README.md

---
dataset_info:
  features:
  - name: id
    dtype: float64
  - name: brand
    dtype: string
  - name: image_id
    dtype: string
  - name: caption
    dtype: string
  - name: image_url
    dtype: string
  - name: width
    dtype: string
  - name: height
    dtype: string
  - name: image
    dtype: image
  - name: num_people
    dtype: int64
  splits:
  - name: train
    num_bytes: 191102948192.0
    num_examples: 569862
  download_size: 179177162005
  dataset_size: 191102948192.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---