def plot_wavelet(self):
d_obs = self.d_obs
d_mod = self.d_mod
d_t_obs = self.d_t_obs
""" plot data wavelet """
scores = []
for j, site in enumerate(self.sitename):
print('Process on Wavelet_' + ''.join(site) + '_No.' + str(j) +
'!')
fig3 = plt.figure(figsize=(8, 8))
# ax3 = fig6.add_subplot(1, 2, 1)
# fig3, ax3 = plt.subplots()
data = d_obs[j, :].compressed()
time_data = d_t_obs[~d_obs[j, :].mask]
# time_data = d_t_obs
result = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Data')
waipy.wavelet_plot('Data', time_data, data, 0.03125, result, fig3)
fig3.savefig(self.filedir + self.variable + '/' + ''.join(site) +
'_Wavelet_' + self.variable + '.png')
for m in range(len(d_mod)):
fig4 = plt.figure(figsize=(8, 8))
data = d_mod[m][j, :][~d_obs[j, :].mask] - d_obs[
j, :].compressed()
# time_data = d_mod[m][j, :]
result = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Data')
waipy.wavelet_plot('Data', time_data, data, 0.03125, result,
fig4)
fig4.savefig(self.filedir + self.variable + '/' +
''.join(site) + 'model' + str(m) + '_wavelet_' +
self.variable + '.png')
plt.close('all')
return scores
def main():
'''
d = np.genfromtxt(INFILE).T
time, data = d
'''
fs, data = wavfile.read(INFILE)
plt.plot(data)
# plt.show()
data = data[10000:11000]
# data = scipy.signal.resample(data, len(data)//4)
# time = np.arange(0, len(data), 1./fs)
time = np.arange(len(data))
data = waipy.normalize(data)
result = waipy.cwt(data,
1,
1,
0.25,
2,
4 / 0.25,
alpha,
6,
mother='Morlet')
waipy.wavelet_plot()
wavelet(time, data, True)
raw_input('')
def main():
n_samples = 1024
# Ruido vermelho
S3 = powernoise(2, n_samples)
# Caos, usado para gerar o sinal S7
rho = 3.85
a0 = 0.001
S4 = logistic(rho, a0, n_samples)
# Soma os sinais e normalizae modo que <A>=0 e std=1
S7 = pre.standardize(S3 + S4)
# Sinal gerado pelo pmodel
S8 = pmodel(noValues=n_samples, p=0.52, slope=-1.66)
data = S8
z = np.linspace(0, 1024, 1024)
data_norm = waipy.normalize(data)
result = waipy.cwt(data_norm,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='h',
name='S8')
waipy.wavelet_plot('S8', z, data_norm, 0.03125, result)
def Plot_Wavelet(obs, mod, site_id):
print('Process on Wavelet ' + 'No.' + str(site_id) + '!')
time_data = mod.time
data = (obs.extractDatasites(lat=obs.lat, lon=obs.lat)).data[:, site_id]
y = (mod.extractDatasites(lat=obs.lat, lon=obs.lat)).data[:, site_id]
fig3 = plt.figure(figsize=(8, 8))
result = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Obs')
ax1, ax2, ax3, ax5 = waipy.wavelet_plot('Obs',
time_data,
data,
0.03125,
result,
fig3,
unit=obs.unit)
change_x_tick(mod, site_id, ax2)
def plot_spectrum_score(self):
import waipy, math
import numpy as np
import matplotlib.pyplot as plt
d_obs = self.d_obs
d_mod = self.d_mod
d_t_obs = self.d_t_obs
scores = []
""" Plot global wavelet spectrum """
for j, site in enumerate(self.sitename):
if self.sitename.mask[j]:
continue
print('Process on Spectrum_' + site + '_No.' + str(j) + '!')
data = d_obs[j, :].compressed()
result = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Data')
loc_o, scale_o = norm.fit(result['global_ws'])
model_score = []
for m in range(len(d_mod)):
data = d_mod[m][j, :][~d_obs[j, :].mask]
result_temp = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Data')
loc_m, scale_m = norm.fit(result_temp['global_ws'])
model_score.append(abs(loc_m - loc_o))
model_score = model_score / max(model_score)
scores.append(model_score)
return scores
from scipy import signal
import numpy as np
import matplotlib.pyplot as plt
import waipy
datasets = (['data1.dat', 'SAR', 1800], ['data2.dat', 'BIK',
1864], ['data3.dat', 'VUS', 1609])
dt = 1
units = 'mm'
for d, name, t0 in datasets:
data = np.loadtxt(d)
N = data.size
time = np.arange(0, N) * dt + t0
label = 'Precipitation data {}'.format(name)
data_norm = waipy.normalize(data)
alpha = abs(np.corrcoef(data_norm[0:-1], data_norm[1:])[0, 1])
result = waipy.cwt(data_norm,
1,
1,
0.25,
2,
7 / 0.25,
alpha,
6,
mother='Morlet',
name=name)
waipy.wavelet_plot(label, time, data_norm, 1.0e-6, result)
def plot_spectrum(self):
import waipy, math
import numpy as np
import matplotlib.pyplot as plt
d_obs = self.d_obs
d_mod = self.d_mod
d_t_obs = self.d_t_obs
scores = []
""" Plot global wavelet spectrum """
# col = ['palevioletred', 'm', 'plum', 'darkorchid', 'blue', 'navy', 'deepskyblue', 'darkcyan', 'seagreen',
# 'darkgreen', 'olivedrab', 'gold', 'tan', 'red']
for j, site in enumerate(self.sitename):
if self.sitename.mask[j]:
continue
print('Process on Spectrum_' + site + '_No.' + str(j) + '!')
