def runAlgorithm(self):
self.alg = EMD(self.refInVector,
tWindowSize=self.tWindowSize,
samplingFrequency=self.samplingFreq,
imfsNo=self.imfsNo)
self.resultVector = self.alg.process()
self.env = self.alg.hilbertAlgorithm.envelope
self.imfs = self.alg.imfs
# self.spline_upper = self.alg.spline_upper
for val in self.resultVector or []:
self.resultFile.write('%d\n' % val)
# for val in self.imf:
# self.imfOutFileFile.write('%f\n' % val)
np.savetxt(self.imfOutFileDir, self.imfs, delimiter=',')
for val in self.env:
self.envOutFileFile.write('%f\n' % val)
def compare_kmeans(C1, C2, unif):
"""
C1 and C2 are obtained from the attribute kmeans.labels_
"""
N = len(C1)
K, J = max(C1)+1, max(C2)+1
zeta = np.zeros((N,K))
gamma = np.zeros((N, J))
for i in range(N):
zeta[i, C1[i]] = 1
gamma[i, C2[i]] = 1
zeta = zeta.T
gamma = gamma.T
if unif:
alpha = [1./K for _ in range(K)]
beta = [1./J for _ in range(J)]
else:
alpha = [len(np.where(C1 == j)[0]) for j in range(K)]
beta = [len(np.where(C2 == j)[0]) for j in range(J)]
# compute EMD
d = lambda x,y: np.linalg.norm(x-y, ord=1)
emd = EMD(zip(zeta, alpha), zip(gamma, beta), d)
emd.write_lp_problem('emd_pb.lp')
emd.solve_lp()
return emd.distance
inst_phase = self._calc_inst_phase(sig)
return np.diff(inst_phase) / (t[1] - t[0])
def show(self):
plt.show()
if __name__ == "__main__":
import numpy as np
from EMD import EMD
# Simple signal example
t = np.arange(0, 3, 0.01)
S = np.sin(13 * t + 0.2 * t**1.4) - np.cos(3 * t)
emd = EMD()
emd.emd(S)
imfs, res = emd.get_imfs_and_residue()
# Initiate visualisation with emd instance
vis = Visualisation(emd)
# Create a plot with all IMFs and residue
vis.plot_imfs(imfs=imfs, residue=res, t=t, include_residue=True)
# Create a plot with instantaneous frequency of all IMFs
vis.plot_instant_freq(t, imfs=imfs)
# Show both plots
vis.show()
plt.figure()
plt.psd(smn, nfft, 1 / dt, label='Signal * Noise', c='#FF8E40')
plt.psd(signal, nfft, 1 / dt, label='Signal', ls=':', c='#9282FF')
plt.xscale('log')
plt.ylim(-64, 9)
plt.legend()
# In[ First IMF ]
# Data plot
fig, ax = plt.subplots(1, 2)
ax[0].plot(signal, 'w-', label='data')
ax[0].grid(linestyle='--', color=([0.2, 0.2, 0.2]))
imfs = EMD(signal, 3)
# IMF plot
for i in range(len(imfs)):
ax[1].plot(imfs[i], label=str(i))
ax[1].grid(linestyle='--', color=([0.2, 0.2, 0.2]))
plt.legend()
# In[ Second IMF ]
max_envs = []
for imf in imfs:
max_env, _ = max_min_env(imf)
max_envs += [max_env]
new_imfs = []
def run_EMD_encoder(x, samples):
emd = EMD(samples)
emd.emd(x, None)
return emd
class EMDTest(unittest.TestCase):
def setUp(self):
refDir = os.path.dirname(os.getcwd())
refNumStr = '103'
refInFileDir = os.path.join(refDir, refNumStr, 'Input.csv')
refOutFileDir = os.path.join(refDir, refNumStr, 'EMDOutput.csv')
resOutFileDir = os.path.join(refDir, refNumStr, 'EMDResultsPython.csv')
self.imfOutFileDir = os.path.join(refDir, refNumStr, 'Imf.csv')
envOutFileDir = os.path.join(refDir, refNumStr, 'Envelope.csv')
self.samplingFreq = 360.0
self.imfsNo = 3
self.tWindowSize = 120.0
self.refInFile = open(refInFileDir, 'r')
self.refOutFile = open(refOutFileDir, 'r')
self.resultFile = open(resOutFileDir, 'w')
# self.imfOutFileFile = open(imfOutFileDir, 'w')
self.envOutFileFile = open(envOutFileDir, 'w')
self.refInVector = []
self.refOutVector = []
self.resultVector = []
self.refIntSignalVector = []
self.refMarksVector = []
def tearDown(self):
self.refInFile.close()
self.refOutFile.close()
self.resultFile.close()
# self.imfOutFileFile.close()
self.envOutFileFile.close()
def loadReferenceData(self):
refInVectorStr = self.refInFile.readline().split(',')
del refInVectorStr[-1]
for item in refInVectorStr:
val = float(item)
self.refInVector.append(val)
refOutVectorStr = self.refOutFile.readline().split(',')
del refOutVectorStr[-1]
for item in refOutVectorStr:
val = int(float(item))
self.refOutVector.append(val)
def runAlgorithm(self):
self.alg = EMD(self.refInVector,
tWindowSize=self.tWindowSize,
samplingFrequency=self.samplingFreq,
imfsNo=self.imfsNo)
self.resultVector = self.alg.process()
self.env = self.alg.hilbertAlgorithm.envelope
self.imfs = self.alg.imfs
# self.spline_upper = self.alg.spline_upper
for val in self.resultVector or []:
