---
dataset_info:
  features:
  - name: images_front
    sequence: image
  - name: images_wrist
    sequence: image
  - name: images_overhead
    sequence: image
  - name: task
    dtype: string
  - name: failed_reason
    dtype: string
  - name: episode
    dtype: string
  - name: wp
    dtype: string
  splits:
  - name: train
    num_bytes: 3065930007.872
    num_examples: 2188
  download_size: 3065555214
  dataset_size: 3065930007.872
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---