/*@
TaoLineSearchComputeObjective - Computes the objective function value at a given point
Collective on TaoLineSearch
Input Parameters:
+ ls - the TaoLineSearch context
- x - input vector
Output Parameter:
. f - Objective value at X
Notes: TaoLineSearchComputeObjective() is typically used within line searches
so most users would not generally call this routine themselves.
Level: developer
.seealso: TaoLineSearchComputeGradient(), TaoLineSearchComputeObjectiveAndGradient(), TaoLineSearchSetObjectiveRoutine()
@*/
PetscErrorCode TaoLineSearchComputeObjective(TaoLineSearch ls, Vec x, PetscReal *f)
{
PetscErrorCode ierr;
Vec gdummy;
PetscReal gts;
PetscFunctionBegin;
PetscValidHeaderSpecific(ls,TAOLINESEARCH_CLASSID,1);
PetscValidHeaderSpecific(x,VEC_CLASSID,2);
PetscValidPointer(f,3);
PetscCheckSameComm(ls,1,x,2);
if (ls->usetaoroutines) {
ierr = TaoComputeObjective(ls->tao,x,f);CHKERRQ(ierr);
} else {
ierr = PetscLogEventBegin(TaoLineSearch_EvalEvent,ls,0,0,0);CHKERRQ(ierr);
if (!ls->ops->computeobjective && !ls->ops->computeobjectiveandgradient && !ls->ops->computeobjectiveandgts) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Line Search does not have objective function set");
PetscStackPush("TaoLineSearch user objective routine");
if (ls->ops->computeobjective) {
ierr = (*ls->ops->computeobjective)(ls,x,f,ls->userctx_func);CHKERRQ(ierr);
} else if (ls->ops->computeobjectiveandgradient) {
ierr = VecDuplicate(x,&gdummy);CHKERRQ(ierr);
ierr = (*ls->ops->computeobjectiveandgradient)(ls,x,f,gdummy,ls->userctx_funcgrad);CHKERRQ(ierr);
ierr = VecDestroy(&gdummy);CHKERRQ(ierr);
} else {
ierr = (*ls->ops->computeobjectiveandgts)(ls,x,ls->stepdirection,f,>s,ls->userctx_funcgts);CHKERRQ(ierr);
}
PetscStackPop;
ierr = PetscLogEventEnd(TaoLineSearch_EvalEvent,ls,0,0,0);CHKERRQ(ierr);
}
ls->nfeval++;
PetscFunctionReturn(0);
}
/*@C
TaoDefaultComputeGradient - computes the gradient using finite differences.
Collective on Tao
Input Parameters:
+ tao - the Tao context
. X - compute gradient at this point
- dummy - not used
Output Parameters:
. G - Gradient Vector
Options Database Key:
+ -tao_fd_gradient - activates TaoDefaultComputeGradient()
- -tao_fd_delta <delta> - change in X used to calculate finite differences
Level: advanced
Notes:
This routine is slow and expensive, and is not currently optimized
to take advantage of sparsity in the problem. Although
TaoDefaultComputeGradient is not recommended for general use
in large-scale applications, It can be useful in checking the
correctness of a user-provided gradient. Use the tao method TAOTEST
to get an indication of whether your gradient is correct.
This finite difference gradient evaluation can be set using the routine TaoSetGradientRoutine() or by using the command line option -tao_fd_gradient
.seealso: TaoSetGradientRoutine()
@*/
PetscErrorCode TaoDefaultComputeGradient(Tao tao,Vec Xin,Vec G,void *dummy)
{
Vec X;
PetscScalar *g;
PetscReal f, f2;
PetscErrorCode ierr;
PetscInt low,high,N,i;
PetscBool flg;
PetscReal h=.5*PETSC_SQRT_MACHINE_EPSILON;
PetscFunctionBegin;
ierr = PetscOptionsGetReal(((PetscObject)tao)->options,((PetscObject)tao)->prefix,"-tao_fd_delta",&h,&flg);CHKERRQ(ierr);
ierr = VecDuplicate(Xin,&X);CHKERRQ(ierr);
ierr = VecCopy(Xin,X);CHKERRQ(ierr);
ierr = VecGetSize(X,&N);CHKERRQ(ierr);
ierr = VecGetOwnershipRange(X,&low,&high);CHKERRQ(ierr);
ierr = VecSetOption(X,VEC_IGNORE_OFF_PROC_ENTRIES,PETSC_TRUE);CHKERRQ(ierr);
ierr = VecGetArray(G,&g);CHKERRQ(ierr);
for (i=0;i<N;i++) {
ierr = VecSetValue(X,i,-h,ADD_VALUES);CHKERRQ(ierr);
ierr = VecAssemblyBegin(X);CHKERRQ(ierr);
ierr = VecAssemblyEnd(X);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao,X,&f);CHKERRQ(ierr);
ierr = VecSetValue(X,i,2.0*h,ADD_VALUES);CHKERRQ(ierr);
ierr = VecAssemblyBegin(X);CHKERRQ(ierr);
ierr = VecAssemblyEnd(X);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao,X,&f2);CHKERRQ(ierr);
ierr = VecSetValue(X,i,-h,ADD_VALUES);CHKERRQ(ierr);
ierr = VecAssemblyBegin(X);CHKERRQ(ierr);
ierr = VecAssemblyEnd(X);CHKERRQ(ierr);
if (i>=low && i<high) {
g[i-low]=(f2-f)/(2.0*h);
}
}
ierr = VecRestoreArray(G,&g);CHKERRQ(ierr);
ierr = VecDestroy(&X);CHKERRQ(ierr);
PetscFunctionReturn(0);
}
/*@C
TaoDefaultComputeGradient - computes the gradient using finite differences.
