metadata
dataset_info:
features:
- name: id
dtype: int64
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 910247553.3707837
num_examples: 39829
download_size: 867395645
dataset_size: 910247553.3707837
configs:
- config_name: default
data_files:
- split: train
path: data/train-*