Phase B.2 create-new: README card

README.md

@@ -0,0 +1,46 @@
+---
+license: apache-2.0
+library_name: lerobot
+tags:
+- lerobot
+- diffusion-policy
+- imitation-learning
+- robotics
+- ur3e
+- bimanual
+datasets:
+- WetLabRoboData/lerobot-data-stir_bar
+pipeline_tag: robotics
+---
+
+# diffusion-stir_bar-scratch
+
+LeRobot diffusion policy for the **stir_bar** task.
+
+| Field              | Value |
+|--------------------|-------|
+| Policy family      | diffusion (LeRobot) |
+| Variant            | Scratch (trained only on this task's data) |
+| Training dataset   | [`WetLabRoboData/lerobot-data-stir_bar`](https://huggingface.co/datasets/WetLabRoboData/lerobot-data-stir_bar) |
+
+| Target task        | stir_bar |
+| Eval episodes      | 20 |
+| Eval successes     | 15 / 20 |
+| Robot              | UR3e bimanual (3 cameras) |
+
+## Evaluation
+
+Rollout videos + per-episode outcomes: [`WetLabRoboData/eval-diffusion-stir_bar-scratch`](https://huggingface.co/datasets/WetLabRoboData/eval-diffusion-stir_bar-scratch).
+
+## Load and deploy
+
+```python
+from lerobot.policies.diffusion.modeling_diffusion import DiffusionPolicy
+policy = DiffusionPolicy.from_pretrained("WetLabRoboData/diffusion-stir_bar-scratch")
+```
+
+## Provenance
+
+Reorganized from `WetLabRoboData/lerobot-data-stir_bar` 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.