.. smooth documentation master file, created by
   sphinx-quickstart on Fri Jan 16 12:46:32 2026.
   You can adapt this file completely to your liking, but it should at least
   contain the root `toctree` directive.

smooth documentation
====================

.. currentmodule:: smooth

Classes
-------

.. autosummary::
   :toctree: _autosummary
   :template: class.rst

   ADAM
   ES

ADAM Methods
------------

.. autosummary::
   :toctree: _autosummary
   :template: method.rst

   ADAM.fit
   ADAM.predict
   ADAM.predict_intervals
   ADAM.select_best_model
   ADAM.summary

ES Methods
----------

.. autosummary::
   :toctree: _autosummary
   :template: method.rst

   ES.fit
   ES.predict
   ES.predict_intervals
   ES.select_best_model
   ES.summary

Utility Functions
-----------------

- :doc:`msdecompose` - Multiple seasonal decomposition for time series
- :doc:`lowess` - LOWESS (Locally Weighted Scatterplot Smoothing)

Optimization Settings
---------------------

The ADAM and ES classes use the NLopt library for parameter optimization. You can
customize the optimization behavior via the ``nlopt_kargs`` parameter:

.. code-block:: python

   from smooth import ADAM

   model = ADAM(
       model="AAN",
       nlopt_kargs={
           "print_level": 1,        # Print optimization progress
           "xtol_rel": 1e-8,        # Relative parameter tolerance
           "algorithm": "NLOPT_LN_SBPLX"  # Use Subplex algorithm
       }
   )
   model.fit(y)

**Available parameters:**

+--------------+--------------------------------------------------------------+--------------------+
| Parameter    | Description                                                  | Default            |
+==============+==============================================================+====================+
| print_level  | Verbosity level. When >0, prints B and CF on every           | 0                  |
|              | iteration.                                                   |                    |
+--------------+--------------------------------------------------------------+--------------------+
| xtol_rel     | Relative tolerance on parameters. Stops when changes         | 1e-6               |
|              | < xtol_rel * \|params\|.                                     |                    |
+--------------+--------------------------------------------------------------+--------------------+
| xtol_abs     | Absolute tolerance on parameters. Stops when changes         | 1e-8               |
|              | < xtol_abs.                                                  |                    |
+--------------+--------------------------------------------------------------+--------------------+
| ftol_rel     | Relative tolerance on cost function. Stops when changes      | 1e-8               |
|              | < ftol_rel * \|CF\|.                                         |                    |
+--------------+--------------------------------------------------------------+--------------------+
| ftol_abs     | Absolute tolerance on cost function. Stops when changes      | 0                  |
|              | < ftol_abs.                                                  |                    |
+--------------+--------------------------------------------------------------+--------------------+
| algorithm    | NLopt algorithm name. Use "LN\_" prefix for derivative-free. | NLOPT_LN_NELDERMEAD|
|              | Options: NLOPT_LN_NELDERMEAD, NLOPT_LN_SBPLX,                |                    |
|              | NLOPT_LN_COBYLA, NLOPT_LN_BOBYQA.                            |                    |
+--------------+--------------------------------------------------------------+--------------------+

.. toctree::
   :maxdepth: 2
   :caption: Contents:
   :hidden:

   api
   msdecompose
   lowess