metadata
license: apache-2.0
library_name: lerobot
tags:
- lerobot
- diffusion-policy
- imitation-learning
- robotics
- ur3e
- bimanual
datasets:
- WetLabRoboData/lerobot-data-cap
pipeline_tag: robotics
base_model: WetLabRoboData/diffusion-multitask_12task_mix-multitask
diffusion-cap-finetuned
LeRobot diffusion policy for the cap task.
| Field | Value |
|---|---|
| Policy family | diffusion (LeRobot) |
| Variant | Finetuned (multitask pretrain, then finetuned on this task's finetune pool) |
| Training dataset | WetLabRoboData/lerobot-data-cap |
| Base model | WetLabRoboData/diffusion-multitask_12task_mix-multitask |
| Target task | cap |
| Eval episodes | 20 |
| Eval successes | 6 / 20 |
| Robot | UR3e bimanual (3 cameras) |
Evaluation
Rollout videos + per-episode outcomes: WetLabRoboData/eval-diffusion-cap-finetuned.
Load and deploy
from lerobot.policies.diffusion.modeling_diffusion import DiffusionPolicy
policy = DiffusionPolicy.from_pretrained("WetLabRoboData/diffusion-cap-finetuned")
Provenance
Reorganized from WetLabRoboData/lerobot-data-rama-lbm_finetune_cap on 2026-10-04. The archived
training outputs (checkpoints, train_config.json, wandb/) are preserved in
the old/ subfolder of that source repo for traceability.