metadata
dataset_info:
features:
- name: problem_id
dtype: string
- name: query
dtype: string
- name: problem_type
dtype: string
- name: data_type
dtype: string
- name: options
list: string
- name: answer
dtype: string
- name: audio_path
dtype: string
- name: video_path
dtype: string
- name: image_path
dtype: string
- name: data_source
dtype: string
- name: cot
dtype: string
splits:
- name: train
num_bytes: 1357034
num_examples: 7853
download_size: 193349
dataset_size: 1357034
configs:
- config_name: default
data_files:
- split: train
path: data/train-*