Changeset f68c68 for src/LevMartester.cpp

Timestamp:

Feb 24, 2013, 12:58:52 PM (12 years ago)

Author:

Frederik Heber <heber@…>

Branches:

Action_Thermostats, Add_AtomRandomPerturbation, Add_FitFragmentPartialChargesAction, Add_RotateAroundBondAction, Add_SelectAtomByNameAction, Added_ParseSaveFragmentResults, AddingActions_SaveParseParticleParameters, Adding_Graph_to_ChangeBondActions, Adding_MD_integration_tests, Adding_ParticleName_to_Atom, Adding_StructOpt_integration_tests, AtomFragments, Automaking_mpqc_open, AutomationFragmentation_failures, Candidate_v1.5.4, Candidate_v1.6.0, Candidate_v1.6.1, ChangeBugEmailaddress, ChangingTestPorts, ChemicalSpaceEvaluator, CombiningParticlePotentialParsing, Combining_Subpackages, Debian_Package_split, Debian_package_split_molecuildergui_only, Disabling_MemDebug, Docu_Python_wait, EmpiricalPotential_contain_HomologyGraph, EmpiricalPotential_contain_HomologyGraph_documentation, Enable_parallel_make_install, Enhance_userguide, Enhanced_StructuralOptimization, Enhanced_StructuralOptimization_continued, Example_ManyWaysToTranslateAtom, Exclude_Hydrogens_annealWithBondGraph, FitPartialCharges_GlobalError, Fix_BoundInBox_CenterInBox_MoleculeActions, Fix_ChargeSampling_PBC, Fix_ChronosMutex, Fix_FitPartialCharges, Fix_FitPotential_needs_atomicnumbers, Fix_ForceAnnealing, Fix_IndependentFragmentGrids, Fix_ParseParticles, Fix_ParseParticles_split_forward_backward_Actions, Fix_PopActions, Fix_QtFragmentList_sorted_selection, Fix_Restrictedkeyset_FragmentMolecule, Fix_StatusMsg, Fix_StepWorldTime_single_argument, Fix_Verbose_Codepatterns, Fix_fitting_potentials, Fixes, ForceAnnealing_goodresults, ForceAnnealing_oldresults, ForceAnnealing_tocheck, ForceAnnealing_with_BondGraph, ForceAnnealing_with_BondGraph_continued, ForceAnnealing_with_BondGraph_continued_betteresults, ForceAnnealing_with_BondGraph_contraction-expansion, FragmentAction_writes_AtomFragments, FragmentMolecule_checks_bonddegrees, GeometryObjects, Gui_Fixes, Gui_displays_atomic_force_velocity, ImplicitCharges, IndependentFragmentGrids, IndependentFragmentGrids_IndividualZeroInstances, IndependentFragmentGrids_IntegrationTest, IndependentFragmentGrids_Sole_NN_Calculation, JobMarket_RobustOnKillsSegFaults, JobMarket_StableWorkerPool, JobMarket_unresolvable_hostname_fix, MoreRobust_FragmentAutomation, ODR_violation_mpqc_open, PartialCharges_OrthogonalSummation, PdbParser_setsAtomName, PythonUI_with_named_parameters, QtGui_reactivate_TimeChanged_changes, Recreated_GuiChecks, Rewrite_FitPartialCharges, RotateToPrincipalAxisSystem_UndoRedo, SaturateAtoms_findBestMatching, SaturateAtoms_singleDegree, StoppableMakroAction, Subpackage_CodePatterns, Subpackage_JobMarket, Subpackage_LinearAlgebra, Subpackage_levmar, Subpackage_mpqc_open, Subpackage_vmg, Switchable_LogView, ThirdParty_MPQC_rebuilt_buildsystem, TrajectoryDependenant_MaxOrder, TremoloParser_IncreasedPrecision, TremoloParser_MultipleTimesteps, TremoloParser_setsAtomName, Ubuntu_1604_changes, stable

Children:

68172a

Parents:

8aa597

git-author:

Frederik Heber <heber@…> (10/15/12 07:43:23)

git-committer:

Frederik Heber <heber@…> (02/24/13 12:58:52)

Message:

Tersoff now also uses TrainingData class.

• shifted calculation of errors into two new TrainingData functions: getL2Error() and getLMaxError().

File:

• src/LevMartester.cpp (modified) (6 diffs)

src/LevMartester.cpp

@@ -239 +239 @@
    */
   void operator()(const range_t &range) {
+    double EnergySum = 0.; //std::numeric_limits<double>::max();
+    size_t counter = 0.;
+    for (HomologyContainer::const_iterator iter = range.first; iter != range.second; ++iter) {
+      const double &energy = iter->second.second;
+//      if (energy <= EnergySum)
+//        EnergySum = energy;
+      EnergySum += energy;
+      ++counter;
+    }
+    EnergySum *= 1./(double)counter;
     for (HomologyContainer::const_iterator iter = range.first; iter != range.second; ++iter) {
       // get distance out of Fragment
@@ -248 +258 @@
       DistanceVector.push_back( args );
       const double &energy = iter->second.second;
-      EnergyVector.push_back( FunctionModel::results_t(1, energy) );
+      EnergyVector.push_back( FunctionModel::results_t(1, energy-EnergySum) );
     }
   }
@@ -266 +276 @@
   const OutputVector_t& getTrainingOutputs() const {
     return EnergyVector;
+  }
+
+  /** Calculate the L2 error of a given \a model against the stored training data.
+   *
+   * \param model model whose L2 error to calculate
+   * \return sum of squared differences at training tuples
+   */
+  const double getL2Error(const FunctionModel &model) const
+  {
+    double L2sum = 0.;
+
+    FunctionApproximation::inputs_t::const_iterator initer = DistanceVector.begin();
+    FunctionApproximation::outputs_t::const_iterator outiter = EnergyVector.begin();
+    for (; initer != DistanceVector.end(); ++initer, ++outiter) {
+      const FunctionModel::results_t result = model((*initer));
+      const double temp = fabs((*outiter)[0] - result[0]);
+      L2sum += temp*temp;
+    }
+    return L2sum;
+  }
+
+  /** Calculate the Lmax error of a given \a model against the stored training data.
+   *
+   * \param model model whose Lmax error to calculate
+   * \return maximum difference over all training tuples
+   */
+  const double getLMaxError(const FunctionModel &model) const
+  {
+    double Lmax = 0.;
+    size_t maxindex = -1;
+    FunctionApproximation::inputs_t::const_iterator initer = DistanceVector.begin();
+    FunctionApproximation::outputs_t::const_iterator outiter = EnergyVector.begin();
+    for (; initer != DistanceVector.end(); ++initer, ++outiter) {
+      const FunctionModel::results_t result = model((*initer));
+      const double temp = fabs((*outiter)[0] - result[0]);
+      if (temp > Lmax) {
+        Lmax = temp;
+        maxindex = std::distance(
+            const_cast<const FunctionApproximation::inputs_t &>(DistanceVector).begin(),
+            initer
+            );
+      }
+    }
+    return Lmax;
   }
 
@@ -499 +553 @@
 
     // Afterwards we go through all of this type and gather the distance and the energy value
-    typedef std::pair<
-        FunctionApproximation::inputs_t,
-        FunctionApproximation::outputs_t> InputOutputVector_t;
-    InputOutputVector_t DistanceEnergyVector;
-    std::pair<HomologyContainer::const_iterator, HomologyContainer::const_iterator> range =
-        homologies.getHomologousGraphs(graph);
-    double EnergySum = 0.; //std::numeric_limits<double>::max();
-    size_t counter = 0.;
-    for (HomologyContainer::const_iterator iter = range.first; iter != range.second; ++iter) {
-      const double &energy = iter->second.second;
-//      if (energy <= EnergySum)
-//        EnergySum = energy;
-      EnergySum += energy;
-      ++counter;
-    }
-    EnergySum *= 1./(double)counter;
-    for (HomologyContainer::const_iterator iter = range.first; iter != range.second; ++iter) {
-      // get distance out of Fragment
-      const double &energy = iter->second.second;
-      const Fragment &fragment = iter->second.first;
-      const Fragment::charges_t charges = fragment.getCharges();
-      const Fragment::positions_t positions = fragment.getPositions();
-      FunctionModel::arguments_t args =
-          Extractors::_detail::gatherAllDistanceArguments(positions, DistanceEnergyVector.first.size());
-      DistanceEnergyVector.first.push_back( args );
-      DistanceEnergyVector.second.push_back( FunctionModel::results_t(1,energy-EnergySum) );
-    }
-    // print training data for debugging
-    {
-      LOG(1, "INFO: I gathered the following (" << DistanceEnergyVector.first.size()
-          << "," << DistanceEnergyVector.second.size() << ") data pairs: ");
-      FunctionApproximation::inputs_t::const_iterator initer = DistanceEnergyVector.first.begin();
-      FunctionApproximation::outputs_t::const_iterator outiter = DistanceEnergyVector.second.begin();
-      for (; initer != DistanceEnergyVector.first.end(); ++initer, ++outiter) {
-        std::stringstream stream;
-        for (size_t index = 0; index < (*initer).size(); ++index)
-           stream << "(" << (*initer)[index].indices.first << "," << (*initer)[index].indices.second
-              << ") " << (*initer)[index].distance;
-        stream << " with energy " << *outiter;
-        LOG(1, "INFO: " << stream.str());
-      }
-    }
+    TrainingData TersoffData(
+        TrainingData::extractor_t(&Extractors::gatherAllDistances) // gather first carbon pair
+        );
+    TersoffData( homologies.getHomologousGraphs(graph) );
+    LOG(1, "INFO: I gathered the following training data: " << TersoffData);
     // NOTICE that distance are in bohrradi as they come from MPQC!
 
     // now perform the function approximation by optimizing the model function
     boost::function< std::vector<FunctionModel::arguments_t>(const argument_t &, const double)> triplefunction =
-        boost::bind(&getTripleFromArgument, DistanceEnergyVector.first, _1, _2);
+        boost::bind(&getTripleFromArgument, boost::cref(TersoffData.getTrainingInputs()), _1, _2);
     srand((unsigned)time(0)); // seed with current time
     LOG(0, "INFO: Initial parameters are " << params << ".");
@@ -553 +570 @@
     FunctionModel &model = saturation;
     FunctionApproximation approximator(
-        DistanceEnergyVector.first.begin()->size(), 
-        DistanceEnergyVector.second.begin()->size(), 
+        TersoffData.getTrainingInputs().begin()->size(),
+        TersoffData.getTrainingOutputs().begin()->size(),
         model);
-    approximator.setTrainingData(DistanceEnergyVector.first,DistanceEnergyVector.second);
+    approximator.setTrainingData(TersoffData.getTrainingInputs(),TersoffData.getTrainingOutputs());
     if (model.isBoxConstraint() && approximator.checkParameterDerivatives())
       approximator(FunctionApproximation::ParameterDerivative);
@@ -583 +600 @@
 
     // check L2 and Lmax error against training set
-    double L2sum = 0.;
-    double Lmax = 0.;
-    size_t maxindex = -1;
-    FunctionApproximation::inputs_t::const_iterator initer = DistanceEnergyVector.first.begin();
-    FunctionApproximation::outputs_t::const_iterator outiter = DistanceEnergyVector.second.begin();
-    for (; initer != DistanceEnergyVector.first.end(); ++initer, ++outiter) {
-      const FunctionModel::results_t result = model((*initer));
-      const double temp = fabs((*outiter)[0] - result[0]);
-      LOG(2, "DEBUG: L2 contribution = " << (*outiter)[0] << "-" << result[0] << "=" << temp);
-      if (temp > Lmax) {
-        Lmax = temp;
-        maxindex = std::distance(const_cast<const FunctionApproximation::inputs_t &>(DistanceEnergyVector.first).begin(), initer);
-      }
-      L2sum += temp*temp;
-    }
-    LOG(1, "INFO: L2sum = " << L2sum << ", LMax = " << Lmax << " from " << maxindex);
+    LOG(1, "INFO: L2sum = " << TersoffData.getL2Error(model)
+        << ", LMax = " << TersoffData.getLMaxError(model) << ".");
   }