metadata
dataset_info:
features:
- name: image
dtype: image
- name: filename
dtype: string
- name: image_id
dtype: string
- name: split
dtype: string
- name: has_labels
dtype: bool
- name: label
dtype: image
splits:
- name: train
num_bytes: 3407001610.75
num_examples: 4333
- name: validation
num_bytes: 164760028
num_examples: 210
- name: test
num_bytes: 384182015
num_examples: 490
download_size: 3955445300
dataset_size: 3955943653.75
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*