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

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

.. _sphx_glr_auto_examples_osr_osr.py:


CIFAR10
-------------------------

Open Set Simulation on CIFAR 10

.. GENERATED FROM PYTHON SOURCE LINES 8-25

.. code-block:: Python
   :lineno-start: 8

    import torch.nn
    from torch.utils.data import DataLoader
    from torchvision.datasets import CIFAR10
    from torch.nn import CrossEntropyLoss
    from tqdm import tqdm
    from torchmetrics import Accuracy

    from pytorch_ood.dataset.ossim import DynamicOSS
    from pytorch_ood.model import WideResNet
    from pytorch_ood.detector import MaxSoftmax
    from pytorch_ood.utils import fix_random_seed, TargetMapping, OODMetrics, is_known

    device = "cuda:0"
    num_epochs = 10

    fix_random_seed(12345)


.. GENERATED FROM PYTHON SOURCE LINES 26-27

Setup preprocessing

.. GENERATED FROM PYTHON SOURCE LINES 27-30

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

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


.. GENERATED FROM PYTHON SOURCE LINES 31-32

Setup datasets

.. GENERATED FROM PYTHON SOURCE LINES 32-36

.. code-block:: Python
   :lineno-start: 32

    dataset_1 = CIFAR10(root="data", train=True, transform=trans, download=True)
    dataset_2 = CIFAR10(root="data", train=False, transform=trans, download=True)
    dataset = dataset_1 + dataset_2


.. GENERATED FROM PYTHON SOURCE LINES 37-39

Create DNN with pre-trained on a downscaled version of the image net, excluding cifar images
adjust it to output 7 logits

.. GENERATED FROM PYTHON SOURCE LINES 39-44

.. code-block:: Python
   :lineno-start: 39

    print("Creating a Model")
    model = WideResNet(num_classes=1000, pretrained="imagenet32-nocifar")
    model.fc = torch.nn.Linear(model.fc.in_features, 7)
    model.to(device)


.. GENERATED FROM PYTHON SOURCE LINES 45-47

Create open set simulation and dataloaders
We use 3 classes as unknown unknown (OOD), and a data split of 90% train and 10% test

.. GENERATED FROM PYTHON SOURCE LINES 47-71

.. code-block:: Python
   :lineno-start: 47

    ossim = DynamicOSS(
        dataset=dataset,
        train_size=0.9,
        val_size=0.0,
        test_size=0.1,
        kuc=0,
        uuc_val=0,
        uuc_test=3,
        seed=1,
    )
    print(f"Known Classes: {ossim.kkc}")
    print(f"Unknown Classes: {ossim.uuc}")

    # create class remapping
    class_mapping = TargetMapping(known=ossim.kkc, unknown=ossim.uuc)

    train_loader = DataLoader(ossim.train_dataset(), batch_size=32, num_workers=12)
    test_loader = DataLoader(ossim.test_dataset(), batch_size=32, num_workers=12)

    criterion = CrossEntropyLoss()

    opti = torch.optim.Adam(model.parameters(), lr=0.001)



.. GENERATED FROM PYTHON SOURCE LINES 72-73

Define function for testing

.. GENERATED FROM PYTHON SOURCE LINES 73-98

.. code-block:: Python
   :lineno-start: 73

    @torch.no_grad()
    def test():
        metrics = OODMetrics()
        acc = Accuracy(task="multiclass", num_classes=7).to(device)
        model.eval()

        for x, y in tqdm(test_loader):
            # do not forget to remap class labels
            y = torch.tensor([class_mapping(i.item()) for i in y])

            y = y.to(device)
            x = x.to(device)

            z = model(x)

            metrics.update(MaxSoftmax.score(z), y)

            known = is_known(y)
            if known.any():
                acc.update(z[known].argmax(dim=1), y[known])

        print(metrics.compute())
        print(acc.compute().item())



.. GENERATED FROM PYTHON SOURCE LINES 99-100

Start training

.. GENERATED FROM PYTHON SOURCE LINES 100-126

.. code-block:: Python
   :lineno-start: 100

    for epoch in range(num_epochs):
        bar = tqdm(train_loader)
        model.train()
        loss_ema = None
        for x, y in bar:
            # do not forget to remap class labels
            y = torch.tensor([class_mapping(i.item()) for i in y])

            y = y.to(device)
            x = x.to(device)

            z = model(x)

            loss = criterion(z, y)
            opti.zero_grad()
            loss.backward()
            opti.step()

            loss_ema = (
                loss_ema * 0.95 + loss.item() * 0.05
                if loss_ema is not None
                else loss.item()
            )
            bar.set_postfix_str(f"loss: {loss_ema:.2f}")

        test()


.. _sphx_glr_download_auto_examples_osr_osr.py:

.. only:: html

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

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

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

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

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


.. only:: html

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

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