[docs] update quick_start and README (#181)

* [docs] update quick_start and README * [docs] remove ENABLE_PADDLE_FRONTEND flags in win docs * [docs] update supported python version for win * [docs] update windows sdk example docs Co-authored-by: Jason <jiangjiajun@baidu.com>
2025-12-24 13:28:13 +08:00 · 2022-09-06 21:11:04 +08:00
parent d9a0dd328b
commit e1ab1104c2
4 changed files with 53 additions and 8 deletions
--- a/README.md
+++ b/README.md
@@ -69,7 +69,10 @@
 ```bash
 pip install numpy opencv-python fastdeploy-gpu-python -f https://www.paddlepaddle.org.cn/whl/fastdeploy.html
 ```
-
+##### [Anaconda配置GPU环境(推荐)](docs/quick_start/Python_prebuilt_wheels.md)
+```bash
+conda config --add channels conda-forge && conda install cudatoolkit=11.2 cudnn=8.2
+```
 ##### 安装CPU版本

 ```bash
--- a/docs/compile/how_to_build_windows.md
+++ b/docs/compile/how_to_build_windows.md
@@ -17,7 +17,7 @@ git clone https://github.com/PaddlePaddle/FastDeploy.git
 cd FastDeploy && git checkout develop
 mkdir build && cd build

-cmake .. -G "Visual Studio 16 2019" -A x64 -DCMAKE_INSTALL_PREFIX=D:\Paddle\FastDeploy\build\fastdeploy-win-x64-0.2.0 -DENABLE_ORT_BACKEND=ON -DENABLE_VISION=ON -DENABLE_VISION_VISUALIZE=ON
+cmake .. -G "Visual Studio 16 2019" -A x64 -DCMAKE_INSTALL_PREFIX=D:\Paddle\FastDeploy\build\fastdeploy-win-x64-0.2.0 -DENABLE_ORT_BACKEND=ON -DENABLE_PADDLE_BACKEND=ON -DENABLE_VISION=ON -DENABLE_VISION_VISUALIZE=ON
 msbuild fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64
 msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64
 ```
@@ -32,7 +32,7 @@ git clone https://github.com/PaddlePaddle/FastDeploy.git
 cd FastDeploy && git checkout develop
 mkdir build && cd build

-cmake .. -G "Visual Studio 16 2019" -A x64 -DCMAKE_INSTALL_PREFIX=D:\Paddle\FastDeploy\build\fastdeploy-win-x64-gpu-0.2.0 -DWITH_GPU=ON -DENABLE_ORT_BACKEND=ON -DENABLE_VISION=ON -DENABLE_VISION_VISUALIZE=ON -DENABLE_PADDLE_FRONTEND=ON -DCUDA_DIRECTORY=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2
+cmake .. -G "Visual Studio 16 2019" -A x64 -DCMAKE_INSTALL_PREFIX=D:\Paddle\FastDeploy\build\fastdeploy-win-x64-gpu-0.2.0 -DWITH_GPU=ON -DENABLE_ORT_BACKEND=ON -DENABLE_PADDLE_BACKEND=ON -DENABLE_VISION=ON -DENABLE_VISION_VISUALIZE=ON -DCUDA_DIRECTORY="C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.2"
 msbuild fastdeploy.sln /m /p:Configuration=Release /p:Platform=x64
 msbuild INSTALL.vcxproj /m /p:Configuration=Release /p:Platform=x64  

@@ -52,9 +52,11 @@ git clone https://github.com/PaddlePaddle/FastDeploy.git
 cd FastDeploy && git checkout develop

