metadata
dataset_info:
features:
- name: index
dtype: int64
- name: image
dtype: image
- name: group_id
dtype: int64
- name: entity_name
dtype: string
- name: image_text
dtype: string
splits:
- name: train
num_bytes: 1011312055
num_examples: 25000
download_size: 510458715
dataset_size: 1011312055
configs:
- config_name: default
data_files:
- split: train
path: data/train-*