Scalar F_cranic(int n, double *wt, Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* titer = ext->fn[0];
Func<Scalar>* tprev = ext->fn[1];
for (int i = 0; i < n; i++)
result += wt[i] * (HEATCAP * (titer->val[i] - tprev->val[i]) * v->val[i] / TAU +
0.5 * lam(titer->val[i]) * (titer->dx[i] * v->dx[i] + titer->dy[i] * v->dy[i]) +
0.5 * lam(tprev->val[i]) * (tprev->dx[i] * v->dx[i] + tprev->dy[i] * v->dy[i]));
return result;
}
Scalar F_cranic(int n, double *wt, Func<Real> *u_ext[], Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* u_prev_newton = u_ext[0];
Func<Scalar>* u_prev_time = ext->fn[0];
for (int i = 0; i < n; i++)
result += wt[i] * ((u_prev_newton->val[i] - u_prev_time->val[i]) * v->val[i] / TAU
+ 0.5 * lam(u_prev_newton->val[i]) * (u_prev_newton->dx[i] * v->dx[i] + u_prev_newton->dy[i] * v->dy[i])
+ 0.5 * lam(u_prev_time->val[i]) * (u_prev_time->dx[i] * v->dx[i] + u_prev_time->dy[i] * v->dy[i])
- heat_src(e->x[i], e->y[i]) * v->val[i]);
return result;
}
static
long image(ZZ& det, mat_ZZ& B, mat_ZZ* U, long verbose)
{
long m = B.NumRows();
long n = B.NumCols();
long force_reduce = 1;
vec_long P;
P.SetLength(m);
vec_ZZ D;
D.SetLength(m+1);
D[0] = 1;
vec_vec_ZZ lam;
lam.SetLength(m);
long j;
for (j = 1; j <= m; j++)
lam(j).SetLength(m);
if (U) ident(*U, m);
long s = 0;
long k = 1;
long max_k = 0;
while (k <= m) {
if (k > max_k) {
IncrementalGS(B, P, D, lam, s, k);
max_k = k;
}
if (k == 1) {
force_reduce = 1;
k++;
continue;
}
if (force_reduce)
for (j = k-1; j >= 1; j--)
reduce(k, j, B, P, D, lam, U);
if (P(k-1) != 0 && P(k) == 0) {
force_reduce = swap(k, B, P, D, lam, U, max_k, verbose);
k--;
}
else {
force_reduce = 1;
k++;
}
}
det = D[s];
return s;
}
TEST(Interface, Simple){
linear_ip::Vector c(5) ,b(3);
linear_ip::Matrix A(3, 5);
b << 6, 12, 10;
c << 2, 1, 0, 0, 0;
A << 1, 1, -1, 0, 0,
1, 4, 0, -1, 0,
1, 2, 0, 0, 1;
linear_ip::lp P(c, A, b);
P.solve();
linear_ip::Vector x(5), s(5), lam(3);
x << 2, 4, 0, 6, 0;
s << 0, 0, 3, 0, 1;
lam << 3, 0, -1;
EXPECT_TRUE(x.isApprox(P.get_x(), 1e-5));
EXPECT_TRUE(s.isApprox(P.get_s(), 1e-5));
EXPECT_TRUE(lam.isApprox(P.get_lam(), 1e-5));
}
void
bar (A b)
{
void (*lam) (A) = [](A) { baz (); };
if (auto c = b)
lam (c);
}
Scalar res(int n, double *wt, Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* u_prev = ext->fn[0];
for (int i = 0; i < n; i++)
result += wt[i] * (lam(u_prev->val[i]) * (u_prev->dx[i] * v->dx[i] + u_prev->dy[i] * v->dy[i])
- heat_src(e->x[i], e->y[i]) * v->val[i]);
return result;
}
Scalar bilinear_form(int n, double *wt, Func<Real> *u_ext[], Func<Real> *u, Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* u_prev = ext->fn[0];
for (int i = 0; i < n; i++)
result += wt[i] * (lam(u_prev->val[i]) * (u->dx[i] * v->dx[i] + u->dy[i] * v->dy[i]));
return result;
}
Scalar res_ss(int n, double *wt, Func<Real> *u_ext[], Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext, double t)
{
Scalar result = 0;
Func<Scalar>* Y_prev_newton = u_ext[0];
for (int i = 0; i < n; i++) {
result += wt[i] * (lam(Y_prev_newton->val[i]) * (Y_prev_newton->dx[i] * v->dx[i] + Y_prev_newton->dy[i] * v->dy[i])
- heat_src(e->x[i], e->y[i], t) * v->val[i]);
}
return result;
}
Scalar jac(int n, double *wt, Func<Real> *u, Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* u_prev = ext->fn[0];
for (int i = 0; i < n; i++)
result += wt[i] * (dlam_du(u_prev->val[i]) * u->val[i] * (u_prev->dx[i] * v->dx[i] + u_prev->dy[i] * v->dy[i])
+ lam(u_prev->val[i]) * (u->dx[i] * v->dx[i] + u->dy[i] * v->dy[i]));
return result;
}
Scalar J_euler(int n, double *wt, Func<Real> *u, Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* titer = ext->fn[0];
for (int i = 0; i < n; i++)
result += wt[i] * (HEATCAP * u->val[i] * v->val[i] / TAU +
dlam_dT(titer->val[i]) * u->val[i] * (titer->dx[i] * v->dx[i] + titer->dy[i] * v->dy[i]) +
lam(titer->val[i]) * (u->dx[i] * v->dx[i] + u->dy[i] * v->dy[i]));
return result;
}
Scalar J_cranic(int n, double *wt, Func<Real> *u_ext[], Func<Real> *u, Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* u_prev_newton = u_ext[0];
for (int i = 0; i < n; i++)
result += wt[i] * (u->val[i] * v->val[i] / TAU +
0.5 * dlam_du(u_prev_newton->val[i]) * u->val[i] * (u_prev_newton->dx[i] * v->dx[i] + u_prev_newton->dy[i] * v->dy[i])
+ 0.5 * lam(u_prev_newton->val[i]) * (u->dx[i] * v->dx[i] + u->dy[i] * v->dy[i]));
return result;
}
Scalar jac_sdirk(int n, double *wt, Func<Real> *u_ext[], Func<Real> *u, Func<Real> *v,
Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* u_prev_newton = u_ext[0];
for (int i = 0; i < n; i++)
result += wt[i] * (u->val[i] * v->val[i] / TAU
+ BUTCHER_A_11 * (dlam_du(u_prev_newton->val[i]) * u->val[i]
* (u_prev_newton->dx[i] * v->dx[i] + u_prev_newton->dy[i] * v->dy[i])
+ lam(u_prev_newton->val[i]) * (u->dx[i] * v->dx[i] + u->dy[i] * v->dy[i])));
return result;
}
Scalar res1(int n, double *wt, Func<Real> *u_ext[], Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* Y1_prev_newton = u_ext[0];
Func<Scalar>* u_prev_time = ext->fn[0];
for (int i = 0; i < n; i++)
result += wt[i] * ((Y1_prev_newton->val[i] - u_prev_time->val[i]) * v->val[i]
+ GAMMA * TAU * (lam(Y1_prev_newton->val[i]) * (Y1_prev_newton->dx[i] * v->dx[i]
+ Y1_prev_newton->dy[i] * v->dy[i])
- heat_src(e->x[i], e->y[i], TIME+GAMMA*TAU) * v->val[i]));
return result;
}
Scalar jac2(int n, double *wt, Func<Real> *u_ext[], Func<Real> *u, Func<Real> *v, Geom<Real> *e, ExtData<Scalar> *ext)
{
Scalar result = 0;
Func<Scalar>* Y2_prev_newton = u_ext[0];
for (int i = 0; i < n; i++)
result += wt[i] * (u->val[i] * v->val[i]
+ GAMMA * TAU * (dlam_du(Y2_prev_newton->val[i]) * u->val[i] * (Y2_prev_newton->dx[i] * v->dx[i]
+ Y2_prev_newton->dy[i] * v->dy[i])
+ lam(Y2_prev_newton->val[i]) * (u->dx[i] * v->dx[i]
+ u->dy[i] * v->dy[i])));
return result;
}
void ProjectToEdge (const Surface * f1, const Surface * f2, Point<3> & hp)
{
Vec<2> rs, lam;
Vec<3> a1, a2;
Mat<2> a;
int i = 10;
while (i > 0)
{
i--;
rs(0) = f1 -> CalcFunctionValue (hp);
rs(1) = f2 -> CalcFunctionValue (hp);
f1->CalcGradient (hp, a1);
f2->CalcGradient (hp, a2);
double alpha = fabs(a1*a2)/sqrt(a1.Length2()*a2.Length2());
if(fabs(1.-alpha) < 1e-6)
{
if(fabs(rs(0)) >= fabs(rs(1)))
f1 -> Project(hp);
else
f2 -> Project(hp);
}
else
{
a(0,0) = a1 * a1;
a(0,1) = a(1,0) = a1 * a2;
a(1,1) = a2 * a2;
a.Solve (rs, lam);
hp -= lam(0) * a1 + lam(1) * a2;
}
if (Abs2 (rs) < 1e-24 && i > 1) i = 1;
}
}
Scalar stac_residual(int n, double *wt, Func<Real> *u_ext[], Func<Real> *v,
Geom<Real> *e, ExtData<Scalar> *ext)
{
Func<Scalar>* u_prev = u_ext[0];
// This is a temporary workaround. The stage time t_n + h * c_i
// can be accessed via u_stage_time->val[0];
// In this particular case the stage time is not needed as
// the form does not depend explicitly on time.
Func<Scalar>* u_stage_time = ext->fn[0];
// Stationary part of the residual (time derivative term left out).
