What is the Intel Distribution for Python and how can it improve data analysis?

[Jamison Lahman]

Hello,
I am relatively new to Python. I would really like to use Python for numerical and data analysis. I recently stumbled upon something called Intel Distribution from this article: http://www.infoworld.com/article/31...n-distribution-turbocharges-data-science.html
I don't really know what a distribution is in this context. It sounds promising and I would like to hear if anyone has more knowledge on this. I have an AMD CPU and MSI GPU on my tower which I hope to replace soon for an Intel chip and Nvidia GPU. Will I be able to utilize this Intel Distribution in the meantime?

Thanks

Edit: I already have anaconda installed (though I'm not sure I have it configured correctly.
when I run $ python, it gives me "Python 3.6.1 |Continuum Analytics, Inc.| (default, May 11 2017, 13:08:54)" so I imagine it's working. I am not yet familiar with applications like SciPy or NumPy so I haven't tested those. How should I go about checking if I have this Intel Distribution or installing via conda?

 

[phinds]

A "distribution" is just the installation package. It's how a company distributes its software, or in this case it's version of a software type. Lots of companies distribute Python.

 

[jedishrfu]

The quantecon website has some lectures on Python and the anaconda distro from Comtinuum

https://lectures.quantecon.org/py/getting_started.html

 

[Jamison Lahman]

A "distribution" is just the installation package. It's how a company distributes its software, or in this case it's version of a software type. Lots of companies distribute Python.

So Anaconda is a distribution from Continuum Analytics and is analogous to this Intel Distribution?
I am sure the answer will be in Jedi's link which I will read in a second, but how might I go about acquiring this Math Kernel Library via anaconda?

 

[phinds]

Generally you acquire a distribution by going to the site and clicking on "download".

 

[Jamison Lahman]

Generally you acquire a distribution by going to the site and clicking on "download".

Haha I meant within anaconda but thanks. It seems that MKL is already installed with anaconda https://www.continuum.io/blog/developer-blog/anaconda-25-release-now-mkl-optimizations

If I write a python script, all I do is start with "import mkl" and it runs faster?