emotone_ar

References:

• Code
• Huggingface

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:emotone_ar')

• Description:

Dataset of 10065 tweets in Arabic for Emotion detection in Arabic text

• License: No known license
• Version: 0.0.0
• Splits:

• Features:

{
    "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"
    }
}