1 2 #include <petsc/private/taolinesearchimpl.h> 3 #include <../src/tao/linesearch/impls/owarmijo/owarmijo.h> 4 5 #define REPLACE_FIFO 1 6 #define REPLACE_MRU 2 7 8 #define REFERENCE_MAX 1 9 #define REFERENCE_AVE 2 10 #define REFERENCE_MEAN 3 11 12 static PetscErrorCode ProjWork_OWLQN(Vec w,Vec x,Vec gv,PetscReal *gdx) 13 { 14 const PetscReal *xptr,*gptr; 15 PetscReal *wptr; 16 PetscErrorCode ierr; 17 PetscInt low,high,low1,high1,low2,high2,i; 18 19 PetscFunctionBegin; 20 ierr=VecGetOwnershipRange(w,&low,&high);CHKERRQ(ierr); 21 ierr=VecGetOwnershipRange(x,&low1,&high1);CHKERRQ(ierr); 22 ierr=VecGetOwnershipRange(gv,&low2,&high2);CHKERRQ(ierr); 23 24 *gdx=0.0; 25 ierr = VecGetArray(w,&wptr);CHKERRQ(ierr); 26 ierr = VecGetArrayRead(x,&xptr);CHKERRQ(ierr); 27 ierr = VecGetArrayRead(gv,&gptr);CHKERRQ(ierr); 28 29 for (i=0;i<high-low;i++) { 30 if (xptr[i]*wptr[i]<0.0) wptr[i]=0.0; 31 *gdx = *gdx + gptr[i]*(wptr[i]-xptr[i]); 32 } 33 ierr = VecRestoreArray(w,&wptr);CHKERRQ(ierr); 34 ierr = VecRestoreArrayRead(x,&xptr);CHKERRQ(ierr); 35 ierr = VecRestoreArrayRead(gv,&gptr);CHKERRQ(ierr); 36 PetscFunctionReturn(0); 37 } 38 39 #undef __FUNCT__ 40 #define __FUNCT__ "TaoLineSearchDestroy_OWArmijo" 41 static PetscErrorCode TaoLineSearchDestroy_OWArmijo(TaoLineSearch ls) 42 { 43 TaoLineSearch_OWARMIJO *armP = (TaoLineSearch_OWARMIJO *)ls->data; 44 PetscErrorCode ierr; 45 46 PetscFunctionBegin; 47 ierr = PetscFree(armP->memory);CHKERRQ(ierr); 48 if (armP->x) { 49 ierr = PetscObjectDereference((PetscObject)armP->x);CHKERRQ(ierr); 50 } 51 ierr = VecDestroy(&armP->work);CHKERRQ(ierr); 52 ierr = PetscFree(ls->data);CHKERRQ(ierr); 53 PetscFunctionReturn(0); 54 } 55 56 #undef __FUNCT__ 57 #define __FUNCT__ "TaoLineSearchSetFromOptions_OWArmijo" 58 static PetscErrorCode TaoLineSearchSetFromOptions_OWArmijo(PetscOptions *PetscOptionsObject,TaoLineSearch ls) 59 { 60 TaoLineSearch_OWARMIJO *armP = (TaoLineSearch_OWARMIJO *)ls->data; 61 PetscErrorCode ierr; 62 63 PetscFunctionBegin; 64 ierr = PetscOptionsHead(PetscOptionsObject,"OWArmijo linesearch options");CHKERRQ(ierr); 65 ierr = PetscOptionsReal("-tao_ls_OWArmijo_alpha", "initial reference constant", "", armP->alpha, &armP->alpha,NULL);CHKERRQ(ierr); 66 ierr = PetscOptionsReal("-tao_ls_OWArmijo_beta_inf", "decrease constant one", "", armP->beta_inf, &armP->beta_inf,NULL);CHKERRQ(ierr); 67 ierr = PetscOptionsReal("-tao_ls_OWArmijo_beta", "decrease constant", "", armP->beta, &armP->beta,NULL);CHKERRQ(ierr); 68 ierr = PetscOptionsReal("-tao_ls_OWArmijo_sigma", "acceptance constant", "", armP->sigma, &armP->sigma,NULL);CHKERRQ(ierr); 69 ierr = PetscOptionsInt("-tao_ls_OWArmijo_memory_size", "number of historical elements", "", armP->memorySize, &armP->memorySize,NULL);CHKERRQ(ierr); 70 ierr = PetscOptionsInt("-tao_ls_OWArmijo_reference_policy", "policy for updating reference value", "", armP->referencePolicy, &armP->referencePolicy,NULL);CHKERRQ(ierr); 71 ierr = PetscOptionsInt("-tao_ls_OWArmijo_replacement_policy", "policy for updating memory", "", armP->replacementPolicy, &armP->replacementPolicy,NULL);CHKERRQ(ierr); 72 ierr = PetscOptionsBool("-tao_ls_OWArmijo_nondescending","Use nondescending OWArmijo algorithm","",armP->nondescending,&armP->nondescending,NULL);CHKERRQ(ierr); 73 ierr = PetscOptionsTail();CHKERRQ(ierr); 74 PetscFunctionReturn(0); 75 } 76 77 #undef __FUNCT__ 78 #define __FUNCT__ "TaoLineSearchView_OWArmijo" 79 static PetscErrorCode TaoLineSearchView_OWArmijo(TaoLineSearch ls, PetscViewer pv) 80 { 81 TaoLineSearch_OWARMIJO *armP = (TaoLineSearch_OWARMIJO *)ls->data; 82 PetscBool isascii; 83 PetscErrorCode ierr; 84 85 PetscFunctionBegin; 86 ierr = PetscObjectTypeCompare((PetscObject)pv, PETSCVIEWERASCII, &isascii);CHKERRQ(ierr); 87 if (isascii) { 88 ierr = PetscViewerASCIIPrintf(pv," maxf=%D, ftol=%g, gtol=%g\n",ls->max_funcs, (double)ls->rtol, (double)ls->ftol);CHKERRQ(ierr); 89 ierr=PetscViewerASCIIPrintf(pv," OWArmijo linesearch",armP->alpha);CHKERRQ(ierr); 90 if (armP->nondescending) { 91 ierr = PetscViewerASCIIPrintf(pv, " (nondescending)");CHKERRQ(ierr); 92 } 93 ierr=PetscViewerASCIIPrintf(pv,": alpha=%g beta=%g ",(double)armP->alpha,(double)armP->beta);CHKERRQ(ierr); 94 ierr=PetscViewerASCIIPrintf(pv,"sigma=%g ",(double)armP->sigma);CHKERRQ(ierr); 95 ierr=PetscViewerASCIIPrintf(pv,"memsize=%D\n",armP->memorySize);CHKERRQ(ierr); 96 } 97 PetscFunctionReturn(0); 98 } 99 100 #undef __FUNCT__ 101 #define __FUNCT__ "TaoLineSearchApply_OWArmijo" 102 /* @ TaoApply_OWArmijo - This routine performs a linesearch. It 103 backtracks until the (nonmonotone) OWArmijo conditions are satisfied. 104 105 Input Parameters: 106 + tao - TAO_SOLVER context 107 . X - current iterate (on output X contains new iterate, X + step*S) 108 . S - search direction 109 . f - merit function evaluated at X 110 . G - gradient of merit function evaluated at X 111 . W - work vector 112 - step - initial estimate of step length 113 114 Output parameters: 115 + f - merit function evaluated at new iterate, X + step*S 116 . G - gradient of merit function evaluated at new iterate, X + step*S 117 . X - new iterate 118 - step - final step length 119 120 Info is set to one of: 121 . 0 - the line search succeeds; the sufficient decrease 122 condition and the directional derivative condition hold 123 124 negative number if an input parameter is invalid 125 - -1 - step < 0 126 127 positive number > 1 if the line search otherwise terminates 128 + 1 - Step is at the lower bound, stepmin. 