Collective on Tao
Input Parameters:
+ tao - the Tao context
. X - compute gradient at this point
- dummy - not used
Output Parameters:
. G - Gradient Vector
Options Database Key:
+ -tao_fd_gradient - Activates TaoDefaultComputeGradient()
- -tao_fd_delta <delta> - change in x used to calculate finite differences
Level: advanced
Note:
This routine is slow and expensive, and is not currently optimized
to take advantage of sparsity in the problem. Although
TaoAppDefaultComputeGradient is not recommended for general use
in large-scale applications, It can be useful in checking the
correctness of a user-provided gradient. Use the tao method TAOTEST
to get an indication of whether your gradient is correct.
Note:
This finite difference gradient evaluation can be set using the routine TaoSetGradientRoutine() or by using the command line option -tao_fd_gradient
.seealso: TaoSetGradientRoutine()
@*/
PetscErrorCode TaoDefaultComputeGradient(Tao tao,Vec X,Vec G,void *dummy)
{
PetscScalar *x,*g;
PetscReal f, f2;
PetscErrorCode ierr;
PetscInt low,high,N,i;
PetscBool flg;
PetscReal h=PETSC_SQRT_MACHINE_EPSILON;
PetscFunctionBegin;
ierr = TaoComputeObjective(tao, X,&f);CHKERRQ(ierr);
ierr = PetscOptionsGetReal(NULL,"-tao_fd_delta",&h,&flg);CHKERRQ(ierr);
ierr = VecGetSize(X,&N);CHKERRQ(ierr);
ierr = VecGetOwnershipRange(X,&low,&high);CHKERRQ(ierr);
ierr = VecGetArray(G,&g);CHKERRQ(ierr);
for (i=0;i<N;i++) {
if (i>=low && i<high) {
ierr = VecGetArray(X,&x);CHKERRQ(ierr);
x[i-low] += h;
ierr = VecRestoreArray(X,&x);CHKERRQ(ierr);
}
ierr = TaoComputeObjective(tao,X,&f2);CHKERRQ(ierr);
if (i>=low && i<high) {
ierr = VecGetArray(X,&x);CHKERRQ(ierr);
x[i-low] -= h;
ierr = VecRestoreArray(X,&x);CHKERRQ(ierr);
}
if (i>=low && i<high) {
g[i-low]=(f2-f)/h;
}
}
ierr = VecRestoreArray(G,&g);CHKERRQ(ierr);
PetscFunctionReturn(0);
}
static PetscErrorCode TaoSolve_NM(Tao tao)
{
PetscErrorCode ierr;
TAO_NelderMead *nm = (TAO_NelderMead*)tao->data;
TaoConvergedReason reason;
PetscReal *x;
PetscInt i;
Vec Xmur=nm->Xmur, Xmue=nm->Xmue, Xmuc=nm->Xmuc, Xbar=nm->Xbar;
PetscReal fr,fe,fc;
PetscInt shrink;
PetscInt low,high;
PetscFunctionBegin;
nm->nshrink = 0;
nm->nreflect = 0;
nm->nincontract = 0;
nm->noutcontract = 0;
nm->nexpand = 0;
if (tao->XL || tao->XU || tao->ops->computebounds) {
ierr = PetscPrintf(((PetscObject)tao)->comm,"WARNING: Variable bounds have been set but will be ignored by NelderMead algorithm\n");CHKERRQ(ierr);
}
ierr = VecCopy(tao->solution,nm->simplex[0]);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao,nm->simplex[0],&nm->f_values[0]);CHKERRQ(ierr);
nm->indices[0]=0;
for (i=1;i<nm->N+1;i++){
ierr = VecCopy(tao->solution,nm->simplex[i]);CHKERRQ(ierr);
ierr = VecGetOwnershipRange(nm->simplex[i],&low,&high);CHKERRQ(ierr);
if (i-1 >= low && i-1 < high) {
ierr = VecGetArray(nm->simplex[i],&x);CHKERRQ(ierr);
x[i-1-low] += nm->lamda;
ierr = VecRestoreArray(nm->simplex[i],&x);CHKERRQ(ierr);
}
ierr = TaoComputeObjective(tao,nm->simplex[i],&nm->f_values[i]);CHKERRQ(ierr);
nm->indices[i] = i;
}
/* Xbar = (Sum of all simplex vectors - worst vector)/N */
ierr = NelderMeadSort(nm);CHKERRQ(ierr);
ierr = VecSet(Xbar,0.0);CHKERRQ(ierr);
for (i=0;i<nm->N;i++) {
ierr = VecAXPY(Xbar,1.0,nm->simplex[nm->indices[i]]);CHKERRQ(ierr);
}
ierr = VecScale(Xbar,nm->oneOverN);CHKERRQ(ierr);
reason = TAO_CONTINUE_ITERATING;
while (1) {
shrink = 0;
ierr = VecCopy(nm->simplex[nm->indices[0]],tao->solution);CHKERRQ(ierr);
ierr = TaoMonitor(tao,tao->niter++,nm->f_values[nm->indices[0]],nm->f_values[nm->indices[nm->N]]-nm->f_values[nm->indices[0]],0.0,1.0,&reason);CHKERRQ(ierr);
if (reason != TAO_CONTINUE_ITERATING) break;
/* x(mu) = (1 + mu)Xbar - mu*X_N+1 */
ierr = VecAXPBYPCZ(Xmur,1+nm->mu_r,-nm->mu_r,0,Xbar,nm->simplex[nm->indices[nm->N]]);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao,Xmur,&fr);CHKERRQ(ierr);
