DescentEngine Class Reference

Collaboration diagram for DescentEngine:

Public Member Functions
	DescentEngine (Communicate *comm, const xmlNodePtr cur)
	Constructor for engine. More...
bool	processXML (const xmlNodePtr cur)
	process xml node More...
void	prepareStorage (const int num_replicas, const int num_optimizables)
	Prepare for taking samples. More...
void	takeSample (const int replica_id, const std::vector< FullPrecValueType > &der_rat_samp, const std::vector< FullPrecValueType > &le_der_samp, const std::vector< FullPrecValueType > &ls_der_samp, ValueType vgs_samp, ValueType weight_samp)
	Function that Take Sample Data from the Host Code. More...
void	sample_finish ()
	Function that reduces all vector information from all processors to the root processor. More...
void	mpi_unbiased_ratio_of_means (int numSamples, std::vector< ValueType > &weights, std::vector< ValueType > &numerSamples, std::vector< ValueType > &denomSamples, ValueType &mean, ValueType &variance, ValueType &stdErr)
	Function for computing ratios of the form <f>/<g> as well as the associated variance and standard error. More...
const std::vector< ValueType > &	getAveragedDerivatives () const
	Returns the derivatives of the cost function we are minimizing. More...
void	updateParameters ()
	helper method for updating parameter values with descent More...
ValueType	setStepSize (int i)
	helper method for seting step sizes for different parameter types in descent optimization More...
void	storeDerivRecord ()
	stores derivatives so they can be used in accelerated descent algorithm on later iterations More...
void	setupUpdate (const optimize::VariableSet &my_vars)
	helper method for transferring information on parameter names and types to the engine More...
void	storeVectors (std::vector< ValueType > &current_params)
	Store a vector of parameter differences to be used by the BLM in a hybrid optimization. More...
void	computeFinalizationUncertainties (std::vector< ValueType > &weights, std::vector< ValueType > &numerSamples, std::vector< ValueType > &denomSamples)
	Compute uncertainties for energy/target function and variance over a history of samples from a set of iterations. More...
int	retrieveStoreFrequency () const
	Returns number of times a parameter difference vector will be stored in the optimization. More...
const std::vector< std::vector< ValueType > > &	retrieveHybridBLM_Input () const
	Returns the set of stored parameter difference vectors that will be given to the BLM. More...
const std::vector< ValueType > &	retrieveNewParams () const
	Returns the current set of parameter values. More...
int	getDescentNum () const
	Returns number of optimization steps that have been taken with descent. More...
ValueType	getOmega () const
	Returns current value of omega. More...
ValueType	getEnergy () const
	Returns value of average energy. More...
ValueType	getVariance () const
	Returns variance of the energy. More...
ValueType	getSD () const
	Returns standard deviation of energy. More...
bool	targetingExcited () const
	Returns whether an excited state is being targeted. More...
int	getFinalDescentNum () const
	Returns the descent iteration number when on a finalizing descent section. More...
void	resetStorageCount ()
	Resets the number of vectors stored to 0 for next hybrid method macro-iteration. More...
void	setDerivs (std::vector< ValueType > &test_derivs)
	Function for setting averaged derivatives, currently only used as part of a unit test of the engine's parameter update. More...
void	setParamVal (int index, ValueType value)
	Function for setting parameter value, used to keep descent parameter values up to date with changes that occur on BLM steps of hybrid method. More...

Private Types
using	FullPrecValueType = qmcplusplus::QMCTraits::FullPrecValueType
using	ValueType = qmcplusplus::QMCTraits::ValueType
using	RealType = qmcplusplus::QMCTraits::RealType
using	FullPrecRealType = qmcplusplus::QMCTraits::FullPrecRealType

Private Attributes
std::vector< FullPrecValueType >	avg_le_der_samp_
	Vector for local energy parameter derivatives. More...
std::vector< std::vector< FullPrecValueType > >	replica_le_der_samp_
	Vector for local energy parameter derivatives on one thread. More...
std::vector< FullPrecValueType >	avg_der_rat_samp_
	Vector for WF parameter derivatives. More...
std::vector< std::vector< FullPrecValueType > >	replica_der_rat_samp_
	Vector for WF parameter derivatives on one thread. More...
std::vector< FullPrecValueType >	avg_numer_der_samp_
	Vector for target function numerator parameter derivatives. More...
std::vector< std::vector< FullPrecValueType > >	replica_numer_der_samp_
	Vector for target function numerator parameter derivatives on one thread. More...
std::vector< FullPrecValueType >	avg_denom_der_samp_
	Vector for target function denominator parameter derivatives. More...
std::vector< std::vector< FullPrecValueType > >	replica_denom_der_samp_
	Vector for target function denominator parameter derivatives on one thread. More...
ValueType	w_sum_
	Total sum of weights. More...
ValueType	e_avg_
	Average energy on a descent step. More...
ValueType	e_var_
	Variance of the energy. More...
ValueType	e_sd_
	Standard deviation of the energy. More...
ValueType	e_err_
	Standard error of the energy. More...
ValueType	numer_avg_
	Average target function numerator on a descent step. More...
ValueType	numer_var_
	Variance of the target function numerator. More...
ValueType	numer_err_
	Standard error of the target function numerator. More...
ValueType	denom_avg_
	Average target function denominator on a descent step. More...
ValueType	denom_var_
	Variance of the target function denominator. More...
ValueType	denom_err_
	Standard error of the target function denominator. More...
ValueType	target_avg_
	Average target function value on a descent step. More...
ValueType	target_var_
	Variance of the target function. More...
ValueType	target_err_
	Standard error of the target function. More...
std::vector< ValueType >	vg_history_
	history of sampled |value/guiding|^2 ratios for one iteration More...
std::vector< std::vector< ValueType > >	replica_vg_history_
std::vector< ValueType >	final_vg_history_
	history of sampled |value/guiding|^2 ratios during the descent finalization phase More...
std::vector< std::vector< ValueType > >	replica_final_vg_history_
std::vector< ValueType >	w_history_
	history of sampled configuration weights for one iteration More...
std::vector< std::vector< ValueType > >	replica_w_history_
std::vector< ValueType >	final_w_history_
	history of sampled configuration weights during descent finalization phase More...
std::vector< std::vector< ValueType > >	replica_final_w_history_
std::vector< ValueType >	lev_history_
	a history of sampled local energies times the |value/guiding|^2 raitos for one iteration More...
std::vector< std::vector< ValueType > >	replica_lev_history_
std::vector< ValueType >	final_lev_history_
	history of sampled local energies times the |value/guiding|^2 raitos during the descent finalization phase More...
std::vector< std::vector< ValueType > >	replica_final_lev_history_
std::vector< ValueType >	final_le_avg_history_
	a vector to store the averages of the energy during the descent finalization phase More...
std::vector< ValueType >	final_var_avg_history_
	a vector to store the variances of the energy during the descent finalization phase More...
std::vector< ValueType >	tnv_history_
	a history of target function numerator times the |value/guiding|^2 ratios for one iteration More...
std::vector< std::vector< ValueType > >	replica_tnv_history_
std::vector< ValueType >	final_tnv_history_
	a history of target function numerator times the |value/guiding|^2 ratios during the descent finalization phase More...
std::vector< std::vector< ValueType > >	replica_final_tnv_history_
std::vector< ValueType >	tdv_history_
	a history of target function denominator times the |value/guiding|^2 ratios for one iteration More...
std::vector< std::vector< ValueType > >	replica_tdv_history_
std::vector< ValueType >	final_tdv_history_
	a history of target function denomerator times the |value/guiding|^2 ratios during the descent finalization phase More...
std::vector< std::vector< ValueType > >	replica_final_tdv_history_
std::vector< ValueType >	final_tar_avg_history_
	a vector to store the averages of the target function during the descent finalization phase More...
std::vector< ValueType >	final_tar_var_history_
	a vector to store the variances of the target function during the descent finalization phase More...
std::vector< ValueType >	lderivs_
	Vector that stores the final averaged derivatives of the cost function. More...
Communicate *	my_comm_
	Communicator handles MPI reduction. More...
bool	engine_target_excited_
	Whether to target excited state. More...
int	num_params_
	Number of optimizable parameters. More...
std::vector< ValueType >	params_copy_
	Vector for storing parameter values from previous optimization step. More...
std::vector< ValueType >	current_params_
	Vector for storing parameter values for current optimization step. More...
std::vector< std::vector< ValueType > >	deriv_records_
	Vector for storing Lagrangian derivatives from previous optimization steps. More...
std::vector< ValueType >	denom_records_
	Vector for storing step size denominator values from previous optimization step. More...
std::vector< ValueType >	numer_records_
	Vector for storing step size numerator values from previous optimization step. More...
ValueType	lambda_ = 0.0
	Parameter for accelerated descent recursion relation. More...
std::vector< ValueType >	taus_
	Vector for storing step sizes from previous optimization step. More...
std::vector< ValueType >	derivs_squared_
	Vector for storing running average of squares of the derivatives. More...
int	descent_num_
	Integer for keeping track of only number of descent steps taken. More...
std::string	flavor_
	What variety of gradient descent will be used. More...
ValueType	tjf_2body_eta_
	Step sizes for different types of parameters. More...
ValueType	tjf_1body_eta_
ValueType	f_eta_
ValueType	gauss_eta_
ValueType	ci_eta_
ValueType	orb_eta_
bool	ramp_eta_
	Whether to gradually ramp up step sizes in descent. More...
int	ramp_num_
	Number of steps over which to ramp up step size. More...
int	store_num_
	Number of parameter difference vectors stored when descent is used in a hybrid optimization. More...
int	store_count_
	Counter of how many vectors have been stored so far. More...
std::vector< std::string >	engine_param_names_
	Vectors of parameter names and types, used in the assignment of step sizes. More...
std::vector< int >	engine_param_types_
std::vector< ValueType >	params_for_diff_
	Vector for storing parameter values for calculating differences to be given to hybrid method. More...
std::vector< std::vector< ValueType > >	hybrid_blm_input_
	Vector for storing the input vectors to the BLM steps of hybrid method. More...
ValueType	omega_
	Value of omega in excited state functional. More...
int	collection_step_
	Iteration to start collecting samples for final average and error blocking analysis. More...
int	compute_step_
	Iteration to start computing averages and errors from the stored values during the finalization phase. More...
bool	collect_count_ = false
	Whether to start collecting samples for the histories in the finalization phase. More...
int	final_descent_num_ = 0
	Counter for the number of descent steps taken in the finalization phase. More...
std::string	print_deriv_
	Whether to print out derivative terms for each parameter. More...

