Matthijs commited on
Commit
6f77ba8
1 Parent(s): b25111d

Upload tokenizer

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +4 -0
  2. tokenizer_config.json +9 -0
  3. vocab.json +43 -0
special_tokens_map.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "pad_token": "<pad>",
3
+ "unk_token": "<unk>"
4
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_blank": true,
3
+ "clean_up_tokenization_spaces": true,
4
+ "language": null,
5
+ "model_max_length": 1000000000000000019884624838656,
6
+ "pad_token": "<pad>",
7
+ "tokenizer_class": "VitsMmsTokenizer",
8
+ "unk_token": "<unk>"
9
+ }
vocab.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ " ": 25,
3
+ "<pad>": 39,
4
+ "<unk>": 40,
5
+ "_": 37,
6
+ "ء": 26,
7
+ "آ": 27,
8
+ "أ": 20,
9
+ "ؤ": 35,
10
+ "إ": 4,
11
+ "ئ": 38,
12
+ "ا": 0,
13
+ "ب": 28,
14
+ "ة": 7,
15
+ "ت": 12,
16
+ "ث": 19,
17
+ "ج": 13,
18
+ "ح": 23,
19
+ "خ": 11,
20
+ "د": 16,
21
+ "ذ": 6,
22
+ "ر": 9,
23
+ "ز": 34,
24
+ "س": 8,
25
+ "ش": 32,
26
+ "ص": 18,
27
+ "ض": 22,
28
+ "ط": 10,
29
+ "ظ": 14,
30
+ "ع": 3,
31
+ "غ": 5,
32
+ "ف": 36,
33
+ "ق": 33,
34
+ "ك": 2,
35
+ "ل": 31,
36
+ "م": 30,
37
+ "ن": 1,
38
+ "ه": 24,
39
+ "و": 29,
40
+ "ى": 21,
41
+ "ي": 15,
42
+ "–": 17
43
+ }