cyrillic-htr-model / processor /tokenizer_config.json
riikkamarttila's picture
Upload 7 files
67f207f verified
raw
history blame
27.9 kB
{
"add_prefix_space": false,
"added_tokens_decoder": {
"0": {
"content": "<s>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"1": {
"content": "<pad>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"2": {
"content": "</s>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"3": {
"content": "<unk>",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"4": {
"content": ".",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"6": {
"content": ",",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"12": {
"content": "-",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"29": {
"content": "s",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"35": {
"content": ":",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"43": {
"content": ")",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"73": {
"content": "/",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"90": {
"content": "t",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"100": {
"content": "I",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"102": {
"content": "a",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"104": {
"content": "S",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"113": {
"content": "\"",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"116": {
"content": "?",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"118": {
"content": "i",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"119": {
"content": "m",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"131": {
"content": ";",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"134": {
"content": "1",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"139": {
"content": "o",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"176": {
"content": "2",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"207": {
"content": "%",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"219": {
"content": "y",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"242": {
"content": "e",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"245": {
"content": "5",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"246": {
"content": "3",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"250": {
"content": "A",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"257": {
"content": "u",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"267": {
"content": "j",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"282": {
"content": "n",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"288": {
"content": "0",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"298": {
"content": "h",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"306": {
"content": "4",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"328": {
"content": "!",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"330": {
"content": "k",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"338": {
"content": "r",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"347": {
"content": "C",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"387": {
"content": "B",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"398": {
"content": "8",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"401": {
"content": "6",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"406": {
"content": "7",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"417": {
"content": "d",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"428": {
"content": "b",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"438": {
"content": "c",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"448": {
"content": "M",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"462": {
"content": "l",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"466": {
"content": "9",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"487": {
"content": "N",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"495": {
"content": "D",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"500": {
"content": "R",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"506": {
"content": "f",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"510": {
"content": "P",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"530": {
"content": "K",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"534": {
"content": "G",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"565": {
"content": "T",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"571": {
"content": "g",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"574": {
"content": "L",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"597": {
"content": "F",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"605": {
"content": "w",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"642": {
"content": "p",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"673": {
"content": "O",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"705": {
"content": "v",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"717": {
"content": "E",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"725": {
"content": "H",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"742": {
"content": "]",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"771": {
"content": "W",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"791": {
"content": "U",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"846": {
"content": "V",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"863": {
"content": "J",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"975": {
"content": "Y",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"1000": {
"content": "X",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"1178": {
"content": "x",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"1215": {
"content": "_",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"1277": {
"content": "Å",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"1301": {
"content": "Z",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"1343": {
"content": "q",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"1640": {
"content": "(",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"2023": {
"content": "»",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"2469": {
"content": "£",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"2744": {
"content": "+",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"4958": {
"content": "«",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"6248": {
"content": "§",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"10809": {
"content": "½",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"11582": {
"content": "¬",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"35423": {
"content": "É",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"42393": {
"content": "å",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"43251": {
"content": "ï",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"46030": {
"content": "Ö",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"48753": {
"content": "ä",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50093": {
"content": "À",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50098": {
"content": "ö",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50264": {
"content": "<mask>",
"lstrip": true,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": true
},
"50265": {
"content": "І",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50266": {
"content": "А",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50267": {
"content": "Б",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50268": {
"content": "В",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50269": {
"content": "Г",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50270": {
"content": "Д",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50271": {
"content": "Е",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50272": {
"content": "Ж",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50273": {
"content": "З",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50274": {
"content": "И",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50275": {
"content": "Й",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50276": {
"content": "К",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50277": {
"content": "Л",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50278": {
"content": "М",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50279": {
"content": "Н",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50280": {
"content": "О",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50281": {
"content": "П",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50282": {
"content": "Р",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50283": {
"content": "С",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50284": {
"content": "Т",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50285": {
"content": "У",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50286": {
"content": "Ф",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50287": {
"content": "Х",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50288": {
"content": "Ц",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50289": {
"content": "Ч",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50290": {
"content": "Ш",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50291": {
"content": "Щ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50292": {
"content": "Ы",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50293": {
"content": "Э",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50294": {
"content": "Ю",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50295": {
"content": "Я",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50296": {
"content": "а",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50297": {
"content": "б",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50298": {
"content": "в",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50299": {
"content": "г",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50300": {
"content": "д",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50301": {
"content": "е",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50302": {
"content": "ж",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50303": {
"content": "з",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50304": {
"content": "и",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50305": {
"content": "й",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50306": {
"content": "к",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50307": {
"content": "л",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50308": {
"content": "м",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50309": {
"content": "н",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50310": {
"content": "о",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50311": {
"content": "п",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50312": {
"content": "р",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50313": {
"content": "с",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50314": {
"content": "т",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50315": {
"content": "у",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50316": {
"content": "ф",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50317": {
"content": "х",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50318": {
"content": "ц",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50319": {
"content": "ч",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50320": {
"content": "ш",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50321": {
"content": "щ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50322": {
"content": "ъ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50323": {
"content": "ы",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50324": {
"content": "ь",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50325": {
"content": "э",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50326": {
"content": "ю",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50327": {
"content": "я",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50328": {
"content": "ё",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50329": {
"content": "і",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50330": {
"content": "ѣ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50331": {
"content": "һ",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50332": {
"content": "–",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50333": {
"content": "“",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50334": {
"content": "…",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
},
"50335": {
"content": "№",
"lstrip": false,
"normalized": true,
"rstrip": false,
"single_word": false,
"special": false
}
},
"bos_token": "<s>",
"clean_up_tokenization_spaces": true,
"cls_token": "<s>",
"eos_token": "</s>",
"errors": "replace",
"mask_token": "<mask>",
"model_max_length": 512,
"pad_token": "<pad>",
"processor_class": "TrOCRProcessor",
"sep_token": "</s>",
"tokenizer_class": "RobertaTokenizer",
"trim_offsets": true,
"unk_token": "<unk>"
}