Checkpoint 5 - incomplete

training/README.md

@@ -7,12 +7,15 @@ policy that runs under LiDAR perception in Webots.
 sim demos (active-scan teacher on tracker output, K=4 frame stack)
     │
     ▼
-bc_pretrain.py  ──►  runs/bc_v3   (deployed policy — beats Strömbom on n≥8)
+bc_pretrain.py  ──►  runs/bc   (BC baseline)
     │
-    ▼  (optional: tools/auto_dagger.sh + tools/dagger_merge_train.py
-    │            if sim-trained doesn't transfer cleanly to Webots)
+    ▼  KL-regularised PPO fine-tune (training/train_ppo.py)
     │
-runs/bc_dagger
+runs/rl         (deployed `rl` mode)
+
+# optional branch — kept for reference, not deployed:
+runs/bc_dagger     (Webots-grounded DAgger refinement, useful if a
+                    modified world breaks sim-to-real transfer)
 ```
 
 ## Files
@@ -42,14 +45,14 @@ rollout collection, not gradient compute.
 # 1. Sim demos with the active-scan + Strömbom teacher under LiDAR
 #    perception. K=4 frame stack so the MLP has temporal context.
 python -m tools.collect_demos --teacher strombom \
-    --out demos_v3.npz --seeds-per-n 15 --subsample 3 --frame-stack 4
+    --out demos.npz --seeds-per-n 15 --subsample 3 --frame-stack 4
 
 # 2. Behavior-clone.
-python -m training.bc_pretrain --demos demos_v3.npz \
-    --out runs/bc_v3 --epochs 60 --net-arch 512,512
+python -m training.bc_pretrain --demos demos.npz \
+    --out runs/bc --epochs 60 --net-arch 512,512
 
 # 3. Evaluate.
-python -m training.eval --policy runs/bc_v3 \
+python -m training.eval --policy runs/bc \
     --max-flock 10 --max-steps 8000 --n-seeds 5
 ```
 
@@ -78,7 +81,7 @@ seat:
 HERDING_POLICY_DIR=$PWD/training/runs/bc_dagger \
 HERDING_DAGGER_DRIVER=student \
 tools/auto_dagger.sh 3 60
-python -m tools.dagger_merge_train --out runs/bc_dagger_v2
+python -m tools.dagger_merge_train --out runs/bc_dagger
 ```
 
 ## Available analytic teachers
@@ -107,6 +110,6 @@ python -m training.eval --policy sequential  --max-flock 10 --max-steps 8000 --n
 tools/run_webots.sh 10 rl
 ```
 
-The dog controller loads the highest-priority policy that exists
-(`bc_dagger_v2` → `bc_dagger` → `bc_v3`). Override with
-`HERDING_POLICY_DIR=…` if you want a specific checkpoint.
+The dog controller loads `runs/bc` for `bc` mode and `runs/rl` for
+`rl` mode. Override with `HERDING_POLICY_DIR=…` for a specific
+checkpoint.