---
dataset_info:
  features:
  - name: source
    dtype: string
  - name: caption_ko
    sequence: string
  splits:
  - name: train
    num_bytes: 45203310
    num_examples: 122256
  download_size: 20493596
  dataset_size: 45203310
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---