Documentation |
On this page… |
---|
Optimization App vs. Command Line |
There are two ways to specify options for the genetic algorithm, depending on whether you are using the Optimization app or calling the functions ga or at the command line:
If you are using the Optimization app (optimtool), select an option from a drop-down list or enter the value of the option in a text field.
If you are calling ga or gamultiobj from the command line, create an options structure using the function gaoptimset, as follows:
options = gaoptimset('Param1', value1, 'Param2', value2, ...);
See Setting Options at the Command Line for examples.
In this section, each option is listed in two ways:
By its label, as it appears in the Optimization app
By its field name in the options structure
For example:
Population type is the label of the option in the Optimization app.
PopulationType is the corresponding field of the options structure.
Plot options let you plot data from the genetic algorithm while it is running. You can stop the algorithm at any time by clicking the Stop button on the plot window.
Plot interval (PlotInterval) specifies the number of generations between consecutive calls to the plot function.
You can select any of the following plot functions in the Plot functions pane:
Best fitness (@gaplotbestf) plots the best function value versus generation.
Expectation (@gaplotexpectation) plots the expected number of children versus the raw scores at each generation.
Score diversity (@gaplotscorediversity) plots a histogram of the scores at each generation.
Stopping (@gaplotstopping) plots stopping criteria levels.
Best individual (@gaplotbestindiv) plots the vector entries of the individual with the best fitness function value in each generation.
Genealogy (@gaplotgenealogy) plots the genealogy of individuals. Lines from one generation to the next are color-coded as follows:
Red lines indicate mutation children.
Blue lines indicate crossover children.
Black lines indicate elite individuals.
Scores (@gaplotscores) plots the scores of the individuals at each generation.
Max constraint (@gaplotmaxconstr) plots the maximum nonlinear constraint violation at each generation.
Distance (@gaplotdistance) plots the average distance between individuals at each generation.
Range (@gaplotrange) plots the minimum, maximum, and mean fitness function values in each generation.
Selection (@gaplotselection) plots a histogram of the parents.
Custom function lets you use plot functions of your own. To specify the plot function if you are using the Optimization app,
Select Custom function.
Enter @myfun in the text box, where myfun is the name of your function.
gamultiobj allows Distance, Genealogy, Score diversity, Selection, Stopping, and Custom function, as well as the following additional choices:
Pareto front (@gaplotpareto) plots the Pareto front for the first two objective functions.
Average Pareto distance (@gaplotparetodistance) plots a bar chart of the distance of each individual from its neighbors.
Rank histogram (@gaplotrankhist) plots a histogram of the ranks of the individuals. Individuals of rank 1 are on the Pareto frontier. Individuals of rank 2 are lower than at least one rank 1 individual, but are not lower than any individuals from other ranks, etc.
Average Pareto spread (@gaplotspread) plots the average spread as a function of iteration number.
To display a plot when calling ga from the command line, set the PlotFcns field of options to be a function handle to the plot function. For example, to display the best fitness plot, set options as follows:
options = gaoptimset('PlotFcns', @gaplotbestf);
To display multiple plots, use the syntax
options = gaoptimset('PlotFcns', {@plotfun1, @plotfun2, ...});
where @plotfun1, @plotfun2, and so on are function handles to the plot functions.
If you specify multiple plot functions, all plots appear as subplots in the same window. Right-click any subplot to obtain a larger version in a separate figure window.
The first line of a plot function has this form:
function state = plotfun(options,state,flag)
The input arguments to the function are
options — Structure containing all the current options settings.
state — Structure containing information about the current generation. The State Structure describes the fields of state.
flag — String that tells what stage the algorithm is currently in.
Passing Extra Parameters in the Optimization Toolbox™ documentation explains how to provide additional parameters to the function.
