Skip to content

Allowing more flexible parametrization of the cost estimate functions #768

Description

@PabloAMC

In the recent cost estimates functions, there are a number of hardcoded parameters in the system:

for logical_data_qubit_distance in range(7, 35, 2):
params = AlgorithmParameters(
physical_error_rate=physical_error_rate,
surface_code_cycle_time=datetime.timedelta(microseconds=1),
logical_data_qubit_distance=logical_data_qubit_distance,
magic_state_factory=factory,
toffoli_count=num_toffoli,
max_allocated_logical_qubits=num_logical_qubits,
factory_count=4,
routing_overhead_proportion=0.5,
proportion_of_bounding_box=portion_of_bounding_box)
cost = params.estimate_cost()
if cost.algorithm_failure_probability > 0.1:

Should we add them as parameters to some class that the user can willingly change?

Metadata

Metadata

Assignees

No one assigned

    Labels

    area/functionalityInvolves the features and operations implemented in OpenFermionpriority/reviewReassess the priority of this item at the next opportunity

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions