Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
D
dune-gfe
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Model registry
Operate
Environments
Monitor
Incidents
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
Sander, Oliver
dune-gfe
Commits
a70dfb7f
Commit
a70dfb7f
authored
11 years ago
by
Oliver Sander
Committed by
sander
11 years ago
Browse files
Options
Downloads
Patches
Plain Diff
Test file to play around with ADOL-C in
[[Imported from SVN: r9377]]
parent
6aa10b53
No related branches found
Branches containing commit
No related tags found
No related merge requests found
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
test/adolctest.cc
+274
-0
274 additions, 0 deletions
test/adolctest.cc
with
274 additions
and
0 deletions
test/adolctest.cc
0 → 100644
+
274
−
0
View file @
a70dfb7f
/*----------------------------------------------------------------------------
ADOL-C -- Automatic Differentiation by Overloading in C++
File: speelpenning.cpp
Revision: $Id: speelpenning.cpp 299 2012-03-21 16:08:40Z kulshres $
Contents: speelpennings example, described in the manual
Copyright (c) Andrea Walther, Andreas Griewank, Andreas Kowarz,
Hristo Mitev, Sebastian Schlenkrich, Jean Utke, Olaf Vogel
This file is part of ADOL-C. This software is provided as open source.
Any use, reproduction, or distribution of the software constitutes
recipient's acceptance of the terms of the accompanying license file.
---------------------------------------------------------------------------*/
/****************************************************************************/
/* INCLUDES */
#include
"config.h"
#include
<adolc/adouble.h>
// use of active doubles
#include
<adolc/drivers/drivers.h>
// use of "Easy to Use" drivers
// gradient(.) and hessian(.)
#include
<adolc/taping.h>
// use of taping
#include
<iostream>
#include
<vector>
#include
<cstdlib>
#include
<math.h>
namespace
std
{
adouble
max
(
adouble
a
,
adouble
b
)
{
return
fmax
(
a
,
b
);
}
adouble
sqrt
(
adouble
a
)
{
return
sqrt
(
a
);
}
adouble
abs
(
adouble
a
)
{
return
fabs
(
a
);
}
adouble
pow
(
const
adouble
&
a
,
const
adouble
&
b
)
{
return
pow
(
a
,
b
);
}
bool
isnan
(
adouble
a
)
{
return
std
::
isnan
(
a
.
value
());
}
bool
isinf
(
adouble
a
)
{
return
std
::
isinf
(
a
.
value
());
}
}
#include
<dune/grid/yaspgrid.hh>
#include
<dune/geometry/quadraturerules.hh>
#include
<dune/localfunctions/lagrange/q1.hh>
#include
<dune/gfe/realtuple.hh>
#include
<dune/gfe/localgeodesicfefunction.hh>
#include
<dune/gfe/harmonicenergystiffness.hh>
using
namespace
Dune
;
#if 1
template
<
class
GridView
,
class
LocalFiniteElement
,
class
TargetSpace
>
double
energy
(
const
typename
GridView
::
template
Codim
<
0
>
::
Entity
&
element
,
const
LocalFiniteElement
&
localFiniteElement
,
const
std
::
vector
<
TargetSpace
>&
localPureSolution
)
{
double
pureEnergy
;
typedef
RealTuple
<
adouble
,
1
>
ADTargetSpace
;
std
::
vector
<
ADTargetSpace
>
localSolution
(
localPureSolution
.
size
());
typedef
typename
TargetSpace
::
template
rebind
<
adouble
>
::
other
ATargetSpace
;
HarmonicEnergyLocalStiffness
<
GridView
,
LocalFiniteElement
,
ATargetSpace
>
assembler
;
trace_on
(
1
);
adouble
energy
=
0
;
for
(
size_t
i
=
0
;
i
<
localSolution
.
size
();
i
++
)
localSolution
[
i
]
<<=
localPureSolution
[
i
];
energy
=
assembler
.
energy
(
element
,
localFiniteElement
,
localSolution
);
energy
>>=
pureEnergy
;
trace_off
(
1
);
return
pureEnergy
;
}
#endif
#if 0
template <class GridView, class LocalFiniteElement, class TargetSpace>
double
energy(const typename GridView::template Codim<0>::Entity& element,
const LocalFiniteElement& localFiniteElement,
const std::vector<TargetSpace>& localPureSolution)
{
typedef RealTuple<adouble,1> ADTargetSpace;
std::vector<ADTargetSpace> localSolution(localPureSolution.size());
trace_on(1);
for (size_t i=0; i<localSolution.size(); i++)
localSolution[i] <<= localPureSolution[i];
assert(element.type() == localFiniteElement.type());
static const int gridDim = GridView::dimension;
typedef typename GridView::template Codim<0>::Entity::Geometry Geometry;
double pureEnergy;
adouble energy = 0;
LocalGeodesicFEFunction<gridDim, adouble, LocalFiniteElement, ADTargetSpace> localGeodesicFEFunction(localFiniteElement,
localSolution);
int quadOrder = (element.type().isSimplex()) ? (localFiniteElement.localBasis().order()-1) * 2
: localFiniteElement.localBasis().order() * 2 * gridDim;
const Dune::QuadratureRule<double, gridDim>& quad
= Dune::QuadratureRules<double, gridDim>::rule(element.type(), quadOrder);
for (size_t pt=0; pt<quad.size(); pt++) {
// Local position of the quadrature point
const Dune::FieldVector<double,gridDim>& quadPos = quad[pt].position();
const double integrationElement = element.geometry().integrationElement(quadPos);
const typename Geometry::JacobianInverseTransposed& jacobianInverseTransposed = element.geometry().jacobianInverseTransposed(quadPos);
double weight = quad[pt].weight() * integrationElement;
// The derivative of the local function defined on the reference element
Dune::FieldMatrix<adouble, TargetSpace::EmbeddedTangentVector::dimension, gridDim> referenceDerivative = localGeodesicFEFunction.evaluateDerivative(quadPos);
// The derivative of the function defined on the actual element
Dune::FieldMatrix<adouble, TargetSpace::EmbeddedTangentVector::dimension, gridDim> derivative(0);
for (size_t comp=0; comp<referenceDerivative.N(); comp++)
jacobianInverseTransposed.umv(referenceDerivative[comp], derivative[comp]);
// Add the local energy density
// The Frobenius norm is the correct norm here if the metric of TargetSpace is the identity.