The output argument state is a state structure as well. Pass the input argument, modified if you like. To stop the iterations, set state.StopFlag to a nonempty string.
ga. The state structure for ga, which is an input argument to plot, mutation, and output functions, contains the following fields:
Population — Population in the current generation
Score — Scores of the current population
Generation — Current generation number
StartTime — Time when genetic algorithm started
StopFlag — String containing the reason for stopping
Selection — Indices of individuals selected for elite, crossover, and mutation
Expectation — Expectation for selection of individuals
Best — Vector containing the best score in each generation
LastImprovement — Generation at which the last improvement in fitness value occurred
LastImprovementTime — Time at which last improvement occurred
NonlinIneq — Nonlinear inequality constraints, displayed only when a nonlinear constraint function is specified
NonlinEq — Nonlinear equality constraints, displayed only when a nonlinear constraint function is specified
gamultiobj. The state structure for gamultiobj, which is an input argument to plot, mutation, and output functions, contains the following fields:
Population — Population in the current generation
Score — Scores of the current population, a Population-by-nObjectives matrix, where nObjectives is the number of objectives
Generation — Current generation number
StartTime — Time when genetic algorithm started
StopFlag — String containing the reason for stopping
Selection — Indices of individuals selected for elite, crossover, and mutation
Rank — Vector of the ranks of members in the population
Distance — Vector of distances of each member of the population to the nearest neighboring member
AverageDistance — The average of Distance
Spread — Vector where the entries are the spread in each generation
Population options let you specify the parameters of the population that the genetic algorithm uses.
Population type (PopulationType) specifies the type of input to the fitness function. Types and their restrictions are:
Double vector ('doubleVector') — Use this option if the individuals in the population have type double. Use this option for mixed integer programming. This is the default.
Bit string ('bitstring') — Use this option if the individuals in the population have components that are 0 or 1.
For Creation function (CreationFcn) and Mutation function (MutationFcn), use Uniform (@gacreationuniform and @mutationuniform) or Custom. For Crossover function (CrossoverFcn), use Scattered (@crossoverscattered), Single point (@crossoversinglepoint), Two point (@crossovertwopoint), or Custom. You cannot use a Hybrid function, and ga ignores all constraints, including bounds, linear constraints, and nonlinear constraints.
Custom — For Crossover function and Mutation function, use Custom. For Creation function, either use Custom, or provide an Initial population. You cannot use a Hybrid function, and ga ignores all constraints, including bounds, linear constraints, and nonlinear constraints.
Population size (PopulationSize) specifies how many individuals there are in each generation. With a large population size, the genetic algorithm searches the solution space more thoroughly, thereby reducing the chance that the algorithm returns a local minimum that is not a global minimum. However, a large population size also causes the algorithm to run more slowly.
If you set Population size to a vector, the genetic algorithm creates multiple subpopulations, the number of which is the length of the vector. The size of each subpopulation is the corresponding entry of the vector. See Migration Options.
Creation function (CreationFcn) specifies the function that creates the initial population for ga. Do not specify a creation function with integer problems because ga overrides any choice you make. Choose from:
[] uses the default creation function for your problem.
Uniform (@gacreationuniform) creates a random initial population with a uniform distribution. This is the default when there are no linear constraints, or when there are integer constraints. The uniform distribution is in the initial population range (PopInitRange). The default values for PopInitRange are [-10;10] for every component, or [-9999;10001] when there are integer constraints. These bounds are shifted and scaled to match any existing bounds lb and ub.
Feasible population (@gacreationlinearfeasible), the default when there are linear constraints and no integer constraints, creates a random initial population that satisfies all bounds and linear constraints. If there are linear constraints, Feasible population creates many individuals on the boundaries of the constraint region, and creates a well-dispersed population. Feasible population ignores Initial range (PopInitRange).
gacreationlinearfeasible calls linprog to create a feasible population with respect to bounds and linear constraints.
For an example showing its behavior, see Linearly Constrained Population and Custom Plot Function.
Nonlinear Feasible population (@gacreationnonlinearfeasible) is the default creation function for the 'penalty' nonlinear constraint algorithm. For details, see Constraint Parameters.
Custom lets you write your own creation function, which must generate data of the type that you specify in Population type. To specify the creation function if you are using the Optimization app,
Set Creation function to Custom.
Set Function name to @myfun, where myfun is the name of your function.
If you are using ga, set
options = gaoptimset('CreationFcn', @myfun);
Your creation function must have the following calling syntax.
function Population = myfun(GenomeLength, FitnessFcn, options)
The input arguments to the function are:
Genomelength — Number of independent variables for the fitness function
FitnessFcn — Fitness function
options — Options structure
The function returns Population, the initial population for the genetic algorithm.
Passing Extra Parameters in the Optimization Toolbox documentation explains how to provide additional parameters to the function.
