metadata
dataset_info:
features:
- name: id
dtype: string
- name: text
dtype: string
- name: img
dtype: image
splits:
- name: train
num_bytes: 2563044
num_examples: 38
download_size: 2564789
dataset_size: 2563044
configs:
- config_name: default
data_files:
- split: train
path: data/train-*