 set ENABLE_ORT_BACKEND=ON
+set ENABLE_PADDLE_BACKEND=ON
 set ENABLE_VISION=ON
+set ENABLE_VISION_VISUALIZE=ON

-% 这里指定用户自己的python解释器 以ptyhon3.8为例 %
+% 这里指定用户自己的python解释器 以python3.8为例 %
 C:\Python38\python.exe setup.py build
 C:\Python38\python.exe setup.py bdist_wheel
 ```
@@ -75,14 +77,13 @@ set ENABLE_ORT_BACKEND=ON
 % 说明：如果不编译TensorRT后端 此项为OFF %
 set ENABLE_TRT_BACKEND=ON
 set ENABLE_PADDLE_BACKEND=ON
-set ENABLE_PADDLE_FRONTEND=ON
 set ENABLE_VISION=ON
 set ENABLE_VISION_VISUALIZE=ON

 git clone https://github.com/PaddlePaddle/FastDeploy.git
 cd FastDeploy && git checkout develop

-% 说明：这里指定用户自己的python解释器 以ptyhon3.8为例 %
+% 说明：这里指定用户自己的python解释器 以python3.8为例 %
 C:\Python38\python.exe setup.py build
 C:\Python38\python.exe setup.py bdist_wheel
 ```
--- a/docs/compile/how_to_use_sdk_on_windows.md
+++ b/docs/compile/how_to_use_sdk_on_windows.md
@@ -40,6 +40,8 @@ cmake .. -G "Visual Studio 16 2019" -A x64 -DFASTDEPLOY_INSTALL_DIR=%cd%\..\..\.
 ```bat
 msbuild infer_demo.sln /m:4 /p:Configuration=Release /p:Platform=x64
 ```
+## 配置依赖库路径
+#### 方式一：命令行设置环境变量
 编译好的exe保存在Release目录下，在运行demo前，需要将模型和测试图片拷贝至该目录。另外，需要在终端指定DLL的搜索路径。请在build目录下执行以下命令。
 ```bat
 set FASTDEPLOY_PATH=%cd%\..\..\..\..\..\..\..\fastdeploy-win-x64-gpu-0.2.0
@@ -49,11 +51,26 @@ set PATH=%FASTDEPLOY_PATH%\lib;%FASTDEPLOY_PATH%\third_libs\install\onnxruntime\
 ```bat
 copy /Y %FASTDEPLOY_PATH%\third_libs\install\onnxruntime\lib\onnxruntime* Release\
 ```
-由于较新的Windows在System32系统目录下自带了onnxruntime.dll，因此就算设置了PATH，系统依然会出现onnxruntime的加载冲突。因此需要先拷贝demo用到的onnxruntime.dll到exe所在的目录。
+由于较新的Windows在System32系统目录下自带了onnxruntime.dll，因此就算设置了PATH，系统依然会出现onnxruntime的加载冲突。因此需要先拷贝demo用到的onnxruntime.dll到exe所在的目录。如下
 ```bat
 where onnxruntime.dll
 C:\Windows\System32\onnxruntime.dll  # windows自带的onnxruntime.dll
 ```
+#### 方式二：拷贝依赖库到exe的目录下
+手动拷贝，或者在build目录下执行以下命令：
+```bat
+set FASTDEPLOY_PATH=%cd%\..\..\..\..\..\..\..\fastdeploy-win-x64-gpu-0.2.0
+copy /Y %FASTDEPLOY_PATH%\lib\*.dll Release\
+copy /Y %FASTDEPLOY_PATH%\third_libs\install\onnxruntime\lib\*.dll Release\
+copy /Y %FASTDEPLOY_PATH%\third_libs\install\opencv-win-x64-3.4.16\build\x64\vc15\bin\*.dll Release\
+copy /Y %FASTDEPLOY_PATH%\third_libs\install\paddle_inference\paddle\lib\*.dll Release\
+copy /Y %FASTDEPLOY_PATH%\third_libs\install\paddle_inference\third_party\install\mkldnn\lib\*.dll Release\
+copy /Y %FASTDEPLOY_PATH%\third_libs\install\paddle_inference\third_party\install\mklml\lib\*.dll Release\
+copy /Y %FASTDEPLOY_PATH%\third_libs\install\paddle2onnx\lib\*.dll Release\
+copy /Y %FASTDEPLOY_PATH%\third_libs\install\tensorrt\lib\*.dll Release\
+copy /Y %FASTDEPLOY_PATH%\third_libs\install\yaml-cpp\lib\*.dll Release\
+```
+
 ## 运行 demo
 ```bat
 cd Release
--- a/docs/quick_start/Python_prebuilt_wheels.md
+++ b/docs/quick_start/Python_prebuilt_wheels.md
@@ -5,7 +5,7 @@ FastDeploy提供了在Windows/Linux/Mac上的预先编译Python Wheel包，开
 目前各平台支持情况如下

 - Linux 支持Python3.6~3.9
- Windows 支持Python3.8~3.9
+- Windows 支持Python3.6~3.9
 - Mac 支持Python3.6~3.9

 ## 安装 CPU Python 版本
@@ -17,6 +17,30 @@ pip install fastdeploy-python -f https://www.paddlepaddle.org.cn/whl/fastdeploy.
 pip install fastdeploy-gpu-python -f https://www.paddlepaddle.org.cn/whl/fastdeploy.html
 ```

+## Anaconda 快速配置 GPU 环境
+使用Anaconda的用户可在命令运行以下命令，快速配置GPU环境。如果是Windows用户，需要先打开`Anaconda Prompt (anaconda3)`命令行终端。
+- 增加 conda-forge 源
+```bash
+conda config --add channels conda-forge
+# 国内用户可以增加国内的源，如
+conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
+conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
+conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge/
+```
+- 新建 python 环境
+```bash
+conda create -n py38 python=3.8
+conda activate py38
+```  
+- 安装 cudatoolkit 11.x 和 cudnn 8.x
+```bash
+conda install cudatoolkit=11.2 cudnn=8.2
+```
+- 安装 FastDeploy GPU 版本 Python 包
+```bash
+pip install fastdeploy-gpu-python -f https://www.paddlepaddle.org.cn/whl/fastdeploy.html
+```
+
 ## 注意事项
 - 不要重复安装`fastdeploy-python`和`fastdeploy-gpu-python`
 - 如已安装CPU版本的`fastdeploy-python`后，在安装GPU版本的`fastdeploy-gpu-python`，请先执行`pip uninstall fastdeploy-python`卸载已有版本