{"url":"/dataset/ucr-time-series-classification-archive","name":"UCR Time Series Classification Archive","full_name":"UCR Time Series Classification Archive","description_markdown":"The UCR Time Series Archive - introduced in 2002,\r\nhas become an important resource in the time series data mining\r\ncommunity, with at least one thousand published papers making\r\nuse of at least one data set from the archive. The original\r\nincarnation of the archive had sixteen data sets but since that\r\ntime, it has gone through periodic expansions. The last expansion\r\ntook place in the summer of 2015 when the archive grew from\r\n45 to 85 data sets. This paper introduces and will focus on the\r\nnew data expansion from 85 to 128 data sets. Beyond expanding\r\nthis valuable resource, this paper offers pragmatic advice to\r\nanyone who may wish to evaluate a new algorithm on the archive.\r\nFinally, this paper makes a novel and yet actionable claim: of the\r\nhundreds of papers that show an improvement over the standard\r\nbaseline (1-nearest neighbor classification), a large fraction may\r\nbe misattributing the reasons for their improvement. Moreover,\r\nthey may have been able to achieve the same improvement with\r\na much simpler modification, requiring just a single line of code.","description_withheld":null,"homepage":"https://www.cs.ucr.edu/~eamonn/time_series_data_2018/","introduced_date":"2018-10-17","introduced_date_note":null,"introduced_by":{"paper":"/paper/the-ucr-time-series-archive","title":"The UCR Time Series Archive","first_author":"Hoang Anh Dau","url":null},"license":null,"modalities":[{"name":"Audio","url":"/datasets/modality/audio"},{"name":"Time series","url":"/datasets/modality/time-series"}],"tasks":[{"name":"Time Series Classification","url":"/task/time-series-classification","datasets_with_task":"/datasets/task/time-series-classification"},{"name":"Audio Classification","url":"/task/audio-classification","datasets_with_task":"/datasets/task/audio-classification"},{"name":"ECG Classification","url":"/task/ecg-classification","datasets_with_task":"/datasets/task/ecg-classification"},{"name":"Time Series Clustering","url":"/task/time-series-clustering","datasets_with_task":"/datasets/task/time-series-clustering"},{"name":"Time Series Averaging","url":"/task/time-series-averaging","datasets_with_task":"/datasets/task/time-series-averaging"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["UCR Time Series Classification Archive"],"data_loaders":[{"repo":"https://github.com/yuezhihan/ts2vec","url":"https://github.com/yuezhihan/ts2vec/blob/main/datasets/preprocess_electricity.py","frameworks":["pytorch"]},{"repo":"https://github.com/WenjieDu/TSDB","url":"https://github.com/WenjieDu/TSDB","frameworks":["pytorch"]}],"num_papers_in_archive":42,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/audio-classification-on-ucr-time-series","task":"Audio Classification","dataset_variant":"UCR Time Series Classification Archive","rows":1,"metrics":["FruitFlies","MosquitoSound","RightWhaleCalls"],"first_row_in_archive_order":{"model":"CDIL","paper":"/paper/classification-of-long-sequential-data-using","metrics":{"FruitFlies":"97.09","MosquitoSound":"91.54","RightWhaleCalls":"91.99"},"code_links":[{"title":"leicheng-no/cdil-cnn","url":"https://github.com/leicheng-no/cdil-cnn"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/ecg-classification-on-ucr-time-series","task":"ECG Classification","dataset_variant":"UCR Time Series Classification Archive","rows":1,"metrics":["Accuracy (Test)"],"first_row_in_archive_order":{"model":"V2Sa","paper":"/paper/voice2series-reprogramming-acoustic-models","metrics":{"Accuracy (Test)":"93.96"},"code_links":[{"title":"huckiyang/Voice2Series-Reprogramming","url":"https://github.com/huckiyang/Voice2Series-Reprogramming"},{"title":"srijith-rkr/kaust-whisper-adapter","url":"https://github.com/srijith-rkr/kaust-whisper-adapter"},{"title":"dodohow1011/speechadvreprogram","url":"https://github.com/dodohow1011/speechadvreprogram"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/classification-of-long-sequential-data-using","title":"Classification of Long Sequential Data using Circular Dilated Convolutional Neural Networks","date":"2022-01-06","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/voice2series-reprogramming-acoustic-models","title":"Voice2Series: Reprogramming Acoustic Models for Time Series Classification","date":"2021-06-17","rows_on_this_dataset":1,"code_links":3,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":7,"samples_ran":1,"samples_unverified":6,"pointer_only_for_licence":3,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":1,"samples_harvested":7,"samples_ran":1,"samples_unverified":6,"pointer_only_for_licence":3,"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."}