-
At the end of on ipopt run (perhaps maximum number of iterations was reached) I can get most of the necessary warm starting information from the call to finalize_solution. In addition, I can pass it back when I want to continue the optimization using get_starting_point. One notable exception is the value of the penalty parameter mu. How do I get the value of mu at the end of a run and how do I pass it in to continue the optimization ? |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments
-
You should be able to get mu from the IpoptData object (https://coin-or.github.io/Ipopt/classIpopt_1_1IpoptData.html#ad83f966070811d55a0c5887df1a854bb). There is option |
Beta Was this translation helpful? Give feedback.
-
@svigerske That works for me (see below). Thanks
|
Beta Was this translation helpful? Give feedback.
You should be able to get mu from the IpoptData object (https://coin-or.github.io/Ipopt/classIpopt_1_1IpoptData.html#ad83f966070811d55a0c5887df1a854bb).
There is option
mu_init
to set the initial mu if you usemu_strategy = monotone
: https://coin-or.github.io/Ipopt/OPTIONS.html#OPT_mu_init