sengzi commited on
Commit
b022516
1 Parent(s): 41c45dd

Update Sejarah.py

Browse files
Files changed (1) hide show
  1. Sejarah.py +8 -9
Sejarah.py CHANGED
@@ -51,8 +51,6 @@ class Sejarah:
51
 
52
  def language_converter(self, content, lang, method):
53
 
54
- content = content.lower()
55
-
56
  if lang == "en":
57
  if method == "question":
58
  tokenized_text = self.en_id_tokenizer.prepare_seq2seq_batch([content], return_tensors='pt')
@@ -65,7 +63,13 @@ class Sejarah:
65
  content = self.id_en_tokenizer.batch_decode(translation, skip_special_tokens=True)[0]
66
 
67
  return content
68
-
 
 
 
 
 
 
69
  def interface(self, question):
70
  language = self.detect_language(question)
71
 
@@ -82,9 +86,4 @@ class Sejarah:
82
  answer = self.language_converter(result['answers'][0].answer, language, "answer")
83
  context = self.language_converter(result['answers'][0].context, language, "answer")
84
 
85
- return answer, context
86
-
87
-
88
- def detect_language(self, content):
89
- lang = langid.classify(content)
90
- return lang[0]
 
51
 
52
  def language_converter(self, content, lang, method):
53
 
 
 
54
  if lang == "en":
55
  if method == "question":
56
  tokenized_text = self.en_id_tokenizer.prepare_seq2seq_batch([content], return_tensors='pt')
 
63
  content = self.id_en_tokenizer.batch_decode(translation, skip_special_tokens=True)[0]
64
 
65
  return content
66
+
67
+
68
+ def detect_language(self, content):
69
+ lang = langid.classify(content)
70
+ return lang[0]
71
+
72
+
73
  def interface(self, question):
74
  language = self.detect_language(question)
75
 
 
86
  answer = self.language_converter(result['answers'][0].answer, language, "answer")
87
  context = self.language_converter(result['answers'][0].context, language, "answer")
88
 
89
+ return answer, context