M1开发深度学习的环境搭建

金玉良缘易配而木石前盟难得|M1 Mac os(Apple Silicon)天生一对Python3开发环境搭建(集成深度学习框架Tensorflow/Pytorch)

首页 - Mac & Linux / 2021-03-04

微信扫一扫：分享

    笔者投入M1的怀抱已经有一段时间了，俗话说得好，但闻新人笑，不见旧人哭，Intel mac早已被束之高阁，而M1 mac已经不能用真香来形容了，简直就是“香透满堂金玉彩，扇遮半面桃花开！”，轻抚M1 mac那滑若柔荑的秒控键盘，别说996了，就是007，我们也能安之若素，也可以笑慰平生。好了，日常吹M1的环节结束，正所谓剑虽利,不厉不断，材虽美，不学不高。本次我们尝试在M1 Mac os 中搭建Python3的开发环境。

    一般情况下，直接Python官网(python.org)下载最新的基于arm架构的python3.9即可，但是由于向下兼容等问题，我们尝试使用Python多版本管理软件conda，conda在业界有三大巨头分别是:Anaconda、Miniconda以及Condaforge，虽然都放出消息要适配M1芯片，但是目前最先放出稳定版的是Condaforge，进入下载页面：https://github.com/conda-forge/miniforge/#download  选择mac arm64位架构:

    该文件并不是安装包，而是一个shell脚本，下载成功后，进入命令行目录:

复制

cd ~/Downloads

    执行命令进行安装：

复制

sudo bash ./Miniforge3-MacOSX-arm64.sh

    随后会有一些条款需要确认，这里按回车之后键入yes:

复制

Welcome to Miniforge3 4.9.2-7

In order to continue the installation process, please review the license
agreement.
Please, press ENTER to continue
>>> 
BSD 3-clause license
Copyright (c) 2019-2020, conda-forge
All rights reserved.

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:

1. Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.

2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.

3. Neither the name of the copyright holder nor the names of its contributors
may be used to endorse or promote products derived from this software without
specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.


Do you accept the license terms? [yes|no]
[no] >>> yes

    安装的默认版本还是3.9，会附带安装35个基础库，这样就不用我们自己手动安装了：

复制

brotlipy                    0.7.0  py39h46acfd9_1001   installed         
  bzip2                       1.0.8  h27ca646_4          installed         
  ca-certificates         2020.12.5  h4653dfc_0          installed         
  certifi                 2020.12.5  py39h2804cbe_1      installed         
  cffi                       1.14.5  py39h702c04f_0      installed         
  chardet                     4.0.0  py39h2804cbe_1      installed         
  conda                       4.9.2  py39h2804cbe_0      installed         
  conda-package-handling      1.7.2  py39h51e6412_0      installed         
  cryptography                3.4.4  py39h6e07874_0      installed         
  idna                         2.10  pyh9f0ad1d_0        installed         
  libcxx                     11.0.1  h168391b_0          installed         
  libffi                        3.3  h9f76cd9_2          installed         
  ncurses                       6.2  h9aa5885_4          installed         
  openssl                    1.1.1j  h27ca646_0          installed         
  pip                        21.0.1  pyhd8ed1ab_0        installed         
  pycosat                     0.6.3  py39h46acfd9_1006   installed         
  pycparser                    2.20  pyh9f0ad1d_2        installed         
  pyopenssl                  20.0.1  pyhd8ed1ab_0        installed         
  pysocks                     1.7.1  py39h2804cbe_3      installed         
  python                      3.9.2  hcbd9b3a_0_cpython  installed         
  python_abi                    3.9  1_cp39              installed         
  readline                      8.0  hc8eb9b7_2          installed         
  requests                   2.25.1  pyhd3deb0d_0        installed         
  ruamel_yaml               0.15.80  py39h46acfd9_1004   installed         
  setuptools                 49.6.0  py39h2804cbe_3      installed         
  six                        1.15.0  pyh9f0ad1d_0        installed         
  sqlite                     3.34.0  h6d56c25_0          installed         
  tk                         8.6.10  hf7e6567_1          installed         
  tqdm                       4.57.0  pyhd8ed1ab_0        installed         
  tzdata                      2021a  he74cb21_0          installed         
  urllib3                    1.26.3  pyhd8ed1ab_0        installed         
  wheel                      0.36.2  pyhd3deb0d_0        installed         
  xz                          5.2.5  h642e427_1          installed         
  yaml                        0.2.5  h642e427_0          installed         
  zlib                       1.2.11  h31e879b_1009       installed

