1 #include <../src/tao/leastsquares/impls/brgn/brgn.h> 2 3 #define BRGN_user 0 4 #define BRGN_l2prox 1 5 #define BRGN_l1dict 2 6 #define BRGNRegTypes 3 7 8 static const char *BRGN_Table[64] = {"user", "l2prox", "l1dict"}; 9 10 static PetscErrorCode GNHessianProd(Mat H,Vec in,Vec out) 11 { 12 TAO_BRGN *gn; 13 PetscErrorCode ierr; 14 15 PetscFunctionBegin; 16 ierr = MatShellGetContext(H,&gn);CHKERRQ(ierr); 17 ierr = MatMult(gn->subsolver->ls_jac,in,gn->r_work);CHKERRQ(ierr); 18 ierr = MatMultTranspose(gn->subsolver->ls_jac,gn->r_work,out);CHKERRQ(ierr); 19 switch (gn->reg_type) { 20 case BRGN_user: 21 ierr = MatMult(gn->Hreg,in,gn->x_work);CHKERRQ(ierr); 22 ierr = VecAXPY(out,gn->lambda,gn->x_work);CHKERRQ(ierr); 23 break; 24 case BRGN_l2prox: 25 ierr = VecAXPY(out,gn->lambda,in);CHKERRQ(ierr); 26 break; 27 case BRGN_l1dict: 28 /* out = out + lambda*D'*(diag.*(D*in)) */ 29 if (gn->D) { 30 ierr = MatMult(gn->D,in,gn->y);CHKERRQ(ierr);/* y = D*in */ 31 } else { 32 ierr = VecCopy(in,gn->y);CHKERRQ(ierr); 33 } 34 ierr = VecPointwiseMult(gn->y_work,gn->diag,gn->y);CHKERRQ(ierr); /* y_work = diag.*(D*in), where diag = epsilon^2 ./ sqrt(x.^2+epsilon^2).^3 */ 35 if (gn->D) { 36 ierr = MatMultTranspose(gn->D,gn->y_work,gn->x_work);CHKERRQ(ierr); /* x_work = D'*(diag.*(D*in)) */ 37 } else { 38 ierr = VecCopy(gn->y_work,gn->x_work);CHKERRQ(ierr); 39 } 40 ierr = VecAXPY(out,gn->lambda,gn->x_work);CHKERRQ(ierr); 41 break; 42 } 43 44 PetscFunctionReturn(0); 45 } 46 47 static PetscErrorCode GNObjectiveGradientEval(Tao tao,Vec X,PetscReal *fcn,Vec G,void *ptr) 48 { 49 TAO_BRGN *gn = (TAO_BRGN *)ptr; 50 PetscInt K; /* dimension of D*X */ 51 PetscScalar yESum; 52 PetscErrorCode ierr; 53 PetscReal f_reg; 54 55 PetscFunctionBegin; 56 /* compute objective *fcn*/ 57 /* compute first term 0.5*||ls_res||_2^2 */ 58 ierr = TaoComputeResidual(tao,X,tao->ls_res);CHKERRQ(ierr); 59 ierr = VecDot(tao->ls_res,tao->ls_res,fcn);CHKERRQ(ierr); 60 *fcn *= 0.5; 61 /* compute gradient G */ 62 ierr = TaoComputeResidualJacobian(tao,X,tao->ls_jac,tao->ls_jac_pre);CHKERRQ(ierr); 63 ierr = MatMultTranspose(tao->ls_jac,tao->ls_res,G);CHKERRQ(ierr); 64 /* add the regularization contribution */ 65 switch (gn->reg_type) { 66 case BRGN_user: 67 ierr = (*gn->regularizerobjandgrad)(tao,X,&f_reg,gn->x_work,gn->reg_obj_ctx);CHKERRQ(ierr); 68 *fcn += gn->lambda*f_reg; 69 ierr = VecAXPY(G,gn->lambda,gn->x_work);CHKERRQ(ierr); 70 break; 71 case BRGN_l2prox: 72 /* compute f = f + lambda*0.5*(xk - xkm1)^T(xk - xkm1) */ 73 ierr = VecAXPBYPCZ(gn->x_work,1.0,-1.0,0.0,X,gn->x_old);CHKERRQ(ierr); /*TODO: no need to use VecAXPBYPCZ for x - xkm1 */ 74 ierr = VecDot(gn->x_work,gn->x_work,&f_reg);CHKERRQ(ierr); 75 *fcn += gn->lambda*0.5*f_reg; 76 /* compute G = G + lambda*(xk - xkm1) */ 77 ierr = VecAXPBYPCZ(G,gn->lambda,-gn->lambda,1.0,X,gn->x_old);CHKERRQ(ierr); 78 break; 79 case BRGN_l1dict: 80 /* compute f = f + lambda*sum(sqrt(y.