Initial population (InitialPopulation) specifies an initial population for the genetic algorithm. The default value is [], in which case ga uses the default Creation function to create an initial population. If you enter a nonempty array in the Initial population field, the array must have no more than Population size rows, and exactly Number of variables columns. In this case, the genetic algorithm calls a Creation function to generate the remaining individuals, if required.
Initial scores (InitialScores) specifies initial scores for the initial population. The initial scores can also be partial. Do not specify initial scores with integer problems because ga overrides any choice you make.
Initial range (PopInitRange) specifies the range of the vectors in the initial population that is generated by the gacreationuniform creation function. You can set Initial range to be a matrix with two rows and Number of variables columns, each column of which has the form [lb;ub], where lb is the lower bound and ub is the upper bound for the entries in that coordinate. If you specify Initial range to be a 2-by-1 vector, each entry is expanded to a constant row of length Number of variables. If you do not specify an Initial range, the default is [-10;10] ([-1e4+1;1e4+1] for integer-constrained problems), modified to match any existing bounds.
See Setting the Initial Range for an example.
Fitness scaling converts the raw fitness scores that are returned by the fitness function to values in a range that is suitable for the selection function. You can specify options for fitness scaling in the Fitness scaling pane.
Scaling function (FitnessScalingFcn) specifies the function that performs the scaling. The options are
Rank (@fitscalingrank) — The default fitness scaling function, Rank, scales the raw scores based on the rank of each individual instead of its score. The rank of an individual is its position in the sorted scores. An individual with rank r has scaled score proportional to $$1/\sqrt{r}$$. So the scaled score of the most fit individual is proportional to 1, the scaled score of the next most fit is proportional to $$1/\sqrt{2}$$, and so on. Rank fitness scaling removes the effect of the spread of the raw scores. The square root makes poorly ranked individuals more nearly equal in score, compared to rank scoring. For more information, see Fitness Scaling.
Proportional (@fitscalingprop) — Proportional scaling makes the scaled value of an individual proportional to its raw fitness score.
Top (@fitscalingtop) — Top scaling scales the top individuals equally. Selecting Top displays an additional field, Quantity, which specifies the number of individuals that are assigned positive scaled values. Quantity can be an integer from 1 through the population size or a fraction from 0 through 1 specifying a fraction of the population size. The default value is 0.4. Each of the individuals that produce offspring is assigned an equal scaled value, while the rest are assigned the value 0. The scaled values have the form [01/n 1/n 0 0 1/n 0 0 1/n ...].
To change the default value for Quantity at the command line, use the following syntax:
options = gaoptimset('FitnessScalingFcn', {@fitscalingtop,quantity})
where quantity is the value of Quantity.
Shift linear (@fitscalingshiftlinear) — Shift linear scaling scales the raw scores so that the expectation of the fittest individual is equal to a constant multiplied by the average score. You specify the constant in the Max survival rate field, which is displayed when you select Shift linear. The default value is 2.
To change the default value of Max survival rate at the command line, use the following syntax
options = gaoptimset('FitnessScalingFcn', {@fitscalingshiftlinear, rate})
where rate is the value of Max survival rate.
Custom lets you write your own scaling function. To specify the scaling function using the Optimization app,
Set Scaling function to Custom.
Set Function name to @myfun, where myfun is the name of your function.
If you are using ga at the command line, set
options = gaoptimset('FitnessScalingFcn', @myfun);
Your scaling function must have the following calling syntax:
function expectation = myfun(scores, nParents)
The input arguments to the function are:
scores — A vector of scalars, one for each member of the population
nParents — The number of parents needed from this population
The function returns expectation, a column vector of scalars of the same length as scores, giving the scaled values of each member of the population. The sum of the entries of expectation must equal nParents.
Passing Extra Parameters in the Optimization Toolbox documentation explains how to provide additional parameters to the function.
See Fitness Scaling for more information.
Selection options specify how the genetic algorithm chooses parents for the next generation. You can specify the function the algorithm uses in the Selection function (SelectionFcn) field in the Selection options pane. Do not use with integer problems. The options are
Stochastic uniform (@selectionstochunif) — The default selection function, Stochastic uniform, lays out a line in which each parent corresponds to a section of the line of length proportional to its scaled value. The algorithm moves along the line in steps of equal size. At each step, the algorithm allocates a parent from the section it lands on. The first step is a uniform random number less than the step size.
