IPOPTSolver.cc 15 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
//=============================================================================
//
//  CLASS IPOPTSolver - IMPLEMENTATION
//
//=============================================================================

//== INCLUDES =================================================================

//== COMPILE-TIME PACKAGE REQUIREMENTS ========================================
#include <CoMISo/Config/config.hh>
#if COMISO_IPOPT_AVAILABLE
//=============================================================================


#include "IPOPTSolver.hh"
16

17 18 19 20 21 22 23 24
//== NAMESPACES ===============================================================

namespace COMISO {

//== IMPLEMENTATION ========================================================== 


// Constructor
25 26
IPOPTSolver::
IPOPTSolver()
27
{
28 29
  // Create an instance of the IpoptApplication
  app_ = IpoptApplicationFactory();
30 31

  // Switch to HSL if available in Comiso
32 33 34
  #if COMISO_HSL_AVAILABLE
    app_->Options()->SetStringValue("linear_solver", "ma57");
  #endif
35 36 37

  // Restrict memory to be able to run larger problems on windows
  // with the default mumps solver
38 39 40
  #ifdef WIN32
    app_->Options()->SetIntegerValue("mumps_mem_percent", 5);
  #endif
41 42

  // set default parameters
43
  app_->Options()->SetIntegerValue("max_iter", 100);
44 45 46 47
  //  app->Options()->SetStringValue("derivative_test", "second-order");
  //  app->Options()->SetIntegerValue("print_level", 0);
  //  app->Options()->SetStringValue("expect_infeasible_problem", "yes");

48
  print_level_ = 5;
49 50
}

51

52
//-----------------------------------------------------------------------------
53

54 55


56 57 58
int
IPOPTSolver::
solve(NProblemInterface* _problem, const std::vector<NConstraintInterface*>& _constraints)
59
{
60 61 62 63 64 65 66 67 68 69 70 71 72
  //----------------------------------------------------------------------------
  // 0. Check whether hessian_approximation is active
  //----------------------------------------------------------------------------
  bool hessian_approximation = false;
  std::string ha, p;
  app().Options()->GetStringValue("hessian_approximation", ha, p);
  if(ha != "exact")
  {
    if(print_level_>=2)
      std::cerr << "Hessian approximation is enabled" << std::endl;
    hessian_approximation = true;
  }

73 74 75
  //----------------------------------------------------------------------------
  // 1. Create an instance of IPOPT NLP
  //----------------------------------------------------------------------------
76
  Ipopt::SmartPtr<Ipopt::TNLP> np = new NProblemIPOPT(_problem, _constraints, hessian_approximation);
77 78 79 80 81 82
  NProblemIPOPT* np2 = dynamic_cast<NProblemIPOPT*> (Ipopt::GetRawPtr(np));

  //----------------------------------------------------------------------------
  // 2. exploit special characteristics of problem
  //----------------------------------------------------------------------------

83 84
  if(print_level_>=2)
    std::cerr << "exploit detected special properties: ";
85 86
  if(np2->hessian_constant())
  {
87 88 89
    if(print_level_>=2)
      std::cerr << "*constant hessian* ";
    app().Options()->SetStringValue("hessian_constant", "yes");
90 91 92 93
  }

  if(np2->jac_c_constant())
  {
94 95 96
    if(print_level_>=2)
      std::cerr << "*constant jacobian of equality constraints* ";
    app().Options()->SetStringValue("jac_c_constant", "yes");
97 98 99 100
  }

  if(np2->jac_d_constant())
  {
101 102 103
    if(print_level_>=2)
      std::cerr << "*constant jacobian of in-equality constraints*";
    app().Options()->SetStringValue("jac_d_constant", "yes");
104
  }
105 106 107

  if(print_level_>=2)
    std::cerr << std::endl;
108 109 110 111 112 113

  //----------------------------------------------------------------------------
  // 3. solve problem
  //----------------------------------------------------------------------------