Detailed Description

Definition at line 27 of file DescentEngine.h.

Member Typedef Documentation

FullPrecRealType

using FullPrecRealType = qmcplusplus::QMCTraits::FullPrecRealType

private

Definition at line 32 of file DescentEngine.h.

FullPrecValueType

using FullPrecValueType = qmcplusplus::QMCTraits::FullPrecValueType

private

Definition at line 29 of file DescentEngine.h.

RealType

using RealType = qmcplusplus::QMCTraits::RealType

private

Definition at line 31 of file DescentEngine.h.

ValueType

using ValueType = qmcplusplus::QMCTraits::ValueType

private

Definition at line 30 of file DescentEngine.h.

Constructor & Destructor Documentation

DescentEngine()

DescentEngine	(	Communicate *	comm,
		const xmlNodePtr	cur
	)

Constructor for engine.

Definition at line 28 of file DescentEngine.cpp.

References DescentEngine::descent_num_, DescentEngine::processXML(), and DescentEngine::store_count_.

    : my_comm_(comm),
      engine_target_excited_(false),
      num_params_(0),
      flavor_("RMSprop"),
      tjf_2body_eta_(.01),
      tjf_1body_eta_(.01),
      f_eta_(.001),
      gauss_eta_(.001),
      ci_eta_(.01),
      orb_eta_(.001),
      ramp_eta_(false),
      ramp_num_(30),
      store_num_(5),
      omega_(0.0),
      collection_step_(-1),
      compute_step_(-1),
      print_deriv_("no")
{
  descent_num_ = 0;
  store_count_ = 0;
  processXML(cur);
}

Member Function Documentation

computeFinalizationUncertainties()

void computeFinalizationUncertainties	(	std::vector< ValueType > &	weights,
		std::vector< ValueType > &	numerSamples,
		std::vector< ValueType > &	denomSamples
	)

Compute uncertainties for energy/target function and variance over a history of samples from a set of iterations.

Definition at line 965 of file DescentEngine.cpp.

References qmcplusplus::abs(), qmcplusplus::app_log(), copy(), DescentEngine::mpi_unbiased_ratio_of_means(), qmcplusplus::n, and qmcplusplus::sqrt().

Referenced by DescentEngine::updateParameters().

{
  // Make copies of input vectors to do recursive blocking for error estimates
  std::vector<ValueType> wtv(weights);
  std::vector<ValueType> nmv(numerSamples);
  std::vector<ValueType> dnv(denomSamples);

  // Reproduce LM engine's crude way of estimating uncertainty in the variance
  int n           = nmv.size();
  int blocks      = 100;
  int section_len = n / 100;

  std::vector<ValueType> bbvars(blocks);
  ValueType bbv  = 0.0;
  ValueType bbv2 = 0.0;

  for (int i = 0; i < blocks; i++)
  {
    std::vector<ValueType> sub_wtv(section_len);
    std::vector<ValueType> sub_nmv(section_len);
    std::vector<ValueType> sub_dnv(section_len);

    ValueType temp_e, temp_v;
    std::copy(wtv.begin() + i * section_len, wtv.begin() + (i + 1) * section_len, sub_wtv.begin());
    std::copy(nmv.begin() + i * section_len, nmv.begin() + (i + 1) * section_len, sub_nmv.begin());
    std::copy(dnv.begin() + i * section_len, dnv.begin() + (i + 1) * section_len, sub_dnv.begin());

    ValueType blockVar;
    ValueType blockMean;
    ValueType blockErr;

    this->mpi_unbiased_ratio_of_means(section_len, sub_wtv, sub_nmv, sub_dnv, blockMean, blockVar, blockErr);

    bbvars[i] = blockVar;

    bbv += bbvars.at(i) / static_cast<ValueType>(blocks);
    bbv2 += bbvars.at(i) * bbvars.at(i) / static_cast<ValueType>(blocks);
    sub_wtv.clear();
    sub_nmv.clear();
    sub_dnv.clear();
  }

  const ValueType bbvov     = bbv2 - bbv * bbv;
  ValueType var_uncertainty = std::sqrt(std::abs(bbvov) / blocks);
  // Depending on when this function is called, this will be the uncertainty in
  // the variance
  // of either the energy or the target function.
  // Which one should be clear from the preceding print statements in the
  // output file.
  app_log() << "Uncertainty in variance of averaged quantity: " << var_uncertainty << std::endl;

  // To compute the standard error on the energy/target function itself,
  // blocking is performed on the whole set of samples in the history of many
  // iterations.
  while (n > 128)
  {
    ValueType currentVar;
    ValueType currentMean;
    ValueType currentErr;
    this->mpi_unbiased_ratio_of_means(n, wtv, nmv, dnv, currentMean, currentVar, currentErr);

    app_log() << "Blocking analysis error for per process length " << n << " is: " << currentErr << std::endl;

    std::vector<ValueType> tmp1, tmp2, tmp3;

    for (int i = 0; i < n; i += 2)
    {
      ValueType avgW     = (wtv[i] + wtv[i + 1]) / static_cast<ValueType>(2.0);
      ValueType avgNumer = (nmv[i] + nmv[i + 1]) / static_cast<ValueType>(2.0);
      ValueType avgDenom = (dnv[i] + dnv[i + 1]) / static_cast<ValueType>(2.0);

      tmp1.push_back(avgW);
      tmp2.push_back(avgNumer);
      tmp3.push_back(avgDenom);
    }
    wtv = tmp1;
    nmv = tmp2;
    dnv = tmp3;
    n   = nmv.size();
  }
}

getAveragedDerivatives()

const std::vector<ValueType>& getAveragedDerivatives ( ) const

inline

Returns the derivatives of the cost function we are minimizing.

Definition at line 293 of file DescentEngine.h.

References DescentEngine::lderivs_.

{ return lderivs_; }

getDescentNum()

int getDescentNum ( ) const

inline

Returns number of optimization steps that have been taken with descent.

Definition at line 332 of file DescentEngine.h.

References DescentEngine::descent_num_.

{ return descent_num_; }

getEnergy()

ValueType getEnergy ( ) const

inline

Returns value of average energy.

Definition at line 338 of file DescentEngine.h.

References DescentEngine::e_avg_.

{ return e_avg_; }

getFinalDescentNum()

int getFinalDescentNum ( ) const

inline

Returns the descent iteration number when on a finalizing descent section.

Definition at line 350 of file DescentEngine.h.

References DescentEngine::final_descent_num_.

{ return final_descent_num_; }

getOmega()

ValueType getOmega ( ) const

inline

Returns current value of omega.

Definition at line 335 of file DescentEngine.h.

References DescentEngine::omega_.

{ return omega_; }

getSD()

ValueType getSD ( ) const

inline

Returns standard deviation of energy.

Definition at line 344 of file DescentEngine.h.

References DescentEngine::e_sd_.

{ return e_sd_; }

getVariance()

ValueType getVariance ( ) const

inline

Returns variance of the energy.

Definition at line 341 of file DescentEngine.h.

References DescentEngine::e_var_.

{ return e_var_; }

mpi_unbiased_ratio_of_means()

void mpi_unbiased_ratio_of_means	(	int	numSamples,
		std::vector< ValueType > &	weights,
		std::vector< ValueType > &	numerSamples,
		std::vector< ValueType > &	denomSamples,
		ValueType &	mean,
		ValueType &	variance,
		ValueType &	stdErr
	)

Function for computing ratios of the form <f>/<g> as well as the associated variance and standard error.

Definition at line 480 of file DescentEngine.cpp.

References Communicate::allreduce(), mean(), qmcplusplus::Units::mass::mp, DescentEngine::my_comm_, qmcplusplus::n, qmcplusplus::Units::time::ns, qmcplusplus::sqrt(), and DescentEngine::w_sum_.

Referenced by DescentEngine::computeFinalizationUncertainties(), and DescentEngine::sample_finish().

