Hyperparameter tuning BigQuery ML
BigQuery ML allows you to quickly train ML models on data in BigQuery. For example, suppose you want to train a linear ML model to predict the duration of a bicycle rental, you can do that using: CREATE OR REPLACE MODEL ch09eu.bicycle_model_linearOPTIONS( model_type='linear_reg', input_label_cols=['duration'])ASSELECT start_station_name, CAST(EXTRACT(DAYOFWEEK FROM start_date) AS STRING) AS dayofweek, CAST(EXTRACT HOUR FROM start_date) AS STRING) AS hourofday, durationFROM `bigquery-public-data.london_bicycles.cycle_hire` BigQuery ML also supports feature engineering that will be automatically repeated during inference.
https://medium.com/google-cloud/hyperparameter-tuning-directly-within-bigquery-ml-a0affb0991ae