Upload tokenizer
Browse files- README.md +0 -1
- added_tokens.json +5 -0
- special_tokens_map.json +14 -2
- tokenizer_config.json +44 -2
- vocab.json +13 -0
README.md
CHANGED
@@ -1,4 +1,3 @@
|
|
1 |
-
|
2 |
---
|
3 |
license: cc-by-nc-4.0
|
4 |
tags:
|
|
|
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
3 |
tags:
|
added_tokens.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"ড়": 87,
|
3 |
+
"ঢ়": 88,
|
4 |
+
"য়": 89
|
5 |
+
}
|
special_tokens_map.json
CHANGED
@@ -1,4 +1,16 @@
|
|
1 |
{
|
2 |
-
"pad_token":
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
}
|
|
|
1 |
{
|
2 |
+
"pad_token": {
|
3 |
+
"content": "6",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"unk_token": {
|
10 |
+
"content": "<unk>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
}
|
16 |
}
|
tokenizer_config.json
CHANGED
@@ -1,10 +1,52 @@
|
|
1 |
{
|
2 |
-
"add_blank":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
"clean_up_tokenization_spaces": true,
|
4 |
"is_uroman": false,
|
5 |
"language": "ben",
|
6 |
"model_max_length": 1000000000000000019884624838656,
|
7 |
-
"normalize":
|
8 |
"pad_token": "6",
|
9 |
"phonemize": false,
|
10 |
"tokenizer_class": "VitsTokenizer",
|
|
|
1 |
{
|
2 |
+
"add_blank": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"0": {
|
5 |
+
"content": "6",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": false,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
},
|
12 |
+
"74": {
|
13 |
+
"content": "<unk>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": false,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": true
|
19 |
+
},
|
20 |
+
"87": {
|
21 |
+
"content": "ড়",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false,
|
26 |
+
"special": false
|
27 |
+
},
|
28 |
+
"88": {
|
29 |
+
"content": "ঢ়",
|
30 |
+
"lstrip": false,
|
31 |
+
"normalized": false,
|
32 |
+
"rstrip": false,
|
33 |
+
"single_word": false,
|
34 |
+
"special": false
|
35 |
+
},
|
36 |
+
"89": {
|
37 |
+
"content": "য়",
|
38 |
+
"lstrip": false,
|
39 |
+
"normalized": false,
|
40 |
+
"rstrip": false,
|
41 |
+
"single_word": false,
|
42 |
+
"special": false
|
43 |
+
}
|
44 |
+
},
|
45 |
"clean_up_tokenization_spaces": true,
|
46 |
"is_uroman": false,
|
47 |
"language": "ben",
|
48 |
"model_max_length": 1000000000000000019884624838656,
|
49 |
+
"normalize": false,
|
50 |
"pad_token": "6",
|
51 |
"phonemize": false,
|
52 |
"tokenizer_class": "VitsTokenizer",
|
vocab.json
CHANGED
@@ -40,7 +40,9 @@
|
|
40 |
"ট": 55,
|
41 |
"ঠ": 73,
|
42 |
"ড": 21,
|
|
|
43 |
"ঢ": 22,
|
|
|
44 |
"ণ": 51,
|
45 |
"ত": 42,
|
46 |
"থ": 14,
|
@@ -53,6 +55,7 @@
|
|
53 |
"ভ": 16,
|
54 |
"ম": 41,
|
55 |
"য": 38,
|
|
|
56 |
"র": 69,
|
57 |
"ল": 10,
|
58 |
"শ": 48,
|
@@ -72,5 +75,15 @@
|
|
72 |
"ৌ": 44,
|
73 |
"্": 36,
|
74 |
"ৎ": 52,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
75 |
"—": 32
|
76 |
}
|
|
|
40 |
"ট": 55,
|
41 |
"ঠ": 73,
|
42 |
"ড": 21,
|
43 |
+
"ড়": 84,
|
44 |
"ঢ": 22,
|
45 |
+
"ঢ়": 85,
|
46 |
"ণ": 51,
|
47 |
"ত": 42,
|
48 |
"থ": 14,
|
|
|
55 |
"ভ": 16,
|
56 |
"ম": 41,
|
57 |
"য": 38,
|
58 |
+
"য়": 86,
|
59 |
"র": 69,
|
60 |
"ল": 10,
|
61 |
"শ": 48,
|
|
|
75 |
"ৌ": 44,
|
76 |
"্": 36,
|
77 |
"ৎ": 52,
|
78 |
+
"০": 74,
|
79 |
+
"১": 75,
|
80 |
+
"২": 76,
|
81 |
+
"৩": 77,
|
82 |
+
"৪": 78,
|
83 |
+
"৫": 79,
|
84 |
+
"৬": 80,
|
85 |
+
"৭": 81,
|
86 |
+
"৮": 82,
|
87 |
+
"৯": 83,
|
88 |
"—": 32
|
89 |
}
|