if (nm->f_values[nm->indices[0]] <= fr && fr < nm->f_values[nm->indices[nm->N-1]]) {
/* reflect */
nm->nreflect++;
ierr = PetscInfo(0,"Reflect\n");CHKERRQ(ierr);
ierr = NelderMeadReplace(nm,nm->indices[nm->N],Xmur,fr);CHKERRQ(ierr);
} else if (fr < nm->f_values[nm->indices[0]]) {
/* expand */
nm->nexpand++;
ierr = PetscInfo(0,"Expand\n");CHKERRQ(ierr);
ierr = VecAXPBYPCZ(Xmue,1+nm->mu_e,-nm->mu_e,0,Xbar,nm->simplex[nm->indices[nm->N]]);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao,Xmue,&fe);CHKERRQ(ierr);
if (fe < fr) {
ierr = NelderMeadReplace(nm,nm->indices[nm->N],Xmue,fe);CHKERRQ(ierr);
} else {
ierr = NelderMeadReplace(nm,nm->indices[nm->N],Xmur,fr);CHKERRQ(ierr);
}
} else if (nm->f_values[nm->indices[nm->N-1]] <= fr && fr < nm->f_values[nm->indices[nm->N]]) {
/* outside contraction */
nm->noutcontract++;
ierr = PetscInfo(0,"Outside Contraction\n");CHKERRQ(ierr);
ierr = VecAXPBYPCZ(Xmuc,1+nm->mu_oc,-nm->mu_oc,0,Xbar,nm->simplex[nm->indices[nm->N]]);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao,Xmuc,&fc);CHKERRQ(ierr);
if (fc <= fr) {
ierr = NelderMeadReplace(nm,nm->indices[nm->N],Xmuc,fc);CHKERRQ(ierr);
} else shrink=1;
} else {
/* inside contraction */
nm->nincontract++;
ierr = PetscInfo(0,"Inside Contraction\n");CHKERRQ(ierr);
ierr = VecAXPBYPCZ(Xmuc,1+nm->mu_ic,-nm->mu_ic,0,Xbar,nm->simplex[nm->indices[nm->N]]);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao,Xmuc,&fc);CHKERRQ(ierr);
if (fc < nm->f_values[nm->indices[nm->N]]) {
ierr = NelderMeadReplace(nm,nm->indices[nm->N],Xmuc,fc);CHKERRQ(ierr);
} else shrink = 1;
}
if (shrink) {
nm->nshrink++;
ierr = PetscInfo(0,"Shrink\n");CHKERRQ(ierr);
for (i=1;i<nm->N+1;i++) {
ierr = VecAXPBY(nm->simplex[nm->indices[i]],1.5,-0.5,nm->simplex[nm->indices[0]]);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao,nm->simplex[nm->indices[i]], &nm->f_values[nm->indices[i]]);CHKERRQ(ierr);
}
ierr = VecAXPBY(Xbar,1.5*nm->oneOverN,-0.5,nm->simplex[nm->indices[0]]);CHKERRQ(ierr);
/* Add last vector's fraction of average */
ierr = VecAXPY(Xbar,nm->oneOverN,nm->simplex[nm->indices[nm->N]]);CHKERRQ(ierr);
ierr = NelderMeadSort(nm);CHKERRQ(ierr);
/* Subtract new last vector from average */
ierr = VecAXPY(Xbar,-nm->oneOverN,nm->simplex[nm->indices[nm->N]]);CHKERRQ(ierr);
}
}
PetscFunctionReturn(0);
}
static PetscErrorCode TaoSolve_NTR(Tao tao)
{
TAO_NTR *tr = (TAO_NTR *)tao->data;
PC pc;
KSPConvergedReason ksp_reason;
TaoConvergedReason reason;
PetscReal fmin, ftrial, prered, actred, kappa, sigma, beta;
PetscReal tau, tau_1, tau_2, tau_max, tau_min, max_radius;
PetscReal f, gnorm;
PetscReal delta;
PetscReal norm_d;
PetscErrorCode ierr;
PetscInt iter = 0;
PetscInt bfgsUpdates = 0;
PetscInt needH;
PetscInt i_max = 5;
PetscInt j_max = 1;
PetscInt i, j, N, n, its;
PetscFunctionBegin;
if (tao->XL || tao->XU || tao->ops->computebounds) {
ierr = PetscPrintf(((PetscObject)tao)->comm,"WARNING: Variable bounds have been set but will be ignored by ntr algorithm\n");CHKERRQ(ierr);
}
tao->trust = tao->trust0;
/* Modify the radius if it is too large or small */
tao->trust = PetscMax(tao->trust, tr->min_radius);
tao->trust = PetscMin(tao->trust, tr->max_radius);
if (NTR_PC_BFGS == tr->pc_type && !tr->M) {
ierr = VecGetLocalSize(tao->solution,&n);CHKERRQ(ierr);
ierr = VecGetSize(tao->solution,&N);CHKERRQ(ierr);
ierr = MatCreateLMVM(((PetscObject)tao)->comm,n,N,&tr->M);CHKERRQ(ierr);
ierr = MatLMVMAllocateVectors(tr->M,tao->solution);CHKERRQ(ierr);
}
/* Check convergence criteria */
ierr = TaoComputeObjectiveAndGradient(tao, tao->solution, &f, tao->gradient);CHKERRQ(ierr);
ierr = VecNorm(tao->gradient,NORM_2,&gnorm);CHKERRQ(ierr);
if (PetscIsInfOrNanReal(f) || PetscIsInfOrNanReal(gnorm)) SETERRQ(PETSC_COMM_SELF,1, "User provided compute function generated Inf or NaN");
needH = 1;
