the_pile_stack_exchange

Bibliografia:

zwykły_tekst

Użyj następującego polecenia, aby załadować ten zestaw danych do TFDS:

ds = tfds.load('huggingface:the_pile_stack_exchange/plain_text')
  • Opis :
This dataset is part of EleutherAI/The Pile dataset and is a dataset for Language Models from processing stackexchange data dump, which is an anonymized dump of all user-contributed content on the Stack Exchange network.
  • Licencja : Brak znanej licencji
  • Wersja : 1.0.0
  • Podziały :
Podział Przykłady
'train' 5096117
  • Cechy :
{
    "domain": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}