Skip to content

Dataset · shunk031

Large Japanese image captioning dataset

shunk031/STAIR-Captions

A large-scale STAIR Captions dataset with 820,310 Japanese image captions for text-to-image and image-to-text generation tasks.

What's inside

Contains 820,310 Japanese (ja-JP) text captions paired with images, bridging text and visual modalities. Designed for training and evaluating models that connect images with the Japanese language. The image source and specific file format are not specified in the dataset card.

What you can do with it

  • Training and fine-tuning (LoRA) image generation models to understand Japanese prompts
  • Building image-to-text captioning and multimodal search systems in Japanese
Tasks
Image → text
Languages
Japanese
Size
100K–1M rows
Downloads this month
2.3K
Likes
7

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("shunk031/STAIR-Captions")