TheAIchemist13 commited on
Commit
4b914c6
·
1 Parent(s): 16c8889

Upload tokenizer

Browse files
Files changed (1) hide show
  1. vocab.json +79 -71
vocab.json CHANGED
@@ -1,78 +1,86 @@
1
  {
 
 
 
 
 
 
 
 
 
 
2
  "</s>": 2,
3
  "<pad>": 0,
4
  "<s>": 1,
5
  "<unk>": 3,
6
  "|": 4,
7
- "ँ": 5,
8
- "ं": 6,
9
- "ः": 7,
10
- "अ": 8,
11
- "आ": 9,
12
- "इ": 10,
13
- "ई": 11,
14
- "उ": 12,
15
- "ऊ": 13,
16
- "ऋ": 14,
17
- "ए": 15,
18
- "ऐ": 16,
19
- "ऑ": 17,
20
- "ओ": 18,
21
- "औ": 19,
22
- "क": 20,
23
- "ख": 21,
24
- "ग": 22,
25
- "घ": 23,
26
- "ङ": 24,
27
- "च": 25,
28
- "छ": 26,
29
- "ज": 27,
30
- "झ": 28,
31
- "ञ": 29,
32
- "ट": 30,
33
- "ठ": 31,
34
- "ड": 32,
35
- "ढ": 33,
36
- "ण": 34,
37
- "त": 35,
38
- "थ": 36,
39
- "द": 37,
40
- "ध": 38,
41
- "न": 39,
42
- "प": 40,
43
- "फ": 41,
44
- "ब": 42,
45
- "भ": 43,
46
- "म": 44,
47
- "य": 45,
48
- "र": 46,
49
- "ल": 47,
50
- "व": 48,
51
- "श": 49,
52
- "ष": 50,
53
- "स": 51,
54
- "ह": 52,
55
- "": 75,
56
- "": 53,
57
- "ि": 54,
58
- "": 55,
59
- "": 56,
60
- "": 57,
61
- "": 58,
62
- "": 59,
63
- "": 50,
64
- "": 61,
65
- "": 62,
66
- "": 63,
67
- "": 64,
68
- "": 65,
69
- "": 66,
70
- "": 67,
71
- "": 68,
72
- "": 69,
73
- "": 70,
74
- "": 71,
75
- "": 72,
76
- "य़": 73,
77
- "।": 74
78
  }
 
1
  {
2
+ "0": 5,
3
+ "1": 6,
4
+ "2": 7,
5
+ "3": 8,
6
+ "4": 9,
7
+ "5": 10,
8
+ "6": 11,
9
+ "7": 12,
10
+ "8": 13,
11
+ "9": 14,
12
  "</s>": 2,
13
  "<pad>": 0,
14
  "<s>": 1,
15
  "<unk>": 3,
16
  "|": 4,
17
+ "ँ": 15,
18
+ "ं": 16,
19
+ "ः": 17,
20
+ "अ": 18,
21
+ "आ": 19,
22
+ "इ": 20,
23
+ "ई": 21,
24
+ "उ": 22,
25
+ "ऊ": 23,
26
+ "ऋ": 24,
27
+ "ए": 25,
28
+ "ऐ": 26,
29
+ "ऑ": 27,
30
+ "ओ": 28,
31
+ "औ": 29,
32
+ "क": 30,
33
+ "ख": 31,
34
+ "ग": 32,
35
+ "घ": 33,
36
+ "ङ": 34,
37
+ "च": 35,
38
+ "छ": 36,
39
+ "ज": 37,
40
+ "झ": 38,
41
+ "ञ": 39,
42
+ "ट": 40,
43
+ "ठ": 41,
44
+ "ड": 42,
45
+ "ढ": 43,
46
+ "ण": 44,
47
+ "त": 45,
48
+ "थ": 46,
49
+ "द": 47,
50
+ "ध": 48,
51
+ "न": 49,
52
+ "प": 50,
53
+ "फ": 51,
54
+ "ब": 52,
55
+ "भ": 53,
56
+ "म": 54,
57
+ "य": 55,
58
+ "र": 56,
59
+ "ल": 57,
60
+ "व": 58,
61
+ "श": 59,
62
+ "ष": 60,
63
+ "स": 61,
64
+ "ह": 62,
65
+ "": 63,
66
+ "ि": 64,
67
+ "": 65,
68
+ "": 66,
69
+ "": 67,
70
+ "": 68,
71
+ "": 69,
72
+ "": 70,
73
+ "": 71,
74
+ "": 72,
75
+ "": 73,
76
+ "": 74,
77
+ "": 75,
78
+ "": 76,
79
+ "": 77,
80
+ "": 78,
81
+ "": 79,
82
+ "": 80,
83
+ "": 81,
84
+ "": 82,
85
+ "": 83
 
 
86
  }