Skip to content

Personalised fitness recommendation by multi-level deep learning approach.

Notifications You must be signed in to change notification settings

bhajagovindu/P3-FitRec

Repository files navigation

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.

About

Personalised fitness recommendation by multi-level deep learning approach.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published