constants.py

LEADERBOARD_HEADER = """
<style>
.header-gradient {
    top: 40%;
    bottom: 40%;
    padding: 10px 0px;
    font-weight: bold;
    font-size: 40px;
    font-family: Inter, Arial, Helvetica, sans-serif;
    background: linear-gradient(to right, #FF705B, #FFB457);
    -webkit-text-fill-color: transparent;
    -webkit-background-clip: text;
}

.header-normal {
    top: 40%;
    bottom: 40%;
    padding: 10px 0px;
    font-weight: bold;
    font-size: 40px;
    font-family: Inter, Arial, Helvetica, sans-serif;
}
</style>

<div align="center">
    <span class="header-gradient"> DD-Ranking </span>
    <span class="header-normal"> Leaderboard </span>
</div>
<p align="center">
| <a href="https://nus-hpc-ai-lab.github.io/DD-Ranking/"><b>Documentation</b></a> | <a href="https://github.com/NUS-HPC-AI-Lab/DD-Ranking"><b>Github</b></a> | <a href="https://arxiv.org/abs/2505.13300"><b>Paper </b> </a> | <a href="https://x.com/Richard91316073/status/1890296645486801230"><b>Twitter/X</b> </a> |
</p>"""

LEADERBOARD_INTRODUCTION = """
# DD-Ranking Leaderboard

🏆 Welcome to the leaderboard of the **DD-Ranking**! 

> DD-Ranking (DD, i.e., Dataset Distillation) is an integrated and easy-to-use benchmark for dataset distillation. It aims to provide a fair evaluation scheme for DD methods that can decouple the impacts from knowledge distillation and data augmentation to reflect the real informativeness of the distilled data.

- **Fair Evaluation**: DD-Ranking provides a fair evaluation scheme for DD methods that can decouple the impacts from knowledge distillation and data augmentation to reflect the real informativeness of the distilled data.
- **Easy-to-use**: DD-Ranking provides a unified interface for dataset distillation evaluation.
- **Extensible**: DD-Ranking supports various datasets and models.
- **Customizable**: DD-Ranking supports various data augmentations and soft label strategies.

**Join Leaderboard**: Please see the [instructions](https://github.com/NUS-HPC-AI-Lab/DD-Ranking/blob/main/CONTRIBUTING.md) to participate.
"""

CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
CITATION_BUTTON_TEXT = r"""
@misc{li2025ddrankingrethinkingevaluationdataset,
      title={DD-Ranking: Rethinking the Evaluation of Dataset Distillation}, 
      author={Zekai Li and Xinhao Zhong and Samir Khaki and Zhiyuan Liang and Yuhao Zhou and Mingjia Shi and Ziqiao Wang and Xuanlei Zhao and Wangbo Zhao and Ziheng Qin and Mengxuan Wu and Pengfei Zhou and Haonan Wang and David Junhao Zhang and Jia-Wei Liu and Shaobo Wang and Dai Liu and Linfeng Zhang and Guang Li and Kun Wang and Zheng Zhu and Zhiheng Ma and Joey Tianyi Zhou and Jiancheng Lv and Yaochu Jin and Peihao Wang and Kaipeng Zhang and Lingjuan Lyu and Yiran Huang and Zeynep Akata and Zhiwei Deng and Xindi Wu and George Cazenavette and Yuzhang Shang and Justin Cui and Jindong Gu and Qian Zheng and Hao Ye and Shuo Wang and Xiaobo Wang and Yan Yan and Angela Yao and Mike Zheng Shou and Tianlong Chen and Hakan Bilen and Baharan Mirzasoleiman and Manolis Kellis and Konstantinos N. Plataniotis and Zhangyang Wang and Bo Zhao and Yang You and Kai Wang},
      year={2025},
      eprint={2505.13300},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2505.13300}, 
}
"""

IPC_INFO = """
Images Per Class
"""

LABEL_TYPE_INFO = """
Hard labels are categorical, having the same format of the real dataset. Soft labels are generated by a teacher model pretrained on the target dataset
"""

WEIGHT_ADJUSTMENT_INTRODUCTION = """
The score for ranking (DD-Ranking Score, DDRS) in the following table is computed by $DDRS = \\frac{e^{w IOR - (1 - w) HLR} - e^{-1}}{e - e^{-1}}$, where $w$ is the weight for the HLR metric.
**You can specify the weight $w$ below.**
"""

METRIC_DEFINITION_INTRODUCTION = """
$\\text{Acc.}$: The accuracy of models trained on different samples.

$\\text{full-hard}$: Full dataset with hard labels.

$\\text{syn-hard}$: Synthetic dataset with hard labels.

$\\text{syn-any}$: Synthetic dataset with personalized evaluation methods (hard or soft labels).

$\\text{rdm-any}$: Randomly selected dataset (under the same compression ratio) with the same personalized evaluation methods.

$\\text{HLR} = \\text{Acc.} \\text{full-hard} - \\text{Acc.} \\text{syn-hard}$: The degree to which the original dataset is recovered under hard labels (hard label recovery).

$\\text{IOR} = \\text{Acc.} \\text{syn-any} - \\text{Acc.} \\text{rdm-any}$: The improvement over random selection when using personalized evaluation methods (improvement over random).
"""

DATASET_LIST = ["CIFAR-10", "CIFAR-100", "Tiny-ImageNet", "ImageNet1K"]
IPC_LIST = ["IPC-1", "IPC-10", "IPC-50"]
DATASET_IPC_LIST = {
    "CIFAR-10": ["IPC-1", "IPC-10", "IPC-50"],
    "CIFAR-100": ["IPC-1", "IPC-10", "IPC-50"],
    "Tiny-ImageNet": ["IPC-1", "IPC-10", "IPC-50"],
    "ImageNet1K": ["IPC-1", "IPC-10", "IPC-50"],
}
LABEL_TYPE_LIST = ["Hard Label", "Soft Label"]

METRICS = ["HLR", "IOR"]
METRICS_SIGN = [1.0, -1.0]
COLUMN_NAMES = ["Ranking", "Method", "Verified", "Date", "Label Type", "HLR%", "IOR%", "LRS"]
DATA_TITLE_TYPE = ['number', 'markdown', 'markdown', 'markdown', 'markdown', 'str', 'str', 'str']

DATASET_MAPPING = {
    "CIFAR-10": 0,
    "CIFAR-100": 1,
    "Tiny-ImageNet": 2,
    "ImageNet1K": 3,
}

IPC_MAPPING = {
    "IPC-1": 0,
    "IPC-10": 1,
    "IPC-50": 2,
}

LABEL_MAPPING = {
    "Hard Label": 0,
    "Soft Label": 1,
}