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

.. only:: html

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

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

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

.. _sphx_glr__gallery_compression_pmd_compression.py:


PMD Compression
===============

Simple PMD compression, the hello world of masknmf

.. GENERATED FROM PYTHON SOURCE LINES 7-76



.. image-sg:: /_gallery/compression/images/sphx_glr_pmd_compression_001.webp
   :alt: pmd compression
   :srcset: /_gallery/compression/images/sphx_glr_pmd_compression_001.webp
   :class: sphx-glr-single-img


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

 .. code-block:: none

    /home/runner/work/masknmf-toolbox/masknmf-toolbox/masknmf/utils/_cuda.py:7: UserWarning: You've explicitly selected to perform computations on the cpu, performance will be significantly slower
      warn(
    [25-11-16 02:56:10]: Starting compression
    [25-11-16 02:56:10]: sampled from the following regions: [0]
    [25-11-16 02:56:10]: We are initializing on a total of 2000 frames
    [25-11-16 02:56:10]: Loading data to estimate complete spatial basis
    [25-11-16 02:56:10]: skipping the pruning step for frame cutoff
    [25-11-16 02:56:10]: Finding spatiotemporal roughness thresholds

      0%|          | 0/250 [00:00<?, ?it/s]
      2%|▏         | 5/250 [00:00<00:05, 42.41it/s]
      4%|▍         | 10/250 [00:00<00:05, 43.21it/s]
      6%|▌         | 15/250 [00:00<00:05, 43.97it/s]
      8%|▊         | 20/250 [00:00<00:05, 44.39it/s]
     10%|█         | 25/250 [00:00<00:05, 42.23it/s]
     12%|█▏        | 30/250 [00:00<00:05, 43.29it/s]
     14%|█▍        | 35/250 [00:00<00:04, 43.98it/s]
     16%|█▌        | 40/250 [00:00<00:04, 44.52it/s]
     18%|█▊        | 45/250 [00:01<00:04, 44.94it/s]
     20%|██        | 50/250 [00:01<00:04, 45.19it/s]
     22%|██▏       | 55/250 [00:01<00:04, 45.33it/s]
     24%|██▍       | 60/250 [00:01<00:04, 45.35it/s]
     26%|██▌       | 65/250 [00:01<00:04, 45.25it/s]
     28%|██▊       | 70/250 [00:01<00:03, 45.51it/s]
     30%|███       | 75/250 [00:01<00:03, 45.58it/s]
     32%|███▏      | 80/250 [00:01<00:03, 45.50it/s]
     34%|███▍      | 85/250 [00:01<00:03, 45.36it/s]
     36%|███▌      | 90/250 [00:02<00:03, 45.51it/s]
     38%|███▊      | 95/250 [00:02<00:03, 45.60it/s]
     40%|████      | 100/250 [00:02<00:03, 45.61it/s]
     42%|████▏     | 105/250 [00:02<00:03, 45.69it/s]
     44%|████▍     | 110/250 [00:02<00:03, 45.83it/s]
     46%|████▌     | 115/250 [00:02<00:02, 45.80it/s]
     48%|████▊     | 120/250 [00:02<00:02, 45.73it/s]
     50%|█████     | 125/250 [00:02<00:02, 45.76it/s]
     52%|█████▏    | 130/250 [00:02<00:02, 45.86it/s]
     54%|█████▍    | 135/250 [00:02<00:02, 45.81it/s]
     56%|█████▌    | 140/250 [00:03<00:02, 45.89it/s]
     58%|█████▊    | 145/250 [00:03<00:02, 45.90it/s]
     60%|██████    | 150/250 [00:03<00:02, 45.88it/s]
     62%|██████▏   | 155/250 [00:03<00:02, 45.75it/s]
     64%|██████▍   | 160/250 [00:03<00:01, 45.77it/s]
     66%|██████▌   | 165/250 [00:03<00:01, 45.86it/s]
     68%|██████▊   | 170/250 [00:03<00:01, 45.96it/s]
     70%|███████   | 175/250 [00:03<00:01, 45.90it/s]
     72%|███████▏  | 180/250 [00:03<00:01, 45.89it/s]
     74%|███████▍  | 185/250 [00:04<00:01, 45.78it/s]
     76%|███████▌  | 190/250 [00:04<00:01, 46.01it/s]
     78%|███████▊  | 195/250 [00:04<00:01, 46.18it/s]
     80%|████████  | 200/250 [00:04<00:01, 46.22it/s]
     82%|████████▏ | 205/250 [00:04<00:00, 46.24it/s]
     84%|████████▍ | 210/250 [00:04<00:00, 46.36it/s]
     86%|████████▌ | 215/250 [00:04<00:00, 44.89it/s]
     88%|████████▊ | 220/250 [00:04<00:00, 45.33it/s]
     90%|█████████ | 225/250 [00:04<00:00, 45.67it/s]
     92%|█████████▏| 230/250 [00:05<00:00, 45.79it/s]
     94%|█████████▍| 235/250 [00:05<00:00, 45.65it/s]
     96%|█████████▌| 240/250 [00:05<00:00, 45.74it/s]
     98%|█████████▊| 245/250 [00:05<00:00, 45.99it/s]
    100%|██████████| 250/250 [00:05<00:00, 46.09it/s]
    100%|██████████| 250/250 [00:05<00:00, 45.47it/s]
    [25-11-16 02:56:16]: Running Blockwise Decompositions
    [25-11-16 02:56:16]: Constructed U matrix. Rank of U is 190
    [25-11-16 02:56:16]: PMD Objected constructed
    /opt/hostedtoolcache/Python/3.12.12/x64/lib/python3.12/site-packages/fastplotlib/graphics/features/_base.py:18: UserWarning: casting float64 array to float32
      warn(f"casting {array.dtype} array to float32")







