| { | |
| "padding": true, | |
| "return_lengths": false, | |
| "return_word_embeddings": false, | |
| "return_casing": false, | |
| "return_features": false, | |
| "return_chars": false, | |
| "return_bert_embeddings": true, | |
| "vocab_char": { | |
| "<PAD>": 0, | |
| "<UNK>": 1, | |
| "\"": 2, | |
| "#": 3, | |
| "%": 4, | |
| "&": 5, | |
| "'": 6, | |
| "(": 7, | |
| ")": 8, | |
| "*": 9, | |
| "+": 10, | |
| ",": 11, | |
| "-": 12, | |
| ".": 13, | |
| "/": 14, | |
| "0": 15, | |
| "1": 16, | |
| "2": 17, | |
| "3": 18, | |
| "4": 19, | |
| "5": 20, | |
| "6": 21, | |
| "7": 22, | |
| "8": 23, | |
| "9": 24, | |
| ":": 25, | |
| ";": 26, | |
| "<": 27, | |
| "=": 28, | |
| ">": 29, | |
| "?": 30, | |
| "@": 31, | |
| "A": 32, | |
| "B": 33, | |
| "C": 34, | |
| "D": 35, | |
| "E": 36, | |
| "F": 37, | |
| "G": 38, | |
| "H": 39, | |
| "I": 40, | |
| "J": 41, | |
| "K": 42, | |
| "L": 43, | |
| "M": 44, | |
| "N": 45, | |
| "O": 46, | |
| "P": 47, | |
| "Q": 48, | |
| "R": 49, | |
| "S": 50, | |
| "T": 51, | |
| "U": 52, | |
| "V": 53, | |
| "W": 54, | |
| "X": 55, | |
| "Y": 56, | |
| "Z": 57, | |
| "[": 58, | |
| "]": 59, | |
| "_": 60, | |
| "`": 61, | |
| "a": 62, | |
| "b": 63, | |
| "c": 64, | |
| "d": 65, | |
| "e": 66, | |
| "f": 67, | |
| "g": 68, | |
| "h": 69, | |
| "i": 70, | |
| "j": 71, | |
| "k": 72, | |
| "l": 73, | |
| "m": 74, | |
| "n": 75, | |
| "o": 76, | |
| "p": 77, | |
| "q": 78, | |
| "r": 79, | |
| "s": 80, | |
| "t": 81, | |
| "u": 82, | |
| "v": 83, | |
| "w": 84, | |
| "x": 85, | |
| "y": 86, | |
| "z": 87, | |
| "{": 88, | |
| "|": 89, | |
| "}": 90, | |
| "~": 91, | |
| "\u0080": 92, | |
| "\u0083": 93, | |
| "\u0089": 94, | |
| "\u008e": 95, | |
| "\u008f": 96, | |
| "\u0091": 97, | |
| "\u0092": 98, | |
| "\u0098": 99, | |
| "\u009f": 100, | |
| "\u00a1": 101, | |
| "\u00a8": 102, | |
| "\u00ae": 103, | |
| "\u00af": 104, | |
| "\u00b0": 105, | |
| "\u00b1": 106, | |
| "\u00b2": 107, | |
| "\u00b3": 108, | |
| "\u00b4": 109, | |
| "\u00b5": 110, | |
| "\u00ba": 111, | |
| "\u00bb": 112, | |
| "\u00bc": 113, | |
| "\u00bd": 114, | |
| "\u00c1": 115, | |
| "\u00c3": 116, | |
| "\u00c5": 117, | |
| "\u00d2": 118, | |
| "\u00d7": 119, | |
| "\u00dc": 120, | |
| "\u00de": 121, | |
| "\u00df": 122, | |
| "\u00e0": 123, | |
| "\u00e1": 124, | |
| "\u00e2": 125, | |
| "\u00e4": 126, | |
| "\u00e7": 127, | |
| "\u00e8": 128, | |
| "\u00e9": 129, | |
| "\u00ea": 130, | |
| "\u00ed": 131, | |
| "\u00ef": 132, | |
| "\u00f1": 133, | |
| "\u00f4": 134, | |
| "\u00f6": 135, | |
| "\u00fa": 136, | |
| "\u00fc": 137, | |
| "\u0131": 138, | |
| "\u01eb": 139, | |
| "\u02c6": 140, | |
| "\u02d8": 141, | |
| "\u02d9": 142, | |
| "\u0393": 143, | |
| "\u0394": 144, | |
| "\u0398": 145, | |
| "\u039b": 146, | |
| "\u03a6": 147, | |
| "\u03a9": 148, | |
| "\u03b1": 149, | |
| "\u03b2": 150, | |
| "\u03b3": 151, | |
| "\u03b4": 152, | |
| "\u03b5": 153, | |
| "\u03b6": 154, | |
| "\u03b7": 155, | |
| "\u03b8": 156, | |
| "\u03ba": 157, | |
| "\u03bb": 158, | |
