Experimental validation result for OASIS software

This dataset comprises 150 reconstructed spectra generated during the high-throughput multimodal experimental validation of the OASIS machine learning framework. The dataset provides a comprehensive record of the automated spectral processing pipeline, encompassing raw spectral inputs alongside their corresponding model outputs for denoising, baseline correction, and peak parameter extraction (including location, intensity, and full width at half maximum, with associated uncertainty quantification). This data is released as a companion to the manuscript, "Universal Autonomous Spectral Analysis for Data-Driven Chemistry and Materials Discovery", to ensure full transparency, facilitate reproducibility, and provide the broader scientific community with robust benchmark data for advanced, data-driven spectral analysis.