metadata
dataset_info:
features:
- name: image
dtype: image
- name: ground_truth
dtype: string
splits:
- name: train
num_bytes: 16831476
num_examples: 100
- name: validation
num_bytes: 2401425
num_examples: 11
- name: test
num_bytes: 2915735
num_examples: 12
download_size: 22077520
dataset_size: 22148636
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*