Calculation of the total fidelity: product of individual reconstruction fidelities, which will show the fidelity of the output tensor network (result).
Floating-point precision, e.g., exatn-gen:float. Looks like already there, just need to register to the plugin registry.
Information on total circuit depth printed (estimate how many cycles of reconstruction it will take)
Each next reconstructing tensor network should be initialized not to a random value, but to the value of the previous tensor network, this should improve convergence. Random guess is not the best.
Calculation of the total fidelity: product of individual reconstruction fidelities, which will show the fidelity of the output tensor network (result).
Floating-point precision, e.g.,
exatn-gen:float
. Looks like already there, just need to register to the plugin registry.Information on total circuit depth printed (estimate how many cycles of reconstruction it will take)
Each next reconstructing tensor network should be initialized not to a random value, but to the value of the previous tensor network, this should improve convergence. Random guess is not the best.