solvers.ErrorMeasurement.ErrorMeasurement Class Reference

Error measurement tool for ExaHyPE 2 solvers, this is the base class that must be specialised for different solvers. More...

Public Member Functions
	__init__ (self, solver, error_measurement_implementation, delta_between_database_flushes, output_file_name, data_delta_between_snapshots, time_delta_between_snapshots, clear_database_after_flush)

Detailed Description

Error measurement tool for ExaHyPE 2 solvers, this is the base class that must be specialised for different solvers.

Used by calling the corresponding specialisation by calling it using the solver instance as an argument. For an example, see /applications/exahype/aderdg i.e. ErrorMeasurement(solver=theSolver)

When called, it adds an additional function analyticalSolution() to the user-defined code snippets, through which the user can define what the solution should be.

The errormeasurement also adds a postprocessing step to the corresponding solver, which computes the L0, L2 and L_inf error between the computed solution and the analytical solution at each data point over the domain. These are then summed up over the entire volume to get the integrated errors over the entire domain.

Finally, these are added to a GlobalDatabase object by the global master thread, and outputted to csv files at regular intervals.

@parameters:

solver: the solver for which the errors should be measured

error_measurement_implementation: allows one to specify the analytical solution directly instead of manually filling it in the user-defined code snippets, analogously to fluxes and such in the solvers.

delta_between_database_flushes: how often should the entire database be printed to a file. By default the database is only printed once at the end of the entire simulation, but this parameters can be used to produce intermediary results

output_file_name: specify what the output .csv files should be named

data_delta_between_snapshots: which data deltas should be used for the database, i.e. if a measurement already exists for the given timestamp but the difference is greater than this value, the existing measurement is replaced.

time_delta_between_snapshots: how often a new measurement should be added to the database. By default the database is updated every timestep, but using this parameter it can be set to only happen at regular intervals, analogously to plotting in ExaHyPE2.

clear_database_after_flush: If the parameter "delta_between_database_flushes" has been set and several database flushes are going to occur, this parameter can be used to flush the database between each flush in order to avoid printing redundant information.

Definition at line 8 of file ErrorMeasurement.py.

Constructor & Destructor Documentation

__init__()

solvers.ErrorMeasurement.ErrorMeasurement.__init__	(		self,
			solver,
			error_measurement_implementation,
			delta_between_database_flushes,
			output_file_name,
			data_delta_between_snapshots,
			time_delta_between_snapshots,
			clear_database_after_flush
	)

Definition at line 60 of file ErrorMeasurement.py.

References solvers.ErrorMeasurement.create_constructor_implementation_for_error_measurement(), solvers.ErrorMeasurement.create_solver_user_abstract_declarations(), solvers.ErrorMeasurement.create_solver_user_abstract_definitions(), solvers.ErrorMeasurement.create_solver_user_declarations(), and solvers.ErrorMeasurement.create_solver_user_definitions().

Here is the call graph for this function:

---

The documentation for this class was generated from the following file:

• python/exahype2/solvers/ErrorMeasurement.py