TimePropagationApproachKrylovConfig#

class cuquantum.densitymat.TimePropagationApproachKrylovConfig(
tolerance: float | None = None,
max_dim: int | None = None,
min_beta: float | None = None,
adaptive_step_size: int | None = None,
)[source]#

Configuration for the Krylov subspace time propagation approach.

Parameters:

  • tolerance – Convergence tolerance. Defaults to 0 when None, resolved to machine epsilon of the compute precision.

  • max_dim – Maximum Krylov subspace dimension. Defaults to 30 when None.

  • min_beta – Minimum beta to proceed with expansion. Defaults to 0 when None, resolved to machine epsilon of the compute precision.

  • adaptive_step_size – Enable adaptive step size control (0 = disabled, 1 = enabled). Defaults to 1 (enabled) when None.