Add config metadata
Browse files- config.json +112 -0
config.json
ADDED
|
@@ -0,0 +1,112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_name": "trankit decompressed weights",
|
| 3 |
+
"version": "1.0.0",
|
| 4 |
+
"library_name": "trankit",
|
| 5 |
+
"source": "https://github.com/nlp-uoregon/trankit",
|
| 6 |
+
"license": "apache-2.0",
|
| 7 |
+
"architectures": [
|
| 8 |
+
"xlm-roberta-base",
|
| 9 |
+
"xlm-roberta-large"
|
| 10 |
+
],
|
| 11 |
+
"tasks": [
|
| 12 |
+
"tokenization",
|
| 13 |
+
"pos-tagging",
|
| 14 |
+
"dependency-parsing",
|
| 15 |
+
"named-entity-recognition",
|
| 16 |
+
"lemmatization"
|
| 17 |
+
],
|
| 18 |
+
"languages": [
|
| 19 |
+
"afrikaans",
|
| 20 |
+
"ancient-greek",
|
| 21 |
+
"ancient-greek-perseus",
|
| 22 |
+
"arabic",
|
| 23 |
+
"armenian",
|
| 24 |
+
"basque",
|
| 25 |
+
"belarusian",
|
| 26 |
+
"bulgarian",
|
| 27 |
+
"catalan",
|
| 28 |
+
"chinese",
|
| 29 |
+
"classical-chinese",
|
| 30 |
+
"croatian",
|
| 31 |
+
"czech",
|
| 32 |
+
"czech-cac",
|
| 33 |
+
"czech-cltt",
|
| 34 |
+
"czech-fictree",
|
| 35 |
+
"danish",
|
| 36 |
+
"dutch",
|
| 37 |
+
"dutch-lassysmall",
|
| 38 |
+
"english",
|
| 39 |
+
"english-gum",
|
| 40 |
+
"english-lines",
|
| 41 |
+
"english-partut",
|
| 42 |
+
"estonian",
|
| 43 |
+
"estonian-ewt",
|
| 44 |
+
"finnish",
|
| 45 |
+
"finnish-ftb",
|
| 46 |
+
"french",
|
| 47 |
+
"french-partut",
|
| 48 |
+
"french-sequoia",
|
| 49 |
+
"french-spoken",
|
| 50 |
+
"galician",
|
| 51 |
+
"galician-treegal",
|
| 52 |
+
"german",
|
| 53 |
+
"german-hdt",
|
| 54 |
+
"greek",
|
| 55 |
+
"hebrew",
|
| 56 |
+
"hindi",
|
| 57 |
+
"hungarian",
|
| 58 |
+
"indonesian",
|
| 59 |
+
"irish",
|
| 60 |
+
"italian",
|
| 61 |
+
"italian-partut",
|
| 62 |
+
"italian-postwita",
|
| 63 |
+
"italian-twittiro",
|
| 64 |
+
"italian-vit",
|
| 65 |
+
"japanese",
|
| 66 |
+
"kazakh",
|
| 67 |
+
"korean",
|
| 68 |
+
"korean-kaist",
|
| 69 |
+
"kurmanji",
|
| 70 |
+
"latin",
|
| 71 |
+
"latin-perseus",
|
| 72 |
+
"latin-proiel",
|
| 73 |
+
"latvian",
|
| 74 |
+
"lithuanian",
|
| 75 |
+
"lithuanian-hse",
|
| 76 |
+
"marathi",
|
| 77 |
+
"norwegian-bokmaal",
|
| 78 |
+
"norwegian-nynorsk",
|
| 79 |
+
"norwegian-nynorsklia",
|
| 80 |
+
"old-french",
|
| 81 |
+
"old-russian",
|
| 82 |
+
"persian",
|
| 83 |
+
"polish",
|
| 84 |
+
"polish-lfg",
|
| 85 |
+
"portuguese",
|
| 86 |
+
"portuguese-gsd",
|
| 87 |
+
"romanian",
|
| 88 |
+
"romanian-nonstandard",
|
| 89 |
+
"russian",
|
| 90 |
+
"russian-gsd",
|
| 91 |
+
"russian-taiga",
|
| 92 |
+
"scottish-gaelic",
|
| 93 |
+
"serbian",
|
| 94 |
+
"slovak",
|
| 95 |
+
"slovenian",
|
| 96 |
+
"slovenian-sst",
|
| 97 |
+
"spanish",
|
| 98 |
+
"spanish-gsd",
|
| 99 |
+
"swedish",
|
| 100 |
+
"swedish-lines",
|
| 101 |
+
"tamil",
|
| 102 |
+
"telugu",
|
| 103 |
+
"traditional-chinese",
|
| 104 |
+
"turkish",
|
| 105 |
+
"ukrainian",
|
| 106 |
+
"urdu",
|
| 107 |
+
"uyghur",
|
| 108 |
+
"vietnamese",
|
| 109 |
+
"vietnamese-vtb"
|
| 110 |
+
],
|
| 111 |
+
"notes": "All files were extracted from uonlp/trankit so that downstream users can download ready-to-use weights without zip archives."
|
| 112 |
+
}
|