---
task_categories:
- text-to-image
- image-to-text
language:
- en
- ko
- ja
- zh
tags:
- art
- text2image
- img2text
- image generation
- nijijoyrney
- nijijoyrneyV5
pretty_name: NijiJoyrneyV5
size_categories:
- 1K<n<10K
license: mit
dataset_info:
  features:
  - name: channel_id
    dtype: string
  - name: id
    dtype: string
  - name: url
    dtype: image
  - name: content
    dtype: string
configs:
- config_name: English
  data_files: NijiJoyrneyV5_English.csv
  default: true
- config_name: Korean
  data_files: NijiJoyrneyV5_Korean.csv
- config_name: Japanese
  data_files: NijiJoyrneyV5_Japenese.csv
- config_name: Chinese
  data_files: NijiJoyrneyV5_Chinese.csv
---

# NijiJoyrney V5

NijiJoyrney V5 is a large-scale dataset scrape from official Discord server NijiJoyrney. All text description and images generated by users who use NijiJoyrney.

## Dataset Structure

```bash
$ tree
├── NijiJoyrneyV5
│   ├── readme.md
│   ├── NijiJoyrneyV5_English.csv
│   ├── NijiJoyrneyV5_Korean.csv
│   ├── NijiJoyrneyV5_Japanese.csv
│   ├── NijiJoyrneyV5_Chinese.csv
```

## How to use

```python
from datasets import load_dataset

Niji_Eng = load_dataset("WiNE-iNEFF/NijiJoyrneyV5", "English")
Niji_Kor = load_dataset("WiNE-iNEFF/NijiJoyrneyV5", "Korean")
Niji_Jap = load_dataset("WiNE-iNEFF/NijiJoyrneyV5", "Japanese")
Niji_Chi = load_dataset("WiNE-iNEFF/NijiJoyrneyV5", "Chinese")
```

## Dataset Details

- **Curated by:** [Artsem Holub (WiNE iNEFF)](https://huggingface.co/WiNE-iNEFF)
- **Language(s) (NLP):** English, Korean, Japanese, Chinese
- **License:** mit

### Dataset Sources

- **Site** [NijiJoyrney](https://nijijourney.com/en/getting-started/)
- **Discord:** [NijiJoyrney](https://discord.com/invite/nijijourney)