Hi Ivo and all others, On 18 Jul., 18:27, Ivo Hedtke <hed...@me.com> wrote: > If the cutoff-parameter is determined by the benchmark: Yes, submit it!
The cutoff-parameter is not hard-wired. So, everyone can find his/her own. > I look forward to review your patch. The patch is at trac ticket #11610, together with some timings and memory data. Thank you for your pointer to the article: It was straight forward to implement a schedule from the article; I chose a non-destructive version (i.e., the input matrices are preserved). With that schedule, Strassen-Winograd could be a real alternative to the default multiplication (school book) of sage.matrix.matrix_modn_dense.Matrix_modn_dense. Cheers, Simon -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
