| { | |
| "arch": "cnn", | |
| "max_len": 128, | |
| "labels": [ | |
| "boundary", | |
| "commercial", | |
| "dedicatory", | |
| "funerary", | |
| "legal", | |
| "ownership", | |
| "votive" | |
| ], | |
| "vocab": { | |
| "char_to_idx": { | |
| "[PAD]": 0, | |
| "[UNK]": 1, | |
| "[MASK]": 2, | |
| "\t": 3, | |
| "\n": 4, | |
| "\r": 5, | |
| " ": 6, | |
| "\"": 7, | |
| "%": 8, | |
| "&": 9, | |
| "'": 10, | |
| "(": 11, | |
| ")": 12, | |
| "+": 13, | |
| ",": 14, | |
| "-": 15, | |
| ".": 16, | |
| "/": 17, | |
| "0": 18, | |
| "1": 19, | |
| "2": 20, | |
| "3": 21, | |
| "4": 22, | |
| "5": 23, | |
| "6": 24, | |
| "7": 25, | |
| "8": 26, | |
| "9": 27, | |
| ":": 28, | |
| ";": 29, | |
| "<": 30, | |
| "=": 31, | |
| ">": 32, | |
| "?": 33, | |
| "[": 34, | |
| "\\": 35, | |
| "]": 36, | |
| "a": 37, | |
| "b": 38, | |
| "c": 39, | |
| "d": 40, | |
| "e": 41, | |
| "f": 42, | |
| "g": 43, | |
| "h": 44, | |
| "i": 45, | |
| "j": 46, | |
| "k": 47, | |
| "l": 48, | |
| "m": 49, | |
| "n": 50, | |
| "o": 51, | |
| "p": 52, | |
| "q": 53, | |
| "r": 54, | |
| "s": 55, | |
| "t": 56, | |
| "u": 57, | |
| "v": 58, | |
| "x": 59, | |
| "y": 60, | |
| "z": 61, | |
| "{": 62, | |
| "|": 63, | |
| "}": 64, | |
| "²": 65, | |
| "·": 66, | |
| "¹": 67, | |
| "ç": 68, | |
| "ê": 69, | |
| "ð": 70, | |
| "ñ": 71, | |
| "þ": 72, | |
| "ą": 73, | |
| "ę": 74, | |
| "į": 75, | |
| "ļ": 76, | |
| "ł": 77, | |
| "ś": 78, | |
| "ş": 79, | |
| "š": 80, | |
| "ţ": 81, | |
| "ǝ": 82, | |
| "ș": 83, | |
| "ț": 84, | |
| "ɛ": 85, | |
| "ί": 86, | |
| "α": 87, | |
| "ε": 88, | |
| "ζ": 89, | |
| "η": 90, | |
| "θ": 91, | |
| "ι": 92, | |
| "κ": 93, | |
| "μ": 94, | |
| "ν": 95, | |
| "ξ": 96, | |
| "ο": 97, | |
| "ρ": 98, | |
| "ς": 99, | |
| "σ": 100, | |
| "υ": 101, | |
| "φ": 102, | |
| "χ": 103, | |
| "а": 104, | |
| "з": 105, | |
| "и": 106, | |
| "м": 107, | |
| "н": 108, | |
| "о": 109, | |
| "п": 110, | |
| "т": 111, | |
| "ч": 112, | |
| "ш": 113, | |
| "я": 114, | |
| "і": 115, | |
| "ִ": 116, | |
| "۰": 117, | |
| "۱": 118, | |
| "۲": 119, | |
| "۷": 120, | |
| "۹": 121, | |
| "ง": 122, | |
| "จ": 123, | |
| "ด": 124, | |
| "ท": 125, | |
| "า": 126, | |
| "เ": 127, | |
| "ḟ": 128, | |
| "ḥ": 129, | |
| "ṇ": 130, | |
| "ṭ": 131, | |
| "ṿ": 132, | |
| "ạ": 133, | |
| "ẹ": 134, | |
| "„": 135, | |
| "‡": 136, | |
| "•": 137, | |
| "…": 138, | |
| "ⅎ": 139, | |
| "∃": 140, | |
| "コ": 141, | |
| "ネ": 142, | |
| "マ": 143, | |
| "ヨ": 144, | |
| "・": 145, | |
| "ㅋ": 146, | |
| "ꞇ": 147, | |
| "・": 148, | |
| "𐌏": 149, | |
| "𐌙": 150 | |
| } | |
| }, | |
| "vocab_size": 151 | |
| } |