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

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

.. _sphx_glr_auto_examples_detectors_openmax.py:


OpenMax
==============================

:class:`OpenMax <pytorch_ood.detector.OpenMax>` was originally proposed
for Open Set Recognition but can be adapted for Out-of-Distribution tasks.

.. warning:: OpenMax requires ``libmr`` to be installed, which is broken at the moment. You can only use it
   by installing ``cython`` and ``numpy``, and ``libmr`` manually afterwards.

.. GENERATED FROM PYTHON SOURCE LINES 13-25

.. code-block:: Python
   :lineno-start: 13

    from torch.utils.data import DataLoader
    from torchvision.datasets import CIFAR10

    from pytorch_ood.dataset.img import Textures
    from pytorch_ood.detector import OpenMax
    from pytorch_ood.model import WideResNet
    from pytorch_ood.utils import OODMetrics, ToUnknown, fix_random_seed

    fix_random_seed(123)

    device = "cuda:0"


.. GENERATED FROM PYTHON SOURCE LINES 26-27

Setup preprocessing and data

.. GENERATED FROM PYTHON SOURCE LINES 27-40

.. code-block:: Python
   :lineno-start: 27

    trans = WideResNet.transform_for("cifar10-pt")

    dataset_train = CIFAR10(root="data", train=True, download=True, transform=trans)
    dataset_in_test = CIFAR10(root="data", train=False, download=True, transform=trans)
    dataset_out_test = Textures(
        root="data", download=True, transform=trans, target_transform=ToUnknown()
    )

    train_loader = DataLoader(dataset_train, batch_size=128, shuffle=True)

    # create data loaders
    test_loader = DataLoader(dataset_in_test + dataset_out_test, batch_size=128)


.. GENERATED FROM PYTHON SOURCE LINES 41-42

Stage 1: Create DNN pre-trained on CIFAR 10

.. GENERATED FROM PYTHON SOURCE LINES 42-44

.. code-block:: Python
   :lineno-start: 42

    model = WideResNet(num_classes=10, pretrained="cifar10-pt").to(device).eval()


.. GENERATED FROM PYTHON SOURCE LINES 45-46

Stage 2: Create and Fit OpenMax

.. GENERATED FROM PYTHON SOURCE LINES 46-49

.. code-block:: Python
   :lineno-start: 46

    detector = OpenMax(model, tailsize=25, alpha=5, euclid_weight=0.5)
    detector.fit(train_loader, device=device)


.. GENERATED FROM PYTHON SOURCE LINES 50-51

Stage 3: Evaluate Detectors

.. GENERATED FROM PYTHON SOURCE LINES 51-57

.. code-block:: Python
   :lineno-start: 51

    metrics = OODMetrics()

    for x, y in test_loader:
        metrics.update(detector(x.to(device)), y)

    print(metrics.compute())


.. _sphx_glr_download_auto_examples_detectors_openmax.py:

.. only:: html

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

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

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

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

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


.. only:: html

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

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