Gallery of the kind of peaks and noise the sequgen package can generate.

In [1]:

# Utilities to plot
import numpy as np
import matplotlib.pyplot as plt

# Time series has 100 points 
t_predict = np.arange(100)

def plot(title, y_predict):
    plt.figure()
    plt.plot(t_predict, y_predict, ".b-")
    plt.title(title)
    plt.grid(True)
    plt.show()

Constant¶

Generate a time series with a constant value of 5.

In [2]:

from sequgen.deterministic.constant import constant

In [3]:

plot('Constant signal = 5', 
     constant(t_predict, value=5)
     )

Normal peak¶

Generate a time series with located in middle of serie and using height of positive height of 5 with a standard deviation of 10

In [4]:

from sequgen.deterministic.normal_peak import normal_peak

In [5]:

plot('Normal peak location=50, stddev=10, height=5, sign=1',
    normal_peak(t_predict, location=50, stddev=10, height=5, sign=1))

Sine signal¶

Generate a single sine signal which starts from origin and has an amplitude of 5

In [6]:

from sequgen.deterministic.sine import sine

In [7]:

plot('Single sine signal', 
    sine(t_predict, wavelength=100, amplitude=5))

Generate a sine which starts at peak and repeats 4 times

In [8]:

plot('4 sines', 
    sine(t_predict, wavelength=25, phase_shift=18.75))

Triangular peak¶

Generate a triangular peak centered at 60 and with height of 5 and base width of 40

In [9]:

from sequgen.deterministic.triangular_peak import triangular_peak

In [13]:

plot('isosceles triangle',
    triangular_peak(t_predict, placement=40, width_base_left=20, width_base_right=20, height=5))

Gaussian noise¶

To generate a random noise which has gaussian distribution centered at 0 and a standard deviation of 5

In [14]:

from sequgen.stochastic.gaussian import gaussian

In [15]:

plot('Gaussian noise', 
    gaussian(t_predict, average=0, stddev=5))

In [ ]: