CODEWITHSUNDEEP
pythonlistnumpyconcatenation
MathArt
MathArt

Reputation: 249

List construction by concatenation after FFT

Is there a simpler Way to construct frequency components after FFT?

I am computing the FFT of a signal and normalizing it as follows:

import numpy as np
from scipy.fftpack import fft

# By calculating FFT...
Y = fft(y) / n  # FFT computing and normalization
# Constructing a new list with one-sided frequency range
Y = [Y[0]] + [2 * Y[i] for i in range(1, n//2)]

This works correctly, but I initially attempted a more compact form:

Y = [Y[0]] + [2 * Y[range(1, n//2)]]  # This does NOT work

which will actually increase to 2D instead of concatenating indices from 0 to 1:n//2.

My question is: Is there a simpler or more Pythonic way to achieve this one-sided frequency range construction? Ideally, something cleaner than my working solution while maintaining efficiency.

Upvotes: -1

Views: 66

Answers (1)

MathArt
MathArt

Reputation: 249

Finally, I found that the following unpack operator can work for this purpose.

 Y= [Y[0],*Y[1:n//2]*2]

Since we can have

 >>>a, *b, c = [1, 2, 3, 4, 5] to 
 >>>b
 >>>[2, 3, 4]

Also if

>>>a = [*'PYTHON']
>>>a
>>>['P', 'Y', 'T', 'H', 'O', 'N']

Upvotes: -1

Related Questions