// Class: ReadH6AONN5MEMLP
// Automatically generated by MethodBase::MakeClass
//

/* configuration options =====================================================

#GEN -*-*-*-*-*-*-*-*-*-*-*- general info -*-*-*-*-*-*-*-*-*-*-*-

Method         : MLP::H6AONN5MEMLP
TMVA Release   : 3.8.6         [198662]
ROOT Release   : 5.10/00       [330240]
Creator        : thompson
Date           : Tue Jan 22 16:30:01 2008
Host           : Linux patlx3.fnal.gov 2.4.21-37.ELsmp #1 SMP Wed Sep 28 12:13:44 CDT 2005 i686 i686 i386 GNU/Linux
Dir            : /data/cdf04/thompson/hww_1fb/hwwcdf6.1.4v3_43/Hww/TMVAAna
Training events: 22756


#OPT -*-*-*-*-*-*-*-*-*-*-*-*- options -*-*-*-*-*-*-*-*-*-*-*-*-

# Set by User:
V: "False" [Verbose mode]
NCycles: "1000" [Number of training cycles]
HiddenLayers: "N+1,N" [Specification of hidden layer architecture]
# Default:
D: "False" [use-decorrelated-variables flag (depreciated)]
Normalise: "True" [Normalise input variables]
VarTransform: "None" [Variable transformation method]
VarTransformType: "Signal" [Use signal or background events for var transform]
NbinsMVAPdf: "60" [Number of bins used to create MVA PDF]
NsmoothMVAPdf: "2" [Number of smoothing iterations for MVA PDF]
VerboseLevel: "Info" [Verbosity level]
H: "False" [Print classifier-specific help message]
CreateMVAPdfs: "False" [Create PDFs for classifier outputs]
TxtWeightFilesOnly: "True" [if True, write all weights as text files]
NeuronType: "sigmoid" [Neuron activation function type]
NeuronInputType: "sum" [Neuron input function type]
TrainingMethod: "BP" [Train with Back-Propagation (BP - default) or Genetic Algorithm (GA - slower and worse)]
LearningRate: "0.02" [ANN learning rate parameter]
DecayRate: "0.01" [Decay rate for learning parameter]
TestRate: "10" [Test for overtraining performed at each #th epochs]
BPMode: "sequential" [Back-propagation learning mode: sequential or batch]
BatchSize: "-1" [Batch size: number of events/batch, only set if in Batch Mode, -1 for BatchSize=number_of_events]
##


#VAR -*-*-*-*-*-*-*-*-*-*-*-* variables *-*-*-*-*-*-*-*-*-*-*-*-

NVar 11
LRHWW                         LRHWW                             'F'    [0,1]
LRWW                          LRWW                              'F'    [0,1]
LRWg                          LRWg                              'F'    [0,0.999932587147]
LRWj                          LRWj                              'F'    [-0.0367123484612,1]
LRZZ                          LRZZ                              'F'    [0,1]
Met                           Met                               'F'    [15.0806331635,275.269592285]
MetDelPhi                     MetDelPhi                         'F'    [0.126148730516,3.13422250748]
MetSpec                       MetSpec                           'F'    [15.0024642944,215.058685303]
dPhiLeptons                   dPhiLeptons                       'F'    [2.5749206543e-05,3.14062619209]
dRLeptons                     dRLeptons                         'F'    [0.397228628397,4.5256857872]
dimass                        dimass                            'F'    [16.0042724609,557.91784668]


============================================================================ */

#include <vector>
#include <cmath>
#include <string>
#include <iostream>

#ifndef IClassifierReader__def
#define IClassifierReader__def

class IClassifierReader {

 public:

   // constructor
   IClassifierReader() {}
   virtual ~IClassifierReader() {}

   // return classifier response
   virtual double GetMvaValue( const std::vector<double>& inputValues ) const = 0;
};

#endif

class ReadH6AONN5MEMLP : public IClassifierReader {

 public:

