{"url":"/dataset/country211","name":"Country211","full_name":"Country211","description_markdown":"Country211 is a dataset released by OpenAI, designed to assess the geolocation capability of visual representations. It filters the YFCC100m dataset (Thomee et al., 2016) to find 211 countries (defined as having an ISO-3166 country code) that have at least 300 photos with GPS coordinates. OpenAI built a balanced dataset with 211 categories, by sampling 200 photos for training and 100 photos for testing, for each country.","description_withheld":null,"homepage":"https://github.com/openai/CLIP/blob/main/data/country211.md","introduced_date":null,"introduced_date_note":null,"introduced_by":null,"license":{"name":"Creative Commons","url":null},"modalities":[{"name":"Images","url":"/datasets/modality/images"}],"tasks":[{"name":"Image Clustering","url":"/task/image-clustering","datasets_with_task":"/datasets/task/image-clustering"},{"name":"Zero-Shot Image Classification","url":"/task/zero-shot-image-classification","datasets_with_task":"/datasets/task/zero-shot-image-classification"}],"languages":[],"variants":["Country211"],"data_loaders":[{"repo":"https://github.com/pytorch/vision","url":"https://pytorch.org/vision/stable/generated/torchvision.datasets.Country211.html","frameworks":["pytorch"]}],"num_papers_in_archive":3,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/image-clustering-on-country211","task":"Image Clustering","dataset_variant":"Country211","rows":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"TURTLE (CLIP + DINOv2)","paper":"/paper/let-go-of-your-labels-with-unsupervised-1","metrics":{"Accuracy":"11.1"},"code_links":[{"title":"mlbio-epfl/turtle","url":"https://github.com/mlbio-epfl/turtle"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/zero-shot-image-classification-on-country211","task":"Zero-Shot Image Classification","dataset_variant":"Country211","rows":1,"metrics":["Top-1 accuracy"],"first_row_in_archive_order":{"model":"OpenClip H/14 (34B)(Laion2B)","paper":"/paper/reproducible-scaling-laws-for-contrastive","metrics":{"Top-1 accuracy":"30.01"},"code_links":[{"title":"mlfoundations/open_clip","url":"https://github.com/mlfoundations/open_clip"},{"title":"laion-ai/scaling-laws-openclip","url":"https://github.com/laion-ai/scaling-laws-openclip"},{"title":"shkarupa-alex/tfclip","url":"https://github.com/shkarupa-alex/tfclip"},{"title":"nahidalam/open_clip","url":"https://github.com/nahidalam/open_clip"},{"title":"eify/open_clip","url":"https://github.com/eify/open_clip"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/let-go-of-your-labels-with-unsupervised-1","title":"Let Go of Your Labels with Unsupervised Transfer","date":"2024-06-11","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":4,"samples_ran":3,"samples_unverified":1,"pointer_only_for_licence":4,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/reproducible-scaling-laws-for-contrastive","title":"Reproducible scaling laws for contrastive language-image learning","date":"2022-12-14","rows_on_this_dataset":1,"code_links":5,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":3,"samples_ran":3,"samples_unverified":0,"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":2,"samples_harvested":7,"samples_ran":6,"samples_unverified":1,"pointer_only_for_licence":7,"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."}