Create tokenizer.json
Browse files- tokenizer.json +117 -0
tokenizer.json
CHANGED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"version": "1.0",
|
| 3 |
+
"truncation": null,
|
| 4 |
+
"padding": null,
|
| 5 |
+
"added_tokens": [
|
| 6 |
+
{
|
| 7 |
+
"id": 3,
|
| 8 |
+
"content": "<unk>",
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"lstrip": false,
|
| 11 |
+
"rstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"special": true
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"id": 0,
|
| 17 |
+
"content": "<pad>",
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"special": true
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"id": 1,
|
| 26 |
+
"content": "<s>",
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"special": true
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"id": 2,
|
| 35 |
+
"content": "</s>",
|
| 36 |
+
"single_word": false,
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"rstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"special": true
|
| 41 |
+
}
|
| 42 |
+
],
|
| 43 |
+
"normalizer": {
|
| 44 |
+
"type": "Sequence",
|
| 45 |
+
"normalizers": [
|
| 46 |
+
{
|
| 47 |
+
"type": "Replace",
|
| 48 |
+
"pattern": {
|
| 49 |
+
"String": "\\s+"
|
| 50 |
+
},
|
| 51 |
+
"content": "|"
|
| 52 |
+
}
|
| 53 |
+
]
|
| 54 |
+
},
|
| 55 |
+
"pre_tokenizer": {
|
| 56 |
+
"type": "Split",
|
| 57 |
+
"pattern": {
|
| 58 |
+
"String": "\\S"
|
| 59 |
+
},
|
| 60 |
+
"behavior": "Isolated",
|
| 61 |
+
"invert": false
|
| 62 |
+
},
|
| 63 |
+
"post_processor": null,
|
| 64 |
+
"decoder": {
|
| 65 |
+
"type": "CTC",
|
| 66 |
+
"pad_token": "<pad>",
|
| 67 |
+
"word_delimiter_token": "|",
|
| 68 |
+
"cleanup": true
|
| 69 |
+
},
|
| 70 |
+
"model": {
|
| 71 |
+
"type": "WordLevel",
|
| 72 |
+
"vocab": {
|
| 73 |
+
"<pad>": 0,
|
| 74 |
+
"<s>": 1,
|
| 75 |
+
"</s>": 2,
|
| 76 |
+
"<unk>": 3,
|
| 77 |
+
"a": 4,
|
| 78 |
+
"b": 5,
|
| 79 |
+
"d": 6,
|
| 80 |
+
"e": 7,
|
| 81 |
+
"f": 8,
|
| 82 |
+
"h": 9,
|
| 83 |
+
"i": 10,
|
| 84 |
+
"j": 11,
|
| 85 |
+
"k": 12,
|
| 86 |
+
"l": 13,
|
| 87 |
+
"m": 14,
|
| 88 |
+
"n": 15,
|
| 89 |
+
"o": 16,
|
| 90 |
+
"p": 17,
|
| 91 |
+
"s": 18,
|
| 92 |
+
"t": 19,
|
| 93 |
+
"u": 20,
|
| 94 |
+
"v": 21,
|
| 95 |
+
"w": 22,
|
| 96 |
+
"z": 23,
|
| 97 |
+
"æ": 24,
|
| 98 |
+
"ð": 25,
|
| 99 |
+
"ŋ": 26,
|
| 100 |
+
"ɑ": 27,
|
| 101 |
+
"ɔ": 28,
|
| 102 |
+
"ə": 29,
|
| 103 |
+
"ɛ": 30,
|
| 104 |
+
"ɡ": 31,
|
| 105 |
+
"ɪ": 32,
|
| 106 |
+
"ɹ": 33,
|
| 107 |
+
"ɾ": 34,
|
| 108 |
+
"ʃ": 35,
|
| 109 |
+
"ʊ": 36,
|
| 110 |
+
"ʌ": 37,
|
| 111 |
+
"ʒ": 38,
|
| 112 |
+
"ʔ": 39,
|
| 113 |
+
"θ": 40
|
| 114 |
+
},
|
| 115 |
+
"unk_token": "<unk>"
|
| 116 |
+
}
|
| 117 |
+
}
|