diff --git a/sdk/c-opr-loaders/mace/dump_model.py b/sdk/c-opr-loaders/mace/dump_model.py
index 19959722..cdc7afee 100644
--- a/sdk/c-opr-loaders/mace/dump_model.py
+++ b/sdk/c-opr-loaders/mace/dump_model.py
@@ -30,7 +30,7 @@ def main():
     parser.add_argument("--input", help="mace model file")
     parser.add_argument("--param", help="mace param file")
     parser.add_argument(
-        "--output", help="converted model that can be fed to dump_with_testcase_mge.py"
+        "--output", help="converted mge model"
     )
     parser.add_argument("--config", help="config file with yaml format")
     args = parser.parse_args()
diff --git a/sdk/xor-deploy/xornet.py b/sdk/xor-deploy/xornet.py
index 5608354f..57104205 100644
--- a/sdk/xor-deploy/xornet.py
+++ b/sdk/xor-deploy/xornet.py
@@ -75,8 +75,8 @@ def main():
     for step, minibatch in enumerate(train_dataset):
         if step > 1000:
             break
-        data = minibatch["data"]
-        label = minibatch["label"]
+        data = mge.tensor(minibatch["data"])
+        label = mge.tensor(minibatch["label"])
         net.train()
         _, loss = train_fun(data, label)
         train_loss.append((step, loss.numpy()))
@@ -128,6 +128,11 @@ def main():
     print("Dump model as {}".format(model_name))
     pred_fun.dump(model_name, arg_names=["data"])
 
+    model_with_testcase_name = "xornet_with_testcase.mge"
+
+    print("Dump model with testcase as {}".format(model_with_testcase_name))
+    pred_fun.dump(model_with_testcase_name, arg_names=["data"], input_data=["#rand(0.1, 0.8, 4, 2)"])
+
 
 if __name__ == "__main__":
     main()