How do I get the confusion matrix for all trained data

How do I get the confusion matrix for all trained data at one set instead of finding a confusion matrix for each set ( training, validation, test) seperatly?

Answer 1

[Jon Cherrie]

You can use the confusionmat command

• https://au.mathworks.com/help/stats/confusionmat.html

When you call it, make sure that you pass the data from all sets that you want to combine.

For example, if you're using patternnet to train a network,

[x,t] = iris_dataset;
net = patternnet(10);
net = train(net,x,t);
y = net(x);

% Decode the predictions from "one hot" to classes
species = ["Setosa" "Versicolour" "Virginica"];
t = onehotdecode(t,species,1,"categorical");
y = onehotdecode(y,species,1,"categorical");

% Compute the confusion matrix and display -- note that 't' and 'y' contain
% results from all three sets
cm = confusionmat(t,y)
confusionchart(cm)

I would be careful about combining train, valiadation and test sets in this way as it might lead to misleading results.

Answer 2

[Image Analyst]

It should be pretty straightforward. Just determine the predicted class, and the actual class, where each class is identified by an integer. Then just increment the confusion matrix there

trueClass = 5; % Whatever
estimatedClass = 4; % For example
% Increment the count.
confusionMatrix(trueClass, estimatedClass) = confusionMatrix(trueClass, estimatedClass) + 1;

Note : you cannot determine a confusion matrix for test data since you don't know the true class - you only know the estimated class. You only know the true class for the training data and validation data, not the test data.