129 @ */ 130 static PetscErrorCode TaoLineSearchApply_OWArmijo(TaoLineSearch ls, Vec x, PetscReal *f, Vec g, Vec s) 131 { 132 TaoLineSearch_OWARMIJO *armP = (TaoLineSearch_OWARMIJO *)ls->data; 133 PetscErrorCode ierr; 134 PetscInt i; 135 PetscReal fact, ref, gdx; 136 PetscInt idx; 137 PetscBool g_computed=PETSC_FALSE; /* to prevent extra gradient computation */ 138 Vec g_old; 139 PetscReal owlqn_minstep=0.005; 140 PetscReal partgdx; 141 MPI_Comm comm; 142 143 PetscFunctionBegin; 144 ierr = PetscObjectGetComm((PetscObject)ls,&comm);CHKERRQ(ierr); 145 fact = 0.0; 146 ls->nfeval=0; 147 ls->reason = TAOLINESEARCH_CONTINUE_ITERATING; 148 if (!armP->work) { 149 ierr = VecDuplicate(x,&armP->work);CHKERRQ(ierr); 150 armP->x = x; 151 ierr = PetscObjectReference((PetscObject)armP->x);CHKERRQ(ierr); 152 } else if (x != armP->x) { 153 ierr = VecDestroy(&armP->work);CHKERRQ(ierr); 154 ierr = VecDuplicate(x,&armP->work);CHKERRQ(ierr); 155 ierr = PetscObjectDereference((PetscObject)armP->x);CHKERRQ(ierr); 156 armP->x = x; 157 ierr = PetscObjectReference((PetscObject)armP->x);CHKERRQ(ierr); 158 } 159 160 /* Check linesearch parameters */ 161 if (armP->alpha < 1) { 162 ierr = PetscInfo1(ls,"OWArmijo line search error: alpha (%g) < 1\n", (double)armP->alpha);CHKERRQ(ierr); 163 ls->reason=TAOLINESEARCH_FAILED_BADPARAMETER; 164 } else if ((armP->beta <= 0) || (armP->beta >= 1)) { 165 ierr = PetscInfo1(ls,"OWArmijo line search error: beta (%g) invalid\n", (double)armP->beta);CHKERRQ(ierr); 166 ls->reason=TAOLINESEARCH_FAILED_BADPARAMETER; 167 } else if ((armP->beta_inf <= 0) || (armP->beta_inf >= 1)) { 168 ierr = PetscInfo1(ls,"OWArmijo line search error: beta_inf (%g) invalid\n", (double)armP->beta_inf);CHKERRQ(ierr); 169 ls->reason=TAOLINESEARCH_FAILED_BADPARAMETER; 170 } else if ((armP->sigma <= 0) || (armP->sigma >= 0.5)) { 171 ierr = PetscInfo1(ls,"OWArmijo line search error: sigma (%g) invalid\n", (double)armP->sigma);CHKERRQ(ierr); 172 ls->reason=TAOLINESEARCH_FAILED_BADPARAMETER; 173 } else if (armP->memorySize < 1) { 174 ierr = PetscInfo1(ls,"OWArmijo line search error: memory_size (%D) < 1\n", armP->memorySize);CHKERRQ(ierr); 175 ls->reason=TAOLINESEARCH_FAILED_BADPARAMETER; 176 } else if ((armP->referencePolicy != REFERENCE_MAX) && (armP->referencePolicy != REFERENCE_AVE) && (armP->referencePolicy != REFERENCE_MEAN)) { 177 ierr = PetscInfo(ls,"OWArmijo line search error: reference_policy invalid\n");CHKERRQ(ierr); 178 ls->reason=TAOLINESEARCH_FAILED_BADPARAMETER; 179 } else if ((armP->replacementPolicy != REPLACE_FIFO) && (armP->replacementPolicy != REPLACE_MRU)) { 180 ierr = PetscInfo(ls,"OWArmijo