Efficient Sampling-Based Kernel Mean Matching Performing Kernel Mean Matching over subsampled source and target datasets to achieve scalability. More details in the publication at IEEE ICDM 2016. Environment Ubuntu v14.04+ Python v2.7 Execution $ python manager.py <dataset_location>