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

.. only:: html

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

        :ref:`Go to the end <sphx_glr_download_examples_decomposition_sketchmap.py>`
        to download the full example code.

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

.. _sphx_glr_examples_decomposition_sketchmap.py:


Sketch-map example using pre-selected high-dimensional landmarks
==================================================================

This example demonstrates a minimal, sphinx-gallery friendly usage of the
`SketchMap` estimator. It loads the provided landmark file, fits the
estimator on the landmark set (using per-sample `sample_weights`) and
plots the resulting 2D embedding.

Notes
-----
- The lightweight Python implementation of Sketch-map lives in
  `skmatter.decomposition._sketchmap`.
- The example fits only the landmark set (callers that want a subset
  should pre-select and pass that array to `fit`).

.. GENERATED FROM PYTHON SOURCE LINES 22-29

.. code-block:: Python

    import numpy as np
    from matplotlib import pyplot as plt
    from matplotlib import cm

    from skmatter.decomposition import SketchMap









.. GENERATED FROM PYTHON SOURCE LINES 30-32

Load example landmark data. The file included with the examples contains
high-dimensional descriptors and the last column holds a per-sample weight.

.. GENERATED FROM PYTHON SOURCE LINES 32-38

.. code-block:: Python

    data_file = "highd-landmarks"
    data = np.loadtxt(data_file)
    X_land = data[:, :-1]
    weights = data[:, -1]









.. GENERATED FROM PYTHON SOURCE LINES 39-41

Fit SketchMap on the landmark set and provide `sample_weights`.
We fit on the landmarks only here (no FPS or internal selection).

.. GENERATED FROM PYTHON SOURCE LINES 41-46

.. code-block:: Python

    sm = SketchMap(n_components=2, auto_histogram=True, preopt_steps=50)
    sm.fit(X_land, sample_weights=weights, mixing_ratio=0.2, n_components=2)
    T = sm.embedding_









.. GENERATED FROM PYTHON SOURCE LINES 47-48

Plot the resulting embedding colored by the per-sample weight.

.. GENERATED FROM PYTHON SOURCE LINES 48-58

.. code-block:: Python

    cmap = cm.get_cmap("viridis")
    fig, ax = plt.subplots(figsize=(6, 5))
    sc = ax.scatter(T[:, 0], T[:, 1], c=weights, s=40, cmap=cmap, edgecolor="k")
    ax.set_title("Sketch-map embedding (example)")
    ax.set_xlabel("Sketchmap 1")
    ax.set_ylabel("Sketchmap 2")
    cbar = fig.colorbar(sc, ax=ax)
    cbar.set_label("landmark weight")
    fig.tight_layout()
    plt.show()



.. image-sg:: /examples/decomposition/images/sphx_glr_sketchmap_001.png
   :alt: Sketch-map embedding (example)
   :srcset: /examples/decomposition/images/sphx_glr_sketchmap_001.png
   :class: sphx-glr-single-img


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

 .. code-block:: none

    /home/docs/checkouts/readthedocs.org/user_builds/scikit-matter/checkouts/284/examples/decomposition/sketchmap.py:48: MatplotlibDeprecationWarning: The get_cmap function was deprecated in Matplotlib 3.7 and will be removed in 3.11. Use ``matplotlib.colormaps[name]`` or ``matplotlib.colormaps.get_cmap()`` or ``pyplot.get_cmap()`` instead.
      cmap = cm.get_cmap("viridis")





.. _sphx_glr_download_examples_decomposition_sketchmap.py:

.. only:: html

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

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

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

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

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

    .. container:: sphx-glr-download sphx-glr-download-zip

      :download:`Download zipped: sketchmap.zip <sketchmap.zip>`


.. only:: html

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

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