{
  std::vector<ValueType> y(7);
  y[0] = 0.0;                   // normalization constant
  y[1] = 0.0;                   // mean of numerator
  y[2] = 0.0;                   // mean of denominator
  y[3] = 0.0;                   // mean of the square of the numerator terms
  y[4] = 0.0;                   // mean of the square of the denominator terms
  y[5] = 0.0;                   // mean of the product of numerator times denominator
  y[6] = ValueType(numSamples); // number of samples

  for (int i = 0; i < numSamples; i++)
  {
    ValueType n      = numerSamples[i];
    ValueType d      = denomSamples[i];
    ValueType weight = weights[i];

    y[0] += weight;
    y[1] += weight * n;
    y[2] += d;
    y[3] += weight * n * n;
    y[4] += weight * d * d;
    y[5] += weight * n * d;
  }

  my_comm_->allreduce(y);

  ValueType mf = y[1] / y[0]; // mean of numerator
  ValueType mg = y[2] / y[0]; // mean of denominator
  ValueType sf = y[3] / y[0]; // mean of the square of the numerator terms
  ValueType sg = y[4] / y[0]; // mean of the square of the denominator terms
  ValueType mp = y[5] / y[0]; // mean of the product of numerator times denominator
  ValueType ns = y[6];        // number of samples

  ValueType vf = (sf - mf * mf) * ns / (ns - static_cast<ValueType>(1.0));
  ValueType vg = (sg - mg * mg) * ns / (ns - static_cast<ValueType>(1.0));
  ValueType cv = (mp - mf * mg) * ns / (ns - static_cast<ValueType>(1.0));

  w_sum_   = y[0];
  mean     = (mf / mg) / (static_cast<ValueType>(1.0) + (vg / mg / mg - cv / mf / mg) / ns);
  variance = (mf * mf / mg / mg) * (vf / mf / mf + vg / mg / mg - static_cast<ValueType>(2.0) * cv / mf / mg);
  stdErr   = std::sqrt(variance / ns);
}

prepareStorage()

void prepareStorage	(	const int	num_replicas,
		const int	num_optimizables
	)

Prepare for taking samples.

Definition at line 102 of file DescentEngine.cpp.

References DescentEngine::avg_denom_der_samp_, DescentEngine::avg_der_rat_samp_, DescentEngine::avg_le_der_samp_, DescentEngine::avg_numer_der_samp_, DescentEngine::collect_count_, DescentEngine::collection_step_, DescentEngine::denom_avg_, DescentEngine::denom_err_, DescentEngine::denom_var_, DescentEngine::e_avg_, DescentEngine::e_err_, DescentEngine::e_sd_, DescentEngine::e_var_, DescentEngine::engine_target_excited_, DescentEngine::final_descent_num_, DescentEngine::lderivs_, DescentEngine::num_params_, DescentEngine::numer_avg_, DescentEngine::numer_err_, DescentEngine::numer_var_, DescentEngine::replica_denom_der_samp_, DescentEngine::replica_der_rat_samp_, DescentEngine::replica_final_lev_history_, DescentEngine::replica_final_tdv_history_, DescentEngine::replica_final_tnv_history_, DescentEngine::replica_final_vg_history_, DescentEngine::replica_final_w_history_, DescentEngine::replica_le_der_samp_, DescentEngine::replica_lev_history_, DescentEngine::replica_numer_der_samp_, DescentEngine::replica_tdv_history_, DescentEngine::replica_tnv_history_, DescentEngine::replica_vg_history_, DescentEngine::replica_w_history_, DescentEngine::target_avg_, DescentEngine::target_err_, DescentEngine::target_var_, and DescentEngine::w_sum_.

Referenced by DescentEngineHandle::prepareSampling().

{
  lderivs_.resize(num_optimizables);

  //Resize the history vectors for the current iteration for the number of threads present
  replica_vg_history_.resize(num_replicas);
  replica_w_history_.resize(num_replicas);
  replica_lev_history_.resize(num_replicas);

  if(engine_target_excited_)
  {
    replica_tnv_history_.resize(num_replicas);
    replica_tdv_history_.resize(num_replicas);
  
  }
  
 
  //Also resize the history vectors for descent finalization if necessary
  if (final_descent_num_ > collection_step_ && collect_count_)
  {
  replica_final_vg_history_.resize(num_replicas);
  replica_final_w_history_.resize(num_replicas);
  replica_final_lev_history_.resize(num_replicas);
  
  

    if(engine_target_excited_)
    {
        replica_final_tnv_history_.resize(num_replicas);
        replica_final_tdv_history_.resize(num_replicas);
    
    }
  }


  // Ground state case
  if (!engine_target_excited_)
  {
    avg_le_der_samp_.resize(num_optimizables);
    avg_der_rat_samp_.resize(num_optimizables);

    num_params_ = num_optimizables;

    std::fill(avg_le_der_samp_.begin(), avg_le_der_samp_.end(), 0.0);
    std::fill(avg_der_rat_samp_.begin(), avg_der_rat_samp_.end(), 0.0);

    replica_le_der_samp_.resize(num_replicas);
    replica_der_rat_samp_.resize(num_replicas);

    for (int i = 0; i < num_replicas; i++)
    {
      replica_le_der_samp_[i].resize(num_optimizables);
      std::fill(replica_le_der_samp_[i].begin(), replica_le_der_samp_[i].end(), 0.0);

      replica_der_rat_samp_[i].resize(num_optimizables);
      std::fill(replica_der_rat_samp_[i].begin(), replica_der_rat_samp_[i].end(), 0.0);
    }
  }
  // Excited state case
  else
  {
    replica_numer_der_samp_.resize(num_replicas);
    replica_denom_der_samp_.resize(num_replicas);

    avg_numer_der_samp_.resize(num_optimizables);
    avg_denom_der_samp_.resize(num_optimizables);

    std::fill(avg_numer_der_samp_.begin(), avg_numer_der_samp_.end(), 0.0);
    std::fill(avg_denom_der_samp_.begin(), avg_denom_der_samp_.end(), 0.0);

    for (int i = 0; i < num_replicas; i++)
    {
      replica_numer_der_samp_[i].resize(num_optimizables);
      replica_denom_der_samp_[i].resize(num_optimizables);

      std::fill(replica_numer_der_samp_[i].begin(), replica_numer_der_samp_[i].end(), 0.0);
      std::fill(replica_denom_der_samp_[i].begin(), replica_denom_der_samp_[i].end(), 0.0);
    }
  }

  w_sum_ = 0;
  e_avg_ = 0;
  e_var_ = 0;
  e_sd_  = 0;
  e_err_ = 0;

  numer_avg_ = 0;
  numer_var_ = 0;
  numer_err_ = 0;

  denom_avg_ = 0;
  denom_var_ = 0;
  denom_err_ = 0;

  target_avg_ = 0;
  target_var_ = 0;
  target_err_ = 0;
}

processXML()

bool processXML

(

const xmlNodePtr

cur

)

process xml node

Definition at line 52 of file DescentEngine.cpp.

References ParameterSet::add(), qmcplusplus::app_log(), DescentEngine::ci_eta_, DescentEngine::collect_count_, DescentEngine::collection_step_, DescentEngine::compute_step_, DescentEngine::engine_target_excited_, DescentEngine::f_eta_, DescentEngine::flavor_, DescentEngine::gauss_eta_, DescentEngine::omega_, DescentEngine::orb_eta_, DescentEngine::print_deriv_, ParameterSet::put(), DescentEngine::ramp_eta_, DescentEngine::ramp_num_, DescentEngine::store_num_, DescentEngine::tjf_1body_eta_, and DescentEngine::tjf_2body_eta_.

Referenced by DescentEngine::DescentEngine(), and qmcplusplus::TEST_CASE().

{
  std::string excited("no");
  std::string ramp_eta_str("no");

  ParameterSet m_param;
  m_param.add(excited, "targetExcited");
  m_param.add(omega_, "omega");
  // Type of descent method being used
  m_param.add(flavor_, "flavor");
  // Step size inputs for different parameter types
  m_param.add(tjf_2body_eta_, "TJF_2Body_eta");
  m_param.add(tjf_1body_eta_, "TJF_1Body_eta");
  m_param.add(f_eta_, "F_eta");
  m_param.add(ci_eta_, "CI_eta");
  m_param.add(gauss_eta_, "Gauss_eta");
  m_param.add(orb_eta_, "Orb_eta");
  // Whether to gradually ramp up step sizes and over how many steps
  m_param.add(ramp_eta_str, "Ramp_eta");
  m_param.add(ramp_num_, "Ramp_num");
  // If using descent as part of hybrid method, how many parameter difference
  // vectors to store
  m_param.add(store_num_, "Stored_Vectors");
  m_param.add(print_deriv_, "print_derivs");

  // When to start storing samples for a final average and when to start
  // computing it
  m_param.add(collection_step_, "collection_step");
  m_param.add(compute_step_, "compute_step");

  //app_log() << "Omega from input file: " << omega_ << std::endl;
  //app_log() << "Current collection step: " << collection_step_ << std::endl;
  // Use -1 as a default value when you don't collect history. Would want to
  // collect only during the descent finalization section
  if (collection_step_ != -1)
  {
    app_log() << "On descent finalization, have collect_count as true" << std::endl;
    collect_count_ = true;
  }

  m_param.put(cur);

  engine_target_excited_ = (excited == "yes");

  ramp_eta_ = (ramp_eta_str == "yes");

  return true;
}

resetStorageCount()

void resetStorageCount ( )

inline

Resets the number of vectors stored to 0 for next hybrid method macro-iteration.

Definition at line 354 of file DescentEngine.h.

References DescentEngine::store_count_.

{ store_count_ = 0; }

retrieveHybridBLM_Input()

const std::vector<std::vector<ValueType> >& retrieveHybridBLM_Input ( ) const

inline

Returns the set of stored parameter difference vectors that will be given to the BLM.

Definition at line 326 of file DescentEngine.h.

References DescentEngine::hybrid_blm_input_.

{ return hybrid_blm_input_; }

retrieveNewParams()

const std::vector<ValueType>& retrieveNewParams ( ) const

inline

Returns the current set of parameter values.

Definition at line 329 of file DescentEngine.h.

References DescentEngine::current_params_.

{ return current_params_; }

retrieveStoreFrequency()

int retrieveStoreFrequency ( ) const

inline

Returns number of times a parameter difference vector will be stored in the optimization.

Definition at line 322 of file DescentEngine.h.

References DescentEngine::store_num_.

{ return store_num_; }

sample_finish()

void sample_finish

(

)

Function that reduces all vector information from all processors to the root processor.

Definition at line 291 of file DescentEngine.cpp.

