Skip to Content

Blue Waters Webinars

Machine Learning on Blue Waters Using Tensorflow
with the Image Feature Detection Problem
Photograph Aaron Saxton
Aaron Saxton

Data Scientist
Blue Waters project

Saxton's Bio

Aaron Saxton is a Data Scientist who works in the Blue Waters project office at the National Center for Super Computing Applications (NCSA). His current interest is in machine learning, data, and migrating popular data/ML techniques to HPC environments. Aaron’s career has shifted back and forth between industry and academic ventures. Most recently he was a data scientist and founding member of the agricultural data company Agrible Inc. Before that, Aaron worked at Neustar Inc, University of Kentucky, and SAIC. In the summer of 2014, shortly after joining Neustar, Aaron graduated from University of Kentucky to earn his PhD in Mathematics by studying Partial Differential Equations, Operator Theory, and Schrödinger’s equation.


Data analysis utilizing Machine Learning (ML) has become ubiquitous in many scientific disciplines that deal with substantial amounts of data. The broad range of possible applications makes ML particularly interesting to researchers working on data-intensive problems. An example of such a problem is image feature detection, a well-understood, yet new problem which requires a significant amount of storage and processing power. In this webinar, I will walk you through the complete pipeline of image feature detection on the Blue Waters supercomputer. We will start with ImageNet annotated data and learn how to do basic data selection and manipulation. Then, TensorFlow will be the framework by which we will ingest data, train, and validate a Convolutional Neural Network model to perform the image feature detection problem. By the end of the webinar, you will have a trained model that you can use to analyze the distribution of trainable parameters or use as a starting model to continue training on your dataset.

Session details

When: 10:00 CST, February 7, 2018

Length of Session: Two hours

Target Audience: Researchers interested in machine learning

Prerequisites: General scientific/high-performance computing background.


Slides: PDF

Survey: Link (Google Forms)