BUGFIX: factorized matrix are not symetric, cholesky factorization should process different types for covariance and decomposition + better format output for cholesky test
This commit is contained in:
parent
e0f691c148
commit
fe2cebc0e8
2 changed files with 110 additions and 54 deletions
|
|
@ -49,7 +49,7 @@ public:
|
|||
|
||||
|
||||
/** Cholesky decomposition, given a matrix V, return a matrix L
|
||||
* such as V = L Lt (Lt being the conjugate transpose of L).
|
||||
* such as V = L L^T (L^T being the transposed of L).
|
||||
*
|
||||
* Need a symmetric and positive definite matrix as an input, which
|
||||
* should be the case of a non-ill-conditionned covariance matrix.
|
||||
|
|
@ -58,7 +58,8 @@ public:
|
|||
class Cholesky
|
||||
{
|
||||
public:
|
||||
typedef ublas::symmetric_matrix< AtomType, ublas::lower > MatrixType;
|
||||
typedef ublas::symmetric_matrix< AtomType, ublas::lower > CovarMat;
|
||||
typedef ublas::matrix< AtomType > FactorMat;
|
||||
|
||||
enum Method {
|
||||
//! use the standard algorithm, with square root @see factorize_LLT
|
||||
|
|
@ -82,7 +83,8 @@ public:
|
|||
*
|
||||
* Use the standard unstable method by default.
|
||||
*/
|
||||
Cholesky(const MatrixType& V, Cholesky::Method use = standard ) : _use(use)
|
||||
Cholesky(const CovarMat& V, Cholesky::Method use = standard ) :
|
||||
_use(use), _L(ublas::zero_matrix<AtomType>(V.size1(),V.size2()))
|
||||
{
|
||||
factorize( V );
|
||||
}
|
||||
|
|
@ -90,14 +92,14 @@ public:
|
|||
|
||||
/** Compute the factorization and return the result
|
||||
*/
|
||||
const MatrixType& operator()( const MatrixType& V )
|
||||
const FactorMat& operator()( const CovarMat& V )
|
||||
{
|
||||
factorize( V );
|
||||
return decomposition();
|
||||
}
|
||||
|
||||
//! The decomposition of the covariance matrix
|
||||
const MatrixType & decomposition() const
|
||||
const FactorMat & decomposition() const
|
||||
{
|
||||
return _L;
|
||||
}
|
||||
|
|
@ -105,18 +107,18 @@ public:
|
|||
protected:
|
||||
|
||||
//! The decomposition is a (lower) symetric matrix, just like the covariance matrix
|
||||
MatrixType _L;
|
||||
FactorMat _L;
|
||||
|
||||
/** Assert that the covariance matrix have the required properties and returns its dimension.
|
||||
*
|
||||
* Note: if compiled with NDEBUG, will not assert anything and just return the dimension.
|
||||
*/
|
||||
unsigned assert_properties( const MatrixType& V )
|
||||
unsigned assert_properties( const CovarMat& V )
|
||||
{
|
||||
unsigned int Vl = V.size1(); // number of lines
|
||||
|
||||
// the result goes in _L
|
||||
_L.resize(Vl);
|
||||
_L = ublas::zero_matrix<AtomType>(Vl,Vl);
|
||||
|
||||
#ifndef NDEBUG
|
||||
assert(Vl > 0);
|
||||
|
|
@ -135,7 +137,6 @@ public:
|
|||
* perform the cholesky factorization
|
||||
* check if all eigenvalues are positives
|
||||
* check if all of the leading principal minors are positive
|
||||
*
|
||||
*/
|
||||
#endif
|
||||
|
||||
|
|
@ -146,7 +147,7 @@ public:
|
|||
/** Actually performs the factorization with the method given at
|
||||
* instanciation. Results are cached in _L.
|
||||
*/
|
||||
void factorize( const MatrixType& V )
|
||||
void factorize( const CovarMat& V )
|
||||
{
|
||||
if( _use == standard ) {
|
||||
factorize_LLT( V );
|
||||
|
|
@ -161,10 +162,12 @@ public:
|
|||
/** This standard algorithm makes use of square root and is thus subject
|
||||
* to round-off errors if the covariance matrix is very ill-conditioned.
