This archive contains the datasets and experimental results associated with the publication "Yield Prediction of Organic Reactions in Biased Datasets via Positive-Unlabeled Learning". It provides both the original literature datasets used for benchmarking and the augmented datasets containing "Reliable Negatives" identified by the PAYN framework. Aditionally, this archive contains:- Regression models predictions for Positive only, PAYN augmented and Fully labeled datasets- Average Tanimoto similarities of Spies, Latent Positives and Unlabeled Negatives to the nearest neighbor in the Known Positives set- Wilcoxon test on paired results using the scipy library and random seeds: a) 42, b) 43, c) 44, d) 45, e) 46- Fully labeled models trained on a subset of the training data to match the size of corresponding Augmented or Positives only datasets (Result files containing MAEs, Result_sizes containing actual size of datasets) These files are intended to be used and were generatd with the PAYN software repository:https://github.com/GloriusGroup/PAYN