^2+epsilon^2) - epsilon), where y = D*x*/ 81 if (gn->D) { 82 ierr = MatMult(gn->D,X,gn->y);CHKERRQ(ierr);/* y = D*x */ 83 } else { 84 ierr = VecCopy(X,gn->y);CHKERRQ(ierr); 85 } 86 ierr = VecPointwiseMult(gn->y_work,gn->y,gn->y);CHKERRQ(ierr); 87 ierr = VecShift(gn->y_work,gn->epsilon*gn->epsilon);CHKERRQ(ierr); 88 ierr = VecSqrtAbs(gn->y_work);CHKERRQ(ierr); /* gn->y_work = sqrt(y.^2+epsilon^2) */ 89 ierr = VecSum(gn->y_work,&yESum);CHKERRQ(ierr);CHKERRQ(ierr); 90 ierr = VecGetSize(gn->y,&K);CHKERRQ(ierr); 91 *fcn += gn->lambda*(yESum - K*gn->epsilon); 92 /* compute G = G + lambda*D'*(y./sqrt(y.^2+epsilon^2)),where y = D*x */ 93 ierr = VecPointwiseDivide(gn->y_work,gn->y,gn->y_work);CHKERRQ(ierr); /* reuse y_work = y./sqrt(y.^2+epsilon^2) */ 94 if (gn->D) { 95 ierr = MatMultTranspose(gn->D,gn->y_work,gn->x_work);CHKERRQ(ierr); 96 } else { 97 ierr = VecCopy(gn->y_work,gn->x_work);CHKERRQ(ierr); 98 } 99 ierr = VecAXPY(G,gn->lambda,gn->x_work);CHKERRQ(ierr); 100 break; 101 } 102 PetscFunctionReturn(0); 103 } 104 105 static PetscErrorCode GNComputeHessian(Tao tao,Vec X,Mat H,Mat Hpre,void *ptr) 106 { 107 TAO_BRGN *gn = (TAO_BRGN *)ptr; 108 PetscErrorCode ierr; 109 110 PetscFunctionBegin; 111 ierr = TaoComputeResidualJacobian(tao,X,tao->ls_jac,tao->ls_jac_pre);CHKERRQ(ierr); 112 113 switch (gn->reg_type) { 114 case BRGN_user: 115 ierr = (*gn->regularizerhessian)(tao,X,gn->Hreg,gn->reg_hess_ctx);CHKERRQ(ierr); 116 break; 117 case BRGN_l2prox: 118 break; 119 case BRGN_l1dict: 120 /* calculate and store diagonal matrix as a vector: diag = epsilon^2 ./ sqrt(x.^2+epsilon^2).^3* --> diag = epsilon^2 ./ sqrt(y.^2+epsilon^2).^3,where y = D*x */ 121 if (gn->D) { 122 ierr = MatMult(gn->D,X,gn->y);CHKERRQ(ierr);/* y = D*x */ 123 } else { 124 ierr = VecCopy(X,gn->y);CHKERRQ(ierr); 125 } 126 ierr = VecPointwiseMult(gn->y_work,gn->y,gn->y);CHKERRQ(ierr); 127 ierr = VecShift(gn->y_work,gn->epsilon*gn->epsilon);CHKERRQ(ierr); 128 ierr = VecCopy(gn->y_work,gn->diag);CHKERRQ(ierr); /* gn->diag = y.^2+epsilon^2 */ 129 ierr = VecSqrtAbs(gn->y_work);CHKERRQ(ierr); /* gn->y_work = sqrt(y.^2+epsilon^2) */ 130 ierr = VecPointwiseMult(gn->diag,gn->y_work,gn->diag);CHKERRQ(ierr);/* gn->diag = sqrt(y.^2+epsilon^2).