symphonym-v8 / training_stats /coverage_stats.json
docuracy's picture
Upload Symphonym v8 model, vocabularies, and evaluation results
c02bc8b verified
Raw History Blame Contribute Delete
3.6 kB
{
"total_toponyms": 73479069,
"in_training_namespaces": 62081769,
"with_ipa": 34210114,
"with_panphon_embedding": 34210114,
"panphon_coverage_pct": 55.10492782510756,
"from_db_cache": 0,
"from_precomputed": 2267714,
"from_epitran": 31942400,
"by_script": {
"LATIN": 60920805,
"CYRILLIC": 4234862,
"ARABIC": 2251220,
"CJK": 3240684,
"THAI": 261989,
"KATAKANA": 358111,
"HANGUL": 416894,
"DEVANAGARI": 184963,
"MYANMAR": 120482,
"ARMENIAN": 165304,
"BOPOMOFO": 3194,
"GURMUKHI": 23399,
"OTHER": 134345,
"HEBREW": 162324,
"GREEK": 245549,
"BENGALI": 125118,
"GEORGIAN": 112180,
"MALAYALAM": 73902,
"HIRAGANA": 153929,
"TIFINAGH": 12563,
"TIBETAN": 17534,
"ORIYA": 8996,
"GUJARATI": 21741,
"MONGOLIAN": 5516,
"TELUGU": 53434,
"THAANA": 1858,
"ETHIOPIC": 10750,
"TAMIL": 54908,
"OL_CHIKI": 13001,
"KHMER": 13897,
"SINHALA": 15523,
"LAO": 7567,
"KANNADA": 45743,
"SYRIAC": 967,
"NKO": 1208,
"CANADIAN_ABORIGINAL": 3667,
"COPTIC": 942
},
"by_script_lang_ipa": {
"LATIN:en": 8887527,
"LATIN:nl": 2617247,
"LATIN:fr": 2430235,
"LATIN:de": 2290310,
"LATIN:sv": 1819495,
"LATIN:es": 1630909,
"CJK:zh": 1345295,
"LATIN:id": 1070569,
"LATIN:tr": 1010536,
"CYRILLIC:ru": 969983,
"LATIN:it": 831880,
"LATIN:pt": 741370,
"LATIN:pl": 704524,
"LATIN:cs": 613932,
"ARABIC:fa": 609405,
"LATIN:fi": 537458,
"CYRILLIC:uk": 509090,
"LATIN:no": 428488,
"ARABIC:ar": 423519,
"LATIN:zh": 419774,
"LATIN:ro": 398698,
"KATAKANA:ja": 320242,
"LATIN:vi": 305191,
"LATIN:da": 299724,
"LATIN:ms": 296819,
"CYRILLIC:sr": 266777,
"LATIN:hu": 250833,
"CYRILLIC:bg": 242123,
"HANGUL:ko": 236220,
"THAI:th": 210566,
"GREEK:el": 172004,
"ARMENIAN:hy": 150927,
"HEBREW:he": 131785,
"LATIN:sw": 118677,
"ARABIC:ur": 112650,
"BENGALI:bn": 92192,
"GEORGIAN:ka": 88896,
"LATIN:la": 78318,
"CYRILLIC:mk": 68026,
"DEVANAGARI:hi": 65639,
"MALAYALAM:ml": 55871,
"CJK:wuu": 49119,
"TELUGU:te": 48622,
"TAMIL:ta": 48365,
"HIRAGANA:ja": 47629,
"CJK:gan": 37112,
"CJK:yue": 31705,
"DEVANAGARI:mr": 24929,
"KANNADA:kn": 21108,
"GUJARATI:gu": 20540,
"LATIN:yue": 13809,
"DEVANAGARI:ne": 10557,
"CJK:ko": 2074,
"LATIN:wuu": 266,
"LATIN:ko": 99,
"KATAKANA:zh": 69,
"LATIN:gan": 59,
"CYRILLIC:zh": 53,
"OTHER:zh": 45,
"LATIN:he": 42,
"HANGUL:zh": 18,
"OTHER:ko": 16,
"HIRAGANA:yue": 15,
"HIRAGANA:zh": 14,
"ARABIC:yue": 14,
"ARABIC:zh": 13,
"GREEK:zh": 10,
"OTHER:he": 9,
"THAI:zh": 7,
"CYRILLIC:yue": 7,
"THAI:yue": 7,
"TIBETAN:zh": 4,
"KATAKANA:ko": 4,
"HEBREW:yue": 4,
"KATAKANA:yue": 3,
"CYRILLIC:ko": 3,
"THAI:wuu": 3,
"ARABIC:ko": 3,
"KHMER:ko": 3,
"BENGALI:yue": 3,
"DEVANAGARI:zh": 3,
"HANGUL:yue": 2,
"CYRILLIC:wuu": 2,
"MONGOLIAN:yue": 2,
"TELUGU:zh": 2,
"TAMIL:yue": 2,
"KHMER:zh": 2,
"ARMENIAN:yue": 2,
"GREEK:ko": 1,
"OTHER:yue": 1,
"GEORGIAN:yue": 1,
"TAMIL:zh": 1,
"MONGOLIAN:zh": 1,
"HEBREW:zh": 1,
"OTHER:gan": 1,
"KHMER:yue": 1,
"SINHALA:yue": 1,
"BOPOMOFO:zh": 1,
"OTHER:wuu": 1
},
"training_namespaces": [
"gn",
"wd",
"tgn"
],
"num_workers": 14,
"db_engine": "DuckDB",
"ipa_backends": [
"epitran",
"phonikud",
"charsiu_g2p"
]
}