This question requires the knowledge of GRAPE. More info can be found here and in this short presentation. In the following jupyter notebook the result of this line
_overlap(result.U_f, U_f_numerical).real, abs(_overlap(result.U_f, U_f_numerical))**2
(which is in the "Test numerical integration of GRAPE pulse") the result obtained seems very low.
Why would the unitary obtained by the propagator function "U_f_numerical" be so different from that derived from the cy_grape_unitary function "result.U_f"?
They are both derived from "result.H_t". I am saying that they seem to be different based on the low value "0.11355125771393167" of the _overlap function. I may be totally wrong about this.
I have raised this as an issue