Hi !
In my previous post I wrote about how to create a custom dataset with images to be used on a Azure Machine Learning Designer project. The second post was about how to use the custom data set and how to train the model. Today let's publish the model to be used as a WebService / HTTP Rest endpoint.
Next step is to select the "real time inference pipeline" option from the [Create inference pipeline] option and press [Publish].
This will create a new designer pipeline project, related to the original one. A closer look at the new pipeline, shows new elements in the designer area
- WebService Input
- WebService Output
And also, on top, we can see how both designer pipelines are related.
Let's Submit this pipeline and wait for the completion of the process.
Important: you can change the default names for WebServiceInput1 and WebServiceOutput1. This will help later with the JSON schema to consume the webservice.
Once the process is complete, we can deploy this running pipeline to be used as an HTTP REST Endpoint. I'll create a new endpoint named [squirrelimageclassification], and a Compute Instance to host this resource.
A couple of minutes later, we are ready to go !
When we open the new endpoint, we can review useful information like the HTTP endpoint !
And we even have a test section and sample code to test the Endpoint.
Tomorrow I'll show how to write a client desktop app to use the the generated endpoint!
Top comments (0)