How Subset Features works
The most rigorous way to assess the quality of an output surface is to compare the predicted values with those measured in the field. It is not always possible to go back to the study area to collect an independent validation dataset. One solution is to divide the original dataset into two parts. One part can be used to model the spatial structure and produce a surface. The other part can be used to compare and validate the output surface.
The Subset Features tool divides the data into two subsets. Subset one will have L features, and subset two will have N - L features (with N being the amount of features in the original dataset). The features are divided by generating random values from a uniform [0,1] distribution. If the random value is less than L/N, the feature is assigned to the first subset. If not, the feature is assigned to the second subset.