ALE for second-order effects

[jklaise]

Current ALE implementation only calculates first order effects, an extension to second order effects is possible.

If n is the number of features then:

• In ALE all n main effects are computed automatically, with 2nd order there are n(n-1)/2 of these, how do we specify which ones we want? This may involve a change to the input API as well as the output Explanation

[jklaise]

Design of the interface would be important especially if we implement PDP for both 1-d and 2-d use cases.