diff options
| author |  Gael Guennebaud <g.gael@free.fr> | 2008-06-07 22:47:11 +0000 |
|---|---|---|
| committer | Gael Guennebaud <g.gael@free.fr> | 2008-06-07 22:47:11 +0000 |
| commit | 4dd57b585d2cd33afcd5900eb740250115e03c4a (patch) | |
| tree | 3fb1fbcc542771388551af764e43bc9954b373f8 /Eigen/src/Core/Fuzzy.h | |
| parent | eb7b7b2cfc4170d5a4697f3be08c1af0559ce25f (diff) | |
* rewrite of the QR decomposition:
- works for complex
- allows direct access to the matrix R
* removed the scale by the matrix dimensions in MatrixBase::isMuchSmallerThan(scalar)
Diffstat (limited to 'Eigen/src/Core/Fuzzy.h')
| -rw-r--r-- | Eigen/src/Core/Fuzzy.h | 7 |
1 files changed, 5 insertions, 2 deletions
diff --git a/Eigen/src/Core/Fuzzy.h b/Eigen/src/Core/Fuzzy.h index e689fc913..0b2305c57 100644 --- a/Eigen/src/Core/Fuzzy.h +++ b/Eigen/src/Core/Fuzzy.h @@ -63,7 +63,10 @@ bool MatrixBase<Derived>::isApprox( * \note The fuzzy compares are done multiplicatively. A vector \f$ v \f$ is * considered to be much smaller than \f$ x \f$ within precision \f$ p \f$ if * \f[ \Vert v \Vert \leqslant p\,\vert x\vert. \f] - * For matrices, the comparison is done using the Hilbert-Schmidt norm. + * + * For matrices, the comparison is done using the Hilbert-Schmidt norm. For this reason, + * the value of the reference scalar \a other should come from the Hilbert-Schmidt norm + * of a reference matrix of same dimensions. * * \sa isApprox(), isMuchSmallerThan(const MatrixBase<OtherDerived>&, RealScalar) const */ @@ -73,7 +76,7 @@ bool MatrixBase<Derived>::isMuchSmallerThan( typename NumTraits<Scalar>::Real prec ) const { - return cwiseAbs2().sum() <= prec * prec * other * other * cols() * rows(); + return cwiseAbs2().sum() <= prec * prec * other * other; } /** \returns \c true if the norm of \c *this is much smaller than the norm of \a other, |