Introduction to H20 AutoML with Python
summary
H2O AutoML runs in Python and Flow, trains ensembles (GLM GBM XGBoost DRF XRT deep nets) on a Freddie Mac loan set, handles imputation and encoding, and returns a leaderboard plus stacked models; a five‑minute run produces many models and a stack whose AUC 0.823 on test matches cross‑val trends; dataset shows class imbalance ~3.6% delinquent; models export as MOJO/POJO for production; demos use Aquarium lab and cloud/Spark integration.
AI-written summary · about
chapters (16)
- 0:00H2O and AutoML intro
- 8:56AutoML core features
- 13:22What AutoML automates
- 18:12AutoML benchmarks
- 18:56Setting up the hands-on environment
- 24:21Load data & start cluster
- 27:18Explore mortgage data
- 31:44Split data & predictors
- 33:31Run AutoML classification
- 40:20Leaderboard & ensemble
- 47:07Test model performance
- 50:17Run AutoML regression
- 57:23Model parameters
- 1:01:01H2O Flow intro
- 1:05:49Models in H2O Flow
- 1:09:54Q&A
related talks
30mScalable Automatic Machine Learning with H2O
44mScalable Automatic Machine Learning with H2O
01h 10mIntroduction to Machine Learning with H2O and Python
01h 26mMachine Learning Algorithms with H2o and PowerShell
01h 10mGetting started with H2O on Python
50mScalable Data Science & Deep Learning with H2O
27mDeploying distributed AI and machine learning in financial services
42mH2O Deep Water with Python early sneek