{"url":"/dataset/pamap2","name":"PAMAP2","full_name":null,"description_markdown":"The PAMAP2 Physical Activity Monitoring dataset contains data of 18 different physical activities (such as walking, cycling, playing soccer, etc.), performed by 9 subjects wearing 3 inertial measurement units and a heart rate monitor. The dataset can be used for activity recognition and intensity estimation, while developing and applying algorithms of data processing, segmentation, feature extraction and classification.\r\n\r\n** Sensors **\r\n3 Colibri wireless inertial measurement units (IMU):\r\n  - sampling frequency: 100Hz\r\n  - position of the sensors:\r\n       - 1 IMU over the wrist on the dominant arm \r\n       - 1 IMU on the chest \r\n       - 1 IMU on the dominant side's ankle \r\nHR-monitor:\r\n  - sampling frequency: ~9Hz\r\n\r\n** Data collection protocol **\r\nEach of the subjects had to follow a protocol, containing 12 different activities. The folder Protocol contains these recordings by subject.\r\nFurthermore, some of the subjects also performed a few optional activities. The folder Optional contains these recordings by subject.\r\n\r\n** Data files **\r\nRaw sensory data can be found in space-separated text-files (.dat), 1 data file per subject per session (protocol or optional). Missing values are indicated with NaN. One line in the data files correspond to one timestamped and labeled instance of sensory data. The data files contain 54 columns: each line consists of a timestamp, an activity label (the ground truth) and 52 attributes of raw sensory data.\r\n\r\nSource: [UCI](https://archive.ics.uci.edu/dataset/231/pamap2+physical+activity+monitoring)","description_withheld":null,"homepage":"","introduced_date":"2012-06-18","introduced_date_note":null,"introduced_by":{"paper":"/paper/introducing-a-new-benchmarked-dataset-for","title":"Introducing a new benchmarked dataset for activity monitoring","first_author":"Attila Reiss","url":null},"license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"modalities":[{"name":"Time series","url":"/datasets/modality/time-series"}],"tasks":[{"name":"Human Activity Recognition","url":"/task/human-activity-recognition","datasets_with_task":"/datasets/task/human-activity-recognition"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["PAMAP2"],"data_loaders":[],"num_papers_in_archive":176,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/human-activity-recognition-on-pamap2","task":"Human Activity Recognition","dataset_variant":"PAMAP2","rows":3,"metrics":["NMI","ARI","Accuracy","Macro F1"],"first_row_in_archive_order":{"model":"Selective HAR Clustering","paper":"/paper/efficient-deep-clustering-of-human-activities","metrics":{"ARI":"0.814","NMI":"0.884"},"code_links":[{"title":"Lou1sM/HAR","url":"https://github.com/Lou1sM/HAR"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/virtual-fusion-with-contrastive-learning-for","title":"Virtual Fusion with Contrastive Learning for Single Sensor-based Activity Recognition","date":"2023-12-01","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/efficient-deep-clustering-of-human-activities","title":"Efficient Deep Clustering of Human Activities and How to Improve Evaluation","date":"2022-09-17","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."}