Introduction

Miniconda official website
Anaconda is a Python distribution for scientific computing, supports Linux, Mac, Windows, and contains many popular scientific computing and data analysis Python packages.
As the foreign distant, slow download, Anaconda installation package can go to Tsinghua mirror Anaconda download.

Miniconda is a lightweight alternative to Anaconda (632M + 3kM). It only includes python and conda by default, with a size of 55M (+250M), but you can install the required packages through pip and conda.
Miniconda installation package to be Tsinghua mirror miniconda download.
A slow download in foreign countries The
content of this article has been tested by Windows 10 pro under VirtualBox. The code is not run in cmd, but in Anaconda Prompot.

Download and install Miniconda

First to mirror miniconda Tsinghua latest version download the corresponding installation package, here I was under the Miniconda3-py38_4.8.3-Windows-x86_64.exe
directly installed on the line all the way, to note that the default conda not added to the system path, it is not here Make a request. After installation, a new column Anaconda3 will be added to the menu. At
Insert picture description here
this time, open Anaconda Prompt, and verify that conda has been installed and running on the system by entering the following command:

conda --version

Conda displays the version number you have installed. A conda 4.8.3similar description appears . If the downloaded version is too old, enter the following:

conda update conda

Conda compares the versions and then displays the versions that can be installed. If there is a newer version of conda, enter y to update:

Proceed ([y]/n)? y

Configure domestic mirror source

For well-known reasons, foreign websites often crash, explode, power outages, and bugs that cannot be connected (dog heads). Therefore, you need to configure the domestic mirror source by yourself. If you do not configure the painting, unless you go online scientifically, you will download other things later. I want fried chicken.

If you want to view the location of the configuration file and the existing mirror source, you can run the following statement

conda config --show-sources

Conda input operation, can be seen in column config location profiles, can be opened by Notepad or Notepad ++ (.condarc if no file to be executed conda config --set show_channel_urls yesafter generating the file again.)
Insert picture description here
After opening below and almost

then Copy all the following Tsinghua mirror configuration list (ali mix-in) into it. (Because the update is too fast and it is difficult to synchronize, Tsinghua does not synchronize pytorch-nightly, pytorch-nightly-cpu, ignite-nightly these three packages)

channels:
  - defaults
show_channel_urls: true
channel_alias: https://mirrors.tuna.tsinghua.edu.cn/anaconda
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
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/pro
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/msys2
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

After saving, you can add the Anaconda Python free warehouse.
As shown below
Insert picture description here
run conda clean -iclear cache index, the index is used to ensure that provided on the mirror.

Install Jupyter

Open Anaconda Prompt, enter and pip install jupyteryou can install Jupyter. I have to say that the download is slow. Fortunately, the mirror is blessed, otherwise you can't download
Jupyter. After installation, you can start Jupyter: jupyter notebook
Insert picture description here
Exit and press Ctrl + C in Prompt.

Environmental management

When you start using conda, there is already a default environment named base. However, you don't want to put the program into the basic environment. Create separate environments to isolate programs from each other.

pytorch

https://blog.csdn.net/weixin_43031092/article/details/108415314

Miniconda 4.8.3 on Windows 10 quickly install and configure the mirror and use Jupyter and usage records-lightweight Anaconda