Published July 2, 2026 | Version v1

Embedded representations of popular image datasets (CIFAR10, CIFAR100, CIFAR100coarse, MNIST, FashionMNIST, SVHN)

  • 1. ROR icon Universidad de Oviedo

Description

Vectorial representations for common image datasets:

  • CIFAR10, CIFAR100, CIFAR100coarse
  • MNIST, FashionMNIST
  • SVHN

Extracted using the scripts in this repo, where more details of the extraction can be consulted.

The embeddings are provided in different modalities:

  • features: the next-to-last representation of the network
  • logits: the pre-softmax values
  • predictions: the post-softmax values (posterior probabilities)
  • targets: the true labels

Each file is stored in format npz (dictionary-like) and contains the numpy representations for the keys "train", "val", and "test" partitions for each dataset. The "train" partition was used to train the corresponding neural model, "val" was used for training monitoring, and "test" was not used during training.

The name of the file uses the format <dataset>_<neural-model>_<modality>.npz; for example: "svhn_resnet18_logits.npz".

See reference list for publications related to each dataset.

Files

Files (728.6 MB)

Name Size
md5:ac417079b7373c169414b87404d9003f
122.9 MB Download
md5:5ea758aa14ca97719f5f2995f7ec0aaf
24.0 MB Download
md5:548751bc94f0b784d971b1a000d1024a
24.0 MB Download
md5:cbe14877695e7302d1f358cdc70f1c4f
480.7 kB Download
md5:ced20628bd354dbae2215bb3ff2208d1
122.9 MB Download
md5:8de1d839dcc7bc0798f88caef829d360
4.8 MB Download
md5:747284632b0a002131a68ec24a52e4b9
4.8 MB Download
md5:705454f06f7c4a2eaa3ba9aa93af1276
480.7 kB Download
md5:ff56a2ce441a77dc12c828a7e7923f62
122.9 MB Download
md5:51a845f14879afb09d9b838802fcf37f
2.4 MB Download
md5:233f09e08702c3d607873b232b75e8fc
2.4 MB Download
md5:6a01bb9b2ae23916c32750d1216ac710
480.7 kB Download
md5:eb2f21edb4caf9a3af0ea7a27c1a2340
35.8 MB Download
md5:02fa457b7d634d0c71f895899ba54fc5
2.8 MB Download
md5:71d937586f818d527e19e984c78a5b67
2.8 MB Download
md5:a7e63337abee86cc8ef9121943c591ee
560.7 kB Download
md5:fe33842d8b3ca471c422ea88b2e01c1b
35.8 MB Download
md5:73c367caf834a6deb216169c4ca4a032
2.8 MB Download
md5:dcd4a87e0b73e1aa03bab63b5ed4de2a
2.8 MB Download
md5:33ddc2cd64ddbead8c74d0a344c42449
560.7 kB Download
md5:cc2f08e21eee932fa27aee97cf9cc226
203.3 MB Download
md5:8e4a28ca236b699e9c16c5daa2010c98
4.0 MB Download
md5:710e035f7565aa9d226085e469d3f46f
4.0 MB Download
md5:c8f215ba84b558377c73b93a75e2382a
795.1 kB Download

Additional details

Software

Repository URL
https://github.com/pglez82/visiondatasets_quapy
Programming language
Python
Development Status
Active

References

  • CIFAR10, CIFAR100, CIFAR100coarse: Alex Krizhevsky and Geoffrey Hinton. Learning multiple layers of features from tiny images. Technical report, University of Toronto, Toronto, Ontario, 2009.
  • MNIST: Yann LeCun, Corinna Cortes, and Christopher J. C. Burges. The MNIST database of handwritten digits. http://yann.lecun.com/exdb/mnist/, 1998.
  • FashionMNIST: Han Xiao, Kashif Rasul, and Roland Vollgraf. Fashion-MNIST: a novel image dataset for benchmarking machine learning algorithms. arXiv preprint arXiv:1708.07747, 2017.
  • SVHN: Yuval Netzer, Tao Wang, Adam Coates, Alessandro Bissacco, Baolin Wu, Andrew Y Ng, et al. Reading digits in natural images with unsupervised feature learning. In NIPS workshop on deep learning and unsupervised feature learning, volume 2011, page 4. Granada, 2011.