OscatttMIT
APPL Simulation Datasets and Pretrained Policies
Release assets for Agent Priors-guided Policy Learning (APPL), containing simulation datasets and trained policies for imitation learning research across MetaWorld and ManiSkill tasks.
Downloads18
Episodes560
Likes1
Why This Matters for Physical AI
Provides reproducible simulation datasets and pretrained policies for imitation learning and diffusion-based policy learning, enabling evaluation of agent prior-guided approaches for robotic manipulation tasks.
Technical Profile
- Modalities
- rgb
- Environment
- simulation
- Task Types
- manipulation
- Episodes
- 560
- Annotation Types
- segmentation
- License
- MIT
Access
Need custom rgb data?
Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.
Request a Sample Pack