30 Golden Rules of Deep Learning Performance

Duration: 50 mins
Siddha Ganju
Self-driving Architect, NVIDIA

“Watching paint dry is faster than training my deep learning model.”

“If only I had ten more GPUs, I could train my model in time.” “I want to run my model on a cheap smartphone, but it’s probably too heavy and slow.”

If this sounds like you, then you might like this talk.

Exploring the landscape of training and inference, we cover a myriad of tricks that step-by-step improve the efficiency of most deep learning pipelines, reduce wasted hardware cycles, and make them cost-effective. We identify and fix inefficiencies across different parts of the pipeline, including data preparation, reading and augmentation, training, and inference.

With a data-driven approach and easy-to-replicate TensorFlow examples, finely tune the knobs of your deep learning pipeline to get the best out of your hardware.

You may also be interested in

50 mins
Rabbit MQ vs Kafka: When, Which, Where?

Rabbit MQ and Kafka are both considered to be the Go-To frameworks for data transmission (messaging). Many developers place them...

50 mins
Mutation Testing

Most developers are familiar with the concept of unit testing, and how this is useful to ensure validity of your...

180 mins
Java and the Semantic Web

Java has always been one of the languages and runtimes with the best support for Semantic Web standards. There have...

50 mins
Java 9 Memory and GC

So your server is having issues? memory? Connections? Limited response? Is the first solution to bounce the server? Perhaps change...

50 mins
Beginning Java 9 Modules

Java 9 modules present a radical shift away from how we thing about modularizing our applications. However, as we have...

50 mins
Java Functional Programming Idioms

A number of developers and organizations are beginning to make use of Functional Programming in Java. With anything that's new,...