Port this snippet from Matlab to Python

Question

I have the following code snippet in Matlab and want to port it to Python.

for i=1:runs;
   tic;
   datmat=importdata('data.txt',' ');
    Limits_Area=[1,261,522,784,1045,1305,1565,1827,2088,2349,2610,2871,3131,3158];
   for k=1:2000;
      for j=1:13;
         Year_Data=datmat.data(Limits_Area(j):Limits_Area(j+1)-1,4:37);
         Min_Year=min(Year_Data);
         Max_Year=max(Year_Data);
         Mean_Year=mean(Year_Data);
         GainLoss_Year=100-(Year_Data(1,:)+0.00000001)./(Year_Data(end,:)+0.00000001)*100;
     end;
   end;

I am having a really hard time with the

Year_Data=datmat.data(Limits_Area(j):Limits_Area(j+1)-1,4:37);

part.... Any directions?

Thank you

It would be nice if you could show some of the data from data.txt, and provide some explanation of what you're trying to accomplish, for those of us who know python but aren't fluid in matlab. Can you show us what you tried and how that failed? Additionally, there seems to be a bug in your code in the sense that the outer loop runs the inner loop 2000 times in exactly the same way. Why? — Roland Smith
– Roland Smith, Commented Oct 22, 2012 at 14:10
Hello Roland, the data seems to be a .txt file containing currency values throughout time with columns: KEY Month Day and then a series of three character currency codes ie AUD ATS etc. What I am trying to accomplish is just take this Matlab code and port it into python, no questions asked. I am not fluent in Matlab either, that's why I am trying to understand that specific line. It's actually a silly "benchmarking" code someone wrote, so what you see as a bug... is a actually a feature. — Thanos
– Thanos, Commented Oct 22, 2012 at 14:14
What's happening in this line is called indexing. You are selecting specific rows and columns in the ´datamat.data´ matrix: rows from Limits_Area(j) to Limits_Area(j+1) and columns 4 to 37. Limits_Area(j) is indexing in a vector and simply means j-th element of Limits_Area. mathworks.com/company/newsletters/articles/… Numpy has a very similar syntax so if it is supposed to look the same in Matlab, use that. — Tobias
– Tobias, Commented Oct 22, 2012 at 14:22

Glorfindel · Accepted Answer · 2022-12-24 20:00:36Z

2

Have you tried numpy ? it is a library for scientific computing which looks like Matlab. For example :

In Matlab : a(1:5,:)

In Numpy : a[0:5] or a[:5] or a[0:5,:]

Check out : Numpy for Matlab Users

if you do not want to use Numpy, try the comprehension lists :

Year_Data = [ [datmat.data(i,j) for j in range (4,38) ] for i in range(j,j+2) ]

EDIT:

for i in range(runs) :
    datamat = numpy.genfromtxt('data.txt',delimiter=' ', newline ='\n' ) 
    // Adapt the previous line to the format of your txt file
    // at this point you should have a numpy.array object with the right shape
    Limits_Area= numpy.array( [1,261,522,784,1045,1305,1565,1827,2088,2349,2610,2871,3131,3158] )
    for k in range(2000):
        for j in range(13):
            Year_Data = datmat[ Limits_Area(j):Limits_Area(j+1)-1 , 4:37 ]
            etc etc ...

NB : Matlab arrays indexes goes from 1 to n whereas numpy arrays indexes goes from 0 to n-1

edited Dec 24, 2022 at 20:00

Glorfindel

22.8k13 gold badges97 silver badges124 bronze badges

answered Oct 22, 2012 at 14:15

lucasg

11k4 gold badges38 silver badges58 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

mgilson Over a year ago

I wouldn't make an attempt to port any serious matlab code to python without numpy.

Thanos Over a year ago

yes I use numpy, so what does this specific line does in Matlab and how would it translate to python? thanks this seems like it's on the right track.

ely Over a year ago

Also worth a look at the Online Matlab to Python Converter. The project appears to be inactive, which is a pity, but there may be something to download that can help. However, I second mgilson's comment. Use NumPy for this.

Collectives™ on Stack Overflow

Port this snippet from Matlab to Python

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related