JayBeams
0.1
Another project to have fun coding.
|
A time-delay estimator (TDE) is an algorithm to compare two families of timeseries and return the estimated delay of the first family vs. More...
#include <tde_result.hpp>
Public Types | |
using | value_type = value_t |
using | record_type = value_type |
Public Member Functions | |
tde_result (container_t const &a) | |
value_type & | operator[] (std::size_t pos) |
tde_result holds only one value, we overload the subscription operator to allow generic usage of the type. More... | |
value_type const & | operator[] (std::size_t pos) const |
tde_result value held by a const object. More... | |
std::size_t | size () const |
size of tde_result record_type, holds only one value More... | |
Private Attributes | |
record_type | value_ |
A time-delay estimator (TDE) is an algorithm to compare two families of timeseries and return the estimated delay of the first family vs.
the second family. When the families of timeseries are represented by an array of dimension K, the last dimension is interpreted as time, and the remaining K-1 dimensions are interpreted as the family parameters. In that case, the output of a TDE is an array of one dimension less than the inputs to the TDE algorithm.
In contrast, when the input into the TDE algorithm is a simple vector, or a one dimensional array then the output is a simple value.
This template class, and its specializations, compute the representation for the TDE result given the input type. Because the TDE algorithms usually have two outputs (the time delay estimate, usually represented by an integer, and the confidence, usually a floating point type), the class is parametric on the output type. This parametrization allows us to use this type in other timeseries computations that return a (1) value out of operating on the last dimension (e.g. sum sqr, average).
container_t | the representation for a timeseries, expected to be a one-dimensional container such as std::vector<> or std::deque<> |
value_t | type to store the result of the timeseries operation |
Definition at line 37 of file tde_result.hpp.
using jb::fftw::tde_result< container_t, value_t >::record_type = value_type |
Definition at line 40 of file tde_result.hpp.
using jb::fftw::tde_result< container_t, value_t >::value_type = value_t |
Definition at line 39 of file tde_result.hpp.
|
inlineexplicit |
Definition at line 41 of file tde_result.hpp.
|
inline |
tde_result holds only one value, we overload the subscription operator to allow generic usage of the type.
Definition at line 47 of file tde_result.hpp.
References jb::fftw::tde_result< container_t, value_t >::value_.
|
inline |
tde_result value held by a const object.
tde_result holds only one value, we overload the subscription operator to allow generic usage of the type.
Definition at line 54 of file tde_result.hpp.
References jb::fftw::tde_result< container_t, value_t >::value_.
|
inline |
size of tde_result record_type, holds only one value
Definition at line 59 of file tde_result.hpp.
|
private |
Definition at line 64 of file tde_result.hpp.
Referenced by jb::fftw::tde_result< container_t, value_t >::operator[](), and jb::fftw::tde_result< boost::multi_array< T, 1, A >, value_t >::operator[]().