Mini-Batch Shuffling Strategies for Multi-class Imbalanced Classification

This repository contains the code for mini-batch shuffling strategies for multi-class imbalanced classification. The code provides the following functions:

• random shuffling strategy code
• different strategies for multi-class imbalanced classification

Installation Requirements

The basic requirement for using the files is a Python 3.8.19 environment with PyTorch 2.3.0

Source Files

Here is a brief description of the files and folder content:

• random_shuffling.py: random shuffling for multi-class imbalanced classification
• class_with_imbalance.py: proposed strategy for multi-class imbalanced classification

Running the code

To generate the dataset, run all .py files to implement different strategies for multi-class imbalanced classification.

Developer Team

The code was developed by Yuwei Mao from the CUCIS group at the Electrical and Computer Engineering Department at Northwestern University.

Publication

1. Mao, Yuwei, Vishu Gupta, Kewei Wang, Wei-keng Liao, Alok Choudhary, and Ankit Agrawal. "To Shuffle or Not To Shuffle: Mini-Batch Shuffling Strategies for Multi-class Imbalanced Classification." In 2022 International Conference on Computational Science and Computational Intelligence (CSCI), pp. 298-301. IEEE, 2022. PDF

Disclaimer

The research code shared in this repository is shared without any support or guarantee on its quality. However, please do raise an issue if you find anything wrong and I will try my best to address it.

email: [email protected]

Copyright (C) 2023, Northwestern University.

See COPYRIGHT notice in top-level directory.

Funding Support

This work is supported in part by the following grants: NIST award 70NANB19H005; DOE awards DE-SC0019358, DE-SC0021399; NSF award CMMI-2053929.