安装Paddle环境

1、首先，确定自己的显卡驱动版本最大支持到哪里

比如这里最高支持CUDA11.6，就可以下载CUDA11.6以下的CUDA。这里统一用CUDA10.2+cudnn7.6.5（有需要我打包上传）

 2、去CUDA官网下载

CUDA Toolkit 10.2 Download | NVIDIA DeveloperSelect Target Platform Click on the green buttons that describe your target platform. Only supported platforms will be shown. Operating System Architecture Distribution Version Installer Type Do you want to cross-compile? Yes No Select Host Platform Click on the green buttons that describe your host platform. Only supported platforms will be shown. Operating Systemhttps://developer.nvidia.com/cuda-10.2-download-archive安装好看看系统环境变量有无添加成功

3、cudnn要稍微科学上网才方便注册下载

cuDNN Archive | NVIDIA DeveloperExplore and download past releases from cuDNN GPU-accelerated primitive library for deep neural networks.https://developer.nvidia.com/rdp/cudnn-archive#a-collapse765-102解压并复制对应的bin、lib、include在C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2的对应文件上

4、验证是否安装成功

打开cmd

cd C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\extras\demo_suite

然后 bandwidthTest.exe 和 deviceQuery.exe，都显示pass就可以了

5、然后创建一个虚拟环境，名字随便（我取了paddle）

然后去paddle的官网查看安装命令（注意是CUDA10.2）

PaddlePaddle-Parallel Distributed Deep Learning, efficient and extensible deep learning framework.PaddlePaddle (Parallel Distributed Deep Learning) is a simple, efficient and extensible deep learning framework.Please refer to PaddlePaddle Github for details, and Release Note for features incorporated in current version.Let’s start with studying basic concept of PaddlePaddle:PaddlePaddle Introduction : Introduction of the new features of PaddlePaddle 2.0 and description of the PaddlePaddle 2.0 upgrade guide.Model Visualization : Introduce VisualDL, a visual tool of PaddlePaddle.Dygraph to Static Graph : Introduce the transformation of dygraph to static graph.Inference and Deployment : Introduce the method of using the trained model to inference.Distributed Training : Introduce how the PaddlePaddle uses distributed training Customize OP : Introduce how to customize OP for PaddlePaddle.Performance Improving : Introduce how to improve performance of PaddlePaddle.Contribution: Introduce how to contribute for PaddlePaddle.FLAGS : Introduce the envirenment flags in paddle.https://www.paddlepaddle.org.cn/en然后cmd进入你的虚拟环境，输入

python -m pip install paddlepaddle-gpu==2.3.1 -i https://mirror.baidu.com/pypi/simple

 6、可安装官网验证

PaddleSeg/install_cn.md at release/2.6 · PaddlePaddle/PaddleSeg · GitHubEasy-to-use image segmentation library with awesome pre-trained model zoo, supporting wide-range of practical tasks in Semantic Segmentation, Interactive Segmentation, Panoptic Segmentation, Image Matting, 3D Segmentation, etc. - PaddleSeg/install_cn.md at release/2.6 · PaddlePaddle/PaddleSeghttps://github.com/PaddlePaddle/PaddleSeg/blob/release/2.6/docs/install_cn.md