metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 94279480.25
num_examples: 3550
download_size: 94231656
dataset_size: 94279480.25
configs:
- config_name: default
data_files:
- split: train
path: data/train-*