Evaluate residuals

84 views
Skip to first unread message

Albert Palomer

unread,
Oct 19, 2016, 7:25:26 AM10/19/16
to ceres-...@googlegroups.com
hello,

I have a ceres::Problem with all the blocks added and ready to be optimized. However, when I optimize it stops far away from the solution because the steps are too small. I have two questions:
1) what can I do to make it continue solving? how can I make its steps bigger?
2) is there a way that I can evaluate the residuals of the problem for a specific solution? I want to check that there are not local minimums around the solution. For that, the easiest for me is to just do brute force and get the residuals for many many solutions and plot the function.

Thank you!

Albert Palomer Vila

Sameer Agarwal

unread,
Oct 19, 2016, 7:55:25 AM10/19/16
to ceres-...@googlegroups.com
Albert,
Ceres uses a trust region algorithm to solve the optimization problem. The algorithm automatically controls the step size. It would be helpful to look at a log from when you call solve to know more, but there are a variety of reasons why it the algorithm is stopping. 

Ill-conditioning of the problem is one and a local minimum is another. 

You can try changing the initial trust region radius (See solver.h) but I doubt it will solve your problem.

As for evaluating residuals see the docs for Problem::Evaluate.

Sameer



--
You received this message because you are subscribed to the Google Groups "Ceres Solver" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ceres-solver...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ceres-solver/CAD3_OrVDyOErwty9Q8z%2B-7sUuAzYskim4Q5Wtm3DV3McNQE%2B%3Dg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
Reply all
Reply to author
Forward
0 new messages