#include <dune/pdelab/newton/newton.hh>
◆ Result
◆ Strategy
◆ Newton() [1/2]
template<class GOS , class S , class TrlV , class TstV = TrlV>
◆ Newton() [2/2]
template<class GOS , class S , class TrlV , class TstV = TrlV>
◆ apply() [1/2]
◆ apply() [2/2]
◆ defect()
|
inlineoverrideprotectedvirtualinherited |
◆ discardMatrix()
Discard the stored Jacobian matrix.
◆ keepMatrix()
Return whether the jacobian matrix is kept across calls to apply().
◆ line_search() [1/2]
|
inlineoverridevirtualinherited |
◆ line_search() [2/2]
|
protectedpure virtualinherited |
◆ prepare_step() [1/2]
|
protectedpure virtualinherited |
◆ prepare_step() [2/2]
|
inlineoverridevirtualinherited |
◆ result()
◆ setAbsoluteLimit()
◆ setFixedLinearReduction()
set a fixed reduction in the linear solver (overwrites setMinLinearReduction)
- Note
- with fixed_linear_reduction > 0, the linear reduction rate will always be fixed to min_linear_reduction.
◆ setForceIteration()
◆ setKeepMatrix()
Set whether the jacobian matrix should be kept across calls to apply().
◆ setLineSearchDampingFactor()
◆ setLineSearchMaxIterations()
◆ setLineSearchStrategy() [1/2]
◆ setLineSearchStrategy() [2/2]
◆ setMaxIterations()
◆ setMinLinearReduction()
set the minimal reduction in the linear solver
- Note
- with min_linear_reduction > 0, the linear reduction will be determined as mininum of the min_linear_reduction and the linear_reduction needed to achieve second order Newton convergence.
◆ setParameters()
template<class GOS , class S , class TrlV , class TstV = TrlV>
interpret a parameter tree as a set of options for the newton solver
example configuration:
[NewtonParameters]
ReassembleThreshold = 0.1
LineSearchMaxIterations = 10
MaxIterations = 7
MinLinearReduction = 1
e-3
LineSearchDamping = 0.9
and invocation in the code:
newton.setParameters(param.sub("NewtonParameters"));
◆ setReassembleThreshold()
set a threshold, when the linear operator is reassembled
We allow to keep the linear operator over several newton iterations. If the reduction in the newton drops below a given threshold the linear operator is reassembled to ensure convergence.
◆ setReduction()
◆ setVerbosityLevel()
◆ strategyFromName()
helper function to get the different strategies from their name
◆ terminate() [1/2]
|
inlineoverridevirtualinherited |
◆ terminate() [2/2]
|
protectedpure virtualinherited |
◆ A_
◆ abs_limit_
◆ gridoperator_
◆ keep_matrix_
◆ linear_reduction_
◆ prev_defect_
◆ r_
◆ reassembled_
◆ reduction_
◆ res_
◆ u_
◆ verbosity_level_
◆ z_
The documentation for this class was generated from the following file: