Loading...
Loading...
General ML benchmarks, tabular data, AutoML, recommendation systems, anomaly detection, evaluation suites
194,257 datasets
HiRISE satellite imagery of Mars's northern mid-latitude craters is annotated for frost presence. The dataset contains labeled 299x299 pixel image tiles at 0.5 m/pixel resolution, with labels determined by a majority vote over polygonal annotations. It was created by S. Diniega of the Jet Propulsion Laboratory using the Labelbox platform.
PsPM-SCRV6 contains skin conductance response measurements from 20 healthy participants in response to discomforting electric shocks. The dataset was authored by Dominik R. Bach at University College London. Stimuli were 0.5ms square pulses repeated at 500Hz for 100ms, with amplitudes averaging 0.78mA.
The Environment Agency holds a Register of Registered Scrap Metal Dealers under the Scrap Metal Dealers Act 2013. This dataset comprises the contents of that Register, with data provided by Local Authorities. The Environment Agency notes it has no control over the accuracy, quality, or completeness of the content.
Environment Agency data comprises office waste produced at qualifying sites for the 2015-16 period. The dataset covers requirements for measurement, monitoring, and disposal arrangements. A qualifying site is defined as having 20 or more Full Time Equivalent (FTE) personnel.
Office waste data from Environment Agency premises in the UK for the 2014-15 period. The dataset covers measurement, monitoring, and disposal arrangements for sites with 20 or more Full Time Equivalent staff. It is provided by the UK Environment Agency and was last updated in July 2026.
The Environment Agency's corporate scorecard provides a high-level overview of performance against environmental and business aims. It is reported every 3 months to executive directors, the board, Defra, and the public on GOV.UK. For each measure, it shows current performance and whether the agency is on track to achieve its annual target.
Will Handley at the University of Cambridge produced these nested sampling inference products and code for a paper published on arXiv in 2021. The dataset likely contains posterior samples and evidence estimates from Bayesian inference runs. It is accompanied by a modified version of the CosmoChord software for nested sampling.
An implementation of the algorithm from Shah (2016) for modeling interactions in high-dimensional data. The algorithm performs sequential Lasso fits on design matrices containing increasing sets of candidate interactions. It was authored by Rajen D. Shah of the University of Cambridge.
REBOUND N-body simulation archives used to train and test the machine learning models in Tamayo et al. (2020). The data supports the SPOCK XGBoost model for predicting the long-term stability of compact multi-planet systems. The dataset is associated with research from Princeton University.
A dataset from Princeton University supporting research on metastable criticality in glassy water-like models. The data, generated by author Thomas E. Gartner, is described as large-scale and is associated with a scientific manuscript. Full structural description and reproduction instructions are provided in the dataset's README file.
SHiELD model simulations of Hurricane Laura using both a regional and a global-nest configuration. The data was generated by Kai‐Yuan Cheng of Princeton University to demonstrate the model's flexibility. Both simulation configurations are available via the Zenodo repository doi.org/10.5281/zenodo.5090124.
2021-2024 is the time frame for this headline customer-focused transport technology strategy from the NSW Government. It outlines Transport for NSW's ambitions, technology toolkit, and six priority programs aimed at transforming the customer experience. The document is a PDF published under an OGL-CA-2.0 license.
Leicester City Council's dataset details recovery activity for Council Tax and Business Rates. It includes the number of properties billed, the volume and value of cases at each recovery stage, and premium charges for long-term vacant and second homes. The data is organized by financial year and updated annually.
Original data from a Nature Communications 2020 study on uranium dioxide formation. The dataset supports the manuscript 'Nanoscale mechanism of UO2 formation through uranium reduction by magnetite' by Zezhen Pan et al. from École Polytechnique Fédérale de Lausanne.
FTIR and AMS spectra from wood and coal burning aerosols generated in an environmental chamber, alongside FTIR spectra of fine particulate matter from the IMPROVE network. The data was analyzed by Amir Yazdani of École Polytechnique Fédérale de Lausanne and published in 2020. For further information, refer to the associated DATASET_README.txt file.
Processed data from the publication 'Retrosplenial cortex enables context-dependent goal-directed sensorimotor transformation' by Bech, Dard et al. in eLife 2026. The dataset is associated with code repositories on GitHub and a data archive on Zenodo, provided by researchers at École Polytechnique Fédérale de Lausanne. It likely contains neural and behavioral recordings related to sensorimotor tasks.
Data supporting a 2022 arXiv publication by researchers from the University of Toronto. The dataset likely contains numerical results related to optimizing quantum measurements of electronic Hamiltonians. It is associated with the paper 'Fluid fermionic fragments for optimizing quantum measurements of electronic Hamiltonians in the variational quantum eigensolver'.
Andrew Styles from the University of Oxford created restart files and namelists for a published ocean modeling study. The data supports the paper 'The Sensitivity of an Idealized Weddell Gyre to Horizontal Resolution' in JGR:Oceans. It is available under an Open Access (green) license.
Samar Khatiwala's repository provides transport matrices and associated data for a specific global ocean model configuration. These matrices represent the advective-diffusive transport operator for the MIT General Circulation Model at a 2.8-degree resolution. The data enables efficient offline simulation of biogeochemical tracers and is linked to the open-source Transport Matrix Method driver code.
London Borough of Barnet's register lists all active and compliant contracts valued at £10,000 or more for the 2018-19 and 2019-20 financial years. The data covers live contracts at the time of publication, providing a snapshot of public procurement and spending. It is published under the Open Government Licence for the United Kingdom (OGL-UK-3.0).