Datasets:
metadata
dataset_info:
features:
- name: image
dtype: image
- name: prompt
dtype: string
- name: label
dtype: string
splits:
- name: train
num_bytes: 23257283614.36
num_examples: 153128
download_size: 23241036646
dataset_size: 23257283614.36
configs:
- config_name: default
data_files:
- split: train
path: data/train-*