Applied Machine Learning (SEG4180)

Weekly assignments completed for Applied Machine Learning (SEG4180 / CEG4195) at the University of Ottawa.

Each week includes a Jupyter notebook implementation and a corresponding formal PDF report.

---

Content Overview

Notebooks

Jupyter notebooks include:

• Data preprocessing and feature engineering
• Model development and training
• Hyperparameter tuning
• Evaluation metrics and visualizations
• Comparative experiments

Technologies used across assignments:

• Python
• NumPy / pandas
• scikit-learn
• HuggingFace
• Keras / TensorFlow
• MLflow / Docker (where applicable)

Reports

PDF reports document:

• Problem formulation
• Methodology and model selection
• Experimental setup
• Results and evaluation
• Technical and engineering considerations

---

Topics Covered

• Supervised Learning (Regression & Classification)
• Unsupervised Learning (Clustering & Dimensionality Reduction)
• Neural Networks & CNNs
• Large Language Models (LLMs)
• Prompt Engineering & Retrieval-Augmented Generation (RAG)
• Model Evaluation & Interpretability (SHAP, LIME)
• ML Pipelines & MLOps
• Deployment and Engineering Constraints