How Tensorflow is Evolving

Andres Rodriguez (Sr. Principal Engineer @Intel) talks about how Tensorflow v2 has evolved, use-cases and applications, frequent usage patterns, and the best ways to begin using Tensorflow.  

SHOW: 466

SHOW SPONSOR LINKS:

• Datadog Security Monitoring Homepage - Modern Monitoring and Analytics
• Try Datadog yourself by starting a free, 14-day trial today. Listeners of this podcast will also receive a free Datadog T-shirt.
• DivvyCloud - Achieve continuous security & compliance. Request a free trial today!
• DivvyCloud - The best mistakes are the ones that don’t happen. Learn how IaC offers preventive cloud security.

CLOUD NEWS OF THE WEEK - http://bit.ly/cloudcast-cnotw

PodCTL Podcast is Back (Enterprise Kubernetes) - http://podctl.com

SHOW NOTES:

• Tensorflow Homepage
• Improving TensorFlow Inference Performance on Intel® Xeon® Processors 
• NASA Frontier Development Lab Case Study 
• Tensorboard: Tensorflow visualization
• Google Cloud TPUs
• Most Popular GitHub Projects (2019)

Topic 1 - Welcome to the show. You’ve had a really interesting mix of industry, government and academic work around Deep Learning. Tell us a little bit about the areas you focus on. 

Topic 2 - Tensorflow is one of the most popular OSS projects on Github. Help us understand the types of data problems where Tensorflow is the best tool/framework (e.g. neural networks). What are some of the most popular capabilities? 

Topic 3 - Tensorflow is focused on looking at how data flows through graphs. Are there common types of ML problems that are more appropriate for using Tensorflow than other ML models? 

Topic 4 - Tensorflow is able to run on a broad set of hardware, but obviously there is a point where specialized hardware is needed for certain performance or scaling. What are some of the things that Intel is doing to help improve the experience with Tensorflow? 

Topic 5 - Tensorflow works primarily with the Python programming language. Beyond having some background with Python, what are some of the skills that are needed to get started and be successful with Tensorflow? 

Topic 6 - Having worked with Tensorflow for a while now, what are some of the learning paths that you’ve found successful (problems areas, communities of interest, tools, new skills, etc.)? 

FEEDBACK?

• Email: show at thecloudcast dot net
• Twitter: @thecloudcastnet