metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: image
splits:
- name: train
num_bytes: 2098989454
num_examples: 14000
- name: validation
num_bytes: 303931270
num_examples: 2000
- name: test
num_bytes: 338668576.772
num_examples: 2419
download_size: 2568233278
dataset_size: 2741589300.772
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*