metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 57183512
num_examples: 1000
- name: validation
num_bytes: 5627284
num_examples: 100
- name: test
num_bytes: 56318653
num_examples: 1000
download_size: 46651813
dataset_size: 119129449
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*