ierr = TaoMonitor(tao, iter, f, gnorm, 0.0, 1.0, &reason);CHKERRQ(ierr);
if (reason != TAO_CONTINUE_ITERATING) PetscFunctionReturn(0);
/* Create vectors for the limited memory preconditioner */
if ((NTR_PC_BFGS == tr->pc_type) &&
(BFGS_SCALE_BFGS != tr->bfgs_scale_type)) {
if (!tr->Diag) {
ierr = VecDuplicate(tao->solution, &tr->Diag);CHKERRQ(ierr);
}
}
switch(tr->ksp_type) {
case NTR_KSP_NASH:
ierr = KSPSetType(tao->ksp, KSPNASH);CHKERRQ(ierr);
if (tao->ksp->ops->setfromoptions) {
(*tao->ksp->ops->setfromoptions)(tao->ksp);
}
break;
case NTR_KSP_STCG:
ierr = KSPSetType(tao->ksp, KSPSTCG);CHKERRQ(ierr);
if (tao->ksp->ops->setfromoptions) {
(*tao->ksp->ops->setfromoptions)(tao->ksp);
}
break;
default:
ierr = KSPSetType(tao->ksp, KSPGLTR);CHKERRQ(ierr);
if (tao->ksp->ops->setfromoptions) {
(*tao->ksp->ops->setfromoptions)(tao->ksp);
}
break;
}
/* Modify the preconditioner to use the bfgs approximation */
ierr = KSPGetPC(tao->ksp, &pc);CHKERRQ(ierr);
switch(tr->pc_type) {
case NTR_PC_NONE:
ierr = PCSetType(pc, PCNONE);CHKERRQ(ierr);
if (pc->ops->setfromoptions) {
(*pc->ops->setfromoptions)(pc);
}
break;
case NTR_PC_AHESS:
ierr = PCSetType(pc, PCJACOBI);CHKERRQ(ierr);
if (pc->ops->setfromoptions) {
(*pc->ops->setfromoptions)(pc);
}
ierr = PCJacobiSetUseAbs(pc);CHKERRQ(ierr);
break;
case NTR_PC_BFGS:
ierr = PCSetType(pc, PCSHELL);CHKERRQ(ierr);
if (pc->ops->setfromoptions) {
(*pc->ops->setfromoptions)(pc);
}
ierr = PCShellSetName(pc, "bfgs");CHKERRQ(ierr);
ierr = PCShellSetContext(pc, tr->M);CHKERRQ(ierr);
ierr = PCShellSetApply(pc, MatLMVMSolveShell);CHKERRQ(ierr);
break;
default:
/* Use the pc method set by pc_type */
break;
}
/* Initialize trust-region radius */
switch(tr->init_type) {
case NTR_INIT_CONSTANT:
/* Use the initial radius specified */
break;
case NTR_INIT_INTERPOLATION:
/* Use the initial radius specified */
max_radius = 0.0;
for (j = 0; j < j_max; ++j) {
fmin = f;
sigma = 0.0;
if (needH) {
ierr = TaoComputeHessian(tao,tao->solution,tao->hessian,tao->hessian_pre);CHKERRQ(ierr);
needH = 0;
}
for (i = 0; i < i_max; ++i) {
ierr = VecCopy(tao->solution, tr->W);CHKERRQ(ierr);
ierr = VecAXPY(tr->W, -tao->trust/gnorm, tao->gradient);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao, tr->W, &ftrial);CHKERRQ(ierr);
if (PetscIsInfOrNanReal(ftrial)) {
tau = tr->gamma1_i;
}
else {
if (ftrial < fmin) {
fmin = ftrial;
sigma = -tao->trust / gnorm;
}
ierr = MatMult(tao->hessian, tao->gradient, tao->stepdirection);CHKERRQ(ierr);
ierr = VecDot(tao->gradient, tao->stepdirection, &prered);CHKERRQ(ierr);
prered = tao->trust * (gnorm - 0.5 * tao->trust * prered / (gnorm * gnorm));
actred = f - ftrial;
if ((PetscAbsScalar(actred) <= tr->epsilon) &&
(PetscAbsScalar(prered) <= tr->epsilon)) {
kappa = 1.0;
}
else {
kappa = actred / prered;
}
tau_1 = tr->theta_i * gnorm * tao->trust / (tr->theta_i * gnorm * tao->trust + (1.0 - tr->theta_i) * prered - actred);
tau_2 = tr->theta_i * gnorm * tao->trust / (tr->theta_i * gnorm * tao->trust - (1.0 + tr->theta_i) * prered + actred);
tau_min = PetscMin(tau_1, tau_2);
tau_max = PetscMax(tau_1, tau_2);
if (PetscAbsScalar(kappa - 1.0) <= tr->mu1_i) {
/* Great agreement */
max_radius = PetscMax(max_radius, tao->trust);
if (tau_max < 1.0) {
tau = tr->gamma3_i;
}
else if (tau_max > tr->gamma4_i) {
tau = tr->gamma4_i;
}
else {
tau = tau_max;
}
}
else if (PetscAbsScalar(kappa - 1.0) <= tr->mu2_i) {
/* Good agreement */
max_radius = PetscMax(max_radius, tao->trust);
if (tau_max < tr->gamma2_i) {
tau = tr->gamma2_i;
}
else if (tau_max > tr->gamma3_i) {
tau = tr->gamma3_i;
}
else {
tau = tau_max;
}
}
else {
/* Not good agreement */
if (tau_min > 1.0) {
tau = tr->gamma2_i;
}
else if (tau_max < tr->gamma1_i) {
tau = tr->gamma1_i;
}
else if ((tau_min < tr->gamma1_i) && (tau_max >= 1.0)) {
tau = tr->gamma1_i;
}
else if ((tau_1 >= tr->gamma1_i) && (tau_1 < 1.0) &&
