Building Machine Learning Pipelines by Hannes Hapke

Author:Hannes Hapke
Language: eng
Format: epub
Publisher: O'Reilly Media
Published: 2020-07-19T16:00:00+00:00

Figure 8-10. Deployment of models produced from TFX pipelines

In “TFX Pusher Component”, we discussed the Pusher component. The TFX component allows us to push validated models to a given location (e.g., a cloud storage bucket). TensorFlow Serving can pick up new model versions from the cloud storage location, unload the earlier model version, and load the latest version for the given model endpoint. This is the default model policy of TensorFlow Serving.

Due to the default model policy, we can build a simple continuous deployment setup with TFX and TensorFlow Serving fairly easily.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.