pusht_image

This dataset was created using [LeRobot](https://github.com/huggingface/lerobot).

## Dataset Description

- **Homepage:** https://diffusion-policy.cs.columbia.edu/
- **Paper:** https://arxiv.org/abs/2303.04137v5
- **License:** mit

## Dataset Structure

[meta/info.json](meta/info.json):
```json
{
"codebase_version": "v2.0",
"robot_type": "unknown",
"total_episodes": 206,
"total_frames": 25650,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 10,
"splits": {
"train": "0:206"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path": null,
"features": {
"observation.image": {
"dtype": "image",
"shape": [
96,
96,
3
],
"names": [
"height",
"width",
"channel"
]
},
"observation.state": {
"dtype": "float32",
"shape": [
2
],
"names": {
"motors": [
"motor_0",
"motor_1"
]
}
},
"action": {
"dtype": "float32",
"shape": [
2
],
"names": {
"motors": [
"motor_0",
"motor_1"
]
}
},
"episode_index": {
"dtype": "int64",
"shape": [
1
],
"names": null
},
"frame_index": {
"dtype": "int64",
"shape": [
1
],
"names": null
},
"timestamp": {
"dtype": "float32",
"shape": [
1
],
"names": null
},
"next.reward": {
"dtype": "float32",
"shape": [
1
],
"names": null
},
"next.done": {
"dtype": "bool",
"shape": [
1
],
"names": null
},
"next.success": {
"dtype": "bool",
"shape": [
1
],
"names": null
},
"index": {
"dtype": "int64",
"shape": [
1
],
"names": null
},
"task_index": {
"dtype": "int64",
"shape": [
1
],
"names": null
}
}
}
```

## Citation

**BibTeX:**

```bibtex
@article{chi2024diffusionpolicy,
author = {Cheng Chi and Zhenjia Xu and Siyuan Feng and Eric Cousineau and Yilun Du and Benjamin Burchfiel and Russ Tedrake and Shuran Song},
title ={Diffusion Policy: Visuomotor Policy Learning via Action Diffusion},
journal = {The International Journal of Robotics Research},
year = {2024},
}
```

本数据集基于[LeRobot](https://github.com/huggingface/lerobot)构建。

## 数据集说明

- **项目主页**：https://diffusion-policy.cs.columbia.edu/
- **论文链接**：https://arxiv.org/abs/2303.04137v5
- **许可证**：MIT

## 数据集结构

`[meta/info.json](meta/info.json)`:
json
{
"代码库版本": "v2.0",
"机器人类型": "未知",
"总回合数": 206,
"总帧数": 25650,
"总任务数": 1,
"总视频数": 0,
"总分块数": 1,
"分块大小": 1000,
"帧率": 10,
"数据集划分": {
"训练集": "0:206"
},
"数据路径": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"视频路径": null,
"特征项": {
"观测.图像": {
"数据类型": "图像",
"形状": [
96,
96,
3
],
"维度说明": [
"高度",
"宽度",
"通道"
]
},
"观测.状态": {
"数据类型": "float32",
"形状": [
2
],
"维度说明": {
"电机": [
"motor_0",
"motor_1"
]
}
},
"动作": {
"数据类型": "float32",
"形状": [
2
],
"维度说明": {
"电机": [
"motor_0",
"motor_1"
]
}
},
"回合索引": {
"数据类型": "int64",
"形状": [
1
],
"维度说明": null
},
"帧索引": {
"数据类型": "int64",
"形状": [
1
],
"维度说明": null
},
"时间戳": {
"数据类型": "float32",
"形状": [
1
],
"维度说明": null
},
"下一时刻奖励": {
"数据类型": "float32",
"形状": [
1
],
"维度说明": null
},
"下一时刻终止标记": {
"数据类型": "bool",
"形状": [
1
],
"维度说明": null
},
"下一时刻成功标记": {
"数据类型": "bool",
"形状": [
1
],
"维度说明": null
},
"全局索引": {
"数据类型": "int64",
"形状": [
1
],
"维度说明": null
},
"任务索引": {
"数据类型": "int64",
"形状": [
1
],
"维度说明": null
}
}
}

## 引用信息

**BibTeX格式：**
bibtex
@article{chi2024diffusionpolicy,
author = {Cheng Chi and Zhenjia Xu and Siyuan Feng and Eric Cousineau and Yilun Du and Benjamin Burchfiel and Russ Tedrake and Shuran Song},
title ={扩散策略：基于动作扩散的视觉运动策略学习},
journal = {The International Journal of Robotics Research},
year = {2024},
}