metadata
dataset_info:
features:
- name: prompt_id
dtype: string
- name: domain
dtype: string
- name: subcategory
dtype: string
- name: text
dtype: string
- name: SC_image
dtype: image
- name: PA_image
dtype: image
- name: split
dtype: string
splits:
- name: train
num_bytes: 13506941245
num_examples: 10000
- name: validation
num_bytes: 2020935807
num_examples: 1500
- name: test
num_bytes: 2031843466
num_examples: 1500
download_size: 17560572165
dataset_size: 17559720518
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*