Added the square to the derivative in the hessian Co-authored-by: Corentin Santos <corentin.santos@iphc.cnrs.fr>
This commit is contained in:
parent
bba6aa74fb
commit
3ef8383d93
@ -52,7 +52,7 @@ If we compute the gradient of said objective function:
|
|||||||
As well as the hessian (the second derivative of the objective):
|
As well as the hessian (the second derivative of the objective):
|
||||||
|
|
||||||
.. math::
|
.. math::
|
||||||
h = \frac{\partial^2{objective}}{\partial{pred}} = \frac{ - \log(pred + 1) + \log(label + 1) + 1}{(pred + 1)^2}
|
h = \frac{\partial^2{objective}}{\partial{pred}^2} = \frac{ - \log(pred + 1) + \log(label + 1) + 1}{(pred + 1)^2}
|
||||||
|
|
||||||
*****************************
|
*****************************
|
||||||
Customized Objective Function
|
Customized Objective Function
|
||||||
|
|||||||
Loading…
x
Reference in New Issue
Block a user