Ok... so here's a routine which gives us all the truth valued arrays
with n axes:
   rntt=: (#2:) $"1 [: #:@i. 2>. 2^2*]

And, here's a test to see if an n-dimensional truth table is degenerate:
   degenerate=: 1 e. (1 ]\. i.@#@$) -:/@|: ]

(Use with "_1 on a result from rntt)

But I don't understand what a dual would be for an arbitrary truth
valued array with an arbitrary number axes, so I'm not quite sure how
I would write that test.

For example, if I implement:

   Glock=: $ #: I.@,
   dual=: (-:&(/:~) -.)&Glock
   selfdual=: dual~

I find four self-dual arrays of rank 2:
   (#~ selfdual"_1) rntt 2
0 0
0 0

0 1
1 0

1 0
0 1

1 1
1 1

What's a dual of a truth table array with arbitrary axes?

Thanks,

-- 
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to