管理 Conda 建立的 Python 虚拟环境

最新推荐文章于 2025-04-06 17:55:18 发布

Hwrn.aou

最新推荐文章于 2025-04-06 17:55:18 发布

阅读量1.4w

点赞数 11

本文链接：https://blog.youkuaiyun.com/weixin_44881875/article/details/104162917

版权

管理 Conda 建立的 Python 虚拟环境

下载安装
- 下载 Anaconda
- 安装 Anaconda
新建虚拟环境
- 更改新虚拟环境的位置.
- 虚拟环境的切换
向虚拟环境中添加包
复制虚拟环境
删除虚拟环境
移动自定义路径

换了新电脑 (Windows 10), 打算用 Anaconda 管理自己的包.

下载安装

下载 Anaconda

下载最新版的 Anaconda.

安装 Anaconda

直接把 Anaconda 安装在目标位置.
这个时候发现安装目录下, Anaconda 大量内容丢失, 我的判断标准是: 没有 \Script 文件夹.

为了安全, 下载安装 VS2019, 勾选 python 组件并安装. 找到 VS2019 下载好的文件夹, 移动到 Anaconda 安装位置即可.

接下来新建环境变量云云.

新建虚拟环境

打开命令行, 输入:

...>conda activate

则进入基础虚拟环境 (base), 环境名会出现在在命令行路径的开头
新建虚拟环境 new_env1, 则输入:

(base) ...>conda create -n new_env1

如果没有 -n 命令, 则:

(base) ...>conda create new_env1
CondaValueError: The target prefix is the base prefix. Aborting.

正常情况下, 应该显示:

Collecting package metadata (current_repodata.json): done
Solving environment: done

## Package Plan ##

  environment location: %Anaconda安装目录%\envs\tmp



Proceed ([y]/n)?

其中 %Anaconda安装目录% 指 Anaconda 的安装目录, 即 \Script 文件夹所在的目录.
输入 y, Enter 创建环境.

更改新虚拟环境的位置.

打开 C:\Users\用户名\.condarc, 其中 用户名 为当前用户的用户名.
向 .condarc 中添加如下内容:

channels:
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
  - defaults
show_channel_urls: true
envs_dirs:
  - %自定义路径%\conda\envs
  - %Anaconda安装目录%\envs
  - C:\Users\用户名\.conda\envs
  - C:\Users\用户名\AppData\Local\conda\conda\envs
pkgs_dirs:
  - %自定义路径%\conda\pkgs
  - %Anaconda安装目录%\pkgs
  - C:\Users\用户名\.conda\pkgs
  - C:\Users\用户名\AppData\Local\conda\conda\pkgs

保存.
默认的 .condarc 文件中仅有 {} 两个字符, 直接删除并粘贴上述内容即可. 不放心的话, 可以先输入

(base) ...>conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
(base) ...>conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge/
(base) ...>conda config --set show_channel_urls yes

来添加 channels 和 show_channel_urls 两个变量.
此时可以 在 base 环境下 查看 Anaconda 环境信息:

(base) ...>conda activate new_env1
(new_env1) ...>conda info

接下来, 创建另一个虚拟环境 new_env2

(new_env1) ...>conda create -n new_env2

可以发现, environment location 已经发生变化, 此时可以放心输入 y, Enter 创建环境.

查阅资料, 还可以使用 --prefix= 命令指定路径.

虚拟环境的切换

虚拟环境内外没有什么差别, 都可以使用 conda [-h] [-V] command ... 等命令, ping 等命令也可以使用, 只是 python 解释器有所不同.
值得一提的是, 每次使用

(last_env) ...>conda activate next_env

来进入下一个虚拟环境 next_env 后再使用

conda deactivate

命令退出时, 会退出到上一个虚拟环境 last_env, 类似出栈入栈.

向虚拟环境中添加包

先把试错的过程放上来

下载 pygame 安装包

先来到和 Anaconda 相同目录的环境 new_env1, 并下载 pygame

(base) ...>conda new_env1
(new_env1) ...>conda install pygame
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.

