diff --git a/nntrainer/models/model_common_properties.cpp b/nntrainer/models/model_common_properties.cpp
index 7599fd3a85..b9589c67f6 100644
--- a/nntrainer/models/model_common_properties.cpp
+++ b/nntrainer/models/model_common_properties.cpp
@@ -44,4 +44,9 @@ ModelTensorDataType::ModelTensorDataType(ModelTensorDataTypeInfo::Enum value) {
 }
 LossScale::LossScale(float value) { set(value); }
 
+bool LossScale::isValid(const float &value) const {
+  ml_loge("Loss scale cannot be 0");
+  return value != 0;
+}
+
 } // namespace nntrainer::props
diff --git a/nntrainer/models/model_common_properties.h b/nntrainer/models/model_common_properties.h
index 0c259f2aa6..8b5d539762 100644
--- a/nntrainer/models/model_common_properties.h
+++ b/nntrainer/models/model_common_properties.h
@@ -238,6 +238,14 @@ class LossScale : public Property<float> {
   LossScale(float value = 1.0f);
   static constexpr const char *key = "loss_scale"; /**< unique key to access */
   using prop_tag = float_prop_tag;                 /**< property type */
+
+  /**
+   * @brief check if valid
+   *
+   * @param value value to check
+   * @return bool true if valid
+   */
+  bool isValid(const float &value) const override;
 };
 
 } // namespace nntrainer::props
diff --git a/test/unittest/integration_tests/integration_test_mixed_precision.cpp b/test/unittest/integration_tests/integration_test_mixed_precision.cpp
index 92fcc999c1..a84a442c86 100644
--- a/test/unittest/integration_tests/integration_test_mixed_precision.cpp
+++ b/test/unittest/integration_tests/integration_test_mixed_precision.cpp
@@ -23,6 +23,11 @@
 #include <optimizer.h>
 using namespace nntrainer;
 
+/**
+ *  Supported Layer list
+ *  [InputLayer, BatchNormalizationLayer, FullyConnectedLayer, Conv1DLayer]
+ */
+
 TEST(mixed_precision, input_only_model_test) {
 
   std::unique_ptr<ml::train::Model> nn =
@@ -44,6 +49,32 @@ TEST(mixed_precision, input_only_model_test) {
   EXPECT_EQ(nn->reinitialize(), ML_ERROR_NONE);
 }
 
+TEST(mixed_precision, conv1d_model_test) {
+
+  std::unique_ptr<ml::train::Model> nn =
+    ml::train::createModel(ml::train::ModelType::NEURAL_NET, {"loss=mse"});
+  nn->setProperty(
+    {"batch_size=1", "model_tensor_type=FP16-FP16", "loss_scale=65536"});
+
+  auto graph = makeGraph({
+    {"input", {"name=in", "input_shape=3:1:3"}},
+    {"conv1d", {"name=conv1d_0", "filters=1", "kernel_size=3"}},
+    {"conv1d", {"name=conv1d_1", "filters=1", "kernel_size=1"}},
+    {"batch_normalization", {"name=bn_0", "epsilon=0.00001", "momentum=0.9"}},
+    {"batch_normalization", {"name=bn_1", "epsilon=0.00001", "momentum=0.9"}},
+  });
+
+  for (auto &node : graph) {
+    nn->addLayer(node);
+  }
+
+  nn->setOptimizer(ml::train::createOptimizer("adam", {"learning_rate = 0.1"}));
+
+  EXPECT_EQ(nn->compile(), ML_ERROR_NONE);
+  EXPECT_EQ(nn->initialize(), ML_ERROR_NONE);
+  EXPECT_EQ(nn->reinitialize(), ML_ERROR_NONE);
+}
+
 TEST(mixed_precision, loss_scale_test) {
   std::unique_ptr<ml::train::Model> nn =
     ml::train::createModel(ml::train::ModelType::NEURAL_NET, {"loss=mse"});
@@ -52,10 +83,8 @@ TEST(mixed_precision, loss_scale_test) {
                                 "loss_scale=0"}),
                std::invalid_argument);
 
-  EXPECT_NO_THROW(
-    nn->setProperty(
-      {"batch_size=1", "model_tensor_type=FP16-FP16", "loss_scale=65536"}),
-    std::invalid_argument);
+  EXPECT_NO_THROW(nn->setProperty(
+    {"batch_size=1", "model_tensor_type=FP16-FP16", "loss_scale=65536"}));
 }
 
 TEST(mixed_precision, model_tensor_type_test) {