Smooth an input signal

smoothSignal.py

#!python3
import numpy as np
from typing import Union

def smooth(inputSignal:Union[list, np.ndarray], windowLength:int= 11, window:str= 'flat') -> np.ndarray:
    """
    Smooth the data using a window with requested size.

    This method is based on the convolution of a scaled window with the signal.
    The signal is prepared by introducing reflected copies of the signal
    (with the window size) in both ends so that transient parts are minimized
    in the begining and end part of the output signal.

    input:
        inputSignal: the input signal
        windowLength: the dimension of the smoothing window; should be an odd integer
        window: the type of window from 'flat', 'hanning', 'hamming', 'bartlett', 'blackman'
            flat window will produce a moving average smoothing.

    output:
        the smoothed signal

    example:

    t = linspace(-2,2,0.1)
    inputSignal = sin(t)+randn(len(t))*0.1
    y = smooth(inputSignal)

    see also:

    numpy.hanning, numpy.hamming, numpy.bartlett, numpy.blackman, numpy.convolve
    scipy.signal.lfilter

    # Adapted from https://scipy-cookbook.readthedocs.io/items/SignalSmooth.html
    # with only minor changes
    """
    #cSpell:ignore hanning, lfilter
    assert inputSignal.ndim == 1, "smooth only accepts 1 dimension arrays."
    assert isinstance(windowLength, int), "windowLength must be an integer"
    assert inputSignal.size > windowLength, "Input vector needs to be bigger than window size."
    if windowLength < 3:
        return inputSignal
    s = np.r_[inputSignal[windowLength - 1:0:-1], inputSignal, inputSignal[-2:-windowLength - 1:-1]]
    if window == 'flat':
        #moving average
        wl = np.ones(windowLength,'d')
    else:
        if window is "hanning":
            wl = np.hanning(windowLength)
        elif window is "hamming":
            wl = np.hamming(windowLength)
        elif window is "bartlett":
            wl = np.bartlett(windowLength)
        elif window is "blackman":
            wl = np.blackman(windowLength)
        else:
            raise ValueError("Window must be one of 'flat', 'hanning', 'hamming', 'bartlett', 'blackman'")
    # Do the convolution
    y = np.convolve(wl / wl.sum(), s, mode= 'valid')
    return y[windowLength // 2 - 1:-windowLength // 2]