   // constructor
   ReadH6AONN5MEMLP( std::vector<std::string>& theInputVars ) 
      : IClassifierReader(),
        fClassName( "ReadH6AONN5MEMLP" ),
        fStatusIsClean( true ),
        fNvars( 11 ),
        fIsNormalised( true )
   {      
      // the training input variables
      const char* inputVars[] = { "LRHWW", "LRWW", "LRWg", "LRWj", "LRZZ", "Met", "MetDelPhi", "MetSpec", "dPhiLeptons", "dRLeptons", "dimass" };

      // sanity checks
      if (theInputVars.size() <= 0) {
         std::cout << "Problem in class \"" << fClassName << "\": empty input vector" << std::endl;
         fStatusIsClean = false;
      }

      if (theInputVars.size() != fNvars) {
         std::cout << "Problem in class \"" << fClassName << "\": mismatch in number of input values: "
                   << theInputVars.size() << " != " << fNvars << std::endl;
         fStatusIsClean = false;
      }

      // validate input variables
      for (size_t ivar = 0; ivar < theInputVars.size(); ivar++) {
         if (theInputVars[ivar] != inputVars[ivar]) {
            std::cout << "Problem in class \"" << fClassName << "\": mismatch in input variable names" << std::endl
                      << " for variable [" << ivar << "]: " << theInputVars[ivar].c_str() << " != " << inputVars[ivar] << std::endl;
            fStatusIsClean = false;
         }
      }

      // initialize min and max vectors (for normalisation)
      fVmin[0] = 0;
      fVmax[0] = 1;
      fVmin[1] = 0;
      fVmax[1] = 1;
      fVmin[2] = 0;
      fVmax[2] = 0.999932587146759;
      fVmin[3] = -0.0367123484611511;
      fVmax[3] = 1;
      fVmin[4] = 0;
      fVmax[4] = 1;
      fVmin[5] = 15.0806331634521;
      fVmax[5] = 275.269592285156;
      fVmin[6] = 0.126148730516434;
      fVmax[6] = 3.13422250747681;
      fVmin[7] = 15.0024642944336;
      fVmax[7] = 215.058685302734;
      fVmin[8] = 2.57492065429688e-05;
      fVmax[8] = 3.1406261920929;
      fVmin[9] = 0.397228628396988;
      fVmax[9] = 4.52568578720093;
      fVmin[10] = 16.0042724609375;
      fVmax[10] = 557.917846679688;

      // initialize input variable types
      fType[0] = 'F';
      fType[1] = 'F';
      fType[2] = 'F';
      fType[3] = 'F';
      fType[4] = 'F';
      fType[5] = 'F';
      fType[6] = 'F';
      fType[7] = 'F';
      fType[8] = 'F';
      fType[9] = 'F';
      fType[10] = 'F';

      // initialize constants
      Initialize();

   }

   // destructor
   virtual ~ReadH6AONN5MEMLP() {
      Clear(); // method-specific
   }

   // the classifier response
   // "inputValues" is a vector of input values in the same order as the 
   // variables given to the constructor
   double GetMvaValue( const std::vector<double>& inputValues ) const
   {
      // classifier response value
      double retval = 0;

      // classifier response, sanity check first
      if (!fStatusIsClean) {
         std::cout << "Problem in class \"" << fClassName << "\": cannot return classifier response"
                   << " because status is dirty" << std::endl;
         retval = 0;
      }
      else {
         if (IsNormalised()) {
            // normalise variables
            std::vector<double> iV;
            int ivar = 0;
            for (std::vector<double>::const_iterator varIt = inputValues.begin();
                 varIt != inputValues.end(); varIt++, ivar++) {
               iV.push_back(NormVariable( *varIt, fVmin[ivar], fVmax[ivar] ));
            }
            retval = GetMvaValue__( iV );
         }
         else {
            retval = GetMvaValue__( inputValues );
         }
      }

      return retval;
   }

 private:

   // method-specific destructor
   void Clear();

   // common member variables
   const char* fClassName;
   bool fStatusIsClean;

   const size_t fNvars;
   size_t GetNvar()           const { return fNvars; }
   char   GetType( int ivar ) const { return fType[ivar]; }

