Lets say we have a training data and we have estimated a fit for a model of square ft of living area vs price of houses. Suppose we know the probability distribution of sq ft and for a fixed sq ft we know the probability distribution of its price. Now, we know the joint probability for each sq ft and price pair. How do I calculate the generalization error(both total and for a particular sq ft)??
$\begingroup$
$\endgroup$
1
-
$\begingroup$ No, MSE is just the mean loss. We need a loss function which in this case in squared function. But generalization error is something else. Its defined as follows $\endgroup$– KAY_YAKJan 26, 2017 at 15:48
Add a comment
|