forte2.helpers.lbfgs
====================

.. py:module:: forte2.helpers.lbfgs




Module Contents
---------------

.. py:class:: LBFGS

   
   Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) optimizer.


   :Parameters:

       **m** : int, optional, default=6
           The number of vectors to keep in memory for the L-BFGS update.

       **epsilon** : float, optional, default=1.0e-5
           Convergence threshold to terminate the minimization: |g| < ε * max(1, |x|).

       **maxiter** : int, optional, default=20
           Maximum number of iterations for the optimization.

       **maxiter_linesearch** : int, optional, default=5
           Maximum number of trials for line search to find the optimal step length.

       **max_dir** : float, optional, default=1.0e15
           Maximum absolute value allowed in the direction vector.

       **line_search_condition** : str, optional, default='strong_wolfe'
           Condition to terminate line search backtracking. Options are 'armijo', 'wolfe',
           or 'strong_wolfe'.

       **step_length_method** : str, optional, default='line_bracketing_zoom'
           Method to determine step lengths. Options are 'max_correction',
           'line_backtracking', or 'line_bracketing_zoom'.

       **min_step** : float, optional, default=1.0e-15
           Minimum step length allowed during line search.

       **max_step** : float, optional, default=1.0e15
           Maximum step length allowed during line search.

       **c1** : float, optional, default=1.0e-4
           Parameter for the Armijo condition in line search.

       **c2** : float, optional, default=0.9
           Parameter for the Wolfe curvature condition in line search.











   .. rubric:: Notes

   Translated into Python from Forte v1: https://github.com/evangelistalab/forte/tree/main/forte/helpers/lbfgs
   For implementation details, see Wikipedia https://en.wikipedia.org/wiki/Limited-memory_BFGS
   and Numerical Optimization 2nd Ed. by Jorge Nocedal and Stephen J. Wright



   ..
       !! processed by numpydoc !!

   .. py:attribute:: m
      :type:  int
      :value: 6



   .. py:attribute:: epsilon
      :type:  float
      :value: 1e-05



   .. py:attribute:: maxiter
      :type:  int
      :value: 20



   .. py:attribute:: maxiter_linesearch
      :type:  int
      :value: 5



   .. py:attribute:: max_dir
      :type:  float
      :value: 1000000000000000.0



   .. py:attribute:: line_search_condition
      :type:  str
      :value: 'strong_wolfe'



   .. py:attribute:: step_length_method
      :type:  str
      :value: 'line_bracketing_zoom'



   .. py:attribute:: min_step
      :type:  float
      :value: 1e-15



   .. py:attribute:: max_step
      :type:  float
      :value: 1000000000000000.0



   .. py:attribute:: c1
      :type:  float
      :value: 0.0001



   .. py:attribute:: c2
      :type:  float
      :value: 0.9



   .. py:attribute:: log_level
      :type:  int


   .. py:attribute:: warn_if_not_converged
      :type:  bool
      :value: False



   .. py:attribute:: dtype
      :type:  type


   .. py:attribute:: converged
      :type:  bool
      :value: False



   .. py:attribute:: iter
      :type:  int
      :value: 0



   .. py:attribute:: p
      :type:  numpy.typing.NDArray


   .. py:method:: minimize(obj, x)

      
      The minimization for the objective function


      :Parameters:

          **obj** : object
              Target function to minimize, it should should be encapsulated in a class that has the following methods:
              ``fx, g = obj.evaluate(x, g, do_g=True)`` where gradient ``g`` is modified by the function,
              ``fx`` is the function return value, and ``g`` is computed when ``do_g==True``.
              If diagonal Hessian is specified, ``h0 = obj.hess_diag(x)`` should be available.

          **x** : NDArray
              The initial value of ``x`` as input, the final value of ``x`` as output.



      :Returns:

          **fx** : float
              The function value at optimized ``x``.











      ..
          !! processed by numpydoc !!


.. py:class:: LBFGS_scipy

   
   A wrapper for the SciPy L-BFGS optimization. For debug use only.
















   ..
       !! processed by numpydoc !!

   .. py:attribute:: epsilon
      :type:  float
      :value: 1e-05



   .. py:attribute:: maxiter
      :type:  int
      :value: 20



   .. py:attribute:: c1
      :type:  float
      :value: 0.0001



   .. py:attribute:: c2
      :type:  float
      :value: 0.9



   .. py:attribute:: print
      :type:  int
      :value: 1



   .. py:attribute:: m
      :type:  int
      :value: 6



   .. py:attribute:: maxiter_linesearch
      :type:  int
      :value: 5



   .. py:attribute:: max_dir
      :type:  float
      :value: 1000000000000000.0



   .. py:attribute:: line_search_condition
      :type:  str
      :value: 'strong_wolfe'



   .. py:attribute:: step_length_method
      :type:  str
      :value: 'line_bracketing_zoom'



   .. py:attribute:: min_step
      :type:  float
      :value: 1e-15



   .. py:attribute:: max_step
      :type:  float
      :value: 1000000000000000.0



   .. py:method:: minimize(obj, x)


.. py:class:: NewtonRaphson

   .. py:attribute:: epsilon
      :type:  float
      :value: 1e-05



   .. py:attribute:: maxiter
      :type:  int
      :value: 20



   .. py:attribute:: c1
      :type:  float
      :value: 0.0001



   .. py:attribute:: c2
      :type:  float
      :value: 0.9



   .. py:attribute:: print
      :type:  int
      :value: 1



   .. py:attribute:: m
      :type:  int
      :value: 6



   .. py:attribute:: maxiter_linesearch
      :type:  int
      :value: 5



   .. py:attribute:: max_dir
      :type:  float
      :value: 1000000000000000.0



   .. py:attribute:: line_search_condition
      :type:  str
      :value: 'strong_wolfe'



   .. py:attribute:: step_length_method
      :type:  str
      :value: 'line_bracketing_zoom'



   .. py:attribute:: min_step
      :type:  float
      :value: 1e-15



   .. py:attribute:: max_step
      :type:  float
      :value: 1000000000000000.0



   .. py:method:: minimize(obj, x)