Note: All images in this directory, unless specified otherwise, are licensed under CC BY-NC 4.0.
| Figure number | Description | Notes |
|---|---|---|
| 9-1 | A high-level overview and comparison of different inference serving options | |
| 9-2 | Navigate to http://localhost:5000/hello within a web browser to view the “Hello World!” web page | |
| 9-3 | Listing page for machine learning models on the Google Cloud ML Engine dashboard | |
| 9-4 | Model creation page on Google Cloud ML Engine | |
| 9-5 | Model listings page on Google Cloud ML Engine | |
| 9-6 | Details page of the just-created Dog/Cat classifier | |
| 9-7 | Creating a new version for a machine learning model | |
| 9-8 | Creating a new Google Cloud Storage bucket within the ML model version creation page | |
| 9-9 | Google Cloud Storage Browser page showing the uploaded Dog/Cat classifier model in TensorFlow format | |
| 9-10 | Add the URI for the model you uploaded to Google Cloud Storage | |
| 9-11 | An end-to-end pipeline illustrated in KubeFlow | |
| 9-12 | Creating a new Jupyter Notebook server on KubeFlow | |
| 9-13 | Creating a KubeFlow deployment on GCP using the browser | |
| 9-14 | Google Cloud ML Engine showing incoming queries and latency of serving the calls, with end-to-end latency at user’s end of about 3.5 seconds | |
| 9-15 | Cost comparison of infrastructure as a service (Google Cloud ML Engine) versus building your own stack over virtual machines (Azure VM) (costs as of August 2019) |