A web application to predict the chances of getting stroke by a patient based on other health factors like hypertension, Smoking habit, etc.
- Live app :- https://stroke--prediction.herokuapp.com/home
- The model was trained and validated on 5100 patient health records.
- Achieved True Positive rate of 0.77 and F1 score of 0.24
- Note Book: Click here!
- Windows 10 OS
- Python 3.7
- Pandas
- Numpy
- Malplotlib
- Seaborn
- Sklearn
- Flask
- HTML
- CSS
- BootStrap
- Deployed on Heroku cloud
Step 1: Download zip file and extract
or
clone this repository
git clone https://github.com/JavSFDev/stroke-prediction-app.git
Step 2: Open your command promt and set app folder as working directory
cd stroke-prediction-app
Step 3: Install dependencies
pip install -r requirements.txt
Step 4: Run the application
python app.py
Step 5: Copy the url from the terminal and open it in your favourite browser.
ex: 127.0.0.1:5000/
------------------------------------------------------------------