.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "auto_examples/plot_3d_to_2d.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_3d_to_2d.py>`
        to download the full example code

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

.. _sphx_glr_auto_examples_plot_3d_to_2d.py:


Sample slices from volumes
==========================

In this example, volumes are padded, scaled, rotated and sometimes flipped.
Then, 2D slices are extracted.

.. GENERATED FROM PYTHON SOURCE LINES 8-45

.. code-block:: Python


    import matplotlib.pyplot as plt
    import torch
    import torchio as tio

    torch.manual_seed(0)
    max_queue_length = 16
    patches_per_volume = 2

    subject = tio.datasets.Colin27()
    subject.remove_image('head')

    subjects = 50 * [subject]
    max_side = max(subject.shape)
    transform = tio.Compose(
        (
            tio.CropOrPad(max_side),
            tio.RandomFlip(),
            tio.RandomAffine(degrees=360),
        )
    )
    dataset = tio.SubjectsDataset(subjects, transform=transform)
    patch_size = (max_side, max_side, 1)  # 2D slices


    def plot_batch(sampler):
        queue = tio.Queue(dataset, max_queue_length, patches_per_volume, sampler)
        loader = torch.utils.data.DataLoader(queue, batch_size=16)
        batch = tio.utils.get_first_item(loader)

        fig, axes = plt.subplots(4, 4, figsize=(12, 10))
        for ax, im in zip(axes.flatten(), batch['t1']['data']):
            ax.imshow(im.squeeze(), cmap='gray')
        plt.suptitle(sampler.__class__.__name__)
        plt.tight_layout()









.. GENERATED FROM PYTHON SOURCE LINES 46-50

Uniform sampler
---------------
When a :class:`torchio.UniformSampler` is used,
some of the patches don't contain much useful information:

.. GENERATED FROM PYTHON SOURCE LINES 50-54

.. code-block:: Python


    sampler = tio.UniformSampler(patch_size)
    plot_batch(sampler)




.. image-sg:: /auto_examples/images/sphx_glr_plot_3d_to_2d_001.png
   :alt: UniformSampler
   :srcset: /auto_examples/images/sphx_glr_plot_3d_to_2d_001.png
   :class: sphx-glr-single-img





.. GENERATED FROM PYTHON SOURCE LINES 55-60

Weighted sampler
----------------
We can use the ``brain`` image contained in the subject as a probability map
for a :class:`torchio.WeightedSampler`. That way, we ensure that the center
of all patches correspond to brain tissue.

.. GENERATED FROM PYTHON SOURCE LINES 60-65

.. code-block:: Python


    sampler = tio.WeightedSampler(patch_size, probability_map='brain')
    plot_batch(sampler)

    plt.show()



.. image-sg:: /auto_examples/images/sphx_glr_plot_3d_to_2d_002.png
   :alt: WeightedSampler
   :srcset: /auto_examples/images/sphx_glr_plot_3d_to_2d_002.png
   :class: sphx-glr-single-img






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

   **Total running time of the script:** (0 minutes 23.530 seconds)


.. _sphx_glr_download_auto_examples_plot_3d_to_2d.py:

.. only:: html

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

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

      :download:`Download Jupyter notebook: plot_3d_to_2d.ipynb <plot_3d_to_2d.ipynb>`

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

      :download:`Download Python source code: plot_3d_to_2d.py <plot_3d_to_2d.py>`


.. only:: html

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

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