cs549-finance-model

Finance model project for CS 549. Loads personal transaction data, trains and evaluates models, and produces the figures and metrics in the report.

---

Environment

• OS: Windows 10/11
• Python: 3.10+
• Key packages: numpy 2.2.6, pandas 2.3.3, scikit-learn 1.7.2, scipy 1.15.3, matplotlib 3.10.8, seaborn 0.13.2, mlxtend 0.23.4, quadprog 0.1.13, torch 2.10.0, JupyterLab 4.5.2

Full dependency list with exact versions is in requirements.txt.

---

Setup

Create and activate a virtual environment, then install dependencies:

python -m venv .venv

Windows:

.\.venv\Scripts\activate

macOS/Linux:

source .venv/bin/activate

pip install -r requirements.txt

---

Running the Notebook

Place aug_personal_transactions_with_UserId.csv in the same directory as the notebook, then launch Jupyter:

jupyter notebook

Open analysis.ipynb and run all cells top to bottom. The notebook covers data loading, preprocessing, model training, evaluation, and generates all figures referenced in the report.