Introduction
Optimize model scoring by choosing the appropriate inferencing strategy within Azure Machine Learning.
Imagine you're a data scientist working for a manufacturing company. Since your company works with heavy-duty equipment, safety is the number one priority.
The safety monitoring team installed cameras in areas where there is a high risk for unsafe situations, like people not wearing helmets near dangerous machinery. The team has asked you to train a model that identifies objects such as helmets and machinery, so that alerts can be sent when a safety risk is detected.
To train compute-intensive models, like a computer vision model, already requires much compute power. When you want to deploy a trained model to get real-time or batch predictions, you have to choose the appropriate deployment configuration for optimal performance.
After this module, you'll be able to design and execute an inference strategy to deploy compute-intensive models.
Learning objectives
In this module, you'll learn:
- To choose the appropriate inference strategy.
- To optimize model scoring with ONNX.
- To deploy Triton as a managed online endpoint.