Hello,
I am on diagnosing bias and variance and using it to tune the model by selecting an nth order polinomial. If you had multiple features, how do you know which one needs to be adjusted? Do you train one feature at a time?
TIA,
Reena
All features are adjusted. You don’t need to pick which one to adjust.
The weights will be learned which minimize the cost. If a new polynomial term isn’t needed, its weight will be very small, so that feature will be mathematically ignored.
1 Like
tx