I did a search and have read that the test time =
total time - training
time. Is this equation correct?
That wasn't clear from your earlier email, it suggested that you
wanted to infer from training time how long predictions will take.
The above holds true, if you do a combined train and test. Different
with cross-validation, of course.
Like Eibe said, use the Experimenter and choose the appropriate
statistic. As far as I remember, you can select train and test times
I am sorry if I didnt get you well. I am looking for
an efficient fast
classification model. Let's say, that I want to know the efficiency of
support vector machine algorithm. How to do that in weka?
How do you define efficient? Memory? Speed? Model size? You will have
to do experiments, unfortunately, as it will depend on your data.
Dept. of Computer Science
University of Waikato, NZ
+64 (7) 858-5174