Model creation

BigQuery ML lets you build and operationalize machine learning (ML)models over data in BigQuery by using SQL.

A typical model development workflow in BigQuery ML looks similarto the following:

1. Create the model using theCREATE MODEL statement.
2. Perform feature preprocessing. Some preprocessing happensautomatically,plus you can usemanual preprocessing functionsinside theTRANSFORM clauseto do additional preprocessing.
3. Refine the model by performinghyperparameter tuning to fit the modelto the training data.
4. Evaluate the model to assess how it mightperform on data outside of the training set, and also to compare it to othermodels if appropriate.
5. Perform inference to analyze data byusing the model.
6. Provideexplainability for the model, toclarify how particular features influenced a given prediction and also themodel overall.
7. Learn more about the components that comprize the model by usingmodel weights.

Because you can use many different kinds of models in BigQuery ML,the functions available for each model vary. For more information aboutsupported SQL statements and functions for each model type, see the followingdocuments:

• End-to-end user journey for generative AI models
• End-to-end user journey for time series forecasting models
• End-to-end user journey for ML models
• End-to-end user journey for imported models
• Contribution analysis user journey