BigQuery ML in a minute

♪ [upbeat music] ♪ Give me one minute, and I'll show you how BigQuery ML can enable you to build machine learning models quickly. BigQuery ML lets you operationalize machine learning models right within BigQuery, using standard SQL. ML usually requires extensive programming and knowledge of specialized frameworks, but BigQuery ML empowers you to use your existing SQL skills to build and use machine learning models in under 10 minutes, saving you from exporting data and managing ML infrastructure.

It supports running models on petabyte scale data in a fraction of the time. It can also handle preprocessing like splitting data, one-hot encoding, and normalization. And it comes with support from many types of ML models and so much more.

You can even use AutoML tables to create best-in-class models while handling feature engineering and model selection. To set up BigQuery ML, create the dataset in BigQuery and just write a simple SQL query with the <i>create model</i> clause to create the model. Execute the query to run and observe the model as it's being trained in the <i>Model Stats</i> tab.

From there, you can use your trained model to make predictions, using the <i>ML Predict</i> function in a new query, and enjoy the results directly in the BigQuery console. Just like that. BIgQuery ML democratizes predictive analytics and makes it easier for anyone to get started with ML.

It's being used for anomaly detection, customer segmentation, product recommendations, predictive forecasting, and more. So, next time you want to build models using your BigQuery data, take a minute to look at BigQuery ML. To learn more check out <i> cloud.

google.