You can model dyad ? as
dyadquery =. (] , ({~ ?...@#)@-.)^:((i...@])`[`($...@0:))
Henry Rich
John Randall wrote:
> Thanks to all for your suggestions.
>
> david alis wrote:
>> I wonder what led John to think of his solution
>> because it is curiously out of character for someone
>> who's been using J for such a long time.
>
> I was working under a restriction I should have stated: you can't use
> dyadic ? . This led me to either accumulating values and checking for
> uniqueness, or making all selections at once using combination index.
>
>
> Roger Hui wrote:
>>> pick2"0 i.5
>>> 2 32 40 45 48 1
>>> 7 28 29 35 54 11
>>> 1 26 31 41 55 3
>>> 3 11 29 41 43 32
>>> 12 20 24 27 41 9
>> John, how can it be random when the first
>> five numbers are always in order? ;-)
>
>
> Sorting lottery ticket numbers is often misunderstood. Students learn
> to use permutations when order matters and combinations when order
> does not matter. The NJ Pick-6 lottery uses 6 numbers from 1-49. The
> number of distinct tickets is 6!49, not (!49x)%(!43x), a common mistake.
>
>
> Best wishes,
>
> John
>
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
