shahules786/OA-cornell-movies-dialog

---
dataset_info:
features:
- name: conversation
dtype: string
splits:
- name: train
num_bytes: 9476338
num_examples: 20959
download_size: 4859997
dataset_size: 9476338
---
# Dataset Card for Open Assistant Cornell Movies Dialog

## Dataset Summary

The dataset was created using [Cornell Movies Dialog Corpus](https://www.cs.cornell.edu/~cristian/Cornell_Movie-Dialogs_Corpus.html) which contains a large metadata-rich collection of fictional conversations extracted from raw movie scripts.
Dialogs and meta-data from the underlying Corpus were used to design a dataset that can be used to InstructGPT based models to learn movie scripts.

Example :

```
User: Assume RICK and ALICE are characters from a fantasy-horror movie, continue the conversation between them
RICK: I heard you screaming. Was it a bad one?
ALICE: It was bad.
RICK: Doesn't the dream master work for you anymore?
Assistant: Sure
ALICE: I can't find him.
RICK: Hey, since when do you play Thomas Edison? This looks like Sheila's.
ALICE: It is...was. It's a zapper, it might help me stay awake.
RICK: Yeah, or turn you into toast.
```

## Citations
```
@InProceedings{Danescu-Niculescu-Mizil+Lee:11a,
author={Cristian Danescu-Niculescu-Mizil and Lillian Lee},
title={Chameleons in imagined conversations:
A new approach to understanding coordination of linguistic style in dialogs.},
booktitle={Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics, ACL 2011},
year={2011}
}
```

原始信息汇总

数据集概述

数据集名称

Open Assistant Cornell Movies Dialog

数据集来源

该数据集基于Cornell Movies Dialog Corpus创建，该语料库包含从原始电影剧本中提取的大量元数据丰富的虚构对话。

数据集用途

用于训练InstructGPT模型，以学习电影剧本。

数据集特征

• conversation：字符串类型

数据集拆分

• train：包含20959个示例，总大小为9476338字节

数据集大小

• 下载大小：4859997字节
• 数据集大小：9476338字节