---
dataset_info:
  features:
  - name: images
    list: image
  - name: problem
    dtype: string
  - name: answer
    dtype: string
  - name: meta
    struct:
    - name: image_path
      dtype: string
  splits:
  - name: train
    num_bytes: 1484647124
    num_examples: 24888
  - name: validation
    num_bytes: 165000034
    num_examples: 2766
  download_size: 1655870571
  dataset_size: 1649647158
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---