import numpy as npimport scipy.special import copyclass MLPNet: def __init__(self, hiddenlayer=(10 ,10 ),classification=False ):0 .0 ; self.xMax = 1 .0 lambda x: scipy.special.expit(x)def _initWeights(self):0 ],self.il) - 0 .5 ))1 ],self.hl[0 ]) - 0 .5 ))1 ]) - 0 .5 ))def _calOut(self,X):0 ]@X.T)1 ]@O1)1 ]@O2).Treturn (y)def predict(self,X):0 ])[:,None ]) ) if self.classification: y = np.round(y) #*\label{code:fullmlpbatch:1} return (y)def fit(self,X,Y,eta=0 .75 ,maxIter=200 ,vareps=10 **-3 ,scale=True ,XT=None ,YT=None ): #*\label{code:fullmlpbatch:4} 0 ) if scale else 0 0 ) if scale else 1 0 ])[:,None ]) ) if len(Y.shape) == 1 :None ] 1 ] 1 ] #*\label{code:fullmlpbatch:morethanone} #*\label{code:fullmlpbatch:3} def train(self,X,Y,XVal=None ,YVal=None ,eta=0 .75 ,maxIter=200 ,vareps=10 **-3 ,XT=None ,YT=None ): if XVal is None : (XVal, YVal, X, Y) = self._divValTrainSet(X,Y) if len(Y.shape) == 1 : Y = Y[:,None ]if len(YVal.shape) == 1 : YVal = YVal[:,None ]if self.il != X.shape[1 ]: X = np.hstack( (X,np.ones(X.shape[0 ])[:,None ]) )if self.il != XVal.shape[1 ]: XVal = np.hstack( (XVal,np.ones(XVal.shape[0 ])[:,None ]) )for i in range(len(self.W)):if self.classification: yp = np.round(yp)2 )/XVal.shape[0 ])/YVal.shape[1 ]#*\label{code:fullmlpbatch:2} 0 ],X.shape[0 ],replace=False )0 while meanE > vareps and counter < maxIter: 1 for m in range(self.ol): #*\label{code:fullmlpbatch:5} for i in mixSet: 0 ]@x.T) 1 ]@O1) 2 ][m,:]*O2*(1 -O2)[None ,:]2 ] = O2 1 ] = temp.T@O1[:,None ].T 0 ] = (O1*(1 -O1)*(temp@self.W[1 ])).T@x[:,None ].T for j in range(len(self.W)): if self.classification: yp = np.round(yp)2 )/XVal.shape[0 ])/YVal.shape[1 ] #*\label{code:fullmlpbatch:6} if meanE < minError: if XT is not None :if len(YT.shape) == 1 : YT = YT[:,None ]; 2 )/XT.shape[0 ])/YT.shape[1 ]if self.classification:2 )/X.shape[0 ])/Y.shape[1 ]def _divValTrainSet(self, X,Y):0 ],int(X.shape[0 ]*0 .25 ),replace=False )0 , Y.shape[0 ] ), self.ValSet) return (XVal, YVal, X, Y)def exportNet(self, filePrefix):"MinMax.csv" , np.array([self.xMin, self.xMax]), delimiter="," )"W0.csv" , self.W[0 ], delimiter="," )"W1.csv" , self.W[1 ], delimiter="," )"W2.csv" , self.W[2 ], delimiter="," )def importNet(self,filePrefix, classification=False ):'MinMax.csv' ,delimiter="," )'W2.csv' ,delimiter="," )'W1.csv' ,delimiter="," ) 'W0.csv' ,delimiter="," ) 0 ], W2.shape[1 ])1 ]0 ]0 ] 1 ]if __name__ == '__main__' :42 )1250 ,2 )1250 ,2 ) )0 ] - 0 .25 )**2 + (X[:,1 ] - 0 .25 )**2 < 0 .2 **2 0 ] = 1 0 ] - 0 .75 )**2 + (X[:,1 ] - 0 .75 )**2 < 0 .2 **2 1 ] = 1 0 ],int(X.shape[0 ]*0 .70 ), replace=False )0 , len(Y) ), TrainSet) 24 ,24 ),classification=True )1200 , XT=XTest , YT=YTest)100 'richtig klassifiziert %0.1f%%' % (100 -fp))'falsch klassifiziert %0.1f%%' % (fp))'foobar' )'foobar' ,classification=True )100 'richtig klassifiziert %0.1f%%' % (100 -fp))'falsch klassifiziert %0.1f%%' % (fp))import matplotlib.pyplot as pltfrom mpl_toolkits.mplot3d import Axes3Dfrom matplotlib import cm'all' )1 )1 ,1 ,1 )0 .25 , 0 .25 ), 0 .2 , color='k' , alpha=0 .3 )0 .75 , 0 .75 ), 0 .2 , color='k' , alpha=0 .3 )0 ] - 0 .25 )**2 + (XTest[:,1 ] - 0 .25 )**2 < 0 .2 **2 , yp[: ,0 ]==0 )0 ],XTest[index1,1 ], marker='v' ,c='r' )0 ] - 0 .75 )**2 + (XTest[:,1 ] - 0 .75 )**2 < 0 .2 **2 , yp[: ,1 ]==0 )0 ],XTest[index2,1 ], marker='^' ,c='r' )0 ]==1 ,0 ],XTest[yp[:,0 ]==1 ,1 ], marker='+' ,c='k' )1 ]==1 ,0 ],XTest[yp[:,1 ]==1 ,1 ], marker='o' ,c='k' )'$x_0$' )'$x_1$' )0 ,1 )0 ,1 )'equal' )3 )1 ,1 ,1 )'r-.' , label='Validierung' ) 'k--' , label='Test' ) 'k:' , label='Training' ) 'Lernzyklus' )'Durchschnittlicher Fehler' )
Messung V0.5 in Prozent C=97 H=88 G=92
¤ Dauer der Verarbeitung: 0.8 Sekunden
(vorverarbeitet am 2026-06-07)
¤ *© Formatika GbR, Deutschland
