metadata
dataset_info:
features:
- name: image
dtype: image
- name: ground_truth
dtype: string
splits:
- name: train
num_bytes: 2840533918.5972295
num_examples: 10322
- name: test
num_bytes: 358183551.4668479
num_examples: 1291
- name: validation
num_bytes: 358702625.69392234
num_examples: 1290
download_size: 3000508409
dataset_size: 3557420095.758
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*