.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "auto_examples\RME_gettingstarted\plot_e05_vectorizing.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        :ref:`Go to the end <sphx_glr_download_auto_examples_RME_gettingstarted_plot_e05_vectorizing.py>`
        to download the full example code.

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_auto_examples_RME_gettingstarted_plot_e05_vectorizing.py:


Using Non-Vectorized Propagation
================================

Here we go over how to use a propagator with the vectorized setting turned off.

.. GENERATED FROM PYTHON SOURCE LINES 9-13

Setup
-----
First lets create a RMEMeas object we can interact with and initialize
a propagator.

.. GENERATED FROM PYTHON SOURCE LINES 13-41

.. code-block:: Python


    from rmellipse.uobjects import RMEMeas
    from rmellipse.propagators import RMEProp
    import xarray as xr
    import numpy as np

    nom = xr.DataArray(np.zeros((4, 2)),
                       dims=('d1', 'd2'),
                       coords={'d1': [0, 1, 2, 3],
                               'd2': np.arange(2)})

    meas = RMEMeas.from_nom(name='meas', nom=nom)

    meas.add_umech(
        name='my_mechanism',
        value=meas.nom + np.ones(meas.nom.shape) * 0.01,
        dof=np.inf,
        category={'Type': 'B', 'Origin': 'Data Sheet'}
    )

    for i in range(100):
        meas.add_mc_sample(meas.nom + np.random.rand(*meas.nom.shape) * 0.01)

    myprop = RMEProp(
        sensitivity=True,
        montecarlo_sims=100,
        verbose=True)








.. GENERATED FROM PYTHON SOURCE LINES 42-53

Non-Vectorized Propagation
--------------------------
Sometimes it isn't possible to vectorize a function. In that case, similar
to the non-vectorized version, the inputs and outputs of the function need
to be a DataArray for the propagator to recognize the object as being
a RMEMeas object. The first dimension of any inputs that were RMEMeas objects will
be 'umech_id' and the coordinates of that dimension will be the
names of the uncertainty mechanism. For non vectorized inputs, the
propagator will loop over that first dimension, so the inputs will have a
length 1 first dimension called 'umech_id'. The output should
also have that dimension and coordinate.

.. GENERATED FROM PYTHON SOURCE LINES 53-70

.. code-block:: Python


    myprop.settings['montecarlo_sims'] = 0
    myprop.settings['sensitivity'] = True
    myprop.settings['vectorize'] = False
    myprop.settings['verbose'] = False


    @myprop.propagate
    def add(x, y):
        """Add 2 numbers."""
        output = x + y
        print(x.umech_id.values, output.umech_id.values)
        return output


    added = add(meas, 2)





.. rst-class:: sphx-glr-script-out

 .. code-block:: none

    ['nominal'] ['nominal']
    353a3892-1917-4c56-a60d-c80aabebf4cf 353a3892-1917-4c56-a60d-c80aabebf4cf




.. GENERATED FROM PYTHON SOURCE LINES 71-79

Vectorized vs. Unvectorized
---------------------------
By default a RME propagator vectorizes all its functions. This is because
vectorized computation is much faster and
will save a lot of time, especially with large data sets and large numbers
of uncertainty mechanisms. For comparison, here is the same function run
once with the vectorize functionality and once without. Note the orders of
magnitude increase in run time for the Monte Carlo propagation.

.. GENERATED FROM PYTHON SOURCE LINES 79-91

.. code-block:: Python



    @myprop.propagate
    def add2(x):
        return x + 2


    myprop.settings['vectorize'] = True
    add2(meas)

    myprop.settings['vectorize'] = False
    add2(meas)




.. rst-class:: sphx-glr-script-out

 .. code-block:: none


    add2 with nominal: 
    <xarray.DataArray (d1: 4, d2: 2)> Size: 64B
    array([[2., 2.],
           [2., 2.],
           [2., 2.],
           [2., 2.]])
    Coordinates:
      * d1       (d1) int64 32B 0 1 2 3
      * d2       (d2) int64 16B 0 1




.. rst-class:: sphx-glr-timing

   **Total running time of the script:** (0 minutes 0.366 seconds)


.. _sphx_glr_download_auto_examples_RME_gettingstarted_plot_e05_vectorizing.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: plot_e05_vectorizing.ipynb <plot_e05_vectorizing.ipynb>`

    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: plot_e05_vectorizing.py <plot_e05_vectorizing.py>`

    .. container:: sphx-glr-download sphx-glr-download-zip

      :download:`Download zipped: plot_e05_vectorizing.zip <plot_e05_vectorizing.zip>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_