Fig. 4From: Investigation of model stacking for drug sensitivity predictionResiduals and best fit line for vertical stacking (left) and horizontal stacking (right) on synthetic data with 1200 training samples and 50 testing samplesBack to article page