You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
|
|
|
|
|
import jieba
|
|
|
|
|
|
import wordcloud
|
|
|
|
|
|
import imageio
|
|
|
|
|
|
# 导入imageio库中的imread函数，并用这个函数读取本地图片，作为词云形状图片
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def generate(file_path):
|
|
|
|
|
|
print("正在生成词云图...")
|
|
|
|
|
|
py = imageio.imread('.\\03.png') # 巴黎铁塔的蒙版图
|
|
|
|
|
|
# 读取文件内容
|
|
|
|
|
|
f = open(file_path, encoding='utf-8')
|
|
|
|
|
|
txt = f.read()
|
|
|
|
|
|
# jiabe 分词 分割词汇
|
|
|
|
|
|
txt_list = jieba.lcut(txt)
|
|
|
|
|
|
string = ' '.join(txt_list)
|
|
|
|
|
|
# 词云图设置
|
|
|
|
|
|
wc = wordcloud.WordCloud(
|
|
|
|
|
|
width=1000, # 图片的宽
|
|
|
|
|
|
height=700, # 图片的高
|
|
|
|
|
|
background_color='white', # 图片背景颜色
|
|
|
|
|
|
font_path='msyh.ttc', # 词云字体
|
|
|
|
|
|
mask=py, # 所使用的词云图片
|
|
|
|
|
|
scale=15,
|
|
|
|
|
|
# contour_color = 'steelblue' ,
|
|
|
|
|
|
contour_color='gold', # 设置轮廓为金色 (使用颜色名称)
|
|
|
|
|
|
contour_width=15, # 轮廓宽度
|
|
|
|
|
|
# 停用词
|
|
|
|
|
|
stopwords={'是', '的', '和', '吗', '这是', '不是', '了', '还', '这个', '哈', '啊'}
|
|
|
|
|
|
)
|
|
|
|
|
|
# 给词云输入文字
|
|
|
|
|
|
wc.generate(string)
|
|
|
|
|
|
search_word = file_path.rsplit(".", 1)[0]
|
|
|
|
|
|
# 保存词云图
|
|
|
|
|
|
wc.to_file(f'{search_word}词云图.png')
|
|
|
|
|
|
print("成功保存词云图！")
|
