04-16-2012 02:23 PM
@ggqshqs wrote:
thanks for your answer ben, but could you explain it in another way because i don't really unerstand it very well
Not without doing the work for you and that would not do you any good LV-wise.
If you don't follow my appraoch then you should develop your own.
Sit down and start "doing the work" by hand with some scratch paper to take notes on any intermidate stuff you need to remember and write down exactly what you are doing so that you have a manual procedure you can hand to a buddy. If the procedure is correct your buddy should be able repeat the steps you outline.
After you have a method outlined that should work manually, THEN turn to converting it into code.
Ben
04-16-2012 03:00 PM - edited 04-16-2012 03:04 PM
Sorry guys, The problem seemed rather interesting so I coded it up. Altenbach's tip on array max-replace was a nice pointer NaN's self inequality proved interesting too.
Yes, I know a FBN would be more elelgant than both SRs.
04-16-2012 03:04 PM
If all of the values are either zero or one why is your data array DBL .... wait! that was supposed to be Christian's line.
Sorry.
Ben
04-16-2012 03:06 PM - edited 04-16-2012 03:10 PM
@Ben wrote:
If all of the values are either zero or one why is your data array DBL .... wait! that was supposed to be Christian's line.
Sorry.
Ben
so you can do the self equality check rather than summing the 3x3 array
And there I go begging fr a RG nomination
04-16-2012 04:10 PM
@Jeff Bohrer wrote:
Sorry guys, The problem seemed rather interesting so I coded it up. Altenbach's tip on array max-replace was a nice pointer NaN's self inequality proved interesting too.
Yes, I know a FBN would be more elelgant than both SRs.
Of course there are some shapes this doesn't quite work with. X's H's.
Try this instead