QuestionQ38

Automating and orchestrating ML pipelines

You are a data scientist at an industrial-equipment manufacturing company. You are building a regression model to estimate power consumption in the company’s manufacturing plants from sensor data gathered from all plants. The sensors gather tens of millions of records each day. You must schedule daily training runs for the model using all data collected through the current date. You want the model to scale smoothly while requiring minimal development effort. What should you do?

Explanation

BigQuery ML supports regression models trained directly against large BigQuery datasets using GoogleSQL, so it avoids data movement and custom model-training infrastructure. Scheduled queries can automate daily retraining, making it the low-development, scalable option for continually accumulating sensor data.

Learn more

Community Discussion

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