='utf8') as file:
        for each_line in file.readlines():
            # 分词
            cut_word = list(jieba.cut(each_line.strip()))
            # 去停用词
            stopwords = [w.strip() for w in open('./stop_words.txt', 'r', encoding='utf8')]
            temp = []
            for each in cut_word:
                if each not in stopwords and each.strip():
                    temp.append(each)
                    word_list.append(each)
            finall.append(temp)
    # 训练模型
    model = Word2Vec(finall, size=100, window=1, min_count=1, workers=4)
    model.save('./word2vec_model.')
    # 查看词向量
    for word in list(set(word_list)):
        content = str(word) + '\t' + str(model[word])
        word2evctor.write(content+'\n')
        print(content)


    print('ok')


if __name__ == '__main__':
    word_vector_gener()