multi_eurlex

参考文献:

jp

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:multi_eurlex/en')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 55000
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

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:multi_eurlex/da')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 55000
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

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:multi_eurlex/de')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 55000
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

nl

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:multi_eurlex/nl')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 55000
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

SV

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:multi_eurlex/sv')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 42490
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

バックグラウンド

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:multi_eurlex/bg')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 15986
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

cs

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:multi_eurlex/cs')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 23187
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

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:multi_eurlex/hr')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 7944
'validation' 2500
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

お願いします

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:multi_eurlex/pl')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 23197
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

スク

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:multi_eurlex/sk')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 22971
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

sl

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:multi_eurlex/sl')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 23184
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

エス

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:multi_eurlex/es')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 52785
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

フランス

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:multi_eurlex/fr')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 55000
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

それ

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:multi_eurlex/it')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 55000
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

ポイント

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:multi_eurlex/pt')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 52370
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

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:multi_eurlex/ro')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 15921
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

など

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:multi_eurlex/et')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 23126
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

フィ

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:multi_eurlex/fi')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 42497
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

ふー

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:multi_eurlex/hu')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 22664
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

それ

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:multi_eurlex/lt')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 23188
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

レベル

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:multi_eurlex/lv')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 23208
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

エル

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:multi_eurlex/el')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 55000
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

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:multi_eurlex/mt')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 17521
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

すべての言語

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:multi_eurlex/all_languages')
  • 説明
MultiEURLEX comprises 65k EU laws in 23 official EU languages (some low-ish resource).
Each EU law has been annotated with EUROVOC concepts (labels) by the Publication Office of EU.
As with the English EURLEX, the goal is to predict the relevant EUROVOC concepts (labels);
this is multi-label classification task (given the text, predict multiple labels).
  • ライセンス: 不明なライセンス
  • バージョン: 1.0.0
  • 分割:
スプリット
'test' 5000
'train' 55000
'validation' 5000
  • 特徴
{
    "celex_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "text": {
        "languages": [
            "en",
            "da",
            "de",
            "nl",
            "sv",
            "bg",
            "cs",
            "hr",
            "pl",
            "sk",
            "sl",
            "es",
            "fr",
            "it",
            "pt",
            "ro",
            "et",
            "fi",
            "hu",
            "lt",
            "lv",
            "el",
            "mt"
        ],
        "id": null,
        "_type": "Translation"
    },
    "labels": {
        "feature": {
            "num_classes": 21,
            "names": [
                "100149",
                "100160",
                "100148",
                "100147",
                "100152",
                "100143",
                "100156",
                "100158",
                "100154",
                "100153",
                "100142",
                "100145",
                "100150",
                "100162",
                "100159",
                "100144",
                "100151",
                "100157",
                "100161",
                "100146",
                "100155"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}