Getting started with H2O on Python
summary
H2O with Python and optional Hadoop moves log analysis from laptop scripts to a distributed cluster: tutorial shows data pipeline, feature engineering (priority, five date formats, tokenized word‑length fractions, IP/MAC counts), an 80/20 train/validation split with 5‑fold CV, and a GBM model tuned by tree count, depth (~24) and learning rate on roughly 700,000 records. Results show decent overall scores but several classes remain poorly separated; model exports as a POJO for deployment and can run via Flow or Sparkling Water.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:47What H2O is
- 9:01Problem and use case
- 16:09H2O Python setup
- 22:46Install H2O server
- 27:56Connect Python to H2O
- 31:09Data prep and features
- 34:23Import data to H2O
- 36:39Split data train/test
- 38:25Train GBM model
- 41:20Run model and results
- 46:12Explore H2O Flow
- 55:59GBM for classification
- 1:01:50Model output analysis
- 1:07:15Productionizing models
related talks
01h 10mIntroduction to Machine Learning with H2O and Python
50mScalable Data Science & Deep Learning with H2O
45mMassive distributed processing with H2O
01h 18mIntroduction to H20 AutoML with Python
32mDifferent Strategies of Scaling H2O Machine Learning on Apache Spark
30mScalable Automatic Machine Learning with H2O
33mProductionizing H2O Models with Apache Spark
33mH2O internals