Guowei-ZouMIT

DMPO Demonstration Datasets

Pre-processed demonstration datasets for DMPO (Dispersive MeanFlow Policy Optimization) policy pre-training. Includes trajectory data and normalization statistics from both D4RL gym tasks and Robomimic manipulation tasks.

Downloads224

Technical Profile

Modalities
proprioceptionrgb
Robot Embodiments
humanoidPandaSawyer
Environment
simulationlab
Task Types
locomotionmanipulationpick_and_placeobject_manipulationkitchen_tasks
Data Format
NPZ
License
MIT
Part of the DMPO Demonstration Datasets family

Access

Need custom proprioception data?

Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.

Request a Sample Pack

Related Datasets