((tau_2 < tr->gamma1_i) || (tau_2 >= 1.0))) {
tau = tau_1;
}
else if ((tau_2 >= tr->gamma1_i) && (tau_2 < 1.0) &&
((tau_1 < tr->gamma1_i) || (tau_2 >= 1.0))) {
tau = tau_2;
}
else {
tau = tau_max;
}
}
}
tao->trust = tau * tao->trust;
}
if (fmin < f) {
f = fmin;
ierr = VecAXPY(tao->solution, sigma, tao->gradient);CHKERRQ(ierr);
ierr = TaoComputeGradient(tao,tao->solution, tao->gradient);CHKERRQ(ierr);
ierr = VecNorm(tao->gradient, NORM_2, &gnorm);CHKERRQ(ierr);
if (PetscIsInfOrNanReal(f) || PetscIsInfOrNanReal(gnorm)) SETERRQ(PETSC_COMM_SELF,1, "User provided compute function generated Inf or NaN");
needH = 1;
ierr = TaoMonitor(tao, iter, f, gnorm, 0.0, 1.0, &reason);CHKERRQ(ierr);
if (reason != TAO_CONTINUE_ITERATING) {
PetscFunctionReturn(0);
}
}
}
tao->trust = PetscMax(tao->trust, max_radius);
/* Modify the radius if it is too large or small */
tao->trust = PetscMax(tao->trust, tr->min_radius);
tao->trust = PetscMin(tao->trust, tr->max_radius);
break;
default:
/* Norm of the first direction will initialize radius */
tao->trust = 0.0;
break;
}
/* Set initial scaling for the BFGS preconditioner
This step is done after computing the initial trust-region radius
since the function value may have decreased */
if (NTR_PC_BFGS == tr->pc_type) {
if (f != 0.0) {
delta = 2.0 * PetscAbsScalar(f) / (gnorm*gnorm);
}
else {
delta = 2.0 / (gnorm*gnorm);
}
ierr = MatLMVMSetDelta(tr->M,delta);CHKERRQ(ierr);
}
/* Have not converged; continue with Newton method */
while (reason == TAO_CONTINUE_ITERATING) {
++iter;
tao->ksp_its=0;
/* Compute the Hessian */
if (needH) {
ierr = TaoComputeHessian(tao,tao->solution,tao->hessian,tao->hessian_pre);CHKERRQ(ierr);
needH = 0;
}
if (NTR_PC_BFGS == tr->pc_type) {
if (BFGS_SCALE_AHESS == tr->bfgs_scale_type) {
/* Obtain diagonal for the bfgs preconditioner */
ierr = MatGetDiagonal(tao->hessian, tr->Diag);CHKERRQ(ierr);
ierr = VecAbs(tr->Diag);CHKERRQ(ierr);
ierr = VecReciprocal(tr->Diag);CHKERRQ(ierr);
ierr = MatLMVMSetScale(tr->M,tr->Diag);CHKERRQ(ierr);
}
/* Update the limited memory preconditioner */
ierr = MatLMVMUpdate(tr->M, tao->solution, tao->gradient);CHKERRQ(ierr);
++bfgsUpdates;
}
while (reason == TAO_CONTINUE_ITERATING) {
ierr = KSPSetOperators(tao->ksp, tao->hessian, tao->hessian_pre);CHKERRQ(ierr);
/* Solve the trust region subproblem */
if (NTR_KSP_NASH == tr->ksp_type) {
ierr = KSPNASHSetRadius(tao->ksp,tao->trust);CHKERRQ(ierr);
ierr = KSPSolve(tao->ksp, tao->gradient, tao->stepdirection);CHKERRQ(ierr);
ierr = KSPGetIterationNumber(tao->ksp,&its);CHKERRQ(ierr);
tao->ksp_its+=its;
tao->ksp_tot_its+=its;
ierr = KSPNASHGetNormD(tao->ksp, &norm_d);CHKERRQ(ierr);
} else if (NTR_KSP_STCG == tr->ksp_type) {
ierr = KSPSTCGSetRadius(tao->ksp,tao->trust);CHKERRQ(ierr);
ierr = KSPSolve(tao->ksp, tao->gradient, tao->stepdirection);CHKERRQ(ierr);
ierr = KSPGetIterationNumber(tao->ksp,&its);CHKERRQ(ierr);
tao->ksp_its+=its;
tao->ksp_tot_its+=its;
ierr = KSPSTCGGetNormD(tao->ksp, &norm_d);CHKERRQ(ierr);
} else { /* NTR_KSP_GLTR */
ierr = KSPGLTRSetRadius(tao->ksp,tao->trust);CHKERRQ(ierr);
ierr = KSPSolve(tao->ksp, tao->gradient, tao->stepdirection);CHKERRQ(ierr);
ierr = KSPGetIterationNumber(tao->ksp,&its);CHKERRQ(ierr);
tao->ksp_its+=its;
tao->ksp_tot_its+=its;
ierr = KSPGLTRGetNormD(tao->ksp, &norm_d);CHKERRQ(ierr);
}
if (0.0 == tao->trust) {
/* Radius was uninitialized; use the norm of the direction */
if (norm_d > 0.0) {
tao->trust = norm_d;
/* Modify the radius if it is too large or small */
tao->trust = PetscMax(tao->trust, tr->min_radius);
tao->trust = PetscMin(tao->trust, tr->max_radius);
}
else {
/* The direction was bad; set radius to default value and re-solve
the trust-region subproblem to get a direction */
tao->trust = tao->trust0;
/* Modify the radius if it is too large or small */
tao->trust = PetscMax(tao->trust, tr->min_radius);
