IMLCV.implementations.tensorflow.CvDiscovery
============================================

.. py:module:: IMLCV.implementations.tensorflow.CvDiscovery


Classes
-------

.. autoapisummary::

   IMLCV.implementations.tensorflow.CvDiscovery.hkFunBase
   IMLCV.implementations.tensorflow.CvDiscovery.TranformerUMAP


Functions
---------

.. autoapisummary::

   IMLCV.implementations.tensorflow.CvDiscovery.umap_function
   IMLCV.implementations.tensorflow.CvDiscovery.umap_encoder


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

.. py:function:: umap_function(x: IMLCV.base.CV.CV, nl: IMLCV.base.CV.NeighbourList, c, enc)

.. py:function:: umap_encoder(x, nlayers, nunits, outdim)

.. py:class:: hkFunBase(*args, **kwargs)

   Bases: :py:obj:`IMLCV.base.CV.CvFunBase`


   Helper class that provides a standard way to create an ABC using
   inheritance.


   .. py:attribute:: _
      :type:  dataclasses.KW_ONLY


   .. py:attribute:: fwd_params
      :type:  dict


   .. py:attribute:: fwd_kwargs
      :type:  dict


   .. py:attribute:: bwd_params
      :type:  dict | None


   .. py:attribute:: bwd_kwargs
      :type:  dict | None


   .. py:method:: _calc(x: IMLCV.base.CV.CV, nl: IMLCV.base.CV.NeighbourList, reverse=False, conditioners: list[IMLCV.base.CV.CV] | None = None, shmap=False) -> IMLCV.base.CV.CV


.. py:class:: TranformerUMAP(*args, **kwargs)

   Bases: :py:obj:`IMLCV.base.CVDiscovery.Transformer`


   Base class for dataclasses that should act like a JAX pytree node.


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



   .. py:attribute:: nunits
      :type:  int
      :value: 256



   .. py:attribute:: nlayers
      :type:  int
      :value: 3



   .. py:attribute:: parametric
      :type:  bool
      :value: True



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



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



   .. py:method:: _fit(x: list[IMLCV.base.CV.CV], x_t: list[IMLCV.base.CV.CV] | None, w: list[jax.Array], dlo: IMLCV.base.rounds.DataLoaderOutput, decoder=False, nunits=256, nlayers=3, parametric=True, densmap=False, n_neighbors=20, chunk_size=None, verbose=True, macro_chunk=1000, **kwargs)