TensorFlow Machine Learning Projects

Running the model on a browser using TensorFlow.js

In this section, we are going to deploy the model on a browser.

The following steps demonstrate how to save the model:

1. Install TensorFlow.js, which will help us format our trained model in accordance with what can be consumed by the browser:

pip install tensorflowjs

2. Save the model in the TensorFlow.js format:

import tensorflowjs as tfjs tfjs.converters.save_keras_model(model, OUTPUT_DIR)

This will create a json file called model.json, which will contain the meta-variables and some other files, such as group1-shard1of1. 

Good job! Deploying the model in the HTML file is a little trickier, however:

For running the code mentioned in the repository, please follow the README.md documentation  ...