- An app developed using
Flutter
to predict the risk of having an heart attack. - Uses a hyperparameter tuned
LogisticRegression
machine learning model with an accuarcy of 86% to predict the result. - Uses a total of
thirteen
paramaters (details can be found in foldermachine_learning_model
) as the basis of scoring. - A Python backend is used to load the trained machine learning model and provide the result in json format to the Flutter frontend
-
In order for the code to work:
- Please ensure to run the python backend to access the result of the machine learning model
-
To run on a real device please run the following commands on the terminal and replace the
url
in the code with the url derived- brew cask install ngrok
- ngrok http 5000