MatLab user thinking of learning Python

Question

I’m considering learning Python with the idea of letting go of MatLab, although I really like MatLab. However, I’m concerned that getting all of the moving and independent pieces to fit together may be a challenge and one that may not be worth it in the end. I’ve also thought about getting into Visual Basic or Visual C++. In the end, I keep coming back to the ease of MatLab. Any thoughts or comments regarding the difficulty of getting going in Python? Is it worth it?

Answer 1

A good place to start is this page: SciPy getting started, which gives an overview of the scientific toolstack that you might be able to use to move towards MatLab from Python: notably the libraries numpy, scipy, matplotlib, and the interactive working environment IPython. In particular, numpy and matplotlib are designed to be very similar to working with MatLab.

NumPy‘s array type augments the Python language with an efficient data structure useful for numerical work, e.g., manipulating matrices. NumPy also provides basic numerical routines, such as tools for finding eigenvectors.

For example in matlab you might write

eye(3)-diag([1 1],1)

and get back

 1    -1     0
 0     1    -1
 0     0     1

In Python/numpy you would write

import numpy as np
np.eye(3)-np.diag([1,1],1)

and get back

array([[ 1., -1.,  0.],
       [ 0.,  1., -1.],
       [ 0.,  0.,  1.]])

With matplotlib

you have full control of line styles, font properties, axes properties, etc, via an object oriented interface or via a set of functions familiar to MATLAB users.

In MatLab for plotting you might write

x=linspace(-pi, pi, 100);
plot(x,sin(x))

In Python/numpy/matplotlib you would write

x=np.linspace(-np.pi, np.pi, 100)
import matplotlib.pyplot as plt
plt.plot(x,np.sin(x))

There is plenty on the web designed for people making the transition, see, e.g. NumPy for Matlab Users.

MATLAB® and NumPy/SciPy have a lot in common. But there are many differences. NumPy and SciPy were created to do numerical and scientific computing in the most natural way with Python, not to be MATLAB® clones. This page is intended to be a place to collect wisdom about the differences, mostly for the purpose of helping proficient MATLAB® users become proficient NumPy and SciPy users. NumPyProConPage is another page for curious people who are thinking of adopting Python with NumPy and SciPy instead of MATLAB® and want to see a list of pros and cons.

You might also like to consider pylab, which brings together numpy and matplotlib into a single namespace, so you don't have to bother with the np and plt prefixes I used above. See, e.g., wikipedia.

There are tags on this site that are worth looking at: e.g. numpy, scipy, matplotlib. There is also a question on Python at stats.se which you might find relevant. If you are interested in statistics, or in reading, writing and manipulating tabular data, you will be interested in pandas, Python's answer to R's data frame.

As to C++, it's a great language, but not in the same category as Python. This is not the right place to discuss their pros and cons, but in short, C++ is much closer to the machine than Python and if you spend the time you can write highly optimized code. In Python you can get code working very quickly, glueing together independent pieces and easily reading and writing data from wherever you want to, but Python code can sometimes run slowly (it's like Matlab -- if you vectorize in numpy it's fast, otherwise it's interpreted and slow). You might occasionally want to speed up slow Python code using the ability for Python to call functions defined in C, see, e.g., this question. (I'll leave Visual Basic to one side as it doesn't seem relevant.)

Finally, as noted in the comments, answering any specifics would involve knowing exactly what your requirements were, not just what you want to do, but who you want to do it with, and how much time and money you have available to invest.

Answer 2

Yes Python is worth learning. It was my first major language which I learned over ten years ago. It's used heavily in Linux systems especially when the systems boot up. Also the libraries for the language are very well developed. If you want to design a game PyGame has been around a long time and makes the process pretty easy. If you want to program for networking Twisted is an excellent library they have. And their web framework Django is a beauty.

I found Python to be a very English like language to get into.

If you'd like to you might take a peak at Ruby as well. Ruby allows a lot of different styles of programming. So you can program just like you did in other languages in Ruby (I strictly mean in style). Also Ruby has the most "free" resources for learning and getting into the language online that I've ever seen.

I love both languages. But my answer to you is "Yes!" Python is a great first language.

It all depends on what you want to do, and what method's you prefer.

---

You can find many mathematical plotting libraries here: https://wiki.python.org/moin/NumericAndScientific/Plotting
Many of which could be an excellent alternative to Matlab.

Answer 3

Python is great for Firmware programming like with Arduino's. C++ is great and very powerful for Programming software & applications. If you want to program hardware, go with python. If you want to program software, go with C++. Im learning C++ and its great.