/*  dist_mvn.cc - MultiVariate Normal distribution class/structure

    Adrian Groves, FMRIB Image Analysis Group

    Copyright (C) 2007-2008 University of Oxford  */

/*  Part of FSL - FMRIB's Software Library
    http://www.fmrib.ox.ac.uk/fsl
    fsl@fmrib.ox.ac.uk
    
    Developed at FMRIB (Oxford Centre for Functional Magnetic Resonance
    Imaging of the Brain), Department of Clinical Neurology, Oxford
    University, Oxford, UK
    
    
    LICENCE
    
    FMRIB Software Library, Release 5.0 (c) 2012, The University of
    Oxford (the "Software")
    
    The Software remains the property of the University of Oxford ("the
    University").
    
    The Software is distributed "AS IS" under this Licence solely for
    non-commercial use in the hope that it will be useful, but in order
    that the University as a charitable foundation protects its assets for
    the benefit of its educational and research purposes, the University
    makes clear that no condition is made or to be implied, nor is any
    warranty given or to be implied, as to the accuracy of the Software,
    or that it will be suitable for any particular purpose or for use
    under any specific conditions. Furthermore, the University disclaims
    all responsibility for the use which is made of the Software. It
    further disclaims any liability for the outcomes arising from using
    the Software.
    
    The Licensee agrees to indemnify the University and hold the
    University harmless from and against any and all claims, damages and
    liabilities asserted by third parties (including claims for
    negligence) which arise directly or indirectly from the use of the
    Software or the sale of any products based on the Software.
    
    No part of the Software may be reproduced, modified, transmitted or
    transferred in any form or by any means, electronic or mechanical,
    without the express permission of the University. The permission of
    the University is not required if the said reproduction, modification,
    transmission or transference is done without financial return, the
    conditions of this Licence are imposed upon the receiver of the
    product, and all original and amended source code is included in any
    transmitted product. You may be held legally responsible for any
    copyright infringement that is caused or encouraged by your failure to
    abide by these terms and conditions.
    
    You are not permitted under this Licence to use this Software
    commercially. Use for which any financial return is received shall be
    defined as commercial use, and includes (1) integration of all or part
    of the source code or the Software into a product for sale or license
    by or on behalf of Licensee to third parties or (2) use of the
    Software or any derivative of it for research with the final aim of
    developing software products for sale or license to a third party or
    (3) use of the Software or any derivative of it for research with the
    final aim of developing non-software products for sale or license to a
    third party, or (4) use of the Software to provide any service to an
    external organisation for which payment is received. If you are
    interested in using the Software commercially, please contact Isis
    Innovation Limited ("Isis"), the technology transfer company of the
    University, to negotiate a licence. Contact details are:
    innovation@isis.ox.ac.uk quoting reference DE/9564. */

#include "dist_mvn.h"
#include "easyoptions.h"
#include "miscmaths/miscmaths.h"

using namespace NEWIMAGE;
using namespace Utilities;
using namespace MISCMATHS;

// Constructors

MVNDist::MVNDist()
{
  Tracer_Plus tr("MVNDist::MVNDist()");
  len = -1;
  precisionsValid = covarianceValid = false;
}

MVNDist::MVNDist(const MVNDist& from1, const MVNDist& from2)
{
  Tracer_Plus tr("MVNDist::MVNDist(from1,from2)");
  len = from1.len + from2.len;
  means = from1.means & from2.means;
  precisionsValid = false;
  
  // Always duplicate the covariances (even if this means some recalculation)
  // Otherwise if we use precisions.i(), zeros won't stay exactly zero
  covariance.ReSize(len);
  covariance = 0;
  covariance.SymSubMatrix(1, from1.len) = from1.GetCovariance();  
  covariance.SymSubMatrix(from1.len+1, from1.len+from2.len) = from2.GetCovariance();
  covarianceValid = true;

  assert(means.Nrows() == len);
}

const MVNDist& MVNDist::operator=(const MVNDist& from)
{
  // Not useful and dominates --debug-running-stack:
  // Tracer_Plus tr("MVNDist::operator=");

  assert(&from != NULL); // yes, this can happen.  References are but pointers in disguise...

  // Special case: assignment to self (is a no-op)
  if (&from == this)
    return *this;

  // Special case: assigned from an uninitialized MVNDist
  if (from.len == -1)
    {
      len = -1;
      precisionsValid = covarianceValid = false;
      // Note, might still be consuming large amounts of memory, even though
      // precisions & covariance are now inaccessible from the outside
      return *this;
    }

  assert(from.len == from.means.Nrows());
  SetSize(from.len);
  //  len = from.len;
  means = from.means;
  precisionsValid = from.precisionsValid;
  covarianceValid = from.covarianceValid;
  if (precisionsValid)
    precisions = from.precisions;
  //  else if (precisions.Nrows() != len)
  //    precisions.ReSize(len);

  if (covarianceValid)
    covariance = from.covariance;
  //  else if (covariance.Nrows() != len)
  //    covariance.ReSize(len);

  assert(means.Nrows() == len);

  return *this;
}

void MVNDist::CopyFromSubmatrix(const MVNDist& from, int first, int last, 
    bool checkIndependence)
{
    Tracer_Plus tr("MVNDist::CopyFromSubmatrix");
    len = last-first+1;
    means = from.means.Rows(first, last);
    precisionsValid = from.precisionsValid;
    covarianceValid = from.covarianceValid;
    if (precisionsValid)
        precisions = from.precisions.SymSubMatrix(first, last);
    else if (precisions.Nrows() != len)
        precisions.ReSize(len);
        
    if (covarianceValid)
        covariance = from.covariance.SymSubMatrix(first, last);
    else if (covariance.Nrows() != len)
        covariance.ReSize(len);
            
    assert(means.Nrows() == len);
    
    if (checkIndependence)
    {
        Matrix deps1 = from.GetCovariance().Rows(first, last).Columns(1, first-1);
        Matrix deps2 = from.GetCovariance().Rows(first, last).Columns(last+1, from.covariance.Ncols());
        if (!deps1.IsZero() || !deps2.IsZero() )
            throw Invalid_option("Covariance found in part of MVN that should be independent from the rest!");
    } 
    return;
}        

void MVNDist::SetSize(int dim)
{
  // Not useful and dominates --debug-running-stack:
  // Tracer_Plus tr("MVNDist::SetSize");
  if (dim<=0)
    throw RBD_COMMON::Logic_error("Can't have dim<=0\n");
    
  assert(means.Nrows() == len || len<0);    
    
  if (len != dim)
  {
    //Tracer_Plus tr("MVNDist::SetSize (actually resizing)");
    len = dim;
    means.ReSize(dim);
    precisions.ReSize(dim);
    covariance.ReSize(dim);
  }
  precisionsValid = false;
  covarianceValid = false;
  // means is also now undefined (or at least out-of-date)  

  assert(means.Nrows() == len);
  assert(precisions.Nrows() == len);
  assert(covariance.Nrows() == len);
}

// Accessors
const SymmetricMatrix& MVNDist::GetPrecisions() const
{
  Tracer_Plus tr("MVNDist::GetPrecisions");
  if (len == -1) throw Logic_error("MVN is uninitialized!\n");
  assert(means.Nrows() == len);
  if (!precisionsValid)
    {
      Tracer_Plus tr("MVNDist::GetPrecisions calculation");        
      assert(covarianceValid);
      // precisions and precisionsValid are mutable, 
      // so we can change them even in a const function
      precisions = covariance.i();
      precisionsValid = true;
    }
  assert(means.Nrows() == len);
  assert(precisions.Nrows() == len);
  return precisions;
}

const SymmetricMatrix& MVNDist::GetCovariance() const
{
  Tracer_Plus tr("MVNDist::GetCovariance");    
  if (len == -1) throw Logic_error("MVN is uninitialized!\n");
  assert(means.Nrows() == len);
  if (!covarianceValid)
    {
      Tracer_Plus tr("MVNDist::GetCovariance calculation");
      assert(precisionsValid);
      // covariance and covarianceValid are mutable, 
      // so we can change them even in a const function
      try 
        {
          covariance = precisions.i();
        } 
      catch (Exception)
        {
	  //          LOG_ERR("Oh dear, it didn't like that\nPrecisions = \n");
	  //          LOG_ERR(precisions);
	  //          LOG_ERR("Rethrowing...\n");
	  //          throw;
	  // Better behaviour: adds a tiny amount to the diagonal and tries again
	  Warning::IssueOnce("MVN precision (len==" + stringify(len) + ") was singular, adding 1e-10 to diagonal");
	  covariance = (precisions + IdentityMatrix(len)*1e-10).i();
        }
      covarianceValid = true;
    }
  assert(means.Nrows() == len);
  assert(covariance.Nrows() == len);
  return covariance;
}

void MVNDist::SetPrecisions(const SymmetricMatrix& from)
{
  Tracer_Plus tr("MVNDist::SetPrecisions");
  assert(from.Nrows() == len);
  assert(means.Nrows() == len);
  precisions = from;
  precisionsValid = true;
  covarianceValid = false;
  assert(means.Nrows() == len);
}

void MVNDist::SetCovariance(const SymmetricMatrix& from)
{
  Tracer_Plus tr("MVNDist::SetCovariance");
  //cout << from.Nrows() << " ---- " << len << endl;  
  assert(from.Nrows() == len);
  assert(means.Nrows() == len);
  covariance = from;
  covarianceValid = true;
  precisionsValid = false;
  assert(means.Nrows() == len);
}

void MVNDist::DumpTo(ostream& out, const string indent) const
{ 
  Tracer_Plus tr("MVNDist::Dump");
  out << indent << "MVNDist, with len == " << len 
       << ", precisionsValid == " << precisionsValid
       << ", covarianceValid == " << covarianceValid << endl;
  out << indent << "  Means: " << means.t();
  if (precisionsValid || covarianceValid)
    {
      out << indent << "  Covariance matrix:" << endl;
      for (int i=1; i<=len; i++)
	out << indent << "  " << GetCovariance().Row(i);
    }
  else
    out << indent << "  Covariance undefined." << endl;

  assert(means.Nrows() == len);
}

void MVNDist::Load(const string& filename)
{
    cout << "Reading MVN from file '" << filename << "'...\n";
    Matrix mat = read_vest(filename);
    
    cout << "Converting to an MVN...\n";
    // Format: [covariance means(:); means(:)' 1.0]
    const int N = mat.Nrows() - 1;
    
    if (N < 1 || mat != mat.t() || mat(N+1,N+1) != 1.0)
    {
        cout << "N == " << N << ", matrix:\n" << mat;
        throw Invalid_option("Inputted MVNs must be symmetric matrices!\nFormat = [covariance means(:); means(:)' 1.0]\n");
    }
    SetSize(N);
    means = mat.Column(len+1).Rows(1, N);
    SymmetricMatrix sym; 
    sym << mat.SubMatrix(1,N,1,N);
    SetCovariance(sym); 
    
    assert(means.Nrows() == len);
}

void MVNDist::Load(vector<MVNDist*>& mvns, const string& filename, const volume<float>& mask)
{
    Tracer_Plus tr("MVNDist::Load (static)");
    
    Matrix vols; 
    LOG_ERR("Reading MVNs from " << filename << endl);
 
    volume4D<float> input;
    read_volume4D(input,filename);
    vols=input.matrix(mask);
    
    const int nVoxels = vols.Ncols();
    for (unsigned i = 0; i < mvns.size(); i++) 
        assert(mvns[i] == NULL); // should've deleted everything first. 
        
    mvns.resize(nVoxels, NULL);
    const int nParams = ((int)sqrt(8*vols.Nrows()+1)-3)/2;
    assert( vols.Nrows() == nParams*(nParams+1)/2 + nParams+1 );
    
    SymmetricMatrix tmp(nParams);
    
//        cout << "--------\n";
//        cout << nParams << endl;
//        cout << nVoxels << endl;
//        cout << vols.Nrows() << endl;
//        cout << tmp.Nrows() << endl;
  
    assert(nVoxels > 0);
  
    for (int vox = 1; vox <= nVoxels; vox++)
    {
        assert(mvns.at(vox-1) == NULL);
        mvns[vox-1] = new MVNDist(nParams);
        //tmp << vols.Column(vox).Rows(1,nParams*(nParams+1)/2); // Doesn't work
	int index = 0;
        for (int r = 1; r <= nParams; r++)
	  for (int c = 1; c <= r; c++)
	    tmp(r,c) = vols(++index,vox);
	assert(index == nParams*(nParams+1)/2);
        mvns[vox-1]->SetCovariance(tmp);
        mvns[vox-1]->means = vols.Column(vox).Rows(nParams*(nParams+1)/2 + 1, nParams*(nParams+1)/2 + nParams);
        assert(vols(vols.Nrows(), vox) == 1);
        assert(mvns[vox-1]->means.Nrows() == mvns[vox-1]->len);
    } 
    
    assert(mvns[0] != NULL);

}

void MVNDist::Save(const vector<MVNDist*>& mvns, const string& filename, const volume<float>& mask)
{    
    Tracer_Plus tr("MVNDist::Save");
     
    // Save the MVNs in a NIFTI file as a single NIFTI_INTENT_SYMMATRIX 
    // last row/col is the means (1 in the corner).
    // Note that I'm using the 4th dim and should really be using the 5th,
    // according to the specification -- but I don't think it really matters.

    Matrix vols;

    const int nVoxels = mvns.size();
    assert(nVoxels > 0 && mvns.at(0) != NULL);   
    const int nParams = mvns.at(0)->means.Nrows();
    
//    means.ReSize(nParams, nVoxels);
//    precisions.ReSize(nParams*(nParams+1)/2, nVoxels);
    vols.ReSize(nParams*(nParams+1)/2 + nParams+1, nVoxels);
    
    ColumnVector aOne(1); aOne = 1.0;
    
    for (int vox = 1; vox <= nVoxels; vox++)
      {
        vols.Column(vox) = 
            mvns.at(vox-1)->GetCovariance().AsColumn()
            & mvns.at(vox-1)->means & aOne;
        // AsColumn uses row ordering on the lower triangular part, 
        // as NIFTI_INTENT_SYMMATRIX specifies: (1,1) (2,1) (2,2) (3,1)...
      }
    // Write the file
    volume4D<float> output(mask.xsize(),mask.ysize(),mask.zsize(),vols.Nrows());
    output.setmatrix(vols,mask);
    output.set_intent(NIFTI_INTENT_SYMMATRIX,0,0,0);
    output.setDisplayMaximumMinimum(output.max(),output.min());
    save_volume4D(output,filename);
}