Aug 21, 2014
Gary Robinson
agent
wrote
What settings have the most impact on run times? I have already seen that if I cap the number of predictors from the default of 50 to 12 this has a big impact. I would assume looking for higher order terms is another one.
1 Answer
Aug 21, 2014
Gary Robinson
agent
wrote
Checking for correlations shortens time generally for large datasets. Smaller number predictors as you say. Increasing the model complexity penalty can shorten times. Not doing binning of numeric shortens times. The tradeoff with everything of course is typically that shorter time = more likelihood of worse model. That's why a good position is to just let the model do the iterative work, even if it takes a little longer.