---
dataset_info:
  features:
  - name: frames
    sequence:
      sequence:
        sequence:
          sequence: float32
  - name: caption
    dtype: string
  splits:
  - name: train
    num_bytes: 9677809244
    num_examples: 1000
  download_size: 2119659863
  dataset_size: 9677809244
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---