metadata
dataset_info:
features:
- name: image
dtype: image
- name: ground_truth
dtype: string
splits:
- name: train
num_bytes: 110978369.457
num_examples: 1199
- name: validation
num_bytes: 25683676
num_examples: 289
- name: test
num_bytes: 26915178
num_examples: 296
download_size: 162694107
dataset_size: 163577223.45700002
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*