|
Limbo 3.5.4
|
A base helper function object to update lagrangian multipliers using subgradient descent. All other schemes can be derived from this class. More...
#include <MultiKnapsackLagRelax.h>
Public Types | |
| typedef T | value_type |
| value type | |
Public Member Functions | |
| LagMultiplierUpdater () | |
| constructor | |
| virtual | ~LagMultiplierUpdater () |
| destructor | |
| virtual value_type | operator() (unsigned int iter, value_type multiplier, value_type slackness)=0 |
| API to update lagrangian multiplier. | |
| virtual void | operator() (unsigned int iter, unsigned int n, value_type const *vSlackness, value_type const *vLagMultiplier, value_type *vNewLagMultiplier)=0 |
| API to update lagrangian multiplier using subgradient descent. | |
A base helper function object to update lagrangian multipliers using subgradient descent. All other schemes can be derived from this class.
| T | coefficient value type |
Definition at line 933 of file MultiKnapsackLagRelax.h.
| typedef T limbo::solvers::LagMultiplierUpdater< T >::value_type |
value type
Definition at line 937 of file MultiKnapsackLagRelax.h.
|
inline |
constructor
Definition at line 940 of file MultiKnapsackLagRelax.h.
|
inlinevirtual |
destructor
Definition at line 942 of file MultiKnapsackLagRelax.h.
|
pure virtual |
API to update lagrangian multiplier using subgradient descent.
| iter | current iteration |
| n | dimension |
| vSlackness | array of slackness |
| vLagMultiplier | array of lagrangian multipliers which will be updated |
| vNewLagMultiplier | array of new lagrangian multipliers |
Implemented in limbo::solvers::SubGradientDescent< T >.
|
pure virtual |
API to update lagrangian multiplier.
| iter | current iteration |
| multiplier | current multiplier value |
| slackness | current slackness value assuming the constraint is in \( Ax \le b \) and compute \( b-Ax \) |
Implemented in limbo::solvers::SubGradientDescent< T >.