Let us consider a binary classification problem with = 1000 observations, consisting of two classes {orange, blue}. We know that blue samples form more than half of the training data. In Experiment 1, we run KNN with = , and in Experiment 2, we run KNN with = 1. The average of training errors (where training errors, in this case, are the number of misclassified observations) of Experiments 1 & 2 is 20%. Do we have enough information to calculate the number of training observations in class orange? If yes, calculate the number. If No, write the additional assumptions we need to calculate the number