The correct answer is: None of the mentioned.
RMSE, RSquared, and Accuracy are all measures of the goodness of fit of a model. They are all continuous measures, which means that they can take on any value on a continuum. Categorical outcomes, on the other hand, are discrete measures that can only take on a limited number of values. For example, the outcome of a coin toss is a categorical outcome, because it can only be heads or tails.
RMSE is the root mean squared error, which is a measure of the average squared difference between the predicted values and the actual values. RSquared is the coefficient of determination, which is a measure of how well the model fits the data. Accuracy is the proportion of the data that is correctly predicted by the model.
All three of these measures are useful for evaluating the performance of a model. However, they are not categorical outcomes.