leaspy.models.utils.initialization.model_initialization
=======================================================

.. py:module:: leaspy.models.utils.initialization.model_initialization


Attributes
----------

.. autoapisummary::

   leaspy.models.utils.initialization.model_initialization.XI_STD
   leaspy.models.utils.initialization.model_initialization.TAU_STD
   leaspy.models.utils.initialization.model_initialization.NOISE_STD
   leaspy.models.utils.initialization.model_initialization.SOURCES_STD


Functions
---------

.. autoapisummary::

   leaspy.models.utils.initialization.model_initialization.initialize_parameters
   leaspy.models.utils.initialization.model_initialization.get_lme_results
   leaspy.models.utils.initialization.model_initialization.lme_init


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

.. py:data:: XI_STD
   :value: 0.5


.. py:data:: TAU_STD
   :value: 5.0


.. py:data:: NOISE_STD
   :value: 0.1


.. py:data:: SOURCES_STD
   :value: 1.0


.. py:function:: initialize_parameters(model, dataset, method='default')

   
   Initialize the model's group parameters given its name & the scores of all subjects.

   Under-the-hood it calls an initialization function dedicated for the `model`:
       * :func:`.initialize_linear` (including when `univariate`)
       * :func:`.initialize_logistic` (including when `univariate`)
       * :func:`.initialize_logistic_parallel`

   It is automatically called during :meth:`.Leaspy.fit`.

   :Parameters:

       **model** : :class:`.AbstractModel`
           The model to initialize.

       **dataset** : :class:`.Dataset`
           Contains the individual scores.

       **method** : str
           Must be one of:
               * ``'default'``: initialize at mean.
               * ``'random'``:  initialize with a gaussian realization with same mean and variance.



   :Returns:

       **parameters** : dict [str, :class:`torch.Tensor`]
           Contains the initialized model's group parameters.




   :Raises:

       :exc:`.LeaspyInputError`
           If no initialization method is known for model type / method







   ..
       !! processed by numpydoc !!

.. py:function:: get_lme_results(df, n_jobs=-1, *, with_random_slope_age=True, **lme_fit_kwargs)

   
   Fit a LME on univariate (per feature) time-series (feature vs. patients' ages with varying intercept & slope)


   :Parameters:

       **df** : :class:`pd.DataFrame`
           Contains all the data (with nans)

       **n_jobs** : int
           Number of jobs in parallel when multiple features to init
           Not used for now, buggy

       **with_random_slope_age** : bool (default True)
           Has LME model a random slope per age (otherwise only a random intercept).

       **\*\*lme_fit_kwargs**
           Kwargs passed to 'lme_fit' (such as `force_independent_random_effects`, default True)



   :Returns:

       dict
           {param: str -> param_values_for_ft: torch.Tensor(nb_fts, \*shape_param)}











   ..
       !! processed by numpydoc !!

.. py:function:: lme_init(model, df, fact_std=1.0, **kwargs)

   
   Initialize the model's group parameters.


   :Parameters:

       **model** : :class:`.AbstractModel`
           The model to initialize (must be an univariate or multivariate linear or logistic manifold model).

       **df** : :class:`pd.DataFrame`
           Contains the individual scores (with nans).

       **fact_std** : float
           Multiplicative factor to apply on std-dev (tau, xi, noise) found naively with LME

       **\*\*kwargs**
           Additional kwargs passed to :func:`.get_lme_results`



   :Returns:

       **parameters** : dict [str, `torch.Tensor`]
           Contains the initialized model's group parameters.




   :Raises:

       :exc:`.LeaspyInputError`
           If model is not supported for this initialization







   ..
       !! processed by numpydoc !!