In my original data set I see a distribution of 70% belongs to label A and 30% belongs to label B. For my train, validation and test set I maintained the same ratio. However, I wonder whether this is correct.
If 70% belongs to label A, is the probability that I classify my sample to label A then also higher than a classification to label B? Or should the model have a 50-50 (label A- label B) distribution so that model looks more at the selected features than at the distribution in the model? Or is that not relevant?
Thanks in advance.