Machine Learning

Scikit-learn pipelines, model selection, ensembles, imbalanced datasets, and interpretability.

Use this subtrack for applied classical ML portfolio work. It is the fastest path in this phase if you want interview-ready projects and strong end-to-end workflow practice without starting from deep learning.

How To Use This Subtrack Well

• Build one full supervised learning project from raw data to stakeholder-facing results.
• Focus on validation, leakage prevention, and interpretability rather than model-count collection.
• Pair this subtrack with ../../16-model-evaluation/README.md and ../../09-mlops/README.md for stronger production habits.

What Comes Next

• Continue to ../README.md for the broader practical data science roadmap.
• Continue to ../../09-mlops/README.md if you want to productionize one of your projects.
• Continue to ../../29-ai-hardware-llm-validation/README.md only if your interest shifts toward systems rather than modeling.