Loading...
Loading...
Student performance, MOOC logs, knowledge tracing, standardized tests, learning analytics
13,909 datasets
Vacant Property Indicators identify likely vacant lots and buildings across Philadelphia. The dataset is produced by the City of Philadelphia, Department of Licenses and Inspections, and was last updated in March 2025. It is derived from an assessment of multiple city administrative datasets.
A dataset comparing FEMA and CRSI risk assessment indices for natural hazards, with methodology for converting between the two. It is associated with published research analyzing hazard frequency, intensity, and spatial patterns in the United States. The dataset originates from the U.S. Environmental Protection Agency and FEMA's National Risk Index.
College Board data provides mean SAT scores for Critical Reading, Mathematics, and Writing from the 2010 graduating seniors in New York City schools. Records include the number of test takers per school and are suppressed for groups of five or fewer students. The dataset was collected and processed by the College Board.
FineMath is a dataset of mathematical educational content filtered from the CommonCrawl web corpus. It contains 34 billion tokens in one version and 54 billion tokens in another, curated by HuggingFaceTB using a classifier trained with Llama-3.1-70B-Instruct annotations. The dataset was last updated in February 2025.
The Black Sea City Territorial Community in Odessa region, Ukraine, provides data on children registered in its communal preschool institutions. The dataset was last updated on February 27, 2025, and originates from the States site of Ukraine. It likely contains administrative records of enrollment applications.
Turkey's university regulatory documents collected from the official mevzuat.gov.tr website. The dataset includes full texts with publication details and unique identifiers, scraped using Selenium and BeautifulSoup in Python. It was uploaded by yusufbaykaloglu and last updated on February 6, 2025.
Smol-SmolTalk contains conversational data specifically curated for training language models with fewer than 1 billion parameters. The dataset was created by HuggingFaceTB and was last updated in February 2025. It was used to develop the instruction-tuned models SmolLM2-360M-Instruct and SmolLM2-135M-Instruct.
District of Columbia's 2018 School Report Card provides STAR performance scores and ratings for 203 schools based on the 2017-18 school year. The dataset includes overall STAR scores on a 0-100 scale and star ratings from 1 to 5. Metrics are calculated from applicable framework scores and student groups, with scores suppressed for groups smaller than 10 students.
1,693 robotic manipulation episodes across four task suites (Spatial, Object, Goal, and Libero-10) using a Panda robot arm. Converted to LeRobot format by Physical Intelligence in 2025, this collection aggregates data from the original Libero project for imitation learning research.
Protein stability landscape prediction is a regression task mapping amino acid sequences to a label measuring extreme conditions for fold stability. The dataset originates from the Evaluating Protein Transfer Learning with TAPE project and uses its original splits. It was uploaded by SaProtHub and last updated on February 4, 2025.
Backnang, Germany's 'Taus' development plan is available via a WFS service from the XPlanung 5.0 standard. The dataset describes a reassessment in the 'Maple Trail' area and was published by the Bundesamt für Kartographie und Geodäsie. It was last updated on February 10, 2025.
Over 64 TiB of ASMR works archived for educational and research purposes. The repository is maintained by DeliberatorArchiver and was last updated in February 2025. Due to its size, direct cloning is not recommended, and selective downloads via Hugging Face tools are advised.
21.6% of students in grades 3-8 took the optional English assessment in 2021, reflecting the pandemic's impact on standardized testing. This dataset from the NYC Department of Education details school-level demographic enrollment and admissions methods for the 2020-21 academic year. It includes counts for student poverty status, disability classification, and English language learner status, sourced from the November 13 audited register and STARS course management software.
24 million image-caption pairs and 30 million image-references extracted from 6 million unique open-source biomedical articles. The dataset is annotated with over 27 unique metadata fields per data point, including article-level information. It was created by BIOMEDICA and last updated on Hugging Face in January 2025.
Data from 2025-01-29 provides information on children waiting for enrollment in preschool educational institutions within the Delyatynska AH (Amalgamated Hromada) in Ukraine. The dataset likely contains details on application numbers, dates, consideration statuses, child information, and preschool institutions. It is published on the States site of Ukraine.
50 episodes of robotic control data totaling 14,478 frames recorded at 15 FPS using the Piper robot platform. The dataset is organized into a single chunk of Parquet files under the LeRobot v2.0 codebase specification for a single task.
A directory of schools participating in New York City's Pre-K for All program for the 2018 school year. The data was collected by the Department of Early Childhood Education and includes offer results from the Fall 2017 admissions process. The dataset includes geospatial columns added to meet Local Law 108 of 2015 standards.
Psycopref is a dataset introduced in the paper 'Preference Learning Unlocks LLMs' Psycho-Counseling Skills'. It was created by Psychotherapy-LLM and uploaded to Hugging Face on March 1, 2025. The dataset is designed for preference learning tasks related to psycho-counseling.
State water quality assessment decisions reported to the U.S. EPA under Clean Water Act Sections 303(d) and 305(b). The dataset provides summary information for each Assessment Unit, as part of the ATTAINS system for tracking water quality and TMDLs. It is maintained by the U.S. Environmental Protection Agency and was last updated in February 2025.
Aime2025 contains competition problems from the 2025 American Invitational Mathematics Examination (AIME) I and II, published by opencompass in February 2025. The collection consists of fewer than 1,000 text-based records designed for high-level mathematical question-answering tasks. It provides a standardized format for evaluating model performance on the most recent invitational-level high school mathematics problems.