CONDA common command summary

One 、conda command

conda Commands can be used to ：

• Create a new conda Environmental Science .
• In the existing conda Installation in the environment / Upgrade package .
• Inquire about / Search for Anaconda The index of the package and the current Anaconda Installation .

Tip: We can put the front with two short lines （--） Of Commonly used Commands are abbreviated （ Not all can be abbreviated ）, The method is to take a short line and the initial of the option . for example --name Can be abbreviated to -n,--envs Can be abbreviated to -e.

Use conda -V To view the current conda edition .

We can ask for help from the command line , namely

conda --help

Of course, abbreviations can also be used

conda -h

If we want to take a closer look install How to use the command , Just execute

conda install -h

Two 、conda info

conda info Can be used to view relevant information , The format is as follows ：

conda info [-a] [--base] [-e] [-s] [--unsafe-channels]

---

-a, -all： View all information （ Not commonly used ）.

---

–base： View the path of the base environment .

conda info --base

---

-e, --envs： List all the current conda Environmental Science .

conda info -e

---

-s, --system： List all environment variables （ Not commonly used ）.

---

–unsafe-channels： View the channel exposed by the token .

conda info --unsafe-channels

3、 ... and 、conda create

conda create Can be used to create new conda Environmental Science , The format is as follows ：

conda create [-n env_name | -p path] [--clone env_name] [-c channel_address]
             [packages]

---

-n, --name： We can use Name of the environment To create a conda Environmental Science , Suppose our environment name is my_env

conda create -n my_env

After creation , We use the following command to enter the environment

conda activate my_env

At this time, there is one more in front of the command line (my_env), This is where we entered the environment sign .

Use conda list Command to see which packages are installed in this environment .

Because we didn't install the package when creating the environment , So the natural result is Nothing there? .

If you want to exit the environment , Just execute

conda deactivate

Be careful ： We can only retreat to the base environment , Continue to exit in the base environment, still in the base environment .

---

-p, --prefix： In addition to creating an environment by name , We can also create... Based on the address conda Environmental Science .

In the example above , We created my_env The address of the environment is ：G:\anaconda3\envs\my_env. in fact , If you don't specify an address , Then all environments will be created in ...\anaconda3\envs\ Next .

To create based on address , Just execute

conda create -p G:\anaconda3\envs\my_env

---

–clone：--clone Mainly for clone （ Copy ） The existing environment .

We can clone according to the name of the existing environment , Suppose the name of the existing environment is original_env, The name of the new environment is new_env, Then just execute

conda create -n new_env --clone original_env

Of course, we can also clone according to the address of the existing environment ：

conda create -n new_env --clone G:\anaconda3\envs\original_env

---

-c, --channel： First of all, let's introduce what is conda channel .

conda Channels are stored package The location of , install package when conda Will search for existing Channel collection , And select one of the channels to install package .

conda The default channel is https://repo.anaconda.com/pkgs/, But the channel needs to pay , We usually use conda-forge This channel to install , It's free .

The channel settings are omitted here , Subsequent chapters will refer to .

---

packages： Our newly created environment cannot be used without installing the necessary packages , Packages that need to be installed Space off , namely

conda create -n my_env python numpy flask

We used this command to create a file called my_env Environment , Three packages are installed at the same time python、numpy、flask.

After installation , have access to conda list Command to view installed packages .

Besides , our python The latest version is installed , Of course, we can also specify the version for installation ：

conda create -n my_env python=3.9.7 numpy flask

Four 、conda install

conda install Can be used in conda Install the corresponding package in the environment , The specific format is as follows ：

conda install [-n env_name | -p path] [-c channel_address] [packages]

The options already introduced will not be introduced here .

for example , We will be having my_env Lower installation 3.9.7 Version of python. We can enter the environment first , Then install ：

conda activate my_env
conda install python=3.9.7

Be careful , there = It's actually a fuzzy constraint , See the figure below for details ：

You can also directly specify the installation environment ：

conda install -n my_env python=3.9.7 #  By name 
conda install -p G:\anaconda3\envs\my_env python=3.9.7 #  According to the address 

Another example is , install cvxpy when , We can go from the designated channel （conda-forge） Installation

conda install -n my_env -c conda-forge cvxpy

5、 ... and 、conda remove

conda remove Can be used to remove conda Some packages in the environment , You can also remove the entire environment , The specific format is as follows ：

conda remove [-n env_name | -p path] [packages] [--all]

for example , If we need to remove my_env Under the numpy, Just execute

conda remove -n my_env numpy

Of course, we can also remove all packages （ The whole conda Environmental Science ）

conda remove -n my_env --all

---

It should be noted that ,conda No command was provided to rename the environment , We can only clone an original environment first , Then delete the original environment , The specific operation is as follows （ For example, the environment a Rename to b）：

conda create -n b --clone a
conda remove -n a --all

---

conda uninstall yes conda remove Of Alias , We can also use conda uninstall To remove the package , The method is the same .

6、 ... and 、conda list

conda list Used to list conda Packages in the environment , The specific format is as follows ：

conda list [-n env_name | -p path] [package]

conda list Only list all packages in the current environment , To list packages for other environments , For example, list my_env All the packages under , Just execute ：

conda list -n my_env

To list my_env China and numpy Related to the package （ To be exact, it contains numpy The bag of words ）, Just execute ：

conda list -n my_env numpy

Based on this , We can use conda list To check whether the corresponding package is installed in the current environment （ Only one... Can be queried at a time ）, If the final result is not listed , That means that... Is not installed .

7、 ... and 、conda search

conda search Used to search the specified package , The specific format is as follows ：

conda search [-c channel_address] [-f] [packages]

We can specify in conda-forge Mid search numpy：

conda search -c conda-forge numpy

in fact , The above command will search for all that contain numpy The bag of words , If we just want to search numpy This package , Need to add -f（--full-name Abbreviation ）, namely

conda search -c conda-forge -f numpy

8、 ... and 、conda update

conda update Used to upgrade a series of packages to the latest version , The specific format is as follows ：

conda update [-n env_name | -p path] [packages] [--all]

for example , If we want to upgrade my_env Under the numpy and scipy, Just execute

conda update -n my_env numpy scipy

If we want to update my_env All packages in , Then we need to carry out

conda update -n my_env --all

---

If you want to update conda In itself , Then we need to carry out

conda update conda

If you want to update anaconda , Then we need to carry out

conda update anaconda

notes ： upgrade Anaconda Need to upgrade before conda.

Nine 、conda config

conda config Used to configure conda Channel , Relevant information will be stored in C:\Users\ Your username \.condarc In file .

The format of adding a channel is ：

conda config --add channels [channel]

for example , To add conda-forge This channel , Just execute

conda config --add channels conda-forge

Because the speed of foreign channels is generally unstable , We usually consider using domestic channels , for example Tsinghua mirror .

in consideration of Windows The user cannot directly create a file named .condarc The file of , We can do it first

conda config --set show_channel_urls yes

Then modify the file directly .

The contents are as follows ：

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/r
  - 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
  pytorch-lts: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  simpleitk: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud

After modification , perform conda clean -i Clear index cache , Make sure you use the index provided by the mirror station .

After setting up , We can execute

conda config --get channels

To view the added channels .