From 7019572f7bd06fcb832b41abc132c6522f0d8efe Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E5=A4=A9=E6=B6=AF=E5=8F=A4=E5=B7=B7?= Date: Mon, 30 Sep 2024 23:59:58 +0800 Subject: [PATCH] Update README.md --- README.md | 65 ++++++++++++++++++++++++++++++------------------------- 1 file changed, 36 insertions(+), 29 deletions(-) diff --git a/README.md b/README.md index d5af1c5..6a17fa8 100755 --- a/README.md +++ b/README.md @@ -80,53 +80,60 @@ conda activate videoEnv
Linux用户 -
(1) 下载CUDA 11.8
- 
wget https://developer.download.nvidia.com/compute/cuda/11.8.0/local_installers/cuda_11.8.0_520.61.05_linux.run
-
(2) 安装CUDA 11.8
- 
sudo sh cuda_11.8.0_520.61.05_linux.run
+
(1) 下载CUDA 11.7
+ 
wget https://developer.download.nvidia.com/compute/cuda/11.7.0/local_installers/cuda_11.7.0_515.43.04_linux.run
+
(2) 安装CUDA 11.7
+ 
sudo sh cuda_11.7.0_515.43.04_linux.run

1. 输入accept

-

2. 选中CUDA Toolkit 11.8（如果你没有安装nvidia驱动则选中Driver，如果你已经安装了nvidia驱动请不要选中driver），之后选中install，回车

+

2. 选中CUDA Toolkit 11.7（如果你没有安装nvidia驱动则选中Driver，如果你已经安装了nvidia驱动请不要选中driver），之后选中install，回车

3. 添加环境变量

在 ~/.bashrc 加入以下内容

# CUDA
-  export PATH=/usr/local/cuda-11.8/bin${PATH:+:${PATH}}
-  export LD_LIBRARY_PATH=/usr/local/cuda-11.8/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
+ export PATH=/usr/local/cuda-11.7/bin${PATH:+:${PATH}} + export LD_LIBRARY_PATH=/usr/local/cuda-11.7/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

使其生效

source ~/.bashrc
-
(3) 下载cuDNN 8.6.0
-

国内：cudnn-linux-x86_64-8.6.0.163_cuda11-archive.tar.xz 提取码：57mg

-

国外：cudnn-linux-x86_64-8.6.0.163_cuda11-archive.tar.xz

-
(4) 安装cuDNN 8.6.0
- 
 tar -xf cudnn-linux-x86_64-8.6.0.163_cuda11-archive.tar.xz
-   mv cudnn-linux-x86_64-8.6.0.163_cuda11-archive cuda
-   sudo cp ./cuda/include/* /usr/local/cuda-11.8/include/
-   sudo cp ./cuda/lib/* /usr/local/cuda-11.8/lib64/
-   sudo chmod a+r /usr/local/cuda-11.8/lib64/*
-   sudo chmod a+r /usr/local/cuda-11.8/include/*
+
(3) 下载cuDNN 8.4.1
+

国内：cudnn-linux-x86_64-8.4.1.50_cuda11.6-archive.tar.xz 提取码：57mg

+

国外：cudnn-linux-x86_64-8.4.1.50_cuda11.6-archive.tar.xz

+
(4) 安装cuDNN 8.4.1
+ 
 tar -xf cudnn-linux-x86_64-8.4.1.50_cuda11.6-archive.tar.xz
+   mv cudnn-linux-x86_64-8.4.1.50_cuda11.6-archive cuda
+   sudo cp ./cuda/include/* /usr/local/cuda-11.7/include/
+   sudo cp ./cuda/lib/* /usr/local/cuda-11.7/lib64/
+   sudo chmod a+r /usr/local/cuda-11.7/lib64/*
+   sudo chmod a+r /usr/local/cuda-11.7/include/*
Windows用户 -
(1) 下载CUDA 11.8
- cuda_11.8.0_522.06_windows.exe -
(2) 安装CUDA 11.8
-
(3) 下载cuDNN 8.6.0
-

国内：cudnn-windows-x86_64-8.6.0.163_cuda11-archive.zip 提取码：57mg

-

国外：cudnn-windows-x86_64-8.6.0.163_cuda11-archive.zip

-
(4) 安装cuDNN 8.6.0
+
(1) 下载CUDA 11.7
+ cuda_11.7.0_516.01_windows.exe +
(2) 安装CUDA 11.7
+
(3) 下载cuDNN 8.4.1
+

cudnn-windows-x86_64-8.4.1.50_cuda11.6-archive.zip

+
(4) 安装cuDNN 8.4.1

- 将cuDNN解压后的cuda文件夹中的bin, include, lib目录下的文件复制到C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8\对应目录下 + 将cuDNN解压后的cuda文件夹中的bin, include, lib目录下的文件复制到C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.7\对应目录下

- 安装GPU版本Paddlepaddle: - ```shell - python -m pip install paddlepaddle-gpu==2.6.1 -i https://pypi.tuna.tsinghua.edu.cn/simple - ``` + - windows: + + ```shell + python -m pip install paddlepaddle-gpu==2.4.2.post117 -f https://www.paddlepaddle.org.cn/whl/windows/mkl/avx/stable.html + ``` + + - Linux: + + ```shell + python -m pip install paddlepaddle-gpu==2.4.2.post117 -f https://www.paddlepaddle.org.cn/whl/linux/mkl/avx/stable.html + ``` - 安装GPU版本Pytorch: @@ -211,7 +218,7 @@ LAMA_SUPER_FAST = False # 保证效果 解决方案：改用cuda 11.8 ```shell -pip install torch==2.1.0 torchvision==0.16.0 --index-url https://download.pytorch.org/whl/cu118 +pip install torch==2.1.0 torchvision==0.15.2 --index-url https://download.pytorch.org/whl/cu118 ``` ## 赞助