Model progress can be saved during and after training. This means a model can resume where it left off and avoid long training times. Saving also means you can share your model and others can recreate your work. When publishing research models and techniques, most machine learning practitioners share:

• code to create the model, and

• the trained weights, or parameters, for the model

Sharing this data helps others understand how the model works and try it themselves with new data.

TensorFlow models are code and it is important to be careful with untrusted code. 

Options

There are different ways to save TensorFlow models depending on the API you’re using. This guide uses tf.keras, a high-level API to build and train models in TensorFlow. For other approaches see the TensorFlow Save and Restore guide or Saving in eager.

Please consider subscribing to my newsletter, and get access to all of the articles and source code. Get access to in-depth analysis and extensive tutorials. Also, you get to download source code which you can add to your resume, in order to get your dream job. Also, I have a podcast coming in a few days, so be ready, and subscribe.