Sorry, one more question:
Suppose we want to calculate the MSE for a node. We have correct responses Y (so to say the labels) for the examples reaching this node. For MSE we now need to sum the squared differences of the responses Y and our predictions, call them P.
How exactly is P calculated? Do we simply use Least Squares to fit a line?![]()
Suppose we want to calculate the MSE for a node. We have correct responses Y (so to say the labels) for the examples reaching this node. For MSE we now need to sum the squared differences of the responses Y and our predictions, call them P.
How exactly is P calculated? Do we simply use Least Squares to fit a line?