   // normalisation of input variables
   const bool fIsNormalised;
   bool IsNormalised() const { return fIsNormalised; }
   double fVmin[11];
   double fVmax[11];
   double NormVariable( double x, double xmin, double xmax ) const {
      // normalise to output range: [-1, 1]
      return 2*(x - xmin)/(xmax - xmin) - 1.0;
   }

   // type of input variable: 'F' or 'I'
   char   fType[11];

   // initialize internal variables
   void Initialize();
   double GetMvaValue__( const std::vector<double>& inputValues ) const;

   // private members (method specific)

   double ActivationFnc(double x) const;

   int fLayers;
   int fLayerSize[4];
   double fWeightMatrix0to1[13][12];   // weight matrix from layer 0 to 1
   double fWeightMatrix1to2[12][13];   // weight matrix from layer 1 to 2
   double fWeightMatrix2to3[1][12];   // weight matrix from layer 2 to 3

   double * fWeights[4];
};

inline void ReadH6AONN5MEMLP::Initialize()
{
   // build network structure
   fLayers = 4;
   fLayerSize[0] = 12; fWeights[0] = new double[12]; 
   fLayerSize[1] = 13; fWeights[1] = new double[13]; 
   fLayerSize[2] = 12; fWeights[2] = new double[12]; 
   fLayerSize[3] = 1; fWeights[3] = new double[1]; 
   // weight matrix from layer 0 to 1
   fWeightMatrix0to1[0][0] = 1.33329445528617;
   fWeightMatrix0to1[1][0] = 1.16277771851443;
   fWeightMatrix0to1[2][0] = 1.01505145721233;
   fWeightMatrix0to1[3][0] = 2.43873945526412;
   fWeightMatrix0to1[4][0] = 0.282207840499145;
   fWeightMatrix0to1[5][0] = 1.97893752772164;
   fWeightMatrix0to1[6][0] = 0.980825847721466;
   fWeightMatrix0to1[7][0] = -0.49744443365359;
   fWeightMatrix0to1[8][0] = 0.335050756305815;
   fWeightMatrix0to1[9][0] = 1.09622508030692;
   fWeightMatrix0to1[10][0] = -0.253988971200609;
   fWeightMatrix0to1[11][0] = -0.267675151556724;
   fWeightMatrix0to1[0][1] = 1.09971141828846;
   fWeightMatrix0to1[1][1] = 0.149953386104817;
   fWeightMatrix0to1[2][1] = 0.843090563190795;
   fWeightMatrix0to1[3][1] = -1.60706900818048;
   fWeightMatrix0to1[4][1] = 1.43681882699378;
   fWeightMatrix0to1[5][1] = -0.202057876952845;
   fWeightMatrix0to1[6][1] = -1.19666911067995;
   fWeightMatrix0to1[7][1] = 1.33915873993917;
   fWeightMatrix0to1[8][1] = 0.602273967674004;
   fWeightMatrix0to1[9][1] = 1.26372118782732;
   fWeightMatrix0to1[10][1] = -0.654516104831529;
   fWeightMatrix0to1[11][1] = 0.0014099523727459;
   fWeightMatrix0to1[0][2] = 0.759196590576108;
   fWeightMatrix0to1[1][2] = 0.815004205907034;
   fWeightMatrix0to1[2][2] = 0.616173541745705;
   fWeightMatrix0to1[3][2] = -1.52849890267838;
   fWeightMatrix0to1[4][2] = 1.96121323113922;
   fWeightMatrix0to1[5][2] = 0.192418399928705;
   fWeightMatrix0to1[6][2] = -0.863394780056049;
   fWeightMatrix0to1[7][2] = -0.371056006997744;
   fWeightMatrix0to1[8][2] = -0.697601965212351;
   fWeightMatrix0to1[9][2] = -0.743056689819698;
   fWeightMatrix0to1[10][2] = 0.193415301314325;
   fWeightMatrix0to1[11][2] = 0.75770864643659;
   fWeightMatrix0to1[0][3] = 0.473593201304116;
   fWeightMatrix0to1[1][3] = 0.393657302073201;
   fWeightMatrix0to1[2][3] = 0.791034796118637;
   fWeightMatrix0to1[3][3] = 1.65996543257492;
