svanhvit commited on
Commit
e35b927
1 Parent(s): 97f7764

add app file

Browse files
Files changed (1) hide show
  1. app.py +33 -0
app.py ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+
3
+ import gradio as gr
4
+ from transformers import pipeline
5
+ import tokenizer
6
+
7
+
8
+ title = "Textaleiðrétting fyrir íslensku"
9
+ description = "Hér má prófa tauganetslíkan sem leiðréttir margs konar málfars- og stafsetningarvillur í íslenskum texta."
10
+ translator = pipeline("translation", model="mideind/yfirlestur-icelandic-correction-byt5", max_length=512)
11
+
12
+ def split_text(text):
13
+ sentence_list = [i for i in tokenizer.split_into_sentences(text, original=True)]
14
+ return sentence_list
15
+
16
+ def predict(text):
17
+ text = text.replace("\n", " ")
18
+ translated = ""
19
+ for i in split_text(text):
20
+ # doesn't keep original string formatting, instead adds a space btw sentences
21
+ translated += translator(i.strip())[0]["translation_text"] + " "
22
+ return translated
23
+
24
+ iface = gr.Interface(
25
+ fn=predict,
26
+ inputs='text',
27
+ outputs='text',
28
+ title=title,
29
+ description=description,
30
+ examples=[["Kvitu fiðrildinn fljua firir utan gluggan."], ["Ég held þetta er ekki góður tími fara heimsókn."], ["Kver a þenan bússtað ja eða nei"]]
31
+ )
32
+ iface.launch()
33
+