Improving representation learning for hit enrichment in low-data drug discovery

This is a processed dataset for the paper "Improving representation learning for hit enrichment in low-data drug discovery". It contains ECFP-6, MACCS, ACSF, ChemBERTa, ChemGPT, and graph representations, as well as PCA-reduced versions of ECFP+MACCS of ALDH1, PKM2 and VDR datasets stored in pytorch tensor formats. To use it, replace the data directory in your project with unzipped data.tar.gz file.More information can be found on GitHub: https://github.com/SokolLonely/Developing_efficient_active_learning