DataTaunew | comments | leaders | submitlogin
Python vs Mathematica. What would you prefer? (dan-erik.com)
7 points by amogh10 3792 days ago | 8 comments


2 points by rohit 3792 days ago | link

I've never seen people use Mathematica for data science. Has anyone here tried using it? Thoughts?

-----

3 points by amogh10 3792 days ago | link

People do use mathematica for datascience. Entire WolframAlpha is build on Mathematica.

-----

1 point by rohit 3792 days ago | link

Very interesting. How does the performance compare to the other languages out there? Any idea?

-----

2 points by superfx 3791 days ago | link

I'm a computational biologist, and I do almost all my work in Mathematica.

-----

1 point by rohit 3791 days ago | link

Any idea how the performance compares to the other languages out there?

-----

2 points by Tomrod 3792 days ago | link

I found it marginal. I think SageMATH is a better product overall.

Some intro materials can be found at http://laits.utexas.edu/compeco/

-----

1 point by mag 3792 days ago | link

From what I infer by reading your post very quickly (I will read it better as soon as I have more time), you are not using Mathematica in a very efficient way.

An example? You claim Mathematica is not using the 4 cores you have available. Try changing Map[] to ParallelMap[] and you win parallelism, for free. Oh btw, For[] loops in Mathematica are quite inefficient (on that, we agree).

Also, you use python+pgsql so why don't you try using Mathematica+pgsql? Needs["DatabaseLink`"] might be beneficial :)

-----

1 point by amogh10 3792 days ago | link

Fwiw, i'm not the author of the blog.

-----




RSS | Announcements