the_pile_stack_exchange

Referencias:

Texto sin formato

Utilice el siguiente comando para cargar este conjunto de datos en TFDS:

ds = tfds.load('huggingface:the_pile_stack_exchange/plain_text')
  • Descripción :
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.
  • Licencia : Sin licencia conocida
  • Versión : 1.0.0
  • Divisiones :
Separar Ejemplos
'train' 5096117
  • Características :
{
    "domain": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}