---
title: OceanInstruct-v0.1
canonical_url: "https://www.modelscope.cn/datasets/ZJUNLP/OceanInstruct-v0.1"
md_url: "https://www.modelscope.cn/datasets/ZJUNLP/OceanInstruct-v0.1.md"
repository: ZJUNLP/OceanInstruct-v0.1
last_updated: 2024-07-05
license: mit
domain:
  - size_categories
  - language
  - pretty_name
tasks:
  - "1K<n<10K"
  - en
  - OceanBench
downloads: 102
stars: 3
---

# OceanInstruct-v0.1

> OceanInstruct-v0.1 - ZJUNLP 在 ModelScope 开源的数据集。We release OceanInstruct, which is part of the instruction data for training OceanGPT.

ZJUNLP/OceanInstruct-v0.1 是 ModelScope 魔搭社区上的1K<n<10K、en、OceanBench数据集，涉及 size_categories、language、pretty_name 领域，采用 mit 许可。

- **Repository**: ZJUNLP/OceanInstruct-v0.1
- **License**: mit
- **Tasks**: 1K<n<10K, en, OceanBench
- **Domain**: size_categories, language, pretty_name
- **Downloads**: 102
- **Stars**: 3
- **Last updated**: 2024-07-05

Source: https://www.modelscope.cn/datasets/ZJUNLP/OceanInstruct-v0.1

---

We release OceanInstruct, which is part of the instruction data for training [OceanGPT](https://github.com/zjunlp/OceanGPT).

## 🛠️ How to use OceanInstruct
We provide the example and you can modify the input according to your needs.

```python
from datasets import load_dataset
dataset = load_dataset("zjunlp/OceanInstruct")
```

### 🚩Citation

Please cite the following paper if you use OceanInstruct in your work.

```bibtex
@article{bi2023oceangpt,
  title={OceanGPT: A Large Language Model for Ocean Science Tasks},
  author={Bi, Zhen and Zhang, Ningyu and Xue, Yida and Ou, Yixin and Ji, Daxiong and Zheng, Guozhou and Chen, Huajun},
  journal={arXiv preprint arXiv:2310.02031},
  year={2023}
}

```