Check for missing argument or incorrect argument data type in call to function 'predict'.

I created my model in two different ways first using fitlm and then fitrlinear. With fitlm everything is okay, but with fitrlinear, when in the last row I use predict(mdl2,Xte), it returns me the error.

I checked all the related documentation but I don' understand why it is wrong.

% Xtr,Ytr = training | Xte,Yte = testing 
mdl1 = fitlm(Xtr,Ytr);
mdl2 = fitrlinear(Xtr,Ytr,'KFold',5,'Learner','leastsquares','Regularization','lasso');
ypred1 = predict(mdl1,Xte);
ypred2 = predict(mdl2,Xte);

Answer 1

[dpb]

I've never used fitrlinear (didn't know it existed, in fact), but it appears that it returns a composite object and that the required RegressionLinear object is contained in the .Trained property of the object. If I try your model from just some random data, I see

>> mdl2
mdl2 = 
  RegressionPartitionedLinear
    CrossValidatedModel: 'Linear'
           ResponseName: 'Y'
        NumObservations: 10000
                  KFold: 5
              Partition: [1×1 cvpartition]
      ResponseTransform: 'none'

  Properties, Methods

Properties for class classreg.learning.partition.RegressionPartitionedLinear:
    ResponseTransform
    CrossValidatedModel
    NumObservations
    Y
    W
    PredictorNames
    CategoricalPredictors
    ResponseName
    Trained
    KFold
    Partition
    ModelParameters
>> mdl2.Trained
ans =
  5×1 cell array
    {1×1 RegressionLinear}
    {1×1 RegressionLinear}
    {1×1 RegressionLinear}
    {1×1 RegressionLinear}
    {1×1 RegressionLinear}
>> mdl2.Trained{1}
ans = 
  RegressionLinear
         ResponseName: 'Y'
    ResponseTransform: 'none'
                 Beta: -0.0320
                 Bias: -0.0072
               Lambda: 1.2500e-04
              Learner: 'leastsquares'

  Properties, Methods

>> 

So, because of the 'Kfold,5' factor, there are five linear models; you can evaluate any one of them by

yhat=predict(mdl2.Trained{1},Xtr);

I've not looked into what the 'Kfold,5' option does, either, so I don't know what that really means to have five returned models; I guess you will. :)

I agree the documentation could be a lot more informative about the content and using it it seems...I dug this out from just copying what an example had done by referencing a property; not by description of the functions.

[Giacomo Cavalca]

Thank you very much