imbbag 1.2

Package contains a collection of bagging ensemble algorithms for imbalanced data

ImbBag

Imbalanced Bagging Ensemble Algorithms

Description

ImbBag is a specialized package that integrates a variety of bagging ensemble methods specifically designed for imbalanced data classification. This package provides a scikit-learn-based framework that simplifies the usage of these methods, making it easier for researchers and practitioners to apply them in their work, whether dealing with binary or multi-class classification problems.

Installation

pip install imbbag

Requirements

The following Python packages are required.

• scikit-learn
• imblearn
• PyGAD
• ARFS
• mlxtend
• scikit-learn-intelex
• ARFS

Also, use Python 3.11

Available Bagging Ensemble Algorithms in the ImbBag Package

• UnderBagging (UnderBag) - ** Multi-class
• Exactly Balanced Bagging (EBBag) - ** Binary-class
• OverBagging (OverBag) - ** Multi-class
• SMOTE Bagging (SMOTEBag) - ** Multi-class
• Roughly Balanced Bagging (RBBag) - ** Binary-class
• Multi-class Roughly Balanced Bagging (MRBBag) - ** Multi-class
• Bagging Ensemble Variation (BEV) - ** Binary-class
• Lazy Bagging (LazyBag) - ** Multi-class
• Multi Random Balance Bagging (MultiRandBalBag) - ** Multi-class
• Neighborhood Balanced Bagging (NBBag) - ** Binary-class
• Probability Threshold Bagging (PTBag) - ** Multi-class
• Adaptive Synthetic Bagging (ADASYNBag) - ** Binary-class
• RSYN Bagging (RSYNBag) - ** Binary-class
• Resampling Ensemble Algorithm (REABag) - ** Multi-class
• Under-bagging K-NN (UnderBagKNN) - ** Multi-class
• Boundary Bagging (BBag) - ** Multi-class
• Bagging of Extrapolation-SMOTE SVM (BEBS) - ** Binary-class
• Evolutionary Under-sampling based Bagging (EUSBag)- ** Binary-class
• Random Balanced Sampling with Bagging (RBSBag) - ** Multi-class
• Cost-sensitive Bagging (CostBag) - ** Multi-class

Get Started

Here is an example:

from sklearn.tree import DecisionTreeClassifier
from sklearn. model_selection import train_test_split
from ImbBag import BBag

dataframe = read_csv('dataset.csv')
data = dataframe.values    
X = data[:,:-1]
Y = data[:,:-1]

# split the dataset into training and test sets
X_train ,X_test ,y_train ,y_test = train_test_split (X, y, test_size =0.2)

# instantiate the imbalance bagging classifier, training, prediction 
cls = BBag(estimator = DecisionTreeClassifier(), n_estimator = 50)
clf.fit(X_train , y_train)
y_pred = clf.predict(X_test)

Version History

• v1.0.0 - Initial release - July 29, 2024

Credits

• **Yousef Abdi
• University of Tabriz

License

This project licensed under the MIT License.

Support

Report issues, ask questions, and provide suggestions using:

• GitHub Issues
• GitHub Discussions
• Email: y.abdi [at] tabrizu [dot] ac [dot] ir

The project can be accessed at https://github.com/yousefabdi/imbbag