Deploying your first model in Python
In the previous recipe, we performed the model evaluation step. In this recipe, we will deploy the Linear Learner model to an inference endpoint using the SageMaker Python SDK. What's an inference endpoint? An inference endpoint is a web application endpoint that (1) accepts a set of values as input (for example, x value/s), (2) loads the trained model, (3) uses the trained model to predict a value using the input, and finally, (4) returns the predicted value in the preferred format.
After we have deployed the model, we will test the inference endpoint with a few test predictions using sample management_experience_months values. We should get the corresponding predicted monthly_salary values within a second or less!
Getting ready
This recipe continues on from the Evaluating the model in Python recipe. Make sure you have completed the steps in that recipe along with the Training your first model in Python recipe as we will need the...
