Prodigy ML Task 01 – House Price Prediction

This repository contains my solution for Task 01 of the Prodigy InfoTech Machine Learning internship.

The goal is to build a linear regression model to predict house prices using the Kaggle House Prices: Advanced Regression Techniques dataset.[web:12]

Features and Target

• GrLivArea – Above-ground living area (square feet)
• BedroomAbvGr – Number of bedrooms above ground
• FullBath – Number of full bathrooms
• Target: SalePrice

Approach

1. Loaded train.csv and test.csv from the Kaggle competition.
2. Selected the three required numerical features and handled missing values using median imputation.
3. Trained a baseline LinearRegression model and evaluated it with RMSE and R².
4. Improved performance by adding PolynomialFeatures and tuning the polynomial degree with GridSearchCV, keeping the model in the linear regression family.[web:66][web:81]
5. Trained the best model on all training data and generated predictions for the Kaggle test set.

Files

• house_price_prediction_task1.ipynb – main Colab notebook (click the badge to open in Colab).
• README.md – overview of the project.

How to run

1. Open the notebook in Google Colab using the "Open in Colab" button.
2. Upload train.csv and test.csv from the Kaggle competition.
3. Run all cells to reproduce the results.

Outputs

• submission.csv – model predictions (Id, SalePrice) for all rows in test.csv.
• The notebook also displays a preview of the predicted prices using submission.head().