| { | |
| "version": "1.0", | |
| "truncation": { | |
| "direction": "Right", | |
| "max_length": 64, | |
| "strategy": "LongestFirst", | |
| "stride": 0 | |
| }, | |
| "padding": { | |
| "strategy": { | |
| "Fixed": 64 | |
| }, | |
| "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 | |
| }, | |
| { | |
| "id": 4, | |
| "content": "[MASK]", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| } | |
| ], | |
| "normalizer": { | |
| "type": "BertNormalizer", | |
| "clean_text": true, | |
| "handle_chinese_chars": true, | |
| "strip_accents": null, | |
| "lowercase": true | |
| }, | |
| "pre_tokenizer": { | |
| "type": "BertPreTokenizer" | |
| }, | |
| "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": { | |
| "type": "WordPiece", | |
| "prefix": "##", | |
| "cleanup": true | |
| }, | |
| "model": { | |
| "type": "WordPiece", | |
| "unk_token": "[UNK]", | |
| "continuing_subword_prefix": "##", | |
| "max_input_chars_per_word": 100, | |
| "vocab": { | |
| "[PAD]": 0, | |
| "[UNK]": 1, | |
| "[CLS]": 2, | |
| "[SEP]": 3, | |
| "[MASK]": 4, | |
| "&": 5, | |
| "'": 6, | |
| "*": 7, | |
| ",": 8, | |
| "-": 9, | |
| ".": 10, | |
| "/": 11, | |
| "0": 12, | |
| "1": 13, | |
| "2": 14, | |
| "3": 15, | |
| "4": 16, | |
| "5": 17, | |
| "6": 18, | |
| "7": 19, | |
| "8": 20, | |
| "9": 21, | |
| ":": 22, | |
| ";": 23, | |
| "_": 24, | |
| "a": 25, | |
| "b": 26, | |
| "c": 27, | |
| "d": 28, | |
| "e": 29, | |
| "f": 30, | |
| "g": 31, | |
| "h": 32, | |
| "i": 33, | |
| "j": 34, | |
| "k": 35, | |
| "l": 36, | |
| "m": 37, | |
| "n": 38, | |
| "o": 39, | |
| "p": 40, | |
| "q": 41, | |
| "r": 42, | |
| "s": 43, | |
| "t": 44, | |
| "u": 45, | |
| "v": 46, | |
| "w": 47, | |
| "x": 48, | |
| "y": 49, | |
| "z": 50, | |
| "ã": 51, | |
| "ä": 52, | |
| "å": 53, | |
| "æ": 54, | |
| "ç": 55, | |
| "é": 56, | |
| "ë": 57, | |
| "í": 58, | |
| "ï": 59, | |
| "ñ": 60, | |
| "ó": 61, | |
| "ô": 62, | |
| "õ": 63, | |
| "ö": 64, | |
| "ø": 65, | |
| "ú": 66, | |
| "ü": 67, | |
| "ń": 68, | |
| "а": 69, | |
| "г": 70, | |
| "д": 71, | |
| "е": 72, | |
| "и": 73, | |
| "н": 74, | |
| "о": 75, | |
| "р": 76, | |
| "с": 77, | |
| "ф": 78, | |
| "я": 79, | |
| "加": 80, | |
| "大": 81, | |
| "学": 82, | |
| "師": 83, | |
| "祥": 84, | |
| "講": 85, | |
| "贝": 86, | |
| "通": 87, | |
| "##a": 88, | |
| "##t": 89, | |
| "##h": 90, | |
| "##l": 91, | |
| "##i": 92, | |
| "##n": 93, | |
| "##e": 94, | |
| "##d": 95, | |
| "##y": 96, | |
| "##o": 97, | |
| "##u": 98, | |
| "##b": 99, | |
| "##6": 100, | |
| "##s": 101, | |
| "##c": 102, | |
| "##r": 103, | |
| "##p": 104, | |
| "##k": 105, | |
| "##j": 106, | |
| "##g": 107, | |
| "##w": 108, | |
| "##m": 109, | |
| "##z": 110, | |
| "##v": 111, | |
| "##q": 112, | |
| "##f": 113, | |
| "##x": 114, | |
| "##8": 115, | |
| "##1": 116, | |
| "##2": 117, | |
| "##0": 118, | |
| "##7": 119, | |
| "##9": 120, | |
| "##3": 121, | |
| "##4": 122, | |
| "##5": 123, | |
| "##ø": 124, | |
| "##æ": 125, | |
| "##ú": 126, | |
| "##ä": 127, | |
| "##ñ": 128, | |
| "##é": 129, | |
| "##е": 130, | |
| "##ç": 131, | |
| "##ã": 132, | |
| "##õ": 133, | |
| "##í": 134, | |
| "##ф": 135, | |
| "##ë": 136, | |
| "##о": 137, | |
| "##с": 138, | |
| "##и": 139, | |
| "##я": 140, | |
| "##г": 141, | |
| "##д": 142, | |
| "##н": 143, | |
| "##通": 144, | |
| "##贝": 145, | |
| "##祥": 146, | |
| "##ń": 147, | |
| "##ï": 148, | |
| "##学": 149, | |
