home-price-prediction

Demo

---

Link:https://home-price-prediction.herokuapp.com/

Overview

---

This is the Streamlit base Machine learning web app to predict the house price in Bangalore by taking different parameter as an input like Location, Square Feet, Bedrooms and Bathrooms and deploy a model on the Heroku platform.

The project contain three components:

1. EDA: EDA stands for Exploratory Data Analysis. It show DataFrame, Shape of the data, Datatype of each columns Etc.
2. Plot: This part show the essential plot of house dataset. Like Correlation of each variable, Show how data is distributed and many more.
3. Prediction: The model takes an Location, Square Feet, Bedrooms, Bathrooms as an input and predict the price.

Technical Aspect

---

1. Building and hosting a Streamlit web app on Heroku.
   

   • Used Numpy and Pandas to manipulate data.

   • Used Matplotlib and Seaborn to explore the data with different plot. I also used streamlit buil-in chart functionality to plotting the plot.

   • I used Multi Linear Regression to predict the price.

Installation

---

The Code is written in Python 3.7. If you don't have Python installed you can find it here. If you are using a lower version of Python you can upgrade using the pip package, ensuring you have the latest version of pip. To install the required packages and libraries, run this command in the project directory after cloning the repository:

• Additional packages that are required are: Numpy, Pandas, Matplotlib, Seaborn, scikit-learn.

• Download all packages from here.

• For streamlit you can just hit below command in cmd.

   pip install streamlit
  

Deployement on Heroku

To deploy streamlit web app, There are some file we need to add like setup.sh, requirements.txt in our repository.

1. Click on this Link
2. Create a new account by click on sign up button.
3. Login with Email and Password.
4. After login, screen redirect to the Dashboard. Click on upper top right corner to create a new app.
5. Give proper name of your app and choose united states america.
6. Heroku create a empty app setup for you. Next step is to go on Deploy menu.You can see that Heroku provide different-different method to deploy the web-app. We deploy by using Github repository. For that, click on github icon and they ask for a username and password of your github account. After that heroku connect with your github account.
7. In same page, go to the App connected to Github section and search repository which is hold all your essential file(Make sure that you have to create repository first then only heroku can able to search that repository).
8. Next step is to click on the connect button and Heroku able to connect with your repository.
9. In same page, go to the Manual Deploy section and click on Deploy Branch. Before clicking onto that button makesure that you need to add all neccessary file in your repository.
10. Heroku installed all packages. BOOM! After sometime heroku able to deploy your app and it will give you link for your app.

Technologies Used