References:
Use the following command to load this dataset in TFDS:
ds = tfds.load('huggingface:medical_questions_pairs')
- Description:
This dataset consists of 3048 similar and dissimilar medical question pairs hand-generated and labeled by Curai's doctors.
- License: No known license
- Version: 0.0.0
- Splits:
Split | Examples |
---|---|
'train' |
3048 |
- Features:
{
"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"
}
}