Todo List

Member indii::ml::filter::FixableStateModel::fix (const unsigned int i, const double value)

Preservation in resize() is not yet implemented in uBLAS for sparse matrices. Copy into dense matrix at present, sparse matrices are used here for computational rather than storage efficiency after all. Review in future if preservation for sparse matrices is implemented in uBLAS.

Member indii::ml::aux::GaussianMixturePdf::GaussianMixturePdf (const unsigned int K, const DiracMixturePdf &p)

This has not been tested thoroughly.

Member indii::ml::filter::KalmanSmoother::smooth (const T tn, const indii::ml::aux::vector &ytn, const indii::ml::aux::GaussianPdf &p_xtn_ytn)

^^^ Subtract sigma_t^-1, see Jordan 15.7.2 end.

Member indii::ml::aux::KernelDensityMixturePdf::densityAt (PartitionTree &tree)

Currently requires KDTree rather than PartitionTree due to apparent link errors related to Boost.MPI and Boost.Serialization.

Member indii::ml::aux::KernelDensityMixturePdf::distributedDensityAt (KDTree< S > &tree)

Currently requires KDTree rather than PartitionTree due to apparent link errors related to Boost.MPI and Boost.Serialization.

Member indii::ml::filter::LinearModel::LinearModel (indii::ml::aux::matrix &A, indii::ml::aux::matrix &G, indii::ml::aux::symmetric_matrix &Q, indii::ml::aux::matrix &C, indii::ml::aux::symmetric_matrix &R)

System noise needn't have the same number of dimensions at the state, and likewise G may be N*P, not N*N, where P is the number of dimensions of the noise.

Member indii::ml::filter::LinearModel::p_xtnp1_ytnp1 (const indii::ml::aux::GaussianPdf &p_xtnp1_ytn, const indii::ml::aux::vector &ytnp1, const unsigned int delta)

Doesn't consider delta currently. Should iterate as many times as specified by delta, for example, in the case that there is a missing observation. Just do this using recursion.

Member indii::ml::aux::Random::orthonormalMatrix (const unsigned int N)

Should check random for full rank, just in case, although it's highly unlikely that the set is linearly dependent.

Class StochasticAdaptiveEulerMaruyama

This class is tightly coupled with the GSL and would benefit from greater independence.

Class StochasticAdaptiveRungeKutta

This class is tightly coupled with the GSL and would benefit from greater independence.

Class StochasticEulerMaruyama

This class is tightly coupled with the GSL and would benefit from greater independence.

---