# Episode 0 of yaak-ai/L2D-v3

- format: LeRobot v3.0
- robot: KIA Niro EV 2023
- robot family, as the index files it: Other robots
- kind, as the index files it: recording; other (inferred from the name)
- frames per second: 10
- frames in this episode: 114
- task: Follow the waypoints while adhering to traffic rules and regulations
- episodes in the dataset: 10,000
- licence: apache-2.0
- dataset page: https://huggingface.co/datasets/yaak-ai/L2D-v3

LeRobot v3 keeps many episodes in one file. Download the data file and keep the rows whose episode_index is 0. Each video file holds several episodes back to back: cut it to episode_seconds. Or load it with LeRobot: LeRobotDataset('yaak-ai/L2D-v3', episodes=[0]). The data file is 37 MB and holds many episodes; data_via_index gives only this one's rows.

Columns of the data file:

- `observation.state.vehicle`: float32, shape 8 (speed, heading, heading_error, hp_loc_latitude, hp_loc_longitude, hp_loc_altitude, acceleration_x, acceleration_y)
- `observation.state.lanes`: string, shape 1
- `observation.state.road`: string, shape 1
- `observation.state.surface`: string, shape 1
- `observation.state.max_speed`: string, shape 1
- `observation.state.precipitation`: string, shape 1
- `observation.state.conditions`: string, shape 1
- `observation.state.lighting`: string, shape 1
- `observation.state.waypoints`: float32, shape 10x2 (way, points)
- `observation.state.timestamp`: int64, shape 1 (unix_epoc_timestamp)
- `task.policy`: string, shape 1
- `task.instructions`: string, shape 1
- `action.continuous`: float32, shape 3 (gas_pedal_normalized, brake_pedal_normalized, steering_angle_normalized)
- `action.discrete`: int32, shape 2 (gear, turn_signal)
- `timestamp`: float32, shape 1

| file | camera | size | this episode | link |
|---|---|---|---|---|
| data |  | 37.1 MB | rows where episode_index == 0 (global frames 0 to 113, 114 frames) | https://huggingface.co/datasets/yaak-ai/L2D-v3/resolve/main/data/chunk-000/file-000.parquet |
| video | observation.images.left_forward | 495.5 MB | seconds 865.1 to 876.5 | https://huggingface.co/datasets/yaak-ai/L2D-v3/resolve/main/videos/observation.images.left_forward/chunk-000/file-025.mp4 |
| video | observation.images.front_left | 510.2 MB | seconds 271.3 to 282.7 | https://huggingface.co/datasets/yaak-ai/L2D-v3/resolve/main/videos/observation.images.front_left/chunk-000/file-024.mp4 |
| video | observation.images.right_forward | 503.7 MB | seconds 521.2 to 532.6 | https://huggingface.co/datasets/yaak-ai/L2D-v3/resolve/main/videos/observation.images.right_forward/chunk-000/file-025.mp4 |
| video | observation.images.left_backward | 509.9 MB | seconds 640.1 to 651.5 | https://huggingface.co/datasets/yaak-ai/L2D-v3/resolve/main/videos/observation.images.left_backward/chunk-000/file-021.mp4 |
| video | observation.images.rear | 523.9 MB | seconds 1240.3 to 1251.7 | https://huggingface.co/datasets/yaak-ai/L2D-v3/resolve/main/videos/observation.images.rear/chunk-000/file-020.mp4 |
| video | observation.images.right_backward | 506.6 MB | seconds 146.1 to 157.5 | https://huggingface.co/datasets/yaak-ai/L2D-v3/resolve/main/videos/observation.images.right_backward/chunk-000/file-024.mp4 |
| video | observation.images.map | 522.5 MB | seconds 28230.7 to 28242.1 | https://huggingface.co/datasets/yaak-ai/L2D-v3/resolve/main/videos/observation.images.map/chunk-000/file-000.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/yaak-ai/L2D-v3?episode=0 (parquet) or https://datasets.gurasees.com/episode-data/yaak-ai/L2D-v3?episode=0&format=json (JSON). Videos are not relayed.

- next episode: https://datasets.gurasees.com/episodes/yaak-ai/L2D-v3.md?episode=1
- last episode: https://datasets.gurasees.com/episodes/yaak-ai/L2D-v3.md?episode=9999