tao->trust = PetscMin(tao->trust, tr->max_radius);
if (NTR_KSP_NASH == tr->ksp_type) {
ierr = KSPNASHSetRadius(tao->ksp,tao->trust);CHKERRQ(ierr);
ierr = KSPSolve(tao->ksp, tao->gradient, tao->stepdirection);CHKERRQ(ierr);
ierr = KSPGetIterationNumber(tao->ksp,&its);CHKERRQ(ierr);
tao->ksp_its+=its;
tao->ksp_tot_its+=its;
ierr = KSPNASHGetNormD(tao->ksp, &norm_d);CHKERRQ(ierr);
} else if (NTR_KSP_STCG == tr->ksp_type) {
ierr = KSPSTCGSetRadius(tao->ksp,tao->trust);CHKERRQ(ierr);
ierr = KSPSolve(tao->ksp, tao->gradient, tao->stepdirection);CHKERRQ(ierr);
ierr = KSPGetIterationNumber(tao->ksp,&its);CHKERRQ(ierr);
tao->ksp_its+=its;
tao->ksp_tot_its+=its;
ierr = KSPSTCGGetNormD(tao->ksp, &norm_d);CHKERRQ(ierr);
} else { /* NTR_KSP_GLTR */
ierr = KSPGLTRSetRadius(tao->ksp,tao->trust);CHKERRQ(ierr);
ierr = KSPSolve(tao->ksp, tao->gradient, tao->stepdirection);CHKERRQ(ierr);
ierr = KSPGetIterationNumber(tao->ksp,&its);CHKERRQ(ierr);
tao->ksp_its+=its;
tao->ksp_tot_its+=its;
ierr = KSPGLTRGetNormD(tao->ksp, &norm_d);CHKERRQ(ierr);
}
if (norm_d == 0.0) SETERRQ(PETSC_COMM_SELF,1, "Initial direction zero");
}
}
ierr = VecScale(tao->stepdirection, -1.0);CHKERRQ(ierr);
ierr = KSPGetConvergedReason(tao->ksp, &ksp_reason);CHKERRQ(ierr);
if ((KSP_DIVERGED_INDEFINITE_PC == ksp_reason) &&
(NTR_PC_BFGS == tr->pc_type) && (bfgsUpdates > 1)) {
/* Preconditioner is numerically indefinite; reset the
approximate if using BFGS preconditioning. */
if (f != 0.0) {
delta = 2.0 * PetscAbsScalar(f) / (gnorm*gnorm);
}
else {
delta = 2.0 / (gnorm*gnorm);
}
ierr = MatLMVMSetDelta(tr->M, delta);CHKERRQ(ierr);
ierr = MatLMVMReset(tr->M);CHKERRQ(ierr);
ierr = MatLMVMUpdate(tr->M, tao->solution, tao->gradient);CHKERRQ(ierr);
bfgsUpdates = 1;
}
if (NTR_UPDATE_REDUCTION == tr->update_type) {
/* Get predicted reduction */
if (NTR_KSP_NASH == tr->ksp_type) {
ierr = KSPNASHGetObjFcn(tao->ksp,&prered);CHKERRQ(ierr);
} else if (NTR_KSP_STCG == tr->ksp_type) {
ierr = KSPSTCGGetObjFcn(tao->ksp,&prered);CHKERRQ(ierr);
} else { /* gltr */
ierr = KSPGLTRGetObjFcn(tao->ksp,&prered);CHKERRQ(ierr);
}
if (prered >= 0.0) {
/* The predicted reduction has the wrong sign. This cannot
happen in infinite precision arithmetic. Step should
be rejected! */
tao->trust = tr->alpha1 * PetscMin(tao->trust, norm_d);
}
else {
/* Compute trial step and function value */
ierr = VecCopy(tao->solution,tr->W);CHKERRQ(ierr);
ierr = VecAXPY(tr->W, 1.0, tao->stepdirection);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao, tr->W, &ftrial);CHKERRQ(ierr);
if (PetscIsInfOrNanReal(ftrial)) {
tao->trust = tr->alpha1 * PetscMin(tao->trust, norm_d);
} else {
/* Compute and actual reduction */
actred = f - ftrial;
prered = -prered;
if ((PetscAbsScalar(actred) <= tr->epsilon) &&
(PetscAbsScalar(prered) <= tr->epsilon)) {
kappa = 1.0;
}
else {
kappa = actred / prered;
}
/* Accept or reject the step and update radius */
if (kappa < tr->eta1) {
/* Reject the step */
tao->trust = tr->alpha1 * PetscMin(tao->trust, norm_d);
}
else {
/* Accept the step */
if (kappa < tr->eta2) {
/* Marginal bad step */
tao->trust = tr->alpha2 * PetscMin(tao->trust, norm_d);
}
else if (kappa < tr->eta3) {
/* Reasonable step */
tao->trust = tr->alpha3 * tao->trust;
}
else if (kappa < tr->eta4) {
/* Good step */
tao->trust = PetscMax(tr->alpha4 * norm_d, tao->trust);
}
else {
/* Very good step */
tao->trust = PetscMax(tr->alpha5 * norm_d, tao->trust);
}
break;
}
}
}
}
else {
/* Get predicted reduction */
if (NTR_KSP_NASH == tr->ksp_type) {
ierr = KSPNASHGetObjFcn(tao->ksp,&prered);CHKERRQ(ierr);
} else if (NTR_KSP_STCG == tr->ksp_type) {
ierr = KSPSTCGGetObjFcn(tao->ksp,&prered);CHKERRQ(ierr);
} else { /* gltr */
ierr = KSPGLTRGetObjFcn(tao->ksp,&prered);CHKERRQ(ierr);
}
if (prered >= 0.0) {
/* The predicted reduction has the wrong sign. This cannot
happen in infinite precision arithmetic. Step should
be rejected! */
tao->trust = tr->gamma1 * PetscMin(tao->trust, norm_d);
}
else {
ierr = VecCopy(tao->solution, tr->W);CHKERRQ(ierr);
ierr = VecAXPY(tr->W, 1.0, tao->stepdirection);CHKERRQ(ierr);
ierr = TaoComputeObjective(tao, tr->W, &ftrial);CHKERRQ(ierr);
if (PetscIsInfOrNanReal(ftrial)) {
tao->trust = tr->gamma1 * PetscMin(tao->trust, norm_d);
}
else {
ierr = VecDot(tao->gradient, tao->stepdirection, &beta);CHKERRQ(ierr);
actred = f - ftrial;
prered = -prered;
if ((PetscAbsScalar(actred) <= tr->epsilon) &&
(PetscAbsScalar(prered) <= tr->epsilon)) {
kappa = 1.0;
}
else {
kappa = actred / prered;
}
tau_1 = tr->theta * beta / (tr->theta * beta - (1.0 - tr->theta) * prered + actred);
tau_2 = tr->theta * beta / (tr->theta * beta + (1.0 + tr->theta) * prered - actred);
tau_min = PetscMin(tau_1, tau_2);
tau_max = PetscMax(tau_1, tau_2);
if (kappa >= 1.0 - tr->mu1) {
/* Great agreement; accept step and update radius */
if (tau_max < 1.0) {
tao->trust = PetscMax(tao->trust, tr->gamma3 * norm_d);
}
else if (tau_max > tr->gamma4) {
tao->trust = PetscMax(tao->trust, tr->gamma4 * norm_d);
}
else {
tao->trust = PetscMax(tao->trust, tau_max * norm_d);
}
break;
}
else if (kappa >= 1.0 - tr->mu2) {
/* Good agreement */
if (tau_max < tr->gamma2) {
tao->trust = tr->gamma2 * PetscMin(tao->trust, norm_d);
}
else if (tau_max > tr->gamma3) {
tao->trust = PetscMax(tao->trust, tr->gamma3 * norm_d);
}
else if (tau_max < 1.0) {
tao->trust = tau_max * PetscMin(tao->trust, norm_d);
}
else {
tao->trust = PetscMax(tao->trust, tau_max * norm_d);
}
break;
}
else {
/* Not good agreement */
if (tau_min > 1.0) {
tao->trust = tr->gamma2 * PetscMin(tao->trust, norm_d);
}
else if (tau_max < tr->gamma1) {
tao->trust = tr->gamma1 * PetscMin(tao->trust, norm_d);
}
else if ((tau_min < tr->gamma1) && (tau_max >= 1.0)) {
tao->trust = tr->gamma1 * PetscMin(tao->trust, norm_d);
}
else if ((tau_1 >= tr->gamma1) && (tau_1 < 1.0) &&
((tau_2 < tr->gamma1) || (tau_2 >= 1.0))) {
tao->trust = tau_1 * PetscMin(tao->trust, norm_d);
}
else if ((tau_2 >= tr->gamma1) && (tau_2 < 1.0) &&
((tau_1 < tr->gamma1) || (tau_2 >= 1.0))) {
tao->trust = tau_2 * PetscMin(tao->trust, norm_d);
}
else {
tao->trust = tau_max * PetscMin(tao->trust, norm_d);
}
}
}
}
}
/* The step computed was not good and the radius was decreased.
Monitor the radius to terminate. */
ierr = TaoMonitor(tao, iter, f, gnorm, 0.0, tao->trust, &reason);CHKERRQ(ierr);
}
/* The radius may have been increased; modify if it is too large */
tao->trust = PetscMin(tao->trust, tr->max_radius);
if (reason == TAO_CONTINUE_ITERATING) {
ierr = VecCopy(tr->W, tao->solution);CHKERRQ(ierr);
f = ftrial;
ierr = TaoComputeGradient(tao, tao->solution, tao->gradient);
ierr = VecNorm(tao->gradient, NORM_2, &gnorm);CHKERRQ(ierr);
if (PetscIsInfOrNanReal(f) || PetscIsInfOrNanReal(gnorm)) SETERRQ(PETSC_COMM_SELF,1, "User provided compute function generated Inf or NaN");
needH = 1;
ierr = TaoMonitor(tao, iter, f, gnorm, 0.0, tao->trust, &reason);CHKERRQ(ierr);
}
}
PetscFunctionReturn(0);
}
PetscErrorCode TaoSolve_BNTR(Tao tao)
{
PetscErrorCode ierr;
TAO_BNK *bnk = (TAO_BNK *)tao->data;
KSPConvergedReason ksp_reason;
PetscReal oldTrust, prered, actred, steplen, resnorm;
PetscBool cgTerminate, needH = PETSC_TRUE, stepAccepted, shift = PETSC_FALSE;
PetscInt stepType, nDiff;
PetscFunctionBegin;
/* Initialize the preconditioner, KSP solver and trust radius/line search */
tao->reason = TAO_CONTINUE_ITERATING;
ierr = TaoBNKInitialize(tao, bnk->init_type, &needH);CHKERRQ(ierr);
if (tao->reason != TAO_CONTINUE_ITERATING) PetscFunctionReturn(0);
/* Have not converged; continue with Newton method */
while (tao->reason == TAO_CONTINUE_ITERATING) {
/* Call general purpose update function */
if (tao->ops->update) {
