Class MultiConditionDataProviderHDF5

Inheritance Relationships

Base Type

Class Documentation

class MultiConditionDataProviderHDF5 : public parpe::MultiConditionDataProvider

The MultiConditionDataProvider class reads simulation data for MultiConditionOptimizationProblem from a HDF5 file.

This class assumes a certain layout of the underlying HDF5 file. Der dataset names can be modified in hdf5*Path members.

Required dimensions:

  • hdf5MeasurementPath, hdf5MeasurementSigmaPath: numObservables x numConditions

  • hdf5ConditionPath: numFixedParameters x numConditions

  • hdf5AmiciOptionPath:

  • hdf5ParameterPath:

NOTE: The following dimensions are determined by the used AMICI model:

  • numObservables := Model::ny

  • numFixedParameters := Model::nk

Public Functions

MultiConditionDataProviderHDF5() = default
MultiConditionDataProviderHDF5(std::unique_ptr<amici::Model> model, const std::string &hdf5Filename)

MultiConditionDataProvider.

Parameters:

  • model – A valid pointer to the amici::Model for which the data is to be provided.

  • hdf5Filename – Path to the HDF5 file from which the data is to be read

MultiConditionDataProviderHDF5(std::unique_ptr<amici::Model> model, std::string const &hdf5Filename, std::string const &rootPath)

See above.

Parameters:

  • model –

  • hdf5Filename –

  • rootPath – The name of the HDF5 group under which the data is stored.

MultiConditionDataProviderHDF5(MultiConditionDataProviderHDF5 const&) = delete
~MultiConditionDataProviderHDF5() override
virtual int getNumberOfSimulationConditions() const override

Get the number of simulations required for objective function evaluation. Currently, this amounts to the number of conditions present in the data.

Returns:

Number of conditions

virtual std::vector<int> getSimulationToOptimizationParameterMapping(int conditionIdx) const override

Get index vector of length of model parameter with indices of optimization parameters for the given condition.

NOTE: This may contain -1 for parameter which are not mapped.

Parameters:

conditionIdx –

Returns:

virtual void mapSimulationToOptimizationGradientAddMultiply(int conditionIdx, gsl::span<double const> simulation, gsl::span<double> optimization, gsl::span<const double> parameters, double coefficient = 1.0) const override
virtual void mapAndSetOptimizationToSimulationVariables(int conditionIdx, gsl::span<double const> optimization, gsl::span<double> simulation, gsl::span<amici::ParameterScaling> optimizationScale, gsl::span<amici::ParameterScaling> simulationScale) const override
virtual std::vector<amici::ParameterScaling> getParameterScaleOpt() const override
virtual amici::ParameterScaling getParameterScaleOpt(int parameterIdx) const override

Get the parameter scale for the given optimization parameter.

Parameters:

simulationIdx –

Returns:

Parameter scale

virtual std::vector<amici::ParameterScaling> getParameterScaleSim(int simulationIdx) const override

Get the parameter scale vector for the given simulation.

Parameters:

simulationIdx –

Returns:

virtual amici::ParameterScaling getParameterScaleSim(int simulationIdx, int modelParameterIdx) const override

Get the parameter scale for the given parameter and simulation.

Parameters:

simulationIdx –

Returns:

virtual void checkDataIntegrity() const

Check if the data in the HDF5 file has consistent dimensions. Aborts if not.

virtual void readFixedSimulationParameters(int conditionIdx, gsl::span<double> buffer) const
virtual std::unique_ptr<amici::ExpData> getExperimentalDataForCondition(int conditionIdx) const override
virtual std::vector<std::vector<double>> getAllMeasurements() const override
virtual std::vector<std::vector<double>> getAllSigmas() const override
std::vector<double> getSigmaForSimulationIndex(int simulationIdx) const
std::vector<double> getMeasurementForSimulationIndex(int conditionIdx) const
virtual void getOptimizationParametersLowerBounds(gsl::span<double> buffer) const

Writes lower parameter bounds into the provided buffer.

Parameters:

buffer – allocated memory to write parameter bounds

virtual void getOptimizationParametersUpperBounds(gsl::span<double> buffer) const

Writes upper parameter bounds into the provided buffer.

Parameters:

buffer – allocated memory to write parameter bounds

virtual int getNumOptimizationParameters() const override

Returns the number of optimization parameters of this problem.

Returns:

Number of parameters

virtual std::unique_ptr<amici::Model> getModel() const override

Returns a pointer to a copy of the underlying AMICI model as provided to the constructor.

Returns:

The model

virtual std::unique_ptr<amici::Solver> getSolver() const override
virtual void updateSimulationParametersAndScale(int simulationIdx, gsl::span<const double> optimizationParams, amici::Model &model) const override

Based on the array of optimization parameters, set the simulation parameters in the given Model object to the ones for simulation index.

Parameters:

  • simulationIdx – Index of the simulation condition for which to set model parameters.

  • optimizationParams – Problem parameters from which to extract simulation parameters.

  • model – Model on which to set parameter values and scale

void copyInputData(const H5::H5File &target)
void getSimAndPreeqConditions(const int simulationIdx, int &preequilibrationConditionIdx, int &simulationConditionIdx) const
std::vector<int> getReinitializationIndices(const int simulationIdx) const
H5::H5File getHdf5File() const

Get a copy of the HDF5 file handle.

Returns:

File handle

void setModel(std::unique_ptr<amici::Model> model)
virtual std::vector<std::string> getProblemParameterIds() const override

Protected Functions

void updateFixedSimulationParameters(int conditionIdx, amici::ExpData &edata) const

Update the constants in AMICI ExpData object. Reads a slab for the given simulation from fixed parameters matrix.

Parameters:

  • simulationIdx – Index of the experimental condition for which the parameters should be taken.

  • edata – The object to be updated.