| "##講": 150, | |
| "##師": 151, | |
| "##ö": 152, | |
| "##ô": 153, | |
| "##å": 154, | |
| "##ó": 155, | |
| "##ü": 156, | |
| "##р": 157, | |
| "##а": 158, | |
| "co": 159, | |
| "com": 160, | |
| "##er": 161, | |
| "##in": 162, | |
| "##ar": 163, | |
| "##ou": 164, | |
| "##es": 165, | |
| "##or": 166, | |
| "##nt": 167, | |
| "##et": 168, | |
| "##al": 169, | |
| "##ro": 170, | |
| "##il": 171, | |
| "##el": 172, | |
| "##ic": 173, | |
| "##ec": 174, | |
| "##at": 175, | |
| "##on": 176, | |
| "##ac": 177, | |
| "##it": 178, | |
| "##is": 179, | |
| "##an": 180, | |
| "##ew": 181, | |
| "##rc": 182, | |
| "pro": 183, | |
| "##ing": 184, | |
| "##out": 185, | |
| "##tp": 186, | |
| "##ts": 187, | |
| "##ap": 188, | |
| "##og": 189, | |
| "re": 190, | |
| "##as": 191, | |
| "##rg": 192, | |
| "##pp": 193, | |
| "org": 194, | |
| "##am": 195, | |
| "##ile": 196, | |
| "##op": 197, | |
| "st": 198, | |
| "##ti": 199, | |
| "##em": 200, | |
| "ma": 201, | |
| "##ol": 202, | |
| "##ri": 203, | |
| "ch": 204, | |
| "##ort": 205, | |
| "##ews": 206, | |
| "##ers": 207, | |
| "##en": 208, | |
| "##ms": 209, | |
| "##ch": 210, | |
| "ab": 211, | |
| "##du": 212, | |
| "about": 213, | |
| "##me": 214, | |
| "##art": 215, | |
| "su": 216, | |
| "##re": 217, | |
| "##nc": 218, | |
| "##net": 219, | |
| "##ad": 220, | |
| "##ent": 221, | |
| "##ourc": 222, | |
| "tel": 223, | |
| "##eam": 224, | |
| "##dap": 225, | |
| "ws": 226, | |
| "se": 227, | |
| "ldap": 228, | |
| "ht": 229, | |
| "do": 230, | |
| "irc": 231, | |
| "http": 232, | |
| "ac": 233, | |
| "##ter": 234, | |
| "sh": 235, | |
| "news": 236, | |
| "bl": 237, | |
| "##pi": 238, | |
| "##erv": 239, | |
| "##ard": 240, | |
| "##ting": 241, | |
| "##bo": 242, | |
| "mail": 243, | |
| "##ervic": 244, | |
| "ad": 245, | |
| "shop": 246, | |
| "uk": 247, | |
| "blog": 248, | |
| "in": 249, | |
| "api": 250, | |
| "##ervices": 251, | |
| "##st": 252, | |
| "##arch": 253, | |
| "##eg": 254, | |
| "##min": 255, | |
| "##act": 256, | |
| "##duc": 257, | |
| "##ash": 258, | |
| "##ount": 259, | |
| "##lo": 260, | |
| "##riv": 261, | |
| "acc": 262, | |
| "br": 263, | |
| "##elp": 264, | |
| "cont": 265, | |
| "ter": 266, | |
| "log": 267, | |
| "##eck": 268, | |
| "reg": 269, | |
| "##acy": 270, | |
| "set": 271, | |
| "##tings": 272, | |
| "##pport": 273, | |
| "prof": 274, | |
| "##ducts": 275, | |
| "##ister": 276, | |
| "search": 277, | |
| "##board": 278, | |
| "settings": 279, | |
| "priv": 280, | |
| "terms": 281, | |
| "services": 282, | |
| "help": 283, | |
| "register": 284, | |
| "support": 285, | |
| "account": 286, | |
| "dash": 287, | |
| "dashboard": 288, | |
| "privacy": 289, | |
| "login": 290, | |
| "##mp": 291, | |
| "contact": 292, | |
| "check": 293, | |
| "admin": 294, | |
| "profile": 295, | |
| "checkout": 296, | |
| "products": 297, | |
| "##io": 298, | |
| "##eb": 299, | |
| "##are": 300, | |
| "##um": 301, | |
| "##po": 302, | |
| "##to": 303, | |
| "##sh": 304, | |
| "##all": 305, | |
| "##vi": 306, | |
| "nl": 307, | |
| "##pe": 308, | |
| "de": 309, | |
| "##her": 310, | |
| "##sy": 311, | |
| "##ies": 312, | |
| "##ial": 313, | |
| "##ric": 314, | |
| "##ob": 315, | |
| "##rou": 316, | |
| "sk": 317, | |
| "sc": 318, | |
| "vi": 319, | |
| "au": 320, | |
| "##cs": 321, | |
| "##ners": 322, | |
| "##ata": 323, | |
| "##dis": 324, | |
| "##roup": 325, | |
| "##vent": 326, | |
| "##ex": 327, | |
