用Keras保存、加载和导出模型

你将学到什么

课程概况

In this 1 hour long project based course, you will learn to save, load and restore models with Keras. In Keras, we can save just the model weights, or we can save weights along with the entire model architecture. We can also export the models to TensorFlow’s Saved Mode format which is very useful when serving a model in production, and we can load models from the Saved Model format back in Keras as well.

In order to be successful in this project, you should be familiar with python programming, and basics of neural networks.

Note: This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions.

课程大纲

Save, Load, and Export Models with Keras

Welcome to this project-based course on how to save, load, and export models with Keras. In this 1 hour long project, you will learn to save, load, and restore models with Keras. In Keras, we can save just the model weights, or we can save weights along with the entire model architecture. We can also export the models to TensorFlow's Saved Mode format which is very useful when serving a model in production, and we can load models from the Saved Model format back in Keras as well.

课程项目

Create a function to return model instances

Data Processing

Keras model checkpoint callback

Create and evaluate a new model instance

Explore the save weights and the save function on model instances

Use the save weights and the save functions to export to and import from the Saved Model format