Massive distributed processing with H2O
summary
H2O is presented as a distributed data-analysis platform for massively parallel processing and was named a Visionary by Gartner. The talk covers H2O’s architecture, features and capabilities, and demonstrates integrations with Spark, TensorFlow and MxNet plus several popular algorithms through interactive demos.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 5:56What is H2O and why use it
- 12:05Features and benefits
- 17:12Distributed architecture
- 20:27Gradient Boosting Machine (GBM)
- 28:30Demo: Iris classification
- 34:58Demo: Neural networks
- 38:14Q&A
related talks
50mScalable Data Science & Deep Learning with H2O
33mH2O internals
01h 10mGetting started with H2O on Python
32mDifferent Strategies of Scaling H2O Machine Learning on Apache Spark
50mFast Analytics On Big Data
33mProductionizing H2O Models with Apache Spark
27mDeploying distributed AI and machine learning in financial services
42mH2O Deep Water with Python early sneek