diff --git a/ppcls/modeling/architectures/mobilenet_v3.py b/ppcls/modeling/architectures/mobilenet_v3.py
index 4afd664d26c20d4e25818d5cbbb1fc096f866a67..95a023c9790e0d7c1593ce2169546c81543e0e2a 100644
--- a/ppcls/modeling/architectures/mobilenet_v3.py
+++ b/ppcls/modeling/architectures/mobilenet_v3.py
@@ -23,8 +23,7 @@ import paddle.nn as nn
 import paddle.nn.functional as F
 from paddle.nn import Conv2d, BatchNorm, Linear, Dropout
 from paddle.nn import AdaptiveAvgPool2d, MaxPool2d, AvgPool2d
-# TODO: need to be removed later!
-from paddle.fluid.regularizer import L2Decay
+from paddle.regularizer import L2Decay
 
 import math
 
@@ -194,11 +193,9 @@ class ConvBNLayer(nn.Layer):
             num_channels=out_c,
             act=None,
             param_attr=ParamAttr(
-                name=name + "_bn_scale",
-                regularizer=L2Decay(regularization_coeff=0.0)),
+                name=name + "_bn_scale", regularizer=L2Decay(0.0)),
             bias_attr=ParamAttr(
-                name=name + "_bn_offset",
-                regularizer=L2Decay(regularization_coeff=0.0)),
+                name=name + "_bn_offset", regularizer=L2Decay(0.0)),
             moving_mean_name=name + "_bn_mean",
             moving_variance_name=name + "_bn_variance")
 
diff --git a/ppcls/modeling/architectures/resnest.py b/ppcls/modeling/architectures/resnest.py
index 18ab76f807a34857df672bf9b773470a4082b6c5..e175a16a33608b65bb34df2b6797e49e47cdaeb2 100644
--- a/ppcls/modeling/architectures/resnest.py
+++ b/ppcls/modeling/architectures/resnest.py
@@ -24,8 +24,7 @@ from paddle import ParamAttr
 from paddle.nn.initializer import MSRA
 from paddle.nn import Conv2d, BatchNorm, Linear, Dropout
 from paddle.nn import AdaptiveAvgPool2d, MaxPool2d, AvgPool2d
-# TODO: need to be removed later!
-from paddle.fluid.regularizer import L2Decay
+from paddle.regularizer import L2Decay
 
 __all__ = ["ResNeSt50_fast_1s1x64d", "ResNeSt50"]
 
@@ -58,11 +57,9 @@ class ConvBNLayer(nn.Layer):
             num_filters,
             act=act,
             param_attr=ParamAttr(
-                name=name + "_scale",
-                regularizer=L2Decay(regularization_coeff=bn_decay)),
+                name=name + "_scale", regularizer=L2Decay(bn_decay)),
             bias_attr=ParamAttr(
-                name + "_offset",
-                regularizer=L2Decay(regularization_coeff=bn_decay)),
+                name + "_offset", regularizer=L2Decay(bn_decay)),
             moving_mean_name=name + "_mean",
             moving_variance_name=name + "_variance")