Opened 10 years ago
Closed 10 years ago
#2245 closed enhancement (done)
Shuffle manipulation treats the training and test partition separately
Reported by: | mkommend | Owned by: | mkommend |
---|---|---|---|
Priority: | medium | Milestone: | HeuristicLab 3.3.11 |
Component: | DataPreprocessing | Version: | 3.3.10 |
Keywords: | Cc: |
Description
The data preprocessing view provides the functionality to shuffle the data. While doing so, the defined training and test partition are shuffled separately, which in most cases has no effect while modeling the data. Therefore, the user should be able to specify if the partitions should be shuffled separately or the whole data should be shuffled, where the later should be the default option.
Change History (10)
comment:1 Changed 10 years ago by bburlacu
comment:2 Changed 10 years ago by bburlacu
r11381: Forgot to commit changes to ManipulationView.cs.
comment:3 Changed 10 years ago by bburlacu
- Status changed from new to accepted
comment:4 Changed 10 years ago by bburlacu
- Owner changed from bburlacu to gkronber
- Status changed from accepted to reviewing
comment:5 Changed 10 years ago by gkronber
- Owner changed from gkronber to bburlacu
- Status changed from reviewing to assigned
Reviewed r11380.
I'm not happy with the methods in interface IManipulationLogic.
void ShuffleWithRanges(bool shuffleRangesSeparately); void ShuffleWithRanges(IEnumerable<HeuristicLab.Data.IntRange> ranges); void ShuffleWithRanges(IEnumerable<HeuristicLab.Data.IntRange> ranges, bool shuffleRangesSeparately);
Instead I think the following would be better:
void Shuffle(); void ShuffleWithRanges(IEnumerable<HeuristicLab.Data.IntRange> ranges);
If reasonable, please adapt the implementation accordingly.
comment:6 Changed 10 years ago by bburlacu
r11403: The Shuffle method needs a bool parameter to be passed from the view (checkbox whether to shuffle with ranges or not). I combined the three methods into just one method void Shuffle(bool shuffleRangesSeparately).
comment:7 Changed 10 years ago by bburlacu
- Owner changed from bburlacu to mkommend
- Status changed from assigned to reviewing
comment:8 Changed 10 years ago by mkommend
- Status changed from reviewing to readytorelease
comment:9 Changed 10 years ago by mkommend
Reviewed r11403.
comment:10 Changed 10 years ago by mkommend
- Resolution set to done
- Status changed from readytorelease to closed
r11380: Added a checkbox in the properties section of the shuffle data tab in the ManipulationView to specify whether the problem data should be shuffled as a whole or separately for each training/test range. Modified ShuffleToIndices methods accordingly. Also fixed very small typo in ManipulationContent.cs