python - 制作简单 '词云图'

import jieba
from jieba.analyse import extract_tags
from scipy.misc import imread
from wordcloud import WordCloud

import matplotlib.pyplot as plt

def generater(**kwargs):
    content_name = kwargs['content_name']
    content_path = './{}.txt'.format(content_name)
    top_k = kwargs['top_k']
    bg_name = kwargs['bg_name']
    color = kwargs['color']
    font_type = kwargs['font_type']

    with open(content_path, mode='r', encoding='utf-8') as f:
        content = f.read()
    '''
    根据TF/IDF提取topK个关键词
    '''
    tags = extract_tags(sentence=content, topK=top_k)
    '''
    得到关键词的词频
    '''
    # 全模式
    words = [word for word in jieba.cut(content, cut_all=True)]
    words_freq = {}
    for tag in tags:
        freq = words.count(tag)
        words_freq[tag] = freq
    '''
    设置背景
    scipy.misc imread()：返回的是 numpy.ndarray 也即 numpy 下的多维数组对象
    '''
    bg_path = './{}.png'.format(bg_name)
    bg_img = imread(bg_path)
    font_path = './{}.ttf'.format(font_type)
    word_cloud = WordCloud(font_path=font_path, # 设置字体
                           background_color=color, # 背景颜色
                           max_words=top_k, # 词云显示的最多词数
                           max_font_size=100, # 字体最大
                           mask=bg_img, # 背景图
                           )
    word_cloud.generate_from_frequencies(words_freq)

    plt.imshow(word_cloud)
    plt.axis('off') # 不显示坐标轴
    plt.show()

    # 保存图片
    word_cloud_img = './{}_word_cloud.jpg'.format(content_name)
    word_cloud.to_file(word_cloud_img)

if __name__ == '__main__':
    generater(content_name='习近平：在庆祝中国共产党成立95周年大会上的讲话',
              top_k=66,
              bg_name='china', # 默认png
              color='black',
              font_type='wryh')

链接: https://pan.baidu.com/s/1bp8dPa3密码: ujxc

windows 安装 wordcloud

下载地址

pip install  wordcloud-1.3.1-cp36-cp36m-win_amd64.whl