python - 歌词词云3

生成词云需要使用wordcloud库，
可以通过以下步骤对文件夹内多个txt格式文件分别进行生成词云

import os
from wordcloud import WordCloud
import imageio  # 图像模块

img = imageio.imread('D:/cat.png')
folder_path = 'G:/lyric/test'
output_path = 'G:/lyric/test/testpic'

for file_name in os.listdir(folder_path):
    if file_name.endswith('.txt'):
        file_path = os.path.join(folder_path, file_name)
        with open(file_path, 'r') as f:
            text = f.read()
        wc = WordCloud(font_path='msyh.ttc',
                       background_color='white',mask=img,
                       max_words=2000)
        wc.generate(text)
        output_file = os.path.join(output_path, file_name.replace('.txt', '.png'))
        wc.to_file(output_file)

存在问题：长句统计在了词云中

解决方式：引入中文分词库jieba，使用cn_stopwords.txt 停用词文本更改设置词云图

import jieba  # 中文分词库
    # 使用jieba进行分词
    words = " ".join(jieba.cut(text))

import os
from wordcloud import WordCloud
import imageio  # 图像模块
import jieba  # 中文分词库

img = imageio.imread('D:/cat.png')
folder_path = 'G:/lyric/test'
output_path = 'G:/lyric/test/testpic'

for file_name in os.listdir(folder_path):
    if file_name.endswith('.txt'):
        file_path = os.path.join(folder_path, file_name)
        with open(file_path, 'r') as f:
            text = f.read()
            txt_list = jieba.lcut(text)
            string = ' '.join(txt_list)    # 使用jieba进行分词words = " ".join(jieba.cut(text))
        wc = WordCloud(font_path='msyh.ttc',
                       background_color='white',mask=img,
                       max_words=2000,
                       stopwords=set([line.strip() for line in open('cn_stopwords.txt', mode='r',
                                                 encoding='utf-8').readlines()])
                       )
        wc.generate(string)
        output_file = os.path.join(output_path, file_name.replace('.txt', '.png'))
        wc.to_file(output_file)

修改后效果优于之前，长句不再出现在词云中