Append results of two 'for loops' into a single variable

2 次查看（过去 30 天）
Jackson Rodrigues
Jackson Rodrigues 2022-4-26
编辑： Davide Masiello 2022-4-26
I'm running an SVM ML template from MATLAB and I need to note down the accuracies for each parameter in 10 runs (with different training and test sets (which i already have prepared). So, I have put this in a 'for loop' which runs for 10 times and saves the accuracies each time (10 accuracies) into a variable named 'a1'. Now, I have to repeat this procedure for another parameter using another 'for loop' and save the results in the same variable 'a1' without rewriting the results obtained from previous loop. Ultimately I want all the accuracies obtained from all the loops/different loops in a single variable. Kindly help me with this.
Thank you in advance.
itr_3 = 1:10;
%%% 1 feature
FeaturesSelected = horzcat(feature1,fs_Y);
colNames = {'Feature1','Label'};
FeaturesSelected_Table = array2table(FeaturesSelected,'VariableNames',colNames);
for i=1:length(itr_3)
X_Train = X(FeaturesSelected_Table(1:80,:));
Y_Train = Y(FeaturesSelected_Table(1:80,:));
X_Test = X(FeaturesSelected_Table(81:100,:));
Y_Test = Y(FeaturesSelected_Table(81:100,:));
t = templateSVM('Standardize',true,'KernelFunction','rbf','SaveSupportVectors',true);
Mdl = fitcecoc(X_Train,Y_Train,'Learners',t,'FitPosterior',true,'Coding','onevsall', 'Verbose',2)
[predictClass, scoresSVM]=predict(Mdl,X_Test);
disp('class predict')
disp([Y_Test predictClass])
result = [Y_Test predictClass];
Accuracy=mean(Y_Test==predictClass)*100;
fprintf('\nAccuracy =%d\n',Accuracy)
a{i}(:,1)=Accuracy;
end
%%% 2 feature
FeaturesSelected = horzcat(feature1,feature2,fs_Y);
colNames = {'Feature1','Feature2','Label'};
FeaturesSelected_Table = array2table(FeaturesSelected,'VariableNames',colNames);
for i=1:length(itr_3)
X_Train = X(FeaturesSelected_Table(1:80,:));
Y_Train = Y(FeaturesSelected_Table(1:80,:));
X_Test = X(FeaturesSelected_Table(81:100,:));
Y_Test = Y(FeaturesSelected_Table(81:100,:));
t = templateSVM('Standardize',true,'KernelFunction','rbf','SaveSupportVectors',true);
Mdl = fitcecoc(X_Train,Y_Train,'Learners',t,'FitPosterior',true,'Coding','onevsall', 'Verbose',2)
[predictClass, scoresSVM]=predict(Mdl,X_Test);
disp('class predict')
disp([Y_Test predictClass])
result = [Y_Test predictClass];
Accuracy=mean(Y_Test==predictClass)*100;
fprintf('\nAccuracy =%d\n',Accuracy)
a{i}(:,1)=Accuracy;
end

请先登录，再回答此问题。

回答（1 个）

Davide Masiello
Davide Masiello 2022-4-26
编辑：Davide Masiello 2022-4-26
You could store the accuracies of the first loop into a variable named a1 and those of the second loop into a variables named a2.
Then, after the loops, you can concatenate the two arrays.
% first loop
for i = 1:length(itr_3)
...
...
a1(i) = Accuracy;
end
% second loop
for i = 1:length(itr_3)
...
...
a2(i) = Accuracy;
end
a = [a1,a2];
Please note that I have changed the indexing so that the accuracies are stored as arrays of doubles and not cell arrays.

类别

在 Help Center 和 File Exchange 中查找有关 Loops and Conditional Statements 的更多信息

标签

产品


版本

R2022a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by