|
|
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
8 @; H6 n5 \% A' e) A; JNSGA2算法特征选择MATLAB实现(多目标)
: {4 X6 b2 ^- Z& f: C2 P& F9 N' t; ~
1 N0 Q- p) T) l" K- y
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。
& s- [% Q8 r7 s+ F# @5 ], |9 E" l
$ v. {( V& g9 z5 H6 w需要优化的两个目标为特征数和精度。
2 l5 B- \' ?0 Z& o) I
( T, l- ^; T: [nsga2是一个多目标优化算法。
5 C, b5 Y. B$ V! {7 I" E% `& a9 J8 ~: C/ Z6 b
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)- O9 U, s) C: f* m: O% _+ ?
, R5 j- q G3 {+ _; o2 M6 J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:
9 m5 ~* b4 w! h
6 o4 e6 `6 r$ U; Y: C9 ]- 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+ q- U& [
1 h( ?2 M5 V' Z) M, c$ }( h. c7 a
$ Y5 S" m! E9 ~0 XMATLAB代码主函数:+ |8 ]% i/ a, b8 {' |) M
; \/ r: m G" H2 v/ r/ J
- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
- %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');
- end' \# T3 a7 S, X3 z0 t3 ^, j. @
9 X1 m$ X: A u
评价函数(利用林志仁SVM进行训练):
4 ?8 k( s; t( g7 M: T' o
: h0 q2 l0 J+ K; E! t5 c+ ~% F4 U4 ^7 u- 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 T4 ] Z1 y3 G* G2 ]
- k/ Q! K( |: V* a' q# R
选的的数据集请从UCI上下载。+ X% F" r9 g0 q: F# ~, z7 P
0 U7 R( s, s' k! y8 ?% g# D
结果:
2 P& w. X' ]- C, v' K6 n$ q# ^% W! W( I& d, b
①pareto面
V6 f0 p& m! e& c' V2 n3 J
- \8 H) V# [$ A
3 N3 x8 `% c' d5 ]! ?$ [- f
2 D w# Q! {) o0 b! P: ^
最后粒子的数据(选出的特征数和精确度)
" Z! A* V) J* M: T7 ^
; b: W- @8 k l+ E, x* R" h: @2 P' x; `# K
2 g6 I7 Y2 i5 e2 S5 r) z; a. M$ H y* ]- t& b
|
|