   fWeightMatrix0to1[4][3] = -0.908458372716504;
   fWeightMatrix0to1[5][3] = -0.440228385776423;
   fWeightMatrix0to1[6][3] = -0.367107450725105;
   fWeightMatrix0to1[7][3] = -1.8717782480327;
   fWeightMatrix0to1[8][3] = -0.971562482515616;
   fWeightMatrix0to1[9][3] = -1.27377717446608;
   fWeightMatrix0to1[10][3] = -0.377500764118501;
   fWeightMatrix0to1[11][3] = 1.07692461516392;
   fWeightMatrix0to1[0][4] = 0.340514893673556;
   fWeightMatrix0to1[1][4] = 0.535632224968392;
   fWeightMatrix0to1[2][4] = 1.21048763668143;
   fWeightMatrix0to1[3][4] = -1.45046362655543;
   fWeightMatrix0to1[4][4] = -0.528070813529872;
   fWeightMatrix0to1[5][4] = -1.55288250963667;
   fWeightMatrix0to1[6][4] = -0.33941337326042;
   fWeightMatrix0to1[7][4] = 0.594978500729115;
   fWeightMatrix0to1[8][4] = 0.335338524301177;
   fWeightMatrix0to1[9][4] = 0.169926658425161;
   fWeightMatrix0to1[10][4] = -0.994075175568608;
   fWeightMatrix0to1[11][4] = -0.705430305572987;
   fWeightMatrix0to1[0][5] = -0.801875085668513;
   fWeightMatrix0to1[1][5] = -1.21749274576568;
   fWeightMatrix0to1[2][5] = -2.52246299848429;
   fWeightMatrix0to1[3][5] = 0.827068159990063;
   fWeightMatrix0to1[4][5] = 1.50517666641559;
   fWeightMatrix0to1[5][5] = -2.12810266933832;
   fWeightMatrix0to1[6][5] = 2.00718292821866;
   fWeightMatrix0to1[7][5] = -0.570436631354959;
   fWeightMatrix0to1[8][5] = -0.527511306318558;
   fWeightMatrix0to1[9][5] = 0.823801929556194;
   fWeightMatrix0to1[10][5] = -0.18848856902848;
   fWeightMatrix0to1[11][5] = -1.60433270681978;
   fWeightMatrix0to1[0][6] = -1.29084169764006;
   fWeightMatrix0to1[1][6] = -0.454405206720752;
   fWeightMatrix0to1[2][6] = 1.87137099410697;
   fWeightMatrix0to1[3][6] = -1.3002037616701;
   fWeightMatrix0to1[4][6] = -0.438387784549448;
   fWeightMatrix0to1[5][6] = -0.285498185854723;
   fWeightMatrix0to1[6][6] = -1.28440451367659;
   fWeightMatrix0to1[7][6] = 1.09429857178614;
   fWeightMatrix0to1[8][6] = -0.0735405486251169;
   fWeightMatrix0to1[9][6] = 0.405604665615793;
   fWeightMatrix0to1[10][6] = 1.32666270235374;
   fWeightMatrix0to1[11][6] = 1.12859317187366;
   fWeightMatrix0to1[0][7] = -0.293485966062245;
   fWeightMatrix0to1[1][7] = -0.155532804419513;
   fWeightMatrix0to1[2][7] = 1.33522605429953;
   fWeightMatrix0to1[3][7] = -0.321402097308763;
   fWeightMatrix0to1[4][7] = 1.73633217738679;
   fWeightMatrix0to1[5][7] = -0.19459731466423;
   fWeightMatrix0to1[6][7] = 0.66640596905167;
   fWeightMatrix0to1[7][7] = -0.677671996568586;
   fWeightMatrix0to1[8][7] = -0.640738880700378;
   fWeightMatrix0to1[9][7] = 2.10759655020123;
   fWeightMatrix0to1[10][7] = -1.51131910273598;
   fWeightMatrix0to1[11][7] = 1.63983468890712;
   fWeightMatrix0to1[0][8] = 0.000281788068855807;
   fWeightMatrix0to1[1][8] = -1.41887267887073;
   fWeightMatrix0to1[2][8] = 2.15206959289121;
   fWeightMatrix0to1[3][8] = -1.75023854572396;
   fWeightMatrix0to1[4][8] = 0.437451262052743;
   fWeightMatrix0to1[5][8] = 0.562726068619443;
   fWeightMatrix0to1[6][8] = 0.246352581970749;
   fWeightMatrix0to1[7][8] = -0.414590186653334;
   fWeightMatrix0to1[8][8] = 2.06962276932526;
