Watch this Ask the Expert session to learn the flexibility, efficiency and scalability of combining SAS Event Stream Processing with ONNX Runtime and Microsoft Azure, addressing common challenges in AI model deployment and execution.
Watch the webinar
You will learn to:
- Streamline AI project workflows by integrating SAS Event Stream Processing with ONNX Runtime and Microsoft Azure.
- Integrate ONNX Runtime with SAS Event Stream Processing to create real-time streaming analytics with simplified testing and deployment on a variety of hardware alignments.
- Jump-start projects with sample projects and code examples on GitHub to deploy ONNX models efficiently and effectively.
The questions from the Q&A segment held at the end of the webinar are listed below and the slides from the webinar are attached.
Q&A
What use cases does this solution provide the most value for?
This question is not easy to answer because there are so many use cases. This presentation was just a selection of a few, but I think it’s particularly useful in computer vision. For example, if you need to not only identify objects but also apply further logic and scale up, this solution can handle multiple cameras across that location. We also have use cases with deployments in multiple different geographic locations, allowing us to manage several factories—about 40 factories with hundreds of cameras each—centrally. This flexibility lets you define your specific needs, because no two businesses are the same.
For instance, in some scenarios, we need to identify vehicles in specific factory spaces, while in others, we need to alert workers' safety. Worker safety is a particularly interesting use case with significant implications for improving employee habits and ensuring a safer workspace. This solution helps to easily spot unsafe behaviors and can be quickly customized for the specific needs of different factories.
For example, we have LOTO (lockout-tagout) deployments that verify the dangerous equipment is properly shut off and not able to be started again prior to completion of the task. We receive sensor feedback from the machinery, so we know if the system is verified safe, or if the conveyor belt is moving. And the best part is, all this is possible without needing to change your existing infrastructure. You can use what is already in place, and this solution complements and enhances your current setup, allowing you to apply the necessary logic.
How can I learn more?
We highly recommend reaching out via email to iotcontact@sas.com. This way, you can get a trial license and use the Azure Marketplace as a very easy way to deploy SAS Event Stream Processing and experiment with it. You'll find a comprehensive list of examples available for download at the welcome screen, covering various use cases such as computer vision object detection.
For further examples, you can refer to our GitHub repository, where you can easily get the code. We also have an online community where you can ask questions and get direct feedback from developers. All these links are on the slide deck provided. We’re very welcoming and happy to assist you.
Recommended Resources
SAS Event Stream Processing Homepage
About SAS
ONNX runtime
Microsoft Azure
Please see additional resources in the attached slide deck.
Want more tips? Be sure to subscribe to the Ask the Expert board to receive follow up Q&A, slides and recordings from other SAS Ask the Expert webinars.
Click image to register for webinar