1

I have a dataset Auto, whose class label is called mpg01, which is a binary variable. I'm using the following code to train and cross validate the model:

classifier <- IBk(Auto$mpg01 ~ Auto$displacement + Auto$horsepower +
                    Auto$weight + Auto$acceleration + Auto$cylinders
                  + Auto$origin, data = Auto, 
                  control = Weka_control(K = 20, X = TRUE))

evaluate_Weka_classifier(classifier, numFolds = 10)

The Summary, however, only gives me:

Correlation coefficient                  0.8062
Mean absolute error                      0.0969
Root mean squared error                  0.3113
Relative absolute error                 19.3296 %
Root relative squared error             62.0734 %
Total Number of Instances              392

It doesn't print out Correctly Classified Instances or Incorrectly Classified Instances. I wonder what maybe the cause.

All the attributes & class label are of type num. Is it because I didn't convert class label (mpg01) to factor type? Thanks in advance.

J Freebird
  • 3,664
  • 7
  • 46
  • 81

0 Answers0