metadata
dataset_info:
features:
- name: images
sequence: image
- name: problem
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 693963764.332
num_examples: 11998
download_size: 645332588
dataset_size: 693963764.332
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
The dataset is conctructed from hiyouga/math12k. The textual data is unrelated to the images. Only used for specific training tasks.