gaoptimset
（不推荐）创建遗传算法 options 结构体
不推荐使用 gaoptimset。请改用 optimoptions。有关详细信息，请参阅版本历史记录。
语法
说明
没有输入或输出参量的 gaoptimset 显示带有其有效值的完整选项列表。一般情况下，默认值出现在括号 {} 中。
注意
指示的默认值并不是所有问题类型的默认值。要获得更新的列表，请评估 optimoptions('ga') 或 optimoptions('gamultiobj')。
options = gaoptimset(Name,Value)options 的结构体，并将 'Name1' 的值设置为 Value1、将 'Name2' 的值设置为 Value2，依此类推。gaoptimset 将任何未指定的选项设置为 []，这意味着求解器使用默认选项值。gaoptimset 可以指定一个选项名称，仅使用足够的前导字符来唯一地定义名称。gaoptimset 会忽略选项名称的大小写。例如，'Display'、'display' 和 'Disp' 是等效的选项名称。
options = gaoptimsetoptions 的结构体，其中包含遗传算法的选项。在这种情况下，gaoptimset 将所有选项值设置为 []，表示使用默认值。
options = gaoptimset(@ga)options = gaoptimset(@gamultiobj)options 结构体，其中包含分别针对 ga 或 gamultiobj 求解器的明确默认值的选项。
options = gaoptimset(oldopts,Name,Value)oldopts 的副本，并使用指定的值修改指定的选项。
示例
要查看选项结构体中 ga 和 gamultiobj 的所有可用选项，请运行不带输入或输出参量的 gamultiobj。
gaoptimset
PopulationType: [ 'bitstring' | 'custom' | {'doubleVector'} ]
PopInitRange: [ matrix | {[-10;10]} ]
PopulationSize: [ positive scalar ]
EliteCount: [ positive scalar | {0.05*PopulationSize} ]
CrossoverFraction: [ positive scalar | {0.8} ]
ParetoFraction: [ positive scalar | {0.35} ]
MigrationDirection: [ 'both' | {'forward'} ]
MigrationInterval: [ positive scalar | {20} ]
MigrationFraction: [ positive scalar | {0.2} ]
Generations: [ positive scalar ]
TimeLimit: [ positive scalar | {Inf} ]
FitnessLimit: [ scalar | {-Inf} ]
StallGenLimit: [ positive scalar ]
StallTest: [ 'geometricWeighted' | {'averageChange'} ]
StallTimeLimit: [ positive scalar | {Inf} ]
TolFun: [ positive scalar ]
TolCon: [ positive scalar | {1e-6} ]
InitialPopulation: [ matrix | {[]} ]
InitialScores: [ column vector | {[]} ]
NonlinConAlgorithm: [ 'penalty' | {'auglag'} ]
InitialPenalty: [ positive scalar | {10} ]
PenaltyFactor: [ positive scalar | {100} ]
CreationFcn: [ function_handle | @gacreationuniform | @gacreationlinearfeasible ]
FitnessScalingFcn: [ function_handle | @fitscalingshiftlinear | @fitscalingprop |
@fitscalingtop | {@fitscalingrank} ]
SelectionFcn: [ function_handle | @selectionremainder | @selectionuniform |
@selectionroulette | @selectiontournament | @selectionstochunif ]
CrossoverFcn: [ function_handle | @crossoverheuristic | @crossoverintermediate |
@crossoversinglepoint | @crossovertwopoint | @crossoverarithmetic |
@crossoverscattered ]
MutationFcn: [ function_handle | @mutationuniform | @mutationadaptfeasible |
@mutationgaussian ]
DistanceMeasureFcn: [ function_handle | {@distancecrowding} ]
HybridFcn: [ @fminsearch | @patternsearch | @fminunc | @fmincon | {[]} ]
Display: [ 'off' | 'iter' | 'diagnose' | {'final'} ]
OutputFcns: [ function_handle | {[]} ]
PlotFcns: [ function_handle | @gaplotbestf | @gaplotbestindiv | @gaplotdistance |
@gaplotexpectation | @gaplotgenealogy | @gaplotselection | @gaplotrange |
@gaplotscorediversity | @gaplotscores | @gaplotstopping |
@gaplotmaxconstr | @gaplotrankhist | @gaplotpareto | @gaplotspread |
@gaplotparetodistance |{[]} ]
PlotInterval: [ positive scalar | {1} ]
Vectorized: [ 'on' | {'off'} ]
UseParallel: [ logical scalar | true | {false} ]
设置 ga 的选项，使其具有包含点 [1,1] 的起始种群，并使用 gaplotbestf 绘图函数。
options = gaoptimset('InitialPopulation',[1 1],... 'PlotFcns',@gaplotbestf);
使用指定的选项，找到函数 rastriginsfcn 的局部最小值，该函数在运行此示例时可用。
rng default % For reproducibility nvar = 2; A = []; b = []; Aeq = []; beq = []; lb = []; ub = []; nlconst = []; [x,fval] = ga(@rastriginsfcn,nvar,A,b,Aeq,beq,lb,ub,nlconst,options)
ga stopped because the average change in the fitness value is less than options.FunctionTolerance.
