how to write a trained fuzzy FIS tree to disk
7 ビュー (過去 30 日間)
古いコメントを表示
Hi all!
I can't write the tree of trained fuzzy inference systems to disk. At the bottom of the program are the options that I used. But they don't work...
Granulator = mamfis('Name','Granulator','NumInputs',2,'NumOutputs',1);
Granulator.Inputs(1).Name = "G_inp1";
Granulator.Inputs(2).Name = "G_inp2";
Granulator.Outputs(1).Name = "G_out";
CTMCM = mamfis('Name','CTMCM','NumInputs',2,'NumOutputs',1);
CTMCM.Inputs(1).Name = "CTMCM_inp1";
CTMCM.Inputs(2).Name = "CTMCM_inp2";
CTMCM.Outputs(1).Name = "CTMCM_out";
OTF = mamfis('Name','OTF','NumInputs',2,'NumOutputs',1);
OTF.Inputs(1).Name = "OTF_inp1";
OTF.Inputs(2).Name = "OTF_inp2";
OTF.Outputs(1).Name = "OTF_out";
con1 = ["Granulator/G_out" "CTMCM/CTMCM_inp1"];
con2 = ["CTMCM/CTMCM_out" "OTF/OTF_inp1"];
tree = fistree([Granulator CTMCM OTF],[con1;con2]);
plotfis(tree) % FIS tree structure
X=rand(1000,5); % imitation of original data
nn=size(X);
nt=round(0.8*nn(1,1)); % training sample size
trnX = X(1:nt,1:4);
trnY = X(1:nt,end);
vldX = X(nt+1:end,1:4);
vldY = X(nt+1:end,end);
options = tunefisOptions;
options.Method = 'particleswarm';
options.OptimizationType = 'learning';
options.NumMaxRules = 10;
options.UseParallel = false;
options.MethodOptions.MaxIterations = 2;
rng('default')
fisout1 = tunefis(tree,[],trnX,trnY,options);
% writeFIS(fisout1,'fis1') % does not work
% writeFIS(tree,'fis1') % does not work
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/1254287/image.png)
0 件のコメント
採用された回答
Govind KM
2024 年 12 月 6 日
Hi Andrey,
Support for saving a FIS tree object to a MAT file using the writeFIS function has been added in MATLAB R2024b. I would recommend upgrading your MATLAB version to R2024b to save FIS tree objects as per your workflow.
その他の回答 (0 件)
参考
カテゴリ
Help Center および File Exchange で Fuzzy Logic Toolbox についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!