C# minasastate类代码示例

        /*************************************************************************
        Returns norm of bounded anti-gradient.

        Bounded antigradient is a vector obtained from  anti-gradient  by  zeroing
        components which point outwards:
            result = norm(v)
            v[i]=0     if ((-g[i]<0)and(x[i]=bndl[i])) or
                          ((-g[i]>0)and(x[i]=bndu[i]))
            v[i]=-g[i] otherwise

        This function may be used to check a stopping criterion.

          -- ALGLIB --
             Copyright 20.03.2009 by Bochkanov Sergey
        *************************************************************************/
        private static double asaboundedantigradnorm(minasastate state)
        {
            double result = 0;
            int i = 0;
            double v = 0;

            result = 0;
            for(i=0; i<=state.n-1; i++)
            {
                v = -state.g[i];
                if( (double)(state.x[i])==(double)(state.bndl[i]) && (double)(-state.g[i])<(double)(0) )
                {
                    v = 0;
                }
                if( (double)(state.x[i])==(double)(state.bndu[i]) && (double)(-state.g[i])>(double)(0) )
                {
                    v = 0;
                }
                result = result+math.sqr(v);
            }
            result = Math.Sqrt(result);
            return result;
        }

        /*************************************************************************
        Obsolete optimization algorithm.
        Was replaced by MinBLEIC subpackage.

          -- ALGLIB --
             Copyright 20.03.2009 by Bochkanov Sergey
        *************************************************************************/
        public static void minasaresultsbuf(minasastate state,
            ref double[] x,
            minasareport rep)
        {
            int i = 0;
            int i_ = 0;

            if( alglib.ap.len(x)<state.n )
            {
                x = new double[state.n];
            }
            for(i_=0; i_<=state.n-1;i_++)
            {
                x[i_] = state.x[i_];
            }
            rep.iterationscount = state.repiterationscount;
            rep.nfev = state.repnfev;
            rep.terminationtype = state.repterminationtype;
            rep.activeconstraints = 0;
            for(i=0; i<=state.n-1; i++)
            {
                if( (double)(state.ak[i])==(double)(0) )
                {
                    rep.activeconstraints = rep.activeconstraints+1;
                }
            }
        }

        /*************************************************************************
        Obsolete optimization algorithm.
        Was replaced by MinBLEIC subpackage.

          -- ALGLIB --
             Copyright 30.07.2010 by Bochkanov Sergey
        *************************************************************************/
        public static void minasarestartfrom(minasastate state,
            double[] x,
            double[] bndl,
            double[] bndu)
        {
            int i_ = 0;

            alglib.ap.assert(alglib.ap.len(x)>=state.n, "MinASARestartFrom: Length(X)<N!");
            alglib.ap.assert(apserv.isfinitevector(x, state.n), "MinASARestartFrom: X contains infinite or NaN values!");
            alglib.ap.assert(alglib.ap.len(bndl)>=state.n, "MinASARestartFrom: Length(BndL)<N!");
            alglib.ap.assert(apserv.isfinitevector(bndl, state.n), "MinASARestartFrom: BndL contains infinite or NaN values!");
            alglib.ap.assert(alglib.ap.len(bndu)>=state.n, "MinASARestartFrom: Length(BndU)<N!");
            alglib.ap.assert(apserv.isfinitevector(bndu, state.n), "MinASARestartFrom: BndU contains infinite or NaN values!");
            for(i_=0; i_<=state.n-1;i_++)
            {
                state.x[i_] = x[i_];
            }
            for(i_=0; i_<=state.n-1;i_++)
            {
                state.bndl[i_] = bndl[i_];
            }
            for(i_=0; i_<=state.n-1;i_++)
            {
                state.bndu[i_] = bndu[i_];
            }
            state.laststep = 0;
            state.rstate.ia = new int[3+1];
            state.rstate.ba = new bool[1+1];
            state.rstate.ra = new double[2+1];
            state.rstate.stage = -1;
            clearrequestfields(state);
        }

        /*************************************************************************

          -- ALGLIB --
             Copyright 20.03.2009 by Bochkanov Sergey
        *************************************************************************/
        public static bool minasaiteration(minasastate state)
        {
            bool result = new bool();
            int n = 0;
            int i = 0;
            double betak = 0;
            double v = 0;
            double vv = 0;
            int mcinfo = 0;
            bool b = new bool();
            bool stepfound = new bool();
            int diffcnt = 0;
            int i_ = 0;

            
            //
            // Reverse communication preparations
            // I know it looks ugly, but it works the same way
            // anywhere from C++ to Python.
            //
            // This code initializes locals by:
            // * random values determined during code
            //   generation - on first subroutine call
            // * values from previous call - on subsequent calls
            //
            if( state.rstate.stage>=0 )
            {
                n = state.rstate.ia[0];
                i = state.rstate.ia[1];
                mcinfo = state.rstate.ia[2];
                diffcnt = state.rstate.ia[3];
                b = state.rstate.ba[0];
                stepfound = state.rstate.ba[1];
                betak = state.rstate.ra[0];
                v = state.rstate.ra[1];
                vv = state.rstate.ra[2];
            }
            else
            {
                n = -983;
                i = -989;
                mcinfo = -834;
                diffcnt = 900;
                b = true;
                stepfound = false;
                betak = 214;
                v = -338;
                vv = -686;
            }
            if( state.rstate.stage==0 )
            {
                goto lbl_0;
            }
            if( state.rstate.stage==1 )
            {
                goto lbl_1;
            }
            if( state.rstate.stage==2 )
            {
                goto lbl_2;
            }
            if( state.rstate.stage==3 )
            {
                goto lbl_3;
            }
            if( state.rstate.stage==4 )
            {
                goto lbl_4;
            }
            if( state.rstate.stage==5 )
            {
                goto lbl_5;
            }
            if( state.rstate.stage==6 )
            {
                goto lbl_6;
            }
            if( state.rstate.stage==7 )
            {
                goto lbl_7;
            }
            if( state.rstate.stage==8 )
            {
                goto lbl_8;
            }
            if( state.rstate.stage==9 )
            {
                goto lbl_9;
            }
            if( state.rstate.stage==10 )
            {
                goto lbl_10;
            }
            if( state.rstate.stage==11 )
            {
//.........这里部分代码省略.........

        /*************************************************************************
        Obsolete optimization algorithm.
        Was replaced by MinBLEIC subpackage.

          -- ALGLIB --
             Copyright 20.03.2009 by Bochkanov Sergey
        *************************************************************************/
        public static void minasaresults(minasastate state,
            ref double[] x,
            minasareport rep)
        {
            x = new double[0];

            minasaresultsbuf(state, ref x, rep);
        }

        /*************************************************************************
        Returns True, if U set is empty.

        * State.X is used as point,
        * State.G - as gradient,
        * D is calculated within function (because State.D may have different
          meaning depending on current optimization algorithm)

        For a meaning of U see 'NEW ACTIVE SET ALGORITHM FOR BOX CONSTRAINED
        OPTIMIZATION' by WILLIAM W. HAGER AND HONGCHAO ZHANG.

          -- ALGLIB --
             Copyright 20.03.2009 by Bochkanov Sergey
        *************************************************************************/
        private static bool asauisempty(minasastate state)
        {
            bool result = new bool();
            int i = 0;
            double d = 0;
            double d2 = 0;
            double d32 = 0;

            d = asad1norm(state);
            d2 = Math.Sqrt(d);
            d32 = d*d2;
            result = true;
            for(i=0; i<=state.n-1; i++)
            {
                if( (double)(Math.Abs(state.g[i]))>=(double)(d2) && (double)(Math.Min(state.x[i]-state.bndl[i], state.bndu[i]-state.x[i]))>=(double)(d32) )
                {
                    result = false;
                    return result;
                }
            }
            return result;
        }

        /*************************************************************************
                      NONLINEAR BOUND CONSTRAINED OPTIMIZATION USING
                                       MODIFIED
                           WILLIAM W. HAGER AND HONGCHAO ZHANG
                                 ACTIVE SET ALGORITHM

        The  subroutine  minimizes  function  F(x)  of  N  arguments  with   bound
        constraints: BndL[i] <= x[i] <= BndU[i]

        This method is  globally  convergent  as  long  as  grad(f)  is  Lipschitz
        continuous on a level set: L = { x : f(x)<=f(x0) }.

        INPUT PARAMETERS:
            N       -   problem dimension. N>0
            X       -   initial solution approximation, array[0..N-1].
            BndL    -   lower bounds, array[0..N-1].
                        all elements MUST be specified,  i.e.  all  variables  are
                        bounded. However, if some (all) variables  are  unbounded,
                        you may specify very small number as bound: -1000,  -1.0E6
                        or -1.0E300, or something like that.
            BndU    -   upper bounds, array[0..N-1].
                        all elements MUST be specified,  i.e.  all  variables  are
                        bounded. However, if some (all) variables  are  unbounded,
                        you may specify very large number as bound: +1000,  +1.0E6
                        or +1.0E300, or something like that.
            EpsG    -   positive number which  defines  a  precision  of  search.  The
                        subroutine finishes its work if the condition ||G|| < EpsG  is
                        satisfied, where ||.|| means Euclidian norm, G - gradient, X -
                        current approximation.
            EpsF    -   positive number which  defines  a  precision  of  search.  The
                        subroutine finishes its work if on iteration  number  k+1  the
                        condition |F(k+1)-F(k)| <= EpsF*max{|F(k)|, |F(k+1)|, 1}    is
                        satisfied.
            EpsX    -   positive number which  defines  a  precision  of  search.  The
                        subroutine finishes its work if on iteration number k+1    the
                        condition |X(k+1)-X(k)| <= EpsX is fulfilled.
            MaxIts  -   maximum number of iterations. If MaxIts=0, the number of
                        iterations is unlimited.

        OUTPUT PARAMETERS:
            State - structure used for reverse communication.

        This function  initializes  State   structure  with  default  optimization
        parameters (stopping conditions, step size, etc.).  Use  MinASASet??????()
        functions to tune optimization parameters.

        After   all   optimization   parameters   are   tuned,   you   should  use
        MinASAIteration() function to advance algorithm iterations.

        NOTES:

        1. you may tune stopping conditions with MinASASetCond() function
        2. if target function contains exp() or other fast growing functions,  and
           optimization algorithm makes too large steps which leads  to  overflow,
           use MinASASetStpMax() function to bound algorithm's steps.

          -- ALGLIB --
             Copyright 25.03.2010 by Bochkanov Sergey
        *************************************************************************/
        public static void minasacreate(int n,
            ref double[] x,
            ref double[] bndl,
            ref double[] bndu,
            ref minasastate state)
        {
            int i = 0;
            int i_ = 0;

            System.Diagnostics.Debug.Assert(n>=1, "MinASA: N too small!");
            for(i=0; i<=n-1; i++)
            {
                System.Diagnostics.Debug.Assert((double)(bndl[i])<=(double)(bndu[i]), "MinASA: inconsistent bounds!");
                System.Diagnostics.Debug.Assert((double)(bndl[i])<=(double)(x[i]), "MinASA: infeasible X!");
                System.Diagnostics.Debug.Assert((double)(x[i])<=(double)(bndu[i]), "MinASA: infeasible X!");
            }
            
            //
            // Initialize
            //
            state.n = n;
            minasasetcond(ref state, 0, 0, 0, 0);
            minasasetxrep(ref state, false);
            minasasetstpmax(ref state, 0);
            minasasetalgorithm(ref state, -1);
            state.bndl = new double[n];
            state.bndu = new double[n];
            state.ak = new double[n];
            state.xk = new double[n];
            state.dk = new double[n];
            state.an = new double[n];
            state.xn = new double[n];
            state.dn = new double[n];
            state.x = new double[n];
            state.d = new double[n];
            state.g = new double[n];
            state.gc = new double[n];
            state.work = new double[n];
            state.yk = new double[n];
            for(i_=0; i_<=n-1;i_++)
            {
//.........这里部分代码省略.........

        /*************************************************************************
        Obsolete optimization algorithm.
        Was replaced by MinBLEIC subpackage.

          -- ALGLIB --
             Copyright 02.04.2010 by Bochkanov Sergey
        *************************************************************************/
        public static void minasasetcond(minasastate state,
            double epsg,
            double epsf,
            double epsx,
            int maxits)
        {
            alglib.ap.assert(math.isfinite(epsg), "MinASASetCond: EpsG is not finite number!");
            alglib.ap.assert((double)(epsg)>=(double)(0), "MinASASetCond: negative EpsG!");
            alglib.ap.assert(math.isfinite(epsf), "MinASASetCond: EpsF is not finite number!");
            alglib.ap.assert((double)(epsf)>=(double)(0), "MinASASetCond: negative EpsF!");
            alglib.ap.assert(math.isfinite(epsx), "MinASASetCond: EpsX is not finite number!");
            alglib.ap.assert((double)(epsx)>=(double)(0), "MinASASetCond: negative EpsX!");
            alglib.ap.assert(maxits>=0, "MinASASetCond: negative MaxIts!");
            if( (((double)(epsg)==(double)(0) && (double)(epsf)==(double)(0)) && (double)(epsx)==(double)(0)) && maxits==0 )
            {
                epsx = 1.0E-6;
            }
            state.epsg = epsg;
            state.epsf = epsf;
            state.epsx = epsx;
            state.maxits = maxits;
        }

        /*************************************************************************
        Obsolete optimization algorithm.
        Was replaced by MinBLEIC subpackage.

          -- ALGLIB --
             Copyright 02.04.2010 by Bochkanov Sergey
        *************************************************************************/
        public static void minasasetxrep(minasastate state,
            bool needxrep)
        {
            state.xrep = needxrep;
        }

 public override alglib.apobject make_copy()
 {
     minasastate _result = new minasastate();
     _result.n = n;
     _result.epsg = epsg;
     _result.epsf = epsf;
     _result.epsx = epsx;
     _result.maxits = maxits;
     _result.xrep = xrep;
     _result.stpmax = stpmax;
     _result.cgtype = cgtype;
     _result.k = k;
     _result.nfev = nfev;
     _result.mcstage = mcstage;
     _result.bndl = (double[])bndl.Clone();
     _result.bndu = (double[])bndu.Clone();
     _result.curalgo = curalgo;
     _result.acount = acount;
     _result.mu = mu;
     _result.finit = finit;
     _result.dginit = dginit;
     _result.ak = (double[])ak.Clone();
     _result.xk = (double[])xk.Clone();
     _result.dk = (double[])dk.Clone();
     _result.an = (double[])an.Clone();
     _result.xn = (double[])xn.Clone();
     _result.dn = (double[])dn.Clone();
     _result.d = (double[])d.Clone();
     _result.fold = fold;
     _result.stp = stp;
     _result.work = (double[])work.Clone();
     _result.yk = (double[])yk.Clone();
     _result.gc = (double[])gc.Clone();
     _result.laststep = laststep;
     _result.x = (double[])x.Clone();
     _result.f = f;
     _result.g = (double[])g.Clone();
     _result.needfg = needfg;
     _result.xupdated = xupdated;
     _result.rstate = (rcommstate)rstate.make_copy();
     _result.repiterationscount = repiterationscount;
     _result.repnfev = repnfev;
     _result.repterminationtype = repterminationtype;
     _result.debugrestartscount = debugrestartscount;
     _result.lstate = (linmin.linminstate)lstate.make_copy();
     _result.betahs = betahs;
     _result.betady = betady;
     return _result;
 }

        /*************************************************************************
        Obsolete optimization algorithm.
        Was replaced by MinBLEIC subpackage.

          -- ALGLIB --
             Copyright 25.03.2010 by Bochkanov Sergey
        *************************************************************************/
        public static void minasacreate(int n,
            double[] x,
            double[] bndl,
            double[] bndu,
            minasastate state)
        {
            int i = 0;

            alglib.ap.assert(n>=1, "MinASA: N too small!");
            alglib.ap.assert(alglib.ap.len(x)>=n, "MinCGCreate: Length(X)<N!");
            alglib.ap.assert(apserv.isfinitevector(x, n), "MinCGCreate: X contains infinite or NaN values!");
            alglib.ap.assert(alglib.ap.len(bndl)>=n, "MinCGCreate: Length(BndL)<N!");
            alglib.ap.assert(apserv.isfinitevector(bndl, n), "MinCGCreate: BndL contains infinite or NaN values!");
            alglib.ap.assert(alglib.ap.len(bndu)>=n, "MinCGCreate: Length(BndU)<N!");
            alglib.ap.assert(apserv.isfinitevector(bndu, n), "MinCGCreate: BndU contains infinite or NaN values!");
            for(i=0; i<=n-1; i++)
            {
                alglib.ap.assert((double)(bndl[i])<=(double)(bndu[i]), "MinASA: inconsistent bounds!");
                alglib.ap.assert((double)(bndl[i])<=(double)(x[i]), "MinASA: infeasible X!");
                alglib.ap.assert((double)(x[i])<=(double)(bndu[i]), "MinASA: infeasible X!");
            }
            
