contrary to decision tree, linear models can only capture linear interaction, so be aware of non-linear relationships in your data.
What does this mean? we do we find out the data set is linear or non-linear?
contrary to decision tree, linear models can only capture linear interaction, so be aware of non-linear relationships in your data.
What does this mean? we do we find out the data set is linear or non-linear?
You could imagine a regression problem where the link between x
and y
is not a linear relationship.
For instance, in the last chapter, we try to model the power of a cyclist from the speed. The relationship is non-linear. The result will be that applying a linear model will result in an underfitted model.
Thanks very much