The impact of K selection in K‑fold cross-validation on bias and variance in supervised learning models. — SciRadar