File size: 6,933 Bytes
4276ff5 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 | {
"tokenizer_config": {
"name": "sudanizer-72k",
"version": "1.0.0",
"vocab_size": 72000,
"min_frequency": 3,
"limit_alphabet": 5000,
"pre_tokenizer": "whitespace",
"dialect_normalization": false,
"hamza_normalization": true
},
"training_metrics": {
"total_samples": 500,
"total_characters": 86872,
"total_tokens": 19153,
"total_words": 16772,
"avg_chars_per_token": 4.536,
"compression_ratio": 0.2205,
"unk_count": 0,
"unk_rate_percent": 0.0,
"avg_subwords_per_word": 1.159,
"vocab_size": 72000,
"max_tokens_in_sample": 868,
"min_tokens_in_sample": 2,
"avg_tokens_per_sample": 38.31
},
"comprehensive_evaluation": {
"by_category": {
"pure_dialect": {
"total_samples": 10,
"total_characters": 311,
"total_tokens": 66,
"total_words": 62,
"avg_chars_per_token": 4.712,
"compression_ratio": 0.2122,
"unk_count": 0,
"unk_rate_percent": 0.0,
"avg_subwords_per_word": 1.064,
"vocab_size": 72000,
"max_tokens_in_sample": 8,
"min_tokens_in_sample": 5,
"avg_tokens_per_sample": 6.6
},
"code_switching": {
"total_samples": 8,
"total_characters": 349,
"total_tokens": 156,
"total_words": 53,
"avg_chars_per_token": 2.237,
"compression_ratio": 0.447,
"unk_count": 0,
"unk_rate_percent": 0.0,
"avg_subwords_per_word": 2.984,
"vocab_size": 72000,
"max_tokens_in_sample": 27,
"min_tokens_in_sample": 13,
"avg_tokens_per_sample": 19.5
},
"social_media": {
"total_samples": 10,
"total_characters": 378,
"total_tokens": 129,
"total_words": 72,
"avg_chars_per_token": 2.93,
"compression_ratio": 0.3413,
"unk_count": 29,
"unk_rate_percent": 22.4806,
"avg_subwords_per_word": 1.853,
"vocab_size": 72000,
"max_tokens_in_sample": 25,
"min_tokens_in_sample": 10,
"avg_tokens_per_sample": 12.9
},
"formal": {
"total_samples": 5,
"total_characters": 356,
"total_tokens": 68,
"total_words": 51,
"avg_chars_per_token": 5.235,
"compression_ratio": 0.191,
"unk_count": 1,
"unk_rate_percent": 1.4706,
"avg_subwords_per_word": 1.315,
"vocab_size": 72000,
"max_tokens_in_sample": 19,
"min_tokens_in_sample": 9,
"avg_tokens_per_sample": 13.6
},
"edge_case": {
"total_samples": 8,
"total_characters": 355,
"total_tokens": 76,
"total_words": 50,
"avg_chars_per_token": 4.671,
"compression_ratio": 0.2141,
"unk_count": 1,
"unk_rate_percent": 1.3158,
"avg_subwords_per_word": 1.525,
"vocab_size": 72000,
"max_tokens_in_sample": 13,
"min_tokens_in_sample": 8,
"avg_tokens_per_sample": 9.5
},
"morphology": {
"total_samples": 8,
"total_characters": 216,
"total_tokens": 48,
"total_words": 40,
"avg_chars_per_token": 4.5,
"compression_ratio": 0.2222,
"unk_count": 0,
"unk_rate_percent": 0.0,
"avg_subwords_per_word": 1.205,
"vocab_size": 72000,
"max_tokens_in_sample": 8,
"min_tokens_in_sample": 4,
"avg_tokens_per_sample": 6.0
},
"cultural": {
"total_samples": 8,
"total_characters": 267,
"total_tokens": 59,
"total_words": 49,
"avg_chars_per_token": 4.525,
"compression_ratio": 0.221,
"unk_count": 0,
"unk_rate_percent": 0.0,
"avg_subwords_per_word": 1.187,
"vocab_size": 72000,
"max_tokens_in_sample": 12,
"min_tokens_in_sample": 5,
"avg_tokens_per_sample": 7.38
},
"mixed_content": {
"total_samples": 8,
"total_characters": 224,
"total_tokens": 81,
"total_words": 33,
"avg_chars_per_token": 2.765,
"compression_ratio": 0.3616,
"unk_count": 1,
"unk_rate_percent": 1.2346,
"avg_subwords_per_word": 3.017,
"vocab_size": 72000,
"max_tokens_in_sample": 18,
"min_tokens_in_sample": 5,
"avg_tokens_per_sample": 10.12
}
},
"by_difficulty": {
"easy": {
"total_samples": 4,
"total_characters": 157,
"total_tokens": 28,
"total_words": 28,
"avg_chars_per_token": 5.607,
"compression_ratio": 0.1783,
"unk_count": 0,
"unk_rate_percent": 0.0,
"avg_subwords_per_word": 1.0,
"vocab_size": 72000,
"max_tokens_in_sample": 9,
"min_tokens_in_sample": 5,
"avg_tokens_per_sample": 7.0
},
"medium": {
"total_samples": 22,
"total_characters": 721,
"total_tokens": 167,
"total_words": 129,
"avg_chars_per_token": 4.317,
"compression_ratio": 0.2316,
"unk_count": 4,
"unk_rate_percent": 2.3952,
"avg_subwords_per_word": 1.349,
"vocab_size": 72000,
"max_tokens_in_sample": 16,
"min_tokens_in_sample": 4,
"avg_tokens_per_sample": 7.59
},
"hard": {
"total_samples": 27,
"total_characters": 1073,
"total_tokens": 315,
"total_words": 180,
"avg_chars_per_token": 3.406,
"compression_ratio": 0.2936,
"unk_count": 21,
"unk_rate_percent": 6.6667,
"avg_subwords_per_word": 1.794,
"vocab_size": 72000,
"max_tokens_in_sample": 27,
"min_tokens_in_sample": 5,
"avg_tokens_per_sample": 11.67
},
"extreme": {
"total_samples": 12,
"total_characters": 505,
"total_tokens": 173,
"total_words": 73,
"avg_chars_per_token": 2.919,
"compression_ratio": 0.3426,
"unk_count": 7,
"unk_rate_percent": 4.0462,
"avg_subwords_per_word": 2.748,
"vocab_size": 72000,
"max_tokens_in_sample": 27,
"min_tokens_in_sample": 6,
"avg_tokens_per_sample": 14.42
}
},
"overall": {
"total_samples": 65,
"total_characters": 2456,
"total_tokens": 683,
"total_words": 410,
"avg_chars_per_token": 3.596,
"compression_ratio": 0.2781,
"unk_count": 32,
"unk_rate_percent": 4.6852,
"avg_subwords_per_word": 1.771,
"vocab_size": 72000,
"max_tokens_in_sample": 27,
"min_tokens_in_sample": 4,
"avg_tokens_per_sample": 10.51
},
"problematic_samples_count": 18
},
"benchmark_results": {
"Sudanizer-72K-v2": {
"avg_chars_per_token": 3.596,
"unk_rate_percent": 4.6852,
"avg_subwords_per_word": 1.771,
"total_tokens": 683
}
}
} |