note the use of transpose v in benchmark
add comment to tutorial
add comment to speed faq
add comment to speed faq
Leave a comment
on 2010-11-10 21:29 *
By james.bergstra
This one is a reminder that the physical transposed-ness of weight matrices (fortran vs c order) can affect the speed of evaluation. I found that the fortran order made the logistic regression example go faster. But for another thing it didn't help.
If we had a list of standard tricks for optimizing computations, then this would be an addition to it.
If we had a list of standard tricks for optimizing computations, then this would be an addition to it.