emotone_ar

参照:

次のコマンドを使用して、このデータセットをTFDSにロードします。

ds = tfds.load('huggingface:emotone_ar')
  • 説明
Dataset of 10065 tweets in Arabic for Emotion detection in Arabic text
  • ライセンス:既知のライセンスはありません
  • バージョン:0.0.0
  • 分割
スプリット
'train' 10065
  • 特徴
{
    "tweet": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "num_classes": 8,
        "names": [
            "none",
            "anger",
            "joy",
            "sadness",
            "love",
            "sympathy",
            "surprise",
            "fear"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    }
}