metadata
dataset_info:
features:
- name: candidate_id
dtype: string
- name: caption
dtype: string
- name: group
sequence: string
- name: split
dtype: string
- name: pairid
dtype: int64
- name: candidate
dtype: image
splits:
- name: test
num_bytes: 421762236.432
num_examples: 4148
download_size: 280725789
dataset_size: 421762236.432
configs:
- config_name: default
data_files:
- split: test
path: data/test-*