    然后编辑配置文件vim  ~/.zshrc，加入如下内容(此处liuyue是笔者用户名，需改成你的Mac当前用户名)：

复制

path=('/Users/liuyue/miniforge3/bin' $path)
export PATH

    存盘之后执行命令：

复制

source ~/.zshrc

    配置好环境变量之后，键入python3：

复制

➜  ~ python3
Python 3.9.2 | packaged by conda-forge | (default, Feb 21 2021, 05:00:30) 
[Clang 11.0.1 ] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>>

    可以看到已经使用conda安装的python版本了。

    这里简单介绍一下conda命令：

    conda info 可以查看当前conda的基本信息内核，平台，下载源以及目录位置：

复制

➜  ~ conda info

     active environment : None
       user config file : /Users/liuyue/.condarc
 populated config files : /Users/liuyue/miniforge3/.condarc
          conda version : 4.9.2
    conda-build version : not installed
         python version : 3.9.2.final.0
       virtual packages : __osx=11.2.2=0
                          __unix=0=0
                          __archspec=1=arm64
       base environment : /Users/liuyue/miniforge3  (read only)
           channel URLs : https://conda.anaconda.org/conda-forge/osx-arm64
                          https://conda.anaconda.org/conda-forge/noarch
          package cache : /Users/liuyue/miniforge3/pkgs
                          /Users/liuyue/.conda/pkgs
       envs directories : /Users/liuyue/.conda/envs
                          /Users/liuyue/miniforge3/envs
               platform : osx-arm64
             user-agent : conda/4.9.2 requests/2.25.1 CPython/3.9.2 Darwin/20.3.0 OSX/11.2.2
                UID:GID : 502:20
             netrc file : None
           offline mode : False

    由于一些众所周知的学术问题，我们需要配置一下国内下载源：

复制

conda config --add channels https://mirrors.ustc.edu.cn/anaconda/pkgs/main/
conda config --add channels https://mirrors.ustc.edu.cn/anaconda/pkgs/free/
conda config --add channels https://mirrors.ustc.edu.cn/anaconda/cloud/conda-forge/
conda config --add channels https://mirrors.ustc.edu.cn/anaconda/cloud/msys2/
conda config --add channels https://mirrors.ustc.edu.cn/anaconda/cloud/bioconda/
conda config --add channels https://mirrors.ustc.edu.cn/anaconda/cloud/menpo/

conda config --set show_channel_urls yes

    随后查看当前下载源：

复制

conda config --show

    可以看到国内源已经被添加进去了：

复制

channel_priority: flexible
channels:
  - https://mirrors.ustc.edu.cn/anaconda/cloud/menpo/
  - https://mirrors.ustc.edu.cn/anaconda/cloud/bioconda/
  - https://mirrors.ustc.edu.cn/anaconda/cloud/msys2/
  - https://mirrors.ustc.edu.cn/anaconda/cloud/conda-forge/
  - https://mirrors.ustc.edu.cn/anaconda/pkgs/free/
  - https://mirrors.ustc.edu.cn/anaconda/pkgs/main/
  - defaults
  - conda-forge
client_ssl_cert: None

    其他的一些conda常用命令：

1. conda --version #查看conda版本，验证是否安装

2. conda update conda #更新至最新版本，也会更新其它相关包

3. conda update --all #更新所有包

4. conda update packag