Configure STL Dataset (STL)

Modified

October 9, 2025

The single-task learning dataset is a single dataset which has training, validation and test data.

STL dataset is a sub-config under the index config of:

Single-task learning experiment and evaluation

To configure a custom STL dataset, you need to create a YAML file in stl_dataset/ folder. Below shows examples of the STL dataset config.

Example

configs
├── __init__.py
├── entrance.yaml
├── index
│   ├── example_stl_expr.yaml
│   └── ...
├── stl_dataset
│   └── mnist.yaml
...

example_configs/index/example_stl_expr.yaml

defaults:
  ...
    - /stl_dataset: mnist.yaml
  ...

example_configs/stl_dataset/mnist.yaml

_target_: clarena.stl_datasets.MNIST
root: data/MNIST
validation_percentage: 0.1
batch_size: 64

Supported STL Datasets & Required Config Fields

In CLArena, we have implemented many STL datasets as Python classes in clarena.stl_datasets module that you can use for your experiments and evaluations.

To choose a STL dataset, assign the _target_ field to the class name of the STL dataset. For example, to use the MNIST dataset, set _target_ field to clarena.stl_datasets.MNIST. Each STL dataset has its own hyperparameters and configurations, which means it has its own required fields. The required fields are the same as the arguments of the class specified by _target_. The arguments of each STL dataset class can be found in API documentation.

API Reference (STL Datasets) Source Code (STL Datasets)

Below is the full list of supported STL datasets. We only support image classification datasets. Note that the “STL Dataset” is exactly the class name that the _target_ field is assigned.

