Brought to you by NumFOCUS Foundation and O’Reilly Media
The official Jupyter Conference
Aug 21-22, 2018: Training
Aug 22-24, 2018: Tutorials & Conference
New York, NY

Scaling notebooks for deep learning workloads (sponsored by IBM Watson)

LUCIANO RESENDE (IBM)
11:05am–11:45am Thursday, August 23, 2018
Sponsored
 Location: Regent Parlor
Average rating: *****
(5.00, 2 ratings)

What you'll learn

  • Learn how to use the Jupyter Notebook along with platforms and services such as Fabric for Deep Learning (FfDL) for cost-effective full dataset training of deep learning models

Description

Deep learning workloads are compute intensive, and training these type of models is better done with specialized hardware like GPUs. Luciano Resende outlines a pattern for building deep learning models using the Jupyter Notebook’s interactive development in commodity hardware and leveraging platforms and services such as Fabric for Deep Learning (FfDL) for cost-effective full dataset training of deep learning models.

This session is sponsored by IBM Watson.

Photo of LUCIANO RESENDE

LUCIANO RESENDE

IBM

Luciano Resende is a senior technical staff manager (STSM) and open source data science and AI platform architect at IBM CODAIT (formerly Spark Technology Center). He’s a member of ASF, where he’s been contributing to open source for over 10 years. He contributes to various big data-related Apache projects around the Apache Spark ecosystem as well as Jupyter ecosystem projects, building a scalable, secure, and flexible enterprise data science platform.