metadata
dataset_info:
features:
- name: image_id
dtype: int64
- name: caption_1
dtype: string
- name: caption_2
dtype: string
- name: caption_3
dtype: string
- name: caption_4
dtype: string
- name: caption_5
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 12630329
num_examples: 250
download_size: 12616966
dataset_size: 12630329
configs:
- config_name: default
data_files:
- split: train
path: data/train-*