{"url":"/dataset/qmnist","name":"QMNIST","full_name":null,"description_markdown":"The exact pre-processing steps used to construct the MNIST dataset have long been lost. This leaves us with no reliable way to associate its characters with the ID of the writer and little hope to recover the full MNIST testing set that had 60K images but was never released. The official MNIST testing set only contains 10K randomly sampled images and is often considered too small to provide meaningful confidence intervals.\nThe **QMNIST** dataset was generated from the original data found in the NIST Special Database 19 with the goal to match the MNIST preprocessing as closely as possible.\nQMNIST is licensed under the BSD-style license.\n\nSource: [https://github.com/facebookresearch/qmnist](https://github.com/facebookresearch/qmnist)\nImage Source: [https://github.com/facebookresearch/qmnist](https://github.com/facebookresearch/qmnist)","description_withheld":null,"homepage":"https://github.com/facebookresearch/qmnist","introduced_date":null,"introduced_date_note":null,"introduced_by":{"paper":"/paper/cold-case-the-lost-mnist-digits","title":"Cold Case: The Lost MNIST Digits","first_author":"Chhavi Yadav","url":null},"license":null,"modalities":[{"name":"Images","url":"/datasets/modality/images"}],"tasks":[{"name":"Image Classification","url":"/task/image-classification","datasets_with_task":"/datasets/task/image-classification"},{"name":"Fine-Grained Image Classification","url":"/task/fine-grained-image-classification","datasets_with_task":"/datasets/task/fine-grained-image-classification"}],"languages":[],"variants":["QMNIST"],"data_loaders":[{"repo":"https://github.com/pytorch/vision","url":"https://pytorch.org/vision/stable/generated/torchvision.datasets.QMNIST.html","frameworks":["pytorch"]},{"repo":"https://github.com/facebookresearch/qmnist","url":"https://github.com/facebookresearch/qmnist","frameworks":["pytorch"]}],"num_papers_in_archive":26,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/fine-grained-image-classification-on-qmnist","task":"Fine-Grained Image Classification","dataset_variant":"QMNIST","rows":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"VGG-5","paper":"/paper/progressivespinalnet-architecture-for-fc","metrics":{"Accuracy":"99.6867"},"code_links":[{"title":"praveenchopra/ProgressiveSpinalNet","url":"https://github.com/praveenchopra/ProgressiveSpinalNet"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/image-classification-on-qmnist","task":"Image Classification","dataset_variant":"QMNIST","rows":1,"metrics":["Accuracy (%)"],"first_row_in_archive_order":{"model":"Deep regularization","paper":"/paper/deep-regularization-and-direct-training-of","metrics":{"Accuracy (%)":"99.67"},"code_links":[{"title":"kernel-enthusiasts/KF_NN2","url":"https://github.com/kernel-enthusiasts/KF_NN2"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/progressivespinalnet-architecture-for-fc","title":"ProgressiveSpinalNet architecture for FC layers","date":"2021-03-21","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/deep-regularization-and-direct-training-of","title":"Deep regularization and direct training of the inner layers of Neural Networks with Kernel Flows","date":"2020-02-19","rows_on_this_dataset":1,"code_links":1,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}