---
title: llava-med-zh-instruct-60k
canonical_url: "https://www.modelscope.cn/datasets/swift/llava-med-zh-instruct-60k"
md_url: "https://www.modelscope.cn/datasets/swift/llava-med-zh-instruct-60k.md"
repository: swift/llava-med-zh-instruct-60k
last_updated: 2024-08-23
license: apache-2.0
storage_size: "6.1 GB"
domain:
  - configs
  - size_categories
  - dataset_info
  - task_categories
  - language
tasks:
  - "{config_name=default, data_files=[{split=train, path=data/train-*}]}"
  - "10K<n<100K"
  - dataset_size
  - image-to-text
  - zh
downloads: 5012
stars: 4
---

# llava-med-zh-instruct-60k

> llava-med-zh-instruct-60k - swift 在 ModelScope 开源的数据集。This Chinese dataset was translated from llava-med using Qwen1.5-14B-Chat and contains 60k medical visual instruction data points.

swift/llava-med-zh-instruct-60k 是 ModelScope 魔搭社区上的{config_name=default, data_files=[{split=train, path=data/train-*}]}、10K<n<100K、dataset_size数据集，涉及 configs、size_categories、dataset_info 领域，存储大小 6.1 GB，采用 apache-2.0 许可。

- **Repository**: swift/llava-med-zh-instruct-60k
- **License**: apache-2.0
- **Tasks**: {config_name=default, data_files=[{split=train, path=data/train-*}]}, 10K<n<100K, dataset_size, image-to-text, zh
- **Domain**: configs, size_categories, dataset_info, task_categories, language
- **Storage size**: 6.1 GB
- **Downloads**: 5012
- **Stars**: 4
- **Last updated**: 2024-08-23

Source: https://www.modelscope.cn/datasets/swift/llava-med-zh-instruct-60k

---

This Chinese dataset was translated from [llava-med](https://github.com/microsoft/LLaVA-Med) using [Qwen1.5-14B-Chat](https://huggingface.co/Qwen/Qwen1.5-14B-Chat) and contains 60k medical visual instruction data points.

You can organize content in the dataset_info.json in [LLaMA Factory](https://github.com/hiyouga/LLaMA-Factory) like this:
```
"llava_med_zh_60k": {
  "hf_hub_url": "BUAADreamer/llava-med-zh-instruct-60k",
  "formatting": "sharegpt",
  "columns": {
    "messages": "messages",
    "images": "images"
  },
  "tags": {
    "role_tag": "role",
    "content_tag": "content",
    "user_tag": "user",
    "assistant_tag": "assistant"
  }
}
```

Then you can use it in [LLaMA Factory](https://github.com/hiyouga/LLaMA-Factory) by specifying --dataset llava_med_zh_60k.
