metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
- name: embedding
list: float32
splits:
- name: train
num_bytes: 283889251
num_examples: 5994
- name: test
num_bytes: 279203822
num_examples: 5794
download_size: 562920993
dataset_size: 563093073
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*