Search params

Tunable parameters for entropy-guided search.

SearchParams dataclass

SearchParams(
    w_d: float = 1.0,
    w_m: float = 0.1,
    alpha: float = 1.0,
    beta: float = 2.0,
    gamma: float = 0.5,
    top_c: int = 3,
    max_candidates: int = 2,
    reversion_steps: int = 1,
    delta_e: int = 1,
    e_max: int = 4,
    max_goal_candidates: int = 1,
)

Tunable parameters for entropy-guided search.

Attributes:

Name	Type	Description
w_d	float	Distance weight in per-qubit-bus scoring.
w_m	float	Mobility weight in per-qubit-bus scoring.
alpha	float	Distance weight in moveset scoring.
beta	float	Arrived-gain weight in moveset scoring.
gamma	float	Mobility weight in moveset scoring.
top_c	int	Top candidate (move_type, bus_id, direction) triples per qubit.
max_candidates	int	Candidates to try per entropy level before regenerating.
reversion_steps	int	Steps to revert up the tree on deadlock.
delta_e	int	Entropy increment per revisit or failed generation.
e_max	int	Entropy threshold that triggers reversion.
max_goal_candidates	int	Number of goal nodes to collect before stopping.