| "##artners": 328, | |
| "##ion": 329, | |
| "##ag": 330, | |
| "sky": 331, | |
| "##wn": 332, | |
| "##ource": 333, | |
| "##cal": 334, | |
| "##ery": 335, | |
| "##mon": 336, | |
| "##ed": 337, | |
| "rt": 338, | |
| "##fy": 339, | |
| "for": 340, | |
| "sf": 341, | |
| "web": 342, | |
| "res": 343, | |
| "mag": 344, | |
| "sv": 345, | |
| "skype": 346, | |
| "##co": 347, | |
| "fa": 348, | |
| "rtmp": 349, | |
| "spo": 350, | |
| "##tsp": 351, | |
| "git": 352, | |
| "gop": 353, | |
| "xm": 354, | |
| "data": 355, | |
| "cat": 356, | |
| "##rome": 357, | |
| "xmpp": 358, | |
| "##tify": 359, | |
| "spotify": 360, | |
| "##ntp": 361, | |
| "gopher": 362, | |
| "sftp": 363, | |
| "##ects": 364, | |
| "source": 365, | |
| "ftp": 366, | |
| "port": 367, | |
| "##sync": 368, | |
| "net": 369, | |
| "rtsp": 370, | |
| "view": 371, | |
| "rsync": 372, | |
| "vnc": 373, | |
| "https": 374, | |
| "chrome": 375, | |
| "##vents": 376, | |
| "ssh": 377, | |
| "ldaps": 378, | |
| "mailto": 379, | |
| "wss": 380, | |
| "nntp": 381, | |
| "ircs": 382, | |
| "scp": 383, | |
| "svn": 384, | |
| "telnet": 385, | |
| "mms": 386, | |
| "##ources": 387, | |
| "magnet": 388, | |
| "redis": 389, | |
| "##esti": 390, | |
| "##fol": 391, | |
| "steam": 392, | |
| "webcal": 393, | |
| "file": 394, | |
| "team": 395, | |
| "##ads": 396, | |
| "events": 397, | |
| "##ials": 398, | |
| "care": 399, | |
| "sit": 400, | |
| "##folio": 401, | |
| "##jects": 402, | |
| "portfolio": 403, | |
| "partners": 404, | |
| "##ure": 405, | |
| "revi": 406, | |
| "cart": 407, | |
| "resources": 408, | |
| "forum": 409, | |
| "testi": 410, | |
| "##monials": 411, | |
| "testimonials": 412, | |
| "reviews": 413, | |
| "gall": 414, | |
| "##emap": 415, | |
| "##ories": 416, | |
| "pric": 417, | |
| "##us": 418, | |
| "sitemap": 419, | |
| "projects": 420, | |
| "##and": 421, | |
| "docs": 422, | |
| "gallery": 423, | |
| "faq": 424, | |
| "down": 425, | |
| "pricing": 426, | |
| "##egories": 427, | |
| "categories": 428, | |
| "##loads": 429, | |
| "downloads": 430, | |
| "careers": 431, | |
| "##ig": 432, | |
| "##ore": 433, | |
| "##ia": 434, | |
| "##ur": 435, | |
| "##sc": 436, | |
| "##eta": 437, | |
| "sec": 438, | |
| "##ww": 439, | |
| "##emo": 440, | |
| "cd": 441, | |
| "mob": 442, | |
| "app": 443, | |
| "##atic": 444, | |
| "www": 445, | |
| "dev": 446, | |
| "mobile": 447, | |
| "secure": 448, | |
| "cdn": 449, | |
| "store": 450, | |
| "demo": 451, | |
| "beta": 452, | |
| "##ech": 453, | |
| "static": 454, | |
| "ho": 455, | |
| "##ra": 456, | |
| "##ons": 457, | |
| "##le": 458, | |
| "ca": 459, | |
| "##ul": 460, | |
| "##be": 461, | |
| "it": 462, | |
| "##ov": 463, | |
| "##ir": 464, | |
| "##group": 465, | |
| "edu": 466, | |
| "fr": 467, | |
| "##th": 468, | |
| "##id": 469, | |
| "##os": 470, | |
| "##om": 471, | |
| "ind": 472, | |
| "##ab": 473, | |
| "##un": 474, | |
| "##ot": 475, | |
| "##est": 476, | |
| "##ay": 477, | |
| "##ation": 478, | |
| "home": 479, | |
| "##ru": 480, | |
| "##anc": 481, | |
| "th": 482, | |
| "sub": 483, | |
| "##ribe": 484, | |
| "index": 485, | |
| "##od": 486, | |
| "##scribe": 487, | |
| "subscribe": 488, | |
| "##of": 489, | |
| "al": 490, | |
| "##oc": 491, | |
| "##ity": 492, | |
| "##iv": 493, | |
| "##ine": 494, | |
| "##inc": 495, | |
| "##im": 496, | |
| "ar": 497, | |
| "tr": 498, | |
| "es": 499 | |
| } | |
| } | |
| } |