All Competitors

Every biological foundation model, evaluated and ranked by the bio.rodeo team

Showing 1–2 of 2 filtered models

  • SelfPAB

    1917—
    NTNUMarch 25, 2024accelerometryfoundation_modelhuman_activity_recognition+3

    A transformer encoder self-supervised on 100,000 hours of dual-accelerometer data, used as a frozen feature extractor for human activity recognition.

    Biosignals
    78Openness

  • SSL-Wearables (HARNet)

    157189—
    University of OxfordMarch 18, 2024accelerometrycnnhuman_activity_recognition+5

    Self-supervised CNN pretrained on 700,000 person-days of UK Biobank accelerometer data for human activity recognition, transferable across devices and cohorts.

    Biosignals
    28Openness