Is there a better way to implement matlab's Logical Indexing using Eigen/C++?

[poster]

I've been looking all over for a more "eigen" way of implementing the functionality of Matlab's Logical indexing. Here's the best I could come up with. (Focusing on an int array here for simplicity)

//an attempt at matlab-style Logical Indexing

//equivalent to the matlab:

//  original = [1,2,3,4]

//  subset = original(original < 3)



using namespace Eigen;

using std::cout;

using std::endl;



IOFormat OctaveFmt(StreamPrecision, 0, ", ", " ", "", "", "[", "]");



ArrayXi original(4);

original << 1,2,3,4;

cout<<"Original with bad values:"<
    <


Array selections = original < 3;

cout<<"One if it's a good value:"<
    <


std::vector picked;

for(int i = 0; i < selections.size(); i++ )

{

    if(selections(i))

    {

        picked.push_back(original(i));

    }

}



//put the vector values back into an eigen array

ArrayXi theGoodStuff = Map

                       (picked.data(), picked.size());



cout<<"Just the good stuff:"<
    <

Here's the output I get:

Original with bad values:

[1  2  3  4]

One if it's a good value:

[1  1  0  0]

Just the good stuff:

[1  2]

Does anyone know how to do this in a more 'eigen' way, or just a faster way than looping through the arrays?

More answer...