Diabetes Prediction using Support Vector Machine

Date created

3rd Sep 2022

 

Project Title

Diabetes Prediction using Support Vector Machine

 

Description

Support Vector Machine is a supervised classification algorithm where we draw a line between two different categories to differentiate between them.

For this project, we will be building a machine learning model to classify whether the person is a Diabetic or a Non-Diabetic based on certain diagnostic measurements included in the dataset provided by the National Institute of Diabetes and Digestive and Kidney Diseases. Our goal is to work through this notebook by collecting data, preprocessing it, splitting it into testing and training datasets, train the model and evaluate the accuracy of our model.

API

We have created an API using FastAPI for users to interact with. You can run the model using uvicorn on your local machine and test the API, refer to diabetes_api.py & api_testing.py.

Also the API is deployed on Heroku with the following URL : https://diabetesmlapi.herokuapp.com/diabetes_prediction

StreamLit Web App

We also created a web app using Streamlit. You can run the app on your local machine using StreamLit and test the web app, refer to diabetes_prediction_web_app.py.

The app is deployed on Heroku. Use the following link to explore: Diabetes Prediction Web App

 

Files used

We used the following dataset available on Kaggle to work on this project:

• Pima Indians Diabetes Database

The datasets consists of several medical predictor variables and one target variable, Outcome. Predictor variables includes the number of pregnancies the patient has had, their BMI, insulin level, age, and so on.

 

Credits

Thanks to Kaggle for teaching me ML