Multiple Linear Regression Project

Project Overview

This project implements a simple Multiple Linear Regression model from scratch in Python using NumPy and Pandas. The model is trained on housing data (houses.csv) to predict a target variable (e.g., house prices) based on selected features (e.g., size, number of rooms, etc.).

The project demonstrates how to:

1. Load and preprocess data using Pandas.
2. Split data into training and testing sets.
3. Build and train a linear regression model using gradient descent.
4. Evaluate the model by predicting on test data.

---

Files in the Repository

• houses.csv: The dataset used for training and testing the model.
• linear_regression.py: The Python script containing the implementation of the linear regression model.
• README.md: Documentation for the project.

---

How It Works

1. Data Preparation

The dataset (houses.csv) is loaded using pandas.read_csv. The target variable y is extracted as the first column, while the features X are selected as the last two columns. The dataset is split into training (80%) and testing (20%) sets.

2. Model Implementation

The MultipleLinearRegression class includes:

• Initialization (__init__): Sets the learning rate and initializes weights and bias.
• Forward Pass (Z): Calculates predictions as the dot product of input features and weights, plus bias.
• Cost Function (cost_function): Computes the Mean Squared Error (MSE).
• Gradient Calculation (gradient): Derives gradients for weights and bias using partial derivatives of MSE.
• Training (fit): Updates weights and bias iteratively using gradient descent.
• Prediction (predict): Predicts the target values for new input data.

3. Training and Prediction

The model is trained on the training set using fit. The predictions on the test set are then rounded to simplify the output.

---

Requirements

To run the project, you need:

• Python 3.7+
• NumPy
• Pandas

Install the dependencies using:

pip install numpy pandas

Usage

Clone the repository:

git clone https://github.com/your-username/linear-regression.git
cd linear-regression

• Place your dataset in the same directory and name it houses.csv.

Run the script:

python linear_regression.py

View the model predictions printed to the console.

Key Functions and Methods

• fit(X_train, y_train, epochs): Trains the model using the provided training data and the specified number of epochs.
• predict(X_test): Returns predictions for the test set.

Notes

• Ensure the dataset houses.csv is correctly formatted with numerical columns, where the first column is the target variable.
• Adjust the learning rate lr and epochs to fine-tune model performance.

License

-This project is open-source under the MIT License. -Feel free to contribute or suggest improvements! 😊