- Project completed during HackUMass 2018 - Completed in 24 hrs with 3 Developers
- A daily health tracker where you get notified to add entries daily. Weekly graphs and summary is posted, once you enter 6 times.
- We gather information from the user on what's his/her daily routine like
- A machine learning model trains a data and tests this on that data predicting the best daily practices for the User
- User is notified every time they open the app and every week to enter their daily activities
- There are graphs and charts which display the overall profile of the User according to the
- Uses recycler views,
- SQLite database - storing the Weekly data for the user - displaying in a History tab
- Graphview to display the User data from the last week - using Android graphs
- Libraries from AndroidMultiSelectDialogy from Abumoallim & GraphView.
- Icons from materialdesignicons.com
- Tensor Flow model created with Python in Jupyter Notebook
- Integrated with Machine Learning kit by Firebase
- Firebase Login, Database implemented
MIT License 2018