Doogie's picture
add tokenizer
2d839b6
raw
history blame
268 Bytes
{"r": 0, "f": 1, "'": 2, "j": 3, "v": 4, "n": 5, "d": 6, "o": 7, "x": 8, "a": 9, "m": 10, "i": 11, "h": 12, "w": 13, "t": 14, "y": 15, "p": 16, "c": 17, "q": 18, "b": 19, "k": 20, "z": 21, "e": 22, "g": 23, "l": 24, "s": 25, "u": 26, "|": 27, "[UNK]": 28, "[PAD]": 29}