参考文献:
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:cornell_movie_dialog')
- 説明：
This corpus contains a large metadata-rich collection of fictional conversations extracted from raw movie scripts:
- 220,579 conversational exchanges between 10,292 pairs of movie characters
- involves 9,035 characters from 617 movies
- in total 304,713 utterances
- movie metadata included:
- genres
- release year
- IMDB rating
- number of IMDB votes
- IMDB rating
- character metadata included:
- gender (for 3,774 characters)
- position on movie credits (3,321 characters)
- ライセンス: 既知のライセンスはありません
- バージョン: 0.1.0
- 分割:
| スプリット | 例 |
|---|---|
'train' | 83097 |
- 特徴：
{
"movieID": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"movieTitle": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"movieYear": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"movieIMDBRating": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"movieNoIMDBVotes": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"movieGenres": {
"feature": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"length": -1,
"id": null,
"_type": "Sequence"
},
"characterID1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"characterID2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"characterName1": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"characterName2": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"utterance": {
"feature": {
"text": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"LineID": {
"dtype": "string",
"id": null,
"_type": "Value"
}
},
"length": -1,
"id": null,
"_type": "Sequence"
}
}