|

.. code-block:: Python


    # test_example = true

    import masknmf
    import torch
    import fastplotlib as fpl
    from urllib.request import urlretrieve
    import tifffile


    urlretrieve(
        "https://github.com/flatironinstitute/CaImAn/raw/refs/heads/main/example_movies/demoMovie.tif",
        "./demo.tif"
    )

    # always lazy load raw data by memmaping or other methods
    data = tifffile.imread("./demo.tif")

    block_sizes = [32, 32]
    max_components = 20

    # it's recommended to use masknmf on a machine with a GPU
    # device = "cuda" if torch.cuda.is_available() else "cpu"
    device = "cpu"
    # number of frames used to estimate the spatial basis in PMD
    num_frames_for_spatial_fit = data.shape[0]

    # perform PMD
    pmd_result = masknmf.compression.pmd_decomposition(
        data,
        block_sizes,
        num_frames_for_spatial_fit,
        max_components=max_components,
        device=device,
        frame_batch_size=1024
    )

    # get the residual
    pmd_residual = masknmf.PMDResidualArray(data, pmd_result)

    # view the movies, note that all these array are LAZY evaluated, allowing you to view extremely large datasets!
    iw = fpl.ImageWidget(
        data=[data, pmd_result, pmd_residual],
        names=["raw", "pmd", "residual"],
        figure_kwargs={"size": (1000, 340), "shape": (1, 3)},
        cmap="gnuplot2",
    )

    iw.show()

    # use the time slider or set the frame index programmatically
    iw.current_index = {"t": 1610}

    # manually set vmin-vmax to emphasize noise in raw video
    # you can also adjust the vmin-vmax using the histogram tool
    # reset the vmin-vmax by clicking the buttons under "ImageWidget Controls"
    for image in iw.managed_graphics:
        image.vmax = 3_200

    # remove toolbar to reduce clutter
    for subplot in iw.figure:
        subplot.toolbar = False


    # ignore the remaining lines these are just for docs generation
    figure = iw.figure
    if __name__ == "__main__":
        print(__doc__)
        fpl.loop.run()


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

   **Total running time of the script:** (0 minutes 51.838 seconds)


.. _sphx_glr_download__gallery_compression_pmd_compression.py:

.. only:: html

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

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

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

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

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


.. only:: html

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

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