gurls::ParamSelLooGPRegr< T > Class Template Reference

ParamSelLooGPRegr is the sub-class of ParamSelection that implements.

#include <loogpregr.h>

Inheritance diagram for gurls::ParamSelLooGPRegr< T >:

Collaboration diagram for gurls::ParamSelLooGPRegr< T >:

List of all members.

Public Member Functions
GurlsOptionsList *	execute (const gMat2D< T > &X, const gMat2D< T > &Y, const GurlsOptionsList &opt)
	Performs parameter selection for Gaussian Process regression.
Static Public Member Functions
static ParamSelection< T > *	factory (const std::string &id) throw (BadParamSelectionCreation)
	Factory function returning a pointer to the newly created object.

---

Detailed Description

template<typename T>
class gurls::ParamSelLooGPRegr< T >

Definition at line 68 of file loogpregr.h.

---

Member Function Documentation

template<typename T >

GurlsOptionsList * gurls::ParamSelLooGPRegr< T >::execute	(	const gMat2D< T > &	X,
		const gMat2D< T > &	Y,
		const GurlsOptionsList &	opt
	)		[virtual]

The leave-one-out approach is used.

Parameters:

X	input data matrix
Y	labels matrix
opt	options with the following: nlambda (default) hoperf (default) split (settable with the class Split and its subclasses) kernel (settable with the class Kernel and its subclasses)

Returns:

paramsel, a GurlsOptionList with the following fields:

• lambdas = array of values of the regularization parameter lambda minimizing the validation error for each class
• guesses = array of guesses for the regularization parameter lambda
• forho = matrix of validation accuracies for each lambda guess and for each class

Implements gurls::ParamSelection< T >.

Definition at line 91 of file loogpregr.h.

{
//    [n,T]  = size(y);
    const unsigned long n = Y.rows();
    const unsigned long t = Y.cols();

    const unsigned long d = X.cols();

//    tot = opt.nlambda;
    int tot = static_cast<int>(opt.getOptAsNumber("nlambda"));

//    K = opt.kernel.K;
    const gMat2D<T> &K = opt.getOptValue<OptMatrix<gMat2D<T> > >("kernel.K");


//    lmax = mean(std(y));

//    T* work = new T[t+n];
//    T* stdY = new T[t];

//    stdDev(Y.getData(), n, t, stdY, work);

//    const T lmax = sumv(stdY, t)/((T)t);

//    delete[] work;
//    delete[] stdY;

//    const T lmin = lmax * (T)1.0e-5;

    T lmin;
    T lmax;

    if(opt.hasOpt("lambdamin"))
        lmin = opt.getOptAsNumber("lambdamin");
    else
        lmin = 0.001;

    if(opt.hasOpt("lambdamax"))
        lmax = opt.getOptAsNumber("lambdamax");
    else
        lmax = 10;

//    guesses = lmin.*(lmax/lmin).^linspace(0,1,tot);
    gMat2D<T> *guesses_mat = new gMat2D<T>(tot, 1);
    T* guesses = guesses_mat->getData();

    T* linspc = new T[tot];
    linspace((T)0.0, (T)1.0, tot, linspc);
    const T coeff = lmax/lmin;

    for(int i=0; i< tot; ++i)
        guesses[i] = lmin* std::pow(coeff, linspc[i]);

    delete[] linspc;


//    perf = zeros(tot,T);
    gMat2D<T> *perf_mat = new gMat2D<T>(tot, t);
    T* perf = perf_mat->getData();
    set(perf, (T)0.0, tot*t);

    const int tr_size = n-1;

    unsigned long* tr = new unsigned long[tr_size+1]; // + 1 cell for convenience
    unsigned long* tr_it = tr;
    for(unsigned long i=1; i< n; ++i, ++tr_it)
        *tr_it = i;

    GurlsOptionsList* nestedOpt = new GurlsOptionsList("nested");
    nestedOpt->copyOpt("singlelambda", opt);


    gMat2D<T>* tmpK = new gMat2D<T>(tr_size, tr_size);
    gMat2D<T>* tmpPredK = new gMat2D<T>(1, tr_size);
    gMat2D<T>* tmpPredKTest = new gMat2D<T>(1, 1);

    GurlsOptionsList* tmpPredKernel = new GurlsOptionsList("predkernel");
    GurlsOptionsList* tmpKernel = new GurlsOptionsList("kernel");
    GurlsOptionsList* tmpParamSel = new GurlsOptionsList("paramsel");

    nestedOpt->addOpt("kernel", tmpKernel);
    nestedOpt->addOpt("predkernel", tmpPredKernel);
    nestedOpt->addOpt("paramsel", tmpParamSel);

    tmpKernel->addOpt("K", new OptMatrix<gMat2D<T> > (*tmpK));
    tmpPredKernel->addOpt("K", new OptMatrix<gMat2D<T> > (*tmpPredK));
    tmpPredKernel->addOpt("Ktest", new OptMatrix<gMat2D<T> > (*tmpPredKTest));

    gMat2D<T> rlsX(tr_size, d);
    gMat2D<T> rlsY(tr_size, t);

//    T* tmpMat = new T[ tr_size * std::max(d, t)];

    gMat2D<T> predX(1, d);
    gMat2D<T> predY(1, t);

    RLSGPRegr<T> rlsgp;
    PredGPRegr<T> predgp;
    Performance<T>* perfClass = Performance<T>::factory(opt.getOptAsString("hoperf"));

    gMat2D<T> *lambda = new gMat2D<T>(1,1);
    tmpParamSel->addOpt("lambdas", new OptMatrix<gMat2D<T> >(*lambda));

