metadata
dataset_info:
features:
- name: feature
dtype: string
- name: class
dtype: string
- name: images
sequence: image
- name: explanation
dtype: string
splits:
- name: train
num_bytes: 2890207
num_examples: 60
download_size: 2672980
dataset_size: 2890207
configs:
- config_name: default
data_files:
- split: train
path: data/train-*