Professional Machine Learning Engineer

By Google · Question Mode

QuestionQ45

Automating and orchestrating ML pipelines

You are building a Kubeflow pipeline on Google Kubernetes Engine. Its first step must run a query against BigQuery, and you plan to use that query’s results as input for the next pipeline step. You want to accomplish this in the simplest possible way. What should you do?

Explanation

Kubeflow Pipelines supports loading reusable components directly from a URL and composing their outputs into downstream pipeline tasks. Reusing the BigQuery Query Component performs the query within the pipeline and supplies its output for the next step without requiring a custom script or custom component implementation.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!