metadata
dataset_info:
features:
- name: image_path
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 14882925
num_examples: 86
download_size: 14661007
dataset_size: 14882925
configs:
- config_name: default
data_files:
- split: train
path: data/train-*