This project is a demo of a linear regression model using neural networks, implemented in JavaScript using TensorFlow.js. It was created as part of the TensorFlow.js course on Platzi.
The project provides a web interface where users can train a linear regression model and visualize its performance. The model is trained using stochastic gradient descent (SGD) and the mean squared error (MSE) loss function. The demo also allows users to load pre-trained models and generate inference curves.
You can try out the demo here, entering the URL of the JSON data to use, and the X and Y variables to plot. Then use the buttons on the page to train the model, save it, load pre-trained models, and generate inference curves.
The project uses the following libraries:
- TensorFlow.js (version 1.7.4)
- tfjs-vis (version 1.0.2) These libraries are included in the HTML file using CDN links.
Contributions to the project are welcome. To contribute, fork the repository, make your changes, and submit a pull request.