diyclassics's picture
v3.10.0: floret lg vectors retrained on v3.10 corpus; updated model card (model files + README)
347c1d9 verified
Raw
History Blame Contribute Delete
1.21 kB
{
"lang":"la",
"name":"vectors_floret_lg",
"version":"3.10.0",
"description":"Floret word vectors for Latin, trained on 14.3M sentences (278M tokens) from 15 sources including UD/LASLA treebanks, CC100, Wikisource, Wikipedia, and other Latin corpora. Part of the LatinCy project.",
"author":"Patrick J. Burns",
"email":"patrick@diyclassics.org",
"url":"https://huggingface.co/latincy",
"license":"MIT",
"spacy_version":">=3.8.11,<3.9.0",
"spacy_git_version":"e7a662a",
"vectors":{
"width":300,
"vectors":200000,
"keys":-1,
"name":"la_vectors_floret_lg.vectors"
},
"labels":{
},
"pipeline":[
],
"components":[
],
"disabled":[
],
"title":"la_vectors_floret",
"sources":[
"UD_Latin-CIRCSE",
"UD_Latin-Perseus",
"UD_Latin-PROIEL",
"UD_Latin-ITTB",
"UD_Latin-LLCT",
"UD_Latin-UDante",
"CC100-Latin",
"Latin Wikisource",
"Latin Wikipedia",
"The Latin Library",
"CAMENA Neo-Latin",
"Patrologia Latina",
"Perseus Digital Library",
"CLTK-Tesserae Latin",
"CSEL",
"digilibLT",
"EDH",
"Formulae",
"Epistolae",
"PTA"
],
"requirements":[
"spacy>=3.8.11,<3.9.0"
]
}