heart_discease_prediction_app

• Developed a comprehensive full-stack machine learning web application designed to predict occurrences of heart disease.
• Implemented an end-to-end heart disease classifier that encompasses three distinct machine learning models: KNeighborsClassifier, LogisticRegression, and RandomForestClassifier. The * most effective model was subsequently chosen for integration into the application's backend.
• Employed Mlflow to monitor and document the progression of machine learning experiments, enhancing the analysis of model performance and facilitating comparisons between different approaches.
• Ensured meticulous dataset versioning and management through the utilization of DVC (Data Version Control), contributing to the reproducibility and consistency of data preprocessing and model training procedures.
• To mitigate the common "works on my machine" dilemma that often arises due to variations in development environments, both the frontend and backend components were encapsulated within Docker containers.
  • The frontend was Dockerized, securing its dependencies and configurations within a containerized environment.
  • The backend was similarly Dockerized, providing an isolated environment tailored for the deployment of the heart disease classifier.
• Tech: Scikit-learn, Pandas, Numpy, Mlflow, Flask, Docker, React.js

Deployed Here

Live Link

Dagshub Repo link: For experiment tracking

Project Link

Download the repository and install the required packages:

Clone this repo

For frontend

1. npm install
2. npm start

For Backend

1. cd Backend
2. python -m venv my-env
3. my-env\Scripts\activate
4. pip install -r requirements.txt
5. python Server.py

For Docker

• Frontend

  • Development

    docker build -f Dockerfile.dev -t react .

    docker run --name react_c -p 3000:3000 react

  • Production

    docker build -t react_pro .

    docker run --name react_proc -p 3000:80 react_pro

• Backend

1. docker build -t my-python-app .
2. docker run -p 8000:8000 -e PORT=8000 my-python-app

For GCP Production:

gcloud auth login

gcloud config set project PROJECT_ID

gcloud builds submit --tag gcr.io/<ProjectName>/<AppName> --project=<ProjectName>

gcloud run deploy --image gcr.io/<ProjectName>/<AppName> --platform managed --project=<ProjectName> --allow-unauthenticated

gcloud iam service-accounts list --project=<ProjectName>

gcloud iam service-accounts keys create ./keys.json --iam-account <Service-account>

gcloud auth activate-service-account --key-file=keys.json

Demo

heart_detect_small.mp4