allahserix's picture
add tokenizer
3678f60
raw
history blame contribute delete
415 Bytes
{
"'": 16,
"[PAD]": 36,
"[UNK]": 35,
"a": 12,
"b": 11,
"c": 18,
"d": 5,
"e": 17,
"f": 10,
"g": 20,
"h": 27,
"i": 1,
"j": 0,
"k": 29,
"l": 6,
"m": 15,
"n": 28,
"o": 23,
"p": 7,
"q": 8,
"r": 26,
"s": 30,
"t": 13,
"u": 19,
"v": 31,
"w": 34,
"x": 3,
"y": 2,
"z": 9,
"|": 22,
"ç": 14,
"ê": 21,
"ŋ": 24,
"ɔ": 25,
"ɛ": 32,
"ɲ": 33,
"̌": 4
}