metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: int32
splits:
- name: train
num_bytes: 1201370389
num_examples: 3000
- name: validation
num_bytes: 401318299
num_examples: 1000
- name: test
num_bytes: 400893761
num_examples: 1000
download_size: 2003730314
dataset_size: 2003582449
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*