Installation - House Price Prediction

This guide covers everything you need to set up the House Price Prediction project on your local machine, including dependency installation, Jupyter setup, and troubleshooting.

System Requirements

Python Version

Python 3.12 or higher is required. This project uses modern Python features and type hints that require at least Python 3.12.

Check your Python version:

python --version

Expected output: Python 3.12.x or higher

Operating System

The project works on:

Linux (Ubuntu, Debian, Fedora, etc.)
macOS (10.15 Catalina or later)
Windows (Windows 10/11 with WSL recommended)

Disk Space

Minimum: 500 MB (dependencies + dataset)
Recommended: 1 GB (includes cache and results)

Installation Methods

Method 1: UV (Recommended)

UV is a fast, modern Python package installer that’s significantly faster than pip.

Install UV

If you don’t have UV installed:

# Linux/macOS
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows (PowerShell)
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

Verify installation:

uv --version

Sync dependencies

Navigate to the project directory and sync all dependencies:

cd ~/workspace/source
uv sync

uv sync automatically:

Creates a virtual environment (.venv/)
Installs all dependencies from pyproject.toml
Locks dependency versions for reproducibility

Expected output:

Using Python 3.12.x
Creating virtualenv at .venv
Resolved 120 packages in 1.2s
Installed 120 packages in 4.3s

Activate the environment

# Linux/macOS
source .venv/bin/activate

# Windows
.venv\Scripts\activate

Your prompt should now show (.venv) prefix.

Method 2: pip

Traditional installation using pip and virtual environments.

Create a virtual environment

cd ~/workspace/source
python -m venv .venv

Activate the environment

source .venv/bin/activate

Install dependencies

pip install -r requirements.txt

This installs all packages listed in requirements.txt (120 packages total).Expected output:

Collecting pandas>=3.0.1
Collecting numpy>=2.4.2
Collecting scikit-learn>=1.8.0
...
Successfully installed pandas-3.0.1 numpy-2.4.2 scikit-learn-1.8.0 ...

Core Dependencies

The project requires these key packages:

Data Processing & Analysis

Package	Version	Purpose
pandas	≥3.0.1	Data manipulation and analysis
numpy	≥2.4.2	Numerical computing and arrays
pathlib	≥1.0.1	File path operations

Machine Learning

Package	Version	Purpose
scikit-learn	≥1.8.0	ML algorithms (regression, preprocessing)
joblib	1.5.3	Model serialization and saving

Visualization

Package	Version	Purpose
matplotlib	≥3.10.8	Static plots and visualizations
seaborn	≥0.13.2	Statistical data visualization
plotly	≥6.5.2	Interactive plots

Jupyter Environment

Package	Version	Purpose
jupyter	≥1.1.1	Jupyter notebook interface
jupyterlab	4.5.5	Modern Jupyter environment
ipykernel	≥7.2.0	Python kernel for notebooks

Data Acquisition

Package	Version	Purpose
kagglehub	≥1.0.0	Automatic Kaggle dataset download

The dataset (Boston Housing) is automatically downloaded from Kaggle using kagglehub on first run. No manual download or API keys required!

Jupyter Setup

Starting Jupyter Notebook

jupyter notebook

Your browser will open automatically to http://localhost:8888

Configuring the Kernel

If Jupyter doesn’t recognize your virtual environment:

Install ipykernel

pip install ipykernel

python -m ipykernel install --user --name=mlproject --display-name="ML House Prediction"

Select the kernel in Jupyter

In your notebook:

Click Kernel → Change Kernel
Select ML House Prediction

Jupyter Extensions (Optional)

Enhance your Jupyter experience:

# Table of contents
pip install jupyter_contrib_nbextensions
jupyter contrib nbextension install --user

# Variable inspector
jupyter nbextension enable varInspector/main

Dataset Download

Automatic Download

The project uses kagglehub to automatically download the Boston Housing dataset:

import kagglehub

# This runs automatically in both notebooks
path = kagglehub.dataset_download("arunjangir245/boston-housing-dataset")
print("Path to dataset files:", path)

First run output:

Downloading dataset...
Path to dataset files: ~/.cache/kagglehub/datasets/arunjangir245/boston-housing-dataset/versions/2

Subsequent runs:

Using cached dataset...
Path to dataset files: ~/.cache/kagglehub/datasets/arunjangir245/boston-housing-dataset/versions/2

The dataset is cached locally after first download, so you don’t need internet connection for subsequent runs.

