Added numericalHessian function for computing the Hessian matrix of a Lie->scalar function
parent
c52d525763
commit
cb999dba58
|
|
@ -16,6 +16,7 @@ check_PROGRAMS =
|
|||
headers += FixedVector.h types.h blockMatrices.h Matrix-inl.h lapack.h
|
||||
sources += Vector.cpp Matrix.cpp cholesky.cpp
|
||||
check_PROGRAMS += tests/testFixedVector tests/testVector tests/testMatrix tests/testCholesky
|
||||
check_PROGRAMS += tests/testNumericalDerivative
|
||||
|
||||
if USE_LAPACK
|
||||
sources += DenseQR.cpp DenseQRUtil.cpp
|
||||
|
|
|
|||
|
|
@ -49,12 +49,12 @@ namespace gtsam {
|
|||
* boost::bind(&SomeClass::bar, ref(instanceOfSomeClass), _1)
|
||||
*
|
||||
* For additional details, see the documentation:
|
||||
* http://www.boost.org/doc/libs/1_43_0/libs/bind/bind.html
|
||||
* http://www.boost.org/doc/libs/release/libs/bind/bind.html
|
||||
*/
|
||||
|
||||
|
||||
/** global functions for converting to a LieVector for use with numericalDerivative */
|
||||
inline LieVector makeLieVector(const Vector& v) { return LieVector(v); }
|
||||
inline LieVector makeLieVector(const Vector& v) { return LieVector(v); }
|
||||
inline LieVector makeLieVectorD(double d) { return LieVector(Vector_(1, d)); }
|
||||
|
||||
/**
|
||||
|
|
@ -68,8 +68,8 @@ namespace gtsam {
|
|||
const size_t n = x.dim();
|
||||
Vector d(n,0.0), g(n,0.0);
|
||||
for (size_t j=0;j<n;j++) {
|
||||
d(j) += delta; double hxplus = h(expmap(x,d));
|
||||
d(j) -= 2*delta; double hxmin = h(expmap(x,d));
|
||||
d(j) += delta; double hxplus = h(x.expmap(d));
|
||||
d(j) -= 2*delta; double hxmin = h(x.expmap(d));
|
||||
d(j) += delta; g(j) = (hxplus-hxmin)*factor;
|
||||
}
|
||||
return g;
|
||||
|
|
@ -456,4 +456,23 @@ namespace gtsam {
|
|||
boost::bind(makeLieVector, boost::bind(h, _1, _2, _3)), x1, x2, x3, delta);
|
||||
}
|
||||
|
||||
/**
|
||||
* Compute numerical Hessian matrix. Requires a single-argument Lie->scalar
|
||||
* function. This is implemented simply as the derivative of the gradient.
|
||||
* @param f A function taking a Lie object as input and returning a scalar
|
||||
* @param x The center point for computing the Hessian
|
||||
* @param delta The numerical derivative step size
|
||||
* @return n*n Hessian matrix computed via central differencing
|
||||
*/
|
||||
template<class X>
|
||||
inline Matrix numericalHessian(const boost::function<double(const X&)>& f, const X& x, double delta=1e-5) {
|
||||
return numericalDerivative11<X>(
|
||||
boost::function<LieVector(const X&)>(boost::bind(numericalGradient<X>, f, _1, delta)), x, delta);
|
||||
}
|
||||
|
||||
template<class X>
|
||||
inline Matrix numericalHessian(double (*f)(const X&), const X& x, double delta=1e-5) {
|
||||
return numericalHessian(boost::function<double(const X&)>(f), x, delta);
|
||||
}
|
||||
|
||||
}
|
||||
|
|
|
|||
|
|
@ -0,0 +1,65 @@
|
|||
/* ----------------------------------------------------------------------------
|
||||
|
||||
* GTSAM Copyright 2010, Georgia Tech Research Corporation,
|
||||
* Atlanta, Georgia 30332-0415
|
||||
* All Rights Reserved
|
||||
* Authors: Frank Dellaert, et al. (see THANKS for the full author list)
|
||||
|
||||
* See LICENSE for the license information
|
||||
|
||||
* -------------------------------------------------------------------------- */
|
||||
|
||||
/**
|
||||
* @file testNumericalDerivative.cpp
|
||||
* @brief
|
||||
* @author Richard Roberts
|
||||
* @created Apr 8, 2011
|
||||
*/
|
||||
|
||||
#include <CppUnitLite/TestHarness.h>
|
||||
|
||||
#include <gtsam/base/numericalDerivative.h>
|
||||
|
||||
using namespace gtsam;
|
||||
|
||||
/* ************************************************************************* */
|
||||
double f(const LieVector& x) {
|
||||
assert(x.size() == 2);
|
||||
return sin(x(0)) + cos(x(1));
|
||||
}
|
||||
|
||||
/* ************************************************************************* */
|
||||
TEST_UNSAFE(testNumericalDerivative, numericalHessian) {
|
||||
LieVector center(2, 1.0, 1.0);
|
||||
|
||||
Matrix expected = Matrix_(2,2,
|
||||
-sin(center(0)), 0.0,
|
||||
0.0, -cos(center(1)));
|
||||
|
||||
Matrix actual = numericalHessian(f, center);
|
||||
|
||||
EXPECT(assert_equal(expected, actual, 1e-5));
|
||||
}
|
||||
|
||||
/* ************************************************************************* */
|
||||
double f2(const LieVector& x) {
|
||||
assert(x.size() == 2);
|
||||
return sin(x(0)) * cos(x(1));
|
||||
}
|
||||
|
||||
/* ************************************************************************* */
|
||||
TEST_UNSAFE(testNumericalDerivative, numericalHessian2) {
|
||||
LieVector center(2, 0.5, 1.0);
|
||||
|
||||
Matrix expected = Matrix_(2,2,
|
||||
-cos(center(1))*sin(center(0)), -sin(center(1))*cos(center(0)),
|
||||
-cos(center(0))*sin(center(1)), -sin(center(0))*cos(center(1)));
|
||||
|
||||
Matrix actual = numericalHessian(f2, center);
|
||||
|
||||
EXPECT(assert_equal(expected, actual, 1e-5));
|
||||
}
|
||||
|
||||
/* ************************************************************************* */
|
||||
int main() { TestResult tr; return TestRegistry::runAllTests(tr); }
|
||||
/* ************************************************************************* */
|
||||
Loading…
Reference in New Issue