I have a problem with the ridge regression based on the function
lfridge <- LMTrainer$new(family = 'gaussian', alpha=0)
When I run the ridge regression and use the function
lfridge$get_importance()
I well obtain the estimated coefficients but the least important variables have coefficient equal to zero whereas they should have a value close to zero with the ridge regression.
First, Thanks for this interesting package!
I have a problem with the ridge regression based on the function lfridge <- LMTrainer$new(family = 'gaussian', alpha=0)
When I run the ridge regression and use the function lfridge$get_importance() I well obtain the estimated coefficients but the least important variables have coefficient equal to zero whereas they should have a value close to zero with the ridge regression.