Scalar result1 = 0, result2 = 0;
for (int i = 0; i < n; i++) {
result1 = result1 - wt[i] * lam(u_prev->val[i]) * (u_prev->dx[i] * v->dx[i] + u_prev->dy[i] * v->dy[i]);
result2 = result2 + wt[i] * heat_src(e->x[i], e->y[i]) * v->val[i];
}
return result1 + result2;
}
Scalar stac_jacobian(int n, double *wt, Func<Real> *u_ext[], Func<Real> *u, Func<Real> *v,
Geom<Real> *e, ExtData<Scalar> *ext)
{
Func<Scalar>* K_sln = u_ext[0];
Func<Scalar>* sln_prev_time = ext->fn[0];
// This is a temporary workaround. The stage time t_n + h * c_i
// can be accessed via u_stage_time->val[0];
// In this particular case the stage time is not needed as
// the form does not depend explicitly on time.
//Func<Scalar>* u_stage_time = ext->fn[1];
// Stationary part of the Jacobian matrix (time derivative term left out).
Scalar result1 = 0, result2 = 0;
for (int i = 0; i < n; i++) {
Scalar sln_val_i = sln_prev_time->val[i] + K_sln->val[i];
Scalar sln_dx_i = sln_prev_time->dx[i] + K_sln->dx[i];
Scalar sln_dy_i = sln_prev_time->dy[i] + K_sln->dy[i];
result1 += -wt[i] * dlam_du(sln_val_i) * u->val[i] * (sln_dx_i * v->dx[i] + sln_dy_i * v->dy[i]);
result2 += -wt[i] * lam(sln_val_i) * (u->dx[i] * v->dx[i] + u->dy[i] * v->dy[i]);
}
return result1 + result2;
}
double ExponentialModel::sim(RNG &rng) const { return rexp_mt(rng, lam()); }
double PoissonModel::logp(int x) const { return dpois(x, lam(), true); }
double PoissonModel::simdat(RNG &rng) const { return rpois_mt(rng, lam()); }
double PoissonModel::pdf(uint x, bool logscale) const{
return dpois(x, lam(), logscale); }
void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[])
{
register int i;
float x;
int max_len = 6;
int p;
int ns0 = 29;
static area_function *af0;
if (CheckArguments(nlhs, plhs, nrhs, prhs) ){
mexErrMsgTxt("AMgetdata argument checking failed.");
return ;
}
/* read nasal tract area function from the file */
//printf("Reading nasal tract file.\n");
read_af(NTAFpath , &nna, &afnt );
//printf("Finished Reading nasal tract file.\n");
/* Initialization */
//read_rad();
// update all constants and variables
if((int)(pTCcfg[0]) == RL_CIRCUIT) rad_boundary = RL_CIRCUIT;
else rad_boundary = SHORT_CIRCUIT;
//printf("Rad_boundary[%d]\n",rad_boundary);
if((int)(pTCcfg[1]) == YIELDING) wall = YIELDING;
else wall = RIGID;
//printf("wall[%d]\n",wall);
if((int)(pTCcfg[2]) == ON) nasal_tract = ON;
else nasal_tract = OFF;
//printf("nasal_tract[%d]\n",nasal_tract);
if((int)(pTCcfg[3]) == CLOSE) glt_boundary = CLOSE;
else glt_boundary = OPEN;
//printf("glt_boundary[%d]\n",glt_boundary);
ro = (float)(pPCcfg[0]);
//printf("Air density[%f]\n",ro);
c = (float)(pPCcfg[1]);
//printf("Sound velocity[%f]\n",c);
wall_resi = (float)(pPCcfg[2]);
//printf("wall_resi[%f]\n",wall_resi);
wall_mass = (float)(pPCcfg[3]);
//printf("wall_mass[%f]\n",wall_mass);
wall_comp = (float)(pPCcfg[4]);
//printf("wall_comp[%f]\n",wall_comp);
for(i=0;i<7;i++)
{
AMpar[i] = (float)(pAMcfg[i]);
//printf("AMpar[%d][%f]\n",i,AMpar[i]);
}
anc = (float)(pAMcfg[7]);
//printf("nasal area[%f]\n",anc);
/* Initialization */
read_model_spec();
af0 = (area_function *) calloc( ns0, sizeof(area_function) );
nph = 9;
nbu = 8;
nss = nbu + nph;
afvt = (area_function *) calloc( nss, sizeof(area_function) );
convert_scale();
semi_polar();
oAf = mxCreateDoubleMatrix(2,nss, mxREAL);
pAf = mxGetPr(oAf);
/* Compute VT profile and area function, and plot them */
lam( AMpar ); /* profile */
sagittal_to_area( &ns0, af0 ); /* area function */
for(i=0;i<ns0;i++)
{
//printf("AreaNS0[%d][%f]\n",i,af0[i]);
}
appro_area_function( ns0, af0, nss, afvt);
for(i=0;i<nss;i++)
{
//printf("AreaAF[%d][%f]\n",i,afvt[i]);
pAf[2*i] = afvt[i].A;
pAf[2*i+1] = afvt[i].x;
}
if( nasal_tract == ON )
{
anc = (float) min( anc, afvt[nph].A );
afvt[nph].A -= anc;
pAf[2*nph] = afvt[nph].A;
pAf[2*nph+1] = afvt[nph].x;
}
calplot_tf_FBA(nfrmmax, frm, bw, amp, &nfrms, tfunc, &ntf);
//printf("Finished calculations.\n");
