High Performance Computing Applied to Logistic Regression: A CPU and GPU Implementation Comparison
This incremental work addresses the problem of slow processing for large datasets in real-time prediction applications like image recognition, spam detection, and fraud detection.
The authors tackled the need for faster logistic regression in binary classification by implementing a GPU-based parallel version, which outperformed CPU implementations in execution time while maintaining comparable f1 scores, with significant acceleration for large datasets.
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