| "\u03bc": 159, | |
| "\u03bd": 160, | |
| "\u03be": 161, | |
| "\u03c0": 162, | |
| "\u03c1": 163, | |
| "\u03c3": 164, | |
| "\u03c4": 165, | |
| "\u03c5": 166, | |
| "\u03c7": 167, | |
| "\u03c8": 168, | |
| "\u03c9": 169, | |
| "\u03d5": 170, | |
| "\u2021": 171, | |
| "\u2022": 172, | |
| "\u2026": 173, | |
| "\u2032": 174, | |
| "\u2044": 175, | |
| "\u2082": 176, | |
| "\u20ac": 177, | |
| "\u2103": 178, | |
| "\u2126": 179, | |
| "\u2192": 180, | |
| "\u21d4": 181, | |
| "\u2200": 182, | |
| "\u2202": 183, | |
| "\u2206": 184, | |
| "\u2207": 185, | |
| "\u2208": 186, | |
| "\u2218": 187, | |
| "\u221a": 188, | |
| "\u221d": 189, | |
| "\u221e": 190, | |
| "\u223c": 191, | |
| "\u2243": 192, | |
| "\u2248": 193, | |
| "\u2261": 194, | |
| "\u2264": 195, | |
| "\u2265": 196, | |
| "\u226a": 197, | |
| "\u226b": 198, | |
| "\u2295": 199, | |
| "\u2299": 200, | |
| "\u22c6": 201, | |
| "\u230a": 202, | |
| "\u230b": 203, | |
| "\u2642": 204, | |
| "\uf020": 205, | |
| "\uf026": 206, | |
| "\uf028": 207, | |
| "\uf029": 208, | |
| "\uf02b": 209, | |
| "\uf02d": 210, | |
| "\uf02e": 211, | |
| "\uf02f": 212, | |
| "\uf030": 213, | |
| "\uf031": 214, | |
| "\uf032": 215, | |
| "\uf033": 216, | |
| "\uf034": 217, | |
| "\uf035": 218, | |
| "\uf036": 219, | |
| "\uf037": 220, | |
| "\uf038": 221, | |
| "\uf03c": 222, | |
| "\uf044": 223, | |
| "\uf057": 224, | |
| "\uf061": 225, | |
| "\uf062": 226, | |
| "\uf06c": 227, | |
| "\uf06d": 228, | |
| "\uf073": 229, | |
| "\uf076": 230, | |
| "\uf0a3": 231, | |
| "\uf0a7": 232, | |
| "\uf0b3": 233, | |
| "\uf0b7": 234, | |
| "\uf0e5": 235, | |
| "\uf0ec": 236, | |
| "\uf0ed": 237, | |
| "\uf0ee": 238, | |
| "\uf0ef": 239, | |
| "\uff0c": 240, | |
| "\ud97b\udf59": 241 | |
| }, | |
| "vocab_tag": { | |
| "<PAD>": 0, | |
| "B-<unitLeft>": 1, | |
| "B-<unitRight>": 2, | |
| "B-<valueAtomic>": 3, | |
| "B-<valueBase>": 4, | |
| "B-<valueLeast>": 5, | |
| "B-<valueList>": 6, | |
| "B-<valueMost>": 7, | |
| "B-<valueRange>": 8, | |
| "I-<unitLeft>": 9, | |
| "I-<unitRight>": 10, | |
| "I-<valueAtomic>": 11, | |
| "I-<valueBase>": 12, | |
| "I-<valueLeast>": 13, | |
| "I-<valueList>": 14, | |
| "I-<valueMost>": 15, | |
| "I-<valueRange>": 16, | |
| "O": 17 | |
| }, | |
| "vocab_case": [ | |
| "<PAD>", | |
| "numeric", | |
| "allLower", | |
| "allUpper", | |
| "initialUpper", | |
| "other", | |
| "mainly_numeric", | |
| "contains_digit" | |
| ], | |
| "max_char_length": 30, | |
| "feature_preprocessor": null, | |
| "indice_tag": { | |
| "0": "<PAD>", | |
| "1": "B-<unitLeft>", | |
| "2": "B-<unitRight>", | |
| "3": "B-<valueAtomic>", | |
| "4": "B-<valueBase>", | |
| "5": "B-<valueLeast>", | |
| "6": "B-<valueList>", | |
| "7": "B-<valueMost>", | |
| "8": "B-<valueRange>", | |
| "9": "I-<unitLeft>", | |
| "10": "I-<unitRight>", | |
| "11": "I-<valueAtomic>", | |
| "12": "I-<valueBase>", | |
| "13": "I-<valueLeast>", | |
| "14": "I-<valueList>", | |
| "15": "I-<valueMost>", | |
| "16": "I-<valueRange>", | |
| "17": "O" | |
| } | |
| } |