End to end AI developer journey with containerized assets using Intel DevCatalog, Intel DevCloud
summary
Shows how to grab containerized AI assets from DevCatalog and run, build and benchmark them on Intel DevCloud using a browser-based container playground with JupyterLab, Git and Docker registry import. Demonstrates a people-counter image (197 FPS, 7018 ms latency reported) and an intelligent-traffic-management stack (three containers: app, InfluxDB, Grafana) deployed to selectable Intel CPUs/GPUs/VPUs/FPGA, and explains queueing, telemetry export and OpenVINO integration.
AI-written summary · about
chapters (4)
related talks
42mAI Everywhere– accelerate your development from PC to cloud
06mLightning Talk: Deploying AI Solutions on the Edge with Containers
05mAccelerate AI development from cloud to client
24mTools & Techniques for a cost optimized Analytics & AI pipeline
31mNavigating Containers: AI Powered by Docker
34mFaster deep learning solutions...
53mVideo Analytics for mere mortals with NVIDIA DeepStream SDK
35mHow does Docker run machine learning on specialized AI