Loading...
Loading...
General ML benchmarks, tabular data, AutoML, recommendation systems, anomaly detection, evaluation suites
193,236 datasets
Jure Leskovec of Stanford University authored a book focusing on practical algorithms for data mining key problems in extremely large datasets from Web and Internet commerce. The third edition includes new and extended coverage on decision trees, deep learning, and mining social-network graphs. It discusses the MapReduce framework, locality-sensitive hashing, stream-processing algorithms, PageRank, frequent itemset discovery, and clustering.
Jure Leskovec of Stanford University authored this conceptual resource on dimensionality reduction. It describes the process of summarizing large matrices, such as web transition, utility, and social network matrices, into narrower approximations. The dataset's specific size, format, and column details are not provided.
Erythroid-specific 5’-Aminolevulinate synthase (ALAS2) catalyzes the first and rate-determining step in haem biosynthesis. This Target Enabling Package from the University of Oxford provides structural biology tools to investigate gain-of-function mutations and explore substrate reduction therapy for porphyria disorders.
Data files and a Python Jupyter notebook to reproduce Figures 2 through 6 from the manuscript 'Turbulent drag at the ice-ocean interface of Europa in simulations of rotating convection: Implications for nonsynchronous rotation of the ice shell'. The package includes analysis code, grid information files, plotting functions, and input files for the MITgcm model. It was authored by Hamish Hay of the University of Oxford.
Quarterly apprenticeship data tables published by the Skills Funding Agency. The data includes breakdowns for starts, participation, and achievements on apprenticeship programmes in England. The latest update was on 2026-07-08 13:33:25.121656.
A Skills Funding Agency register details subcontractors holding at least £100,000 in aggregate contracts with lead education providers for an academic year. The list is compiled from Subcontractor Declaration Forms submitted by lead providers and includes individual contract values. This data was last updated on 2026-07 08.
Camden Licensing Conditions Beta is a companion dataset designed to be joined with a primary licensing register. Its structure suggests it contains specific regulatory stipulations attached to individual licenses. The dataset is explicitly marked as being in a beta phase of development. Multiple platforms list it under an Open Government Licence, indicating its public sector origin and intended reuse.
Brian Trammell of ETH Zurich authored a paper proposing an architectural solution for data exchange. The paper discusses improving the rigor of large-scale computer and network security studies through better data sharing and analysis. The dataset likely contains information related to the proposed framework for cooperative security research.
Sentinel-1 A, B, and C satellites have collected directional ocean swell wave measurements globally since 2014. This dataset provides homogenised swell wave number spectra, spectral partitions, and partition bulks for the Australasian region, including the Pacific islands, with data available from July 2015. The Australian Ocean Data Network presents these measurements in a consistent, near real-time format.
GPS tracks of little penguins (Eudyptula minor) tagged at three colonies around Storm Bay, Tasmania during the 2015/16 summer breeding season. This data was collected as part of an honours thesis submitted in 2016 and is hosted by the Australian Ocean Data Network.
Government of Nova Scotia publishes summaries of recent disciplinary actions regarding non-compliant matters brought before the Executive Director of the Alcohol, Gaming, Fuel and Tobacco Division. The data includes notices of proposed action and details on sanctions, which may vary in length according to case specifics. It was last updated on 2026-06-29 and is available in XML, CSV, HTML, and RSS formats under the OGL-CA-2.0 license.
NASA provides estimates of carbon dioxide and methane diffusive fluxes from approximately 17,000 waterbodies in the central-interior Yukon-Kuskokwim Delta, Alaska. A gradient boosting machine learning model predicted dissolved gas concentrations using Sentinel-2, Sentinel-1, and DEM remote sensing data, trained on ~300 field samples collected in July from 2016 to 2019. The resulting data, provided as GeoTIFF and shapefile formats, include watershed landcover and scaled flux estimates.
Experimental results for State of Charge (SOC) estimation errors in lithium-ion batteries under different temperature conditions. The 5.5 KB XLS file was authored by Huipin Lin and last updated on June 4, 2026. It contains results from experiments using the University of Maryland’s Dynamic Stress Test (DST), US06 test, and Federal Urban Driving Scheme (FUDS) datasets.
Huipin Lin published experimental results on June 4, 2026, comparing a hybrid time-series convolutional network with Unscented Kalman Filter against other deep learning models for battery state-of-charge estimation. The dataset likely contains error metrics from tests using the University of Maryland's Dynamic Stress Test (DST), US06, and Federal Urban Driving Scheme (FUDS) datasets. The proposed method achieved mean absolute error values between 1.015% and 1.470% under different driving conditions.
Northern Ireland's camping, glamping, and caravan sites are mapped in this geospatial point dataset. Locations were digitized from the Camping NI website and verified against the latest OSNI Orthophotography. The data supports tourism planning and outdoor recreation analysis.
63 tiles form the digital state map of the entire Austrian territory at a basic scale of 1:50,000. The cartographic model KM50-R is free of blades, georeferenced, and updated nationwide at periodic intervals or for significant infrastructural changes. It is produced through a cooperation between OGD Österreich and Wikimedia Österreich.
A digital state map of the entire Austrian territory at a basic scale of 1:50,000. The cartographic model is free of blades and georeferenced, and it is updated nationwide at periodic intervals and for significant infrastructural changes. This specific file is one of 63 tiles that together cover Austria.
KM50-R is the digital state map covering the entire territory of Austria at a scale of 1:50,000. It is a georeferenced, seamless raster dataset updated periodically and used as the basis for other official Austrian maps. The data is provided as one of 63 tiles covering the country and is published under a CC-BY-4.0 license by a cooperation including Wikimedia Österreich.
A 50 km x 50 km tile from the KM50-R cartographic model, which is the digital state map of Austria at a 1:50,000 scale. The dataset is one of 63 tiles covering the entire Austrian territory, is free of blades and georeferenced, and is updated periodically and for significant infrastructural changes. It is provided by Cooperation OGD Österreich and Wikimedia Österreich under a CC-BY-4.0 license.
A digital state map covering the entire Austrian territory at a basic scale of 1:50,000, free of map sheet edges and georeferenced. The cartographic model is updated nationwide at periodic intervals and also on occasions of significant infrastructural change. This specific 50 km x 50 km tile is one of 63 tiles that together form the complete coverage of Austria.