The "covertype" data (ub learns) was obtained from the US Forest Service and

are concerned with seven different types of forest cover. The data can be found

on UB learns. There are 581,012 observations (each a 30x30 meter cell) on 54

input variables (10 quantitative variables, 4 binary wilderness areas, and 40 binary

soil type variables). Divide these data randomly into a training set and a test set.

Use any of the methods we discussed to develop a model that predicts the forest

cover type. Report the training and test error rates.