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

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

.. _sphx_glr_auto_examples_detectors_softmax.py:


MSP
==============================

Uses MSP based on a pre-trained model from the Hendrycks baseline paper.

.. GENERATED FROM PYTHON SOURCE LINES 9-46

.. code-block:: Python
   :lineno-start: 9

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

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

    fix_random_seed(123)
    device = "cuda:0"

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

    # setup IN test data
    dataset_in_test = CIFAR10(root="data", train=False, download=True, transform=trans)
    # setup OOD test data
    dataset_out_test = Textures(
        root="data", download=True, transform=trans, target_transform=ToUnknown()
    )

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

    # Stage 1: Create DNN
    model = WideResNet(num_classes=10, pretrained="cifar10-pt").to(device).eval()

    # Stage 2: Create Detector
    detector = MaxSoftmax(model)

    # Stage 3: Evaluate Detectors
    metrics = OODMetrics()

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

    print(metrics.compute())


.. _sphx_glr_download_auto_examples_detectors_softmax.py:

.. only:: html

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

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

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

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

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


.. only:: html

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

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