Qualcomm AI Hub Tutorial 3: Targeting a specific device
Welcome back to the Qualcomm AI Hub video series! In this tutorial, we'll guide you through targeting a specific edge device when submitting a job to Qualcomm AI Hub, enabling you to deploy your AI models to edge devices within minutes. Let's dive right in!
Targeting a Specific Edge Device
Find Available Devices:
Open your Google Colab notebook, which is already set up for launching jobs to AI Hub.
Run QAI Hub list devices to get the list of available devices, including device name, OS, vendor type, chipset, and how to invoke it.
Select Your Target Device:
For this walkthrough, we'll target the Qualcomm Robotics RB3 Gen 2 platform powered by the Qualcomm QCS 6490 chipset.
Scroll down the list to find QCS 6490 and get the API call for this device.
Prepare Your Model:
Choose the model you want to run. In this example, we'll use the pre-trained YOLOv7 object detection model from AI Hub models.
Trace the model to make it ready for submission to Qualcomm AI Hub.
Submit the Job:
Add the API call to submit the job and customize it to target your chosen device.
Specify the device by adding device=hub_device and then the device name, in this case, QCS6490_proxy.
Run the Compile and Profile APIs:
Run the APIs to compile and profile YOLOv7 on the Qualcomm Robotics RB3 Gen 2.
You'll receive a link to view the results on Qualcomm AI Hub, indicating the job was submitted successfully.
Check out the demo of YOLOv7 performing object detection on the Qualcomm Robotics RB3 Gen 2, optimized using Qualcomm AI Hub and deployed on-device.
Try this out on your own, and if you encounter any issues or have questions, join our Qualcomm AI Hub Community on Slack: https://join.slack.com/t/qualcomm-ai-...
Stay tuned for more videos in this series!
Subscribe: More QDN videos: http://tinyurl.com/2p8xmcw6
Join our developer community at Discord: / discord
ABOUT Qualcomm Developer Network (QDN) is a comprehensive program designed to equip the next generation of mobile pioneers to develop what’s next. Our collection of software and hardware tools and resources is designed so you can build upon our foundational technologies in new and innovative ways, creating the power to build products, enrich lives and even transform entire industries. At Qualcomm Developer Network, we aim to help you kickstart your development by being the catalyst for your vision, today, tomorrow, and in the future.
#QualcommAIHub #AIModelDeployment #DeveloperTutorial #EdgeAI #YOLOv7