Hugging Face's logo

Datasets:
yjang43
/
lp2-pointmaze-medium

License:
Dataset card Files
xet
 Community
Dataset Viewer
The dataset viewer is not available for this subset.
Job manager crashed while running this job (missing heartbeats).

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Datasets for the PointMaze medium-maze navigation task used in LP² (Latent Projection for Latent Planning), collected in gymnasium-robotics PointMaze_Medium-v3 with the goal marker hidden, using our stable-worldmodel fork.

  • d4rl_pointmaze_medium_random.tar.zst — 10,000 random-policy episodes. The world model and the state dynamics are fit on this set.
  • d4rl_pointmaze_medium_expert.tar.zst — 4,752 D4RL expert episodes, re-rendered through the same environment. Evaluation start and goal pairs are drawn from this set.

Extract both into $STABLEWM_HOME/datasets.
Downloads last month
35
Total file size:
1.25 GB

Models trained or fine-tuned on yjang43/lp2-pointmaze-medium