‹ BackHN Continuity

Thread

TabPFN and TabICL vs. tuned XGBoost: the model that doesn't train won 14/14

23 points · 12 comments · EfrainGaray

  1. not_a_feature · · focus · HN ↗
    If you want a better (and Biomedical) benchmark that evaluates a whole range of models: <a href="https:&#x2F;&#x2F;tabbench-bio.eu" rel="nofollow">https:&#x2F;&#x2F;tabbench-bio.eu

    or a general one: <a href="https:&#x2F;&#x2F;tabarena.ai" rel="nofollow">https:&#x2F;&#x2F;tabarena.ai

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.