def Plot_Button_Clicked(self):
Parameters = oct2py.Struct()
FilePath = self.FilePath_LineEdit.text()
RawPath = FeatAConfig.RawDataFunctions
Mu_Min = self.MuMin_SpinBox.value()
Mu_Max = self.MuMax_SpinBox.value()
Beta_Min = self.BetaMin_SpinBox.value()
Beta_Max = self.BetaMax_SpinBox.value()
Time_Min = self.TimeRangeStart_SpinBox.value()
Time_Max = self.TimeRangeEnd_SpinBox.value()
if (self.RemoveNoise_CheckBox.isChecked()):
RemoveNoise = 'true'
else:
RemoveNoise = 'false'
cmd = "DataOut = GetDataAnalysis('{0}', '{1}', {2}, {3}, {4}, {5}, {6}, {7}, {8});".format(
FilePath, RawPath, Mu_Min, Mu_Max, Beta_Min, Beta_Max, Time_Min,
Time_Max, RemoveNoise)
self.octave.eval(cmd)
self.NavigateWindow.show()
self.ui_navigate.plot()
def detect(self, TrialData):
DetectIn = oct2py.Struct()
DetectIn['TrialData'] = TrialData
DetectIn['TrainOut'] = self.TrainOut
self.DetectThread.startDetect(DetectIn)
def Struct(self):
return oct2py.Struct()
def struct(self, dict):
return oct2py.Struct(dict)
def run(self):
#"../Osama Mohamed.csv",1,1,0,0,0,0,0,0,1,0,0,0,4
#TrainOut = KNN_Generic(directory, noiseFlag, f1FLag,f2FLag,f3FLag,f4FLag,f5FLag,f6FLag,LDAFLag,PCAFlag,CSP_LDAFlag,NoneFlag,startD,endD)
#TODO: change f*FLag to f*Flag!
#settig the feature selection method flags
if (self.selectedFeatureExtractionMethod == "mean"):
self.f1FLag = 1
else:
self.f1FLag = 0
if (self.selectedFeatureExtractionMethod == "Min MU and Max Beta"):
self.f3FLag = 1
else:
self.f3FLag = 0
if (self.selectedFeatureExtractionMethod ==
"Min Mu, Max Beta, Mean Mu, Mean Beta"):
self.f4FLag = 1
else:
self.f4FLag = 0
if (self.selectedFeatureExtractionMethod ==
"Min Mu Max Mu Min Beta Max Beta"):
self.f5FLag = 1
else:
self.f5FLag = 0
if (self.selectedFeatureExtractionMethod ==
"Min Mu, max Mu, Min Beta, Max Beta, Mean Mu, Mean Beta"):
self.f6FLag = 1
else:
self.f6FLag = 0
#preprocessing is not done yet
self.idealFlag = 0
self.butterFlag = 0
self.NoneFilterFlag = 0
if (self.selectedPreprocessingMethod == "ideal"):
self.idealFlag = 1
print("Ideal")
elif (self.selectedPreprocessingMethod == "butter"):
self.butterFlag = 1
print("BUTTER")
elif (self.selectedPreprocessingMethod == "None"):
self.NoneFilterFlag = 1
print("None")
else:
print "undetermined pre-processing type"
#setting the feature enhancement flags
if (self.FeatureEnhancementSelectedMethod == "PCA"):
self.PCAFlag = 1
else:
self.PCAFlag = 0
if (self.FeatureEnhancementSelectedMethod == "LDA"):
self.LDAFlag = 1
else:
self.LDAFlag = 0
if (self.FeatureEnhancementSelectedMethod == "None"):
self.NoneFlag = 1
else:
self.NoneFlag = 0
#unused flags for now
self.CSP_LDAFlag = 0
self.f2FLag = 0
#define the output structures
self.knnTrainOut = oct2py.Struct()
self.fisherTrainOut = oct2py.Struct()
self.leastSquaresTrainOut = oct2py.Struct()
self.likelihoodTrainOut = oct2py.Struct()
self.likelihoodClass = oct2py.Struct()
self.knnResult = oct2py.Struct()
self.knnResultInput = oct2py.Struct()
self.fisherResult = oct2py.Struct()
self.fisherResultInput = oct2py.Struct()
self.likelihoodResult = oct2py.Struct()
self.likelihoodResultInput = oct2py.Struct()
self.leastSquaresResult = oct2py.Struct()
self.leastSquaresResultInput = oct2py.Struct()
#calling the classifer selected according
#if (self.trainTestFlag == True):
if (self.classifierFile == "KNN"):
self.knnTrainOut = self.octave.feval(
'KNN_Generic.m', self.dataFile, self.removeNoiseFlag,
self.idealFlag, self.butterFlag, self.NoneFilterFlag,
self.f1FLag, self.f2FLag, self.f3FLag, self.f4FLag,
self.f5FLag, self.f6FLag, self.LDAFlag, self.PCAFlag,
self.CSP_LDAFlag, self.NoneFlag, self.SignalStart,
self.SignalEnd)
self.LDAData = self.knnTrainOut.LDAData
self.PCAData = self.knnTrainOut.PCAData
self.NoneData = self.knnTrainOut.NoneData
self.dataLength = self.knnTrainOut.datalength
print("KNN training done!")
###>--- using either of them is ok for debugging ---<###
#print(self.knnTrainOut.KPCA)
#print(self.knnTrainOut['KPCA'])
elif (self.classifierFile == "Fisher"):
print(self.idealFlag)
print(self.butterFlag)
self.fisherTrainOut = self.octave.feval(
'Fisher_Generic.m', self.dataFile, self.removeNoiseFlag,
self.idealFlag, self.butterFlag, self.NoneFilterFlag,
self.f1FLag, self.f2FLag, self.f3FLag, self.f4FLag,
self.f5FLag, self.f6FLag, self.LDAFlag, self.PCAFlag,
self.CSP_LDAFlag, self.NoneFlag, self.SignalStart,
self.SignalEnd)
self.LDAData = self.fisherTrainOut.LDAData
self.PCAData = self.fisherTrainOut.PCAData
self.NoneData = self.fisherTrainOut.NoneData
self.dataLength = self.fisherTrainOut.datalength
print("Fisher training done!")
elif (self.classifierFile == "Likelihood"):
self.likelihoodTrainOut = self.octave.feval(
'Likelihood_Generic.m', self.dataFile, self.removeNoiseFlag,
self.idealFlag, self.butterFlag, self.NoneFilterFlag,
self.f1FLag, self.f2FLag, self.f3FLag, self.f4FLag,
self.f5FLag, self.f6FLag, self.LDAFlag, self.PCAFlag,
self.CSP_LDAFlag, self.NoneFlag, self.SignalStart,
self.SignalEnd)
self.LDAData = self.likelihoodTrainOut.LDAData
self.PCAData = self.likelihoodTrainOut.PCAData
self.NoneData = self.likelihoodTrainOut.NoneData
self.dataLength = self.likelihoodTrainOut.datalength
print("Likelihood training done!")
elif (self.classifierFile == "Least Squares"):
self.leastSquaresTrainOut = self.octave.feval(
'Leastsquares_Generic.m', self.dataFile, self.removeNoiseFlag,
self.idealFlag, self.butterFlag, self.NoneFilterFlag,
self.f1FLag, self.f2FLag, self.f3FLag, self.f4FLag,
self.f5FLag, self.f6FLag, self.LDAFlag, self.PCAFlag,
self.CSP_LDAFlag, self.NoneFlag, self.SignalStart,
self.SignalEnd)
self.LDAData = self.leastSquaresTrainOut.LDAData
self.PCAData = self.leastSquaresTrainOut.PCAData
self.NoneData = self.leastSquaresTrainOut.NoneData
self.dataLength = self.leastSquaresTrainOut.datalength
print("Least Square training done!")
if (self.trainTestFlag == False):
if (self.detectFile != None):
self.detectData()
else:
print "Detection failed cause of a missing detection file!"
def run(self):
octave = oct2py.Oct2Py()
octave.addpath('octave2')
#"../Osama Mohamed.csv",1,1,0,0,0,0,0,0,1,0,0,0,4
#TrainOut = KNN_Generic(directory, noiseFlag, f1FLag,f2FLag,f3FLag,f4FLag,f5FLag,f6FLag,LDAFLag,PCAFlag,CSP_LDAFlag,CSPFlag,startD,endD)
#settig the feature selection method flags
if(self.selectedFeatureExtractionMethod =="mean"):
self.f1FLag = 1
else:
self.f1FLag = 0
if(self.selectedFeatureExtractionMethod =="Min MU and Max Beta"):
self.f3FLag = 1
else:
self.f3FLag = 0
if(self.selectedFeatureExtractionMethod =="Min Mu, Max Beta, Mean Mu, Mean Beta"):
self.f4FLag = 1
else:
self.f4FLag = 0
if(self.selectedFeatureExtractionMethod =="Min Mu Max Mu Min Beta Max Beta"):
self.f5FLag = 1
else:
self.f5FLag = 0
if(self.selectedFeatureExtractionMethod =="Min Mu, max Mu, Min Beta, Max Beta, Mean Mu, Mean Beta"):
self.f6FLag = 1
else:
self.f6FLag = 0
#preprocessing is not done yet
if(self.selectedPreprocessingMethod == ""):
pass
#setting the feature enhancement flags
if(self.FeatureEnhancementSelectedMethod == "PCA"):
self.PCAFlag = 1
else:
self.PCAFlag = 0
if(self.FeatureEnhancementSelectedMethod == "LDA"):
self.LDAFlag = 1
else:
self.LDAFlag = 0
if(self.FeatureEnhancementSelectedMethod == "CSP"):
self.CSPFlag = 1
else:
self.CSPFlag = 0
#unused flags for now
self.CSP_LDAFlag =0
self.f2FLag =0
#define the output structures
self.knnTrainOut = oct2py.Struct()
self.fisherTrainOut = oct2py.Struct()
self.leastSquaresTrainOut = oct2py.Struct()
self.likelihoodTrainOut = oct2py.Struct()
#calling the classifer selected according
if(self.classifierFile == "KNN"):
self.knnTrainOut = octave.call('KNN_Generic.m', self.dataFile, self.removeNoiseFlag, self.f1FLag,self.f2FLag,self.f3FLag,self.f4FLag,self.f5FLag,self.f6FLag,self.LDAFlag,self.PCAFlag,self.CSP_LDAFlag,self.CSPFlag,self.SignalStart,self.SignalEnd)
print("KNN training done!")
###>--- using either of them is ok for debugging ---<###
#print(self.knnTrainOut.KPCA)
#print(self.knnTrainOut['KPCA'])
elif (self.classifierFile == "Fisher"):
self.fisherTrainOut = octave.call('Fisher_Generic.m', self.dataFile, self.removeNoiseFlag, self.f1FLag,self.f2FLag,self.f3FLag,self.f4FLag,self.f5FLag,self.f6FLag,self.LDAFlag,self.PCAFlag,self.CSP_LDAFlag,self.CSPFlag,self.SignalStart,self.SignalEnd)
print("Fisher training done!")
elif(self.classifierFile == "Likelihood"):
self.likelihoodTrainOut = octave.call('Likelihood_Generic.m', self.dataFile, self.removeNoiseFlag, self.f1FLag,self.f2FLag,self.f3FLag,self.f4FLag,self.f5FLag,self.f6FLag,self.LDAFlag,self.PCAFlag,self.CSP_LDAFlag,self.CSPFlag,self.SignalStart,self.SignalEnd)
print("Likelihood training done!")
elif(self.classifierFile == "Least Squares"):
self.leastSquaresTrainOut = octave.call('Leastsquares_Generic.m', self.dataFile, self.removeNoiseFlag, self.f1FLag,self.f2FLag,self.f3FLag,self.f4FLag,self.f5FLag,self.f6FLag,self.LDAFlag,self.PCAFlag,self.CSP_LDAFlag,self.CSPFlag,self.SignalStart,self.SignalEnd)
print("Least Square training done!")
"double_compress_params_octave.sav",verbose=False)
params=loadout.pdc
Nlinmax=Nlinac
else:#curently not working set readfile to True
Nlinmax=10
gun=linac.Gun_Param()
linp_arr=linac.Linac_Param_Array(Nlinmax)
for l in range(Nlinmax):
lin=linac.Linac_Param()
print 'here'
linp_arr[l]=lin
params=oct2py.Struct()
params.Ev=[None]
params.lamv=[None]
params.Lv=[None]
params.av=[None]
params.R56v=[None]
params.T566v=[None]
params.phiv=[None]
params.s0v=[None]
params.Ev[0]=Nlinmax*[None]
params.lamv[0]=Nlinmax*[None]
params.Lv[0]=Nlinmax*[None]
params.av[0]=Nlinmax*[None]
params.R56v[0]=Nlinmax*[None]
params.T566v[0]=Nlinmax*[None]