Prediction error Vs model complexity graph

JustaLearner

Is there a way in RapidMiner to plot the trade-off between the above two - contrasting training data with test data? Similar to the conceptual chart shown:

MartinLiebig

Hey,

sure. How do you want to define model complexity? It usually depends on the model.

Best,

Martin

JustaLearner

Hi Martin, let's say I'm using Random Forest and only increasing the tree depth in 10 increments from 10 to 100. How would that work, please?