COMP20008 Assignment 2

Project Structure

Project/
├── main.py                     # Main execution script
├── pyproject.toml              # Project dependencies
├── uv.lock                     # Dependency lock file
├── README.md                   # Project documentation
├── .gitignore                  # Git ignore rules
│
├── src/                        # Source code modules
│   ├── __init__.py
│   ├── DataPreprocessor.py     # Data loading and preprocessing
│   ├── ModelTrainer.py         # Machine learning model training
│   └── Cluster.py              # Clustering analysis
│
├── notebook/                   # Jupyter notebooks
│   └── COMP20008_A2.ipynb      # notebook same with Colab
│
└── images/                     # Generated visualizations
    ├── waste_time_distribution.png
    ├── model_performance_comparison.png
    ├── confusion_matrices.png
    ├── clustering_features_correlation.png
    ├── cluster_visualization.png
    └── cluster_commuter_profiles.png

---

Installation

1. Install dependencies:

pip install -e .

Or

uv sync

Or install packages directly:

pip install matplotlib numpy pandas scikit-learn seaborn xgboost

---

Usage

Colab Notebook

We highly recommend using Google Colab to review our code and results, as it provides a convenient way to display and interact with the outputs.

Access the notebook here: COMP20008 Assignment 2 Colab

Or

Run the main analysis:

python main.py > output.log 2>&1

This will:

• Load and preprocess Victoria transport data
• Perform correlation analysis
• Train and evaluate classification models
• Generate clustering analysis
• Save all visualizations to ./images/

Note: There are a difference between notebook and script, notebook will not save result in the folder, but the script will.