metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype:
class_label:
names:
'0': observation.images.cam_high_episode_000000
'1': observation.images.cam_high_episode_000001
'2': observation.images.cam_left_wrist_episode_000000
'3': observation.images.cam_left_wrist_episode_000001
'4': observation.images.cam_low_episode_000000
'5': observation.images.cam_low_episode_000001
'6': observation.images.cam_right_wrist_episode_000000
'7': observation.images.cam_right_wrist_episode_000001
splits:
- name: train
num_bytes: 4188427288
num_examples: 9620
download_size: 4000640708
dataset_size: 4188427288
configs:
- config_name: default
data_files:
- split: train
path: data/train-*