You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
All solvers in JAXopt support implicit differentiation out-of-the-box. Most solvers have an implicit_diff=True|False option. When set to False, autodiff of unrolled iterates is used instead of implicit differentiation.
Is there any example of explicit differentiation by setting implicit_diff=False and unroll=True?
If I use explicit differentiation and calculate jvp, does this mean that I need to call the solver twice?
First, solve the problem to get the optimal solution.
Second, use jax.jvp and the optimal solution to calculate the jacobian and vector product.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
According to the documentation,
Is there any example of explicit differentiation by setting
implicit_diff=False
andunroll=True
?If I use explicit differentiation and calculate
jvp
, does this mean that I need to call the solver twice?jax.jvp
and the optimal solution to calculate the jacobian and vector product.Thanks in advance.
Beta Was this translation helpful? Give feedback.
All reactions