A helper class for local assemblers using numeric differentiation to determine the epsilon.  
 More...
#include <dumux/assembly/numericepsilon.hh>
template<class Scalar, int numEq>
class Dumux::NumericEpsilon< Scalar, numEq >
- Template Parameters
- 
  
    | Scalar | the scalar type |  | numEq | the number of primary variables / equations |  
 
|  | 
|  | NumericEpsilon (const std::string ¶mGroup="") | 
|  | 
| Scalar | operator() (Scalar priVar, int priVarIdx) const noexcept | 
|  | get the epsilon 
 | 
|  | 
◆ NumericEpsilon()
template<class Scalar, int numEq> 
 
 
◆ operator()()
template<class Scalar, int numEq> 
 
- Note
- If no user input was specified -> try to estimate magnitude from primary variable value else -> use given magnitude for the primary variable times the base epsilon (default 1e-10) 
 
 
The documentation for this class was generated from the following file: