publish hill: perpetual, policy.onnx

README.md

@@ -28,5 +28,4 @@ The observation normalizer is baked into `policy.onnx`; feed raw observations.
 
 - **repo**: `pollen-robotics/microduck_rl`
 - **branch**: `main`
-- **commit**: `fe529223f`
-- exported from a checkout with uncommitted changes
+- **commit**: `9fe37534b`

manifest.json

@@ -29,9 +29,9 @@
   "slot": "walk",
   "training": {
     "repo": "pollen-robotics/microduck_rl",
-    "commit": "fe529223f",
+    "commit": "9fe37534b",
     "branch": "main",
-    "dirty": true,
+    "dirty": false,
     "source_file": "policy.onnx"
   },
   "arena": {

policy.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:454f4a82b40efca65a6ee610af3c1e210d2630949581be646c1797dad5f1d7af
-size 793722
+oid sha256:3b97facc314f09b367891e0494aa5860b48dd17423fe8815ec690e0ca3e6a2a0
+size 793721