Orange Forum • View topic - learning vs. classifying modules

learning vs. classifying modules

A place to ask questions about methods in Orange and how they are used and other general support.

learning vs. classifying modules

Postby re832003 » Wed Jul 18, 2012 15:15

I know that Orange has a learning module for each classifier but does it have a classification module can be used to apply the learned model to new examples?

Re: learning vs. classifying modules

Postby Ales » Thu Jul 19, 2012 12:26

The trained classifiers can be applied to new examples. If you want serialization, simply use python's pickle module.
Code: Select all
import Orange, cPickle
data = Orange.data.Table("train.tab")
c = Orange.classification.bayes.NaiveLearner(data)
cPickle.dump(c, open("bayes.pck", "wb"))
And later in a new process
Code: Select all
import Orange, cPickle
c = cPickle.load(open("bayes.pck", "rb"))
data = Orange.data.Table("new.tab")
c(data[0])


Return to Questions & Support