High Performance Computing Applied to Logistic Regression: A CPU and GPU Implementation Comparison

We present a versatile GPU-based parallel version of Logistic Regression (LR), aiming to address the increasing demand for faster algorithms in binary classification due to large data sets. Our implementation is a direct translation of the parallel Gradient Descent Logistic Regression algorithm proposed by X. Zou et al. [12]. Our experiments demonstrate that our GPU-based LR outperforms existing CPU-based implementations in terms of execution time while maintaining comparable f1 score. The significant acceleration of processing large datasets makes our method particularly advantageous for real-time prediction applications like image recognition, spam detection, and fraud detection. Our algorithm is implemented in a ready-to-use Python library available at : https://github.com/NechbaMohammed/SwiftLogisticReg.

I. Introduction

The advent of machine learning algorithms has brought about significant changes in the field of data analysis and has become a critical component of many industries. One particular application of machine learning is binary classification, which has gained widespread use in various fields, such as image recognition, spam detection, and fraud detection. Logistic regression (LR) is one of the most widely employed algorithms for binary classification and has proven to be highly effective in real-world scenarios. However, the exponential increase in dataset sizes has led to a demand for faster algorithms to process this data.