https://github.com/ncordon/feature-selection
Repo de resolución de prácticas de la asignatura Metaheurísticas, UGR
https://github.com/ncordon/feature-selection
Last synced: 4 months ago
JSON representation
Repo de resolución de prácticas de la asignatura Metaheurísticas, UGR
- Host: GitHub
- URL: https://github.com/ncordon/feature-selection
- Owner: ncordon
- License: gpl-3.0
- Created: 2016-03-11T17:18:10.000Z (over 10 years ago)
- Default Branch: master
- Last Pushed: 2016-07-16T03:36:47.000Z (almost 10 years ago)
- Last Synced: 2025-10-10T10:46:58.794Z (8 months ago)
- Language: TeX
- Size: 1.3 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Feature Selection
Contains implementations from scratch of several metaheuristics adjusted for feature selection, and its application to three
different datasets:
- [Wdbc](data/wdbc.arff)
- [Arrhythmia](data/arrhythmia.arff)
- [Movement Libras](data/movement_libras.arff)
## License
GNU licensed
## Files
* `bin` contains the implemented code.
* Simulated Annealing `ES.r`
* Local Search `BL.r`
* Tabu Search: `BT.r`, `BText.r`
* Basic Multirestart `BMB.r`
* Genetic algorithms `AGs.r`
* GRASP `GRASP.r`
* ILS `ILS.r`
* Ant Colony Optimization `OCH.r`
* Memetic algorithms `AGs.r`
* `data` contains datasets, graphs and boxplots.
* `pscode` contains pseudocodes that are compiled to `.tex` doing `make`
* `tex` contains documentation, only available in Spanish:
* `trajectories.tex`
* `multi-restart.tex`
* `genetic.tex`
* `och.tex`
* `memetic.tex`
## Usage
Execute `src/main.r` to get the results
of each algorithm.
For example, to gather the results from local search algorithm applied to all three datasets:
```r
BL.results <- cross.eval(BL)
```
`params.r` contains settings used by the algorithms.
