House Price Prediction

A short, easy-to-follow project that trains a regression model to predict house prices using the (classic) Boston housing dataset. The primary work is in the colab notebook:

• Project_4_House_Price_Prediction.ipynb Table of contents
• About
• Notebook(s)
• Requirements
• Quick start (Colab and local)
• Reproducible run (commands/snippets)
• Notes about the dataset
• How the model is trained
• Evaluation & results

About This project demonstrates end-to-end steps for a supervised regression task:

• Loading a housing dataset
• Exploratory data analysis and visualization
• Train/test split
• Train an XGBoost regression model
• Evaluate model performance

Notebook(s)

• Project_4_House_Price_Prediction.ipynb — the primary notebook with all the code and visualizations.

Requirements Minimal Python packages (example versions that are known to work):

• python >= 3.8
• numpy
• pandas
• matplotlib
• seaborn
• scikit-learn
• xgboost

You can install the essentials with pip:

pip install numpy pandas matplotlib seaborn scikit-learn xgboost jupyter

Quick start

Run in Google Colab (recommended if you don't want to configure locally)

1. Open the notebook in Colab:
   • The notebook includes a "Open in Colab" badge; or open: https://colab.research.google.com/github/meeks627/House_price_Prediction/blob/main/Project_4_House_Price_Prediction.ipynb
2. Run the notebook cells in order.

Run locally

1. Clone the repository:

git clone https://github.com/meeks627/House_price_Prediction.git
cd House_price_Prediction

2. Install dependencies (see Requirements).
3. Start Jupyter and open the notebook:

jupyter notebook Project_4_House_Price_Prediction.ipynb

4. Run the cells top-to-bottom.

Reproducible run / key snippets

• The notebook uses sklearn.datasets.load_boston() to load the dataset:

import sklearn.datasets
house_price_dataset = sklearn.datasets.load_boston()

Note: sklearn.datasets.load_boston is deprecated/removed in recent scikit-learn versions. If you encounter an error, either:

• Install a scikit-learn version that still includes load_boston (e.g., pip install scikit-learn==1.1.3), OR
• Use fetch_openml to retrieve the Boston dataset:

from sklearn.datasets import fetch_openml
boston = fetch_openml(name="boston", version=1, as_frame=True)
X = boston.data
y = boston.target

Model training (as implemented in the notebook)

• The notebook trains an XGBoost regressor:

from xgboost import XGBRegressor
model = XGBRegressor()
model.fit(X_train, Y_train)
preds = model.predict(X_test)

• Evaluation metrics commonly shown in the notebook: MAE, MSE, RMSE, R^2 (scikit-learn metrics module).

Notes about the dataset

• The project uses the Boston housing dataset (13 features) and price as target.
• The dataset historically contains a capped value of 50.0 for some entries — check the notebook for handling and interpretation.
• The Boston dataset has been deprecated in scikit-learn due to ethical concerns; for production work consider using alternative datasets (e.g., California housing) or a custom dataset.

Evaluation & results

• The notebook includes visualizations, correlation heatmap, train/test split, model training and evaluation.