Remainder (@selectionremainder) — Remainder selection assigns parents deterministically from the integer part of each individual's scaled value and then uses roulette selection on the remaining fractional part. For example, if the scaled value of an individual is 2.3, that individual is listed twice as a parent because the integer part is 2. After parents have been assigned according to the integer parts of the scaled values, the rest of the parents are chosen stochastically. The probability that a parent is chosen in this step is proportional to the fractional part of its scaled value.
Uniform (@selectionuniform) — Uniform selection chooses parents using the expectations and number of parents. Uniform selection is useful for debugging and testing, but is not a very effective search strategy.
Roulette (@selectionroulette) — Roulette selection chooses parents by simulating a roulette wheel, in which the area of the section of the wheel corresponding to an individual is proportional to the individual's expectation. The algorithm uses a random number to select one of the sections with a probability equal to its area.
Tournament (@selectiontournament) — Tournament selection chooses each parent by choosing Tournament size players at random and then choosing the best individual out of that set to be a parent. Tournament size must be at least 2. The default value of Tournament size is 4.
To change the default value of Tournament size at the command line, use the syntax
options = gaoptimset('SelectionFcn',... {@selectiontournament,size})
where size is the value of Tournament size.
When Constraint parameters > Nonlinear constraint algorithm is Penalty, ga uses Tournament with size 2.
Custom enables you to write your own selection function. To specify the selection function using the Optimization app,
Set Selection function to Custom.
Set Function name to @myfun, where myfun is the name of your function.
If you are using ga at the command line, set
options = gaoptimset('SelectionFcn', @myfun);
Your selection function must have the following calling syntax:
function parents = myfun(expectation, nParents, options)
The input arguments to the function are
expectation — Expected number of children for each member of the population
nParents— Number of parents to select
options — Genetic algorithm options structure
The function returns parents, a row vector of length nParents containing the indices of the parents that you select.
Passing Extra Parameters in the Optimization Toolbox documentation explains how to provide additional parameters to the function.
See Selection for more information.
Reproduction options specify how the genetic algorithm creates children for the next generation.
Elite count (EliteCount) specifies the number of individuals that are guaranteed to survive to the next generation. Set Elite count to be a positive integer less than or equal to the population size. The default value is 2 for noninteger problems, and 0.05*min(max(10*nvars,40),100) for integer problems.
Crossover fraction (CrossoverFraction) specifies the fraction of the next generation, other than elite children, that are produced by crossover. Set Crossover fraction to be a fraction between 0 and 1, either by entering the fraction in the text box or moving the slider. The default value is 0.8.
See Setting the Crossover Fraction for an example.
Mutation options specify how the genetic algorithm makes small random changes in the individuals in the population to create mutation children. Mutation provides genetic diversity and enables the genetic algorithm to search a broader space. You can specify the mutation function in the Mutation function (MutationFcn) field in the Mutation options pane. Do not use with integer problems. You can choose from the following functions:
Gaussian (mutationgaussian) — The default mutation function for unconstrained problems, Gaussian, adds a random number taken from a Gaussian distribution with mean 0 to each entry of the parent vector. The standard deviation of this distribution is determined by the parameters Scale and Shrink, which are displayed when you select Gaussian, and by the Initial range setting in the Population options.
The Scale parameter determines the standard deviation at the first generation. If you set Initial range to be a 2-by-1 vector v, the initial standard deviation is the same at all coordinates of the parent vector, and is given by Scale*(v(2)-v(1)).
If you set Initial range to be a vector v with two rows and Number of variables columns, the initial standard deviation at coordinate i of the parent vector is given by Scale*(v(i,2) - v(i,1)).
The Shrink parameter controls how the standard deviation shrinks as generations go by. If you set Initial range to be a 2-by-1 vector, the standard deviation at the kth generation, σ_{k}, is the same at all coordinates of the parent vector, and is given by the recursive formula
$${\sigma}_{k}={\sigma}_{k-1}\left(1-\text{Shrink}\frac{k}{\text{Generations}}\right).$$
If you set Initial range to be a vector with two rows and Number of variables columns, the standard deviation at coordinate i of the parent vector at the kth generation, σ_{i,k}, is given by the recursive formula
$${\sigma}_{i,k}={\sigma}_{i,k-1}\left(1-\text{Shrink}\frac{k}{\text{Generations}}\right).$$
If you set Shrink to 1, the algorithm shrinks the standard deviation in each coordinate linearly until it reaches 0 at the last generation is reached. A negative value of Shrink causes the standard deviation to grow.
