Not a valid predictor error when trying to predict

Question

Or Shem Tov 2020-5-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/528108-not-a-valid-predictor-error-when-trying-to-predict

回答： Aditya Patil 2021-2-17

Hi,

I've trained a few models using the classification learner, they work perfectly fine when I export them to the workspace.

But when I'm running them using the function generated by the classification learner, they produce this error after trying to predict:

Error using classreg.learning.internal.table2PredictMatrix>makeXMatrix (line 97)
Table variable Market_Cap is not a valid predictor.
Error in classreg.learning.internal.table2PredictMatrix (line 47)
Xout = makeXMatrix(X,CategoricalPredictors,vrange,pnames);
Error in classreg.learning.classif.CompactClassificationEnsemble/score (line 79)
            X = classreg.learning.internal.table2PredictMatrix(X,[],[],...
Error in classreg.learning.classif.CompactClassificationEnsemble/predict (line 159)
            scores = score(this,X,varargin{:});
Error in trainClassifier4>@(x)predict(classificationEnsemble,x) (line 73)
ensemblePredictFcn = @(x) predict(classificationEnsemble, x);
Error in trainClassifier4>@(x)ensemblePredictFcn(featureSelectionFcn(predictorExtractionFcn(x))) (line 74)
trainedClassifier.predictFcn = @(x) ensemblePredictFcn(featureSelectionFcn(predictorExtractionFcn(x)));
Error in MakePredictions (line 39)
predictions = trainedModel4.predictFcn(T); 

I've had a similar error before but was able to fix it by turning one of the variables to categorical, this isn't the case here, market_cap is the right datatype.

Let me say that I've trained models to be with different features, and I have a code like this at the beginning of the file that trains a model if it doesn't exist in the workspace:

%% Train the models
if ~exist('trainedModel')
    trainingData = readtable('historical_data_tagged.csv');
	trainedModel = trainClassifier(trainingData);
end
if ~exist('trainedModel2')
    trainingData = readtable('historical_data_tagged.csv');
	trainedModel2 = trainClassifier2(trainingData);
end
if ~exist('trainedModel3')
    trainingData = readtable('historical_data_tagged.csv');
	trainedModel3 = trainClassifier3(trainingData);
end
if ~exist('trainedModel4')
    trainingData = readtable('historical_data_tagged.csv');
	trainedModel4 = trainClassifier4(trainingData);
end