Upload tokenizer
Browse files- added_tokens.json +3 -0
- tokenizer_config.json +35 -0
- vocab.json +43 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<unk>": 41
|
| 3 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_blank": true,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "0",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"41": {
|
| 13 |
+
"content": "<unk>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
}
|
| 20 |
+
},
|
| 21 |
+
"additional_special_tokens": null,
|
| 22 |
+
"backend": "custom",
|
| 23 |
+
"clean_up_tokenization_spaces": true,
|
| 24 |
+
"extra_special_tokens": [],
|
| 25 |
+
"is_local": true,
|
| 26 |
+
"is_uroman": false,
|
| 27 |
+
"language": "ceb",
|
| 28 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 29 |
+
"normalize": true,
|
| 30 |
+
"pad_token": "0",
|
| 31 |
+
"phonemize": false,
|
| 32 |
+
"tokenizer_class": "VitsTokenizer",
|
| 33 |
+
"unk_token": "<unk>",
|
| 34 |
+
"verbose": false
|
| 35 |
+
}
|
vocab.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
" ": 27,
|
| 3 |
+
"'": 10,
|
| 4 |
+
"-": 9,
|
| 5 |
+
"0": 0,
|
| 6 |
+
"1": 23,
|
| 7 |
+
"2": 15,
|
| 8 |
+
"3": 35,
|
| 9 |
+
"4": 32,
|
| 10 |
+
"5": 18,
|
| 11 |
+
"6": 4,
|
| 12 |
+
"7": 37,
|
| 13 |
+
"8": 22,
|
| 14 |
+
"9": 11,
|
| 15 |
+
"_": 28,
|
| 16 |
+
"a": 24,
|
| 17 |
+
"b": 38,
|
| 18 |
+
"c": 8,
|
| 19 |
+
"d": 3,
|
| 20 |
+
"e": 16,
|
| 21 |
+
"f": 39,
|
| 22 |
+
"g": 6,
|
| 23 |
+
"h": 20,
|
| 24 |
+
"i": 33,
|
| 25 |
+
"j": 1,
|
| 26 |
+
"k": 34,
|
| 27 |
+
"l": 29,
|
| 28 |
+
"m": 31,
|
| 29 |
+
"n": 30,
|
| 30 |
+
"o": 36,
|
| 31 |
+
"p": 25,
|
| 32 |
+
"q": 17,
|
| 33 |
+
"r": 14,
|
| 34 |
+
"s": 5,
|
| 35 |
+
"t": 19,
|
| 36 |
+
"u": 12,
|
| 37 |
+
"v": 13,
|
| 38 |
+
"w": 7,
|
| 39 |
+
"x": 26,
|
| 40 |
+
"y": 40,
|
| 41 |
+
"z": 2,
|
| 42 |
+
"—": 21
|
| 43 |
+
}
|