            //
            // Initialize
            //
            state.n = n;
            minasasetcond(state, 0, 0, 0, 0);
            minasasetxrep(state, false);
            minasasetstpmax(state, 0);
            minasasetalgorithm(state, -1);
            state.bndl = new double[n];
            state.bndu = new double[n];
            state.ak = new double[n];
            state.xk = new double[n];
            state.dk = new double[n];
            state.an = new double[n];
            state.xn = new double[n];
            state.dn = new double[n];
            state.x = new double[n];
            state.d = new double[n];
            state.g = new double[n];
            state.gc = new double[n];
            state.work = new double[n];
            state.yk = new double[n];
            minasarestartfrom(state, x, bndl, bndu);
        }

        /*************************************************************************
        This function sets maximum step length

        INPUT PARAMETERS:
            State   -   structure which stores algorithm state between calls and
                        which is used for reverse communication. Must be
                        initialized with MinCGCreate()
            StpMax  -   maximum step length, >=0. Set StpMax to 0.0,  if you don't
                        want to limit step length.

        Use this subroutine when you optimize target function which contains exp()
        or  other  fast  growing  functions,  and optimization algorithm makes too
        large  steps  which  leads  to overflow. This function allows us to reject
        steps  that  are  too  large  (and  therefore  expose  us  to the possible
        overflow) without actually calculating function value at the x+stp*d.

          -- ALGLIB --
             Copyright 02.04.2010 by Bochkanov Sergey
        *************************************************************************/
        public static void minasasetstpmax(ref minasastate state,
            double stpmax)
        {
            System.Diagnostics.Debug.Assert((double)(stpmax)>=(double)(0), "MinASASetStpMax: StpMax<0!");
            state.stpmax = stpmax;
        }

        /*************************************************************************
        This function sets optimization algorithm.

        INPUT PARAMETERS:
            State   -   structure which stores algorithm state between calls and
                        which is used for reverse communication. Must be
                        initialized with MinASACreate()
            UAType  -   algorithm type:
                        * -1    automatic selection of the best algorithm
                        * 0     DY (Dai and Yuan) algorithm
                        * 1     Hybrid DY-HS algorithm

          -- ALGLIB --
             Copyright 02.04.2010 by Bochkanov Sergey
        *************************************************************************/
        public static void minasasetalgorithm(ref minasastate state,
            int algotype)
        {
            System.Diagnostics.Debug.Assert(algotype>=-1 & algotype<=1, "MinASASetAlgorithm: incorrect AlgoType!");
            if( algotype==-1 )
            {
                algotype = 1;
            }
            state.cgtype = algotype;
        }

        /*************************************************************************
        Returns norm of GI(x).

        GI(x) is  a  gradient  vector  whose  components  associated  with  active
        constraints are zeroed. It  differs  from  bounded  anti-gradient  because
        components  of   GI(x)   are   zeroed  independently  of  sign(g[i]),  and
        anti-gradient's components are zeroed with respect to both constraint  and
        sign.

          -- ALGLIB --
             Copyright 20.03.2009 by Bochkanov Sergey
        *************************************************************************/
        private static double asaginorm(minasastate state)
        {
            double result = 0;
            int i = 0;

            result = 0;
            for(i=0; i<=state.n-1; i++)
            {
                if( (double)(state.x[i])!=(double)(state.bndl[i]) && (double)(state.x[i])!=(double)(state.bndu[i]) )
                {
                    result = result+math.sqr(state.g[i]);
                }
            }
            result = Math.Sqrt(result);
            return result;
        }

        /*************************************************************************
        Obsolete optimization algorithm.
        Was replaced by MinBLEIC subpackage.

          -- ALGLIB --
             Copyright 02.04.2010 by Bochkanov Sergey
        *************************************************************************/
        public static void minasasetalgorithm(minasastate state,
            int algotype)
        {
            alglib.ap.assert(algotype>=-1 && algotype<=1, "MinASASetAlgorithm: incorrect AlgoType!");
            if( algotype==-1 )
            {
                algotype = 1;
            }
            state.cgtype = algotype;
        }