ylacombe HF staff commited on
Commit
3ad57c7
1 Parent(s): f906d18

Upload tokenizer

Browse files
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<unk>": 45
3
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "pad_token": {
3
+ "content": "7",
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 ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_blank": true,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "7",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "45": {
13
+ "content": "<unk>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ }
20
+ },
21
+ "clean_up_tokenization_spaces": true,
22
+ "is_uroman": false,
23
+ "language": null,
24
+ "model_max_length": 1000000000000000019884624838656,
25
+ "normalize": true,
26
+ "pad_token": "7",
27
+ "phonemize": false,
28
+ "tokenizer_class": "VitsTokenizer",
29
+ "unk_token": "<unk>",
30
+ "verbose": false
31
+ }
vocab.json ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ " ": 27,
3
+ "0": 5,
4
+ "1": 24,
5
+ "2": 36,
6
+ "3": 32,
7
+ "4": 12,
8
+ "5": 6,
9
+ "6": 44,
10
+ "7": 0,
11
+ "8": 8,
12
+ "9": 23,
13
+ "_": 22,
14
+ "a": 1,
15
+ "b": 31,
16
+ "c": 3,
17
+ "d": 29,
18
+ "e": 39,
19
+ "f": 25,
20
+ "g": 34,
21
+ "h": 19,
22
+ "i": 37,
23
+ "j": 33,
24
+ "k": 15,
25
+ "l": 35,
26
+ "m": 13,
27
+ "n": 20,
28
+ "o": 41,
29
+ "p": 9,
30
+ "q": 18,
31
+ "r": 43,
32
+ "s": 16,
33
+ "t": 26,
34
+ "u": 38,
35
+ "v": 2,
36
+ "x": 28,
37
+ "y": 10,
38
+ "z": 11,
39
+ "á": 17,
40
+ "é": 21,
41
+ "í": 30,
42
+ "ñ": 42,
43
+ "ó": 7,
44
+ "ú": 40,
45
+ "ü": 14,
46
+ "—": 4
47
+ }