I use m/ it sometimes. I think most places I've used it, I could have
used the insert-from-the back definition equally well. I'm sure there
are good cases for each definition.
For my optimization problem, +`<./\. is a no-go, because it starts a
verb on each scalar, which is a lot of overhead.
Henry Rich
On 2/22/2016 5:16 PM, Raul Miller wrote:
Note also that if (3*4+5) -: +`*/3 4 5 this gerund mechanism would be
useful for implementing a solution to your minimization problem, using
something like
C#inv!._1 (#~ 0 1$~#)|.+`<./\.|. (C=:D~:_1)#D,.P
That's untested code, for obvious reasons, but I think it would be in
the right ballpark.
So my question is: is the existing implementation useful for anything?
Thanks,
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
