Benchmarking Deep Learning Models for Raman Spectroscopy Across Open-Source Datasets

TL;DR

This paper systematically benchmarks multiple deep learning models for Raman spectroscopy classification across open-source datasets, providing a fair comparison of their performance and highlighting the current state of deep learning in this domain.

Contribution

It offers one of the first comprehensive benchmarks of Raman-specific deep learning classifiers on multiple datasets using a unified evaluation protocol.

Findings

Deep learning models outperform classical methods in Raman classification.

Performance varies significantly across different architectures and datasets.

The study establishes a reproducible evaluation framework for future research.

Abstract

Deep learning classifiers for Raman spectroscopy are increasingly reported to outperform classical chemometric approaches. However their evaluations are often conducted in isolation or compared against traditional machine learning methods or trivially adapted vision-based architectures that were not originally proposed for Raman spectroscopy. As a result, direct comparisons between existing deep learning models developed specifically for Raman spectral analysis on shared open-source datasets remain scarce. To the best of our knowledge, this study presents one of the first systematic benchmarks comparing three or more published Raman-specific deep learning classifiers across multiple open-source Raman datasets. We evaluate five representative deep learning architectures under a unified training and hyperparameter tuning protocol across three open-source Raman datasets selected to support standard evaluation, fine-tuning, and explicit distribution-shift testing. We report classification accuracies and macro-averaged F1 scores to provide a fair and reproducible comparison of deep learning models for Raman spectra based classification.