https://github.com/gugarosa/mh_feature_selection
📄 Official implementation regarding the paper "A Survey on Metaheuristic Approaches to Feature Selection".
https://github.com/gugarosa/mh_feature_selection
feature-selection implementation machine-learning meta-heuristic optimization paper
Last synced: 7 months ago
JSON representation
📄 Official implementation regarding the paper "A Survey on Metaheuristic Approaches to Feature Selection".
- Host: GitHub
- URL: https://github.com/gugarosa/mh_feature_selection
- Owner: gugarosa
- License: gpl-3.0
- Created: 2020-06-16T21:18:17.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2022-12-22T13:15:34.000Z (almost 3 years ago)
- Last Synced: 2024-10-18T07:40:02.581Z (12 months ago)
- Topics: feature-selection, implementation, machine-learning, meta-heuristic, optimization, paper
- Language: Python
- Homepage:
- Size: 85 KB
- Stars: 1
- Watchers: 6
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: history/.gitignore
- License: LICENSE
Awesome Lists containing this project
README
# A Survey on Metaheuristic Approaches to Feature Selection
*This repository holds all the necessary code to run the very-same experiments described in the paper "A Survey on Metaheuristic Approaches to Feature Selection".*
---
## References
If you use our work to fulfill any of your needs, please cite us:
```
```
---
## Structure
* `data`: Folder containing the OPF file format datasets;
* `history`: Folder for saving the output files, such as `.pkl` and `.txt`;
* `models`
* `classifiers.py`: Defines the possible classifiers that can be used;
* `heuristics.py`: Defines the possible meta-heuristics that can be used;
* `transfers.py`: Defines the possible transfer functions that can be used;
* `utils`
* `loader.py`: Loads OPF file format datasets;
* `optimizer.py`: Wraps the optimization task into a single method;
* `outputter.py`: Converts the optimization history into readable output files;
* `targets.py`: Implements the objective functions to be optimized.
---
## Package Guidelines
### Installation
Install all the pre-needed requirements using:
```Python
pip install -r requirements.txt
```
### Data configuration
Please [download](https://www.recogna.tech/files/mh_feature_selection/data.tar.gz) the datasets in the OPF file format and then put them on the `data/` folder.
---
## Usage
### Select Features using Meta-Heuristic Optimization
The first step is to select features using a meta-heuristic optimization. To accomplish such a step, one needs to use the following scripts:
```Python
python bool_select_features.py -h
```
or
```Python
python select_features.py -h
```
*Note that `-h` invokes the script helper, which assists users in employing the appropriate parameters.*
### Process Optimization History
After conducting the optimization task, one needs to process its history into readable outputs. Please, use the following scripts to accomplish such a procedure:
```Python
python bool_process_history.py -h
```
or
```Python
python process_history.py -h
```
*Note that this script converts the .pkl optimization history into readable .txt outputs.*
### Classify Selected Features
With the readable outputs in hands, one can now classify the testing set using the selected features:
```Python
python classify_selected_features.py -h
```
### Classify Baseline Features (Optional)
Additionally, it is possible to perform the classification task over the testing set using baseline features (all features), as follows:
```Python
python classify_baseline_features.py -h
```
### Bash Script
Instead of invoking every script to conduct the experiments, it is also possible to use the provided shell scripts, as follows:
```Bash
./bool_pipeline.sh
```
or
```Bash
./pipeline.sh
```
Such a script will conduct every step needed to accomplish the experimentation used throughout this paper. Furthermore, one can change any input argument that is defined in the script.
---
## Support
We know that we do our best, but it is inevitable to acknowledge that we make mistakes. If you ever need to report a bug, report a problem, talk to us, please do so! We will be available at our bests at this repository.
---