diff --git a/test_tipc/config/PP-ShiTu/PPShiTu_general_rec_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt b/test_tipc/config/PP-ShiTu/PPShiTu_general_rec_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt
index 56dcff845be57473ca04a4a53b6c84b686b3a1ef..652203bcb5f9569cace42b82daca2421c33be9d4 100644
--- a/test_tipc/config/PP-ShiTu/PPShiTu_general_rec_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt
+++ b/test_tipc/config/PP-ShiTu/PPShiTu_general_rec_linux_gpu_normal_normal_paddle2onnx_python_linux_cpu.txt
@@ -1,5 +1,5 @@
 ===========================paddle2onnx_params===========================
-model_name:PP-ShiTu_general_rec
+model_name:GeneralRecognition_PPLCNet_x2_5
 python:python3.7
 2onnx: paddle2onnx
 --model_dir:./deploy/models/general_PPLCNet_x2_5_lite_v1.0_infer/
@@ -9,8 +9,8 @@ python:python3.7
 --opset_version:10
 --enable_onnx_checker:True
 inference_model_url:https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/rec/models/inference/general_PPLCNet_x2_5_lite_v1.0_infer.tar
-inference:./python/predict_cls.py
+inference:./python/predict_rec.py
 Global.use_onnx:True
-Global.inference_model_dir:./models/general_PPLCNet_x2_5_lite_v1.0_infer
+Global.rec_inference_model_dir:./models/general_PPLCNet_x2_5_lite_v1.0_infer
 Global.use_gpu:False
--c:configs/inference_cls.yaml
\ No newline at end of file
+-c:configs/inference_rec.yaml
\ No newline at end of file
diff --git a/test_tipc/prepare.sh b/test_tipc/prepare.sh
index 1cb53ea4d919f237eb0f8421127ee3bbd16bc9a6..99575fa3f529d75daaf529e6325e31d5ef8584a0 100644
--- a/test_tipc/prepare.sh
+++ b/test_tipc/prepare.sh
@@ -85,12 +85,18 @@ if [[ ${MODE} = "cpp_infer" ]]; then
     if [[ ! -d "./deploy/cpp/paddle_inference/" ]]; then
         pushd ./deploy/cpp/
         PADDLEInfer=$3
-        if [ "" = "$PADDLEInfer" ];then
+        if [ "" = "$PADDLEInfer" ]; then
             wget -nc https://paddle-inference-lib.bj.bcebos.com/2.2.2/cxx_c/Linux/GPU/x86-64_gcc8.2_avx_mkl_cuda10.1_cudnn7.6.5_trt6.0.1.5/paddle_inference.tgz --no-check-certificate
+            tar xf paddle_inference.tgz
         else
             wget -nc ${PADDLEInfer} --no-check-certificate
+            tar_name=$(func_get_url_file_name "$PADDLEInfer")
+            tar xf ${tar_name}
+            paddle_inference_install_dir=${tar_name%.*}
+            if [ ! -d "paddle_inference" ]; then
+                ln -s ${paddle_inference_install_dir} paddle_inference
+            fi
         fi
-        tar xf paddle_inference.tgz
         popd
     fi
     if [[ $FILENAME == *infer_cpp_linux_gpu_cpu.txt ]]; then