metadata
dataset_info:
features:
- name: id
dtype: int64
- name: image
dtype: string
- name: ground_truth
dtype: string
- name: category
dtype: float64
splits:
- name: train
num_bytes: 594752
num_examples: 91
download_size: 221180
dataset_size: 594752
configs:
- config_name: default
data_files:
- split: train
path: data/train-*