The default value of both Scale and Shrink is 1. To change the default values at the command line, use the syntax
options = gaoptimset('MutationFcn', ... {@mutationgaussian, scale, shrink})
where scale and shrink are the values of Scale and Shrink, respectively.
Uniform (mutationuniform) — Uniform mutation is a two-step process. First, the algorithm selects a fraction of the vector entries of an individual for mutation, where each entry has a probability Rate of being mutated. The default value of Rate is 0.01. In the second step, the algorithm replaces each selected entry by a random number selected uniformly from the range for that entry.
To change the default value of Rate at the command line, use the syntax
options = gaoptimset('MutationFcn', {@mutationuniform, rate})
where rate is the value of Rate.
Adaptive Feasible (mutationadaptfeasible), the default mutation function when there are constraints, randomly generates directions that are adaptive with respect to the last successful or unsuccessful generation. The mutation chooses a direction and step length that satisfies bounds and linear constraints.
Custom enables you to write your own mutation function. To specify the mutation function using the Optimization app,
Set Mutation function to Custom.
Set Function name to @myfun, where myfun is the name of your function.
If you are using ga, set
options = gaoptimset('MutationFcn', @myfun);
Your mutation function must have this calling syntax:
function mutationChildren = myfun(parents, options, nvars, FitnessFcn, state, thisScore, thisPopulation)
The arguments to the function are
parents — Row vector of parents chosen by the selection function
options — Options structure
nvars — Number of variables
FitnessFcn — Fitness function
state — Structure containing information about the current generation. The State Structure describes the fields of state.
thisScore — Vector of scores of the current population
thisPopulation — Matrix of individuals in the current population
The function returns mutationChildren—the mutated offspring—as a matrix where rows correspond to the children. The number of columns of the matrix is Number of variables.
Passing Extra Parameters in the Optimization Toolbox documentation explains how to provide additional parameters to the function.
Crossover options specify how the genetic algorithm combines two individuals, or parents, to form a crossover child for the next generation.
Crossover function (CrossoverFcn) specifies the function that performs the crossover. Do not use with integer problems. You can choose from the following functions:
Scattered (@crossoverscattered), the default crossover function for problems without linear constraints, creates a random binary vector and selects the genes where the vector is a 1 from the first parent, and the genes where the vector is a 0 from the second parent, and combines the genes to form the child. For example, if p1 and p2 are the parents
p1 = [a b c d e f g h] p2 = [1 2 3 4 5 6 7 8]
and the binary vector is [1 1 0 0 1 0 0 0], the function returns the following child:
child1 = [a b 3 4 e 6 7 8]
Single point (@crossoversinglepoint) chooses a random integer n between 1 and Number of variables and then
Selects vector entries numbered less than or equal to n from the first parent.
Selects vector entries numbered greater than n from the second parent.
Concatenates these entries to form a child vector.
For example, if p1 and p2 are the parents
p1 = [a b c d e f g h] p2 = [1 2 3 4 5 6 7 8]
and the crossover point is 3, the function returns the following child.
child = [a b c 4 5 6 7 8]
Two point (@crossovertwopoint) selects two random integers m and n between 1 and Number of variables. The function selects
Vector entries numbered less than or equal to m from the first parent
Vector entries numbered from m+1 to n, inclusive, from the second parent
Vector entries numbered greater than n from the first parent.
The algorithm then concatenates these genes to form a single gene. For example, if p1 and p2 are the parents
p1 = [a b c d e f g h] p2 = [1 2 3 4 5 6 7 8]
and the crossover points are 3 and 6, the function returns the following child.
child = [a b c 4 5 6 g h]
Intermediate (@crossoverintermediate), the default crossover function when there are linear constraints, creates children by taking a weighted average of the parents. You can specify the weights by a single parameter, Ratio, which can be a scalar or a row vector of length Number of variables. The default is a vector of all 1's. The function creates the child from parent1 and parent2 using the following formula.
child = parent1 + rand * Ratio * ( parent2 - parent1)
If all the entries of Ratio lie in the range [0, 1], the children produced are within the hypercube defined by placing the parents at opposite vertices. If Ratio is not in that range, the children might lie outside the hypercube. If Ratio is a scalar, then all the children lie on the line between the parents.
