.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "examples/06_we8there/run-01-demo=mix_vb_single_run-model=mix+mult.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

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

        Click :ref:`here <sphx_glr_download_examples_06_we8there_run-01-demo=mix_vb_single_run-model=mix+mult.py>`
        to download the full example code

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

.. _sphx_glr_examples_06_we8there_run-01-demo=mix_vb_single_run-model=mix+mult.py:


=================================================
VB coordinate descent for Mixture of Multinomials
=================================================

.. GENERATED FROM PYTHON SOURCE LINES 8-24

.. code-block:: default

    import bnpy
    import numpy as np
    import os

    from matplotlib import pylab
    import seaborn as sns

    FIG_SIZE = (3, 3)
    SMALL_FIG_SIZE = (1,1)
    pylab.rcParams['figure.figsize'] = FIG_SIZE

    top_word_kws = dict(
        wordSizeLimit=15,
        ncols=4,
        Ktop=10)


.. GENERATED FROM PYTHON SOURCE LINES 25-26

Read text dataset from file

.. GENERATED FROM PYTHON SOURCE LINES 26-37

.. code-block:: default


    dataset_path = os.path.join(bnpy.DATASET_PATH, 'we8there', 'raw')
    dataset = bnpy.data.BagOfWordsData.read_npz(
        os.path.join(dataset_path, 'dataset.npz'),
        vocabfile=os.path.join(dataset_path, 'x_csc_colnames.txt'))

    # Filter out documents with less than 20 words
    doc_ids = np.flatnonzero(
        dataset.getDocTypeCountMatrix().sum(axis=1) >= 20)
    dataset = dataset.make_subset(docMask=doc_ids, doTrackFullSize=False)


.. GENERATED FROM PYTHON SOURCE LINES 38-39

Make a simple plot of the raw data

.. GENERATED FROM PYTHON SOURCE LINES 40-46

.. code-block:: default

    bnpy.viz.PrintTopics.plotCompsFromWordCounts(
        dataset.getDocTypeCountMatrix()[:10],
        vocabList=dataset.vocabList,
        prefix='doc',
        **top_word_kws)


.. GENERATED FROM PYTHON SOURCE LINES 47-51

Train with K=1 cluster
----------------------

This is a simple baseline.

.. GENERATED FROM PYTHON SOURCE LINES 52-65

.. code-block:: default


    trained_model, info_dict = bnpy.run(
        dataset, 'DPMixtureModel', 'Mult', 'VB',
        output_path='/tmp/we8there/helloworld-model=dp_mix+mult-K=1/',
        nLap=1000, convergeThr=0.0001, nTask=1,
        K=1, initname='bregmankmeans+lam1+iter1',
        gamma0=50.0, lam=0.1)

    bnpy.viz.PrintTopics.plotCompsFromHModel(
        trained_model,
        vocabList=dataset.vocabList,
        **top_word_kws)


.. GENERATED FROM PYTHON SOURCE LINES 66-70

Train with K=3 clusters
-----------------------

Take the best of 10 initializations

.. GENERATED FROM PYTHON SOURCE LINES 71-85

.. code-block:: default


    trained_model, info_dict = bnpy.run(
        dataset, 'DPMixtureModel', 'Mult', 'VB',
        output_path='/tmp/we8there/helloworld-model=dp_mix+mult-K=3/',
        nLap=1000, convergeThr=0.0001, nTask=10,
        K=3, initname='bregmankmeans+lam1+iter1',
        gamma0=50.0, lam=0.1)

    bnpy.viz.PrintTopics.plotCompsFromHModel(
        trained_model,
        vocabList=dataset.vocabList,
        **top_word_kws)



.. GENERATED FROM PYTHON SOURCE LINES 86-90

Train with K=10 clusters
------------------------

Take the best of 10 initializations

.. GENERATED FROM PYTHON SOURCE LINES 91-104

.. code-block:: default


    trained_model, info_dict = bnpy.run(
        dataset, 'DPMixtureModel', 'Mult', 'VB',
        output_path='/tmp/we8there/helloworld-model=dp_mix+mult-K=10/',
        nLap=1000, convergeThr=0.0001, nTask=10,
        K=10, initname='bregmankmeans+lam1+iter1',
        gamma0=50.0, lam=0.1)

    bnpy.viz.PrintTopics.plotCompsFromHModel(
        trained_model,
        vocabList=dataset.vocabList,
        **top_word_kws)


.. GENERATED FROM PYTHON SOURCE LINES 105-109

Train with K=30 clusters
------------------------

Take the best of 10 initializations

.. GENERATED FROM PYTHON SOURCE LINES 110-122

.. code-block:: default


    trained_model, info_dict = bnpy.run(
        dataset, 'DPMixtureModel', 'Mult', 'VB',
        output_path='/tmp/we8there/helloworld-model=dp_mix+mult-K=30/',
        nLap=1000, convergeThr=0.0001, nTask=10,
        K=30, initname='bregmankmeans+lam1+iter1',
        gamma0=50.0, lam=0.1)

    bnpy.viz.PrintTopics.plotCompsFromHModel(
        trained_model,
        vocabList=dataset.vocabList,
        **top_word_kws)


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

   **Total running time of the script:** ( 0 minutes  0.000 seconds)


.. _sphx_glr_download_examples_06_we8there_run-01-demo=mix_vb_single_run-model=mix+mult.py:

.. only:: html

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


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

      :download:`Download Python source code: run-01-demo=mix_vb_single_run-model=mix+mult.py <run-01-demo=mix_vb_single_run-model=mix+mult.py>`

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

      :download:`Download Jupyter notebook: run-01-demo=mix_vb_single_run-model=mix+mult.ipynb <run-01-demo=mix_vb_single_run-model=mix+mult.ipynb>`


.. only:: html

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

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