meta.json

{
  "lang":"fa",
  "name":"floret_400k",
  "version":"0.1.0",
  "description":"Persian floret static vector table: 50,000 rows x 300 dimensions, floret mode, minn=maxn=5, hash_count=2, trained with floret-torch on the first 400,000 articles of the Persian Wikipedia dump. Vectors only, no pipeline components. This is the table used by the fa_*_news_md tier.",
  "author":"Kiyarash Fazeli",
  "email":"kiyarash@nlogn.ir",
  "url":"https://github.com/Fazel94/spacy-persian",
  "license":"CC BY-SA 4.0",
  "sources":[
    {
      "name":"fa_floret static vectors (50k rows x 300d, floret mode, first 400,000 Persian Wikipedia articles, trained with floret-torch)",
      "url":"https://github.com/Fazel94/spacy-persian",
      "author":"Kiyarash Fazeli",
      "license":"CC BY-SA 4.0"
    },
    {
      "name":"Persian Wikipedia dump (fawiki), the text the vector table is trained on",
      "url":"https://dumps.wikimedia.org/fawiki/",
      "author":"Wikipedia contributors",
      "license":"CC BY-SA 4.0"
    }
  ],
  "spacy_version":">=3.8.15,<3.9.0",
  "spacy_git_version":"f69c32f",
  "vectors":{
    "width":300,
    "vectors":50000,
    "keys":-1,
    "name":"floret"
  },
  "labels":{

  },
  "pipeline":[

  ],
  "components":[

  ],
  "disabled":[

  ],
  "requirements":[
    "spacy>=3.8.15,<3.9.0",
    "spacy>=3.8.15,<3.9.0"
  ],
  "vectors_training":{
    "corpus":"first 400,000 articles of the Persian Wikipedia (`fawiki`) dump, extracted with WikiExtractor `--no-templates`, sentence-tokenized with spaCy `blank(\"fa\")`",
    "tokens":"~163M (project estimate for 400,000 articles; not measured for this run)",
    "epochs":5,
    "minn":5,
    "maxn":5,
    "hash_count":2,
    "min_count":20,
    "tool":"[floret-torch](https://github.com/Fazel94/floret-torch) (GPU port of [explosion/floret](https://github.com/explosion/floret)), cbow, on a Colab T4",
    "cli":"python -m floret_torch.train --input fa.txt --output fa \\\n  --model cbow --mode floret --dim 300 --minn 5 --maxn 5 \\\n  --neg 10 --epoch 5 --hashCount 2 --bucket 50000 \\\n  --minCount 20 --lr 0.05 --batch 8192 --dtype fp32 --device cuda --seed 0",
    "dump_date":"not pinned: the run used `wikipedia_date: latest`"
  }
}