// (And if the metric of the domain space is the identity, which it always is here.)
energy += weight * derivative.frobenius_norm2();
}
energy *= 0.5;
energy >>= pureEnergy;
trace_off(1);
return pureEnergy;
}
#endif
/****************************************************************************/
/* MAIN PROGRAM */
int
main
()
{
size_t
n
=
4
;
//std::cout << className< decltype(adouble() / double()) >() << std::endl;
const
int
dim
=
2
;
typedef
YaspGrid
<
dim
>
GridType
;
FieldVector
<
double
,
dim
>
l
(
1
);
std
::
array
<
int
,
dim
>
elements
=
{{
1
,
1
}};
GridType
grid
(
l
,
elements
);
typedef
Q1LocalFiniteElement
<
double
,
double
,
dim
>
LocalFE
;
LocalFE
localFiniteElement
;
typedef
RealTuple
<
double
,
1
>
TargetSpace
;
std
::
vector
<
TargetSpace
>
localSolution
(
n
);
localSolution
[
0
]
=
TargetSpace
(
0
);
localSolution
[
1
]
=
TargetSpace
(
0
);
localSolution
[
2
]
=
TargetSpace
(
1
);
localSolution
[
3
]
=
TargetSpace
(
1
);
double
laplaceEnergy
=
energy
<
GridType
::
LeafGridView
,
LocalFE
,
TargetSpace
>
(
*
grid
.
leafbegin
<
0
>
(),
localFiniteElement
,
localSolution
);
std
::
cout
<<
"Laplace energy: "
<<
laplaceEnergy
<<
std
::
endl
;
std
::
vector
<
double
>
xp
(
n
);
for
(
size_t
i
=
0
;
i
<
n
;
i
++
)
xp
[
i
]
=
1
;
double
**
H
=
(
double
**
)
malloc
(
n
*
sizeof
(
double
*
));
for
(
size_t
i
=
0
;
i
<
n
;
i
++
)
H
[
i
]
=
(
double
*
)
malloc
((
i
+
1
)
*
sizeof
(
double
));
hessian
(
1
,
n
,
xp
.
data
(),
H
);
// H equals (n-1)g since g is
std
::
cout
<<
"Hessian:"
<<
std
::
endl
;
for
(
size_t
i
=
0
;
i
<
n
;
i
++
)
{
for
(
size_t
j
=
0
;
j
<
n
;
j
++
)
{
double
value
=
(
j
<=
i
)
?
H
[
i
][
j
]
:
H
[
j
][
i
];
std
::
cout
<<
value
<<
" "
;
}
std
::
cout
<<
std
::
endl
;
}
// Get gradient
#if 0
int n,i,j;
size_t tape_stats[STAT_SIZE];
cout << "SPEELPENNINGS PRODUCT (ADOL-C Documented Example)\n\n";
cout << "number of independent variables = ? \n";
cin >> n;
std::vector<double> xp(n);
double yp = 0.0;
std::vector<adouble> x(n);
adouble y = 1;
for(i=0; i<n; i++)
xp[i] = (i+1.0)/(2.0+i); // some initialization
trace_on(1); // tag = 1, keep = 0 by default
for(i=0; i<n; i++) {
x[i] <<= xp[i]; // or x <<= xp outside the loop
y *= x[i];
} // end for
y >>= yp;
trace_off(1);
tapestats(1,tape_stats); // reading of tape statistics
cout<<"maxlive "<<tape_stats[NUM_MAX_LIVES]<<"\n";
// ..... print other tape stats
double* g = new double[n];
gradient(1,n,xp.data(),g); // gradient evaluation
double** H = (double**) malloc(n*sizeof(double*));
for(i=0; i<n; i++)
H[i] = (double*)malloc((i+1)*sizeof(double));
hessian(1,n,xp.data(),H); // H equals (n-1)g since g is
double errg = 0; // homogeneous of degree n-1.
double errh = 0;
for(i=0; i<n; i++)
errg += fabs(g[i]-yp/xp[i]); // vanishes analytically.
for(i=0; i<n; i++) {
for(j=0; j<n; j++) {
if (i>j) // lower half of hessian
errh += fabs(H[i][j]-g[i]/xp[j]);
} // end for
} // end for
cout << yp-1/(1.0+n) << " error in function \n";
cout << errg <<" error in gradient \n";
cout << errh <<" consistency check \n";
#endif
return
0
;
}
// end main
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment