| license: cc-by-sa-4.0 | |
| dataset_info: | |
| features: | |
| - name: dialogue_id | |
| dtype: int32 | |
| - name: conversations | |
| list: | |
| - name: from | |
| dtype: string | |
| - name: value | |
| dtype: string | |
| splits: | |
| - name: train | |
| num_bytes: 5561473 | |
| num_examples: 3646 | |
| - name: validation | |
| num_bytes: 463729 | |
| num_examples: 300 | |
| - name: test | |
| num_bytes: 471851 | |
| num_examples: 300 | |
| download_size: 2171621 | |
| dataset_size: 6497053 | |
| configs: | |
| - config_name: default | |
| data_files: | |
| - split: train | |
| path: data/train-* | |
| - split: validation | |
| path: data/validation-* | |
| - split: test | |
| path: data/test-* | |
| 以下のデータセットから、dialogue_idとturnsのspeakerとutteranceを抽出し、形式を変更して作成しました。 | |
| https://huggingface.co/datasets/nu-dialogue/jmultiwoz |