Description
Filmed at PyData London 2017
Description This talk will be an exposition of machine learning with random forests for Python programmers. The talk will cover the internals of how random forests are implemented, applications that are well suited to the use of random forests, and Python code samples to demonstrate their use.
Abstract Outline
- Intro (5 minutes) What are random forests, how are they used, and what Python software is available for using them? What strengths do they have relative to other models (scalability and applicability to a broad range of problems)?
- Forest Internals (15 minutes)
- Decision Trees (5 minutes)
- Presentation of the decision tree model, the building block of random forests.
- Entropy Minimization (5 minutes)
- Explanation of how decision trees are tuned using entropy minimization.
- Building Forests from Decision Trees (5 minutes)
- Explanation of how decision trees are aggregated to form random forests.
- Illustrative Examples (10 minutes)
- Regression on non-linear functions (5 minutes)
- Classification with unscaled features (5 minutes)