x = 1×2
0.0227 0.0656
fval = 0.9442
使用 gaoptimset 为 ga 创建一个选项结构体，并将其值设置为默认值。
options = gaoptimset(@ga)
options = struct with fields:
PopulationType: 'doubleVector'
PopInitRange: []
PopulationSize: 'auto'
EliteCount: '0.05*PopulationSize'
CrossoverFraction: 0.8000
ParetoFraction: []
MigrationDirection: 'forward'
MigrationInterval: 20
MigrationFraction: 0.2000
Generations: '100*numberOfVariables'
TimeLimit: Inf
FitnessLimit: -Inf
StallGenLimit: 50
StallTest: 'averageChange'
StallTimeLimit: Inf
TolFun: 1.0000e-06
TolCon: 1.0000e-03
InitialPopulation: []
InitialScores: []
NonlinConAlgorithm: 'auglag'
InitialPenalty: 10
PenaltyFactor: 100
PlotInterval: 1
CreationFcn: []
FitnessScalingFcn: @fitscalingrank
SelectionFcn: []
CrossoverFcn: []
MutationFcn: []
DistanceMeasureFcn: []
HybridFcn: []
Display: 'final'
PlotFcns: []
OutputFcns: []
Vectorized: 'off'
IntegerTolerance: 1.0000e-05
UseParallel: 0
创建遗传算法选项以返回迭代显示。
oldopts = gaoptimset('Display','iter');
修改 oldopts 以包含 gaplotbestf 绘图函数和 250 的种群规模。
options = gaoptimset(oldopts,'PlotFcns',@gaplotbestf,... 'PopulationSize',250)
options = struct with fields:
PopulationType: []
PopInitRange: []
PopulationSize: 250
EliteCount: []
CrossoverFraction: []
ParetoFraction: []
MigrationDirection: []
MigrationInterval: []
MigrationFraction: []
Generations: []
TimeLimit: []
FitnessLimit: []
StallGenLimit: []
StallTest: []
StallTimeLimit: []
TolFun: []
TolCon: []
InitialPopulation: []
InitialScores: []
NonlinConAlgorithm: []
InitialPenalty: []
PenaltyFactor: []
PlotInterval: []
CreationFcn: []
FitnessScalingFcn: []
SelectionFcn: []
CrossoverFcn: []
MutationFcn: []
DistanceMeasureFcn: []
HybridFcn: []
Display: 'iter'
PlotFcns: @gaplotbestf
OutputFcns: []
Vectorized: []
IntegerTolerance: []
UseParallel: []
创建两组遗传算法选项，oldopts 和 newopts。为 gaplotbestf 指定迭代显示和 oldopts 绘图函数。为 newopts 指定无显示且种群规模为 300。
oldopts = gaoptimset('Display','iter','PlotFcns',@gaplotbestf); newopts = gaoptimset('Display','off','PopulationSize',300);
将这些选项结合起来，使 newopts 优先。
options = gaoptimset(oldopts,newopts)
options = struct with fields:
PopulationType: []
PopInitRange: []
PopulationSize: 300
EliteCount: []
CrossoverFraction: []
ParetoFraction: []
MigrationDirection: []
MigrationInterval: []
MigrationFraction: []
Generations: []
TimeLimit: []
FitnessLimit: []
StallGenLimit: []
StallTest: []
StallTimeLimit: []
TolFun: []
TolCon: []
InitialPopulation: []
InitialScores: []
NonlinConAlgorithm: []
InitialPenalty: []
PenaltyFactor: []
PlotInterval: []
CreationFcn: []
FitnessScalingFcn: []
SelectionFcn: []
CrossoverFcn: []
MutationFcn: []
DistanceMeasureFcn: []
HybridFcn: []
Display: 'off'
PlotFcns: @gaplotbestf
OutputFcns: []
Vectorized: []
IntegerTolerance: []
UseParallel: []
请注意，Display 选项的值是 newopts 中指定的值。
