TwentyNine commited on
Commit
a77af43
β€’
1 Parent(s): 908620a

streamline operation and add example of incorrect conversion

Browse files
Files changed (1) hide show
  1. README.md +11 -7
README.md CHANGED
@@ -44,8 +44,8 @@ fix_tokenizer(tokenizer, 'ain_Latn')
44
 
45
  def convert(
46
  text,
47
- model,
48
- tokenizer,
49
  src_lang='ain_Japn',
50
  tgt_lang='ain_Latn',
51
  max_length='auto',
@@ -71,10 +71,14 @@ def convert(
71
  return out[0]
72
  return
73
 
74
- convert("ポむ セタ クコン ルスむ", model=model, tokenizer=tokenizer)
75
- # 'pon seta ku=kor rusuy'
76
 
77
- convert("γ‚Ώγƒ³γƒˆ γŒγ£γ“γ†γ€€γ‚ͺルン パむェ", model=model, tokenizer=tokenizer)
78
- # 'tanto γŒγ£γ“γ† or un paye'
79
- # ideal: 'tanto GAKKO or un paye' or 'tanto GAKKOU or un paye'
 
 
 
 
80
  ```
 
44
 
45
  def convert(
46
  text,
47
+ model=model,
48
+ tokenizer=tokenizer,
49
  src_lang='ain_Japn',
50
  tgt_lang='ain_Latn',
51
  max_length='auto',
 
71
  return out[0]
72
  return
73
 
74
+ convert("ポむ セタ クコン ルスむ")
75
+ # GOOD: 'pon seta ku=kor rusuy'
76
 
77
+ convert("γ‚Ώγƒ³γƒˆ γŒγ£γ“γ†γ€€γ‚ͺルン パむェ")
78
+ # OK: 'tanto γŒγ£γ“γ† or un paye'
79
+ # IDEAL: 'tanto GAKKO or un paye' or 'tanto GAKKOU or un paye'
80
+
81
+ convert("γ‚Ώγƒ³γƒˆ γŒγ£γ“γ†γ€€γ‚ͺルン パむェ")
82
+ # OK: 'tanto γŒγ£γ“γ† or un paye'
83
+ # IDEAL: 'tanto GAKKO or un paye' or 'tanto GAKKOU or un paye'
84
  ```