The way the rfinterval method is currently set up, the only way to get predictions for new data is to input them as test_data while training the model. This makes it much more difficult to use in a real-life application where you want to produce new prediction intervals as new data comes in. Is there any solution for this?
The way the
rfinterval
method is currently set up, the only way to get predictions for new data is to input them astest_data
while training the model. This makes it much more difficult to use in a real-life application where you want to produce new prediction intervals as new data comes in. Is there any solution for this?