Is Mathematica a good choice for a scientific project with Machine Learning?

I'm a chemist with some rudimentary programming skills and in the middle of the year I'll be starting a project concerning machine learning, so, I'm sorry if I'm going to ask two questions in this thread (but I believe they are related).

At this point, I'm still undecided if Mathematica is the right choice for a Ph.D. project or if it is best to focus on Python entirely. I'd like to ask your opinions on this matter.

Because I work with molecular magnetism I know that one of the hardest tasks in this field is to predict the nature of magnetic interactions so, I'll try to tackle this problem with Machine Learning (because such interactions are dependent, basically, of distances, angles and number of unpaired electrons). The idea is to feed the computer with a large training set (with number of unpaired electrons, XYZ coordinates of each molecule and experimental magnetic couplings) and see if it can predict the magnetic couplings (J(AB)) of new systems:

$\left[\text{number of unpaired electrons},\:\left[\begin{array}{cccc} Cu & 0.0 & 0.0 & 0.0\\ C & 0.0 & 0.0 & 0.1\\ H & 1.0 & 0.0 & 0.1\\ \vdots & \vdots & \vdots & \vdots\\ H & 2.0 & 1.0 & 3.0 \end{array}\right],\:J_{\left(AB\right)}\right]$

Can Mathematica (using the Predict function) handle the task? Or is this a job better suited for another software?

P. S.: The molecules (and consequently the matrices that represent them) will have a variable number of atoms most of the time.

• Definitely, NO. There are many bugs and black magic to defeat you. – HyperGroups Aug 29 '17 at 7:43
• Thank you, @HyperGroups. I've decided to move on with Python (and I'm glad I did). – Henrique Junior Aug 30 '17 at 9:25