Introduction to

machine learning

with scikit-learn

Credit: Benjamin Bengfort

Machine learning

• Subfield of computer science
• Evolved from pattern recognition & learning theory in AI
• Provides computers with the ability to learn without having being explicitly programmed
• Learn from and make predictions from data

• Classification

• Regression
• Clustering
• Semi-supervised learning
• Feature selection
• Feature extraction
• Dimensionality reduction
• ...

scikit-learn: About

• Extensions of SciPy are called scikits
• scikit-learn provides ML algorithms
• Supervised & unsupervised learning
• Built on Numpy and SciPy
• Sits on top of C libraries
• Standard Python API interface
• Open Source (BSD License)
• scikit-learn 0.17 (stable)
• http://scikit-learn.org/