[PYTHON] How to quickly create a machine learning environment using Jupyter Notebook on macOS Sierra with anaconda

Introduction

I wrote a similar article about Ubuntu Server before, but it was easy on macOS, so I will organize the procedure as well.

[Ubuntu Server Edition] http://qiita.com/mix_dvd/items/29dfb8d47a596b4df36d

Install Git

Check the version.

$ git --version

If Git is not installed, the screen asking for installation will be displayed as shown below, so select "Install".

If it is installed, it will be displayed as follows for macOS Sierra.

git version 2.8.4 (Apple Git-73)

Install pyenv

https://github.com/yyuu/pyenv

pyenv is a tool for installing multiple versions of Python in one environment.

$ git clone https://github.com/yyuu/pyenv.git ~/.pyenv
$ echo 'export PYENV_ROOT="$HOME/.pyenv"' >> ~/.bashrc
$ echo 'export PATH="$PYENV_ROOT/bin:$PATH"' >> ~/.bashrc
$ echo 'eval "$(pyenv init -)"' >> ~/.bashrc
$ exit

Restart the terminal and check the version.

$ pyenv -v
pyenv 20160629-19-ga761ff6

• If pyenv does not work, create a file called ".bash_profile" in your home directory so that ".bashrc" will be read when you start the terminal.

$ vi .bash_profile

When the editor starts, press the "i" key to copy and paste the following code, press the "esc" key, enter ": wq" and press the "enter" key to save the file and exit the editor.

.bashrc

if [ -f ~/.bashrc ] ; then
. ~/.bashrc
fi

Installation of Anaconda

https://www.continuum.io/why-anaconda

Version confirmation

Check the version of Anaconda that can be installed.

$ pyenv install -l | grep anaconda

As of July 16, 2016, the latest version was 4.1.0. Since there are many differences between version 2.x and version 3.x of Python, please select one of the following according to your environment. Unless otherwise specified, I think you can choose the 3.x series.

Python 2 series

$ pyenv install anaconda2-4.1.0
$ pyenv rehash
$ pyenv global anaconda2-4.1.0
$ echo 'export PATH="$PYENV_ROOT/versions/anaconda2-4.1.0/bin/:$PATH"' >> ~/.bashrc
$ exit

Python 3 series

$ pyenv install anaconda3-4.1.0
$ pyenv rehash
$ pyenv global anaconda3-4.1.0
$ echo 'export PATH="$PYENV_ROOT/versions/anaconda3-4.1.0/bin/:$PATH"' >> ~/.bashrc
$ exit

Restart the terminal and check the Python version.

$ python -V
Python 3.5.1 :: Anaconda 4.1.0 (x86_64)

Seaborn installation

Headquarters site https://web.stanford.edu/~mwaskom/software/seaborn/

$ pip install seaborn

Install XGBoost

Headquarters site https://xgboost.readthedocs.io/

Reference URL https://xgboost.readthedocs.io/en/latest/build.html#python-package-installation

Installation instructions for Mac https://github.com/dmlc/xgboost/blob/master/doc/build.md#building-on-osx

If Xcode is not installed, install Xcode.

After installing Xcode, execute the following command.

$ git clone --recursive https://github.com/dmlc/xgboost
$ cd xgboost
$ cp make/minimum.mk ./config.mk
$ make -j4
$ cd python-package
$ python setup.py install

Japanese environment settings for matplotlib

Download the font file "IPAfont00303.zip (19.1 MB)" of 4 typeface pack (Ver.003.03) from the following site.

http://ipafont.ipa.go.jp/old/ipafont/download.html

After downloading, execute the following command.

$ cd Download
$ unzip IPAfont00303.zip
$ mv IPAfont00303/*.ttf ~/.pyenv/versions/anaconda3-4.1.0/lib/python3.5/site-packages/matplotlib/mpl-data/fonts/ttf/
$ rm ~/.matplotlib/fontList.*.cache 

Jupyter Notebook

Headquarters site http://jupyter.org

Execute the following command in the home directory of the initial user.

$ jupyter notebook

The default browser should start automatically and you should see the Jupyter screen.

However, if you start it in your home directory, you will see various extra things and there is a risk of erroneous operation, so it is recommended to create a working directory and use it.

$ mkdir notebook
$ cd notebook
$ jupiter notebook

Thank you for your support.