README.md

metadata

license: cc-by-4.0
dataset_info:
  features:
    - name: id
      dtype: int64
    - name: file_name
      dtype: string
    - name: instruction
      dtype: string
    - name: output
      dtype: string
  splits:
    - name: train
      num_bytes: 30613024
      num_examples: 82783
    - name: validation
      num_bytes: 14889077
      num_examples: 40504
  download_size: 20826123
  dataset_size: 45502101
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*

Dataset Card for Dataset Name

This dataset was created to train a model that generates diverse image generation prompts from simple Japanese descriptions.

Dataset Details

Dataset Description

STAIR Captions + Microsoft COCO Caption + Microsoft COCO Images caption interpreted by UForm

This dataset was created by having UForm explain Microsoft COCO Images of source Microsoft COCO based on the pair of translation inforThis dataset was created to train a model that generates diverse image generation prompts from simple Japanese descriptions.mation STAIR Captions and source information Microsoft COCO Caption.