YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

dp_isaaclab_ur7e_3task_fixed

Diffusion Policy (lerobot) checkpoints for the UR7e + RH5DG2 Isaac Lab tasks, trained on the FIXED datasets (DexSteer/isaaclab_ur7e_3task_fixed: joint-space absolute arm-target actions; pour_cup with the repaired wrist video). One folder per task = a lerobot pretrained_model dir (config.json, model.safetensors, policy_pre/postprocessor, train_config.json).

folder trained on steps notes
grasp_pan/ isaaclab_ur7e_3task_fixed/grasp_pan (199 ep) 60000
cup_hang/ isaaclab_ur7e_3task_fixed/cup_hang (200 ep) 60000
pour_cup/ isaaclab_ur7e_3task_fixed/pour_cup (150 ep, wrist-fixed) 60000 retrained after the wrist-video fix

Training: lerobot diffusion policy, 2-GPU DDP, effective batch 32, seed 1000, resize 224. Eval with isaac-tasks/scripts/pipeline/eval/run_dp_ur7e.sh <task> (arm action = absolute JointPositionAction; see the dataset README).

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support