  // Initialize the IpoptApplication and process the options
114 115 116 117 118 119 120
  Ipopt::ApplicationReturnStatus status;
  status = app_->Initialize();
  if (status != Ipopt::Solve_Succeeded)
  {
    if(print_level_>=2)
      printf("\n\n*** Error IPOPT during initialization!\n");
  }
121

122
  status = app_->OptimizeTNLP( np);
123 124 125 126

  //----------------------------------------------------------------------------
  // 4. output statistics
  //----------------------------------------------------------------------------
127 128 129 130 131 132 133 134 135 136 137 138
  if(print_level_>=2)
    if (status == Ipopt::Solve_Succeeded || status == Ipopt::Solved_To_Acceptable_Level)
    {
      // Retrieve some statistics about the solve
      Ipopt::Index iter_count = app_->Statistics()->IterationCount();
      printf("\n\n*** IPOPT: The problem solved in %d iterations!\n", iter_count);

      Ipopt::Number final_obj = app_->Statistics()->FinalObjective();
      printf("\n\n*** IPOPT: The final value of the objective function is %e.\n", final_obj);
    }

  return status;
139 140 141 142 143 144
}


//-----------------------------------------------------------------------------


145 146 147 148

int
IPOPTSolver::
solve(NProblemInterface*                        _problem,
149 150 151 152 153 154
      const std::vector<NConstraintInterface*>& _constraints,
      const std::vector<NConstraintInterface*>& _lazy_constraints,
      const double                              _almost_infeasible,
      const int                                 _max_passes        )
{
  //----------------------------------------------------------------------------
155 156 157 158 159 160 161 162 163 164 165 166
  // 0. Check whether hessian_approximation is active
  //----------------------------------------------------------------------------
  bool hessian_approximation = false;
  std::string ha, p;
  app().Options()->GetStringValue("hessian_approximation", ha, p);
  if(ha != "exact")
  {
    if(print_level_>=2)
      std::cerr << "Hessian approximation is enabled" << std::endl;
    hessian_approximation = true;
  }

167
  //----------------------------------------------------------------------------
168 169 170 171
  // 0. Initialize IPOPT Applicaiton
  //----------------------------------------------------------------------------

  StopWatch sw; sw.start();
172 173 174

  // Initialize the IpoptApplication and process the options
  Ipopt::ApplicationReturnStatus status;
175
  status = app_->Initialize();
176
  if (status != Ipopt::Solve_Succeeded)
177 178 179
  {
    printf("\n\n*** Error IPOPT during initialization!\n");
  }
180 181 182 183 184 185 186 187 188 189 190 191 192 193

  bool feasible_point_found = false;
  int  cur_pass = 0;
  double acceptable_tolerance = 0.01; // hack: read out from ipopt!!!
  // copy default constraints
  std::vector<NConstraintInterface*> constraints = _constraints;
  std::vector<bool> lazy_added(_lazy_constraints.size(),false);

  // cache statistics of all iterations
  std::vector<int> n_inf;
  std::vector<int> n_almost_inf;

  while(!feasible_point_found && cur_pass <(_max_passes-1))
  {
194

195 196 197 198
    ++cur_pass;
    //----------------------------------------------------------------------------
    // 1. Create an instance of current IPOPT NLP
    //----------------------------------------------------------------------------
199
    Ipopt::SmartPtr<Ipopt::TNLP> np = new NProblemIPOPT(_problem, constraints, hessian_approximation);
200 201 202 203 204 205 206 207
    NProblemIPOPT* np2 = dynamic_cast<NProblemIPOPT*> (Ipopt::GetRawPtr(np));
    // enable caching of solution
    np2->store_solution() = true;

    //----------------------------------------------------------------------------
    // 2. exploit special characteristics of problem
    //----------------------------------------------------------------------------