References Communicate::allreduce(), qmcplusplus::app_log(), DescentEngine::avg_denom_der_samp_, DescentEngine::avg_der_rat_samp_, DescentEngine::avg_le_der_samp_, DescentEngine::avg_numer_der_samp_, DescentEngine::collect_count_, DescentEngine::collection_step_, DescentEngine::denom_avg_, DescentEngine::denom_err_, DescentEngine::denom_var_, DescentEngine::e_avg_, DescentEngine::e_err_, DescentEngine::e_sd_, DescentEngine::e_var_, DescentEngine::engine_target_excited_, DescentEngine::final_descent_num_, DescentEngine::final_le_avg_history_, DescentEngine::final_lev_history_, DescentEngine::final_tar_avg_history_, DescentEngine::final_tar_var_history_, DescentEngine::final_tdv_history_, DescentEngine::final_tnv_history_, DescentEngine::final_var_avg_history_, DescentEngine::final_vg_history_, DescentEngine::final_w_history_, DescentEngine::lderivs_, DescentEngine::lev_history_, DescentEngine::mpi_unbiased_ratio_of_means(), DescentEngine::my_comm_, DescentEngine::numer_avg_, DescentEngine::numer_err_, DescentEngine::numer_var_, DescentEngine::print_deriv_, DescentEngine::replica_denom_der_samp_, DescentEngine::replica_der_rat_samp_, DescentEngine::replica_final_lev_history_, DescentEngine::replica_final_tdv_history_, DescentEngine::replica_final_tnv_history_, DescentEngine::replica_final_vg_history_, DescentEngine::replica_final_w_history_, DescentEngine::replica_le_der_samp_, DescentEngine::replica_lev_history_, DescentEngine::replica_numer_der_samp_, DescentEngine::replica_tdv_history_, DescentEngine::replica_tnv_history_, DescentEngine::replica_vg_history_, DescentEngine::replica_w_history_, qmcplusplus::sqrt(), DescentEngine::target_avg_, DescentEngine::target_err_, DescentEngine::target_var_, DescentEngine::tdv_history_, DescentEngine::tnv_history_, DescentEngine::vg_history_, DescentEngine::w_history_, and DescentEngine::w_sum_.

Referenced by DescentEngineHandle::finishSampling().

{

  //After a potentially multithreaded section, cocatenate the replica history vectors
  int num_threads = replica_vg_history_.size();
  for(int i =0; i< num_threads; i++)
  {
    vg_history_.insert(vg_history_.end(), replica_vg_history_[i].begin(),replica_vg_history_[i].end());
    w_history_.insert(w_history_.end(), replica_w_history_[i].begin(),replica_w_history_[i].end());
    lev_history_.insert(lev_history_.end(),replica_lev_history_[i].begin(),replica_lev_history_[i].end());
  
    //Clear the individual thread history vectors for the next iteration after their values have been collected
    replica_vg_history_[i].clear();
    replica_w_history_[i].clear();
    replica_lev_history_[i].clear();


    if(engine_target_excited_)
    {
        
        tnv_history_.insert(tnv_history_.end(), replica_tnv_history_[i].begin(),replica_tnv_history_[i].end());
        tdv_history_.insert(tdv_history_.end(), replica_tdv_history_[i].begin(),replica_tdv_history_[i].end());
   
       replica_tnv_history_[i].clear();
      replica_tdv_history_[i].clear(); 
    }
  }

  //Do the same for finalization histories if necessary
  if (final_descent_num_ > collection_step_ && collect_count_)
  {
     for(int i =0; i< num_threads; i++)
     {
        final_vg_history_.insert(final_vg_history_.end(), replica_final_vg_history_[i].begin(),replica_final_vg_history_[i].end());
        final_w_history_.insert(final_w_history_.end(), replica_final_w_history_[i].begin(),replica_final_w_history_[i].end());
        final_lev_history_.insert(final_lev_history_.end(),replica_final_lev_history_[i].begin(),replica_final_lev_history_[i].end());

     replica_final_vg_history_[i].clear();
     replica_final_w_history_[i].clear();
     replica_final_lev_history_[i].clear();


        if(engine_target_excited_)
        {
            
            final_tnv_history_.insert(final_tnv_history_.end(), replica_final_tnv_history_[i].begin(),replica_final_tnv_history_[i].end());
            final_tdv_history_.insert(final_tdv_history_.end(), replica_final_tdv_history_[i].begin(),replica_final_tdv_history_[i].end());
 
            replica_final_tnv_history_[i].clear();
            replica_final_tdv_history_[i].clear();       
        }
     }  
  }


  int numSamples = lev_history_.size();

  //Compute average energy and variance for this iteration
  this->mpi_unbiased_ratio_of_means(numSamples, w_history_, lev_history_, vg_history_, e_avg_, e_var_, e_err_);

  app_log() << "Energy Average: " << std::setprecision(9) << e_avg_ << std::endl;
  app_log() << "Energy Variance: " << std::setprecision(9) << e_var_ << std::endl;
  app_log() << "Weight Total: " << std::setprecision(9) << w_sum_ << std::endl;
  e_sd_ = std::sqrt(e_var_);
  app_log() << "Energy Standard Deviation: " << std::setprecision(9) << e_sd_ << std::endl;
  app_log() << "Energy Standard Error: " << std::setprecision(9) << e_err_ << std::endl;

  //Store average values during descent finalization
  if (final_descent_num_ > collection_step_ && collect_count_)
  {
    final_le_avg_history_.push_back(e_avg_);
    final_var_avg_history_.push_back(e_var_);
  }

  if (!engine_target_excited_)
  {
    for (int i = 0; i < replica_le_der_samp_.size(); i++)
    {
      for (int j = 0; j < lderivs_.size(); j++)
      {
        avg_le_der_samp_[j] += replica_le_der_samp_[i].at(j);
        avg_der_rat_samp_[j] += replica_der_rat_samp_[i].at(j);
      }
    }

    my_comm_->allreduce(avg_le_der_samp_);
    my_comm_->allreduce(avg_der_rat_samp_);
  }
  else
  {
    this->mpi_unbiased_ratio_of_means(numSamples, w_history_, tnv_history_, vg_history_, numer_avg_, numer_var_,
                                      numer_err_);

    this->mpi_unbiased_ratio_of_means(numSamples, w_history_, tdv_history_, vg_history_, denom_avg_, denom_var_,
                                      denom_err_);

    this->mpi_unbiased_ratio_of_means(numSamples, w_history_, tnv_history_, tdv_history_, target_avg_, target_var_,
                                      target_err_);

    app_log() << "Target Function Average: " << std::setprecision(9) << target_avg_ << std::endl;
    app_log() << "Target Function Variance: " << std::setprecision(9) << target_var_ << std::endl;
    app_log() << "Target Function Error: " << std::setprecision(9) << target_err_ << std::endl;

    if (final_descent_num_ > collection_step_ && collect_count_)
    {
      final_tar_avg_history_.push_back(target_avg_);
      final_tar_var_history_.push_back(target_var_);
    }
    for (int i = 0; i < replica_numer_der_samp_.size(); i++)
    {
      for (int j = 0; j < lderivs_.size(); j++)
      {
        avg_numer_der_samp_[j] += replica_numer_der_samp_[i].at(j);
        avg_denom_der_samp_[j] += replica_denom_der_samp_[i].at(j);
      }
    }

    my_comm_->allreduce(avg_numer_der_samp_);
    my_comm_->allreduce(avg_denom_der_samp_);
  }

  int num_optimizables = lderivs_.size();

  // Vectors for parts of excited state functional derivatives
  std::vector<ValueType> numer_term1(num_optimizables, 0.0);
  std::vector<ValueType> numer_term2(num_optimizables, 0.0);
  std::vector<ValueType> denom(num_optimizables, 0.0);

  ValueType gradNorm = 0.0;

  // Compute contribution to derivatives
  for (int i = 0; i < num_optimizables; i++)
  {
    // Ground state case
    if (!engine_target_excited_)
    {
      avg_le_der_samp_.at(i)  = avg_le_der_samp_.at(i) / static_cast<FullPrecValueType>(w_sum_);
      avg_der_rat_samp_.at(i) = avg_der_rat_samp_.at(i) / static_cast<FullPrecValueType>(w_sum_);

      lderivs_.at(i) = static_cast<FullPrecValueType>(2.0) *
          (avg_le_der_samp_.at(i) - static_cast<FullPrecValueType>(e_avg_) * avg_der_rat_samp_.at(i));
      if (print_deriv_ == "yes")
      {
        app_log() << "Parameter # " << i << " Hamiltonian term: " << avg_le_der_samp_.at(i) << std::endl;
        app_log() << "Parameter # " << i << " Overlap term: " << avg_der_rat_samp_.at(i) << std::endl;
        app_log() << "Derivative for param # " << i << " : " << lderivs_.at(i) << std::endl;
      }
    }
    // Excited state case
    else
    {
      avg_numer_der_samp_.at(i) = avg_numer_der_samp_.at(i) / static_cast<FullPrecValueType>(w_sum_);
      avg_denom_der_samp_.at(i) = avg_denom_der_samp_.at(i) / static_cast<FullPrecValueType>(w_sum_);

      if (print_deriv_ == "yes")
      {
        app_log() << "Parameter # " << i << " Numer Deriv: " << avg_numer_der_samp_.at(i) << std::endl;
        app_log() << "Parameter # " << i << " Denom Deriv: " << avg_denom_der_samp_.at(i) << std::endl;
      }

      numer_term1.at(i) = avg_numer_der_samp_.at(i) * static_cast<FullPrecValueType>(denom_avg_);

      numer_term2.at(i) = avg_denom_der_samp_.at(i) * static_cast<FullPrecValueType>(numer_avg_);

      denom.at(i) = denom_avg_ * denom_avg_;

      lderivs_.at(i) = (numer_term1.at(i) - numer_term2.at(i)) / denom.at(i);
    }

    gradNorm += lderivs_.at(i) * lderivs_.at(i);
  }

  gradNorm = std::sqrt(gradNorm);
  app_log() << "Norm of gradient vector is: " << gradNorm << std::endl;

  // Clear the history vectors for next iteration once sample_finish is done
  lev_history_.clear();
  vg_history_.clear();
  w_history_.clear();

  if (engine_target_excited_)
  {
    tnv_history_.clear();
    tdv_history_.clear();
  }
}

setDerivs()

void setDerivs ( std::vector< ValueType > &  test_derivs )

inline

Function for setting averaged derivatives, currently only used as part of a unit test of the engine's parameter update.

