In [46]:
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import statsmodels.api as sm
%matplotlib inline
In [86]:
w1 = 0.3
w2 = 1.0
dt = 0.1
T = 200
Nt = int(T / dt)
Tperiod1 = 21.0
w1 = (2 * np.pi) / Tperiod1
Tperiod2 = 6.2
w2 = (2 * np.pi) / Tperiod2
t = np.arange(start=0, stop=T, step=dt)
original_signal = np.sin(w1 * t) + np.sin(w2 * t)
# Now let's calculate a moving average
window_size = 5.0
Nwindow_size = int(window_size / dt)
a = np.ones(Nwindow_size)
In [87]:
y1 = np.zeros(Nt)
y2 = np.zeros(Nt)
aux = np.convolve(original_signal, a / Nwindow_size, mode='valid')
y2[Nwindow_size:] = aux[:-1]
for index in range(Nwindow_size, Nt):
x_windowed = original_signal[index - Nwindow_size:index]
product = np.dot(x_windowed, a) / Nwindow_size
y1[index] = product
In [88]:
plt.plot(t, original_signal, label='Original')
plt.plot(t, y1, label='Own Method')
plt.plot(t, y2, label='Convolution')
plt.ylim([-3, 3])
plt.legend()
Out[88]:
In [89]:
print(np.corrcoef(original_signal, y1))
print(np.corrcoef(original_signal, y2))
In [90]:
nlags = 200
t = np.arange(0, int((nlags) * dt) + dt, dt)
# t = np.linspace(0, int(nlags * dt), num=nlags)
acf_original = sm.tsa.stattools.acf(original_signal, nlags=nlags)
acf_y1 = sm.tsa.stattools.acf(y1, nlags=nlags)
acf_y2 = sm.tsa.stattools.acf(y2, nlags=nlags)
In [91]:
plt.plot(t, acf_original)
Out[91]:
In [92]:
plt.plot(t, acf_y1)
Out[92]:
In [93]:
plt.plot(t, acf_y2)
Out[93]:
In [98]:
import sys
sys.path.append("../")
from inputs.sensors import Sensor, PerceptualSpace
from inputs.lag_structure import LagStructure
# Visualization libraries
from visualization.sensor_clustering import visualize_cluster_matrix
from visualization.sensors import visualize_SLM
from visualization.sensors import visualize_STDM_seaborn
from visualization.time_cluster import visualize_time_cluster_matrix
from visualization.code_vectors import visualize_code_vectors
from nexa.nexa import Nexa
In [99]:
Tperiod = Tperiod1
lag_times = np.arange(0, 2 * Tperiod) # Go two times the period
tau = 2 * Tperiod
window_size = 1 * Tperiod
Nwindowsize = int(window_size / dt)
# weights = np.exp( -np.arange(Nwindowsize) / tau)
weights = None
lag_structure = LagStructure(lag_times=lag_times, weights=weights, window_size=window_size)
sensor1 = Sensor(original_signal, dt, lag_structure)
sensor2 = Sensor(y1, dt, lag_structure)
sensors = [sensor1, sensor2]
perceptual_space = PerceptualSpace(sensors, lag_first=True)
Nspatial_clusters = 2 # Number of spatial clusters
Ntime_clusters = 2 # Number of time clusters
Nembedding = 3 # Dimension of the embedding space
# Now the Nexa object
nexa_object = Nexa(perceptual_space, Nspatial_clusters,
Ntime_clusters, Nembedding)
# Make all the calculations
nexa_object.calculate_all()
In [100]:
%matplotlib inline
fig = visualize_SLM(nexa_object)
plt.show(fig)
In [ ]: