Includes ONNX runtime support on Node.js.
Tested on wokwi.com ESP32 C++ and VitCon IoT-MODLINK, for remote AIoT lab sessions.

Web Panel Screenshot	Wokwi Simulator Diagram

• ESP32 C++ example code
• Arduino Mega C++ example code
• IoT web server with customizable control panels
• JSON-based DB (powered by lowdb)
• On-server ML inference (powered by onnxruntime-node)

1. Install node.js v16 or higher on your server.
2. Clone the repository on your server. git clone https://github.com/wldh-g/iot-lab-for-beginners.git
3. Install the dependencies. npm install
4. Run the server. npm run dev
5. Go to https://wokwi.com/projects/326257884268593746.
6. Edit line 14, your-ip-or-domain to your server's IP address or domain.
   If you are using port rather than 3000 on your server, also edit line 14, 3000 to your port number.
7. Run the simulation using the green start button.
8. Go to http:https://your-ip-or-domain:port/panel/example1.
9. Enjoy!

Note: If you are a lecturer and if you want to use wokwi.com in your class, first contact the administrator of wokwi.com to get dedicated build servers and gateways. You may have to pay for them.

1. Install node.js v16 or higher on your server.
2. Clone the repository on your server. git clone https://github.com/wldh-g/iot-lab-for-beginners.git
3. Install the dependencies. npm install
4. Run the server. npm run dev
5. Run the broker server. npm run broker
6. Install DHT sensor, U8g2, SimplyAtomic, ArduinoJson, AsyncTimer library from Arduino Library Manager.
7. Install SoftPWM, Adafruit_Sensor-master library from here.
8. Install Vegemite library at example/ArduinoMega-cpp/Vegemite.
9. Configure VitCon IoT-MODLINK WiFi-LINK with this Android app or this iOS app. Host is your-ip-or-domain and port is 3010.
10. Open example/ArduinoMega-cpp/sketch.ino and compile it.
11. Upload the program.
12. Go to http:https://your-ip-or-domain:port/panel/example2.
13. Enjoy!

1. If you want to use a pre-trained model, skip to step 5.
2. Gather data named humidity and temperature from your IoT device (or wokwi simulator) into db.json.
3. Run example/inference/dht_predict.ipynb with your own db.json.
4. Save humi_predict.onnx and temp_predict.onnx.
5. Copy .onnx files and model_map.yaml to the root directory of this repository.
   For the pre-trained .onnx files and model_map.yaml, please check example/inference directory.
6. Uncomment "H/T Inference" section of pages/panel/example.tsx.
   i.e. delete two lines({/* and */}).
7. Re-run the server. npm run dev
8. Go to http:https://your-ip-or-domain:port/panel/example1 or example2.
9. Enjoy!

Note that the pre-trained model in this repository is not accurate. It is just a simple example to show how inference can work.

Made with ♥️ by Jio Gim.