Opened 7 years ago
Closed 4 years ago
#2806 closed defect (duplicate)
Positive class value setting from classification problems is not set correctly in classification ensemble solutions
Reported by: | gkronber | Owned by: | bburlacu |
---|---|---|---|
Priority: | medium | Milestone: | HeuristicLab 3.3.17 |
Component: | Algorithms.DataAnalysis | Version: | |
Keywords: | Cc: |
Description
Steps to reproduce:
- New RF for classification
- Load UCI problem 'Iris'
- Set positive class value to 'Class 2'
- Load algorithm into cross-validation and run
- The positive class value for the resulting ensemble solution is 'Class 0'
Change History (8)
comment:1 Changed 7 years ago by gkronber
- Milestone changed from HeuristicLab 3.3.15 to HeuristicLab 3.3.16
comment:2 Changed 7 years ago by abeham
- Version 3.3.14 deleted
comment:3 Changed 5 years ago by gkronber
- Milestone changed from HeuristicLab 3.3.16 to HeuristicLab 3.3.17
comment:4 Changed 5 years ago by gkronber
comment:5 Changed 4 years ago by bburlacu
- Owner set to bburlacu
- Status changed from new to accepted
The problem was that the CrossValidation class does not use the regular cloning mechanism to obtain a copy of the problem data, because it also needs to shuffle the dataset. the positive class parameter was missing from the ad-hoc cloning code. Fixed with r17833.
Version 0, edited 4 years ago
by bburlacu
(next)
comment:6 Changed 4 years ago by bburlacu
- Owner changed from bburlacu to gkronber
- Status changed from accepted to reviewing
comment:7 Changed 4 years ago by bburlacu
- Owner changed from gkronber to bburlacu
- Status changed from reviewing to assigned
comment:8 Changed 4 years ago by bburlacu
- Resolution set to duplicate
- Status changed from assigned to closed
Superseeded by #3102
Note: See
TracTickets for help on using
tickets.
Duplicate of #3053?