metadata
dataset_info:
features:
- name: caption
sequence: string
- name: positive_caption
sequence: string
- name: negative_caption
sequence: string
- name: positive_imgs
sequence: image
- name: negative_imgs
sequence: image
- name: scene_scoring
dtype: string
- name: real_scoring
dtype: string
- name: category
dtype: string
- name: law
dtype: string
splits:
- name: train
num_bytes: 1424885413
num_examples: 102
- name: test
num_bytes: 13789
num_examples: 12
download_size: 1424857492
dataset_size: 1424899202
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*