|
||||
*
|
||||
* Compute L such that V = L L^T
|
||||
*
|
||||
* When compiled in debug mode and called on ill-conditionned matrix,
|
||||
* will raise an assert before calling the square root on a negative number.
|
||||
*/
|
||||
void factorize_LLT( const MatrixType& V)
|
||||
void factorize_LLT( const CovarMat& V)
|
||||
{
|
||||
unsigned int N = assert_properties( V );
|
||||
|
||||
|
|
@ -210,7 +213,7 @@ public:
|
|||
* Be aware that this increase round-off errors, this is just a ugly
|
||||
* hack to avoid crash.
|
||||
*/
|
||||
void factorize_LLT_abs( const MatrixType & V)
|
||||
void factorize_LLT_abs( const CovarMat & V)
|
||||
{
|
||||
unsigned int N = assert_properties( V );
|
||||
|
||||
|
|
@ -247,19 +250,21 @@ public:
|
|||
}
|
||||
|
||||
|
||||
/** This alternative algorithm do not use square root.
|
||||
/** This alternative algorithm do not use square root in an inner loop,
|
||||
* but only for some diagonal elements of the matrix D.
|
||||
*
|
||||
* Computes L and D such as V = L D Lt
|
||||
* Computes L and D such as V = L D L^T.
|
||||
* The factorized matrix is (L D^1/2), because V = (L D^1/2) (L D^1/2)^T
|
||||
*/
|
||||
void factorize_LDLT( const MatrixType& V)
|
||||
void factorize_LDLT( const CovarMat& V)
|
||||
{
|
||||
// use "int" everywhere, because of the "j-1" operation
|
||||
int N = assert_properties( V );
|
||||
// example of an invertible matrix whose decomposition is undefined
|
||||
assert( V(0,0) != 0 );
|
||||
|
||||
MatrixType L(N,N);
|
||||
MatrixType D = ublas::zero_matrix<AtomType>(N,N);
|
||||
FactorMat L = ublas::zero_matrix<AtomType>(N,N);
|
||||
FactorMat D = ublas::zero_matrix<AtomType>(N,N);
|
||||
D(0,0) = V(0,0);
|
||||
|
||||
for( int j=0; j<N; ++j ) { // each columns
|
||||
|
|
@ -281,12 +286,12 @@ public:
|
|||
} // for i in rows
|
||||
} // for j in columns
|
||||
|
||||
// now compute the final symetric matrix: from _LD_LT to _L_LT
|
||||
// remember that V = (_LD^1/2)(_LD^1/2)^T
|
||||
// now compute the final symetric matrix: _L = L D^1/2
|
||||
// remember that V = ( L D^1/2) ( L D^1/2)^T
|
||||
|
||||
// square root of a diagonal matrix is the square root of all its
|
||||
// scalars
|
||||
MatrixType D12 = D;
|
||||
// fortunately, the square root of a diagonal matrix is the square
|
||||
// root of all its elements
|
||||
FactorMat D12 = D;
|
||||
for(int i=0; i<N; ++i) {
|
||||
D12(i,i) = sqrt(D(i,i));
|
||||
}
|
||||
|
|
@ -295,7 +300,6 @@ public:
|
|||
_L = ublas::prod( L, D12);
|
||||
}
|
||||
|
||||
|
||||
}; // class Cholesky
|
||||
|
||||
|
||||
|
|
@ -309,14 +313,10 @@ public:
|
|||
unsigned int size = distrib.size();
|
||||
assert(size > 0);
|
||||
|
||||
// Cholesky factorisation gererating matrix L from covariance
|
||||
// matrix V.
|
||||
// We must use cholesky.decomposition() to get the resulting matrix.
|
||||
//
|
||||
// L = cholesky decomposition of varcovar
|
||||
const typename Cholesky::MatrixType& L = _cholesky( distrib.varcovar() );
|
||||
const typename Cholesky::FactorMat& L = _cholesky( distrib.varcovar() );
|
||||
|
||||
// T = vector of size elements drawn in N(0,1) rng.normal(1.0)
|
||||
// T = vector of size elements drawn in N(0,1)
|
||||
ublas::vector< AtomType > T( size );
|
||||
for ( unsigned int i = 0; i < size; ++i ) {
|
||||
T( i ) = rng.normal();
|
||||
|
|
|
|||
Reference in a new issue