README.md

---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: mask
    dtype: image
  - name: segments_info
    list:
    - name: area
      dtype: int64
    - name: bbox
      sequence: float32
    - name: category_id
      dtype: int64
    - name: iscrowd
      dtype: int64
  splits:
  - name: train
    num_bytes: 36227495.2
    num_examples: 200
  - name: validation
    num_bytes: 9056873.8
    num_examples: 50
  download_size: 44124458
  dataset_size: 45284369.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---