youzaina001/ML_supervised

Implementing and Comparing Regression Models Using Scikit-Learn & PyCaret

# Predictive Regression Model using Scikit-Learn and PyCaret

## Project Overview

This project develops a predictive regression model to estimate the "Selling_Price" based on different factors. We leverage the power of scikit-learn and PyCaret, two of the most popular machine learning libraries in Python, to preprocess data, select features, train models, and evaluate performance.

### Objectives

- To preprocess and analyze the dataset for regression analysis.
- To select the most significant features affecting the "Selling_Price".
- To compare multiple regression models and select the best performer based on different metrics (MSE, RMSE, etc).
- To fine-tune the model parameters for optimal performance.
- etc.

## Dataset

The dataset used for our regression models is public and can be found in the Kaggle datasets section (https://www.kaggle.com/datasets/nehalbirla/vehicle-dataset-from-cardekho).

### Features

- **Age**: [Age of the vehicule]
- **Present_Price**: [Present price for the vehicule]
- **Kms_Driven**: [Kilometers driven using the car]
- **Fuel_Type**: [Fuel type : Diesel or Petrol or CNG]
- **Seller_Type**: [Seller type : Dealer or individual]
- **Transmission**: [Transmission : Manual or Automatic]
- **Owner**: [Owner of the car]

### Target Variable

- **[Selling_Price]**: [Description]

## Installation

This project is executed using Python 3.11 and the following Python libraries:

- NumPy
- Pandas
- scikit-learn
- PyCaret
- matplotlib (for data visualization)
- seaborn (for data visualization)
- etc.

To run the jupyter notebook in a container, follow the steps below:

- build the docker image using the provided dockerfile by running the command "docker build -t regression ."
- run a container using the image you just created by running the command "docker run -p 8888:8888 regression"
- to access your virtual environment, use the link obtained via the command line.