| { | |
| "version": "1.0", | |
| "truncation": null, | |
| "padding": { | |
| "strategy": "BatchLongest", | |
| "direction": "Right", | |
| "pad_to_multiple_of": null, | |
| "pad_id": 0, | |
| "pad_type_id": 0, | |
| "pad_token": "[PAD]" | |
| }, | |
| "added_tokens": [ | |
| { | |
| "id": 0, | |
| "content": "[PAD]", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| }, | |
| { | |
| "id": 1, | |
| "content": "[UNK]", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| }, | |
| { | |
| "id": 2, | |
| "content": "[CLS]", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| }, | |
| { | |
| "id": 3, | |
| "content": "[SEP]", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| } | |
| ], | |
| "normalizer": null, | |
| "pre_tokenizer": { | |
| "type": "WhitespaceSplit" | |
| }, | |
| "post_processor": { | |
| "type": "TemplateProcessing", | |
| "single": [ | |
| { | |
| "SpecialToken": { | |
| "id": "[CLS]", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "Sequence": { | |
| "id": "A", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "SpecialToken": { | |
| "id": "[SEP]", | |
| "type_id": 0 | |
| } | |
| } | |
| ], | |
| "pair": [ | |
| { | |
| "SpecialToken": { | |
| "id": "[CLS]", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "Sequence": { | |
| "id": "A", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "SpecialToken": { | |
| "id": "[SEP]", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "Sequence": { | |
| "id": "B", | |
| "type_id": 1 | |
| } | |
| }, | |
| { | |
| "SpecialToken": { | |
| "id": "[SEP]", | |
| "type_id": 1 | |
| } | |
| } | |
| ], | |
| "special_tokens": { | |
| "[CLS]": { | |
| "id": "[CLS]", | |
| "ids": [ | |
| 2 | |
| ], | |
| "tokens": [ | |
| "[CLS]" | |
| ] | |
| }, | |
| "[SEP]": { | |
| "id": "[SEP]", | |
| "ids": [ | |
| 3 | |
| ], | |
| "tokens": [ | |
| "[SEP]" | |
| ] | |
| } | |
| } | |
| }, | |
| "decoder": null, | |
| "model": { | |
| "type": "WordLevel", | |
| "vocab": { | |
| "[PAD]": 0, | |
| "[UNK]": 1, | |
| "[CLS]": 2, | |
| "[SEP]": 3, | |
| "n": 4, | |
| "h": 5, | |
| "t": 6, | |
| "c": 7, | |
| "i": 8, | |
| "g": 9, | |
| "a": 10, | |
| "u": 11, | |
| "đ": 12, | |
| "m": 13, | |
| "o": 14, | |
| "r": 15, | |
| "v": 16, | |
| "à": 17, | |
| "l": 18, | |
| "p": 19, | |
| ",": 20, | |
| "ư": 21, | |
| "y": 22, | |
| "á": 23, | |
| "s": 24, | |
| "b": 25, | |
| "k": 26, | |
| ".": 27, | |
| "d": 28, | |
| "ế": 29, | |
| "ệ": 30, | |
| "ạ": 31, | |
| "ô": 32, | |
| "e": 33, | |
| "ả": 34, | |
| "ê": 35, | |
| "ộ": 36, | |
| "ó": 37, | |
| "ố": 38, | |
| "ớ": 39, | |
| "ấ": 40, | |
| "ờ": 41, | |
| "ị": 42, | |
| "â": 43, | |
| "q": 44, | |
| "ề": 45, | |
| "ủ": 46, | |
| "1": 47, | |
| "ể": 48, | |
| "2": 49, | |
| "ợ": 50, | |
| "0": 51, | |
| "ơ": 52, | |
