The most recent update of Apple’s OS X 10.10 Yosemite makes its mark with a bit of splendour. But, under the hood, not too much has changed. The same Unix shell, the same way to perform our daily programming at the same level. Setting up Python in Yosemite appears today to be easier than ever (compare my post on performing the same task in OS X 10.9 Mavericks a year ago).
Download and Installation
The evolution of Continuum Analytics with its unique mission to deliver Python to everyone started to pay off. They created Anaconda, a completely free enterprise-ready Python distribution for large-scale data processing, predictive analytics, and scientific computing. It’s free including its application for commercial purposes. You can get a working version for Yosemite (as well for Linux and Microsoft Windows). Just go to:
fill in your email details, and choose next with grand version (2.7 vs 3.4) you prefer,
and download the .dmg file to begin the installation on your Mac in a graphical mode.
It installs into a single directory and does not affect other Python installations on your system. No admin/root privileges are required too. You can switch among Python interpreters in versions of 2.6, 2.7, 3.3, and 3.4. After installation you will get an instant access to 195+ of the most popular Python packages for science, math, engineering, data analysis including NumPy, SciPy, Pandas, IPython, Matplotlib, Numba, Blaze, and others.
Don’t rush with payment for an additional Anaconda Accelerate package making use of GPUs. As I will address this thread in my next post, Python using CUDA within Numbapro module offered by Continuum Analytics disappoints.