Matlab to Python conversion: undefined variable

Question

I am trying to convert a Matlab program into Python. It's not giving me the results I'd like.

Matlab Code:

for jj=1:data_length            %for each symbol in the input symbol sequence

    [a,b]= min(abs(phase_recovered(jj)-U_alphabets));    

    quantized(jj)=U_alphabets(b);     
end

Here quantized is initialized in the for loop for the first time. However, in Python it was giving the error

quantized is undefined.

So I defined quantized as zeros of length equal to data_length.

Python Code:

quantized=zeros(data_length,dtype='complex')
for jj in arange(0,data_length):
    diff=np.absolute((phase_recovered[jj]-u_alphabets))
    a=diff.argmin()
    b=diff[a]
    quantized[jj]=u_alphabets[b]

You also need to include what input you are giving this, what output you are expecting and what you got instead. — Martijn Pieters, Aug 15 '13 at 14:00
*"So I defined `quantized` as zeros of length equal to `data_length`."* Ok, so what's the problem? — arshajii, Aug 15 '13 at 14:03

score 1 · Accepted Answer · edited May 23 '17 at 11:49

1

I think you don't want the line b = diff[a], you want something more like:

quantized=zeros(data_length,dtype='complex')
for jj in arange(0,data_length):
    diff=np.absolute((phase_recovered[jj]-u_alphabets))
    b = diff.argmin()
    quantized[jj] = u_alphabets[b]

Also if u_alphabets happens to be sorted you could use a solution like the one described here.

edited May 23 '17 at 11:49

Community

1
1

answered Aug 15 '13 at 14:09

Bi Rico

25,283
3
52
75

Matlab to Python conversion: undefined variable

1 Answers1