Skip to content

Latest commit

 

History

History
2 lines (2 loc) · 506 Bytes

README.md

File metadata and controls

2 lines (2 loc) · 506 Bytes

P3-FitRec

A multi-level deep learning approach that extracts insightful fitness recommendations from a large-scale real fitness dataset gathered from wearable IoT devices. Our model uses machine learning techniques to accurately estimate speed sequence, distance and heart rate sequence based on personalized fitness profiles. We use a convolution neural network (CNN) to separate out the individual components of the running profile, allowing us to more accurately predict and recommend training goals.