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

.. only:: html

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

        Click :ref:`here <sphx_glr_download_auto_examples_model_selection_plot_successive_halving_iterations.py>`
        to download the full example code or to run this example in your browser via Binder

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

.. _sphx_glr_auto_examples_model_selection_plot_successive_halving_iterations.py:


Successive Halving Iterations
=============================

This example illustrates how a successive halving search
(:class:`~sklearn.model_selection.HalvingGridSearchCV` and
:class:`~sklearn.model_selection.HalvingRandomSearchCV`)
iteratively chooses the best parameter combination out of
multiple candidates.

.. GENERATED FROM PYTHON SOURCE LINES 12-25

.. code-block:: default

    import pandas as pd
    from sklearn import datasets
    import matplotlib.pyplot as plt
    from scipy.stats import randint
    import numpy as np

    from sklearn.experimental import enable_halving_search_cv  # noqa
    from sklearn.model_selection import HalvingRandomSearchCV
    from sklearn.ensemble import RandomForestClassifier


    print(__doc__)








.. GENERATED FROM PYTHON SOURCE LINES 26-28

We first define the parameter space and train a
:class:`~sklearn.model_selection.HalvingRandomSearchCV` instance.

.. GENERATED FROM PYTHON SOURCE LINES 28-48

.. code-block:: default


    rng = np.random.RandomState(0)

    X, y = datasets.make_classification(n_samples=700, random_state=rng)

    clf = RandomForestClassifier(n_estimators=20, random_state=rng)

    param_dist = {"max_depth": [3, None],
                  "max_features": randint(1, 11),
                  "min_samples_split": randint(2, 11),
                  "bootstrap": [True, False],
                  "criterion": ["gini", "entropy"]}

    rsh = HalvingRandomSearchCV(
        estimator=clf,
        param_distributions=param_dist,
        factor=2,
        random_state=rng)
    rsh.fit(X, y)






.. raw:: html

    <div class="output_subarea output_html rendered_html output_result">
    <style>#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 {color: black;background-color: white;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 pre{padding: 0;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-toggleable {background-color: white;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.2em 0.3em;box-sizing: border-box;text-align: center;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;margin: 0.25em 0.25em;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-parallel-item::after {content: "";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-serial::before {content: "";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-item {z-index: 1;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-parallel-item {display: flex;flex-direction: column;position: relative;background-color: white;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-parallel-item:only-child::after {width: 0;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0.2em;box-sizing: border-box;padding-bottom: 0.1em;background-color: white;position: relative;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-label label {font-family: monospace;font-weight: bold;background-color: white;display: inline-block;line-height: 1.2em;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-label-container {position: relative;z-index: 2;text-align: center;}#sk-73182f3b-d64d-4ac0-accb-103b78f594d3 div.sk-container {display: inline-block;position: relative;}</style><div id="sk-73182f3b-d64d-4ac0-accb-103b78f594d3" class"sk-top-container"><div class="sk-container"><div class="sk-item sk-dashed-wrapped"><div class="sk-label-container"><div class="sk-label sk-toggleable"><input class="sk-toggleable__control sk-hidden--visually" id="6685b19c-b294-4f4a-a657-187b2676fca7" type="checkbox" ><label class="sk-toggleable__label" for="6685b19c-b294-4f4a-a657-187b2676fca7">HalvingRandomSearchCV</label><div class="sk-toggleable__content"><pre>HalvingRandomSearchCV(estimator=RandomForestClassifier(n_estimators=20,
                                                           random_state=RandomState(MT19937) at 0x7EFBFF87EC40),
                          factor=2,
                          param_distributions={'bootstrap': [True, False],
                                               'criterion': ['gini', 'entropy'],
                                               'max_depth': [3, None],
                                               'max_features': <scipy.stats._distn_infrastructure.rv_frozen object at 0x7efbffa6ee20>,
                                               'min_samples_split': <scipy.stats._distn_infrastructure.rv_frozen object at 0x7efbff60bc10>},
                          random_state=RandomState(MT19937) at 0x7EFBFF87EC40,
                          refit=<function _refit_callable at 0x7efc209d88b0>)</pre></div></div></div><div class="sk-parallel"><div class="sk-parallel-item"><div class="sk-item"><div class="sk-serial"><div class="sk-item"><div class="sk-estimator sk-toggleable"><input class="sk-toggleable__control sk-hidden--visually" id="2848a189-6134-4d51-a8d4-8625feec7e00" type="checkbox" ><label class="sk-toggleable__label" for="2848a189-6134-4d51-a8d4-8625feec7e00">RandomForestClassifier</label><div class="sk-toggleable__content"><pre>RandomForestClassifier(n_estimators=20,
                           random_state=RandomState(MT19937) at 0x7EFBFF87EC40)</pre></div></div></div></div></div></div></div></div></div></div>
    </div>
    <br />
    <br />

.. GENERATED FROM PYTHON SOURCE LINES 49-51

We can now use the `cv_results_` attribute of the search estimator to inspect
and plot the evolution of the search.

.. GENERATED FROM PYTHON SOURCE LINES 51-73

.. code-block:: default


    results = pd.DataFrame(rsh.cv_results_)
    results['params_str'] = results.params.apply(str)
    results.drop_duplicates(subset=('params_str', 'iter'), inplace=True)
    mean_scores = results.pivot(index='iter', columns='params_str',
                                values='mean_test_score')
    ax = mean_scores.plot(legend=False, alpha=.6)

    labels = [
        f'iter={i}\nn_samples={rsh.n_resources_[i]}\n'
        f'n_candidates={rsh.n_candidates_[i]}'
        for i in range(rsh.n_iterations_)
    ]

    ax.set_xticks(range(rsh.n_iterations_))
    ax.set_xticklabels(labels, rotation=45, multialignment='left')
    ax.set_title('Scores of candidates over iterations')
    ax.set_ylabel('mean test score', fontsize=15)
    ax.set_xlabel('iterations', fontsize=15)
    plt.tight_layout()
    plt.show()




.. image:: /auto_examples/model_selection/images/sphx_glr_plot_successive_halving_iterations_001.png
    :alt: Scores of candidates over iterations
    :class: sphx-glr-single-img





.. GENERATED FROM PYTHON SOURCE LINES 74-88

Number of candidates and amount of resource at each iteration
-------------------------------------------------------------

At the first iteration, a small amount of resources is used. The resource
here is the number of samples that the estimators are trained on. All
candidates are evaluated.

At the second iteration, only the best half of the candidates is evaluated.
The number of allocated resources is doubled: candidates are evaluated on
twice as many samples.

This process is repeated until the last iteration, where only 2 candidates
are left. The best candidate is the candidate that has the best score at the
last iteration.


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

   **Total running time of the script:** ( 0 minutes  13.323 seconds)


.. _sphx_glr_download_auto_examples_model_selection_plot_successive_halving_iterations.py:


.. only :: html

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


  .. container:: binder-badge

    .. image:: images/binder_badge_logo.svg
      :target: https://mybinder.org/v2/gh/scikit-learn/scikit-learn/0.24.X?urlpath=lab/tree/notebooks/auto_examples/model_selection/plot_successive_halving_iterations.ipynb
      :alt: Launch binder
      :width: 150 px


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

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



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

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


.. only:: html

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

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