add tokenizer

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

vocab.json

@@ -1 +1 @@
-{"x": 0, "u": 2, "y": 3, "m": 4, "d": 5, "s": 6, "k": 7, "a": 8, "b": 9, "z": 10, "t": 11, "i": 12, "f": 13, "h": 14, "v": 15, "w": 16, "g": 17, "c": 18, "e": 19, "p": 20, "l": 21, "o": 22, "n": 23, "r": 24, "|": 1, "[UNK]": 25, "[PAD]": 26}
+{"x": 0, "b": 1, "c": 3, "w": 4, "s": 5, "f": 6, "h": 7, "n": 8, "g": 9, "e": 10, "a": 11, "z": 12, "u": 13, "k": 14, "r": 15, "o": 16, "v": 17, "m": 18, "i": 19, "t": 20, "l": 21, "p": 22, "y": 23, "d": 24, "|": 2, "[UNK]": 25, "[PAD]": 26}