|
EDA365欢迎您登录!
您需要 登录 才可以下载或查看,没有帐号?注册
x
* u3 I, b" { a& ^5 X6 M
NSGA2算法特征选择MATLAB实现(多目标) ) _1 ~' ]! B) r% k4 j1 p5 R. L- a
1 G/ b- R) O4 R+ G4 X
8 |/ `( `5 t& z! z' f1 }利用nsga2进行进行特征选择其主要思想是:将子集的选择看作是一个搜索寻优问题(wrapper方法),生成不同的组合,对组合进行评价,再与其他的组合进行比较。这样就将子集的选择看作是一个是一个优化问题。& x& ]" n1 j* A& I: q
3 W$ z" A, V! `# N
需要优化的两个目标为特征数和精度。; C) t$ x7 x; n) r7 O" o/ E0 @; m; j
/ Q% F7 ~% Z$ `9 l+ Unsga2是一个多目标优化算法。1 k: O: t0 x% f( S
7 V$ Q/ F$ J9 K; n% H3 K% D& E具体的nsga2通用算法请看:NSGA2算法MATLAB实现(能够自定义优化函数)$ x1 g5 P. ~+ u
+ f4 w* N- F& [
具体的特征选择代码在上述代码的基础上改了两个①主函数②评价函数,增加了一个数据分成训练集和测试集的函数:
. j# J2 R& ~5 h7 w9 d& o. L7 |7 [- O9 P1 k0 A. S; J- D( \
- function divide_datasets()
- load Parkinson.mat;
- dataMat=Parkinson_f;
- len=size(dataMat,1);
- %归一化
- maxV = max(dataMat);
- minV = min(dataMat);
- range = maxV-minV;
- newdataMat = (dataMat-repmat(minV,[len,1]))./(repmat(range,[len,1]));
- Indices = crossvalind('Kfold', length(Parkinson_label), 10);
- site = find(Indices==1|Indices==2|Indices==3);
- train_F = newdataMat(site,:);
- train_L = Parkinson_label(site);
- site2 = find(Indices~=1&Indices~=2&Indices~=3);
- test_F = newdataMat(site2,:);
- test_L =Parkinson_label(site2);
- save train_F train_F;
- save train_L train_L;
- save test_F test_F;
- save test_L test_L;
- end
- %what doesn't kill you makes you stronger, stand a little taller,doesn't mean i'm over cause you're gonw. v' ^5 v7 S" x9 `
X; ]# g9 V$ R/ |! J
' }: {' \8 y& E0 a/ P C* ~
MATLAB代码主函数:
8 d1 |+ R: A% @! ]
5 e2 d2 I1 L8 B/ M" S- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
- %此处可以更改
- %更多机器学习内容请访问omegaxyz.com
- clc;
- clear;
- pop = 500; %种群数量
- gen = 100; %迭代次数
- M = 2; %目标数量
- V = 22; %维度
- min_range = zeros(1, V); %下界
- max_range = ones(1,V); %上界
- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
- %特征选择
- divide_datasets();
- global answer
- answer=cell(M,3);
- global choice %选出的特征个数
- choice=0.8;
- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
- chromosome = initialize_variables(pop, M, V, min_range, max_range);
- chromosome = non_domination_sort_mod(chromosome, M, V);
- for i = 1 : gen
- pool = round(pop/2);
- tour = 2;
- parent_chromosome = tournament_selection(chromosome, pool, tour);
- mu = 20;
- mum = 20;
- offspring_chromosome = genetic_operator(parent_chromosome,M, V, mu, mum, min_range, max_range);
- [main_pop,~] = size(chromosome);
- [offspring_pop,~] = size(offspring_chromosome);
- clear temp
- intermediate_chromosome(1:main_pop,:) = chromosome;
- intermediate_chromosome(main_pop + 1 : main_pop + offspring_pop,1 : M+V) = offspring_chromosome;
- intermediate_chromosome = non_domination_sort_mod(intermediate_chromosome, M, V);
- chromosome = replace_chromosome(intermediate_chromosome, M, V, pop);
- if ~mod(i,100)
- clc;
- fprintf('%d generations completed\n',i);
- end
- end
- if M == 2
- plot(chromosome(:,V + 1),chromosome(:,V + 2),'*');
- xlabel('f_1'); ylabel('f_2');
- title('Pareto Optimal Front');
- elseif M == 3
- plot3(chromosome(:,V + 1),chromosome(:,V + 2),chromosome(:,V + 3),'*');
- xlabel('f_1'); ylabel('f_2'); zlabel('f_3');
- title('Pareto Optimal SuRFace');
- end2 N; n2 b/ V2 Q% I, Q7 u: A6 {3 n
) S2 f9 T! j$ e) O. `/ N评价函数(利用林志仁SVM进行训练):5 e2 B5 Q8 Z M/ _
3 U% ], b+ N! `* q- function f = evaluate_objective(x, M, V, i)
- f = [];
- global answer
- global choice
- load train_F.mat;
- load train_L.mat;
- load test_F.mat;
- load test_L.mat;
- temp_x = x(1:V);
- inmodel = temp_x>choice;%%%%%设定恰当的阈值选择特征
- f(1) = sum(inmodel(1,:));
- answer(i,1)={f(1)};
- model = libsvmtrain(train_L,train_F(:,inmodel), '-s 0 -t 2 -c 1.2 -g 2.8');
- [predict_label, ~, ~] = libsvmpredict(test_L,test_F(:,inmodel),model,'-q');
- error=0;
- for j=1:length(test_L)
- if(predict_label(j,1) ~= test_L(j,1))
- error = error+1;
- end
- end
- error = error/length(test_L);
- f(2) = error;
- answer(i,2)={error};
- answer(i,3)={inmodel};
- end
. H# Q& H1 h8 ?& g & m# b0 R$ c8 Z* ^
选的的数据集请从UCI上下载。8 p7 {8 Z4 O$ d. u1 V( o+ H5 t
2 q% [: l [5 C0 [
结果:
1 v6 a8 `2 m" G+ z2 D7 ?
, Q5 V! M2 G" w; t①pareto面
# C; ^' S1 T9 N" l& i8 ~2 \) J9 R; @, h& x& C- A2 C" v& L
1 H3 r0 ?3 @" ]
7 C h( x" k9 n; U+ _; i最后粒子的数据(选出的特征数和精确度)
7 ?4 Z, g: ~! E8 a) O" R
# N; Z( v1 g. Q& C2 g- R, J5 P; R z7 z8 Q
+ g. m- x) F; l6 H7 M$ h9 O( y
* A! x1 t, I% f( `% K* U |
|