oPdat1 = mxCreateDoubleMatrix(4,NP, mxREAL);
pPdat1 = mxGetPr(oPdat1);
for(i=0;i<NP;i++)
{
pPdat1[4*i] = ivt[i].x;
pPdat1[4*i+1] = ivt[i].y;
pPdat1[4*i+2] = evt[i].x;
pPdat1[4*i+3] = evt[i].y;
}
oPdat2 = mxCreateDoubleMatrix(1,2,mxREAL);
pPdat2 = mxGetPr(oPdat2);
pPdat2[0] = lip_w;
pPdat2[1] = lip_h;
oTf = mxCreateDoubleMatrix(ntf, 1, mxREAL);
pTf = mxGetPr(oTf);
for(i=0;i<ntf;i++)
{
pTf[i] = tfunc[i];
}
oFmt = mxCreateDoubleMatrix(nfrms, 1, mxREAL);
pFmt = mxGetPr(oFmt);
oBw = mxCreateDoubleMatrix(nfrms, 1, mxREAL);
pBw = mxGetPr(oBw);
oAmp = mxCreateDoubleMatrix(nfrms, 1, mxREAL);
pAmp = mxGetPr(oAmp);
for(i=0;i<nfrms;i++)
{
pFmt[i] = frm[i];
pBw[i] = bw[i];
pAmp[i] = amp[i];
}
/* Assign output pointers */
plhs[0] = oAf;
plhs[1] = oTf;
plhs[2] = oFmt;
plhs[3] = oBw;
plhs[4] = oAmp;
plhs[5] = oPdat1;
plhs[6] = oPdat2;
// do cleanup
//free( rad_re );
//free( rad_im );
free( afnt );
free( af0 );
free( afvt );
}
bool Mesh::interpolateElementCentered(uint elementIndex, double x, double y, double* value, const ElementOutput* output, const ValueAccessor* accessor) const
{
const Mesh* mesh = output->dataSet->mesh();
const Element& elem = mesh->elements()[elementIndex];
if (elem.eType() == Element::ENP)
{
const Node* nodes = projectedNodes();
QPolygonF pX(elem.nodeCount());
QVector<double> lam(elem.nodeCount());
for (int i=0; i<elem.nodeCount(); i++) {
pX[i] = nodes[elem.p(i)].toPointF();
}
if (!ENP_physicalToLogical(pX, QPointF(x, y), lam))
return false;
*value = accessor->value(elementIndex);
return true;
}
else if (elem.eType() == Element::E4Q)
{
int e4qIndex = mE4QtmpIndex[elementIndex];
E4Qtmp& e4q = mE4Qtmp[e4qIndex];
double Lx, Ly;
if (!E4Q_mapPhysicalToLogical(e4q, x, y, Lx, Ly, *mE4Qnorm))
return false;
if (Lx < 0 || Ly < 0 || Lx > 1 || Ly > 1)
return false;
*value = accessor->value(elementIndex);
return true;
}
else if (elem.eType() == Element::E3T)
{
const Node* nodes = projectedNodes();
double lam1, lam2, lam3;
if (!E3T_physicalToBarycentric(nodes[elem.p(0)].toPointF(),
nodes[elem.p(1)].toPointF(),
nodes[elem.p(2)].toPointF(),
QPointF(x, y),
lam1, lam2, lam3))
return false;
// Now interpolate
*value = accessor->value(elementIndex);
return true;
}
else if (elem.eType() == Element::E2L)
{
const Node* nodes = projectedNodes();
double lam;
if (!E2L_physicalToLogical(nodes[elem.p(0)].toPointF(),
nodes[elem.p(1)].toPointF(),
QPointF(x, y),
lam))
return false;
// Now interpolate
*value = accessor->value(elementIndex);
return true;
}
else
{
Q_ASSERT(0 && "unknown element type");
return false;
}
}
bool Mesh::interpolate(uint elementIndex, double x, double y, double* value, const NodeOutput* output, const ValueAccessor* accessor) const
{
const Mesh* mesh = output->dataSet->mesh();
const Element& elem = mesh->elements()[elementIndex];
if (elem.eType() == Element::ENP)
{
const Node* nodes = projectedNodes();
QPolygonF pX(elem.nodeCount());
QVector<double> lam(elem.nodeCount());
for (int i=0; i<elem.nodeCount(); i++) {
pX[i] = nodes[elem.p(i)].toPointF();
}
if (!ENP_physicalToLogical(pX, QPointF(x, y), lam))
return false;
*value = 0;
for (int i=0; i<elem.nodeCount(); i++) {
*value += lam[i] * accessor->value( elem.p(i) );
}
return true;
}
else if (elem.eType() == Element::E4Q)
{
int e4qIndex = mE4QtmpIndex[elementIndex];
E4Qtmp& e4q = mE4Qtmp[e4qIndex];
double Lx, Ly;
if (!E4Q_mapPhysicalToLogical(e4q, x, y, Lx, Ly, *mE4Qnorm))
return false;
if (Lx < 0 || Ly < 0 || Lx > 1 || Ly > 1)
return false;
double q11 = accessor->value( elem.p(2) );
double q12 = accessor->value( elem.p(1) );
double q21 = accessor->value( elem.p(3) );
double q22 = accessor->value( elem.p(0) );
*value = q11*Lx*Ly + q21*(1-Lx)*Ly + q12*Lx*(1-Ly) + q22*(1-Lx)*(1-Ly);
return true;
}
else if (elem.eType() == Element::E3T)
{
/*
So - we're interpoalting a 3-noded triangle
The query point must be within the bounding box of this triangl but not nessisarily
within the triangle itself.
