Scalable Data Science and Deep Learning with H2O

Explore scalable data science and deep learning with H2O in this 50-minute conference talk from GOTO Chicago 2015. Dive into the capabilities of H2O, an open-source machine learning platform that can handle billions of data rows in-memory, even on small compute clusters. Learn about its interfaces for R, Python, Scala, Java, JS, and JSON, as well as the interactive graphical Flow interface. Discover how H2O integrates with Hadoop and Spark, and witness demonstrations of its applications in normal digit detection, crime detection, and Kaggle challenges. Gain insights into H2O's performance metrics, deep learning capabilities, and world record achievements. Follow along as the speaker covers topics such as downloading H2O, data parsing, date manipulation, and using various programming languages with the platform. Understand how H2O can be leveraged to create smarter applications and streamline analytic workflows for both engineers and non-engineers alike.

Intro
Agenda
H2O
What is H2O
H2O customers
Downloading H2O
Demo
Metrics
Deep Learning
World Record
Normal Digit Detection
Crime Detection
Datanami
Spark Memory
Parse Dataset
Date Manipulation
Spark
Scala
Flow
State Up Table
Python
IPython Notebook
Kaggle Challenge
GBM
Outlook