^3 */ 131 ierr = VecReciprocal(gn->diag);CHKERRQ(ierr); 132 ierr = VecScale(gn->diag,gn->epsilon*gn->epsilon);CHKERRQ(ierr); 133 break; 134 } 135 136 PetscFunctionReturn(0); 137 } 138 139 static PetscErrorCode GNHookFunction(Tao tao,PetscInt iter) 140 { 141 TAO_BRGN *gn = (TAO_BRGN *)tao->user_update; 142 PetscErrorCode ierr; 143 144 PetscFunctionBegin; 145 /* Update basic tao information from the subsolver */ 146 gn->parent->nfuncs = tao->nfuncs; 147 gn->parent->ngrads = tao->ngrads; 148 gn->parent->nfuncgrads = tao->nfuncgrads; 149 gn->parent->nhess = tao->nhess; 150 gn->parent->niter = tao->niter; 151 gn->parent->ksp_its = tao->ksp_its; 152 gn->parent->ksp_tot_its = tao->ksp_tot_its; 153 ierr = TaoGetConvergedReason(tao,&gn->parent->reason);CHKERRQ(ierr); 154 /* Update the solution vectors */ 155 if (iter == 0) { 156 ierr = VecSet(gn->x_old,0.0);CHKERRQ(ierr); 157 } else { 158 ierr = VecCopy(tao->solution,gn->x_old);CHKERRQ(ierr); 159 ierr = VecCopy(tao->solution,gn->parent->solution);CHKERRQ(ierr); 160 } 161 /* Update the gradient */ 162 ierr = VecCopy(tao->gradient,gn->parent->gradient);CHKERRQ(ierr); 163 /* Call general purpose update function */ 164 if (gn->parent->ops->update) { 165 ierr = (*gn->parent->ops->update)(gn->parent,gn->parent->niter);CHKERRQ(ierr); 166 } 167 PetscFunctionReturn(0); 168 } 169 170 static PetscErrorCode TaoSolve_BRGN(Tao tao) 171 { 172 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 173 PetscErrorCode ierr; 174 175 PetscFunctionBegin; 176 ierr = TaoSolve(gn->subsolver);CHKERRQ(ierr); 177 /* Update basic tao information from the subsolver */ 178 tao->nfuncs = gn->subsolver->nfuncs; 179 tao->ngrads = gn->subsolver->ngrads; 180 tao->nfuncgrads = gn->subsolver->nfuncgrads; 181 tao->nhess = gn->subsolver->nhess; 182 tao->niter = gn->subsolver->niter; 183 tao->ksp_its = gn->subsolver->ksp_its; 184 tao->ksp_tot_its = gn->subsolver->ksp_tot_its; 185 ierr = TaoGetConvergedReason(gn->subsolver,&tao->reason);CHKERRQ(ierr); 186 /* Update vectors */ 187 ierr = VecCopy(gn->subsolver->solution,tao->solution);CHKERRQ(ierr); 188 ierr = VecCopy(gn->subsolver->gradient,tao->gradient);CHKERRQ(ierr); 189 PetscFunctionReturn(0); 190 } 191 192 static PetscErrorCode TaoSetFromOptions_BRGN(PetscOptionItems *PetscOptionsObject,Tao tao) 193 { 194 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 195 PetscErrorCode ierr; 196 197 PetscFunctionBegin; 198 ierr = PetscOptionsHead(PetscOptionsObject,"least-squares problems with regularizer: ||f(x)||^2 + lambda*g(x), g(x) = ||xk-xkm1||^2 or ||Dx||_1 or user defined function.");CHKERRQ(ierr); 199 ierr = PetscOptionsReal("-tao_brgn_lambda","regularizer weight","",gn->lambda,&gn->lambda,NULL);CHKERRQ(ierr); 200 ierr = PetscOptionsReal("-tao_brgn_epsilon","L1-norm smooth approximation parameter: ||x||_1 = sum(sqrt(x.^2+epsilon^2)-epsilon)","",gn->epsilon,&gn->epsilon,NULL);CHKERRQ(ierr); 201 ierr = PetscOptionsEList("-tao_brgn_reg_type","regularization type", "",BRGN_Table,BRGNRegTypes,BRGN_Table[gn->reg_type],&gn->reg_type,NULL);CHKERRQ(ierr); 202 ierr = PetscOptionsTail();CHKERRQ(ierr); 203 ierr = TaoSetFromOptions(gn->subsolver);CHKERRQ(ierr); 204 PetscFunctionReturn(0); 205 } 206 207 static PetscErrorCode TaoView_BRGN(Tao tao,PetscViewer viewer) 208 { 209 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 210 PetscErrorCode ierr; 211 212 PetscFunctionBegin; 213 ierr = PetscViewerASCIIPushTab(viewer);CHKERRQ(ierr); 214 ierr = TaoView(gn->subsolver,viewer);CHKERRQ(ierr); 215 ierr = PetscViewerASCIIPopTab(viewer);CHKERRQ(ierr); 216 PetscFunctionReturn(0); 217 } 218 219 static PetscErrorCode TaoSetUp_BRGN(Tao tao) 220 { 221 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 222 PetscErrorCode ierr; 223 PetscBool is_bnls,is_bntr,is_bntl; 224 PetscInt i,n,N,K; /* dict has size K*N*/ 225 /*PetscScalar v; */ /* XH: hack to set value of matrix */ 226 227 PetscFunctionBegin; 228 if (!tao->ls_res) SETERRQ(PetscObjectComm((PetscObject)tao),PETSC_ERR_ORDER,"TaoSetResidualRoutine() must be called before setup!"); 229 ierr = PetscObjectTypeCompare((PetscObject)gn->subsolver,TAOBNLS,&is_bnls);CHKERRQ(ierr); 230 ierr = PetscObjectTypeCompare((PetscObject)gn->subsolver,TAOBNTR,&is_bntr);CHKERRQ(ierr); 231 ierr = PetscObjectTypeCompare((PetscObject)gn->subsolver,TAOBNTL,&is_bntl);CHKERRQ(ierr); 232 if ((is_bnls || is_bntr || is_bntl) && !tao->ls_jac) SETERRQ(PetscObjectComm((PetscObject)tao),PETSC_ERR_ORDER,"TaoSetResidualJacobianRoutine() must be called before setup!"); 233 if (!tao->gradient){ 234 ierr = VecDuplicate(tao->solution,&tao->gradient);CHKERRQ(ierr); 235 } 236 if (!gn->x_work){ 237 ierr = VecDuplicate(tao->solution,&gn->x_work);CHKERRQ(ierr); 238 } 239 if (!gn->r_work){ 240 ierr = VecDuplicate(tao->ls_res,&gn->r_work);CHKERRQ(ierr); 241 } 242 if (!gn->x_old) { 243 ierr = VecDuplicate(tao->solution,&gn->x_old);CHKERRQ(ierr); 244 ierr = VecSet(gn->x_old,0.0);CHKERRQ(ierr); 245 } 246 247 /*ierr = VecGetSize(tao->solution,&N);CHKERRQ(ierr);*/ 248 /* TODO: Safeguard against NULL matrix */ 249 /*if (!gn->D)*/ 250 ierr = MatGetSize(gn->D,&K,&N);CHKERRQ(ierr); /* Shell matrices still must have sizes defined */ 251 /* K = N for identity matrix, K=N-1 or N for gradient matrix */ 252 if (!gn->y){ 253 ierr = VecCreate(PETSC_COMM_SELF,&gn->y);CHKERRQ(ierr); 254 ierr = VecSetSizes(gn->y,PETSC_DECIDE,K);CHKERRQ(ierr); 255 ierr = VecSetFromOptions(gn->y);CHKERRQ(ierr); 256 ierr = VecSet(gn->y,0.0);CHKERRQ(ierr); 257 258 } 259 if (!gn->y_work){ 260 ierr = VecDuplicate(gn->y,&gn->y_work);CHKERRQ(ierr); 261 } 262 if (!gn->diag){ 263 ierr = VecDuplicate(gn->y,&gn->diag);CHKERRQ(ierr); 264 ierr = VecSet(gn->diag,0.0);CHKERRQ(ierr); 265 } 266 267 /* XH: debug: check matrix */ 268 #if 0 269 ierr = PetscPrintf(PETSC_COMM_SELF,"-------- Check D matrix: -------- \n"); CHKERRQ(ierr); 270 ierr = MatView(gn->D,PETSC_VIEWER_STDOUT_WORLD);CHKERRQ(ierr); 