Motion Sensor for Occupancy Detection

Dashboard

To visit Dashboard deployed on vercel click here

To see the dashboard on your localhost, follow these steps

Note: Make sure npm, node are installed on your pc.

cd Dashboard
npm i
npm start

Now open http://localhost:3000/ on your machine to see the dashboard.

ESP32 Code

Deployment

The final code used while deployment is in the folder final and named as final.ino. Click here to see the code

Training

The code used for training is in two folders

One for zero people in the room. It is in the folder for0values and named as for0values.ino. Click here to see the code.
One for non-zero people in the room. It is in the folder forvalues and named as forvalues.ino. Click here to see the code.

ESP32 Cam code

The code for ESP32 cam is in the folder Esp_32_cam_drive. It is in three files. Click here to see the code for cam.

ML model code

The code for ML model is in classifier.py which is in the folder python/MLAlgo. Click here to see the code.

Python server

We need to run a python server parallely and it's code is in the folder python/MLAlgo and the file name is app.py, Click here to see the code.

To run the python server follow these steps

Note: Make sure you have python and pip installed on your pc.

Run the following commands to install dependencies

pip install --user --upgrade catboost
pip install --user --upgrade ipywidgets
pip install shap
pip install sklearn
pip install --upgrade numpy
jupyter nbextension enable --py widgetsnbextension

Run this command to start the server.

cd python/MLAlgo
python app.py

Run Dashboard code, final.ino and app.py (python server) at the same time.

srisatyavinay/occupancy-detection