---
dataset_info:
  features:
  - name: episode_id
    dtype: int64
  - name: image
    dtype: image
  - name: step
    dtype: int64
  - name: action
    sequence: int64
  splits:
  - name: train
    num_bytes: 26492183.0
    num_examples: 569
  download_size: 25290481
  dataset_size: 26492183.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---