ierr = (*tao->ops->update)(tao, tao->niter, tao->user_update);CHKERRQ(ierr);
}
++tao->niter;
if (needH && bnk->inactive_idx) {
/* Take BNCG steps (if enabled) to trade-off Hessian evaluations for more gradient evaluations */
ierr = TaoBNKTakeCGSteps(tao, &cgTerminate);CHKERRQ(ierr);
if (cgTerminate) {
tao->reason = bnk->bncg->reason;
PetscFunctionReturn(0);
}
/* Compute the hessian and update the BFGS preconditioner at the new iterate */
ierr = (*bnk->computehessian)(tao);CHKERRQ(ierr);
needH = PETSC_FALSE;
}
/* Store current solution before it changes */
bnk->fold = bnk->f;
ierr = VecCopy(tao->solution, bnk->Xold);CHKERRQ(ierr);
ierr = VecCopy(tao->gradient, bnk->Gold);CHKERRQ(ierr);
ierr = VecCopy(bnk->unprojected_gradient, bnk->unprojected_gradient_old);CHKERRQ(ierr);
/* Enter into trust region loops */
stepAccepted = PETSC_FALSE;
while (!stepAccepted && tao->reason == TAO_CONTINUE_ITERATING) {
tao->ksp_its=0;
/* Use the common BNK kernel to compute the Newton step (for inactive variables only) */
ierr = (*bnk->computestep)(tao, shift, &ksp_reason, &stepType);CHKERRQ(ierr);
/* Temporarily accept the step and project it into the bounds */
ierr = VecAXPY(tao->solution, 1.0, tao->stepdirection);CHKERRQ(ierr);
ierr = TaoBoundSolution(tao->solution, tao->XL,tao->XU, 0.0, &nDiff, tao->solution);CHKERRQ(ierr);
/* Check if the projection changed the step direction */
if (nDiff > 0) {
/* Projection changed the step, so we have to recompute the step and
the predicted reduction. Leave the trust radius unchanged. */
ierr = VecCopy(tao->solution, tao->stepdirection);CHKERRQ(ierr);
ierr = VecAXPY(tao->stepdirection, -1.0, bnk->Xold);CHKERRQ(ierr);
ierr = TaoBNKRecomputePred(tao, tao->stepdirection, &prered);CHKERRQ(ierr);
} else {
/* Step did not change, so we can just recover the pre-computed prediction */
ierr = KSPCGGetObjFcn(tao->ksp, &prered);CHKERRQ(ierr);
}
prered = -prered;
/* Compute the actual reduction and update the trust radius */
ierr = TaoComputeObjective(tao, tao->solution, &bnk->f);CHKERRQ(ierr);
if (PetscIsInfOrNanReal(bnk->f)) SETERRQ(PETSC_COMM_SELF,1, "User provided compute function generated Inf or NaN");
actred = bnk->fold - bnk->f;
oldTrust = tao->trust;
ierr = TaoBNKUpdateTrustRadius(tao, prered, actred, bnk->update_type, stepType, &stepAccepted);CHKERRQ(ierr);
if (stepAccepted) {
/* Step is good, evaluate the gradient and flip the need-Hessian switch */
steplen = 1.0;
needH = PETSC_TRUE;
++bnk->newt;
ierr = TaoComputeGradient(tao, tao->solution, bnk->unprojected_gradient);CHKERRQ(ierr);
ierr = TaoBNKEstimateActiveSet(tao, bnk->as_type);CHKERRQ(ierr);
ierr = VecCopy(bnk->unprojected_gradient, tao->gradient);CHKERRQ(ierr);
ierr = VecISSet(tao->gradient, bnk->active_idx, 0.0);CHKERRQ(ierr);
ierr = TaoGradientNorm(tao, tao->gradient, NORM_2, &bnk->gnorm);CHKERRQ(ierr);
} else {
/* Step is bad, revert old solution and re-solve with new radius*/
steplen = 0.0;
needH = PETSC_FALSE;
bnk->f = bnk->fold;
ierr = VecCopy(bnk->Xold, tao->solution);CHKERRQ(ierr);
ierr = VecCopy(bnk->Gold, tao->gradient);CHKERRQ(ierr);
ierr = VecCopy(bnk->unprojected_gradient_old, bnk->unprojected_gradient);CHKERRQ(ierr);
if (oldTrust == tao->trust) {
/* Can't change the radius anymore so just terminate */
tao->reason = TAO_DIVERGED_TR_REDUCTION;
}
}
/* Check for termination */
ierr = VecFischer(tao->solution, bnk->unprojected_gradient, tao->XL, tao->XU, bnk->W);CHKERRQ(ierr);
ierr = VecNorm(bnk->W, NORM_2, &resnorm);CHKERRQ(ierr);
if (PetscIsInfOrNanReal(resnorm)) SETERRQ(PETSC_COMM_SELF,1, "User provided compute function generated Inf or NaN");
ierr = TaoLogConvergenceHistory(tao, bnk->f, resnorm, 0.0, tao->ksp_its);CHKERRQ(ierr);
ierr = TaoMonitor(tao, tao->niter, bnk->f, resnorm, 0.0, steplen);CHKERRQ(ierr);
ierr = (*tao->ops->convergencetest)(tao, tao->cnvP);CHKERRQ(ierr);
}
}
PetscFunctionReturn(0);
}
