Referencias:
Utilice el siguiente comando para cargar este conjunto de datos en TFDS:
ds = tfds.load('huggingface:medical_questions_pairs')
- Descripción :
This dataset consists of 3048 similar and dissimilar medical question pairs hand-generated and labeled by Curai's doctors.
- Licencia : Sin licencia conocida
- Versión : 0.0.0
- Divisiones :
| Separar | Ejemplos |
|---|---|
'train' | 3048 |
- Características :
{
"dr_id": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"question_1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"question_2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"label": {
"num_classes": 2,
"names": [
0,
1
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
}
}