Definition at line 358 of file DescentEngine.h.

References DescentEngine::lderivs_.

{ lderivs_ = test_derivs; }

setParamVal()

void setParamVal ( int  index, ValueType  value  )

inline

Function for setting parameter value, used to keep descent parameter values up to date with changes that occur on BLM steps of hybrid method.

Definition at line 362 of file DescentEngine.h.

References DescentEngine::current_params_.

{ current_params_[index] = value; }

setStepSize()

DescentEngine::ValueType setStepSize

(

int

i

)

helper method for seting step sizes for different parameter types in descent optimization

Definition at line 843 of file DescentEngine.cpp.

References DescentEngine::ci_eta_, DescentEngine::descent_num_, DescentEngine::engine_param_names_, DescentEngine::engine_param_types_, DescentEngine::f_eta_, DescentEngine::gauss_eta_, DescentEngine::orb_eta_, DescentEngine::ramp_eta_, DescentEngine::ramp_num_, DescentEngine::tjf_1body_eta_, and DescentEngine::tjf_2body_eta_.

Referenced by DescentEngine::updateParameters().

{
  ValueType type_eta;

  std::string name = engine_param_names_[i];

  int type = engine_param_types_[i];

  // Step sizes are assigned according to parameter type identified from the
  // variable name.
  // Other parameter types could be added to this section as other wave function
  // ansatzes are developed.
  if ((name.find("uu") != std::string::npos) || (name.find("ud") != std::string::npos))
  {
    type_eta = tjf_2body_eta_;
  }
  // If parameter name doesn't have "uu" or "ud" in it and is of type 1, assume
  // it is a 1 body Jastrow parameter.
  else if (type == 1)
  {
    type_eta = tjf_1body_eta_;
  }
  else if (name.find("F_") != std::string::npos)
  {
    type_eta = f_eta_;
  }
  else if (name.find("CIcoeff_") != std::string::npos || name.find("CSFcoeff_") != std::string::npos)
  {
    type_eta = ci_eta_;
  }
  else if (name.find("orb_rot_") != std::string::npos)
  {
    type_eta = orb_eta_;
  }
  else if (name.find("g") != std::string::npos)
  {
    // Gaussian parameters are rarely optimized in practice but the descent code
    // allows for it.
    type_eta = gauss_eta_;
  }
  else
  {
    // If there is some other parameter type that isn't in one of the categories
    // with a default/input, use a conservative default step size.
    type_eta = .001;
  }

  if (ramp_eta_ && descent_num_ < ramp_num_)
  {
    type_eta = type_eta * static_cast<ValueType>((descent_num_ + 1) / (double)ramp_num_);
  }

  return type_eta;
}

setupUpdate()

void setupUpdate

(

const optimize::VariableSet &

my_vars

)

helper method for transferring information on parameter names and types to the engine

Definition at line 900 of file DescentEngine.cpp.

References qmcplusplus::app_log(), DescentEngine::current_params_, DescentEngine::engine_param_names_, DescentEngine::engine_param_types_, VariableSet::getType(), VariableSet::name(), DescentEngine::num_params_, DescentEngine::params_copy_, DescentEngine::params_for_diff_, VariableSet::size(), and VariableSet::where().

{
  // omega_ = omega_input;

  num_params_ = my_vars.size();
  app_log() << "This is num_params_: " << num_params_ << std::endl;
  for (int i = 0; i < num_params_; i++)
  {
    // app_log() << "Variable #" << i << ": " << my_vars[i] << " with index val:
    // " << my_vars.where(i) << std::endl;
    if (my_vars.where(i) != -1)
    {
      engine_param_names_.push_back(my_vars.name(i));
      engine_param_types_.push_back(my_vars.getType(i));
      params_copy_.push_back(my_vars[i]);
      current_params_.push_back(my_vars[i]);
      params_for_diff_.push_back(my_vars[i]);
    }
  }
}

storeDerivRecord()

void storeDerivRecord ( )

inline

stores derivatives so they can be used in accelerated descent algorithm on later iterations

Definition at line 304 of file DescentEngine.h.

References DescentEngine::deriv_records_, and DescentEngine::lderivs_.

{ deriv_records_.push_back(lderivs_); }

storeVectors()

void storeVectors

(

std::vector< ValueType > &

current_params

)

Store a vector of parameter differences to be used by the BLM in a hybrid optimization.

Definition at line 923 of file DescentEngine.cpp.

References qmcplusplus::app_log(), DescentEngine::hybrid_blm_input_, DescentEngine::params_for_diff_, and DescentEngine::store_count_.

{
  std::vector<ValueType> row_vec(current_params.size(), 0.0);

  // Take difference between current parameter values and the values from some
  // number of
  // iterations before to be stored as input to BLM.
  // The current parameter values are then copied to params_for_diff_ to be used
  // if storeVectors is called again later in the optimization.
  for (int i = 0; i < current_params.size(); i++)
  {
    row_vec[i]          = current_params[i] - params_for_diff_[i];
    params_for_diff_[i] = current_params[i];
  }

  // If on first store of descent section, clear anything that was in vector
  if (store_count_ == 0)
  {
    hybrid_blm_input_.clear();
    hybrid_blm_input_.push_back(row_vec);
  }
  else
  {
    hybrid_blm_input_.push_back(row_vec);
  }

#if !defined(QMC_COMPLEX)
  for (int i = 0; i < hybrid_blm_input_.size(); i++)
  {
    std::string entry = "";
    for (int j = 0; j < hybrid_blm_input_.at(i).size(); j++)
    {
      entry = entry + std::to_string(hybrid_blm_input_.at(i).at(j)) + ",";
    }
    app_log() << "Stored Vector: " << entry << std::endl;
  }
#endif
  store_count_++;
}

takeSample()

void takeSample	(	const int	replica_id,
		const std::vector< FullPrecValueType > &	der_rat_samp,
		const std::vector< FullPrecValueType > &	le_der_samp,
		const std::vector< FullPrecValueType > &	ls_der_samp,
		ValueType	vgs_samp,
		ValueType	weight_samp
	)

Function that Take Sample Data from the Host Code.

Parameters

[in]	der_rat_samp	<n|Psi_i>/<n|Psi> (i = 0 (|Psi>), 1, ... N_var )
[in]	le_der_samp	<n|H|Psi_i>/<n|Psi> (i = 0 (|Psi>), 1, ... N_var )
[in]	ls_der_samp	<|S^2|Psi_i>/<n|Psi> (i = 0 (|Psi>), 1, ... N_var )
[in]	vgs_samp	|<n|value_fn>/<n|guiding_fn>|^2
[in]	weight_samp	weight for this sample

Definition at line 213 of file DescentEngine.cpp.

References DescentEngine::collect_count_, DescentEngine::collection_step_, DescentEngine::engine_target_excited_, DescentEngine::final_descent_num_, qmcplusplus::n, DescentEngine::omega_, DescentEngine::replica_denom_der_samp_, DescentEngine::replica_der_rat_samp_, DescentEngine::replica_final_lev_history_, DescentEngine::replica_final_tdv_history_, DescentEngine::replica_final_tnv_history_, DescentEngine::replica_final_vg_history_, DescentEngine::replica_final_w_history_, DescentEngine::replica_le_der_samp_, DescentEngine::replica_lev_history_, DescentEngine::replica_numer_der_samp_, DescentEngine::replica_tdv_history_, DescentEngine::replica_tnv_history_, DescentEngine::replica_vg_history_, and DescentEngine::replica_w_history_.

Referenced by DescentEngineHandle::takeSample().

