Using Random Forests in Python
summary
Shows concrete results: out‑of‑the‑box scikit-learn regressor fits a noisy xsin(x) curve with no tuning, a random forest classifier* perfectly separates quadrants when X spans -0.1..1 and Y spans -5000..5000, and a learned search policy on arrays of size 100 slightly outperforms binary search across most quantiles while admitting tail cases where it fails. Notes parallel deployment via Spark/MLlib and explains bagging, entropy minimization, tree depth and feature‑subset choices.
AI-written summary · about
chapters (7)
- 0:00Random forests intro
- 5:23Building the forest
- 8:41Why random forests are effective
- 13:02Random forests versus neural networks
- 15:35Scaling random forests with distributed systems
- 16:18RFs AI example
- 20:18Q&A
related talks
36mAdventures in scikit-learn's Random Forest
44mUnderstanding Random Forests
57mFrom Bonsais to Random Forests
40mCART: Not only Classification and Regression Trees
53mA Taste of Random Decision Forests on Apache Spark
37mAttacking Clustered Data with a Mixed Effects Random Forests Model in Python
33mQuantifying uncertainty in machine learning models
42mCreating correct and capable classifiers