PackagesNotFoundError: The following packages are not available from current channels:

  - pygame

Current channels:

  - (URL)
  - 

To search for alternate channels that may provide the conda package you're
looking for, navigate to

    https://anaconda.org

and use the search bar at the top of the page.

(URL) 是conda的源服务器
无法下载, 只能使用 pip 命令.

(new_env1) ...>pip install pygame
Collecting pygame
  Downloading pygame-1.9.6-cp36-cp36m-win_amd64.whl (4.3 MB)

只不过网络不太好, 下载了好几次后系统才显示安装成功.
为了验证是否仅在当前虚拟环境下安装, 进入 base 环境查看:

(new_env1) ...>conda activate base
(base) pip list
Package                            Version
---------------------------------- ---------
...
pygame                             1.9.6
...

很明显, 此时 pip 仍然是全局命令.
于是删除这个包

(base) ...>pip uninstall pygame

输入 y, Enter 后成功删除.
然后在又在 new_env2, base 环境中反复尝试:

(base) ...>conda activate new_env2
(new_env2) ...>pip install pygame
Collecting pygame
  Using cached pygame-1.9.6-cp37-cp37m-win_amd64.whl (4.3 MB)
Installing collected packages: pygame
Successfully installed pygame-1.9.6

不用下载了, 但 pip 命令都是全局的. 在网上搜索众多解决方案无果.

`conda install` 命令安装包

pip 命令无效后, 先尝试用 conda install 命令安装其他包, 比如说 matplotlib, 安装在创建于自定义路径下的环境 new_env2 中:

(new_env2) ...>conda install matplotlib
Collecting package metadata (current_repodata.json): done
Solving environment: done

## Package Plan ##

  environment location: %自定义路径%\conda\envs\new_env2

  added / updated specs:
    - matplotlib


The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    certifi-2019.11.28         |           py38_0         149 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    cycler-0.10.0              |             py_2           9 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    freetype-2.10.0            |       h563cfd7_1         476 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    icu-64.2                   |       he025d50_1        14.1 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    jpeg-9c                    |    hfa6e2cd_1001         314 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    kiwisolver-1.1.0           |   py38he980bc4_0          61 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    libclang-9.0.1             |default_hf44288c_0        20.8 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    libpng-1.6.37              |       h7602738_0         1.3 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    matplotlib-3.1.3           |           py38_0           7 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    matplotlib-base-3.1.3      |   py38h2981e6d_0         6.5 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    numpy-1.17.5               |   py38hc71023c_0         4.7 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    pip-20.0.2                 |           py38_1         1.9 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    pyparsing-2.4.6            |             py_0          59 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    pyqt-5.12.3                |   py38h6538335_1         4.7 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    python-3.8.1               |       he1f5543_2        18.9 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    python-dateutil-2.8.1      |             py_0         220 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    qt-5.12.5                  |       h7ef1ec2_0       104.4 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    setuptools-45.1.0          |           py38_0         664 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    six-1.14.0                 |           py38_0          23 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    sqlite-3.30.1              |       hfa6e2cd_0         1.1 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    tornado-6.0.3              |   py38hfa6e2cd_0         641 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    vs2015_runtime-14.16.27012 |       hf0eaf9b_1         1.1 MB  defaults
    wheel-0.34.2               |           py38_0          61 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    wincertstore-0.2           |        py38_1003          13 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    zlib-1.2.11                |    h2fa13f4_1006         236 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    ------------------------------------------------------------
                                           Total:       182.5 MB