{

    const size_t num_optimizables = der_rat_samp.size() - 1;

  ValueType etmp = static_cast<ValueType>(le_der_samp.at(0));


  // Store a history of samples for the current iteration
  replica_lev_history_[replica_id].push_back(etmp * vgs_samp);
  replica_vg_history_[replica_id].push_back(vgs_samp);
  replica_w_history_[replica_id].push_back(static_cast<ValueType>(1.0));

  // If on descent finalizing section and past the collection step, store each
  // local energy for this iteration
  if (final_descent_num_ > collection_step_ && collect_count_)
  {
    replica_final_lev_history_[replica_id].push_back(etmp * vgs_samp);
    replica_final_vg_history_[replica_id].push_back(vgs_samp);
    replica_final_w_history_[replica_id].push_back(1.0);
  }

  // Ground State Case
  if (!engine_target_excited_)
  {
    for (int i = 0; i < num_optimizables; i++)
    {
      replica_le_der_samp_[replica_id].at(i) += le_der_samp.at(i + 1) * static_cast<FullPrecValueType>(vgs_samp);
      replica_der_rat_samp_[replica_id].at(i) += der_rat_samp.at(i + 1) * static_cast<FullPrecValueType>(vgs_samp);
    }
  }
  // Excited State Case
  else
  {
    ValueType n;
    ValueType d;

    n = (omega_ - etmp) * vgs_samp;
    d = (omega_ * omega_ - static_cast<ValueType>(2) * omega_ * etmp + etmp * etmp) * vgs_samp;

    replica_tnv_history_[replica_id].push_back(n);
    replica_tdv_history_[replica_id].push_back(d);

    if (final_descent_num_ > collection_step_ && collect_count_)
    {
      replica_final_tnv_history_[replica_id].push_back(n);
      replica_final_tdv_history_[replica_id].push_back(d);
    }


    for (int i = 0; i < num_optimizables; i++)
    {
      // Combination of derivative ratios for target function numerator <psi |
      // omega -H | psi>/<psi | psi>
      replica_numer_der_samp_[replica_id].at(i) += static_cast<FullPrecValueType>(2) *
          (static_cast<FullPrecValueType>(omega_) * der_rat_samp.at(i + 1) - le_der_samp.at(i + 1)) *
          static_cast<FullPrecValueType>(vgs_samp);

      // Combination of derivative ratios for target function denominator <psi |
      // (omega -H)^2 | psi>/<psi | psi>
      replica_denom_der_samp_[replica_id].at(i) += static_cast<FullPrecValueType>(2) *
          ((static_cast<FullPrecValueType>(omega_) * der_rat_samp.at(i + 1) - le_der_samp.at(i + 1)) *
           (static_cast<FullPrecValueType>(omega_) * der_rat_samp.at(0) - le_der_samp.at(0))) *
          static_cast<FullPrecValueType>(vgs_samp);
    }
  }
}

targetingExcited()

bool targetingExcited ( ) const

inline

Returns whether an excited state is being targeted.

Definition at line 347 of file DescentEngine.h.

References DescentEngine::engine_target_excited_.

{ return engine_target_excited_; }

updateParameters()

void updateParameters

(

)

helper method for updating parameter values with descent

Definition at line 531 of file DescentEngine.cpp.

References qmcplusplus::abs(), qmcplusplus::app_log(), DescentEngine::ci_eta_, DescentEngine::collect_count_, DescentEngine::collection_step_, DescentEngine::compute_step_, DescentEngine::computeFinalizationUncertainties(), DescentEngine::current_params_, DescentEngine::denom_records_, DescentEngine::deriv_records_, DescentEngine::derivs_squared_, DescentEngine::descent_num_, qmcplusplus::Units::charge::e, DescentEngine::engine_target_excited_, qmcplusplus::exp(), DescentEngine::f_eta_, DescentEngine::final_descent_num_, DescentEngine::final_le_avg_history_, DescentEngine::final_lev_history_, DescentEngine::final_tar_avg_history_, DescentEngine::final_tar_var_history_, DescentEngine::final_tdv_history_, DescentEngine::final_tnv_history_, DescentEngine::final_var_avg_history_, DescentEngine::final_vg_history_, DescentEngine::final_w_history_, DescentEngine::flavor_, qmcplusplus::isnan(), DescentEngine::lambda_, DescentEngine::num_params_, DescentEngine::numer_records_, DescentEngine::orb_eta_, DescentEngine::params_copy_, qmcplusplus::pow(), DescentEngine::setStepSize(), sign(), qmcplusplus::sqrt(), DescentEngine::taus_, DescentEngine::tjf_1body_eta_, and DescentEngine::tjf_2body_eta_.

{
  app_log() << "Number of Parameters: " << num_params_ << std::endl;
  app_log() << "Descent Number: " << descent_num_ << std::endl;

  app_log() << "Finalization Descent Num (should be zero if not on last section): " << final_descent_num_ << std::endl;
  if (final_descent_num_ > collection_step_ && collect_count_)
  {
    app_log() << "Should be storing history, length of history on one process is: " << final_lev_history_.size()
              << std::endl;
  }

  app_log() << "Parameter Type step sizes: "
            << " TJF_2Body_eta=" << tjf_2body_eta_ << " TJF_1Body_eta=" << tjf_1body_eta_ << " F_eta=" << f_eta_
            << " CI_eta=" << ci_eta_ << " Orb_eta=" << orb_eta_ << std::endl;

  // Get set of derivatives for current (kth) optimization step
  std::vector<ValueType> cur_deriv_set = deriv_records_.at(deriv_records_.size() - 1);
  std::vector<ValueType> prev_deriv_set;
  if (!taus_.empty())
  {
    // Get set of derivatives for previous (k-1th) optimization step
    prev_deriv_set = deriv_records_.at(deriv_records_.size() - 2);
  }

  ValueType denom;
  ValueType numer;
  ValueType v;
  ValueType cor_numer;
  ValueType cor_v;

  ValueType epsilon  = 1e-8;
  ValueType type_eta = 0;

  ValueType tau = 0;
  // Update parameters according to specified flavor of gradient descent method

  // RMSprop corresponds to the method used by Booth and co-workers
  if (flavor_.compare("RMSprop") == 0)
  {
    app_log() << "Using RMSprop" << std::endl;

    // To match up with Booth group paper notation, prevLambda is lambda_k-1,
    // curLambda is lambda_k, nextLambda is lambda_k+1
    ValueType cur_lambda = static_cast<ValueType>(.5) +
        static_cast<ValueType>(.5) *
            std::sqrt(static_cast<ValueType>(1.0) + static_cast<ValueType>(4.0) * lambda_ * lambda_);
    ValueType next_lambda = static_cast<ValueType>(.5) +
        static_cast<ValueType>(.5) *
            std::sqrt(static_cast<ValueType>(1.0) + static_cast<ValueType>(4.0) * cur_lambda * cur_lambda);
    ValueType gamma = (static_cast<ValueType>(1.0) - cur_lambda) / next_lambda;

    // Define damping factor that turns off acceleration of the algorithm
    // small value of d corresponds to quick damping and effectively using
    // steepest descent
    ValueType d            = 100;
    ValueType decay_factor = std::exp(-(static_cast<ValueType>(1.0) / d) * (static_cast<ValueType>(descent_num_)));
    gamma                  = gamma * decay_factor;

    ValueType rho = .9;

    for (int i = 0; i < num_params_; i++)
    {
      ValueType cur_square = std::pow(cur_deriv_set.at(i), static_cast<RealType>(2));

      // Need to calculate step size tau for each parameter inside loop
      // In RMSprop, the denominator of the step size depends on a a running
      // average of past squares of the parameter derivative
      if (derivs_squared_.size() < num_params_)
      {
        cur_square = std::pow(cur_deriv_set.at(i), static_cast<RealType>(2));
      }
      else if (derivs_squared_.size() >= num_params_)
      {
        cur_square = rho * derivs_squared_.at(i) +
            (static_cast<ValueType>(1.0) - rho) * std::pow(cur_deriv_set.at(i), static_cast<RealType>(2));
      }

      denom = std::sqrt(cur_square + epsilon);

      // The numerator of the step size is set according to parameter type based
      // on input choices
      type_eta = this->setStepSize(i);
      tau      = type_eta / denom;

      // Include an additional factor to cause step size to eventually decrease
      // to 0 as number of steps taken increases
      ValueType step_lambda = .1;

      ValueType step_decay_denom = static_cast<ValueType>(1.0) + step_lambda * static_cast<ValueType>(descent_num_);
      tau                        = tau / step_decay_denom;

      // Update parameter values
      // If case corresponds to being after the first descent step
      if (descent_num_ > 0)
      {
        ValueType old_tau = taus_.at(i);

        current_params_[i] = (static_cast<ValueType>(1.0) - gamma) * (current_params_[i] - tau * cur_deriv_set[i]) +
            gamma * (params_copy_[i] - old_tau * prev_deriv_set[i]);
      }
      else
      {
        tau = type_eta;

        current_params_[i] = current_params_[i] - tau * cur_deriv_set[i];
      }

      if (taus_.size() < num_params_)
      {
        // For the first optimization step, need to add to the vectors
        taus_.push_back(tau);
        derivs_squared_.push_back(cur_square);
      }
      else
      {
        // When not on the first step, can overwrite the previous stored values
        taus_[i]           = tau;
        derivs_squared_[i] = cur_square;
      }

      params_copy_[i] = current_params_[i];
    }

    // Store current (kth) lambda value for next optimization step
    lambda_ = cur_lambda;
  }
  // Random uses only the sign of the parameter derivatives and takes a step of
  // random size within a range.
  else if (flavor_.compare("Random") == 0)
  {
    app_log() << "Using Random" << std::endl;

    for (int i = 0; i < num_params_; i++)
    {
      denom           = 1;
      ValueType alpha = (static_cast<ValueType>(rand() / RAND_MAX));
      ValueType sign  = std::abs(cur_deriv_set[i]) / cur_deriv_set[i];
      if (qmcplusplus::isnan(std::real(sign)))
      {
        app_log() << "Got a nan, choosing sign randomly with 50-50 probability" << std::endl;