Dataset Details

Source: Kaggle - Boston Housing Dataset
Size: ~25 KB (BostonHousing.csv)
Samples: 506 rows
Features: 13 input features + 1 target variable
Cache Location: ~/.cache/kagglehub/

Manual Download (Optional)

If automatic download fails, you can manually download:

Visit: https://www.kaggle.com/datasets/arunjangir245/boston-housing-dataset
Download BostonHousing.csv
Update notebook to point to your local file:

# Replace in notebooks/analyze.ipynb and notebooks/train.ipynb
df = pd.read_csv('path/to/your/BostonHousing.csv')

Verifying Installation

Run this verification script to ensure everything is installed correctly:

import sys
import importlib.metadata

# Check Python version
print(f"Python version: {sys.version}")
assert sys.version_info >= (3, 12), "Python 3.12+ required"

# Check key packages
packages = {
    'pandas': '3.0.1',
    'numpy': '2.4.2',
    'scikit-learn': '1.8.0',
    'jupyter': '1.1.1',
    'kagglehub': '1.0.0',
    'matplotlib': '3.10.8',
    'seaborn': '0.13.2'
}

print("\n✓ Python version check passed")
print("\nPackage versions:")

for package, min_version in packages.items():
    try:
        version = importlib.metadata.version(package)
        print(f"  {package}: {version}")
    except importlib.metadata.PackageNotFoundError:
        print(f"  ❌ {package}: NOT INSTALLED")

print("\n✅ All core dependencies installed!")

Expected output:

Python version: 3.12.x

✓ Python version check passed

Package versions:
  pandas: 3.0.1
  numpy: 2.4.2
  scikit-learn: 1.8.0
  jupyter: 1.1.1
  kagglehub: 1.0.0
  matplotlib: 3.10.8
  seaborn: 0.13.2

✅ All core dependencies installed!

Troubleshooting

Python Version Issues

Python 3.12 not available

Problem: Your system has an older Python version.Solutions:

Using pyenv (Linux/macOS):

curl https://pyenv.run | bash
pyenv install 3.12.0
pyenv global 3.12.0

Using apt (Ubuntu/Debian):

sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt update
sudo apt install python3.12 python3.12-venv

Download from python.org: Visit https://www.python.org/downloads/ and install Python 3.12+

Dependency Installation Issues

pip install fails with build errors

Problem: C++ compilation errors during installation (especially for numpy/scipy)Solution 1 - Install build tools:

# Ubuntu/Debian
sudo apt install build-essential python3-dev

# macOS
xcode-select --install

# Fedora/RHEL
sudo dnf install gcc gcc-c++ python3-devel

Solution 2 - Use pre-built wheels:

pip install --only-binary :all: numpy scikit-learn

SSL certificate verification failed

Problem: kagglehub fails to download dataset due to SSL errorsSolutions:

# Upgrade certifi package
pip install --upgrade certifi

# Set environment variable (temporary workaround)
export SSL_CERT_FILE=$(python -c "import certifi; print(certifi.where())")

Out of memory during installation

Problem: Installation crashes with memory errorsSolution:

# Install packages one at a time
pip install pandas
pip install numpy
pip install scikit-learn
pip install jupyter
pip install kagglehub matplotlib seaborn plotly

Jupyter Issues

Jupyter command not found

Problem: jupyter command not available after installationSolution:

# Ensure virtual environment is activated
source .venv/bin/activate

# Verify jupyter is installed
pip list | grep jupyter

# If not installed
pip install jupyter

Kernel keeps dying

Problem: Jupyter kernel crashes during executionCauses & Solutions:

Out of memory: Close other applications, or reduce dataset size for testing

Corrupted environment: Recreate virtual environment

deactivate
rm -rf .venv
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Conflicting packages: Clear pip cache

pip cache purge
pip install --force-reinstall jupyter ipykernel

Can't import modules in notebook

Problem: ModuleNotFoundError even after installationSolution:

Check that you’re using the correct kernel (see Configuring the Kernel)
Restart the Jupyter kernel: Kernel → Restart

Verify installation in notebook:

import sys
print(sys.executable)  # Should point to .venv/bin/python

Dataset Issues

Dataset download is very slow

Problem: kagglehub download takes a long timeSolutions:

Be patient - first download can take 1-2 minutes depending on connection
Check your internet speed: speedtest-cli
Use manual download method (see Manual Download)

Permission denied when caching dataset

Problem: Can’t write to ~/.cache/kagglehub/Solution:

# Fix permissions
mkdir -p ~/.cache/kagglehub
chmod 755 ~/.cache/kagglehub

# Or set custom cache directory
export KAGGLEHUB_CACHE_DIR=/path/to/writable/directory

Platform-Specific Notes

Windows

Windows users should use WSL (Windows Subsystem for Linux) for best compatibility. Some packages may have issues with native Windows Python.

Using WSL:

# Install WSL
wsl --install

# Install Python in WSL
sudo apt update
sudo apt install python3.12 python3.12-venv python3-pip

# Follow Linux installation instructions

Native Windows:

Use Anaconda for easier dependency management
Some visualization libraries may render differently
File paths use backslashes: results\model.joblib

macOS

M1/M2 Apple Silicon: Some packages need Rosetta or native ARM builds:

# Install Rosetta (if needed)
softwareupdate --install-rosetta

# Use conda for ARM-native packages (alternative)
brew install miniforge
conda create -n mlproject python=3.12
conda activate mlproject
conda install pandas numpy scikit-learn jupyter

Linux

Ubuntu 20.04 / Debian 10 users: Python 3.12 not in default repos - use deadsnakes PPA:

sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt update
sudo apt install python3.12 python3.12-venv python3.12-dev

Next Steps

Quick Start

Run your first analysis and train models

Project Structure

Understand the codebase organization

Data Exploration

Learn about the Boston Housing dataset

Model Training

Deep dive into model training workflow

Getting Help

If you encounter issues not covered here:

Check the GitHub Issues
Search existing discussions
Create a new issue with:
- Your Python version (python --version)
- Operating system
- Full error message
- Steps to reproduce

Still stuck? Open an issue with the label installation and we’ll help you get up and running!

Get Started

Core Concepts

Workflows

Model Guide

​System Requirements

​Python Version

​Operating System

​Disk Space

​Installation Methods

​Method 1: UV (Recommended)

​Method 2: pip

​Core Dependencies

​Data Processing & Analysis

​Machine Learning

​Visualization

​Jupyter Environment

​Data Acquisition

​Jupyter Setup

​Starting Jupyter Notebook

​Configuring the Kernel

​Jupyter Extensions (Optional)

​Dataset Download

​Automatic Download

​Dataset Details

​Manual Download (Optional)

​Verifying Installation

​Troubleshooting

​Python Version Issues

​Dependency Installation Issues

​Jupyter Issues

​Dataset Issues

​Platform-Specific Notes

​Windows

​macOS

​Linux

​Next Steps

Quick Start

Project Structure

Data Exploration

Model Training

​Getting Help

Build docs developers (and LLMs) love

System Requirements

Python Version

Operating System

Disk Space

Installation Methods

Method 1: UV (Recommended)

Method 2: pip

Core Dependencies

Data Processing & Analysis

Machine Learning

Visualization

Jupyter Environment

Data Acquisition

Jupyter Setup

Starting Jupyter Notebook

Configuring the Kernel

Jupyter Extensions (Optional)

Dataset Download

Automatic Download

Dataset Details

Manual Download (Optional)

Verifying Installation

Troubleshooting

Python Version Issues

Dependency Installation Issues

Jupyter Issues

Dataset Issues

Platform-Specific Notes

Windows

macOS

Linux

Next Steps

Getting Help