metadata
dataset_info:
features:
- name: model_name
dtype: string
- name: id_prompt
dtype: string
- name: frame_prompt
dtype: string
- name: Image
dtype: image
- name: sub_images
list:
- name: bytes
dtype: binary
- name: path
dtype: 'null'
- name: score_list
sequence: float64
- name: same_list
sequence: string
- name: max_score
dtype: float64
- name: all_same
dtype: bool
splits:
- name: train
num_bytes: 1522597282
num_examples: 98
download_size: 1522665680
dataset_size: 1522597282
configs:
- config_name: default
data_files:
- split: train
path: data/train-*