Fair replacement policy
Fair replacement policy#
New in version 2.11.
#include <pagmo/r_policies/fair_replace.hpp>

class fair_replace#
This userdefined replacement policy (UDRP) will replace individuals in a group only if the candidate replacement individuals are better than the original individuals.
In this context, better means the following:
in singleobjective unconstrained problems, an individual is better than another one if its fitness is lower,
in singleobjective constrained problems, individuals are ranked via
sort_population_con()
,in multiobjective unconstrained problems, individuals are ranked via
sort_population_mo()
.
See the documentation of
replace()
for more details on the replacement algorithm implemented by this UDRP.Note that this userdefined replacement policy currently does not support multiobjective constrained problems.

fair_replace()#
Default constructor.
The default constructor initialises a policy with an absolute migration rate of 1 (that is, 1 individual in the original population is considered for replacement).

template<typename T>
explicit fair_replace(T x)# Constructor from a migration rate.
This constructor participates in overload resolution only if
T
is a C++ integral or a floatingpoint type. The input migration rate, x, is used to indicate how many individuals will be replaced in an input population by thereplace()
member function.If x is a floating point value in the \(\left[0,1\right]\) range, then it represents a fractional migration rate. That is, it indicates, the fraction of individuals that may be replaced in the input population: a value of 0 means that no individuals will be replaced, a value of 1 means that all individuals may be replaced.
If x is an integral value, then it represents an absolute migration rate, that is, the exact number of individuals that may be replaced in the input population.
 Parameters
x – the fractional or absolute migration rate.
 Throws
std::invalid_argument – if the supplied fractional migration rate is not finite or not in the \(\left[0,1\right]\) range.
unspecified – any exception raised by
boost::numeric_cast()
while trying to convert the input absolute migration rate topop_size_t
.

individuals_group_t replace(const individuals_group_t &inds, const vector_double::size_type&, const vector_double::size_type&, const vector_double::size_type &nobj, const vector_double::size_type &nec, const vector_double::size_type &nic, const vector_double &tol, const individuals_group_t &mig) const#
This member function will replace individuals in inds with individuals from mig.
The replacement algorithm determines first how many individuals in inds can be replaced. This depends both on the migration rate specified upon construction, and on the size \(S\) of inds.
After having established the number \(N\) of individuals that can be replaced in inds, the algorithm then selects the top \(N\) individuals from mig, merges them with inds into a new population, and returns the top \(S\) individuals from the new population. The ranking of individuals in mig and in the new population depends on the problem’s properties:
in singleobjective unconstrained problems, the individuals are ranked according to their (scalar) fitnesses,
in singleobjective constrained problems, the ranking of individuals is done via
sort_population_con()
,in multiobjective unconstrained problems, the ranking of individuals is done via
sort_population_mo()
.
Note that this userdefined replacement policy currently does not support multiobjective constrained problems.
 Parameters
inds – the input individuals.
nobj – the number of objectives of the problem the individuals in inds and mig refer to.
nec – the number of equality constraints of the problem the individuals in inds and mig refer to.
nic – the number of inequality constraints of the problem the individuals in inds and mig refer to.
tol – the vector of constraint tolerances of the problem the individuals in inds and mig refer to.
mig – the individuals that may replace individuals in inds.
 Returns
the new population resulting from replacing individuals in inds with individuals from mig.
 Throws
std::invalid_argument –
in the following cases:
the problem the individuals in inds and mig refer to is multiobjective and constrained,
an absolute migration rate larger than the number of input individuals was specified.
unspecified – any exception raised by one of the invoked ranking functions or by memory allocation errors in standard containers.

std::string get_name() const#
Get the name of the policy.
 Returns
"Fair replace"
.

std::string get_extra_info() const#
 Returns
Humanreadable extra info about this replacement policy.