metadata
dataset_info:
features:
- name: questionId
dtype: int64
- name: images
dtype: image
- name: question
dtype: string
- name: answers
sequence: string
- name: query_analysis
dtype: string
- name: plan_create
dtype: string
- name: final_answer
dtype: string
- name: plan_exec
dtype: string
splits:
- name: train
num_bytes: 7326730749
num_examples: 12713
download_size: 1545104645
dataset_size: 7326730749
configs:
- config_name: default
data_files:
- split: train
path: data/train-*