pandalow commited on
Commit
cb0b221
·
1 Parent(s): c81d8de

fix: font

Browse files
Files changed (1) hide show
  1. app.py +12 -0
app.py CHANGED
@@ -1,10 +1,21 @@
1
  import re
 
 
 
2
  import jieba
3
  from PIL import Image
4
  from wordcloud import WordCloud
5
  import gradio as gr
6
  from typing import List
7
 
 
 
 
 
 
 
 
 
8
 
9
  def tokenize_text(text:str) -> List:
10
  """
@@ -40,6 +51,7 @@ def generate_world_cloud(text:str) -> Image.Image:
40
  segmented = tokenize_text(text)
41
 
42
  wc = WordCloud(
 
43
  width=800,
44
  height=400,
45
  background_color='white'
 
1
  import re
2
+ import os
3
+ import requests
4
+
5
  import jieba
6
  from PIL import Image
7
  from wordcloud import WordCloud
8
  import gradio as gr
9
  from typing import List
10
 
11
+ FONT_PATH = "NotoSansCJK-Regular.otf"
12
+ FONT_URL = "https://github.com/adobe-fonts/source-han-sans/raw/release/OTF/SimplifiedChinese/SourceHanSansSC-Regular.otf"
13
+
14
+
15
+ if not os.path.exists(FONT_PATH):
16
+ print("Downloading font...")
17
+ with open(FONT_PATH, "wb") as f:
18
+ f.write(requests.get(FONT_URL).content)
19
 
20
  def tokenize_text(text:str) -> List:
21
  """
 
51
  segmented = tokenize_text(text)
52
 
53
  wc = WordCloud(
54
+ font_path=FONT_PATH,
55
  width=800,
56
  height=400,
57
  background_color='white'