Machine Learning Fundamentals
Build your first ML models and understand the theory behind them.
About this course
A thorough introduction to machine learning theory and practice. You will implement supervised and unsupervised algorithms from scratch, learn to evaluate and tune models correctly, and complete three real-world projects in classification, regression, and clustering. Python and scikit-learn are used throughout.
Target audience: Software engineers moving into ML, analysts building predictive models
What you will learn
- Supervised learning
- Unsupervised learning
- Model evaluation
- scikit-learn
- Feature engineering
Course syllabus
10 modules · video + projects
- 1The ML problem formulation: inputs, outputs, loss functions
- 2Linear regression: from normal equations to gradient descent
- 3Logistic regression and classification metrics
- 4Decision trees and random forests
- 5Support vector machines and kernel methods
- 6Clustering: k-means, hierarchical, and DBSCAN
- 7Dimensionality reduction: PCA and t-SNE
- 8Model selection, cross-validation, and hyperparameter tuning
- 9Building a production-ready ML pipeline
- 10Capstone: predict customer churn on a real dataset
Prerequisites
- –Python basics
- –Basic statistics
Frequently asked questions
Does this course cover deep learning?
This course covers classical ML only. For neural networks and deep learning, see the "Deep Learning with PyTorch" course, which this course is a prerequisite for.
How much maths do I need to know?
You need a working understanding of high-school algebra and basic statistics (mean, variance, probability). The course builds up the calculus intuition you need without requiring formal proof-writing.
Ready to start Machine Learning Fundamentals?
Join 5,300+ learners already enrolled. Self-paced, certificate on completion.