GiftEvalPretrain

# GIFT-Eval Pre-training Datasets

Pretraining dataset aligned with [GIFT-Eval](https://huggingface.co/datasets/Salesforce/GiftEval) that has 71 univariate and 17 multivariate datasets, spanning seven domains and 13 frequencies, totaling 4.5 million time series and 230 billion data points. Notably this collection of data has no leakage issue with the train/test split and can be used to pretrain foundation models that can be fairly evaluated on GIFT-Eval.

[📄 Paper](https://arxiv.org/abs/2410.10393)

[🖥️ Code](https://github.com/SalesforceAIResearch/gift-eval)

[📔 Blog Post]()

[🏎️ Leader Board](https://huggingface.co/spaces/Salesforce/GIFT-Eval)

## Ethical Considerations

This release is for research purposes only in support of an academic paper. Our models, datasets, and code are not specifically designed or evaluated for all downstream purposes. We strongly recommend users evaluate and address potential concerns related to accuracy, safety, and fairness before deploying this model. We encourage users to consider the common limitations of AI, comply with applicable laws, and leverage best practices when selecting use cases, particularly for high-risk scenarios where errors or misuse could significantly impact people’s lives, rights, or safety. For further guidance on use cases, refer to our AUP and AI AUP.

## Citation

<!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. -->

If you find this benchmark useful, please consider citing:
```
@article{aksu2024giftevalbenchmarkgeneraltime,
title={GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation},
author={Taha Aksu and Gerald Woo and Juncheng Liu and Xu Liu and Chenghao Liu and Silvio Savarese and Caiming Xiong and Doyen Sahoo},
journal = {arxiv preprint arxiv:2410.10393},
year={2024},
```

# GIFT-Eval 预训练数据集

本预训练数据集与[GIFT-Eval](https://huggingface.co/datasets/Salesforce/GiftEval)对齐，包含71个单变量（univariate）数据集与17个多变量（multivariate）数据集，覆盖7个领域与13种采样频率，总计包含450万条时间序列与2300亿个数据点。值得注意的是，该数据集集合的训练/测试划分不存在数据泄露问题，可用于预训练可在GIFT-Eval上进行公平评估的基础模型（foundation model）。

[📄 论文](https://arxiv.org/abs/2410.10393)
[🖥️ 代码](https://github.com/SalesforceAIResearch/gift-eval)
[📔 博客文章]()
[🏎️ 排行榜](https://huggingface.co/spaces/Salesforce/GIFT-Eval)

## 伦理考量

本数据集仅用于支持学术论文的研究用途。我们的模型、数据集与代码并非针对所有下游任务专门设计或评估。我们强烈建议用户在部署该模型前，对其准确性、安全性与公平性相关的潜在问题进行评估与处理。我们鼓励用户考量人工智能的普遍局限性，遵守适用法律法规，并在选择应用场景时采用最佳实践，尤其是在错误或滥用可能严重影响民众生命、权利或安全的高风险场景中。如需了解应用场景的进一步指导，请参阅我们的AUP与AI AUP。

## 引用

若您认为本基准测试具有参考价值，请考虑引用以下文献：

@article{aksu2024giftevalbenchmarkgeneraltime,
title={"GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation"},
author={Taha Aksu and Gerald Woo and Juncheng Liu and Xu Liu and Chenghao Liu and Silvio Savarese and Caiming Xiong and Doyen Sahoo},
journal = {arxiv preprint arxiv:2410.10393},
year={2024},