This repository contains the code, documentation, and datasets for a comprehensive exploration of machine learning techniques to address class imbalance. The project investigates the impact of various methods, like ADASYN, KMeansSMOTE, and Deep Learning Generator, on classification performance while effectively demonstrating benefits of pipelining.
deep-learning pipeline cross-validation eda feature-selection classification data-preprocessing class-imbalance data-cleaning hyperparameter-tuning data-pipeline oversampling synthetic-data adasyn kmeanssmote
-
Updated
Jul 15, 2024 - Jupyter Notebook