State of the art on ChatPPG

#1 on RealWorld Activity recognition

0.661Macro F1 (higher is better)

Method: ActiNet
Reported in: ActiNet: an open-source tool for activity intensity classification of wrist-worn accelerometry using self-supervised deep learning · Table 2
Compared: 2 results under the same protocol; runner-up Random forest + HMM (accelerometer package) 0.531
Protocol: 4-class activity intensity · whole dataset as external test · cross-dataset (trained on CAPTURE-24) · fine-tuned + HMM smoothing · 30 s windows

Share this resultEmail LinkedIn Post on X

Embed the SOTA badge

SOTA on ChatPPG: Activity recognition · RealWorld

Markdown

[![SOTA on ChatPPG: Activity recognition · RealWorld](https://chatppg.com/research/badges/actinet--har-realworld-har-macro-f1-har.svg)](https://chatppg.com/research/models/actinet#key-results)

HTML

<a href="https://chatppg.com/research/models/actinet#key-results"><img src="https://chatppg.com/research/badges/actinet--har-realworld-har-macro-f1-har.svg" alt="SOTA on ChatPPG: Activity recognition · RealWorld" height="20"></a>

Results as reported by the authors; not independently reproduced. Awarded 2026-10-20; re-checked on every atlas sync.