詞雲 wordcloud
阿新 • • 發佈:2020-09-16
import jieba # 分詞 import numpy as np # 矩陣運算 import pymysql from PIL import Image # 圖片處理 from matplotlib import pyplot as plt # 繪圖,資料視覺化 from wordcloud import WordCloud # 詞雲 conn = pymysql.connect(host="localhost", port=3306, user="pig", passwd="123456", db="runoob", charset='utf8') cursor = conn.cursor() cursor.execute('select `inq` from `t_movie`') data = cursor.fetchall() text = ' '.join([_[0] for _ in data]) cursor.close() conn.close() fenci_text = jieba.cut(text, cut_all=False) # words = [] # for t in fenci_text: # if t not in words: # words.append(t) print(text) print('-' * 30) show_text = ' '.join(fenci_text) print(show_text) img = Image.open(r'./static/img/123.jpg') img_array = np.array(img) wc = WordCloud(background_color='white', # 設定背景色 mask=img_array, # 設定背景圖片 font_path=r'C:\Windows\Fonts\FZSTK.TTF', # 字型,微軟雅黑一直找不到 collocations=False, # 避免重複單詞 max_words=1000, # 詞雲顯示的最大詞數 max_font_size=500, # 字型最大值 min_font_size=20, # 字型最小值 ) wc.generate_from_text(show_text) # 繪製圖片 fig = plt.figure(1) plt.imshow(wc) plt.axis('off') # 隱藏座標軸 plt.show() wc.to_file('ttt.png') #儲存圖片