testimonial

ScottishPower and Metacoda present at SAS Forum UK 2019

Learn how to manage and optimise your SAS platform at #SASForumUK!

The SAS Forum UK website states…

“Analytics informs. It inspires.
It transforms the world around us.
KNOW how to put your AI & Analytics in Action.”

#SASForumUK Charles Senabulya invitation

With a theme of inspiration and transformation, we’re looking forward to SAS Forum UK on the 25th June 2019 and hope to see you there.

After all, as Oliver Schabenberger (SAS CTO and COO and keynote speaker at the upcoming conference) has been saying for years; “Data without analytics is value not yet realised“.

SAS Institute has been sharing the power of analytics for more than 40 years, and now – with advancements in technology – more businesses than ever are realising the true value of data.

In his compelling invitation Charles Senabulya, VP and Country Manager for SAS UKI, explains that SAS Forum UK will focus on stories from customers about HOW they are bringing analytics to life in extraordinary ways every day.

It’s an opportunity not to be missed!
Read more

Paul demonstrating Metacoda software to Tamara Barker at SASGF 2019

Power to do Good: bringing analytics to life at #SASGF

Chris Hemedinger SASGF Facial Recognition Demo Tweet

When more than 6,000 SAS users came together at SAS Global Forum 2019 in Dallas, it was clear that we are living in exciting times, full of potential!

Dr Jim Goodnight, SAS CEO, set the scene perfectly in the opening session:
“We are witnessing a renewed focus on data and analytics, driven by the combination of increased computing power, a more connected world and more powerful technologies, like artificial intelligence, computer vision, natural language processing and machine learning.”

To support his words, he blew us away with a live facial recognition demonstration, using an enthusiastic group of volunteers (including Metacoda’s Michelle Homes) and a model TinyYOLO v2. The model essentially converts each face into a vector of 512 floating point numbers, looks it up using nearest neighbour, and then returns a name. But obviously behind the scenes there’s a lot more going on: the facial recognition uses approximately 11 million parameters… with more than 22 million floating point operations that have to occur on a workstation with 2 GPUs… which sends the data to SAS Viya running ESP on 4 GPUs… which sends the pixels in the bounding box into a model that has 55 million parameters from FaceNet! If you’re intrigued, please see Dr Goodnight’s opening session presentation for yourself (and notice that even funny faces can’t fool that clever underlying algorithm!) 😉

Read more