The following NEW packages will be INSTALLED:

  ca-certificates    anaconda/cloud/conda-forge/win-64::ca-certificates-2019.11.28-hecc5488_0
  certifi            anaconda/cloud/conda-forge/win-64::certifi-2019.11.28-py38_0
  cycler             anaconda/cloud/conda-forge/noarch::cycler-0.10.0-py_2
  freetype           anaconda/cloud/conda-forge/win-64::freetype-2.10.0-h563cfd7_1
  icu                anaconda/cloud/conda-forge/win-64::icu-64.2-he025d50_1
  intel-openmp       pkgs/main/win-64::intel-openmp-2019.4-245
  jpeg               anaconda/cloud/conda-forge/win-64::jpeg-9c-hfa6e2cd_1001
  kiwisolver         anaconda/cloud/conda-forge/win-64::kiwisolver-1.1.0-py38he980bc4_0
  libblas            anaconda/cloud/conda-forge/win-64::libblas-3.8.0-14_mkl
  libcblas           anaconda/cloud/conda-forge/win-64::libcblas-3.8.0-14_mkl
  libclang           anaconda/cloud/conda-forge/win-64::libclang-9.0.1-default_hf44288c_0
  liblapack          anaconda/cloud/conda-forge/win-64::liblapack-3.8.0-14_mkl
  libpng             anaconda/cloud/conda-forge/win-64::libpng-1.6.37-h7602738_0
  matplotlib         anaconda/cloud/conda-forge/win-64::matplotlib-3.1.3-py38_0
  matplotlib-base    anaconda/cloud/conda-forge/win-64::matplotlib-base-3.1.3-py38h2981e6d_0
  mkl                pkgs/main/win-64::mkl-2019.4-245
  numpy              anaconda/cloud/conda-forge/win-64::numpy-1.17.5-py38hc71023c_0
  openssl            anaconda/cloud/conda-forge/win-64::openssl-1.1.1d-hfa6e2cd_0
  pip                anaconda/cloud/conda-forge/win-64::pip-20.0.2-py38_1
  pyparsing          anaconda/cloud/conda-forge/noarch::pyparsing-2.4.6-py_0
  pyqt               anaconda/cloud/conda-forge/win-64::pyqt-5.12.3-py38h6538335_1
  python             anaconda/cloud/conda-forge/win-64::python-3.8.1-he1f5543_2
  python-dateutil    anaconda/cloud/conda-forge/noarch::python-dateutil-2.8.1-py_0
  qt                 anaconda/cloud/conda-forge/win-64::qt-5.12.5-h7ef1ec2_0
  setuptools         anaconda/cloud/conda-forge/win-64::setuptools-45.1.0-py38_0
  six                anaconda/cloud/conda-forge/win-64::six-1.14.0-py38_0
  sqlite             anaconda/cloud/conda-forge/win-64::sqlite-3.30.1-hfa6e2cd_0
  tornado            anaconda/cloud/conda-forge/win-64::tornado-6.0.3-py38hfa6e2cd_0
  vc                 pkgs/main/win-64::vc-14.1-h0510ff6_4
  vs2015_runtime     pkgs/main/win-64::vs2015_runtime-14.16.27012-hf0eaf9b_1
  wheel              anaconda/cloud/conda-forge/win-64::wheel-0.34.2-py38_0
  wincertstore       anaconda/cloud/conda-forge/win-64::wincertstore-0.2-py38_1003
  zlib               anaconda/cloud/conda-forge/win-64::zlib-1.2.11-h2fa13f4_1006


Proceed ([y]/n)?

输入 Enter 以下载.
奇怪, 居然要下载 python 和 pip, 难道之前一直在用 base 环境下的 python 解释器? 那就说的通了. 打开 new_env2 的安装目录, 的确多了不少东西, 而 new_env1 的安装目录中空空荡荡, 我的判断标准是: 没有 python.exe 程序文件.
那么再试试下载 pygame:

(new_env2) ...>pip install pygame
Collecting pygame
  Downloading pygame-1.9.6-cp36-cp36m-win_amd64.whl (4.3 MB)

发现又需要重新从网上下载, 网络状况不好, 下载中断, 安装失败.
那么回到 new_env1 再试试, 这次下载 python:

  (new_env2) ...>conda activate new_env1
  (new_env1) ...>conda install conda install python
Collecting package metadata (current_repodata.json): done
Solving environment: done

