Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

about use of two TMB templates as C++ files with headers in a new R package #317

Open
oevkaya opened this issue Jul 16, 2020 · 0 comments
Open
Labels

Comments

@oevkaya
Copy link

oevkaya commented Jul 16, 2020

Urgent help about the use of TMB templates in R package with Rcpp

Description:

For my project, I am trying to use two TMB templates (LocalLikelihood1 and LocalLikelihood2) with some C files also. LocalLikelihood1 is same as the LocalLikelihood.cpp in https://github.com/mlysy/LocalCop/tree/master/src. I created a second one because of two different estimations. Besides, I followed the steps given here for creating two templates correctly (in that sense it is different what it was done in the package LocalCop it seems, for detail: https://github.com/mlysy/LocalCop). Since I have limited knowledge on C++ I tried to stick previous codes and also mimicked what is discussed in different discussions but still I could not figure out why I got errors related to RcppEigen ?

Reproducible Steps:

I just created a testVDAG package with Locallikelihood1.cpp with the header files like Clayton.h etc.
When I check the package I got error messages related to RcppEigen files like Meta.h, DenseBase.h telling me that
-Line 83 no members matching and similar

This is my main .cpp file

/// @file LocalLikelihood1.cpp
///
/// Local Likelihood functions.

#define TMB_LIB_INIT R_init_tsVine
#include <TMB.hpp>
#include "FrankCopula.h"
#include "GaussCopula.h"
#include "GumbelCopula.h"
#include "JoeCopula.h"
#include "ClaytonCopula.h"

template
Type objective_function::operator() () {
DATA_VECTOR(y1); // first response vector
DATA_VECTOR(y2); // second response vector
DATA_VECTOR(wgt); // weights
DATA_VECTOR(yc); // centered covariates, i.e., Y - y
DATA_INTEGER(family); // copula family: 1-3-4-5-6.
PARAMETER_VECTOR(beta); // dependence parameter: eta = beta[0] + beta[1] * yc
Type nll = 0.0;
if(family == 1) {
// Gaussian copula
nll = GaussNLL(y1, y2, yc, beta, wgt);
} else if(family == 3) {
// Clayton copula
nll = ClaytonNLL(y1, y2, yc, beta, wgt);
} else if(family == 4) {
// Gumbel copula
nll = GumbelNLL(y1, y2, yc, beta, wgt);
} else if(family == 5) {
// Frank copula
nll = FrankNLL(y1, y2, yc, beta, wgt);
} else if(family == 6) {
// Joe copula
nll = JoeNLL(y1, y2, yc, beta, wgt);
}
return nll;
}

This is one of the .h files
/// @file ClaytonCopula.h

/// Local likelihood for the Clayton copula.
///
/// @param[in] lu1 First log-uniform response vector.
/// @param[in] lu2 Second log-uniform response vector.
/// @param[in] yc Covariate vector in centered form yc = Y-y.
/// @param[in] beta Length-two vector of dependence parameters,
/// such that eta(yc) = beta[0] + beta[1] * yc.
/// @param[in] wgt Local likelihood weights.
/// @return Negative local loglikelihood (scalar).
template
Type ClaytonNLL(vector lu1, vector lu2,
vector yc, vector beta, vector wgt) {
// parameter on regular scale
vector theta = (beta[0] + beta[1] * yc).exp();
vector ll = (1.0 + theta).log() - (1.0 + theta) * (lu1 + lu2);
ll -= (2.0 + 1.0/theta) * ((-theta * lu1).exp() + (-theta * lu2).exp() - 1.0).log();
ll *= wgt;
return -ll.sum();
}

This is my DESCRIPTION file (you can use the given mail address if you want)

Package: testVDAG
Type: Package
Title: What the Package Does (Title Case)
Version: 0.1.0
Author: OzanEvkaya [email protected]
Maintainer: The package maintainer [email protected]
Description: More about what it does (maybe more than one line)
Use four spaces when indenting paragraphs within the Description.
License: GPL-3
Encoding: UTF-8
LazyData: true
RoxygenNote: 7.1.0
LinkingTo:
Rcpp, RcppEigen, TMB
Imports:
Rcpp,
TMB

I created .R file for this part named as testVDAG.R

#'Test package for semVineDAG
#' @useDynLib testVDAG, .registration = TRUE
#' @importFrom Rcpp sourceCpp

Sorry for above long lines, to reproduce you can look at the attached file for the testVDAG project in R

TMB Version:

packageVersion("TMB")
[1] ‘1.7.16’

R Version:

R.Version()
$platform
[1] "x86_64-w64-mingw32"

$arch
[1] "x86_64"

$os
[1] "mingw32"

$system
[1] "x86_64, mingw32"

$version.string
[1] "R version 3.6.2 (2019-12-12)"

Operating System:

Windows 10 Enterprise 64-bit operating system, x64 based processor

testVDAG.zip

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

2 participants