Ticket #223 (closed bug: fixed)

Opened 6 years ago

Last modified 6 years ago

Exception: 'BayesLearner': threshold can only be optimized for binary classes

Reported by: gregor Owned by: janez
Milestone: 1.0 Component: orng_modules
Severity: minor Keywords:
Cc: Blocking:
Blocked By:

Description

Janez, a mogoce ti ves kako popravit ta bug:

File: OWBaseWidget.py in line 827

Function name: randomlyChangeSettings Code: callback()

File: OWTestLearners.py in line 115 Function name: <lambda> Code: callback=lambda p=2: self.conditionalRecompute(p))

File: OWTestLearners.py in line 445 Function name: conditionalRecompute Code: self.recompute(False)

File: OWTestLearners.py in line 436 Function name: recompute Code: self.score([l.id for l in self.learners.values()])

File: OWTestLearners.py in line 255 Function name: score Code: times=self.pRepeat, callback=pb.advance, storeExamples = True)

File: orngTest.py in line 171 Function name: proportionTest Code: learnAndTestOnTestData(learners, (learnset, weight), (testset, weight), testResults, time, pps, argkw)

File: orngTest.py in line 436 Function name: learnAndTestOnTestData Code: classifiers = [learner(learnset, learnweight) for learner in learners] Exception type: <class 'orange.KernelException'> Exception value: 'BayesLearner': threshold can only be optimized for binary classes

dobis ga z "debugWidgets.py simple-evaluate and bayes.pyw"

Change History

comment:1 Changed 6 years ago by janez

  • Status changed from new to closed
  • Resolution set to fixed

Tole je pri meni že dva tedna popravljeno, comital pa sem kaksno uro, preden si poslal report. Sprememba je v C, tako da bos to videl sele jutri.

Note: See TracTickets for help on using tickets.