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
In my experience, using find_zero() with with bracketing methods always returns an error if the supplied interval is not in fact bracketing. Today, however, I discovered that Roots.Chandrapatla() does not do that, and instead silently fails and returns a non-sential result. I believe it would be good to make its behavior consistent with that of the other methods in order to avoid silent errors.
For example, if we define a function that does not have a root f(x) = 1 + x^2
and apply Roots.Chandrapatla() to it on some interval find_zero(f, (10.0, 20.0), Roots.Chandrapatla())
we get a non-sential result 2.149362113533461.
Apologies if I missed something in the discussion/documentation, and this behaviour is in fact intentional.
The text was updated successfully, but these errors were encountered:
In my experience, using
find_zero()
with with bracketing methods always returns an error if the supplied interval is not in fact bracketing. Today, however, I discovered thatRoots.Chandrapatla()
does not do that, and instead silently fails and returns a non-sential result. I believe it would be good to make its behavior consistent with that of the other methods in order to avoid silent errors.For example, if we define a function that does not have a root
f(x) = 1 + x^2
and apply
Roots.Chandrapatla()
to it on some intervalfind_zero(f, (10.0, 20.0), Roots.Chandrapatla())
we get a non-sential result 2.149362113533461.
Apologies if I missed something in the discussion/documentation, and this behaviour is in fact intentional.
The text was updated successfully, but these errors were encountered: