.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "auto_examples/plot_nonparametric_clustering.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_plot_nonparametric_clustering.py>`
        to download the full example code

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

.. _sphx_glr_auto_examples_plot_nonparametric_clustering.py:


==========================
Non parametric clustering
==========================

This example illustrates how we can run nonparametric clustering using GEMINI.

.. GENERATED FROM PYTHON SOURCE LINES 8-16

.. code-block:: default


    import numpy as np
    from matplotlib import pyplot as plt
    from sklearn import metrics

    from gemclus import data
    from gemclus.nonparametric import CategoricalMMD








.. GENERATED FROM PYTHON SOURCE LINES 17-19

Draw samples from a GMM
--------------------------------------------------------------

.. GENERATED FROM PYTHON SOURCE LINES 21-29

.. code-block:: default


    # Generate samples on that are simple to separate
    N = 100  # Number of nodes in the graph
    # GMM parameters
    means = np.array([[1, -1], [1, 1], [-1, -1], [-1, 1]])*2
    covariances = [np.eye(2)*0.5]*4
    X, y = data.draw_gmm(N, means, covariances, np.ones(4) / 4, random_state=1789)








.. GENERATED FROM PYTHON SOURCE LINES 30-34

Train the model
--------------------------------------------------------------
Create the Non parametric GEMINI clustering model and call the .fit method to optimise the cluster
assignment of the nodes

.. GENERATED FROM PYTHON SOURCE LINES 36-40

.. code-block:: default


    model = CategoricalMMD(n_clusters=4, ovo=True, random_state=0, learning_rate=1e-2)
    y_pred = model.fit_predict(X)








.. GENERATED FROM PYTHON SOURCE LINES 41-43

Final Clustering
-----------------

.. GENERATED FROM PYTHON SOURCE LINES 45-52

.. code-block:: default


    plt.scatter(X[:, 0], X[:, 1], c=y_pred)
    plt.show()

    ari_score = metrics.adjusted_rand_score(y, y_pred)
    gemini_score = model.score(X)
    print(f"Final ARI score: {ari_score:.3f}")
    print(f"GEMINI score is {gemini_score:.3f}")


.. image-sg:: /auto_examples/images/sphx_glr_plot_nonparametric_clustering_001.png
   :alt: plot nonparametric clustering
   :srcset: /auto_examples/images/sphx_glr_plot_nonparametric_clustering_001.png
   :class: sphx-glr-single-img


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

 .. code-block:: none

    Final ARI score: 0.975
    GEMINI score is 3.328





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

   **Total running time of the script:** (0 minutes 0.393 seconds)


.. _sphx_glr_download_auto_examples_plot_nonparametric_clustering.py:

.. only:: html

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




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

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

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

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


.. only:: html

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

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