swaroop-p / ckd-fullstack-application

Fullstack web application to predict chrnoic kidney disease given certain body parameters

Chronic Kidney Disease Prediction

Machine learning + Angular2 + Python Tornado + Docker

Watch it live at - http://ckd-website.s3-website.ap-south-1.amazonaws.com/

The front end is developed on Angular Framework, deployed on AWS The Back end is a Tornado web application, deployed on AWS elastic beanstalk

Running & Building the Apps via Docker

• Make sure to install docker on your machine
• For windows 7 users,
  • download docker toolbox - https://docs.docker.com/toolbox/toolbox_install_windows/
  • Open docker quickstart terminal
    • cd /ckd-app
    • docker-compose build frontend-s & docker-compose build backend-s
    • OR simply run - docker-compose up (this will build and run both the frontend and backend applications).
  • Launch Powershell or cmd
    • docker-machine ls
    • docker-machine env <docker-name> is mostly default
    • Copy and execute the last command displayed after running the above command.
    • Docker is now hooked. You can now run all the docker commands.
• For windows 10 users,
  • Download and install docker
  • Start powershell or cmd and change directory to the ckd-app folder.
  • Run - docker-compose up
• you can run - docker ps -a to view the running containers
• Open browser and go to - localhost:4200 to view the website
• For Win. 7 users, the url to give in the browser will be IP mentioned at the start of docker qucikstart terminal - for eg: "docker is configured to use the default machine with IP 192.168.99.100". In this case browse the website at - 192.168.99.100:4200

Running the Apps on their own

The web application consists of -

1. Web application (web site) - Angular2 application - ckd-angular-app
2. Tornado web server application - Tornado python server - health_predictor_tornado_app

Steps to set up and run Tornado Web application

• Install anaconda 3 for python
• Switch to the directory with the python application. Install the python packages from requirements.txt using the terminal.

  conda install --file requirements.txt
  

• Run main.py file from health_predictor_tornado_app directory
  • To run main.py, either open the project in pycharm, set anaconda python3 as the interpretor, and run main.py
  • Else, type anaconda in start menu -
    • click on anaconda prompt menu
    • change directory to health_predictor_tornado_app , cd path/health_predictor_tornado_app
    • Run - python main.py
    • This will start the tornado server at - localhost:8080 ​

Steps to set up and run the Web App (website)

• The web application is built on Angular2. angular-cli was used to create the web app

• Steps to set up angular2 & angular-cli

  • First, download and install npm (node package manager)

    • npm is installed with Node.js

    • To install Node.js, go to https://nodejs.org/en/download/

    • Download the windows installer - node-v8.9.4-x86.msi (choose which ever is the latest version)

    • Check that you have node and npm installed

      To check if you have Node.js installed, run this command in your terminal/command prompt:

      node -v
      

      To confirm that you have npm installed you can run this command in your terminal:

      npm -v
      

  • Install angular-cli

    • Run the following command in the terminal/command prompt:

      npm install -g @angular/cli

      The Angular CLI makes it easy to create an application that already works, right out of the box.

      Eg: (optional steps - not necessary)

      To create a new app. - ng new your-app-name

      To run the app. -

      • change directory cd your-app-name
      • run the app ng serve
      • open browser and hit - localhost:4200, the app will run

• Now, that angular-cli is set up, open command prompt, navigate to ckd-angular-app -

  cd ckd-angular-app

• Run the following command -

  npm install - to install the necessary node modules (need to be run only once)

  ng serve - this will start the angular app.

  Open browser and go to -> localhost:4200

Steps to deploy Angular app to AWS (Static hosting)

Steps to deploy Tornado app to AWS Elastic Beanstalk