metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 341408978.0229583
num_examples: 8000
- name: test
num_bytes: 42705606.500202514
num_examples: 1000
download_size: 339516203
dataset_size: 384114584.5231608
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*