Benchmarks
==========
Monitizer currently supports 9 ID-datasets:
`MNIST `__, `CIFAR-10 `_, `CIFAR-100 `_, `German Traffic Sign Recognition Benchmark (GTSRB) `_, `SVHN `__, `Describable Textures Dataset (DTD) `_, `Fashion-MNIST `_, `Kuzushiji-MNIST `_, `ImageNet `__.
Note that it is fairly simple to include any other dataset, especially if it already exists in `Torchvision `_.
Check out :ref:`dataset-implementation` to see how you can include another dataset. There you will see how you can include your own custom dataset.
.. _cifar10:
CIFAR-10
--------
The CIFAR-10 dataset consists of 60,000 32x32 color images in 10 different classes. It is widely used for object recognition tasks.
`CIFAR-10 Dataset `_
.. image:: ../images/CIFAR10.png
:alt: CIFAR-10 example image
:width: 400px
Monitizer provides four specific OOD classes for CIFAR-10:
* NewWorld/GTSRB: (handselected) GTSRB images share no similarity with any CIFAR-100 image
* NewWorld/DTD: DTD images share no similarity with any CIFAR-10 image
* UnseenObject/Cifar100: the general type of pictures is similar, but the (handselected) CIFAR-100 images contain objects that do not appear in CIFAR-10
* WrongPrediction/FGSM: automatically generated OOD images by using a provided NN to generate adversarial examples
.. _cifar100:
CIFAR-100
---------
Similar to CIFAR-10 but with 100 classes containing 600 images each, CIFAR-100 offers a more challenging classification problem.
`CIFAR-100 Dataset `_
.. image:: ../images/CIFAR100.png
:alt: CIFAR-100 example image
:width: 400px
Monitizer provides three specific OOD classes for CIFAR-100:
* NewWorld/GTSRB: (handselected) GTSRB images share no similarity with any CIFAR-100 image
* NewWorld/DTD: DTD images share no similarity with any CIFAR-100 image
* WrongPrediction/FGSM: automatically generated OOD images by using a provided NN to generate adversarial examples
.. _fashion_mnist:
Fashion-MNIST
-------------
Fashion-MNIST contains 70,000 grayscale images of 28x28 pixels representing 10 different fashion categories.
`Fashion-MNIST Dataset `_
.. image:: ../images/FashionMNIST.png
:alt: Fashion-MNIST example image
:width: 400px
Monitizer provides one specific OOD classes for FashionMNIST
* WrongPrediction/FGSM: automatically generated OOD images by using a provided NN to generate adversarial examples
.. _gtsrb:
GTSRB (German Traffic Sign Recognition Benchmark)
--------------------------------------------------
GTSRB is a multi-class traffic sign recognition dataset with more than 50,000 images of traffic signs in different conditions.
`GTSRB Dataset `_
.. image:: ../images/GTSRB.png
:alt: GTSRB example image
:width: 400px
Monitizer provides four specific OOD classes for GTSRB:
* NewWorld/CIFAR10: (handselected) CIFAR-10 images share no similarity with any GTSRB image
* NewWorld/DTD: DTD images share no similarity with any GTSRB image
* NewWorld/SVHN: (handselected) SVHN images that share no similarity with any GTSRB image
* WrongPrediction/FGSM: automatically generated OOD images by using a provided NN to generate adversarial examples
.. _imagenet:
ImageNet
--------
ImageNet is a large-scale dataset with over 14 million images across more than 20,000 categories, commonly used for deep learning.
`ImageNet Dataset `_
.. image:: ../images/Imagenet.png
:alt: ImageNet example image
:width: 400px
.. _kmnist:
KMNIST (Kuzushiji-MNIST)
------------------------
KMNIST is a dataset of 70,000 grayscale images of Japanese Kuzushiji characters, designed as a drop-in replacement for MNIST.
`KMNIST Dataset `_
.. image:: ../images/KMNIST.png
:alt: KMNIST example image
:width: 400px
Monitizer provides one specific OOD classes for FashionMNIST
* WrongPrediction/FGSM: automatically generated OOD images by using a provided NN to generate adversarial examples
.. _mnist:
MNIST
-----
MNIST contains 70,000 grayscale images of handwritten digits (0-9) and is a classic benchmark for image classification.
`MNIST Dataset `_
.. image:: ../images/MNIST.png
:alt: MNIST example image
:width: 400px
Monitizer provides six specific OOD classes for MNIST:
* NewWorld/CIFAR10: CIFAR-10 images share no similarity with any MNIST image
* NewWorld/DTD: DTD images share no similarity with any MNIST image
* UnseenEnvironment/SVHN: SVHN also shows numbers but in a different style than MNIST
* UnseenObject/FashionMNIST: the type of images is the same, but the objects differ
* UnseenObject/KMNIST: the type of images is the same, but the objects differ
* WrongPrediction/FGSM: automatically generated OOD images by using a provided NN to generate adversarial examples
.. _svhn:
SVHN (Street View House Numbers)
--------------------------------
SVHN consists of real-world house number images obtained from Google Street View, used for digit recognition tasks.
`SVHN Dataset `_
.. image:: ../images/SVHN.png
:alt: SVHN example image
:width: 400px
Monitizer provides one specific OOD classes for FashionMNIST
* WrongPrediction/FGSM: automatically generated OOD images by using a provided NN to generate adversarial examples
.. _dtd:
DTD (Describing Textures in the Wild)
-------------------------------------
DTD contains 5,640 images categorized into 47 texture classes and is used for texture recognition and classification.
`DTD Dataset `_
.. image:: ../images/DTD.png
:alt: DTD example image
:width: 400px
Monitizer provides one specific OOD classes for FashionMNIST
* WrongPrediction/FGSM: automatically generated OOD images by using a provided NN to generate adversarial examples