octave_idx_type is the correct index type in this context. What it the purpose of your proposed changes?

These are the types of warnings I'm seeing:

editDistance.cc:189:35: warning: comparison of integer expressions of different signedness: ‘octave_idx_type’ {aka ‘long int’} and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
editDistance.cc:203:21: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::vector<int> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]

You would think that octave_idx_type would be an appropriate index type, but it's signed instead of unsigned.

The more concerning warnings are these:

fcnn.cpp:127:17: warning: ‘i’ is used uninitialized [-Wuninitialized]
fcnn.cpp:213:15: warning: ‘i’ is used uninitialized [-Wuninitialized]

The more concerning warnings are these:

fcnn.cpp:127:17: warning: ‘i’ is used uninitialized [-Wuninitialized]
fcnn.cpp:213:15: warning: ‘i’ is used uninitialized [-Wuninitialized]

Thanks for catching this

I also made changes to editDistance to consistently use size_t instead of interchangeable using octave_idx_type and/or int with methods returning size_t. I assume that casting octave_idx_type (usually returned by methods of Octave classes, eg numel()) to size_t does not produce any warning.
Can you check whether any warning are emitted yet?