metadata
dataset_info:
features:
- name: object_id
dtype: int64
- name: caption
dtype: string
- name: image
dtype: image
- name: object_name
dtype: string
- name: artist
dtype: string
- name: title
dtype: string
- name: date
dtype: string
- name: medium
dtype: string
splits:
- name: train
num_bytes: 1431658975
num_examples: 6608
download_size: 1400472252
dataset_size: 1431658975
configs:
- config_name: default
data_files:
- split: train
path: data/train-*