metadata
dataset_info:
features:
- name: caption
dtype: string
- name: label
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 823769261.79
num_examples: 10805
download_size: 866531077
dataset_size: 823769261.79
configs:
- config_name: default
data_files:
- split: train
path: data/train-*