🧠 ML Learning Lab

A hands-on machine learning lab built from scratch in Python.
This project showcases simple and advanced regression models using real-world datasets, complete with training visuals and evaluation metrics.

---

🚀 Project Highlights

• ✅ Implemented Linear Regression from scratch
• 📊 Visualized training performance using RMSE and MSE over epochs
• 🔍 Compared red vs white wine quality prediction models
• ⏱️ Integrated tqdm progress bars with live terminal updates
• 💾 Interactive chart saving and user input via CLI
• 🎓 Included a student exam score predictor with a single-feature model

---

📁 Folder Structure

ml-learning-lab/ ├── data/ │ ├── student_scores.csv # Simple dataset (Hours vs Scores) │ ├── winequality-red.csv # Red wine data │ └── winequality-white.csv # White wine data │ ├── experiments/ │ ├── experiment_1.py # High learning rate test │ ├── experiment_2.py # Low learning rate test │ ├── wine_experiment_red.py # Red wine regression test │ ├── wine_experiment_white.py # White wine regression test │ └── compare_wines.py # Side-by-side RMSE/MSE analysis │ ├── model_utils.py # Core ML logic (train, predict, plot) ├── main.py # Run student score model ├── requirements.txt # Python dependencies └── README.md # You're here!

---

🧪 Datasets Used

• Student Scores – [simple CSV] hours studied → exam score
• Wine Quality (UCI) – Multifeature dataset:
  Predicts quality based on acidity, sugar, alcohol %, etc.

---

📈 Sample Output

Terminal: Red Wine Epoch 100/100 | MSE: 26.30 | RMSE: 5.13 White Wine Epoch 100/100 | MSE: 29.91 | RMSE: 5.45

Chart:

• RMSE and MSE plotted side-by-side
• Red vs White wine training comparison

---

🧠 Learnings

• Difference between MSE (Mean Squared Error) and RMSE (Root MSE)
• Role of hyperparameters (learning rate, epochs)
• Importance of feature scaling (via StandardScaler)
• How model loss behaves over training

---

▶️ Getting Started

git clone https://github.com/R56123/ml-learning-lab.git
cd ml-learning-lab
pip install -r requirements.txt
python main.py                         # Run student regression
python experiments/wine_experiment_red.py   # Run wine experiments
python experiments/compare_wines.py         # Compare RMSE & MSE