model parser: use check missing for mhp checks (#36020)

* model parser: use check missing for mhp checks

* lint + support re

* lint...

* no walrus

* just remove

selfdrive/modeld/modeld.py

@@ -116,7 +116,7 @@ class ModelState:
     self.vision_output = np.zeros(vision_output_size, dtype=np.float32)
     self.policy_inputs = {k: Tensor(v, device='NPY').realize() for k,v in self.numpy_inputs.items()}
     self.policy_output = np.zeros(policy_output_size, dtype=np.float32)
-    self.parser = Parser()
+    self.parser = Parser(ignore_missing=('desired_curvature',))
 
     with open(VISION_PKL_PATH, "rb") as f:
       self.vision_run = pickle.load(f)

selfdrive/modeld/parse_model_outputs.py

@@ -22,9 +22,10 @@ class Parser:
     self.ignore_missing = ignore_missing
 
   def check_missing(self, outs, name):
-    if name not in outs and not self.ignore_missing:
+    missing = name not in outs
+    if missing and not self.ignore_missing:
       raise ValueError(f"Missing output {name}")
-    return name not in outs
+    return missing
 
   def parse_categorical_crossentropy(self, name, outs, out_shape=None):
     if self.check_missing(outs, name):
@@ -84,6 +85,13 @@ class Parser:
     outs[name] = pred_mu_final.reshape(final_shape)
     outs[name + '_stds'] = pred_std_final.reshape(final_shape)
 
+  def is_mhp(self, outs, name, shape):
+    if self.check_missing(outs, name):
+      return False
+    if outs[name].shape[1] == 2 * shape:
+      return False
+    return True
+
   def parse_vision_outputs(self, outs: dict[str, np.ndarray]) -> dict[str, np.ndarray]:
     self.parse_mdn('pose', outs, in_N=0, out_N=0, out_shape=(ModelConstants.POSE_WIDTH,))
     self.parse_mdn('wide_from_device_euler', outs, in_N=0, out_N=0, out_shape=(ModelConstants.WIDE_FROM_DEVICE_WIDTH,))
@@ -94,23 +102,18 @@ class Parser:
     self.parse_categorical_crossentropy('desire_pred', outs, out_shape=(ModelConstants.DESIRE_PRED_LEN,ModelConstants.DESIRE_PRED_WIDTH))
     self.parse_binary_crossentropy('meta', outs)
     self.parse_binary_crossentropy('lead_prob', outs)
-    if outs['lead'].shape[1] == 2 * ModelConstants.LEAD_MHP_SELECTION *ModelConstants.LEAD_TRAJ_LEN * ModelConstants.LEAD_WIDTH:
+    lead_mhp = self.is_mhp(outs, 'lead', ModelConstants.LEAD_MHP_SELECTION * ModelConstants.LEAD_TRAJ_LEN * ModelConstants.LEAD_WIDTH)
-      self.parse_mdn('lead', outs, in_N=0, out_N=0,
+    lead_in_N, lead_out_N = (ModelConstants.LEAD_MHP_N, ModelConstants.LEAD_MHP_SELECTION) if lead_mhp else (0, 0)
-                     out_shape=(ModelConstants.LEAD_MHP_SELECTION, ModelConstants.LEAD_TRAJ_LEN,ModelConstants.LEAD_WIDTH))
+    self.parse_mdn(
-    else:
+      'lead', outs, in_N=lead_in_N, out_N=lead_out_N,
-      self.parse_mdn('lead', outs, in_N=ModelConstants.LEAD_MHP_N, out_N=ModelConstants.LEAD_MHP_SELECTION,
+      out_shape=(ModelConstants.LEAD_MHP_SELECTION, ModelConstants.LEAD_TRAJ_LEN, ModelConstants.LEAD_WIDTH)
-                   out_shape=(ModelConstants.LEAD_TRAJ_LEN,ModelConstants.LEAD_WIDTH))
+      )
     return outs
 
   def parse_policy_outputs(self, outs: dict[str, np.ndarray]) -> dict[str, np.ndarray]:
-    if outs['plan'].shape[1] == 2 * ModelConstants.IDX_N * ModelConstants.PLAN_WIDTH:
+    plan_mhp = self.is_mhp(outs, 'plan',  ModelConstants.IDX_N * ModelConstants.PLAN_WIDTH)
-      self.parse_mdn('plan', outs, in_N=0, out_N=0,
+    plan_in_N, plan_out_N = (ModelConstants.PLAN_MHP_N, ModelConstants.PLAN_MHP_SELECTION) if plan_mhp else (0, 0)
-                     out_shape=(ModelConstants.IDX_N,ModelConstants.PLAN_WIDTH))
+    self.parse_mdn('plan', outs, in_N=plan_in_N, out_N=plan_out_N, out_shape=(ModelConstants.IDX_N,ModelConstants.PLAN_WIDTH))
-    else:
-      self.parse_mdn('plan', outs, in_N=ModelConstants.PLAN_MHP_N, out_N=ModelConstants.PLAN_MHP_SELECTION,
-                     out_shape=(ModelConstants.IDX_N,ModelConstants.PLAN_WIDTH))
-    if 'desired_curvature' in outs:
-      self.parse_mdn('desired_curvature', outs, in_N=0, out_N=0, out_shape=(ModelConstants.DESIRED_CURV_WIDTH,))
     self.parse_categorical_crossentropy('desire_state', outs, out_shape=(ModelConstants.DESIRE_PRED_WIDTH,))
     return outs