Dataset · IPEC-COMMUNITY
Language Table large robotics dataset in LeRobot
IPEC-COMMUNITY/language_table_lerobot
Over 442k episodes and 7M video frames and xArm robot telemetry for training robotic models.
What's inside
A large-scale robotics dataset converted to the LeRobot v2.0 framework format. Includes 442,226 episodes (equal number of 10 fps videos), 7,045,476 frames, and 127,605 tasks for an xArm type robot. Tabular data and telemetry are chunked and saved in parquet format.
What you can do with it
- Training multimodal robotic manipulator control models using video and text tasks
- Developing and testing imitation learning policies within the LeRobot ecosystem
- Tasks
- Robotics
- Downloads this month
- 175.6K
- Likes
- 38
How to load it
from datasets import load_dataset
ds = load_dataset("IPEC-COMMUNITY/language_table_lerobot")