   fWeightMatrix0to1[9][8] = 0.256146400355865;
   fWeightMatrix0to1[10][8] = 0.239500935769585;
   fWeightMatrix0to1[11][8] = -1.58492959703701;
   fWeightMatrix0to1[0][9] = -0.0522454373278077;
   fWeightMatrix0to1[1][9] = -2.05763379793288;
   fWeightMatrix0to1[2][9] = 1.5859927233195;
   fWeightMatrix0to1[3][9] = -1.07010558536526;
   fWeightMatrix0to1[4][9] = 0.0701147722504132;
   fWeightMatrix0to1[5][9] = 0.531604693028227;
   fWeightMatrix0to1[6][9] = -0.433565092839553;
   fWeightMatrix0to1[7][9] = 0.690502300858756;
   fWeightMatrix0to1[8][9] = -1.29837092802056;
   fWeightMatrix0to1[9][9] = -2.24191200711996;
   fWeightMatrix0to1[10][9] = -1.23887399704555;
   fWeightMatrix0to1[11][9] = 0.373215951522473;
   fWeightMatrix0to1[0][10] = -0.897658732948844;
   fWeightMatrix0to1[1][10] = 0.439953983306433;
   fWeightMatrix0to1[2][10] = 1.2534558640764;
   fWeightMatrix0to1[3][10] = -1.72575060797909;
   fWeightMatrix0to1[4][10] = 0.961837004154293;
   fWeightMatrix0to1[5][10] = -1.51164487904294;
   fWeightMatrix0to1[6][10] = -1.23026529261941;
   fWeightMatrix0to1[7][10] = 0.741087031991318;
   fWeightMatrix0to1[8][10] = -0.617199897527471;
   fWeightMatrix0to1[9][10] = -0.210046874393532;
   fWeightMatrix0to1[10][10] = -0.631020364736678;
   fWeightMatrix0to1[11][10] = -1.36993628524005;
   fWeightMatrix0to1[0][11] = -1.31198962802525;
   fWeightMatrix0to1[1][11] = -0.0417159035193038;
   fWeightMatrix0to1[2][11] = -0.165321151560166;
   fWeightMatrix0to1[3][11] = 0.406910838319176;
   fWeightMatrix0to1[4][11] = 0.679438129940624;
   fWeightMatrix0to1[5][11] = 1.45629672496869;
   fWeightMatrix0to1[6][11] = -0.0481718219663419;
   fWeightMatrix0to1[7][11] = -1.48067373060391;
   fWeightMatrix0to1[8][11] = 2.16147366365734;
   fWeightMatrix0to1[9][11] = -2.616389259031;
   fWeightMatrix0to1[10][11] = -0.519732966284467;
   fWeightMatrix0to1[11][11] = -0.162615898643527;
   // weight matrix from layer 1 to 2
   fWeightMatrix1to2[0][0] = -1.71872838464772;
   fWeightMatrix1to2[1][0] = 0.00681896639457651;
   fWeightMatrix1to2[2][0] = -0.893030943526098;
   fWeightMatrix1to2[3][0] = -0.815347606218253;
   fWeightMatrix1to2[4][0] = -1.55848121688902;
   fWeightMatrix1to2[5][0] = 0.175192612328059;
   fWeightMatrix1to2[6][0] = 1.89387699593054;
   fWeightMatrix1to2[7][0] = 0.901110551713271;
   fWeightMatrix1to2[8][0] = 1.57342117177967;
   fWeightMatrix1to2[9][0] = -2.18327721696925;
   fWeightMatrix1to2[10][0] = -0.178107847557894;
   fWeightMatrix1to2[0][1] = 0.752668169073905;
   fWeightMatrix1to2[1][1] = -1.77733599707027;
   fWeightMatrix1to2[2][1] = -0.782339356014479;
   fWeightMatrix1to2[3][1] = -0.011385910001889;
   fWeightMatrix1to2[4][1] = 1.54528470348543;
   fWeightMatrix1to2[5][1] = -2.43510255296474;
   fWeightMatrix1to2[6][1] = -0.947644263447778;
   fWeightMatrix1to2[7][1] = -0.279134489678471;
   fWeightMatrix1to2[8][1] = -1.37309764889297;
   fWeightMatrix1to2[9][1] = -1.24768277227113;
   fWeightMatrix1to2[10][1] = -0.446389550129488;
   fWeightMatrix1to2[0][2] = 0.487213013175192;
   fWeightMatrix1to2[1][2] = -0.248037649189283;
   fWeightMatrix1to2[2][2] = -0.920344923037634;
