I'm looking to perform numerical computations for topological materials, including computations involving the Chern number.  So, I came across Z2Pack and it seemed intuitive enough for me to use.  However, as I've been going through examples and changing parameters/trying to replicate results from the literature, I find myself a little confused about the Haldane model example.  I suspect it's some confusion with defining the reciprocal lattice.  In particular, in the Z2Pack example reads in a lattice of k-values, then makes arrays for the nearest neighbor and next nearest neighbor k_a and k_b values, then makes those directly the arguments of cosines and sines to determine the Hamiltonian.  However, a lot of the literature simplifies the sums of those cosines and sines, where I am performing those calculations to ensure I understand.  However, if I directly put in that Hamiltonian with the simplified expressions, (for example assuming k=(kx,ky) and a vector from the honeycomb lattice being (sqrt(3)/2, 1/2), then inner(k,a) = sqrt(3)/2*kx + 1/2*ky), I am unable to get the Chern number to be correct when taking the lattice of k-values that is the argument of the Hamiltonian in Z2Pack.  So, where are the k_a and k_b values coming from? 

Any help would be much appreciated.

Thank you,

Michael Warnock