line search error: replacement_policy invalid\n");CHKERRQ(ierr); 181 ls->reason=TAOLINESEARCH_FAILED_BADPARAMETER; 182 } else if (PetscIsInfOrNanReal(*f)) { 183 ierr = PetscInfo(ls,"OWArmijo line search error: initial function inf or nan\n");CHKERRQ(ierr); 184 ls->reason=TAOLINESEARCH_FAILED_BADPARAMETER; 185 } 186 187 if (ls->reason != TAOLINESEARCH_CONTINUE_ITERATING) PetscFunctionReturn(0); 188 189 /* Check to see of the memory has been allocated. If not, allocate 190 the historical array and populate it with the initial function 191 values. */ 192 if (!armP->memory) { 193 ierr = PetscMalloc1(armP->memorySize, &armP->memory );CHKERRQ(ierr); 194 } 195 196 if (!armP->memorySetup) { 197 for (i = 0; i < armP->memorySize; i++) { 198 armP->memory[i] = armP->alpha*(*f); 199 } 200 armP->current = 0; 201 armP->lastReference = armP->memory[0]; 202 armP->memorySetup=PETSC_TRUE; 203 } 204 205 /* Calculate reference value (MAX) */ 206 ref = armP->memory[0]; 207 idx = 0; 208 209 for (i = 1; i < armP->memorySize; i++) { 210 if (armP->memory[i] > ref) { 211 ref = armP->memory[i]; 212 idx = i; 213 } 214 } 215 216 if (armP->referencePolicy == REFERENCE_AVE) { 217 ref = 0; 218 for (i = 0; i < armP->memorySize; i++) { 219 ref += armP->memory[i]; 220 } 221 ref = ref / armP->memorySize; 222 ref = PetscMax(ref, armP->memory[armP->current]); 223 } else if (armP->referencePolicy == REFERENCE_MEAN) { 224 ref = PetscMin(ref, 0.5*(armP->lastReference + armP->memory[armP->current])); 225 } 226 227 if (armP->nondescending) { 228 fact = armP->sigma; 229 } 230 231 ierr = VecDuplicate(g,&g_old);CHKERRQ(ierr); 232 ierr = VecCopy(g,g_old);CHKERRQ(ierr); 233 234 ls->step = ls->initstep; 235 while (ls->step >= owlqn_minstep && ls->nfeval < ls->max_funcs) { 236 /* Calculate iterate */ 237 ierr = VecCopy(x,armP->work);CHKERRQ(ierr); 238 ierr = VecAXPY(armP->work,ls->step,s);CHKERRQ(ierr); 239 240 partgdx=0.0; 241 ierr = ProjWork_OWLQN(armP->work,x,g_old,&partgdx);CHKERRQ(ierr); 242 ierr = MPI_Allreduce(&partgdx,&gdx,1,MPIU_REAL,MPIU_SUM,comm);CHKERRQ(ierr); 243 244 /* Check the condition of gdx */ 245 if (PetscIsInfOrNanReal(gdx)) { 246 ierr = PetscInfo1(ls,"Initial Line Search step * g is Inf or Nan (%g)\n",(double)gdx);CHKERRQ(ierr); 247 ls->reason=TAOLINESEARCH_FAILED_INFORNAN; 248 PetscFunctionReturn(0); 249 } 250 if (gdx >= 0.0) { 251 ierr = PetscInfo1(ls,"Initial Line Search step is not descent direction (g's=%g)\n",(double)gdx);CHKERRQ(ierr); 252 ls->reason = TAOLINESEARCH_FAILED_ASCENT; 253 PetscFunctionReturn(0); 254 } 255 256 /* Calculate function at new iterate */ 257 ierr = TaoLineSearchComputeObjectiveAndGradient(ls,armP->work,f,g);CHKERRQ(ierr); 258 g_computed=PETSC_TRUE; 