metadata
dataset_info:
features:
- name: images
dtype: image
- name: doc_id
dtype: string
- name: problem
dtype: string
- name: solution
dtype: string
- name: data_source
dtype: string
splits:
- name: train
num_bytes: 16188773826
num_examples: 5729
download_size: 12536788970
dataset_size: 16188773826
configs:
- config_name: default
data_files:
- split: train
path: data/train-*