We have a dataset about bottles of wine, with Wine Type (Red, White, Rose) and measurements of chemical analysis of each wine. Our training set has 900 rows with equal numbers of each type of wine, and our validation set has 500 rows. We run an SVM model. We see that the generated model predicts that all the wine is Red. We can conclude that A. We should have used a Cluster analysis model B. The validation set has only Red wine C. The training data was not balanced D. The SVM kernel cannot distinguish between wine types