IMP
2.4.0
The Integrative Modeling Platform
|
A simple steepest descent optimizer. More...
#include <IMP/core/SteepestDescent.h>
Inherits AttributeOptimizer.
A simple steepest descent optimizer.
Follow the gradient based on the partial derivatives. Multiply by the current step size. If the score gets worse, reduce the step size. If the score gets better, increase the step size if we are sufficiently far from a score of zero. If the score reaches the threshold, quit.
Definition at line 23 of file SteepestDescent.h.
Public Member Functions | |
SteepestDescent (kernel::Model *m, std::string name="SteepestDescent%1%") | |
virtual Float | do_optimize (unsigned int max_steps) |
virtual std::string | get_type_name () const |
virtual ::IMP::base::VersionInfo | get_version_info () const |
void | set_maximum_step_size (Float t) |
The maximum step size. More... | |
void | set_step_size (Float t) |
The starting step size. More... | |
void | set_threshold (Float t) |
Set the minimum gradient threshold. More... | |
void IMP::core::SteepestDescent::set_maximum_step_size | ( | Float | t | ) |
The maximum step size.
Definition at line 35 of file SteepestDescent.h.
void IMP::core::SteepestDescent::set_step_size | ( | Float | t | ) |
The starting step size.
Definition at line 33 of file SteepestDescent.h.
void IMP::core::SteepestDescent::set_threshold | ( | Float | t | ) |
Set the minimum gradient threshold.
Definition at line 31 of file SteepestDescent.h.