datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
cleaned_nhanes_1988_2018Description: The National Health and Nutrition Examination Survey (NHANES) provides data and have considerable potential to study the health and environmental exposure of the non-institutionalized US population. However, as NHANES data are plagued with multiple inconsistencies, processing these data is required before deriving new insights through large-scale analyses. Thus, we developed a set of curated and unified datasets by merging 614 separate files and harmonizing unrestricted data… See the full description on the dataset page: https://huggingface.co/datasets/nguyenvy/cleaned_nhanes_1988_2018.nhanes-merged
NHANES merged participant-level dataset
This is the output of a pipeline that merges the 341 public NHANES tables into one
row per participant per survey cycle. It accompanies the code and documentation in
OpenFIM-Bench, under
pipelines/nhanes-merge/, where the methodology and every judgment call are written up.
The files here are the ones too large for a git repository. Everything smaller (the
column dictionary, the participant spine, the training tiers, the synthetic outputs… See the full description on the dataset page: https://huggingface.co/datasets/gracemyu/nhanes-merged.nhanes-dataset-feature-engineerednhanes-diabetes-selected-attributes-386Selected 386 attributes. Diabetes type is not included.
nhanes-dataset-diabetes-classified-v3nhanes-2021-2023-diabetes
NHANES August 2021–August 2023 — diabetes and demographics
Unofficial mirror of two real CDC/NCHS public-use files. This is not the complete
NHANES cycle, a monogenic-diabetes dataset or a custodian deployment.
The train labels are hosting conventions, not study-design splits.
Attribution and permitted use
Centers for Disease Control and Prevention (CDC). National Center for Health
Statistics (NCHS). National Health and Nutrition Examination Survey Data,
August… See the full description on the dataset page: https://huggingface.co/datasets/edithatogo/nhanes-2021-2023-diabetes.nhanes-accel-cardiometabolic-benchmarknhanes-dataset-prediction-selection-stage-1nhanes-training-mergednhanes-validation-data
NHANES Data for Validation
Place NHANES XPT files here to run tests/validate_on_nhanes.py.
Expected structure
nhanes_data_dump/
├── 2017-2020/ # Primary cycle (P_BIOPRO.XPT, P_MCQ.XPT, P_CBC.XPT, etc.)
└── extended_data/
├── 2003-04/
├── 2005-06/
├── 2007-08/
├── 2009-10/
├── 2011-12/
├── 2013-14/
└── 2015-16/
Download
Get data from NHANES. The validation script uses the same file merges as the training notebook.
nhanes-data-convertedThe final 74 attribute dataset (73 features + 1 Label) for all prediction, severity, and treatment T2D research.
nhanes-diabetes-classified-training-datasetnhanes-dataset-selected-raw-attributes-v3NHANES_Structured_Health_Profiles_2021-2023dqs-nhanes-footnotes
DQS NHANES Footnotes
Description
List of footnotes, notes, and source information for NHANES Summary Statistics. Each row of this dataset contains the accompanying text for a footnote found in the NHANES Datasets.
Dataset Details
Publisher: Centers for Disease Control and Prevention
Temporal Coverage: 1999-2018
Geographic Coverage: United States
Last Modified: 2025-04-21
Contact: National Center for Health Statistics (CDCINFO@CDC.GOV)
Source… See the full description on the dataset page: https://huggingface.co/datasets/HHS-Official/dqs-nhanes-footnotes.nhanes-dataset-diabetes-classifiednhanes-dataset-selected-raw-attributesdqs-nhanes-select-chronic-conditions-prevalence-es
DQS NHANES Select Chronic Conditions Prevalence Estimates
Description
These data represent prevalence estimates of select chronic conditions from the National Health and Nutrition Examination Survey (NHANES). This version of the dataset is specific for use by the NCHS DQS.
Search, visualize, and download these and other estimates from over 120 health topics with the NCHS Data Query System (DQS), available from: https://www.cdc.gov/nchs/dataquery/index.htm.… See the full description on the dataset page: https://huggingface.co/datasets/HHS-Official/dqs-nhanes-select-chronic-conditions-prevalence-es.nhanes-training-merged-newcalm-nhanes-2017-2018-results
CALM
Calibrated Menopausal-status benchmark — an open, reproducible benchmark for predicting
menopausal status from routine, low-cost health markers.
0.1.4 · Python 3.14 · uv · MIT
CALM provides a rigorously specified benchmark for:
premenopausal vs. postmenopausal from routine markers (labs, body measures, demographics), with
calibrated confidence, an explicit no-call option, honest generalization reporting, and
end-to-end provenance. It is a benchmark engine — a library and… See the full description on the dataset page: https://huggingface.co/datasets/meltus-bwire/calm-nhanes-2017-2018-results.nhanes_celiac_data