metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 30035788
num_examples: 1000
- name: validation
num_bytes: 581255
num_examples: 20
- name: test
num_bytes: 3025580
num_examples: 100
download_size: 30283506
dataset_size: 33642623
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*