208
    if(print_level_>=2) std::cerr << "detected special properties which will be exploit: ";
209 210
    if(np2->hessian_constant())
    {
211 212
      if(print_level_>=2) std::cerr << "*constant hessian* ";
      app().Options()->SetStringValue("hessian_constant", "yes");
213 214 215 216
    }

    if(np2->jac_c_constant())
    {
217 218
      if(print_level_>=2) std::cerr << "*constant jacobian of equality constraints* ";
      app().Options()->SetStringValue("jac_c_constant", "yes");
219 220 221 222
    }

    if(np2->jac_d_constant())
    {
223 224
      if(print_level_>=2) std::cerr << "*constant jacobian of in-equality constraints*";
      app().Options()->SetStringValue("jac_d_constant", "yes");
225
    }
226
    if(print_level_>=2) std::cerr << std::endl;
227 228 229 230

    //----------------------------------------------------------------------------
    // 3. solve problem
    //----------------------------------------------------------------------------
231
    status = app_->OptimizeTNLP( np);
232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296

    // check lazy constraints
    n_inf.push_back(0);
    n_almost_inf.push_back(0);
    feasible_point_found = true;
    for(unsigned int i=0; i<_lazy_constraints.size(); ++i)
      if(!lazy_added[i])
      {
        NConstraintInterface* lc = _lazy_constraints[i];
        double v = lc->eval_constraint(&(np2->solution()[0]));
        bool inf        = false;
        bool almost_inf = false;

        if(lc->constraint_type() == NConstraintInterface::NC_EQUAL)
        {
          v = std::abs(v);
          if(v>acceptable_tolerance)
            inf = true;
          else
            if(v>_almost_infeasible)
              almost_inf = true;
        }
        else
          if(lc->constraint_type() == NConstraintInterface::NC_GREATER_EQUAL)
          {
            if(v<-acceptable_tolerance)
              inf = true;
            else
              if(v<_almost_infeasible)
                almost_inf = true;
          }
          else
            if(lc->constraint_type() == NConstraintInterface::NC_LESS_EQUAL)
            {
              if(v>acceptable_tolerance)
                inf = true;
              else
                if(v>-_almost_infeasible)
                  almost_inf = true;
            }

        // infeasible?
        if(inf)
        {
          constraints.push_back(lc);
          lazy_added[i] = true;
          feasible_point_found = false;
          ++n_inf.back();
        }

        // almost violated or violated? -> add to constraints
        if(almost_inf)
        {
          constraints.push_back(lc);
          lazy_added[i] = true;
          ++n_almost_inf.back();
        }
      }
  }

  // no termination after max number of passes?
  if(!feasible_point_found)
  {
    ++cur_pass;

297
    std::cerr << "*************** could not find feasible point after " << _max_passes-1 << " -> solving with all lazy constraints..." << std::endl;
298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313
    for(unsigned int i=0; i<_lazy_constraints.size(); ++i)
      if(!lazy_added[i])
        constraints.push_back(_lazy_constraints[i]);

    //----------------------------------------------------------------------------
    // 1. Create an instance of current IPOPT NLP
    //----------------------------------------------------------------------------
    Ipopt::SmartPtr<Ipopt::TNLP> np = new NProblemIPOPT(_problem, constraints);
    NProblemIPOPT* np2 = dynamic_cast<NProblemIPOPT*> (Ipopt::GetRawPtr(np));
    // enable caching of solution
    np2->store_solution() = true;

    //----------------------------------------------------------------------------
    // 2. exploit special characteristics of problem
    //----------------------------------------------------------------------------

314
    if(print_level_>=2) std::cerr << "exploit detected special properties: ";
315 316
    if(np2->hessian_constant())
    {
317 318
      if(print_level_>=2) std::cerr << "*constant hessian* ";
      app().Options()->SetStringValue("hessian_constant", "yes");
319 320 321 322
    }

    if(np2->jac_c_constant())
    {
323 324
      if(print_level_>=2) std::cerr << "*constant jacobian of equality constraints* ";
      app().Options()->SetStringValue("jac_c_constant", "yes");
325 326 327 328
    }

    if(np2->jac_d_constant())
    {
329 330
      if(print_level_>=2) std::cerr << "*constant jacobian of in-equality constraints*";
      app().Options()->SetStringValue("jac_d_constant", "yes");
331
    }
332
    if(print_level_>=2) std::cerr << std::endl;
333 334 335 336