data = d_obs[j, :].compressed()
if len(data) > 0:
result = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Data')
loc_o, scale_o = norm.fit(result['global_ws'])
fig4 = plt.figure(figsize=(4, 4))
ax4 = fig4.add_subplot(1, 1, 1)
# f1, sxx1 = waipy.fft(data)
# ax.plot(np.log2(1 / f1 * result['dt']), sxx1, 'red', label='Fourier spectrum')
# plt.suptitle(self.variable + ' ( ' + self.d_unit_obs + ' )', fontsize=8)
ax4.plot(np.log2(result['period']),
result['global_ws'],
'k-',
label='Wavelet spectrum')
ax4.plot(np.log2(result['period']),
result['global_signif'],
'r--',
label='95% confidence spectrum')
model_score = []
for m in range(len(d_mod)):
data2 = d_mod[m][j, :][~d_obs[j, :].mask]
data = data2.compressed()
# data = d_mod[m][j, :][~d_obs[j, :].mask]
if len(data) > 0:
result_temp = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Data')
loc_m, scale_m = norm.fit(result_temp['global_ws'])
model_score.append(abs(loc_m - loc_o))
ax4.plot(np.log2(result_temp['period']),
result_temp['global_ws'],
label='Model' + str(m),
c=col[m])
else:
model_score.append(0.5)
model_score = [i / max(model_score) for i in model_score]
scores.append(model_score)
ax4.legend(bbox_to_anchor=(1.05, 1),
loc=2,
fontsize=lengendfontsize)
# ax4.set_ylim(0, 1.25 * np.max(result['global_ws']))
ax4.set_ylabel('Power', fontsize=fontsize)
ax4.set_title('Global Wavelet Spectrum', fontsize=fontsize)
y_min = int(
min(np.log2(result['period'][0]),
np.log2(result_temp['period'][0])))
y_max = int(
max(
np.log2(result['period'][-1]) + 1,
np.log2(result_temp['period'][-1]) + 1))
yt = range(y_min, y_max, 3) # create the vector of period
Yticks = [float(math.pow(2, p)) for p in yt] # make 2^periods
ax4.set_xticks(yt)
ax4.set_xticklabels(Yticks)
ax4.set_xlim(xmin=(np.log2(np.min(result['period']))),
xmax=(np.log2(np.max(result['period']))))
plt.tight_layout()
fig4.savefig(self.filedir + self.variable + '/' + site +
'_spectrum_' + self.variable + '.png',
bbox_inches='tight')
plt.close('all')
return scores
def plot_wavelet(self):
d_obs = self.d_obs
d_mod = self.d_mod
d_t_obs = self.d_t_obs
""" plot data wavelet """
scores = []
for j, site in enumerate(self.sitename):
if self.sitename.mask[j]:
continue
print('Process on Wavelet_' + site + '_No.' + str(j) + '!')
data = d_obs[j, :].compressed()
fig3 = plt.figure(figsize=(8, 8))
if len(data) > 0:
time_data = d_t_obs[~d_obs[j, :].mask]
# time_data = d_t_obs
result = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Obs')
waipy.wavelet_plot('Obs',
time_data,
data,
0.03125,
result,
fig3,
unit=self.d_unit_obs)
# plt.tight_layout()
for m in range(len(d_mod)):
fig4 = plt.figure(figsize=(8, 8))
data2 = d_mod[m][j, :][~d_obs[j, :].mask]
data = data2.compressed() - d_obs[
j, :].compressed()[~data2.mask]
if len(data) > 0:
result = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Obs - Mod' + str(m + 1))
waipy.wavelet_plot('Obs - Mod' + str(m + 1),
time_data[~data2.mask],
data,
0.03125,
result,
fig4,
unit=self.d_unit_obs,
m=m)
# plt.tight_layout()
fig4.savefig(self.filedir + self.variable + '/' + site +
'model' + str(m) + '_wavelet_' + self.variable +
'.png',
bbox_inches='tight')
fig3.savefig(self.filedir + self.variable + '/' + site +
'_Wavelet_' + self.variable + '.png',
bbox_inches='tight')
plt.close('all')
return scores
def plot_spectrum(self):
import waipy, math
import numpy as np
import matplotlib.pyplot as plt
d_obs = self.d_obs
d_mod = self.d_mod
d_t_obs = self.d_t_obs
scores = []
""" Plot global wavelet spectrum """
for j, site in enumerate(self.sitename):
print('Process on Spectrum_' + ''.join(site) + '_No.' + str(j) +
'!')
