GeomInputConvertor.cpp

Go to the documentation of this file.

// Copyright (c) 2005 - 2014 Marc de Kamps
// All rights reserved.
//
// Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
//
//    * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
//    * Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation 
//      and/or other materials provided with the distribution.
//    * Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software 
//      without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 
// WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY 
// DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF 
// USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING 
// NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
//      If you use this software in work leading to a scientific publication, you should include a reference there to
//      the 'currently valid reference', which can be found at http://miind.sourceforge.net
#include "../MPILib/include/AlgorithmInterface.hpp"
#include "GeomInputConvertor.hpp"
#include "MuSigmaScalarProduct.hpp"
#include "NeuronParameter.hpp"
#include "GeomLibException.hpp"


using namespace GeomLib;

GeomInputConvertor::GeomInputConvertor
(
    const NeuronParameter&                  par_neuron,
    const DiffusionParameter&               par_diff,
    const CurrentCompensationParameter&     par_curr,
    const std::vector<MPILib::Potential>&   vec_int,
    bool                                    b_force_small
):
_par_neuron         (par_neuron),
_par_diff           (par_diff),
_par_curr           (par_curr),
_vec_interpretation (vec_int),
_vec_direct         (0),
_vec_diffusion      (0),
_force_small_bins   (b_force_small)
{
}

MPILib::Number GeomInputConvertor::NumberDirect() const
{
    return _vec_direct.size();
}

void GeomInputConvertor::SetDiffusionParameters
(
    const MuSigma& par,
    InputParameterSet& set
) const
{
    double mu    = par._mu;
    double sigma = par._sigma;

    if (mu == 0.0 && sigma ==0.0){
        set._h_exc = 0.0;
        set._h_inh = 0.0;
        set._rate_exc = 0.0;
        set._rate_inh = 0.0;
        return;
    }

    MPILib::Potential h = (mu != 0.0) ? sigma*sigma/mu : std::numeric_limits<double>::max();

    if (mu != 0 && IsSingleDiffusionProcess(h) ){

        if (fabs(h) < 2*_min_step && _force_small_bins)
            throw GeomLibException("Increase the number of bins.");

        MPILib::Rate rate = mu*mu/(sigma*sigma*_par_neuron._tau);

        if ( h > 0 ){
            set._h_exc    = h;
            set._h_inh    = 0.0;
            set._rate_exc = rate;
            set._rate_inh = 0.0;
        }
        else {
            set._h_exc    = 0.0;
            set._h_inh    =  h; // removed - sign, MdK 04/08/2012; the step size need to be forwarded as a negative number
            set._rate_exc = 0.0;
            set._rate_inh = rate;
        }
    }
    else {
        double h = _par_diff._diffusion_jump*(_par_neuron._theta - _vec_interpretation[0]);

        if (fabs(h) < 2*_min_step && _force_small_bins)
            throw GeomLibException("Increase the number of bins.");

        double tau =  _par_neuron._tau;
        set._h_exc =  h;
        set._h_inh = -h;

        set._rate_exc = (sigma*sigma + h*mu)/(2*h*h*tau);
        set._rate_inh = (sigma*sigma - h*mu)/(2*h*h*tau);

        assert( set._rate_exc >= 0.0);
        assert( set._rate_inh >= 0.0);
    }
}

bool GeomInputConvertor::IsSingleDiffusionProcess(MPILib::Potential h) const
{
    return (fabs(h) < _par_diff._diffusion_limit*(_par_neuron._theta - _vec_interpretation[0]));
}

void GeomInputConvertor::SortConnectionvector
(
    const std::vector<MPILib::Rate>&                vec_rates,
    const std::vector<MPILib::DelayedConnection>&   vec_con,
    const std::vector<MPILib::NodeType>&            vec_type
)
{
    assert(vec_rates.size() == vec_con.size());
    assert(vec_type.size()  == vec_rates.size());

    // it is guaranteed that there is exist a parameter vector that is large enough
    if (_vec_set.size() == 0)
        // If all inputs are direct, still one extra input is needed for diffusion
        _vec_set = std::vector<InputParameterSet>(vec_type.size() + 1);

    _vec_direct.clear();
    _vec_diffusion.clear();

    for (MPILib::Index i = 0; i < vec_type.size(); i++)
        if (vec_type[i] == MPILib::EXCITATORY_GAUSSIAN ||
            vec_type[i] == MPILib::INHIBITORY_GAUSSIAN)
            _vec_diffusion.push_back(i);
        else
            _vec_direct.push_back(i);

    this->AddDiffusionParameter (vec_con, vec_rates);
    this->AddBurstParameters    (vec_con, vec_rates);
}

void GeomInputConvertor::AddBurstParameters
(
    const std::vector<MPILib::DelayedConnection>& vec_con,
    const std::vector<MPILib::Rate>& vec_rates
)
{
    MPILib::Index start = 1;
    for(auto i: _vec_direct){
        
        double h = vec_con[i]._efficacy;
        double N = vec_con[i]._number_of_connections;
        double rate = vec_rates[i];

        if (h >= 0 ){
            _vec_set[start]._h_exc    = h;
            _vec_set[start]._h_inh    = 0;
            _vec_set[start]._rate_exc = rate*N;
            _vec_set[start]._rate_inh = 0.0;
        } else
        {
            _vec_set[start]._h_exc   = 0;
            _vec_set[start]._h_inh   = h;
            _vec_set[start]._rate_exc = 0.0;
            _vec_set[start]._rate_inh = rate*N;
        }
        start++;
    }
}

void GeomInputConvertor::SortDiffusionInput
(
    const std::vector<MPILib::DelayedConnection>& vec_con,
    const std::vector<MPILib::Rate>& vec_rates,
    std::vector<MPILib::DelayedConnection>* p_vec_con_diff,
    std::vector<MPILib::Rate>* p_vec_rates_diff
)
{
    for (auto i: _vec_diffusion){
        p_vec_con_diff->push_back(vec_con[i]);
        p_vec_rates_diff->push_back(vec_rates[i]);
    }
}


void GeomInputConvertor::AddDiffusionParameter
(
    const std::vector<MPILib::DelayedConnection>& vec_con,
    const std::vector<MPILib::Rate>& vec_rates
)
{   
    std::vector<MPILib::DelayedConnection> vec_diff_con;
    std::vector<MPILib::Rate> vec_diff_rates;

    SortDiffusionInput(vec_con,vec_rates, &vec_diff_con,&vec_diff_rates);

    MuSigmaScalarProduct<MPILib::DelayedConnection> scalar;
    MuSigma par = 
        scalar.Evaluate
        (
            vec_diff_rates,
            vec_diff_con,
            _par_neuron._tau
        );

    par._mu    += _par_curr._I;
    par._sigma += _par_curr._sigma;

    SetDiffusionParameters(par,_vec_set[0]);
}

MPILib::Potential GeomInputConvertor::MinVal
(
    const std::vector<MPILib::Potential>& vec_interpretation
) const
{
    assert (vec_interpretation.size() > 0);
    double min = std::numeric_limits<double>::max();
    for (MPILib::Index i = 0; i < vec_interpretation.size() - 1; i++){
        double dif = fabs(vec_interpretation[i+1] - vec_interpretation[i]);
        if (min > dif)
            min = dif;
    }
    return min;
}