foscat.xarray.parameters
========================

.. py:module:: foscat.xarray.parameters


Classes
-------

.. autoapisummary::

   foscat.xarray.parameters.Parameters


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

.. py:class:: Parameters

   parameters for the scattering covariance transform

   :Parameters: * **n_orientations** (:py:class:`int`) -- The number of orientations of the kernel
                * **kernel_size** (:py:class:`int`) -- The size of the kernel in cells / pixels
                * **jmax_delta** (:py:class:`float`) -- Compute the scattering covariance coefficients for this many refinement levels,
                  starting with the data level (``level``, which is defined as ``log2(nside)``) in
                  decreasing order. This means that the levels for which coefficients are computed
                  are in the range

                  .. math::
                      [level - relative_level, level]

                  If ``relative_level`` is None, compute all levels until ``level=0``.
                * **dtype** (:py:class:`str` or :py:class:`numpy.dtype`, *default*: ``"float64"``) -- The dtype to use for the transform.
                * **backend** (``{"numpy", "torch", "tensorflow"}``, *default*: ``"tensorflow"``) -- The compute backend.


   .. py:attribute:: n_orientations
      :type:  int


   .. py:attribute:: kernel_size
      :type:  int


   .. py:attribute:: jmax_delta
      :type:  float


   .. py:attribute:: dtype
      :type:  str | numpy.dtype
      :value: 'float64'



   .. py:attribute:: backend
      :type:  str
      :value: 'tensorflow'



   .. py:attribute:: cache
      :type:  foscat.scat_cov.funct | None
      :value: None