        ValueType t = (static_cast<ValueType>(rand() / RAND_MAX));
        if (std::real(t) > std::real(.5))
        {
          sign = 1;
        }
        else
        {
          sign = -1;
        }
      }
      app_log() << "This is random alpha: " << alpha << " with sign: " << sign << std::endl;

      current_params_.at(i) = current_params_.at(i) - tau * alpha * sign;
    }
  }

  else
  {
    // ADAM method
    if (flavor_.compare("ADAM") == 0)
    {
      app_log() << "Using ADAM" << std::endl;

      for (int i = 0; i < num_params_; i++)
      {
        ValueType cur_square = std::pow(cur_deriv_set.at(i), static_cast<RealType>(2));
        ValueType beta1      = .9;
        ValueType beta2      = .99;
        if (descent_num_ == 0)
        {
          numer_records_.push_back(0);
          denom_records_.push_back(0);
        }
        numer = beta1 * numer_records_[i] + (static_cast<ValueType>(1.0) - beta1) * cur_deriv_set[i];
        v     = beta2 * denom_records_[i] + (static_cast<ValueType>(1.0) - beta2) * cur_square;

        cor_numer = numer / (static_cast<ValueType>(1.0) - std::pow(beta1, static_cast<RealType>(descent_num_ + 1)));
        cor_v     = v / (static_cast<ValueType>(1.0) - std::pow(beta2, static_cast<RealType>(descent_num_ + 1)));

        denom = std::sqrt(cor_v) + epsilon;

        type_eta = this->setStepSize(i);
        tau      = type_eta / denom;

        current_params_.at(i) = current_params_.at(i) - tau * cor_numer;

        if (taus_.size() < num_params_)
        {
          // For the first optimization step, need to add to the vectors
          taus_.push_back(tau);
          derivs_squared_.push_back(cur_square);
          denom_records_[i] = v;
          numer_records_[i] = numer;
        }
        else
        {
          // When not on the first step, can overwrite the previous stored
          // values
          taus_[i]           = tau;
          derivs_squared_[i] = cur_square;
          denom_records_[i]  = v;
          numer_records_[i]  = numer;
        }

        params_copy_[i] = current_params_.at(i);
      }
    }
    // AMSGrad method, similar to ADAM except for form of the step size
    // denominator
    else if (flavor_.compare("AMSGrad") == 0)
    {
      app_log() << "Using AMSGrad" << std::endl;

      for (int i = 0; i < num_params_; i++)
      {
        ValueType cur_square = std::pow(cur_deriv_set.at(i), static_cast<RealType>(2));
        ValueType beta1      = .9;
        ValueType beta2      = .99;
        if (descent_num_ == 0)
        {
          numer_records_.push_back(0);
          denom_records_.push_back(0);
        }

        numer = beta1 * numer_records_[i] + (static_cast<ValueType>(1.0) - beta1) * cur_deriv_set[i];
        v     = beta2 * denom_records_[i] + (static_cast<ValueType>(1.0) - beta2) * cur_square;
        v     = std::max(std::real(denom_records_[i]), std::real(v));

        denom    = std::sqrt(v) + epsilon;
        type_eta = this->setStepSize(i);
        tau      = type_eta / denom;

        current_params_.at(i) = current_params_.at(i) - tau * numer;

        if (taus_.size() < num_params_)
        {
          // For the first optimization step, need to add to the vectors
          taus_.push_back(tau);
          derivs_squared_.push_back(cur_square);
          denom_records_[i] = v;
          numer_records_[i] = numer;
        }
        else
        {
          // When not on the first step, can overwrite the previous stored
          // values
          taus_[i]           = tau;
          derivs_squared_[i] = cur_square;
          denom_records_[i]  = v;
          numer_records_[i]  = numer;
        }

        params_copy_[i] = current_params_.at(i);
      }
    }
  }

  descent_num_++;
  if (collect_count_)
  {
    final_descent_num_++;

    // Start computing averages and uncertainties from stored history.
    // This should be done only near the end of the descent finalization section
    // as it is unnecessary earlier.
    if (final_descent_num_ >= compute_step_)
    {
      app_log() << "Computing average energy and its variance over stored "
                   "steps and its standard error"
                << std::endl;

      ValueType collected_steps = final_le_avg_history_.size();
      ValueType final_e_sum =
          std::accumulate(final_le_avg_history_.begin(), final_le_avg_history_.end(), static_cast<ValueType>(0.0));
      ValueType final_e_avg = final_e_sum / collected_steps;

      ValueType final_var_sum =
          std::accumulate(final_var_avg_history_.begin(), final_var_avg_history_.end(), static_cast<ValueType>(0.0));
      ValueType final_var_avg = final_var_sum / collected_steps;

      app_log() << "Final average energy: " << final_e_avg << std::endl;
      app_log() << "Final varaince: " << final_var_avg << std::endl;

      this->computeFinalizationUncertainties(final_w_history_, final_lev_history_, final_vg_history_);

      // Do the same for the target function during excited state optimizations.
      if (engine_target_excited_)
      {
        app_log() << "Computing average target function over stored steps and "
                     "its standard error"
                  << std::endl;

        ValueType final_tar_sum =
            std::accumulate(final_tar_avg_history_.begin(), final_tar_avg_history_.end(), static_cast<ValueType>(0.0));

        ValueType final_tar_avg = final_tar_sum / collected_steps;

        ValueType final_tar_var_sum =
            std::accumulate(final_tar_var_history_.begin(), final_tar_var_history_.end(), static_cast<ValueType>(0.0));
        ValueType final_tar_var_avg = final_tar_var_sum / collected_steps;

        app_log() << "Final average target function: " << final_tar_avg << std::endl;
        app_log() << "Final target function varaince: " << final_tar_var_avg << std::endl;
        this->computeFinalizationUncertainties(final_w_history_, final_tnv_history_, final_tdv_history_);
      }
    }
  }
}

Member Data Documentation

avg_denom_der_samp_

std::vector<FullPrecValueType> avg_denom_der_samp_

private

Vector for target function denominator parameter derivatives.

Definition at line 51 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

avg_der_rat_samp_

std::vector<FullPrecValueType> avg_der_rat_samp_

private

Vector for WF parameter derivatives.

Definition at line 41 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

avg_le_der_samp_

std::vector<FullPrecValueType> avg_le_der_samp_

private

Vector for local energy parameter derivatives.

Definition at line 36 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

avg_numer_der_samp_

std::vector<FullPrecValueType> avg_numer_der_samp_

private

Vector for target function numerator parameter derivatives.

Definition at line 46 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

ci_eta_

ValueType ci_eta_

private

Definition at line 196 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), DescentEngine::setStepSize(), and DescentEngine::updateParameters().

collect_count_

bool collect_count_ = false

private

Whether to start collecting samples for the histories in the finalization phase.

Definition at line 236 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::processXML(), DescentEngine::sample_finish(), DescentEngine::takeSample(), and DescentEngine::updateParameters().

collection_step_

int collection_step_

private

Iteration to start collecting samples for final average and error blocking analysis.

Definition at line 228 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::processXML(), DescentEngine::sample_finish(), DescentEngine::takeSample(), and DescentEngine::updateParameters().

compute_step_

int compute_step_

private

Iteration to start computing averages and errors from the stored values during the finalization phase.

Definition at line 232 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), and DescentEngine::updateParameters().

current_params_

std::vector<ValueType> current_params_

private

Vector for storing parameter values for current optimization step.

Definition at line 165 of file DescentEngine.h.

Referenced by DescentEngine::retrieveNewParams(), DescentEngine::setParamVal(), DescentEngine::setupUpdate(), and DescentEngine::updateParameters().

denom_avg_

ValueType denom_avg_

private

Average target function denominator on a descent step.

Definition at line 75 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

denom_err_

ValueType denom_err_

private

Standard error of the target function denominator.

Definition at line 79 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

denom_records_

std::vector<ValueType> denom_records_

private

Vector for storing step size denominator values from previous optimization step.

Definition at line 172 of file DescentEngine.h.

Referenced by DescentEngine::updateParameters().

denom_var_

ValueType denom_var_

private

Variance of the target function denominator.

Definition at line 77 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

deriv_records_

std::vector<std::vector<ValueType> > deriv_records_

private

Vector for storing Lagrangian derivatives from previous optimization steps.

Definition at line 168 of file DescentEngine.h.

Referenced by DescentEngine::storeDerivRecord(), and DescentEngine::updateParameters().

derivs_squared_

std::vector<ValueType> derivs_squared_

private

Vector for storing running average of squares of the derivatives.

Definition at line 183 of file DescentEngine.h.

Referenced by DescentEngine::updateParameters().

descent_num_

int descent_num_

private

Integer for keeping track of only number of descent steps taken.

Definition at line 186 of file DescentEngine.h.

Referenced by DescentEngine::DescentEngine(), DescentEngine::getDescentNum(), DescentEngine::setStepSize(), and DescentEngine::updateParameters().

e_avg_

ValueType e_avg_

private

Average energy on a descent step.

Definition at line 59 of file DescentEngine.h.

Referenced by DescentEngine::getEnergy(), DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

e_err_

ValueType e_err_

private

Standard error of the energy.

Definition at line 65 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

e_sd_

ValueType e_sd_

private

Standard deviation of the energy.

Definition at line 63 of file DescentEngine.h.

Referenced by DescentEngine::getSD(), DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

e_var_

ValueType e_var_

private

Variance of the energy.

Definition at line 61 of file DescentEngine.h.

Referenced by DescentEngine::getVariance(), DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

engine_param_names_

std::vector<std::string> engine_param_names_

private

Vectors of parameter names and types, used in the assignment of step sizes.

Definition at line 213 of file DescentEngine.h.

Referenced by DescentEngine::setStepSize(), and DescentEngine::setupUpdate().

engine_param_types_

std::vector<int> engine_param_types_

private

Definition at line 214 of file DescentEngine.h.

Referenced by DescentEngine::setStepSize(), and DescentEngine::setupUpdate().

engine_target_excited_

bool engine_target_excited_

private

Whether to target excited state.

Definition at line 156 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::processXML(), DescentEngine::sample_finish(), DescentEngine::takeSample(), DescentEngine::targetingExcited(), and DescentEngine::updateParameters().

f_eta_

ValueType f_eta_

private

Definition at line 194 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), DescentEngine::setStepSize(), and DescentEngine::updateParameters().

final_descent_num_

int final_descent_num_ = 0

private

Counter for the number of descent steps taken in the finalization phase.

Definition at line 239 of file DescentEngine.h.