271 #endif 272 273 if (!tao->setupcalled) { 274 /* Hessian setup */ 275 ierr = VecGetLocalSize(tao->solution,&n);CHKERRQ(ierr); 276 ierr = VecGetSize(tao->solution,&N);CHKERRQ(ierr); 277 ierr = MatSetSizes(gn->H,n,n,N,N);CHKERRQ(ierr); 278 ierr = MatSetType(gn->H,MATSHELL);CHKERRQ(ierr); 279 ierr = MatSetUp(gn->H);CHKERRQ(ierr); 280 ierr = MatShellSetOperation(gn->H,MATOP_MULT,(void (*)(void))GNHessianProd);CHKERRQ(ierr); 281 ierr = MatShellSetContext(gn->H,(void*)gn);CHKERRQ(ierr); 282 /* Subsolver setup,include initial vector and dicttionary D */ 283 ierr = TaoSetUpdate(gn->subsolver,GNHookFunction,(void*)gn);CHKERRQ(ierr); 284 ierr = TaoSetInitialVector(gn->subsolver,tao->solution);CHKERRQ(ierr); 285 if (tao->bounded) { 286 ierr = TaoSetVariableBounds(gn->subsolver,tao->XL,tao->XU);CHKERRQ(ierr); 287 } 288 ierr = TaoSetResidualRoutine(gn->subsolver,tao->ls_res,tao->ops->computeresidual,tao->user_lsresP);CHKERRQ(ierr); 289 ierr = TaoSetJacobianResidualRoutine(gn->subsolver,tao->ls_jac,tao->ls_jac,tao->ops->computeresidualjacobian,tao->user_lsjacP);CHKERRQ(ierr); 290 ierr = TaoSetObjectiveAndGradientRoutine(gn->subsolver,GNObjectiveGradientEval,(void*)gn);CHKERRQ(ierr); 291 ierr = TaoSetHessianRoutine(gn->subsolver,gn->H,gn->H,GNComputeHessian,(void*)gn);CHKERRQ(ierr); 292 /* Propagate some options down */ 293 ierr = TaoSetTolerances(gn->subsolver,tao->gatol,tao->grtol,tao->gttol);CHKERRQ(ierr); 294 ierr = TaoSetMaximumIterations(gn->subsolver,tao->max_it);CHKERRQ(ierr); 295 ierr = TaoSetMaximumFunctionEvaluations(gn->subsolver,tao->max_funcs);CHKERRQ(ierr); 296 for (i=0; i<tao->numbermonitors; ++i) { 297 ierr = TaoSetMonitor(gn->subsolver,tao->monitor[i],tao->monitorcontext[i],tao->monitordestroy[i]);CHKERRQ(ierr); 298 ierr = PetscObjectReference((PetscObject)(tao->monitorcontext[i]));CHKERRQ(ierr); 299 } 300 ierr = TaoSetUp(gn->subsolver);CHKERRQ(ierr); 301 } 302 PetscFunctionReturn(0); 303 } 304 305 static PetscErrorCode TaoDestroy_BRGN(Tao tao) 306 { 307 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 308 PetscErrorCode ierr; 309 310 PetscFunctionBegin; 311 if (tao->setupcalled) { 312 ierr = VecDestroy(&tao->gradient);CHKERRQ(ierr); 313 ierr = VecDestroy(&gn->x_work);CHKERRQ(ierr); 314 ierr = VecDestroy(&gn->r_work);CHKERRQ(ierr); 315 ierr = VecDestroy(&gn->x_old);CHKERRQ(ierr); 316 ierr = VecDestroy(&gn->diag);CHKERRQ(ierr); 317 ierr = VecDestroy(&gn->y);CHKERRQ(ierr); 318 ierr = VecDestroy(&gn->y_work);CHKERRQ(ierr); 319 } 320 ierr = MatDestroy(&gn->H);CHKERRQ(ierr); 321 ierr = MatDestroy(&gn->D);CHKERRQ(ierr); 322 ierr = TaoDestroy(&gn->subsolver);CHKERRQ(ierr); 323 gn->parent = NULL; 324 ierr = PetscFree(tao->data);CHKERRQ(ierr); 325 PetscFunctionReturn(0); 326 } 327 328 /*MC 329 