Cross Validation Improvements in TMVA 

TMVA is a machine learning framework primarily targeted at enabling physics research and is distributed as part of ROOT. This project sets out to improve the validation tools, and cross validation of TMVA. Cross validation is an important tool for making the most out of limited data sets trading increased data efficiency for decreased computation efficiency. This is relevant to HEP applications since experiments are limited by the prohibitive cost of running detector and reconstruction simulation. Tasks include: implement and evaluate the performance of different cross validation splitting functions; extend the current implementation to seamlessly handle nested cross validation; evaluate and compare the performance of the TMVA implementation to industry standard libraries using physics datasets; parallelise training and evaluation.