Upload tokenizer
8fad86d
|
|
{ |
|
|
"ar_eg": { |
|
|
"+": 1, |
|
|
"/": 2, |
|
|
"0": 3, |
|
|
"1": 4, |
|
|
"2": 5, |
|
|
"3": 6, |
|
|
"4": 7, |
|
|
"5": 8, |
|
|
"6": 9, |
|
|
"7": 10, |
|
|
"8": 11, |
|
|
"9": 12, |
|
|
"[PAD]": 102, |
|
|
"[UNK]": 101, |
|
|
"a": 13, |
|
|
"b": 14, |
|
|
"c": 15, |
|
|
"d": 16, |
|
|
"e": 17, |
|
|
"f": 18, |
|
|
"g": 19, |
|
|
"h": 20, |
|
|
"i": 21, |
|
|
"j": 22, |
|
|
"k": 23, |
|
|
"l": 24, |
|
|
"m": 25, |
|
|
"n": 26, |
|
|
"o": 27, |
|
|
"p": 28, |
|
|
"r": 29, |
|
|
"s": 30, |
|
|
"t": 31, |
|
|
"u": 32, |
|
|
"v": 33, |
|
|
"w": 34, |
|
|
"x": 35, |
|
|
"y": 36, |
|
|
"z": 37, |
|
|
"|": 0, |
|
|
"½": 38, |
|
|
"¾": 39, |
|
|
"×": 40, |
|
|
"õ": 41, |
|
|
"،": 42, |
|
|
"؛": 43, |
|
|
"ء": 44, |
|
|
"آ": 45, |
|
|
"أ": 46, |
|
|
"ؤ": 47, |
|
|
"إ": 48, |
|
|
"ئ": 49, |
|
|
"ا": 50, |
|
|
"ب": 51, |
|
|
"ة": 52, |
|
|
"ت": 53, |
|
|
"ث": 54, |
|
|
"ج": 55, |
|
|
"ح": 56, |
|
|
"خ": 57, |
|
|
"د": 58, |
|
|
"ذ": 59, |
|
|
"ر": 60, |
|
|
"ز": 61, |
|
|
"س": 62, |
|
|
"ش": 63, |
|
|
"ص": 64, |
|
|
"ض": 65, |
|
|
"ط": 66, |
|
|
"ظ": 67, |
|
|
"ع": 68, |
|
|
"غ": 69, |
|
|
"ـ": 70, |
|
|
"ف": 71, |
|
|
"ق": 72, |
|
|
"ك": 73, |
|
|
"ل": 74, |
|
|
"م": 75, |
|
|
"ن": 76, |
|
|
"ه": 77, |
|
|
"و": 78, |
|
|
"ى": 79, |
|
|
"ي": 80, |
|
|
"ً": 81, |
|
|
"ٌ": 82, |
|
|
"ٍ": 83, |
|
|
"َ": 84, |
|
|
"ُ": 85, |
|
|
"ِ": 86, |
|
|
"ّ": 87, |
|
|
"ْ": 88, |
|
|
"٠": 89, |
|
|
"١": 90, |
|
|
"٢": 91, |
|
|
"٣": 92, |
|
|
"٥": 93, |
|
|
"٦": 94, |
|
|
"٧": 95, |
|
|
"٨": 96, |
|
|
"٩": 97, |
|
|
"٪": 98, |
|
|
"چ": 99, |
|
|
"—": 100 |
|
|
} |
|
|
} |
|
|
|
