Loading...
Loading...
Image classification, object detection, segmentation, face recognition, OCR, image generation, video understanding
15,368 datasets
5.5 KB of training configuration data for standalone and ensemble machine learning models, authored by Vijay U. Rathod and shared on figshare under a CC-BY-4.0 license. The dataset was last updated on May 8, 2026. It contains configurations for models including MLP, CNN, FT-Transformer, and their ensembles with Autoencoders.
28 focus group discussions with 210 participants and 25 semi-structured interviews were conducted across over 70 villages in two provinces. The dataset, authored by Adelina Chandra, is available in XLS format under a CC-BY-4.0 license and was last updated on 2026-05 08.
Monthly statistics from Canada's Incentives for Medium- and Heavy-Duty Zero-Emission Vehicles (iMHZEV) Program, launched July 11, 2022. The program offers incentives up to $200,000 per vehicle, with data searchable by province, postal code, vehicle make/model, engine type, and time period.
UNHCR collected 875 household responses regarding relocation intentions from Congolese refugees in the Central African Republic's Toko-kota area. The survey, conducted in 2022, assessed intentions to relocate to Kouango or return home, and the reasons for fearing their current location. The data represents households hosting approximately 3628 individuals according to a report from 31 May 2022.
December 2022 exploration drilling site survey for the UK oil and gas industry, acquired under licence P2500. The British Geological Survey (BGS) provides this record of a survey traversing block 42/15a in the UK Continental Shelf. Data was last updated in the platform on 2026-05-28.
Chronicles-OCR is the first benchmark designed to evaluate the cross-temporal visual perception capabilities of Visual Language Models across the complete evolutionary trajectory of Chinese characters, known as the 'Seven Chinese Scripts'. The dataset was curated in collaboration with institutional domain experts and is hosted on HuggingFace by author VirtualLUO. It was last updated on May 18, 2026.
5.5 KB Excel file authored by Yanran Chen and last updated in April 2026. It contains counts of sentences used in Universal Dependencies (UD) treebanks and for an OCR spelling attack. The test UD treebank size is the sum of all treebanks, and the OCR attack value totals six levels at 2k sentences each.
Avinash Bansal published this 9.5 KB Excel file on figshare in April 2026. The table provides detailed training configurations for pose estimation models, corresponding to results in a published research paper. Its small size suggests it is a focused, supplementary dataset.
Andrew M. K. Nassief proposes a distributed computing architecture for generating synthetic biomedical data to address statistical scarcity. The approach leverages regressional Generative Adversarial Networks (GANs) and platforms like BOINC and the Decentralized-Internet SDK to increase data pools for rare diseases. This conceptual proposal is published on paperswithcode under an Open Access license.
Prior to the mid-twentieth century, this dataset contains a geospatial database of 1,562 Buddhist monasteries, temples, hermitages, shrines, and Bonpo religious establishments. It includes point locations, reconstructed regional systems, transportation corridors, and GIS data, compiled by Karl Ryavec for the Tibetan Monastic Ritual Economy Model Database.
Sixteen DC-8 missions collected data from 27 May to 24 June 2017 during the NASA Convective Processes Experiment (CPEX) field campaign. The dataset contains satellite products from MetOp-A, MetOp-B, NOAA-18, and NOAA-19, covering the North Atlantic-Gulf of America-Caribbean Sea region. Data are available from 26 May through 15 July 2017 in netCDF-4 format.
Over 15 columns detail mailing and principal addresses, phone numbers, websites, and entity types for charitable organizations registered in Colorado. The Colorado Department of State's Charities Program collects and maintains this data, which was last updated in April 2026.
535 bacterial isolates were analyzed from two lagoon wastewater treatment plants in Settat and Ouled Said, Morocco, focusing on antibiotic-resistant bacteria (ARB) prevalence and physicochemical water parameters. The study evaluates treatment efficiency by measuring reductions in total bacterial counts, coliforms, Enterococcus, and Pseudomonas, while correlating resistance patterns with factors like dissolved oxygen and pH. It provides data on resistance rates to specific antibiotics, including ampicillin and ciprofloxacin.
Colchester Borough Council funded local projects to improve community health and wellbeing in the 2016/17 financial year. The funding was allocated by CCVS in grants of up to £1,000 per project. The dataset is published by the Government Digital Service under an open government license.
China's restored wetlands show slow recovery of stable carbon pools, constraining long-term soil carbon restoration. The dataset, authored by Min Luo, is a 50.4 KB XLSX file last updated in May 2026. It is shared under a CC-BY-4.0 license on figshare.
Organizations and entities from the cultural sector that have participated in calls from the Ministry of Cultures, Arts, and Knowledge. The dataset aims to identify and characterize institutional and organizational actors active in fostering, managing, and promoting culture in the country. Information includes territorial distribution, participation level in institutional offerings, and the role of cultural liaisons in municipalities and departments.
50 episodes of robot teleoperation data created using LeRobot. The dataset contains 85,602 frames across 150 videos, focusing on two distinct manipulation tasks. It was authored by YOLO2431 and last updated on Hugging Face in May 2026.
A dataset for object detection, likely containing images of nuts and bolts, created by rllab-postech. The dataset page was last updated on 2026-05-22 05:53:30. It is associated with a YOLO26 model for detecting these industrial components.
49,326 cases of building energy data organized for graph-based learning. The dataset includes 5,481 unique buildings and 64 unique weather IDs, with temporal sequences ranging from 968 to 8,760 steps. It was created by ArchEGraph and last updated on 2026-05-07.
42 samples from the Lower Ordovician Nambeet Formation (1354.80–2435.04 mRT) in the Barnicarndy 1 well, Canning Basin, Western Australia. Geoscience Australia's Exploring for the Future program produced this palynological reconnaissance study to assess microfossil yield, preservation, and utility for regional and international correlation. The record includes digital images of identified microfossils such as acritarchs, algae, cryptospores, and chitinozoans.