As title says, I have a shiny app where I introduce datasets, split and preprocess them to then train the models and compare their result/prediction sucess.
For categorical target variables I use a confusion matrix to visually display predicted vs observed results, but when It comes to numeric target variables I got suggested to use RMSE to compare them, so the plan is to plot the observed vs predicted and also add that RMSE to the plot in some way.
For numeric variables I plan to use the LM NNET models if that changes anything.