## Package Plan ##

  environment location: %自定义路径%\conda\envs\new_env1

  added / updated specs:
    - python


The following packages will be downloaded:

    package                    |            build
    ---------------------------|-----------------
    certifi-2019.11.28         |           py38_0         149 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    pip-20.0.2                 |           py38_1         1.9 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    python-3.8.1               |       he1f5543_2        18.9 MB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    setuptools-45.1.0          |           py38_0         664 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    vs2015_runtime-14.16.27012 |       hf0eaf9b_1         1.1 MB  defaults
    wheel-0.34.2               |           py38_0          61 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    wincertstore-0.2           |        py38_1003          13 KB  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge
    ------------------------------------------------------------
                                           Total:        22.8 MB

The following NEW packages will be INSTALLED:

  ca-certificates    anaconda/cloud/conda-forge/win-64::ca-certificates-2019.11.28-hecc5488_0
  certifi            anaconda/cloud/conda-forge/win-64::certifi-2019.11.28-py38_0
  openssl            anaconda/cloud/conda-forge/win-64::openssl-1.1.1d-hfa6e2cd_0
  pip                anaconda/cloud/conda-forge/win-64::pip-20.0.2-py38_1
  python             anaconda/cloud/conda-forge/win-64::python-3.8.1-he1f5543_2
  setuptools         anaconda/cloud/conda-forge/win-64::setuptools-45.1.0-py38_0
  sqlite             anaconda/cloud/conda-forge/win-64::sqlite-3.30.1-hfa6e2cd_0
  vc                 pkgs/main/win-64::vc-14.1-h0510ff6_4
  vs2015_runtime     pkgs/main/win-64::vs2015_runtime-14.16.27012-hf0eaf9b_1
  wheel              anaconda/cloud/conda-forge/win-64::wheel-0.34.2-py38_0
  wincertstore       anaconda/cloud/conda-forge/win-64::wincertstore-0.2-py38_1003


Proceed ([y]/n)?


Downloading and Extracting Packages
pip-20.0.2           | 1.9 MB    | ######################################################################### | 100%
python-3.8.1         | 18.9 MB   | ######################################################################### | 100%
wincertstore-0.2     | 13 KB     | ######################################################################### | 100%
setuptools-45.1.0    | 664 KB    | ######################################################################### | 100%
wheel-0.34.2         | 61 KB     | ######################################################################### | 100%
certifi-2019.11.28   | 149 KB    | ######################################################################### | 100%
vs2015_runtime-14.16 | 1.1 MB    | ######################################################################### | 100%
Preparing transaction: done
Verifying transaction: done
Executing transaction: done

(new_env1) ...>pip install pygame
Collecting pygame
  Using cached pygame-1.9.6-cp37-cp37m-win_amd64.whl (4.3 MB)
Installing collected packages: pygame
Successfully installed pygame-1.9.6

进入其他环境查看,

(new_env1) ...>conda activate base
(base) ...>pip list

就没有发现 pygame 了.

通过已下载的包安装

还不会.

复制虚拟环境

后来发现, 完全可以用克隆的方式新建一个虚拟环境, 这样也免得要从头下载了

(base) ...>conda create -n new_env3 --clone base
Source:      %Anaconda安装目录%
Destination: %自定义路径%\conda\envs\mylibrary
The following packages cannot be cloned out of the root environment:
 - defaults/win-64::conda-4.8.1-py37_0
 - defaults/win-64::conda-build-3.18.9-py37_3
 - defaults/win-64::conda-env-2.6.0-1
Packages: 275
Files: 508
Preparing transaction: done
Verifying transaction: done
Executing transaction: done

方便了许多.

另一种复制虚拟环境的方式是使用 --file 参数, 其后输入一个文件名

输入:

(base)...> conda create -n new_env3

即可

删除虚拟环境

为了保证系统盘的空间充裕, 删除 new_env1 环境:

(base) ...>conda remove -n new_env1 --all

Remove all packages in environment %Anaconda安装目录%\envs\new_env1:


## Package Plan ##

  environment location: %Anaconda安装目录%\envs\new_env1


The following packages will be REMOVED:

  ...


Proceed ([y]/n)? y

Preparing transaction: done
Verifying transaction: done
Executing transaction: done