To change the default value of Ratio at the command line, use the syntax
options = gaoptimset('CrossoverFcn', ... {@crossoverintermediate, ratio});
where ratio is the value of Ratio.
Heuristic (@crossoverheuristic) returns a child that lies on the line containing the two parents, a small distance away from the parent with the better fitness value in the direction away from the parent with the worse fitness value. You can specify how far the child is from the better parent by the parameter Ratio, which appears when you select Heuristic. The default value of Ratio is 1.2. If parent1 and parent2 are the parents, and parent1 has the better fitness value, the function returns the child
child = parent2 + R * (parent1 - parent2);
To change the default value of Ratio at the command line, use the syntax
options=gaoptimset('CrossoverFcn',... {@crossoverheuristic,ratio});
where ratio is the value of Ratio.
Arithmetic (@crossoverarithmetic) creates children that are the weighted arithmetic mean of two parents. Children are always feasible with respect to linear constraints and bounds.
Custom enables you to write your own crossover function. To specify the crossover function using the Optimization app,
Set Crossover function to Custom.
Set Function name to @myfun, where myfun is the name of your function.
If you are using ga, set
options = gaoptimset('CrossoverFcn',@myfun);
Your crossover function must have the following calling syntax.
xoverKids = myfun(parents, options, nvars, FitnessFcn, ... unused,thisPopulation)
The arguments to the function are
parents — Row vector of parents chosen by the selection function
options — options structure
nvars — Number of variables
FitnessFcn — Fitness function
unused — Placeholder not used
thisPopulation — Matrix representing the current population. The number of rows of the matrix is Population size and the number of columns is Number of variables.
The function returns xoverKids—the crossover offspring—as a matrix where rows correspond to the children. The number of columns of the matrix is Number of variables.
Passing Extra Parameters in the Optimization Toolbox documentation explains how to provide additional parameters to the function.
Note: Subpopulations refer to a form of parallel processing for the genetic algorithm. ga currently does not support this form. In subpopulations, each worker hosts a number of individuals. These individuals are a subpopulation. The worker evolves the subpopulation independently of other workers, except when migration causes some individuals to travel between workers. Because ga does not currently support this form of parallel processing, there is no benefit to setting PopulationSize to a vector, or to setting the MigrationDirection, MigrationInterval, or MigrationFraction options. |
Migration options specify how individuals move between subpopulations. Migration occurs if you set Population size to be a vector of length greater than 1. When migration occurs, the best individuals from one subpopulation replace the worst individuals in another subpopulation. Individuals that migrate from one subpopulation to another are copied. They are not removed from the source subpopulation.
You can control how migration occurs by the following three fields in the Migration options pane:
Direction (MigrationDirection) — Migration can take place in one or both directions.
If you set Direction to Forward ('forward'), migration takes place toward the last subpopulation. That is, the nth subpopulation migrates into the (n+1)th subpopulation.
If you set Direction to Both ('both'), the n^{th} subpopulation migrates into both the (n–1)th and the (n+1)th subpopulation.
Migration wraps at the ends of the subpopulations. That is, the last subpopulation migrates into the first, and the first may migrate into the last.
Interval (MigrationInterval) — Specifies how many generation pass between migrations. For example, if you set Interval to 20, migration takes place every 20 generations.
Fraction (MigrationFraction) — Specifies how many individuals move between subpopulations. Fraction specifies the fraction of the smaller of the two subpopulations that moves. For example, if individuals migrate from a subpopulation of 50 individuals into a subpopulation of 100 individuals and you set Fraction to 0.1, the number of individuals that migrate is 0.1*50=5.
Constraint parameters refer to the nonlinear constraint solver. For details on the algorithm, see Nonlinear Constraint Solver Algorithms.
Choose between the nonlinear constraint algorithms by setting the NonlinConAlgorithm option to 'auglag' (Augmented Lagrangian) or 'penalty' (Penalty algorithm).