//    for k = 1:n;
    for(unsigned long k = 0; k<n; ++k)
    {
//        tr = setdiff(1:n,k);

//        opt.kernel.K = K(tr,tr);
        copy_submatrix(tmpK->getData(), K.getData(), K.rows(), tr_size, tr_size, tr, tr);

//        opt.predkernel.K = K(k,tr);
        copy_submatrix(tmpPredK->getData(), K.getData(), K.rows(), 1, tr_size, &k , tr);

//        opt.predkernel.Ktest = K(k,k);
        tmpPredKTest->getData()[0] = K.getData()[(k*K.rows()) + k];

//        for i = 1:tot
        for(int i=0; i< tot; ++i)
        {
//            opt.paramsel.noises = guesses(i);
            lambda->getData()[0] = guesses[i];

//            opt.rls = rls_gpregr(X(tr,:),y(tr,:),opt);
            subMatrixFromRows(X.getData(), n, d, tr, tr_size, rlsX.getData());

            subMatrixFromRows(Y.getData(), n, t, tr, tr_size, rlsY.getData());

            GurlsOptionsList* ret_rlsgp = rlsgp.execute(rlsX, rlsY, *nestedOpt);

            nestedOpt->removeOpt("optimizer");
            nestedOpt->addOpt("optimizer", ret_rlsgp);

//            tmp = pred_gpregr(X(k,:),y(k,:),opt);
            getRow(X.getData(), n, d, k, predX.getData());
            getRow(Y.getData(), n, t, k, predY.getData());

            GurlsOptionsList * pred_list = predgp.execute(predX, predY, *nestedOpt);

//            opt.pred = tmp.means;
            nestedOpt->removeOpt("pred");
            nestedOpt->addOpt("pred", pred_list->getOpt("means"));

            pred_list->removeOpt("means", false);

            delete pred_list;

//            opt.perf = opt.hoperf([],y(k,:),opt);
            GurlsOptionsList * perf_list = perfClass->execute(predX, predY, *nestedOpt);

            gMat2D<T>& forho = perf_list->getOptValue<OptMatrix<gMat2D<T> > >("forho");

//            for t = 1:T
            for(unsigned long j = 0; j<t; ++j)
//                perf(i,t) = opt.perf.forho(t)+perf(i,t)./n;
//                perf(i,t) = opt.perf.forho(t)./n+perf(i,t);
                perf[i+(tot*j)] += forho.getData()[j]/n;
                //perf[i+(tot*j)] = forho.getData()[j]/n+perf[i+(tot*j)];


            delete perf_list;
        }

        tr[k] = k;

    }

    delete perfClass;

    delete nestedOpt;


    GurlsOptionsList* paramsel;

    if(opt.hasOpt("paramsel"))
    {
        GurlsOptionsList* tmp_opt = new GurlsOptionsList("tmp");
        tmp_opt->copyOpt("paramsel", opt);

        paramsel = tmp_opt->getOptAs<GurlsOptionsList>("paramsel");
        tmp_opt->removeOpt("paramsel", false);
        delete tmp_opt;

        paramsel->removeOpt("lambdas");
        paramsel->removeOpt("perf");
        paramsel->removeOpt("guesses");
    }
    else
        paramsel = new GurlsOptionsList("paramsel");


//    [dummy,idx] = max(perf,[],1);
    unsigned long* idx = new unsigned long[t];
    T* work = NULL;
    indicesOfMax(perf, tot, t, idx, work, 1);


//    vout.noises =     guesses(idx);
    gMat2D<T> *lambdas = new gMat2D<T>(1, t);
    copyLocations(idx, guesses, t, tot, lambdas->getData());

    delete[] idx;

    paramsel->addOpt("lambdas", new OptMatrix<gMat2D<T> >(*lambdas));

//    vout.perf =   perf;
    paramsel->addOpt("perf", new OptMatrix<gMat2D<T> >(*perf_mat));

//    vout.guesses = guesses;
    paramsel->addOpt("guesses", new OptMatrix<gMat2D<T> >(*guesses_mat));

    return paramsel;
}

template<typename T>

static ParamSelection<T>* gurls::ParamSelection< T >::factory

(

const std::string &

id

)

throw (BadParamSelectionCreation) [inline, static, inherited]

Warning:

The returned pointer is a plain, un-managed pointer. The calling function is responsible of deallocating the object.

Definition at line 146 of file paramsel.h.

    {
        if(id == "loocvprimal")
            return new ParamSelLoocvPrimal<T>;
        if(id == "loocvdual")
            return new ParamSelLoocvDual<T>;
        if(id == "fixlambda")
            return new ParamSelFixLambda<T>;
        if(id == "calibratesgd")
            return new ParamSelCalibrateSGD<T>;
        if(id == "siglam")
            return new ParamSelSiglam<T>;
        if(id == "siglamho")
            return new ParamSelSiglamHo<T>;
        if(id == "hodual")
            return new ParamSelHoDual<T>;
        if(id == "hodualr")
            return new ParamSelHoDualr<T>;
        if(id == "hoprimal")
            return new ParamSelHoPrimal<T>;
        if(id == "hoprimalr")
            return new ParamSelHoPrimalr<T>;
        if(id == "fixsiglam")
            return new ParamSelFixSigLam<T>;
        if(id == "loogpregr")
            return new ParamSelLooGPRegr<T>;
        if(id == "hogpregr")
            return new ParamSelHoGPRegr<T>;
        if(id == "siglamloogpregr")
            return new ParamSelSiglamLooGPRegr<T>;
        if(id == "siglamhogpregr")
            return new ParamSelSiglamHoGPRegr<T>;

        throw BadParamSelectionCreation(id);
    }

---

The documentation for this class was generated from the following file:

• loogpregr.h