From 0d36145952f8e077b6bd056114b6be003152e693 Mon Sep 17 00:00:00 2001 From: Julien Langou Date: Thu, 24 Nov 2016 09:41:52 +0100 Subject: [PATCH] defficient spells deficient --- SRC/cgejsv.f | 6 +++--- SRC/dgejsv.f | 6 +++--- SRC/sgejsv.f | 6 +++--- SRC/zgejsv.f | 6 +++--- 4 files changed, 12 insertions(+), 12 deletions(-) diff --git a/SRC/cgejsv.f b/SRC/cgejsv.f index f69ab9d..0641e42 100644 --- a/SRC/cgejsv.f +++ b/SRC/cgejsv.f @@ -85,7 +85,7 @@ *> rows, then using this condition number gives too pessimistic *> error bound. *> = 'A': Small singular values are the noise and the matrix is treated -*> as numerically rank defficient. The error in the computed +*> as numerically rank deficient. The error in the computed *> singular values is bounded by f(m,n)*epsilon*||A||. *> The computed SVD A = U * S * V^* restores A up to *> f(m,n)*epsilon*||A||. @@ -469,7 +469,7 @@ *> The rank revealing QR factorization (in this code: CGEQP3) should be *> implemented as in [3]. We have a new version of CGEQP3 under development *> that is more robust than the current one in LAPACK, with a cleaner cut in -*> rank defficient cases. It will be available in the SIGMA library [4]. +*> rank deficient cases. It will be available in the SIGMA library [4]. *> If M is much larger than N, it is obvious that the initial QRF with *> column pivoting can be preprocessed by the QRF without pivoting. That *> well known trick is not used in CGEJSV because in some cases heavy row @@ -1021,7 +1021,7 @@ ELSE IF ( L2RANK ) THEN * .. similarly as above, only slightly more gentle (less agressive). * Sudden drop on the diagonal of R1 is used as the criterion for -* close-to-rank-defficient. +* close-to-rank-deficient. TEMP1 = SQRT(SFMIN) DO 3401 p = 2, N IF ( ( ABS(A(p,p)) .LT. (EPSLN*ABS(A(p-1,p-1))) ) .OR. diff --git a/SRC/dgejsv.f b/SRC/dgejsv.f index 3488f26..64f0908 100644 --- a/SRC/dgejsv.f +++ b/SRC/dgejsv.f @@ -87,7 +87,7 @@ *> rows, then using this condition number gives too pessimistic *> error bound. *> = 'A': Small singular values are the noise and the matrix is treated -*> as numerically rank defficient. The error in the computed +*> as numerically rank deficient. The error in the computed *> singular values is bounded by f(m,n)*epsilon*||A||. *> The computed SVD A = U * S * V^t restores A up to *> f(m,n)*epsilon*||A||. @@ -428,7 +428,7 @@ *> The rank revealing QR factorization (in this code: DGEQP3) should be *> implemented as in [3]. We have a new version of DGEQP3 under development *> that is more robust than the current one in LAPACK, with a cleaner cut in -*> rank defficient cases. It will be available in the SIGMA library [4]. +*> rank deficient cases. It will be available in the SIGMA library [4]. *> If M is much larger than N, it is obvious that the initial QRF with *> column pivoting can be preprocessed by the QRF without pivoting. That *> well known trick is not used in DGEJSV because in some cases heavy row @@ -967,7 +967,7 @@ ELSE IF ( L2RANK ) THEN * .. similarly as above, only slightly more gentle (less agressive). * Sudden drop on the diagonal of R1 is used as the criterion for -* close-to-rank-defficient. +* close-to-rank-deficient. TEMP1 = DSQRT(SFMIN) DO 3401 p = 2, N IF ( ( DABS(A(p,p)) .LT. (EPSLN*DABS(A(p-1,p-1))) ) .OR. diff --git a/SRC/sgejsv.f b/SRC/sgejsv.f index 4054a59..a52e39b 100644 --- a/SRC/sgejsv.f +++ b/SRC/sgejsv.f @@ -87,7 +87,7 @@ *> rows, then using this condition number gives too pessimistic *> error bound. *> = 'A': Small singular values are the noise and the matrix is treated -*> as numerically rank defficient. The error in the computed +*> as numerically rank deficient. The error in the computed *> singular values is bounded by f(m,n)*epsilon*||A||. *> The computed SVD A = U * S * V^t restores A up to *> f(m,n)*epsilon*||A||. @@ -428,7 +428,7 @@ *> The rank revealing QR factorization (in this code: SGEQP3) should be *> implemented as in [3]. We have a new version of SGEQP3 under development *> that is more robust than the current one in LAPACK, with a cleaner cut in -*> rank defficient cases. It will be available in the SIGMA library [4]. +*> rank deficient cases. It will be available in the SIGMA library [4]. *> If M is much larger than N, it is obvious that the initial QRF with *> column pivoting can be preprocessed by the QRF without pivoting. That *> well known trick is not used in SGEJSV because in some cases heavy row @@ -967,7 +967,7 @@ ELSE IF ( L2RANK ) THEN * .. similarly as above, only slightly more gentle (less agressive). * Sudden drop on the diagonal of R1 is used as the criterion for -* close-to-rank-defficient. +* close-to-rank-deficient. TEMP1 = SQRT(SFMIN) DO 3401 p = 2, N IF ( ( ABS(A(p,p)) .LT. (EPSLN*ABS(A(p-1,p-1))) ) .OR. diff --git a/SRC/zgejsv.f b/SRC/zgejsv.f index 95da5e6..fa85af0 100644 --- a/SRC/zgejsv.f +++ b/SRC/zgejsv.f @@ -85,7 +85,7 @@ *> rows, then using this condition number gives too pessimistic *> error bound. *> = 'A': Small singular values are the noise and the matrix is treated -*> as numerically rank defficient. The error in the computed +*> as numerically rank deficient. The error in the computed *> singular values is bounded by f(m,n)*epsilon*||A||. *> The computed SVD A = U * S * V^* restores A up to *> f(m,n)*epsilon*||A||. @@ -470,7 +470,7 @@ *> The rank revealing QR factorization (in this code: ZGEQP3) should be *> implemented as in [3]. We have a new version of ZGEQP3 under development *> that is more robust than the current one in LAPACK, with a cleaner cut in -*> rank defficient cases. It will be available in the SIGMA library [4]. +*> rank deficient cases. It will be available in the SIGMA library [4]. *> If M is much larger than N, it is obvious that the initial QRF with *> column pivoting can be preprocessed by the QRF without pivoting. That *> well known trick is not used in ZGEJSV because in some cases heavy row @@ -1026,7 +1026,7 @@ ELSE IF ( L2RANK ) THEN * .. similarly as above, only slightly more gentle (less agressive). * Sudden drop on the diagonal of R1 is used as the criterion for -* close-to-rank-defficient. +* close-to-rank-deficient. TEMP1 = DSQRT(SFMIN) DO 3401 p = 2, N IF ( ( ABS(A(p,p)) .LT. (EPSLN*ABS(A(p-1,p-1))) ) .OR. -- 2.7.4