   fWeightMatrix1to2[3][2] = -0.00121375920445563;
   fWeightMatrix1to2[4][2] = 1.52581020178261;
   fWeightMatrix1to2[5][2] = -0.633153402510593;
   fWeightMatrix1to2[6][2] = 1.2350739663797;
   fWeightMatrix1to2[7][2] = -2.21623064196065;
   fWeightMatrix1to2[8][2] = 0.316531923088552;
   fWeightMatrix1to2[9][2] = -0.400959279579671;
   fWeightMatrix1to2[10][2] = -0.022172166628122;
   fWeightMatrix1to2[0][3] = -1.54740998736963;
   fWeightMatrix1to2[1][3] = 0.102605812655679;
   fWeightMatrix1to2[2][3] = 0.892459900138467;
   fWeightMatrix1to2[3][3] = -2.35873371820945;
   fWeightMatrix1to2[4][3] = 0.853940655872569;
   fWeightMatrix1to2[5][3] = 0.140474250974906;
   fWeightMatrix1to2[6][3] = 1.23794292555137;
   fWeightMatrix1to2[7][3] = 0.213296182894952;
   fWeightMatrix1to2[8][3] = -1.74252450992404;
   fWeightMatrix1to2[9][3] = -1.87881757233646;
   fWeightMatrix1to2[10][3] = -0.928239715774229;
   fWeightMatrix1to2[0][4] = 0.312159226199657;
   fWeightMatrix1to2[1][4] = 0.136673388660527;
   fWeightMatrix1to2[2][4] = -0.0443044029868572;
   fWeightMatrix1to2[3][4] = -1.01978095989112;
   fWeightMatrix1to2[4][4] = 0.550797714557982;
   fWeightMatrix1to2[5][4] = 0.262547276071448;
   fWeightMatrix1to2[6][4] = -1.27549182578611;
   fWeightMatrix1to2[7][4] = -0.53905039228941;
   fWeightMatrix1to2[8][4] = 0.81154783115659;
   fWeightMatrix1to2[9][4] = 0.80988208344329;
   fWeightMatrix1to2[10][4] = -0.788272091119694;
   fWeightMatrix1to2[0][5] = -0.850481179242301;
   fWeightMatrix1to2[1][5] = -0.170008409713353;
   fWeightMatrix1to2[2][5] = -0.416361919395077;
   fWeightMatrix1to2[3][5] = -0.804376119281284;
   fWeightMatrix1to2[4][5] = -2.0937032895512;
   fWeightMatrix1to2[5][5] = -1.51165889238655;
   fWeightMatrix1to2[6][5] = -0.81439771823417;
   fWeightMatrix1to2[7][5] = 0.106555116678542;
   fWeightMatrix1to2[8][5] = -2.09912996940905;
   fWeightMatrix1to2[9][5] = 1.55180723500744;
   fWeightMatrix1to2[10][5] = -0.755654513139985;
   fWeightMatrix1to2[0][6] = -1.21903179169395;
   fWeightMatrix1to2[1][6] = 1.10901580193308;
   fWeightMatrix1to2[2][6] = -1.32942426903596;
   fWeightMatrix1to2[3][6] = -1.88418904716011;
   fWeightMatrix1to2[4][6] = 0.887477824604427;
   fWeightMatrix1to2[5][6] = -0.63546351026751;
   fWeightMatrix1to2[6][6] = 0.497551987483262;
   fWeightMatrix1to2[7][6] = -2.44358723828672;
   fWeightMatrix1to2[8][6] = -0.952489395894645;
   fWeightMatrix1to2[9][6] = 0.789113329979111;
   fWeightMatrix1to2[10][6] = -3.01897589585839;
   fWeightMatrix1to2[0][7] = 0.722595097523398;
   fWeightMatrix1to2[1][7] = 0.0584832035546435;
   fWeightMatrix1to2[2][7] = -0.980204356385037;
   fWeightMatrix1to2[3][7] = -1.37147952558853;
   fWeightMatrix1to2[4][7] = -1.2202412604078;
   fWeightMatrix1to2[5][7] = -1.07959915623576;
   fWeightMatrix1to2[6][7] = -0.480868892656051;
   fWeightMatrix1to2[7][7] = -1.91239912477056;
   fWeightMatrix1to2[8][7] = -0.152264489073025;
   fWeightMatrix1to2[9][7] = -1.2507324604474;
   fWeightMatrix1to2[10][7] = 1.24114656067732;
   fWeightMatrix1to2[0][8] = -1.56441916010942;
   fWeightMatrix1to2[1][8] = -0.646199091985705;
   fWeightMatrix1to2[2][8] = 1.05772292389528;
