metadata
dataset_info:
features:
- name: image
dtype: string
- name: title
dtype: string
- name: origin
dtype: string
- name: date
dtype: string
- name: size
dtype: string
- name: inventory_no
dtype: string
- name: embedding
sequence: float64
splits:
- name: train
num_bytes: 11919351
num_examples: 1895
download_size: 3028879
dataset_size: 11919351
configs:
- config_name: default
data_files:
- split: train
path: data/train-*