Airbnb Rental Price Prediction API

This is a Flask-based API that predicts Airbnb rental prices based on several factors like bedrooms, bathrooms, accommodation capacity, and neighborhood. The API has two main endpoints:

• /reload: Reloads the data and trains the model.
• /predict: Predicts the rental price for a given listing.

Data Source and Prediction Process

Data Source

The data used for this project comes from the Inside Airbnb dataset, which provides detailed information about Airbnb listings in various cities. For this particular app, the data for Boston, MA is used.

The dataset includes important features such as:

• Price: The rental price of the listing.
• Bedrooms: The number of bedrooms in the listing.
• Bathrooms: The number of bathrooms in the listing.
• Accommodates: The maximum number of guests the listing can accommodate.
• Neighbourhood: The neighborhood where the listing is located.

The full dataset can be accessed and downloaded from the Inside Airbnb website at Inside Airbnb - Get the Data.

Prediction Process

The application makes use of a simple Linear Regression Model to predict the rental price of an Airbnb listing based on various input features such as the number of bedrooms, bathrooms, accommodation capacity, and the neighborhood.

The process of prediction is as follows:

1. Data Preprocessing: The data is cleaned and processed. Non-numeric values are removed or converted, and categorical variables like neighbourhood are one-hot encoded to make them suitable for machine learning models.
2. Model Training: A linear regression model is trained on the cleaned dataset using features like bedrooms, bathrooms, accommodates, and one-hot encoded neighborhood values.
3. Prediction: Once trained, the model can predict the rental price based on user input, such as the number of bedrooms, bathrooms, and neighborhood.

By using this model, the app can provide quick rental price predictions for Airbnb listings in Boston based on historical data.

Prerequisites

Before you can set up and run this app, ensure you have the following software installed:

• Python 3.9+
• pip (Python package installer)
• Virtualenv (Optional but recommended)

Setting up on macOS and Windows

1. Clone the Repository

First, clone this repository to your local machine:

git clone https://github.com/tjhoranumass/airbnb.git
cd airbnb

2. Create a Virtual Environment (Optional but Recommended)

You can create a virtual environment to isolate the project dependencies.

For macOS:

python3 -m venv venv
source venv/bin/activate

For Windows:

python -m venv venv
venv\Scripts\activate

3. Install the Dependencies

Install the required Python dependencies using pip:

pip install -r requirements.txt

4. Set Up Environment Variables

Flask requires some environment variables to run the app correctly. Create a .env file in the project root with the following content:

FLASK_APP=app.py
FLASK_ENV=development

For macOS, you can set the environment variables using the following commands:

export FLASK_APP=app.py
export FLASK_ENV=development

For Windows, you can set the environment variables using the following commands in PowerShell:

$env:FLASK_APP = "app.py"
$env:FLASK_ENV = "development"

5. Initialize the SQLite Database

To set up the SQLite database for the first time, run:

flask shell

Inside the shell, run:

from app import db
db.create_all()
exit()

6. Running the Application

Once everything is set up, you can run the application with the following command:

flask run

By default, the app will run on http://127.0.0.1:5000.

7. Swagger Documentation

You can access the Swagger documentation for the API at:

http://127.0.0.1:5000/apidocs/

8. Testing the Endpoints

Reload Data

To reload the data and train the model, use the /reload endpoint:

curl -X POST http://127.0.0.1:5000/reload

Predict Price

To predict a rental price, you can use the /predict endpoint. Here's an example request:

curl -X POST http://127.0.0.1:5000/predict \
  -H 'Content-Type: application/json' \
  -d '{
    "bedrooms": 2,
    "bathrooms": 1.5,
    "accommodates": 4,
    "neighbourhood_cleansed": "South Boston"
}'

9. Stopping the Application

To stop the Flask app, you can press Ctrl + C in the terminal window where the app is running.

---

Troubleshooting

Common Issues

• Environment variables not being set: Ensure you have set the environment variables correctly, especially when switching between macOS and Windows.

• Database initialization issues: If the app crashes because of database-related errors, make sure you have run the flask shell commands to initialize the database properly.

• Dependency issues: Ensure that you are using the correct version of Python (3.9+) and have installed the dependencies using pip install -r requirements.txt.

---

License

This project is licensed under the MIT License.

Running Tests

We use pytest for running tests on this application. Before running the tests, ensure all dependencies are installed and the application is properly set up.

Setting up for Testing

1. Install the required dependencies by running:

pip install -r requirements.txt

2. Export the PYTHONPATH environment variable to ensure Python can locate the app module.

For macOS/Linux:

export PYTHONPATH=.

For Windows (PowerShell):

$env:PYTHONPATH="."
pytest

3. Run the tests:

pytest

This will execute all the tests located in the tests/ folder and provide feedback on the application behavior.

Deploying to Heroku

1. Install the Heroku CLI

Before deploying the application to Heroku, you need to install the Heroku CLI. You can follow the steps below to install it on your machine.

macOS:

You can install the Heroku CLI using Homebrew:

brew tap heroku/brew && brew install heroku

Windows:

Download and run the Heroku installer from the Heroku Dev Center.

Verify Installation:

Once installed, verify the installation by running:

heroku --version

You should see the version of Heroku CLI installed.

2. Log in to Heroku

Log in to your Heroku account from the terminal:

heroku login

This will open a web browser for you to log in to your Heroku account.

3. Prepare the App for Deployment

Ensure your requirements.txt and Procfile are present in the project root.

• Procfile: Create a Procfile in the root directory with the following content to tell Heroku how to run the app:

web: flask run --host=0.0.0.0 --port=$PORT

4. Create a Heroku App

Run the following command to create a new Heroku app:

heroku create

5. Deploy to Heroku

After you've created your Heroku app, deploy your app using Git:

git add .
git commit -m "Initial commit"
git push heroku main

6. Scale the Application

Heroku apps require at least one running dyno. Scale your app to run one web dyno:

heroku ps:scale web=1

7. Open the App

Once your app is deployed, you can open it in the browser using:

heroku open

Your app should now be live on Heroku!