    //----------------------------------------------------------------------------
    // 3. solve problem
    //----------------------------------------------------------------------------
337
    status = app_->OptimizeTNLP( np);
338 339
  }

340 341
  const double overall_time = sw.stop()/1000.0;

342 343 344
  //----------------------------------------------------------------------------
  // 4. output statistics
  //----------------------------------------------------------------------------
345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367
  if (status == Ipopt::Solve_Succeeded || status == Ipopt::Solved_To_Acceptable_Level)
  {
    if(print_level_>=2)
    {
      // Retrieve some statistics about the solve
      Ipopt::Index iter_count = app_->Statistics()->IterationCount();
      printf("\n\n*** IPOPT: The problem solved in %d iterations!\n", iter_count);

      Ipopt::Number final_obj = app_->Statistics()->FinalObjective();
      printf("\n\n*** IPOPT: The final value of the objective function is %e.\n", final_obj);
    }
  }

  if(print_level_>=2)
  {
    std::cerr <<"############# IPOPT with lazy constraints statistics ###############" << std::endl;
    std::cerr << "overall time: " << overall_time << "s" << std::endl;
    std::cerr << "#passes     : " << cur_pass << "( of " << _max_passes << ")" << std::endl;
    for(unsigned int i=0; i<n_inf.size(); ++i)
      std::cerr << "pass " << i << " induced " << n_inf[i] << " infeasible and " << n_almost_inf[i] << " almost infeasible" << std::endl;
  }

  return status;
368 369 370 371 372 373
}


//-----------------------------------------------------------------------------


374 375 376
int
IPOPTSolver::
solve(NProblemInterface*    _problem)
377 378
{
  std::vector<NConstraintInterface*> constraints;
379
  return this->solve(_problem, constraints);
380 381 382 383 384 385
}


//-----------------------------------------------------------------------------


386 387 388
int
IPOPTSolver::
solve(NProblemGmmInterface* _problem, std::vector<NConstraintInterface*>& _constraints)
389
{
390
  std::cerr << "****** Warning: NProblemGmmInterface is deprecated!!! -> use NProblemInterface *******\n";
391 392 393 394 395 396 397 398 399 400 401

  //----------------------------------------------------------------------------
  // 1. Create an instance of IPOPT NLP
  //----------------------------------------------------------------------------
  Ipopt::SmartPtr<Ipopt::TNLP> np = new NProblemGmmIPOPT(_problem, _constraints);

  //----------------------------------------------------------------------------
  // 2. solve problem
  //----------------------------------------------------------------------------

  // Initialize the IpoptApplication and process the options
402 403
  Ipopt::ApplicationReturnStatus status;
  status = app_->Initialize();
404
  if (status != Ipopt::Solve_Succeeded)
405 406 407
  {
    printf("\n\n*** Error IPOPT during initialization!\n");
  }
408 409 410 411

  //----------------------------------------------------------------------------
  // 3. solve problem
  //----------------------------------------------------------------------------
412
  status = app_->OptimizeTNLP(np);
413 414 415 416

  //----------------------------------------------------------------------------
  // 4. output statistics
  //----------------------------------------------------------------------------
417 418 419 420 421
  if (status == Ipopt::Solve_Succeeded || status == Ipopt::Solved_To_Acceptable_Level)
  {
    // Retrieve some statistics about the solve
    Ipopt::Index iter_count = app_->Statistics()->IterationCount();
    printf("\n\n*** IPOPT: The problem solved in %d iterations!\n", iter_count);
422

423 424 425
    Ipopt::Number final_obj = app_->Statistics()->FinalObjective();
    printf("\n\n*** IPOPT: The final value of the objective function is %e.\n", final_obj);
  }
426

427
  return status;
428 429 430 431 432 433 434 435
}


//=============================================================================
} // namespace COMISO
//=============================================================================
#endif // COMISO_IPOPT_AVAILABLE
//=============================================================================