   fWeightMatrix1to2[3][8] = 0.584592995828214;
   fWeightMatrix1to2[4][8] = -1.75704192377107;
   fWeightMatrix1to2[5][8] = 1.51209593605402;
   fWeightMatrix1to2[6][8] = -1.83549026089717;
   fWeightMatrix1to2[7][8] = 0.636964274781543;
   fWeightMatrix1to2[8][8] = -0.746098488239318;
   fWeightMatrix1to2[9][8] = -1.64738649018105;
   fWeightMatrix1to2[10][8] = 1.60542332916382;
   fWeightMatrix1to2[0][9] = -0.540696004838309;
   fWeightMatrix1to2[1][9] = 1.12716930001639;
   fWeightMatrix1to2[2][9] = 1.27784409681559;
   fWeightMatrix1to2[3][9] = 0.0344973498980295;
   fWeightMatrix1to2[4][9] = 1.12108883871905;
   fWeightMatrix1to2[5][9] = -1.21393632945652;
   fWeightMatrix1to2[6][9] = -1.17221378528003;
   fWeightMatrix1to2[7][9] = -0.884248224434228;
   fWeightMatrix1to2[8][9] = -0.537078951070785;
   fWeightMatrix1to2[9][9] = -1.13013561828007;
   fWeightMatrix1to2[10][9] = -0.907177986727912;
   fWeightMatrix1to2[0][10] = 0.734845086736402;
   fWeightMatrix1to2[1][10] = -0.984971592615846;
   fWeightMatrix1to2[2][10] = -1.94304400303567;
   fWeightMatrix1to2[3][10] = 1.06969883076261;
   fWeightMatrix1to2[4][10] = -1.29629285537926;
   fWeightMatrix1to2[5][10] = -1.08185374438838;
   fWeightMatrix1to2[6][10] = -1.08128842422598;
   fWeightMatrix1to2[7][10] = 0.14859655579043;
   fWeightMatrix1to2[8][10] = -0.363623972281679;
   fWeightMatrix1to2[9][10] = -0.461770241632295;
   fWeightMatrix1to2[10][10] = -1.27544686951188;
   fWeightMatrix1to2[0][11] = 0.747852426862887;
   fWeightMatrix1to2[1][11] = -0.645409924979703;
   fWeightMatrix1to2[2][11] = 0.315277743822886;
   fWeightMatrix1to2[3][11] = -1.40536341467642;
   fWeightMatrix1to2[4][11] = -0.682632009981741;
   fWeightMatrix1to2[5][11] = -1.1657951591496;
   fWeightMatrix1to2[6][11] = 0.8875389635865;
   fWeightMatrix1to2[7][11] = 1.32121199713091;
   fWeightMatrix1to2[8][11] = -0.310851983166818;
   fWeightMatrix1to2[9][11] = 0.875157334599399;
   fWeightMatrix1to2[10][11] = 0.514513537819907;
   fWeightMatrix1to2[0][12] = -0.350378240785207;
   fWeightMatrix1to2[1][12] = -0.305957278816555;
   fWeightMatrix1to2[2][12] = -0.627651439402512;
   fWeightMatrix1to2[3][12] = -0.591259287407712;
   fWeightMatrix1to2[4][12] = 1.59418447515226;
   fWeightMatrix1to2[5][12] = 0.901035863653301;
   fWeightMatrix1to2[6][12] = -2.08125714041752;
   fWeightMatrix1to2[7][12] = -2.25070969848801;
   fWeightMatrix1to2[8][12] = -0.0169959355374658;
   fWeightMatrix1to2[9][12] = -0.130909906042466;
   fWeightMatrix1to2[10][12] = 0.86301824135679;
   // weight matrix from layer 2 to 3
   fWeightMatrix2to3[0][0] = -0.424084550266778;
   fWeightMatrix2to3[0][1] = -0.0838696657063759;
   fWeightMatrix2to3[0][2] = -0.836238550933598;
   fWeightMatrix2to3[0][3] = 0.236185483210546;
   fWeightMatrix2to3[0][4] = 0.617507627850387;
   fWeightMatrix2to3[0][5] = -1.37988173662972;
   fWeightMatrix2to3[0][6] = -1.3601258740386;
   fWeightMatrix2to3[0][7] = -0.771420915715125;
   fWeightMatrix2to3[0][8] = 0.472600954754963;
   fWeightMatrix2to3[0][9] = -0.474831637841767;
   fWeightMatrix2to3[0][10] = -0.726630996950985;
   fWeightMatrix2to3[0][11] = 0.975397725271345;
}

