df_history_t Struct Reference

#include <gromacs/mdtypes/df_history.h>

Description

Free-energy sampling history struct.

Note that an intended invariant of this structure is that all vectors (and the dimensions of the matrices) have the same size, ie. nlambda.

Todo:

Split out into microstate and observables history.

Public Member Functions
	df_history_t (int numLambdaValues)
	df_history_t (const df_history_t &src)=default
	Default copy constructor.
df_history_t &	operator= (const df_history_t &v)=default
	Default copy assignment operator.
	df_history_t (df_history_t &&src) noexcept=default
	Default move constructor.
df_history_t &	operator= (df_history_t &&v) noexcept=default
	Default move assignment operator.
template<gmx::CheckpointDataOperation operation>
void	doCheckpoint (gmx::CheckpointData< operation > checkpointData, LambdaWeightCalculation elamstats)
	Allows to read and write checkpoint within modular simulator. More...

Public Attributes
int	nlambda = 0
	total number of lambda states - useful to history as the number of lambdas determines the size of arrays.
bool	bEquil = false
	Have we reached equilibration yet, where the weights stop updating?
std::vector< int >	numSamplesAtLambdaForStatistics
	The number of points observed at each lambda up to the current time, in this simulation, for calculating statistics.
std::vector< int >	numSamplesAtLambdaForEquilibration
	The number of points observed at each lambda up to the current time, over a set of simulations, for determining equilibration.
std::vector< real >	wl_histo
	The histogram for WL flatness determination. Can be preserved between simulations with input options.
real	wl_delta
	The current wang-landau delta, used to increment each state when visited.
std::vector< real >	sum_weights
	Sum of weights of each state over all states.
std::vector< real >	sum_dg
	Sum of the free energies of the states – not actually used for weighting, but informational.
std::vector< real >	sum_minvar
	corrections to weights for minimum variance
std::vector< real >	sum_variance
	variances of the states
TwoDLambdaArray	accum_p
	accumulated bennett weights for n+1
TwoDLambdaArray	accum_m
	accumulated bennett weights for n-1
TwoDLambdaArray	accum_p2
	accumulated squared bennett weights for n+1
TwoDLambdaArray	accum_m2
	accumulated squared bennett weights for n-1
TwoDLambdaArray	Tij
	Transition matrix, estimated from probabilities of transitions.
TwoDLambdaArray	Tij_empirical
	Empirical transition matrix, estimated from only counts of transitions.

Member Function Documentation

template<gmx::CheckpointDataOperation operation>

template void df_history_t::doCheckpoint	(	gmx::CheckpointData< operation >	checkpointData,
		LambdaWeightCalculation	elamstats
	)

Allows to read and write checkpoint within modular simulator.

Template Parameters

operation

Whether we're reading or writing

Parameters

checkpointData	The CheckpointData object
elamstats	How the lambda weights are calculated

---

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

• src/gromacs/mdtypes/df_history.h
• src/gromacs/mdtypes/df_history.cpp