CIFAR-10N

displayName: CIFAR-10N (Real-World Human Annotations)
labelTypes:
- Classification
license:
- CC BY-NC 4.0
mediaTypes:
- Image
paperUrl: https://arxiv.org/pdf/2110.12088v2.pdf
publishDate: "2022"
publishUrl: https://github.com/UCSC-REAL/cifar-10-100n
publisher:
- University of California
- University of Sydney
- RIKEN Center for Advanced Intelligence Project
tags:
- Text
taskTypes:
- Image Classification

---
# 数据集介绍
## 简介
这项工作提出了两个新的基准数据集（CIFAR-10N、CIFAR-100N），为 CIFAR-10 和 CIFAR-100 的训练数据集配备了我们从 Amazon Mechanical Turk 收集的人工注释的真实世界噪声标签。
## 引文
```
@article{wei2021learning,
title={Learning with noisy labels revisited: A study using real-world human annotations},
author={Wei, Jiaheng and Zhu, Zhaowei and Cheng, Hao and Liu, Tongliang and Niu, Gang and Liu, Yang},
journal={arXiv preprint arXiv:2110.12088},
year={2021}
}

```

## Download dataset
:modelscope-code[]{type="git"}

displayName: CIFAR-10N（真实世界人工标注）
labelTypes:
- 分类
license:
- CC BY-NC 4.0
mediaTypes:
- 图像
paperUrl: https://arxiv.org/pdf/2110.12088v2.pdf
publishDate: "2022"
publishUrl: https://github.com/UCSC-REAL/cifar-10-100n
publisher:
- 加利福尼亚大学
- 悉尼大学
- 日本理化学研究所先进智能项目中心
tags:
- 文本
taskTypes:
- 图像分类

---

# 数据集介绍
## 简介
本工作提出了两个全新的基准数据集（CIFAR-10N、CIFAR-100N），为CIFAR-10与CIFAR-100的训练集配备了我们从Amazon Mechanical Turk（亚马逊众包标注平台）采集的真实世界人工标注噪声标签。
## 引文

@article{wei2021learning,
title={重探带噪标签学习：基于真实世界人工标注的研究},
author={Wei, Jiaheng and Zhu, Zhaowei and Cheng, Hao and Liu, Tongliang and Niu, Gang and Liu, Yang},
journal={arXiv preprint arXiv:2110.12088},
year={2021}
}

## Download dataset
:modelscope-code[]{type="git"}