CppADCodeGen
HEAD
A C++ Algorithmic Differentiation Package with Source Code Generation
|
Public Member Functions | |
virtual std::set< std::string > | getModelNames ()=0 |
virtual std::unique_ptr< GenericModel< Base > > | model (const std::string &modelName)=0 |
virtual void | setThreadPoolDisabled (bool disabled)=0 |
virtual bool | isThreadPoolDisabled () const =0 |
virtual unsigned int | getThreadNumber () const =0 |
virtual void | setThreadNumber (unsigned int n)=0 |
virtual ThreadPoolScheduleStrategy | getThreadPoolSchedulerStrategy () const =0 |
virtual void | setThreadPoolSchedulerStrategy (ThreadPoolScheduleStrategy s)=0 |
virtual void | setThreadPoolVerbose (bool v)=0 |
virtual bool | isThreadPoolVerbose () const =0 |
virtual void | setThreadPoolGuidedMaxWork (float v)=0 |
virtual float | getThreadPoolGuidedMaxWork () const =0 |
virtual void | setThreadPoolNumberOfTimeMeas (unsigned int n)=0 |
virtual unsigned int | getThreadPoolNumberOfTimeMeas () const =0 |
Abstract class used to load models
Definition at line 28 of file model_library.hpp.
|
pure virtual |
Provides the model names in the dynamic library.
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Provides the maximum number of threads used to determine sparse Jacobians and sparse Hessians for the models in this library. This value is only used by the models if they were compiled with multithreading support.
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Provides the number of time measurements taken by each computational task during multithreaded model evaluations. This is used to schedule work accross threads. The higher the value the more accurate the time estimations are but it requires additional calls to retrieve times. This value is only used by the models if they were compiled with multithreading support.
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Provides the thread scheduling strategy used to determine sparse Jacobians and sparse Hessians for the models in this library. This value is only used by the models if they were compiled with multithreading support.
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Determines whether or not multithreaded model evaluations are disabled.
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Creates a new GenericModel object that can be used to evaluate the model.
modelName | The model name. |
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Defines the maximum number of threads used to determine sparse Jacobians and sparse Hessians for the models in this library. This value is only used by the models if they were compiled with multithreading support. It should be defined before using the models.
n | the maximum number of threads |
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Defines whether or not to disable multithreaded model evaluations. This only works if the models if they were compiled with multithreading support.
disabled | true to only use the current thread to evaluate models. |
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Defines the number of time measurements taken by each computational task during multithreaded model evaluations. This is used to schedule work across threads. The higher the value the more accurate the time estimations are but it requires additional calls to retrieve times. This value is only used by the models if they were compiled with multithreading support.
n | the number of time measurements to take per task. |
Implemented in CppAD::cg::FunctorModelLibrary< Base >.
|
pure virtual |
Defines the thread scheduling strategy used to determine sparse Jacobians and sparse Hessians for the models in this library. This value is only used by the models if they were compiled with multithreading support. It should be defined before using the models.
s | the thread scheduling strategy |
Implemented in CppAD::cg::FunctorModelLibrary< Base >.