Hi everyone, I’m currently working on deploying a lightweight CNN model on edge devices for an IoT-based object detection project. The tricky part is optimizing PyTorch-trained models to run efficiently on constrained hardware, such as ESP32, in conjunction with external processors. While going through resources, I found a helpful overview about IoT basics https://www.theengineeringprojects.com/2023/06/top-iot-starter-kits-for-the-beginners-to-learn-programming.html which nicely explains the broader IoT setup I’m using, but it doesn’t cover the machine learning side in depth. Has anyone here successfully converted and deployed a model for real-time IoT applications? Would love any practical tips or recommended workflows.