Mark deploy patches as applied; add RealSense camera setup and gripper-convention check

README.md

@@ -96,17 +96,24 @@ the demonstrated range, so it does **not** need the gripper clamp the B-spline v
 
 ## Deployment
 
-Environment setup, arm bring-up, and the five repo patches for joint-space actions are identical
-to the [B-spline model card](https://huggingface.co/Dimios45/yam-duster-bspline-dp) — follow
-steps 1, 3 and 4 there. The clamp in `decode_action_vector` is harmless here and worth keeping
-anyway.
-
-Those patches are needed because the upstream rollout path decodes **end-effector** actions (its
-iPhone teleop records EE poses), while this model predicts **joints**. That is a difference in
-what was recorded, not in the LeRobot file format. Note also that a missing `top_image` key does
-not raise — `policy_local_bspline.py:625-629` feeds a black frame instead, so skipping the camera
-edit produces a half-blind policy rather than an error. Image resolution needs no change; the
-rollout resizes from the checkpoint's `shape_meta` automatically.
+Environment setup, camera configuration, arm bring-up, and the five repo patches for joint-space
+actions are identical to the
+[B-spline model card](https://huggingface.co/Dimios45/yam-duster-bspline-dp) — follow steps 1–4
+there. Those patches are **applied and verified** in the working tree these models were trained
+from, but are **not** in upstream `B-spline-policy/bspline-policy`; apply them yourself if you
+start from upstream. The gripper clamp in `decode_action_vector` is unnecessary for this model
+(see below) but harmless and worth keeping.
+
+They are needed because the upstream rollout path decodes **end-effector** actions (its iPhone
+teleop records EE poses), while this model predicts **joints** — a difference in what was
+recorded, not in the LeRobot file format. Two things carry over unchanged: a missing `top_image`
+does not raise (`policy_local_bspline.py:625-629` feeds a black frame, so a dead camera yields a
+half-blind policy rather than an error), and image resolution needs no change because the rollout
+resizes from the checkpoint's `shape_meta`. The top camera is an Intel RealSense; use
+`TOP_CAMERA_TYPE = 'realsense'` so the RGB pipeline matches training.
+
+The gripper-convention check in the B-spline card applies here too — verify it before the first
+rollout, or the policy will open to grasp and close to release.
 
 ```bash
 hf download Dimios45/yam-duster-dp deploy_ema.ckpt --local-dir ./ckpt