[Scipy-svn] r6674 - trunk/scipy/sparse/linalg/isolve

scipy-svn at scipy.org scipy-svn at scipy.org
Sat Sep 4 18:53:26 EDT 2010


Author: ptvirtan
Date: 2010-09-04 17:53:26 -0500 (Sat, 04 Sep 2010)
New Revision: 6674

Modified:
   trunk/scipy/sparse/linalg/isolve/iterative.py
Log:
BUG: sparse.linalg: clarify that the sparse solver tolerance is both absolute and relative

Modified: trunk/scipy/sparse/linalg/isolve/iterative.py
===================================================================
--- trunk/scipy/sparse/linalg/isolve/iterative.py	2010-09-04 20:54:05 UTC (rev 6673)
+++ trunk/scipy/sparse/linalg/isolve/iterative.py	2010-09-04 22:53:26 UTC (rev 6674)
@@ -26,7 +26,8 @@
 x0  : {array, matrix}
     Starting guess for the solution.
 tol : float
-    Relative tolerance to achieve before terminating.
+    Tolerance to achieve. The algorithm terminates when either the relative
+    or the absolute residual is below `tol`.
 maxiter : integer
     Maximum number of iterations.  Iteration will stop after maxiter
     steps even if the specified tolerance has not been achieved.
@@ -318,8 +319,9 @@
     x0  : {array, matrix}
         Starting guess for the solution.
     tol : float
-        Relative tolerance to achieve before terminating.
-    restart : integer, optional
+        Tolerance to achieve. The algorithm terminates when either the relative
+        or the absolute residual is below `tol`.
+    restart : integer
         Number of iterations between restarts. Larger values increase
         iteration cost, but may be necessary for convergence.
         (Default: 20)
@@ -463,7 +465,8 @@
     x0  : {array, matrix}
         Starting guess for the solution.
     tol : float
-        Relative tolerance to achieve before terminating.
+        Tolerance to achieve. The algorithm terminates when either the relative
+        or the absolute residual is below `tol`.
     maxiter : integer
         Maximum number of iterations.  Iteration will stop after maxiter
         steps even if the specified tolerance has not been achieved.




More information about the Scipy-svn mailing list