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

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

.. _sphx_glr_auto_examples_plot_cluster_user.py:


===========================================
DL8.5 used to perform predictive clustering
===========================================
This example illustrates how to use a user-specified error function to perform predictive
clustering. The PyDL8.5 library also provides an implementation of predictive clustering
that does not require the use of user-specified error function. 
Check the DL85Cluster class for this implementation.

The main purpose of this example is to show how users of the library can implement their
own decision tree learning task using PyDL8.5's interface for writing error functions.

.. GENERATED FROM PYTHON SOURCE LINES 13-61




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

 .. code-block:: none

    ############################################################################################
    #      DL8.5 clustering : user specific error function and leaves' values assignment       #
    ############################################################################################
    Model building...
    Model built. Duration of the search = 4.4355






|

.. code-block:: default

    import numpy as np
    from sklearn.model_selection import train_test_split
    from sklearn.metrics import DistanceMetric
    import time
    from pydl85 import DL85Predictor

    dataset = np.genfromtxt("../datasets/anneal.txt", delimiter=' ')
    X = dataset[:, 1:]
    X_train, X_test = train_test_split(X, test_size=0.2, random_state=0)


    print("############################################################################################\n"
          "#      DL8.5 clustering : user specific error function and leaves' values assignment       #\n"
          "############################################################################################")

    # The quality of every cluster is determined using the Euclidean distance.
    eucl_dist = DistanceMetric.get_metric('euclidean')


    # user error function
    def error(tids):
        # collect the complete examples identified using the tids. 
        X_subset = X_train[list(tids), :]
        # determine the centroid of the cluster
        centroid = np.mean(X_subset, axis=0)
        # calculate the distances towards centroid
        distances = eucl_dist.pairwise(X_subset, [centroid])
        # return the sum of distances as the error
        return float(sum(distances))


    # user leaf assignment
    def leaf_value(tids):
        # The prediction for every leaf is the centroid of the cluster
        return np.mean(X.take(list(tids)))


    # Change the parameters of the algorithm as desired.
    clf = DL85Predictor(max_depth=2, min_sup=5, error_function=error, leaf_value_function=leaf_value, time_limit=600)

    start = time.perf_counter()
    print("Model building...")
    clf.fit(X_train)
    duration = time.perf_counter() - start
    print("Model built. Duration of the search =", round(duration, 4))
    predicted = clf.predict(X_test)




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

   **Total running time of the script:** ( 0 minutes  4.474 seconds)


.. _sphx_glr_download_auto_examples_plot_cluster_user.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_cluster_user.py <plot_cluster_user.py>`

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

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


.. only:: html

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

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