*/
/*
First determine if the point of interest lies within the triangle using Barycentric techniques
described here: http://www.blackpawn.com/texts/pointinpoly/
*/
const Node* nodes = projectedNodes();
double lam1, lam2, lam3;
if (!E3T_physicalToBarycentric(nodes[elem.p(0)].toPointF(),
nodes[elem.p(1)].toPointF(),
nodes[elem.p(2)].toPointF(),
QPointF(x, y),
lam1, lam2, lam3))
return false;
// Now interpolate
double z1 = accessor->value( elem.p(0));
double z2 = accessor->value( elem.p(1));
double z3 = accessor->value( elem.p(2));
*value = lam1 * z3 + lam2 * z2 + lam3 * z1;
return true;
}
else if (elem.eType() == Element::E2L)
{
/*
So - we're interpoalting a 2-noded line
*/
const Node* nodes = projectedNodes();
double lam;
if (!E2L_physicalToLogical(nodes[elem.p(0)].toPointF(),
nodes[elem.p(1)].toPointF(),
QPointF(x, y),
lam))
return false;
// Now interpolate
double z1 = accessor->value( elem.p(0) );
double z2 = accessor->value( elem.p(1) );
*value = z1 + lam * (z2 - z1);
return true;
}
else
{
Q_ASSERT(0 && "unknown element type");
return false;
}
}
double PoissonModel::pdf(const Data * dp, bool logscale) const{
return dpois(DAT(dp)->value(), lam(), logscale); }
double PoissonModel::mean()const{return lam();}
double PoissonModel::var()const{return lam();}
double PoissonModel::sd()const{return sqrt(lam());}
double PoissonModel::simdat()const{
return rpois(lam());}
void NLPSolverInternal::init(){
// Read options
verbose_ = getOption("verbose");
gauss_newton_ = getOption("gauss_newton");
// Initialize the functions
casadi_assert_message(!F_.isNull(),"No objective function");
if(!F_.isInit()){
F_.init();
log("Objective function initialized");
}
if(!G_.isNull() && !G_.isInit()){
G_.init();
log("Constraint function initialized");
}
// Get dimensions
n_ = F_.input(0).numel();
m_ = G_.isNull() ? 0 : G_.output(0).numel();
parametric_ = getOption("parametric");
if (parametric_) {
casadi_assert_message(F_.getNumInputs()==2, "Wrong number of input arguments to F for parametric NLP. Must be 2, but got " << F_.getNumInputs());
} else {
casadi_assert_message(F_.getNumInputs()==1, "Wrong number of input arguments to F for non-parametric NLP. Must be 1, but got " << F_.getNumInputs() << " instead. Do you perhaps intend to use fixed parameters? Then use the 'parametric' option.");
}
// Basic sanity checks
casadi_assert_message(F_.getNumInputs()==1 || F_.getNumInputs()==2, "Wrong number of input arguments to F. Must be 1 or 2");
if (F_.getNumInputs()==2) parametric_=true;
casadi_assert_message(getOption("ignore_check_vec") || gauss_newton_ || F_.input().size2()==1,
"To avoid confusion, the input argument to F must be vector. You supplied " << F_.input().dimString() << endl <<
" We suggest you make the following changes:" << endl <<
" - F is an SXFunction: SXFunction([X],[rhs]) -> SXFunction([vec(X)],[rhs])" << endl <<
" or F - -> F = vec(F) " <<
" - F is an MXFunction: MXFunction([X],[rhs]) -> " << endl <<
" X_vec = MX(\"X\",vec(X.sparsity())) " << endl <<
" F_vec = MXFunction([X_flat],[F.call([X_flat.reshape(X.sparsity())])[0]]) " << endl <<
" or F - -> F = vec(F) " <<
" You may ignore this warning by setting the 'ignore_check_vec' option to true." << endl
);
casadi_assert_message(F_.getNumOutputs()>=1, "Wrong number of output arguments to F");
casadi_assert_message(gauss_newton_ || F_.output().scalar(), "Output argument of F not scalar.");
casadi_assert_message(F_.output().dense(), "Output argument of F not dense.");
casadi_assert_message(F_.input().dense(), "Input argument of F must be dense. You supplied " << F_.input().dimString());
if(!G_.isNull()) {
if (parametric_) {
casadi_assert_message(G_.getNumInputs()==2, "Wrong number of input arguments to G for parametric NLP. Must be 2, but got " << G_.getNumInputs());
} else {
