The Habana® team is excited to be at re:Invent 2022, November 28 – December 1. We’re proud that Amazon EC2 DL1 instances featuring Habana Labs Gaudi deep learning accelerators are providing an alternative to GPU-based EC2 instances, delivering a new level of performance and efficiency in developing, training and deploying deep learning models and workloads.
This week we will provide demonstrations and present sessions to help customers learn how to train more and spend less with DL1 instances. Please check out the following opportunities to learn how Habana can deliver greater deep learning value to customers on the AWS cloud.
Mobileye Demo
Come to Intel booth #1617 to see how Gaudi-based EC2 DL1 instances are helping Mobileye make roadways safer by detecting, segmenting and tracking thousands of images per second to enable real-time, on-road perception. This is an example of how Gaudi is optimized for large scale distributed training in the most challenging environments.
Breakout session: Mobileye: Smarter, safer autonomous solutions with Habana & Granulate
Monday, November 28, 2:30 – 3:30pm, Room PRT279
See You There
You can find us at Intel booth #1617 and at #2741 in re:Invent’s AI Neighborhood or come see us at any of the following sessions to learn more about Gaudi DL accelerators.
Location: Venetian Lido 3106
Speakers:
Pini Reisman, Director of REM Cloud Application, Mobileye
Ohad Shitrit, Senior Director of AI Engineering and Algorithms, Mobileye
Gilberto Vargas, Vice President & General Manager, DCAI Category Sales, Intel Corporation
Mobileye, a global leader in automotive and driving assistance systems, is revolutionizing how we drive. Intel is committed to bringing Mobileye the best in hardware and software solutions, allowing them to continually innovate, reduce collisions, and protect road users. In this session, learn how Mobileye utilizes Amazon EC2 DL1 instances powered by Gaudi accelerators and Granulate’s optimization technology to realize cost and performance outcomes. This presentation is brought to you by Intel, an AWS Partner.
Breakout Session: Accelerate Deep Learning Training with Habana Gaudi-Based Amazon EC2 DL1 Instances
Thursday, December 1, 11:45am – 12:45pm, Room PRT280
Speakers:
Dvij Bajpai, Senior Product Manager – Technical, AWS and Greg Serochi, Applications Engineering, Habana Labs
Learn how to maximize training throughput and efficiencies to address the growing number of applications and services that can improve our daily lives based on image and gesture recognition, speech recognition, natural language processing, and more with Amazon EC2 DL1 instances powered by Gaudi accelerators.This session includes a live demonstration to highlight the cost-effective and easy-to-use AI solution that is democratizing deep learning (DL) by empowering more DL users, increasing usage frequency, and creating more DL usage models.
We also look forward to sharing the latest performance advances with Habana’s second-generation deep learning training accelerator, Gaudi2, along with demonstrating our collaborations with partners like Hugging Face that will help accelerate model implementation on DL1 instances and showing how easy it is to build new or migrate models based on Gaudi solutions.
Text-to-image Diffusion Demo
See Gaudi2 DL accelerators execute a state-of-the-art text-to-image diffusion model that can spontaneously generate batches of 16 unique images in roughly 3.5 seconds. This blazing-fast deployment is executed on a single Gaudi2 DL accelerator. With this new and evolving text-to-image technology, we are making creativity more accessible to businesses and individuals, which is opening up an entirely new realm of creative possibilities. Come by and try it out yourself.
Hugging Face Meet & Greet
Join Habana for an exclusive meet & greet with Hugging Face, featuring Philipp Schmid, Technical Lead with Hugging Face.
Wednesday, November 30, 5:30 – 7:00 p.m. Venetian Expo | Casanova, Level 1, Room #603
During this meet & greet, participants will learn how Habana and Hugging Face can accelerate model implementation on DL1 instances with Text-to-Image Stable Diffusion, in particular. And, we’ll dive a little deeper with a Q&A/chalk talk designed to help you get started on Gaudi, DL1 and Hugging Face.
Want to Connect?
The Habana team would love to meet you and learn about your deep learning compute requirements. Feel free to come by our booth or drop us your info here and we’ll respond asap.
Habana Labs re:Invent ’22 Schedule:
Monday, November 28 | |
2:30 – 3:30pm | Mobileye: Smarter, safer autonomous solutions with Habana & Granulate (sponsored by Intel), room #PRT279 |
4:00pm – 7:00pm | Expo Welcome reception, Intel booth #1617 and #2741 AI Neighborhood, Venetian Expo |
Tuesday, November 29 | |
10:00am – 6:00pm | Expo hours, Intel booth #1617 and #2741 AI Neighborhood |
Wednesday, November 30 | |
10:00am – 6:00pm | Expo hours, Intel booth #1617 and #2741 AI Neighborhood |
5:30 – 7:00pm | Hugging Face Meet & Greet, Venetian Expo | Casanova, Level 1, Room #603 |
Thursday, December 1 | |
10:00am – 4:00pm | Expo hours, Intel booth #1617 and #2741 AI Neighborhood |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.