how to test the data after trained in classification learner

i have data in xlsx. file with 35 raws and 6 columns.. i already trained the data in classification learner and export the model to work space but i do not know how to test it with new data

any one can guiding me please ?

[shamsah alotibe]

i wrot this command in command window :

yfit = C.predictFcn(T)

where T a new data and C the model name but there are some errors that said :

Unable to use a value of type 'cell' as an index.

Error in mlearnapp.internal.model.DatasetSpecification>@(t)t(:,predictorNames) (line 163)

extractPredictorsFromTableFcn = @(t) t(:,predictorNames);

Error in

mlearnapp.internal.model.DatasetSpecification>@(x)extractPredictorsFromTableFcn(splitMatricesInTableFcn(x))

(line 164)

predictorExtractionFcn = @(x)

extractPredictorsFromTableFcn(splitMatricesInTableFcn(x));

Error in

mlearnapp.internal.model.DatasetSpecification>@(x)exportableModel.predictFcn(predictorExtractionFcn(x))

(line 167)

newExportableModel.predictFcn = @(x)

exportableModel.predictFcn(predictorExtractionFcn(x));

Answer 1

[Shivam Sardana]

The error message in your comment indicates that something is trying to be indexed using a cell array as a subscript, which is a valid indexing method for tables but not an indexing method for a double array.

I may not know about your test dataset. There is a similar question which may be of relevance to you: https://www.mathworks.com/matlabcentral/answers/264459-failed-to-call-classification-learner-s-testing-function

[shamsah alotibe]

Thanks Shivam, it is work.

but how i can get the accuracy percentage ?

the output of this function is a matrix of the result of each columns.

[Shivam Sardana]

You have got the predicted values. For test dataset, you have actual values(ground truth). You can calculate RMSE to get acuracy.

[shamsah alotibe]

sorry, I did not understand, I am new in matlab

can you clarify or post some useful linkes please ?

[anita jasmine]

To test the model,MD Predictedlabels=predict(MD,testfeaturesmatrix) Accuracy=mean (Predictedlabels==testlabels)

[Hussein Hasan Mohsen]

It's not working Can you provide us with any link

Answer 2

[Fatma HM]

Hi @Shivam Sardana I need to estimate the test data I have trained, with (ANN,NARX and Ridge regression). Do you have any idea?

Answer 3

[Arnab Banerjee]

@shamsah alotibe kindly check one weather your TestData heading are same as TrainingData or not.. if same then hopefully you will not get this error message, because i have just now resolved this issue, by keeping the coloum name same.