---
dataset_info:
  features:
  - name: question_id
    dtype: int64
  - name: problem
    dtype: string
  - name: image
    dtype: image
  - name: mask
    dtype: image
  - name: category_type
    dtype: string
  splits:
  - name: train
    num_bytes: 396791370.0
    num_examples: 2000
  download_size: 396036455
  dataset_size: 396791370.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---