AdvancedContent in Spanish
Machine learning with scikit-learn
Move from statistical theory to models that work. You will learn the full workflow of a machine learning project with scikit-learn: train/test split, preprocessing, the essential supervised and unsupervised algorithms (regression, trees, random forest, SVM, kNN, k-means, PCA), how to measure performance with the right metrics and how to fight overfitting with cross-validation, regularization, pipelines and hyperparameter tuning. It is the third course of the Professional Certificate in Data Science.
10 lessons 14 hours 1000 XP
Start the courseSyllabus
Module 1 · Fundamentals and workflow of an ML project
- 1What machine learning is and its paradigms45 min
- 2Train, test and validation50 min
- 3Preprocessing: encoding and scaling55 min
Module 2 · Supervised models
- 4Linear and logistic regression50 min
- 5Decision trees and random forest55 min
- 6SVM and k-nearest neighbors55 min
Module 3 · Unsupervised learning and evaluation
- 7Clustering (k-means) and dimensionality reduction (PCA)55 min
- 8Evaluation metrics55 min
Module 4 · Validation, generalization and production
- 9Cross-validation, overfitting and regularization55 min
- 10Pipelines and hyperparameter tuning55 min