Thoughts - Ambarish

Azure ML Pipelines

Different types of pipelines are

In Azure Machine Learning, a pipeline is a workflow of machine learning tasks in which each task is implemented as a step.

  • Steps can be arranged sequentially or in parallel
  • Each step can be run on a specific compute target, making it possible to combine different types of processing as required to achieve an overall goal
  • A pipeline can be executed as a process by running the pipeline as an experiment
  • Each step in the pipeline runs on its allocated compute target as part of the overall experiment run.

We can publish a pipeline as a REST endpoint, enabling client applications to initiate a pipeline run.
We can also define a schedule for a pipeline, and have it run automatically at periodic intervals.

References