Scikit-learn compatible library for molecular fingerprints and chemoinformatics.
Molecular fingerprints with scikit-learn compatible interface, built on RDKit. Also includes molecular filters, distances and similarities, applicability domain checks, and much more. Easy, efficient, and feature-rich tool for ML in chemoinformatics.
Software suite for Bayesian statistical tests for fair model comparison.
Easy to use library providing Bayesian statistical tests, providing interpretable probabilities instead of raw p-values. Purpose-built for benchmarking and fair comparison of ML models over many datasets. Based on PyMC, with visualization utilities.
Large-scale, diverse dataset for pretraining molecular models.
Standardized, high-quality small molecules dataset, built from public chemical databases. Ready for training foundation models for molecular representation learning.
Benchmark dataset of small molecule toxicity to honey bees.
Curated dataset of pesticide toxicity data to honey bees. Predefined train-test splits, ready for benchmarking QSAR/QSPR models.