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

    Click :ref:`here <sphx_glr_download_examples_classifiers_frnn.py>` to download the full example code
.. rst-class:: sphx-glr-example-title

.. _sphx_glr_examples_classifiers_frnn.py:


===================================
Multiclass classification with FRNN
===================================

The figures contain the training instances within a section of the selected feature space.
The training instances are coloured according to their true labels,
while the feature space is coloured according to predictions on the basis of the training instances,
making the decision boundaries visible.

Two subfigures are displayed: the first represents strict FRNN (`k == 1`),
while the second represents FRNN with linear OWA weights and `k == 20`.



.. image:: /examples/classifiers/images/sphx_glr_frnn_001.png
    :class: sphx-glr-single-img


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

 Out:

 .. code-block:: none


    /home/oliver/code/fuzzy-rough-learn/examples/classifiers/frnn.py:59: MatplotlibDeprecationWarning: shading='flat' when X and Y have the same dimensions as C is deprecated since 3.3.  Either specify the corners of the quadrilaterals with X and Y, or pass shading='auto', 'nearest' or 'gouraud', or set rcParams['pcolor.shading'].  This will become an error two minor releases later.
      plt.pcolormesh(xx, yy, Z, cmap=cmap_light)
    /home/oliver/code/fuzzy-rough-learn/examples/classifiers/frnn.py:59: MatplotlibDeprecationWarning: shading='flat' when X and Y have the same dimensions as C is deprecated since 3.3.  Either specify the corners of the quadrilaterals with X and Y, or pass shading='auto', 'nearest' or 'gouraud', or set rcParams['pcolor.shading'].  This will become an error two minor releases later.
      plt.pcolormesh(xx, yy, Z, cmap=cmap_light)
    /home/oliver/code/fuzzy-rough-learn/examples/classifiers/frnn.py:76: UserWarning: Matplotlib is currently using agg, which is a non-GUI backend, so cannot show the figure.
      plt.show()





|


.. code-block:: default

    print(__doc__)

    import numpy as np
    import matplotlib.pyplot as plt
    from matplotlib.colors import ListedColormap
    from sklearn import datasets

    from frlearn.base import select_class
    from frlearn.classifiers import FRNN
    from frlearn.weights import LinearWeights

    # Import example data and reduce to 2 dimensions.
    iris = datasets.load_iris()
    X = iris.data[:, :2]
    y = iris.target

    # Define color maps.
    cmap_light = ListedColormap(['#FFAAAA', '#AAFFAA', '#AAAAFF'])
    cmap_bold = ListedColormap(['#FF0000', '#00FF00', '#0000FF'])

    # Initialise figure with wide aspect for two side-by-side subfigures.
    plt.figure(figsize=(8, 4))

    for i, owa_weights, k, title in [
        (1, None, 1, 'Strict'),
        (2, LinearWeights(), 20, 'With linear weights and k = 20')
    ]:
        axes = plt.subplot(1, 2, i)

        # Create an instance of the FRNN classifier and construct the model.
        clf = FRNN(upper_weights=owa_weights, lower_weights=owa_weights, upper_k=k, lower_k=k)
        model = clf(X, y)

        # Create a mesh of points in the attribute space.
        step_size = .02
        x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1
        y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1
        xx, yy = np.meshgrid(np.arange(x_min, x_max, step_size), np.arange(y_min, y_max, step_size))

        # Query mesh points to obtain class values and select highest valued class.
        Z = model(np.c_[xx.ravel(), yy.ravel()])
        Z = select_class(Z, labels=model.classes)

        # Plot mesh.
        Z = Z.reshape(xx.shape)
        plt.pcolormesh(xx, yy, Z, cmap=cmap_light)

        # Plot training instances.
        plt.scatter(X[:, 0], X[:, 1], c=y, cmap=cmap_bold,
                    edgecolor='k', s=20)

        # Set subplot aspect to standard aspect ratio.
        axes.set_aspect(1.0 / axes.get_data_ratio() * .75)

        # Set plot dimensions.
        plt.xlim(xx.min(), xx.max())
        plt.ylim(yy.min(), yy.max())

        # Describe the subfigures.
        plt.title(title)

    plt.suptitle('FRNN applied to iris dataset', fontsize=14)
    plt.show()



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

   **Total running time of the script:** ( 0 minutes  1.027 seconds)


.. _sphx_glr_download_examples_classifiers_frnn.py:


.. only :: html

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



  .. container:: sphx-glr-download

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



  .. container:: sphx-glr-download

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


.. only:: html

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

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