mariusbock/dl-for-har

Improving Deep Learning for HAR with shallow LSTMs (best paper award), Bock et al., presented at ISWC 2021 (online)

45
/ 100
Emerging

This project helps researchers and engineers who develop systems that automatically detect human activities using sensor data from wearables. It provides a refined deep learning architecture that processes raw sensor data and outputs identified human activities, offering improved accuracy and faster training times compared to previous methods. The primary users are researchers or practitioners in human activity recognition.

No commits in the last 6 months.

Use this if you are building or evaluating human activity recognition models based on sensor data and want to achieve higher performance with reduced training time and computational resources.

Not ideal if you are working with non-sequential data, require a black-box solution without understanding the underlying architecture, or are not focused on sensor-based human activity recognition.

human-activity-recognition wearable-technology sensor-data-analysis behavior-monitoring machine-learning-research
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

62

Forks

33

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 21, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mariusbock/dl-for-har"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.