# Episode 0 of cadene/droid_1.0.1

- format: LeRobot v2.1
- robot: Franka
- robot family, as the index files it: Franka
- kind, as the index files it: recording; teleop (inferred from the name)
- frames per second: 15
- episodes in the dataset: 95,600
- licence: apache-2.0
- dataset page: https://huggingface.co/datasets/cadene/droid_1.0.1

LeRobot v2.1 keeps one episode per file, so each file listed is the whole episode. Or load it with LeRobot: LeRobotDataset('cadene/droid_1.0.1', episodes=[0]).

Columns of the data file:

- `is_first`: bool, shape 1
- `is_last`: bool, shape 1
- `is_terminal`: bool, shape 1
- `language_instruction`: string, shape 1
- `language_instruction_2`: string, shape 1
- `language_instruction_3`: string, shape 1
- `observation.state.gripper_position`: float32, shape 1 (gripper)
- `observation.state.cartesian_position`: float32, shape 6 (x, y, z, roll, pitch, yaw)
- `observation.state.joint_position`: float32, shape 7 (x, y, z, roll, pitch, yaw)
- `observation.state`: float32, shape 8 (joint_0, joint_1, joint_2, joint_3, joint_4, joint_5, joint_6, gripper)
- `action.gripper_position`: float32, shape 1 (gripper)
- `action.gripper_velocity`: float32, shape 1 (gripper)
- `action.cartesian_position`: float32, shape 6 (x, y, z, roll, pitch, yaw)
- `action.cartesian_velocity`: float32, shape 6 (x, y, z, roll, pitch, yaw)
- `action.joint_position`: float32, shape 7 (joint_0, joint_1, joint_2, joint_3, joint_4, joint_5, joint_6)
- `action.joint_velocity`: float32, shape 7 (joint_0, joint_1, joint_2, joint_3, joint_4, joint_5, joint_6)
- `action.original`: float32, shape 7 (x, y, z, roll, pitch, yaw, gripper)
- `action`: float32, shape 8 (joint_0, joint_1, joint_2, joint_3, joint_4, joint_5, joint_6, gripper)
- `discount`: float32, shape 1
- `reward`: float32, shape 1
- `task_category`: string, shape 1
- `building`: string, shape 1
- `collector_id`: string, shape 1
- `date`: string, shape 1
- `camera_extrinsics.wrist_left`: float32, shape 6 (x, y, z, roll, pitch, yaw)
- `camera_extrinsics.exterior_1_left`: float32, shape 6 (x, y, z, roll, pitch, yaw)
- `camera_extrinsics.exterior_2_left`: float32, shape 6 (x, y, z, roll, pitch, yaw)
- `is_episode_successful`: bool, shape 1
- `timestamp`: float32, shape 1

| file | camera | size | this episode | link |
|---|---|---|---|---|
| data |  | 79.6 KB | the whole file | https://huggingface.co/datasets/cadene/droid_1.0.1/resolve/main/data/chunk-000/episode_000000.parquet |
| video | observation.images.wrist_left | 594.9 KB | the whole file | https://huggingface.co/datasets/cadene/droid_1.0.1/resolve/main/videos/chunk-000/observation.images.wrist_left/episode_000000.mp4 |
| video | observation.images.exterior_1_left | 925.8 KB | the whole file | https://huggingface.co/datasets/cadene/droid_1.0.1/resolve/main/videos/chunk-000/observation.images.exterior_1_left/episode_000000.mp4 |
| video | observation.images.exterior_2_left | 1.0 MB | the whole file | https://huggingface.co/datasets/cadene/droid_1.0.1/resolve/main/videos/chunk-000/observation.images.exterior_2_left/episode_000000.mp4 |

If your network cannot reach Hugging Face, this site fetches the episode's rows for you, and only that episode's: https://datasets.gurasees.com/episode-data/cadene/droid_1.0.1?episode=0 (parquet) or https://datasets.gurasees.com/episode-data/cadene/droid_1.0.1?episode=0&format=json (JSON). Videos are not relayed.

- next episode: https://datasets.gurasees.com/episodes/cadene/droid_1.0.1.md?episode=1
- last episode: https://datasets.gurasees.com/episodes/cadene/droid_1.0.1.md?episode=95599