Attend the Women in ML Symposium on December 7 Register now

dialog_re

参考:

dialog_re

使用以下命令在 TFDS 中加载此数据集:

ds = tfds.load('huggingface:dialog_re/dialog_re')
  • 说明
DialogRE is the first human-annotated dialogue based relation extraction (RE) dataset aiming
to support the prediction of relation(s) between two arguments that appear in a dialogue.
The dataset annotates all occurrences of 36 possible relation types that exist between pairs
of arguments in the 1,788 dialogues originating from the complete transcripts of Friends.
拆分 样本
'test' 357
'train' 1073
'validation' 358
  • 特征
{
    "dialog": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "relation_data": {
        "feature": {
            "x": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            },
            "y": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            },
            "x_type": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            },
            "y_type": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            },
            "r": {
                "feature": {
                    "dtype": "string",
                    "id": null,
                    "_type": "Value"
                },
                "length": -1,
                "id": null,
                "_type": "Sequence"
            },
            "rid": {
                "feature": {
                    "dtype": "int32",
                    "id": null,
                    "_type": "Value"
                },
                "length": -1,
                "id": null,
                "_type": "Sequence"
            },
            "t": {
                "feature": {
                    "dtype": "string",
                    "id": null,
                    "_type": "Value"
                },
                "length": -1,
                "id": null,
                "_type": "Sequence"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}