TAOBRGN - Bounded Regularized Gauss-Newton method for solving nonlinear least-squares 330 problems with bound constraints. This algorithm is a thin wrapper around TAOBNTL 331 that constructs the Guass-Newton problem with the user-provided least-squares 332 residual and Jacobian. The problem is regularized with an L2-norm proximal point 333 term. 334 335 Options Database Keys: 336 + -tao_bqnk_max_cg_its - maximum number of bounded conjugate-gradient iterations taken in each Newton loop 337 . -tao_bqnk_init_type - trust radius initialization method ("constant", "direction", "interpolation") 338 . -tao_bqnk_update_type - trust radius update method ("step", "direction", "interpolation") 339 - -tao_bqnk_as_type - active-set estimation method ("none", "bertsekas") 340 341 Level: beginner 342 M*/ 343 PETSC_EXTERN PetscErrorCode TaoCreate_BRGN(Tao tao) 344 { 345 TAO_BRGN *gn; 346 PetscErrorCode ierr; 347 348 PetscFunctionBegin; 349 ierr = PetscNewLog(tao,&gn);CHKERRQ(ierr); 350 351 tao->ops->destroy = TaoDestroy_BRGN; 352 tao->ops->setup = TaoSetUp_BRGN; 353 tao->ops->setfromoptions = TaoSetFromOptions_BRGN; 354 tao->ops->view = TaoView_BRGN; 355 tao->ops->solve = TaoSolve_BRGN; 356 357 tao->data = (void*)gn; 358 gn->lambda = 1e-4; 359 gn->epsilon = 1e-6; 360 gn->parent = tao; 361 362 ierr = MatCreate(PetscObjectComm((PetscObject)tao),&gn->H);CHKERRQ(ierr); 363 ierr = MatSetOptionsPrefix(gn->H,"tao_brgn_hessian_");CHKERRQ(ierr); 364 365 ierr = TaoCreate(PetscObjectComm((PetscObject)tao),&gn->subsolver);CHKERRQ(ierr); 366 ierr = TaoSetType(gn->subsolver,TAOBNLS);CHKERRQ(ierr); 367 ierr = TaoSetOptionsPrefix(gn->subsolver,"tao_brgn_subsolver_");CHKERRQ(ierr); 368 PetscFunctionReturn(0); 369 } 370 371 /*@C 372 TaoBRGNGetSubsolver - Get the pointer to the subsolver inside BRGN 373 374 Collective on Tao 375 376 Level: developer 377 378 Input Parameters: 379 + tao - the Tao solver context 380 - subsolver - the Tao sub-solver context 381 @*/ 382 PetscErrorCode TaoBRGNGetSubsolver(Tao tao,Tao *subsolver) 383 { 384 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 385 386 PetscFunctionBegin; 387 *subsolver = gn->subsolver; 388 PetscFunctionReturn(0); 389 } 390 391 /*@C 392 TaoBRGNSetL1RegularizerWeight - Set the L1-norm regularizer weight for the Gauss-Newton least-squares algorithm 393 394 Collective on Tao 395 396 Level: developer 397 398 Input Parameters: 399 + tao - the Tao solver context 400 - lambda - L1-norm regularizer weight 401 @*/ 402 PetscErrorCode TaoBRGNSetRegularizerWeight(Tao tao,PetscReal lambda) 403 { 404 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 405 406 /* Initialize lambda here */ 407 408 PetscFunctionBegin; 409 gn->lambda = lambda; 410 PetscFunctionReturn(0); 411 } 412 413 /*@C 414 TaoBRGNSetL1SmoothEpsilon - Set the L1-norm smooth approximation parameter for L1-regularized least-squares algorithm 415 416 Collective on Tao 417 418 Level: developer 419 420 Input Parameters: 421 + tao - the Tao solver context 422 - epsilon - L1-norm smooth approximation parameter 423 @*/ 424 PetscErrorCode TaoBRGNSetL1SmoothEpsilon(Tao tao,PetscReal epsilon) 425 { 426 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 427 428 /* Initialize epsilon here */ 429 430 PetscFunctionBegin; 431 gn->epsilon = epsilon; 432 PetscFunctionReturn(0); 433 } 434 435 /*@C 436 TaoBRGNSetDictionaryMatrix - bind the dictionary matrix from user application context to gn->D, for compressed sensing (with least-squares problem) 437 438 Input Parameters: 439 + tao - the Tao context 440 . dict - the user specified dictionary matrix 441 442 Level: developer 443 @*/ 444 PetscErrorCode TaoBRGNSetDictionaryMatrix(Tao tao,Mat dict) 445 { 446 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 447 PetscErrorCode ierr; 448 PetscFunctionBegin; 449 PetscValidHeaderSpecific(tao,TAO_CLASSID,1); 450 if (dict) { 451 PetscValidHeaderSpecific(dict,MAT_CLASSID,2); 452 PetscCheckSameComm(tao,1,dict,2); 453 ierr = PetscObjectReference((PetscObject)dict);CHKERRQ(ierr); 454 } 455 ierr = MatDestroy(&gn->D);CHKERRQ(ierr); 456 gn->D = dict; /* We allow to set a null dictionary, which means we just use default identity matrix? */ 457 PetscFunctionReturn(0); 458 } 459 460 /*@C 461 @*/ 462 PetscErrorCode TaoBRGNSetRegularizerObjectiveAndGradientRoutine(Tao tao,PetscErrorCode (*func)(Tao, Vec, PetscReal *, Vec, void*),void *ctx) 463 { 464 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 465 466 PetscFunctionBegin; 467 PetscValidHeaderSpecific(tao,TAO_CLASSID,1); 468 if (ctx) { 469 gn->reg_obj_ctx = ctx; 470 } 471 if (func) { 472 gn->regularizerobjandgrad = func; 473 } 474 PetscFunctionReturn(0); 475 } 476 477 /*@C 478 @*/ 479 PetscErrorCode TaoBRGNSetRegularizerHessianRoutine(Tao tao,Mat Hreg,PetscErrorCode (*func)(Tao, Vec, Mat, void*),void *ctx) 480 { 481 TAO_BRGN *gn = (TAO_BRGN *)tao->data; 482 PetscErrorCode ierr; 483 484 PetscFunctionBegin; 485 PetscValidHeaderSpecific(tao,TAO_CLASSID,1); 486 if (Hreg) { 487 PetscValidHeaderSpecific(Hreg,MAT_CLASSID,2); 488 PetscCheckSameComm(tao,1,Hreg,2); 489 } else { 490 SETERRQ(PetscObjectComm((PetscObject)tao),PETSC_ERR_ARG_WRONG,"NULL Hessian detected! User must provide valid Hessian for the regularizer."); 491 } 492 if (ctx) { 493 gn->reg_hess_ctx = ctx; 494 } 495 if (func) { 496 gn->regularizerhessian = func; 497 } 498 if (Hreg) { 499 ierr = PetscObjectReference((PetscObject)Hreg);CHKERRQ(ierr); 500 ierr = MatDestroy(&gn->Hreg);CHKERRQ(ierr); 501 gn->Hreg = Hreg; 502 } 503 PetscFunctionReturn(0); 504 }