README.md

---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: filename
    dtype: string
  - name: caption
    sequence: string
  splits:
  - name: train
    num_bytes: 829880166.0
    num_examples: 6000
  - name: test
    num_bytes: 136929806.0
    num_examples: 1000
  - name: validation
    num_bytes: 138076299.0
    num_examples: 1000
  download_size: 1102895899
  dataset_size: 1104886271.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---