data = d_obs[j, :].compressed()
result = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Data')
fig4 = plt.figure(figsize=(8, 8))
ax4 = fig4.add_subplot(1, 1, 1)
# f1, sxx1 = waipy.fft(data)
# ax.plot(np.log2(1 / f1 * result['dt']), sxx1, 'red', label='Fourier spectrum')
# plt.suptitle(self.variable + ' ( ' + self.d_unit_obs + ' )', fontsize=8)
ax4.plot(np.log2(result['period']),
result['global_ws'],
'k-',
label='Wavelet spectrum')
ax4.plot(np.log2(result['period']),
result['global_signif'],
'r--',
label='95% confidence spectrum')
for m in range(len(d_mod)):
data = d_mod[m][j, :][~d_obs[j, :].mask]
result_temp = waipy.cwt(data,
1,
1,
0.125,
2,
4 / 0.125,
0.72,
6,
mother='Morlet',
name='Data')
ax4.plot(np.log2(result_temp['period']),
result_temp['global_ws'],
label='Model' + str(m))
ax4.legend(loc=0)
ax4.set_ylim(0, 1.25 * np.max(result['global_ws']))
ax4.set_ylabel('Power', fontsize=12)
ax4.set_title('Global Wavelet Spectrum', fontsize=12)
yt = range(int(np.log2(result['period'][0])),
int(np.log2(result['period'][-1]) +
1)) # create the vector of periods
Yticks = [float(math.pow(2, p)) for p in yt] # make 2^periods
ax4.set_xticks(yt)
ax4.set_xticklabels(Yticks)
ax4.set_xlim(xmin=(np.log2(np.min(result['period']))),
xmax=(np.log2(np.max(result['period']))))
plt.tight_layout()
fig4.savefig(self.filedir + self.variable + '/' + ''.join(site) +
'_spectrum_' + self.variable + '.png')
plt.close('all')
return scores
import waipy
import numpy as np
z = np.linspace(0,2048,2048)
x = np.sin(50*np.pi*z)
y = np.cos(50*np.pi*z)
data_norm = waipy.normalize(x)
result = waipy.cwt(data_norm, 1, 1, 0.125, 2, 4/0.125, 0.72, 6,
mother='Morlet',name='x')
waipy.wavelet_plot('Sine', z, data_norm, 0.03125, result)
data_norm1 = waipy.normalize(y)
result1 = waipy.cwt(data_norm1, 1, 1, 0.125, 2, 4/0.125, 0.72, 6,
mother='Morlet',name='y')
waipy.wavelet_plot('Cosine', z, data_norm1, 0.03125, result1)
cross_power, coherence, phase_angle = waipy.cross_wavelet(result['wave'],
result1['wave'])
waipy.plot_cross('Crosspower sine and cosine', cross_power, phase_angle,
z, result, result1)
##############################################################################
################################### MAIN #####################################
##############################################################################
namefile = "daily-cases-covid-19.csv"
l = pd.read_csv(namefile)
codes = list(set(l["Entity"]))
codes = codes[1:]
l = l.set_index("Entity")
values = []
countries = ["Brazil", "India", "Iran", "South Africa", "Egypt"]
for i in codes:
y = list(l.filter(like=i, axis=0)["Daily confirmed cases (cases)"])
if i in countries:
result = waipy.cwt(y, 1, 1, 0.125, 2, 4 / 0.125, 0.72, 6, 'DOG', "x")
waipy.wavelet_plot(i, range(len(y)), y, 0.03125, result, savefig=True)
if len(y) > 50:
alfa, xdfa, ydfa, reta = funcs.dfa1d(y, 1)
freqs, power, xdata, ydata, amp, index, powerlaw, INICIO, FIM = funcs.psd(
y)
values.append([
funcs.variance(y),
funcs.skewness(y),
funcs.kurtosis(y), alfa, index,
mfdfa.makemfdfa(y), i
])
skew2 = []
alfa = []
kurt = []
from waipy import wavelet_plot
tick_spacing = 4
data = np.genfromtxt("data/batch_4.txt", delimiter=",")[20000:30000]
time = (data[:, 0])#/1000
x = 9.8*(data[:, 1])/2048
y = 9.8*(data[:, 2])/2048
z = 9.8*((data[:, 3]))/2048 # -- factory offset of this particular accelerometer!
var = z
dt = 5
data_norm = waipy.normalize(var)
alpha = np.corrcoef(data_norm[0:-1], data_norm[1:])[0,1];
result = waipy.cwt(data_norm, dt, 0, 0.0625, 1, 36, alpha, 6, mother='Morlet', name="name")
waipy.wavelet_plot("img/batch_4_z_start_osc", time, data_norm, 0.03125, result);
"""
CONTINUOUS WAVELET TRANSFORM
pad = 1 # pad the time series with zeroes (recommended)
dj = 0.25 # this will do 4 sub-octaves per octave
s0 = 2*dt # this says start at a scale of 6 months
j1 = 7/dj # this says do 7 powers-of-two with dj sub-octaves each
lag1 = 0.72 # lag-1 autocorrelation for red noise background
param = 6
mother = 'Morlet'
"""