Referencias:
Utilice el siguiente comando para cargar este conjunto de datos en TFDS:
ds = tfds.load('huggingface:quora')
- Descripción :
The Quora dataset is composed of question pairs, and the task is to determine if the questions are paraphrases of each other (have the same meaning).
- Licencia : Sin licencia conocida
- Versión : 0.0.0
- Divisiones :
Separar | Ejemplos |
---|---|
'train' | 404290 |
- Características :
{
"questions": {
"feature": {
"id": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"text": {
"dtype": "string",
"id": null,
"_type": "Value"
}
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"is_duplicate": {
"dtype": "bool",
"id": null,
"_type": "Value"
}
}