Use Miniconda line of code configuration Pytorch environment

Installation miniconda

Similarly anaconda installation, only need to download the corresponding anaconda. I mainly do not need anaconda own scientific computing package, so we chose miniconda

1. Tsinghua mirror station, download miniconda

   # 清华镜像站地址
   https://mirrors.tuna.tsinghua.edu.cn/help/anaconda/

2. Installation miniconda, first enter the installation files folder

   bash Miniconda3-latest-Linux-x86_64.sh

   During the installation, it will ask whether to agree to the license, enter yes, then let you choose where to install, as no special requirements, direct the transport, installed in the default location.

Set mirror source

• By modification in the user directory .condarc file:

  channels:
    - defaults
  show_channel_urls: true
  default_channels:
    - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
    - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free
    - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/r
  custom_channels:
    conda-forge: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
    msys2: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
    bioconda: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
    menpo: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
    pytorch: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
    simpleitk: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud

• If the file does not exist, you can create a direct, further modification

  touch .condarc

Specific installation depth learning environment

Installation tensorflow

• tensorflow-gpu version

  # 使用哪个tensorflow版本都行
  conda create -n tf-gpu python=3.6 tensorflow-gpu
  
  # 比如使用tensorflow1.13版本
  conda create -n tf-gpu python=3.6 tensorflow-gpu=1.13

• tensorflow-cpu version

  conda create -n tf-cpu python=3.6 tensorflow

• Description:

  • create: to create a virtual environment in conda in
  • -n: behind the tf-gpu is the name of the virtual environment

Installation keras

• hard

  conda install keras

Installation Pytorch

• Pytorch-gpu version

   conda create -n pyt-gpu python=3.6 pytorch torchvision cudatoolkit=10.1 -c pytorch

• pytorch-cpu version

   conda create -n pyt-cpu python=3.6 pytorch-cpu torchvision-cpu -c pytorch