STL Dataset	Description	Required Config Fields
ArabicHandwrittenDigits	Arabic Handwritten Digits dataset. The Arabic Handwritten Digits Dataset (AHDD) is a collection of handwritten Arabic digits (0-9). It consists of 60,000 training and 10,000 test images of handwritten Arabic digits (10 classes), each 28x28 grayscale image (similar to MNIST).	Same as ArabicHandwrittenDigits class arguments
Caltech101	Caltech 101 dataset. The Caltech 101 dataset is a collection of pictures of objects. It consists of 9,146 images of 101 classes, each color image.	Same as Caltech101 class arguments
Caltech256	Caltech 256 dataset. The Caltech 256 dataset is a collection of pictures of objects. It consists of 30,607 images of 256 classes, each color image.	Same as Caltech256 class arguments
CelebA	CelebA dataset. The CelebFaces Attributes Dataset (CelebA) is a large-scale celebrity faces dataset. It consists of 202,599 face images of 10,177 celebrity identities (classes), each 178x218 color image. Note that the original CelebA dataset is not a classification dataset but a attributes dataset. We only use the identity of each face as the class label for classification.	Same as CelebA class arguments
CIFAR10	CIFAR-10 dataset. The CIFAR-10 dataset is a subset of the 80 million tiny images dataset. It consists of 50,000 training and 10,000 test images of 10 classes, each 32x32 color image.	Same as CIFAR10 class arguments
CIFAR100	CIFAR-100 dataset. The CIFAR-100 dataset is a subset of the 80 million tiny images dataset. It consists of 50,000 training and 10,000 test images of 100 classes, each 32x32 color image.	Same as CIFAR100 class arguments
Country211	Country211 dataset. The Country211 dataset is a collection of geolocation pictures of different countries. It consists of 62,200 images of 211 countries (classes), each 256x256 color image.	Same as Country211 class arguments
CUB2002011	CUB-200-2011 dataset. The CUB (Caltech-UCSD Birds)-200-2011) is a bird image dataset. It consists of 11,788 images of 200 bird species (classes), each 64x64 color image.	Same as CUB2002011 class arguments
DTD	DTD dataset. The Describable Textures Dataset (DTD) is a collection of describable texture pictures. It consists of 5,640 images of 47 kinds of textures (classes), each 300x300-640x640 color image.	Same as DTD class arguments
EMNIST	EMNIST dataset. The EMNIST dataset is a collection of handwritten letters and digits (including A-Z, a-z, 0-9). It consists of 814,255 images in 62 classes, each 28x28 grayscale image. EMNIST has 6 different splits: `byclass`, `bymerge`, `balanced`, `letters`, `digits` and `mnist`, each containing a different subset of the original collection. We support all of them in Permuted EMNIST.	Same as EMNIST class arguments
EuroSAT	EuroSAT dataset. The EuroSAT dataset is a collection of satellite images of lands. It consists of 27,000 images of 10 classes, each 64x64 color image.	Same as EuroSAT class arguments
FaceScrub	FaceScrub dataset. The original FaceScrub dataset is a collection of human face images. It consists 106,863 images of 530 people (classes), each high resolution color image. To make it simple, this version uses subset of the official Megaface FaceScrub challenge, cropped and resized to 32x32. We have FaceScrub-10, FaceScrub-20, FaceScrub-50, FaceScrub-100 datasets where the number of classes are 10, 20, 50 and 100 respectively.	Same as FaceScrub class arguments
FashionMNIST	Fashion-MNIST dataset. The Fashion-MNIST dataset is a collection of fashion images. It consists of 60,000 training and 10,000 test images of 10 types of clothing (classes), each 28x28 grayscale image (similar to MNIST).	Same as FashionMNIST class arguments
FER2013	FER2013 dataset. The FER2013 dataset is a collection of facial expression images. It consists of 35,887 images of 7 facial expressions (classes), each 48x48 grayscale image.	Same as FER2013 class arguments
FGVCAircraft	FGVC-Aircraft dataset. The FGVC-Aircraft dataset is a collection of aircraft images. It consists of 10,200 images, each color image. FGVC-Aircraft has 3 different class labels by variant, family and manufacturer, which has 102, 70, 41 classes respectively. We support all of them in Permuted FGVC-Aircraft.	Same as FGVCAircraft class arguments
Flowers102	Oxford 102 Flower dataset. The Oxford 102 Flower dataset is a collection of flower pictures. It consists of 8,189 images of 102 kinds of flowers (classes), each color image.	Same as Flowers102 class arguments
Food101	Food-101 dataset. The Food-101 dataset is a collection of food images. It consists of 101,000 images of 101 classes, each color image.	Same as Food101 class arguments
GTSRB	GTSRB dataset. The GTSRB dataset is a collection of traffic sign images. It consists of 51,839 images of 43 different traffic signs (classes), each color image.	Same as GTSRB class arguments
Imagenette	Imagenette dataset. The Imagenette dataset is a subset of 10 easily classified classes from Imagenet. Permuted Linnaeus 5 dataset. The Linnaeus 5 dataset is a collection of flower images. It consists of 8,000 images of 5 flower species (classes). It provides 256x256, 128x128, 64x64, and 32x32 color images. We support all of them in Permuted Linnaeus 5. We support all of them in Permuted Imagenette.	Same as Imagenette class arguments
KannadaMNIST	Kannada-MNIST dataset. The Kannada-MNIST dataset is a collection of handwritten Kannada digits (0-9). It consists of 60,000 training and 10,000 test images of handwritten Kannada digits (10 classes), each 28x28 grayscale image (similar to MNIST).	Same as KannadaMNIST class arguments
KMNIST	Kuzushiji-MNIST dataset. The Kuzushiji-MNIST dataset is a collection of Japanese Kuzushiji character images. It consists of 60,000 training and 10,000 test images of Japanese Kuzushiji images (10 classes), each 28x28 grayscale image (similar to MNIST).	Same as KMNIST class arguments
Linnaeus5	Linnaeus 5 dataset. The Linnaeus 5 dataset is a collection of flower images. It consists of 8,000 images of 5 flower species (classes). It provides 256x256, 128x128, 64x64, and 32x32 color images. We support all of them in Permuted Linnaeus 5.	Same as Linnaeus5 class arguments
MNIST	MNIST dataset. The MNIST dataset is a collection of handwritten digits. It consists of 60,000 training and 10,000 test images of handwritten digit images (10 classes), each 28x28 grayscale image.	Same as MNIST class arguments
NotMNIST	NotMNIST dataset. The NotMNIST dataset is a collection of letters (A-J). Permuted MNIST dataset. This version uses the smaller set, which consists of about 19,000 images of 10 classes, each 28x28 grayscale image.	Same as NotMNIST class arguments
OxfordIIITPet	Oxford-IIIT Pet dataset. The Oxford-IIIT Pet dataset is a collection of cat and dog pictures. It consists of 7,349 images of 37 breeds (classes), each color image. It also provides a binary classification version with 2 classes (cat or dog). We support both versions in Permuted Oxford-IIIT Pet.	Same as OxfordIIITPet class arguments
PCAM	PCAM dataset. The PCAM dataset is a collection of medical images of breast cancer. It consists of 327,680 images in 2 classes (benign and malignant), each 96x96 color image.	Same as PCAM class arguments
RenderedSST2	Rendered SST2 dataset. The Rendered SST2 dataset is a collection of optical character recognition images. It consists of 9,613 images in 2 classes (positive and negative sentiment), each 448x448 color image.	Same as RenderedSST2 class arguments
SEMEION	SEMEION dataset. The SEMEION dataset is a collection of handwritten digits. It consists of 1,593 handwritten digit images (10 classes), each 16x16 grayscale image.	Same as SEMEION class arguments
SignLanguageMNIST	Sign Language MNIST dataset. The Sign Language MNIST dataset is a collection of hand gesture images representing ASL letters (A-Y, excluding J). It consists of 34,627 images of 24 classes, each 28x28 grayscale image.	Same as SignLanguageMNIST class arguments
StanfordCars (download link expired)	Stanford Cars dataset. The Stanford Cars dataset is a collection of car images. It consists of 16,185 images in 196 classes, each color image.	Same as StanfordCars class arguments
SUN397	SUN397 dataset. The SUN397 dataset is a collection of scene images. It consists of 108,754 images of 397 classes, each color image.	Same as SUN397 class arguments
SVHN	SVHN dataset. The SVHN dataset is a collection of street view house number images. It consists 73,257 training and 26,032 test images of 10 classes, each 32x32 color image.	Same as SVHN class arguments
TinyImageNet	TinyImageNet dataset. The TinyImageNet dataset is smaller, more manageable version of the Imagenet dataset. It consists of 100,000 training, 10,000 validation and 10,000 test images of 200 classes, each 64x64 color image.	Same as TinyImageNet class arguments
USPS	USPS dataset. The USPS dataset is a collection of handwritten digits. It consists of 9,298 handwritten digit images (10 classes), each 16x16 grayscale image.	Same as USPS class arguments