Zero-crossings

[DominiqueMakowski]

Hi Raph,

Was doing some cross-checking and I have a quick question to disperse a doubt in my mind regarding the counting of the number of inversions:

antropy/antropy/entropy.py

Line 908 in 88fea89

nzc = np.diff(np.signbit(x), axis=axis).sum(axis=axis)

Shouldn't it be: np.diff(np.signbit(np.diff( here? I.e., counting the changes in sign of the consecutive differences, rather than the difference of the sign of the consecutive samples 🤔

[raphaelvallat]

Hi @DominiqueMakowski!

You mean using np.diff(np.signbit(x), n=2).sum() instead of np.diff(np.signbit(x)).sum(), correct?

Taking a sine wave example, I think that the current method implemented in AntroPy is correct:

import numpy as np
import matplotlib.pyplot as plt
sf, f, dur = 100, 1, 4
N = sf * dur # Total number of discrete samples
t = np.arange(N) / sf # Time vector
x = np.sin(2 * np.pi * f * t)
plt.plot(t, x)
plt.xlim(t[0], t[-1])
plt.axhline(0);

There are 7 zero-crossings in x.

>>> def nzc(x, n):
>>>   return np.diff(np.signbit(x), n=n).sum()
>>> nzc(x, n=1), nzc(x, n=2)
(7, 14)

The n=1 method (default) gives the correct answer. The double differentiation leads to twice the expected number of zero-crossings.

Thanks,
Raphael

[DominiqueMakowski]

Ah sorry yes I must have been asleep and missed one step ^^ (I was looking at here but I didn't see the that the diff was inside here) thanks!