metadata
dataset_info:
features:
- name: source_dataset
dtype: string
- name: category
dtype: string
- name: input_image
dtype: image
- name: edit_instruction
dtype: string
- name: output_reasoning_text
dtype: string
- name: output_image
dtype: image
- name: score
dtype: int64
- name: reasoning
dtype: string
splits:
- name: quality
num_bytes: 160275260.66666666
num_examples: 120
- name: instruction_follow
num_bytes: 160348718.33333334
num_examples: 120
- name: reasoning_following
num_bytes: 160353843.33333334
num_examples: 120
- name: consistency_score
num_bytes: 80417060
num_examples: 65
- name: consistency
num_bytes: 156943453
num_examples: 120
download_size: 864809193
dataset_size: 718338335.3333334
configs:
- config_name: default
data_files:
- split: quality
path: data/quality-*
- split: consistency_score
path: data/consistency_score-*
- split: instruction_follow
path: data/instruction_follow-*
- split: reasoning_following
path: data/reasoning_following-*
- split: consistency
path: data/consistency-*