---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: annotation
    dtype: image
  - name: semantic_class_to_id
    struct:
    - name: background
      dtype: int64
    - name: car
      dtype: int64
    - name: person
      dtype: int64
  splits:
  - name: train
    num_bytes: 284160.0
    num_examples: 8
  - name: validation
    num_bytes: 283458.0
    num_examples: 8
  download_size: 579951
  dataset_size: 567618.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---