[英]Tensorflow-gpu doesn't working with Nvidia driver 455.45 & CUDA version - 11.1 on UBUNTU 20.04
I'm trying to install the latest version of TensorFlow 2 using pip install tensorflow
as of mentioned in official website .我正在尝试使用官方网站中提到的
pip install tensorflow
安装最新版本的 TensorFlow 2 。 I've the latest version of Ubuntu 20.04 LTS .我有最新版本的Ubuntu 20.04 LTS 。 And my laptop has Nvidia GeForce GTX 1050ti graphics card.
我的笔记本电脑有Nvidia GeForce GTX 1050ti显卡。 But when I install Nvidia driver, it only installs the latest version - 455.45 and its CUDA version - 11.1.
但是当我安装 Nvidia 驱动程序时,它只安装最新版本 - 455.45 及其CUDA 版本 - 11.1。
Then I installed cuda-toolkit 10.1 and CuDNN7.6.然后我安装了 cuda-toolkit 10.1 和 CuDNN7.6。 But still TensorFlow doesn't detect GPU on my system.
但是 TensorFlow 在我的系统上仍然没有检测到 GPU。
Help me to install TensorFlow on the latest version of Ubuntu and Nvidia Driver.帮我在最新版本的Ubuntu和Nvidia驱动上安装TensorFlow。
If you find any blogpost which shows the proper way to install it, please share it.如果您发现任何显示正确安装方法的博文,请分享。 Thank you.
谢谢你。
Just simply do a clean install of CUDA (First delete all of its traces and then try to reinstall the correct version, which is 10.1
).只需简单地安装 CUDA (首先删除其所有痕迹,然后尝试重新安装正确的版本,即
10.1
)。
OR:- you can install Miniconda or Anaconda on your computer and then run this command on cmd
:-或者:-您可以在您的计算机上安装 Miniconda 或 Anaconda,然后在
cmd
上运行此命令:-
conda create --name tf_gpu tensorflow-gpu
which would automatically download and install all the correct software needed to run TensorFlow on GPU.这将自动下载并安装在 GPU 上运行 TensorFlow 所需的所有正确软件。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.