---
dataset_info:
  features:
  - name: images
    dtype: image
  - name: problem
    dtype: string
  - name: answer
    dtype: string
  splits:
  - name: train
    num_bytes: 2922916590.0
    num_examples: 15000
  download_size: 401663461
  dataset_size: 2922916590.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---