Vehicle Suggestion System

This project aims to provide vehicle recommendations using unsupervised learning via clustering. It is built with Python, Flask, and scikit-learn.

Project Structure

• app.py: Main application file that includes the Flask API for vehicle recommendations.
• generate_vehicles.py: Script to generate a dataset of vehicles.

Installation

1. Clone the repository:

   git clone https://github.com/Yassim-B/deep-vehicle-suggest.git
   cd VehicleSuggestionSystem

2. Create a virtual environment and activate it:

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install the required packages:

   pip install -r requirements.txt

4. Generate the dataset:

   python generate_vehicles.py

Running the Application

1. Start the Flask application:

   python app.py

2. The API will be available at http://127.0.0.1:5000/.

API Endpoints

• GET /coucou: Test endpoint to check if the API is running.

  • Example:

    curl http://127.0.0.1:5000/coucou

• POST /recommander: Endpoint to get vehicle recommendations based on user input.

  • Request body example (JSON):

    {
        "sieges": 5,
        "consommation": 8.5,
        "prix": 30000,
        "design": 4,
        "kilometrage_maxi": 50000,
        "motorisation": "Essence"
    }

  • Response example:

    {
        "count": 50,
        "vehicles": [
            {
                "Marque": "Toyota",
                "Modele": "Corolla",
                "Sieges": 5,
                "Consommation": 8.0,
                "Prix": 25000,
                "Design": 4.5,
                "Motorisation": "Essence",
                "Kilometrage": 45000,
                "Distance": 0.12345
            },
            ...
        ]
    }

Explanation

app.py

The app.py file contains the main Flask application which handles the following:

1. Data Loading and Preprocessing:

   • Loads vehicle data from vehicles.csv.
   • Fills missing values with the mean of their respective columns.
   • Encodes the categorical variable Motorisation using LabelEncoder.
   • Standardizes numerical features.

2. KMeans Clustering:

   • Trains a KMeans model with 5 clusters using the preprocessed data.

3. Vehicle Recommendation:

   • Defines a function to find the nearest vehicles to the user's preferences using Euclidean distances.
   • Provides an endpoint /recommander to get recommendations.

generate_vehicles.py

The generate_vehicles.py script generates a synthetic dataset of 1000 vehicles with random attributes and saves it as vehicles.csv.

Contributing

Feel free to submit issues or pull requests if you have suggestions or improvements.