Open discussion regarding features, bugs, issues, vendors, etc.
by mpakala » Thu Jan 04, 2007 6:05 pm
Hi
I was wondering if I can use matrices with integer data with CBLAS level 3 routines. Is it possible to do this? If so, how ? Please let me know.
thanks
kiran
-
mpakala
-
- Posts: 6
- Joined: Thu Nov 09, 2006 7:57 pm
by Julien Langou » Thu Jan 04, 2007 6:09 pm
No way. You need to convert the input data to floating-point representation. Julien.
-
Julien Langou
-
- Posts: 835
- Joined: Thu Dec 09, 2004 12:32 pm
- Location: Denver, CO, USA
-
by mpakala » Thu Jan 04, 2007 6:18 pm
hi julien
thanks..so i can type cast the int to a higher data type i.e. supported by cblas??
regards
kiran
-
mpakala
-
- Posts: 6
- Joined: Thu Nov 09, 2006 7:57 pm
by Julien Langou » Thu Jan 04, 2007 6:24 pm
Yep you cast your int matrix element by element in double (or single) format. Then you call CBLAS and the output will be double (or single), then you cast back to int. There is no igemm :) in the BLAS.
-
Julien Langou
-
- Posts: 835
- Joined: Thu Dec 09, 2004 12:32 pm
- Location: Denver, CO, USA
-
Return to User Discussion
Who is online
Users browsing this forum: No registered users and 6 guests