inline double ReadH6AONN5MEMLP::GetMvaValue__( const std::vector<double>& inputValues ) const
{
   if (inputValues.size() != (unsigned int)fLayerSize[0]-1) {
      std::cout << "Input vector needs to be of size " << fLayerSize[0]-1 << std::endl;
      return 0;
   }

   for (int l=0; l<fLayers; l++)
      for (int i=0; i<fLayerSize[l]; i++) fWeights[l][i]=0;

   for (int l=0; l<fLayers-1; l++)
      fWeights[l][fLayerSize[l]-1]=1;

   for (int i=0; i<fLayerSize[0]-1; i++)
      fWeights[0][i]=inputValues[i];

   // layer 0 to 1
   for (int o=0; o<fLayerSize[1]-1; o++) {
      for (int i=0; i<fLayerSize[0]; i++) {
         double inputVal = fWeightMatrix0to1[o][i] * fWeights[0][i];
         fWeights[1][o] += inputVal;
      }
      fWeights[1][o] = ActivationFnc(fWeights[1][o]);
   }
   // layer 1 to 2
   for (int o=0; o<fLayerSize[2]-1; o++) {
      for (int i=0; i<fLayerSize[1]; i++) {
         double inputVal = fWeightMatrix1to2[o][i] * fWeights[1][i];
         fWeights[2][o] += inputVal;
      }
      fWeights[2][o] = ActivationFnc(fWeights[2][o]);
   }
   // layer 2 to 3
   for (int o=0; o<fLayerSize[3]; o++) {
      for (int i=0; i<fLayerSize[2]; i++) {
         double inputVal = fWeightMatrix2to3[o][i] * fWeights[2][i];
         fWeights[3][o] += inputVal;
      }
   }

   return fWeights[3][0];
}

double ReadH6AONN5MEMLP::ActivationFnc(double x) const {
   // sigmoid
   return 1.0/(1.0+exp(-x));
}
   
// Clean up
inline void ReadH6AONN5MEMLP::Clear() 
{
   // nothing to clear
}