README.md

---
dataset_info:
  features:
  - name: image_id
    dtype: int64
  - name: image_path
    dtype: string
  - name: coco_url
    dtype: string
  - name: image
    dtype: image
  - name: alt_text
    dtype: string
  - name: objects
    sequence: string
  splits:
  - name: train
    num_bytes: 9314018.0
    num_examples: 16
  - name: test
    num_bytes: 8750613.0
    num_examples: 16
  - name: validation
    num_bytes: 7180609.0
    num_examples: 16
  download_size: 25250957
  dataset_size: 25245240.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---