How to export predictions?

Hello,

I created a predictive model with the classification learner but I can't actually see the predictions. What do I need to do in order to see them and how can I export them to excel?

Thanks

Answer 1

[Ridwan Alam]

On the Classification Learner app, there is "export model" button.

After you have the model exported to your workspace, you can use

yhat = model.predictFcn(x)

where x is your input or test data.

If you want to write the variable yhat to csv, use csvwrite() or any similar function.

A helpful link:

https://www.mathworks.com/help/stats/export-classification-model-for-use-with-new-data.html?s_tid=answers_rc2-1_p4_MLT

[Tiago Afonso]

I managed to get predictions on new data but it shows only a categorical response. The target variable uses 3 categories. How can I turn the predictions into a percentage? For instance, 50% category A, 30% category B, 20% category C.

[Ridwan Alam]

[yhat, yscores] = predict(model, x);
% yscores are the posterior probabilities for each class
% yscores is a matrix of size N-by-K, where N is the number of observations (rows) in x, 
% and K is the number of classes (in model.ClassNames). 
% yscores(i,j) is the posterior probability that row i of x is of class j.

% to convert the probabilities into percentage
ypercentage = 100*yscores;

I believe this answers your question.

[Tiago Afonso]

I tried it and got this message:

'model' requires Simulink.

I'm using a trial version of MatLab by the way.

[Ridwan Alam]

'model' should be replaced by the variable name you assigned to your exported model from the Classification Learner app.

Btw, please accept this as an answer and vote up if you liked to converse. Thanks!

[Tiago Afonso]

This is what I get now:

>> [yhat, yscores] = predict(trainedModel, gggggggggggg)

Undefined function 'predict' for input arguments of type 'table'.

"gggggggggggg" is the name of the new data file.

[Ridwan Alam]

[yhat, yscores] = predict(trainedModel, table2array(gggggggggggg))

[Tiago Afonso]

I get this now:

>> [yhat, yscores] = predict(trainedModel, table2array(gggggggggggg))

Error using table2array (line 37)

Unable to concatenate the specified table variables.

Caused by:

Error using categorical/cat (line 119)

Unable to concatenate a string array and a categorical array.