casadi_assert_message(G_.getNumInputs()==1, "Wrong number of input arguments to G for non-parametric NLP. Must be 1, but got " << G_.getNumInputs() << " instead. Do you perhaps intend to use fixed parameters? Then use the 'parametric' option.");
}
casadi_assert_message(G_.getNumOutputs()>=1, "Wrong number of output arguments to G");
casadi_assert_message(G_.input().numel()==n_, "Inconsistent dimensions");
casadi_assert_message(G_.input().sparsity()==F_.input().sparsity(), "F and G input dimension must match. F " << F_.input().dimString() << ". G " << G_.input().dimString());
}
// Find out if we are to expand the objective function in terms of scalar operations
bool expand_f = getOption("expand_f");
if(expand_f){
log("Expanding objective function");
// Cast to MXFunction
MXFunction F_mx = shared_cast<MXFunction>(F_);
if(F_mx.isNull()){
casadi_warning("Cannot expand objective function as it is not an MXFunction");
} else {
// Take use the input scheme of G if possible (it might be an SXFunction)
vector<SXMatrix> inputv;
if(!G_.isNull() && F_.getNumInputs()==G_.getNumInputs()){
inputv = G_.symbolicInputSX();
} else {
inputv = F_.symbolicInputSX();
}
// Try to expand the MXFunction
F_ = F_mx.expand(inputv);
F_.setOption("number_of_fwd_dir",F_mx.getOption("number_of_fwd_dir"));
F_.setOption("number_of_adj_dir",F_mx.getOption("number_of_adj_dir"));
F_.init();
}
}
// Find out if we are to expand the constraint function in terms of scalar operations
bool expand_g = getOption("expand_g");
if(expand_g){
log("Expanding constraint function");
// Cast to MXFunction
MXFunction G_mx = shared_cast<MXFunction>(G_);
if(G_mx.isNull()){
casadi_warning("Cannot expand constraint function as it is not an MXFunction");
} else {
// Take use the input scheme of F if possible (it might be an SXFunction)
vector<SXMatrix> inputv;
if(F_.getNumInputs()==G_.getNumInputs()){
inputv = F_.symbolicInputSX();
} else {
inputv = G_.symbolicInputSX();
}
// Try to expand the MXFunction
G_ = G_mx.expand(inputv);
G_.setOption("number_of_fwd_dir",G_mx.getOption("number_of_fwd_dir"));
G_.setOption("number_of_adj_dir",G_mx.getOption("number_of_adj_dir"));
G_.init();
}
}
// Find out if we are to expand the constraint function in terms of scalar operations
bool generate_hessian = getOption("generate_hessian");
if(generate_hessian && H_.isNull()){
casadi_assert_message(!gauss_newton_,"Automatic generation of Gauss-Newton Hessian not yet supported");
log("generating hessian");
// Simple if unconstrained
if(G_.isNull()){
// Create Hessian of the objective function
FX HF = F_.hessian();
HF.init();
// Symbolic inputs of HF
vector<MX> HF_in = F_.symbolicInput();
// Lagrange multipliers
MX lam("lam",0);
// Objective function scaling
MX sigma("sigma");
// Inputs of the Hessian function
vector<MX> H_in = HF_in;
H_in.insert(H_in.begin()+1, lam);
H_in.insert(H_in.begin()+2, sigma);
// Get an expression for the Hessian of F
MX hf = HF.call(HF_in).at(0);
// Create the scaled Hessian function
H_ = MXFunction(H_in, sigma*hf);
log("Unconstrained Hessian function generated");
} else { // G_.isNull()
// Check if the functions are SXFunctions
SXFunction F_sx = shared_cast<SXFunction>(F_);
SXFunction G_sx = shared_cast<SXFunction>(G_);
// Efficient if both functions are SXFunction
if(!F_sx.isNull() && !G_sx.isNull()){
// Expression for f and g
SXMatrix f = F_sx.outputSX();
SXMatrix g = G_sx.outputSX();
// Numeric hessian
bool f_num_hess = F_sx.getOption("numeric_hessian");
bool g_num_hess = G_sx.getOption("numeric_hessian");
// Number of derivative directions
int f_num_fwd = F_sx.getOption("number_of_fwd_dir");
int g_num_fwd = G_sx.getOption("number_of_fwd_dir");
int f_num_adj = F_sx.getOption("number_of_adj_dir");
int g_num_adj = G_sx.getOption("number_of_adj_dir");
// Substitute symbolic variables in f if different input variables from g
if(!isEqual(F_sx.inputSX(),G_sx.inputSX())){
f = substitute(f,F_sx.inputSX(),G_sx.inputSX());
}
// Lagrange multipliers
SXMatrix lam = ssym("lambda",g.size1());
