|
|
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
" F. v* n& x# ]$ VNSGA2算法特征选择MATLAB实现(多目标) ! q. N. P3 x6 }& U; f0 P0 v% \0 H
9 y% |: g* q7 \6 l( M
( L( d0 Y) h. e8 x T* l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。/ A: P2 J& D. \0 B9 p
- k7 F5 r) H" X. O+ T4 M需要优化的两个目标为特征数和精度。
9 t( S6 z) A& X4 N2 G0 k' k3 @9 {4 v+ e) P) }2 K) d6 I
nsga2是一个多目标优化算法。
5 |% B k% o6 ~2 F- Y
$ T! W8 h5 x$ A6 F$ k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)8 H$ T7 _, f3 a* G/ }$ X$ u2 v) x
# R" i9 u6 R% [/ J3 M2 {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:
' a& }0 x* ^1 l+ V5 o) ^& f0 N, R3 C) Q3 p* q6 D+ R$ ~9 J3 N5 ~$ L+ F9 q6 [
- 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.6 h: Q9 q- N; x2 o1 j/ p
! |& @# w0 d3 c5 @5 ?8 ]. t% Z$ b$ K2 f
# n+ u# {. ~. i+ d& \MATLAB代码主函数:. _* q2 z! z! Z! g( H4 j: K% T- L
2 O5 D% ^$ d, p2 f/ [3 l: |
- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
- %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');
- end5 g) @! S, ~* C3 d4 c+ i; T
3 l3 C: b* Z0 @" H% F% t- {评价函数(利用林志仁SVM进行训练):
4 [, Y. w! q$ p1 s0 v" J# [; i0 D6 J, M) E
- 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};
- end9 w$ @+ P% m# N; B
! H; {. N* M. ^5 v; {# a
选的的数据集请从UCI上下载。
' U: w- L" o8 {
' Q8 k7 {. q4 ]% F" D) F结果:
% ?$ ^; g: g6 F6 M0 j
1 y( I- {" D% o, Y- ]①pareto面% Z" c+ h% O5 L+ j d3 c
7 T: n) C3 b }4 C3 E
6 [ S, e) p. D" Z) i' r- L0 z% F) D/ ?" E2 C$ Z
最后粒子的数据(选出的特征数和精确度)
) o, q; P! V% F8 s$ S
# C& I8 e3 e2 Y: K/ b5 x7 {; v9 R3 s1 ?0 d7 w
6 T7 P; g6 E2 v# c8 U6 P& |7 \0 h" a1 x' l: }$ p4 X! M U/ ?
|
|