metadata
dataset_info:
features:
- name: images
sequence: image
- name: actions
sequence: string
splits:
- name: train
num_bytes: 14340602
num_examples: 23
download_size: 4232
dataset_size: 14340602
configs:
- config_name: default
data_files:
- split: train
path: data/train-*