On 2012-09-22 17:13:24 -0300, firstname.lastname@example.org said:
> Given a 3 column array A of x,y,z numbers, defining a set of spatial > coordinates, and a binary or logical selection vector S (with as many > elements as rows in A) : > > How do I express the operation of selecting an array R with the set of > those co-ordinates in A where S is 1 (or true) ? > > In Apl I write this as R ? S / A . > > But a referee of a paper is asking that I express this in standard > maths, and I have not yet managed to find an appropriate mathematical > expression. > Cheers for any help, > Beau
Let S be the set of indices you want. The you can talk about its entries as subscripted values. To get the rows of A you then needed subscripted subscripts. The question is where do you do the colapsing. S does the job!