【ELM】动态自适应可变加权极限学习机ELM预测(Matlab代码实现)

 欢迎来到本博客❤️❤️

博主优势:博客内容尽量做到思维缜密,逻辑清晰,为了方便读者。

⛳️座右铭:行百里者,半于九十。

目录

1 概述

2 运行结果

3 参考文献

4 Matlab代码实现

1 概述

该文提出了一种新的长度可变极限学习机。该方法旨在提高单隐藏层前馈神经网络(SLFN)在丰富的动态不平衡数据下的学习性能。粒子群优化 (PSO) 涉及增量学习期间的超参数调整和更新。该算法使用燃气涡轮发动机C-MAPSS(商用模块化航空推进系统仿真)数据集的子集进行评估,并与其衍生物进行比较。结果表明,新算法具有较好的学习效果。

文献来源:

【ELM】动态自适应可变加权极限学习机ELM预测(Matlab代码实现)_第1张图片

2 运行结果

【ELM】动态自适应可变加权极限学习机ELM预测(Matlab代码实现)_第2张图片

【ELM】动态自适应可变加权极限学习机ELM预测(Matlab代码实现)_第3张图片

部分代码:

function [net] = OP_W_LCI_ELM(xtr,ytr,xts,yts,sample,Options)
% Length changeabale incremental ELM :
% hyper parameters
k=Options.k;
lambda=Options.lambda;
MaxHiddenNeurons=Options.MaxHiddenNeurons;
epsilon=Options.epsilon;
ActivationFunType=Options.ActivationFunType;
C=Options.C;
Weighted=Options.Weighted;
maxite=Options.maxite;
epsilonPSO=Options.epsilonPSO;
%%%%%%%%%%%%%%%%
[~,InputDimension] = size(xtr);% get dimensions
start_time_training = cputime;% calculating time
W = [];% initialize input weights 
b = [];% initialize hidden layer biases 
TrainingAccuracy = [];% initialize
OutputWeights = [];% initialize
num = 0;  % initialize number of hidden nodes
count = 0;% initialize ITERATION COUNTER
E=[];     % nitialize vector of RMSEs by iteration
Ets=[];   % initialize vector of RMSEs of testing
Tolerance=1;% initialize tolerance value 
%Training phase
node=0;

while num < MaxHiddenNeurons & Tolerance > epsilon
    len = floor(k*exp(-lambda*count))+1; %calculate the number of newly added hidden nodes for this loop
    wi = rand(InputDimension,len)*2-1;
    W = [W wi];
    switch ActivationFunType
        case 'sig'
            bi = rand(1,len)*2-1;
            b = [b bi];
            H = activeadd(xtr,W,b);
        case 'radbas'
            bi = rand(1,len)*0.5;
            b = [b bi];
            H = activerbf(xtr,W,b);
        otherwise
    end
    % particl swarm optimization for best regularization and weightes
    % Adjustments
    % dfine initial population
    population=[C Weighted];
    %  random search
    [~,population,~,fit_behavior]=PSO(population,H,ytr,Options);
    % save best solutions
    Reg(count+1,:)=population;
    % determine beta
    [beta] = R(H,ytr, population);
    % calculate target
    f = H*beta; 
    %update the output function
    OutputWeights = beta; %update the output weights
    TrainingAccuracy = sqrt(mse(ytr-f));
    E=[E TrainingAccuracy];% save training accuraccy
    if count>1
    Tolerance= sqrt(mse(E(count)-E(count-1))); % tolerance function 
    Tol(count-1,1)=Tolerance;
    end
    count = count+1;% counter
    % testing phase
    [O,TestingAccuracy,TestingTime]=ELMtest(xts,yts,ActivationFunType,W,b,OutputWeights);
    % testing accuracy
    Ets=[Ets TestingAccuracy];
    % number of used hiddden nodes
    num = num+len;
    % used hidden nodes in each step
    nodes(count)=node+len;
    % stor
    node=nodes(count);
end
end_time_training = cputime;
TrainingTime = end_time_training - start_time_training;
clear wi bi H beta ;
% Testing phase
% sample
switch ActivationFunType
    case 'sig'
        y_sample = activeadd(sample,W,b)*OutputWeights;
    case 'radbas'
        y_sample = activerbf(sample,W,b)*OutputWeights;
    otherwise
end
y_sample=scaledata(y_sample,7,125);

 num=num-1;
% performances 
[SCORE,S,d,er]=Score(O,yts);
% stor results
%%%
net.E=E;
net.Ets=Ets;
net.num=num;
net.nodes=nodes;
net.Tol=Tol;
net.Tr_time=TrainingTime;
net.Ts_time=TestingTime;
net.Tr_acc=TrainingAccuracy;
net.Ts_acc=TestingAccuracy;
net.Opt_solution=population;
net.Score=SCORE;
net.S=S;
net.d=d;
net.er=er;
net.reg=Reg;
net.y_sample=y_sample;
net.fit_behavior=fit_behavior;

end

3 参考文献

部分理论来源于网络,如有侵权请联系删除。

[1] Y. X. Wu, D. Liu, and H. Jiang, “Length-Changeable Incremental Extreme Learning Machine,” J. Comput. Sci. Technol., vol. 32, no. 3, pp. 630–643, 2017.
[2] A. Saxena, M. Ieee, K. Goebel, D. Simon, and N. Eklund, “Damage Propagation Modeling for Aircraft Engine Prognostics,” Response, 2008.
[3] M. N. Alam, “Codes in MATLAB for Particle Swarm Optimization Codes in MATLAB for Particle Swarm Optimization,” no. March, 2016.
[4] J. Cao, K. Zhang, M. Luo, C. Yin, and X. Lai, “Extreme learning machine and adaptive sparse representation for image classification,” Neural Networks, vol. 81, no. 61773019, pp. 91–102, 2016.

4 Matlab代码实现

你可能感兴趣的:(#,神经网络预测预测与分类,#,matlab,开发语言,ELM,动态自适应)