In this tutorial, you are going to learn how to deploy a simple linear regression model. The model predicts the number of users of bike-sharing system. It is modeled in Python from scratch. Then, you are going to deploy this model as a TensorFlow.js application. You will also learn how to host this application on ESP32. You are going to deploy the model in C Arduino code, so the linear regression is run on the ESP32. Finally, you are going to use sensor data as input to the model and plot predictions in real-time chart.
This tutorial teaches you how to build a simple linear regression model from the scratch. This process is quite useful, rather than you just read a little bit of the theory and then start using libraries. Because you will better understand how the algorithm works. Once you understand the algorithm, it is relatively easy to use any libraries. This tutorial is divided into several parts as follows. So, let’s get started.
- Part 1 – Simple Linear Regression Model for Predicting Number of Bike-Sharing Users
- Part 2 – Simple Linear Regression in TensorFlow.js with Bootstrap
- Part 3 – Hosting TensorFlow.js Web Application in ESP32 Arduino
- Part 4 – Simple Linear Regression in C/C++ (ESP32 Arduino) with Embedded Web Application
- Part 5 – Using Sensor Data as Input for Simple Linear Regression and Plot Predictions in Real Time Charts
- Get the full source code: edge-ai repository.
- Andrew Ng’s machine learning lecture (lecture 2.1 – 2.7).
- ESP32 Web Server using SPIFFS (SPI Flash File System).
If you benefit from this page, please make a small donation. I suggest $5, but you can choose the amount. We need your help to pay for web hosting and buy new components and equipment for new tutorials. Alternately, you can make a donation by sending me Bitcoin, at address 1JN3zEw8NqCr8bn4UQa3TfZXaz7UZnb5bH