| "ậ": 53, | |
| "x": 54, | |
| "ì": 55, | |
| "ầ": 56, | |
| "ự": 57, | |
| "ă": 58, | |
| "ứ": 59, | |
| "ở": 60, | |
| "í": 61, | |
| "ã": 62, | |
| "ụ": 63, | |
| "ọ": 64, | |
| "ồ": 65, | |
| "ữ": 66, | |
| "-": 67, | |
| "ắ": 68, | |
| "ú": 69, | |
| "ổ": 70, | |
| "ừ": 71, | |
| "ò": 72, | |
| "ù": 73, | |
| "ặ": 74, | |
| "3": 75, | |
| "ỉ": 76, | |
| "9": 77, | |
| ")": 78, | |
| "(": 79, | |
| "ũ": 80, | |
| "5": 81, | |
| "\"": 82, | |
| "ễ": 83, | |
| ":": 84, | |
| "ử": 85, | |
| "4": 86, | |
| "/": 87, | |
| "f": 88, | |
| "ẽ": 89, | |
| "ý": 90, | |
| "ỏ": 91, | |
| "6": 92, | |
| "ẩ": 93, | |
| "é": 94, | |
| "8": 95, | |
| "7": 96, | |
| "ẫ": 97, | |
| "ằ": 98, | |
| "ỗ": 99, | |
| "ĩ": 100, | |
| "w": 101, | |
| "ẻ": 102, | |
| ";": 103, | |
| "'": 104, | |
| "ỹ": 105, | |
| "ẹ": 106, | |
| "ỷ": 107, | |
| "%": 108, | |
| "ỳ": 109, | |
| "z": 110, | |
| "j": 111, | |
| "ỡ": 112, | |
| "õ": 113, | |
| "è": 114, | |
| "ẳ": 115, | |
| "?": 116, | |
| "ẵ": 117, | |
| "–": 118, | |
| "&": 119, | |
| "!": 120, | |
| "*": 121, | |
| "’": 122, | |
| "+": 123, | |
| "‘": 124, | |
| ">": 125, | |
| "|": 126, | |
| "_": 127, | |
| "ỵ": 128, | |
| "=": 129, | |
| "@": 130, | |
| "[": 131, | |
| "]": 132, | |
| "•": 133, | |
| "#": 134, | |
| "●": 135, | |
| "·": 136, | |
| "ð": 137, | |
| "°": 138, | |
| "<": 139, | |
| "ö": 140, | |
| "≥": 141, | |
| "$": 142, | |
| "ü": 143, | |
| "о": 144, | |
| "⁄": 145, | |
| "а": 146, | |
| "\\": 147, | |
| "т": 148, | |
| "е": 149, | |
| "и": 150, | |
| "~": 151, | |
| "с": 152, | |
| "н": 153, | |
| "ä": 154, | |
| "—": 155, | |
| "ç": 156, | |
| "р": 157, | |
| "ø": 158, | |
| "≤": 159, | |
| "ë": 160, | |
| "к": 161, | |
| "п": 162, | |
| "ć": 163, | |
| "л": 164, | |
| "‐": 165, | |
| "μ": 166, | |
| "«": 167, | |
| "š": 168, | |
| "ь": 169, | |
| "×": 170, | |
| "ā": 171, | |
| "м": 172, | |
| "у": 173, | |
| "å": 174, | |
| "β": 175, | |
| "ч": 176, | |
| "я": 177, | |
| "в": 178, | |
| "`": 179, | |
| "ō": 180, | |
| "ï": 181, | |
| "č": 182, | |
| "ŋ": 183, | |
| "̣": 184, | |
| "−": 185, | |
| "α": 186, | |
| "ы": 187, | |
| "�": 188, | |
| "ß": 189, | |
| "д": 190, | |
| "′": 191, | |
| "ń": 192, | |
| "́": 193, | |
| "б": 194, | |
| "̀": 195, | |
| "з": 196, | |
| "ş": 197, | |
| "̉": 198, | |
| "‰": 199, | |
| "»": 200, | |
| "æ": 201, | |
| "δ": 202, | |
| "ν": 203, | |
| "ж": 204, | |
| "ğ": 205, | |
| "ı": 206, | |
| "ɛ": 207, | |
| "虎": 208, | |
| "ǎ": 209, | |
| "г": 210, | |
| "ᅲ": 211, | |
| "人": 212, | |
| "家": 213, | |
| "空": 214, | |
| "ġ": 215, | |
| "ī": 216, | |
| "ł": 217, | |
| "ū": 218, | |
| "ɪ": 219, | |
| "ς": 220, | |
| "ц": 221, | |
| "子": 222, | |
| "礼": 223, | |
| "精": 224, | |
| "自": 225, | |
| "花": 226 | |
| }, | |
| "unk_token": "[UNK]" | |
| } | |
| } |