I tried to build an environment for machine learning with Python (Mac OS X)

At the CROSS 2015 python session the other day, I thought that machine learning would be interesting, so I tried to prepare the environment for the time being, so I'm sorry.

The other day's article on CROSS 2015 is here.

: black_square_button: Overview

--Building an environment where machine learning can be done with python ――I haven't done much, so I thought it was necessary to give it to Qiita, but I personally got into various things, so I will give it to you.

: black_square_button: Premise

• MacOS X

: black_square_button: Various installations

• Python

$ brew update
$ brew upgrade
$ brew install python

For the time being, check the version

$ python -V 
Python 2.7.5

pip

$ pip install --upgrade setuptools
$ pip install --upgrade pip

Set the path in .bash_profile. I forgot. .. If you don't do this, you will get the following error when importing scikit-learn.

ImportError: No module named sklearn

So add the path to .bash_profile.

export PATH=/usr/local/bin:$PATH
export PATH=/usr/local/share/python:$PATH

Update after adding the above to .bash_profile.

source ~/.bash_profile

--Install various math libraries

$ pip install numpy
$ pip install scipy
$ pip install matplotlib
$ pip install scikit-learn
$ pip install pandas
$ pip install pydot
$ pip install rpy2
$ pip install ipython
$ pip install pyzmq
$ pip install tornado
$ pip install jinja2

I ran into the following error when installing rpy2.

$ pip install rpy2
Collecting rpy2
  Using cached rpy2-2.5.5.tar.gz
    Error: Tried to guess R's HOME but no R command in the PATH.
    Complete output from command python setup.py egg_info:
    Error: Tried to guess R's HOME but no R command in the PATH.

    ----------------------------------------
    Command "python setup.py egg_info" failed with error code 1 in /private/var/folders/jl/h7wk98516f1_tj7zg87r08b00000gn/T/pip-build-Cs7fDa/rpy2

You don't have R installed. .. That's right. ..
Refer to the following for how to install R on MacOSX. Thank you. Thank you.
Install R with homebrew on MacOSX I was quite addicted to it. ..

: black_square_button: Miscellaneous feelings

――I feel like I somehow understand why python is attracting attention in machine learning. ――Because there are abundant libraries in mathematics, engineering, and statistics? ――Is there so much Ruby? I can't say anything because I haven't investigated it properly. .. --It's great to be able to call R-related libraries with rpy2 and pandas ――Is it a feature of Python that there are Scipy and scikit-learn? --This completes the environment for boxplots, histograms, regression analysis, cluster analysis, etc. ――By the way, I remembered that I was doing statistical things when I was a student, and I felt a little nostalgic. .. .. ――If you have a good library other than the above, please let me know. ――After all, I think it wasn't enough to give to Qiita. ..

: black_square_button: Reference

-I put ipython in Mavericks -Install R with homebrew on MacOSX