Initial penalty (InitialPenalty) — Specifies an initial value of the penalty parameter that is used by the nonlinear constraint algorithm. Initial penalty must be greater than or equal to 1, and has a default of 10.
Penalty factor (PenaltyFactor) — Increases the penalty parameter when the problem is not solved to required accuracy and constraints are not satisfied. Penalty factor must be greater than 1, and has a default of 100.
The penalty algorithm uses the gacreationnonlinearfeasible creation function by default. This creation function uses fmincon to find feasible individuals. gacreationnonlinearfeasible starts fmincon from a variety of initial points within the bounds from the PopInitRange option. Optionally, gacreationnonlinearfeasible can run fmincon in parallel on the initial points.
You can specify tuning parameters for gacreationnonlinearfeasible using the following name-value pairs.
Name | Value |
---|---|
SolverOpts | fmincon options, created using optimoptions or optimset. |
UseParallel | When true, run fmincon in parallel on initial points; default is false. |
NumStartPts | Number of start points, a positive integer up to sum(PopulationSize) in value. |
Include the name-value pairs in a cell array along with @gacreationnonlinearfeasible.
options = gaoptimset('CreationFcn',{@gacreationnonlinearfeasible,... 'UseParallel',true,'NumStartPts',20});
Multiobjective options define parameters characteristic of the multiobjective genetic algorithm. You can specify the following parameters:
DistanceMeasureFcn — Defines a handle to the function that computes distance measure of individuals, computed in decision variable or design space (genotype) or in function space (phenotype). For example, the default distance measure function is distancecrowding in function space, or {@distancecrowding,'phenotype'}.
ParetoFraction — Sets the fraction of individuals to keep on the first Pareto front while the solver selects individuals from higher fronts. This option is a scalar between 0 and 1.
A hybrid function is another minimization function that runs after the genetic algorithm terminates. You can specify a hybrid function in Hybrid function (HybridFcn) options. Do not use with integer problems. The choices are
[] — No hybrid function.
fminsearch (@fminsearch) — Uses the MATLAB^{®} function fminsearch to perform unconstrained minimization.
patternsearch (@patternsearch) — Uses a pattern search to perform constrained or unconstrained minimization.
fminunc (@fminunc) — Uses the Optimization Toolbox function fminunc to perform unconstrained minimization.
fmincon (@fmincon) — Uses the Optimization Toolbox function fmincon to perform constrained minimization.
You can set separate options for the hybrid function. Use optimset for fminsearch, psoptimset for patternsearch, or optimoptions for fmincon or fminunc. For example:
hybridopts = optimoptions('fminunc','Display','iter','Algorithm','quasi-newton');
Include the hybrid options in the Genetic Algorithm options structure as follows:
options = gaoptimset(options,'HybridFcn',{@fminunc,hybridopts});
hybridopts must exist before you set options.
See Include a Hybrid Function for an example.
Stopping criteria determine what causes the algorithm to terminate. You can specify the following options:
Generations (Generations) — Specifies the maximum number of iterations for the genetic algorithm to perform. The default is 100*numberOfVariables.
Time limit (TimeLimit) — Specifies the maximum time in seconds the genetic algorithm runs before stopping, as measured by cputime.
Fitness limit (FitnessLimit) — The algorithm stops if the best fitness value is less than or equal to the value of Fitness limit.
Stall generations (StallGenLimit) — The algorithm stops if the average relative change in the best fitness function value over Stall generations is less than or equal to Function tolerance. (If the StallTest option is 'geometricWeighted', then the test is for a geometric weighted average relative change.) For a problem with nonlinear constraints, Stall generations applies to the subproblem (see Nonlinear Constraint Solver Algorithms).
For gamultiobj, if the weighted average relative change in the spread of the Pareto solutions over Stall generations is less than Function tolerance, and the spread is smaller than the average spread over the last Stall generations, then the algorithm stops. The spread is a measure of the movement of the Pareto front.
Stall time limit (StallTimeLimit) — The algorithm stops if there is no improvement in the best fitness value for an interval of time in seconds specified by Stall time limit, as measured by cputime.
Function tolerance (TolFun) — The algorithm stops if the average relative change in the best fitness function value over Stall generations is less than or equal to Function tolerance. (If the StallTest option is 'geometricWeighted', then the test is for a geometric weighted average relative change.)
