AI-Powered Business Intelligence by Tobias Zwingmann

Author:Tobias Zwingmann [Tobias Zwingmann]
Language: eng
Format: epub, pdf
Publisher: O'Reilly Media, Inc.
Published: 2022-10-24T16:00:00+00:00

Model Deployment with Microsoft Azure Walkthrough

After we trained, validated and understood our machine learning model it is time to deploy it so we can use it to make predictions for new data.

In your Azure Machine Learning Studio, choose the model you want to make available and hit “deploy”. The “Deploy a model” pane will pop up on the right side as shown in Figure XX. Give your model deployment a name. I suggest using a combination of the model purpose and the model type your are using. In our case, a good model name could be “arrdel-votingensemble”. Next, select the compute type. You can choose here between Azure Kubernetes Services and Azure Container Instance. Both resources are typically used for real-time inference. The Kubernetes service is typically used for high-scale production deployments where fast response time and autoscaling are needed. That’s just the opposite of our prototyping requirements. We’re good with Azure Container service which is used for low-scale CPU-based workloads that require less than 48 GB of RAM. Make sure that authentication is disabled for now and hit “Deploy”.

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.