Emrys::Batched< F, Mark > Struct Template Reference

A function that takes every point at once, as batched() marks it (decisions.md [batched-functions]): it receives the points as the columns of a matrix X and writes into one output Y, a BatchedOutput the library provides, last. More...

#include <EmrysFilter/Models/Batched.h>

Public Types
using	FunctionType = F
	the function
using	MarkType = Mark
	the Jacobians it writes

Public Member Functions
template<typename In, typename Out, MatrixExpression X, typename... Rest>
auto	operator() (In const &in, Out const &out, X const &x, Rest const &... rest) const

Public Attributes
F	function
	the function

Static Public Attributes
static constexpr bool	GivesJacobians = !std::is_same_v<Mark, internal::NoJacobians>
	whether the function writes Jacobians
static constexpr bool	PerColumn = std::is_same_v<Mark, jacobians::PerColumn>
	whether it writes one Jacobian per column

Detailed Description

template<typename F, typename Mark = internal::NoJacobians>
struct Emrys::Batched< F, Mark >

A function that takes every point at once, as batched() marks it (decisions.md [batched-functions]): it receives the points as the columns of a matrix X and writes into one output Y, a BatchedOutput the library provides, last.

Template Parameters

F	the function
Mark	the Jacobians it writes: internal::NoJacobians, jacobians::PerColumn, or jacobians::Shared

See also

batched()

Member Function Documentation

operator()()

template<typename F, typename Mark>

template<typename In, typename Out, MatrixExpression X, typename... Rest>

auto Emrys::Batched< F, Mark >::operator() ( In const & in, Out const & out, X const & x, Rest const &... rest ) const

nodiscard

Returns

the function at one point, x, a vector in its input space in, as a batch of one column, with rest, the step of a process model or nothing, before its output: the value, a vector in the output space out, and with a mark that gives Jacobians a WithJacobian of it and the Jacobian over the input's differences, every block wanted. This is how Linearized and the iterated updates call a batched function, and how a function differentiated in the autodiff scalar is called (decisions.md [batched-functions]).

---

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

• EmrysFilter/Models/Batched.h