A hands-on laboratory for building, evaluating, and comparing machine learning models across diverse datasets — from fast baselines to fully tuned pipelines.
This repository documents my learning-by-building journey in applied machine learning, with an emphasis on clean experimentation, reproducibility, and fair model comparison.
- Classical ML models: Logistic Regression, KNN, SVM, Naive Bayes, Decision Trees, Random Forests
- Data preprocessing & feature engineering
- Pipeline-based training (scikit-learn style)
- Hyperparameter tuning
- Model evaluation & benchmarking
- Visualizations and experiment tracking
- Notes on failures, fixes, and insights