How to handle Arrays that has functional parameters:

• To: mathgroup at smc.vnet.net
• Subject: [mg68924] How to handle Arrays that has functional parameters:
• From: Gopinath Venkatesan <gopinathv at ou.edu>
• Date: Wed, 23 Aug 2006 07:15:57 -0400 (EDT)
• Sender: owner-wri-mathgroup at wolfram.com

Dear Friends

If we have to define an array but also work as a function, how do we define them?

For example, consider the following:

\!\(\(func1 = Table[a[i, j], {i, 3}, {j, 3}];\)\[IndentingNewLine]
\(func2[s_, t_, i_, j_] := \(s\^i\) t\^j;\)\[IndentingNewLine]
\(Do[func1[\([i, j]\)] = func2[s, t,
i, j], {i, 3}, {j, 3}];\)\[IndentingNewLine]
func1 // MatrixForm\)

Where I defined func1 to be an array, and func2 to be a function, and then assigned the func1 elements using func2 relation.

I wanted to know if we can build this relation directly into the array definition?

Like say, func1[[i,j]][s_,t_]:=s^i t^j, inside the Do loop, or other alternatives.

The reason is I have some functions of a variable, which I dont want to evaluate until the last step of the solution - i.e., I dont want to evaluate them during the do loop operations, and keep it safe and then evaluate in another do loop operation where only the variables will change to evaluate the lagrangian and hence the individual solution for the variable value. I can use the following method, it takes some time to fix it, but would like to know if we can do it some other way.

Thanks,

Gopinath Venkatesan