metadata
dataset_info:
features:
- name: sample_id
dtype: int64
- name: input_path
dtype: string
- name: category
dtype: string
- name: edit_instruction
dtype: string
- name: target_path
dtype: string
- name: output_path_bagel_nothink
dtype: string
- name: output_path_hidream_e1
dtype: string
- name: output_path_qwenedit
dtype: string
- name: input_image
dtype: image
- name: target_image
dtype: image
- name: output_image_bagel_nothink
dtype: image
- name: output_image_hidream_e1
dtype: image
- name: output_image_qwenedit
dtype: image
- name: new_column
dtype: string
- name: qwenedit_list
list:
- name: Correction
dtype: bool
- name: answer
dtype: string
- name: ground_truth_answer
dtype: string
- name: question
dtype: string
- name: bagel_nothink_list
list:
- name: Correction
dtype: bool
- name: answer
dtype: string
- name: ground_truth_answer
dtype: string
- name: question
dtype: string
- name: hidream_e1_list
list:
- name: Correction
dtype: bool
- name: answer
dtype: string
- name: ground_truth_answer
dtype: string
- name: question
dtype: string
- name: qa_list
list:
- name: ground_truth_answer
dtype: string
- name: question
dtype: string
- name: output_image_gpt_cot
dtype: image
- name: gpt_cot_list
list:
- name: Correction
dtype: bool
- name: answer
dtype: string
- name: ground_truth_answer
dtype: string
- name: question
dtype: string
splits:
- name: train
num_bytes: 295381485.25
num_examples: 1001
download_size: 246367980
dataset_size: 295381485.25
configs:
- config_name: default
data_files:
- split: train
path: data/train-*