Learn what machine learning actually is, how the end-to-end workflow fits together, and how to frame a real problem as a machine-learning task. You will leave able to decide whether ML is the right tool, name the task type, identify features and labels, and set up a working Python environment with scikit-learn and Jupyter.
Watch the free preview
What machine learning is (and is not) — free to watch, no account needed.
What you'll learn
- Frame a real problem as a machine-learning task, naming the target, the inputs and the task type
- Distinguish supervised, unsupervised and reinforcement learning and pick the right one
- Walk the end-to-end ML workflow from problem framing to deployment
- Split data correctly into training, validation and test sets and explain why
- Set up a reproducible Python ML environment with scikit-learn and Jupyter
Syllabus
What Machine Learning Really Is
What machine learning is (and is not)Free preview
Supervised, unsupervised and reinforcement learning
The End-to-End ML Workflow
The ML workflow from problem to deployment
Features and labels — the shape of ML data
Splitting Data and Fitting Models Well
Training, validation and test splits
Overfitting, underfitting and the bias-variance idea
Environment and Framing the Problem
Setting up the Python ML environment
Framing a problem as a machine-learning task
Lab — 10 Exercises & Solutions
Exercises 1–5
Exercises 6–10