🧠 Machine Learning Portfolio by Wei Tong

Welcome! This is my curated collection of hands-on machine learning projects built during my learning journey and practical experience.

Each project focuses on a different concept or real-world use case—ranging from traditional models like logistic regression to deep learning applications using CNNs, model compression, and optimization techniques.

---

🎯 Goals of This Portfolio

• Apply machine learning concepts in real scenarios
• Demonstrate technical skills in Python, data preprocessing, modeling, and evaluation
• Explore practical tools like Scikit-learn, TensorFlow, PyTorch, and deployment techniques
• Build a personal learning log to track progress and share with peers or potential employers

---

🗂️ Projects Overview

#	Project	Type	Description	Tags
1	CIFAR-10 Image Classifier	🧠 Deep Learning	Classify 32x32 color images using CNN with Keras	CNN, Image, Keras
2	Boston Housing Regression	📈 Supervised	Predict housing prices with linear regression and feature analysis	Regression, EDA
3	Model Distillation Practice	🧪 Optimization	Apply knowledge distillation to compress large models	Distillation, Model Compression
4	Titanic Survival Prediction	🛳️ Classification	Classic Kaggle binary classification starter project	Classification, EDA, Kaggle
5	Sentiment Analysis (IMDB)	💬 NLP	Build an RNN-based model for text sentiment classification	NLP, RNN, Text

---

🔧 Tech Stack & Tools

• Languages: Python 3.x
• Core Libraries: NumPy, Pandas, Scikit-learn, Matplotlib, Seaborn
• Deep Learning: TensorFlow, Keras, PyTorch
• Experiment Tools: TensorBoard, Weights & Biases
• Version Control: Git, GitHub
• Environment: Conda, Jupyter Notebook, VS Code

---

🗃️ Folder Structure

ml-portfolio/
├── projects/
│   ├── 01\_cifar10\_cnn/
│   ├── 02\_boston\_regression/
│   ├── 03\_knowledge\_distillation/
│   └── ...
├── docs/                  # Diagrams, notes, visualizations
├── utils/                 # Shared helper functions
├── requirements.txt
└── README.md              # This file

---

🧭 Learning Milestones

✅ Linear Models →
✅ Tree-based Models (RF, XGBoost) →
✅ Neural Networks (CNN, RNN) →
✅ Model Evaluation & Tuning →
🔜 Deployment (Flask/Gradio), MLOps Tools

---

📌 How to Use

You can clone this repository to explore individual projects:

git clone https://github.com/yourname/ml-portfolio.git
cd ml-portfolio/projects/01_cifar10_cnn
jupyter notebook

Each project includes:

• README.md with project description and results
• Notebook or Python scripts
• Plots and evaluation metrics

---

📬 Contact

• Name: Wei Tong
• Email: weitong.dev@example.com
• GitHub: github.com/yourname

---

🌟 Acknowledgements

Thanks to the open-source community and educational platforms like:

• Google MLCC
• Kaggle Learn
• Fast.ai

---

📝 License

This portfolio is released under the MIT License. Feel free to explore, fork, and reuse for learning purposes.