\(\newcommand{\W}[1]{ \; #1 \; }\) \(\newcommand{\R}[1]{ {\rm #1} }\) \(\newcommand{\B}[1]{ {\bf #1} }\) \(\newcommand{\D}[2]{ \frac{\partial #1}{\partial #2} }\) \(\newcommand{\DD}[3]{ \frac{\partial^2 #1}{\partial #2 \partial #3} }\) \(\newcommand{\Dpow}[2]{ \frac{\partial^{#1}}{\partial {#2}^{#1}} }\) \(\newcommand{\dpow}[2]{ \frac{ {\rm d}^{#1}}{{\rm d}\, {#2}^{#1}} }\)

base2ad

View page source

Create an AD<Base> Function From a Base Function

Syntax

af = f . base2ad ()

See Also

mul_level

Base

This is the base type used to recorded the operation sequence in f and af ; i.e., the type AD < Base > was used to record the operation sequence.

f

This object has prototype

ADFun < Base > f

It does it’s derivative calculations using the type Base .

af

This object has prototype

ADFun< AD< Base > , Base > af

It has the same operation sequence as f , but it does it’s derivative calculations using the type AD < Base> . This enables one to record new functions that are defined using derivatives of the function f . Initially, there are no Taylor coefficients stored in af and af.size_order() is zero.

Example

The file base2ad.cpp contains an example and test of this operation.

VecAD

Forward mode on a base2ad function does not preserve VecAD operations (which might be expected); see the base2vec_ad.cpp example.