前言

距上一次搭建树莓派平台已经两年了，保存的镜像出了问题，重新搭建一下。

系统

下载

从官网下载树莓派系统镜像，有以下几种可选

Raspberry Pi — Teach, Learn, and Make with Raspberry Pi

  1. Raspbian & Raspbian Lite，基于Debian
  2. Noobs & Noobs Lite
  3. Ubuntu MATE
  4. Snappy Ubuntu Core
  5. Windows 10 IOT

其余不太了解，之前安装的是Raspbian，对于Debian各种不适，换上界面优雅的Ubuntu Mate玩一下
老老实实玩Raspbian，笑脸:-)

安装

比较简单，准备micro-SD卡，用Win32 Disk Imager烧写镜像

Win32 Disk Imager download | SourceForge.net

Win32DiskImager

安装完软件后可点击Read备份自己的镜像。

注意第二次开机前需要配置config.txt文件，否则hdmi无法显示

树莓派配置文档 config.txt 说明 | 树莓派实验室

1
2
3
4
5
6
disable_overscan=1 
hdmi_force_hotplug=1
hdmi_group=2    # DMT
hdmi_mode=32    # 1280x960
hdmi_drive=2
config_hdmi_boost=4

修改交换分区

Ubuntu Mate

查看交换分区

1
$ free -m

未设置时如下

1
2
3
4
total     used     free   shared  buffers   cached
Mem:           435       56      379        0        3       16
-/+ buffers/cache:       35      399
Swap:            0        0        0

创建和挂载

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
# 获取权限
$ sudo -i

# 创建目录
$ mkdir /swap
$ cd /swap

# 指定一个大小为1G的名为“swap”的交换文件
$ dd if=/dev/zero of=swap bs=1M count=1k
# 创建交换文件
$ mkswap swap
# 挂载交换分区
$ swapon swap

# 卸载交换分区
# $ swapoff swap

查看交换分区

1
$ free -m

未设置时如下

1
2
3
4
total     used     free   shared  buffers   cached
Mem:           435       56      379        0        3       16
-/+ buffers/cache:       35      399
Swap:         1023        0     1023

Raspbian

We will change the configuration in the file /etc/dphys-swapfile:

1
$ sudo nano /etc/dphys-swapfile

The default value in Raspbian is:

1
CONF_SWAPSIZE=100

We will need to change this to:

1
CONF_SWAPSIZE=1024

Then you will need to stop and start the service that manages the swapfile own Rasbian:

1
2
$ sudo /etc/init.d/dphys-swapfile stop
$ sudo /etc/init.d/dphys-swapfile start

You can then verify the amount of memory + swap by issuing the following command:

1
$ free -m

The output should look like:

1
2
3
4
total     used     free   shared  buffers   cached
Mem:           435       56      379        0        3       16
-/+ buffers/cache:       35      399
Swap:         1023        0     1023

软件

安装指令

  • apt-get

    • 安装软件
      apt-get install softname1 softname2 softname3 ...
    • 卸载软件
      apt-get remove softname1 softname2 softname3 ...
    • 卸载并清除配置
      apt-get remove --purge softname1
    • 更新软件信息数据库
      apt-get update
    • 进行系统升级
      apt-get upgrade
    • 搜索软件包
      apt-cache search softname1 softname2 softname3 ...
    • 修正（依赖关系）安装：
      apt-get -f insta

  • dpkg

    • 安装.deb软件包
      dpkg -i xxx.deb

    • 删除软件包
      dpkg -r xxx.deb

    • 连同配置文件一起删除
      dpkg -r --purge xxx.deb

    • 查看软件包信息
      dpkg -info xxx.deb

    • 查看文件拷贝详情
      dpkg -L xxx.deb

    • 查看系统中已安装软件包信息
      dpkg -l

    • 重新配置软件包
      dpkg-reconfigure xx

    • 卸载软件包及其配置文件，但无法解决依赖关系！
      sudo dpkg -p package_name

    • 卸载软件包及其配置文件与依赖关系包
      sudo aptitude purge pkgname

    • 清除所有已删除包的残馀配置文件
      dpkg -l |grep ^rc|awk '{print $2}' |sudo xargs dpkg -P

软件源

  1. 备份原始文件

    1
    $ sudo cp /etc/apt/sources.list /etc/apt/sources.list.backup

  2. 修改文件并添加国内源

    1
    $ vi /etc/apt/sources.list

  3. 注释元文件内的源并添加如下地址

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    #Mirror.lupaworld.com 源更新服务器（浙江省杭州市双线服务器，网通同电信都可以用，亚洲地区官方更新服务器）：
    deb http://mirror.lupaworld.com/ubuntu gutsy main restricted universe multiverse
    deb http://mirror.lupaworld.com/ubuntu gutsy-security main restricted universe multiverse
    deb http://mirror.lupaworld.com/ubuntu gutsy-updates main restricted universe multiverse
    deb http://mirror.lupaworld.com/ubuntu gutsy-backports main restricted universe multiverse
    deb-src http://mirror.lupaworld.com/ubuntu gutsy main restricted universe multiverse
    deb-src http://mirror.lupaworld.com/ubuntu gutsy-security main restricted universe multiverse
    deb-src http://mirror.lupaworld.com/ubuntu gutsy-updates main restricted universe multiverse
    deb-src http://mirror.lupaworld.com/ubuntu gutsy-backports main restricted universe multiverse

    #Ubuntu 官方源 
    deb http://archive.ubuntu.com/ubuntu/ gutsy main restricted universe multiverse
    deb http://archive.ubuntu.com/ubuntu/ gutsy-security main restricted universe multiverse
    deb http://archive.ubuntu.com/ubuntu/ gutsy-updates main restricted universe multiverse
    deb http://archive.ubuntu.com/ubuntu/ gutsy-proposed main restricted universe multiverse
    deb http://archive.ubuntu.com/ubuntu/ gutsy-backports main restricted universe multiverse
    deb-src http://archive.ubuntu.com/ubuntu/ gutsy main restricted universe multiverse
    deb-src http://archive.ubuntu.com/ubuntu/ gutsy-security main restricted universe multiverse
    deb-src http://archive.ubuntu.com/ubuntu/ gutsy-updates main restricted universe multiverse
    deb-src http://archive.ubuntu.com/ubuntu/ gutsy-proposed main restricted universe multiverse
    deb-src http://archive.ubuntu.com/ubuntu/ gutsy-backports main restricted universe multiverse

    或者

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    22
    23
    #阿里云
    deb http://mirrors.aliyun.com/ubuntu/ trusty main restricted universe multiverse
    deb http://mirrors.aliyun.com/ubuntu/ trusty-security main restricted universe multiverse
    deb http://mirrors.aliyun.com/ubuntu/ trusty-updates main restricted universe multiverse
    deb http://mirrors.aliyun.com/ubuntu/ trusty-proposed main restricted universe multiverse
    deb http://mirrors.aliyun.com/ubuntu/ trusty-backports main restricted universe multiverse
    deb-src http://mirrors.aliyun.com/ubuntu/ trusty main restricted universe multiverse
    deb-src http://mirrors.aliyun.com/ubuntu/ trusty-security main restricted universe multiverse
    deb-src http://mirrors.aliyun.com/ubuntu/ trusty-updates main restricted universe multiverse
    deb-src http://mirrors.aliyun.com/ubuntu/ trusty-proposed main restricted universe multiverse
    deb-src http://mirrors.aliyun.com/ubuntu/ trusty-backports main restricted universe multiverse

    #网易163
    deb http://mirrors.163.com/ubuntu/ trusty main restricted universe multiverse
    deb http://mirrors.163.com/ubuntu/ trusty-security main restricted universe multiverse
    deb http://mirrors.163.com/ubuntu/ trusty-updates main restricted universe multiverse
    deb http://mirrors.163.com/ubuntu/ trusty-proposed main restricted universe multiverse
    deb http://mirrors.163.com/ubuntu/ trusty-backports main restricted universe multiverse
    deb-src http://mirrors.163.com/ubuntu/ trusty main restricted universe multiverse
    deb-src http://mirrors.163.com/ubuntu/ trusty-security main restricted universe multiverse
    deb-src http://mirrors.163.com/ubuntu/ trusty-updates main restricted universe multiverse
    deb-src http://mirrors.163.com/ubuntu/ trusty-proposed main restricted universe multiverse
    deb-src http://mirrors.163.com/ubuntu/ trusty-backports main restricted universe multiverse

  4. 放置非官方源的包不完整，可在为不添加官方源

    1
    deb http://archive.ubuntu.org.cn/ubuntu-cn/ feisty main restricted universe multiverse

  5. 更新源

    1
    $ sudo apt-get update

  6. 更新软件

    1
    $ sudo apt-get dist-upgrade

  7. 常见的修复安装命令

    1
    $ sudo apt-get -f install

Python

主要是Python和相关依赖包的安装，使用以下指令可导出已安装的依赖包

1
$ pip freeze > requirements.txt

并使用指令安装到树莓派

1
$ pip install -r requirements.txt

注意pip更新

1
python -m pip install --upgrade pip

最新版本会报错

1
ImportError: cannot import name main

修改文件/usr/bin/pip

1
2
3
from pip import main
if __name__ == '__main__':
    sys.exit(main())

改为

1
2
3
from pip import __main__
if __name__ == '__main__':
    sys.exit(__main__._main())

成功!!!
失败了，笑脸:-)，手动安装吧。。。

  • 部分包可使用pip3

    1
    2
    3
    $ pip3 install numpy
    $ pip3 install pandas
    $ pip3 install sklearn

    若需要权限，加入--user

  • 部分包用apt-get，但是优先安装到Python2.7版本，笑脸:-)

    1
    2
    3
    $ sudo apt-get install python-scipy
    $ sudo apt-get install python-matplotlib
    $ sudo apt-get install python-opencv

  • 部分从PIPY下载.whl或.tar.gz文件

    PyPI – the Python Package Index · PyPI

    • tensorboardX-1.4-py2.py3-none-any.whl
    • visdom-0.1.8.5.tar.gz

    安装指令为

    1
    $ pip3 install xxx.whl
    1
    2
    $ tar -zxvf xxx.tar.gz
    $ python setup.py install

  • Pytorch源码安装

    pytorch/pytorch: Tensors and Dynamic neural networks in Python with strong GPU acceleration

    安装方法Installation - From Source

    需要用到miniconda，安装方法如下，注意中间回车按慢一点，有两次输入。。。。。(行我慢慢看条款不行么。。笑脸:-))

    • 第一次是是否同意条款，yes
    • 第二次是添加到环境变量，yes，否则自己修改/home/pi/.bashrc添加到环境变量
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    $ wget http://repo.continuum.io/miniconda/Miniconda3-latest-Linux-armv7l.sh
    $ sudo md5sum Miniconda3-latest-Linux-armv7l.sh # (optional) check md5
    $ sudo /bin/bash Miniconda3-latest-Linux-armv7l.sh 
    # -> change default directory to /home/pi/miniconda3
    $ sudo nano /home/pi/.bashrc 
    # -> add: export PATH="/home/pi/miniconda3/bin:$PATH"
    $ sudo reboot -h now

    $ conda 
    $ python --version
    $ sudo chown -R pi miniconda3

    然后就可以安装了没有对应版本的mkl，笑脸:-)

    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    export CMAKE_PREFIX_PATH="$(dirname $(which conda))/../" # [anaconda root directory]

    # Disable CUDA
    export NO_CUDA=1

    # Install basic dependencies
    conda install numpy pyyaml mkl mkl-include setuptools cmake cffi typing
    conda install -c mingfeima mkldnn

    # Install Pytorch
    git clone --recursive https://github.com/pytorch/pytorch
    cd pytorch
    python setup.py install

  • tensorflow
    安装tensorflow需要的一些依赖和工具

    1
    2
    3
    4
    5
    6
    7
    $ sudo apt-get update

    # For Python 2.7
    $ sudo apt-get install python-pip python-dev

    # For Python 3.3+
    $ sudo apt-get install python3-pip python3-dev

    安装tensorflow

    若下载失败，手动打开下面网页下载.whl包

    1
    2
    3
    4
    5
    6
    7
    # For Python 2.7
    $ wget https://github.com/samjabrahams/tensorflow-on-raspberry-pi/releases/download/v1.1.0/tensorflow-1.1.0-cp27-none-linux_armv7l.whl
    $ sudo pip install tensorflow-1.1.0-cp27-none-linux_armv7l.whl

    # For Python 3.4
    $ wget https://github.com/samjabrahams/tensorflow-on-raspberry-pi/releases/download/v1.1.0/tensorflow-1.1.0-cp34-cp34m-linux_armv7l.whl
    $ sudo pip3 install tensorflow-1.1.0-cp34-cp34m-linux_armv7l.whl

    卸载，重装mock

    1
    2
    3
    4
    5
    6
    7
    # For Python 2.7
    $ sudo pip uninstall mock
    $ sudo pip install mock

    # For Python 3.3+
    $ sudo pip3 uninstall mock
    $ sudo pip3 install mock

    安装的版本tensorflow v1.1.0没有models，因为1.0版本以后models就被Sam Abrahams独立出来了，例如classify_image.py就在models/tutorials/image/imagenet/里

    tensorflow/models

其余

  1. 输入法

    1
    2
    $ sudo apt-get install fcitx fcitx-googlepinyin 
    $ fcitx-module-cloudpinyin fcitx-sunpinyin

  2. git

    1
    $ sudo apt-get install git

    配置git和ssh

    1
    2
    3
    4
    5
    $ git config --global user.name "Louis Hsu"
    $ git config --global user.email is.louishsu@foxmail.com

    $ ssh-keygen -t rsa -C "is.louishsu@foxmail.com"
    $ cat ~/.ssh/id_rsa.pub  # 添加到github