Orange Forum • View topic - Orange Newbie question

Orange Newbie question

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

Orange Newbie question

Postby wagnerfelix » Mon Mar 04, 2013 17:47

Hello,

I'm new to data mininig and orange. I want to load a dataset from a sqlite/mysql/postgres database.
So far so good. I followed the instructions:

Code: Select all
        r = SQLReader('sqlite://dcmt_db/')
        r.connect()
        r.execute("SELECT * FROM devices_measurement;")
        d = r.data()


Now i got my data. However if I want to create a classifier:
Code: Select all
classifier = orange.BayesLearner(d)

I get orange.KernelException: 'orange.BayesLearner': class-less domain. This is obvious since:
Code: Select all
...
for x in d.domain.features:
     print x
...
Orange.feature.String 'id'
Orange.feature.String 'subdevice_id'
Orange.feature.String 'value'
Orange.feature.String 'begin'
Orange.feature.String 'end'

there is no class domain. I have to admit that I'm confused about the domains. so lets say I want to analyse 'value' then value should be a Class. How can I do that?

Re: Orange Newbie question

Postby Ales » Tue Mar 05, 2013 17:58

See this thread for some tips


Return to Questions & Support



cron