How To Export a TensorFlow Model

Introduction

In this guide you will learn how to export a TensorFlow model and register it in the Model Registry.

Save in SavedModel format

Make sure the model is saved in the SavedModel format to be able to deploy it on TensorFlow Serving.

Code

Step 1: Connect to Hopsworks

import hopsworks

project = hopsworks.login()

# get Hopsworks Model Registry handle
mr = project.get_model_registry()

Step 2: Train

Define your TensorFlow model and run the training loop.

# Define a model
model = tf.keras.Sequential()

# Add layers
model.add(..)

# Compile the model.
model.compile(..)

# Train the model
model.fit(..)

Step 3: Export to local path

Export the TensorFlow model to a directory on the local filesystem.

model_dir = "./model"

tf.saved_model.save(model, model_dir)

Step 4: Register model in registry

Use the ModelRegistry.tensorflow.create_model(..) function to register a model as a TensorFlow model. Define a name, and attach optional metrics for your model, then invoke the save() function with the parameter being the path to the local directory where the model was exported to.

# Model evaluation metrics
metrics = {'accuracy': 0.92}

tf_model = mr.tensorflow.create_model("tf_model", metrics=metrics)

tf_model.save(model_dir)

Conclusion

In this guide you learned how to export a TensorFlow model to the Model Registry. You can also try attaching an Input Example and a Model Schema to your model to document the shape and type of the data the model was trained on.