matlab 代码复现问题

情况是这样的，我下载了一篇文献的代码，是动态多目标问题的，连接放这里，有兴趣的老哥可以看看

第一个，基于历史的，问题是他给的代码是函数，如何在platemo平台上跑呢，我试着处理了一下，报错，原来代码是这样的

1. function [PopX,Pareto,POF_iter,POS_iter]=NSGAII22( Problem,popSize,MaxIt,t,T_parameter,group,Mp,hisPareto,Mm,ni,bi,ci,numi,init_pop)
   
2. 
   
3.     %% Generate random population
   
4. %     Population = Global.Initialization();
   
5. CostFunction=Problem.FObj;  % Cost Function
   
6. nVar=size(Problem.XLow,1);             % Number of Decision Variables
   
7. VarSize=[nVar 1];   % Decision Variables Matrix Size
   
8. detect_py=0;
   
9. VarMin=Problem.XLow;
   
10. VarMax=Problem.XUpp;
    
11. T=size(hisPareto,2)+1;
    
12.    
    
13. nObj=Problem.NObj;
    
14. nPop=popSize;    % Population Size (Number of Sub-Problems)
    
15. nArchive=100;
    
16. Lower = repmat(VarMin,1,nPop);
    
17. Upper = repmat(VarMax,1,nPop);
    
18. crossover_params.gamma=0.5;
    
19. crossover_params.VarMin=VarMin;
    
20. crossover_params.VarMax=VarMax;
    
21. empty_individual.Position=[];
    
22. empty_individual.Cost=[];
    
23. pop=repmat(empty_individual,nPop,1);
    
24. Off=repmat(empty_individual,nPop,1);
    
25. Offspring=repmat(empty_individual,nPop,1);
    
26. gailv=0.5;
    
27. gailv2=0.1;
    
28. zhixing1=1;
    
29. zhixing2=1;
    
30. traindata=[];
    
31. testdata=[];
    
32. if nargin == 6
    
33.     MaxIt=50;
    
34. %     MaxIt=1000;
    
35.     for i=1:nPop
    
36.         pop(i).Position=unifrnd(VarMin,VarMax,VarSize);
    
37.         pop(i).Cost=CostFunction(pop(i).Position',t,T_parameter,group);
    
38.     end

复制代码

然后在 platemo 上是这样的：

1. classdef NSGAII_HEL < ALGORITHM
   
2. % <multi> <real/integer/label/binary/permutation> <constrained/none><dynamic>
   
3. % Nondominated sorting genetic algorithm II
   
4. 
   
5. %------------------------------- Reference --------------------------------
   
6. % K. Deb, A. Pratap, S. Agarwal, and T. Meyarivan, A fast and elitist
   
7. % multiobjective genetic algorithm: NSGA-II, IEEE Transactions on
   
8. % Evolutionary Computation, 2002, 6(2): 182-197.
   
9. %------------------------------- Copyright --------------------------------
   
10. % Copyright (c) 2023 BIMK Group. You are free to use the PlatEMO for
    
11. % research purposes. All publications which use this platform or any code
    
12. % in the platform should acknowledge the use of "PlatEMO" and reference "Ye
    
13. % Tian, Ran Cheng, Xingyi Zhang, and Yaochu Jin, PlatEMO: A MATLAB platform
    
14. % for evolutionary multi-objective optimization [educational forum], IEEE
    
15. % Computational Intelligence Magazine, 2017, 12(4): 73-87".
    
16. %--------------------------------------------------------------------------
    
17. 
    
18.     methods
    
19.         function main(Algorithm,Problem)
    
20.             %% Generate random population
    
21. %     Population = Global.Initialization();
    
22. CostFunction=Problem.FObj;  % Cost Function
    
23. nVar=size(Problem.XLow,1);             % Number of Decision Variables
    
24. VarSize=[nVar 1];   % Decision Variables Matrix Size
    
25. detect_py=0;
    
26. VarMin=Problem.XLow;
    
27. VarMax=Problem.XUpp;
    
28. T=size(hisPareto,2)+1;
    
29.    
    
30. nObj=Problem.NObj;
    
31. nPop=popSize;    % Population Size (Number of Sub-Problems)
    
32. nArchive=100;
    
33. Lower = repmat(VarMin,1,nPop);
    
34. Upper = repmat(VarMax,1,nPop);
    
35. crossover_params.gamma=0.5;
    
36. crossover_params.VarMin=VarMin;
    
37. crossover_params.VarMax=VarMax;
    
38. empty_individual.Position=[];
    
39. empty_individual.Cost=[];
    
40. pop=repmat(empty_individual,nPop,1);
    
41. Off=repmat(empty_individual,nPop,1);
    
42. Offspring=repmat(empty_individual,nPop,1);
    
43. gailv=0.5;
    
44. gailv2=0.1;
    
45. zhixing1=1;
    
46. zhixing2=1;
    
47. traindata=[];
    
48. testdata=[];
    
49. if nargin == 6
    
50.     MaxIt=50;
    
51. %     MaxIt=1000;
    
52.     for i=1:nPop
    
53.         pop(i).Position=unifrnd(VarMin,VarMax,VarSize);
    
54.         pop(i).Cost=CostFunction(pop(i).Position',t,T_parameter,group);
    
55.     end
    
56. elseif nargin > 6
    
57.     if size(init_pop,1)< popSize
    
58.         newpop = addNoise(init_pop, popSize, size(init_pop,2));
    
59.         init_pop = [init_pop' newpop'];
    
60.     else
    
61.         init_pop = init_pop';
    
62.     end
    
63.     init_pop  = max(min(init_pop,Upper),Lower);
    
64.     for i=1:size(init_pop,2)
    
65.         pop(i).Position=init_pop(:,i);
    
66.         pop(i).Cost=CostFunction(pop(i).Position',t,T_parameter,group);
    
67.     end
    
68. end

复制代码

这可咋办

补充内容 (2023-11-4 17:58):
报错图片没出来，这里补上
未识别类 'FDA1' 的方法、属性或字段 'FObj'。

出错 NSGAII_HEL/main (第 22 行)
CostFunction=Problem.FObj;  % Cost Function

出错 ALGORITHM/Sol...

写成函数，老老实实按参数调用不就好了

windyer 发表于 2023-11-4 18:14
写成函数，老老实实按参数调用不就好了

不是啊，platemo要写成类

周鱼佬 发表于 2023-11-4 18:44
不是啊，platemo要写成类

这就不是matlab问题，而是platemo这个包的问题，你应该去该项目的github问问，或者对应群。

windyer 发表于 2023-11-4 22:52
这就不是matlab问题，而是platemo这个包的问题，你应该去该项目的github问问，或者对应群。 ...

对 我不会用platemo框架