nsmc

Người giới thiệu:

Sử dụng lệnh sau để tải tập dữ liệu này trong TFDS:

ds = tfds.load('huggingface:nsmc')
  • Mô tả :
This is a movie review dataset in the Korean language. Reviews were scraped from Naver movies. The dataset construction is based on the method noted in Large movie review dataset from Maas et al., 2011.
  • Giấy phép : CC0 1.0 Universal (CC0 1.0)
  • Phiên bản : 1.1.0
  • Tách :
Tách ra Các ví dụ
'test' 50000
'train' 150000
  • Các tính năng :
{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "document": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 2,
        "names": [
            "negative",
            "positive"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}