self.resultFile.write('%d\n' % val)
# for val in self.imf:
# self.imfOutFileFile.write('%f\n' % val)
np.savetxt(self.imfOutFileDir, self.imfs, delimiter=',')
for val in self.env:
self.envOutFileFile.write('%f\n' % val)
def isResultIdentical(self):
result = True
# self.assertTrue(len(self.resultVector) > 0)
print 'Output size: {}'.format(len(self.resultVector))
for ref, res in zip(self.refOutVector, self.resultVector):
# print '{} {}'.format(ref, res)
if ref != res:
print 'reference ({}) and result ({}) are not equal'.format(ref, res)
result = False
return result
def testAlgorithm(self):
self.loadReferenceData()
self.runAlgorithm()
self.assertTrue(self.isResultIdentical(), 'Reference vector and result vector are not identical')
import os
from EMD import EMD
refDir = os.path.dirname(os.getcwd())
refNumStr = '101'
refInFileDir = os.path.join(refDir, refNumStr, 'Input.csv')
resOutFileDir = os.path.join(refDir, refNumStr, 'EMDResultsPython.csv')
refInFile = open(refInFileDir, 'r')
resultFile = open(resOutFileDir, 'w')
refInVector = []
refInVectorStr = refInFile.readline().split(',')
del refInVectorStr[-1]
for item in refInVectorStr:
val = float(item)
refInVector.append(val)
alg = EMD(refInVector, tWindowSize=120.0, samplingFrequency=360, imfsNo=3)
resultVector = alg.process()
for val in resultVector or []:
resultFile.write('%d\n' % val)
refInFile.close()
resultFile.close()
import os
from EMD import EMD
refDir = os.path.dirname(os.getcwd())
refNumStr = '101'
refInFileDir = os.path.join(refDir, refNumStr, 'Input.csv')
resOutFileDir = os.path.join(refDir, refNumStr, 'EMDResultsPython.csv')
refInFile = open(refInFileDir, 'r')
resultFile = open(resOutFileDir, 'w')
refInVector = []
refInVectorStr = refInFile.readline().split(',')
del refInVectorStr[-1]
for item in refInVectorStr:
val = float(item)
refInVector.append(val)
alg = EMD(refInVector,
tWindowSize=120.0,
samplingFrequency=360,
imfsNo=3)
resultVector = alg.process()
for val in resultVector or []:
resultFile.write('%d\n' % val)
refInFile.close()
resultFile.close()
def get_emd(self):
self.imf = EMD().emd(self.sig, self.t)
inst_phase = self._calc_inst_phase(sig)
return np.diff(inst_phase)/(t[1]-t[0])
def show(self):
plt.show()
if __name__ == "__main__":
import numpy as np
from EMD import EMD
# Simple signal example
t = np.arange(0, 3, 0.01)
S = np.sin(13*t + 0.2*t**1.4) - np.cos(3*t)
emd = EMD()
emd.emd(S)
imfs, res = emd.get_imfs_and_residue()
# Initiate visualisation with emd instance
vis = Visualisation(emd)
# Create a plot with all IMFs and residue
vis.plot_imfs(imfs=imfs, residue=res, t=t, include_residue=True)
# Create a plot with instantaneous frequency of all IMFs
vis.plot_instant_freq(t, imfs=imfs)
# Show both plots
vis.show()
plt.plot(x, y, 'b', label="Actual Trajectory")
plt.plot([x[0], x[-1]], [y[0], y[-1]], 'go')
plt.plot([x[0], x[-1]], [y[0], y[-1]], 'r--', label="Shortest Path")
plt.text(x[0], y[0], "(%i,%i) Start" % (int(x[0]), int(y[0])))
plt.text(x[-1], y[-1], "(%i,%i) End" % (int(x[-1]), int(y[-1])))
plt.xlabel("X-axis on screen")
plt.ylabel("Y-axis on screen")
#plt.title("Mouse_movement sequence")
#plt.title("User 2 : mouse signal trajectory")
plt.legend(loc='upper right')
plt.show()
# Execute EMD on signal
S = s
print type(S)
eIMFs = EMD().emd(S, t)
nIMFs = eIMFs.shape[0]
print eIMFs.shape
print s
"""
##### Retrieval #############
p1 = s1.get_start()
p2 = s1.get_end()
slope = (p2.y - p1.y)/(p2.x - p1.x)
cosx = -(p2.x - p1.x)/dist(p2,p1)
sinx = (p2.y - p1.y)/dist(p2,p1)
ret = eIMFs[1:].sum(axis=0)
print "Retrieved"
print ret
[val,t1] = s1.get_p()
from EMD import EMD import numpy as np import pylab as plt # Define signal t = np.linspace(0, 1, 200) sin = lambda x, p: np.sin(2 * np.pi * x * t + p) S = 3 * sin(18, 0.2) * (t - 0.2)**2 S += 5 * sin(11, 2.7) S += 3 * sin(14, 1.6) S += 1 * np.sin(4 * 2 * np.pi * (t - 0.8)**2) S += t**2.1 - t s = S # Execute EMD on signal IMF = EMD().emd(s, t) N = IMF.shape[0] + 1 print IMF.shape imf = sum(IMF) print imf.shape plt.figure(figsize=(12, 9)) plt.subplot(N, 1, 1) plt.plot(t, imf, 'r') plt.show() # Plot results """ plt.figure(figsize=(12,9)) plt.subplot(N, 1, 1) plt.plot(t, s, 'r')
