diff --git a/model_zoo/official/cv/centerface/src/centerface.py b/model_zoo/official/cv/centerface/src/centerface.py
index b023ada4c6c..aae19169a39 100644
--- a/model_zoo/official/cv/centerface/src/centerface.py
+++ b/model_zoo/official/cv/centerface/src/centerface.py
@@ -310,8 +310,8 @@ class TrainingWrapper(nn.Cell):
         else:
             cond = self.less_equal(self.base, flag_sum)
 
-        self.optimizer(grads)
-        return (loss, cond, sens)
+        ret = (loss, cond, sens)
+        return F.depend(ret, self.optimizer(grads))
 
 
 class CenterFaceWithNms(nn.Cell):
diff --git a/model_zoo/official/recommend/ncf/src/ncf.py b/model_zoo/official/recommend/ncf/src/ncf.py
index c48af973ca7..6a9bb21059f 100644
--- a/model_zoo/official/recommend/ncf/src/ncf.py
+++ b/model_zoo/official/recommend/ncf/src/ncf.py
@@ -20,6 +20,7 @@ from mindspore.nn.layer.activation import get_activation
 import mindspore.common.dtype as mstype
 from mindspore.ops import operations as P
 from mindspore.common.initializer import initializer
+from mindspore.ops import functional as F
 from mindspore.ops import composite as C
 from mindspore.context import ParallelMode
 from mindspore.nn.wrap.grad_reducer import DistributedGradReducer
@@ -260,8 +261,7 @@ class TrainStepWrap(nn.Cell):
         if self.reducer_flag:
             # apply grad reducer on grads
             grads = self.grad_reducer(grads)
-        self.optimizer(grads)
-        return loss
+        return F.depend(loss, self.optimizer(grads))
 
 
 class PredictWithSigmoid(nn.Cell):