Ok, how would I form the cartesian cross-product of these 4 vectors:
age =: i. 2 NB. young, pre-presbyopic, presbyopic
spec =: i. 1 NB. myope, hypermetrope
astig =: i. 1 NB. no, yes
tear =: i. 1 NB. reduced, normal
I presume the resulting cross product will be a rank-4 array. Call it r4a
Ok, now what if I wanted to take one "slice" of this array and pick off the age,
spec, astig, and tear for this slice using
age_i =: [ spec_i =: 0 [ astig_i =: 0 [ tear_i =: 0
tear =: tear_i { ( astig_i { ( spec_i { ( age_i { r4a ) ) )
But there must be a simpler way to extract each value at the indices age_i,
spec_i, astig_i and tear_i
Ok, so given that age_i is zero, how would we come up with English for each
value on each axis?
Here, I store the English associations for the indexes:
age_tbl =: 3 2 $ ( 0;'young' ) , (1;'pre-presbyopic') , ( 2;'presbyopic' )
NB. how to avoid having to specify the "depth" of the array. In other words,
NB. how to say take this data and make it into an n by 2 array?
I'm sorry about being impatient, if someone wants to refer me to the appropriate
chapter in "Learning J" I will go study it myself.
I happen to be reading a book on Machine Learning and want to munge my data with
J instead of Perl.
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
