word2vec如何对关键词进行转换。将关键词转换为向量。句子也同样可以使用。（累加法）

# -*- coding: utf-8 -*-
import gensim.models.word2vec as word2vec
import gensim.models.doc2vec as doc2vec
from nltk.tokenize import LineTokenizer,SpaceTokenizer,TweetTokenizer
from nltk import word_tokenize
import gensim
import logging
import numpy as np
from nltk.corpus import stopwords

# word2vec Text8 的训练
def train_save_model():
    # logging.basicConfig(format='%(asctime)s : %(levelname)s :%(message)',level=logging.INFO)
    # 加载预料
    sentences = word2vec.Text8Corpus('text8')
    model = word2vec.Word2Vec(sentences, size=200)
    model.save('text2020.model')
# 加载模型
def load_model():
    model = word2vec.Word2Vec.load('text2020.model')
    # simi = model.similar_by_vector('women', 'men')
    # print(simi)
    #print(model.most_similar('man'))
    address = 'C:\\Users\\Administrator\\Desktop\\过度用\\keyword.txt'
    f = open(address, 'r', encoding='utf-8')
    z=np.zeros((200,))
    str = ''
    for read in f.readlines():
        read=read.lower()
        #print(read)
        stop_words = stopwords.words('english')
        for readperword in read:
            if readperword!='\n':
                # print(read)
                t = word_tokenize(readperword)
                for k in t:
                    z = z + model[k]
                    # print(count)
            else:
                print(z)
                continue






        str=str+read
        #print(str)
        #for i in str:
            #print(i)


    #print(model(list1))
"""
def get_keyword(path):
    f=open(path,'r',encoding='utf-8')
    list=[]
    for read in f.readlines():
        list.append(read)
        """
# 执行代码
#load_model()
if __name__ == '__main__':
    #train_save_model()



    # 执行代码
    #address='C:\\Users\\Administrator\\Desktop\\过度用\\keyword.txt'
    #get_keyword(address)
    load_model()