ubuntu配置NVIDIA384.130+cuda9.0+cudnn+anaconda3+TensorFlow1.8+pycharm

1、ubuntu安装
删除EFI
https://blog.csdn.net/weixin_39837402/article/details/79961912
安装教程
https://blog.csdn.net/s717597589/article/details/79117112
https://blog.csdn.net/yeluohanchan/article/details/74999091
https://blog.csdn.net/Meditator_hkx/article/details/52626077

3、ubuntu安装nvidia显卡
下载run文件
https://blog.csdn.net/xunan003/article/details/81665835

4、cuda9.0安装
https://docs.nvidia.com/cuda/cuda-installation-guide-linux/#ubuntu-installation
在最后sudo apt-get update和sudo apt-get install cuda要在home目录下进行
https://www.jianshu.com/p/71bc5f02ecd2
5、cudnn安装
https://blog.csdn.net/CosmosHua/article/details/76644029
6、anaconda3.4.2安装
Python默认版本切换
https://blog.csdn.net/wzyaiwl/article/details/81263004
anaconda安装
https://blog.csdn.net/u012243626/article/details/82469174
7、anaconda配制tensorflow虚拟环境
https://blog.csdn.net/pursuit_zhangyu/article/details/79362128
安装pytorch的方式和TensorFlow一样
建立conda虚拟环境
pip install pytorch安装包名字
pip install torchvision

8、pycharm环境配置
pycharm安装
https://jingyan.baidu.com/article/c45ad29c030102051753e20f.html
环境配置
https://blog.csdn.net/qq_33039859/article/details/81365294
选择已经存在的环境，进入虚拟环境的Python目录进行建立，建立好后，后期需要什么软件只需在软件里面下载