LBFGSB

class pmrf.optimize.LBFGSB(gtol: float = 0.001, stepsize: float = 1.0, linesearch: str = 'zoom')

Bases: AbstractBoundedMinimizer

A L-BFGS-B optimizer in JAX.

Wrapper around jaxopt.LBFGSB.

Parameters:

  • gtol (float, default=1e-3) – The gradient norm tolerance for termination.

  • stepsize (float, default=1.0) – Initial step size for the line search.

  • linesearch (str, default="zoom") – Type of line search to use.

run(fn: Callable[[PyTree, Any], Any], y0: PyTree, args: Any = None, bounds: tuple[PyTree, PyTree] | None = None, max_iter: int = 1024, **kwargs) → tuple[MinimizeResult, PyTree]

Execute the minimization algorithm.

Parameters:

  • fn (callable) – The objective function to minimize.

  • y0 (PyTree) – The initial parameter guess.

  • args (Any) – Args to pass to fn.

  • bounds (PyTree) – Bounds for y0, if any.

  • max_iter (int = 1024) – The maximum number of iterations to take.

  • **kwargs – Runtime arguments forward to the solver backend.

Returns:

A tuple of (pmrf.optimize.MinimizeResult, metrics)`.

Return type:

tuple

gtol: float = 0.001
linesearch: str = 'zoom'
stepsize: float = 1.0