259 260 if (ls->step == ls->initstep) { 261 ls->f_fullstep = *f; 262 } 263 264 if (PetscIsInfOrNanReal(*f)) { 265 ls->step *= armP->beta_inf; 266 } else { 267 /* Check descent condition */ 268 if (armP->nondescending && *f <= ref - ls->step*fact*ref) break; 269 if (!armP->nondescending && *f <= ref + armP->sigma * gdx) break; 270 ls->step *= armP->beta; 271 } 272 } 273 ierr = VecDestroy(&g_old);CHKERRQ(ierr); 274 275 /* Check termination */ 276 if (PetscIsInfOrNanReal(*f)) { 277 ierr = PetscInfo(ls, "Function is inf or nan.\n");CHKERRQ(ierr); 278 ls->reason = TAOLINESEARCH_FAILED_BADPARAMETER; 279 } else if (ls->step < owlqn_minstep) { 280 ierr = PetscInfo(ls, "Step length is below tolerance.\n");CHKERRQ(ierr); 281 ls->reason = TAOLINESEARCH_HALTED_RTOL; 282 } else if (ls->nfeval >= ls->max_funcs) { 283 ierr = PetscInfo2(ls, "Number of line search function evals (%D) > maximum allowed (%D)\n",ls->nfeval, ls->max_funcs);CHKERRQ(ierr); 284 ls->reason = TAOLINESEARCH_HALTED_MAXFCN; 285 } 286 if (ls->reason) PetscFunctionReturn(0); 287 288 /* Successful termination, update memory */ 289 armP->lastReference = ref; 290 if (armP->replacementPolicy == REPLACE_FIFO) { 291 armP->memory[armP->current++] = *f; 292 if (armP->current >= armP->memorySize) { 293 armP->current = 0; 294 } 295 } else { 296 armP->current = idx; 297 armP->memory[idx] = *f; 298 } 299 300 /* Update iterate and compute gradient */ 301 ierr = VecCopy(armP->work,x);CHKERRQ(ierr); 302 if (!g_computed) { 303 ierr = TaoLineSearchComputeGradient(ls, x, g);CHKERRQ(ierr); 304 } 305 ierr = PetscInfo2(ls, "%D function evals in line search, step = %10.4f\n",ls->nfeval, (double)ls->step);CHKERRQ(ierr); 306 PetscFunctionReturn(0); 307 } 308 309 #undef __FUNCT__ 310 #define __FUNCT__ "TaoLineSearchCreate_OWArmijo" 311 PETSC_EXTERN PetscErrorCode TaoLineSearchCreate_OWArmijo(TaoLineSearch ls) 312 { 313 TaoLineSearch_OWARMIJO *armP; 314 PetscErrorCode ierr; 315 316 PetscFunctionBegin; 317 PetscValidHeaderSpecific(ls,TAOLINESEARCH_CLASSID,1); 318 ierr = PetscNewLog(ls,&armP);CHKERRQ(ierr); 319 320 armP->memory = NULL; 321 armP->alpha = 1.0; 322 armP->beta = 0.25; 323 armP->beta_inf = 0.25; 324 armP->sigma = 1e-4; 325 armP->memorySize = 1; 326 armP->referencePolicy = REFERENCE_MAX; 327 armP->replacementPolicy = REPLACE_MRU; 328 armP->nondescending=PETSC_FALSE; 329 ls->data = (void*)armP; 330 ls->initstep=0.1; 331 ls->ops->setup=0; 332 ls->ops->reset=0; 333 ls->ops->apply=TaoLineSearchApply_OWArmijo; 334 ls->ops->view = TaoLineSearchView_OWArmijo; 335 ls->ops->destroy = TaoLineSearchDestroy_OWArmijo; 336 ls->ops->setfromoptions = TaoLineSearchSetFromOptions_OWArmijo; 337 PetscFunctionReturn(0); 338 } 339 340