For gamultiobj, if the weighted average relative change in the spread of the Pareto solutions over Stall generations is less than Function tolerance, and the spread is smaller than the average spread over the last Stall generations, then the algorithm stops. The spread is a measure of the movement of the Pareto front.
Constraint tolerance (TolCon) — The Constraint tolerance is not used as stopping criterion. It is used to determine the feasibility with respect to nonlinear constraints. Also, max(sqrt(eps),sqrt(TolCon)) determines feasibility with respect to linear constraints.
See Set Maximum Number of Generations for an example.
Output functions are functions that the genetic algorithm calls at each generation. To specify the output function using the Optimization app,
Select Custom function.
Enter @myfun in the text box, where myfun is the name of your function. Write myfun with appropriate syntax.
To pass extra parameters in the output function, use Anonymous Functions.
For multiple output functions, enter a cell array of output function handles: {@myfun1,@myfun2,...}.
At the command line, set
options = gaoptimset('OutputFcns',@myfun);
For multiple output functions, enter a cell array:
options = gaoptimset('OutputFcns',{@myfun1,@myfun2,...});
To see a template that you can use to write your own output functions, enter
edit gaoutputfcntemplate
at the MATLAB command line.
The output function has the following calling syntax.
[state,options,optchanged] = myfun(options,state,flag)
The function has the following input arguments:
options — Options structure
state — Structure containing information about the current generation. The State Structure describes the fields of state.
flag — String indicating the current status of the algorithm as follows:
'init' — Initial stage
'iter' — Algorithm running
'interrupt' — Intermediate stage
'done' — Algorithm terminated
Passing Extra Parameters in the Optimization Toolbox documentation explains how to provide additional parameters to the function.
The output function returns the following arguments to ga:
state — Structure containing information about the current generation. The State Structure describes the fields of state. To stop the iterations, set state.StopFlag to a nonempty string.
options — Options structure modified by the output function. This argument is optional.
optchanged — Flag indicating changes to options
Level of display ('Display') specifies how much information is displayed at the command line while the genetic algorithm is running. The available options are
Off ('off') — No output is displayed.
Iterative ('iter') — Information is displayed at each iteration.
Diagnose ('diagnose') — Information is displayed at each iteration. In addition, the diagnostic lists some problem information and the options that have been changed from the defaults.
Final ('final') — The reason for stopping is displayed.
Both Iterative and Diagnose display the following information:
Generation — Generation number
f-count — Cumulative number of fitness function evaluations
Best f(x) — Best fitness function value
Mean f(x) — Mean fitness function value
Stall generations — Number of generations since the last improvement of the fitness function
When a nonlinear constraint function has been specified, Iterative and Diagnose do not display the Mean f(x), but will additionally display:
Max Constraint — Maximum nonlinear constraint violation
The default value of Level of display is
Off in the Optimization app
'final' in an options structure created using gaoptimset
You can choose to have your fitness and constraint functions evaluated in serial, parallel, or in a vectorized fashion. These options are available in the User function evaluation section of the Options pane of the Optimization app, or by setting the 'Vectorized' and 'UseParallel' options with gaoptimset.
When Evaluate fitness and constraint functions ('Vectorized') is in serial ('off'), ga calls the fitness function on one individual at a time as it loops through the population. (At the command line, this assumes 'UseParallel' is at its default value of false.)
When Evaluate fitness and constraint functions ('Vectorized') is vectorized ('on'), ga calls the fitness function on the entire population at once, i.e., in a single call to the fitness function.
If there are nonlinear constraints, the fitness function and the nonlinear constraints all need to be vectorized in order for the algorithm to compute in a vectorized manner.
See Vectorize the Fitness Function for an example.
When Evaluate fitness and constraint functions (UseParallel) is in parallel (true), ga calls the fitness function in parallel, using the parallel environment you established (see How to Use Parallel Processing). At the command line, set UseParallel to false to compute serially.
Note: You cannot simultaneously use vectorized and parallel computations. If you set 'UseParallel' to true and 'Vectorized' to 'on', ga evaluates your fitness and constraint functions in a vectorized manner, not in parallel. |
How Fitness and Constraint Functions Are Evaluated
Vectorized = 'Off' | Vectorized = 'On' | |
---|---|---|
UseParallel = false | Serial | Vectorized |
UseParallel = true | Parallel | Vectorized |