metadata
dataset_info:
features:
- name: image
dtype: image
- name: image_path
dtype: string
- name: query
dtype: string
splits:
- name: train
num_bytes: 173387499010
num_examples: 200000
download_size: 173271900653
dataset_size: 173387499010
configs:
- config_name: default
data_files:
- split: train
path: data/train-*