diff --git a/deploy/cpp/docs/compile/tensorrt/trt.md b/deploy/cpp/docs/compile/tensorrt/trt.md
index eedae48ce8..2fcf4ec748 100644
--- a/deploy/cpp/docs/compile/tensorrt/trt.md
+++ b/deploy/cpp/docs/compile/tensorrt/trt.md
@@ -25,9 +25,7 @@ $ docker run -it --gpus=all --name tensorrt-onnx  -v ~/paddle2onnx/:/paddle2onnx
 
 ```
 $ git clone https://github.com/PaddlePaddle/PaddleX.git
-$ cd PaddleX
-$ git checkout deploykit
-$ cd deploy/cpp
+$ cd PaddleX/deploy/cpp
 $ git clone https://github.com/NVIDIA/TensorRT.git
 # 如果不是其他版本的容器， 将cuda_dir路径换成自己的cuda路径即可
 $ sh scripts/tensorrt_build.sh --tensorrt_dir=/usr/lib/x86_64-linux-gnu/ --cuda_dir=/usr/local/cuda-11.1/targets/x86_64-linux/ --tensorrt_header=./TensorRT/
diff --git a/deploy/cpp/docs/compile/triton/docker.md b/deploy/cpp/docs/compile/triton/docker.md
index 171c0c3ca5..a0e31da857 100644
--- a/deploy/cpp/docs/compile/triton/docker.md
+++ b/deploy/cpp/docs/compile/triton/docker.md
@@ -168,7 +168,6 @@ curl -v localhost:8000/v2/health/ready
 ```
 git clone https://github.com/PaddlePaddle/PaddleX.git
 cd PaddleX
-git checkout deploykit
 ```
 
 ### 3.2 启动Triton客户端容器
diff --git a/paddlex/cv/models/classifier.py b/paddlex/cv/models/classifier.py
index 8a623aa24c..58bff9f493 100644
--- a/paddlex/cv/models/classifier.py
+++ b/paddlex/cv/models/classifier.py
@@ -158,7 +158,7 @@ def default_optimizer(self, parameters, learning_rate, warmup_steps,
                     exit=False)
                 logging.error(
                     "See this doc for more information: "
-                    "https://github.com/PaddlePaddle/PaddleX/blob/develop/docs/appendix/parameters.md#notice",
+                    "https://github.com/PaddlePaddle/PaddleX/blob/release/2.0.0/docs/parameters.md",
                     exit=False)
                 logging.error(
                     "warmup_steps should less than {} or lr_decay_epochs[0] greater than {}, "
diff --git a/paddlex/cv/models/detector.py b/paddlex/cv/models/detector.py
index 5440203803..38d9929654 100644
--- a/paddlex/cv/models/detector.py
+++ b/paddlex/cv/models/detector.py
@@ -127,7 +127,7 @@ def default_optimizer(self, parameters, learning_rate, warmup_steps,
                     exit=False)
                 logging.error(
                     "See this doc for more information: "
-                    "https://github.com/PaddlePaddle/PaddleX/blob/develop/docs/appendix/parameters.md#notice",
+                    "https://github.com/PaddlePaddle/PaddleX/blob/release/2.0.0/docs/parameters.md",
                     exit=False)
 
             scheduler = paddle.optimizer.lr.LinearWarmup(