 |
Simple conjugate gradients optimizer.
Algorithm is as per Shanno and Phua, ACM Transactions On Mathematical Software 6 (December 1980), 618-622
Conjugate gradients optimization is sensitive to the scales of the derivatives of the various attributes being optimized. By default, the scales are estimated from the range of values found for the attribute upon initialization. These estimates can be viewed either by calling Model::get_range(my_float_key) or by turning on TERSE logging and looking at logged messages. If this estimate does not accurately reflect the scale, then you can use Model::set_range to set a more accurate range for the parameters.
Examples: symmetry, bipartite nonbonded interactions, nonbonded interactions, rigid collisions, dock with crosslinks, analyze convergence, restrain diameter, optimize balls
Inheritance diagram for IMP::core::ConjugateGradients:Public Member Functions | |
| ConjugateGradients (Model *m=nullptr) | |
| void | set_gradient_threshold (Float t) |
| Set the threshold for the minimum gradient. | |
| void | set_max_change (Float t) |
| Limit how far anything can change each time step. | |
Static Public Member Functions | |
| static ConjugateGradients * | get_from (IMP::base::Object *o) |
| void IMP::core::ConjugateGradients::set_gradient_threshold | ( | Float | t | ) |
Set the threshold for the minimum gradient.
| void IMP::core::ConjugateGradients::set_max_change | ( | Float | t | ) |
Limit how far anything can change each time step.
