Database for multi-tier machine learning optimization of organic photovoltaic photoactive layer fabrication

This dataset provides a standardized, multi-tiered resource for optimizing organic photovoltaic devices through data-driven methods. It integrates donor/acceptor molecular structures (with SMILES, CDK fingerprints, and electronic descriptors) with nine key photoactive layer processing parameters and device efficiency.The data is uniquely structured in three tiers to guide optimization: from single-parameter analysis, through stage-combined synergies (solution preparation, spin-coating, post-processing), to global optimization with all parameters. This structure facilitates the development and validation of both interpretable and high-performance machine learning models.For complete details on the dataset structure, file descriptions, and exact data dictionary, please refer to the README.txt file included in the download.