---
dataset_info:
  features:
  - name: episode_id
    dtype: int32
  - name: goal
    dtype: string
  - name: num_steps
    dtype: int32
  - name: steps
    dtype: string
  - name: screenshots
    list: image
  splits:
  - name: train
    num_bytes: 1429185740
    num_examples: 1573
  download_size: 1299604326
  dataset_size: 1429185740
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---