metadata
dataset_info:
features:
- name: images
sequence: image
- name: positive_caption
sequence: string
- name: negative_caption
sequence: string
- name: original_file_name
dtype: string
- name: dataset
dtype: string
- name: key
dtype: string
- name: linguistic_phenomena
dtype: string
- name: original_split
dtype: string
splits:
- name: test
num_bytes: 879276294.706
num_examples: 7702
download_size: 820884492
dataset_size: 879276294.706
configs:
- config_name: default
data_files:
- split: test
path: data/test-*