This is for a Image orientation project. I'm trying to implement adaboost with four perceptrons. The training data has four labels 0, 90, 180, 270. Each of the perceptron identifies one of the labels like 0 or not 0, 90 or not 90 etc. Now the problem arises, for an image with label 180 if the expected out put is [0, 0, 1, 0], and the perceptrons output [0,0,1,1]. In this case three of the perceptrons have given correct output. But one of them is wrong. Shall we identify this one as right or wrong for adaboost. If not, then how do we handle this.