Referenced by DescentEngine::getFinalDescentNum(), DescentEngine::prepareStorage(), DescentEngine::sample_finish(), DescentEngine::takeSample(), and DescentEngine::updateParameters().

final_le_avg_history_

std::vector<ValueType> final_le_avg_history_

private

a vector to store the averages of the energy during the descent finalization phase

Definition at line 117 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

final_lev_history_

std::vector<ValueType> final_lev_history_

private

history of sampled local energies times the |value/guiding|^2 raitos during the descent finalization phase

Definition at line 112 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

final_tar_avg_history_

std::vector<ValueType> final_tar_avg_history_

private

a vector to store the averages of the target function during the descent finalization phase

Definition at line 144 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

final_tar_var_history_

std::vector<ValueType> final_tar_var_history_

private

a vector to store the variances of the target function during the descent finalization phase

Definition at line 147 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

final_tdv_history_

std::vector<ValueType> final_tdv_history_

private

a history of target function denomerator times the |value/guiding|^2 ratios during the descent finalization phase

Definition at line 139 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

final_tnv_history_

std::vector<ValueType> final_tnv_history_

private

a history of target function numerator times the |value/guiding|^2 ratios during the descent finalization phase

Definition at line 130 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

final_var_avg_history_

std::vector<ValueType> final_var_avg_history_

private

a vector to store the variances of the energy during the descent finalization phase

Definition at line 121 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

final_vg_history_

std::vector<ValueType> final_vg_history_

private

history of sampled |value/guiding|^2 ratios during the descent finalization phase

Definition at line 94 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

final_w_history_

std::vector<ValueType> final_w_history_

private

history of sampled configuration weights during descent finalization phase

Definition at line 102 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish(), and DescentEngine::updateParameters().

flavor_

std::string flavor_

private

What variety of gradient descent will be used.

Definition at line 189 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), and DescentEngine::updateParameters().

gauss_eta_

ValueType gauss_eta_

private

Definition at line 195 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), and DescentEngine::setStepSize().

hybrid_blm_input_

std::vector<std::vector<ValueType> > hybrid_blm_input_

private

Vector for storing the input vectors to the BLM steps of hybrid method.

Definition at line 221 of file DescentEngine.h.

Referenced by DescentEngine::retrieveHybridBLM_Input(), and DescentEngine::storeVectors().

lambda_

ValueType lambda_ = 0.0

private

Parameter for accelerated descent recursion relation.

Definition at line 179 of file DescentEngine.h.

Referenced by DescentEngine::updateParameters().

lderivs_

std::vector<ValueType> lderivs_

private

Vector that stores the final averaged derivatives of the cost function.

Definition at line 150 of file DescentEngine.h.

Referenced by DescentEngine::getAveragedDerivatives(), DescentEngine::prepareStorage(), DescentEngine::sample_finish(), DescentEngine::setDerivs(), and DescentEngine::storeDerivRecord().

lev_history_

std::vector<ValueType> lev_history_

private

a history of sampled local energies times the |value/guiding|^2 raitos for one iteration

Definition at line 107 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish().

my_comm_

Communicate* my_comm_

private

Communicator handles MPI reduction.

Definition at line 153 of file DescentEngine.h.

Referenced by DescentEngine::mpi_unbiased_ratio_of_means(), and DescentEngine::sample_finish().

num_params_

int num_params_

private

Number of optimizable parameters.

Definition at line 159 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::setupUpdate(), and DescentEngine::updateParameters().

numer_avg_

ValueType numer_avg_

private

Average target function numerator on a descent step.

Definition at line 68 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

numer_err_

ValueType numer_err_

private

Standard error of the target function numerator.

Definition at line 72 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

numer_records_

std::vector<ValueType> numer_records_

private

Vector for storing step size numerator values from previous optimization step.

Definition at line 176 of file DescentEngine.h.

Referenced by DescentEngine::updateParameters().

numer_var_

ValueType numer_var_

private

Variance of the target function numerator.

Definition at line 70 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

omega_

ValueType omega_

private

Value of omega in excited state functional.

Definition at line 224 of file DescentEngine.h.

Referenced by DescentEngine::getOmega(), DescentEngine::processXML(), and DescentEngine::takeSample().

orb_eta_

ValueType orb_eta_

private

Definition at line 197 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), DescentEngine::setStepSize(), and DescentEngine::updateParameters().

params_copy_

std::vector<ValueType> params_copy_

private

Vector for storing parameter values from previous optimization step.

Definition at line 162 of file DescentEngine.h.

Referenced by DescentEngine::setupUpdate(), and DescentEngine::updateParameters().

params_for_diff_

std::vector<ValueType> params_for_diff_

private

Vector for storing parameter values for calculating differences to be given to hybrid method.

Definition at line 218 of file DescentEngine.h.

Referenced by DescentEngine::setupUpdate(), and DescentEngine::storeVectors().

print_deriv_

std::string print_deriv_

private

Whether to print out derivative terms for each parameter.

Definition at line 242 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), and DescentEngine::sample_finish().

ramp_eta_

bool ramp_eta_

private

Whether to gradually ramp up step sizes in descent.

Definition at line 200 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), and DescentEngine::setStepSize().

ramp_num_

int ramp_num_

private

Number of steps over which to ramp up step size.

Definition at line 203 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), and DescentEngine::setStepSize().

replica_denom_der_samp_

std::vector<std::vector<FullPrecValueType> > replica_denom_der_samp_

private

Vector for target function denominator parameter derivatives on one thread.

Definition at line 53 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_der_rat_samp_

std::vector<std::vector<FullPrecValueType> > replica_der_rat_samp_

private

Vector for WF parameter derivatives on one thread.

Definition at line 43 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_final_lev_history_

std::vector<std::vector<ValueType> > replica_final_lev_history_

private

Definition at line 113 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_final_tdv_history_

std::vector<std::vector<ValueType> > replica_final_tdv_history_

private

Definition at line 140 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_final_tnv_history_

std::vector<std::vector<ValueType> > replica_final_tnv_history_

private

Definition at line 131 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_final_vg_history_

std::vector<std::vector<ValueType> > replica_final_vg_history_

private

Definition at line 95 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_final_w_history_

std::vector<std::vector<ValueType> > replica_final_w_history_

private

Definition at line 103 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_le_der_samp_

std::vector<std::vector<FullPrecValueType> > replica_le_der_samp_

private

Vector for local energy parameter derivatives on one thread.

Definition at line 38 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_lev_history_

std::vector<std::vector<ValueType> > replica_lev_history_

private

Definition at line 108 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_numer_der_samp_

std::vector<std::vector<FullPrecValueType> > replica_numer_der_samp_

private

Vector for target function numerator parameter derivatives on one thread.

Definition at line 48 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_tdv_history_

std::vector<std::vector<ValueType> > replica_tdv_history_

private

Definition at line 136 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_tnv_history_

std::vector<std::vector<ValueType> > replica_tnv_history_

private

Definition at line 126 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_vg_history_

std::vector<std::vector<ValueType> > replica_vg_history_

private

Definition at line 90 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

replica_w_history_

std::vector<std::vector<ValueType> > replica_w_history_

private

Definition at line 100 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), DescentEngine::sample_finish(), and DescentEngine::takeSample().

store_count_

int store_count_

private

Counter of how many vectors have been stored so far.

Definition at line 210 of file DescentEngine.h.

Referenced by DescentEngine::DescentEngine(), DescentEngine::resetStorageCount(), and DescentEngine::storeVectors().

store_num_

int store_num_

private

Number of parameter difference vectors stored when descent is used in a hybrid optimization.

Definition at line 207 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), and DescentEngine::retrieveStoreFrequency().

target_avg_

ValueType target_avg_

private

Average target function value on a descent step.

Definition at line 82 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

target_err_

ValueType target_err_

private

Standard error of the target function.

Definition at line 86 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

target_var_

ValueType target_var_

private

Variance of the target function.

Definition at line 84 of file DescentEngine.h.

Referenced by DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

taus_

std::vector<ValueType> taus_

private

Vector for storing step sizes from previous optimization step.

Definition at line 181 of file DescentEngine.h.

Referenced by DescentEngine::updateParameters().

tdv_history_

std::vector<ValueType> tdv_history_

private

a history of target function denominator times the |value/guiding|^2 ratios for one iteration

Definition at line 135 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish().

tjf_1body_eta_

ValueType tjf_1body_eta_

private

Definition at line 193 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), DescentEngine::setStepSize(), and DescentEngine::updateParameters().

tjf_2body_eta_

ValueType tjf_2body_eta_

private

Step sizes for different types of parameters.

Definition at line 192 of file DescentEngine.h.

Referenced by DescentEngine::processXML(), DescentEngine::setStepSize(), and DescentEngine::updateParameters().

tnv_history_

std::vector<ValueType> tnv_history_

private

a history of target function numerator times the |value/guiding|^2 ratios for one iteration

Definition at line 125 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish().

vg_history_

std::vector<ValueType> vg_history_

private

history of sampled |value/guiding|^2 ratios for one iteration

Definition at line 89 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish().

w_history_

std::vector<ValueType> w_history_

private

history of sampled configuration weights for one iteration

Definition at line 99 of file DescentEngine.h.

Referenced by DescentEngine::sample_finish().

w_sum_

ValueType w_sum_

private

Total sum of weights.

Definition at line 56 of file DescentEngine.h.

Referenced by DescentEngine::mpi_unbiased_ratio_of_means(), DescentEngine::prepareStorage(), and DescentEngine::sample_finish().

---

The documentation for this class was generated from the following files:

• /home/pk7/projects/qmc/for_cron_doxygen/qmcpack/src/QMCDrivers/Optimizers/DescentEngine.h
• /home/pk7/projects/qmc/for_cron_doxygen/qmcpack/src/QMCDrivers/Optimizers/DescentEngine.cpp