def create_fake_data(N):
''' Given number of signal components N, return signal object containing N components with random frquencies, amplitudes and phases.
'''
signal = seismo.signal()
print("Fake numbers")
for i in range(N):
amplitude = 0.1*np.random.rand()
frequency = np.random.rand()*300
phase = np.random.rand()*np.pi
component = seismo.sinewave(amplitude, frequency, phase, 0)
# add component to signal
signal.add_component(component)
print("{}, {}, {}".format(amplitude, frequency, phase))
return signal
def create_fake_data(N):
''' Given number of signal components N, return signal object containing N components with random frquencies, amplitudes and phases.
'''
signal = seismo.signal()
print("Fake numbers")
for i in range(N):
amplitude = 0.1 * np.random.rand()
frequency = np.random.rand() * 300
phase = np.random.rand() * np.pi
component = seismo.sinewave(amplitude, frequency, phase, 0)
# add component to signal
signal.add_component(component)
print("{}, {}, {}".format(amplitude, frequency, phase))
return signal
import matplotlib.pyplot as plt
import matplotlib
import seismo
matplotlib.style.use('ggplot')
if __name__ == "__main__":
# create some fake data
t = np.linspace(0, 20, 10000)
# let's add some random sine waves
N_signals = 5
signal = seismo.signal()
for i in range(N_signals):
amplitude = 0.01*np.random.rand()
frequency = np.random.rand()*200
phase = np.random.rand()*np.pi
component = seismo.sinewave(amplitude, frequency, phase, 0)
# add component to signal
signal.add_component(component)
# evaluate signal at times
s = signal.evaluate(t) + 0.025*np.random.randn(t.size)
# now calculate the DFT
# Need to know Nyquist (more or less)
import matplotlib
from matplotlib.pyplot import subplot, plot, show
from numpy import linspace, sin, pi, random
import seismo
matplotlib.style.use('ggplot')
x = linspace(0, 0.05, 500)
y = 0.6*sin(2*pi*240*x) + 0.2*random.randn(x.size)
f, a = seismo.deeming(x, y)
fmax, amax = seismo.find_peak(f, a)
signal = seismo.signal()
comp1 = seismo.sinewave(amax, fmax, 0)
signal.add_component(comp1)
signal.fit(x, y)
plot(x, y, '.')
plot(x, signal.evaluate(x), lw=2)
show()