// Objective function scaling
SXMatrix sigma = ssym("sigma");
// Lagrangian function
vector<SXMatrix> lfcn_in(parametric_? 4: 3);
lfcn_in[0] = G_sx.inputSX();
lfcn_in[1] = lam;
lfcn_in[2] = sigma;
if (parametric_) lfcn_in[3] = G_sx.inputSX(1);
SXFunction lfcn(lfcn_in, sigma*f + inner_prod(lam,g));
lfcn.setOption("verbose",getOption("verbose"));
lfcn.setOption("numeric_hessian",f_num_hess || g_num_hess);
lfcn.setOption("number_of_fwd_dir",std::min(f_num_fwd,g_num_fwd));
lfcn.setOption("number_of_adj_dir",std::min(f_num_adj,g_num_adj));
lfcn.init();
// Hessian of the Lagrangian
H_ = static_cast<FX&>(lfcn).hessian();
H_.setOption("verbose",getOption("verbose"));
log("SX Hessian function generated");
} else { // !F_sx.isNull() && !G_sx.isNull()
// Check if the functions are SXFunctions
MXFunction F_mx = shared_cast<MXFunction>(F_);
MXFunction G_mx = shared_cast<MXFunction>(G_);
// If they are, check if the arguments are the same
if(!F_mx.isNull() && !G_mx.isNull() && isEqual(F_mx.inputMX(),G_mx.inputMX())){
casadi_warning("Exact Hessian calculation for MX is still experimental");
// Expression for f and g
MX f = F_mx.outputMX();
MX g = G_mx.outputMX();
// Lagrange multipliers
MX lam("lam",g.size1());
// Objective function scaling
MX sigma("sigma");
// Inputs of the Lagrangian function
vector<MX> lfcn_in(parametric_? 4:3);
lfcn_in[0] = G_mx.inputMX();
lfcn_in[1] = lam;
lfcn_in[2] = sigma;
if (parametric_) lfcn_in[3] = G_mx.inputMX(1);
// Lagrangian function
MXFunction lfcn(lfcn_in,sigma*f+ inner_prod(lam,g));
lfcn.init();
log("SX Lagrangian function generated");
/* cout << "countNodes(lfcn.outputMX()) = " << countNodes(lfcn.outputMX()) << endl;*/
bool adjoint_mode = true;
if(adjoint_mode){
// Gradient of the lagrangian
MX gL = lfcn.grad();
log("MX Lagrangian gradient generated");
MXFunction glfcn(lfcn_in,gL);
glfcn.init();
log("MX Lagrangian gradient function initialized");
// cout << "countNodes(glfcn.outputMX()) = " << countNodes(glfcn.outputMX()) << endl;
// Get Hessian sparsity
CRSSparsity H_sp = glfcn.jacSparsity();
log("MX Lagrangian Hessian sparsity determined");
// Uni-directional coloring (note, the hessian is symmetric)
CRSSparsity coloring = H_sp.unidirectionalColoring(H_sp);
log("MX Lagrangian Hessian coloring determined");
// Number of colors needed is the number of rows
int nfwd_glfcn = coloring.size1();
log("MX Lagrangian gradient function number of sensitivity directions determined");
glfcn.setOption("number_of_fwd_dir",nfwd_glfcn);
glfcn.updateNumSens();
log("MX Lagrangian gradient function number of sensitivity directions updated");
// Hessian of the Lagrangian
H_ = glfcn.jacobian();
} else {
// Hessian of the Lagrangian
H_ = lfcn.hessian();
}
log("MX Lagrangian Hessian function generated");
} else {
casadi_assert_message(0, "Automatic calculation of exact Hessian currently only for F and G both SXFunction or MXFunction ");
}
} // !F_sx.isNull() && !G_sx.isNull()
} // G_.isNull()
} // generate_hessian && H_.isNull()
if(!H_.isNull() && !H_.isInit()) {
H_.init();
log("Hessian function initialized");
}
// Create a Jacobian if it does not already exists
bool generate_jacobian = getOption("generate_jacobian");
if(generate_jacobian && !G_.isNull() && J_.isNull()){
log("Generating Jacobian");
J_ = G_.jacobian();
// Use live variables if SXFunction
if(!shared_cast<SXFunction>(J_).isNull()){
J_.setOption("live_variables",true);
}
log("Jacobian function generated");
}
if(!J_.isNull() && !J_.isInit()){
J_.init();
log("Jacobian function initialized");
}
if(!H_.isNull()) {
if (parametric_) {
casadi_assert_message(H_.getNumInputs()>=2, "Wrong number of input arguments to H for parametric NLP. Must be at least 2, but got " << G_.getNumInputs());
} else {
casadi_assert_message(H_.getNumInputs()>=1, "Wrong number of input arguments to H for non-parametric NLP. Must be at least 1, but got " << G_.getNumInputs() << " instead. Do you perhaps intend to use fixed parameters? Then use the 'parametric' option.");
}
casadi_assert_message(H_.getNumOutputs()>=1, "Wrong number of output arguments to H");
casadi_assert_message(H_.input(0).numel()==n_,"Inconsistent dimensions");
casadi_assert_message(H_.output().size1()==n_,"Inconsistent dimensions");
casadi_assert_message(H_.output().size2()==n_,"Inconsistent dimensions");
}
if(!J_.isNull()){
if (parametric_) {
casadi_assert_message(J_.getNumInputs()==2, "Wrong number of input arguments to J for parametric NLP. Must be at least 2, but got " << G_.getNumInputs());
} else {
casadi_assert_message(J_.getNumInputs()==1, "Wrong number of input arguments to J for non-parametric NLP. Must be at least 1, but got " << G_.getNumInputs() << " instead. Do you perhaps intend to use fixed parameters? Then use the 'parametric' option.");
}
casadi_assert_message(J_.getNumOutputs()>=1, "Wrong number of output arguments to J");
casadi_assert_message(J_.input().numel()==n_,"Inconsistent dimensions");
casadi_assert_message(J_.output().size2()==n_,"Inconsistent dimensions");
}
if (parametric_) {
sp_p = F_->input(1).sparsity();
if (!G_.isNull()) casadi_assert_message(sp_p == G_->input(G_->getNumInputs()-1).sparsity(),"Parametric NLP has inconsistent parameter dimensions. F has got " << sp_p.dimString() << " as dimensions, while G has got " << G_->input(G_->getNumInputs()-1).dimString());
if (!H_.isNull()) casadi_assert_message(sp_p == H_->input(H_->getNumInputs()-1).sparsity(),"Parametric NLP has inconsistent parameter dimensions. F has got " << sp_p.dimString() << " as dimensions, while H has got " << H_->input(H_->getNumInputs()-1).dimString());
if (!J_.isNull()) casadi_assert_message(sp_p == J_->input(J_->getNumInputs()-1).sparsity(),"Parametric NLP has inconsistent parameter dimensions. F has got " << sp_p.dimString() << " as dimensions, while J has got " << J_->input(J_->getNumInputs()-1).dimString());
}
// Infinity
double inf = numeric_limits<double>::infinity();
// Allocate space for inputs
input_.resize(NLP_NUM_IN - (parametric_? 0 : 1));
input(NLP_X_INIT) = DMatrix(n_,1,0);
input(NLP_LBX) = DMatrix(n_,1,-inf);
input(NLP_UBX) = DMatrix(n_,1, inf);
input(NLP_LBG) = DMatrix(m_,1,-inf);
input(NLP_UBG) = DMatrix(m_,1, inf);
input(NLP_LAMBDA_INIT) = DMatrix(m_,1,0);
if (parametric_) input(NLP_P) = DMatrix(sp_p,0);
// Allocate space for outputs
output_.resize(NLP_NUM_OUT);
output(NLP_X_OPT) = DMatrix(n_,1,0);
output(NLP_COST) = DMatrix(1,1,0);
output(NLP_LAMBDA_X) = DMatrix(n_,1,0);
output(NLP_LAMBDA_G) = DMatrix(m_,1,0);
output(NLP_G) = DMatrix(m_,1,0);
if (hasSetOption("iteration_callback")) {
callback_ = getOption("iteration_callback");
if (!callback_.isNull()) {
if (!callback_.isInit()) callback_.init();
casadi_assert_message(callback_.getNumOutputs()==1, "Callback function should have one output, a scalar that indicates wether to break. 0 = continue");
casadi_assert_message(callback_.output(0).size()==1, "Callback function should have one output, a scalar that indicates wether to break. 0 = continue");
casadi_assert_message(callback_.getNumInputs()==NLP_NUM_OUT, "Callback function should have the output scheme of NLPSolver as input scheme. i.e. " <<NLP_NUM_OUT << " inputs instead of the " << callback_.getNumInputs() << " you provided." );
for (int i=0;i<NLP_NUM_OUT;i++) {
casadi_assert_message(callback_.input(i).sparsity()==output(i).sparsity(),
"Callback function should have the output scheme of NLPSolver as input scheme. " <<
"Input #" << i << " (" << getSchemeEntryEnumName(SCHEME_NLPOutput,i) << " aka '" << getSchemeEntryName(SCHEME_NLPOutput,i) << "') was found to be " << callback_.input(i).dimString() << " instead of expected " << output(i).dimString() << "."
);
callback_.input(i).setAll(0);
}
}
}
callback_step_ = getOption("iteration_callback_step");
// Call the initialization method of the base class
FXInternal::init();
}
