fix RMSProp one_dim_param_no_weight_decay

PaddlePaddle · Apr 17, 2023 · 55e2114 · 55e2114
1 parent a283b51
commit 55e2114
Showing 1 changed file with 17 additions and 17 deletions.
diff --git a/ppcls/optimizer/optimizer.py b/ppcls/optimizer/optimizer.py
@@ -228,26 +228,26 @@ def __init__(self,
     def __call__(self, model_list):
         # model_list is None in static graph
         parameters = None
-        if len(self.no_weight_decay_name_list) > 0:
+        if model_list:
             params_with_decay = []
             params_without_decay = []
             for m in model_list:
-                params = [p for n, p in m.named_parameters() \
-                          if not any(nd in n for nd in self.no_weight_decay_name_list)]
-                params_with_decay.extend(params)
-                params = [p for n, p in m.named_parameters() \
-                          if any(nd in n for nd in self.no_weight_decay_name_list) or (self.one_dim_param_no_weight_decay and len(p.shape) == 1)]
-                params_without_decay.extend(params)
-            parameters = [{
-                "params": params_with_decay,
-                "weight_decay": self.weight_decay
-            }, {
-                "params": params_without_decay,
-                "weight_decay": 0.0
-            }]
-        else:
-            parameters = sum([m.parameters() for m in model_list],
-                             []) if model_list else None
+                for n, p in m.named_parameters():
+                    if any(nd in n for nd in self.no_weight_decay_name_list) \
+                        or (self.one_dim_param_no_weight_decay and len(p.shape) == 1):
+                        params_without_decay.append(p)
+                    else:
+                        params_with_decay.append(p)
+            if params_without_decay:
+                parameters = [{
+                    "params": params_with_decay,
+                    "weight_decay": self.weight_decay
+                }, {
+                    "params": params_without_decay,
+                    "weight_decay": 0.0
+                }]
+            else:
+                parameters = params_with_decay
         opt = optim.RMSProp(
             learning_rate=self.learning_rate,
             momentum=self.momentum,